LncRNA LUCAT1/miR-181a-5p axis promotes proliferation and invasion of breast cancer via targeting KLF6 and KLF15

Long non-coding RNAs (lncRNAs) are novel regulatory molecules in breast cancer development. LncRNA LUCAT1 is a potential tumor promoter in human cancers. In this study, we aimed to explore the role of LUCAT1 in human breast cancer tissues and cells. A total of 31 breast cancer patients who underwent tumor resection, but without chemo- or radiotherapy or acute lung/heart/kidney diseases, provided tumor and adjacent normal tissues. Bioinformatic analysis, qRT-PCR, and luciferase reporter assay were carried out during the study. qRT-PCR analysis indicated that, compared with the adjacent tissues and MCF-10A normal breast epithelial cells, LUCAT1 was markedly up-regulated in the breast cancer tissues and five BC cell lines, including MDA-MB-231, MDA-MB-468, MDA-MB-435, SKBR3, and MCF-7. The knockdown of LUCAT1, through the transfection of small interfering RNA (siRNA) specific to LUCAT1, resulted in inhibition of proliferation in breast cancer cells. The expression levels of miR-181a-5p were decreased in the breast cancer tissues and five BC cell lines. Bioinformatic analysis and luciferase reporter assay suggested the interaction between miR-181a-5p and LUCAT1. In addition, the effects of LUCAT1 on promoting cell proliferation were attenuated by overexpression of miR-181a-5p through the transfection of miR-181a-5p mimic. Moreover, bioinformatics and luciferase reporter assay confirmed that miR-181a-5p targeted the 3′-UTR region of KLF6 and KLF15 mRNA, which were two tumor suppressor genes. LUCAT1/miR-181a-5p axis regulated the expression of KLF6 and KLF15 both in vitro and in vivo. Our data indicate that LUCAT1/miR-181a-5p axis can serve as a novel therapeutic target in breast cancer.

cancer, a number of lncRNAs and miRNAs are found as potent regulators, which are utilized as both diagnostic and therapeutic targets, providing promising targets of novel molecular therapies in advancing the BC clinical management [4,5].
LncRNA LUCAT1 is recently characterized as a tumor-promoting lncRNA in a wide variety of cancers, including osteosarcoma [6], non-small cell lung cancer [7], esophageal cancer [8], and renal cell carcinoma [9]. Importantly, LUCAT1 was shown to be associated with tumorigenesis [8] and poor prognosis of cancer patients [7] and participated in chemoresistance [7]. This study aimed to investigate the role of LUCAT1 in BC.
MiR-181a-5p is a miRNA shown to play an essential role in multiple cancers, such as hepatic cancer [10], non-small cell lung cancer [11,12] and human diseases, such as osteoarthritis [13]. The interaction between lncRNA CCAT1 and miR-181a-5p has been demonstrated in endometrial carcinoma [14]. However, the interaction between miR-181a-5p and LUCAT1 has not been verified in BC and the underlying mechanism of their regulation has not been fully elucidated.
KLF6 and KLF15 are two genes of the class of Kruppel-like factors. In BC, KLF6 was shown to inhibit cancer growth by inactivating estrogen-receptor mediated pathways [15]. KLF15 is also a putative BC suppressor gene [16]. KLFs could modulate the expression of ERα. And they all could inhibit the expression of ERα. Moreover, KLF6 and KLF15 are also targets of miRNAs [17,18].
The present study aimed to investigate the role of the LUCAT1/miR-181a-5p axis in BC and the mechanism of their regulation. Both clinical tissues and BC cell lines were used to explore the potential of LUCAT1 and miR-181a-5p as biomarkers of BC. Overexpression and knockdown of LUCAT1 were induced through the transfection of overexpression plasmid and small-interfering RNAs (siRNAs) specific to LUCAT1, to explore the role of LUCAT1 in regulating BC proliferation in vitro and in vivo. Bioinformatic analysis and luciferase reporter assay were utilized to investigate the binding of LUCAT1 and miR-181a-5p to their targets.

LUCAT1 is upregulated in breast cancer
The expression levels of LUCAT1 in BC tissues were compared with that of adjacent tissues, which showed a significant upregulation of LUCAT1 in BC tissues (p < 0.01, Fig. 1a and b). Further, the expression of LUCAT1 in different BC cells (MCF-7, MDA-MB-231, MDA-MB-468, MDA-MB-435, SKBR3) and normal breast cells (MCF-10A) were also evaluated. The results showed a considerable overexpression of LUCAT1 in the BC cell lines compared to that in MCF-10A cells (p < 0.05, Fig. 1c). Due to the prominent upregulation of LUCAT1 in MDA-MB-231 cells compared with that in other cell lines, MDA-MB-231 cells were used in the subsequent experiments. The MDA-MB-231 cell line used in our experiment was genotyped by STR genotype detection. The DNA typing of this cell line found a perfect match in the cell line search, and the ATCC database showed that the cell was named MDA-MB-231. No multiple alleles were found in this cell line (Table 1). In addition, miR-181a-5p was downregulated in BC tissues and BC cell lines (Fig. 1d and e).
Downregulation of LUCAT1 reduces cell proliferation, migration, and invasion of breast cancer LUCAT1 was knocked-down and qRT-PCR results showed that the expression of LUCAT1 was downregulated (p < 0.01, Fig. 3a). MTT and colony formation assay revealed that knockdown of LUCAT1 reduced cell viability and proliferation, respectively ( Fig. 3b and c). Moreover, knockdown of LUCAT1 significantly suppressed cell migratory and invasive capacities of MDA-MB-231 cells (Fig. 3d-f).

MiR-181a-5p mediates the tumor-promoting effects of LUCAT1 in vivo
The role of LUCAT1/miR-181a-5p axis in BC was explored in vivo. It was shown that overexpression of LUCAT1 could promote the growth of mice tumors, whereas overexpression of miR-181a-5p could attenuate tumor growth. The tumor volume of mice in pc-LUCAT1 + miR-181a-5p mimic group was significantly smaller than that in the pc-LUCAT1 group (P < 0.05, Fig. 5a-d).

LUCAT1/miR-181a-5p axis regulates the expression of KLF6 and KLF15
To seek the potential targets of miR-181a-5p, bioinformatic analysis was conducted and the output of miRDB showed that the 3′-UTR region of seven consecutive bases in KLF6 and KLF15 mRNA sequences completely complement with the seed of miR-181a-5p ( Fig. 6a and c). The 3′-UTR luciferase reporter gene assay showed that miR-181a-5p pronouncedly inhibited the luciferase activity of pcDNA-KLF6 or pcDNA-KLF15 (p < 0.05, Fig. 6b and d). To further validate that KLF6 and KLF15 were targets of miR-181a-5p, western blotting analysis of KLF6 and KLF15 were performed, which showed increased expression levels of KLF6 and KLF15 in cells with pc-LUCAT1 transfection. In contrast, the expression levels of KLF6 and KLF15 were sharply reduced by transfection of miR-181a-5p mimic. The reduction of KLF6 and KLF15 by miR-181a-5p could be reversed by the co-transfection of pc-LUCAT1 and miR-181a-5p mimic together (p < 0.01, Fig. 6e). The same results were also observed in mice tumor tissues (p < 0.05, Fig. 6f). The expression of KLF6, KLF15 and ERα were detected in in MDA-MB-231 cells treated with miR-181a-5p mimic, miR-181a-5p mimic + inhibitor, miR-181a-5p mimic + pc-LUCAT1, and miR-181a-5p mimic + pc-LUCAT1 + inhibitor. The results showed that the expression levels of KLF6 and KLF15 were increasing, while the expression levels of ERα were decreasing. It suggested that LUCAT1/miR-181a-5p axis regulated the expression of KLF6 and KLF15, and KLFs modulated the expression of ERα (p < 0.05, p < 0.01, Fig. 6g).

Discussions
In this study, we explored a novel lncRNA, LUCAT1, as a potential biomarker in BC. Using qRT-PCR analysis, we showed that LUCAT1 was overexpressed in both BC tissues and cells. This finding was consistent with previous reports that LUCAT1 is overexpressed in multiple human cancers [6][7][8][9]. Our findings indicated the potential use of LUCAT1 as a diagnostic marker of BC. To further validate the expression of LUCAT1 in predicting BC prognosis, correlation of the expression levels of LUCAT1 with clinical stages is needed to examine whether the expression of LUCAT1 elevated with higher-grade cancer. Instead of analysis on tissue-level LUCAT1, the quantification of serum level LUCAT1 is more convenient and of greater translational potential, which portends further studies. Further, the value of a LUAT1 as a therapeutic target in BC was demonstrated as the impaired BC cell proliferation by si-LUCAT1 and enhanced BC cell proliferation by pc-LUCAT1. This finding agrees with previous data showing that knockdown of LUCAT1 was able to exert a potent anti-tumor effect [19]. Given the potential of LUCAT1 as a therapeutic target in reversing chemoresistance [6], further studies are also needed to investigate if knockdown of LUCAT1 is a viable strategy to treat chemoresistance BC, which is critical for improving the prognosis of late-stage BC patients.
Contrary to the upregulation of LUCAT1 in BC, miR-181a-5p was shown to be downregulated in BC, and a negative correlation between miR-181a-5p and LUCAT1 was observed. Using bioinformatic analysis and luciferase reporter assay, we confirmed that miR-181a-5p was a target of LUCAT1, which was in agreement with the sponging effects of lncRNAs in regulating expression of miRNAs [20]. This interaction between LUCAT1 and miR-181a-5p translated to the observation that miR-181a-5p mediated the tumor-promoting effects of ectopic overexpression of LUCAT1, as evidenced by the attenuated in vitro BC cell proliferation and in vivo tumor growth after co-transfection of miR-181a-5p mimic and pc-LUCAT1. Previously, LUCAT1 was shown to interact with an array of miRNAs such as miR-375 [19] and miR-200c [6] Therefore, our data cannot rule out that the interaction between LUCAT1 and other miRNAs also contribute to the tumor-promoting effects of LUCAT1.
MiR-181a-5p was previously shown as a therapeutic target in a number of cancers [11,12]. Here we showed that miR-181a-5p inhibited KLF6 and KLF15. We focused on KLF6 and KLF15 as the interaction between miR-181a-5p and KLF6 has been previously demonstrated in gastric cancer [21], and KLF6 and KLF15 were putative tumor suppressors in BC [15,16,22,23]. Our study provides clear evidence that KLF6 and KLF15 are targets of miR-181a-5p and the regulation of KLF6 and KL15 by miR-181a-5p could be a mechanism of the regulation of miR-181a-5p and LUCAT1 in BC.

Conclusion
In summary, we demonstrated that the LUCAT1/miR-181a-5p axis constitutes an important regulatory mechanism in BC, whereby LUCAT1 is a tumor inducer and miR-181a-5p is a tumor suppressor. A direct interaction between LUCAT1 and miR-181a-5p exists. The downregulation of KLF6 and KLF15 are also identified as a mechanism of the regulation of miR-181a-5p in BC. LUCAT1/miR-181a-5p is a promising therapeutic target in BC.

Colony formation assay
Cells (1 × 10 3 cells per well) were seeded in a 6-well plate and incubated for 1 week. After washing with PBS, cells were fixed with 4% formaldehyde for 15 min and stained for 10-30 min with 2.5% Giemsa stain, followed by counting the colonies with a diameter of over 100 μm.

Luciferase reporter assay
For Luciferase reporter assay, 100 ng plasmids and 200 nmol/L miR-181a-5p mimic or their negative control were used to transfect cells (1 × 10 5 per ml) using Attractene Transfection Reagent (Qiagen). After transfection for 2 d, the promoters of LUCAT1, KLF6 and KLF15 were amplified and inserted into a psiCHECK™-2 vector (Promega), the luciferase activity was evaluated by determining the ratio of firefly to Renilla luciferase activity with a dual-luciferase reporter system (Promega).

Tumor xenograft in vivo
A total of 30 tumor-bearing nude mice (18-22 g, 6-8-week-old) were purchased from the Animal Center of Tongji Hospital. Animal experiments were conducted following the guidelines for the Protection and Use of Experimental Animals issued by the National Institutes   follows: volume = (long diameter × short diameter 2 )/2. After 28 d, the mice were euthanized with isoflurane, the mice were sacrificed and the tumor tissues were weighed. Tissue samples were stored in − 80°C for further analysis.

Bioinformatics analyses
The prediction of target genes of miR-181a-5p was performed by Starbase (http://starbase.sysu.edu.cn), TargetScan (http://www.targetscan.org/) and miRDB (mirdb.org/ miRDB/) bioinformatics analyses, followed by further analysis with the KEGG system. Twenty predicted targets were identified during the first screening. Next, the secondary screening was performed by selection of genes with functional change that induced any changes in LUCAT1, KLF15 and KLF6 in previous studies, which were selected prior to other candidates.

STR genotype detection
DNA was extracted by Axygen genome extraction kit (Corning) and amplified by 21-STR amplification scheme. STR point and sex gene Amelogenin were detected on the ABI3730XL genetic analyzer.

Statistical analyses
GraphPad Prism 6.0 (GraphPad Software) was used for data analysis. Data were expressed as mean ± standard deviation (SD). Differences between individual groups were analyzed by Student's t-test. The criterion of statistical significance was p < 0.05.