Novel insights into the mechanisms by which lncRNAHOTAIR regulates migration and invasion in HeLa cells

Background HOTAIR, as one of the few well-studied oncogenic lncRNAs, is involved in human tumorigenesis and is dysregulated in most human cancers. The transcription co-activator factor YAP1 is broadly expressed in many tissues, and promotes cancer metastasis and progression. However, the precise biological roles of HOTAIR and YAP1 in cancer cells remain unclear. The expression levels of HOTAIR and YAP1 were measured by quantitative PCR (qPCR), immunoblotting. Wound-healing and transwell assays were used to examine the invasive abilities of HeLa cells. Luciferase reporter assays and CHIP were used to determine how YAP1 regulates RPL23. A xenograft mouse mode was used to assess the correlation between HOTAIR and YAP1 in vivo. In this study, we showed that HOTAIR regulates H3K27 histone modication in the promoter of miR-200a to mediate miR-200a expression byrecruiting EZH2. YAP1, as a potential target gene of miR-200a, aggravated the effects of miR-200a on the migration and invasion of HeLa cells. YAP1 activated the transcription of RPL23, which is a novel target of YAP1 transcriptional regulation. Agreement with this, the expression of YAP1 and RPL23 was dramatically decreased after injecting HeLa cells transfected with siHOTAIR in a xenograft mouse model. These elucidates that HOTAIR, as an oncogenic lncRNA, recruits EZH2 to reduce miR-200a-3p expression via H3k27 trimethylation in the miR-200a-3p promoter. As a target gene of miR-200a-3p,YAP1 then promotes the migration and invasion of HeLa cells by mediating the downstream transcription of RPL23 which normally functions as a cancer-promoting factor. Accordingly, we propose a novel model of the molecular mechanism by which HOTAIR promotes the migration and invasion of cancer cells involving the miR-200a-3p/YAP1/RPL23 axis.


Background
In recent years, lncRNAs have gained widespread attention as a group of non-coding transcripts of > 200 nucleotides that are involved in a wide range of biological processes (1)(2)(3)(4)(5)(6)(7)(8). lncRNAs modulate gene expression at the transcriptional, post-transcriptional, translational, or post-translational levels (9)(10)(11)(12). Within the last decade, HOTAIR has emerged as a key regulator of carcinogenesis and metastasis, a crucial oncogenic lncRNA contributing to different processes in several cancers (13)(14)(15). Howardet.al. rstuncovered a compelling mechanistic basis for HOTAIR in cancer, showing that it interacts with PRC2 to enhance H3K27 trimethylation, and decreases the expression of a large number of genes (16). HOTAIR, a negative prognostic factor, has been correlated with cancer cell proliferation, apoptosis, invasion, and metastasis in various cancer cell lines (17)(18)(19)(20). Moreover, we showed that it may be involved in a diverse range of biological processes by mediating the expression of proteins, such as MKL1, OGFR, and vimentin (21)(22)(23). In our previous work, the well-established oncogene YAP1 was found to be signi cantly decreased after HOTAIR inhibition in HepG2 cells using proteomics technology (23). However, there is no evidence investigating whether HOTAIR can regulate YAP1 expression.
YAP is a transcription co-activator with a potential C-terminal transactivation domain (24) and an N-terminal region responsible for interaction with TEAD (25). However, YAP1 itself has no DNA binding activity; therefore, it must bind to DNAbinding transcription factors, such as TEAD1-4, RUNX2, and the ErbB4 cytoplasmic domain, to stimulate gene expression (24)(25)(26). Four TEAD genes are widely but variedly expressed in most human tissues. Moreover, YAP1 is widely overexpressed in many human tumors, and plays an essential role in cancer initiation, progression, and metastasis (27)(28)(29)(30). YAP1 is phosphorylated and inhibited by large tumor suppressor 1/2 kinases, which are important components of the Hippo pathway (31,32). YAP1 phosphorylation leads to its cytoplasmic retention and reduced nuclear localization, preventing it from stimulating gene expression. The observed YAP1 nuclear-cytoplasmic shuttling indicates that the nuclear abundance of YAP1 plays a key role in tumor growth controlin vitro (33).
MiRNAs are a family of small endogenous non-coding RNAs molecular found in a diverse range of organisms, whichinvolved in many important biological processes such as cell growth, apoptosisand cancer progression (34)(35)(36)(37)(38). MiR-200a, as a member of miR-200 family, located in chromosome 1 (39,40) and played a key role in the growth and development of tumors (7,(41)(42)(43)(44)(45)(46)(47). This miRNA has been implicated in epithelial to mesenchymal transition and tumor invasion by targeting the transcriptional factors of β-catenin, ZEB1, and ZEB2 (48,49). Wang et.al.reported that miR-200a is a likely master regulator that affects the metastatic potential of cervical cancer by coordinately suppressing the expression of multiple genes important to cell motility (50). Epigenetic mechanisms are crucial regulators of cell type-speci c genes, including miRNAs. Lukas et.al.found that the promoter of miR-200a is occupied by the polycomb-speci c marker H3K27me3 (51), while the silencing of miR-200a by histone methylation was shown to promote tumor growth and invasion in numerous types of human cancer (52)(53)(54)(55). Enhancer of zeste homolog 2 (EZH2), as a methyltransferase and critical component of PRC2, is primarily responsible for H3K27 trimethylation. Meanwhile, HOTAIR recruits the PRC2 complex to regulate target gene expression via histone modi cations. However, the precise relationship among HOTAIR, miR-200a, and H3K27 methylation was unclear.
In this study, we analyzed the correlations and potential role of HOTAIR and YAP1 in the migration and invasion of HeLa cells. We found that HOTAIR inhibition increased the expression of miR-200a-3p by preventingrecruiting of EZH2 to the miR-200a-3p promoter, thus decreasing H3K27 methylation. Furthermore, YAP1, as the targeted gene of miR-200a-3p, was investigated to promote the migration and invasion of HeLa cells. Additionally, RPL23 was rst veri ed as the downstream transcriptionally regulated gene of YAP1 that promoted migration and invasion through regulating the expression of mutant p53 in HeLa cells. In agreement with these results, we showed that the expression of both YAP1 and RPL23 was decreased after injecting HeLa cells transfected with siHOTAIR in a xenograft mouse model.Taking all ndings into account, we deduced that HOTAIR may promote migration and invasion of HeLa cells through the miR-200a-3p/YAP1/RPL23 axis both in vitro and in vivo.

Cell culture and siRNAtransfection
The human cervical cancer cell line HeLa obtained from the American Type Culture Collection (ATCC) was grown in DMEM containing 10% FBS, 2 mM glutamine, 50U/ml penicillin, and 50mg/ml streptomycin at 37°C with 5% CO 2 . A total of 40 nM siRNA targeted against HOTAIR (siHOTAIR-I or siHOTAIR-II) or negative control siRNA (siNC) were transfected into HeLa cells using RNAi-mate. All siRNAs were purchased from GenePharma Co., Ltd., and siRNA sequences are listed in Table S1-1. HOTAIR expression levels were measured 48 h after transfection by qPCR (see below).
HeLa cells were also transfected with 40nM of siRNA targeted against YAP1 or RPL23 to silence gene expression. Again, cells were harvested 48 h after transfection, and protein expression was assessed by western blotting (see below).
MiR-200a-3p mimics, inhibitors, and negative controls were obtained from GenePharma, and their sequences are listed in Table S1-1. Cells cultured in 6-well plates were transfected miR-200a-3p mimics or inhibitors with 40nM/well according to the manufacturer's protocol. The negative controls consisting of random sequences had no detectable effects on human cell lines or tissues.

RNA isolation and qRT-PCR
Total RNA was extracted from cultured HeLa cells using TRIzol reagent (Invitrogen) according to the manufacturer's protocol. A Nanodrop 2000 spectrophotometer was used to measure the concentration of total RNA. RNA was then reverse-transcribed into rst strand cDNA using the Revert Aid First Strand cDNA Synthesis Kit (Invitrogen). Quantitative PCR was carried out using the SYBR Green PCR Master Mix (Roche) and Light Cycler 480 Real-Time PCR system (Roche). GAPDH was used as the endogenous control gene to normalize expression of the target genes. Each sample was analyzed in triplicate. The thermal cycling program consisted of 95°C for 5 min followed by 40 cycles of 95°C for 10 s, 62°C for 45 s, and 72°C for 30 s. Melting curve data were then collected to verify the PCR speci city and the absence of primer dimers. All primer sequences are listed in Table S1-2.
MiRNAs were isolated using the mirVana miRNA isolation kit (Ambion) according to the manufacturer's protocol, and quanti ed using a Nanodrop 2000 spectrophotometer. Quantitative analysis of miR-200a-3p expression was performed using a Hairpin-it miRNA real-time PCR quantitation kit (GenePharma). Small nuclear RNA U6 was used as an internal control. Each sample was analyzed in triplicate.

Plasmid constructs and expression
The full-length YAP1 (NM_001130145) and RPL23 (NM_000978) cDNA were ampli ed by RT-PCR from total RNA isolated from HeLa cells, and inserted into the mammalian expression vector pCDNA3.1/myc-his B (Invitrogen, USA). The BamH I and XhoI restriction sites were designed in the forward and reverse primers respectively. All the sequences of primers were listed in Table S1-3.
Plasmid was transfected into HeLa cells by Lipofection 2000 (Invitrogen) according to the manufacturer's instructions. After incubation for 6 h, the medium was removed and replaced with normal culture medium for 48 hr. And the plasmid pCDNA3.1/myc-his B was used as the negative control. Proteins expression was assessed by Western blotting.

Isolation of nuclear and cytoplasmic extracts
Nuclear extraction was performed using an NE-PER Nuclear Cytoplasmic Extraction Reagent kit (Pierce, Rockford, IL, USA) according to the manufacturer's instructions. In brief, HeLa cells were washed twice with cold PBS and centrifuged at 500 ´g for 5 min. The cell pellet was suspended in 200 μl of cytoplasmic extraction reagent I, and vortexed vigorously on the highest setting for 15 s, then incubated on ice for 10 min. Next, 11 μl CER II was added, vortexed for 5 s, incubated on ice for 1 min, and centrifuged at 16 000 ´g for 5 min. The supernatant (cytoplasmic extract) was immediately transferred to a clean pre-chilled tube. The insoluble pellet fraction, containing crude nuclei, was resuspended in 100 μl nuclear extraction reagent by vortexing for 15 s, incubating on ice for 10 min, then centrifuging at 16 000 ´g for 10 min. The supernatant (nuclear extract) was immediately transferred to a clean pre-chilled tube and used for subsequent experiments.
Wound healing assay HeLa cells were seeded into 6-well plates and allowed to grow to 70% con uency. Cell monolayers were then wounded by scratching a plastic pipette tip (1 mm) across the plate.

Western blotting
Protein extracts (10 μg) prepared with RIPA lysis buffer were resolved on a 12% SDS-PAGE gel, and transferred to an Immobilon-P PVDF transfer membrane (Millipore) by electro-blotting. After blocking with 5% non-fat milk, membranes were incubated overnight at 4°C with a 1:1000 dilution of the following primary antibodies: rabbit anti-YAP1 polyclonal, rabbit anti-RPL23 polyclonal, rabbit anti-mutant p53 polyclonal, rabbit anti-TBP (TATA binding protein) polyclonal, or mouse anti-GAPDH polyclonal. Blots were then incubated with peroxidaseconjugated IgG (ABclonal) diluted for 1 h at room temperature and then developed using a Super Signal West Pico kit (Pierce).
Immunoblots were scanned using an Image Scanner. Blot densitometry analysis was performed using Image J software. All analyses were performed in triplicate.

Assay of luciferase activity
Site-directed mutagenesis was used to change the binding sites of miR-200a-3p in the 3¢UTR of YAP1. The sequences of primer were shown in Table S1-4. For the reporter assay, a total of 20 nM miR-200a-3p mimics or control miRNA were co-transfected with 0.1μg pGL3-YAP1-3¢UTR or pGL3-YAP1-3¢UTR-mut into HeLa cells using Lipofectamine2000 (Invitrogen). Cells were harvested 48 h post-transfection using lysis buffer. Luciferase activities in cell lysates were determined using the Dual-Luciferase Reporter Assay System (Promega).

ChIP
The ChIP assay was performed using ChIP Assay Kit, following the manufacturer's instructions. Brie y, 1×10 7 HeLa cells were cross-linked with 1% formaldehyde for 10 min at 37°C. Then, cells were scraped and resuspended with ice-cold PBS containing protease inhibitor cocktails (Pierce). Cells were then lysed and sonicated to shear DNA to an average length of 200-1000 bp. All procedures were performed on ice. Lysates were immunoprecipitated with an anti-YAP1 or anti-H3K27me3 antibody at 4ºC overnight. Immunoprecipitation with an irrelevant normal IgG was used as a negative control. Immune complexes were then isolated with Protein A/G Sepharose beads at 4ºC for 1 h. After washing, DNA fragments contained in immune complexes were puri ed and ampli ed by PCR. The sequences of primer were shown in Table S1-5.

Immuno uorescence staining
HeLa cells were grown on sterile glass-bottomed culture dishes, xed in 4% formaldehyde for 30 min, and permeabilized in 1% FBS, 0.2% Triton-X100 on ice for 5 min. After washing, cells were blocked with 1% BSA for 1 h at room temperature, and then incubated with antibody at a dilution of 1:200 overnight. They were then incubated with Dylight 488-conjugated IgG (H+L) at a dilution of 1:100 for 2h in the dark, then stained with DAPI for 15 min. Subsequently, cells were thoroughly washed three times with PBS and examined using an LSM 710 laser scanning confocal microscope (Zeiss, Germany).

Xenograftmouse model
Four-week-old female athymic BALB/c mice were purchased from Vital River Laboratories. For xenograft models, 2 ´10 6 HeLa cells transfected with siNC or siHOTAIR were subcutaneously injected in the right ank of BALB/c nude mice (n=5 per group). Three weeks later, mice were sacri ced, tumors were homogenized, and proteins were extracted for western blotting. All animal procedures were performed in accordance with protocols approved by the Institutional Animal Care and Use Committee at the Institute of Wuhan University of Science and Technology.

Statistical analysis
Statistical analysis was performed using SPSS standard version 13.0 software. The independent Student's t-test was used to compare continuous variables between two groups. Data were expressed as means ± SD from at least three independent determinations. Values of P<0.05 (or P<0.01) were considered statistically signi cant.

HOTAIR inhibition affected the expression of YAP1 in HeLa cells
In our previous work (23), quantitative proteomics identi ed hundreds of differential expression proteins after HOTAIR inhibition in HepG2 cells, which contains the well-established oncogene YAP1. So, western blotting was used to analyze the whole protein expression of YAP1 after HOTAIR silencing in HeLa cells. As shown in Fig. 1A&1B, YAP1 protein expression was signi cantly decreased after transfecting siHOTAIR. Additionally, YAP1 levels were dramatically reduced in both the cytoplasm and nucleus after HOTAIR knockdown (Fig. 1A&1C   &1D). GAPDH and TATA-binding protein served as loading controls.
The migration and invasiveness of HeLa cells were further evaluated by the wound healing assay and Matrigel invasion assay. HOTAIR inhibition signi cantly decreased HeLa cell migration as determined by the wound healing assay (Fig. 1E& 1F), while YAP1 inhibition clearly reduced cell migration compared with the negative control (Fig. 1E &1F). Interestingly, co-transfection of HeLa cells with siHOTAIR and siYAP1further suppressed the effect compared with the silencing of HOTAIR or YAP1 alone (Fig. 1E &1F).
The number of invaded cells decreased signi cantly in HOTAIR knockdown cells compared with the negative control, while YAP1 inhibition reduced the invasion capability (Fig. 1G &1H). Furthermore, the co-transfection of HeLa cells with siHOTAIR and siYAP1 aggravated the effect of HOTAIR knockdown on cell invasion. Based on these results, we propose that YAP1 plays a critical role in the biological effects of HOTAIR on the migration and invasion of HeLa cells.
Functional effects of YAP1 in HeLa cells YAP1, as a well-established oncogene, is a major effecter of the Hippo pathway and is closely associated with cancer (59)(60)(61)(62). We evaluated the role of YAP1 in HeLa cells by the silencing/overexpression of YAP1 using the transient transfection of validated siRNAs and cDNAs. Plasmid pYAP1 was used to express YAP1 in HeLa cells,pCDNA3.1 was the negative control, and GAPDH served as the loading control. Western blotting was performed to determine YAP1 expression 48 h after transfection. As shown in Fig. 2A, YAP1 was signi cantly increased in HeLa cells compared with the control.Increased cell adhesion (Fig. 2B), cell migration, and invasion (Fig. 2C & 2D& 2E & 2F)were all observed after YAP1 overexpression.YAP1 expression was signi cantly decreased 48 h after the transfection of YAP1 siRNA (Fig. 2G), resulting in a reduction of cell adhesion (Fig. 2H), migration, and invasion ( Fig. 2G-J). These results suggest that YAP1 has the potential to promote HeLa cell migration and invasion in HeLa cells.

YAP1 is a potential target of miR-200a-3p
We nextevaluated whether YAP1 was regulated by the defection of a speci c miRNA in cervical carcinoma. A human miRNA chip microarray was used to detect differentially expression miRNAs following the transfection of HeLa cells with siRNA against HOTAIR. The expression of miR-200a-3p was signi cantly increased after HOTAIR inhibition (data not shown).
To investigate miRNA stargeting YAP1 that may be involved in the modulation of cell migration and invasion, we used the two common prediction algorithms TargetScan and PicTarto analyze the 3′UTR of YAP1. MiR-200a-3p was identi ed as targeting the 3′UTR of YAP1 by both algorithms. To further study whether miR-200a-3p expression was associated with the migration and invasion of cancer cells, we transfected miR-200a-3p mimics or an inhibitor intoHeLa cells. After transfecting miR-200a-3p mimics, miR-200a-3p expression levelswere signi cantly increased (Fig. 3A) and YAP1 protein expression was obviously decreased (Fig. 3B).Conversely, the miR-200a-3p inhibitor decreased miR-200a-3p RNA expression (Fig. 3C) and increased YAP1 expression (Fig. 3D).To further demonstrate the direct regulation of YAP1 by miR-200a-3p, we constructed luciferase reporters with the target sequences of wild-type (WT-UTR) and mutated YAP1 3′UTRs (mut-UTR). We mutated six bases in the predicted sites of the YAP1 3′UTR (Fig. 3E). As shown in Fig. 3F, the luciferase activities of YAP1-mut-luc were dramatically increased compared with the negative control WT-UTR. All data suggested that miR-200a-3p, as a potential miRNAtargeting YAP1, degraded YAP1 by targeting speci c sites in HeLa cells.

Effect of miR-200a-3p on migration and invasion in HeLa cells
To investigate whether miR-200a-3p has a role in mediating the biological roles between HOTAIR and YAP1, we rstly determined the functional roles of miR-200a-3p in the migration and invasion of HeLa cells. Compared with the negative control, the transfection of miR-200a-3p mimics signi cantlyreduced migration and invasion in HeLa cells (Fig. 4A-D). Moreover, the co-transfection of HeLa cells with miR-200a-3p mimics and YAP1 siRNA further decreased these effects compared with miR-200a-3p mimicsalone (Fig. 4A-D). In contrast, miR-200a-3p inhibition dramatically increased cell migration and invasion (Fig. 4E-H). What is more, co-transfection of HeLa cells with anmiR-200a-3p inhibitor and siYAP1 abrogated the effects of miR-200a-3p mimics on cell migration (Fig. 4E-H). These data suggested that miR-200a-3p promotes migration and invasion in HeLa cells by regulating YAP1 expression.
HOTAIR suppressed miR-200a-3p expression by H3 lysine 27 trimethylation HOTAIR is known to bind PRC2 (EZH2, EED, and SUZ12) at the 5′domain of HOTAIR(16), while EZH2 was reported to bind the miR-200a promoter and repress miR-200aexpression (54). We next investigated the relationship between HOTAIR and miR-200a-3p by examining the expression of miR-200a-3p after HOTAIR inhibition. As shown in Fig. 5A, miR-200a-3p was clearly increased after HOTAIR knockdown compared with negative control, and H3K27me3 expression was signi cantly decreased (Fig. 5B). We then performed ChIP assays to examine the effects of HOTAIR on the expression of H3K27me3 in the miR-200a-3p promoter. We found that H3K27me3 expression in the miR-200a-3p promoterwas signi cantly decreased to approximately 40% after HOTAIR inhibition in HeLa cells (Fig. 5C & 5D). These data suggested that HOTAIR increases H3K27me3 levels at the miR-200a-3p promoter to suppress miR-200a-3p expression by recruiting EZH2 in HeLa cells.
YAP1 promoted cancer cell migration and invasion by activating RPL23 transcription YAP1 and its paralog TAZ are co-transcriptional regulators downstream of the Hippo pathway. YAP1, as a co-activation factor, binds with TEAD to regulate the transcription of downstream genes (63). Here, the Cistrome Data Browser identi ed RPL23 as a novel target downstream of the transcriptional regulator YAP1. To evaluate whether YAP1 regulates RPL23 transcription, we rst altered the expression of YAP1 in HeLa cells by silencing/overexpressing using the transient transfection of validated siRNAs and cDNAs. Protein and mRNA levels of RPL23 were dramatically increased after the overexpression of YAP1 compared with controls( Fig. 6A & 6B), while YAP1 inhibition by siRNA suppressed RPL23 protein and mRNA expression (Fig. 6C& 6D). These ndings were consistent with immuno uorescence staining after altering YAP1 expression in HeLa cells (Fig. 6G&6H). We then used dual luciferase and ChIP assays to further investigate the relationship between YAP1 and RPL23. As shown in Fig. 6E, relative luciferase levels were clearly decreased after mutating the TEAD binding site on the RPL23 promoter, while YAP1 was highly enriched in the DNA fragments compared with negative control IgG immunoprecipitates (Fig. 6F).
RPL23 is a protein component of the 60S large ribosomal subunit that was previously shown to be up-regulated in many cancers (64,65), and was also emerging as a metastasis-related gene (66). As shown in Fig. 7A & 7B, the wound width was signi cantly decreased after overexpressing RPL23 compared with controls. In contrast, RPL23 inhibition clearly increased the wound width of HeLa cells after transiently transfecting siRPL23 (Fig. 7C& 7D). We next examined the effect of RPL23 on the invasion of HeLa cells using a Matrigel invasion assay. The number of invaded cells increased signi cantly in RPL23-overexpressingcells compared with the control, while RPL23 knockdown reduced the invasion capability (Fig. 7E-H).
Based on these results, we propose that YAP1 activates RPL23 transcription through TEAD binding with speci c sequences on the RPL23 promoter to promote the migration and invasion of cancer cells.

HOTAIR knock down reduced RPL23 expression
To further investigate the role of HOTAIR in the migration and invasion of cancer cells, we measured the expression of YAP1 and RPL23 after HOTAIR knockdown in vitro and in vivo. The expression of YAP1, RPL23, and mutant p53 was signi cantly decreased after HOTAIR knockdown compared with the negative control (Fig. 8A), while YAP1 inhibition by siRNA reduced RPL23 and mutant p53 expression (Fig. 8B). Conversely, RPL23 and mutant p53 expression was clearly increased after overexpressing YAP1 (Fig. 8C). Furthermore, RPL23 overexpression dramatically increased the expression of mutant p53 (Fig. 8D),while the mutant p53 was signi cantly reduced after RPL23 inhibition compared with the negative control (Fig. 8E). A xenograft mouse model was used to further validate in vitro ndings. The transient transfection of siHOTAIR was used to signi cantly suppress the tumorigenicity of HeLa cells. Subsequently, western blot analysis veri ed the reduction of YAP1 and RPL23 expression in HOTAIR knockdown xenografts compared with control xenografts (Fig. 8G), indicating that HOTAIR regulates the expression of YAP1 and RPL23 in vivo.
Based on these results, we propose that HOTAIR promotes migration and invasion in HeLacells by the miR-200a-3p/YAP1/RPL23 axis.

Discussion
As a new class of non-coding RNAs, lncRNAs have emerged as a crucial layer of gene regulation (13)(14)(15). HOTAIR is one of the few wellstudied lncRNAs and considerable attention has been given to determining its functions and identifying its target genes (15,67,68).
Previous studies showed that HOTAIR regulates a diverse range of biological processes by modulating RNA or protein levels of hundreds of genes (21,69). To better understand the precise molecular mechanism underlying the potential role of HOTAIR in cancer cells, it is essential to investigate the relationship between HOTAIR and its target genes.
We previously used quantitative proteomics technology to identify that HOTAIR up-regulates YAP1 expression in HeLa and HepG2 cells (11,21). As shown in Fig. 1, the present study showed that YAP1 expression was signi cantly decreased after HOTAIR inhibition, especially in the nucleus. Moreover, the capability of migration and invasion were aggravated after co-transfecting with siYAP1 and siHOTAIR compared with HOTAIR alone in HeLa cells (Fig. 1E & 1F). These results imply that YAP1 plays a pivotal role in the effects of HOTAIR on cell migration and invasion.
As a transcription co-activator factor, YAP1 is broadly expressed in many tissues, and promotes proliferation, migration, and invasion in several cancers by binding with the TEAD family to stimulate gene expression (25). YAP1 is overexpressed in numerous human cancers, such as gastric cancer, lung adenocarcinoma, ovarian cancer, and prostate cancer (30,70,71). It exerts its tumor-supporting properties by up-regulating Jag-1 expression and activating the Notch pathway (28). Moreover, YAP1 has been de ned as a prognostic biomarker and potential therapeutic target for gastric cancer, which promotes colony formation, cell growth, and metastasis both in vitro and in vivo (30). In agreement with these published data, we found that the overexpression of YAP1 increasedthe invasion and migration of HeLa cells, whilethe inhibition of YAP1 dramatically suppressed them. We con rmed this positive correlation between YAP1 and HOTAIR in a xenograft mouse model (Fig. 8).By investigating the impact of HOTAIR inhibition and YAP1 knockdown in HeLa cells, we show that the effects of HOTAIR knockdown oncancer cell migration and invasion can be mimicked by the respective manipulation of YAP1 expression. Importantly, HeLa cell migration and invasiveness were further decreased by the simultaneous inhibition of HOTAIR and YAP1.
MiR-200a, as a tumor suppressor, regulates the expression of several oncogenes in various cancers, while miR-200a overexpression signi cantly decreased cell motility in cervical cancer (50,72,73). We previously used the Affymetrix miRNA assay to show that miR-200a was dramatically increased after HOTAIR knockdown in HeLa cells. In the present study, we further veri ed that HOTAIR knockdown enhanced miR-200a-3p expression in HeLa cells. Recent research has shown that the histone modi cation of H3K27 occupies the promoter of miR-200a and silences its expression (51,52,54,55). EZH2, as a methyltransferase and critical component of PRC2, is primarily responsible for H3K27 methylation. Intriguingly, HOTAIR recruits the PRC2 complex to regulate target gene expression via histone modi cations (16). However, the precise relationship between HOTAIR, miR-200a, and H3K27me3 was unclear.
To investigate this in the present study, we determined H3K27me3 expression using ChIP combined with RNA interference. As shown in Fig. 3, H3K27me3 expression in the promoter of miR-200a-3p was clearly decreased after HOTAIR inhibition compared with the negative control. We therefore speculated that HOTAIR regulates H3K27 trimethylationin the promoter of miR-200a-3p to mediate miR-200a-3p expression by recruiting EZH2, representing a novel compensation mechanism between HOTAIR and miRNAs. Our results also showed that increased miR-200a-3p expression signi cantly reduced the migration and invasion of HeLa cells. These effects were enhanced after inhibiting miR-200a-3p. Furthermore, YAP1 was down-regulated by miR-200a-3p targeting its 3′UTR, con rming it as a target gene of miR-200a-3p in HeLa cells. Together, these data suggest that HOTAIR exerts its effects on the migration and invasion of cancer cells, at least in part, through the regulation of YAP1 by inhibiting miR-200a-3p in HeLa cells.
Ribosomal proteins (RPs) are components of ribosomal subunits that are ubiquitous RNAbinding proteins carrying out multiple auxiliary extraribosomal functions, and are moderately related to tumorigenesis (74,75). When cancer cells undergo uncontrolled growth and proliferation, they increase the biosynthesis of ribosomal biogenesis and the production of RPs. Growing evidence suggests the existence of an association between RPL23 expression and tumor invasiveness and aggressiveness (64,65,76). Indeed, RPL23 was reported to be up-regulated in human prostate cancer, lung cancer, and hepatocellular carcinoma (66,77,78). RPL23 promoted the invasionof the metastatic lung adenocarcinoma cell line Anip973 (66). However, the precise role of RPL23 in various fundamental processes remains elusive, and its tissue speci city may be more of a determinant o ts functions in the stressresponse (74). It is thus of great interest that we show that the role of YAP1 as an upstream transcriptional regulator of RPL23, at least in part, are mediated by the regulation of RPL23 expression in HeLa cells. The p53 mutation, which loses tumor suppressive functions and gains tumor-promoting activities (79), was reported to be associated with aggressive growth and increased recurrence rates for certain tumors (80). Here, we showed that RPL23 increased the expression of mutant p53 in HeLa cells. We therefore speculate that RPL23 enhances the migration and invasiveness of HeLa cells through promoting mutant p53 expression.

Conclusion
we propose a novel model depicting the precise molecular mechanism of HOTAIR in regulating the migration and invasion of cancer cells.
We suggest that HOTAIR, as an oncogenic lncRNA, recruits EZH2 to reduce miR-200a-3p expression via H3k27 trimethylation in the miR-200a-3p promoter. As a target gene of miR-200a-3p,YAP1 thenpromotes the migration and invasion of HeLa cells by mediating the downstream transcriptionof RPL23 which normally functions as a cancer-promoting factor. So we speculated that HOTAIR exerts its effects on the migration and invasion of cancer cells via the miR-200a/YAP1/RPL23/ signaling axis. Moreover, YAP1 may be a potential therapeutic target for the treatment of metastatic cervical cancer.

Availability of supporting data
All data generated during this study are included in this published article and its Additional les.