Sage Journals: Discover world-class research

Abstract

BACKGROUND:

Long-non-coding RNAs, a class of transcripts with lengths $>$ 200 nt, play key roles in tumour progression. Previous reports revealed that LINC00052 (long intergenic non-coding RNA 00052) was strongly downregulated during breast cancer multicellular spheroids formation and suggested a role in cell migration and oxidative metabolism.

OBJECTIVE:

To examine the function of LINC00052 in MCF-7 breast cancer cells.

METHODS:

Loss-of-function studies were performed to evaluate LINC00052 role on MCF-7 breast cancer cells. Microarray expression assays were performed to determine genes and cellular functions modified after LINC00052 knockdown. Next, the impact of LINC00052 depletion on MCF-7 cell respiration and migration was evaluated.

RESULTS:

1,081 genes were differentially expressed upon LINC00052 inhibition. Gene set enrichment analysis, Gene Ontology and Key Pathway Advisor analysis showed that signalling networks related to cell migration and oxidative phosphorylation were enriched. However, whereas LINC00052 knockdown in MCF-7 cells revealed marginal difference in oxygen consumption rates when compared with control cells, LINC00052 inhibition enhanced cell migration in vitro and in vivo, as observed using a Zebrafish embryo xenotransplant model.

CONCLUSION:

Our data show that LINC00052 modulates MCF-7 cell migration. Genome-wide microarray experiments suggest that cancer cell migration is affected by LINC00052 through cytoskeleton modulation and Notch/ $\beta$ -catenin/NF- $\kappa$ B signalling pathways.

Keywords

LINC00052 cell migration transcriptome analysis breast cancer

List of abbreviations

dpf	days post-fertilization
ECM	Extracellular Matrix

EMT	Epithelial-Mesenchymal Transition
GSEA	Gene Set Enrichment Analysis
INMEGEN	Instituto Nacional de Medicina Genómica
KPA	Key Pathway Advisor
lnc	long non-coding
LINC00052	Long intergenic non-coding RNA-00052
MCTS	Multicellular Tumour Spheroid
OxPhos	Oxidative Phosphorylation
RT-qPCR	Real Time quantitative PCR

shRNA	short hairpin
TAC	Transcriptome Analysis Console
TCGA	The Cancer Genome Atlas
TNBC	Triple Negative Breast Cancer

1. Introduction

Breast cancer is a heterogeneous group of diseases that represent the first cause of cancer-related death among women worldwide. Currently, breast cancer is classified on different subtypes based on the presence of oestrogen and, progesterone receptors and human epidermal growth factor receptor 2 [1]. Although great advances have been made in breast cancer research, patient mortality has remained high and still merits attention. Therefore, it is crucial to find novel relevant breast cancer regulators, among which the knowledge about epigenetic regulators has elicited high interest [2].

Genome-wide sequencing technology has shown that only 2% of the human genome encodes for proteins while the remaining genome includes a broad diversity of non-coding genes [3]. The latter have been linked to key cellular processes that are often altered in cancer. In particular, current research has focused in long ( $>$ 200 nt) noncoding RNAs (lncRNA) [4, 5]. lncRNA regulate gene expression through diverse transcriptional and posttranscriptional strategies, such as modulators of chromatin remodelling complexes, and acting as scaffolds, decoys and as signalling platform components [2, 6]. Interestingly, the expression of lncRNA is more tightly linked to specific tissues and development stages than their coding counterparts [7, 8].

lncRNA have been directly implicated in breast carcinogenesis, where they could play an important role in subgrouping breast cancer patients and are associated with known clinical phenotypes. Indeed, a recent study identified a group of 4 lncRNA that are specifically modulated in TNBC (triple-negative breast cancer) [9]. In addition, lncRNA differential expression in both tissue samples and cell lines has been associated with hierarchical groups related to subclassification system PAM50 [10]. Moreover, a differential co-expression between mRNA and lncRNA was observed when normal and tumoral breast tissues were compared [11]. In this sense, particular lncRNA have been shown to play a key role in MCF-7 breast cancer cells proliferation and in cell cycle-related gene expression upon oestrogen treatment [12]. In addition, lncRNA have been considered as predictors of breast cancer metastasis and recurrence risk. Accordingly, lncRNA-ROR has been related to Epithelial-Mesenchymal Transition (EMT) in breast cancer tissues [13]. NKILA, on the other hand, was identified as a lncRNA that suppresses breast cancer cell metastasis through regulation of the NF- $\kappa$ B pathway [14]. Even though several lncRNA have been implicated in breast cancer, for most of them precise molecular mechanisms of action and functional roles remain undescribed.

Previously, we evaluated the transcriptome of MCF-7 breast cancer cells in MCTS (Multicellular Tumoral Spheroids), a 3D model that mimics the tumour behaviour of cancer cells in vivo [15, 16]. Signalling pathways analysis identified gene signatures related to metastasis, oxidative mitochondrial metabolism and cell cycle. In these experiments, we found that LINC00052, an intergenic lncRNA, was strongly downregulated during MCTS growth. Interestingly, this lncRNA has also been associated to cancer cell proliferation, differentiation, and migration [17, 18, 19, 20, 21, 22]. LINC00052 promotes gastric and head and neck squamous cell carcinoma; however, it acts as a tumour suppressor in hepatocarcinoma, glioma, pancreatic as well as in colorectal cancer. Nevertheless, the role of LINC00052 in breast cancer still remains poorly understood. Salameh and colleagues have shown that HER3 expression correlates with LINC00052 levels and that when BC cells overexpress HER3, LINC00052 promotes growth and survival [17]. Though, LINC00052 is included within a lncRNA-non-TNBC cancer signature proposed by Lv et al. [9]. In the present study, microarray expression analyses showed that cell migration and mitochondrial metabolism-related pathways were enriched after LINC00052 knockdown. Consistently, we evaluated the function of LINC00052 in MCF-7 cells and found that its inhibition enhances cell migration in vitro and in vivo, and marginally affects mitochondrial metabolism. Our results suggest that LINC00052 regulates cellular processes that may allow tumour progression and provide further insight into the function of lncRNA in breast cancer.

2. Materials and methods

2.1 Analysis in silico of LINC00052 expression in breast cancer tissues

The Km-Plotter platform was used to analyse patient outcomes indicating LINC00052 best cut-off expression levels, using mRNA gene chip databases for breast cancer samples [23].

2.2 Cell culture

The MCF-7 cell line was obtained by ATCC (Cat. No. HTB-22) and was genotyped at INMEGEN high-tech sequencing unit. Cells were grown at 37 ${}^{\circ}$ C, in 5% CO ${}_{2}$ and supplemented with 5% FBS in RPMI medium (Cat. No. 10-040 CMR 1L Corning, USA).

2.3 Multicellular tumour spheroid generation

MCTS were formed as previously described [16, 24]. Briefly, cells were detached once they reached 80% confluence, washed with PBS 1X and separated with trypsin (Cat. No. 25200-056 Gibco) in PBS 1X EDTA 0.05 mM. Pelleted cells were resuspended in 3 mL of L-15 medium supplemented with 5% FBS and were counted in Neubauer chambers. One million cells were seeded in 12.5 cm ${}^{2}$ -low-adhesion bottles (Cat. No. 10-045 CVR 500 ML Corning, USA) in 4 mL L-15 medium-5% FBS and immediately placed in an incubator under stirring (58 rpm) at 37 ${}^{\circ}$ C without CO ${}_{2}$ . During MCTS growth, fresh medium was replaced each three days.

2.4 shRNA design and stable cell line establishment

Two shRNAs for LINC00052 (lnc-AGBL1-4:1) were designed using the E-RNAi software [25]. The sequences used were: A) 5 ${}^{\prime}$ -CCTGGAAAGCTGTCATCTT-3 ${}^{\prime}$ , and B) 5 ${}^{\prime}$ -CTCTCATCCTTCCCATGTT-3 ${}^{\prime}$ . The aligned oligonucleotides were cloned in vector pSIREN-RetroQ (Clontech, CA, USA) between EcoR1 and XbaI restriction sites. shLuc-pSIREN-RetroQ construct was employed as control. MCF-7 cells transfection was performed using Lipofectamine 2000 reagent (Cat No. 11668019 Thermofisher Scientific) as indicated by the manufacturer. Cells were selected with puromycin (Cat. No. P96-20-10ML Sigma-Aldrich) at 0.6 $\mu$ g/mL for 14 days to generate stable cell lines. Inhibition was verified by RT-qPCR with specific primers (see following sections).

2.5 Total RNA extraction and reverse transcription (RT) reaction

Total RNA was extracted from shLuc, shA and shB MCF-7 cells grown as monolayers using TriReagent (Cat. No. AM9738 Ambion), as suggested by the provider. Isolated RNA was resuspended in 40 $\mu$ L nuclease-free H ${}_{2}$ O. After DNAse treatment (Cat. No. M610A Promega), cDNA synthesis was performed using 1.5 $\mu$ g with High-Capacity Kit (Cat. No. 4368814 Thermofisher Scientific) as indicated by the supplier.

2.6 qPCR

Specific primers were designed to analyse the expression of TBP and LINC00052 genes (Supp. Table 1). qPCR was performed with SYBR ${}^{\text{\textregistered}}$ Select Master Mix (Cat No. 4472908 Applied Biosystems).

2.7 Transcriptome analysis by microarrays

HuGene-2_0-st-v1 Arrays from Affimetrix were used to analyse gene expression of shLuc and shB MCF-7 cells. RNA was extracted as previously described from 3 biological samples of each cell line [24]. RNA quality was evaluated with a Bioanalyzer from Illumina (RIN $=$ 9) at INMEGEN’s microarray facility. Microarray data were submitted to the Gene Expression Omnibus (GEO) database (accession number GSE125357).

2.8 Gene expression bioinformatic analyses

Microarray data (.cel files) were analysed with Transcriptome Analysis Console (TAC 4.0) software from Affimetrix (Fold change $\geqslant 2$ or $\leqslant-2$ , $p$ -value $<$ 0.05, FDR $<$ 0.05) and Limma protocol. The value of differentially expressed genes from all biological replicates was used to make heatmaps with the help of the Shinyheatmap tool available online [26]. A cluster between the rows and the columns was made, employing a Euclidean comparation and a single linkage algorithm.

Data intensity from differentially expressed genes (FC $2>$ / $<-2$ and $P$ value $<$ 0.05) were used to perform enrichment and network analysis, using Network Analyst 3.0 platform and Gene Ontology: Biological Processes (GO:BP) databases [27].

Additionally, differentially expressed genes data were explored with Key Pathway Advisor (KPA, Version 17.4) platform (http://ipscience.thomsonreuters.com/ product/metacore/.2016). Total expression matrix from TAC was used to perform a Gene Set Enrichment Analysis (GSEA) from Broad Institute online tool, employing gene-set as permutation type and default parameters [28].

Finally, to analyse miRNA and mRNA associations, fold change and $P$ value data, as well as their intensities, were used to generate interactions with the miRTarvis $+$ v.0.8.11 platform. GenMiR $++$ , TargetScan and microRNA.org tools were employed to generate predictions and Pearson correlation was used to build interaction nodes [29].

2.9 Microarray data validation by qPCR

Gene expression changes identified by microarray experiments were additionally evaluated by qPCR as previously described, employing specific primers (Supp. Table 1). For each gene, mRNA relative expression was calculated normalizing to TBP expression and fold change (FC) was estimated on reference to the shLuc MCF-7 cells found expression. Gene expression changes were assessed in three independent biological samples.

2.10 Oxygen consumption and determination of OxPhos-dependent flux

To measure the rate of oxygen consumption, shLuc, shA and shB-MCF-7 (80% monolayer confluence) cells were detached with 0.025 mM of trypsin PBS 1x and 0.05 mM EDTA for 5 minutes at 37 ${}^{\circ}$ C. Cells were pelleted, gently washed with fresh medium and centrifuged at 1,200 rpm for 5 minutes. Cells were suspended in 500 $\mu$ L Krebs-Ringer medium (125 mM NaCl, 5 mM KCl, 1 mM MgCl ${}_{2}$ , 1 mM KH ${}_{2}$ PO ${}_{4}$ , 25 mM HEPES, 1.4 mM CaCl ${}_{2}$ , pH $=$ 7.2). The rates of oxygen consumption were measured in a Strathkelvin Instruments 782 Oxygen Meter (North Lanarkshire, Scotland, UK) interfaced with a computer. Samples were placed in a water-jacketed chamber at 37 ${}^{\circ}$ C. Basal respiration rates were obtained by adding 1 $\times$ 10 ${}^{6}$ cells to 200 $\mu$ L of air-saturated Krebs-Ringer medium. 5 $\mu$ M oligomycin was added to inhibit ATP synthesis by mitochondrial complex V. In addition, 1 mM NaCN was added to fully inhibit mitochondrial oxygen consumption [30]. At least two measurements were performed for each cell line in three independent biological assays.

2.11 In vitro migration assay

shLuc, shA and shB MCF-7 cells were grown as monolayers, detached with Trypsin and counted. 30,000 cells of each cell clone were seeded in serum-free RPMI medium inside upper chambers of Transwell cell culture plates (Cat. No. 3422 Corning, USA). The lower chambers of plates were filled with RPMI medium with 15% FBS as a chemoattractant. After 48 h cell migration was assessed by staining the cells at filter bottom surfaces with 0.05% Crystal Violet. Five fields from each filter were imaged with a Leica microscope (10x) and a mean cell number was calculated for each condition. Three measurements were performed for all cell lines in all experiments. Assays were performed three independent times.

2.12 Gelatine degradation assay

shLuc, shA and shB MCF-7 cells were seeded on FITC-conjugated gelatine (Cat. No. G-13187 Molecular Probes) coated coverslips for 7 h. Cells were fixed and stained for F-actin to quantify gelatine degradation as previously described [24, 31]. Three independent experiments were performed.

2.13 Zebrafish husbandry

Zebrafish (Danio rerio) of the Tab Wik genetic background (donated by Dr. Ernesto Maldonado from the Institute of marine sciences and limnology-UNAM and by Dr. Francisco Carmona from the Instituto de Fisiología Celular-UNAM), were maintained at a temperature of 28.5 ${}^{\circ}$ C on a pH of 7.2–7.4 with a 14 hr-on and 10 hr-off light cycle. Embryos were raised in liquid Embryo Medium (EM).

Care and feeding of fish and larvae followed the standard procedures described in “The zebrafish book. A guide for the laboratory uses of zebrafish (Danio rerio)”. All experiments and procedures in animals are carried out following the guidelines of “THE INSTITUTIONAL ANIMAL CARE AND USE COMMITTE (IACUC) UNIVERSITY OF OREGON”, available for consultation in “The zebrafish book. A guide for the laboratory use of zebrafish (Danio rerio)” [32]. In turn, at the Instituto Nacional de Medicina Genómica, all procedures in zebrafish followed the guidelines established by the local committee on ethics and management of laboratory animals (CICUAL, Comite Interno para el Cuidado y Uso de Animales de Laboratorio; Approved Verdict No. 427). Zebrafish embryo and larvae were euthanized with 0.4% tricaine (MS-222; Sigma) for 1 h (until heart beating stops).

Figure 1.

Regulation of LINC00052 expression. LINC00052 expression was analysed by KMplots bioinformatic tool. (A) Graph shows relapse-free survival and (B) graph depicts distant metastasis-free survival, regarding LINC00052 expression. (C) LINC00052 expression decreases early during MTCS formation. MCF-7 cells were grown as a monolayer (M) and as multicellular aggregates during 4 h, 2 and 4 days. LINC00052 expression was evaluated in these growth conditions by RT-qPCR. Graph shows LINC00052 average fold change in three independent experiments. (D) LINC00052 expression knockdown. MCF-7 cells were stably transfected with 2 independent shRNA, shA and shB as well as with a control shRNA, shLuc. LINC00052 expression was assessed in these cell lines by RT-qPCR in 3 independent samples. Statistical difference was calculated by a One-way ANOVA test (PRISMA Graph software). Asterisks indicate a $p$ -value of $<$ 0.05. Bars show standard errors.

2.14 In vivo migration assays in a xenograft zebrafish model

Two days post-fertilization (dpf) zebrafish embryos were dechorionated and anesthetized with 0.04% tricaine, and then microinjected with MCF7 cancer cells into the middle of the embryonic yolk sac region. For the subsequent detection in zebrafish body, either MCF7-shLuc, MCF7-shA or MCF7-shB cells were transiently transfected with pSiren-EGFP vector 48 hours prior to injection. On the day of the xenografts, cells were re-suspended at a concentration of 10 ${}^{6}$ cells/50 $\mu$ l and about 300 cells were microinjected into the yolk of Tab-wik zebrafish embryos. Microinjections were carried out loading the cells into a borosilicate glass needle (Sutter Instrument) using a vertical pipette puller (Model 51210, Stoelting). The cells were injected using a Pneumatic Pico-Liter Injector (PLI-100A; Warner Instruments). After injection, each zebrafish was washed once with EM egg water, and incubated at 34 ${}^{\circ}$ C. Only embryos with a fluorescent cell mass were considered for each experiment. For cell migration observation, 24 hours post injection embryos were analysed and photographed using an epifluorescence microscope (Zeiss) equipped with a Cannon camera. Images were analysed with FIJI-ImageJ and the number of fluorescent cells in both posterior trunk and tail region (outside the yolk sac) were counted in each embryo. Five independent experiments were performed.

2.15 Statistical analysis

All experiments were performed at least by triplicate and included at least two technical replicates. One-way ANOVA test was employed in all assays for independent groups to compare mean values ( $\pm$ standard error, SE) to assess statistical significance with Prism Graph V5.0. For Oxygen consumption and determination of OxPhos-dependent flux a Two-way ANOVA test was used (first factor evaluated differences between experiments and second factor determined the effect of LINC00052 inhibition).

Figure 2.

LINC00052 knockdown affects MCF-7 cells gene expression. (A) Gene expression analysis was done in shLuc-MCF-7 and shB-MCF-7 cells using Affymetrix chip HuGene-2_0-st-v1. Graph shows the number of differentially regulated transcripts in shB-MCF-7 cells compared to control shLuc-MCF-7 cells. (B) Number of LINC00052 modulated coding and non-coding transcripts. (C) A heatmap was generated using a Euclidean correlation method and shows differentially expressed genes in all biological samples.

3. Results

3.1 LINC00052 expression decreases in breast tumour from patients with worse outcome

LINC00052 is an intergenic lncRNA located at chromosome 15, between the AGBL1 and NTRK3 genes and intronic-antisense to lnc-NTRK3. LINC00052 was selected for further investigation based on its low expression profile during MCTS development, as previously reported [15, 16]. To determine the clinical relevance of LINC00052, we queried breast cancer patient samples data using the Kaplan-Meier plotter (KM-plotter) web-tool [23]. KM-plotter analysis revealed a worse prognosis for patients with low LINC00052 expression, showing a decreased relapse-free and distant metastasis-free survival (Fig. 1A and B).

3.2 LINC00052 expression is downregulated upon MCTS formation

To further explore LINC00052 regulation, we analysed its transcript levels by RT-qPCR in MCF-7 cells growing as a monolayer (M) and during the early phases of MCTS formation, at 4 h, 2 and 4 days. Interestingly, we found that its expression was maintained 4 h after cell detachment (Fig. 1C). However, after two days of MCTS growth, when cell-cell interactions began to be established, LINC00052 expression was significantly downregulated (FC-15 and $p$ -value $<$ 0.0050). Four days later, its expression was further decreased in comparison to MCF-7 cells growing in monolayer (FC-53.80 and $p$ -value $<$ 0.0009).

Our previous research demonstrated higher migration capacity and increased respiration rates when MCF-7 cells were grown as MCTS since early stages (6 days) [15, 16]. As previously mentioned, LINC00052 expression was also downregulated, suggesting that this lncRNA could facilitate a more aggressive breast cancer cells phenotype. To assess these possibilities and to mimic LINC00052 downregulation in breast cancer cells as in MCTS, we utilized two specific shRNA targeting LINC00052 to generate stably transfected MCF-7 cell clones, hereafter named shA and shB. These cell clones displayed significantly lower levels of LINC00052 expression in comparison with control MCF-7 cells (control shRNA, shLuc) when cells are grown as monolayers (Fig. 1D).

3.3 Gene expression changes in LINC00052-knockdown MCF-7 cells

We performed microarray expression analyses (Affimetrix chip HuGene-2_0-st-v1) to gain further insight regarding the genes and signalling pathways modulated by LINC00052 in MCF-7 cells. We found 1,081 differentially regulated genes (Fold change $\geqslant 2$ or $\leqslant-2$ , gene-level $p<$ 0.05), from which 738 were upregulated and 343 were downregulated (Fig. 2A). We found that 1,000 of these were coding genes whereas 81 were classified as non-coding genes (21 miRNAs and 60 lncRNAs, Fig. 2B and S1A Fig). A global gene expression heatmap, describing the transcriptome changes after LINC00052 knockdown, was generated using the Shiny heatmap tool (Fig. 2C).

Figure 3.

Enrichment network analysis upon LINC00052 inhibition. Enrichment nodes were generated with GO:BP database and obtained with significant $P$ value $<$ 0.05 and FDR $<$ 0.05. (A) Diagram of microtubules-based movement node (B) Diagram of cytoskeleton organization node, and (C) Heatmap shows the expression intensities of genes from (A) and (B) enriched nodes. Colour code shows the molecules with a positive FC $>$ 2 in red and in green are the molecules with a FC $<-2$ .

We observed that the top up- and downregulated coding genes were associated with known cellular processes modulated in cancer cells (Table 1). Among these, the highest upregulated gene was FOLH1 (FC 16.790) whereas the lowest was SERPINA3 (FC-13.010). Gene expression changes were confirmed by RT-qPCR in shLuc, shA and shB MCF-7 clones for 5 of the top modulated genes (Table 2). We observed that the sense of expression change was the same as in microarrays.

Table 1

Differentially expressed coding genes

Top 10 expressed molecules
Up-regulated molecules	Fold change	Down-regulated molecules	Fold change
FOLH1	16.790	SERPINA3	$-$ 13.010
FAM111B	9.160	IGFBP5	$-$ 12.740
CDC6	6.530	CP	$-$ 12.570
GINS2	5.730	S100A7	$-$ 12.440
PLEKHS1	5.130	CEACAM5	$-$ 9.920
PEG10	5.120	mir-23c	$-$ 9.890
HIST1H1B	4.830	CFB	$-$ 8.460
NCAPG	4.650	CD68	$-$ 6.570
DTL	4.540	FOS	$-$ 6.390
FAM72D	4.500	PIP	$-$ 6.330

Table 2

Validation of gene expression changes

Gene	FC microarray	FC qPCR shA	FC qPCR shB
FOLH1	16.79	6.081	4.983
PEG10	5.12	2.5	3.3
SERPINA3	$-$ 13.01	$-$ 4.94	$-$ 2.36
S100A7	$-$ 12.44	$-$ 10.65	$-$ 4.7
IGFBP5	$-$ 12.74	$-$ 3.2	$-$ 5.39
LINC00052	$-$ 3.8	$-$ 22.4	$-$ 10.5

Gene expression changes (mRNA accumulation) were analysed by RT-qPCR using specific primers for each gene in shLuc, shA and shB MCF-7 cell lines. Values are expressed as Fold Change (FC) and TBP gene expression was employed as an internal expression control. Gene expression changes were statistically significant ( $p$ -value $<$ 0.05, One-way ANOVA test) in three independent experiments.

Table 3

“Cytoskeleton organization” node molecules modulated by LINC00052 knockdown

Molecular entity	Molecular function in breast cancer cell migration	FC	Reference
ANG	Ang II/AT1R/PI3K/Akt/NF-kB signaling	$-$ 3.061680	[55]
ANL	Cytoskeletal scaffolding protein, stemness and EMT	2.923748	[58]
ARHGAP21	Associates with PK1, inhibits Rho-A, alters focal adhesions	2.243166	[59]
CKS2	Correlates with poor prognosis of survival, invasion and metastasis	2.307450	[60]
CLDN1	Tight junction protein, collective migration, EMT, PKC signalling	2.072788	[61, 62]
ECT2	Rho GEF factor, expression and prognostic significance	2.594068	[63]
EDN1	Cell-cell interactions, chemotaxis, invasiveness	2.800048	[64]
EHD2	Tumour suppressor, Endocytic trafficking, migration and invasion through Rac-1 and EMT	$-$ 3.049510	[65, 66]
EPBL41L2	Assembly/stability of protein complexes in plasma membranes, tumour suppressor	$-$ 2.089851	[67]
MARK1 (Par-1)	G-protein CR, integrin and cytoskeletal interactions, chemotaxis and invasion	2.116690	[68]
MYBL2 (B-Myb)	Regulates cell-cell junctions, modulates E-cadherin and EMT	3.558057	[69]
PLK1/4	Polo kinase, cytoskeletal organization and cell polarity, ECT2, RhoA and DIAPH1 interactions	2.668497	[70]
PREX1	Rac-GEF, Actin cytoskeleton, cell migration, ERB and TRK receptors	3.675649	[71, 72]
SKA2	SKA complex, EMT, migration and invasion, Microtubules and E-cadherin organization	2.639506	[73]
SPAG5	Promotes Wnt3 and $\beta$ -catenin expression; activates $\beta$ -catenin/TCF4	2.907918	[74]
TACC3	Reduces PLK1 expression, activates Aurora A kinase of PI3K/Akt and ERK signalling pathways	2.806607	[75, 76]
TPX2	Microtubule-associated protein, migration and invasion, PI3K/AKT pathway	2.224906	[77, 78]

Table 4

MCF-7 cells oxygen consumption rates

	shLuc	shA	shB
Basal respiration (B.R)	3.94 $\pm$ 1.30	8.15 $\pm$ 4.8	5.93 $\pm$ 0.91
$+$ Oligomycin	1.98 $\pm$ 0.56	3.65 $\pm$ 2.3	2.10 $\pm$ 0.41
OxPhos-dependent flux	1.96 $\pm$ 0.50	4.50 $\pm$ 2.5	3.82 $\pm$ 1.16

Basal respiration rates and OxPhos-dependent flux was evaluated in shLuc, shA and shB MCF-7 cells grown as monolayers. Values are shown in natgO/min $\times$ 10 ${}^{6}$ cells. Data from 3 independent experiments are expressed as the mean $\pm$ SE. Two-way ANOVA showed that cell respiration was marginally significantly different ( $p$ -value $=$ 0.07). Fisher pos-hoc analysis indicated statistically significant differences between shLuc and shA ( $p$ -value $=$ 0.035) and marginally significant between shLuc and shB ( $p$ -value $=$ 0.07). Basal respiration was determined in the absence of exogenous electron donors or carbon sources. OxPhos $=$ Basal respiration rate minus rate of oxygen consumption in the presence of oligomycin.

3.4 Functional and pathway enrichment analysis

The microarray data was first explored using the Gene Set Enrichment Analysis (GSEA) tool [28], looking for enriched cellular processes. The KEGG platform evaluation demonstrated the positive enrichment of oxidative phosphorylation and mitochondrial protein complexes in shB MCF-7 cells ( $p<$ 0.05; FDR $<$ 0.05) (S2A Fig).

3.5 Gene Ontology microarray data analysis

In parallel, a network enrichment analysis was performed with Gene Ontology (BP: Biological Processes) database [33]. On the modelling of microarray data this analysis showed that two of the main significant nodes were related to “Microtubules-based movement” (Fig. 3A) and “Cytoskeleton organization” (Fig. 3B). Among node components we observed key molecules for cell motion that were positively enriched in both processes. As well, and in consistence with our functional assays, these molecules participate in breast cancer cell motility and include: Receptors/ligands (ANG, MARK1); Scaffolding proteins (ANL, EBPL41L2, SKA2), RHO-GAP/GEF (ARHGAP1, ECT2, PREX1), Junction components (CLDN1, END1, MYBL2), Trafficking protein (EHD2) and Kinases (CKS2, PLK1 and 4) (Table 3). Moreover, this analysis showed that 11 Kinesins, key for breast cancer pathogenesis were modulated by LINC00052 inhibition (Table S2). Gene expression changes related to these nodes are more specifically depicted by a heatmap, generated by the Shiny heatmap tool (Fig. 3C).

An additional microarray data analysis was performed with the Key Pathway Advisor (KPA) tool. KPA predicted 12 significantly modified signalling pathways related to cell migration based on the observed differentially expressed genes (Table S3). Among them, “Signal transduction_Angiotensin II/AGTR1 signalling via Notch, $\beta$ -catenin and NF- $\kappa$ B pathways” (S2B Fig), “Canonical Notch in CRC”, “Developmental TGF- $\beta$ 1”, “Inhibition of TGF- $\beta$ 1 in early Colorectal cancer”, and “Tumour-stroma interactions in Pancreatic cancer” signalling pathways were forecasted.

Finally, we hypothesized that LINC00052 could influence breast cancer cells behaviour by miRNA regulation (S1A Fig). In this sense, the analysis regarding the observed set of miRNAs and mRNAs showed that hsa-miR-130b, which targets the DIAPH3 mRNA, was downregulated. In parallel, hsa-miR-21, a microRNA that binds to FOXM1 mRNA, was also inhibited (S1B Fig).

3.6 Effects of LINC00052 downregulation on cell respiration rates

Microarray data analysis suggested that LINC00052 downregulation triggered gene expression changes associated with mitochondrial functions. This result is in agreement with our previous findings where we observed that mitochondrial respiration increased rates were associated with more aggressive MCTS characteristics, together with LINC00052 downregulation [15, 16]. Afterwards, we analysed the oligomycin sensitive respiration rate in shA and shB MCF-7 cells, which represents mitochondrial ATP synthesis by complex V coupled with the respiratory chain activity [34]. We found a small increment in shA and shB clones’ basal respiration when compared to control shLuc MCF-7 cells (Table 4). The OxPhos-dependent flux, determined by exposing cells to oligomycin (a specific F1 FO-ATP synthase inhibitor), revealed that shA and shB cells had a higher ATP synthesis-coupled respiration rate when compared to control shLuc cells (Fisher poshoc analysis indicated statistically significant differences between shLuc and shA ( $p$ -value $=$ 0.035) and marginally significant between shLuc and shB ( $p$ -value $=$ 0.07). These changes were consistent throughout all biological and technical replicates.

3.7 LINC00052 knockdown results in greater MCF-7 cell migration capacity in vitro and in vivo

Enhanced cell migration capacity was observed previously in cells of MCTS model [16]. In agreement, transcriptome changes triggered upon LINC00052 knockdown were associated with cell migration. Therefore, we tested whether LINC00052 inhibition could change this phenomenon. Using in vitro Boyden chamber assays, we observed a significant migration increment in shA and shB cells in comparison to shLuc control cells after 48 h of culture (Fig. 4A). In parallel, we analysed the ability of these cells to degrade the extracellular matrix (ECM) through a fluorescent-gelatine assay. We observed that both shA and shB cells had a significantly greater capacity to degrade ECM than shLuc control cells (Fig. 4B). Finally, we examined the participation of LINC00052 in cell migration in vivo using a zebrafish model (2-day old embryos), comparing the migration rate of fluorescent shA, shB, and shLuc MCF7 cells to the posterior trunk and tail region (Fig. 4C). We counted the number of cells inside the tail area of all embryos [35], finding a significantly higher number of migrating cells in the embryos injected with shA and shB MCF-7 clones than in those inoculated with shLuc MCF-7 control cells (Fig. 4D). Of note, we examined whether LINC00052 inhibition affected MCTS formation in shA, shB cell lines in comparison with control shLuc and we did not observe differences between the three cell lines (Fig. 4E).

Figure 4.

LINC00052 knockdown modulates MCF-7 cells migration. (A) shLuc, shA and shB MCF-7 cells in vitro migration capacity. Cells were allowed to migrate in commercial migration chambers for 24 h as indicated formerly. Graph displays cell average per field assessed in three independent experiments. (B) FITC-gelatine degradation capacity of shLuc, shA and shB MCF-7 cells. Cells were grown on top of FITC-gelatine during 7 h to assess their extracellular matrix degradation ability as previously described. Chart shows the average cell degradation area per field evaluated in three independent experiments. In (A) and (B) Statistical difference was calculated by a One-way ANOVA test (PRISMA Graph software). Asterisks indicate a $p$ -value of $<$ 0.05. Error bars shown standard error value. (C) and (D) In vivo Zebrafish embryo cell migration assay. (C) Pictures show representative 2 day-embryos after 24 h injection (hpi) of shLuc, shA and shB MCF-7 cells (300 each). Embryos are shown as bright field images (Left) and epifluorescence images (Right). Scale bars $=$ 100 $\mu$ m. (D) Graph depicts fluorescent cells outside the yolk sac from all embryos. Error bars show standard error means, asterisks indicate a $p$ -value of $<$ 0.05. (E) MCTS formation assay. Graph displays shLuc, shA and shB MCF-7 cells aggregates average diameter ( $\mu$ m) in three independent experiments. Error bars shown standard error value. Statistical difference was calculated by a One-way ANOVA test (PRISMA Graph software); no difference was observed.

Altogether these findings suggest that downregulation of LINC00052 expression is associated with enhanced cell migration and invasion abilities and thus, to a more aggressive phenotype in MCF-7 cells.

4. Discussion

The present study provides evidence that LINC00052 plays a regulatory role in altered cellular processes commonly observed in breast cancer. Previous work from our laboratory showed a transcriptome change when MCF-7 cells were grown as MCTS in comparison with a monolayer counterpart. Further, we observed an expression shift not only in coding genes but also in non-coding ones, which were subsequently reflected by the hyperactivation of cellular processes such as metabolism (oxidative phosphorylation and cholesterol signalling), cell cycle, and migration.

Interestingly, among the most inhibited molecules in MCTS we found a non-coding RNA, LINC00052 [15, 16]. These findings led us to evaluate whether LINC00052 inhibition triggers some of breast cancer cells characteristics. First, when we analysed LINC00052 expression in MCF-7 breast cancer cells at early time points of MCTS formation, we found that its expression becomes downregulated after 48 and 96 hours under non-hypoxic conditions and when the establishment of cell-cell junctions is a main process. This data suggests that the downregulation of LINC00052 could be associated with cell-cell interactions during the free substratum growth of 3D cultures.

In agreement, the analysis of a breast cancer patient database revealed that relapse-free survival and distant metastasis-free survival rates are poorer in breast cancer patients with lower LINC00052 expression. This observation is consistent with a previous report showing that LINC00052 expression was lower in basal breast cancer samples and cell lines compared with a non-basal counterpart [9] and support the idea that LINC00052 could act as a tumour suppressor during breast cancer development. Even though, further patient data analysis would be required to support this hypothesis. In this regard, the work of Salameh and colleagues also showed that LINC00052 together with HER3 mRNA expression is increased in luminal BC samples in comparison to normal tissue [17], supporting the idea that LINC00052 delicate regulation acts during breast cancer development and progression.

Nonetheless these findings [9], the function of LINC00052 in breast cancer cells remains poorly understood. In this sense, Xiong et al. [19] showed that the overexpression of LINC00052 results in the inhibition of cell migration, invasion and proliferation in hepatocellular carcinoma SMMC7721 cells [19]. Furthermore, there is evidence suggesting that LINC00052 acts as sponge for miR-128, miR-485-3p and miR-101-3p miRNAs, which respectively regulate the NTRK3, EPB41L3 and SOX9 expression, thus modulating cell proliferation and migration in several cancers including breast cancer [20, 21]. Consistent with these findings, we observed that some key miRNA-mRNA pairs were modulated in MCF-7 cells after LINC00052 knockdown. It is possible then to suggest that LINC00052 could regulate breast cancer cell migration and invasion, by working as a miRNA sponge, as has been the case for several other lncRNA.

The microarray transcriptome analysis showed that LINC00052 regulated the expression of 1,081 genes. In agreement with our previous work, gene enrichment analysis of our microarray data suggested that LINC00052 knockdown affects oxidative phosphorylation pathways [15, 16]. In this study, we observed a minor increase in the aerobic metabolism after LINC00052 knockdown in MCF-7 cells as we observed acceleration of both the basal respiration and the oligomycin-resistant respiration. However, we currently do not know if these results are associated with higher efficiency of mitochondria in these cells as we did not observe any response to the uncoupling agent CCCP (data not shown), as has been reported for other cancer cell lines [36]. It remains to be answered if LINC00052 knockdown promotes an increase in the mitochondrial number and/or an increase in the concentration/activity of mitochondrial respiratory chain complexes.

Afterwards, we validated the expression changes of five of the top regulated genes that are known to have a role in breast cancer development and cell migration. Among them, we observed FOLH1 as the highest upregulated gene. FOLH1, also known as prostate-specific membrane antigen, modulates breast cancer progression and metastasis by PI3K signalling [37]. Another upregulated gene by LINC00052 inhibition was PEG10 (Paternally expressed gene 10). Overexpression of this gene correlates with decreased survival of breast cancer patients. Moreover, in breast cancer cell lines PEG10 overexpression promotes cell proliferation, migration and metastasis through ERK/MMP-7 signalling [38, 39]. Additionally, we validated a reduced S100A7 (Psoriasin) expression when LINC00052 is inhibited. Interestingly, when this calcium-binding protein is overexpressed by MCF-7 cells, Rac-1 pathway and MMP-9 secretion are inhibited, thereby cell migration and invasion are repressed [40]. We also confirmed the IGFBP5 downregulation upon LINC00052 knockdown. In agreement with our results, previous work has shown that IGFBP5 induces MCF-7 cells adhesion while it inhibits cell migration, by its interaction with $\alpha$ 2 $\beta$ 1 integrins and EMT modulation [41, 42]. Finally, we proved inhibition of SERPINA3 expression by qPCR. Intriguingly, this molecule is also implicated in cell migration, invasion and metastasis of several tumour types, including breast cancer [43, 44].

Moreover, in vitro and in vivo assays allowed us to determine that the major changes triggered by LINC00052 inhibition were associated with enhanced cell migration and invasion. Cell migration has been widely studied in zebrafish embryos, since cancer cells preserve their phenotype within the zebrafish [45, 46]. Remarkably, we observed that apparent cell groups moved together within zebrafish embryos. These observations are consistent with the fact that cancer cells commonly migrate as clusters from the primary tumour to reach other tissues [47]. Interestingly, LINC00052 inhibition induced the expression of genes linked to cell dissemination pathways, which was reflected by the increased cell migration capacity in the zebrafish embryo xenograft assay.

Consistent with our findings of a role of LINC00052 in cell migration, gene enrichment analysis showed a “Cytoskeleton organization” node that contains molecules directly related to metastasis. In agreement, within the Microtubule-based movement node, we observed that a group of these genes belong to the Kinesin superfamily (KIF). Kinesins role is mainly related to intracellular transport (vesicles and organelles as well as chromosomes); therefore, and as expected, alteration of Kinesins contributes to cell transformation, including invasive and metastatic behaviour [48, 49]. Particular or collective functions of these LINC00052-modulated genes associate with breast cancer progression and cell migration.

In agreement with these observations, integral bioinformatic analysis inferred signalling pathways that modulate cell migration. First, we observed the route related to PTHR1 and Angiotensin II, “Pro- oncogenic action of PTHrP in breast cancer”. Remarkably, PTHrP signalling pathway participates in breast cancer cell adhesion, migration and metastasis [50, 51]. In particular, we observed the upregulation of heterotrimeric G proteins, G-alpha after LINC00052 knockdown. Through this signalling pathway we could hypothesize that the observed increase in PLK1/4 expression could lead to its activation and the enrichment of cell protrusion, that phosphorylates and activates RacGAP1/ECT2 GEF [52], which is also upregulated by LINC00052 knockdown. GEF proteins lead to RhoA activation and thus trigger several signalling pathways that may promote cell migration [53, 54]. For instance, RhoA activates ROCK kinase and Diaphanous formins, from which DIAPH1 and DIAPH3 are upregulated after LINC00052 inhibition.

Interestingly, we noticed that downstream Angiotensin II, another signalling pathway, “Signal transduction_Angiotensin II/AGTR1 signaling via Notch, $\beta$ -catenin and NF- $\kappa$ B pathways”, is activated. Remarkably, Angiotensin II treated breast cancer cells induce PI3K/Akt phosphorylation and increased NF- $\kappa$ B activity, thus supporting cell migration [55]. On the other hand, NOTCH signalling is dysregulated in breast cancer and crosstalk to other major pathways such as NF- $\kappa$ B [56, 57]. In this regard, we observed that LINC00052 inhibition triggers the expression of NOTCH and NF- $\kappa$ B signalling pathway components. Therefore, it is possible that, in collaboration with aforementioned “Cytoskeleton organization” and “Microtubule-based movement” routes, NOTCH-mTOR/AKT-NF- $\kappa$ B pathways may induce EMT and promote cell migration in LINC00052-inhibited MCF-7 cells.

In conclusion, the present study demonstrates that LINC00052 causes changes in the transcriptome of MCF-7 cells and inhibits cell migration. Although the precise LINC00052 action and molecular mechanisms await further investigation, these phenotypic changes are possibly associated with cytoskeleton modulation and signal transduction pathways NOTCH-mTOR/AKT-NF- $\kappa$ B.

Supplementary data

The supplementary files are available to download from http://dx.doi.org/10.3233/CBM-200337.

Footnotes

Acknowledgments

José Manuel Sánchez López thanks the Postgraduate program of Biological Sciences (Posgrado en Ciencias Biológicas, Doctorado en Ciencias Biológicas) and Universidad Nacional Autonoma de Mexico (Mexico’s Autonomous National University), UNAM, Faculty of Medicine. JMSL additionally thanks CONACYT national fellowship program, for support during his graduate training (CVU No. 777479). We acknowledge Microarray High Technology Unit from Instituto Nacional de Medicina Genomica. We thank Dr. Julio Perez-Carreon and Dr. Marcela Lizano Soberon for kind manuscript review and for support during JMSL graduate training. We are in debt with Dr. Edgar Krötzch and Dr. J. Perez-Carreon for allowing the use of microscopes.

References

Dai

Bai

Yang

Liu

Zhan

and Shi

, Breast cancer intrinsic subtype classification, clinical use and future trends, Am J Cancer Res 5 (2015), 2929–2943.

Gutschner

and Diederichs

, The hallmarks of cancer: A long non-coding RNA point of view, RNA Biol 9 (2012), 703–719.

Derrien

Johnson

Bussotti

Tanzer

Djebali

Tilgner

Guernec

Martin

Merkel

Knowles

D.G.

Lagarde

Veeravalli

Ruan

Lassmann

Carninci

Brown

J.B.

Lipovich

Gonzalez

J.M.

Thomas

Davis

C.A.

Shiekhattar

Gingeras

T.R.

Hubbard

T.J.

Notredame

Harrow

and Guigo

, The GENCODE v7 catalog of human long noncoding RNAs: Analysis of their gene structure, evolution, and expression, Genome Res 22 (2012), 1775–1789.

Gloss

B.S.

and Dinger

M.E.

, The specificity of long noncoding RNA expression, Biochim Biophys Acta 1859 (2016), 16–22.

Karlsson

and Baccarelli

A.A.

, Environmental health and long non-coding RNAs, Curr Environ Health Rep 3 (2016), 178–187.

Cao

, The functional role of long non-coding RNAs and epigenetics, Biol Proced Online 16 (2014), 11.

Gao

Y.F.

Wang

Z.B.

Zhu

Mao

C.X.

Mao

X.Y.

Yin

J.Y.

Zhou

H.H.

and Liu

Z.Q.

, A critical overview of long non-coding RNA in glioma etiology 2016: An update, Tumour Biol 37 (2016), 14403–14413.

Wang

Shao

Ding

Tan

Song

Wang

Jia

Ling

and Cheng

, Crosstalk between transforming growth factor-beta signaling pathway and long non-coding RNAs in cancer, Cancer Lett 370 (2016), 296–301.

Huang

Zeng

Shen

Jia

Yin

Yuan

Xie

and Fu

, LncRNAs as new biomarkers to differentiate triple negative breast cancer from non-triple negative breast cancer, Oncotarget 7 (2016), 13047–13059.

10.

Chia

S.K.

Bramwell

V.H.

Shepherd

L.E.

Jiang

Vickery

Mardis

Leung

Ung

Pritchard

K.I.

Parker

J.S.

Bernard

P.S.

Perou

C.M.

Ellis

M.J.

and Nielsen

T.O.

, A 50-gene intrinsic subtype classifier for prognosis and prediction of benefit from adjuvant tamoxifen, Clin Cancer Res 18 (2012), 4465–4472.

11.

Ning

Wang

Zhou

Sun

and Yu

, The evolution and expression pattern of human overlapping lncrna and protein-coding gene pairs, Sci Rep 7 (2017), 42775.

12.

Sun

Gadad

S.S.

Kim

D.S.

and Kraus

W.L.

, Discovery, annotation, and functional analysis of long noncoding RNAs controlling cell-cycle gene expression and proliferation in breast cancer cells, Mol Cell 59 (2015), 698–711.

13.

Hou

Zhao

Yao

Gao

Zhao

Zhang

Huang

and Lu

, LincRNA-ROR induces epithelial-to-mesenchymal transition and contributes to breast cancer tumorigenesis and metastasis, Cell Death Dis 5 (2014), e1287.

14.

Liu

Sun

Liu

Gong

Yao

Lin

Yao

Zeng

and Song

, A cytoplasmic NF-kappaB interacting long noncoding RNA blocks IkappaB phosphorylation and suppresses breast cancer metastasis, Cancer Cell 27 (2015), 370–381.

15.

Munoz-Galindo

Melendez-Zajgla

Pacheco-Fernandez

Rodriguez-Sosa

Mandujano-Tinoco

E.A.

Vazquez-Santillan

Castro-Oropeza

Lizarraga

, J.M. Sanchez-Lopez and Maldonado

, Changes in the transcriptome profile of breast cancer cells grown as spheroids, Biochem Biophys Res Commun 516 (2019), 1258–1264.

16.

Pacheco-Marin

Melendez-Zajgla

Castillo-Rojas

Mandujano-Tinoco

Garcia-Venzor

Uribe-Carvajal

Cabrera-Orefice

Gonzalez-Torres

Gaytan-Cervantes

, I.B. Mitre-Aguilar and Maldonado

, Transcriptome profile of the early stages of breast cancer tumoral spheroids, Sci Rep 6 (2016), 23373.

17.

Salameh

Fan

Choi

B.K.

Zhang

and An

, HER3 and LINC00052 interplay promotes tumor growth in breast cancer, Oncotarget 8 (2017), 6526–6539.

18.

Shan

Ying

Jia

Kong

Zheng

and Jin

, LINC00052 promotes gastric cancer cell proliferation and metastasis via activating the wnt/beta-catenin signaling pathway, Oncol Res 25 (2017), 1589–1599.

19.

Xiong

Sheng

Ding

Chen

Tan

Zeng

Qin

Zhu

Huang

and Tang

, LINC00052 regulates the expression of NTRK3 by miR-128 and miR-485-3p to strengthen HCC cells invasion and migration, Oncotarget 7 (2016), 47593–47608.

20.

Yan

Shan

Chen

Liu

Chen

Zeng

Zhu

Dang

Chen

Ling

Huang

and Tang

, LINC00052/miR-101-3p axis inhibits cell proliferation and metastasis by targeting SOX9 in hepatocellular carcinoma, Gene 679 (2018), 138–149.

21.

Zhu

Yang

Chen

Zeng

Yan

Liu

Chen

Pan

Zhou

Huang

and Tang

, LINC00052 upregulates EPB41L3 to inhibit migration and invasion of hepatocellular carcinoma by binding miR-452-5p, Oncotarget 8 (2017), 63724–63737.

22.

Ouyang

Zhang

Tang

and Wang

, Long non-coding RNA LINC00052 regulates miR-608/EGFR axis to promote progression of head and neck squamous cell carcinoma, Experimental and Molecular Pathology 111 (2019), 104321.

23.

Nagy

Lanczky

Menyhart

and Gyorffy

, Validation of miRNA prognostic power in hepatocellular carcinoma using expression data of independent datasets, Sci Rep 8 (2018), 9227.

24.

Mandujano-Tinoco

E.A.

Garcia-Venzor

Munoz-Galindo

Lizarraga-Sanchez

Favela-Orozco

Chavez-Gutierrez

Krotzsch

Salgado

R.M.

, J. Melendez-Zajgla and Maldonado

, miRNA expression profile in multicellular breast cancer spheroids, Biochim Biophys Acta Mol Cell Res 1864 (2017), 1642–1655.

25.

Horn

and Boutros

, E-RNAi: A web application for the multi-species design of RNAi reagents – 2010 update, Nucleic Acids Res 38 (2010), W332–W339.

26.

Khomtchouk

B.B.

Hennessy

J.R.

and Wahlestedt

, shinyheatmap: Ultra fast low memory heatmap web interface for big data genomics, PLoS One 12 (2017), e0176334.

27.

Zhou

Soufan

Ewald

Hancock

R.E.W.

Basu

and Xia

, NetworkAnalyst 3.0: A visual analytics platform for comprehensive gene expression profiling and meta-analysis, Nucleic Acids Res 47 (2019), W234–W241.

28.

Subramanian

Tamayo

Mootha

V.K.

Mukherjee

Ebert

B.L.

Gillette

M.A.

Paulovich

Pomeroy

S.L.

Golub

T.R.

Lander

E.S.

and Mesirov

J.P.

, Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles, Proc Natl Acad Sci USA 102 (2005), 15545–15550.

29.

Jung

Kim

Freishtat

R.J.

Giri

Hoffman

and Seo

, miRTarVis: An interactive visual analysis tool for microRNA-mRNA expression profile data, BMC Proc 9 (2015), S2.

30.

Mandujano-Tinoco

E.A.

Gallardo-Perez

J.C.

Marin-Hernandez

, R. Moreno-Sanchez and Rodriguez-Enriquez

, Anti-mitochondrial therapy in human breast cancer multi-cellular spheroids, Biochim Biophys Acta 1833 (2013), 541–551.

31.

Lizarraga

Poincloux

Romao

Montagnac

, G. Le Dez Bonne

Rigaill

Raposo

and Chavrier

, Diaphanous-related formins are required for invadopodia formation and invasion of breast tumor cells, Cancer Res 69 (2009), 2792–2800.

32.

Westerfield

, The zebrafish book, A guide for the laboratory use of zebrafish (Danio rerio), University of Oregon Press, Eugene, 4th edition, 2000.

33.

Muruganujan

Ebert

Huang

and Thomas

P.D.

, PANTHER version 14: More genomes, a new PANTHER GO-slim and improvements in enrichment analysis tools, Nucleic Acids Res 47 (2019), D419–D426.

34.

Moran

Rivera

Sanchez-Arago

Blazquez

Merinero

Ugalde

Arenas

Cuezva

J.M.

and Martin

M.A.

, Mitochondrial bioenergetics and dynamics interplay in complex I-deficient fibroblasts, Biochim Biophys Acta 1802 (2010), 443–453.

35.

Schindelin

Arganda-Carreras

Frise

Kaynig

Longair

Pietzsch

Preibisch

Rueden

Saalfeld

Schmid

Tinevez

J.Y.

White

D.J.

Hartenstein

Eliceiri

Tomancak

and Cardona

, Fiji: An open-source platform for biological-image analysis, Nat Methods 9 (2012), 676–682.

36.

Thomas

K.J.

and Jacobson

M.R.

, Defects in mitochondrial fission protein dynamin-related protein 1 are linked to apoptotic resistance and autophagy in a lung cancer model, PLoS One 7 (2012), e45319.

37.

Kasoha

Unger

Solomayer

E.F.

Bohle

R.M.

Zaharia

Khreich

Wagenpfeil

and Juhasz-Boss

, Prostate-specific membrane antigen (PSMA) expression in breast cancer and its metastases, Clin Exp Metastasis 34 (2017), 479–490.

38.

Yan

Huang

and Tian

, Prognostic value of PEG10 in Asian solid tumors: A meta-analysis, Clin Chim Acta 483 (2018), 197–203.

39.

Xiao

Tembo

Hao

Xiong

Pan

Yang

Yuan

Xiong

and Zhang

, PEG10 promotes human breast cancer cell proliferation, migration and invasion, Int J Oncol 48 (2016), 1933–1942.

40.

Sneh

Deol

Y.S.

Ganju

Shilo

Rosol

T.J.

Nasser

M.W.

and Ganju

R.K.

, Differential role of psoriasin (S100A7) in estrogen receptor alpha positive and negative breast cancer cells occur through actin remodeling, Breast Cancer Res Treat 138 (2013), 727–739.

41.

Sureshbabu

Tonner

Allan

G.J.

and Flint

D.J.

, Relative roles of TGF-beta and IGFBP-5 in idiopathic pulmonary fibrosis, Pulm Med 2011 (2011), 517687.

42.

Vijayan

Guha

Ameer

Kaziri

Mooney

C.C.

Bennett

Sureshbabu

Tonner

Beattie

Allan

G.J.

Edwards

and Flint

D.J.

, IGFBP-5 enhances epithelial cell adhesion and protects epithelial cells from TGFbeta1-induced mesenchymal invasion, Int J Biochem Cell Biol 45 (2013), 2774–2785.

43.

Cimino

Fuso

Sfiligoi

Biglia

Ponzone

Maggiorotto

Russo

Cicatiello

Weisz

Taverna

Sismondi

and De Bortoli

, Identification of new genes associated with breast cancer progression by gene expression analysis of predefined sets of neoplastic tissues, Int J Cancer 123 (2008), 1327–1338.

44.

Sano

Wada

Eguchi

Osaki

Saeki

and Nishiyama

, Quantitative prediction of tumor response to neoadjuvant chemotherapy in breast cancer: Novel marker genes and prediction model using the expression levels, Breast Cancer 19 (2012), 37–45.

45.

Marques

I.J.

Weiss

F.U.

Vlecken

D.H.

Nitsche

Bakkers

Lagendijk

A.K.

Partecke

L.I.

Heidecke

C.D.

Lerch

M.M.

and Bagowski

C.P.

, Metastatic behaviour of primary human tumours in a zebrafish xenotransplantation model, BMC Cancer 9 (2009), 128.

46.

Murphy

D.A.

Diaz

Bromann

P.A.

Tsai

J.H.

Kawakami

Maurer

Stewart

R.A.

, Izpisua-Belmonte

J.C.

and Courtneidge

S.A.

, A Src-Tks5 pathway is required for neural crest cell migration during embryonic development, PLoS One 6 (2011), e22499.

47.

Bronsert

Enderle-Ammour

Bader

Timme

Kuehs

Csanadi

Kayser

Kohler

Bausch

Hoeppner

Hopt

U.T.

Keck

Stickeler

Passlick

Schilling

Reiss

C.P.

Vashist

Brabletz

Berger

Lotz

Olesch

Werner

and Wellner

U.F.

, Cancer cell invasion and EMT marker expression: A three-dimensional study of the human cancer-host interface, J Pathol 234 (2014), 410–422.

48.

Lucanus

A.J.

and Yip

G.W.

, Kinesin superfamily: Roles in breast cancer, patient prognosis and therapeutics, Oncogene 37 (2018), 833–838.

49.

Singel

S.M.

Cornelius

Zaganjor

Batten

Sarode

V.R.

Buckley

D.L.

Peng

John

G.B.

H.C.

Sadeghi

Wright

W.E.

Lum

Corson

T.W.

and Shay

J.W.

, KIF14 promotes AKT phosphorylation and contributes to chemoresistance in triple-negative breast cancer, Neoplasia 16 (2014), 247–256.

50.

Assaker

Camirand

Abdulkarim

Omeroglu

Deschenes

Joseph

Noman

A.S.M.

, A.V. Ramana Kumar Kremer

and Sabri

, PTHrP, a biomarker for CNS metastasis in triple-negative breast cancer and selection for adjuvant chemotherapy in node-negative disease, JNCI Cancer Spectr 4 (2020), pkz063.

51.

Shen

Qian

and Falzon

, PTH-related protein enhances MCF-7 breast cancer cell adhesion, migration, and invasion via an intracrine pathway, Exp Cell Res 294 (2004), 420–433.

52.

Wolfe

B.A.

Takaki

Petronczki

and Glotzer

, Polo-like kinase 1 directs assembly of the HsCyk-4 RhoGAP/Ect2 RhoGEF complex to initiate cleavage furrow formation, PLoS Biol 7 (2009), e1000110.

53.

Jansen

Gosens

Wieland

and Schmidt

, Paving the Rho in cancer metastasis: Rho GTPases and beyond, Pharmacol Ther 183 (2018), 1–21.

54.

Lawson

C.D.

and Ridley

A.J.

, Rho GTPase signaling complexes in cell migration and invasion, J Cell Biol 217 (2018), 447–457.

55.

Zhao

Wang

Cao

Meng

Pang

Nadolny

Dong

and Cai

, Ang II-AT1R increases cell migration through PI3K/AKT and NF-kappaB pathways in breast cancer, J Cell Physiol 229 (2014), 1855–1862.

56.

Liu

Zhang

Zhao

X.Y.

Min

P.X.

Y.D.

Wang

Y.Y.

Chen

Tang

S.J.

Zhang

Y.J.

and Gu

, Non-canonical notch signaling regulates actin remodeling in cell migration by activating PI3K/AKT/Cdc42 pathway, Front Pharmacol 10 (2019), 370.

57.

Shao

Zhao

Zhang

Luo

Zuo

Huang

Wang

and Zhao

, Notch1 signaling regulates the epithelial-mesenchymal transition and invasion of breast cancer in a slug-dependent manner, Mol Cancer 14 (2015), 28.

58.

Wang

Naydenov

N.G.

Dozmorov

M.G.

Koblinski

J.E.

and Ivanov

A.I.

, Anillin regulates breast cancer cell migration, growth, and metastasis by non-canonical mechanisms involving control of cell stemness and differentiation, Breast Cancer Res 22 (2020), 3.

59.

Zhang

Luga

Armitage

S.K.

Musiol

Won

Yip

C.M.

Plotnikov

S.V.

and Wrana

J.L.

, A lateral signalling pathway coordinates shape volatility during cell migration, Nat Commun 7 (2016), 11714.

60.

Huang

Hong

Wang

Yang

and Li

, Overexpression of CKS2 is associated with a poor prognosis and promotes cell proliferation and invasion in breast cancer, Mol Med Rep 19 (2019), 4761–4769.

61.

Blanchard

A.A.

Wang

Hombach-Klonisch

Penner

Ozturk

Klonisch

Pitz

Murphy

Leygue

and Myal

, Claudin 1 is highly upregulated by PKC in MCF7 human breast cancer cells and correlates positively with PKCepsilon in patient biopsies, Transl Oncol 12 (2019), 561–575.

62.

Zhou

Blanchard

Wang

Han

Schroedter

Leygue

and Myal

, Claudin 1 promotes migration and increases sensitivity to tamoxifen and anticancer drugs in luminal-like human breast cancer cells MCF7, Cancer Invest 33 (2015), 429–439.

63.

Wang

H.K.

Liang

J.F.

Zheng

H.X.

and Xiao

, Expression and prognostic significance of ECT2 in invasive breast cancer, J Clin Pathol 71 (2018), 442–445.

64.

Bartella

, E.M. De Francesco Perri

M.G.

Curcio

Dolce

Maggiolini

and Vivacqua

, The G protein estrogen receptor (GPER) is regulated by endothelin-1 mediated signaling in cancer cells, Cell Signal 28 (2016), 61–71.

65.

Shi

Liu

Sun

Qiu

Cheng

and Wang

, Decreased expression and prognostic role of EHD2 in human breast carcinoma: Correlation with E-cadherin, J Mol Histol 46 (2015), 221–231.

66.

Yang

Ren

Yao

Chen

and He

, Role of EHD2 in migration and invasion of human breast cancer cells, Tumour Biol 36 (2015), 3717–3726.

67.

Feng

Guo

Yan

Shen

Ruan

and Qiu

, Expression of protein 4.1 family in breast cancer: Database mining for 4.1 family members in malignancies, Med Sci Monit 25 (2019), 3374–3389.

68.

Wojtukiewicz

M.Z.

Hempel

Sierko

Tucker

S.C.

and Honn

K.V.

, Protease-activated receptors (PARs) – biology and role in cancer invasion and metastasis, Cancer Metastasis Rev 34 (2015), 775–796.

69.

Tao

Pan

Jiang

Zheng

Lin

and Cao

, B-Myb regulates snail expression to promote epithelial-to-mesenchymal transition and invasion of breast cancer cell, Med Oncol 32 (2015), 412.

70.

Rosario

C.O.

Kazazian

Zih

F.S.

Brashavitskaya

Haffani

R.S.

George

Dennis

J.W.

and Swallow

C.J.

, A novel role for Plk4 in regulating cell spreading and motility, Oncogene 34 (2015), 3441–3451.

71.

Barrio-Real

Wertheimer

Garg

Abba

M.C.

and Kazanietz

M.G.

, Characterization of a P-Rex1 gene signature in breast cancer cells, Oncotarget 7 (2016), 51335–51348.

72.

Liu

H.J.

Ooms

L.M.

Srijakotre

Man

Vieusseux

Waters

J.E.

Feng

Bailey

C.G.

Rasko

J.E.

Price

J.T.

and Mitchell

C.A.

, PtdIns(3, 4, 5)P3-dependent rac exchanger 1 (PREX1) rac-guanine nucleotide exchange factor (GEF) activity promotes breast cancer cell proliferation and tumor growth via activation of extracellular signal-regulated kinase 1/2 (ERK1/2) signaling, J Biol Chem 291 (2016), 17258–17270.

73.

Ren

Yang

Zhang

Liu

and Wang

, SKA2 mediates invasion and metastasis in human breast cancer via EMT, Mol Med Rep 19 (2019), 515–523.

74.

Jiang

Wang

Sun

Zhou

and Yu

, High expression of SPAG5 sustains the malignant growth and invasion of breast cancer cells through the activation of Wnt/beta-catenin signalling, Clin Exp Pharmacol Physiol 46 (2019), 597–606.

75.

Campo

and Breuer

E.K.

, Inhibition of TACC3 by a small molecule inhibitor in breast cancer, Biochem Biophys Res Commun 498 (2018), 1085–1092.

76.

G.H.

Park

J.S.

and Breuer

E.K.

, TACC3 promotes epithelial-mesenchymal transition (EMT) through the activation of PI3K/Akt and ERK signaling pathways, Cancer Lett 332 (2013), 63–73.

77.

Chen

Zhang

Zhong

Kai

Tong

Xie

Wang

Zheng

Guo

and Lu

, Targeting TPX2 suppresses proliferation and promotes apoptosis via repression of the PI3k/AKT/P21 signaling pathway and activation of p53 pathway in breast cancer, Biochem Biophys Res Commun 507 (2018), 74–82.

78.

Yang

D.P.

Shen

X.C.

J.B.

Song

and Zhang

J.H.

, TPX2 promotes migration and invasion of human breast cancer cells, Asian Pac J Trop Med 8 (2015), 1064–1070.

79.

Zhou

Chen

W.R.

Yang

L.C.

Wang

Sun

J.Y.

Zhang

W.W.

Z.Y.

and Wu

S.G.

, KIF11 functions as an oncogene and is associated with poor outcomes from breast cancer, Cancer Res Treat 51 (2019), 1207–1221.

80.

Ahmed

S.M.

Theriault

B.L.

Uppalapati

Chiu

C.W.

Gallie

B.L.

Sidhu

S.S.

and Angers

, KIF14 negatively regulates Rap1a-Radil signaling during breast cancer progression, J Cell Biol 199 (2012), 951–967.

81.

Yang

Yan

Liu

Zhu

and Gu

, KIF14 promotes tumor progression and metastasis and is an independent predictor of poor prognosis in human gastric cancer, Biochim Biophys Acta Mol Basis Dis 1865 (2019), 181–192.

82.

Zou

J.X.

Duan

Wang

Sokolov

Chen

C.Z.

J.J.

and Chen

H.W.

, Kinesin family deregulation coordinated by bromodomain protein ANCCA and histone methyltransferase MLL for breast cancer cell growth, survival, and tamoxifen resistance, Mol Cancer Res 12 (2014), 539–549.

83.

Pan

Shang

D.T.

and Wu

F.L.

, KIF18B as a regulator in microtubule movement accelerates tumor progression and triggers poor outcome in lung adenocarcinoma, Tissue Cell 61 (2019), 44–50.

84.

Yan

Zhu

and Zhang

, Kinesin family member 18B: A contributor and facilitator in the proliferation and metastasis of cutaneous melanoma, J Biochem Mol Toxicol 33 (2019), e22409.

85.

Yang

Wang

Xie

Wang

Gao

Rong

Liu

and Lu

, KIF18B promotes hepatocellular carcinoma progression through activating Wnt/beta-catenin-signaling pathway, J Cell Physiol (2020).

86.

Chong

Yang

and Chen

, Kinesin motor protein KIFC1 is a target protein of miR-338-3p and is Abbreviations associated with poor prognosis and progression of renal cell carcinoma, Oncol Res 27 (2018), 125–137.

87.

Ogden

Garlapati

X.B.

Turaga

R.C.

Oprea-Ilies

Wright

Bhattarai

Mittal

Wetherilt

C.S.

Krishnamurti

Reid

M.D.

Jones

Gupta

Osan

Pattni

Riaz

Klimov

Rao

Cantuaria

Rida

P.C.

and Aneja

, Multi-institutional study of nuclear KIFC1 as a biomarker of poor prognosis in African American women with triple-negative breast cancer, Sci Rep 7 (2017), 42289.

88.

Xiao

K.H.

Teng

Y.L.

Tan

Chen

M.K.

Liang

H.T.

Feng

Z.H.

Duan

J.L.

Deng

M.H.

Wei

W.S.

Luo

J.H.

Qin

Z.K.

and Xie

, Kinesin family member C1 accelerates bladder cancer cell proliferation and induces epithelial-mesenchymal transition via Akt/GSK3beta signaling, Cancer Sci 110 (2019), 2822–2833.

89.

Mandal

Pogoda

Nandi

Mathieu

Kasri

Klein

Radvanyi

Goud

Janmey

P.A.

and Manneville

J.B.

, Role of a kinesin motor in cancer cell mechanics, Nano Lett 19 (2019), 7691–7702.

90.

Stangel

Erkan

Buchholz

Gress

Michalski

Raulefs

Friess

and Kleeff

, Kif20a inhibition reduces migration and invasion of pancreatic cancer cells, J Surg Res 197 (2015), 91–100.

91.

Xie

Gao

Yang

Qiao

and Fang

, KIF20A silence inhibits the migration, invasion and proliferation of non-small cell lung cancer and regulates the JNK pathway, Clin Exp Pharmacol Physiol 47 (2020), 135–142.

92.

Lin

W.F.

Lin

X.L.

S.W.

Yang

Tang

C.T.

Gao

Y.J.

Chen

H.Y.

and Ge

Z.Z.

, Pseudopod-associated protein KIF20B promotes Gli1-induced epithelial-mesenchymal transition modulated by pseudopodial actin dynamic in human colorectal cancer, Mol Carcinog 57 (2018), 911–925.

93.

Z.Y.

Jiang

X.Y.

Zhao

R.R.

Qin

J.J.

Luo

C.J.

Ren

Y.X.

Ren

Z.J.

and Jiao

Z.Y.

, Effect of KIF22 on promoting proliferation and migration of gastric cancer cells via MAPK-ERK pathways, Chin Med J (Engl) 133 (2020), 919–928.

94.

Shimo

Tanikawa

Nishidate

Lin

M.L.

Matsuda

Park

J.H.

Ueki

Ohta

Hirata

Fukuda

Nakamura

and Katagiri

, Involvement of kinesin family member 2C/mitotic centromere-associated kinesin overexpression in mammary carcinogenesis, Cancer Sci 99 (2008), 62–70.

95.

Hou

Dong

Liu

Chen

Liu

Wang

and Zhou

, Upregulate KIF4A enhances proliferation, invasion of hepatocellular carcinoma and indicates poor prognosis across human cancer types, Sci Rep 7 (2017), 4148.

Integrative analysis of transcriptional profile reveals LINC00052 as a suppressor of breast cancer cell migration

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSION:

Keywords

List of abbreviations

1. Introduction

2. Materials and methods

2.1 Analysis in silico of LINC00052 expression in breast cancer tissues

2.2 Cell culture

2.3 Multicellular tumour spheroid generation

2.4 shRNA design and stable cell line establishment

2.5 Total RNA extraction and reverse transcription (RT) reaction

2.6 qPCR

2.7 Transcriptome analysis by microarrays

2.8 Gene expression bioinformatic analyses

2.9 Microarray data validation by qPCR

2.10 Oxygen consumption and determination of OxPhos-dependent flux

2.11 In vitro migration assay

2.12 Gelatine degradation assay

2.13 Zebrafish husbandry

2.15 Statistical analysis

3.1 LINC00052 expression decreases in breast tumour from patients with worse outcome

3.2 LINC00052 expression is downregulated upon MCTS formation

3.3 Gene expression changes in LINC00052-knockdown MCF-7 cells

3.5 Gene Ontology microarray data analysis

3.6 Effects of LINC00052 downregulation on cell respiration rates

3.7 LINC00052 knockdown results in greater MCF-7 cell migration capacity in vitro and in vivo

Supplementary data

Footnotes

Acknowledgments

References