oncogenic long noncoding rna landscape in breast cancer

Xu et al. Molecular Cancer (2017) 16:129 DOI 10.1186/s12943-017-0696-6

RESEARCH Open Access

Oncogenic long noncoding RNA landscapein breast cancer
Shouping Xu1, Dejia Kong1, Qianlin Chen1, Yanyan Ping2 and Da Pang1,3*
Abstract

Background: Few long noncoding RNAs (lncRNAs) that act as oncogenic genes in breast cancer have been identified.

Methods: Oncogenic lncRNAs associated with tumourigenesis and worse survival outcomes were examined andvalidated in Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA), respectively. Then, the potentialbiological functions and expression regulation of these lncRNAs were studied via bioinformatics and genome dataanalysis. Moreover, progressive breast cancer subtype-specific lncRNAs were investigated via high-throughputsequencing in our cohort and TCGA validation. To elucidate the mechanisms of the regulation of these lncRNAs,genomic alterations from the TCGA, Broad, Sanger and BCCRC data, as well as epigenetic modifications from GEOdata, were then applied and examined to meet this objective. Finally, cell proliferation assays, flow cytometryanalyses and TUNEL assays were applied to validate the oncogenic roles of these lncRNAs in vitro.

Results: A cluster of oncogenic lncRNAs that was upregulated in breast cancer tissue and was associated withworse survival outcomes was identified. These oncogenic lncRNAs are involved in regulating immune systemactivation and the TGF-beta and Jak-STAT signalling pathways. Moreover, TINCR, LINC00511, and PPP1R26-AS1were identified as subtype-specific lncRNAs associated with HER-2, triple-negative and luminal B subtypes ofbreast cancer, respectively. The up-regulation of these oncogenic lncRNAs is mainly caused by gene amplificationin the genome in breast cancer and other solid tumours. Finally, the knockdown of TINCR, DSCAM-AS1 or HOTAIRinhibited breast cancer cell proliferation, increased apoptosis and inhibited cell cycle progression in vitro.

Conclusions: These findings enhance the landscape of known oncogenic lncRNAs in breast cancer and provideinsights into their roles. This understanding may potentially aid in the comprehensive management of breast cancer.

Keywords: LncRNAs, Prognosis, Tumourigenesis, Breast cancer, TINCR

BackgroundAlthough great advances have been made in the manage-ment of breast cancer over the past decade, patient out-comes still merit consideration due to the high rate oftumour-specific death [1, 2]. The molecular mechanismsof tumourigenesis are still unclear in breast cancer. Thus,identifying new genes related to tumourigenesis and pa-tient prognosis, as well as elucidating the molecularmechanisms underlying these oncogenic processes, areurgently required.

* Correspondence: [email protected] of Breast Surgery, Harbin Medical University Cancer Hospital,150 Haping Road, Harbin 150081, China3Heilongjiang Academy of Medical Sciences, 157 Baojian Road, Harbin150086, ChinaFull list of author information is available at the end of the article

© The Author(s). 2017 Open Access This articInternational License (http://creativecommonsreproduction in any medium, provided you gthe Creative Commons license, and indicate if(http://creativecommons.org/publicdomain/ze

It is certified that less than 2% of the genome encodesproteins, but at least 75% are transcribed into noncodingRNAs [3]. Generally, long noncoding RNAs (lncRNAs)are defined as transcripts that are longer than 200 nucle-otides and have no protein-coding capacity [4]. Thereare more than 120,000 transcripts annotated as lncRNAsin the human genome as released by the Encyclopediaof DNA Elements (ENCODE) Project Consortium(release 28) [5]. LncRNAs share common features withmRNAs: many of them are transcribed by RNA poly-merase II and undergo 5′-capping, polyadenylation andsplicing [3]. In addition, the histone profiles of lncRNAsare similar to protein-coding genes for the active histonemarkers H3K4me2, H3K4me3, H3K9ac and H3K27ac [6].On the other hand, lncRNAs have several distinct featuresthat distinguish them from protein-coding mRNAs.

le is distributed under the terms of the Creative Commons Attribution 4.0.org/licenses/by/4.0/), which permits unrestricted use, distribution, andive appropriate credit to the original author(s) and the source, provide a link tochanges were made. The Creative Commons Public Domain Dedication waiverro/1.0/) applies to the data made available in this article, unless otherwise stated.

http://crossmark.crossref.org/dialog/?doi=10.1186/s12943-017-0696-6&domain=pdf

http://en.wikipedia.org/wiki/Transcription_(genetics)

http://en.wikipedia.org/wiki/Nucleotide

http://en.wikipedia.org/wiki/Nucleotide

mailto:[email protected]

http://creativecommons.org/licenses/by/4.0/

http://creativecommons.org/publicdomain/zero/1.0/

Xu et al. Molecular Cancer (2017) 16:129 of 15

Generally, lncRNAs lack coding potential due to fewerexons [7]. However, there are some exceptions with thedevelopment of lncRNA investigations. LINC00961 regu-lates mTORC1 and muscle regeneration by encodingSPAR polypeptide [8]. Functional polypeptides of myore-gulin encoded by LINC00948 and Dworf encoded byNONMMUG026737 have been reported to modulateSERCA pump activity [9, 10]. LncRNA CRNDE encodes anuclear peptide, CRNDEP, which is overexpressed inhighly proliferating tissues and is involved in cell turnover[11]. Moreover, many investigations have shown thatlncRNAs are of relatively lower expression levels thanprotein-coding genes; however, they exhibit a more celltype-specific pattern [12–14]. Unlike mRNAs, mostlncRNAs are localized to the nucleus, while mRNA is inthe cytoplasm [6]. Finally, lncRNAs belong to evolutionaryconserved families evolving faster than mRNAs, where se-quence similarity is likely to be preserved mainly in re-gions with secondary structure formation [6, 12].LncRNAs play an important role in regulating gene ex-

pression at various levels, including alternative splicing,regulation of protein activity, and alteration of proteinlocalization, as well as chromatin modification, transcrip-tion, and posttranscriptional processing [3, 15–21]. Themechanisms by which lncRNAs contribute to the regula-tory networks that underpin cancer development are di-verse [22, 23]. Unlike microRNA or piwi-interacting RNA(piRNA), lncRNAs drive many important cancer pheno-types through their interactions with other cellular macro-molecules such as DNA, protein, and RNA [22–26].Regarding their role in malignancy, lncRNAs are associ-ated with immortal-associated characteristics, includingcell cycle regulation, survival, immune response orpluripotency in cancer cells [27–32]. Several commonlncRNAs have been investigated in cancers. For ex-ample, HOTAIR promotes cancer metastasis by repro-gramming the chromatin state in a manner dependenton PRC2 [33]. SChLAP1 contributes to the develop-ment of lethal cancer by antagonizing the tumour-suppressive functions of the SWI/SNF complex [34].Another lncRNA, SAMMSON, increases the pro-oncogenic function via interacting with p32 in melan-oma [35]. Despite growing knowledge regarding themolecular mechanisms of lncRNA functions in malig-nancy, the modes of action of most lncRNAs in breastcancer remain unclear. Aberrant expression of oncogeniclncRNAs may confer capacities for tumour initiation,growth, and metastasis in breast cancer cells, thus leadingto a worse prognosis for patients. Similarly, applying anti-sense oligonucleotides can inhibit the expression of thecandidate therapeutic target Malat1 successfully in anMMTV-PyMT mouse model with breast cancer [36].Thus, unravelling the landscape of these oncogeniclncRNAs in breast cancer is essential and urgent.

The aim of this study was to identify the predictivecapability of oncogenic lncRNAs for the tumourigenesisand prognosis of breast cancer. In this study, 1088 breastcancer patients from GEO data were selected and analysedto investigate the lncRNAs associated with tumourigenesisand outcomes. The aberrant long noncoding RNAs werethen validated using TCGA data and high-throughput se-quencing in our cohort. Genome-wide in silico analysisand in vitro assays revealed the potential biological func-tions of these lncRNAs, including their mechanisms ofregulation and roles in tumourigenesis. Moreover, identifi-cation and validation of breast cancer subtype-specificlncRNAs and the determination of their association withclinical outcomes were also investigated.

MethodsPublic data access and analysisGEO data (GSE21653, GSE31448, GSE10810, GSE29431,GSE23177, GSE42568, and GSE48391) were downloadedand processed (http://www.ncbi.nlm.nih.gov/geo/). Thegenome-wide lncRNA expression profiles for breast can-cer, renal cancer, and lung cancer were downloaded fromTCGA (https://tcga-data.nci.nih.gov/). For the micro-array analysis, we adjusted the signal values for low-abundance genes. A signal value lower than log3 was set tolog3. Moreover, the invariant genes (i.e., same expressionvalue across all samples) and low-variation genes were fil-tered. Genes that were detected in less than 50% of the pro-filed samples were also filtered. The SAM method wasapplied, and we implemented a series of steps to estimatethe significance of difference and false discovery rate foreach filtered gene as previously described [37]. A methodto estimate the significance of difference and false discoveryrate for each filtered gene was implemented as follows:

(1)Calculate the exchange factor s0. First, we calculatedthe standard deviation of geneexpression for allgenes si, denoted sα, as the α percentile for si. Therelative difference in gene expression (d Score) atthe α percentile is calculated as dα

i ¼ ri= si þ sαð Þ,where ri is the fold change of gene expression forgene i between two conditions. Next, each intervalof the percentile value q1 < q2 < ··· < q100 of the siand the mean absolute deviation of dα

i vj ¼ maddαi ¼ ri= si þ sαð Þjsi∈ qj; qjþ1

h �n owere calculated.

Finally, we selected the α (denote as α̂ ) to make theCV (coefficient of variation) of the vj achieve aminimum, and set the exchange factor s0 ¼ sα̂ .

(2)Calculate the statistical value (d Score) for every gene i:

di = ri/(si + s0),

http://www.ncbi.nlm.nih.gov/geo

https://tcga-data.nci.nih.gov


(3)Calculate the order statistic according to di:d(1) ≤ d(2) ≤ ··· ≤ d(i) ≤ ··· ≤ d(p),

(4)Perform 1000 permutations to estimate the expecteddistribution of the d Score. We denote the estimatedstatistical values:d�

1ð Þ≤d�2ð Þ≤…≤d�

ið Þ≤…≤d�pð Þ,

(5)Obtain the order statistic value under thepermutation:

d�

ið Þ ¼P1000

i¼1d ið Þ�

1000 ,

(6)By calculating the maximum distance between theorder statistic d(i) and the expected order statisticd�

ið Þ, we would construct a series of reject regionsfor the q-value.

(7)For a fixed delta value, we computed the differenceΔ ið Þ ¼ d ið Þ−d

�ið Þ and found the nearest Δ(i) for gene i.

The cut-up was marked as min{Δ(i) ≥ delta} for thepositive gene and the cut-down as max{Δ(i) ≤ − delta}as for negative gene. Next, we called these genessignificantly positive genes, whose difference waslarger than the cut-up value, and significantlynegative genes, whose difference was smaller than thecut-down value.

(8)Estimate the false discovery rate (FDR):FDR ¼ V pð Þ

R pð Þ, where V

(p) is the number of positivegenes called in the 1000 permutations, and R(p) isthe median of the number of false-positive genes forthe above permutations.

(9)Obtain the q-value for gene i by selecting theminimum of the FDR for the 50 delta valuesdetermined in step (7).

Welch’s t-test (unequal variances) and analysis of vari-ance were also applied for two-group and multiple-group analyses, respectively. For multiple-comparisonanalysis, the q-value was used to control the false discoveryrate. Clustering heatmap analysis was performed accordingto previous study [38].

Guilt-by-association analysisTo identify a list of lncRNAs positively and negatively cor-related with the target genes, data from the TCGA wereevaluated to compute a pairwise Pearson correlation be-tween the expression of the target lncRNA and all thegenes. Only associated genes with an absolute r ≥ 0.4 anda significant correlation (P < 0.05) were retained. Geneontology term enrichment (GO) and Kyoto Encyclopediaof Genes and Genomes (KEGG) pathway analysis wereanalysed using DAVID as previously described [39, 40].

Patients and clinical samplesThirty invasive breast cancer and adjacent non-cancerous tissues were obtained from patients who

had not received either chemotherapy or radiotherapyand who were treated at the Department of BreastSurgery at Harbin Medical University Cancer Hospitalin 2014. This study was approved by the researchethics committee of the Harbin Medical UniversityCancer Hospital. Written informed consent was ob-tained from all the patients who participated in thestudy. The clinicopathological characteristics of thepatients are presented in Additional file 1: Table S1.

Library preparation for lncRNA sequencingA total of 3 μg of RNA per sample was used for down-stream RNA sample preparations. Ribosomal RNA wasremoved using the Ribo-Zero™ Gold kit (Epicentre,Wisconsin, USA). Subsequently, sequencing librarieswere generated according to the manufacturer’s recom-mendations. The libraries were sequenced on an Illu-mina HiSeq 2500 platform, and 100-bp paired-end readswere generated. Raw sequencing and processed RNA-Seq data of this study have been deposited to the NCBIGene Expression Omnibus database under accession num-ber GSE71651 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi? token = obcxosaur xoppwx & acc = GSE71651).

Cell culture experimentsMDA-MB-453 and MCF-7 cells were cultured inDMEM (Invitrogen, Carlsbad, CA) containing 10% foetalbovine serum and 100 units/ml penicillin/streptomycinat 37 °C in an atmosphere containing 5% CO2.UACC812 and T47D cells were cultured in 1640medium (Invitrogen, Carlsbad, CA) containing 10%foetal bovine serum and 100 units/ml penicillin/streptomycin at 37 °C in an atmosphere containing5% CO2. All the cell lines were obtained from theChinese Type Culture Collection, Chinese Academy ofSciences. Cells were used during their logarithmicgrowth phase.

Small interfering RNA (siRNA) and qRT-PCRBreast cells were grown in complete medium beforetransfection with siRNAs using Lipofectamine 2000(Life Technologies, 11,668–019) according to the man-ufacturer’s protocol. Two siRNAs each were designedto target TINCR (siR_1:5′- GCAUGAAGUAGCAGGUAUUUU-3′ and siR_2: 5′-GAUCCCGAGUGAGUCAGAA UU-3′), HOTAIR (siR_1: 5′-CCACAUGAACGCCCAGAGAUU-3′ and siR_2: 5′-GAACGGGAGUACAGAGAGAUU-3′) and DSCAM-AS1 (siR_1: 5′- ACUCAUCCAUGUACCCAUUUCUUAA-3′ and siR_2: 5′-CCUCCUCCAACUGCCAUU UAUUUAU-3′). qRT-PCRanalysis was performed using the SYBR-Green method, andthe specific sequences of the primers used were asfollows: 5′-TGTGGCCCAAACTCAGGGATACAT-3′ (for-ward) and 5′-AGATGACAGTGGCTGGAGTTGTCA-

http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi

http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi


3′(reverse) for TINCR, 5′- GTCCCTAATATCCCGGAGGT-3′ (forward) and 5′-GCAGGCTTCTAAATCCGTTC-3′ (reverse) for HOTAIR; 5′-GATCCTTGTTTGGTCTCACTCC-3′(forward) and 5′-ATGCCTATGTGGGTGATTGG-3′(reverse) for DSCAM-AS1; and 5′-TTTGATGGTGACCTGGGAAT-3′ (forward) and 5′-GAACATCTGGCTGGTTCACA-3′ (reverse) for ERBB2; 5′-ACCACAGTCCATGCCATCAC-3′ (forward) and 5′- TCCACCCTGTTGCTGTA-3′ (reverse) for GAPDH. Primers forMiR-125b and U6 were obtained as previously described[41]. Quantitative normalization of target cDNA was per-formed for each sample using GAPDH/U6 expression as aninternal control. The relative levels of TINCR, HOTAIR,DSCAM-AS1, ERBB2, MiR-125b vs. GAPDH/U6 were de-termined by the comparative CT (2−ΔΔCT) method.

Cell proliferation assaysCell proliferation assays were performed using the CellCounting Kit-8 according to the manufacturer’s instruc-tions (Beyotime, Shanghai, China). Briefly, 2 × 103 cellswere seeded in a 96-well plate. Cell proliferation wasassessed at 24, 48, and 72 h. After the addition of 20 μlof WST-1 reagents per well, cultures were incubated for2 h, and the absorbance was measured at 450 nm usinga microplate reader (BioTek, VT, United States).

Flow cytometryAn Annexin-PE Apoptosis detection kit (BD Biosciences,San Jose, CA) was used to examine cell apoptosis ac-cording to the manufacturer’s instructions. Briefly, cellswere washed twice in cold PBS and then harvested andresuspended in 1× binding buffer. Next, 100 μl of thecell solution (1 × 105 cells) was transferred to a 5-mlculture tube, and 5 μl of annexin V-PE and 5 μl of 7-AAD were added into culture tube. The cells were gentlyvortexed and incubated for 15 min at RT (25 °C) in thedark. Next, 400 μl of 1× binding buffer was added toeach tube, and apoptosis analysis was performed using aFACScan instrument (Becton Dickinson, MountainView, CA, USA). For cell cycle analysis, the CycleTEST™PLUS DNA Reagent Kit (BD. Cat No.340242) wasused, and the experiment was performed as previouslydescribed [42].

TUNEL assaysApoptosis-induced DNA fragmentation was performedusing the transfers-mediated deoxyuridine triphosphate(dUTP)-digoxigenin nick end-labelling (TUNEL) assay.UACC812 and MDA-MB-453 cells were plated in 24-well flat-bottomed plates at a density of 1 × 105 cells perwell, and cells were fixed in 4% (w/v) paraformaldehydeat 4 °C for 25 min. TUNEL staining was examined usingthe in situ cell death detection kit (Roche), and the nu-clei were stained with DAPI for 10 min according to the

manufacturer’s instructions. The numbers of TUNEL-positive cells were captured with a fluorescence micro-scope (Olympus), and the ratio of apoptosis cells wasdetermined with ImagePro Plus software.

Statistical analysesThe times of OS and RFS were calculated as the timefrom surgery until the occurrence of death or relapse,respectively. The expression of lncRNA was dichoto-mized using a study-specific median expression as thecut-off to define “high value” at or above the medianversus “low value” below the median. The differencesbetween the groups in our in vitro experiments wereanalysed using Student’s t-test. Spearman correlation co-efficients were calculated for correlation analysis. All theexperiments were performed in triplicate, and SPSS 16.0software (SPSS, Chicago, IL) was used for statistical ana-lysis. All statistical tests were two-sided, and P < 0.05was considered to be statistically significant.

ResultsAberrant gene expression landscape in breast cancerTwo cohorts containing seven datasets, including 1088breast cancer patients, were selected from GEO data andwere analysed in this study. In cohort I (GSE21653 andGSE31448), 137 lncRNAs and 8914 coding genesshowed significantly aberrant expression between breastcancer and normal tissue (P < 0.05); however, in cohortII (GSE10810, GSE29431, GSE23177, GSE42568, andGSE48391), 164 lncRNAs and 9685 coding genes weredifferentially expressed in breast cancer and normal tis-sues (P < 0.05) (Fig. 1a, b and Additional file 2: Table S2and Additional file 3: Table S3). Next, the overlappinggenes were compared across these two cohorts. Therewere 30 and 25 lncRNAs upregulated and downregu-lated, respectively, in both of these cohorts (P < 0.05)(Fig. 1c and Additional file 2: Table S2). For the codinggenes, there were 1712 and 1269 mRNAs upregulatedand downregulated, respectively (P < 0.05) (Fig. 1d andAdditional file 3: Table S3). To further confirm the resultin a third independent cohort, the aberrant lncRNAswere validated using TCGA data. As expected, all exceptone lncRNA (WASIRS1) showed similar expression pro-files as those from the in silico analysis (Fig. 1e, f ).

Potential biological functions of aberrantly expressedlncRNAs in breast cancerTo explore the possible functions of these lncRNAs,guilt-by-association analyses were used, as described inthe Methods section. GO and KEGG pathway analysiswere also investigated for the identified lncRNAs.GOTERM_BP_DIRECT was selected to investigate therelated biological process of enriched genes. The mainrelated functions of the upregulated lncRNAs were

DSCAM

-AS1

DLUE2

DYNLL

1-AS1

GATA

3-AS1

LINC00

263

LINC00

511

LINC00

869

LINC00

665

LINC00

922

LOXL

1-AS1

MIA

T

RHPN1-AS1

SNHG

11

SNHG

3

HOTA

IR

LINC00

115

MCM

3AP-

AS1

PCAT6

PPP1

R26-A

S1

ST8S

IA6-AS1

TINCR

HCG18

LINC00

537

LINC00

623

LINC00

662

MIR

181A

2HG

RUSC1-AS1

TUG1

WASI

R1

WRAP7

3-20

-10

0

10

Exp

ressio

n(l

og

2)

C17or

f76-

AS1

PGM5-

AS1

ADAMTS

9-AS2

EPB41L4

A-AS1

LINC00

341

LINC00

086

MIR

143H

G

FGF1

4-AS2

ZNF6

67-A

S1

TPT1

-AS1

HCG11

LINC00

478

RBPMS-A

S1

LINC00

924

HLA-F

-AS1

LINC00

667

MBNL1

-AS1

CHL1-A

S2

HOTAIR

M1

LPP-A

S2

LIPE-A

S1

LINC00

961

EMX2O

S

CECR7

FGF1

3-AS1

-20

-15

-10

-5

0

5

10

Exp

ressio

n( l

og

2)

Cancer Normal Cancer Normal

Cancer Normal

Cancer Normal

High

Low

A B

C

E

F

D

Fig. 1 Aberrant gene expression profile in breast cancer. a Hierarchical clustering of differentially expressed genes in breast cancer relative tonormal tissue in cohort I (N = 623). b Hierarchical clustering of differentially expressed genes in breast cancer relative to normal tissue in cohort II(N = 465). Red through blue colour indicates high to low expression levels, respectively. c Upregulated (left) and downregulated (right) lncRNAsfrom cohort I and cohort II identified via Venny online software analysis (http://bioinfogp.cnb.csic.es/tools/venny/). d Upregulated (left) anddownregulated coding genes (right) from cohort I and cohort II. e Expression levels of 30 upregulated lncRNAs in cancer relative to normal tissue(left) and hierarchical clustering of these genes (right) in the TCGA. f Expression levels of 25 downregulated lncRNAs in cancer relative to normaltissue (left) and hierarchical clustering of these genes (right) in the TCGA. The red columns indicate the expression levels of lncRNAs in cancertissues, and the blue columns represent normal tissues


regulation of lymphocyte activation, extracellular matrixorganization, and cell adhesion, and the main pathways orproteins that were involved included cell adhesion mole-cules, the T-cell receptor signalling pathway, ECM-receptor interaction and the Jak-STAT signalling pathway(Fig. 2a, b and Additional file 4: Table S4). The results of

GOTERM_CC_DIRECT and GOTERM_MF_DIRECT forupregulated lncRNAs were also performed (Additionalfile 5: Table S5). For the downregulated lncRNAs, celladhesion, extracellular matrix organization, extracellu-lar structure organization and angiogenesis were identi-fied in the GO function analysis, and focal adhesion,

http://bioinfogp.cnb.csic.es/tools/venny/

0 1 2 3 4 5 6 7 8 9

GO:0045058~T cell selectionGO:0043062~extracellular structure organization

GO:0045449~regulation of transcriptionGO:0016310~phosphorylation

GO:0051251~positive regulation of lymphocyte activationGO:0006350~transcription

GO:0006955~immune responseGO:0001775~cell activation

GO:0050870~positive regulation of T cell activationGO:0002684~positive regulation of immune system process

GO:0002696~positive regulation of leukocyte activationGO:0045321~leukocyte activation

GO:0042110~T cell activationGO:0046649~lymphocyte activation

GO:0050867~positive regulation of cell activationGO:0030199~collagen fibril organization

GO:0002694~regulation of leukocyte activationGO:0022610~biological adhesion

GO:0007155~cell adhesionGO:0030198~extracellular matrix organization

GO:0051249~regulation of lymphocyte activationGO:0050865~regulation of cell activation

GO:0050863~regulation of T cell activation

-log10(p val)

0 1 2 3 4 5 6 7

hsa05222:Small cell lung cancerhsa04062:Chemokine signaling pathway

hsa04630:Jak-STAT signaling pathwayhsa04660:T cell receptor signaling pathway

hsa05310:Asthmahsa05340:Primary immunodeficiency

hsa03050:Proteasomehsa05416:Viral myocarditis

hsa04260:Cardiac muscle contractionhsa04810:Regulation of actin cytoskeleton

hsa05330:Allograft rejectionhsa05320:Autoimmune thyroid disease

hsa04940:Type I diabetes mellitushsa05332:Graft-versus-host disease

hsa05012:Parkinson's diseasehsa04670:Leukocyte transendothelial migration

hsa04672:Intestinal immune network for IgA productionhsa04512:ECM-receptor interaction

hsa05010:Alzheimer's diseasehsa04510:Focal adhesion

hsa00190:Oxidative phosphorylationhsa05016:Huntington's disease

hsa04514:Cell adhesion molecules (CAMs)

-log10(p val)

0 7 14 21 28 35

GO:0030324~lung developmentGO:0007389~pattern specification process

GO:0009792~embryonic development ending in birth or egg hatchingGO:0007517~muscle organ developmentGO:0030334~regulation of cell migration

GO:0043009~chordate embryonic developmentGO:0030323~respiratory tube development

GO:0051270~regulation of cell motionGO:0040012~regulation of locomotion

GO:0007167~enzyme linked receptor protein signaling pathwayGO:0030199~collagen fibril organization

GO:0001525~angiogenesisGO:0007160~cell-matrix adhesion

GO:0031589~cell-substrate adhesionGO:0035295~tube development

GO:0048514~blood vessel morphogenesisGO:0043062~extracellular structure organization

GO:0001501~skeletal system developmentGO:0030198~extracellular matrix organization

GO:0001568~blood vessel developmentGO:0001944~vasculature development

GO:0022610~biological adhesionGO:0007155~cell adhesion

-log10(p val)

0 2 4 6 8 10 12 14

hsa05410:Hypertrophic cardiomyopathy (HCM)hsa04060:Cytokine-cytokine receptor interaction

hsa00564:Glycerophospholipid metabolismhsa04520:Adherens junction

hsa00565:Ether lipid metabolismhsa03320:PPAR signaling pathwayhsa05414:Dilated cardiomyopathy

hsa04670:Leukocyte transendothelial migrationhsa04360:Axon guidance

hsa05200:Pathways in cancerhsa04020:Calcium signaling pathway

hsa04810:Regulation of actin cytoskeletonhsa04080:Neuroactive ligand-receptor interaction

hsa04350:TGF-beta signaling pathwayhsa02010:ABC transporters

hsa05218:Melanomahsa04270:Vascular smooth muscle contraction

hsa04514:Cell adhesion molecules (CAMs)hsa04610:Complement and coagulation cascades

hsa04512:ECM-receptor interactionhsa04510:Focal adhesion

-log10(p val)

0 4 8 12 16 20 24 28 32

GO:0051321~meiotic cell cycleGO:0007051~spindle organization

GO:0051726~regulation of cell cycleGO:0007017~microtubule-based process

GO:0006974~response to DNA damage stimulusGO:0006281~DNA repair

GO:0000075~cell cycle checkpointGO:0000819~sister chromatid segregation

GO:0000070~mitotic sister chromatid segregationGO:0006260~DNA replication

GO:0051276~chromosome organizationGO:0006259~DNA metabolic process

GO:0007059~chromosome segregationGO:0051301~cell division

GO:0022402~cell cycle processGO:0007049~cell cycle

GO:0000278~mitotic cell cycleGO:0048285~organelle fissionGO:0000280~nuclear division

GO:0007067~mitosisGO:0000087~M phase of mitotic cell cycle

GO:0022403~cell cycle phaseGO:0000279~M phase

-log10(p val)

0 2 4 6 8 10

hsa04620:Toll-like receptor signaling pathwayhsa05211:Renal cell carcinoma

hsa00250:Alanine, aspartate and glutamate metabolismhsa04660:T cell receptor signaling pathwayhsa04810:Regulation of actin cytoskeleton

hsa03450:Non-homologous end-joininghsa04514:Cell adhesion molecules (CAMs)

hsa05215:Prostate cancerhsa04115:p53 signaling pathway

hsa04520:Adherens junctionhsa00512:O-Glycan biosynthesis

hsa05212:Pancreatic cancerhsa05223:Non-small cell lung cancer

hsa03440:Homologous recombinationhsa05340:Primary immunodeficiency

hsa00240:Pyrimidine metabolismhsa04114:Oocyte meiosis

hsa04914:Progesterone-mediated oocyte maturationhsa03410:Base excision repair

hsa05219:Bladder cancerhsa05322:Systemic lupus erythematosus

hsa03030:DNA replicationhsa04110:Cell cycle

-log10(p val)

0 2 4 6 8 10 12

GO:0007169~protein tyrosine kinase signaling pathwayGO:0048545~response to steroid hormone stimulus

GO:0050873~brown fat cell differentiationGO:0022610~biological adhesion

GO:0007155~cell adhesionGO:0040017~positive regulation of locomotion

GO:0001525~angiogenesisGO:0043434~response to peptide hormone stimulusGO:0008285~negative regulation of cell proliferation

GO:0051272~positive regulation of cell motionGO:0048514~blood vessel morphogenesis

GO:0019216~regulation of lipid metabolic processGO:0032101~regulation of response to external stimulus

GO:0007167~enzyme linked receptor protein signaling pathwayGO:0001944~vasculature development

GO:0030334~regulation of cell migrationGO:0040012~regulation of locomotion

GO:0001568~blood vessel developmentGO:0042127~regulation of cell proliferation

GO:0009719~response to endogenous stimulusGO:0009725~response to hormone stimulus

GO:0051270~regulation of cell motionGO:0010033~response to organic substance

-log10(p val)

0 1 2 3 4 5 6

hsa00564:Glycerophospholipid metabolismhsa04530:Tight junction

hsa04960:Aldosterone-regulated sodium reabsorptionhsa05221:Acute myeloid leukemia

hsa00010:Glycolysis / Gluconeogenesishsa00980:Metabolism of xenobiotics by cytochrome P450

hsa00561:Glycerolipid metabolismhsa00680:Methane metabolism

hsa05210:Colorectal cancerhsa00340:Histidine metabolism

hsa00650:Butanoate metabolismhsa00380:Tryptophan metabolism

hsa00330:Arginine and proline metabolismhsa04512:ECM-receptor interaction

hsa04520:Adherens junctionhsa04610:Complement and coagulation cascades

hsa00280:Valine, leucine and isoleucine degradationhsa05200:Pathways in cancer

hsa03320:PPAR signaling pathwayhsa00620:Pyruvate metabolism

hsa00071:Fatty acid metabolismhsa00640:Propanoate metabolism

hsa04510:Focal adhesion

-log10(p val)

BA

DC

FE

HG

Fig. 2 (See legend on next page.)


(See figure on previous page.)Fig. 2 Potential biological function of aberrantly expressed lncRNAs in breast cancer. a & b Gene ontology enrichment analysis (left) and KEGGanalysis (right) for 30 upregulated lncRNAs from cohort I and cohort II. c & d Gene ontology enrichment analysis (left) and KEGG analysis (right) for25 downregulated lncRNAs from cohort I and cohort II. e & f Gene ontology enrichment analysis for 1712 upregulated coding genes (left) and1269 downregulated coding genes (right) from cohort I and cohort II. g & h KEGG analysis for 1712 upregulated coding genes (left) and 1269downregulated coding genes (right) from cohort I and cohort II. P-values <0.05 were defined as statistically significant. The vertical axis representsthe biological procession or pathway category, and the horizontal axis represents the –log10 (P value) of these significant biological processionsor pathways


ECM-receptor interaction, ABC transporter activityand the TGF-beta signalling pathway were identified inthe KEGG pathway analysis (Fig. 2c, d and Additionalfile 4: Table S4). Moreover, GO and KEGG analyseswere performed for the coding genes. The results indi-cated that the upregulated genes were associated withcell cycle phase and mitosis, whereas the downregu-lated genes were associated with the regulation of cellmotility, cell proliferation, cell adhesion, and angiogen-esis (Fig. 2e, f and Additional file 6: Table S6). ForKEGG analysis, cell cycle, DNA replication, the p53 sig-nalling pathway, and the Toll-like receptor signallingpathway were identified for the upregulated genes, andfocal adhesion, the PPAR signalling pathway, ECM-receptor interaction and glycolysis/gluconeogenesiswere identified for the downregulated genes (Fig. 2g, hand Additional file 6: Table S6).

Oncogenic lncRNAs associated with progressive breastcancerBreast cancer is highly heterogeneous and can be dividedinto the following five subclasses: luminal A, luminal B(HER-2 positive), luminal B (HER-2 negative), HER-2subtype and triple negative [1]. Owing to the aggressivenature of the triple-negative and HER-2 subtypes, wewondered whether aggressive breast cancer subtype-specific lncRNAs (BCSPLs) might exist. Thus, 33 speci-mens, including each type of breast cancer tissue(N = 15) and adjacent normal tissues (N = 15), as well asbreast tissue from non-cancer patients (N = 3), were ob-tained and subjected to RNA-Seq analysis performed byour research group (Fig. 3a and Additional file 7: Table S7).Among the 30 upregulated lncRNAs from the in silicoanalysis, three lncRNAs were confirmed to be subtype spe-cific in our RNA-Seq results: TINCR, LINC00511, andPPP1R26-AS1 represented the HER-2, triple negative andluminal B subtypes, respectively (Fig. 3b–d). To avoid se-lection bias in the samples for BCSPL, TCGA data wereused to validate these results. As expected, we obtainedconsistent results from the analysis of the public data(Fig. 3e–g).Considering triple-negative breast cancer as one of the

worst subtypes in breast cancer, LINC00511, as one ofthe triple-negative-subtype-specific lncRNAs, was se-lected for further analysis. By guilt-by-association

analysis in the TCGA set, LINC00511 was associatedwith cell adhesion, cell migration, ERK1 and ERK2 cas-cade regulation and so on by GO analysis (Additionalfile 8: Table S8). Moreover, it was involved in glyco-sphingolipid biosynthesis, cell adhesion molecules, pro-teoglycans and the p53 signalling pathway by KEGGanalysis (Additional file 8: Table S8). Next, LINC00511expression was higher in patients with BRCA1 mutationthan in those with wild-type expression (Additional file 9:Figure S1a). Moreover, this phenomenon was observed inpatients with RB1 mutation or TP53 mutation (Additionalfile 9: Figure S1b, c). Finally, the StarBase v2.0 databasewas applied to examine the relationship of LINC00511and tumour suppressor microRNAs [43]. In this process,LINC00511 had binding sites of miR-29, miR-16, miR-195and miR-497, respectively (Additional file 10: Table S9).Meanwhile, these microRNAs were reported to be tumoursuppressors [44–47]. Thus, LINC00511 may be a bonafide therapy target in triple-negative breast cancer.

Oncogenic lncRNAs are correlated with worse survivaloutcomesNext, we investigated the relationship between breastcancer prognosis and the 30 aberrant upregulatedlncRNAs obtained via in silico analysis from cohorts I &II. GEO was rechecked to identify datasets that includedall such genes and the patients’ long-term follow-up in-formation. Finally, 26 datasets with 4140 breast cancerpatients met these criteria and were used to identifybreast cancer prognosis-associated lncRNAs (BCPALs)(Additional file 11: Table S10). Among these lncRNAs,HOTAIR, LINC00115, MCM3AP-AS1, TINCR, PPP1R26-AS1, and DSCAM-AS1 were confirmed to be BCPALs vialog-rank overall survival analysis (Fig. 4a–f ). In additionto overall survival, increased expression of HOTAIR,MCM3AP-AS1, and PCAT6 indicated worse relapse-free survival in these patients (Fig. 4g–i). Among thesix BCPALs, only the increased expression of HOTAIRwas associated with a worse prognosis; the remainingfive BCPALs were first reported in this study.

Regulation of the oncogenic lncRNAsThe mechanisms of the upregulation of 30 lncRNAswere next explored via a genomic approach. Mutation,deletion, amplification, and multiple alterations were

A

LA_C

_0LA

_C_1

LA_C

_2LB

1_C

_0LB

1_C

_1LB

1_C

_2LB

2_C

_1LB

2_C

_0H

ER

2_C

_0H

ER

2_C

_1H

ER

2_C

_2LB

2_C

_2TN

B_C

_1TN

B_C

_2TN

B_C

_0H

ER

2_P

_0H

ER

2_P

_2LA

_P_2

LB2_

P_0

HE

R2_

P_1

Nor

mal

_2LA

_P_1

Nor

mal

_0LB

1_P

_2LA

_P_0

LB2_

P_1

LB1_

P_0

TNB

_P_0

Nor

mal

_1LB

2_P

_2LB

1_P

_1TN

B_P

_1TN

B_P

_2

TCONS_00017550 chr1 31205315 31230667 −TCONS_00057122 chr11 1773981 1785222 −TCONS_00142481 chr17 42422647 42430470 +TCONS_00226714 chr22 32340446 32353590 +TCONS_00023684 chr1 152004981 152009511 −TCONS_00306957 chr6 120277014 120277727 .TCONS_00340760 chr8 128675816 128677684 .TCONS_00041227 chr10 63166404 63213208 −TCONS_00058110 chr11 9041070 9113150 −TCONS_00296252 chr6 151773421 151791236 +TCONS_00033723 chr10 8982799 8991633 +TCONS_00098033 chr14 43421894 43443169 +TCONS_00342982 chr9 80912058 80945009 +TCONS_00203469 chr2 192699027 192711981 −TCONS_00224280 chr21 28387440 28392667 .TCONS_00157038 chr17 66768001 66777212 .TCONS_00259313 chr4 129579770 129583152 +TCONS_00031816 chr1 170877987 170882122 .

HER-2 HER-2 HER-2

Luminal A

Luminal A

Luminal A

Luminal B

Luminal B

Luminal B

TNB TNB TNB

Normaltissue

Normaltissue

Normaltissue

E F G

B C D Scalechr19:

2 kb hg19

5,559,000 5,560,000 5,561,000 5,562,000 5,563,000 5,564,000 5,565,000 5,566,000 5,567,000 5,568,00020 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

Scalechr17:

10 kb hg19

70,600,000 70,605,000 70,610,000 70,615,000 70,620,000 70,625,000 70,630,000 70,635,0005 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

1 _5 _

Scalechr9:

2 kb hg19

138,356,000 138,357,000 138,358,000 138,359,000 138,360,000 138,361,000 138,362,000 138,363,000 138,364,00020 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

1 _20 _

Fig. 3 Progressive breast cancer subtype-specific lncRNA identification and validation. a Hierarchical clustering of differentially expressed genes inbreast cancer tissues (N = 15), adjacent normal tissues (N = 15) and normal breast tissue from non-cancer patients (N = 3) via RNA-Seq performedby our research group. Red through blue colour indicates high to low expression levels, respectively. b, c and d RNA-Seq data visualization ofTINCR (b) and Linc00511 (c) and PPP1R26-AS1 (d) representing HER-2, triple-negative and luminal B subtype-specific lncRNAs, respectively. e, f and gValidation of TINCR (e), Linc00511 (f) and PPP1R26- AS1 (g) as progressive breast cancer subtype-specific lncRNAs in the TCGA. The expression level ofeach gene was measured by log2 FPKM. **** P < 0. 0001


investigated for these lncRNAs in the TCGA, Broadand Sanger data. The results showed that the upregula-tion of these 30 lncRNAs is mainly caused by geneamplification in breast cancer (Fig. 5a). Moreover, gene

amplification may lead to the upregulation of theselncRNAs in pan-cancer, particularly ovarian cancer,bladder cancer and uterine cancer (Fig. 5b). Next, thepotential upstream regulation of these lncRNAs via

A B C

D E F

G H I

0 50 100 150

0.0

0.2

0.4

0.6

0.8

1.0

Time (Months)

Ove

rall

surv

ival

Low expressionHigh expression logrank P = 0.002

0 50 100 150 200 250 300

0.0

0.2

0.4

0.6

0.8

1.0

Time (Months)

Ove

rall

surv

ival


0 50 100 150 200 250 300

0.0

0.2

0.4

0.6

0.8

1.0

Time (Months)

Ove

rall

surv

ial

Low expressionHigh expression logrank P < 0.001

0 50 100 150

0.0

0.2

0.4

0.6

0.8

1.0

Time (Months)

Ove

rall

surv

ial


0 50 100 150

0.0

0.2

0.4

0.6

0.8

1.0

Time (Months)

Ove

rall

surv

ial


0 50 100 1500.0

0.2

0.4

0.6

0.8

1.0

Time (Months)

Ove

rall

surv

iva

l


0 50 100 150 200 250

0.0

0.2

0.4

0.6

0.8

1.0

Time (Months)

Re

lap

s fr

ee

su

rviv

al


0 50 100 150 200 250

0.0

0.2

0.4

0.6

0.8

1.0

Time (Months)

Re

lap

s fr

ee

su

rviv

al

Low expressionHigh expression logr

0 50 100 150 200 250

0.0

0.2

0.4

0.6

0.8

1.0

Time (Months)

Rel

aps

free

sur

viva

l


HOTAIR LINC00115 MCM3AP-AS1

TINCR PPP1R26-AS1 DSCAM-AS1

HOTAIR PCAT6 PPP1R26-AS1

Fig. 4 Upregulated lncRNAs and clinical outcomes. Overall survival rates of breast cancer patients compared with the levels of HOTAIR (a),LINC00115 (b), MCM3AP-AS1 (c), TINCR (d), PPP1R26-AS1 (e), and DSCAM-AS1 (f) in 26 datasets from GEO. Relapse-free survival of breast cancerpatients compared with the levels of HOTAIR (g), PCAT6 (h) and PPP1R26-AS1 (i) in 26 datasets from GEO (N = 4140)


transcription factor and epigenetic modifications wasnext identified in breast cancer (http://www.ncbi.nlm.-nih.gov/gds/). The results indicated that DYNLL1-AS1,GATA3-AS1, LINC00263, LINC00922, MIAT, RHPN1-AS1, SNHG11, HOTAIR, MCM3AP-AS1, PCAT6,PPP1R26-AS1 and TINCR are positively regulated byoestrogen, whereas other lncRNAs, including DLEU2,LINC00511, LOXL1-AS1, SNHG3, LINC00115, MIR181-A2HG, and WRAP73, are negatively regulated by

oestrogen (Fig. 5c). Moreover, WRAP73, LINC00115,and PCAT6 may be negatively regulated by theKruppel-like zinc finger protein ZNF217, whereasLOXL1-AS1 and MIAT may be positively regulated bythis protein (Additional file 12: Table S11). DNAmethylation and other transcription factors, such asGATA3 and LIM-only protein 4, were found to be in-volved in the regulation of similar lncRNAs (Additionalfile 12: Table S11).

http://www.ncbi.nlm.nih.gov/gds

http://www.ncbi.nlm.nih.gov/gds

A C

Cancer type

Mutation data

CNA data

0%

10%

20%

30%

40%

50%

60%

70%

Alte

ratio

n Fr

eque

ncy

+ + + + + + +

+ + + + - - -

Breast (BCCRC Xenograft)

Breast (TCGA 2015)

Breast (TCGA)Breast (TCGA pub)

Breast (BCCRC 2012)

Breast (Broad)Breast (Sanger)

Cancer type

Mutation data

CNA data

0%

10%

20%

30%

40%

50%

60%

70%

Alte

ratio

n Fr

eque

ncy

+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + - + + -

+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +

Breast (BCC

RC

Xenograft)

Pancreas (UTSW

)

ACyC

(MSKC

C)

CC

LE (Novartis/Broad 2012)

Lung adeno (TCG

A pub)

Prostate (FHC

RC

)

Ovarian (TC

GA)

Cholangiocarcinom

a (TCG

A)

DLBC

(TCG

A)Breast (TC

GA 2015)

Breast (TCG

A)Liver (TC

GA)

Lung adeno (TCG

A)

Bladder (TCG

A 2014)

Sarcoma (TC

GA)

Stomach (TC

GA pub)

Uveal m

elanoma (TC

GA)

Melanom

a (TCG

A)

Ovarian (TC

GA pub)

Uterine C

S (TCG

A)

Breast (TCG

A pub)

Uterine (TC

GA pub)

Prostate (SU2C

)AC

C (TC

GA)

Bladder (TCG

A)PC

PG (TC

GA)

Bladder (MSKC

C 2012)

Prostate (TCG

A 2015)

Gliom

a (TCG

A)M

PNST (M

SKCC

)

Stomach (TC

GA)

Lung adeno (Broad)

Esophagus (TCG

A)

Uterine (TC

GA)

Head & neck (TC

GA pub)

Mesotheliom

a (TCG

A)

Mutation Deletion Amplification Multiple alterations

With oestrogen Without oestrogen

B

Fig. 5 Regulation of the oncogenic lncRNAs. a Gene amplification was the main cause of the upregulation of these genes in breast cancer andother cancers, particularly ovarian cancer, bladder cancer and uterine cancer, based on the data from the TCGA, Broad, Sanger and BCCRC databases(c). b Hierarchical clustering of aberrantly expressed lncRNAs in breast cancer with (GSM678805, GSM678806 and GSM678807) or without (GSM678802,GSM678803 and GSM678804) oestrogen stimulation. Red through blue colour indicates high to low expression levels, respectively


Validation of oncogenic roles of lncRNAsTo interpret the potential biological functions of theselncRNAs in tumourigenesis, we chose TINCR, HOTAIRand DSCAM-AS1 for further biological function re-search because these lncRNAs showed the most signifi-cant differential expression in cohort I and cohort II.Due to the increased expression of such oncogenic

lncRNAs being associated with worse outcomes, wesought to determine whether these lncRNAs might beinvolved in breast cancer tumourigenesis. The expres-sion level of TINCR was examined in breast cancer lines(Fig. 6a). UACC812 and MDA-MB-453 cells were se-lected for the biological function study of TINCR. Twodifferent siRNAs were designed and used to knock down

A

Si-NC Si-TINCR#1 Si-TINCR#20

5

10

15

Apo

ptos

isce

lls(%

) ******

Si-NC

Si-TINCR#1

Si-TINCR#2

DAPI TUNEL MERGE

MDA

-MB-

453

Si-NC

Si-TINCR#1

Si-TINCR#2

UAC

C812


5

10

15

*** ***

Apo

ptos

isce

lls(%

)

Si-NC Si-TINCR#1 Si-TINCR#20.0

0.5

1.0

1.5

*** ***

Rela

tive

exp

ressio

no

fT

INC

R

UACC812

Si-NC Si-TINCR#1 Si-TINCR#20.0

0.5

1.0

1.5

*** ***

Rela

tive

exp

ressio

no

fT

INC

R

MDA-MB-453

0h 24h 48h 72h0.0

0.5

1.0

1.5

Cell

vit

ality

(OD

450n

m)

Si-TINCR#2

Scr

**

Si-TINCR#1

MDA-MB-453

Time

NS

FITC-A

PE

-A


5

10

15

20

UACC812

Q2+

Q4(

%)

***


5

10

15

MDA-MB-453

Q2+

Q4(

%)

***

G0/G1 S G2/M0

20

40

60

80

MDA-MB-453

Perc

ent(

%)

Si-NCSi-TINCR#1Si-TINCR#2

****

****

NS*

G0/G1 S G2/M0

20

40

60

80

UACC812

Perc

ent(

%)

Si-NCSi-TINCR#1Si-TINCR#2

***

**

NS

**

**

NS

MDA-MB-45

3

UACC812

BT549

MCF7

MDA-MB-2

31

HS578T

0

50

100

150

Rel

ativ

eex

pres

sion

ofT I

NC

R

0h 24h 48h 72h0.0

0.5

1.0

1.5

Cell

vit

al it

y(O

D450n

m)

Si-TINCR#1 *Si-TINCR#2

Si-NC

UACC812

Time

*

B

C D

E

F

Fig. 6 (See legend on next page.)


(See figure on previous page.)Fig. 6 Oncogenic functions of TINCR. a The expression level of TINCR was examined in breast cancer lines. b Knockdown efficiency of siRNAstargeting TINCR in UACC812 and MDA-MB-453 cells, as determined by qRT-PCR. Quantitative normalization of TINCR was performed in eachsample using GAPDH expression as an internal control. The relative levels of TINCR vs. GAPDH were determined by the comparative CT (2−ΔΔCT)method. c The CCK-8 assay was conducted to measure cell proliferation in UACC812 and MDA-MB-453 cells after the knockdown of TINCR.d TUNEL apoptosis analysis of UACC812 and MDA-MB-453 cells after the knockdown of TINCR. e Flow cytometry apoptosis analysis of UACC812and MDA-MB-453 cells after the knockdown of TINCR. f Determination of the cell cycle distribution in UACC812 and MDA-MB-453 cells after theknockdown of TINCR via flow cytometry. *** P < 0.0 01;** P < 0. 01; * P < 0.05. Each assay was performed in triplicate


TINCR. The knockdown efficiency was validated via qRT-PCR in vitro (Fig. 6b), and the CCK-8 assay results indi-cated that knockdown of TINCR inhibited UACC-812 orMDA-MB-453 cell proliferation, respectively (Fig. 6c).Moreover, increased apoptosis was observed after thedownregulation of TINCR in UACC-812 and MDA-MB-453 cells via flow cytometry apoptosis detection andTUNEL assay, respectively (Fig. 6d, e). Finally, knock-down of TINCR, compared with controls, increased thepercentage of the G0/G1 stage and decreased the per-centage of S stage in UACC-812 and MDA-MB-453cells (Fig. 6f ).Next, molecular interaction and potential oncogenic

mechanism of TINCR were explored subsequently.Firstly, knockdown of TINCR increases the expressionof Bax and decreases the expression of Bcl-2 in UACC-812 cell (Additional file 13: Figure S2a, b). This resultwas next validated using breast cancer data in TCGA(Additional file 13: Figure S2c, d). Secondly, co-expressed genes with TINCR were investigated (Add-itional file 14: Table S12) and cell proliferation functionof these genes were detected via GO analysis (Additionalfile 13: Figure S2e). Finally, TargetScan database (release3.1) was applied to predict mircroRNAs regulated byTINCR. MiR-125b was selected and validated in vitrousing UACC-812 cell (Additional file 13: Figure S2f, g).It was reported that ERBB2 was a target of MiR-125b insolid cancers [41, 48, 49]. Thus, expression of MiR-125band ERBB2 in breast cancer was investigated. As ex-pected, MiR-125b was upregulated and ERBB2 wasdownregulated with TINCR knockdown (Additional file13: Figure S2g, h). Taken together, TINCR promotestumourigenesis partly via TINCR-MiR-125b-ERBB2 axisand its negative regulation of apoptosis pathway inbreast cancer. For HOTAIR and DSCAM-AS1, such ex-periments to examine their oncogenic roles were carriedout in breast cancer cell lines. As expected, the knock-down of DSCAM-AS1 or HOTAIR inhibits cell prolifer-ation, promotes cell apoptosis, and increases the ratio ofG0/G1 stage in MCF7 and T47D cells (Additional file15: Figure S3a, b, c, d, e). Together, these results indicatethat these lncRNAs may play oncogenic roles in thetumourigenesis of breast cancer.

DiscussionOncogenic lncRNAs in breast cancer were identified andvalidated comprehensively via genome-wide in silicoanalysis and biological experiments in this study. To ourknowledge, only a few lncRNAs that regulate the processof tumourigenesis and that are associated with the prog-nosis in breast cancer have been identified. In this study,the identified lncRNAs were found to be associated withessential biological functions in cancer, including theregulation of immune system activation, cell adhesion,angiogenesis, ABC transporter activity, and TGF-betaand Jak-STAT signalling. Moreover, TINCR, LINC00511,and PPP1R26-AS1 were identified as HER-2, triple-negative and luminal B subtype-specific lncRNAs, re-spectively. In addition, BCPALs, including HOTAIR,LINC00115, MCM3AP-AS1, TINCR, PPP1R26-AS1, andDSCAM-AS1, were identified and confirmed via log-rank analysis. Next, gene amplification in the genomeappeared to be the main underlying mechanisms for theupregulation of these oncogenic lncRNAs. Finally, theoncogenic roles of TINCR, HOTAIR and DSCAM-AS1,which showed the most significant differential expres-sion in cohort I and cohort II, were selected and per-formed in vitro. Knockdown of each of the abovelncRNA inhibited proliferation in breast cancer celllines, increased apoptosis and inhibited cell cycle pro-gression. Together, these results indicated that thelncRNAs identified and validated in this study playoncogenic roles in breast cancer.Among the 30 aberrantly upregulated lncRNAs identi-

fied in this study, only 7 have previously been reportedto be associated with malignancy, and the potential func-tions of the other 23 lncRNAs remain a mystery. In-creased SNHG3 expression is associated with malignantstatus and worse overall survival, recurrence-free sur-vival and disease-free survival in hepatocellular carcin-oma patients [50]. PCAT6 may play an oncogenic role inlung cancer progression, and it negatively correlates withthe overall survival of lung cancer patients [51]. MIATmay promote the development of lung adenocarcinomavia the MIAT-miR-106-MAPK signalling pathway loop[52]. Oestrogen increases HOTAIR levels via GPER-mediated miR-148a inhibition and is an independent


prognostic marker of metastasis in breast cancer [53, 54].Moreover, increased expression of DSCAM-AS1 was asso-ciated with a worse overall survival in breast cancer pa-tients according to our in silico analysis, and knockdownof DSCAM-AS1 inhibited breast cancer cell proliferation,increased apoptosis and inhibited cell cycle progression inthis study. We obtained similar results in a previous report[55]. Antisense lncRNAs can regulate their correspondingsense mRNAs at different levels, including transcriptionalinterference, imprinting, alternative splicing, translation orRNA editing [56]. However, the sense mRNA DSCAMwas not regulated by DSCAM-AS1 [55]. Thus, trans-regulation may be associated with the role of DSCAM-AS1 in breast cancer. Regarding DLEU2 and TINCR, itappears to be controversial that DLEU2 and TINCRwould act differently in various malignancies. DLEU2 in-hibits cell proliferation and tumour progression throughthe regulation of miR-15a/miR-16-1 in chronic lympho-cytic leukaemia [57, 58]. However, our in silico analysissuggest that, in breast cancer, it may be oncogenic, in con-trast to its reported role as a tumour suppressor inchronic lymphocytic leukaemia. For TINCR, this lncRNAis induced by SP1 and promotes cell proliferation in gas-tric cancer and oesophageal squamous cell carcinoma, afinding that is in accordance with our results in breastcancer [59, 60]. However, in colorectal cancer, the loss ofTINCR expression promotes proliferation and metastasisby activating EpCAM cleavage [61]. The functional diver-sity of DLEU2 and TINCR in various malignancies mayexplain this discrepancy, and subsequent research is ne-cessary to reveal the potential roles of these two lncRNAs.The identification of these oncogenic lncRNAs in this

study may add vital significance for clinical practice. Thestudy addresses three issues that are not present in otherstudies. First, a panel of oncogenic lncRNAs was identi-fied via genome-wide in silico analysis. It provides amore efficient step in research methodology and ex-pands the scope of studied lncRNAs that are associatedwith patient outcome. Hence, the drawbacks of a singlelncRNA study model were avoided. Second, the breastcancer prognosis-associated lncRNAs identified via dataintegration and analysis are definite in this study, andthey can be applied for the evaluation of cancer charac-teristics and patient survival potential. In this way, clin-ical examination of HOTAIR, LINC00115, MCM3AP-AS1, TINCR, PPP1R26-AS1, and DSCAM-AS1 may bebeneficial for comprehensive management. Third, breastcancer subtype-specific lncRNAs were also obtained viahigh-throughput sequencing. Such oncogenic lncRNAsshould be not underestimated for their clinical signifi-cance. From a clinical viewpoint, owing to the varyingobjectivity of pathologists and different qualities of anti-bodies to ER, PR, HER-2 and Ki-67 used in immunohis-tochemical methods, determining the molecular subtype

for an individual’s breast cancer may be influenced bysubjective and objective factors. Many obstacles, includ-ing economic and technical factors, must be overcometo achieve the standardization of immunohistochemicalmethods among laboratories and greater accessibility ofquality assurance programmes. From our sequencingresults and TCGA data, TINCR, LINC00511, andPPP1R26-AS1 were identified and were found to repre-sent HER-2, triple-negative and luminal B subtypes ofbreast cancer, respectively. Thus, examination of theseBCSPLs may complement the use of the aforementionedparameters, and this approach may decrease the mis-diagnosis rate and optimize costs in clinical practice.We acknowledge several limitations of our study. First,

not all the datasets that include lncRNA expression andfollow-up information in the GEO could be accessed.Thus, the oncogenic lncRNA landscape of breast cancerindicated in this study would not be sufficiently compre-hensive. Second, the biological functions and detailedmechanisms of these oncogenic lncRNAs were not eluci-dated. Additionally, further in-depth efforts will beneeded to investigate other oncogenic lncRNAs besidesTINCR, HOTAIR and DSCAM-AS1 with large samplesizes and a focus on molecular mechanisms.

ConclusionsTaken together, these findings broaden the oncogeniclncRNA landscape of breast cancer and provide insightsinto the roles of these oncogenic lncRNAs. The resultsmay aid in the comprehensive management of breastcancer.

Additional files

Additional file 1: Table S1. The clinicopathological characteristics ofthe patients. (XLS 19 kb)

Additional file 2: Table S2. Differentially expressed genes in Cohort I.(XLS 96 kb)

Additional file 3: Table S3. Differentially expressed genes in Cohort II.(XLS 8892 kb)

Additional file 4: Table S4. Gene ontology term enrichment (GO) andKEGG pathway analysis for upregulated lncRNAs. (XLS 326 kb)

Additional file 5: Table S5. GOTERM_CC_DIRECT andGOTERM_MF_DIRECT for upregulated lncRNAs. (XLSX 41 kb)

Additional file 6: Table S6. Gene ontology term enrichment (GO) andKEGG pathway analysis for differentially expressed mRNAs. (XLS 909 kb)

Additional file 7: Table S7. Differentially expressed genes of RNA-seqin our Cohort. (XLS 2231 kb)

Additional file 8: Table S8. Gene ontology term enrichment (GO) andKEGG pathway analysis for linc00511 in TCGA. (XLSX 21 kb)

Additional file 9: Figure S1. LINC00511 expression was high inpatients with BRCA1 or RB1 or TP53 mutation than those with wild type.(PDF 116 kb)

Additional file 10: Table S9. StarBase v2.0 database prediction for therelationship betwwen LINC00511 and tumour suppressor microRNAs.(XLSX 9 kb)

dx.doi.org/10.1186/s12943-017-0696-6

dx.doi.org/10.1186/s12943-017-0696-6

dx.doi.org/10.1186/s12943-017-0696-6

dx.doi.org/10.1186/s12943-017-0696-6

dx.doi.org/10.1186/s12943-017-0696-6

dx.doi.org/10.1186/s12943-017-0696-6

dx.doi.org/10.1186/s12943-017-0696-6

dx.doi.org/10.1186/s12943-017-0696-6

dx.doi.org/10.1186/s12943-017-0696-6

dx.doi.org/10.1186/s12943-017-0696-6


Additional file 11: Table S10. 26 datasets with 4140 breast cancerpatients for identifying prognosis-associated lncRNAs. (XLS 24 kb)

Additional file 12: Table S11. DNA methylation and transcriptionfactors regulation for lncRNAs. (XLS 27 kb)

Additional file 13: Figure S2. TINCR mediates the tumourigenesis ofbreast cancer partly via regulation of apoptosis pathway and TINCR-MiR-125-ERBB2 axis in UACC-812 cell line. (PDF 1571 kb)

Additional file 14: Table S12. TINCR is negatively expressed withseveral genes involved in apoptosis pathway via analysis of TCGA breastcancer data. (XLSX 657 kb)

Additional file 15: Figure S3. Oncogenic functions of DSCAM-AS1 andHOTAIR. (PDF 462 kb)

AbbreviationsBCPALs: Breast cancer prognosis-associated lncRNAs; BCSPLs: Breast cancersubtype-specific lncRNAs; CV: Coefficient of variation; ENCODE: Encyclopediaof DNA Elements; FDR: False discovery rate; GEO: Gene expression omnibus;GO: Gene ontology term enrichment; KEGG: Kyoto encyclopedia of genesand genomes; LncRNA: Long noncoding RNAs; piRNA: Piwi-interacting RNA;siRNA: Small interfering RNA; TCGA: The cancer genome atlas;TUNEL: Transferase mediated deoxyuridine triphosphate-digoxigenin nickend-labelling assay

AcknowledgementsThe authors thank the patients, study investigators, and staff whoparticipated in this study.

FundingThis work was supported by funding from the specific research fund forpublic service sector, National Health and Family Planning Commission ofthe People’s Republic of China (Grant Number 201402003), the National KeyTechnology Support Programme (Grant Number 2014BAI09B08), the Projectof Heilongjiang Province Applied Technology Research and Development(Grant Number GA13C201), and the National Natural Science Foundation ofChina (Grant Number 81602323), China Postdoctoral Science FoundationGrant (Grant Number 2016 M600262), Heilong Jiang postdoctoral Fund(Grant Number LBH-Z16163), Heilong Jiang province Health and FamilyPlanning Commission Foundation Grant (Grant Number 2016-087), GraduateInnovation Fund of China and Russia (Grant Number YJSCX2015-65HYD).

Availability of data and materialsThe authors declare that the data supporting the findings of this study areavailable within the article and its supplementary information files.

Authors’ contributionsSX and YP contributed to the study design, RNA sequencing data and publicdata interpretation, and manuscript draft. DP contributed to the studydesign and supervision. DK, QC and contributed to the molecular biologyexperiments. All authors contributed to the review and revision of themanuscript. All authors read and approved the final manuscript.

Ethics approval and consent to participateThis study protocol conformed to the clinical research guidelines and wasapproved by the research ethics committee of the Harbin Medical UniversityCancer Hospital. Written informed consent was obtained from all thepatients who participated in this study.

Consent for publicationNot applicable.

Competing interestsThe authors declare that they have no competing interests.

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.

Author details1Department of Breast Surgery, Harbin Medical University Cancer Hospital,150 Haping Road, Harbin 150081, China. 2College of Bioinformatics Scienceand Technology, Harbin Medical University, Harbin, China. 3HeilongjiangAcademy of Medical Sciences, 157 Baojian Road, Harbin 150086, China.

Received: 7 April 2017 Accepted: 10 July 2017

References1. Santana-Davila R, Perez EA. Treatment options for patients with triple-

negative breast cancer. J Hematol Oncol. 2010;3:42.2. Smith AD, Roda D, Yap TA. Strategies for modern biomarker and drug

development in oncology. J Hematol Oncol. 2014;7:70.3. Huarte M. The emerging role of lncRNAs in cancer. Nat Med. 2015;21(11):

1253–61.4. Ponting CP, Oliver PL, Reik W. Evolution and functions of long noncoding

RNAs. Cell. 2009;136(4):629–41.5. Jalali S, Gandhi S, Scaria V. Navigating the dynamic landscape of long

noncoding RNA and protein-coding gene annotations in GENCODE. HumGenomics. 2016;10(1):35.

6. Derrien T, Johnson R, Bussotti G, Tanzer A, Djebali S, Tilgner H, et al. TheGENCODE v7 catalog of human long noncoding RNAs: analysis of theirgene structure, evolution, and expression. Genome Res. 2012;22(9):1775–89.

7. Batista PJ, Chang HY. Long noncoding RNAs: cellular address codes indevelopment and disease. Cell. 2013;152(6):1298–307.

8. Matsumoto A, Pasut A, Matsumoto M, Yamashita R, Fung J, Monteleone E,et al. mTORC1 and muscle regeneration are regulated by the LINC00961-encoded SPAR polypeptide. Nature. 2017;541(7636):228–32.

9. Anderson DM, Anderson KM, Chang CL, Makarewich CA, Nelson BR,McAnally JR, et al. A micropeptide encoded by a putative long noncodingRNA regulates muscle performance. Cell. 2015;160(4):595–606.

10. Nelson BR, Makarewich CA, Anderson DM, Winders BR, Troupes CD, Wu F,et al. A peptide encoded by a transcript annotated as long noncoding RNAenhances SERCA activity in muscle. Science. 2016;351(6270):271–5.

11. Szafron LM, Balcerak A, Grzybowska EA, Pienkowska-Grela B, Felisiak-GolabekA, Podgorska A, et al. The novel gene CRNDE encodes a nuclear peptide(CRNDEP) which is overexpressed in highly proliferating tissues. PLoS One.2015;10(5):e0127475.

12. Cabili MN, Trapnell C, Goff L, Koziol M, Tazon-Vega B, Regev A, et al.Integrative annotation of human large intergenic noncoding RNAs revealsglobal properties and specific subclasses. Genes Dev. 2011;25(18):1915–27.

13. Djebali S, Davis CA, Merkel A, Dobin A, Lassmann T, Mortazavi A, et al.Landscape of transcription in human cells. Nature. 2012;489(7414):101–8.

14. Ravasi T, Suzuki H, Pang KC, Katayama S, Furuno M, Okunishi R, et al.Experimental validation of the regulated expression of large numbers ofnon-coding RNAs from the mouse genome. Genome Res. 2006;16(1):11–9.

15. Beltran M, Puig I, Pena C, Garcia JM, Alvarez AB, Pena R, et al. A naturalantisense transcript regulates Zeb2/Sip1 gene expression during Snail1-induced epithelial-mesenchymal transition. Genes Dev. 2008;22(6):756–69.

16. Hah N, Murakami S, Nagari A, Danko CG, Kraus WL. Enhancer transcripts markactive estrogen receptor binding sites. Genome Res. 2013;23(8):1210–23.

17. Heo JB, Sung S. Vernalization-mediated epigenetic silencing by a longintronic noncoding RNA. Science. 2011;331(6013):76–9.

18. Kwek KY, Murphy S, Furger A, Thomas B, O'Gorman W, Kimura H, et al. U1snRNA associates with TFIIH and regulates transcriptional initiation. NatStruct Biol. 2002;9(11):800–5.

19. Nagano T, Mitchell JA, Sanz LA, Pauler FM, Ferguson-Smith AC, Feil R, et al.The Air noncoding RNA epigenetically silences transcription by targetingG9a to chromatin. Science. 2008;322(5908):1717–20.

20. Penny GD, Kay GF, Sheardown SA, Rastan S, Brockdorff N. Requirement forXist in X chromosome inactivation. Nature. 1996;379(6561):131–7.

21. Rinn JL, Kertesz M, Wang JK, Squazzo SL, Xu X, Brugmann SA, et al.Functional demarcation of active and silent chromatin domains in humanHOX loci by noncoding RNAs. Cell. 2007;129(7):1311–23.

22. Huang X, Xiao R, Pan S, Yang X, Yuan W, Tu Z, et al. Uncovering the roles oflong non-coding RNAs in cancer stem cells. J Hematol Oncol. 2017;10(1):62.

23. Zhang H, Chen Z, Wang X, Huang Z, He Z, Chen Y. Long non-coding RNA: anew player in cancer. J Hematol Oncol. 2013;6:37.

24. Moyano M, Stefani G. piRNA involvement in genome stability and humancancer. J Hematol Oncol. 2015;8:38.

dx.doi.org/10.1186/s12943-017-0696-6

dx.doi.org/10.1186/s12943-017-0696-6

dx.doi.org/10.1186/s12943-017-0696-6

dx.doi.org/10.1186/s12943-017-0696-6

dx.doi.org/10.1186/s12943-017-0696-6


25. Naidu S, Magee P, Garofalo M. MiRNA-based therapeutic intervention ofcancer. J Hematol Oncol. 2015;8:68.

26. Sanchez-Mejias A, Tay Y. Competing endogenous RNA networks: tying theessential knots for cancer biology and therapeutics. J Hematol Oncol. 2015;8:30.

27. Guttman M, Amit I, Garber M, French C, Lin MF, Feldser D, et al. Chromatinsignature reveals over a thousand highly conserved large non-coding RNAsin mammals. Nature. 2009;458(7235):223–7.

28. Kawasaki Y, Komiya M, Matsumura K, Negishi L, Suda S, Okuno M, et al.MYU, a target lncRNA for Wnt/c-Myc signaling, mediates induction of CDK6to promote cell cycle progression. Cell Rep. 2016;16(10):2554–64.

29. Li Z, Gao B, Hao S, Tian W, Chen Y, Wang L, et al. Knockdown of lncRNA-PANDAR suppresses the proliferation, cell cycle and promotes apoptosis inthyroid cancer cells. EXCLI J. 2017;16:354–62.

30. Li Z, Rana TM. Decoding the noncoding: prospective of lncRNA-mediatedinnate immune regulation. RNA Biol. 2014;11(8):979–85.

31. Suravajhala P, Kogelman LJ, Mazzoni G, Kadarmideen HN. Potential role oflncRNA cyp2c91-protein interactions on diseases of the immune system.Front Genet. 2015;6:255.

32. Zhou M, Hou Y, Yang G, Zhang H, Tu G, Du YE, et al. LncRNA-Hh strengthencancer stem cells generation in twist-positive breast cancer via activation ofhedgehog signaling pathway. Stem Cells. 2016;34(1):55–66.

33. Gupta RA, Shah N, Wang KC, Kim J, Horlings HM, Wong DJ, et al. Long non-coding RNA HOTAIR reprograms chromatin state to promote cancermetastasis. Nature. 2010;464(7291):1071–6.

34. Prensner JR, Iyer MK, Sahu A, Asangani IA, Cao Q, Patel L, et al. The longnoncoding RNA SChLAP1 promotes aggressive prostate cancer andantagonizes the SWI/SNF complex. Nat Genet. 2013;45(11):1392–8.

35. Leucci E, Vendramin R, Spinazzi M, Laurette P, Fiers M, Wouters J, et al.Melanoma addiction to the long non-coding RNA SAMMSON. Nature. 2016;531(7595):518–22.

36. Mendell JT. Targeting a long noncoding RNA in breast cancer. N Engl JMed. 2016;374(23):2287–9.

37. Tusher VG, Tibshirani R, Chu G. Significance analysis of microarrays appliedto the ionizing radiation response. Proc Natl Acad Sci U S A. 2001;98(9):5116–21.

38. Eisen MB, Spellman PT, Brown PO, Botstein D. Cluster analysis and display ofgenome-wide expression patterns. Proc Natl Acad Sci U S A. 1998;95(25):14863–8.

39. Huang da W, Sherman BT, Lempicki RA. Systematic and integrative analysisof large gene lists using DAVID bioinformatics resources. Nat Protoc. 2009;4(1):44–57.

40. Zhang B, Horvath S. A general framework for weighted gene co-expressionnetwork analysis. Stat Appl Genet Mol Biol. 2005;4:Article17.

41. Tang XY, Zheng W, Ding M, Guo KJ, Yuan F, Feng H, et al. miR-125b acts asa tumor suppressor in chondrosarcoma cells by the sensitization todoxorubicin through direct targeting the ErbB2-regulated glucosemetabolism. Drug Des Devel Ther. 2016;10:571–83.

42. Garcia-Castro B, Alvarez-Zavala M, Riveros-Magana AR, Ortiz-Lazareno PC,Ratkovich-Gonzalez S, Hernandez-Flores G, et al. Restoration of WNT4inhibits cell growth in leukemia-derived cell lines. BMC Cancer. 2013;13:557.

43. Li JH, Liu S, Zhou H, Qu LH, Yang JH. starBase v2.0: decoding miRNA-ceRNA,miRNA-ncRNA and protein-RNA interaction networks from large-scale CLIP-Seq data. Nucleic Acids Res. 2014;42(Database issue):D92–7.

44. Rivas MA, Venturutti L, Huang YW, Schillaci R, Huang TH, Elizalde PV.Downregulation of the tumor-suppressor miR-16 via progestin-mediatedoncogenic signaling contributes to breast cancer development. BreastCancer Res. 2012;14(3):R77.

45. Rostas JW 3rd, Pruitt HC, Metge BJ, Mitra A, Bailey SK, Bae S, et al.microRNA-29 negatively regulates EMT regulator N-myc interactor in breastcancer. Mol Cancer. 2014;13:200.

46. Wang Y, Zhang X, Zou C, Kung HF, Lin MC, Dress A, et al. miR-195 inhibitstumor growth and angiogenesis through modulating IRS1 in breast cancer.Biomed Pharmacother. 2016;80:95–101.

47. Wu Z, Li X, Cai X, Huang C, Zheng M. miR-497 inhibits epithelialmesenchymal transition in breast carcinoma by targeting Slug. Tumour Biol.2016;37(6):7939–50.

48. Fassan M, Pizzi M, Realdon S, Balistreri M, Guzzardo V, Zagonel V, et al. TheHER2-miR125a5p/miR125b loop in gastric and esophageal carcinogenesis.Hum Pathol. 2013;44(9):1804–10.

49. Yagishita S, Fujita Y, Kitazono S, Ko R, Nakadate Y, Sawada T, et al.Chemotherapy-regulated microRNA-125-HER2 pathway as a noveltherapeutic target for Trastuzumab-mediated cellular cytotoxicity in smallcell lung cancer. Mol Cancer Ther. 2015;14(6):1414–23.

50. Zhang T, Cao C, Wu D, Liu L. SNHG3 correlates with malignant status and poorprognosis in hepatocellular carcinoma. Tumour Biol. 2016;37(2):2379–85.

51. Wan L, Zhang L, Fan K, Cheng ZX, Sun QC, Wang JJ. Knockdown of LongNoncoding RNA PCAT6 Inhibits Proliferation and Invasion in Lung CancerCells. Oncol Res. 2016;24(3):161–70.

52. Li DS, Ainiwaer JL, Sheyhiding I, Zhang Z, Zhang LW. Identification of keylong non-coding RNAs as competing endogenous RNAs for miRNA-mRNAin lung adenocarcinoma. Eur Rev Med Pharmacol Sci. 2016;20(11):2285–95.

53. Tao S, He H, Chen Q. Estradiol induces HOTAIR levels via GPER-mediatedmiR-148a inhibition in breast cancer. J Transl Med. 2015;13:131.

54. Yu X, Li Z. Long non-coding RNA HOTAIR: A novel oncogene (Review). MolMed Rep. 2015;12(4):5611–8.

55. Niknafs YS, Han S, Ma T, Speers C, Zhang C, Wilder-Romans K, et al. ThelncRNA landscape of breast cancer reveals a role for DSCAM-AS1 in breastcancer progression. Nat Commun. 2016;7:12791.

56. Magistri M, Faghihi MA, St Laurent G 3rd, Wahlestedt C. Regulation ofchromatin structure by long noncoding RNAs: focus on natural antisensetranscripts. Trends Genet. 2012;28(8):389–96.

57. Kasar S, Underbayev C, Yuan Y, Hanlon M, Aly S, Khan H, et al. Therapeuticimplications of activation of the host gene (Dleu2) promoter for miR-15a/16-1 in chronic lymphocytic leukemia. Oncogene. 2014;33(25):3307–15.

58. Klein U, Lia M, Crespo M, Siegel R, Shen Q, Mo T, et al. The DLEU2/miR-15a/16-1 cluster controls B cell proliferation and its deletion leads to chroniclymphocytic leukemia. Cancer Cell. 2010;17(1):28–40.

59. Xu TP, Liu XX, Xia R, Yin L, Kong R, Chen WM, et al. SP1-inducedupregulation of the long noncoding RNA TINCR regulates cell proliferationand apoptosis by affecting KLF2 mRNA stability in gastric cancer. Oncogene.2015;34(45):5648–61.

60. Xu Y, Qiu M, Chen Y, Wang J, Xia W, Mao Q, et al. Long noncoding RNA,tissue differentiation-inducing nonprotein coding RNA is upregulated andpromotes development of esophageal squamous cell carcinoma. DisEsophagus. 2016;29(8):950–8.

61. Zhang ZY, Lu YX, Zhang ZY, Chang YY, Zheng L, Yuan L, et al. Loss of TINCRexpression promotes proliferation, metastasis through activating EpCAMcleavage in colorectal cancer. Oncotarget. 2016;7(16):22639–49.

• We accept pre-submission inquiries

• Our selector tool helps you to find the most relevant journal

• We provide round the clock customer support

• Convenient online submission

• Thorough peer review

• Inclusion in PubMed and all major indexing services

• Maximum visibility for your research

Submit your manuscript atwww.biomedcentral.com/submit

Submit your next manuscript to BioMed Central and we will help you at every step:

oncogenic long noncoding rna landscape in breast cancer

Documents