Lung cancers are the leading cause of cancer-related mortality worldwide, and the majority of lung cancers are non–small cell lung carcinoma (NSCLC). Overexpressed or activated EGFR has been associated with a poor prognosis in NSCLC. We previously identified a circular noncoding RNA, hsa_circ_0000190 (C190), as a negative prognostic biomarker of lung cancer. Here, we attempted to dissect the mechanistic function of C190 and test the potential of C190 as a therapeutic target in NSCLC. C190 was upregulated in both NSCLC clinical samples and cell lines. Activation of the EGFR pathway increased C190 expression through a MAPK/ERK-dependent mechanism. Transient and stable overexpression of C190 induced ERK1/2 phosphorylation, proliferation, and migration in vitro and xenograft tumor growth in vivo. RNA sequencing and Expression2Kinases (X2K) analysis indicated that kinases associated with cell-cycle and global translation are involved in C190-activated networks, including CDKs and p70S6K, which were further validated by immunoblotting. CRISPR/Cas13a-mediated knockdown of C190 decreased proliferation and migration of NSCLC cells in vitro and suppressed tumor growth in vivo. TargetScan and CircInteractome databases predicted that C190 targets CDKs by sponging miR-142-5p. Analysis of clinical lung cancer samples showed that C190, CDK1, and CDK6 expressions were significantly higher in advanced-stage lung cancer than in early-stage lung cancer. In summary, C190 is directly involved in EGFR–MAPK–ERK signaling and may serve as a potential therapeutic target for the treatment of NSCLC.

Significance:

The circRNA C190 is identified as a mediator of multiple pro-oncogenic signaling pathways in lung cancer and can be targeted to suppress tumor progression.

Lung cancer has consistently been responsible for the highest cancer-related mortality worldwide (1, 2). Despite the tremendous advancement in chemotherapeutic and targeted therapeutic drugs development, lung cancer global mortality rate has almost doubled from 1.03 million cases in 1990 to approximately 1.88 million cases in 2017 (3). Of note, non–small cell lung carcinoma (NSCLC) is responsible for about 85% of lung cancer cases (2). Abnormal activation of EGFR, which may be due to overexpression or kinase activation mutations, is one of the most important driving forces for the malignancy development of cancers (4). Though there have been three generations of EGFR tyrosine kinase inhibitors (TKI) developed and commonly used for clinical treatment of NSCLC, these treatments are nevertheless challenged by innate or acquired drug resistance in less than 12 months of drug administration to patients (5). Theoretically, targeting the major intracellular molecular effectors of EGFR activation might present an alternative to control NSCLC progression in a manner independent of EGFR, which constantly changes its targeted mutation. Despite of developing new TKIs targeting emerging EGFR mutations, elucidating the downstream modulators of the EGFR pathway may give insights to the development of NSCLC therapeutic methods that target EGFR signaling pathways without targeting EGFR protein itself.

Several noncoding RNAs such as long noncoding RNAs (lncRNA), miRNAs, and more recently characterized circular RNAs (circRNA) have been extensively studied and proven to partake in biological processes during healthy or disease states. CircRNAs are the products of noncanonical pre-mRNA splicing called backsplicing that results in the formation of a closed loop RNA transcript (6). RNA-binding proteins (RBP) were discovered to mediate circRNA biogenesis whereas RNase L mediates protein kinase R–bound circRNAs degradation (7, 8). CircRNA dysregulation has been implicated in various pathological processes such as lung cancer development (9). For example, circPRKC1 was found to promote lung adenocarcinoma (10) and circSDHC was shown to promote proliferation and metastasis of renal cell carcinoma (11). Likewise, several studies have implicated circRNA hsa_circ_0000190 in gastric cancer (12), myeloma (13), and lung adenocarcinoma (14). On the same note, Liu and colleagues (15) also reported that epigenetic modification of circIGF2BP3 promotes immune evasion in NSCLC. We previously reported an elevated expression level of circRNA hsa_circ_0000190 (C190) in the blood samples of patients with advanced-stage lung cancer, and that the persistent expression of C190 in blood may predict a poor outcome of anti-PD-L1 treatment in patients (14). However, the functional involvements of C190 in lung cancer progression are unclear. In view of understanding circRNA-mediated oncogenic development of lung cancer, it is important to dissect the roles of C190 in modulating key oncogenic pathways in lung cancer.

Clustered regularly interspaced short palindromic repeats (CRISPR) and CRISPR-associated protein 13 (Cas13) techniques can precisely edit RNA and have recently been discovered to possess great potential for targeting pathologically overexpressed circRNAs (16). The CRISPR/Cas13 system was originally identified to be a part of the CRISPR–Cas-adaptive immunity of microbes that defends them against foreign DNA or RNA through its RNA interference ability. It belongs to the class 2 type-VI-A and was previously known as c2c2 (16). Cas13 proteins, comprising Cas13a, b, c, and d, are ribonuclease enzymes that complex with a short CRISPR RNA (crRNA) to target single-stranded RNA transcripts complementary to the crRNA sequence, and thus facilitate their degradation (16, 17). More interesting is the fact that CRISPR/Cas13 system-mediated RNA degradation is characterized by little to no off-target effects with collateral cleavage (16). Its extremely precise ability to detect nucleic acid transcripts and target them for degradation has been harnessed in various nonimmunity-related applications (18). In mammalian systems, Cas13-mediated RNAi therapeutic-related potential has been studied for diseases such as malaria, HIV, and several cancer types, including bladder, hematologic, pancreatic, cervical, and brain cancers (18). To date, all Cas13-mediated detection and RNAi studies focused on the detection and/or depletion of mRNAs (19), miRNAs or lncRNAs (20). In a recent study, CRISPR/Cas13d was used to target oncogenic circFAM120A and several other circRNAs in colon adenocarcinoma, lymphoma, as well as normal human cell-lines (21). Likewise, Zhang and colleagues (22) also used CRISPR/Cas13d to knock down circZKSCAN1 in hepatocellular carcinoma (HCC). This evidence is supportive for the CRISPR/Cas13 system to be used as a targeting mechanism for circRNAs (21, 22).

In the current study, we report the characterization of C190 circRNA as an important mediator of EGFR-induced signaling in NSCLC. Specifically, C190 was shown to regulate the MAPK/ERK pathway regulated by EGFR activation and to be involved in such pathological processes as increased proliferation, migration/invasion, and in vivo tumor growth, thus making it a promising target for NSCLC therapy. Consequently, we successfully applied the CRISPR/Cas13 system to knock down C190 expression in cells, and demonstrated that such knockdown significantly curbed its tumorigenic properties in vitro and in vivo. Invariably, our data reported the roles of C190 in the EGFR-mediated tumor progression pathway, and further support that CRISPR/Cas13 system could be a novel strategy to target oncogenic circRNAs in lung cancer.

Patients' tissue sample collection

Primary lung adenocarcinoma and adjacent normal tissues were collected from patients who had undergone tumor resection at the Department of Thoracic Surgery, Taiwan Veterans General Hospital (TVGH; Taipei, Taiwan). We first enrolled 21 patients with lung cancer to compare the expression of C190 in the NSCLC tumor samples and adjacent normal lung tissue. To compare the expression of C190 and its downstream microRNAs and kinases among patients with lung cancer of different stages, we further enrolled additional 23 patients with lung cancer for comparing the molecule expression in early-stage and advanced-stage lung cancer samples. Experienced pathologists confirmed all tumor samples. Written informed consent was obtained from all patients. Collection of human tissue samples was conducted in accordance with the International Ethical Guidelines for Biomedical Research Involving Human Subjects. This study was approved by the Ethics Committee of the National Yang-Ming University. For the detail of other Materials and Methods, please refer to the Supplementary Methods and Materials and Supplementary Tables S1–S6 in the Supplementary Information.

Plasmid construction and cloning

C190-coding sequence (exons 3 and 4 of the CNIH4 gene) was designed to be flanked by Alu repeat sequences as per Kramer and colleagues (7) 2015 containing restriction sites KpnI and EcoRI, then was subcloned into pcDNA3.1 vector (RRID:Addgene_79663). Subsequently, KpnI and EcoRI sites were used to clone synthesized gene into pcDNA(3.1+) ZKSCAN1 MCS vector to make pcDNA(3.1+)ZKSCAN1-C190 circRNA overexpression plasmid. To make stable clone cell lines, C190 with Alu repeats was amplified from pcDNA(3.1+)ZKSCAN1-C190 and cloned into pLAS2w.Ppuro lentiviral vector (Academia Sinica). The plasmids used in this study are listed in Supplementary Table S2.

RNA pulldown assay

RNA pulldown was carried out according to the modified protocol by Bai and colleagues (23). Briefly, A549 monolayer cell culture was washed in 1X PBS and subsequently crosslinked by incubating with 1% paraformaldehyde for 10 minutes at room temperature followed by inactivation using 1.25 mol/L glycine addition. Cells were washed in 1X PBS and collected by scraping with PBS before lysis in RNA pulldown lysis buffer. Streptavidin magnetic beads were pre-washed and RNase-inactivated before 2 hours binding with biotinylated C190 targeting or luciferase control denatured probes. Probe beads conjugate was then incubated overnight with 2–4 mg protein lysate overnight. Several washings of the RNA pulldown complex were done before final elution in elution buffer or SDS loading dye for downstream analysis of RNA or protein. miRNA extraction was performed using the miReasy Mini Kit (QIAGEN) according to the manufacturer's protocol. The sequence of RIP probes is listed in Supplementary Table S3.

RNA-FISH

RNA-FISH was done using a modified protocol by Wang and colleagues (24). Briefly, 4–6-μm thick sections of tumor, neighboring normal tissue or xenograft tumor samples were cut from either frozen specimens or FFPE (formalin-fixed, paraffin-embedded) samples. Frozen sections were fixed in acetone, air-dried, and subsequently bathed in TBS solution (pH 7.6). FFPE tissue sections were first melted on a heating block before deparaffination and rehydrated in a series of xylene and gradient alcohol concentrations. Subsequent proteinase K (0.6 U/mL) treatment was performed for 20 minutes at 37°C in a humidified chamber. Samples were subsequently incubated overnight with 200 nmol/L FAM and Cy3-tagged fluorescent probe (IDT) at 42°C followed by stringent washing with 2X SSC buffer. Slides were then immersed in 100 μg/mL RNase A diluted in 2X SSC at 37°C for 30 minutes, washed in 1X PBS and then incubated in 1X PBS containing DAPI (1:10,000 dilution). Slides were finally rinsed in 1X PBS followed by mounting in mounting medium before imaging. The sequence of the RNA-FISH probes are listed in Supplementary Table S4

Animals and tumor cell transplantation

All animals used in this study were bred and maintained according to the Guidelines for Laboratory Animal Welfare in the Taipei Veterans General Hospital under the supervision of Department of Medical Research of Taipei Veterans General Hospital. For subcutaneous xenograft mouse model, the NSCLC cell lines were harvested, washed, suspended in PBS and subjected to subcutaneous implantation into the dorsolateral side of the flank region of 8-week-old male BALB/c nude mice (National Laboratory Animal Center, Taipei, Taiwan). Tumor size in the subcutaneous xenograft model was measured every two days using a caliper. The average tumor volume was calculated using the following equation: V = A*B2*0.5 (A, long diameter; B, short diameter). Xenograft tumors were removed, imaged, and weighed after euthanizing the animals at the end of the study. Tumor samples were then processed for histopathological analysis.

Hsa_circ_0000190 is overexpressed in NSCLC clinical tissue samples and cell lines

EGFR overexpression or activated EGFR pathway has been found to be highly associated with NSCLC with distal prognosis. Recently, by analyzing the transcriptomes of lung cancer cell lines using next-generation sequencing, we identified C190 as one of the most significantly upregulated circRNAs in lung cancer cell lines. Moreover, C190 was found to be detectable in lung cancer patients' blood samples, and elevated C190 levels were highly associated with tumor size, aggressiveness, metastasis, and low survival rate in patients with advanced lung cancers (14). However, in addition to being a noninvasive biomarker for advanced lung cancers, it remains unclear whether C190 plays a direct role in the EGFR signaling cascade and can be used as a therapeutic target in lung cancers. In the current study, we first enrolled 21 patients with lung cancer and compared the expression of C190 in the NSCLC tumor samples and adjacent normal lung tissue by FISH and determined that it mainly resided in the cytoplasm and its expression was markedly higher in the cancerous tissue (Fig. 1A). Similarly, by using droplet digital RT-PCR (RT-ddPCR), C190 expression was determined to be significantly higher in tumor samples as compared with normal samples, whereas the expression of CNIH4 mRNA, the linear product of C190-encoding gene, did not significantly change between tumor and normal samples (Fig. 1B). CircRNAs are uniquely known to be resistant to exonuclease degradation by RNase R (7). Therefore, we treated the total RNA extracted from NSCLC clinical samples with RNase R and identified that C190 expression was enriched after such treatment, whereas the expression of CNIH4 mRNA was depleted, thus corroborating the circular nature of C190 (Fig. 1C). Furthermore, we applied qRT-PCR to validate the expression of C190 in in vitro models of lung carcinoma represented by the cell lines A549, HCC827, and H1299. C190 was significantly overexpressed in all of these cell lines as compared with normal bronchial epithelial cells, BEAS-2B, particularly in HCC827, which is characterized by the constitutively active tyrosine kinase domain of EGFR (Fig. 1D). Of note, the divergent pair of primers that specifically amplifies the backspliced junction of circRNA form was used, as was validated by successful amplification from the cDNA rather than genomic DNA (gDNA) of BEAS-2B, A549 and HCC827 cells (Fig. 1E). Furthermore, when such amplicon from A549 cell line was subjected to Sanger sequencing, it demonstrated a perfect match with the C190 backspliced junction sequence from the circBase database (http://www.circbase.org/; Fig. 1F). As was demonstrated by actinomycin D treatment, C190 exhibited much higher stability than its linear CNIH4 mRNA counterpart in A549 cells (Fig. 1G). To summarize, these data indicated that C190 circRNA, rather than the linear product of the CNIH4 gene, was upregulated in lung cancer clinical samples and cell lines.

Figure 1.

Hsa_circ_0000190 (C190) is overexpressed in NSCLC clinical tissue samples and cell lines. A, RNA-FISH analysis showing the expression of C190 in two clinical samples of NSCLC tissue and paired normal lung tissue. Tissue morphology shown by hematoxylin and eosin (H&E) staining and nuclei stained with DAPI. B, Droplet digital RT-PCR (RT-ddPCR) analysis of the expression of C190 (left) and CNIH4 linear mRNA (right) in NSCLC tissues and paired normal lung tissues. Data are shown as means (N = 21) with SD error bars; **, P < 0.01, ns, not significant (Student t test). C, qRT-PCR analysis of the expression of C190 and CNIH4 linear mRNA in clinical samples of NSCLC with or without prior treatment of total RNA with RNase R. Mean fold changes relative to –RNase R are shown with SD error bars (N = 2); **, P < 0.01 (Student t test). D, qRT-PCR analysis (left) of the expression of C190 in BEAS-2B, A549, HCC827 and H1299 cell lines. Mean fold changes relative to BEAS-2B are shown with SD error bars (N = 2) *, P < 0.05; **, P < 0.01 (Student t test). The EGFR expression and activation status in the indicated cell lines are shown by immunoblotting (right). E, RT-PCR electrophoregram showing amplification of C190 with the divergent primers from cDNA, but not gDNA of the indicated cell lines. F, Sanger sequencing of C190 RT-PCR fragment amplified from A549 cells aligned to C190 sequence retrieved from the circBase database. G, qRT-PCR analysis of the stability of C190 and CNIH4 transcripts in A549 cells treated with actinomycin D for the indicated time.

Figure 1.

Hsa_circ_0000190 (C190) is overexpressed in NSCLC clinical tissue samples and cell lines. A, RNA-FISH analysis showing the expression of C190 in two clinical samples of NSCLC tissue and paired normal lung tissue. Tissue morphology shown by hematoxylin and eosin (H&E) staining and nuclei stained with DAPI. B, Droplet digital RT-PCR (RT-ddPCR) analysis of the expression of C190 (left) and CNIH4 linear mRNA (right) in NSCLC tissues and paired normal lung tissues. Data are shown as means (N = 21) with SD error bars; **, P < 0.01, ns, not significant (Student t test). C, qRT-PCR analysis of the expression of C190 and CNIH4 linear mRNA in clinical samples of NSCLC with or without prior treatment of total RNA with RNase R. Mean fold changes relative to –RNase R are shown with SD error bars (N = 2); **, P < 0.01 (Student t test). D, qRT-PCR analysis (left) of the expression of C190 in BEAS-2B, A549, HCC827 and H1299 cell lines. Mean fold changes relative to BEAS-2B are shown with SD error bars (N = 2) *, P < 0.05; **, P < 0.01 (Student t test). The EGFR expression and activation status in the indicated cell lines are shown by immunoblotting (right). E, RT-PCR electrophoregram showing amplification of C190 with the divergent primers from cDNA, but not gDNA of the indicated cell lines. F, Sanger sequencing of C190 RT-PCR fragment amplified from A549 cells aligned to C190 sequence retrieved from the circBase database. G, qRT-PCR analysis of the stability of C190 and CNIH4 transcripts in A549 cells treated with actinomycin D for the indicated time.

Close modal

C190 expression is predominantly driven by MAPK/ERK pathway activation

In lung cancer, the activation of EGFR by EGF or by catalytic domain mutation leads to the activation of several downstream signaling pathways, including MAPK/ERK, PI3K/Akt, and JAK/STAT, resulting in pro-oncogenic effects such as increased proliferation, resistance to apoptosis, increased migration/invasion potential (Fig. 2A). Given our observation that C190 was more upregulated in the HCC827 cell line with constitutive activation of EGFR tyrosine kinase domain as compared with nonmutated A549 cells (Fig. 1D; ref. 14), we opted to dissect the role of EGFR activation and its downstream pathways in C190 expression. Indeed, upon treatment of A549 and H1299 cells with EGF in a time course of 15 minutes to 3 hours, EGFR was activated as was monitored by Western blot detection of Tyr1068 phosphorylation, with maximum activation at 15–30 minutes (Fig. 2B and C). The downstream MAPK/ERK and PI3K/Akt signaling pathways were activated with a similar pattern as determined by detecting phosphorylation of ERK1/2 on Thr202/Tyr204 and Akt on Ser473, respectively (Fig. 2B and C). On the other hand, the JAK/STAT pathway was not activated by treatment of A549 and H1299 with EGF, as no changes of STAT3 Tyr705 phosphorylation were observed (Fig. 2C and D). In HCC827 cells, EGFR, Akt, ERK1/2, but not STAT3, were abundantly phosphorylated as compared with A549 cells not treated with EGF (Fig. 2BD). Furthermore, we have shown by qRT-PCR that stimulation of EGFR in A549 and H1299 cells with EGF led to increased expression of C190 (Fig. 2E and F). Interestingly, whereas the activation of EGFR and its downstream targets (Akt and ERK1/2) peaked at earlier stages of the time course of treatment with EGF (15–60 minutes) and then receded, the induction of C190 expression persisted at later time points (60–180 minutes; Fig. 2E and F). To further dissect which major intracellular signaling pathway is responsible for EGF-dependent C190 expression, small-molecule inhibitors, LY294002 and PD98059, were used to block PI3K and MEK1/2 pathways, respectively, in the absence or presence of EGF stimulation (Fig. 2G and H). The block of MEK1/2, but not PI3K, significantly reduced C190 expression in A549 and H1299 cells treated with EGF for 30 minutes (Fig. 2I and J). Meanwhile, neither the expression of C190′s parental gene-encoded linear mRNA (Fig. 2K and L) nor the expression of CNIH4 protein encoded by it (Fig. 2G and H) were significantly affected by modulation of these pathways. Consistently, the blockade of EGFR tyrosine kinase domain with gefitinib blocked EGF-induced upregulation of C190, indicating that EGFR is indeed upstream of MAPK/ERK-dependent regulation of C190 expression (Fig. 2M). To summarize, we showed that C190 is upregulated by EGFR activation via the MAPK/ERK pathway in A549 and H1299 cell lines.

Figure 2.

C190 expression is predominantly driven by MAPK/ERK pathway activation. A, Schematic of downstream signaling pathways induced by EGFR activation. The inhibitors used in this study are shown in red font. B–D, Western blot analysis of the expression of EGFR, p-EGFR, and the indicated components of downstream MAPK/ERK, PI3K/Akt (B) and JAK/STAT (C and D) pathways and their phosphorylated forms in A549 (B and D) and H1299 (C) cells treated with EGF. GAPDH (B and C) and β-actin (D) were used as loading controls. E and F, qRT-PCR analysis of the expression of C190 in A549 (E) and H1299 (F) cells treated with EGF. Data are shown as the mean relative values from three independent experiments with SD error bars; *, P < 0.05; ns, not significant (Kruskal–Wallis test). G and H, Western blot analysis of the expression of EGFR, p-EGFR, and the indicated components of downstream MAPK/ERK and PI3K/Akt pathways and their phosphorylated forms in A549 (G) and H1299 (H) cells treated with EGF and small inhibitors of PI3K (LY294002) and MEK1/2 (PD98059). GAPDH was used as a loading control. I–L, qRT-PCR analysis of the expression of C190 (I and J) and CNIH4 mRNA (K and L) in A549 (I and K) and H1299 (J and L) cells treated with EGF and small inhibitors of PI3K (LY294002) and MEK1/2 (PD98059). Data are shown as the mean relative values from three independent experiments with SD error bars; *, P < 0.05; **, P < 0.01; ns, not significant (Student t test). M, qRT-PCR analysis of the expression of C190 in A549 cells treated with EGF and inhibitors of EGFR (gefitinib) and MEK1/2 (PD98059). Data shown as the mean relative values from three independent experiments with SD error bars, **, P < 0.01; ***, P < 0.001 (Kruskal–Wallis test).

Figure 2.

C190 expression is predominantly driven by MAPK/ERK pathway activation. A, Schematic of downstream signaling pathways induced by EGFR activation. The inhibitors used in this study are shown in red font. B–D, Western blot analysis of the expression of EGFR, p-EGFR, and the indicated components of downstream MAPK/ERK, PI3K/Akt (B) and JAK/STAT (C and D) pathways and their phosphorylated forms in A549 (B and D) and H1299 (C) cells treated with EGF. GAPDH (B and C) and β-actin (D) were used as loading controls. E and F, qRT-PCR analysis of the expression of C190 in A549 (E) and H1299 (F) cells treated with EGF. Data are shown as the mean relative values from three independent experiments with SD error bars; *, P < 0.05; ns, not significant (Kruskal–Wallis test). G and H, Western blot analysis of the expression of EGFR, p-EGFR, and the indicated components of downstream MAPK/ERK and PI3K/Akt pathways and their phosphorylated forms in A549 (G) and H1299 (H) cells treated with EGF and small inhibitors of PI3K (LY294002) and MEK1/2 (PD98059). GAPDH was used as a loading control. I–L, qRT-PCR analysis of the expression of C190 (I and J) and CNIH4 mRNA (K and L) in A549 (I and K) and H1299 (J and L) cells treated with EGF and small inhibitors of PI3K (LY294002) and MEK1/2 (PD98059). Data are shown as the mean relative values from three independent experiments with SD error bars; *, P < 0.05; **, P < 0.01; ns, not significant (Student t test). M, qRT-PCR analysis of the expression of C190 in A549 cells treated with EGF and inhibitors of EGFR (gefitinib) and MEK1/2 (PD98059). Data shown as the mean relative values from three independent experiments with SD error bars, **, P < 0.01; ***, P < 0.001 (Kruskal–Wallis test).

Close modal

C190 overexpression activates ERK1/2 and promotes tumor cell growth and migration in vitro

Here, we found a close link between the pro-oncogenic EGFR pathway and C190 circRNA. To further characterize the functional effects of C190 and its molecular mechanisms in lung cancer, we performed its transient overexpression in A549 cells (Fig. 3A). Interestingly, whereas the overexpression of C190 did not have any effect on EGFR activation (p-EGFR; Fig. 3B and C), it nevertheless resulted in increased phosphorylation of ERK1/2 (Fig. 3C and D). Overexpression of hsa_circ_0001649 (C1649), another circRNA previously identified by us as a biomarker of lung cancer (14), did not have any effect on either EGFR or ERK1/2 activation (Fig. 3BD). When C190 was overexpressed in the HCC827 cell line, it also noticeably increased the phosphorylation level of ERK1/2, which was already high due to constitutively activated EGFR (Fig. 3D and F). For further studies, we selected stable clones of A549, H1299, and HCC827 stably overexpressing C190 (Supplementary Fig. S1). As increased proliferation is the major consequence of MAPK/ERK pathway activation, we measured the proliferation rate of A549 and HCC827 cells overexpressing C190 by the Alamar Blue assay. Indeed, we found that overexpression of C190 resulted in increased proliferation rate in both cell lines (Fig. 3G and H). Moreover, we found that ERK1/2 was significantly phosphorylated in C190-overexpressing clones of H1299 and A549 cells (Fig. 3IK). Furthermore, it was found that stable overexpression of C190 resulted in increased cell migration as was tested by wound healing (Fig. 3L and M) and Transwell migration (Fig. 3N and O) assays in A549 and H1299 cell lines. To summarize, we have shown that whereas C190 expression is driven by MAPK/ERK pathway activation (Fig. 2), it can also augment MAPK/ERK activation when overexpressed and, at the same time, C190 overexpression increases proliferation and migration capacity of lung cancer cells.

Figure 3.

C190 overexpression activates ERK1/2 and promotes tumor cell growth and migration in vitro. A, qRT-PCR showing overexpression of C190 in transiently transfected A549 cells. B–F, Western blot analysis of the expression of the indicated components of the MAPK/ERK pathway upon transient overexpression of C190 or C1649 circRNAs in A549 (B–D) and HCC827 (E and F) cells. GAPDH (B) and β-actin (C and D) were used as loading controls. D and F, The densitometry quantification of p-ERK1/2 to total ERK1/2 ratio in A549 and HCC827, respectively. G and H, Alamar Blue assay demonstrating the dynamics of growth of A549 (G) and HCC827 (H) clones overexpressing C190 as compared with the vector-transfected stable clones. Mean values (N = 3) are shown with SD error bars; **, P < 0.001; ns, not significant (Student t test). I–K, Western blot analysis of the expression of the indicated components of the MAPK/ERK pathway in stable C190-overexpressing clones of A549 (I) and H1299 (J) cell lines and the vector-transfected stable clones. GAPDH was used as loading control. K, Densitometry quantification of the expression of p-ERK1/2 in A549 (top) and H1299 (bottom) clones, respectively. Data expressed as mean ratios of p-ERK1/2 to total ERK1/2, N = 3; **, P < 0.01 (t test). L, Wound-healing assay showing the migration capacity of A549 and H1299 vector-transfected and C190-expressing stable clones. M, Quantification of wound closure in L at 24 hours time point. Mean values with SD error bars are shown, N = 3; *, P < 0.05 (Student t test). N and O, Transwell migration assay showing the migration capacity of A549 (N) and H1299 (O) vector-transfected and C190-expressing stable clones. Left, representative images of migrated cells; right, quantification of the number of migrated cells. Mean values with SD error bars are shown N = 3; *, P < 0.05 (Student t test).

Figure 3.

C190 overexpression activates ERK1/2 and promotes tumor cell growth and migration in vitro. A, qRT-PCR showing overexpression of C190 in transiently transfected A549 cells. B–F, Western blot analysis of the expression of the indicated components of the MAPK/ERK pathway upon transient overexpression of C190 or C1649 circRNAs in A549 (B–D) and HCC827 (E and F) cells. GAPDH (B) and β-actin (C and D) were used as loading controls. D and F, The densitometry quantification of p-ERK1/2 to total ERK1/2 ratio in A549 and HCC827, respectively. G and H, Alamar Blue assay demonstrating the dynamics of growth of A549 (G) and HCC827 (H) clones overexpressing C190 as compared with the vector-transfected stable clones. Mean values (N = 3) are shown with SD error bars; **, P < 0.001; ns, not significant (Student t test). I–K, Western blot analysis of the expression of the indicated components of the MAPK/ERK pathway in stable C190-overexpressing clones of A549 (I) and H1299 (J) cell lines and the vector-transfected stable clones. GAPDH was used as loading control. K, Densitometry quantification of the expression of p-ERK1/2 in A549 (top) and H1299 (bottom) clones, respectively. Data expressed as mean ratios of p-ERK1/2 to total ERK1/2, N = 3; **, P < 0.01 (t test). L, Wound-healing assay showing the migration capacity of A549 and H1299 vector-transfected and C190-expressing stable clones. M, Quantification of wound closure in L at 24 hours time point. Mean values with SD error bars are shown, N = 3; *, P < 0.05 (Student t test). N and O, Transwell migration assay showing the migration capacity of A549 (N) and H1299 (O) vector-transfected and C190-expressing stable clones. Left, representative images of migrated cells; right, quantification of the number of migrated cells. Mean values with SD error bars are shown N = 3; *, P < 0.05 (Student t test).

Close modal

Overexpression of C190 increases tumor growth in vivo

After demonstrating that C190 overexpression enhances pro-oncogenic properties of NSCLC cells by modulating the MAPK/ERK pathway, we proceeded to investigate whether it would have pathological consequences in vivo. For this purpose, A549 cells stably overexpressing C190 were xenografted into immunocompromised mice, and the tumor growth was monitored in vivo during the course of four weeks (Fig. 4A). The tumors derived from C190-overexpressing A549 cells demonstrated higher growth rate than that of tumors derived from the empty vector-transfected control cells (Fig. 4B). At the final day of the time course, the mass of C190-overexpressing tumors was two times bigger than that of the control tumors (Fig. 4C and D). Overexpression of C190 in the tumors at the final day of the time course was demonstrated by RNA FISH of the tumor cross sections (Fig. 4E). Moreover, immunofluorescent staining of the tumor cross sections revealed higher expression of the proliferative antigen Ki-67 in C190-overexpressing tumors, further corroborating higher proliferative status of such tumors (Fig. 4F and G). To summarize, consistently with its pro-oncogenic effect in vitro, C190 overexpression increased tumor growth in vivo.

Figure 4.

Overexpression of C190 increases tumor growth in vivo. A, Schematic showing the experimental design. A549 vector and A549-C190 cells were injected into BALB/c nude mice. At day 8 after injection, tumor size started to be measured by external palpation until day 31, when mice were sacrificed for histological analysis. B, Dynamics of growth of tumors derived from A549_vector and A549_C190 cells. Mean tumor volumes are shown (N = 4); **, P < 0.01 (Student t test). C, Side-by-side photographs (top) and GFP fluorescent images (bottom) of the tumors from the indicated experimental groups. D, Mean tumor masses at the day of sacrifice, **, P < 0.01. E, RNA-FISH analysis of the expression of C190 in the tumors derived from A549 vector and A549-C190 cells. Nuclei stained with DAPI. F, Immunofluorescent staining of Ki-67 antigen in the tumors derived from A549 vector and A549-C190 cells. Tissue morphology shown by hematoxylin and eosin (H&E) staining. G, Quantification of the number Ki-67–positive cells; **, P < 0.01.

Figure 4.

Overexpression of C190 increases tumor growth in vivo. A, Schematic showing the experimental design. A549 vector and A549-C190 cells were injected into BALB/c nude mice. At day 8 after injection, tumor size started to be measured by external palpation until day 31, when mice were sacrificed for histological analysis. B, Dynamics of growth of tumors derived from A549_vector and A549_C190 cells. Mean tumor volumes are shown (N = 4); **, P < 0.01 (Student t test). C, Side-by-side photographs (top) and GFP fluorescent images (bottom) of the tumors from the indicated experimental groups. D, Mean tumor masses at the day of sacrifice, **, P < 0.01. E, RNA-FISH analysis of the expression of C190 in the tumors derived from A549 vector and A549-C190 cells. Nuclei stained with DAPI. F, Immunofluorescent staining of Ki-67 antigen in the tumors derived from A549 vector and A549-C190 cells. Tissue morphology shown by hematoxylin and eosin (H&E) staining. G, Quantification of the number Ki-67–positive cells; **, P < 0.01.

Close modal

C190 overexpression activates cell-cycle and global translation-associated kinases

Because we identified the functional link between C190 and the MAPK/ERK pathway, and given the fact that any modulation of the MAPK/ERK pathway causes massive scale transcriptional response, we next aimed to investigate how C190 overexpression would affect the transcriptome. C190 was transiently overexpressed in A549 cells and these cells were subjected to the RNA-seq analysis. As compared with control A549 cells transfected with an empty vector, 600 genes were upregulated (FC>2) and 447 genes were downregulated (FC<0.5) in C190-overexpressing A549 cells (Fig. 5A). To identify the pathways that may be modulated by C190 to regulate the expression of these genes, we performed eXpression2Kinases (X2K) network analysis (25). Apart from the components of the MAPK/ERK signaling pathway, X2K analysis predicted the enrichment of cyclin-dependent kinases (CDK) such as CDK1, CDK2, and CDK4, which are important for G1–S and G2–M phase transitions of the cell cycle (Fig. 5B; ref. 26). Further validation by immunoblotting confirmed the increased expression of CDK1, 4 and 6, as well as hyperphosphorylation of their principal target, retinoblastoma (Rb) protein, during C190 overexpression (Fig. 5C). Moreover, consistently with the fact that elevated cell proliferation is associated with elevated global translation rates, we also observed increased phosphorylation of p70S6K and its target, ribosomal protein S6 (RPS6), which normally constitute an effector module linking EGFR activation with modulation of global translation (Fig. 5D). To corroborate the potential clinical relevance of regulation of CDKs by C190 in lung cancer, we analyzed The Cancer Genome Atlas Lung Adenocarcinoma Dataset (TCGA-LUAD) and found that higher levels of CDK1 and CDK6 were clearly associated with poorer survival of patients with NSCLC (Fig. 5E). To summarize, we demonstrate that C190 overexpression results in activation of kinases associated with cell-cycle and global translation, and the networks of these kinases are associated with global changes in gene expression.

Figure 5.

C190 overexpression activates cell-cycle and global translation-associated kinases. A, Diagram showing the number of upregulated (FC > 2) and downregulated genes (FC<0.5) in A549 cells overexpressing C190 among the total number of the expressed genes. B, X2K network analysis showing the kinases predicted to regulate the differentially expressed genes in A. Data are expressed as enrichment scores of −log10(P value). C and D, Western blot analysis of the expression of the indicated CDKs (C) and components of the MAPK/ERK signaling pathway and total and phosphorylated p70S6K and RPS6 proteins (D). C, Bottom, densitometry quantification of the CDK signals in C190-overexpressing cells relative to the control. Mean relative values (N = 3) are shown with SD error bars; *, P < 0.05 (t test). E, Survival curves of patients with high or low levels of the indicated CDKs from the TCGA-LUAD dataset.

Figure 5.

C190 overexpression activates cell-cycle and global translation-associated kinases. A, Diagram showing the number of upregulated (FC > 2) and downregulated genes (FC<0.5) in A549 cells overexpressing C190 among the total number of the expressed genes. B, X2K network analysis showing the kinases predicted to regulate the differentially expressed genes in A. Data are expressed as enrichment scores of −log10(P value). C and D, Western blot analysis of the expression of the indicated CDKs (C) and components of the MAPK/ERK signaling pathway and total and phosphorylated p70S6K and RPS6 proteins (D). C, Bottom, densitometry quantification of the CDK signals in C190-overexpressing cells relative to the control. Mean relative values (N = 3) are shown with SD error bars; *, P < 0.05 (t test). E, Survival curves of patients with high or low levels of the indicated CDKs from the TCGA-LUAD dataset.

Close modal

CRISPR/Cas13a-mediated C190 knockdown inhibits MAPK activation, growth, and survival of NSCLC cells

RNA targeting by RNAi-based methods, such as siRNA, miRNA, and shRNA, is the most widely used approach to knock down gene expression on RNA level (27). This strategy has been successfully used to knock down mRNAs and noncoding RNAs such as lncRNAs and even circRNAs. One major drawback of RNAi is its relatively high rate of off-target effects (28). Therefore, in this study, we used more precise CRISPR/Cas13a RNA editing system guided by single-stranded RNA, called crRNA, to precisely target C190 and increase C190 degradation through Cas13a RNA endonuclease activity (29). Two crRNAs were designed to specifically target C190 at its backspliced junction (Fig. 6A). Indeed, their transfection into Cas13a-expressing stable clone of A549 cells led to lower expression of C190 (Fig. 6B; Supplementary Table S6). Cas13a-mediated C190 repression resulted in significantly reduced MAPK/ERK pathway activation manifested in decreased phosphorylation of ERK1/2 and RPS6 in A549 (Fig. 6C) and HCC827 (Fig. 6D) cells. Interestingly, the phosphorylation of RPS6 associated with global translation regulation was also reduced upon C190 knockdown in both cell lines (Fig. 6C and D). Given that HCC827 cells are characterized by constitutively active EGFR, these effects of C190 knockdown on ERK1/2 and RPS6 were independent of EGFR activation status. The biological phenotype resulting from Cas13a-mediated C190 repression included lower proliferation rate shown in both A549 and HCC827 cell lines (Fig. 6E and F), decreased migration (Fig. 6G), and anchorage-free survival (Fig. 6H and I). To summarize, we have shown that the knockdown of C190 resulted in suppression of pro-oncogenic properties of NSCLC cells, which was an opposite effect as compared with its overexpression.

Figure 6.

CRISPR/Cas13a-mediated C190 knockdown inhibits MAPK activation, growth, and survival of NSCLC cells. A, Schematic diagram showing targeting of C190 by CRISPR/Cas13a and crRNAs. B, Top, qRT-PCR analysis of the expression of C190 in A549 cells stably expressing Cas13a upon knockdown with crRNAs 1 and 2. Bottom, Western blot showing the expression of Cas13a in transfected A549 cells. C and D, Western blot analysis of the expression of the indicated components of MAPK/ERK signaling pathway in A549 (C) and HCC827 (D) cells transfected with the indicated crRNAs. Right, densitometry quantification of the expression of p-ERK1/2. Data expressed as mean ratios of p-ERK1/2 to total ERK1/2, N = 3, *, P < 0.05; **, P < 0.01; ***, P < 0.001 (one-way ANOVA test). GAPDH was used as loading control. E and F, Alamar Blue assay demonstrating the dynamics of growth of A549 (E) and HCC827 (F) cells transfected with the indicated crRNAs. Mean values (N = 3) are shown with SD error bars; *, P < 0.05; n.s., not significant (Kruskal–Wallis test). G, Wound-healing assay showing the migration capacity of Cas13a-expressing A549 cells transfected with the indicated crRNAs. Left, representative wound images; right, quantification of wound closure at 48 hours time point. Mean values with SD error bars are shown, N = 3; *, P < 0.05 (ANOVA test). H and I, Colony formation assay of A549 (H) and HCC827 (I) cells transfected with the indicated crRNAs; *, P < 0.05; **, P < 0.01.

Figure 6.

CRISPR/Cas13a-mediated C190 knockdown inhibits MAPK activation, growth, and survival of NSCLC cells. A, Schematic diagram showing targeting of C190 by CRISPR/Cas13a and crRNAs. B, Top, qRT-PCR analysis of the expression of C190 in A549 cells stably expressing Cas13a upon knockdown with crRNAs 1 and 2. Bottom, Western blot showing the expression of Cas13a in transfected A549 cells. C and D, Western blot analysis of the expression of the indicated components of MAPK/ERK signaling pathway in A549 (C) and HCC827 (D) cells transfected with the indicated crRNAs. Right, densitometry quantification of the expression of p-ERK1/2. Data expressed as mean ratios of p-ERK1/2 to total ERK1/2, N = 3, *, P < 0.05; **, P < 0.01; ***, P < 0.001 (one-way ANOVA test). GAPDH was used as loading control. E and F, Alamar Blue assay demonstrating the dynamics of growth of A549 (E) and HCC827 (F) cells transfected with the indicated crRNAs. Mean values (N = 3) are shown with SD error bars; *, P < 0.05; n.s., not significant (Kruskal–Wallis test). G, Wound-healing assay showing the migration capacity of Cas13a-expressing A549 cells transfected with the indicated crRNAs. Left, representative wound images; right, quantification of wound closure at 48 hours time point. Mean values with SD error bars are shown, N = 3; *, P < 0.05 (ANOVA test). H and I, Colony formation assay of A549 (H) and HCC827 (I) cells transfected with the indicated crRNAs; *, P < 0.05; **, P < 0.01.

Close modal

CRISPR/Cas13a-mediated knockdown of C190 suppresses tumor growth in vivo

To explore the effects of CRISPR/Cas13a-mediated C190 knockdown in in vivo tumor xenograft model, A549 cells stably expressing Cas13a with a GFP marker were transfected with control nontargeting crRNA (NT_crRNA) or two C190-targeting crRNAs (crRNA1 and crRNA2; Fig. 7A; Supplementary Table S6). These cells were subcutaneously injected into immunocompromised mice followed by subsequent intratumor crRNA injection every four days 3 times (Fig. 7A). Following that, the tumor growth was monitored in the course of the next two weeks, and it was determined that the tumors derived from C190-targteting crRNA-transfected cells grew with a slower rate as compared with the control tumors (Fig. 7B). Likewise, terminal tumor size and weight were also significantly reduced in the C190-targeting crRNA group (Fig. 7CE). Both RT-ddPCR (Fig. 7F) and RNA-FISH (Fig. 7G) analyses confirmed lower expression of C190 in the final tumors of C190-targeting group. Moreover, the expression of Ki-67 proliferation marker was markedly lower in the tumors subjected to C190 knockdown as compared with NT-crRNA and untransfected controls (Fig. 7F and I). To summarize, we successfully knocked down C190 by CRISPR/Cas13a in in vivo mouse model and thus confirmed its prooncogenic role from the loss of function perspective (Fig. 7J).

Figure 7.

CRISPR/Cas13a-mediated knockdown of C190 suppresses tumor growth in vivo. A, Schematic showing the design of xenograft experiment. B, Dynamics of growth of tumors derived from the indicated experimental cells. C, Right, the tumor volume at the final day of the time course. Mean values are shown with SD error bars, N = 4; *, P < 0.05; **, P < 0.01 (Kruskal–Wallis test). D, Side-by-side photographs (left) and GFP fluorescent images (right) of the tumors from the indicated experimental groups. E, Mean tumor masses at the day of sacrifice, *, P < 0.05; ns, not significant (Kruskal–Wallis test). F, RT-ddPCR analysis of the expression of C190 in the indicated experimental groups. *, P < 0.05 (Kruskal–Wallis test). G, RNA-FISH analysis of the expression of C190 in the tumors from the indicated experimental groups. Nuclei stained with DAPI. H, Immunofluorescent staining of Ki-67 antigen in the tumors from the indicated experimental groups. I, Quantification of the number of Ki-67–positive cells in (H), ***, P < 0.001; ns, not significant (ANOVA test). J, Summary of the experiment.

Figure 7.

CRISPR/Cas13a-mediated knockdown of C190 suppresses tumor growth in vivo. A, Schematic showing the design of xenograft experiment. B, Dynamics of growth of tumors derived from the indicated experimental cells. C, Right, the tumor volume at the final day of the time course. Mean values are shown with SD error bars, N = 4; *, P < 0.05; **, P < 0.01 (Kruskal–Wallis test). D, Side-by-side photographs (left) and GFP fluorescent images (right) of the tumors from the indicated experimental groups. E, Mean tumor masses at the day of sacrifice, *, P < 0.05; ns, not significant (Kruskal–Wallis test). F, RT-ddPCR analysis of the expression of C190 in the indicated experimental groups. *, P < 0.05 (Kruskal–Wallis test). G, RNA-FISH analysis of the expression of C190 in the tumors from the indicated experimental groups. Nuclei stained with DAPI. H, Immunofluorescent staining of Ki-67 antigen in the tumors from the indicated experimental groups. I, Quantification of the number of Ki-67–positive cells in (H), ***, P < 0.001; ns, not significant (ANOVA test). J, Summary of the experiment.

Close modal

C190 targets CDKs by sponging miR-142-5p

The best-characterized mechanism of circRNA action is by sequestering miRNAs, which alleviates their target mRNAs from miRNA-mediated inhibition. Given the fact that we demonstrated the positive effect of C190 overexpression on the protein levels of CDKs (Fig. 5), we tested the possibility that C190 regulates them by such miRNA sponging mechanism. First, TargetScan tool (http://www.targetscan.org/vert_72/) was used to predict 131 miRNAs broadly conserved among vertebrates targeting CDK1, CDK4 and CDK6 (Fig. 8A and B). To increase the probability of identifying functional miRNAs, we narrowed down this list to 15 miRNAs that target at least two CDKs (Fig. 8A). This list was then compared with the CircInteractome database (http://circinteractome.nia.nih.gov) prediction of 9 miRNAs targeted by C190 (Fig. 8A). As a result, miR-142-5p was selected as a miRNA targeting both CDK4 and CDK6 and being a sponging target of C190 (Fig. 8B). Interestingly, the alternative maturation product of the miR-142 pre-miRNA, miR-142-3p, was predicted to target CDK1 (Fig. 8B). To validate the binding between miR-142-5p and C190, we performed RNA pulldown of C190 in A549 cells. As shown by qRT-PCR analysis, both C190 and miR-142-5p were concomitantly purified indicative of their direct interaction (Fig. 8C and D). To validate the targeting of CDK4 and CDK6 by miR-142-5p, the 3′-UTRs of these CDKs were cloned into the luciferase-reporter vector and transfected to A549 cells with or without miR-142-5p. Indeed, cotransfection of the CDK6 3′-UTR-luciferase construct (Fig. 8E), but not CDK4 3′-UTR-luciferase (Supplementary Fig. S2A) with miR-142-5p resulted in the suppression of the luciferase activity, indicating that the former is the bona fide target of miR-142-5p. At the same time, the concomitant cotransfection of C190 with miR-142-5p reversed the suppression of the luciferase activity (Fig. 8E). Moreover, qRT-PCR analysis of patients' tumor samples revealed that whereas the higher expression of C190, CDK1, and CDK6 was associated with the later stages of lung cancer, the higher expression of miR-142-5p was associated with the early stage, which is consistent with the antagonistic roles of C190 and miR-142-5p (Fig. 8F).

Figure 8.

C190 targets CDKs by sponging miR-142-5p. A, Schematic showing the selection of the potential miRNAs that can mediate the effect of C190 on CDKs. TargetScan predictions of miRNAs targeting at least two CDKs are shown in the red circles. B, Top, CircInteractome prediction of the interaction site between C190 and hsa–miR-142-5p. Bottom, TargetScan prediction of the CDK 3′-UTR sites targeted by miR-142-5p and miR-142-3p. C, RNA pulldown assay showing precipitation of C190 by the C190-spedific probe. D, RNA pulldown assay showing precipitation of miR-142-5p by the C190-spedific probe. C and D, mean relative enrichment as compared with the control probe is shown, N = 3. *, P < 0.05 (Mann–Whitney U test). E, Dual luciferase reporter assay demonstrating the effect of miR-142-5p on CDK6 3′-UTR (CDK6UTR). CDK6 3′-UTR–luciferase construct was transfected into A549 cells alone or in combination with miR-142-5p or nontargeting miRNA (miR-NC) encoding plasmid and relative luminescence was determined 48 hours post-transfection. Mean values (N = 2) are shown with SD error bars. **, P < 0.01; ***, P < 0.001; ns, not significant (Kruskal–Wallis test). F, qRT-PCR showing the expression of the indicated transcripts in the early- and late-stage lung cancer tumor samples. G, Summary of the effects of C190 on the pathways implicated in NSCLC and its targeting by CRISPR/Cas13a system.

Figure 8.

C190 targets CDKs by sponging miR-142-5p. A, Schematic showing the selection of the potential miRNAs that can mediate the effect of C190 on CDKs. TargetScan predictions of miRNAs targeting at least two CDKs are shown in the red circles. B, Top, CircInteractome prediction of the interaction site between C190 and hsa–miR-142-5p. Bottom, TargetScan prediction of the CDK 3′-UTR sites targeted by miR-142-5p and miR-142-3p. C, RNA pulldown assay showing precipitation of C190 by the C190-spedific probe. D, RNA pulldown assay showing precipitation of miR-142-5p by the C190-spedific probe. C and D, mean relative enrichment as compared with the control probe is shown, N = 3. *, P < 0.05 (Mann–Whitney U test). E, Dual luciferase reporter assay demonstrating the effect of miR-142-5p on CDK6 3′-UTR (CDK6UTR). CDK6 3′-UTR–luciferase construct was transfected into A549 cells alone or in combination with miR-142-5p or nontargeting miRNA (miR-NC) encoding plasmid and relative luminescence was determined 48 hours post-transfection. Mean values (N = 2) are shown with SD error bars. **, P < 0.01; ***, P < 0.001; ns, not significant (Kruskal–Wallis test). F, qRT-PCR showing the expression of the indicated transcripts in the early- and late-stage lung cancer tumor samples. G, Summary of the effects of C190 on the pathways implicated in NSCLC and its targeting by CRISPR/Cas13a system.

Close modal

Furthermore, given the fact that circRNAs are also known to execute their functions by forming complexes with proteins, we performed RNA pulldown followed by mass spectrometry analysis (Supplementary Fig. S2B). Consistently with our results, we observed association of a number of cell-cycle and translation-related proteins with C190 (Supplementary Table S4, Supplementary Fig. S2C). Such an analysis revealed the enrichment of C190-associated proteins in cell proliferation and gene translation networks (Supplementary Fig. S2D and S2E).

CircRNA species of the mammalian noncoding transcriptome resultant from nonlinear splicing are now recognized as important factors involved in a plethora of biological processes in a highly regulated manner (30). Therefore, it comes as no surprise that the dysregulated expression of circRNAs has been shown to be associated with carcinogenesis (9). In our previous study, we identified hsa_circ_0000190 (C190) as a prognostic biomarker of NSCLC that can be easily detected in the patients' blood samples (14). In the same study, we noticed that the higher expression of C190 correlated with the dominant EGFR mutation status, which is the key determinant of lung cancer malignancy and therapeutic strategies (14). Here, we confirm that C190 is highly overexpressed in NSCLC tumor as compared with normal lung tissue, and its expression is higher in a cell line with the constitutively active EGFR (Fig. 1). At the same time, we notice that only the backspliced circular RNA product of the CNIH4 gene, rather than the conventionally spliced mRNA, is highly overexpressed in the EGFR-active NSCLC cells (Fig. 1). Whereas the function of the protein product of the CNIH4 gene is related to basic G protein-coupled receptor endoplasmic reticulum (ER) and its membrane trafficking (31), no evidence exists on its direct involvement in the signaling pathways controlling cancer cell proliferation, survival or metastasis. Therefore, we speculate that this gene has been adapted in its noncoding circular form product for cellular signaling pathways. Given that C190 and CNIH4 are expressed from the same promoter, the mechanism of their differential expression in lung cancer is expected to be posttranscriptional. Generally, our result is consistent with the previous studies demonstrating that the correlation between the expression levels of linear and circRNA isoforms of the same gene is weak and subject to high variability in different tissues and cell lines (32–34). This is indicative of the independent mechanisms of the biogenesis of linear and circular isoforms. Indeed, certain RBPs can promote formation of circular isoforms in response to signaling cues regulating various biological processes. Quaking RBP was shown to facilitate the biogenesis of circRNAs in response to TGFβ-induced epithelial–mesenchymal transition (35). In another study, the direct phosphorylation of DHX9 RBP by PI3K was shown to promote the formation of oncogenic circCCDC66 implicated in drug resistance in colon cancer (36). Given the rapid induction of C190 expression in response to EGFR stimulation (Fig. 2), we can speculate the existence of an RBP facilitating C190 biogenesis that is directly regulated by the MAPK/ERK pathway.

In treating NSCLC, patients often benefit more in the form of achieving stable disease and/or progression-free survival (PFS) from targeted therapy rather than standard chemotherapy (37) and targeting receptor tyrosine kinases such as EGFR results in the best PFS outcome (38). Inhibiting EGFR activation is a potent NSCLC therapeutic strategy, but secondary mutation development drives drug resistance that prevents long-term treatment efficacy despite consistent new EGFR–TKI generation (39). In the current study, we discovered that EGFR activation by its ligand (EGF) uniquely increased C190 expression as well as previously known key EGFR downstream signaling pathways (Fig. 2). Of note, small-molecule inhibition of only the MAPK/ERK pathway significantly suppressed C190 expression in a manner similar to EGFR–TKI treatment (Fig. 2). This is a strong indication of the involvement of C190 circRNA in EGFR intracellular signaling. Our observation is similar to a previous study, which reported that circ-ZKSCAN1 can modulate MAPK signaling in NSCLC (40). MAPK signaling activation is known to drive proliferation, metastasis, and survival of cancer cells (41). Sun and colleagues (42) previously reported that circRNA CEP128 promotes bladder cancer progression by increasing MAPK signaling pathway activation through miR-145-5p/MYD88. Likewise, in lung adenocarcinoma, circRNA UBAP2 overexpression also increased MAPK activation and siRNA-mediated knockdown suppressed p-MAPK expression and lung cancer proliferation (43). Consistently, our result also showed that ectopic C190 overexpression could increase MAPK activation in NSCLC cells, which subsequently increased cell proliferation and migration (Fig. 3), which are the major hallmarks of cancer progression. Surprisingly, exogenic C190 overexpression did not significantly increase EGFR phosphorylation (Fig. 3), which suggests that C190 acts independently of EGFR activation or mutation status. Consistently, the knockdown of endogenous C190 neither had any effect on EGFR phosphorylation status (Fig. 6). On the other hand, we showed that overexpression (Fig. 2) or knockdown (Fig. 6) of C190 had drastic effect on ERK1/2 phosphorylation status; in contrast, phosphorylation of MEK1/2 was not or only weakly affected. Whereas MEK1/2 exhibits substrate specificity toward ERK1/2 only (42), ERK1/2, as the most downstream tier of the MAPK/ERK signaling cascade, is known to phosphorylate a wide range of around 200 downstream effector targets (44). This makes C190 a highly promising therapeutic target that can be used to combat lung cancer resistant not only to TKIs, but also to therapeutic agents targeting all other members of the cascade, including RAS, RAF, and MEK (45). Moreover, because we observe both induction of C190 expression by the MAPK/ERK pathway and positive regulation of the MAPK/ERK pathway by C190, we can hypothesize the existence of a positive feedback regulation loop. Whereas the interplay of positive and negative regulation loops normally ensures the dynamics and oscillatory nature of MAPK/ERK pathway activation (46), the dysregulation of C190 by such positive feedback mechanism may lead to the chronic activation of MAPK/ERK signaling that contributes to tumorigenesis in NSCLC. This further justifies the importance of considering C190 as an important therapeutic target in addition to classical TKI-based therapies.

Normally, growth factors such as EGF induce rapid transient changes in cell signaling pathways followed by long-term effects that require changes in the transcriptome or epigenetic status. Indeed, EGF signaling is characterized by distinct waves of transcription starting with immediate early genes and culminating in fate-determining genes (47). Consistently with this notion, we observed that upon EGF stimulation the levels of p-EGFR, p-ERK, and p-Akt initially increased after 15–60 minutes but then receded (Fig. 2B and C). Interestingly, C190 upregulation followed the early response pattern, as its level started to increase concomitantly with EGFR phosphorylation; however, it increased continuously in the later time points (Fig. 2E and F). Previously, it was shown that circRNAs in general do not respond to EGF stimulation as efficiently as linear mRNAs or miRNAs (32), which makes C190 an unusual case. Consistently with the increased stability of circRNAs, it comes as no surprise that high C190 expression was sustained after the transient activation of EGFR-related signaling pathways (Fig. 2E and F) indicative of its putative function in establishment of the long-term biological effects.

To gain more global insight on the molecular functions of C190, we performed differential RNA-seq and expression-to-kinase (X2K) analysis (25) to identify the kinases that lead to gene expression profile resultant from C190 overexpression (Fig. 5). Consistently with our evidence of C190 being a regulator of the MAPK/ERK pathway, we observed the enrichment of this pathway's component kinases, including ERK1/2 (Fig. 5). Interestingly, we also found the enrichment of a number of cell-cycle–dependent kinases, particularly CDK1, CDK2 and CDK4 and validated their upregulation in C190-overexpressing cells (Fig. 5C). On this note, de Leeuw and colleagues (26) have previously identified a link between CDK4/6 and MAPK signaling in cancer cells. Likewise, CDK1 inhibitor was reported to reduce lung cell proliferation by targeting MAPK signaling (48). Similarly, our result that shows that C190 overexpression increases MAPK activation and CDK1, 2, 4, and 6 expression (Fig. 5) is in line with these previous observations. In addition, Chen and colleagues (49) previously suggested that circRNA_100290 downregulates CDK6 expression through sponging miR-29b in oral carcinoma.

To explore the possibility of C190 to regulate CDK levels by such sponging mechanism, we applied bioinformatic prediction and identified miR-142-5p–binding sites present on C190 and predicted that it could target CDK4 and CDK6 (Fig. 8A). C190 RNA pulldown assay result proved that C190 could actually bind to miR-142-5p (Fig. 8C and D). Indeed, both strands of miR-142 have been widely implicated in cell-cycle regulation (50, 51); for instance, miR-142-3p was discovered to target G1–S phase transition-implicated CDK4 protein to suppress colorectal cancer (52). In meanwhile, CDK6-associated miRNAs such as miR-142-5p have also been implicated in cancer pathogenesis and treatment (53). More precisely, miR-142-5p was discovered to suppress osteosarcoma proliferation by targeting the ERK1/2 signaling pathway (54). In addition, Lou and colleagues (55) reported that miR-142-5p overexpression could suppress HCC by blocking G1–S phase transition. On the other hand, miR-142-5p upregulation was reported to promote breast cancer (56). However, in the case of NSCLC, Wang and colleagues (57) discovered that miR-142-5p functions as a tumor suppressor by inhibiting PIK3CA proliferative signaling. Therefore, it is evident that miR-142-5p can have both pro-oncogenic and tumor suppressor effects depending on the tumor type (50). In the current study, we proved that miR-142-5p targets CDK6 using dual luciferase assay (Fig. 8E), which is consistent with its tumor suppressor role. Therefore, the sponging of miR-142-5p by pro-oncogenic C190 may be a mechanism of the increased CDK6 expression and the resultant increase in Rb phosphorylation (Fig. 5C), which may eventually result in cell-cycle progression and tumor growth (Fig. 8G). Moreover, given the fact that miR-142-5p was shown to suppress ERK phosphorylation in osteosarcoma (54), such mechanism may also be implicated in C190-mediated phosphorylation of ERK1/2 in our NSCLC experimental system (Figs. 3 and 5).

After proving that C190 is an oncogenic circRNA, we hypothesized that targeting C190 for degradation in NSCLC could reverse its tumorigenic effects. Currently, in circRNA field, several studies have used RNAi technique to downregulate the expression of oncogenic circRNAs. For instance, Zong and colleagues (58) used siRNA to downregulate circRNA_102231 in NSCLC cell lines. siRNA-mediated knockdown of CCDC66 circRNA was also reported in the study of drug resistance of NSCLC cells (59). On the same note, shRNA targeting hsa_circ_0020123 was used to reduce NSCLC growth in vivo (60). Meanwhile, the major concern of RNAi technique is its considerable off-target effects (61). To avoid this pitfall, we adopted the CRISPR/Cas13a RNA-targeting system to target a circRNA for degradation for the first time, because CRISPR/Cas13a is known to possess no off-target effects in mammalian cells (16). Our results showed that approximately 70% C190 downregulation was achieved using CRISPR/Cas13a system (Fig. 6B). Such knockdown efficiency was comparable with that in other studies, such as by Saifullah and colleagues (62), where CRISPR/Cas13a system was used to knock down the EML4–ALK hybrid transcript in the H3122 lung cancer cell line. More interestingly, MAPK activation and global translation-associated RPS6 activation were suppressed by CRISPR/Cas13a-mediated C190 targeting (Fig. 6C and D). Phenotypically, proliferation, migration, and survival of NSCLC cell lines were also significantly reduced in vitro (Fig. 6EI). Finally, we further used in vivo xenograft model to prove that targeting C190 for degradation with CRISPR/Cas13a effectively reduced tumor growth and proliferative marker in vivo (Fig. 7). All these observations corroborate the potential of the novel CRISPR/Cas13a system to be applied as a cancer therapeutic tool.

In conclusion, in our current study, we confirmed that C190 is an oncogenic circRNA implicated in NSCLC progression. We identified an intimate connection between C190 and the MAPK/ERK signaling pathway, as C190 expression was both regulated by it, at the same time, C190 exerted positive effect on ERK1/2 activation. C190 was shown to regulate the important effectors of cell-cycle and global translation, CDKs and RPS6, respectively. Whereas such effects on these proteins could be mediated by the MAPK/ERK pathway, we provide evidence of the pleiotropic mode of action of C190, as its effect on CDKs could also be mediated by independent miRNA (miR-142-5p) sponging mechanism. Meanwhile, we applied the CRISPR/Cas13a system to target circRNA for the first time in NSCLC, and demonstrated its efficiency both in vitro and in vivo. Most importantly, we believe that C190 can be an important target for NSCLC therapeutics given its direct involvement in the EGFR-directed MAPK/ERK pathway and other pro-oncogenic effects, and CRISPR/Cas13a system can be a promising therapeutic tool (Fig. 8G).

No disclosures were reported.

A.A. Ishola: Conceptualization, investigation, writing–original draft. C.-S. Chien: Investigation. Y.-P. Yang: Conceptualization, data curation. Y. Chien: Data curation, validation. A.A. Yarmishyn: Writing–original draft, writing–review and editing. P.-H. Tsai: Software, formal analysis. J.C.-Y. Chen: Validation. P.-K. Hsu: Resources. Y.-H. Luo: Resources. Y.-M. Chen: Resources. K.-H. Liang: Software. Y.-T. Lan: Resources. T.-I. Huo: Resources. H.-I. Ma: Resources. M.-T. Chen: Resources. M.-L. Wang: Resources, supervision. S.-H. Chiou: Resources, supervision.

This study was funded by the Ministry of Science and Technology in Taiwan (MOST 109-2320-B-075-001 and MOST 110-2320-B-075-006-MY3), Taipei Veterans General Hospital (V108D46-004-MY2-2, V109E-007-4, V109E-007-5, V110C-187), the IBMS CRC Research Program of the Institute of Biomedical Sciences, Academia Sinica (IBMS-CRC109-P04), the Veterans Affairs Council (110VACS-003, 110VACS-007), the SPROUT Project-Center For Intelligent Drug Systems and Smart Bio-devices (IDS2B) of National Chiao Tung University, and the “Cancer Progression Research Center, National Yang-Ming Chiao-Tung University” from The Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education (MOE), Taiwan.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Bray
F
,
Ferlay
J
,
Soerjomataram
I
,
Siegel
RL
,
Torre
LA
,
Jemal
A
. 
Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries
.
CA Cancer J Clin
2018
;
68
:
394
424
.
2.
Gridelli
C
,
Rossi
A
,
Carbone
DP
,
Guarize
J
,
Karachaliou
N
,
Mok
T
, et al
Non-small-cell lung cancer
.
Nat Rev Dis Primers
2015
;
1
:
15009
.
3.
Ritchie MRaH
. 
2018 Cancer
.
Available from:
<https://ourworldindata.org/cancer>.
4.
Thomas
R
,
Weihua
Z
. 
Rethink of EGFR in cancer with its kinase independent function on board
.
Front Oncol
2019
;
9
:
800
.
5.
Huang
L
,
Fu
L
. 
Mechanisms of resistance to EGFR tyrosine kinase inhibitors
.
Acta Pharmaceutica Sinica B
2015
;
5
:
390
401
.
6.
Barrett
SP
,
Salzman
J
. 
Circular RNAs: analysis, expression and potential functions
.
Development
2016
;
143
:
1838
47
.
7.
Kramer
MC
,
Liang
D
,
Tatomer
DC
,
Gold
B
,
March
ZM
,
Cherry
S
, et al
Combinatorial control of Drosophila circular RNA expression by intronic repeats, hnRNPs, and SR proteins
.
Genes Dev
2015
;
29
;
2168
82
.
8.
Liu
CX
,
Li
X
,
Nan
F
,
Jiang
S
,
Gao
X
,
Guo
SK
, et al
Structure and degradation of circular RNAs regulate PKR activation in innate immunity
.
Cell
2019
;
177
:
865
80
.
9.
Chen
T
,
Luo
J
,
Gu
Y
,
Huang
J
,
Luo
Q
,
Yang
Y
. 
Comprehensive analysis of circular RNA profiling in AZD9291-resistant non–small cell lung cancer cell lines
.
Thoracic Cancer
2019
;
10
;
930
41
.
10.
Qiu
M
,
Xia
W
,
Chen
R
,
Wang
S
,
Xu
Y
,
Ma
Z
, et al
The circular RNA circPRKCI promotes tumor growth in lung adenocarcinoma
.
Cancer Res
2018
;
78
:
2839
51
.
11.
Cen
J
,
Liang
Y
,
Huang
Y
,
Pan
Y
,
Shu
G
,
Zheng
Z
, et al
Circular RNA circSDHC serves as a sponge for miR-127-3p to promote the proliferation and metastasis of renal cell carcinoma via the CDKN3/E2F1 axis
.
Mol Cancer
2021
;
20
:
19
.
12.
Chen
S
,
Li
T
,
Zhao
Q
,
Xiao
B
,
Guo
J
. 
Using circular RNA hsa_circ_0000190 as a new biomarker in the diagnosis of gastric cancer
.
Clin Chim Acta
2017
;
466
:
167
71
.
13.
Feng
Y
,
Zhang
L
,
Wu
J
,
Khadka
B
,
Fang
Z
,
Gu
J
, et al
CircRNA circ_0000190 inhibits the progression of multiple myeloma through modulating miR-767-5p/MAPK4 pathway
.
J Exp Clin Cancer Res
2019
;
38
:
54
.
14.
Luo
Y-H
,
Yang
Y-P
,
Chien
C-S
,
Yarmishyn
AA
,
Ishola
AA
,
Chien
Y
, et al
Plasma level of circular RNA hsa_circ_0000190 correlates with tumor progression and poor treatment response in advanced lung cancers
.
Cancers
2020
;
12
:
1740
.
15.
Liu
Z
,
Wang
T
,
She
Y
,
Wu
K
,
Gu
S
,
Li
L
, et al
N (6)-methyladenosine-modified circIGF2BP3 inhibits CD8(+) T-cell responses to facilitate tumor immune evasion by promoting the deubiquitination of PD-L1 in non–small cell lung cancer
.
Mol Cancer
2021
;
20
:
105
.
16.
Abudayyeh
OO
,
Gootenberg
JS
,
Essletzbichler
P
,
Han
S
,
Joung
J
,
Belanto
JJ
, et al
RNA targeting with CRISPR–Cas13
.
Nature
2017
;
550
:
280
.
17.
East-Seletsky
A
,
O'Connell
MR
,
Burstein
D
,
Knott
GJ
,
Doudna
JA
. 
RNA targeting by functionally orthogonal type VI-A CRISPR-Cas enzymes
.
Mol Cell
2017
;
66
:
373
83
.
18.
Zhang
J
,
You
Y
. 
CRISPR-Cas13a system: a novel approach to precision oncology
.
Cancer Biol Med
2020
;
17
:
6
8
.
19.
Zhao
X
,
Liu
L
,
Lang
J
,
Cheng
K
,
Wang
Y
,
Li
X
, et al
A CRISPR-Cas13a system for efficient and specific therapeutic targeting of mutant KRAS for pancreatic cancer treatment
.
Cancer Lett
2018
;
431
:
171
81
.
20.
Xu
D
,
Cai
Y
,
Tang
L
,
Han
X
,
Gao
F
,
Cao
H
, et al
A CRISPR/Cas13-based approach demonstrates biological relevance of vlinc class of long non-coding RNAs in anticancer drug response
.
Sci Rep
2020
;
10
:
1794
.
21.
Li
S
,
Li
X
,
Xue
W
,
Zhang
L
,
Yang
LZ
,
Cao
SM
, et al
Screening for functional circular RNAs using the CRISPR-Cas13 system
.
Nat Methods
2021
;
18
:
51
9
.
22.
Zhang
Y
,
Nguyen
TM
,
Zhang
XO
,
Wang
L
,
Phan
T
,
Clohessy
JG
, et al
Optimized RNA-targeting CRISPR/Cas13d technology outperforms shRNA in identifying functional circRNAs
.
Genome Biol
2021
;
22
:
41
.
23.
Bai
Q
,
Bai
Z
,
Sun
L
. 
Detection of RNA-binding proteins by In vitro RNA pull-down in adipocyte culture
.
JoVE
2016
:
e54207
.
doi: 10.3791/54207
.
24.
Wang
DO
,
Matsuno
H
,
Ikeda
S
,
Nakamura
A
,
Yanagisawa
H
,
Hayashi
Y
, et al
A quick and simple FISH protocol with hybridization-sensitive fluorescent linear oligodeoxynucleotide probes
.
RNA
2012
;
18
:
166
75
.
25.
Clarke
DJB
,
Kuleshov
MV
,
Schilder
BM
,
Torre
D
,
Duffy
ME
,
Keenan
AB
, et al
eXpression2Kinases (X2K) Web: linking expression signatures to upstream cell signaling networks
.
Nucleic Acids Res
2018
;
46
:
W171
W9
.
26.
de Leeuw
R
,
McNair
C
,
Schiewer
MJ
,
Neupane
NP
,
Brand
LJ
,
Augello
MA
, et al
MAPK Reliance via acquired CDK4/6 inhibitor resistance in cancer
.
Clin Cancer Res
2018
;
24
:
4201
14
.
27.
Chang
HJ
. 
RNAi-mediated knockdown of target genes: a promising strategy for pancreatic cancer research
.
Cancer Gen Ther
2007
;
14
:
677
85
.
28.
Jackson
AL
,
Bartz
SR
,
Schelter
J
,
Kobayashi
SV
,
Burchard
J
,
Mao
M
, et al
Expression profiling reveals off-target gene regulation by RNAi
.
Nat Biotechnol
2003
;
21
:
635
7
.
29.
Abudayyeh
OO
,
Gootenberg
JS
,
Konermann
S
,
Joung
J
,
Slaymaker
IM
,
Cox
DBT
, et al
C2c2 is a single-component programmable RNA-guided RNA-targeting CRISPR effector
.
Science
2016
;
10
:
aaf5573
.
30.
Hansen
TB
,
Jensen
TI
,
Clausen
BH
,
Bramsen
JB
,
Finsen
B
,
Damgaard
CK
, et al
Natural RNA circles function as efficient microRNA sponges
.
Nature
2013
;
495
:
384
.
31.
Sauvageau
E
,
Rochdi
MD
,
Oueslati
M
,
Hamdan
FF
,
Percherancier
Y
,
Simpson
JC
, et al
CNIH4 interacts with newly synthesized GPCR and controls their export from the endoplasmic reticulum
.
Traffic
2014
;
15
:
383
400
.
32.
Enuka
Y
,
Lauriola
M
,
Feldman
ME
,
Sas-Chen
A
,
Ulitsky
I
,
Yarden
Y
. 
Circular RNAs are long-lived and display only minimal early alterations in response to a growth factor
.
Nucleic Acids Res
2016
;
44
:
1370
83
.
33.
Salzman
J
,
Chen
RE
,
Olsen
MN
,
Wang
PL
,
Brown
PO
. 
Cell-type specific features of circular RNA expression
.
PLos Genet
2013
;
9
:
e1003777
.
34.
Liang
D
,
Wilusz
JE
. 
Short intronic repeat sequences facilitate circular RNA production
.
Genes Dev
2014
;
28
:
2233
47
.
35.
Conn
SJ
,
Pillman
KA
,
Toubia
J
,
Conn
VM
,
Salmanidis
M
,
Phillips
CA
, et al
The RNA binding protein quaking regulates formation of circRNAs
.
Cell
2015
;
160
:
1125
34
.
36.
Lin
YC
,
Yu
YS
,
Lin
HH
,
Hsiao
KY
. 
Oxaliplatin-induced DHX9 phosphorylation promotes oncogenic circular RNA CCDC66 expression and development of chemoresistance
.
Cancers (Basel)
2020
;
12
:
697
.
37.
Zhang
Q
,
Wang
Z
,
Guo
J
,
Liu
L
,
Han
X
,
Li
M
, et al
Comparison of single-agent chemotherapy and targeted therapy to first-line treatment in patients aged 80 years and older with advanced non–small cell lung cancer
.
Onco Targets Ther
2015
;
8
:
893
8
.
38.
Middleton
G
,
Fletcher
P
,
Popat
S
,
Savage
J
,
Summers
Y
,
Greystoke
A
, et al
The National Lung Matrix Trial of personalized therapy in lung cancer
.
Nature
2020
;
583
:
807
12
.
39.
Huang
CY
,
Hsu
LH
,
Chen
CY
,
Chang
GC
,
Chang
HW
,
Hung
YM
, et al
Inhibition of alternative cancer cell metabolism of EGFR mutated non-small cell lung cancer serves as a potential therapeutic strategy
.
Cancers
2020
;
12
:
181
.
40.
Wang
Y
,
Xu
R
,
Zhang
D
,
Lu
T
,
Yu
W
,
Wo
Y
, et al
Circ-ZKSCAN1 regulates FAM83A expression and inactivates MAPK signaling by targeting miR-330-5p to promote non–small cell lung cancer progression
.
Transl Lung Cancer Res
2019
;
8
:
862
75
.
41.
McCubrey
JA
,
Steelman
LS
,
Chappell
WH
,
Abrams
SL
,
Wong
EW
,
Chang
F
, et al
Roles of the Raf/MEK/ERK pathway in cell growth, malignant transformation and drug resistance
.
Biochim Biophys Acta
2007
;
1773
:
1263
84
.
42.
Sun
M
,
Zhao
W
,
Chen
Z
,
Li
M
,
Li
S
,
Wu
B
, et al
Circular RNA CEP128 promotes bladder cancer progression by regulating mir-145-5p/Myd88 via MAPK signaling pathway
.
Int J Cancer
2019
;
145
:
2170
81
.
43.
Yin
Y
,
Gao
H
,
Guo
J
,
Gao
Y
. 
Effect of circular RNA UBAP2 silencing on proliferation and invasion of human lung cancer A549 cells and its mechanism
.
Zhongguo Fei Ai Za Zhi
2017
;
20
;
800
7
.
44.
Yoon
S
,
Seger
R
. 
The extracellular signal-regulated kinase: multiple substrates regulate diverse cellular functions
.
Growth Factors
2006
;
24
:
21
44
.
45.
Yu
Z
,
Ye
S
,
Hu
G
,
Lv
M
,
Tu
Z
,
Zhou
K
, et al
The RAF-MEK-ERK pathway: targeting ERK to overcome obstacles to effective cancer therapy
.
Future Med Chem
2015
;
7
:
269
89
.
46.
Shin
SY
,
Rath
O
,
Choo
SM
,
Fee
F
,
McFerran
B
,
Kolch
W
, et al
Positive- and negative-feedback regulations coordinate the dynamic behavior of the Ras–Raf–MEK–ERK signal transduction pathway
.
J Cell Sci
2009
;
122
:
425
35
.
47.
Avraham
R
,
Yarden
Y
. 
Feedback regulation of EGFR signalling: decision making by early and delayed loops
.
Nat Rev Mol Cell Biol
2011
;
12
:
104
17
.
48.
Knockaert
M
,
Lenormand
P
,
Gray
N
,
Schultz
P
,
Pouyssegur
J
,
Meijer
L
. 
p42/p44 MAPKs are intracellular targets of the CDK inhibitor purvalanol
.
Oncogene
2002
;
21
:
6413
24
.
49.
Chen
L
,
Zhang
S
,
Wu
J
,
Cui
J
,
Zhong
L
,
Zeng
L
, et al
circRNA_100290 plays a role in oral cancer by functioning as a sponge of the miR-29 family
.
Oncogene
2017
;
36
:
4551
.
50.
Shrestha
A
,
Mukhametshina
RT
,
Taghizadeh
S
,
Vásquez-Pacheco
E
,
Cabrera-Fuentes
H
,
Rizvanov
A
, et al
MicroRNA-142 is a multifaceted regulator in organogenesis, homeostasis, and disease. Developmental dynamics: an official publication of the American Association of
Anatomists
2017
;
246
:
285
90
.
51.
Su
Y
,
Zhou
Z
,
Yang
K
,
Wang
X
,
Zhu
Y
,
Fa
XE
, et al
MIR-142-5p and miR-9 may be involved in squamous lung cancer by regulating cell-cycle related genes
.
Eur Rev Med Pharmacol Sci
2013
;
17
:
3213
20
.
52.
Zhu
X
,
Ma
SP
,
Yang
D
,
Liu
Y
,
Wang
YP
,
Lin
T
, et al
miR-142-3p suppresses cell growth by targeting CDK4 in colorectal cancer
.
Cell Physiol Biochem
2018
;
51
:
1969
81
.
53.
Batool
S
,
Hasan Mahmood
MS
,
Furqan
T
,
Batool
S
. 
Highlighting the role of Cdk6 associated MicroRNAs in cancer treatment using in silico approaches
.
J Surg Oncol
2020
.
DOI: 10.31487/j.JSO.2020.04.11
.
54.
Cheng
D
,
Li
J
,
Zhang
L
,
Hu
L
. 
miR-142-5p suppresses proliferation and promotes apoptosis of human osteosarcoma cell line, HOS, by targeting PLA2G16 through the ERK1/2 signaling pathway
.
Oncol Lett
2019
;
17
:
1363
71
.
55.
Lou
K
,
Chen
N
,
Li
Z
,
Zhang
B
,
Wang
X
,
Chen
Y
, et al
MicroRNA-142-5p overexpression inhibits cell growth and induces apoptosis by regulating FOXO in hepatocellular carcinoma cells
.
Oncol Res
2017
;
25
:
65
73
.
56.
Xu
W
,
Wang
W
. 
MicroRNA1425p modulates breast cancer cell proliferation and apoptosis by targeting phosphatase and tensin homolog
.
Mol Med Rep
2018
;
17
:
7529
36
.
57.
Wang
Z
,
Liu
Z
,
Fang
X
,
Yang
H
. 
MiR-142-5p suppresses tumorigenesis by targeting PIK3CA in non–small cell lung cancer
.
Cell Physiol Biochem
2017
;
43
:
2505
15
.
58.
Zong
L
,
Sun
Q
,
Zhang
H
,
Chen
Z
,
Deng
Y
,
Li
D
, et al
Increased expression of circRNA_102231 in lung cancer and its clinical significance
.
Biomed Pharmacother
2018
;
102
:
639
44
.
59.
Joseph
NA
,
Chiou
S-H
,
Lung
Z
,
Yang
C-L
,
Lin
T-Y
,
Chang
H-W
, et al
The role of HGF-MET pathway and CCDC66 cirRNA expression in EGFR resistance and epithelial-to-mesenchymal transition of lung adenocarcinoma cells
.
J Hematol Oncol
2018
;
11
:
74
.
60.
Qu
D
,
Yan
B
,
Xin
R
,
Ma
T
. 
A novel circular RNA hsa_circ_0020123 exerts oncogenic properties through suppression of miR-144 in non–small cell lung cancer
.
Am J Cancer Res
2018
;
8
:
1387
.
61.
Rao
DD
,
Senzer
N
,
Cleary
MA
,
Nemunaitis
J
. 
Comparative assessment of siRNA and shRNA off target effects: what is slowing clinical development
.
Cancer Gene Ther
2009
;
16
:
807
9
.
62.
Saifullah
SM
,
Suzuki
T
,
Yano
S
,
Tsukahara
T
. 
Effective RNA knockdown using CRISPR-Cas13a and molecular targeting of the EML4–ALK transcript in H3122 lung cancer cells
.
Int J Mol Sci
2020
;
21
:
8904
.