LINC02418 promotes malignant behaviors in lung adenocarcinoma cells by sponging miR-4677-3p to upregulate KNL1 expression

Human LAD cell lines (A549, SPC-A1, H1299 and PC-9) and normal lung epithelial cells BEAS-2B were bought from the Cell Bank of the Chinese Academy of Sciences and incubated with Dulbecco’s modified Eagle’s medium (DMEM; Invitrogen Life Technology Inc., Carlsbad, CA), containing 10% fetal bovine serum (FBS) with 5% CO₂ at 37 °C in humid air. All cell lines were authenticated via STR profiling before using.

The suppression of LINC02418 expression was achieved by sh-LINC02418#1/2. Sh-LINC02418#1 and sh-LINC02418#2 were obtained from GenePharma (Shanghai, China). LINC02418 and KNL1 were overexpressed with pcDNA3.1 vectors (Invitrogen, Carlsbad, USA). MiR-4677-3p mimics were applied to elevate miR-4677-3p expression. MiR-4677-3p mimics and NC mimics were also bought from GenePharma (Shanghai, China). The transfection of above plasmids was conducted by use of Lipofectamine® 2000 agent (Invitrogen).

Three pairs of LAD and matched non-cancerous tissue samples were acquired from patients (including two male, one female; two patients < 50 years old, one patient > 50 years old; one patient at stage of I-II; two patients at stage of III-IV) who received operation in the Second Affiliated Hospital of Air Force Medical University. Patients enrolled in this study signed the informed consents. Ethic Committee of the Second Affiliated Hospital of Air Force Medical University has approved sample collection of this research. Microarray analysis was implemented to profile the expression of lncRNAs in LAD. In detail, total RNA was isolated from three pairs of tissues and quantified utilizing NanoDrop 2000 (Thermo, Waltham, MA, USA), followed by quality-checking by use of Agilent Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA, USA). Subsequently, the GeneChip 3’IVT express kit (Affymetrix, Santa Clara, CA, USA) was utilized to label the qualified RNA samples and then Affymetrix GeneChip Primeview Human cDNA microarray was used for hybridization in line with the manufacturer’s guides. After that, data were analyzed using GeneChip Scanner 3000 (Affymetrix).

Total RNA was extracted utilizing TRIzol (Invitrogen; Thermo Fisher Scientific, Inc.) and diluted to 200 ng/ml. Complementary DNA (cDNA) synthesis was conducted via applying Taqman Advanced miRNA cDNA Synthesis Kit or Pyrobest DNA Polymerase and M-MLV Reverse Transcriptase (Thermo Fisher Scientific, USA). RT-qPCR was then operated using One Step SYBR® Prime Script™ RT-PCR Kit II (Takara Biotechnology Co., Ltd., Dalian, China) based on the producer’s protocol. Gene expression relative to GAPDH or U6 was assessed using the 2^-ΔΔCt method.

Cell proliferation was estimated utilizing the CCK-8 kit (Boster) based on the manufacturer’s requirements. Briefly, cells (1 × 10³) were supplemented into 96-well plates. After cell adhesion, each well received 10 μl CCK-8 solution and then cells were further incubated for 1 h at 37 °C. Cell proliferation ability was monitored by detecting absorbance at 450 nm utilizing microplate reader (EL340; Bio-Tek Instruments, Hopkinton, MA, USA) at the indicated time points (0, 24, 48, 72 and 96 h).

Cells were seeded in six-well plates and grown in media with 10% FBS. Two weeks later, the colonies were fixated using methanol and dyed using 0.1% crystal violet for half an hour. Colonies with over 50 cells were counted manually.

For invasion estimation, 5 × 10⁴ cells were added onto the upper chambers (BD Biosciences, San Jose, CA, USA) with Matrigel-coating and incubated in DMEM. DMEM supplementing with 10% FBS was put into the bottom chambers. Twenty-four hours later, cotton swabs were used to scrape off cells in the upper chamber. The methanol and 0.5% crystal violet were separately use to fasten and color the cells in the lower chamber. For the migration assays, transfected cells were seeded into the upper chambers with no Matrigel-coating, while other steps were similar to that in invasion assays. Finally, the invaded or migrated cells were counted using an inverted biological microscope (magnification, Å ~ 200).

Cells were lysed by use of RIPA (Beyotime, Shangahi, China) containing protease inhibitor cocktail (Roche, Pleasanton, CA) and phenylmethylsulfonyl fluoride (Roche). Protein samples were then subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), followed by transferring onto nitrocellulose (NC) membranes (Sigma-Aldrich). After blocking via 5%-skim milk, the membranes were cultured with primary antibodies (dilution 1:1000) against Bax, Bcl-2, E-cadherin, N-cadherin, MRP2, MRP9, KNL1 (Cell Signaling Technology, Danvers, MA), followed by incubation with secondary antibodies (dilution 1:10000) for an hour at room temperature. β-actin or GAPDH was the loading control. Thereafter, signals were captured with the employment of the ECL chromogenic substrate.

After fixation and permeabilization, cells were processed with dUTP-end labeling (Clontech, Mountain View, CA) and (4′,6-diamidino-2-phenylindole) DAPI, in succession. After that, cells were observed and analyzed under fluorescent microscope (Olympus, Tokyo, Japan).

A549 and SPC-A1 cells (2.0 × 10⁴) grown in a 96-well plates were co-transfected with 150 ng of LINC02418-WT or LINC02418-Mut reporters (Sangon Biotech, Shanghai, China) and miR-4677-3p mimic or NC mimics into LAD cells by use of Lipofectamine 2000 (Invitrogen, Carlsbad, California, USA). KNL1-WT or KNL1-Mut reporters (Sangon Biotech, Shanghai, China) and miR-4677-3p mimics or NC mimics were also co-transfected into indicated LAD cells. After transfection for 2 days, the luciferase activity normalized to Renilla luciferase activity was examined by luciferase reporter assay system (Promega, Madison WI).

RIP assay was conducted utilizing the EZ-Magna RIP kit (Millipore, Billerica, MA) according to the manufacturer’s protocol. A549 and SPCA1 cells at 80–90% confluence were obtained, and then lysed in complete RIP lysis buffer. Cell extract was processed at 4 °C for about 6 h with RIP buffer which contained human Ago2 antibody or control IgG (Millipore) coated magnetic beads. After beads washed, the RNA complexes were cultured with Proteinase K to digest proteins. RNA concentration was measured though employing a NanoDrop spectrophotometer (Thermo Scientific), with the quality assessed by a bioanalyser (Agilent, Santa Clara, CA). Finally, the immunoprecipitated RNAs were purified and analysed by RT-qPCR.

RNA pull-down assay was used to detect the probable interaction among miR-4677-3p, LINC02418 and KNL1. MiR-4677-3p was biotinylated to be miR-4677-3p biotin probe by GenePharma Company (Shanghai, China). MiR-4677-3p biotin probe and miR-4677-3p no-biotin probe were added into the lysates of A549 and SPC-A1 cells. After 48 h incubation, Dynabeads M-280 Streptavidin (Invitrogen, CA) were put into above mixture. Two hours later, RNA in the pulled down complexes was examined using RT-qPCR analysis after purification.

To identify the molecular mechanisms underlying LAD development, we first examined the differentially expressed lncRNAs in LAD samples relative to paired non-tumor ones. The upregulated lncRNAs in LAD tissues in comparison with adjacent normal tissues were displayed using a heatmap (Fig. 1a). Then, the top ten upregulated lncRNAs in LAD tissues were further analyzed in LAD cell lines via qRT-PCR. Results manifested that compared to the normal BEAS-2B cells, LINC02418 was expressed higher in four LAD cell lines (Fig. 1b; ^**P < 0.01). However, other 9 lncRNAs were not significantly differential expressed in the four kinds of LAD cells relative to control cells (Figure S1A; ^*P < 0.05, ^**P < 0.01, n.s. was no significance). Thus, we chose LINC02418 as the research object in subsequent experiments. A549 and SPC-A1 cells possessing highest LINC02418 level were used for loss-of function assays. To probe the biological role of LINC02418, LINC02418 was knocked down in A549 and SPC-A1 cells by transfection with sh-LINC02418#1/2, with sh-NC-transfected cells as the scramble control. The results showed that knockdown of LINC02418 resulted in an obvious decline in LINC02418 expression compared with control group (Fig. 1c; ^*P < 0.05). Additionally, CCK-8 and colony formation assays indicated that LINC02418 absence inhibited the proliferation of these two LAD cells (Fig. 1d-e; ^**P < 0.01). TUNEL assay illustrated that cell apoptosis was promoted by LINC02418 depletion (Figure S2A; ^**P < 0.01). As shown in Fig. 1f (^**P < 0.01) and S2B (^**P < 0.01), the migration and invasion of sh-LINC02418#1/2-transfected cells were notably blocked in contrast to that in sh-NC group, suggesting that LINC02418 silencing weakened LAD cell migration and invasion abilities. Furthermore, western blot analyses displayed that silencing LINC02418 suppressed the expression of migration-related proteins (MRP2 and MRP9), which indicated that LINC02418 depletion could inhibit cell migration (Fig. 1g). Further, when downregulating LINC02418, the levels of Bax and E-cadherin were augmented while Bcl-2 and N-cadherin expression was declined (Figure S2C). In total, LINC02418 is upregulated and LINC02418 knockdown suppresses cell proliferation and motility in LAD.

Thereafter, bioinformatics analyses were employed to predict the miRNAs that could possibly bind with LINC02418. As a result, miR-4677-3p was then identified from starBase (Fig. 2a). In order to investigate whether LINC02418 could regulate miR-4677-3p expression, RT-qPCR analysis was performed. Interestingly, an observable increase in the expression of miR-4677-3p was observed in LAD cells with LINC02418 deficiency (Fig. 2b; ^**P < 0.01). Additionally, miR-4677-3p was found to be with a low expression trend in LAD cell lines relative to BEAS-2B cells (Fig. 2c; ^**P < 0.01). Subsequently, we observed a significantly heightened level of miR-4677-3p in miR-4677-3p mimics group compared with NC mimics group (Fig. 2d; ^**P < 0.01). Moreover, miR-4677-3p mimics suppressed cell proliferation in two LAD cells (Fig. 2e-f; ^**P < 0.01). Furthermore, overexpression of miR-4677-3p notably impaired LAD cell migration capability (Fig. 2g; ^**P < 0.01). Moreover, as displayed in Fig. 2h, miR-4677-3p upregulation decreased the expression of migration-related proteins (MRP2 and MRP9). Taken together, miR-4677-3p is downregulated and serves a tumor-restraining part in LAD.

Based on research over the past decades, miRNAs are known to have a vital impact on cancer progression by directly modulating the expression of its target genes [24‐26]. Herein, 8 candidates (MAP 1LC3B, IGSF3, KNL1, MKL2, EFTUD2, CA12, MOCS1 and SLC31A1) were identified as the potential targets of miR-4677-3p following analyses of the RNA22 and miRmap databases (Fig. 3a). RT-qPCR was then employed to assess the influence of miR-4677-3p on the expression of these genes in A549 and SPC-A1 cells. Results indicated that only KNL1 expression was notably reduced by miR-4677-3p mimics compared with that in NC mimics control group in both the two LAD cells, while the level of other genes not (Fig. 3b; ^*P < 0.05, ^**P < 0.01). Therefore, KNL1 was selected to carry out the following assays. As illustrated in Fig. 3c, the level of KNL1 protein was also decreased by miR-4677-3p mimics. Besides, both mRNA and protein expressions of KNL1 were also hampered in face of LINC02418 knockdown (Fig. 3d; ^**P < 0.01). In addition, LINC02418, miR-4677-3p and KNL1 were all concentrated in anti-Ago2 group instead of anti-IgG group (Fig. 3e; ^***P < 0.001). Furthermore, the binding sequences between miR-4677-3p and LINC02418 were obtained from starBase (Fig. 3f, left). Then we found that the luciferase activity of pmirGLO-LINC02418-WT was reduced by miR-4677-3p mimics whereas that of pmirGLO-LINC02418-Mut showed no obvious alteration between miR-4677-3p mimic group and NC mimic group (Fig. 3f, right; ^**P < 0.01). Similarly, starBase predicted the sequences of miR-4677-3p and KNL1 where the binding occurred (Fig. 3g, upper). The luciferase activity of pmirGLO-KNL1-WT was also lowered by enhanced miR-4677-3p while that of pmirGLO-KNL1-Mut wasn’t affected (Fig. 3g, lower; ^**P < 0.01). Moreover, the results of RNA pull-down assay testified that both LINC02418 and KNL1 were enriched in miR-4677-3p biotin probe group (Fig. 3h; ^**P < 0.01). In sum, KNL1 is the downstream molecule of miR-4677-3p in LAD.

Subsequently, we planned to determine whether LINC02418 influenced LAD development through miR-4677-3p/KNL1 pathway. RT-qPCR and western blot analyses indicated that co-transfection of pcDNA3.1/KNL1 recovered LINC02418 depletion-lessened KNL1 expression in A549 cells (Fig. 4a-b; ^**P < 0.01, n.s. was no significance). Then, it was verified that KNL1 upregulation reversed the hindering impact of LINC02418 deficiency on A549 cell proliferation (Fig. 4c-d; ^**P < 0.01, n.s. was no significance). Additionally, the stimulated cell apoptosis induced by LINC02418 silencing was rescued by KNL1 overexpression (Figure S3A; ^**P < 0.01, n.s. was no significance). Subsequently, it was confirmed that overexpression of KNL1 offset the obstructive function caused by LINC02418 depletion on A549 cell motility (Fig. 4e and S3B; ^**P < 0.01, n.s. was no significance). Additionally, KNL1 upregulation neutralized the influence of LINC02418 depletion on the levels of Bax, Bcl-2, E-cadherin and N-cadherin (Figure S3C). These findings suggest that LINC02418 aggravates malignant behaviors in LAD via miR-4677-3p/KNL1 pathway.