Circ_PIP5K1A regulates cisplatin resistance and malignant progression in non-small cell lung cancer cells and xenograft murine model via depending on miR-493-5p/ROCK1 axis

Background Chemoresistance limits the therapeutic effect of cisplatin (DDP) on non-small cell lung cancer (NSCLC). Circular RNAs (circRNAs) function as important regulators in chemoresistance. This study aimed to explore the regulation of circRNA Phosphatidylinositol-4-Phosphate 5-Kinase Type 1 Alpha (circ_PIP5K1A) in DDP resistance. Methods The expression analysis of circ_PIP5K1A, micoRNA-493-5p (miR-493-5p) and Rho Associated Coiled-Coil Containing Protein Kinase 1 (ROCK1) was conducted through reverse transcription-quantitative polymerase chain reaction (RT-qPCR). Cell sensitivity was determined using 3-(4,5-dimethylthiazol-2-y1)-2,5-diphenyl tetrazolium bromide (MTT) assay. Cell proliferation and cell viability were evaluated by colony formation assay and MTT assay, respectively. Cell cycle and apoptosis detection was performed via flow cytometry. Cell motility was examined by transwell migration or invasion assay. Dual-luciferase reporter assay was applied to confirm the target binding. ROCK1 protein level was assayed via Western blot. In vivo assay was carried out using xenograft model in mice. Results Circ_PIP5K1A level was abnormally increased in DDP-resistant NSCLC tissues and cells. Silencing circ_PIP5K1A reduced DDP resistance, proliferation, cell cycle progression and cell motility in DDP-resistant NSCLC cells. Circ_PIP5K1A directly interacted with miR-493-5p in NSCLC cells. The function of circ_PIP5K1A was dependent on the negative regulation of miR-493-5p. MiR-493-5p directly targeted ROCK1 and circ_PIP5K1A regulated the ROCK1 level via acting as a sponge of miR-493-5p. Overexpression of miR-493-5p inhibited chemoresistance and cancer progression by downregulating ROCK1 expression in DDP-resistant NSCLC cells. Circ_PIP5K1A regulated DDP sensitivity in vivo via the miR-493-5p/ROCK1 axis. Conclusion These findings suggested that circ_PIP5K1A upregulated the ROCK1 expression to promote DDP resistance and cancer progression in NSCLC by sponging miR-493-5p.


Background
Non-small cell lung cancer (NSCLC) is a familiar fatal malignancy that accounts for more than 80% cases of lung cancer [1]. Cisplatin (DDP) is an effective chemotherapeutic drug for various kinds of cancer, but drug resistance usually leads to treatment failure [2]. DDPbased chemotherapy has also been a first-line strategy for metastatic NSCLC [3]. Reducing DDP resistance is essential for the better treatment of NSCLC patients. Circular RNAs (circRNAs) play important roles in cancer biology by functioning as molecular sponges of microRNAs (miRNAs) and inducing expression changes of downstream genes [4,5]. CircRNA Phosphatidylinositol-4-Phosphate 5-Kinase Type 1 Alpha (circ_PIP5K1A, hsa_circ_0014130) contributed to the malignant progression of colon cancer via sponging miR-1273a [6] and promoted the developing process of gastric cancer by the miR-376c-3p/zinc finger protein 146 (ZNF146) network [7]. NSCLC research indicated that circ_PIP5K1A facilitated carcinogenesis and development by regulating different miRNA/mRNA axes, including miR-600/ HIF-1α, miR-142-5p/insulin-like growth factor-1 (IGF-1) and miR-136-5p/B-cell lymphoma-2 (Bcl-2) [8][9][10]. The potential effect of circ_PIP5K1A on DDP resistance in NSCLC is still unclear.
In this study, circ_PIP5K1A was hypothesized as a miR-493-5p sponge to result in the expression change of ROCK1 in NSCLC. The aim of this research was to investigate the circ_PIP5K1A/miR-493-5p/ROCK1 axis in chemoresistance and carcinogenesis of NSCLC.

Tumor tissues
This research was authorized by the Ethics Committee of The Second Affiliated Hospital of Xi'an Jiaotong University. Tumor tissues from NSCLC patients were acquired at The Second Affiliated Hospital of Xi'an Jiaotong University. Patients without recurrence during primary DDP therapy and with recurrence beyond 6 months after chemotherapy were defined as Tumor-sensitive (n = 33). Patients with tumor progression during primary DDP therapy and recurrence within 6 months were defined as Tumor-resistant (n = 23). The physiopathologic diagnoses were affirmed by two experienced pathologists. 56 patients have afforded the informed consent for this study. Tissue samples were all preserved at − 70 °C for later use.

Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) assay
RNA isolation was carried out by TRIzol ™ Reagent (Invitrogen). SuperScript ™ IV First-Strand Synthesis System and SYBR ™ Green One-Step qPCR Kit (Invitrogen) were used to determine the levels of circ_PIP5K1A and ROCK1. The miR-493-5p expression was quantified through TaqMan Advanced miRNA cDNA Synthesis Kit and TaqMan ™ Advanced miRNA Assay (Applied Biosystems, Foster City, CA, USA). The calculation of relative expression was performed through 2 −∆∆Ct method [18]. Additionally, circ_PIP5K1A stability was analyzed via RT-qPCR following treatment of RNase R (GENESEED) and Actinomycin D (Sigma). The primer sequences were shown in Table 1. Glyceraldehyde-phosphate dehydrogenase (GAPDH) and U6 served as the housekeeping genes for circ_PIP5K1A/ROCK1 and miR-493-5p, respectively.

Colony formation assay
Cell proliferation was assessed using colony formation assay. Transfection with different oligonucleotides and vectors was conducted for 24 h. Subsequently, cells were collected and transplanted into 12-well plates for 2 weeks. 4% paraformaldehyde and 0.1% crystal violet (Sigma) were incubated to cell colonies for 15 min, then Image J software (NIH, Bethesda, MD, USA) was applied for number counting.

Flow cytometry
1 × 10 5 DDP-resistant cells were harvested by trypsin (Beyotime). Cell cycle analysis and apoptosis detection were respectively performed using Cell Cycle Analysis Kit (Beyotime) and Annexin V-FITC Apoptosis Detection Kit (Beyotime), according to the producer's instruction books. Cell determination was conducted through a flow cytometer (BD Biosciences, San Diego, CA, USA), followed by the analysis of cell distribution at different phases and the calculation of apoptotic rate.

Transwell assay
Cell motility was determined through transwell chamber (Corning Inc., Corning, NY, USA). 5 × 10 4 cells were pipetted into the top chamber for migration assay, and equal number of cells were seeded into the top chamber enveloped with matrigel (Corning Inc.) for invasion assay. The chamber was incubated at 37 °C for 24 h following the addition of cell medium into the bottom chamber. Then cells passed across to the membranes were measured on the inverted microscope (Olympus, Tokyo, Japan). Cell images were obtained under ×100 magnification, and the migrated or invaded cells were counted under three view of fields.

Dual-luciferase reporter assay
The binding sites between targets were predicted via the online starbase (http:// starb ase. sysu. edu. cn). Circ_ PIP5K1A sequence was inserted into the pmirGLO vector (Promega, Madison, WI, USA) to construct the wild-type plasmid (circ_PIP5K1A-WT). The miR-493-5p binding sites in circ_PIP5K1A sequence were mutated and the mutant control (circ_PIP5K1A-MUT) was obtained. Also, the luciferase plasmids for ROCK1 were defined as ROCK1-WT and ROCK1-MUT. DDP-resistant cells were transfected with miR-493-5p or miR-NC and each luciferase plasmid. After cell incubation at 37 °C for 48 h, luciferase activity analysis was carried out through Dual-luciferase Reporter Assay Kit (Promega).

Western blot
Radioimmunoprecipitation assay (RIPA) containing protease inhibitor was applied for extraction of total protein. 50 μg proteins of each sample were electrophoresed on 12% TruPAGE ™ Precast Gels using Sigma-Aldrich ® Dual Run and Blot System (Sigma) and transferred to Immobilon-E Polyvinylidene Fluoride Membrane (Sigma) through the Trans-Blot Turbo Transfer System (Bio-Rad). The non-specific proteins were blocked and the membranes were incubated with primary antibodies of ROCK1 (Abcam, Cambridge, UK; ab97592, 1:1000) or GAPDH (Abcam, ab128915, 1:1000) at 4 °C overnight. After the incubation of Goat Anti-Rabbit IgG H&L secondary antibody (ab205718, 1:3000), immunoreactive blots were visualized via Enhanced Chemiluminescence (ECL) Substrate (Bio-Rad). GAPDH were used as the internal reference, and the protein intensity was analyzed by ImageJ software (NIH).

Xenograft model in mice
BALB/c male nude mice were bought from Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China). Lentiviral vectors (RIBOBIO) was used for stable transfection, including lentivirus negative control (lenti-NC), lentiviral circ_PIP5K1A (lenti-circ_PIP5K1A), and lentiviral short hairpin RNA of circ_PIP5K1A (sh-circ_PIP5K1A). A549/DDP cells were respectively transfected with these lentiviral vectors, followed by cell injection (2 × 10 6 cells) into the mice with 6 mice/ group. 10 days later, mice were treated with intraperitoneal injection of 6 mg/kg DDP once 2 days and tumor Table 1 Primer sequences for RT-qPCR volume (length × width 2 × 0.5) was determined every 5 days. Mice were sacrificed through the flow rate of CO 2 after cell injection for 30 days, then tumors were dissected from mice and weighed on the electronic scale. Circ_PIP5K1A, miR-493-5p and ROCK1 levels were examined using RT-qPCR and Western blot assays. Ki67 (Sigma, SAB5600249) and Cleaved-caspase 3 (Sigma, SAB1305630) protein detection was performed by Immunohistochemistry (IHC) assay. All programs of animals were ratified by Animal Ethical Committee of The Second Affiliated Hospital of Xi'an Jiaotong University and in consistent with the Management and Use Guidelines of Laboratory Animals of NIH.

Statistical analysis
Data were displayed as mean ± standard deviation (SD) and statistical analysis was carried out through SPSS 22.0 (SPSS Inc., Chicago, IL, USA). The relationships between gene levels in tumor samples were determined through Pearson's correlation coefficient. The group difference was compared using Student's t-test and one-way analysis of variance (ANOVA) followed by Tukey's test. Statistically, P < 0.05 was termed as a significant difference.

Circ_PIP5K1A was upregulated in DDP-resistant NSCLC tissues and cells
Cell viability analysis showed that IC 50 value was higher in A549/DDP (IC 50 = 24.530) and H460/DDP (IC 50 = 18.150) cells than that in A549 (IC 50 = 9.626) and H460 (IC 50 = 8.663) cells, indicating that DDP resistance was presented in A549/DDP and H460/DDP cells (Fig. 1A, B). The results of RT-qPCR assay demonstrated that circ_PIP5K1A expression was obviously increased in tumor-resistant tissues relative to tumor-sensitive tissues (P < 0.001) and A549/DDP or H460/DDP cells compared with the A549 or H460 cells (P < 0.001, P = 0.001) (Fig. 1C, D). RNase R treatment significantly decreased the PIP5K1A mRNA level compared to the Mock group (P < 0.001), while the difference of circ_PIP5K1A expression between Mock and RNase R groups was not conspicuous (Fig. 1E, F). Also, the half-life of circ_PIP5K1A was much longer than PIP5K1A (P < 0.001) after cells were incubated to Actinomycin D (Fig. 1G, H). Circ_ PIP5K1A was an upregulated circRNA in DDP-resistant NSCLC samples and cells.

Circ_PIP5K1A knockdown inhibited DDP resistance and malignant behaviors in DDP-resistant NSCLC cells
The specific siRNA was used to inhibit the circ_ PIP5K1A level. As shown in Fig. 2A and H460/DDP cells (Fig. 2C, D). By performing colony formation assay (Fig. 2E) and MTT assay (Fig. 2F, G), we found that cell proliferation ability and cell viability were suppressed in si-circ_PIP5K1A group compared with si-NC group (P < 0.001). Flow cytometry revealed that si-circ_PIP5K1A resulted in cell cycle arrest from G0/G1 to S phase (Fig. 2H, I) but elevated cell apoptotic rate (Fig. 2J), by contrast with si-NC group (P < 0.001).
The migrated and invaded cells were reduced by si-circ_PIP5K1A transfection compared with si-NC transfection (P < 0.001), suggesting that circ_PIP5K1A expression repression restrained cell motility in A549/ DDP and H460/DDP cells (Fig. 2K, L). Thus, inhibition of circ_PIP5K1A repressed the chemoresistance to DDP and the malignant phenotypes in DDP-resistant cells.

MiR-493-5p acted as a sensitizer of DDP and tumor inhibitor in DDP-resistant NSCLC cells by inducing ROCK1 downregulation
The role and mechanism of miR-493-5p were explored in DDP-resistant cells. The miR-493-5p transfection induced the ROCK1 mRNA and protein expression reduction by comparison with miR-NC transfection (P < 0.001), then this expression change was relieved by transfection of ROCK1 relative to pcDNA transfection (P < 0.001) (Fig. 6A, B). IC 50 of DDP (Fig. 6C), cell proliferation and viability (Fig. 6D-F) in A549/DDP and H460/DDP cells were inhibited by miR-493-5p compared with miR-NC group (P < 0.001), which was notably counterbalanced by ROCK1 but not pcDNA (P < 0.001). The introduction of ROCK1 also attenuated the miR-493-5p-mediated cell cycle retardation (Fig. 6G, H), cell apoptosis enhancement (Fig. 6I) and migration or invasion suppression (Fig. 6J, K) contrasted to the introduction of pcDNA (P < 0.001). All in all, the inhibitory effects of miR-493-5p on DDP resistance and NSCLC development were associated with the downregulation of ROCK1.

Discussion
The therapeutic outcomes of cancer patients have been largely affected by chemoresistance. Herein, our results in vitro and in vivo manifested that knockdown of circ_PIP5K1A could enhance sensitivity of NSCLC to DDP. Circ_PIP5K1A might be applied as a biomarker to improve the DDP therapy for NSCLC patients. The were evaluated using transwell assay. ***P < 0.001 molecular mechanism of circ_PIP5K1A in the regulation of resistance was also disclosed for the first time.
The covalent closed structures confer circRNAs the high stability in human eukaryotes [19]. Our data showed that circ_PIP5K1A was more stable than linear PIP5K1A after treatment with RNase R and Actinomycin D in both A549/DDP and H460/DDP cells. CircR-NAs have pivotal roles in various kinds of biological processes, including carcinogenesis, cancer progression and drug resistance [20]. Qian et al. discovered that circ-G004213 significantly elevated the DDP sensitivity in liver cancer via regulating the miR-513b-5p/ pre-mRNA processing factor 39 (PRPF39) levels [21]. Wei et al. reported that circSAMD4A increased doxorubicin resistance in osteosarcoma cells through the effect on miR-218-5p/krüppel-like factor 8 (KLF8) Fig. 7 Circ_PIP5K1A regulated DDP sensitivity to NSCLC in vivo by the expression regulation of miR-493-5p and ROCK1. A, B tumor volume (A) and weight (B) were determined in DDP + lenti-NC, DDP + lenti-circ-PIP5K1A and DDP + sh-circ_PIP5K1A groups. C, D the levels of circ_PIP5K1A and miR-493-5p in tumor tissues were examined by RT-qPCR. E, F ROCK1 mRNA and protein detection was performed via RT-qPCR and Western blot in tumor tissues. G Ki67 and Cleaved caspase3 levels in tumor tissues were analyzed using IHC assay. *P < 0.05, **P < 0.01, ***P < 0.001 axis [22]. Circ_CELSR1 overexpression also promoted the chemoresistance of ovarian cancer cells to paclitaxel via mediating the salt inducible kinase 2 (SIK2) level by targeting miR-149-5p [23]. The expression analysis revealed that circ_PIP5K1A was significantly upregulated in tumor-resistant tissue samples and DDP-resistant NSCLC cells, which implied that circ_ PIP5K1A might be involved in the resistance of DDP in NSCLC. As expected, we found that circ_PIP5K1A downregulation repressed the IC 50 of DDP in resistant NSCLC cells. Cellular behavior analysis has shown that silence of circ_PIP5K1A induced the inhibitory effects on cell proliferation, cell viability, cell cycle progression and cell migration/invasion but the stimulative effect on apoptosis in DDP-resistant cells. All these findings suggested that circ_PIP5K1A inhibition reduced DDP resistance to further impede the progression of NSCLC.
Many studies have validated that the regulatory functions of circRNAs were associated with miRNA/ mRNA signal networks. For instance, circFBXW7 played a tumor-inhibitory role in lung adenocarcinoma via sponging miR-942-5p and increasing the BARX homeobox 2 (BARX2) expression [24]. Cir-cUBE2D2 enhanced cell proliferation and doxorubicin resistance in triple-negative breast cancer cells by controlling miR-512-3p/cell division cycle associated protein-3 (CDCA3) axis [25]. CircRNA_001275 increased the wingless-type protein 7a (Wnt7a) expression by competitively sponging miR-370-3p to facilitate the resistance of DDP to esophageal cancer cells [26]. In this study, circ_PIP5K1A upregulated the ROCK1 expression through sponging miR-493-5p. The regulation of circ_PIP5K1A knockdown was ascribed to the miR-493-5p level upregulation. Also, miR-493-5p inhibited chemoresistance and oncogenesis of DDPresistant cells by targeting ROCK1. Animal assay further revealed that circ_PIP5K1A expression reduction contributed to DDP sensitivity by affecting the miR-493-5p/ROCK1 axis in vivo.
There are still some limitations in the current study. For example, the further experiment of circ_PIP5K1A/ miR-493-5p/ROCK1 axis in regulating DDP resistance in vivo needs to be performed. Additionally, whether circ_PIP5K1A could regulate the signaling pathways via the miR-493-5p/ROCK1 axis remains unknown. ROCK1 has inactivated the LATS2-JNK pathway and PTEN/PI3K/FAK pathway in NSCLC progression regulation [27,28]. Exploring the signaling pathways in the downstream of circ_PIP5K1A/miR-493-5p/ROCK1 will be beneficial for the better understanding of the functional mechanism behind circ_PIP5K1A.

Conclusion
In conclusion, circ_PIP5K1A/miR-493-5p/ROCK1 axis was implicated in the regulation of DDP resistance and malignant behaviors in NSCLC cells and murine model. This study might show a novel perspective for increasing DDP sensitivity, with circ_PIP5K1A as a potential biological marker.