microRNA-9 Targets Matrix Metalloproteinase 14 to Inhibit ... · Huanyu Zhang 1, Meng Qi , Shiwang Li 1, Teng Qi , Hong Mei , Kai Huang2,4, Liduan Zheng3,4, and Qiangsong Tong1,4

Therapeutic Discovery

microRNA-9 Targets Matrix Metalloproteinase 14to Inhibit Invasion, Metastasis, and Angiogenesis ofNeuroblastoma Cells

Huanyu Zhang1, Meng Qi1, Shiwang Li1, Teng Qi1, Hong Mei1, Kai Huang2,4,Liduan Zheng3,4, and Qiangsong Tong1,4

AbstractMatrix metalloproteinase (MMP)-14 is the only membrane-anchored MMP that plays a critical role in

tumor metastasis and angiogenesis. However, the mechanisms underlying MMP-14 expression in tumors

still remain largely unknown. In this study, MMP-14 immunostaining was identified in 29/42 neuroblas-

toma tissues, which was correlated with clinicopathologic features and shorter patients’ survival. In

subtotal 20 neuroblastoma cases, microRNA 9 (miR-9) was downregulated and inversely correlated with

MMP-14 expression. Bioinformatics analysis revealed a putative miR-9–binding site in the 30-untranslatedregion (30-UTR) of MMP-14 mRNA. Overexpression or knockdown of miR-9 responsively altered both the

mRNA and protein levels of MMP-14 and its downstream gene, vascular endothelial growth factor, in

cultured neuroblastoma cell lines SH-SY5Y and SK-N-SH. In anMMP-14 30-UTR luciferase reporter system,

miR-9 downregulated the luciferase activity, and these effects were abolished by a mutation in the putative

miR-9–binding site. Overexpression of miR-9 suppressed the invasion, metastasis, and angiogenesis of SH-

SY5Y and SK-N-SH cells in vitro and in vivo. In addition, the effects of miR-9 on MMP-14 expression,

adhesion, migration, invasion, and angiogenesis were rescued by overexpression of MMP-14 in these cells.

Furthermore, anti-miR-9 inhibitor or knockdown of MMP-14 respectively increased or inhibited the

migration, invasion, and angiogenesis of neuroblastoma cells. These data indicate that miR-9 suppresses

MMP-14 expression via the binding site in the 30-UTR, thus inhibiting the invasion, metastasis, and

angiogenesis of neuroblastoma. Mol Cancer Ther; 11(7); 1454–66. �2012 AACR.

IntroductionNeuroblastoma, an embryonic malignancy derived

from the neural crest, is characterized by heterogeneousbiologic behaviors, including spontaneous regression oraggressive progression (1). Because metastasis is the lead-ing cause of death in this disease, better elucidation of theunderlying mechanisms is important for improving thetherapeutic efficiencies (1). Themetastaticprocessof tumor

cells is highly complex and consists ofmultiple steps, suchas local invasion, intravasation into the circulatory system,migration to a distant site, and colonization in a differentmicroenvironment (2). Matrix metalloproteinase (MMP)-14, also named as membrane type-1 MMP (MT1-MMP),plays a critical role in facilitating the tumor cells to remodelandpenetrate extracellularmatrix (ECM; ref. 3). It has beenestablished that MMP-14 promotes tumor invasion byfunctioning as a pericellular collagenase and an activatorof proMMP-2, and is directly linked to tumorigenesis,metastasis, and angiogenesis (4, 5). However, the expres-sion of MMP-14 and underlying mechanisms in neuro-blastoma still remain largely unknown.

In recent years, emerging evidence has indicated thatmiRNAs, highly conserved and small noncoding RNAmolecules, participate in the metastasis processes byinterfering with the expression of tumor- and metasta-sis-associated genes through posttranscriptional repres-sion or mRNA degradation (2). For example, miR-21 hasbeen established as one of the most intensively studiedmetastasis-promoting miRNAs by targeting multipletumor suppressor genes, such as tissue inhibitor ofmetalloproteinase 3 (6), phosphatase, and tensin homo-log deleted on chromosome 10 (7), tropomyosin 1 (8),and programmed cell death 4 (9). The let-7 family,

Authors' Affiliations:Departments of 1Pediatric Surgery, 2Cardiology, and3Pathology, and 4Clinical Center of Human Genomic Research, UnionHospital of Tongji Medical College, Huazhong University of Science andTechnology, Wuhan, PR China

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

H. Zhang, M. Qi, and S. Li contributed equally to this work.

Corresponding Authors: Qiangsong Tong, Department of Pediatric Sur-gery, Union Hospital of Tongji Medical College, Huazhong University ofScienceandTechnology,Wuhan430022,HubeiProvince,PRChina.Phone:86-27-85726005; Fax: 86-27-85726821; E-mail: [email protected];and Liduan Zheng, Department of Pathology, Union Hospital of TongjiMedical College, Huazhong University of Science and Technology, Wuhan430022, Hubei Province, PR China. Phone: 86-27-85627129; Fax: 86-27-85726821; E-mail: [email protected]

doi: 10.1158/1535-7163.MCT-12-0001

�2012 American Association for Cancer Research.

MolecularCancer

Therapeutics

Mol Cancer Ther; 11(7) July 20121454

on November 22, 2020. © 2012 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst May 7, 2012; DOI: 10.1158/1535-7163.MCT-12-0001

http://mct.aacrjournals.org/

which comprises 13 miRNA members located on 9different chromosomes, is downregulated in many can-cer types, such as lung cancer, ovarian cancer, head andneck squamous cell carcinoma, and regulates the metas-tasis process of cancer cells (10). miR-146a inhibitscancer cell migration and invasion by targeting Rho-associated coiled-coil containing protein kinase 1 (11).Thus, it is currently urgent to investigate the roles ofmiRNAs and target genes in tumor metastasis by exper-imental models.miR-9 is a kind of miRNA selectively expressed in

neuron tissues that plays essential roles in developingneurons, neural carcinogenesis, and other diseases ofthe nervous system (12–15). In brain cancers, miR-9 iselevated and used to distinguish primary or metastaticbrain tumors with very high accuracy (16). Meanwhile,miR-9 is downregulated in lung and ovarian cancer (17,18), and is regarded as a biomarker in recurrent ovariancancer (18). These findings indicate that the miR-9expression profile in cancer relies on tissue distribu-tion. Recent evidence shows that miR-9 is closelycorrelated with metastasis of several kinds of cancer(19–21). It has been indicated that miR-9 is downregu-lated in 50% of primary neuroblastoma tumors, sug-gesting its potential function as an oncosuppressorgene (22). However, the exact roles of miR-9 and targetgene in neuroblastoma still remain elusive. In thisstudy, we showed, for the first time, that miR-9 atten-uated the expression of MMP-14 through directlytargeting the 30-untranslated region (30-UTR), and sup-pressed the invasion, metastasis, and angiogenesis ofneuroblastoma cells in vitro and in vivo.

Materials and MethodsPatient tissue samplesApproval to conduct this study was obtained from the

Institutional Review Board of Tongji Medical College.Paraffin-embedded specimens from 42 well-establishedprimary neuroblastoma cases were obtained from theDepartment of Pediatric Surgery, Union Hospital ofTongji Medical College (23). The pathologic diagnosisof neuroblastoma was confirmed by at least 2 patholo-gists. On the basis of Shimada classification system,including the mitosis karyorrhexis index (MKI), degreeof neuroblastic differentiation and stromal maturation,and patient’s age, 19 patients were classified as favor-able histology and 23 as unfavorable histology. Accord-ing to the International Neuroblastoma Staging System(INSS), 7 patients were classified as stage 1, 7 as stage 2,9 as stage 3, 11 as stage 4, and 8 as stage 4. In subtotal 20neuroblastoma patients, fresh tumor specimens werecollected at surgery and stored at �80�C until use.Protein and RNAs of normal human dorsal gangliawere obtained from Clontech.

ImmunohistochemistryImmunohistochemical staining was done as previous-

ly described (23), with antibodies specific for MMP-14

(Abcam Inc.; Santa Cruz Biotechnology; 1:200 dilu-tions), VEGF and CD31 (Santa Cruz Biotechnology;1:200 dilutions). The negative controls included parallelsections treated with omission of the primary antibody,in addition to an adjacent section of the same block inwhich the primary antibody was replaced by rabbitpolyclonal IgG (Abcam Inc.) as an isotype control. Theimmunoreactivity in each tissue section was assessed byat least 2 pathologists without knowledge of the clini-copathologic features of tumors or patients’ survival.The degree of positivity was initially classified accord-ing to the percentage of positive tumor cells as thefollowing: (�) <5% cells positive, (1þ) 6% to 25% cellspositive, (2þ) 26% to 50% cells positive, and (3þ) >50%cells positive.

Western blotTissue or cellular protein was extracted with 1� cell

lysis buffer (Promega). Western blotting was done aspreviously described (24), with antibodies specificfor MMP-14, VEGF, and glyceraldehyde-3-phosphatedehydrogenase (GAPDH; Santa Cruz Biotechnology).Enhanced chemiluminescence substrate kit (Amer-sham) was used for the detection of signals with auto-radiography film (Amersham).

Reverse transcription PCR (RT-PCR) and real-timequantitative RT-PCR

Total RNA was isolated with RNeasy Mini Kit (QiagenInc.). The reverse transcription reactions were conductedwith Transcriptor First Strand cDNA Synthesis Kit(Roche). The PCR primers for MMP-14, VEGF, andGAPDH were designed by Premier Primer 5.0 software(Supplementary Table S1). RT-PCR was done as previ-ously described (24). Real-time quantitative RT-PCRwithSYBR Green PCR Master Mix (Applied Biosystems) wasdone using ABI Prism 7700 Sequence Detector (AppliedBiosystems). The fluorescent signals were collected dur-ing extension phase, Ct values of the sample were calcu-

lated, and the transcript levels were analyzed by 2�DDCt

method.

Quantification of miR-9 expressionThe levels of mature miR-9 in primary tissues and cell

lines were determined using Bulge-Loop miRNAsqPCR Primer Set (RiboBio Co. Ltd.). After cDNA wassynthesized with a miRNA-specific stem-loop primer,the quantitative PCR was conducted with the specificprimers. The miR-9 levels were normalized to those ofU6 snRNA.

miRNA target predictionmiRNA targets were predicted using the algorithms,

including miRanda Human miRNA Targets (25), miRDB(26), RNA22 (27), and TargetScan (28). To identify thegenes commonly predicted by these 4 different algo-rithms, results of predicted targets were intersected usingmiRWalk (29).

miR-9 Targets MMP-14 in Neuroblastoma

www.aacrjournals.org Mol Cancer Ther; 11(7) July 2012 1455




Cell culture and transfectionHuman neuroblastoma cell lines IMR32 (CCL-127), SK-

N-AS (CRL-2137), SH-SY5Y (CRL-2266), and SK-N-SH(HTB-11), and human endothelial cell line HUVEC(human umbilical vein endothelial cell; CRL-1730) werepurchased from American Type Culture Collection.Cells were grown in RPMI-1640 medium (Life Technol-ogies, Inc.) supplemented with 10% FBS (Life Technolo-gies, Inc.), penicillin (100 U/mL) and streptomycin(100 mg/mL). Cells were maintained at 37�C in a humid-ified atmosphere of 5% CO2. Anti-miR-9 or negativecontrol inhibitors (RiboBio Co. Ltd.) were transfected intoconfluent cells with Lipofectamine 2000 (Life Technolo-gies, Inc.).

Pre-miR-9 construct and stable transfectionAccording to the pre-miR-9 (50-TCTTTGGTTATC-

TAGCTGTATG-30) sequence documented in miRNARegistry database (30), oligonucleotides encoding miR-9precursor (Supplementary Table S2) were subcloned intothe BamHI and XhoI restrictive sites of pPG/miR/EGFP(GenePharma Co., Ltd.), and verified by DNA sequenc-ing. The plasmids pPG/miR/EGFP and pPG-miR-9-EGFP were transfected into tumor cells, and stable celllines were screened by administration of Blasticidin(Invitrogen).

Luciferase reporter assayHuman MMP-14 30-UTR (1,555 bp) containing the

putative binding site of miR-9, and its identicalsequence with a mutation of the miR-9 seed sequence(mutant) were amplified by PCR (Supplementary TableS2), inserted between the restrictive sites XhoI and NotIof firefly/Renilla luciferase reporter vector pmiR-RB-REPOR (RiboBio Co. Ltd.), and validated by sequenc-ing. Tumor cells were plated at 1 � 105 cells/well on24-well plates, and transfected with pmiR-RB-MMP-1430-UTR (30 ng) or its mutant construct. Twenty-fourhours posttransfection, firefly and Renilla luciferaseactivities were consecutively measured, according tothe dual-luciferase assay manual (Promega). The Renillaluciferase signal was normalized to the firefly luciferasesignal for each individual analysis.

MMP-14 overexpression and knockdownHuman MMP-14 cDNA (1,749 bp) was amplified

from neuroblastoma tissue (Supplementary Table S2),subcloned into the HindIII and XhaI restrictive sites ofpcDNA3.1/Zeo(þ) (Invitrogen), and validated bysequencing. To restore the miR-9–induced downregula-tion of MMP-14, stable cell lines were transfected withthe recombinant vector pcDNA3.1-MMP14. The 21-nucleotide small interfering RNA (siRNA) targeting theencoding region of MMP-14 (31) was chemically syn-thesized (RiboBio Co. Ltd.) and transfected with Gene-silencer Transfection Reagent (Genlantis). The scramblesiRNA (si-Scb) was applied as controls (SupplementaryTable S2).

Cell adhesion assayHomogeneous single-cell suspensions (2 � 104 tumor

cells) were inoculated into each well of 96-well platesthat were precoated with 100 mL of 20 mg/mL Matrigel(BD Biosciences) or 50 mL of 10 mg/L fibronectin (BDBiosciences), and incubated at 37�C in serum-freecomplete medium (pH 7.2) for 1 hour. After incubation,the wells were washed 3 times with PBS and theremaining cells were fixed in 4% paraformaldehydefor 20 minutes at room temperature. The cells werestained with 0.1% crystal violet and washed 3 timeswith PBS to remove free dye. After extraction with 10%acetic acid, absorbance of the samples was measured at570 nm.

Scratch migration assayTumor cellswere cultured in 24-well plates and scraped

with the fine end of 1-mL pipette tips (time 0). Plates werewashed twice with PBS to remove detached cells, andincubatedwith the complete growthmedium. Cellmigra-tion was photographed using 10 high-power fields, at 0and 24 hours postinduction of injury. Remodeling wasmeasured as diminishing distance across the inducedinjury, normalized to the 0 hour control, and expressedas outgrowth (mm).

Matrigel invasion assayThe Boyden chamber technique (Transwell analysis)

was done as previously described (24). Homogeneoussingle-cell suspensions (1 � 105 cells/well) were addedto the upper chambers and allowed to invade for 24 hoursat 37�C in a CO2 incubator. Migrated cells were stainedwith 0.1% crystal violet for 10 minutes at room temper-ature and examined by light microscopy. Quantificationofmigrated cells was done according to published criteria(24).

Tube formation assayFifty microliters of growth factor-reduced Matrigel

were polymerized on 96-well plates. HUVECs wereserum starved in RPMI-1640 medium for 24 hours, sus-pended in RPMI-1640 medium preconditioned withtumor cells, added to the Matrigel-coated wells at thedensity of 5� 104 cells/well, and incubated at 37�C for 18hours. Quantification of antiangiogenic activity was cal-culated by measuring the length of tube walls formedbetween discrete endothelial cells in each well relative tothe control (24).

In vivo growth and metastasis assayAll animal experiments were approved by the Animal

Care Committee of Tongji Medical College (approvalnumber: Y20080290). For the in vivo tumor growthstudies, 2-month-old male nude mice (n ¼ 6 per group)were injected subcutaneously in the upper back with1 � 106 tumor cells stably transfected with empty vectoror pPG-miR-9-EGFP. Six weeks later, mice were sacri-ficed and examined for tumor weight, gene expression,

Zhang et al.

Mol Cancer Ther; 11(7) July 2012 Molecular Cancer Therapeutics1456




and angiogenesis. The experimental metastasis (0.4 � 106

tumor cells per mouse) studies were conducted with2-month-old male nude mice as previously described(32).

Statistical analysisUnless otherwise stated, all data were shown as mean

� SEM. The SPSS 12.0 statistical software (SPSS Inc.)was applied for statistical analysis. The c2 analysis andFisher exact probability analysis were applied for com-parison among the expression of MMP-14, miR-9, andindividual clinicopathologic features. Pearson’s coeffi-cient correlation was applied for analyzing the relation-ship between miR-9 and MMP-14 transcripts. TheKaplan–Meier method was used to estimate survivalrates, and the log-rank test was used to assess survivaldifference. Difference of tumor cells was determined byt test or ANOVA.

ResultsHigh levels of MMP-14 were inversely correlatedwith endogenous miR-9 expression inneuroblastoma tissues and cell linesTo investigate the expression of MMP-14 in neuroblas-

toma, paraffin-embedded sections from 42 well-established primary cases were collected (23). Immuno-histochemical staining with antibodies from differentcompanies revealed that MMP-14 was expressed in thecytoplasm or at the membrane of tumor cells within theneuroblastic nests (Fig. 1A). MMP-14 was also expressedin some ganglionic differentiated tumor cells (Fig. 1A).MMP-14 was detected in 29/42 cases (69.0%) and thestaining was weak in 8, moderate in 8, and intense in13 (Supplementary Table S3). The MMP-14 immunoreac-tivity was significantly higher in neuroblastoma caseswith age more than 1 year (P ¼ 0.005), poorer differenti-ation (P ¼ 0.01), higher MKI (P ¼ 0.03), and higher INSSstages (P ¼ 0.035; Supplementary Table S3). The mediansurvival time (20.6 months) of MMP-14–positive patients(n ¼ 29) was significantly shorter than that (35.4 months)of MMP-14–negative patients (n ¼ 13, P ¼ 0.005; Fig. 1B).In addition, Western blotting and real-time quantitativeRT-PCR were applied to measure the expression levels ofMMP-14 and mature miR-9 in subtotal 20 neuroblastomaspecimens, normal dorsal ganglia, and cultured IMR32,SK-N-AS, SH-SY5Y, and SK-N-SH cell lines. As shownin Fig. 1C and D, higher levels of MMP-14 were observedin neuroblastoma tissues and cell lines than those innormal dorsal ganglia. In contrast, mature miR-9 wasdownregulated in the neuroblastoma tissues and cell linescomparedwith normal dorsal ganglia (Fig. 1E). Therewasan inverse correlation between miR-9 expression andMMP-14 mRNA levels in neuroblastoma tissues (Fig.1F). In situ hybridization further revealed that the miR-9 expression mainly located at the cytoplasm of tumorcells in the neuroblastoma tissues (Supplementary Fig.S1). These results indicated high MMP-14 expression in

primary neuroblastoma tissues and cell lines, which wasinversely correlated with endogenous miR-9 levels.

MMP-14 was a candidate target of miR-9To investigate the hypothesis that miR-9 may influence

theMMP-14 expression in neuroblastoma, computationalpredictionwasdone bymiRNAdatabases. In theMMP-1430-UTR, therewas one potential binding site ofmiR-9withhigh complementarity (Fig. 2A),whichwas coincidentallypredicted by at least 3 independent sources. The miR-9–binding site was located at bases 1,336 to 1,349 of theMMP-14 30-UTR (Fig. 2A).

miR-9 downregulated MMP-14 expression throughposttranscriptional repression

To investigate the direct effects of miR-9 on MMP-14expression in neuroblastoma cell lines, we conducted themiRNA overexpression experiments. Stable transfectionof miR-9 precursor into neuroblastoma cells resulted inincrease of miR-9 levels (Fig. 2B). Western blotting, RT-PCR, and real-time quantitative RT-PCR showed thatstable transfection of miR-9 precursor resulted indecreased protein and transcriptional levels of MMP-14in neuroblastoma cells, when compared with untrans-fected parental cells and those stably transfected withempty vector (mock; Fig. 2C–E). The expression levels ofVEGF, an MMP-14 downstream target gene (33), weresignificantly downregulated in miR-9–overexpressingneuroblastoma cells, consistent with the MMP-14 reduc-tion (Fig. 2C–E). Because the analysis ofmiRNAdatabasesfrom at least 3 independent sources revealed no potentialbinding site of miR-9 within 30-UTR of VEGF, and com-bining the evidence that overexpression or knockdown ofMMP-14 promoted or suppressed theVEGF expression inneuroblastoma cells, respectively (Supplementary Fig.S2), we ruled out the possibility that miR-9 may directlyregulate the VEGF expression. To further examine thesuppressive role of miR-9 on MMP-14 expression, weconducted the miR-9 knockdown experiments by trans-fection of anti-miR-9 or negative control (anti-NC) inhi-bitors into SH-SY5Y and SK-N-SH cells. Transfection ofanti-miR-9 obviously downregulated the miR-9 expres-sion (Supplementary Fig. S3A) and upregulated MMP-14and VEGF protein levels than those of anti-NC (Fig. 2F).Both RT-PCR and real-time quantitative RT-PCR analysesshowed the enhanced transcriptional levels of MMP-14and VEGF in neuroblastoma cells transfected with anti-miR-9, when compared with those transfected with anti-NC (Fig. 2G). Overall, these results showed that miR-9considerably inhibitedMMP-14 expression through post-transcriptional repression.

miR-9 interacted with a putative binding site in theMMP-14 30-UTR

To determine whether or not miR-9 could repressMMP-14 expression by targeting its binding sites in theMMP-14 30-UTR, the PCR products containing intacttarget sites or a mutation of miR-9 seed recognition






Figure 1. MMP-14 was highly expressed and inversely correlated with endogenous levels of miR-9 in neuroblastoma (NB) tissues and cell lines. A, hematoxylin/eosin (HE) and immunohistochemical staining revealed that MMP-14 was expressed in the cytoplasm or at the membrane of tumor cells within theneuroblastic nests andwas also expressed in some ganglionic differentiated tumor cells. Scale bars, 100 mm. B, the Kaplan–Meier methodwas used to estimatesurvival rates and indicated that the median survival time (20.6 months) of MMP-14–positive NB patients was significantly shorter than that (35.4 months)of MMP-14–negative patients. C, Western blotting indicated higher protein levels of MMP-14 in NB tissues (n ¼ 20) and cultured cell lines (IMR32, SK-N-AS,SH-SY5Y, and SK-N-SH) than those in normal dorsal ganglia (DG). D, real-time quantitative RT-PCR revealed higher transcription levels of MMP-14 inNB tissues (n¼ 20) and cultured cell lines (IMR32, SK-N-AS, SH-SY5Y, and SK-N-SH) than those in DG. E, real-time quantitative RT-PCR indicated lowermiR-9levels inNBtissues (n¼20) andculturedcell lines (IMR32,SK-N-AS,SH-SY5Y,andSK-N-SH) than those inDG.F, therewasan inversecorrelationbetweenmiR-9levels and MMP-14 transcription in NB tissues. The symbols (� and D) indicate a significant decrease and a significant increase from DG, respectively.

Zhang et al.





Figure 2. miR-9 downregulated MMP-14 expression through posttranscriptional repression in neuroblastoma cells. A, scheme of the potentialbinding site of miR-9 in the MMP-14 30-UTR, locating at bases 1,336 to 1,349. B, real-time quantitative RT-PCR revealed that stable transfection of miR-9precursor into cultured SH-SY5Y and SK-N-SH cells resulted in enhanced miR-9 levels, when compared with untransfected parental cells and thosestably transfected with empty vector (mock). C, Western blotting indicated that stable transfection of miR-9 precursor resulted in decreased MMP-14 andVEGF protein levels in SH-SY5Y and SK-N-SH cells. D and E, RT-PCR and real-time quantitative RT-PCR revealed the decreased MMP-14 and VEGFtranscription levels in miR-9 precursor–transfected SH-SY5Y and SK-N-SH cells, but not in control and mock cells. F, Western blotting indicated thattransfection of anti-miR-9 inhibitor (100 nmol/L), but not of negative control inhibitor (anti-NC, 100 nmol/L), resulted in increased MMP-14 and VEGF proteinlevels in SH-SY5Y and SK-N-SH cells. G, RT-PCR and real-time quantitative RT-PCR revealed the increasedMMP-14 and VEGF expression in SH-SY5Y andSK-N-SHcells transfectedwith anti-miR-9 inhibitor (100nmol/L), but not in those transfectedwith anti-NC.The symbols � andD indicate a significant decreaseand a significant increase from control or anti-NC, respectively.






sequence (Fig. 3A)were inserted into the luciferase report-er vector. The plasmids were transfected into neuroblas-toma cells stably transfected with empty vector (mock) ormiR-9 precursor. The Renilla luciferase activity normal-ized to that of firefly was significantly reduced inthe tumor cells stably transfected with miR-9 precursor(Fig. 3B), and the effect was abolished by mutating theputativemiR-9–binding sitewithin the 30-UTRofMMP-14(Fig. 3B). Moreover, knockdown ofmiR-9with anti-miR-9

inhibitor increased the luciferase activity in SH-SY5Y andSK-N-SH cells (Fig. 3C), whereas mutation of miR-9 rec-ognition site abolished these effects (Fig. 3C). These resultsindicated that miR-9 directly and specifically interactedwith the target site in the MMP-14 30-UTR.

miR-9 suppressed the adhesion, migration, invasion,and angiogenesis of neuroblastoma cells in vitro

Because previous studies indicate that MMP-14 pro-motes the invasion and angiogenesis of tumor cells (4, 5),we further investigated the effects of miR-9 overexpres-sion and MMP-14 restoration on cultured neuroblastomacells. Western blotting indicated that transfection ofMMP-14 rescued the miR-9–induced downregulation ofMMP-14 (Fig. 4A). In adhesion assay, tumor cells stablytransfectedwithmiR-9 precursor possessed thedecreasedability in adhesion to the precoated Matrigel or fibronec-tin, when compared with those stably transfectedwith empty vector (mock; Fig. 4B). In scratch migrationassay, miR-9 overexpression attenuated the migrationcapabilities of SH-SY5Y and SK-N-SH cells (Fig. 4C; Sup-plementary Fig. S4). Transwell analysis showed that neu-roblastoma cells stably transfected with miR-9 precursorpresented an impaired invasion capacity than mock cells(Fig. 4D). The tube formation of endothelial cells wassuppressed by treatment with the medium precondi-tioned by stable transfection of neuroblastoma cells withmiR-9 precursor (Fig. 4E). In addition, transfection ofMMP-14 into SH-SY5Y and SK-N-SH cell lines restoredthe decrease in adhesion, migration, invasion, and angio-genesis induced by stable overexpression of miR-9 (Fig.4B–E; Supplementary Fig. S4). On the other hand, weexamined the effects of miR-9 knockdown on neuroblas-toma cells. Introduction of anti-miR-9 inhibitor intoneuroblastoma cells resulted in enhanced abilities inadhesion (Supplementary Fig. S3B), migration (Supple-mentary Fig. S3C), invasion (Supplementary Fig. S3D),and angiogenic capabilities (Supplementary Fig. S3E).These results indicated that miR-9 remarkably decreasedthe adhesion, migration, invasion, and angiogenesis ofneuroblastoma cells in vitro.

Overexpression of miR-9 attenuated the growth,metastasis, and angiogenesis of neuroblastoma cellsin vivo

We next investigated the efficacy of miR-9 againsttumor growth, metastasis, and angiogenesis in vivo.Stable transfection of miR-9 precursor into SH-SY5Ycells resulted in decreased growth and tumor weightof subcutaneous xenograft tumors in athymic nudemice, when compared with those stably transfectedwith empty vector (mock; Fig. 5A and B). In addition,the expression of MMP-14 and downstream VEGF wasalso reduced by stable transfection of miR-9 precursor(Fig. 5C). Moreover, stable transfection of miR-9 pre-cursor resulted in a decrease in CD31-positive meanvessel density within tumors (Fig. 5C). In the experi-mental metastasis studies, SH-SY5Y cells stably

Figure 3. miR-9 directly interactedwith a putative binding site in theMMP-14 30-UTR. A, scheme and sequence of the intact miR-9–binding site(wild-type; WT) and its mutation (Mut) within the luciferase reportervector. B, stable transfection of miR-9 precursor into SH-SY5Y and SK-N-SH cells resulted in decreased luciferase activities of MMP-14 30-UTRreporter, whencomparedwith those stably transfectedwith empty vector(mock). These effectswere abolishedbyamutation in theputativemiR-9–binding site within the 30-UTR of MMP-14. C, transfection of anti-miR-9(100 nmol/L) inhibitor into SH-SY5Y and SK-N-SH increased theluciferase activity when compared with those transfected with negativecontrol inhibitor (anti-NC, 100 nmol/L), whereas mutation of miR-9recognition site abolished these effects. The symbols � and D indicate asignificant decrease and a significant increase from mock or anti-NC,respectively.

Zhang et al.





Figure 4. Overexpression of miR-9 abolished the adhesion, migration, invasion, and angiogenesis of neuroblastoma cells in vitro. A, Western blottingindicated that transfection of MMP-14 restored the downregulation of MMP-14 induced by stable miR-9 overexpression. B, stable transfection ofmiR-9 precursor into SH-SY5Y and SK-N-SH cells resulted in decreased adhesion to Matrigel or fibronectin, when compared with those stablytransfected with empty vector (mock). In addition, transfection of MMP-14 restored the cell adhesion in miR-9–overexpressing tumor cells. C, inscratch migration assay, the migration of miR-9–overexpressing SH-SY5Y and SK-N-SH cells was significantly reduced when compared with mock.Transfection of MMP-14 rescued the migration of miR-9–overexpressing cells. D, Matrigel invasion assay indicated the decreased invasioncapabilities of miR-9–overexpressing SH-SY5Y and SK-N-SH cells than those of mock cells. However, transfection of MMP-14 restored theinvasion of miR-9–overexpressing cells. E, the tube formation of endothelial HUVEC cells was suppressed by treatment with the mediumpreconditioned by miR-9–overexpressing SH-SY5Y and SK-N-SH cells, when compared with that of mock cells. Transfection of MMP-14 rescuedthe angiogenic capabilities of miR-9–overexpressing cells. The symbols � and D indicate a significant decrease and a significant increase from mock,respectively.






transfected with miR-9 precursor established statistical-ly fewer lung metastatic colonies than mock group (Fig.5D). These results suggested that miR-9 could inhibitthe growth, metastasis, and angiogenesis of neuroblas-toma cells in vivo.

Knockdown of MMP-14 suppressed the adhesion,migration, invasion, and angiogenesis ofneuroblastoma cells

Because above results indicated the negative regula-tion of MMP-14 expression by miR-9, we hypothesizedthat knockdown of MMP-14 should have a similar effecton cultured neuroblastoma cells. siRNAs targeting the

encoding region of MMP-14 were designed and trans-fected into SH-SY5Y and SK-N-SH cells. Transfection ofsi-MMP-14, but not of si-Scb, resulted in decreasedMMP-14 expression in neuroblastoma cells (Fig. 6A).Knockdown of MMP-14 suppressed the adhesion,migration, invasion, and angiogenesis of SH-SY5Yand SK-N-SH cells (Fig. 6B–E). These results were con-sistent with the findings that overexpression of miR-9suppressed the adhesion, migration, invasion, andangiogenesis of neuroblastoma cells in vitro, providingfurther evidence that MMP-14 was involved in miR-9-mediated suppression of neuroblastoma. Accordingly,identification of MMP-14 as a miR-9 target gene may

Figure 5. Overexpression ofmiR-9 attenuated the growth,metastasis, and angiogenesis ofneuroblastoma cells in vivo. A andB, hypodermic injection of SH-SY5Y cells into athymic nude miceestablished subcutaneousxenograft tumors. Six weeks later,mice (n ¼ 6) from each group weresacrificed. Stable transfection oftumor cells with miR-9 precursorresulted in decreased tumor size,and the mean tumor weight formedfrom miR-9–overexpressing cellswas significantly decreased.C, hematoxylin/eosin (HE) andimmunohistochemical stainingrevealed that stable transfection ofmiR-9 precursor resulted indecreased expression of MMP-14,VEGF, and CD31 within tumors.The mean vessel density withintumors decreased after stabletransfection of miR-9 precursor.Scale bars, 100 mm. D, SH-SY5Ycells were injected into the tail veinof athymic nude mice (0.4 � 106

cells per mouse, n ¼ 6 for eachgroup). Tumor cells stablytransfected with miR-9 precursorestablished significantly fewermetastatic colonies. Scale bars,100 mm. The asterisk indicates asignificant decrease from emptyvector (mock).

Zhang et al.





Figure 6. Knockdown of MMP-14 suppressed the adhesion, migration, invasion, and angiogenesis of neuroblastoma cells. A, Westernblotting indicated that transfection of si-MMP-14 (100 nmol/L), but not of si-Scb (100 nmol/L), into SH-SY5Y and SK-N-SH cells, resulted indecreased MMP-14 expression when compared with untransfected cells (control). B, the adhesion of SH-SY5Y and SK-N-SH cells to Matrigel orfibronectin was significantly attenuated by transfection of si-MMP-14, but not of si-Scb, when compared with control cells. C, in scratch migrationassay, the migration of si-MMP-14–transfected SH-SY5Y and SK-N-SH cells was significantly reduced when compared with control andthose transfected with si-Scb. D, Matrigel invasion assay indicated the decreased invasion capabilities of MMP-14–knockdown SH-SY5Y andSK-N-SH cells than those of control cells. E, the tube formation of endothelial HUVEC cells was suppressed by treatment with the mediumpreconditioned by MMP-14–knockdown SH-SY5Y and SK-N-SH cells, when compared with that of control cells. The asterisk indicates asignificant decrease from control.






explain, at least in part, why overexpression of miR-9suppressed the migration, invasion, and angiogenesis ofneuroblastoma cells.

DiscussionThe mature human miR-9 transcript is encoded by 3

independent genes,miR-9-1,miR-9-2, andmiR-9-3, whichlocate on chromosomes 1, 5, and15, respectively (34).miR-9 is one of the crucial regulators of neuronal development,neuronal stem cell fate determination, and migration ofneural progenitors (12), and is upregulated in oligoden-droglioma (13), glioblastoma (14), and gliomas (15). How-ever, in colon cancer, lung cancer, breast cancer, andmelanoma, the promoters of 2 miR-9 genes are aberrantlyhypermethylated, resulting in undetectable miR-9 tran-scripts (21, 35). Downregulation of miR-9 is also noted inpancreatic cancer (36), gastric cancer (37), and ovariancancer (18). Altered miR-9 expression is associated withthe malignant progression and metastasis in liver cancer(19), breast cancer (20), and colorectal cancer (21). In thisstudy, we showed the downregulation of miR-9 in pri-mary neuroblastoma tissues and cell lines, which was inline with previous studies (22). It is proposed that thedownregulated miR-9 expression may be not because ofchromosome aberrations, because the miR-9 gene family(miR-9-1, miR-9-2, and miR-9-3) is not positioned withinthe chromosomal regions rearranged in neuroblastoma(38). Importantly, we noted the inverse correlationbetween miR-9 levels and MMP-14 expression in neuro-blastoma tissues, suggesting that MMP-14 expressionmay be negatively regulated by miR-9.

On the basis of the base pairing betweenmiRNAand 30-UTR of target gene, computational algorithms have beenthe major methods in predicting miRNA targets (39).Although bioinformatics reveal over 1,000 target geneswith predicted miR-9 seed sites in their 30-UTR, the pres-ence of a conserved seed is not sufficient to indicate a truebiologic interaction between miR-9 and putative targetgenes (40). In this study, our experimental evidenceshowed that MMP-14 was a target of miR-9. First, theability ofmiR-9 to regulateMMP-14 expressionwas likelydirect because of its high complementarity to the 30-UTRof MMP-14 mRNA. Second, the activities of MMP-14 30-UTR luciferase reporter were responsive to miR-9 over-expression. Third, mutation of the miR-9–binding siteabolished the regulatory effects of miR-9 on the MMP-14 30-UTR luciferase reporter. Fourth, knockdown ofmiR-9 with anti-miR-9 inhibitor increased the activities ofMMP-14 30-UTR luciferase reporter. Finally, endogenousMMP-14 expression, both mRNA and protein, wasdecreased in miR-9 precursor-transfected neuroblastomacells, suggesting thatmiR-9may regulateMMP-14 expres-sion by inducing mRNA degradation and/or translation-al suppression.

The adhesion of tumor cells to ECM is a key step in theinitial process of migration and invasion. MMP-14 en-hances cell attachment tomatrix, suggesting thatMMP-14

can facilitate cell attachment at a new site for metastasis,aside from its leading role in matrix degradation (41).Matrigel, a solubilized basement membrane preparationextracted from EHS mouse sarcoma, resembles the bio-logically active matrix, and its major components includelaminin, collagen IV, heparan sulfate proteoglycans, andentactin (42). In this study, we showed that miR-9 inhib-ited the MMP-14 expression and reduced the adhesion ofneuroblastoma cells to Matrigel or fibronectin. It has beenestablished that MMP-14–mediated ECM degradation atcell–matrix adhesion facilitates the focal adhesion turn-over, which regulates integrin-generated signal trans-duction and subsequent cell migration (43). MMP-14also regulates cell–ECM interaction by processing celladhesion molecule CD44, and eventually promotes cellmigration (44). We believe that the dynamic attachmentand detachment of neuroblastoma cells to ECM is intri-cately regulated for cell migration, and the underlyingmechanisms for MMP-14–promoted cell–ECM adhesionwarrant our further investigation. MMP-14, but notother collagenases, can promote the invasion of epithelialcells, fibroblasts, and cancer cells (5). MMP-14–mediateddegradation of ECM occurs throughout the angiogenicprocess and contributes to vascular regression (4). Fur-thermore, MMP-14 promotes tumor growth and angio-genesis through upregulating the protein and mRNAexpression of VEGF (33). Multiple lines of preclinicalevidence have shown the linkage between high MMP-14 expression and cancer progression, such as lymphnode metastases, invasion, poor clinical stage, largertumor size, and increasing tumor stage (45). In this study,we showed that high MMP-14 expression in neuroblas-toma was correlated with clinicopathologic featuresand shorter patients’ survival time. Thus, in light of theemerging pivotal role of MMP-14 in cancer progression,the miR-9–mediated MMP-14 inhibition seems attractiveas a strategy to suppress the tumor growth, invasion,and metastases of neuroblastoma.

Previous studies suggest that the function of miR-9 istumor-type specific. Overexpression of miR-9 suppressesthe in vitro and in vivo growth of ovarian cancer cellsthrough downregulating the expression of nuclear factorkB1 (46). However, miR-9 can directly target its bindingsite in the caudal-type homeobox 2 (CDX2) 30-UTR, result-ing in blockage of CDX2 protein translation in gastriccancer cells (47), although knockdown of miR-9 inhibitsthe proliferation of gastric cancer cells, which is similarto the effects of CDX2 overexpression (47). In addition,miR-9 increases the invasion and epithelial mesenchymaltransition through targeting E-cadherin in hepatic cancer(48) and breast cancer cell lines (32). When breast cancercells overexpressing miR-9 are implanted in mice, thetumors show enhanced angiogenesis and growth thanthose formed by miR-9 low-expressing cells (32). In colo-rectal cancer cells, overexpression of miR-9 promotescell migration and cytoskeleton reorganization throughdownregulating the a-catenin expression (49). Thus,miR-9 seems to be a useful marker for tumor metastasis, but

Zhang et al.





its role in this process is also dependent on the type ofcancer. In this study, we showed that overexpressionof miR-9 attenuated the invasion, metastasis, and angio-genesis of neuroblastoma cells, which was similar to thatof MMP-14 knockdown, suggesting the potential appli-cation of miR-9 as a target for the therapeutics ofneuroblastoma.In summary, we have shown that miR-9 expression is

downregulated in human neuroblastoma, and overex-pression of miR-9 inhibits the invasion, metastasis, andangiogenesis of neuroblastoma cells in vitro and in vivo.Furthermore, miR-9 suppresses the MMP-14 expressionvia the binding site in the 30-UTR in neuroblastomacell lines. This study extends our knowledge aboutthe regulation of MMP-14 at the posttranscriptionallevel by miRNA, and suggests that miR-9 may be ofpotential values as novel therapeutic target for humanneuroblastoma.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: L. Zheng, Q. TongDevelopment of methodology: H. Zhang, M. Qi, S. Li, T. Qi, H. Mei, K.HuangAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): H. Zhang, M. Qi, S. LiAnalysis and interpretation of data (e.g., statistical analysis, biostatis-tics, computational analysis): H. Zhang, M. Qi, S. Li, L. Zheng, Q. TongWriting, review, and/or revision of the manuscript: L. Zheng, Q. TongAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): T. Qi, H. Mei, K. HuangStudy supervision: L. Zheng, Q. Tong

Grant SupportL. Zheng is supported by the National Natural Science Foundation of

China (30600278, 81071997). Q. Tong is supported by the National NaturalScience Foundation of China (30200284, 30772359, 81072073), Program forNew Century Excellent Talents in University (NCET-06-0641), ScientificResearch Foundation for the Returned Overseas Chinese Scholars (2008-889), and Fundamental Research Funds for the Central Universities(2010JC025).

The costs of publication of this article were defrayed in part by the pay-mentofpagecharges.Thisarticlemust thereforebeherebymarked advertise-ment in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received January 1, 2012; revised April 2, 2012; accepted April 17, 2012;published OnlineFirst May 7, 2012.

References1. Ara T, DeClerck YA. Mechanisms of invasion andmetastasis in human

neuroblastoma. Cancer Metastasis Rev 2006;25:645–7.2. Le XF, Merchant O, Bast RC, Calin GA. The roles of microRNAs in the

cancer invasion-metastasis cascade. Cancer Microenviron 2010;3:137–47.

3. Chun TH, Sabeh F, Ota I, Murphy H, McDonagh KT, Holmbeck K,et al. MT1-MMP-dependent neovessel formation within the con-fines of the three-dimensional extracellular matrix. J Cell Biol2004;167:757–67.

4. Aplin AC, Zhu WH, Fogel E, Nicosia RF. Vascular regression andsurvival are differentially regulated by MT1-MMP and TIMPs in theaortic ringmodel of angiogenesis. Am J Physiol Cell Physiol 2009;297:C471–80.

5. Sabeh F, Ota I, Holmbeck K, Birkedal-Hansen H, Soloway P, BalbinM,et al. Tumor cell traffic through the extracellular matrix is controlled bythe membrane-anchored collagenase MT1-MMP. J Cell Biol2004;167:769–81.

6. Gabriely G, Wurdinger T, Kesari S, Esau CC, Burchard J, Linsley PS,et al. MicroRNA 21 promotes glioma invasion by targeting matrixmetalloproteinase regulators. Mol Cell Biol 2008;28:5369–80.

7. Meng F, Henson R, Wehbe-Janek H, Ghoshal K, Jacob ST, Patel T.MicroRNA-21 regulates expression of the PTEN tumor suppressorgene in human hepatocellular cancer. Gastroenterology 2007;133:647–58.

8. Zhu S, Si ML, Wu H, Mo YY. MicroRNA-21 targets the tumorsuppressor gene tropomyosin 1 (TPM1). J Biol Chem 2007;282:14328–36.

9. Lu Z, Liu M, Stribinskis V, Klinge CM, Ramos KS, Colburn NH, et al.MicroRNA-21 promotes cell transformation by targeting the pro-grammed cell death 4 gene. Oncogene 2008;27:4373–9.

10. Aigner A. MicroRNAs (miRNAs) in cancer invasion and metastasis:therapeutic approaches based on metastasis-related miRNAs. J MolMed 2011;89:445–57.

11. Lin SL, Chiang A, Chang D, Ying SY. Loss of mir-146a function inhormone-refractory prostate cancer. RNA 2008;14:417–24.

12. Krichevsky AM, Sonntag KC, Isacson O, Kosik KS. Specific micro-RNAs modulate embryonic stem cell-derived neurogenesis. StemCells 2006;24:857–64.

13. Nelson PT, Baldwin DA, Kloosterman WP, Kauppinen S, Plasterk RH,Mourelatos Z. RAKE and LNA-ISH reveal microRNA expression andlocalization in archival human brain. RNA 2006;12:187–91.

14. Schraivogel D, Weinmann L, Beier D, Tabatabai G, Eichner A, Zhu JY,et al. CAMTA1 is a novel tumour suppressor regulated by miR-9/9� inglioblastoma stem cells. EMBO J 2011;30:4309–22.

15. Malzkorn B,WolterM, Liesenberg F, Grzendowski M, St€uhler K,MeyerHE, et al. Identification and functional characterization of microRNAsinvolved in the malignant progression of gliomas. Brain Pathol2010;20:539–50.

16. Nass D, Rosenwald S, Meiri E, Gilad S, Tabibian-Keissar H,Schlosberg A, et al. MiR-92b and miR-9/9� are specificallyexpressed in brain primary tumors and can be used to differentiateprimary from metastatic brain tumors. Brain Pathol 2009;19:375–83.

17. Yanaihara N, Caplen N, Bowman E, Seike M, Kumamoto K, Yi M, et al.Unique microRNA molecular profiles in lung cancer diagnosis andprognosis. Cancer Cell 2006;9:189–98.

18. Laios A, O'Toole S, Flavin R, Martin C, Kelly L, Ring M, et al. Potentialrole of miR-9 and miR-223 in recurrent ovarian cancer. Mol Cancer2008;7:35.

19. Budhu A, Jia HL, Forgues M, Liu CG, Goldstein D, Lam A, et al.Identification of metastasis-related microRNAs in hepatocellular car-cinoma. Hepatology 2008;47:897–907.

20. Iorio MV, Ferracin M, Liu CG, Veronese A, Spizzo R, Sabbioni S, et al.MicroRNA gene expression deregulation in human breast cancer.Cancer Res 2005;65:7065–70.

21. Bandres E, Agirre X, Bitarte N, Ramirez N, Zarate R, Roman-Gomez J,et al. Epigenetic regulation of microRNA expression in colorectalcancer. Int J Cancer 2009;125:2737–43.

22. Laneve P, Di Marcotullio L, Gioia U, Fiori ME, Ferretti E, Gulino A,et al. The interplay between microRNAs and the neurotrophinreceptor tropomyosin-related kinase C controls proliferation ofhuman neuroblastoma cells. Proc Natl Acad Sci U S A 2007;104:7957–62.

23. Zheng LD, TongQS, Tang ST, Du ZY, Liu Y, JiangGS, et al. Expressionand clinical significance of heparanase in neuroblastoma. World JPediatr 2009;5:206–10.

24. Zheng L, Jiang G, Mei H, Pu J, Dong J, Hou X, et al. Small RNAinterference-mediated gene silencing of heparanase abolishes theinvasion, metastasis and angiogenesis of gastric cancer cells. BMCCancer 2010;10:33.

25. John B, Enright AJ, Aravin A, Tuschl T, Sander C, Marks DS. HumanmicroRNA targets. PLoS Biol 2004;2:e363.






26. Wang X. miRDB: a microRNA target prediction and functionalannotation database with a wiki interface. RNA 2008;14:1012–7.

27. Miranda KC, Huynh T, Tay Y, Ang YS, Tam WL, Thomson AM, et al. Apattern-based method for the identification of microRNA bindingsites and their corresponding heteroduplexes. Cell 2006;126:1203–17.

28. Garcia DM, Baek D, Shin C, Bell GW, Grimson A, Bartel DP. Weakseed-pairing stability and high target-site abundance decrease theproficiency of lsy-6 and other microRNAs. Nat Struct Mol Biol2011;18:1139–46.

29. Dweep H, Sticht C, Pandey P, Gretz N. miRWalk-database: predictionof possible miRNA binding sites by "walking" the genes of threegenomes. J Biomed Inform 2011;44:839–47.

30. Griffiths-Jones S. The microRNA registry. Nucleic Acids Res 2004;32:D109–11.

31. Ries C, Egea V, Karow M, Kolb H, Jochum M, Neth P. MMP-2, MT1-MMP, and TIMP-2 are essential for the invasive capacity of humanmesenchymal stem cells: differential regulation by inflammatory cyto-kines. Blood 2007;109:4055–63.

32. Ma L, Young J, Prabhala H, Pan E, Mestdagh P, Muth D, et al. miR-9, aMYC/MYCN-activated microRNA, regulates E-cadherin and cancermetastasis. Nat Cell Biol 2010;12:247–56.

33. Sounni NE, Devy L, Hajitou A, Frankenne F, Munaut C, Gilles C, et al.MT1-MMP expression promotes tumor growth and angiogenesisthrough an up-regulation of vascular endothelial growth factor expres-sion. FASEB J 2002;16:555–64.

34. Hildebrandt MA, Gu J, Lin J, Ye Y, Tan W, Tamboli P, et al. Hsa-miR-9methylation status is associated with cancer development and met-astatic recurrence in patients with clear cell renal cell carcinoma.Oncogene 2010;29:5724–8.

35. Lujambio A, Calin GA, Villanueva A, Ropero S, S�anchez-C�espedes M,Blanco D, et al. A microRNA DNA methylation signature for humancancer metastasis. Proc Natl Acad Sci U S A 2008;105:13556–61.

36. Omura N, Li CP, Li A, Hong SM, Walter K, Jimeno A, et al. Genome-wide profiling atmethylated promoters in pancreatic adenocarcinoma.Cancer Biol Ther 2008;7:1146–56.

37. Luo H, Zhang H, Zhang Z, Zhang X, Ning B, Guo J, et al. Down-regulated miR-9 and miR-433 in human gastric carcinoma. J Exp ClinCancer Res 2009;28:82.

38. Tonini GP, Romani M. Genetic and epigenetic alterations in neuro-blastoma. Cancer Lett 2003;197:69–73.

39. Lewis BP, Shih IH, Jones-Rhoades MW, Bartel DP, Burge CB.Prediction of mammalian microRNA targets. Cell 2003;115:787–98.

40. Zhang J, Chintalgattu V, Shih T, Ai D, Xia Y, Khakoo AY.MicroRNA-9 is an activation-induced regulator of PDGFR-beta expression in cardiomyocytes. J Mol Cell Cardiol 2011;51:337–46.

41. Ip YC, Cheung ST, Leung KL, Fan ST. Mechanism of metastasis bymembrane type 1-matrix metalloproteinase in hepatocellular carcino-ma. World J Gastroenterol 2005;11:6269–76.

42. Benton G, Kleinman HK, George J, Arnaoutova I. Multipleuses of basement membrane-like matrix (BME/Matrigel) in vitroand in vivo with cancer cells. Int J Cancer 2011;128:1751–7.

43. Takino T, Saeki H, Miyamori H, Kudo T, Sato H. Inhibition of mem-brane-type 1 matrix metalloproteinase at cell-matrix adhesions. Can-cer Res 2007;67:11621–9.

44. Kajita M, Itoh Y, Chiba T, Mori H, Okada A, Kinoh H, et al. Membrane-type 1 matrix metalloproteinase cleaves CD44 and promotes cellmigration. J Cell Biol 2001;153:893–904.

45. Jiang WG, Davies G, Martin TA, Parr C, Watkins G, Mason MD, et al.Expression of membrane type-1 matrix metalloproteinase, MT1-MMPin human breast cancer and its impact on invasiveness of breastcancer cells. Int J Mol Med 2006;17:583–90.

46. Guo LM, Pu Y, Han Z, Liu T, Li YX, Liu M, et al. MicroRNA-9 inhibitsovarian cancer cell growth through regulation of NF-kB1. FEBS J2009;276:5537–46.

47. Rotkrua P, Akiyama Y, Hashimoto Y, Otsubo T, Yuasa Y. MiR-9downregulates CDX2 expression in gastric cancer cells. Int J Cancer2011;129:2611–20.

48. Tan HX, Wang Q, Chen LZ, Huang XH, Chen JS, Fu XH, et al. Micro-RNA-9 reduces cell invasion and E-cadherin secretion in SK-Hep-1cell. Med Oncol 2010;27:654–60.

49. Zhu L, Chen H, Zhou D, Li D, Bai R, Zheng S, et al. MicroRNA-9 up-regulation is involved in colorectal cancer metastasis via promotingcell motility. Med Oncol 2012;29:1037–43.


Zhang et al.




2012;11:1454-1466. Published OnlineFirst May 7, 2012.Mol Cancer Ther Huanyu Zhang, Meng Qi, Shiwang Li, et al. Metastasis, and Angiogenesis of Neuroblastoma CellsmicroRNA-9 Targets Matrix Metalloproteinase 14 to Inhibit Invasion,

Updated version

10.1158/1535-7163.MCT-12-0001doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2012/05/07/1535-7163.MCT-12-0001.DC1

Access the most recent supplemental material at:

Cited articles

http://mct.aacrjournals.org/content/11/7/1454.full#ref-list-1

This article cites 49 articles, 13 of which you can access for free at:

Citing articles

http://mct.aacrjournals.org/content/11/7/1454.full#related-urls

This article has been cited by 5 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://mct.aacrjournals.org/content/11/7/1454To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-12-0001

http://mct.aacrjournals.org/content/suppl/2012/05/07/1535-7163.MCT-12-0001.DC1

http://mct.aacrjournals.org/content/11/7/1454.full#ref-list-1

http://mct.aacrjournals.org/content/11/7/1454.full#related-urls

http://mct.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mct.aacrjournals.org/content/11/7/1454


Documents

microRNA-9 Targets Matrix Metalloproteinase 14 to Inhibit ... · Huanyu Zhang 1, Meng Qi , Shiwang Li 1, Teng Qi , Hong Mei , Kai Huang2,4, Liduan Zheng3,4, and Qiangsong Tong1,4