The Axis Regulates the Proliferation, Apoptosis and Metastasis of Human Colorectal Cancer Cell Line HCT116

Background: Circular RNAs (circRNAs) have gained wide attention as a class of potential biomarkers for the early detection of multiple cancers. However, the functions and mechanisms of circRNAs in the oncogenesis of human colorectal cancer (CRC) remain to be elucidated. This study aimed to investigate the roles of hsa_circ_0000523 and its parental gene METTL3 in regulating cell proliferation, apoptosis and metastasis in a human CRC cell line (HCT116). Methods: HCT116 cells were left untreated, transfected with hsa_circ_0000523- or METTL3-expressing plasmid, transfected with siRNA oligo against hsa_circ_0000523 or METTL3, or transfected with negative control vector or siRNA oligo. All transfections were performed with Lipofectamine 2000. Transcriptional levels of hsa_circ_0000523, miR-let-7b, and METTL3 were measured by real-time quantitative polymerase chain reaction. Cell proliferation was assessed by CCK-8 assay. Apoptosis was evaluated by ow cytometry after staining for annexin V and propidium iodide. Cellular potential for migration was detected by transwell assay. Results: In HCT116 cells, hsa_circ-0000523 indirectly regulated METTL3 expression by suppressing the transcription of miR-let-7b. The expression of METTL3 promoted cell proliferation and suppressed apoptosis. Higher levels of METTL3 expression were associated with more aggressive tumor invasion. Conclusion: The hsa_circ_0000523/miR-let-7b/METLL3 axis functions in the tumorigenesis and pathogenesis of human CRC. Our results suggest that circRNAs and METTL3 may be used for the highly sensitive diagnosis of CRC and predicting prognosis in patients who have undergone therapy. to be a microRNA In the pattern of of METTL3, which be a target gene miR-let-7b, was similar to that of hsa_circ_0000523. Through transfection of METLL3-expressing plasmid and METTL3-specic siRNA oligo, we demonstrated that METLL3 promoted the proliferation, apoptosis, and migration of HCT116 cells. Our work suggested a potential role for the hsa_circ_0000523/miR-let-7b/METLL3 axis in the tumorigenesis and pathogenesis of human colorectal cancer.

Jiang et al. identi ed a large set of circRNA with signi cantly differential expression in a primary colorectal cancer cell line (SW480) and a metastatic colorectal cancer cell line (SW620), relative to a normal colon cell line (NCM460) [15]. Hsa_circ_000984 has been reported to promote colon cancer growth and metastasis by sponging miR-106b [16]. Moreover, Bachmayr-Heyda [17] et al. demonstrated a global reduction in cirRNA abundance in colorectal cancer cell lines and clinical colorectal cancer specimens, compared to normal tissues. The authors described ve speci c cirRNAs (circ0817, circ3203, circ6229, circ7374, and circ7780) and proposed a potentially negative correlation between global circular RNA abundance and cell proliferation.
Among 5 cirRNAs identi ed by Bachmayr-Heyda et al., we were interested in circ6229 (also known as hsa_circ_0000523) and its corresponding gene/linear mRNA, METTL3 (methyltransferase-like 3). As a major RNA N6-adenosine methyltransferase (m 6 A), METTL3 is widely implicated in mRNA biogenesis, decay, and translation control [18]. METTL3 has recently been shown to promote the growth, survival, and invasion of many human cancers [19], such as lung cancer [20], hepatocellular carcinoma [21], breast cancer [22], and pancreatic cancer [23]. However, the exact role of METTL3 in the tumorigenesis and pathogenesis of colorectal cancer is largely unknown.
HCT116 cells are a population of malignant cells isolated from the primary cell culture of a single human colonic carcinoma [24]. In this study, using HCT116 cells, we identi ed a negative correlation between expression levels of hsa_circ_0000523 and miR-let-7b, which is considered to be a microRNA with tumor suppressor ability [25]. In HCT116 cells, the pattern of expression of METTL3, which may be a target gene of miR-let-7b, was similar to that of hsa_circ_0000523. Through transfection of METLL3-expressing plasmid and METTL3-speci c siRNA oligo, we demonstrated that METLL3 promoted the proliferation, apoptosis, and migration of HCT116 cells. Our work suggested a potential role for the hsa_circ_0000523/miR-let-7b/METLL3 axis in the tumorigenesis and pathogenesis of human colorectal cancer.
Transfection of plasmids and siRNA oligos HCT116 cells were transfected with the empty pcDNA3.1 vector, METTL3-expressing pcDNA3.1 vector (pcDNA3.1-METTL3), negative control siRNA oligo (siR-NC), METTL3-speci c siRNA oligo (siR-METTL3), negative control oligo for hsa_circ_0000523 (siR-circ-NC), or circ-0000523-speci c siRNA oligo (siR-circ-0000523), using Lipofectamine 2000 reagent (Thermo Fisher Scienti c), according to the manufacturer's instructions. HCT116 cells seeded in 6-well plates were maintained at con uence of 70% to 80% in complete culture medium. Then 2.5 μg plasmid or 100 pmol siRNA oligo was added to 250 μL Opti-MEM (Thermo Fisher Scienti c), and 5 μL Lipofectamine 2000 reagent was added to 250 μL Opti-MEM. The diluted plasmid or siRNA oligo was mixed with diluted Lipofectamine 2000 reagent and kept at room temperature for 20 min. After 500 μL serum-free medium had been placed in each well, Opti-MEM mixture was added. After incubation of transfected cells at 37˚C for 4 h, the culture medium was changed to regular DMEM supplemented with 10% FBS.

RNA isolation and reverse transcription
Total RNA was extracted from cells with Trizol reagent (Thermo Fisher Scienti c), according to the manufacturer's protocols. Brie y, after treatment, cell pellets were re-suspended in 1 mL Trizol and mixed well. Then, 0.2 mL of chloroform was added to the cell suspension, which was shaken vigorously for 15 s, then incubated for 3 min at room temperature. After centrifugation of the suspension at 12,000´g for 10 min at 4˚C, supernatant was collected and subjected to RNA precipitation by adding 0.5 mL of isopropanol at 4˚C. The RNA pellet was obtained by centrifugation at 12,000´g for 10 min at 4˚C, then washed with 75% ethanol. The appropriate amount of RNase-free distilled water was used to dissolve extracted RNA. The concentration of RNA was measured with a NanoDrop 2000 Spectrophotometer (Thermo Fisher Scienti c). For each sample, 1 μg of RNA was reverse-transcribed to cDNA with the ReverTra Ace qPCR RT Kit (catalogue number FSQ-101; Toyobo Co., Ltd.; Osaka, Japan), according to the speci cations in the product manual.

Real-time quantitative PCR (RT-qPCR)
For RT-qPCR, SYBR Green Realtime PCR Master Mix (catalogue number QPK-212; Toyobo Co., Ltd.; Osaka, Japan) was used to measure the expression of targeted genes, according to the manufacturer's instructions. For quanti cation of hsa_circ_0000523 and let-7b, U6 was used as an internal reference gene. GAPDH was used as an internal reference gene for quanti cation of METTL3. The conditions for PCR were set as follows: pre-denaturation at 95°C for 1 min, followed by 40 cycles of denaturation at 95°C for 15 s, and annealing and elongation at 60°C for 30 s. The 7900HT Real-Time PCR System (Applied Biosystems; Thermo Fisher Scienti c) was used to perform the assay. The 2 -ΔΔCt method was used to calculate differences between the experimental and control groups in expression of the target gene. The calculation formula was as follows: ΔΔCt = ΔCt experimental group -ΔCt normal group , ΔCt = Ct target gene -Ct internal reference . The sequences of primers used for RT-qPCR are presented in Table 1. Cell apoptosis HCT116 cells seeded in 6-well plates were transfected with empty pcDNA3.1 vector, METTL3-expressing pcDNA3.1 vector, siR-NC oligo, or siR-METTL3 oligo. At 48 h after transfection, cells were detached by incubation with 0.25% trypsin-EDTA solution (Gibco Laboratories), then harvested. Flow cytometric analysis was performed to detect apoptosis using the Annexin V-uorescein isothiocyanate (FITC) /propidium iodide (PI) Apoptosis Kit (Beyotime Biotech Inc., Shanghai, China), according to the manufacturer's protocol. Brie y, after one wash in phosphate-buffered saline and one wash in binding buffer, cells were stained with Annexin V-FITC/PI for 20 min at room temperature, in the dark. After another wash in binding buffer, labeled cells were detected immediately by a ow cytometer (CytoFLEX S, Beckman Coulter; Brea, CA, USA). Data were analyzed with CytExpert Software (Beckman Coulter).

Cell migration
The cellular potential for migration was determined using a 24-well transwell plate with 8.0-μm Pore Polyester Membrane Inserts (Corning; Thermo Fisher Scienti c). Matrigel (Corning) was melted overnight at 4˚C and diluted to a nal concentration of 1 mg/mL in pre-cooled serum-free medium. Then, 100 mL of diluted matrigel was added to the bottom of the upper chamber. The plate was then incubated at 37˚C for 4-5 h to dry the matrigel. At 24 h after transfection, HCT116 cells in the logarithmic growth phase were seeded in triplicate at a density of 1×10 6 cells/well. Cells were seeded on top of the transwell plate, in 100 μL DMEM supplemented with 0.1% bovine serum albumin. Then 0.8 mL of DMEM supplemented with 10% FBS was added to the lower chamber as a chemoattractant. After 24 h of incubation, cells on the top surface of the insert were removed with a cotton swab. Cells that had migrated to the lower surface of the membrane were xed with 4% paraformaldehyde, stained with 800 μL Giemsa solution (Beyotime Biotech Inc., Shanghai, China). Cells were visualized by a microscope (CKX-41; Olympus, Japan). Counts were obtained for three randomly selected optical elds.

Statistical analysis
All data were analyzed using SPSS Version 21.0 software (IBM Corp., USA). These data are presented as mean ± standard deviation (SD). Quantitative data were compared using the chi-square test.
Comparisons between groups were performed with Student's t-test. P < 0.05 was accepted as an indication of statistical signi cance.

Results
Hsa_circ-0000523 regulates METTL3 expression by modulating levels of miR-let-7b in HCT116 cells To explore the role of hsa_circ_0000523 circRNA in the tumorigenesis and pathogenesis of CRC, we established an in vitro model by manipulating the expression of hsa_circ_0000523 in the HCT116 human CRC cell line. As shown in Figure 1A, the mammalian expression vector pcDNA3.1 mediated overexpression of hsa_circ_0000523. The transcription of hsa_circ_0000523 was increased by more than 20fold in HCT116 cells, compared to cells transfected with empty vector. E cient knockdown of hsa_circ_0000523 was observed in HCT116 cells transfected with hsa_circ_0000523-speci c siRNA oligo, which displayed a decrease in expression of hsa_circ_0000523 of > 80%, compared with levels observed in HCT116 cells transfected with control siRNA oligo (Fig. 1A).
We then evaluated the impact of hsa_circ_0000523 expression on the expression of miR-let-7b, a microRNA that may be a taret of hsa_circ_0000523 in HCT116 cells. A strongly negative correlation between the expression of hsa_circ_0000523 and miR-let-7b was observed. Among the various treatment conditions, HCT116 cells transfected with hsa_circ_0000523-expressing plasmid displayed the highest level of hsa_circ_0000523 and the lowest level of miR-let-7b. Among all treatment groups, HCT116 cells with signi cant knockdown of hsa_circ_0000523 had the highest expression of miR-let-7b (Fig. 1B). In HCT116 cells, expression of METTL3, one of the targets of miR-let-7b, was similar to that of hsa_circ_0000523 (Fig. 1C). HCT116 cells with the highest levels of hsa_circ_0000523 transcription also had the highest levels of METTL3 mRNA. HCT116 cells with the lowest levels of hsa_circ_0000523 transcription had the lowest levels of METTL3 mRNA (Fig. 1C). An hsa_circ_0000523/miR-let-7b/METTL3 axis was thus identi ed in HCT116 cells.
Positive correlation between METTL3 expression and cell proliferation in HCT116 cells Because METTL3 is the endpoint effector molecule for the hsa_circ_0000523/miR-let-7b/METLL3 axis, we focused on elucidating the effects of METTL3 expression on the proliferation, apoptosis, and migration of HCT116 cells. Similarly, HCT116 cells with over-expression or knockdown of METTL3 were generated by transfection of METTL3-expressing pcDNA3.1 plasmid or METTL3-speci c siRNA oligo, respectively. Levels of METTL3 mRNA were quanti ed by RT-qPCR. The results demonstrated that overexpression mediated by transfection with plasmid resulted in a > 200-fold increase in levels of METTL3 transcription. Compared with untreated wild-type HCT116, siRNA oligo-transfection-mediated silence led to a > 70% decrease in levels of METTL3 transcription ( Fig. 2A).
After transfection with plasmids or siRNA oligos, HCT116 cells were further cultured for 24 h, 48 h, or 72 h. Cell proliferation was measured at these time-points using the CCK-8 assay. As expected, there was no signi cant difference in cell proliferation among untreated wild-type HCT116 cells, empty vectortransfected HCT116 cells, and negative control siRNA oligo-transfected HCT116 cells (Fig. 2B). Compared with the empty vector-transfected HCT116 cells, HCT116 cells with ectopic expression of METTL3 exhibited a signi cantly higher rate of proliferation at each time-point. Cell proliferation was decreased in HCT116 cells with knockdown of METTL3 cells transfected with negative control siRNA oligos (Fig. 2B). Taken together, these results indicate a positive correlation between METTL3 expression and the rate of cell proliferation in HCT116 cells.

METTL3 regulates apoptosis in HCT116 cells
We evaluated whether and how the expression of METTL3 impacts apoptosis in HCT116 cells after transfection of plasmid or siRNA oligo with annexin V and PI staining. As expected, there was no signi cant difference on apoptotic rates among the untreated wild type HCT116 cells, empty vectortransfected HCT116 cells and negative control siRNA oligo-transfected HCT116 cells, all of which demonstrated an apoptotic rate of around 15% (Fig. 3A and Fig. 3B). The rate of apoptosis was signi cantly decreased in HCT116 cells with ectopic expression of METTL3, compared with HCT116 cells transfected with empty vector (12% vs, 15%). The rate of apoptosis was much higher in HCT116 cells with knockdown of METTL3 than cells transfected with negative control siRNA oligos (20% vs. 15%; Fig.  3A and Fig. 3B). Therefore, an evidently negative correlation between METTL3 expression and apoptotic rate was observed, which suggested that METTL3 inhibited apoptosis in HCT116 cells.

Higher expression of METTL3 is associated with more aggressive tumor invasion in HCT116 cells
To explore the role of METTL3 in regulating the metastasis of CRC, we performed transwell assays to examine the migration ability of HCT116 cells with various levels of METTL3 expression. After plasmid or siRNA oligo transfection, HCT116 cells were seeded in the upper chamber of transwell plates, then cultured in medium containing a very low concentration of serum (0.5% BSA). The number of cells that had migrated to the surface of the lower chamber lled with DMEM containing 10% FBS can be used as an index for the metastatic ability of HCT116 cells. As shown by Giemsa staining of migrating cells in Figure 4A, more positively stained cells were observed in the group with over-expression of METTL3, compared with the group with knockdown of METTL3. Statistical analysis con rmed the positive correlation between the expression of METTL3 and the invasion abilities of HCT116 cells. As expected, there was no signi cant difference in migration ability among untreated wild-type HCT116 cells, empty vector-transfected HCT116 cells, and negative control siRNA oligo-transfected HCT116 cells (Fig. 4B). Compared with empty vector-transfected HCT116 cells, HCT116 cells with ectopic expression of METTL3 exhibited signi cantly higher migration ability. Consistently, migration ability was decreased in HCT116 cells with knockdown of METTL3, compared with cells transfected with negative control siRNA oligos (Fig. 4A and Fig. 4B). These results suggest that higher expression of METTL3 was associated with more aggressive tumor invasion in HCT116 cells.

Discussion
CircRNAs modulate the expression of parental genes by regulating alternative splicing or transcription and acting as competitive sponges for endogenous RNA or miRNA. CircRNAs may provide more comprehensive information on key genes involved in the oncogenesis and development of many cancers. They are potential accessible and noninvasive biological markers for the early detection of CRC. However, very few studies have reported on the speci c functions of circRNAs in the tumorigenesis and pathogenesis of human CRC. In this study, we explored the expression patterns of hsa_circ_0000523 and its parental gene METTL3 in the human CRC cell line HCT116. Our results identi ed a potential hsa_circ_0000523/miR-let-7b/METLL3 axis. The results of gain-of-function and loss-of-function assays showed that expression of METTL3 promoted cell proliferation and migration and inhibited apoptosis in HCT116 cells.
Accumulating evidence has con rmed the abnormal expression and important biological functions of circRNAs in CRC [12][13][14]. Interestingly, the expression patterns of most cirRNAs identi ed by bioinformatics approaches have been reported to be signi cantly down-regulated in CRC cell lines and clinical CRC tissues [26]. Huang et al. demonstrated that the expression of circ-ITCH was much lower in CRC tissues, compared with adjacent noncancerous tissues [27]. Cir-ITCH was found to sponge tumorigenic miR-7 and miR-20a, which contribute to the malignancy of CRC. A positive correlation between transcription of circ-ITCH and parental gene ITCH has been identi ed [27]. More importantly, expression of circ-ITCH was found to suppress cell proliferation in CRC cells [27]. Similarly, Zhu et al. reported that circ-BANP was expressed in 35 CRC tissues, and knock down of circ-BANP signi cantly decreased the proliferation of CRC cells [28].
We evaluated the expression levels of hsa_circ_0000523, its potential target miR-let-7b, and its parental gene METTL3 in HCT116 cells with varied expression levels of hsa_circ_0000523. Similar to the expression patterns of cir-ITCH and ITCH, the transcriptional level of hsa_circ_0000523 was positively associated with mRNA levels of linear METTL3 in HCT116 cells. Although we did not measure the impacts of hsa_circ_0000523 expression on cell proliferation, the ability of METTL3 expression to increase HCT116 proliferation suggest a similar proliferative role for hsa_circ_0000523. However, a recent report from Jin et al. demonstrated that hsa_circ_0000523 exerted anti-proliferative effects and promoted apoptosis in two other human CRC cell lines, SW480 and SW620 [29]. Therefore, although hsa_circ_0000523 can modulate the expression of METTL3 by acting on miR-let-7b, additional studies will be necessary to determine whether the expression of METTL3 can signi cantly impact the transcription of hsa_circ_0000523 in HCT116 cells.
CircRNA may regulate the proliferation of CRC by sequestering multiple miRNAs. For instance, circHIPK3 was shown to be over-expressed in CRC cells, compared with normal tissue, and regulates cell proliferation by sponging 9 miRNAs with 18 potential binding sites [30]. Speci cally, circ-HIPK3 was shown to bind to and inhibit the activity of miR-124, a tumor suppressor usually down-regulated in CRC [31]. Therefore, circRNAs may modulate the tumorigenic proliferation of CRC cells by diminishing the antitumor effects of certain tumor suppressive miRNAs.
MiR-let-7b, shown here to be a potential target of hsa_circ_0000523, is also a tumor suppressor in multiple cancers [25]. Human miR-let-7b was found to be down-regulated in various cancers, and the induction of tumorigenesis can be inhibited in normal cells by ectopic expression of miR-let-7b [32,33]. Human miR-let-7 can inhibit cancer growth by targeting various oncogenes and inhibiting key regulators of several mitogenic pathways in cancer. These results point to the therapeutic potential of human miRlet-7 in cancer therapy [25]. Therefore, high levels of hsa_circ_0000523 and low levels of miR-let-7 may contribute to tumorigenesis in CRC.
On the other hand, circRNA may regulate the proliferation of CRC by indirectly targeting oncogenes. In this study, METTL3, the parental gene of hsa_circ_0000523, and a target gene for miR-let-7b, was shown to promote the growth, survival, and invasion of human lung cancer [20]. Moreover, the oncogenic roles of METTL3 have been demonstrated in many solid tumors and hematopoietic malignancies [34,35]. To the best of our knowledge, this is the rst report to describe the role of METTL3 in CRC cells. The expression of METTL3 in HCT116 cells was found to be positively correlated with cell proliferation and cancer invasion, but negatively correlated with apoptosis. These ndings suggest that METTL3 functions like an oncogene in CRC, possibly by promoting the translation of other oncogenes. In contrast, METTL3 was found to be a tumor suppressor in renal cell carcinoma [36]. METTL3 has been demonstrated to promote cell proliferation, migration, and invasion in two human renal cell carcinoma cell lines, CAKI-1 and CAKI-2. These effects are mediated by modulation of the epithelial-to-mesenchymal transition and PI3K-Akt-mTOR signaling [36]. The molecular mechanisms underlying the diverse roles of METTL3 in renal cell carcinoma and CRC remain to be elucidated.

Conclusion
Through mammalian expression vector transfection-mediated gain-of-function studies and siRNA oligo transfection-mediated loss-of-function studies, we identi ed a potential hsa_circ_0000523/miR-let-7b/METLL3 axis that contributes to tumorigenesis and pathogenesis in the human CRC cell line HCT116. METTL3 was shown to promote cell proliferation and migration and to inhibit apoptosis in HCT116 cells. Our work on the hsa_circ_0000523/miR-let-7b/METLL3 axis may facilitate highly sensitive diagnosis of CRC and prognosis prediction in patients after therapy. Table 1 The sequences of primers used for RT-qPCR in this study. At 48 h after transfection, HCT116 cells were harvested, and transcriptional levels of hsa_circ-0000523 (a), miR-let-7b (b) and METTL3 (c) were quanti ed by RT-qPCR. WT, untreated wild-type HCT116 cells; OE,