let-7a

Carcinogenesis vol.32no.5pp.713–722,2011

doi:10.1093/carcin/bgr035

Advance Access publication February24,2011

Low-level expression of let-7a in gastric cancer and its involvement in tumorigenesis by targeting RAB40C

Qiaoyuan Yang1,2,y,Zhigang Jie3,y,Hong Cao4,

Anne R.Greenlee5,Chengfeng Yang6,Fei Zou1and Yiguo Jiang1,2,?

1School of Public Health and Tropical Medicine,Southern Medical University,18318North Guangzhou Avenue,Guangzhou510515,People’s Republic of China,2Institute for Chemical Carcinogenesis,State Key Laboratory of Respiratory Disease,Guangzhou Medical University,195 Dongfengxi Road,Guangzhou510182,People’s Republic of China,

3Department of General Surgery,First Af?liated Hospital,Nanchang University,Nanchang330006,People’s Republic of China,4Department of General Surgery,People’s Hospital of Jiangxi Province,Nanchang330006, People’s Republic of China,5The Jackson Laboratory,Bar Harbor,ME04609, USA and6Department of Physiology and Center for Integrative Toxicology, Michigan State University,East Lansing,MI48824,USA

?To whom correspondence should be addressed.Tel:t862081340186; Fax:t862081340724;

Email:jiangyiguo@https://www.wendangku.net/doc/896168516.html,

Correspondence may also be addressed to Fei Zou.Tel:t862061648301; Fax:t862061648324;

Email:zoufei26@https://www.wendangku.net/doc/896168516.html,

Gastric cancer is the fourth most common cancer and the second leading cause of cancer mortality worldwide but the underlying molecular mechanism is not entirely clear.The objective of this study was to explore the role of let-7a microRNA(miRNA)in gastric tumorigenesis and the possible correlation between RAB40C and let-7a miRNA in gastric cancer.We found that ex-pression of let-7a is reduced in human gastric cancer tissues and cell lines and there was a signi?cant correlation between the level of let-7a expression and the stage of differentiation.Overexpres-sion of let-7a resulted in a decrease in cell proliferation and G1arrest,signi?cantly suppressed anchorage-dependent growth in vitro and the tumorigenicity of gastric cancer cells in a nude mouse xenograft model.Furthermore,we demonstrated that RAB40C is regulated directly by let-7a and plays an essential role as a mediator of the biological effects of let-7a in gastric tumori-genesis.This study revealed that let-7a is signi?cant in suppress-ing gastric cancer growth in vivo and in vitro and provided the?rst evidence that RAB40C is negatively regulated by let-7a at the posttranscriptional level via binding to the3#-untranslated region of RAB40C messenger RNA in gastric cancer.The results of this study suggest that let-7a and RAB40C are potentially useful targets for gastric cancer diagnosis and therapy.

Introduction

Gastric cancer is the fourth most common cancer and the second leading cause of cancer mortality worldwide despite a decreasing incidence in recent decades(1).It remains an important public health burden worldwide,especially in developing countries.In China,gastric cancer has the highest mortality among all cancers and the overall mortality rate has increased steadily in the past20 years(2).However,the molecular mechanisms involved in gastric cancer are diverse,complex and not fully understood.

New opportunities in the study of cancer molecular mechanisms have been provided by the discovery of microRNAs(miRNAs),a class of short non-coding endogenous RNAs that function as nega-tive regulators of gene expression(3).As the major endogenous trig-gers for posttranscriptional silencing,miRNAs can negatively regulate the expression of a protein-coding gene by binding with the3#-untranslated regions(3#-UTRs)of their messenger RNA (mRNA)targets and then repressing expression of the target gene through mRNA degradation or translational inhibition(4,5).miRNAs are predicted to target more than one-third of human genes and each miRNA can control hundreds of target genes(6).Moreover,miRNAs have been demonstrated to be evolutionarily conserved and to perform regulatory functions in numerous biological processes,including developmental timing,cell proliferation,apoptosis,metabolism,cell differentiation and morphogenesis(7–9).

Recently acquired evidence demonstrates that miRNAs can be regulators in carcinogenesis.Calin et al.(10)showed that.50% of the known mature human miRNA genes are located in cancer-associated genomic regions or in fragile sites,suggesting that miRNAs might have an important role in the pathogenesis of human cancers.Moreover,different cancer types have distinct miRNA ex-pression pro?les,and an increasing number of miRNAs have been suggested to have important roles in tumor progression or in tumor suppression(11–13).Increased expressions of some miRNAs,such as miR-21and miR-27a,have been found to play crucial roles in gastric tumors(14,15).In addition,the miR-106b-25cluster,which is upregulated in human gastric tumors,is involved in the posttran-scriptional regulation of transcription factor E2F1(16)and miR-15b and miR-16modulate multi-drug resistance by targeting B-cell lymphoma/leukmia-2(BCL2)in human gastric cancer cells(17). In contrast,miR-9,miR-141,miR-143,miR-145,miR-433and miR-451are downregulated in gastric cancer and these miRNAs act as anti-oncogenic miRNAs with a signi?cant growth inhibitory effect on gastric cancer(18–21).

Among all human cancer-related miRNAs,the let-7family has attracted the most interest because its family members have been noted to express aberrantly in human cancers(22,23).The family was discovered initially in Caenorhabditis elegans and is currently one of the most important members of the miRNA family.The let-7 family consists of11very closely related genes and many human let-7genes map to regions that are altered or deleted in human tumors, indicating that these genes might function as tumor suppressors(22). Moreover,when overexpressed in colon cancer cells,let-7miRNA leads to growth proliferation associated with a reduced level of RAS protein(24).let-7a is downregulated in Burkitt’s lymphoma and it has been shown to be an anticancer miRNA that repressed C-MYC expression at the translational level(25).Recently,the implication of let-7in carcinogenesis has been extended to the repression of high-mobility group A2,thus preventing oncogenic transformation in many tumors(26,27).These?ndings suggest that let-7miRNAs participate actively in tumorigenic processes and the targets in-volved in the regulation of let-7miRNAs have been associated with various tumorigenic processes in addition to the miRNAs them-selves.However,the data for the relationship between gastric carcinogenesis and the expression of let-7a miRNA are very limited. Evidence collected to date shows let-7a was linked to the modula-tion of different target genes,the most well-known being the RAS family.The RAS proteins function as the critical molecular switch for various signaling pathways controlling the diverse biological processes.RAB40C is a member of the RAS family,which plays important roles in tumorigenesis.With the help of a bioinformatic analysis,we found RAB40C contained the let-7a binding site and was evolutionarily conserved across10species.To our knowledge, there is no report of work investigating the role of let-7a or a possible correlation between RAB40C and let-7a miRNA in gastric cancers.

Abbreviations:CCK-8,cell counting kit-8;DMEM,Dulbecco’s modi?ed

Eagle’s medium;inhibitor NC,inhibitor non-speci?c control miRNA;mimic

NC,mimic non-speci?c control miRNA;mRNA,messenger RNA;miRNA,

microRNA;PCR,polymerase chain reaction;RT,reverse transcription;siRNA,

small interfering RNA;3#-UTR,3#-untranslated region.

y These authors contributed equally to this work.

óThe Author2011.Published by Oxford University Press.All rights reserved.For Permissions,please email:journals.permissions@https://www.wendangku.net/doc/896168516.html,713

In this study,we used the quantitative real-time polymerase chain reaction(PCR)to examine the expression of let-7a mature miRNA in 27matched pairs of normal and gastric cancer tissues from patients.In comparison with normal tissues,we found that the expression of let-7a was signi?cantly lower in gastric tumor specimens.Importantly,in-creased expression of let-7a suppressed cell growth in vitro and tumor growth in vivo and was associated with decreased rates of cell pro-liferation and the cell cycle.This is the?rst report that let-7a is involved in tumorigenesis of gastric cancer both in vitro and in vivo.Further-more,it is con?rmed for the?rst time that let-7a targets RAB40C di-rectly.By targeting RAB40C,let-7a suppresses proliferation and anchorage-independent growth of human gastric cancer cells.Our?nd-ings suggest that let-7a and RAB40C might be valuable tools for de-veloping interventions aimed at treating and diagnosing gastric cancer. Materials and methods

Patients and tissue samples

In2008–09,27pairs of gastric cancer tissues and matched normal gastric tissues were obtained with informed consent from patients in the First Af?liated Hospital of Nanchang University(Nanchang,China).All tissue samples were con?rmed histopathologically and snap-frozen in liquid nitrogen.The non-cancerous gastric tissues were taken3cm away from the tumor.Clinicopatho-logic information was available for all samples and the study was approved by the Medical Ethics Committee of Nanchang University and the Medical Ethics Committee of Guangzhou Medical University.

Cell lines and reagents

The human gastric epithelial cell line GES-1and human gastric cancer cell lines SGC-7901,BGC-823and HGC-27were purchased from the Cell Resource Center of XiangYa Central Laboratory(Changsha,China).Human gastric cancer cell line MKN-28was provided by the Laboratory Animal Center of the Fourth Military Medical University(Xian,China).Human gastric cancer cell line AGS was donated by the Sun Yat-Sen University Cancer Center(Guangzhou,China). All cells,except AGS cells,were grown in Dulbecco’s modi?ed Eagle’s medium (DMEM)(Gibco,Carlsbad,CA)supplemented with10%(vol/vol)fetal bovine serum,50U/ml penicillin and50l g/ml streptomycin.The AGS cells were grown in F12k medium(Sigma,Louis,MO)supplemented with10%fetal bovine serum,with50U/ml penicillin and50l g/ml streptomycin.All cell lines were incubated at37°C in a humidi?ed5%(vol/vol)CO2atmosphere.

RNA extraction and quantitative real-time–PCR-based detection of let-7a and RAB40C mRNA

Total RNA from tissue samples and cell lines was obtained with the TRIzolòisolation reagent(Invitrogen,Carlsbad,CA)following the manufacturer’s in-structions.About100mg of tissue was homogenized in1ml of TRIzol reagent and then entered into the same step of the analysis as cell samples.The concentration,purity and amount of total RNA were determined by ultraviolet spectrometry(ND-1000spectrophotometer;NanoDrop Technologies, Wilmington,DE).

The reverse transcription(RT)–PCR was used to detect the expression of let-7a and the RAB40C gene at the transcript level as described(28).Brie?y,this method uses two-step RT–PCR.In the RT step,complementary DNA was re-verse transcribed from total RNA samples using the ReverTra Ace qPCR RT kit (TOYOBO,Tokyo,Japan).The quantitative real-time–PCR for detecting the expression of let-7a using the TaqMan MicroRNA assay(Applied Biosystems, Foster City,CA)together with the TaqMan Universal PCR Master Mix(Applied Biosystems)were done with an Applied Biosystems7500real-time PCR system (Applied Biosystems).The relative quanti?cation of let-7a was calculated using the2àDD C t method normalized with respect to RNU6B as the internal control and relative to a calibrator sample as the external control.The SYBR Premix Ex Taq TM Kit(TaKaRa,Dalian,China)was used for detecting the expression of RAB40C mRNA following the manufacturer’s instructions.The data were also calculated using the2–DD C t method normalized to the individual b-actin level. All primers were synthesized by TaKaRa.The primers were b-actin forward: CCCAGATCATGTTTGAGACCT and reverse:GAGTCCATCACGATGC-CAGT and RAB40C forward:TCATCGACAAGCTTCCACTG and reverse: TTGGACCTCTTGAGGCTGTT.

miRNA and small interfering RNA transfections

The let-7a mimic was an RNA duplex with the sequence:5#-UGAGGUA-GUAGGUUGUAUAGUU-3#and5#-CUAUACAACCUACUACCUCAUU-3#. The mimic non-speci?c control miRNAs duplex(named mimic NC)with a sequence of:5#-UUCUCCGAACGUGUCACGUTT-3#and5#-ACGUGA-CACGUUCGGAGAATT-3#was not homologous to any human genome sequence.For the in vivo tumorigenicity assay,all pyrimidine nucleotides in the let-7a mimic or mimic NC were substituted by their2#-O-methyl analogs to improve RNA stability.The anti-let-7a:5#-AACUAUACAACCUACUAC-CUCA-3#was a2#-O-methyl-modi?ed oligoribonucleotide designed as let-7a inhibitor.The inhibitor non-speci?c control miRNAs(named inhibitor NC), with the sequence:5#-CAGUACUUUUGUGUAGUACAA-3#was used as a negative control for anti-let-7a in the antagonism experiment.All the RNA oligoribonucleotides were purchased from GenePharma(Shanghai,China).

A blank control treated with only the transfection reagent was used in every transfection experiment.Small interfering RNAs(siRNAs)for human RAB40C sequence sense:5#-GGGACAUUGACCACUCAAATT-3#and antisense: 5#-UUUGAGUGGUCAAUGUCCCTT-3#and scrambled siRNA were de-signed and synthesized by GenePharma.

Cells were seeded onto six-well plates(3?105cells per well)the day before transfections were performed.Cells($60%con?uent)were transfected with let-7a mimic(50nmol/L),mimic NC(50nmol/L),inhibitor(100nmol/L)or inhibitor NC(100nmol/L)using Lipofectamine TM2000(Invitrogen)and transfection ef?ciency(.90%)was con?rmed with the use of the Silencer 5-carboxy?uorescein-labeled Negative Control(GenePharma).For the miRNA and siRNA combination experiments,cells were transfected with let-7a mimic (50nmol/L)for24h.These cells were then cotreated with RAB40C siRNA (100nmol/L)or siRNA NC(100nmol/L)for another24h.Total RNA and protein were prepared1or2days after transfection and were used for RT–PCR or western blot analysis to validate the knockdown ef?ciency.

Cell proliferation assay

The cell proliferation assay was done with a cell counting kit-8(CCK-8; Dojindo,Tokyo,Japan)at24h after transfection.Brie?y,5000transfected GES-1,AGS or BGC-823cells were plated per well in96-well plates at 24h after transfection and cultured in100l l of cell culture medium per well for24h in normal conditions.After incubation for24h,20l l of CCK-8 reagent was added to each well and incubated at37°C for1h.The absorbance at450nm(A450)and at650nm(A650)were measured with a Synergy2micro-plate reader(BioTek,Winooski,VT).The?nal absorbance was calculated as A650àA450,and cell viability was normalized as:

eFinal absorbance treated=final absorbance controlT?100%:

Colony formation assay

Transfected GES-1,AGS and BGC-823cells were trypsinized,counted and seeded at a density of1000cells per60mm culture dish in normal culture medium for colony formation assay and incubated for10days at37°C in a humidi?ed5%CO2atmosphere.During colony growth,the culture medium was replaced every3days and a colony was counted only if it contained.50 cells.Colony formation and growth were visualized by staining with crystal violet and the colony formation rate was calculated as:

eNumber of colonies=number of seeded cellsT?100%:

Cell cycle analysis

Cells were harvested at24h after transfection and?xed in70%ice-cold ethanol,treated with RNAse A,stained with50mg/ml propidium iodide and 0.1mg/ml RNase A for DNA content analysis by?ow cytometry with a FACS Calibur system(Becton Dickinson,Franklin,NJ).The percentage of cell pop-ulation in each phase was calculated with FlowJo FACS analysis software (Tree Star,Ashland,OR).

Soft agar assay

Soft agar plates were prepared in triplicate in60mm dishes with a bottom layer of0.6%(wt/vol)Noble agar(Sigma)in serum-free DMEM.Thereafter,trans-fected cells were trypsinized and1000cells per dish were seeded onto the bottom agar layer after mixing with0.3%Noble agar in DMEM supplemented with10%fetal bovine serum and allowed to harden.To assess cell viability before plating in soft agar,the number of cells was determined by staining with trypan blue.The dishes were incubated for3weeks at37°C in a humidi?ed5% CO2atmosphere and scored for clones.The result is expressed as the number of colonies containing at least50cells per well.

Tumorigenicity assays in nude mice

Five-week-old Balb/c nude mice were provided by Guangzhou University of Traditional Chinese Medicine(Guangzhou,China).All experimental proce-dures involving animals were in accordance with the Guide for the Care and Use of Laboratory Animals and were in accord with the institutional ethical guidelines for experiments with animals.Transfected and control BGC-823 cells were trypsinized,collected by centrifugation and suspended in DMEM.

Q.Yang et al. 714

A0.2ml sample of culture medium containing5?106cells was injected subcutaneously into the right-hand side of the posterior?ank of each mouse. The mice were housed in a pathogen-free environment and monitored every5 days for tumor growth.The mice were killed after40days and the weight and volume of each tumor were determined.The tumor xenografts were excised and weighed,then used for the extraction of total RNA and immunohistochem-ical assays.The tumor latency time was determined as the time to appearance of palpable tumor and the volume(V)was calculated as described(29). Immunohistochemistry

Ki-67protein expression in the cancer tissues of nude mice was detected using the streptavidin–peroxidase complex method with a Histostain-plus kit (Zhongshan Golden Bridge Biotechnology,Beijing,China).Rabbit anti-human Ki-67antibodies(Boster,Wuhan,China)were used at a dilution of1:400as primary antibodies.Color development was achieved with3#,3#-diaminobenzidine,which stained positive cells brown.Normal rabbit serum and secondary antibody alone were used as negative controls.

Western blotting

Total proteins were prepared by standard procedures and quanti?ed by the BCA method(Jiancheng,Nanjing,China).A20l g sample of protein was mixed with5?sodium dodecyl sulfate/polyacrylamide gel electrophoresis sample buffer(Weijia,Guangzhou,China)and boiled for5min before sodium dodecyl sulfate–polyacrylamide gel electrophoresis(15%polyacrylamide gel) and transfer to a polyvinylidene di?uoride membrane(Millipore Corp., Bedford,MA).The membrane was blocked with5%non-fat dry milk in phosphate-buffered saline for45min at37°C with agitation.RAB40C protein and glyceraldehyde-3-phosphate dehydrogenase on the same membrane were quanti?ed by dividing the membrane into two pieces according to the molec-ular mass of prestained protein standards(Weijia).The piece of the membrane with the greater molecular mass proteins was incubated with a primary anti-body for rat anti-human RAB40C(Epitomics,Burlingame,CA)at a concentra-tion of1:1000at37°C for1h.The other piece of the membrane was incubated with a primary antibody for mouse anti-human glyceraldehyde-3-phosphate dehydrogenase(Epitomics)at a concentration of1:1500.Signals were detected by secondary antibodies labeled with IRDye800(Rockland Immunochemi-cals,Gilbertsville,PA)and signal intensity was determined with an Odyssey Infrared Imaging System(Li-Cor Biosciences,Lincoln,NE).

Luciferase reporter assay

The full-length3#-UTR of RAB40C gene(GenBank accession number NM_021168;length1650bp)was ampli?ed from GES-1cells complementary DNA and cloned into the Notl and Xhol sites in a psiCHECK-2vector(Prom-ega,Madison,WI)downstream of the reporter gene.The primers for RAB40C were forward:5#-AGCTTTGTTTAAACCGGGATGGGCGCGGGGATG-3# and reverse:5#-AGCTTTGTTTAAACTGTGGGGACCACAGCTGAAT-TAC-3#.To introduce mutations into let-7a target sites in the RAB40C coding region,primers were designed for site-directed mutagenesis based on wild psiCHECK-2-RAB40C3#-UTR plasmid that resulted in the destruction of the let-7a target site without altering the amino acid sequence of RAB40C. The sites were mutated as follows:before mutagenesis,CUACCUC;after mutagenesis,AUCGAGC.The primers for RAB40C mt were forward:5#-CA-CAGCACTGGTGATCACCTATCGCTCCTGTCCTCAGGCCGTGCGGC-3# and reverse:5#-GCCGCACGGCCTGAGGACAGGATCGATAGGTGAG-CACCAGTGCTGTG-3#.The segment of the RAB40C3#-UTR containing the mutated let-7a target sequence.was also cloned into the psiCHECK-2 Luciferase vector(Promega).The resulting wild and mutated RAB40C expres-sion vectors were con?rmed by sequencing.The construct of reporter vector was performed by Landbiology(Guangzhou,China).

For the luciferase reporter assay,GES-1cells(50%con?uent in24-well plates)were cotransfected with0.5l g of psiCHECK-2vector constructs with or without20l M let-7a mimic or inhibitor for48h by Lipofectamine2000 (Invitrogen).At48h after transfection,the activity of Renilla luciferase and ?re?y luciferase was measured with the Dual-luciferase Reporter Assay System(Promega).Relative luciferase activity was normalized with?re?y luciferase activity and then compared with the psiCHECK-2vector control. Statistical analysis

All statistical analysis was done with SPSS13.0software.Values are expressed as mean±SD.Differences between groups were analyzed by Student’s t-test or the non-parametric Mann–Whitney U-test for comparison of two groups and one-way analysis of variance or the non-parametric Kruskal–Wallis H test for multiple comparisons.Nonlinear regression analysis between groups was used with logarithmic regression model by method of curve estimation.All experi-ments were done at least in triplicate and the level of statistical signi?cance was set at P,0.05for all tests.Results

Expression of let-7a is reduced in human gastric cancer tissues and cell lines

The clinicopathologic data for27patients are given in Table I.Based on the results of TaqMan real-time PCR,we found the expression of let-7a in21cancer tissues was lower than that in the matched normal tissues (Figure1A).The expression of let-7a was signi?cantly lower in gastric cancer tissues than that in normal tissues,with a median change of0.38-fold(Figure1B).Next,we examined the correlation of the tumor tissue (T)/normal tissue(N)ratios for let-7a expression with the clinicopath-ologic factors given in Table1.The T/N ratios in patients with poor differentiation were signi?cantly lower than those with moderate or good differentiation(Table I;Figure1C).However,there was no sig-ni?cant difference between the groups divided by any other clinico-pathologic feature,including gender,tumor size,histologic cell type, lymph node metastasis or venous invasion.To con?rm the association between the expression of let-7a and gastric cancer,?ve cell lines de-rived from gastric cancers with various degrees of differentiation were selected to detect let-7a expression:MKN-28(well differentiated), AGS(well differentiated),SGC-7901(moderately differentiated), BGC-823(poorly differentiated or undifferentiated)and HGC-27(un-differentiated).The data showed that the expression of let-7a was sig-ni?cantly downregulated in MKN-28(0.63-fold),AGS(0.33-fold), SGC-7901(0.28-fold),BGC-823(0.23-fold)and HGC-27(0.13-fold) cells compared with normal human gastric epithelial cell line GES-1 (Figure1D).

let-7a regulates cell proliferation and the cell cycle

We evaluated the transfection ef?ciency by determining the percentage of cells containing the5-carboxy?uorescein-labeled non-speci?c miRNA control and the results indicated successful transfection of 5-carboxy?uorescein-labeled non-speci?c miRNA control into.90% of cells of all the cell lines used in this study.TaqMan real-time PCR revealed the transfected let-7a mimic signi?cantly increased the levels of let-7a(Figure2A),whereas the transfected let-7a inhibitor effec-tively inhibited the expression of let-7a(Figure2B)compared with mimic NC or inhibitor NC cells or with untransfected cells in all six cell lines,which demonstrated that the transfected let-7a mimic and inhibitor were functional.

Table I.The Clinicopathologic factors of27gastric cancer patients and comparison of let-7a expression in different groups

Clinicopathologic factor No.of

patients

Relative

let-7a

expression

P

Min–max

(median)

Age(years)

.60190.04–8.97(0.47)0.313 6080.08–2.00(0.25)

Gender

Male170.04–8.97(0.47)0.335 Female100.04–2.63(0.25)

Tumor size(cm)

.370.10–2.72(0.82)0.580 3200.04–8.97(0.31) Histologic cell type

Well differentiated60.10–8.97(2.06)0.018 Moderately differentiated80.16–2.63(0.45)

Poorly differentiated130.04–2.00(0.19)

Lymph node metastasis

Absent120.04–2.72(0.51)0.727 Present150.04–8.97(0.35)

Venous invasion

Absent90.04–8.97(0.20)0.862 Present180.04–2.72(0.49)

let-7a in gastric cancer

715

We used the CCK-8assay to measure cell proliferation in vitro .The AGS and BGC-823cell lines,which had the highest level of let-7a expression and the smallest standard deviation (Figure 2A),were used in an overexpression study.AGS and BGC-823cells transfected with the let-7a mimic showed a signi?cant reduction in the number of metabolically active cells compared with those transfected with mimic NC or the untransfected control (Figure 2C).The GES-1cell line,which had the lowest level of let-7a expression (Figure 2B),was used for an inhibition study.An increase of 58.8%in cell growth was observed in GES-1cells after transfection with a let-7a inhibitor compared with inhibitor NC (Figure 2C).

As shown in Figure 2D,DNA content analysis by ?ow cytometry revealed let-7a restoration induced an accumulation of AGS cells in the percentage of cells at the G 1phase from 55.38to 64.87%and a reduction of cells in S phase from 37.94to 26.55%compared with the mimic NC group and this effect on cell cycle was observed also for BGC-823cells (Figure 2E).In contrast,the percentage of GES-1cells transfected with let-7a inhibitor in the G 1phase was decreased from

75.01to 60.35%,whereas the percentage of cells in the S phase was increased from 20.61to 35.76%compared with the inhibitor NC group (Figure 2F).These ?ndings,together with the cell proliferation results (Figure 1C),illustrated that let-7a could be a tumor suppressor gene in gastric cancer.

let-7a suppresses anchorage-dependent growth in vitro and tumor growth in vivo

The signi?cant reduction of let-7a on cell growth in vitro prompted us to explore the possible biological signi?cance of let-7a in tumorigen-esis.AGS and BGC-823cells were used in these assays.To test how long the promotion of let-7a by the let-7a mimic can be sustained,we measured the let-7a levels after 1,2,3,4,7,10,14and 21days from transfection by TaqMan real-time PCR.We found that the promotion effect lasted up to at least 10days in let-7a mimic-transfected AGS cells (Figure 3A)and 14days in let-7a mimic-transfected BGC-823cells (Figure 3B)compared with mimic NC-transfected cells.The col-ony numbers of AGS and BGC-823cells that formed in soft agar medium were decreased at 3weeks after transfection with the let-7a mimic (Figure 3C and D).Furthermore,AGS and BGC-823cells trans-fected with the let-7a mimic displayed fewer and smaller colonies compared with the mimic NC-transfected and untransfected cells (Fig-ure 3E).Taken together,these results suggest that the initial promotion of let-7a might be suf?cient to inhibit tumor growth and prompted us to investigate the role of let-7a on tumor growth in vivo .

Poorly differentiated BGC-823cells were included in this step to further con?rm the promotion effect on tumor growth in vivo induced by transfection of the let-7a mimic.The emerging rate of tumor and tumor latency time showed a signi?cant difference between the mice injected with let-7a mimic-transfected cells and the mimic NC group (Figure 4A).It is of considerable interest that tumors derived from cells transfected with the let-7a mimic grew substantially more slowly than the mimic NC group throughout tumor growth (Figure 4B).The average weight of tumors derived from cells transfected with the let-7a mimic was only 63.2%of that derived from the cells transfected with the mimic NC (Figure 4C and D).There was no signi?cant difference of tumor volume,tumor average weight or other indices between the mimic NC-transfected cell group and the untransfected cell group.Additionally,the expression of the proliferation marker Ki-67was signi?cantly lower in tumor xenografts of the let-7a mimic group compared with the mimic NC and untransfected groups (Figure 4E).These results indicate that the reduced tumor growth is probably to be due to decreased proliferation induced by let-7a .

RAB40C is a direct target of let-7a

A bioinformatic analysis identi?ed RAB40C as a hypothetical target gene of let-7a ,as identi?ed by TargetScan alghorithms (http://www https://www.wendangku.net/doc/896168516.html,/).The let-7a targeting sites in RAB40C 3#-UTR are conserved among mammals and the corresponding sequences of the RAB40C mutated 3#-UTRs are shown in supplementary Figure S1,available at Carcinogenesis Online.The putative secondary RNA hy-brid (http://bibiserv.techfak.uni-bielefeld.de/rnahybrid/)for human let-7a and RAB40C or N-Ras mRNA with minimal free energy is shown in supplementary Figures S2and S3,available at Carcinogenesis Online.Our analysis indicates that let-7a has a more stable secondary structure with lower free energy with RAB40C compared with the validated target N-Ras .

Next,we found overexpression of let-7a in AGS and BGC-823cells effectively decreased the level of the RAB40C protein.A blocking strategy was further adapted by introducing the let-7a inhibitor into GES-1cells,which increased the level of the RAB40C protein (Figure 5A).In both cases,neither the let-7a mimic nor the inhibitor affected the mRNA of RAB40C (supplementary Figure S4is available at Carcinogenesis Online),suggesting posttranscriptional regulation of RAB40C by let-7a in gastric cancer cell lines.A further hint about the potential role of let-7a in the regulation of RAB40C expression came from the analysis of ?ve different gastric cancer cell lines (MKN-28,AGS,SGC-7901,BGC-823and HGC-27)that

showed

Fig.1.Level of expression of mature let-7a in gastric cancer tissue

specimens and cell lines.Mature let-7a expression was analyzed by TaqMan real-time PCR and normalized to U6B expression.(A )The comparison of let-7a expression between matched normal gastric and gastric cancer tissues in 27patients.T,gastric cancer tumor tissue;N,pair-matched adjacent non-tumor tissue.(B )The data for let-7a expression were analyzed using the Mann–Whitney U-test.(C )The ratios of tumor to normal tissue for let-7a expression are presented as the relative tumor tissue/normal tissue ratio (T/N ratio)of let-7a expression.The T/N ratios were analyzed statistically in patients of different pathological grade (Kruskal–Wallis H test,P 50.018).Horizontal lines show the median value of each group.W,well differentiated;M,moderately differentiated;P,poorly differentiated.(D )The expression of let-7a in ?ve gastric cancer cell lines relative to the normal gastric cell line GES-1(Kruskal–Wallis H test,P 50.016).N,normal gastric epithelial cell line;W,well differentiated cancer cell line;M,moderately differentiated cancer cell line;P,poorly differentiated cancer cell line;U,undifferentiated cancer cell line.The data are shown as median with interquartile range for three independent experiments.

Q.Yang et al.

716

a signi?cant reversed nonlinear correlation between let-7a levels and the RAB40C protein (Figures 5B and C).To assess the clinical rele-vance of these ?ndings,we examined the correlation between the level of the RAB40C protein with let-7a expression in the 27matched normal and cancer tissues (Figure 5D).There was a reversed nonlinear corre-lation between the level of the RAB40C protein and let-7a expression (Figure 5E).These results support the premise that down-regulation of let-7a increases the level of RAB40C gene in gastric cancer.

To con?rm the direct interaction between let-7a and its binding site within RAB40C mRNA,a human RAB40C 3#-UTR fragment contain-ing a wild type or mutant let-7a -binding sequence was cloned down-stream of the luciferase reporter gene.The psiCHECK-2vector was cotransfected in GES-1cells in association with let-7a mimic or in-hibitor.GES-1cells transfected with psiCHECK-2vector showed $50%decrease or 30%increase of the relative luciferase activity when cotransfected with let-7a mimic or inhibitor (Figure 5F and G).The let-7a seed sequence was mutagenized in the cloned RAB40C 3#-UTR region to generate the psiCHECK-2-RAB40C mt vector.No signi?cant change of the relative luciferase activity was observed following the cotransfection of this mutagenized vector with let-7a mimic or inhibitor (Figure 5F and G).These ?ndings showed a direct interaction between let-7a and RAB40C mRNA and indicated that let-7a might suppress gene expression through the let-7a binding se-quence at the 3#-UTR of RAB40C .

RAB40C mediates the let-7a effect on cell proliferation and anchorage-independent growth

To con?rm that the let-7a effect on cell proliferation and anchorage-independent growth is associated with its modulation of RAB40C ,GES-1cells were transfected with siRNA targeting RAB40C or control siRNA.Transfection of AGS cells with siRNA for RAB40C effectively suppressed RAB40C mRNA expression (Figure 6A)and RAB40C pro-tein (Figure 6B).A similar effect was observed in BGC-823cells (Figure 6A and B).RAB40C depletion signi?cantly alleviated the anti-proliferative effect of let-7a upregulation in AGS cells and BGC-823cells as determined by the CCK-8assay (Figure 6C).Furthermore,RAB40C siRNA attenuated the plating ef?ciency of cells in soft agar,which was decreased upon the transfection of let-7a mimic in AGS cells and BGC-823cells (Figure 6D).These data provided further evidence that RAB40C was targeted by let-7a and therefore suggested that RAB40C mediates the let-7a effect on cell proliferation and an-chorage-independent growth in gastric cancer cells.Discussion

miRNAs are becoming increasingly recognized as regulatory mole-cules in human cancers,which have been demonstrated to function as oncogenes or tumor suppressors.let-7a has been suggested to function as a tumor suppressor and have a strong correlation

with

Fig.2.Expression of let-7a after transfection with the let-7a mimic or inhibitor in six gastric cell lines and overexpression or knockdown of let-7a in gastric cells affected cell growth in vitro .Total RNA was isolated 24h after transfection for the detection of let-7a and mature let-7a expression was analyzed by TaqMan real-time PCR and normalized to U6B expression.(A )Expression of let-7a in let-7a mimic-transfected,mimic NC-transfected and untransfected gastric cells.(B )Expression of let-7a in let-7a inhibitor-transfected,inhibitor NC-transfected and untransfected gastric cells.(C )Overexpression or knockdown of signi?cant let-7a -affected cell proliferation in AGS,BGC-823or GES-1cells as measured by the CCK-8assay.(D and E )Overexpression of let-7a resulted in G 1arrest in AGS (D)and BGC-823(E)cells.(F )Knockdown of let-7a enhanced S phase cells in GES-1cells.(A–F)The data are shown as mean ±SD for three independent experiments,each done in triplicate.?P ,0.05,??P ,0.01,compared with the let-7a mimic NC-or inhibitor NC-transfected group or with untransfected group.P values were obtained by one-way analysis of variance or the non-parametric Kruskal–Wallis H test for multiple comparisons.

let-7a in gastric cancer

717

clinicopathologic factors or prognosis in cancer patients (22,23,30).In the present study,the 75%let-7a expression of gastric carcinoma samples was signi?cantly lower than that of matched normal tissues,suggesting that reduced let-7a expression is a frequent event in gastric cancer.Furthermore,we found that the expression level of let-7a was associated with the differentiation stage in patients and the same results were observed in the cells.Our analytical results showed that tumor tissue of patients and cell lines with lower levels of let-7a tended to have poor differentiation.This observation was consistent with an earlier report that identi?ed the loss of let-7expression as a marker for less differentiated cancer (31).These results indicated that let-7a expression might be an important in-dicator for gastric cancer diagnosis and clinical stage.

miRNAs have been shown to be important in the development and maintenance of normal cellular function,and alteration in expression of miRNAs can result in human cancer initiation and tumor progres-sion.Some studies have reported that let-7miRNA was a master regulator of cell proliferation and cell cycle pathways in lung,co-lorectal and hepatic cells (22,23,32).However,no information is available on the impact of altered let-7a expression on gastric cancer cell proliferation pathways.Here,we show that overexpression of let-7a resulted in 40–50%decrease in cell proliferation rate and the

effect

Fig.3.The change of let-7a expression and inhibition of anchorage-dependent growth by overexpression of let-7a .(A and B )The change of let-7a expression after 1,2,3,4,7,10,14and 21days from let-7a mimic and let-7a mimic NC transfected or untransfected AGS (A)or BGC-823(B)cells.(C and D )Inhibition of anchorage-dependent growth by the let-7a mimic in AGS (C)and BGC-823(D)cells.(E )Representative results of colony formation of untransfected,NC-transfected and mimic-transfected AGS and BGC-823cells.The data in (A–D)are shown as mean ±SD from three independent experiments.?P ,0.05;??P ,0.01,compared with the let-7a mimic NC-transfected group.P values were obtained by one-way analysis of variance or the non-parametric Kruskal–Wallis H test for multiple comparisons.

Q.Yang et al.

718

of let-7a was attenuated by the inhibition of let-7a .More importantly,decreased levels of protein Ki-67,which has been used as a marker for cell proliferation (33),were detected in the xenografts of the let-7a mimic-transfected cells.It further con?rmed that cell proliferation was affected by let-7a miRNA in gastric cancer.Flow cytometry-based cell cycle analysis revealed a trend toward an accumulation of cells in G 1phase in let-7a mimic-transfected gastric cancer cells.Conversely,let-7a inhibitor caused a signi?cant increase in the frac-tion of cells in S phase.However,in the present study,we did not observe any apparent increase of the sub-G 1population or any apo-ptosis-related morphological changes,such as nuclear blebbing or condensation,under the phase contrast microscope (data not shown).This suggests that growth suppression induced by let-7a transfection was caused by induction of G 1arrest rather than apoptosis.This study con?rmed the results of that earlier work and extended our knowledge of the inhibitory effect of let-7a on cell proliferation and cell cycle control in gastric cancer cells.All the results demonstrated that the consistently reduced let-7a expression in gastric cancer should be a factor contributing to the development of the tumor rather than being affected as a consequence of the disease.

Gastric tumor xenograft models were established to investigate the antitumor effect of let-7a in vivo relevant to our ?ndings in vitro .Mice injected with BGC-823cells transfected with the let-7a mimic showed a signi?cant inhibition of tumor xenografts,implying that overexpression of let-7a suppressed tumor growth of gastric cancer signi?cantly in nude mice.Therefore,therapeutic strategies to intro-duce let-7a into cancer cells might be useful for retarding the process of tumorigenesis.Experiments with the xenograft carcinoma model indicated that one transient transfection with let-7a mimic is suf?cient to cause substantial inhibition of tumor growth,which raises the pos-sibility that let-7a mimic might have potential therapeutic value,con-sistent with the earlier studies (34).

It is generally accepted that miRNAs exert their function through regulating the expression of their downstream target genes (35).Thus,putative let-7a targets were predicted using TargetScan https://www.wendangku.net/doc/896168516.html,putational predictions suggest that let-7a can target

819

Fig.4.Suppression of tumor growth by overexpression of let-7a .(A )The tumor latency was the number of days to the onset of palpable tumor and the values are given as mean ±SD (n 54per group).?P ,0.05,compared with let-7a mimic NC-transfected or untransfected cells-injected group.(B )Tumor growth curves were measured after injection of untransfected BGC-823cells,let-7a mimic NC and let-7a mimic-transfected BGC-823cells.The tumor volume V (in cubic centimeter)was calculated using the formula volume V 50.5L W 2,where L is the length of the tumor (in centimeter),W is the width of the tumor (in centimeter)and V is the mean tumor volume (n 54per group).(C )Tumor weight;the values are given as mean ±SD (n 54per group).?P ,0.05,compared with let-7a mimic NC-transfected cells-injected group.(D )Effect of let-7a on tumorigenesis in vivo .The photograph shows representative features of tumor xenografts 40days after inoculation.(E)Representative immunohistochemical assay of Ki-67in tumor xenografts of mice injected with untransfected cells,mimic NC-transfected cells or let-7a mimic-transfected cells.P values were obtained by one-way analysis of variance for multiple comparisons.

let-7a in gastric cancer

719

transcripts,with a total of 905conserved binding sites and 125poorly conserved binding sites.RAB40C is a member of the RAS family,which plays important roles in the regulation of immune responses,embryo and cell lineage development,cell cycle progression,in?am-mation and oncogenesis and is involved in the execution of important steps in tumorigenesis (36,37).Moreover,the dysregulation of RAS family members is frequently observed in human cancers (38–40).In this study,RAB40C was signi?cantly overexpressed in gastric tumor samples compared with the matched normal tissue.Based on the bioinformatic analysis,let-7a binds to the 3#-UTR of RAB40C mRNA with lower free energy than N-Ras mRNA.It might stabilize a favor-able conformation that facilitates the binding between let-7a and 3#-UTRs of RAB40C (41,42),which prompted us to explore whether RAB40C is a functional target of let-7a .

Here,we demonstrate that RAB40C is regulated directly by let-7a .This is evident at the level of the RAB40C protein is dysregulated by the change of let-7a expression without any change in the amount of RAB40C mRNA.In addition,there was a signi?cant inverse correla-tion of let-7a expression with the level of RAB40C protein in gastric cancer cell lines and gastric tumor tissues.To the best of our knowl-edge,this is the ?rst report showing a correlation between RAB40C mRNA and let-7a miRNA in clinical samples of human cancer.More-over,mutation of the let-7a -binding site abolished the effect of let-7a on the regulation of RAB40C ?uorescence intensity.Importantly,de-pletion of RAB40C by siRNA transfection partly rescues the reduced cellular proliferation and the plating ef?ciency of cells in soft agar observed upon let-7a up-regulation,further demonstrating that RAB40C is a target of let-7a and suggesting an essential role for RAB40C as a mediator of the biological effects of let-7a in gastric tumorigenesis.This is the ?rst study to show that RAB40C is nega-tively regulated by let-7a at the posttranscriptional level via binding to the 3#-UTR of RAB40C mRNA in gastric cancer cells.

In summary,we report that let-7a expression was decreased in human gastric cancer tissues and cell lines.Our study extended

the

Fig.5.let-7a negatively regulates RAB40C by binding to the RAB40C 3#-UTR.(A )let-7a mimic reduced RAB40C protein in AGS and BGC-823cells and let-7a inhibitor increased RAB40C protein levels in GES-1cells.(B –C )Analysis of the protein expression levels of RAB40C and correlation with let-7a expression in ?ve gastric cancer cell lines and the GES-1cell line.Protein expression was determined by western blotting (B).The expression of let-7a is plotted against protein expression (C).Coef?cient of determination from logarithmic regression model (R 2)and P values are given (R 250.813;P 50.014).(D –E )Analysis of the protein expression levels of RAB40C and correlation with let-7a expression in 27pairs of gastric cancer tissues and matched normal gastric tissues.Protein expression was determined by western blotting (D)and the expression of let-7a is plotted against protein expression (E).Coef?cient of determination from logarithmic regression model (R 2)and P values are given (R 250.451;P 50.020).The value on each lane in (B and D)indicates the relative expression level of RAB40C ,which is represented by the intensity ratio between RAB40C and glyceraldehyde-3-phosphate dehydrogenase fragments in each lane.(F –G )Analysis of luciferase activity.let-7a mimic (F)or inhibitor (G)inhibited or enhanced wild type but not mutant RAB40C 3#-UTR reporter activity.??P ,0.01,compared with the mimic or inhibitor NC group.P values were obtained by two-tailed Student’s t -test.Each reporter plasmid was transfected at least twice (on different days),and each sample was assayed in triplicate.

Q.Yang et al.

720

impact of let-7a to gastric cancer by showing for the ?rst time that let-7a decreased proliferation and cell cycle in gastric cancer cells and inhibited tumor growth in nude mice.These effects are possibly due to down-regulation of RAB40C by let-7a .Therefore,miRNAs,in par-ticular let-7a ,might serve as potentially useful targets for gastric cancer diagnosis and therapy.Future studies will be aimed at devel-oping strategies to use let-7a and RAB40C as a therapeutic regimen.Supplementary material

Supplementary Figures S1–S4can be found at http://carcin https://www.wendangku.net/doc/896168516.html,/Funding

National Natural Science Foundation of China (30771780,30972443to J.Y .);Natural Science Foundation of Guangdong Province (07117550,9251018201000004to J.Y .);Science and Technology Program of Guangzhou Bureau of Education (08A093to J.Y .,08A092to Y .Q.);Science and Technology Pillar Program of Jiangxi Province (2009zDS16000to C.H.).Acknowledgements

We thank Xin Zhou and Wei Li (Department of General Surgery,First Af?l-iated Hospital of Nanchang University)for collecting the clinicopathologic information for the patients.We thank Rong Liu (Institute of Respiratory Diseases,First Af?liated Hospital of Guangzhou Medical University)for as-sistance with the nude mice assays.

Con?ict of Interest Statement:None declared.

References

1.Parkin,D.M.et al .(2005)Global cancer statistics,200

2.CA Cancer J.Clin.,55,74–108.

2.Yang,L.et al .(2005)Estimates of cancer incidence in China for 2000and projections for 2005.Cancer Epidemiol.Biomarkers Prev.,14,243–250.

3.Ambros,V .(2004)The functions of animal microRNAs.Nature ,431,350–355.

4.Engels,B.M.et al .(2006)Principles and effects of microRNA-mediated post-transcriptional gene regulation.Oncogene ,25,6163–6169.

5.Lim,L.P.et al .(2005)Microarray analysis shows that some microRNAs-downregulate large numbers of target mRNAs.Nature ,433,769–773.

6.Lewis,B.P.et al .(2005)Conserved seed pairing,often ?anked by adeno-sines,indicates that thousands of human genes are microRNA targets.Cell ,120,15–20.

7.Hwang,H.W.et al .(2006)MicroRNAs in cell proliferaftion,cell death,and tumorigenesis.Br.J.Cancer.,94,776–780.

8.Alvarez-Garcia,I.et al .(2005)MicroRNA functions in animal development and human disease.Development ,132,4653–4662.

9.Krutzfeldt,J.et al .(2006)Strategies to determine the biological function of microRNAs.Nat.Genet.,38,S14–S19.

10.Calin,G.A.et al .(2004)Human microRNA genes are frequently located at fragile sites and genomic regions involved in cancers.Proc.Natl https://www.wendangku.net/doc/896168516.html,A.,101,2999–3004.

11.Iorio,M.V .et al .(2005)MicroRNA gene expression deregulation in human breast cancer.Cancer Res.,65,7065–7070.

12.Calin,G.A.et al .(2005)A MicroRNA signature associated with prognosis and progression in chronic lymphocytic leukemiap.N.Engl.J.Med.,353,1793–1801.

13.Yanaihara,N.et al .(2006)Unique microRNA molecular pro?les in lung cancer diagnosis and prognosis.Cancer Cell ,9,189–198.

14.Zhang,Z.et al .(2008)MiR-21plays a pivotal role in gastric cancer path-ogenesis and https://www.wendangku.net/doc/896168516.html,b.Invest.,88,1358–1366.

15.Liu,T.et al .(2008)MicroRNA-27a functions as an oncogene in gastric asenocarcinoma by targeting prohibitin.Cancer Lett.,273,233–242.

16.Petrocca,F.et al .(2008)E2F1-regulated microRNAs impair TGFbeta de-pendent cell-cycle arrest and apoptosis in gastric cancer.Cancer Cell ,13,272–286.

17.Xia,L.et al .(2008)miR-15b and miR-16modulate multidrug resistance by targeting BCL2in human gastric cancer cells.Int.J.Cancer.,123,372–

379.

Fig.6.The effect of RAB40C siRNA and let-7a mimic on cell proliferation and anchorage-independent growth.(A )Suppression of RAB40C mRNA expression levels by RAB40C siRNA transfection.(B )Suppression of RAB40C protein expression levels by RAB40C siRNA transfection.(C )The effect of RAB40C siRNA and let-7a mimic on cell proliferation.Relative cell growth is compared with the untransfected group.(D )The effect of RAB40C siRNA and let-7a mimic on anchorage-independent growth of cells.Data are given as mean ±SD from three independent experiments.?P ,0.05;??P ,0.01,compared with the siRNA NC group.P values were obtained by one-way analysis of variance or the non-parametric Kruskal–Wallis H test for multiple comparisons.

let-7a in gastric cancer

721

18.Du,Y.et al.(2009)Down-regulation of miR-141in gastric cancer and its

involvement in cell growth.J.Gastroenterol.,44,556–561.

19.Bandres,E.et al.(2009)MicroRNA-451regulates macrophage migration in-

hibitory factor production and proliferation of gastrointestinal cancer cells.

Clin.Cancer Res.,15,2281–2290.

20.Takagi,T.et al.(2009)Decreased expression of microRNA-143and-145in

human gastric cancers.Oncology,77,12–21.

21.Luo,H.et al.(2009)Down-regulated miR-9and miR-433in human gastric

carcinoma.J.Exp.Clin.Cancer Res.,16,28–42.

22.Takamizawa,J.et al.(2004)Reduced expression of the let-7microRNAs in

human lung cancers in association with shortened postoperative survival.

Cancer Res.,64,3753–3756.

23.Akao,Y.et al.(2006)let-7microRNA functions as a potential growth

suppressor in human colon cancer cells.Biol.Pharm.Bull.,29, 903–906.

24.Johnson,S.M.et al.(2005)RAS is regulated by the let-7microRNA family.

Cell,120,635–647.

25.Sampson,V.B.et al.(2007)MicroRNA let-7a downregulates MYC and

reverts MYC-induced growth in Burkitt lymphoma cells.Cancer Res., 67,9762–9770.

26.Mayr,C.et al.(2007)Disrupting the pairing between let-7and HMGA2

nhances oncogenic transformation.Science,315,1576–1579.

27.Motoyama,K.et al.(2008)Clinical signi?cance of high mobility group A2

in human gastric cancer and its relationship to let-7microRNA family.Clin.

Cancer Res.,14,2334–2340.

28.Pfaf?M.W(2001)A new mathematical model for relative quanti?cation in

real-time RT-PCR.Nucleic Acids Res.,29,e45.

29.Si,M.L.et al.(2007)miR-21-mediated tumor growth.Oncogene,26,2799–

2803.

30.Long,X.B.et al.(2009)Let-7a microRNA functions as a potential tumor

suppressor in human laryngeal cancer.Oncol.Rep.,22,1189–1195.31.Shell,S.et al.(2007)Let-7expression de?nes two differentiation stages of

cancer.Proc.Natl https://www.wendangku.net/doc/896168516.html,A.,104,11400–11405.

32.Johnson,C.D.et al.(2007)The let-7microRNA represses cell proliferation

pathways in human cells.Cancer Res.,67,7713–7722.

33.Schlu¨ter,C.et al.(1993)The cell proliferation-associated antigen of

antibody Ki-67:a very large,ubiquitous nuclear protein with numerous-repeated elements,representing a new kind of cell cycle-maintaining pro-teins.J.Cell Biol.,123,513–522.

34.Bandres,E.et al.(2007)MicroRNAs as cancer players:potential clinical

and biological effects.DNA Cell Biol.,26,273–282.

35.Su,H.et al.(2009)MicroRNA-101,down-regulated in hepatocellular car-

cinoma,promotes apoptosis and suppresses tumorigenicity.Cancer Res., 69,1135–1142.

36.Ng,E.L.et al.(2008)Rab GTPases and their roles in brain neurons and glia.

Brain Res.Rev.,58,236–246.

37.Rodriguez-Gabin,A.G.et al.(2004)Vesicle transport in oligodendrocytes:

probable role of Rab40c protein.J.Neurosci.Res.,76,758–770.

38.Chen,X.et al.(2009)Role of miR-143targeting KRAS in colorectal tu-

morigenesis.Oncogene,28,1385–1392.

39.Leung,P.S.(2010)Current research of the RAS in pancreatitis and pancre-

atic cancer.Adv.Exp.Med.Biol.,690,179–199.

40.Cortot,A.B.et al.(2010)KRAS mutation status in primary nonsmall

cell lung cancer and matched metastases.Cancer,116,2682–2687.

41.Lindow,M.et al.(2007)Principles and limitations of computational micro-

RNA gene and target?nding.DNA Cell Biol.,26,339–351.

42.Rehmsmeier,M.et al.(2004)Fast and effective prediction of microRNA/

target duplexes.RNA,10,1507–1517.

Received September29,2010;revised January21,2011;

accepted February12,2011

Q.Yang et al. 722