Original Article Array analysis for potential biomarker of ... · Array analysis for potential biomarker of gemcitabine identification in non-small cell ... cell lines to gem-citabine

Int J Clin Exp Pathol 2013;6(9):1734-1746www.ijcep.com /ISSN:1936-2625/IJCEP1305003

Original ArticleArray analysis for potential biomarker of gemcitabine identification in non-small cell lung cancer cell lines

Hai-Hong Zhang1, Zhi-Yi Zhang1, Chun-Li Che2, Yi-Fang Mei2, Yu-Zhi Shi2

1Department of rheumatism and immunology, First Clinical Medical College affiliated to Harbin Medical University, Harbin, China; 2Department of respiratory medicine, First Clinical Medical College affiliated to Harbin Medical University, Harbin, China

Received May 6, 2013; Accepted August 5, 2013; Epub August 15, 2013; Published September 1, 2013

Abstract: Gemcitabine is one of the most widely used drugs for the treatment of advanced Non-small cell lung cancer (NSCLC), but modest objective response rate of patients to gemcitabine makes it necessary to identify novel biomarkers for patients who can benefit from gemcitabine-based therapy and to improve the effect of clinical therapy. In this work, 3 NSCLC cell lines displaying different sensitivities to gemcitabine were applied for mRNA and microRNA (miR) expression chips to figure out the biomarkers for gemcitabine sensitivity. Genes whose expression increased dramatically in sensitive cell lines were mainly enriched in cell adhesion (NRP2, CXCR3, CDK5R1, IL32 and CDH2) and secretory granule (SLC11A1, GP5, CD36 and IGF1), while genes with significantly upregulated expression in resistant cell line were mainly clustered in methylation modification (HIST1H2BF, RAB23 and TP53) and oxidoreductase (TP53I3, CYP27B1 and SOD3). The most intriguing is the activation of Wnt/β-catenin signal-ing in gemcitabine resistant NSCLC cell lines. The miR-155, miR-10a, miR-30a, miR-24-2* and miR-30c-2* were upregulated in sensitive cell lines, while expression of miR-200c, miR-203, miR-885-5p, miR-195 and miR-25* was increased in resistant cell line. Genes with significantly altered expression and putatively mediated by the expression-changed miRs were mainly enriched in chromatin assembly (MAF, HLF, BCL2, and IGSF3), anti-apoptosis (BCL2, IGF1 and IKBKB), protein kinase (NRP2, PAK7 and CDK5R1) (all the above genes were upregulated in sensi-tive cells) and small GTPase mediated signal transduction (GNA13, RAP2A, ARHGAP5 and RAB23, down-regulated in sensitive cells). Our results might provide potential biomarkers for gemcitabine sensitivity prediction and putative targets to overcome gemcitabine resistance in NSCLC patients.

Keywords: Non-small cell lung cancer, gemcitabine, gene expression profiles, miR expression profiles

Introduction

Non-small cell lung cancer (NSCLC), the leading cause of cancer-related death in the world, accounts for about 80-85% of all cases of lung cancer [1]. Gemcitabine is one of the most widely used drugs for the treatment of NSCLC. Based on two clinical trials [2, 3], the cisplatin plus gemcitabine regimen has become a com-monly used combination for advanced NSCLC. The favorable tolerability profile of gemcitabine makes it an ideal choice for evaluation as main-tenance therapy [4]. When used as first-line monotherapy, the objective response rate to gemcitabine is 16-22% [5-7]. As the majority of patients do not response to gemcitabine, it is urgent to figure out the biomarkers that influ-

ence the sensitivity to gemcitabine for individu-al therapy.

A substantial number of potential biomarkers for sensitivity or resistance to gemcitabine have been proposed, including gene - expression of ribonucleotide reductase subunit 1 (RRM1) [8-10], cN - II nucleotidase [11], multidrug resis-tance protein 5 (ABCC5) [12], BRCA1 [10], human equilibrate transporter - 1 [13], trans-forming growth factor beta-induced protein (TGFBI) [14], Rad51 [15] and clusterin (CLU) [16]. Owing to the different experimental mate-rials and methods used in each research, the results varied case by case, sometimes were contradictory. For example, it was reported that the mRNA expression level of RRM2 in tissues

http://www.ijcep.com

Identifying biomarker in non-small cell lung cancer cell lines

1735 Int J Clin Exp Pathol 2013;6(9):1734-1746

of NSCLC patients was markedly correlated to response to gemcitabine [10], but another study suggested that no significant change in the expression of RRM2 was observed between H358, H460 cell lines and their corresponding gemcitabine-resistant cell lines [8], respective-ly. Therefore, it is necessary to investigate sys-temically the effect of gene expressions on che-mosensitivity to gemcitabine.

MicroRNAs (miRs) are 19 - 23 nucleotides long RNAs found in multiple organisms that regulate gene expression and have been shown to play important roles in tumorigenesis. In the context of lung cancer, numerous studies have shown that tumor suppressor genes and oncogenes that play crucial roles in lung tumor develop-ment and progression are targets of miRs regu-lation. It has been documented that miRs regu-late cell viability and drug sensitivity in lung cancer [17]. In the A549 model, miR-29b down-regulation of DMNT3b reduced promoter meth-ylation of tumor suppressor genes such as Cell adhesion molecule 1 (CADM1), Ras associated (RalGDS/AF-6) domain family member 1 (RASSF1), and Fragile histidine triad gene (FHIT), increasing their expression [18]. Yet, there are few reports that miRs expression is directly related to the sensitivity to gemcitabine in NSCLC cell lines.

In this work, we employed 3 NSCLC cell lines, which are hypersensitive, moderate sensitive and resistant, respectively to gemcitabine. The mRNA and miR expression chips were carried out and bioinformatics analysis was performed to figure out the biomarkers for gemcitabine sensitivity. The results showed that hundreds of genes and 10 miRs were found to be mark-edly differentially expressed and the relation-ships between these expression-altered genes and miRs were investigated.

Materials and methods

Cell culture

NSCLC cell lines, NCI-H1975, NCI-H460 and SK-MES-1, were purchased from ATCC and maintained in DMEM medium supplemented with 10% FBS (Hyclone), penicillin (100 IU/ml) and Streptomycin (100 μg/ml) (Life Tech- nologies) in a humidified atmosphere of 95% air and 5% CO2 at 37°C. Cells in the exponential growth phase were used for all the experiments.

IC50 detection of NSCLC cell lines to gem-citabine by MTS assay

NCI-H1975, NCI-H460 and SK-MES-1 cells (4×103) were grown in 100 μl of DMEM medium containing serum per well in a 96-well plate. After 24 h, the cells were treated with gem-citabine (0, 2, 6.3, 20, 63, 200, 630, 2000 nmol/L, respectively) for 72 h. Every treatment was triplicate in the same experiment. Then 20 μl of MTS (CellTiter 96 AQueous One Solution Reagent; Promega) was added to each well for 1 h at 37°C. After incubation, the absorbance was read at a wavelength of 490 nm according to the manufacturer’s instructions. The IC50 calculation was performed with GraphPad Prism 5.0 software.

Microarray analysis: gene expression profile, miR expression profile

The basal gene expression profiles of NCI-H1975, NCI-H460 and SK-MES-1 cell lines were derived from Cell line genetic (mutation and copy number) and gene expression data used for EN analysis on Cancer Centre of Sanger Institute website (http://www.cancerrxgene.org/downloads/).

Total RNA from NCI-H1975, NCI-H460 and SK-MES-1 cell lines was processed and hybrid-ized to Affymetrix GeneChip® miRNA 2.0 Array, which recognizes 1,105 separate human miRs in accordance with the Sanger Institute miR-Base version 15. Each sample was duplicated for miR expression profile.

Genes that expression level increased or decreased by higher than 50% gradually from NCI-H1975 to NCI-H460 to SK-MES-1 were thought as significantly altered genes. MiRs whose expression level altered by higher than 100% (P<0.05) gradually from NCI-H1975 to NCI-H460 to SK-MES-1 were considered as sig-nificantly altered miRs. Gene set enrichment analysis (GSEA) was performed for the differen-tially expressed genes with DAVID 6.7 online software and Ingenuity Pathway Analysis (IPA).

miR target prediction and miRNA target cor-relation

miR target prediction was performed with miR-Walk online software. The comparative analysis was done by 5 prediction programs: miRanda, miRDB, miRWalk, RNA22 and TargetScan.



Genes predicted by greater than or equal to 3 programs were selected as putative target of some miR. The selected putative genes were done intersection with genes that expression level altered bigger than 50% gradually from NCI-H1975 to NCI-H460 to SK-MES-1. Candi- date genes that predicted by miRs of increased expression were done intersection with down-regulated genes determined by microarray, while candidate genes that mediated by miRs of decreased expression were done intersec-tion with upregulated genes determined by microarray. The overlapped genes that mediat-ed by at least 2 miRs were selected to construct miR-gene networks with the aid of Cytoscape 2.8 software.

Quantitative real-time PCR (qPCR)

Total RNA above isolated was synthesized to cDNA using PrimeScript RT reagent kit with gDNA Eraser (Takara, RR074A) for RT-PCR with mixture of oli-go-dT and Random Primer (9 mer). The primers used for qPCR validation were list in Table 1. Real-time qPCR was performed on CFX-96 (Bio-lab), with endogenous control hActb. Gene expression was calculated relative to expression of hActb endogenous control and adjusted relative to expression in SK-MES-1 cells.

Statistical analysis

R2 values were calculated using Pearson’s correlation coefficient. The significant difference was calculated using Student’s t-test.

Results

The IC50s of gemcitabine examination in three NSCLC cell lines

Three NSCLC cell lines, NCI-H1975, NCI-H460 and

Table 1. Primers for qPCR validationGENE Forward ReverseActb GCATCCCCCAAAGTTCACAA GGACTTCCTGTAACAACGCATCTEIF2S1 GAAGGCGTATCCGTTCTATCAAC AGCAACATGACGAAGAATGCTATPITPNB CAACAAAACCCACATTTCATACATC CAGTGCCCAGAGAGTAAACATCACYB5A AGCTGGAGGTGACGCTACTGA ACATTTCCCTGGCATCTGTAGAGCDK4 AAAGCCTCTCTTCTGTGGAAACTC GCAGCCCAATCAGGTCAAAGHDGF CTCTTCCCTTACGAGGAATCCA CCTTGACAGTAGGGTTGTTCTCIFITM2 ATGAACCACATTGTGCAAACCT CCCAGCATAGCCACTTCCTEIF5A CAAGGAGATTGAGCAGAAGTACGA TGGCAGACAGCACCGTGATGOLM1 GGCCCCAGAGATCGTTTGA CTCTGTAACTTCTAGGCCCCATTTTMYC GGCGAACACACAACGTCTTG TGGTCACGCAGGGCAAAPTEN GCAGTTGGCTAAGAGAGGTTTCC CTGAGCATTCCCTCCATTCC

SK-MES-1, were treated with 7 different con-centrations of gemcitabine as indicated for 72 h. Then the cell viability was determined by MTS assay (Figure 1) and IC50s of these three cell lines were calculated. IC50 dose of NCI-H1975, NCI-H460 and SK-MES-1 cells to gem-citabine at 72 h were 471 (R2=0.97), 64 (R2=0.98) and 4.7 nmol/L (R2=0.90), respec-tively. According to data reported in DTP Data Search (http://dtp.nci.nih.gov/dtpstandard/dwindex/index.jsp), the mean IC50 of NCI-60 cell panel to gemcitabine is 92-240 nmol/L, NCI-H1975 was resistant to gemcitabine, NCI-H460 was moderately sensitive to gemcitabine, while SK-MES-1 was hypersensitive to gemcitabine.

Figure 1. Three NSCLC cells showed different sensitivity to gemcitabine. MTS assay was used to determine cell viability of NCI-H1975, NCI-H460 and SK-MES-1 cells treated with gemcitabine (0, 2, 6.3, 20, 63, 200, 630, 2000 nmol/L, respectively) for 72 h. Every treatment was triplicate in the same experiment. Error bars represent the standard deviation (SD).



Differential gene expression profiles of the three NSCLC cell lines

To find differential gene expression profiles of 3 NSCLC cell lines, cell line genetic (mutation and copy number) and gene expression data used for EN analysis on Cancer Centre of Sanger Institute website (http://www.cancerrxgene.org/downloads/) were download and used for analysis. Genes whose expression increased or decreased by higher than 50% gradually from SK-MES-1 to NCI-H460 to NCI-H1975 were considered as significantly altered genes. There were 1,797 genes of increased expression and 3,042 genes of decreased expression from SK-MES-1 to NCI-H460 to NCI-H1975. When the selection criterion was set as higher than 100%, there were 159 genes of increased expression and 716 genes of decreased expression. The most significantly differential-ly-expressed genes were BCL11B, MAP2, BCL2, IGF1, MYCN, BCL10, POLE3 and ZNF468 (Table 2). Interestingly, TP53, Myc and PTEN are upregulated in gemcitabine-resistant NSCLC cell lines. Expression of TP53, Myc and PTEN was increased by 4.0-, 2.6- and 2.0-fold

in NCI-H460, increased by 8.0-, 11.2- and 3.1-fold in NCI-H1975, respectively, compared to that in SK-MES-1.

GSEA were done to the differential gene expres-sion profiles of 3 NSCLC cell lines with DAVID 6.7 online software (Figure 2 and Table 3). The 716 genes with decreased expression were mainly enriched in cell adhesion (NRP2, CXCR3, CDK5R1, IL32 and CDH2), disease mutation (SCN1A, FGFR3, NFKBIA, MAF and JAK2), ECM-receptor interaction (COL4A2, COL4A1 and ITGA8) and cell motion (NRP2, CDK5R1, CXCR3, CDH2, POU4F1, IGF1, ISL1, JAK2 and BMP7). The 159 expression-increased genes were mainly enriched in histone-nucleosome-chro-matin assembly (H1F0 and HIST1H2AC), meth-ylation modification (HIST1H2AC, RAB8A and TP53), response to extracellular stimulus (MUC1, TP53, and IGFBP2) and tight junction (CSNK2A1 and GNAI1).

And then, the 875 genes (159 upregulated genes and 716 down-regulated genes in gem-citabine-resistant cell line) were applied to inge-nuity pathway analysis (IPA) systems. IPA

Table 2. The most significantly differentially-expressed genes in 3 NSCLC cell lines

geneRelative expression/Fold

DescriptionH1975 H460 SK-MES-1

BCL11B 1.0 22.0 40.9 B-cell CLL/lymphoma 11BCBX4 1.0 21.4 43.2 chromobox homolog 4MAP2 1.0 16.1 39.3 microtubule-associated protein 2HLF 1.0 15.0 37.0 hepatic leukemia factorMPDZ 1.0 13.5 43.3 multiple PDZ domain proteinATXN7L1 1.0 13.4 31.3 ataxin 7-like 1RAPGEF1 1.0 12.8 26.0 Rap guanine nucleotide exchange factor (GEF) 1BCL2 1.0 4.1 6.5 B-cell CLL/lymphoma 2IGF1 1.0 2.6 8.1 insulin-like growth factor 1MYCN 1.0 1.6 6.5 v-myc myelocytomatosis viral related oncogene, neuroblastoma derived

(avian)BCL10 1.0 -1.7 -2.9 B-cell CLL/lymphoma 10SUCLG2 1.0 -2.4 -5.5 succinate-CoA ligase, GDP-forming, beta subunitRAB23 1.0 -2.4 -34.2 RAB23, member RAS oncogene familyPOLE3 1.0 -2.5 -6.4 polymerase (DNA directed), epsilon 3, accessory subunitLHFPL2 1.0 -2.9 -8.2 lipoma HMGIC fusion partner-like 2RPS6KA3 1.0 -3.2 -8.9 ribosomal protein S6 kinase, 90 kDa, polypeptide 3CPD 1.0 -3.5 -9.5 carboxypeptidase DLIPG 1.0 -4.9 -16.2 lipase, endothelialKCTD14 1.0 -6.4 -22.4 potassium channel tetramerization domain containing 14ZNF468 1.0 -6.8 -15.9 zinc finger protein 468‘‘-’’ in Relative expression columns means down-regulated compared to that in NCI-H1975.



results showed that Wnt/β-catenin signaling, tight junction signaling, aryl hydrocarbon recep-tor (AHR) signaling and p38MAPK signaling were dramatically altered. Within these altered signaling, Wnt/β-catenin signaling and p38MAPK signaling were predicted to be acti-vated and repressed, respectively. The upregu-lation of β-catenin (CTNNB1, increased by 2.4-fold in NCI-H1975 compared to that in

SK-MES-1), downregulation of SOX (including SOX3, SOX10 and SOX14, decreased by 52.5-, 26.2- and 8.8-fold in NCI-H1975 compared to that in SK-MES-1, respectively.) and predicted inactivation of NLK by downregulation of MAP3K7IP1 (TAB1, decreased by 21.1-fold in NCI-H1975 compared to that in SK-MES-1) con-currently activated the β-catenin/LEF/CBP transcription complex and hence promote the

Figure 2. GSEA of genes highly expressed in gemcitabine sensitive cell lines and resistant cell line, respectively, according to pathway/cell process using online software DAVID 6.7. The black column represents the upregulated pathway/cell process in sensitive cell lines, while the grey column represents the upregulated pathway/cell process in resistant cell line.

Table 3. GSEA for differentially-expressed genes

pathway/cell process Up/Down* Count P value Genes

cell adhesion down 56 1.25E-07 NRP2, IGFBP7, CLDN5, CXCR3, CDH22, CDH2, CNTNAP2, CDK5R1, COL2A1, IL32, LY6D,

COL7A1, COL8A2, LAMA4, DSG3, CDON, etcdisease mutation down 85 4.46E-06 FGF8, SLC7A7, IHH, SOX10, CYP11A1, NCF2, TF,

SCN1A, FGFR3, ABCD1, SOX3, NFKBIA, COL2A1, FOXP3, TSC2, JAK2, MAPK8IP1, etc.

ECM-receptor interaction down 14 2.66E-05 IBSP, COL4A2, COL4A1, TNXB, HSPG2, COL2A1, GP9, LAMA4, GP5, CD36, GP1BB, ITGA8, SV2A,

SV2Csecretory granule down 17 9.26E-04 TF, A2M, ARC, ICA1, NCF2, SEMG1, IGF1, GP9,

AZU1, SLC11A1, GP5, CD36, F5, CAPN11, FGA, GP1BB, CHGB

cell motion down 32 0.002 NRP2, DRD1, CDK5R1, PLXNA3, CXCR3, CDH2, DOCK2, CXCR4, TEKT2, POU3F2, POU4F1,

SPON2, VNN2, EFNB3, IGF1, ISL1, CD34, JAK2, BMP7, etc.

nucleosome up 6 1.24E-04 H1F0, HIST1H2AC, HIST1H2BD, H2BFS, HIST1H2BE, HIST1H2BF, HIST1H3H

methylation up 8 0.002 HIST1H2AC, RAB8A, HIST1H2BD, H2BFS, HIST1H2BE, HIST1H2BF, RAB23, TP53,

HIST1H3Hoxidoreductase up 11 0.01 TP53I3, CYP27B1, EHHADH, AKR1B1, HS-

D17B6, OXNAD1, PRDX2, CP, DHRS7B, SC5DL, SOD3

response to extracellular stimulus up 7 0.01 MUC1, CYP27B1, LIPG, TP53, CP, IGFBP2, AQP3Tight junction up 6 0.01 CSNK2A1, CLDN4, GNAI1, EXOC4, CASK, TJP2FDR<20%. *, up means upregulated in gemcitabine resistant cell lines, while down means down-regulated in gemcitabine resis-tant cell lines.



expression of a series of downstream genes, including OCT4, Myc, Cyclin D1, TCF1, MMP7 et al (Figure 3). Highly expression of OCT4 may confer the cancer cell pluripotency.

The confirmation of microarray data by qPCR

10 genes whose expression level increased gradually from SK-MES-1 to H460 to H1975 were selected to perform qPCR to validate the microarray data. The qPCR results showed that

expression trends of majority of genes were consistent with the microarray data, except for that of EIF2S1 and PIPTNB (Figure 4). Although the expression of these two genes were decreased gradu-ally from H1975 to H460 to SK-MES-1, but the expression changes had no significant differ-ence (EIF2S1: p=0.06 and 0.3, respectively, for H1975 and H460 compared to SK-MES-1; PIPTNB: p=0.06 and 0.01, respectively, for H1975 and H460 compared to SK-MES-1). In spite of this, expression trends of majority of genes (8/10) were consistent with the microarray data, sug-gesting that online data from Sanger Institute is reliable for

Figure 3. Wnt/β-catenin signaling was activated in gemcitabine resistant NSCLC cell lines. The 875 expression-altered genes were applied to IPA. IPA results showed that Wnt/β-catenin signaling was one of the most significantly activated pathway in gemcitabine resistant NSCLC cell lines. P value is 0.002 and 15 genes in the 875 gene-set were in-volved in this signaling pathway.

Figure 4. qPCR validation of microarray results. 10 genes whose ex-pression increased in gemcitabine resistant cell line were chosen for qPCR. The expression of each gene was adjusted relative to expres-sion in SK-MES-1 cells. Error bars represent the standard deviation (SD).

gene expression profile analysis. Intriguingly, the upregulation of Myc and PTEN in gem-citabine-resistant NSCLC cell lines was vali- dated.

The miR expression patterns in the three NSCLC cell lines

Differential expression analysis showed that there were only 10 miRs whose expression altered higher than 100% (p<0.05) gradually



from NCI-H1975 to NCI-H460 to SK-MES-1 (Table 4). Within these miRs, expression of miR-155, miR-10a, miR-30a, miR-24-2* and miR-30c-2* was down-regulated in resistant cell lines, while expression of miR-200c, miR-203, miR-885-5p, miR-195 and miR-25* was increased gradually from SK-MES-1 to NCI-H460 to NCI-H1975. The putative targets of these 10 miRs were also partially list in Table 4.

MiRs-genes network construction

To observe the interactions between signifi-cantly altered miRs and genes in these 3 NSCLC cell lines, the predicted targets of the 10 miRs were done intersections, respectively, with genes whose expression significantly altered to find the putative targets of these miRs in NSCLC cell lines. Candidate genes predicted by miRs of increased expression were done intersection with down-regulated genes determined by microarray, while candidate genes mediated by miRs decreased expression were done inter-section with upregulated genes determined by microarray. Amounts of overlapped genes for each miR were showed in Table 4. Expression

of 604 genes was mediated by at least one of the 10 miRs. And then the putative targets of each miR of increased or decreased expres-sion were done intersection with the other four miRs of increased or decreased expression, respectively. The overlapped genes mediated by more than or equal to 2 miRs were used for miRs-genes network construction with the aid of Cytoscape 2.8 software (Figure 5). Within these genes, expression of 95 genes decreased gradually and that of 42 genes increased grad-ually from SK-MES-1 to H460 to H1975. Within the 95 genes, amounts of genes whose decreased expression was mediated by miRs of increased expression were as follow: only one gene (KCNMA1) by 5 miRs; 14 genes (AKAP13, BCL2, IKBKB, MMP16, etc.) by 4 miRs; 29 genes (BCL11B, IGF1, IGSF3, KSR1, PAK7, etc.) by 3 miRs; the remaining was medi-ated by 2 miRs. Within the 42 genes, amounts of genes whose increased expression was mediated by miRs of decreased expression were as follow: 2 genes (IMPAD1 and ZNF264) by 5 miRs; 2 genes (SEMA5A and SUCLG2) by 4 miRs; 16 genes (CHM, GNA13, POLE3, RAB23, etc.) by 3 miRs; the remaining 22 genes (BCL10,

Table 4. The most differentially expressed miRs in 3 NSCLC cell lines

miRH460 vs H1975 SK-MES-1 vs H1975

Hits** Putative prediction targetsFold change P Fold change P

miR-155 +4.4 0.005 +1699.6 0.0002 60 ARHGAP29, BCL10, CCNJL, CHM, DDAH1, EXOC5, FDX1, GOLPH3L

miR-10a +4.7 0.03 +40.6 0.001 41 CHM, DDAH1, ELOVL6, EXOC5, FDX1, GOLPH3L, IMPAD1, OSMR, RAP2A, SSR1,

ZNF264miR-24-2* +5.1 0.003 +12.1 0.001 30 ARHGAP5, ENTPD7, IMPAD1, SEMA5A,

SSR1, SUCLG2, ZNF264miR-30c-2* +2.5 0.11 +8.2 0.03 108 ARHGAP5, CHM, DDX51, FDX1, GNA13,

GOLPH3LmiR-30a +2.7 0.02 +6.4 0.002 82 ARHGAP29, BCL10, CCNJL, DDAH1, GNA13,

IMPAD1, miR-200c -117.2 0.0002 -1719.0 0.0001 147 AKAP13, BCL2, BCL11B, CDK5R1, DISC1,

FGF23, IGF1, miR-203 -33.8 0.001 -218.9 0.0003 160 AKAP13, ATRX, BCL2, BCL11B, CNTNAP2,

DISC1, HLF, IGF1, miR-885-5p -12.7 0.0002 -30.3 0.001 81 AKAP13, BCL2, DISC1, DZIP1, IGSF3, IKBKB,

MMP16miR-195 -4.7 0.01 -26.2 0.002 190 AKAP13, BCL2, BCL11B, CDK5R1, FGF23,

IGF1, IGSF3, IKBKB, MMP16, PAK7miR-25* -2.1 0.01 -5.9 0.0003 58 ASB1, CA5B, CCNL2, CDON, DZIP1, KLK7,

MYCN, POLH“+” in fold change columns means upregulated in sensitive cell lines; “-” in fold change columns means down-regulated in sensi-tive cell lines. **, hits means amount of overlapped genes between predicted target genes of some miR and genes whose expres-sion was increased or decreased bigger than 50% gradually from H1975 to H460 to SK-MES-1. For miR-24-2*, miR-30c-2* and miR-25*, genes predicted by greater than or equal to 2 programs were selected as putative targets.



RAP2A, etc.) was mediated by 2 miRs of decreased expression. The genes of increased

or decreased expression were applied for func-tional analysis (Table 5). The 95 genes of

Figure 5. MiRs-genes network. Genes whose expression was putatively mediated by more than or equal to 2 miRs were selected to construct network with corresponding miRs with the aid of Cytoscape 2.8. The yellow square rep-resents miR, and the pink cycle represents mediated gene. A: Network of miRs whose expression was increased with genes whose expression was decreased in gemcitabine sensitive NSCLC cell lines. There were 42 genes in this figure. B: Network of miRs whose expression was increased with genes whose expression was decreased in gem-citabine resistant NSCLC cell lines. There were 95 genes in this figure.



decreased expression were mainly enriched in chromosomal rearrangement (MAF, HLF, BCL2, IGSF3, CNTNAP2 and DISC1), protein kinase activity (NRP2, PAK7, PRKCQ, RNASEL, CDK5R1, PLK2, AKAP13, IKBKB and KSR1), cell-cell junction (SCN1A, INADL, MPDZ, TLN2 and ASH1L), Proto-oncogene (MAF, HLF, BCL2, AKAP13 and MYCN) and anti-apoptosis (PAK7, BCL2, CBX4, IGF1 and IKBKB). The 42 genes of increased expression were mainly enriched in small GTPase mediated signal transduction (GNA13, RAP2A, ARHGAP5, RAB23 and ARH- GAP29), oncostatin-M receptor activity (OSMR and LIFR) and myosin pathway (GNA13, ARHGAP5).

Discussion

Gemcitabine is a widely used drug in the treat-ment of advanced NSCLC, yet the modest objective response rate in patients to gem-citabine suggests identifying novel biomarkers for gemcitabine sensitivity prediction could do benefit to patients with gemcitabine-based therapy and improving the therapy effect. In this work, we used DNA microarray for gene and miR expression profile analysis to find the differentially expressed genes and miRs in 3 NSCLC cell lines. Hundreds of genes and 10 miRs were found to be markedly differentially expressed and the relationships between these expression-altered genes and miRs were investigated.

3 NSCLC cell lines, NCI-H1975, NCI-H460 and SK-MES-1, are resistant, moderate sensitive and hypersensitive, respectively, to gem-

citabine. In these 3 cell lines, there are 875 sig-nificantly differentially-expressed genes (expression difference>100%). Within these genes, expression of 716 genes decreased and that of 159 genes increased gradually from gem-citabine sensitive cell lines (NCI-H460 and SK-MES-1) to resistant cell line (NCI-H1975). It was suggested that resistance to treatment is strongly related to tumor cell adhesion to the extracellular matrix [14], and some genes asso-ciated to cell adhesion, such as transforming growth factor beta-induced protein (TGFBI) [14] and DSP [19], have been reported that their ectopic expression increases the sensitivity of NSCLC cells to apoptosis induced by gem-citabine. Our data showed the most significant-ly down-regulated pathway in the 716 genes was cell adhesion, and expression of 56 mem-bers of this pathway markedly increased in rela-tively sensitive cell lines (H460 and SK-MES-1), suggesting the importance of this pathway in mediating sensitivity of NSCLC cells to gem-citabine. Moreover, it was reported that pancre-atic tumor cells with a low secretory activity show a better response to gemcitabine [20], while our data showed that sensitive NSCLC cell lines had higher level of genes linked to secretory granule. This difference may be caused by the intrinsic characteristic of the two types of cancer and need to be further studied.

Genes highly expressed in gemcitabine resis-tant NSCLC cell line were not only potential bio-markers but also putative targets to overcome gemcitabine resistance by knockdown of these genes. These 159 genes were mainly enriched

Table 5. GSEA for differentially expressed genes mediated by significantly altered miRspathway/cell process up/down* P value FDR/% Geneschromosomal rearrangement down 0.008 9.2 MAF, HLF, BCL2, IGSF3, CNTNAP2,

DISC1protein kinase activity down 0.01 13.5 NRP2, PAK7, PRKCQ, RNASEL, CD-

K5R1, PLK2, AKAP13, IKBKB, KSR1cell-cell junction down 0.02 17.8 SCN1A, INADL, MPDZ, TLN2, ASH1LProto-oncogene down 0.02. 20.9 MAF, HLF, BCL2, AKAP13, MYCNanti-apoptosis down 0.02 26.3 PAK7, BCL2, CBX4, IGF1, IKBKBsmall GTPase mediated signal transduction up 0.006 8.5 GNA13, RAP2A, ARHGAP5, RAB23,

ARHGAP29oncostatin-M receptor activity up 0.007 7.6 OSMR, LIFRmyosin Pathway up 0.03 14.5 GNA13, ARHGAP5*, up means upregulated in gemcitabine resistant cell lines, while down means down-regulated in gemcitabine resistant cell lines.



in histone-nucleosome-chromatin assembly (H1F0 and HIST1H2AC), methylation modifica-tion (HIST1H2AC, RAB8A and TP53), oxidore-ductase (TP53I3, CYP27B1 and EHHADH), response to extracellular stimulus (MUC1, TP53, and IGFBP2) and tight junction (CSNK2A1 and GNAI1). Previous study has showed that methylation of Wnt antagonists like WIF and secreted frizzled related proteins plays roles in NSCLC stem cell maintenance and drug resis-tance [21]. Recently, it was demonstrated that methylation of TMS1 [22] and DKK3 [23] made pancreatic cancer cell resistant to gemcitabine. Since methylation of some genes can influence gemcitabine resistance, increased expression of genes linked to methylation modification, showed in our results, may enable NSCLC cells resistant to gemcitabine. Furthermore, it was proposed that silencing of oxidoreductase-linked genes, such as Heme oxygenase-1 in cholangiocarcinoma cells [24] and ribonucleo-tide reductase in pancreatic carcinoma cells [25], sensitized cancer cells to gemcitabine. These previous reports are in accordance to our data, which showed that oxidoreductases were overexpressed in gemcitabine resistant NSCLC cell line, H1975, and therefore may con-fer NSCLC cells resistance to gemcitabine.

In this work, two types of software, DAVID 6.7 and IPA, were used for GSEA. Results by DAVID 6.7 have been discussed above. IPA results showed some different pattern with DAVID 6.7, except for the tight junction signaling, which was showed to be significantly altered in resis-tant cell lines following both DAVID 6.7 and IPA analysis. IPA suggested that Wnt/β-catenin sig-naling and p38MAPK signaling were predicted to be activated and repressed, respectively, in gemcitabine resistant cell lines. Wnt/β-catenin signaling is demonstrated to play crucial roles in caner initiation and progression, its activa-tion confers cancers resistance to multiple drugs, including chemotherapy [26-28] and tar-geted therapy drugs [29-32]. So its activation in gemcitabine-resistant NSCLC cell line is not surprising. Furthermore, activation of Wnt/β-catenin signaling confers cancer cells stem-ness and hence acquires multi-drug resistance [29, 33]. In our IPA results (Figure 3), upregula-tion of OCT4, a stemness marker, may be the reason for gemcitabine resistance. Taken together, Wnt/β-catenin signaling may serve as a potential therapeutic target for NSCLC patients resistant to gemcitabine or other ther-

apy, which should be further studied in more NSCLC cell lines and patients.

It has been reported that upregulated miR-155 increases chemo-sensitivity to cisplatin in human Caski cervical cancer cells [34]. MiR-10a was showed that its inhibition enhanced sensitivity to apoptosis in myelodysplastic syn-drome [35]. MiR-30a was proposed that it sen-sitizes tumor cells to cis-platinum via suppress-ing beclin 1-mediated autophagy [36], but another report showed that overexpression of miR-30a in A549 lung cancer or BEAS-2B nor-mal lung cell lines does not alter drug sensitivi-ty to doxorubicin and cisplatin [37]. MiR-200c restores paclitaxel sensitivity of cancer cells [38] and was showed be expression-decreased in prostate cancer cells resistant to paclitaxel [39]. Dai et al reported that miR-195 was upregulated in docetaxel resistant head and neck squamous cell carcinoma cells [40], but another study showed miR-195 sensitizes human hepatocellular carcinoma cells to 5-FU by targeting BCL-w [41]. There were no report on effects of miR-24-2*, miR-30c-2*, miR-203, miR-885-5p and miR-25* on drug sensitivity. Our data also partially supported the above results, but for more detailed mechanism of actions, it is necessary to validate our data by functional experiments.

The highly expressed genes putatively mediat-ed by significantly expression-decreased miRs in gemcitabine sensitive cell lines, H460 and SK-MES-1, were mainly enriched in chromo-somal rearrangement, protein kinase activity, cell-cell junction, proto-oncogene and anti-apoptosis. The highly expressed genes puta-tively mediated by markedly down-regulated miRs in gemcitabine resistant cell line, H1975, were mainly enriched in small GTPase mediat-ed signal transduction, oncostatin-M receptor activity and myosin pathway. BCL2, linked to anti-apoptosis and chromosomal rearrange-ment, was highly expressed in sensitive cell lines in our data, suggesting its lower expres-sion is associated with gemcitabine resistance. This result is in line with the report that lower expression of BNIP3, a Bcl-2 family protein, may play an important role in the poor response to gemcitabine treatment in pancreatic cancer patients [42]. Furthermore, many studies have proposed that inhibition of protein kinase activ-ity can sensitize NSCLC cells to gemcitabine, but our data showed that gemcitabine sensi-



tive cell lines, H460 and SK-MES-1, harbored higher expression of 9 protein kinase genes (NRP2, PAK7, PRKCQ, RNASEL, CDK5R1, PLK2, AKAP13, IKBKB and KSR1) putatively mediat-ed by some miRs. This inconsistency may be caused by the intrinsic characteristic of gem-citabine. Gemcitabine, a deoxycytidine analog, is phosphorylated by deoxycytidine kinase (DCK) to active nucleotides. It was suggested that increased DCK and thymidine kinase 2 activities enhanced the sensitivity of NSCLC cells to gemcitabine [43]. So, it is possible that the upregulated 9 protein kinases in sensitive NSCLC cell lines were associated with the acti-vation of gemcitabine. Moreover, previous study has shown that inhibition of RAS, mem-ber of small GTPases and kinases, chemosen-sitized PANC-1 and SW480 cells to gemcitabine [44]. Our data showed that the genes linked to small GTPase mediated signal transduction were down-regulated and putatively repressed by some upregulated miRs in gemcitabine sen-sitive NSCLC cell lines. Additionally, amount (=604) of genes putatively mediated by altered miRs was far less than that (=4839) of genes differentially expressed in NSCLC cell lines, suggesting that gene expression alterations mediated by miRs only play partial roles in regu-lating gemcitabine sensitivity of NSCLC cells, there must be some genes themselves that are critical in regulating gemcitabine sensitivity of NSCLC cells.

Collectively, our data showed that hundreds of genes and 10 miRs were markedly differentially expressed in 3 NSCLC cell lines with different sensitivities to gemcitabine. Genes whose expression were increased dramatically in sen-sitive cell lines were mainly enriched in cell adhesion and secretory granule, while genes whose expression were increased significantly in resistant cell lines were mainly enriched in methylation modification and oxidoreductase. The most intriguing is the activation of Wnt/β-catenin signaling in gemcitabine resistant NSCLC cell lines. Expression of miR-155, miR-10a, miR-30a, miR-24-2* and miR-30c-2* was upregulated in sensitive cell lines, while expres-sion of miR-200c, miR-203, miR-885-5p, miR-195 and miR-25* was increased in resistant cell line. After miR target prediction and miRNA target correlation, genes whose expression altered significantly and putatively mediated by expression-altered miRs were mainly enriched

in chromatin assembly (upregulated in sensi-tive cells), anti-apoptosis (upregulated in sensi-tive cells), protein kinase (upregulated in sensi-tive cells) and small GTPase mediated signal transduction (down-regulated in sensitive cells). The above pathways and the involved genes are worth of further study by functional validation in more NSCLC cell lines and tissue samples, which might provide valuable biomarker(s) for Gemcitabine resistance pre-diction in clinical therapy.

Disclosure of conflict of interest

None.

Address correspondence to: Dr. Yu-Zhi Shi, Department of respiratory medicine, First Clinical Medical College affiliated to Harbin Medical University, Harbin, China. E-mail: [email protected]

References

[1] Gridelli C, Aapro M, Ardizzoni A, Balducci L, De Marinis F, Kelly K, Le Chevalier T, Manegold C, Perrone F, Rosell R, Shepherd F, De Petris L, Di Maio M and Langer C. Treatment of advanced non-small-cell lung cancer in the elderly: re-sults of an international expert panel. J Clin Oncol 2005; 23: 3125-3137.

[2] Sandler AB, Nemunaitis J, Denham C, von Pawel J, Cormier Y, Gatzemeier U, Mattson K, Manegold C, Palmer MC, Gregor A, Nguyen B, Niyikiza C and Einhorn LH. Phase III trial of gemcitabine plus cisplatin versus cisplatin alone in patients with locally advanced or met-astatic non-small-cell lung cancer. J Clin Oncol 2000; 18: 122-130.

[3] Cardenal F, Lopez-Cabrerizo MP, Anton A, Alberola V, Massuti B, Carrato A, Barneto I, Lomas M, Garcia M, Lianes P, Montalar J, Vadell C, Gonzalez-Larriba JL, Nguyen B, Artal A and Rosell R. Randomized phase III study of gemcitabine-cisplatin versus etoposide-cispla-tin in the treatment of locally advanced or met-astatic non-small-cell lung cancer. J Clin Oncol 1999; 17: 12-18.

[4] Brodowicz T, Krzakowski M, Zwitter M, Tzekova V, Ramlau R, Ghilezan N, Ciuleanu T, Cucevic B, Gyurkovits K, Ulsperger E, Jassem J, Grgic M, Saip P, Szilasi M, Wiltschke C, Wagnerova M, Oskina N, Soldatenkova V, Zielinski C, Wenczl M; Central European Cooperative Oncology Group CECOG. Cisplatin and gem-citabine first-line chemotherapy followed by maintenance gemcitabine or best supportive care in advanced non-small cell lung cancer: a

mailto:[email protected]

mailto:[email protected]



phase III trial. Lung Cancer 2006; 52: 155-163.

[5] Gridelli C, Perrone F, Gallo C, Cigolari S, Rossi A, Piantedosi F, Barbera S, Ferrau F, Piazza E, Rosetti F, Clerici M, Bertetto O, Robbiati SF, Frontini L, Sacco C, Castiglione F, Favaretto A, Novello S, Migliorino MR, Gasparini G, Galetta D, Iaffaioli RV, Gebbia V; MILES Investigators. Chemotherapy for elderly patients with ad-vanced non-small-cell lung cancer: the Multicenter Italian Lung Cancer in the Elderly Study (MILES) phase III randomized trial. J Natl Cancer Inst 2003; 95: 362-372.

[6] Ricci S, Antonuzzo A, Galli L, Tibaldi C, Bertuccelli M, Lopes Pegna A, Petruzzelli S, Algeri R, Bonifazi V, Fioretto ML, Orlandini C and Conte PF. Gemcitabine monotherapy in el-derly patients with advanced non-small cell lung cancer: a multicenter phase II study. Lung Cancer 2000; 27: 75-80.

[7] Gridelli C, Cigolari S, Gallo C, Manzione L, Ianniello GP, Frontini L, Ferrau F, Robbiati SF, Adamo V, Gasparini G, Novello S, Perrone F; MILES Investigators. Activity and toxicity of gemcitabine and gemcitabine + vinorelbine in advanced non-small-cell lung cancer elderly patients: Phase II data from the Multicenter Italian Lung Cancer in the Elderly Study (MILES) randomized trial. Lung Cancer 2001; 31: 277-284.

[8] Davidson JD, Ma L, Flagella M, Geeganage S, Gelbert LM and Slapak CA. An increase in the expression of ribonucleotide reductase large subunit 1 is associated with gemcitabine resis-tance in non-small cell lung cancer cell lines. Cancer Res 2004; 64: 3761-3766.

[9] Tooker P, Yen WC, Ng SC, Negro-Vilar A and Hermann TW. Bexarotene (LGD1069, Targretin), a selective retinoid X receptor ago-nist, prevents and reverses gemcitabine resis-tance in NSCLC cells by modulating gene am-plification. Cancer Res 2007; 67: 4425-4433.

[10] Boukovinas I, Papadaki C, Mendez P, Taron M, Mavroudis D, Koutsopoulos A, Sanchez-Ronco M, Sanchez JJ, Trypaki M, Staphopoulos E, Georgoulias V, Rosell R and Souglakos J. Tumor BRCA1, RRM1 and RRM2 mRNA expression levels and clinical response to first-line gem-citabine plus docetaxel in non-small-cell lung cancer patients. PLoS One 2008; 3: e3695.

[11] Seve P, Mackey JR, Isaac S, Tredan O, Souquet PJ, Perol M, Cass C and Dumontet C. cN-II ex-pression predicts survival in patients receiving gemcitabine for advanced non-small cell lung cancer. Lung Cancer 2005; 49: 363-370.

[12] Oguri T, Achiwa H, Sato S, Bessho Y, Takano Y, Miyazaki M, Muramatsu H, Maeda H, Niimi T and Ueda R. The determinants of sensitivity and acquired resistance to gemcitabine differ

in non-small cell lung cancer: a role of ABCC5 in gemcitabine sensitivity. Mol Cancer Ther 2006; 5: 1800-1806.

[13] Danesi R, Altavilla G, Giovannetti E and Rosell R. Pharmacogenomics of gemcitabine in non-small-cell lung cancer and other solid tumors. Pharmacogenomics 2009; 10: 69-80.

[14] Irigoyen M, Pajares MJ, Agorreta J, Ponz-Sarvise M, Salvo E, Lozano MD, Pio R, Gil-Bazo I and Rouzaut A. TGFBI expression is associat-ed with a better response to chemotherapy in NSCLC. Mol Cancer 2010; 9: 130.

[15] Tsai MS, Kuo YH, Chiu YF, Su YC and Lin YW. Down-regulation of Rad51 expression over-comes drug resistance to gemcitabine in hu-man non-small-cell lung cancer cells. J Pharmacol Exp Ther 2010; 335: 830-840.

[16] Cheng CY, Cherng SH, Wu WJ, Yang TY, Huang XY, Liao FT, Wu MF and Sheu GT. Regulation of chemosensitivity and migration by clusterin in non-small cell lung cancer cells. Cancer Chemother Pharmacol 2012; 69: 145-154.

[17] Du L and Pertsemlidis A. microRNA regulation of cell viability and drug sensitivity in lung can-cer. Expert Opin Biol Ther 2012; 12: 1221-1239.

[18] Samakoglu S, Deevi DS, Li H, Wang S, Murphy M, Bao C, Bassi R, Prewett M and Tonra JR. Preclinical rationale for combining an EGFR antibody with cisplatin/gemcitabine for the treatment of NSCLC. Cancer Genomics Proteomics 2012; 9: 77-92.

[19] Yang L, Chen Y, Cui T, Knosel T, Zhang Q, Albring KF, Huber O and Petersen I. Desmoplakin acts as a tumor suppressor by inhibition of the Wnt/beta-catenin signaling pathway in human lung cancer. Carcinogenesis 2012; 33: 1863-1870.

[20] Wissniowski TT, Meister S, Hahn EG, Kalden JR, Voll R and Ocker M. Mucin production de-termines sensitivity to bortezomib and gem-citabine in pancreatic cancer cells. Int J Oncol 2012; 40: 1581-1589.

[21] Reya T and Clevers H. Wnt signalling in stem cells and cancer. Nature 2005; 434: 843-850.

[22] Ramachandran K, Miller H, Gordian E, Rocha-Lima C and Singal R. Methylation-mediated si-lencing of TMS1 in pancreatic cancer and its potential contribution to chemosensitivity. Anticancer Res 2010; 30: 3919-3925.

[23] Gu YM, Ma YH, Zhao WG and Chen J. Dickkopf3 overexpression inhibits pancreatic cancer cell growth in vitro. World J Gastroenterol 2011; 17: 3810-3817.

[24] Kongpetch S, Kukongviriyapan V, Prawan A, Senggunprai L, Kukongviriyapan U and Buranrat B. Crucial role of heme oxygenase-1 on the sensitivity of cholangiocarcinoma cells



to chemotherapeutic agents. PLoS One 2012; 7: e34994.

[25] Funamizu N, Kamata Y, Misawa T, Uwagawa T, Lacy CR, Yanaga K and Manome Y. Hydroxyurea decreases gemcitabine resistance in pancre-atic carcinoma cells with highly expressed ribo-nucleotide reductase. Pancreas 2012; 41: 107-113.

[26] Flahaut M, Meier R, Coulon A, Nardou KA, Niggli FK, Martinet D, Beckmann JS, Joseph JM, Muhlethaler-Mottet A and Gross N. The Wnt receptor FZD1 mediates chemoresistance in neuroblastoma through activation of the Wnt/beta-catenin pathway. Oncogene 2009; 28: 2245-2256.

[27] Shen DY, Zhang W, Zeng X and Liu CQ. Inhibition of Wnt/beta-catenin signaling down-regulates P-glycoprotein and reverses multi-drug resistance of cholangiocarcinoma. Cancer Sci 2013; [Epub ahead of print].

[28] Gao Y, Liu Z, Zhang X, He J, Pan Y, Hao F, Xie L, Li Q, Qiu X and Wang E. Inhibition of cytoplas-mic GSK-3beta increases cisplatin resistance through activation of Wnt/beta-catenin signal-ing in A549/DDP cells. Cancer Lett 2013; 336: 231-239.

[29] Yeung J, Esposito MT, Gandillet A, Zeisig BB, Griessinger E, Bonnet D and So CW. beta-Catenin mediates the establishment and drug resistance of MLL leukemic stem cells. Cancer Cell 2010; 18: 606-618.

[30] Zhang B, Li M, McDonald T, Holyoake TL, Moon RT, Campana D, Shultz L and Bhatia R. Microenvironmental protection of CML stem and progenitor cells from tyrosine kinase in-hibitors through N-cadherin and Wnt-beta-catenin signaling. Blood 2013; 121: 1824-1838.

[31] Wu Y, Ginther C, Kim J, Mosher N, Chung S, Slamon D and Vadgama JV. Expression of Wnt3 activates Wnt/beta-catenin pathway and promotes EMT-like phenotype in trastuzumab-resistant HER2-overexpressing breast cancer cells. Mol Cancer Res 2012; 10: 1597-1606.

[32] Tenbaum SP, Ordonez-Moran P, Puig I, Chicote I, Arques O, Landolfi S, Fernandez Y, Herance JR, Gispert JD, Mendizabal L, Aguilar S, Ramon y Cajal S, Schwartz S Jr, Vivancos A, Espin E, Rojas S, Baselga J, Tabernero J, Munoz A and Palmer HG. β-catenin confers resistance to PI3K and AKT inhibitors and subverts FOXO3a to promote metastasis in colon cancer. Nat Med 2012; 18: 892-901.

[33] He K, Xu T, Xu Y, Ring A, Kahn M and Goldkorn A. Cancer cells acquire a drug resistant, highly tumorigenic, cancer stem-like phenotype through modulation of the PI3K/Akt/beta-catenin/CBP pathway. Int J Cancer 2013; [Epub ahead of print].

[34] Lei C, Wang Y, Huang Y, Yu H, Huang Y, Wu L and Huang L. Up-regulated miR155 reverses the epithelial-mesenchymal transition induced by EGF and increases chemo-sensitivity to cis-platin in human Caski cervical cancer cells. PLoS One 2012; 7: e52310.

[35] Li X, Xu F, Chang C, Byon J, Papayannopoulou T, Deeg HJ and Marcondes AM. Transcriptional regulation of miR-10a/b by TWIST-1 in myelo-dysplastic syndromes. Haematologica 2013; 98: 414-419.

[36] Zou Z, Wu L, Ding H, Wang Y, Zhang Y, Chen X, Chen X, Zhang CY, Zhang Q and Zen K. MicroRNA-30a sensitizes tumor cells to cis-platinum via suppressing beclin 1-mediated autophagy. J Biol Chem 2012; 287: 4148-4156.

[37] Patnaik SK, Kannisto E and Yendamuri S. Overexpression of microRNA miR-30a or miR-191 in A549 lung cancer or BEAS-2B normal lung cell lines does not alter phenotype. PLoS One 2010; 5: e9219.

[38] Cochrane DR, Spoelstra NS, Howe EN, Nordeen SK and Richer JK. MicroRNA-200c mitigates invasiveness and restores sensitivity to micro-tubule-targeting chemotherapeutic agents. Mol Cancer Ther 2009; 8: 1055-1066.

[39] Singh S, Chitkara D, Mehrazin R, Behrman SW, Wake RW and Mahato RI. Chemoresistance in prostate cancer cells is regulated by miRNAs and Hedgehog pathway. PLoS One 2012; 7: e40021.

[40] Dai Y, Xie CH, Neis JP, Fan CY, Vural E and Spring PM. MicroRNA expression profiles of head and neck squamous cell carcinoma with docetaxel-induced multidrug resistance. Head Neck 2011; 33: 786-791.

[41] Yang X, Yin J, Yu J, Xiang Q, Liu Y, Tang S, Liao D, Zhu B, Zu X, Tang H and Lei X. miRNA-195 sensitizes human hepatocellular carcinoma cells to 5-FU by targeting BCL-w. Oncol Rep 2012; 27: 250-257.

[42] Akada M, Crnogorac-Jurcevic T, Lattimore S, Mahon P, Lopes R, Sunamura M, Matsuno S and Lemoine NR. Intrinsic chemoresistance to gemcitabine is associated with decreased ex-pression of BNIP3 in pancreatic cancer. Clin Cancer Res 2005; 11: 3094-3101.

[43] Sigmond J, Bergman AM, Leon LG, Loves WJ, Hoebe EK and Peters GJ. Staurosporine in-creases toxicity of gemcitabine in non-small cell lung cancer cells: role of protein kinase C, deoxycytidine kinase and ribonucleotide re-ductase. Anticancer Drugs 2010; 21: 591-599.

[44] Gana-Weisz M, Halaschek-Wiener J, Jansen B, Elad G, Haklai R and Kloog Y. The Ras inhibitor S-trans,trans-farnesylthiosalicylic acid chemo-sensitizes human tumor cells without causing resistance. Clin Cancer Res 2002; 8: 555-565.

Documents

Original Article Array analysis for potential biomarker of ... · Array analysis for potential biomarker of gemcitabine identification in non-small cell ... cell lines to gem-citabine