Downregulation of long noncoding RNA breast cancer anti-estrogen resistance 4 inhibits cell proliferation, invasion, and migration in esophageal squamous cell carcinoma by regulating the microRNA-181c-5p/LIM and SH3 protein 1 axis.

Recently, abnormal expression of long non-coding RNAs (lncRNAs) has been observed in esophageal squamous cell carcinoma (ESCC). In various human cancers, breast cancer anti‑estrogen resistance 4 (BCAR4) was reported to be highly expressed, while the biological roles of BCAR4 in ESCC remain unclear. In ESCC cells and tissues, BCAR4 and microRNA -181c-5p (miR-181c-5p) expression, and phosphorylated signal transducer and activator of transcription (p-STAT3) and COX2 expression were evaluated by real-time reverse transcription PCR (qRT-PCR) and Western blot analysis. Cell function was evaluated by colony formation, CCK-8 assay, transwell and flow cytometer assays. Interactions between BCAR4 and miR-181c-5p, as well as miR-181c-5p and LIM and SH3 protein 1 (LASP1) were evaluated by RIP and luciferase reporter assay. ESCC cell malignancy with inhibition of BCAR4 was confirmed by a tumor xenograft model in vivo. In both ESCC tissues and cell lines, BCAR4 was upregulated. Downregulation of BCAR4 effectively induced cell apoptosis and inhibited invasion and migration in vitro, and reduced tumorigenesis in nude mice. BCAR4 was a sponge of miR-181c-5p to upregulate LASP1. Moreover, knockdown of BCAR4 and overexpression of miR-181c-5p inhibited the activation of the STAT3/COX2 signaling, which was reversed by overexpression of LASP1. In conclusion, BCAR4 promotes ESCC tumorigenesis by targeting the miR-181c-5p/LASP1 axis, which may act as a treatment and diagnosis biomarker for ESCC.

Highlights

BCAR4 was upregulated in LSCC and acted as a sponge of miR-181c-5p.

Knockdown of BCAR4 promoted miR-181c-5p to upregulate LASP1.

si-BCAR4 inhibited invasion and migration via the miR-181c-5p/LASP1 axis.

si-BCAR4 decreased levels of p-STST3 and COX2 via the miR-181c-5p/LASP1 axis.

The miR-181c-5p/LASP1 axis mediated the function of BCAR4 in tumor growth in vivo.

Background

As the common malignance, esophageal squamous cell carcinoma (ESCC) has over 90% patient mortality [1]. In China, ESCC even ranks the fourth leading cause of cancer-related deaths due to the lack of actionable targets [2,3]. The 5-year survival rate of ESCC patients remains low at only 15–25% because of local invasion and distant metastasis despite of intensive clinical efforts by combining multiple therapeutic approaches [4]. Therefore, it is urgent to understand the molecular mechanisms underlying ESCC tumor progression.

Long non-coding RNAs (lncRNAs) are known to lack protein-coding function [5]. However, aberrant expression of lncRNAs during tumor development was reported to participate in poor prognosis of patients and aggressive tumor behavior [6,7]. In ESCC, a large number of lncRNAs, including MALAT, ZFAS1, CASC9, and EZR-AS1, are upregulated in tumors and promote cancer progression by regulating invasion, apoptosis, migration, EMT, and metastasis [8-11]. BCAR4 was firstly identified in breast cancer and reported to be involved in the progression of different types of cancer [12]. For example, in colon cancer, BCAR4 promotes cell migration and proliferation [13]. Similarly, BCAR4 is upregulated in colorectal cancer cells and overexpression of BCAR4 facilitates ALDH positive cell stemness maintenance, ALDH+ cell proliferation, and migration [14]. These reports suggest that BCAR4 functions as an oncogene to participate in various human cancer tumorigenesis. However, how BCAR4 promotes carcinogenesis in ESCC remains elusive.

In many types of human cancer, LIM and SH3 protein 1 (LASP1) is always highly expressed and exerts essential roles in cancer development [15]. Increasing evidence has demonstrated that LASP1 enhances breast cancer cell proliferation and tumorigenesis, and high LASP1 expression is markedly associated with the poor prognosis [16] and gastric cancer patients [17]. In ESCC, overexpression of LASP1 is closely involved in tumor cell invasion, proliferation, and migration [18]. Recently, LASP1 was identified to participate in tumor progression by targeting microRNAs (miRNAs). For example, LASP1 promotes the glioma cell migration and proliferation by regulating miR-377-3p [19]. MiR-133b suppresses proliferation, cellular migration, and invasion through targeting and downregulating LASP1 in hepatocarcinoma cells [20]. These reports suggest that LASP1 may be an important target gene of miRNAs to regulate biological behaviors of ESCC cells. However, the role of LASP1 in ESCC remains unknown. In human cancers, lncRNAs exert their functions by targeting miRNAs [20]. Here, lncRNA BCAR4 was found to have a potential binding site for miR-181c-5p. Interestingly, miR-181c-5p was firstly found to target LASP1. Based on these findings, we hypothesized that lncRNA BCAR4 might regulate LASP1 by sponging miR-181c-5p, indicating their involvement in ESCC development and progression. This study was to explore the function of lncRNA BCAR4 and its regulatory network in ESCC, and to find potential new drug targets.

Methods

Tissues samples

A total of 40 ESCC patients who were admitted at Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology in 2019 were enrolled in this study. They all signed the written informed consent. Prior to surgery, none of them received chemotherapy or radiotherapy. By surgical excision, ESCC tissues and matched adjacent normal tissues were collected. The human Ethics Committee of this hospital approved all experimental procedures (Supplemental file 1).

Cell culture

ESCC lines (TE-1, EC8712, Ec-9706, Eca-109 and KYSE-150) and HEEC line were purchased from American Type Culture Collection (Manassas). Cells were cultured in DMEM medium with 10% FBS with 5% CO2 at 37°C.

Cell transfection

SiRNA targeting BCAR4 (si-BCAR4), miR-181c-5p inhibitor and mimics and their controls (inhibitor NC and miR-NC) were obtained from GenePharma. Primer sequences of si-BCAR4 were: 5’-CCAAGUGUUGACUCAACAATT-3’, 5’-UUGGUUAGUCCACACUUGGTT-3’. To overexpress BCAR4 or LASP1, the full length of BCAR4 or LASP1 was synthesized and cloned into pcDNA3.1 vector to generate pc-BCAR4 and pcDNA-LASP1, respectively. As a negative control (pc-NC), pcDNA3.1 empty vector was used. The transfection of Ec-9706 and TE-1 (5 × 104 cells/well) cells using Lipofectamine 2000 reagent.

qRT-PCR

Total RNAs were extracted from tissue samples or cultured cells. Approximately 1.2 μg RNA sample was reversely transcribed into cDNA. Then qRT-PCR was performed on the Roche LightCycler 480 System using SYBR® Premix Ex Taq™. GAPDH and U6 were the internal reference for BCAR4 and miR-181c-5p, respectively. Gene expression was calculated using the 2−ΔΔCt method [21]. PCR primers were:

BCAR4 forward: 5′-CCTTATTACTTGGCCCGGA-3′, reverse: 5′- CCGTTAGGATTCCCAGG-3′; miR-181c-5p forward: 5′- CTTGATGGTGAGAGGATGT-3′, reverse: 5′-GTTCTTCGACCATTCCGCGG-3′; LASP1 forward: 5′-CCACCAGAATCCCGGAACT-3′, reverse: 5′-GAAGTTCTTACAAGGTTGCC-3′; GAPDH forward: 5′-CGAGAGAATCCGCGGACAT-3′, reverse: 5′-TTGTGCAATACAGCGTGGAC-3′; U6 forward: 5′-GACAGATTCGGTCTGTGGCAC-3′, reverse: 5′-GATTACCCGTCGGCCATCGATC-3′.

Western blot analysis

From tissue samples or cultured cells, total proteins were extracted using RIPA Lysis Buffer (Thermo Scientific, USA) [22]. A total of 20 ul proteins were separated and transferred onto PVDF membranes. Then, primary antibody (Amyjet, Wuhan, China) including anti-LASP1, anti-p-STAT3, anti-COX-2 was added at 1:1,000 dilatation and incubated at 4°C overnight, with GAPDH (1:10,000, Invitrogen, USA) as the internal reference. HRP-conjugated secondary antibody was then added the next day and incubated for 2 h. The ECL detection kit was used to detect protein signals.

Cell proliferation

CCK-8 assay was conducted as previously described [23]. Briefly, cells were cultured for different time periods including 24, 48, 72 and 96 h. In each cell well, CCK-8 solution (10 μl) was added at 37°C and incubated for 2 h. The optical density at 450 nm was detected using a microplate reader. Approximately 1,000 cells were cultured for colony formation assay. Next, 1% crystal violet dye was used to stain cells after they were fixed with 10% formalin for 30 min. Finally, colony numbers were counted using a light microscope.

Cell migration and invasion assays

Transwell assay was conducted as previously described [24]. In brief, in the 8.0-μm-pore Matrigel™-coated membranes (for invasion) or upper uncoated (for migration) chamber, approximately 1 × 105 cells were added with serum-free medium. The lower wells were filled for 24 h with DMEM medium containing 20% FBS, cells that had invaded to the lower wells were gently swabbed, fixed and stained. Subsequently, cells were counted in five randomly selected fields under a microscope.

Apoptosis analysis

The apoptosis detection kit (Maokang, Shanghai, China) was used to detect cell apoptosis following the manufacturer’s instructions. Briefly, 1 × Annexin V binding buffer (195 μL) was used to suspend the cells, followed by centrifugation of cells at 1,500 rpm for 5 min. Then 10 μL PI and Annexin-V/FITC (5 μL) were added for incubation in the dark for 15 min. Finally, FACScalibur cytometer was used to detect cell apoptosis.

Luciferase reporter assay

Luciferase reporter assay as performed as previously described [22]. BCAR4-WT, BCAR4-MUT, LASP1-WT and LASP1-MUT, the mutant (MUT) or wild type (WT) of BCAR4 or LASP1 3′-UTR fragments containing the putative miR-181c-5p binding sites were synthesized and cloned into the pmirGLO dual-luciferase plasmid. Cells were co-transfected with miR-NC or miR-181c-5p mimics and different luciferase reporter vectors using Lipofectamine 2000 reagent. Cells were collected and lysed and analyzed by a dual luciferase reporter assay system.

RIP assay

RIP assay kit was used as previously described [25]. In brief, cells were lyzed and incubated for 5 min on ice. Next, Ago2 antibodies (5 μg) or corresponding immunoglobulin G (IgG) (Abcam, USA) was incubated with 50 μl magnetic beads for 30 min. Next, each tube with 100 μl lysate was incubated at 4°C overnight. Using qRT-PCR to purify the enriched RNA.

Construction of stably transfected Ec-9706 cells

Lentiviral vectors carrying the knockdown of BCAR4 (sh-BCAR4) or control (sh-NC) were obtained from GenePharm (Shanghai, China). sh-BCAR4 sequence was as follows: 5’-AATGGGAGCTGTGTCCCATTA-3’. In the presence of polybrene (6 µg/ml), the recombinant lentivirus-transducing units were transfected into EC109 cells as previously described [26]. And to screen stably transfected Ec-9706 cells, 2.5 µg/ml puromycin was added for 2 weeks. To generate Ec-9706 cells stably transfected with sh-BCAR4 and miR-181c-5p inhibitor, sh-BCAR4 and miR-181c-5p inhibitor were co-transfected into EC109 cells using Lipofectamine 2000 reagent (Invitrogen), and puromycin was applied to screen stably transfected cells.

Xenograft model construction

Athymic male BALB/c nude mice were obtained from Tongji Medical College. following the protocols as previously reported [27], 5 × 106 Ec-9706 cells were inoculated subcutaneously into the right side of the back of nude mice to produce 4 groups of mice (sh-NC, sh-BCAR4, sh-BCAR4 + miR-181c-5p, and miR-181c-5p inhibitor) (n = 5). Tumor volume was evaluated according to the formula (length × width2)/2 every week for a total of 5 weeks. The xenograft tumors of different groups were isolated and weighed after 5 weeks.

Immunohistochemistry assay

Immunohistochemistry assay was conducted as previously described [28]. Briefly, after weighing, xenograft tumors were cut into 3 µm thick slices. With human Ki-67 antibody (1:200, Abcam), slices were incubated overnight at 4°C, Next, secondary antibody was added for 2 h. All images were acquired using Leica TCS SP8X confocal microscope (magnification ×100 or ×400).

Statistical analysis

Data were shown as means ± standard deviation (SD). All data were analyzed by SPSS 20.0 software. One-way or two-way ANOVA was used to compare differences among multi-groups, while student’s t test was used to compare the differences between two groups. P < 0.05 was considered as the significant threshold.

Results

The expression of BCAR4 in ESCC patients

To explore the function of BCAR4 in ESCC, its expression in ESCC tissues (n = 40) was detected. Compared with adjacent normal tissues, BCAR4 was markedly upregulated in ESCC tissues (p < 0.01, Figure 1(a)). As expected, the expression of BCAR4 was also markedly regulated in TE-1, KYSE-150, Ec-9706, Eca-109 and EC8712 cells (p < 0.01, p < 0.001, Figure 1(b)). The highest expression level of BCAR4 was observed in Ec-9706 and TE-1 cells, which were selected in the subsequent experiments. Moreover, ESCC patients with higher expression levels of BCAR4 had a poor prognosis (p < 0.01, Figure 1(c)). These results indicated that BCAR4 might be involved in ESCC.

Figure 1.

The potential oncogenic role of BCAR4 in ESCC. (a) BCAR4 expression in ESCC tissues was detected (n = 40). (b) BCAR4 expression in ESCC and normal cell lines was detected. (c) The overall survival of ESCC patients was evaluated. ** p < 0.01, *** p < 0.001.

Downregulation of BCAR4 inhibited ESCC cell growth

To determine the oncogenic role of BCAR4 in ESCC,, si-BCAR4 (knockdown of BCAR4) was designed and transfected into TE-1 and Ec-9706 cells. si-BCAR4 significantly reduced the expression levels of BCAR4 in TE-1 and Ec-9706 cells (p < 0.001, Figure 2(a)). In addition, si-BCAR4 reduced Ec-9706 and TE-1 cell proliferation (p < 0.01, Figure 2(b)) and colony formation. Simultaneously, knockdown of BCAR4 significantly inhibited the invasion (p < 0.01, Figure 2(d)) and migration capacity (p < 0.01, Figure 2(e)). Furthermore, knockdown of BCAR4 significantly enhanced apoptosis of the two cell lines (p < 0.01, Figure 2(f)). Taken together, downregulation of BCAR4 efficiently induced ESCC cell apoptosis and inhibited ESCC cell proliferation.

Figure 2.

BCAR4 in ESCC cell growth in vitro. (a) The transfection efficiency of si-BCAR4 in Ec-9706 and TE-1 cells was detected by qRT-PCR. (b) CCK-8 assay was performed to detect cell proliferation. (c) Colony formation assay was performed. (d and e) Transwell assay was performed to detect invasion and migration. (f)Flow cytometry was performed to detect cell apoptosis. ** p < 0.01, *** p < 0.001.

BCAR4 acted as a sponge for miR-181c-5p

Potential base pairs formed between BCAR4 and miR-181c-5p were predicted by the miRanda database. We found that miR-181c-5p might be the potential target (Figure 3(a) and Supplementary data 1). We observed downregulation of miR-181c-5p in ESCC cell lines and tissues (p < 0.01, Figure 3(b,c)). Moreover, Ec-9706 and TE-1 cells transfected with miR-181c-5p mimics showed increased expression levels of miR-181c-5p (p < 0.01, Figure 3(d)). Furthermore, in the two cell lines, overexpression of miR-181c-5p decreased the BCAR4-WT vector mediated luciferase activity (p < 0.01), while not BCAR4-MUT vector mediated luciferase activity (Figure 3(e)). RIP assay also showed that, compared with anti-IgG immune-precipitates, BCAR4 and miR-181c-5p were enriched preferentially in miRNPs containing Ago2 (p < 0.01, Figure 3(f)). Downregulation of BCAR4 increased the expression levels of mIR-181c-5p (p < 0.01, Figure 3(g)), while overexpression of BCAR4 decreased the expression levels of miR-181c-5p (p < 0.01, Figure 3(h)). Finally, a strong negative correlation between the expression of BCAR4 and miR-181c-5p was revealed in ESCC tissues (n = 40, r = −0.062, p < 0.0001, Figure 3(i)). Taken together, BCAR4 might function as a sponge of miR-181c-5p in ESCC.

Figure 3.

BCAR4 was a sponge for miR-181c-5p. (a) Starbase predicted binding site between BCAR4 and miR-181c-5p. (b) In ESCC cell lines and normal cell line, miR-181c-5p expression was detected. (c) In ESCC tissues and adjacent normal tissues, miR-181c-5p expression was detected (n = 40). (d) MiR-181c-5p expression was detected after transfection of miR-181c-5p mimics or miR-NC into Ec-9706 and TE-1 cells. (e) The relative luciferase activity was detected. (f) The enrichment of BCAR4 and miR-181c-5p was determined by RIP assay. (g) MiR-181c-5p expression was detected after transfection of si-BCAR4 or si-NC into Ec-9706 and TE-1 cells. (h) MiR-181c-5p expression was detected after transfection of pc-BCAR4 or pc-NC into Ec-9706 and TE-1 cells. (i) In ESCC tissues, correlation between BCAR4 and miR-181c-5p expression was assessed (n = 40). ** p < 0.01, *** p < 0.001.

The effect of knockdown of BCAR4 was attenuated by miR-181c-5p

Whether miR-181c-5p mediated the oncogenic role of BCAR4 in ESCC was then explored. It was found that si-BCAR4 inhibited Ec-9706 and TE-1 cell proliferation (Figure 4(a,b), p < 0.01), which was attenuated by miR-181c-5p inhibitor (p < 0.05). The si-BCAR4 significantly decreased Ec-9706 and TE-1 cell invasion and migration (p < 0.05), which was attenuated by miR-181c-5p inhibitor (p < 0.05, Figure 4(c,d)). si-BCAR4 markedly increased Ec-9706 and TE-1 cell apoptosis (p < 0.05), which was attenuated by miR-181c-5p inhibitor (p < 0.05, Figure 4(e)). Taken together, miR-181c-5p partially mediated the oncogenic role of BCAR4 in ESCC.

Figure 4.

MiR-181c-5p in ESCC function in vitro. (a) Cell viability was evaluated. (b) Cell proliferation was evaluated. (c and d) Cell invasion (c) and migration (d) were evaluated. (e) Cell apoptosis was detected. * p < 0.05, ** p < 0.01 vs si-NC group; # p < 0.05 vs si-BCAR4 group.

LASP1 was a target of miR-181c-5p

Next, potential base pairs formed by LASP1 and miR-181c-5p were predicted by the miRanda database. It was found that 3′-UTR of LASP1 might be the potential target of miR-181c-5p (Figure 5(a) and Supplementary data 2). In addition, overexpression of miR-181c-5p significantly decreased LASP1-WT vector mediated luciferase activity, but not LASP1-MUT vector mediated luciferase activity compared with miR-NC (Figure 5(b), p < 0.01). In addition, overexpression of miR-181c-5p decreased the expression levels of LASP1, while miR-181c-5p inhibition increased the expression levels of LASP1 at both mRNA and protein levels (Figure 5(c,d), p < 0.01). Simultaneously, si-BCAR4 could promote the expression of LAPS1 at both mRNA and protein levels (S1 A, B, p < 0.01). Moreover, LASP1 was upregulated (n = 40, p < 0.01, Figure 5(e)) in ESCC tissues, and the expression of miR-181c-5p was negative correlated with LASP1 (n = 40, r = 0.669, p < 0.001, Figure 5(f)). Taken together, these results indicated that miR-181c-5p targeted LASP1 in ESCC.

Figure 5.

LASP1 was a target of miR-181c-5p. (a) Targetscan predicted putative binding site between miR-181c-5p and LASP1. (b) The relative luciferase activity was detected. (c and d) LASP1 expression was detected by qRT-PCR (c) and western blot (d). (e) in ESCC tissues and adjacent normal tissues, LASP1 expression was detected (n = 40). (f) The correlation between miR-181c-5p and LASP1 expression in ESCC tissues was analyzed (n = 40). ** p < 0.01 vs miR-NC group; ## p < 0.01 vs inhibitor NC group.

The effect of miR-181c-5p mimics was attenuated by overexpression of LASP1 in ESCC

Whether LASP1 mediated the function of miR-181c-5p in ESCC was then investigated. As shown in Figure 6(a,b), compared with miR-NC, miR-181c-5p mimics significantly inhibited Ec-9706 and TE-1 cell growth (p < 0.01), which was attenuated by overexpression of LASP1 (p < 0.05). miR-181c-5p mimics decreased Ec-9706 and TE-1 cell migration and invasion (p < 0.01), which was attenuated by overexpression of LASP1 (p < 0.05, Figure 6(c,d)). miR-181c-5p mimics significantly promoted Ec-9706 and TE-1 cell apoptosis (p < 0.01), which was attenuated by overexpression of LASP1 (p < 0.01, Figure 6(e)). These results demonstrated that LASP1 partially mediated the function of miR-181c-5p in ESCC.

Figure 6.

Overexpression of LASP1 in ESCC in vitro. (a) Cell viability was evaluated. (b) Cell proliferation was evaluated. (c and d) Cell invasion (c) and migration (d) were evaluated. (e) Cell apoptosis was detected. ** p < 0.01 vs mi-NC group; # p < 0.05, ## p < 0.01 vs miR-181c-5p mimics group.

The BCAR4/miR-181c-5p/LASP1 axis participated in ESCC progression by regulating the p-STAT3 and COX-2 signaling

We next evaluated the function of the BCAR4/miR-181c-5p/LASP1 axis in the STAT3 and COX-2 signaling. We found that downregulation of BCAR4 significantly decreased the expression levels of p-STST3 and COX2 in Ec-9706 and TE-1 cells, (p <0.01), but not the expression of STAT3 (Figure 7(a)). The si-BCAR4 significantly inhibited the expression of p-STST3and COX2 (p < 0.05), which was attenuated by co-transfection of miR-181c-5p mimics (p < 0.05) (Figure 7(b)). Overexpression of miR-181c-5p also decreased the expression levels of p-STST3 (p < 0.01) and COX2 (p < 0.01), which was attenuated by co-transfection of pcDNA-LASP1 (p < 0.05, Figure 7(c)). Taken together, p-STAT3 and COX-2 signaling might mediated the function of BCAR4/miR-181c-5p/LASP1 axis in ESCC progression.

Figure 7.

BCAR4/miR-181c-5p/LASP1 participated in ESCC progression by modulating p-STAT3 and COX-2 signaling. (a) si-BCAR4 or si-NC were transfected into Ec-9706 and TE-1 cells. (b) si-NC, si-BCAR4, miR-181c-5p inhibitor were transfected or co-transfected into Ec-9706 and TE-1 cells. (c) miR-NC, miR-181c-5p mimics were transfected or co-transfected into Ec-9706 and TE-1 cells. p-STAT3, STAT3 and COX2 expression was detected. * p < 0.05, ** p < 0.01 vs si-NC or miR-NC group; # p < 0.05, ## p < 0.01 vs si-BCAR4 or miR-181c-5p mimics group.

MiR-181c-5p/LASP1 axis in tumor growth in vivo

Above results confirmed that si-BCAR4 induced promotion of LSCC progression through promoting the miR-181c-5p/EGFR axis. Next, the effect of overexpression of miR-181c-5p on tumor promotion induced by BCAR4 was explored. The representative images of 5-week subcutaneous tumors from different groups are shown in Figure 8(a). sh-BCAR4 decreased tumor weight and volume (p < 0.01), which was attenuated by miR-181c-5p inhibitor (Figure 8(b,c)). Moreover, sh-BCAR4 reduced cell proliferation, which was inhibited by miR-181c-5p inhibitor (Figure 8(d)). In addition, in tumor tissues, sh-BCAR4 decreased the expression levels of LASP1 (p < 0.01), which was attenuated by miR-181c-5p inhibitor (p < 0.05). Taken together, the miR-181c-5p/LASP1 axis mediated the function of BCAR4 in tumor growth in vivo.

Figure 8.

BCAR4 in tumor growth in vivo. (a) From four groups, representative images of subcutaneous tumors were shown. (b) Tumor weight. (c) Tumor volume. (d) By immunohistochemistry staining using anti-Ki-67 antibody, cell proliferation in vivo was evaluated. (e) LASP1 expression in tumor tissues was detected. n = 5 in each group. * p < 0.05, ** p < 0.01 vs sh-NC group; # p < 0.05 vs sh-BCAR4 group.

Discussion

ESCC is known to have a high incidence and often leads to poor prognosis [29]. Therefore, the development of effective targeted therapies for ESCC is urgent. Here, we revealed that knockdown of BCAR4 effectively inhibited cell proliferation, invasion and migration in vitro, and the tumorigenesis by targeting miR-181c-5p/LASP1 followed by decreased expression levels of p -STAT3 and COX2 in vivo.

Abnormally expressed lncRNAs in cell growth and tumor development in ESCC have been well studied. For example, lncRNA Erbb4-IR is highly upregulated in tumor tissues of ESCC patients and overexpression of Erbb4-IR results in a significantly increased rates of proliferation and decreased rates of apoptosis in ESCC cells [30]. In ESCC cells and tissues, the expression SNHG16 is markedly upregulated, which is strongly associated with lymph nodes metastasis, clinical stage and tumor stage of ESCC patients [31]. MNX1-AS1 is upregulated in ESCC tissues and involved in ESCC cell function [32]. Even with the identification of these important lncRNAs involved in ESCC progression, more specific biomarkers for ESCC are still necessary. Although lncRNA BCAR4 is involved in bladder cancer [12], NLCLC [33], colon cancer [13], colorectal cancer [14], and cervical cancer [34], the function of BCAR4 in ESCC remains unknown. Here, we found for the first time that the BCAR4 was upregulated in ESCC cells and tissues. Furthermore, knockdown of BCAR4 effectively inhibited ESCC cell growth in vitro, and suppressed ESCC cell malignancy in vivo. Our study identified a novel lncRNA in ESCC progression, which might be a potential biomarker for the diagnosis and treatment of ESCC.

During disease progression, lncRNAs often inhibit target miRNA through directly binding [35,36]. One recent study revealed several direct target miRNAs of BCAR4 in various cancers including miR-665 in colorectal cancer [14], miR‑370‑3p and miR-644a in bladder cancer [12,37], miR-2276 in glioma [38], and miR-1261 in liver cancer [39]. In the present study miR-181c-5p was identified as a BCAR4 target. miR-181c-5p exerts important functions in different types of cancer. For example, downregulation of miR-181c-5p inhibits pancreatic cancer cell viability [40]. However, the function of miR-181c-5p in ESCC remains unclear. Here, miR-181c-5p was downregulated in ESCC cell and tissues. BCAR4 was a sponge of miR-181c-5p to inhibit the expression of miR-181c-5p. Moreover, overexpression of miR-181c-5p showed a similar protective role as knockdown of BCAR4. Downregulation of miR-181c-5p enhanced cell proliferation, invasion and migration, while attenuated the effect of knockdown of BCAR4 in mediating ESCC cell growth. Our study elucidated the miR-181c-5p function in ESCC and contributed to the understanding of ESCC pathogenesis from the network of lnRNAs/miRNAs.

To study the mechanism of the BCAR4/miR-181c-5p axis in ESCC, we identified that LASP1 was a target of miR-181c-5p. Overexpression of LASP1 promotes ESCC cell function [18]. Here, overexpression of LASP1 attenuated miR-181c-5p mimics mediated effect on cell growth and apoptosis. It was reported that LASP1 affected the tumor development partially through activating the STAT3 and COX2 signaling [41,42]. STAT3 is a transcription factor that can regulate gene expression associated with cell cycle, survival, and immune response involved in cancer progression of a large number of cancer types [43]. COX2 is a an inducible enzyme and high expression levels of COX2 is positively related with tumor proliferation and metastasis of various human malignancies [44]. Here, knockdown of BCAR4 and overexpression of miR-181c-5p decreased the expression levels of p-STAT3 and COX2, downregulation of miR-181c-5p increased the expression levels of p-STAT3 and COX2. Furthermore, overexpression of LASP1 attenuated miR-181c-5p mimics mediated effect on p-STAT3 and COX2 expression, suggesting that STAT3 and COX2 signaling partially mediated BCAR4/miR-181c-5p/LASP1 effect on ESCC.

Conclusion

In conclusion, BCAR4 was found to promote the development of ESCC by targeting the miR-181c-5p/LASP1 axis, followed by upregulation of p-STAT3 and COX2 both in vitro and in vivo, suggesting that BCAR4 might be a potential therapeutic target for ESCC.

Click here for additional data file.