Long Non-Coding RNA BLACAT1 Promotes the Tumorigenesis of Gastric Cancer by Sponging microRNA-149-5p and Targeting KIF2A

Introduction

Gastric cancer (GC) is a gastrointestinal tumor and the third cancer-related death cause worldwide.^1,2 GC is asymptomatic in early stages, and diagnosed at an advanced stage in most patients.³ The mortality of GC is high and successful therapeutic strategies are limited.⁴ Thus, it is indispensable to explore critical molecules involved in GC cell tumorigenesis and develop new effective therapies.

Long noncoding RNAs (lncRNAs) take part in biological processes of gastrointestinal malignancies, such as colorectal cancer,⁵ esophageal cancer⁶ and GC.⁷ LncRNA bladder cancer-associated transcript 1 (BLACAT1) participates in multiple biological processes of cancers. For instance, BLACAT1 exhibits tumor pro-oncogenic activity and is a negative prognostic factor in colorectal cancer.⁸ BLACAT1 contributes to tumorigenesis of cervical cancer cells.⁹ BLACAT1 high-expression acts as a poor outcome biomarker for overall survival in small-cell lung cancer.¹⁰ Importantly, BLACAT1 elevates the oxaliplatin-resistance of GC by enhancing ABCB1 protein expression through inhibiting miR-361.¹¹ However, the detailed mechanism of BLACAT1 on the occurrence and development of GC needs to be further explored. Some microRNAs (miRNAs) have anti-cancer effect on GC. For example, miR-449 functions as anti-onco-miR in human GC.¹² miR-126 retards cell growth and is down-regulated in GC.¹³ Researchers have indicated that miR-149 is associated with the processes of GC. Wang et al have proved that miR-149 attenuates cell-cycle progression and proliferation in GC.¹⁴ Zhang et al have showed that miR-149-5p reduces the proliferation rate of GC cells.¹⁵ Notably, lncRNA GACAT1 increases GC cell growth, invasion and migration via inhibiting miR-149.¹⁶ However, it is still unknown whether the suppressive effect of miR-149 on the development of GC is regulated by BLACAT1.

Kinesin superfamily proteins 2A (KIF2A) belongs to KIFs, and serves as a role of microtubule stabilizer.¹⁷ KIF2A takes part in the progression of several malignancies, including ovarian cancer,¹⁸ glioma¹⁹ and colorectal cancer.²⁰ Additionally, up-regulation of KIF2A is associated with poor prognoses and clinicopathological features in GC patients.²¹ Nevertheless, the potential regulatory mechanism of BLACAT1 related to KIF2A in GC remains undefined.

Herein, we determined the expression and functional role of BLACAT1 in GC. The regulatory relationship between BLACAT1 and miR-149-5p was further evaluated. Then, we explored whether KIF2A was a target of miR-149-5p. A BLACAT1/miR-149-5p/KIF2A cascade in GC pathogenesis was identified. Our research may uncover a novel therapeutic target for GC.

Materials and Methods

Tissue Samples

In total 52 GC patients (26 males and 26 females) were collected from our hospital from January 2017 to January 2019. Patients had not received preoperative systemic or local treatment before this study. Tumor tissues and adjacent tissues were collected. We separated our patient cohort according to the median BLACAT1 expression into high BLACAT1 (n = 26) and low BLACAT1 groups (n = 26). This study was permitted by our hospital ethics committee, and the informed consents were obtained from each patient.

Cell Culture

Four human GC cell lines (AGS, MNK-45, MGC803 and SNU-5) and human normal gastric epithelial cell line GES-1 were obtained from the American Type Culture Collection (Manassas, VA, USA). Cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM, Invitrogen, Carlsbad, CA, USA) with 10% fetal bovine serum (FBS, Invitrogen) at 37°C.

Cell Transfection

The siRNA-negative control (si-NC), si-BLACAT1-1, si-BLACAT1-2, miR-NC, miR-149-5p mimics, miR-149-5p inhibitor, pcDNA3.1-NC (pcDNA-NC) and pcDNA3.1-KIF2A (pcDNA-KIF2A) were synthesized by Shanghai Genepharma (Shanghai, China). AGS and MKN45 cells grown to 80% confluence were transfected or co-transfected with the above agents using Lipofectamine 3000 reagent (Invitrogen) according to the manufacturer’s protocol.

Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR)

Total RNA was extracted from tissues and cells using the TRIzol reagent (Invitrogen) and was reverse-transcribed into cDNA by PrimeScript RT reagent kit (Takara, Otsu, Japan). PCR reaction was performed on ABI 7500HT Fast Real-Time PCR System (Applied Biosystems, Waltham, MA, USA) with the following conditions: 95°C for 5 min, 40 cycles of 72°C for 34 s and 60°C for 20 s. Relative expression was calculated by the 2^−ΔΔCt method. GAPDH, U6 or β-actin were used for the normalization of BLACAT1, miR-149-5p and KIF2A, respectively. The primer sequences are shown in Table 1.

Table 1 Primer Sequences

Western Blot

AGS and MKN45 cells were lysed by ice-cold lysis buffer to obtain total proteins. Protein samples were size-fractionated in SDS-PAGE, and transferred to polyvinylidene fluoride membranes. Then, membranes were blocked with 5% skimmed milk and incubated overnight at 4°C with primary antibody anti-KIF2A (1:1000, SAB2101262MSDS, Sigma, St. Louis, MO, USA). Afterwards, the membranes were probed with HRP conjugate secondary antibody (1:5000, 12–348MSDS, Sigma) for 1 h at 25°C. The protein blots were visualized by ECL chemiluminescent substrate reagent kit (Invitrogen), and quantified by ImageLab software (Bio-Rad, Hercules, CA, USA). β-actin (1:5000, SAB2701711MSDS, Sigma) was introduced as the internal reference.

MTT Assay

AGS and MKN45 cells were seeded into 96-well plates (2 × 10³ cells/well) and cultured with 5% CO₂ at 37°C. Cell proliferation was measured using the MTT cell proliferation assay kit (Sigma) according to the guidelines.

Wound-Healing Assay

AGS and MKN45 cells were seeded into a 6-well plate and cultured until 80% confluence. A sterile pipette tip was used to draw a vertical line on the cell surface. Cells were then cultured in DMEM without FBS for 24 h. The scratch width was measured at 0 and 24 h post-incubation, respectively. Cells were observed under a light microscope and the wound-healing rate was calculated as: (1–scratch width at 24 h/scratch width at 0 h) × 100%.

Transwell Assay

The transwell assay was used to measure the cell invasion by using transwell chambers (8 nm pore size) (Corning Inc., Corning, NY, USA). AGS and MKN45 cells (2 × 10⁵) in serum-free medium were placed to upper chambers pre-coated with matrigel (Sigma). Medium with 10% FBS was added to the lower chambers. Cells remaining in the upper chamber were wiped with cotton swab after 48 h of incubation at 37°C, while cells in lower chambers were fixed in methanol, and stained with 0.1% crystal violet. Five random fields were selected in each well for counting under an inverted microscope (Olympus, Tokyo, Japan).

Tumor Xenograft in Nude Mice

Female BALB/c nude mice (weighing 20 ± 2g) were purchased from Cavens Lab, Ltd (Changzhou, China) and fed in sterile environment at 25°C. Si-BLACAT1 or si-NC was integrated into lentiviral vector and then transfected into AGS cells. After that, the transfected AGS cells (2 × 10⁶ cells/100 ul, s.c.) were injected into the right flanks of the nude mice (n = 5). Tumor volumes were measured every other week. Four weeks later, mice were anesthetized with pentobarbital sodium (50 mg/kg) and then sacrificed by cervical dislocation. The tumor xenograft was separated from mice and weighted. The study was performed in the Animal Experimental Center of our hospital and obtained the approval of ethical committee in our hospital.

Dual-Luciferase Reporter Assay

The potential binding sites of miR-149-5p and BLACAT1 or miR-149-5p and KIF2A were predicted by Starbase or TargetScan, respectively. BLACAT1 with WT or MUT miR-149-5p-binding sites were generated and fused to the psiCHECK-2 vector (Promega, Madison, WI, USA). The 3ʹ-UTR of KIF2A containing the predicted miR-149-5p targeting site (WT), and MUT KIF2A were amplified and cloned into the psiCHECK-2 vector. Then, BLACAT1-Wt and BLACAT1-Mut vectors were co-transfected with miR-NC or miR-149-5p mimics into AGS and MKN45 cells. KIF2A-Wt and KIF2A-Mut vectors were co-transfected with miR-NC + pcDNA-NC, miR-149-5p mimics + pcDNA-NC or miR-149-5p mimics + pcDNA-BLACAT1 into AGS and MKN45 cells. Cell transfection was performed using Lipofectamine 3000 following the guidelines.

Statistical Analysis

Statistical analysis was performed on GraphPad Prism 7.0 (GraphPad Software, Inc., La Jolla, CA, USA). Data were displayed as mean ± standard deviation (SD). The differences between two groups or among multiple groups were assessed by Student’s t-test or one-way ANOVA followed by Tukey’s post hoc test. The correlation significance was determined by Pearson correlation analysis. A P value <0.05 was considered statistically significant.

Results

The Expression of BLACAT1 Was Increased in GC Tissues and Cells

Through TCGA database, BLACAT1 expression in stomach adenocarcinoma (STAD) and normal tissues was analyzed. The results showed that BLACAT1 expression was increased in STAD tissues in contrast to that in normal tissues (P < 0.05) (Figure 1A). Then we measured BLACAT1 expression in tissues of GC patients by qRT-PCR. The BLACAT1 expression in tumor tissues was obviously increased by contrast to the adjacent tissues in GC patients (P < 0.001) (Figure 1B). And its level in tissues in WHO grade III/IV was visibly higher than that in WHO grade I/II (P < 0.001) (Figure 1C). The BLACAT1 expression was visibly enhanced in AGS, MKN45, MGC803 and SNU-5 cells by contrast to the GES-1 cells (P < 0.01) (Figure 1D). AGS and MKN45 cells were used for the subsequent assays. As shown in Table 2, BLACAT1 expression had no connections with age, gender, diameter and resection degree in GC patients, while the BLACAT1 expression was positively correlated with WHO grade (P < 0.01) and TNM stage (P < 0.05).

Table 2 Correlation Between BLACAT1 Expression and Clinicopathological Features in Gastric Cancer (GC)

Figure 1 The expression of BLACAT1 was increased in gastric cancer (GC). (A) The expression of BLACAT1 in STAD tissues and normal tissues was analyzed in TCGA. *P < 0.05. (B) The expression of BLACAT1 in adjacent tissues and tumor tissues of GC patients was detected by qRT-PCR. P < 0.001 vs Adjacent tissues. (C) The expression of BLACAT1 in GC patients at the TNM I/II and TNM III/IV was detected by qRT-PCR. P < 0.001 vs I/II. (D) The expression of BLACAT1 in human normal gastric epithelial cell line GES-1 and GC cell lines (AGS, MKN45, MGC803 and SNU-5) was detected by qRT-PCR. **P < 0.01 vs GES-1.

Silencing of BLACAT1 Inhibited the Proliferation, Migration and Invasion of GC Cells in vitro and the Growth of Tumor Xenograft in vivo

To evaluate the effect of BLACAT1 on tumorigenesis of GC, BLACAT1 was silenced by the transfection of si-BLACAT1-1 and −2 (P < 0.01) (Figure 2A). Then, si-BLACAT1-1 was used for the subsequent assays. MTT assay discovered that si-BLACAT1-1 significantly decreased the proliferation of AGS and MKN45 cells at 96 h post-culturing (P < 0.01) (Figure 2B). Wound-healing assay and transwell assay disclosed that the wound-healing rate and number of invasion GC cells were markedly reduced in the si-BLACAT1-1 group compared with the si-NC group (P < 0.01) (Figure 2C and D). Additionally, we also investigated the effect of BLACAT1 knockdown on the growth of tumor xenograft in vivo. Four weeks later, the results demonstrated that in comparison to the si-NC group, tumor volume (P < 0.05) and tumor weight (P < 0.01) were both distinctly decreased in the si- BLACAT-1 group (Figure 2E).

Figure 2 Silencing of BLACAT1 inhibited proliferation, migration and invasion of gastric cancer (GC) cells. (A) The expression of BLACAT1 in AGS and MKN45 cells was detected by qRT-PCR. (B) The proliferation of AGS and MKN45 cells was detected by MTT assay. (C) The wound-healing rate of AGS and MKN45 cells was detected by wound-healing assay. (D) The invasion rate of AGS and MKN45 cells were detected by transwell assay. **P < 0.01 vs si-NC. (E) The tumor volume was monitored at different time points and tumor weight was measured at 4 weeks later. *P < 0.05; **P < 0.01 vs sh-NC.

BLACAT1 Was Negatively Correlated with miR-149-5p

Starbase was used to predict a binding site of miR-149-5p on the 3ʹ UTR of BLACAT1 (Figure 3A). Silencing of BLACAT1 markedly enhanced the miR-149-5p expression in AGS and MKN45 cells (P < 0.01) (Figure 3B). Next, the relative luciferase activity was obviously declined in AGS and MKN45 cells co-transfected with miR-149-5p mimics and BLACAT1 Wt by contrast to the cells co-transfected with miR-NC and BLACAT1 Wt (P < 0.01, Figure 3C). Moreover, miR-149-5p expression in tumor tissues was notably decreased by contrast to the adjacent normal tissues (P < 0.001) (Figure 3D). There was a negative correlation between BLACAT1 and miR-149-5p expression in GC tissues (N = 52, r = −0.5319, P < 0.001) (Figure 3E). Furthermore, miR-149-5p expression was obviously lower in AGS, MKN45, MGC803 and SNU-5 cells than that in GES-1 cells (P < 0.01) (Figure 3F).

Figure 3 BLACAT1 was negatively correlated with miR-149-5p. (A) Starbase showed the predicted binding site between BLACAT1 and miR-149-5p. (B) The expression of miR-149-5p in AGS and MKN45 cells was detected by qRT-PCR. **P < 0.01 vs si-NC. (C) Relative luciferase activity in AGS and MKN45 cells was measured by dual-luciferase reporter assay. **P < 0.01 vs miR-NC. (D) The expression of miR-149-5p in adjacent tissues and tumor tissues in GC patients was detected by qRT-PCR. (E) The expression of BLACAT1 was negatively correlated with miR-149-5p. (F) The expression of miR-149-5p in GES-1, AGS, MKN45, MGC803 and SNU-5 cells was detected by qRT-PCR. **P < 0.01 vs GES-1.

miR-149-5p Inhibited Proliferation, Migration and Invasion of GC Cells

To determine the biological function of miR-149-5p in GC, miR-149-5p was overexpressed by the transfection of miR-149-5p mimics, and inhibited by the transfection of miR-149-5p inhibitor (P < 0.01) (Figure 4A). miR-149-5p up-regulation markedly attenuated the proliferation of AGS and MKN45 cells at 96 h post-culturing (P < 0.01) (Figure 4B). Additionally, the wound-healing rate and number of invasion GC cells were lower in the miR-149-5p mimics group than those in the miR-NC group (P < 0.01) (Figure 4C and D).

Figure 4 miR-149-5p inhibited proliferation, migration and invasion of gastric cancer (GC) cells. (A) The expression of miR-149-5p in AGS and MKN45 cells was detected by qRT-PCR. **P < 0.01 vs miR-NC. (B) The proliferation of AGS and MKN45 cells was detected by MTT assay. (C) The wound-healing rate of AGS and MKN45 cells was detected by wound-healing assay. (D) The invasion rate of AGS and MKN45 cells was detected by transwell assay. **P < 0.01 vs miR-NC.

BLACAT1 Positively Regulated KIF2A Expression via Binding miR-149-5p

TargetScan predicted a binding site for miR-149-5p on the 3ʹ UTR of KIF2A (Figure 5A). The relative luciferase activity was visibly decreased in AGS and MKN45 cells co-transfected with miR-149-5p mimics + pcDNA-NC and KIF2A Wt compared with the cells co-transfected with miR-NC + pcDNA-NC and KIF2A Wt (P < 0.01). Overexpression of BLACAT1 reversed the reducing effect of miR-149-5p mimics on the relative luciferase activity (P < 0.05) (Figure 5B). Furthermore, the KIF2A expression in tumor tissues was markedly increased by contrast to the adjacent tissues in GC patients (P < 0.001) (Figure 5C). The expression of KIF2A and miR-149-5p was negatively correlated in GC tissues (N = 52, r = −0.6011, P < 0.001) (Figure 5D), but the expression of KIF2A and BLACAT1 was positively correlated (N = 52, r = 0.3872, P = 0.0046) (Figure 5E). Additionally, the KIF2A expression was obviously higher in AGS, MKN45, MGC803 and SNU-5 cells than that in GES-1 cells (P < 0.01) (Figure 5F). Western blot assay demonstrated that the protein level of KIF2A in GC cells (SNU-5, MGC803, AGS and MKN45) was distinctly elevated in comparison to that in GES-1 cells (P < 0.01) (Figure 5G).

Figure 5 BLACAT1 positively regulated KIF2A expression via binding miR-149-5p. (A) TargetScan showed the predicted binding site between miR-149-5p and KIF2A. (B) Relative luciferase activity in AGS and MKN45 cells was measured by dual-luciferase reporter assay. *P < 0.05, **P < 0.01 vs miR-NC + pcDNA-NC, #P < 0.05 vs miR-149-5p mimics + pcDNA-NC. (C) The expression of KIF2A in adjacent tissues and tumor tissues in GC patients was detected by qRT-PCR. P < 0.001 vs Adjacent tissues. (D) The expression of KIF2A was negatively correlated with miR-149-5p. (E) The expression of KIF2A was positively correlated with BLACAT1. (F) The expression of KIF2A in GES-1, AGS, MKN45, MGC803 and SNU-5 cells was detected by qRT-PCR. **P < 0.01 vs GES-1. (G) The protein level of KIF2A in GES-1, AGS, MKN45, MGC803 and SNU-5 cells was detected by Western blot. **P < 0.01 vs GES-1.

Knockdown of BLACAT1 Attenuated the Proliferation, Migration and Invasion of GC Cells via Targeting miR-149-5p/KIF2A Axis

Western blot assay revealed that in AGS cells, miR-149-5p overexpression down-regulated the protein expression of KIF2A and BLACAT1 overexpression up-regulated the protein expression of KIF2A (P < 0.01). At the same time, BLACAT1 reversed the suppressive effect of miR-149-5p on KIF2A protein expression (P < 0.05) (Figure 6A). The proliferation, wound-healing rate and invasion rate of AGS cells were obviously declined in the si-BLACAT1-1 group by contrast to the si-NC group (P < 0.01). Overexpression of KIF2A or inhibition of miR-149-5p strikingly reversed the reduction effect of si-BLACAT1-1 on the proliferation, migration and invasion of AGS cells (P < 0.05) (Figure 6B-D).

Figure 6 Knockdown of BLACAT1 attenuated the proliferation, migration and invasion of gastric cancer (GC) cells via targeting miR-149-5p/KIF2A axis. (A) The protein expression of KIF2A in AGS cells was measured by Western blot. *P < 0.05, **P < 0.01 vs miR-NC + pcDNA-NC, #P < 0.05 vs miR-149-5p mimics + pcDNA-NC. (B) The proliferation of AGS cells was detected by MTT assay. (C) The wound-healing rate of AGS cells was detected by wound-healing assay. (D) The invasion rate of AGS was detected by transwell assay. *P < 0.05, **P < 0.01 vs si-NC, #P < 0.05 vs si-BLACAT1-1.

Discussion

Dysregulation of lncRNAs involves in the tumorigenesis of GC.^22,23 LncRNAs such as lncRNA HOX11-AS,²⁴ lncRNA PVT1²⁵ and lncRNA Sox2ot²⁶ are all up-regulated in patients with GC. Here, BLACAT1 expression was enhanced in GC tissues and cells, and was correlated with TNM stage in GC patients. The function of BLACAT1 was similar to some previously reported lncRNAs. Yao et al have pointed out that lncRNA CASC15 expression is related to the TNM stage in GC patients.²⁷ Peng et al have proved that high-expressed lncRNA EGOT is associated with metastasis and TNM stage in GC patients.²⁸ In addition, we also demonstrated that BLACAT1 expression was correlated with WHO grade. Therefore, our data implied that BLACAT1 might act as a pathogenic-factor role in GC.

Prior researches have displayed that BLACAT1 takes part in the biological processes of various cancers.²⁹ For instance, BLACAT1 accelerates the cell proliferation and invasion in glioma.³⁰ BLACAT1 contributes to cell growth and inhibits cell cycle arrest in ovarian cancer.³¹ In the present study, we found that BLACAT1 knockdown suppressed the proliferation, migration and invasion of GC cells in vitro. Consistent with our results, a previous study has revealed that BLACAT1 knockdown attenuates the proliferation, migration and invasion of GC cells in vitro.³⁰ However, the prior study only investigated the inhibitory effect of BLACAT1 knockdown on GC cells in vitro. Our results further indicated that BLACAT1 knockdown also inhibited the growth of tumor xenograft in vivo. The above results suggested that BLACAT1 knockdown retarded the progression of GC.

Numerous studies have proved that miR-149-5p serves as anti-onco-miR in multiple cancers. Jin et al have shown that the expression of miR-149-5p is downregulated in renal cell carcinoma and overexpression of miR-149-5p has an anti-tumor effect on the progression of renal cell carcinoma.³² Chen et al have reported that miR-149-5p expression is distinctly reduced in melanoma and miR-149-5p up-regulation can hinder the cell proliferation of melanoma.³³ Notably, miR-149 expression is markedly decreased in GC tissues and overexpression of miR-149 diminishes the viability of GC cells.³⁴ In the current study, miR-149-5p expression was declined in GC tissues and cells, and miR-149-5p overexpression attenuated the proliferation, migration and invasion of GC cells. The results implied that miR-149-5p might serve as a suppressor in GC. LncRNAs can function as ceRNAs or molecular sponges in modulating miRNAs in the progression of GC.^35,36 BLACAT1 has been reported to act as an onco-lncRNA by targeting miRNAs. For example, BLACAT1 interacts with miR-485-5p to accelerate the tumorigenesis of hepatocellular carcinoma.^37,38 BLACAT1 contributes to the glioma development through binding to miR-605-3p.³⁹ BLACAT1 enhances the cell survival and metastasis of breast cancer by miR-150-5p/CCR2 axis.⁴⁰ We suspected that BLACAT1 mediated the progression of GC by similar mechanisms above. In this study, miR-149-5p was proved to be the target of BLACAT1. At the same time, down-regulation of miR-149-5p reversed the inhibitory effect of BLACAT1 knockdown on GC cells. The above results further indicated that BLACAT1 might promote tumorigenesis of GC by modulating miR-149-5p.

KIF2A has been reported to be taken part in the occurrence and development of GC.^41–43 KIF2A acts as a molecular target for the therapeutics of squamous cell carcinoma of the oral tongue.⁴¹ Knockdown of KIF2A reduces the cell viability of diffuse large B cell lymphoma.⁴² Importantly, KIF2A is increased in GC and KIF2A silencing can retard the proliferation and invasion of GC cells.⁴³ In this study, KIF2A expression was increased in GC tumor, which is consistent with the previous studies.^43–45 The results suggested that KIF2A might be a pathogenic factor in GC. KIF2A usually exerts its role in regulating the tumorigenesis as a target of miRNAs, such as miR-206a-KIF2A in ovarian cancer and miR-451a-KIF2A in lung squamous cell carcinoma.^18,46 In the current study, KIF2A was a target of miR-149-5p and was negatively regulated by miR-149-5p. We suspected that miR-149-5p was involved in the development of GC by targeting KIF2A. Additionally, the KIF2A expression was positively related to BLACAT1 in GC. We further assumed that BLACAT1 might serve as miR-149-5p sponge to inhibit KIF2A expression and then elevate KIF2A expression in GC. To further validate these conjectures, the feedback verification experiment showed that KIF2A up-regulation and miR-149-5p down-regulation both reversed the anti-tumor effect of BLACAT1 silencing on GC cells. To sum up, we indicated that BLACAT1 interacted with miR-149-5p to promote the tumorigenesis of GC via targeting KIF2A.

Conclusions

In conclusion, BLACAT1 expression was up-regulated in GC tissues and cells. miR-149-5p was targeted by BLACAT1 and KIF2A was targeted by miR-149-5p. BLACAT1 up-regulated KIF2A expression by interacting with miR-149-5p, and then promoted the proliferation, migration and invasion of GC cells. Thus, BLACAT1 may be a promising therapeutic target for GC.