miR-181b modulates multidrug resistance by targeting BCL2 inhuman cancer cell lines

Wei Zhu1, Xia Shan2, Tongshan Wang1, Yongqian Shu1,3 and Ping Liu1,3

1 Department of Oncology, First Affiliated Hospital of Nanjing Medical University, Nanjing, China2 Cancer Center of Nanjing Medical University, Nanjing, China3 The Fourth Class of the Combined Bachelor’s/Master’s Degree Program of Clinic Medicine, First Clinical Medical College of Nanjing Medical University,

Nanjing, China

MicroRNAs (miRNAs) are short noncoding RNA molecules, which posttranscriptionally regulate genes expression and play

crucial roles in diverse biological processes, such as development, differentiation, apoptosis and proliferation. Here, we

investigated the possible role of miRNAs in the development of multidrug resistance (MDR) in human gastric and lung cancer

cell lines. We found that miR-181b was downregulated in both multidrug-resistant human gastric cancer cell line SGC7901/

vincristine (VCR) and multidrug-resistant human lung cancer cell line A549/cisplatin (CDDP), and the downregulation of miR-

181b in SGC7901/VCR and A549/CDDP cells was concurrent with the upregulation of BCL2 protein, compared with the

parental SGC7901 and A549 cell lines, respectively. In vitro drug sensitivity assay demonstrated that overexpression of

miR-181b sensitized SGC7901/VCR and A549/CDDP cells to anticancer drugs, respectively. The luciferase activity of a BCL2

30-untranslated region-based reporter construct in SGC7901/VCR and A549/CDDP cells suggests that a new target site in the

30UTR of BCL2 of the mature miR-181s (miR-181a, miR-181b, miR-181c and miR-181d) was found. Enforced miR-181b

expression reduced BCL2 protein level and sensitized SGC7901/VCR and A549/CDDP cells to VCR-induced and CDDP-induced

apoptosis, respectively. Taken together, our findings suggest that miR-181b could play a role in the development of MDR in

both gastric and lung cancer cell lines, at least in part, by modulation of apoptosis via targeting BCL2.

Multidrug resistance (MDR) constitutes a major obstacle tosuccessful chemotherapy in cancer patients. In many cases,chemotherapies fail because of MDR of cancer cells eitherintrinsic or acquired after an initial round of treatment.1 Accu-mulating studies indicate that there are 3 major mechanismsof drug resistance in cells: first, decreased uptake of water-soluble drugs; second, various changes in cells that affect thecapacity of cytotoxic drugs to kill cells, including alterations in

cell cycle, enhanced DNA repair activity, defective apoptosispathway, altered metabolism of drugs, etc.2–4; third, increasedenergy-dependent efflux of hydrophobic drugs, represented byoverexpression of a family of energy-dependent transporters,known as ATP-binding cassette transporters, such as P-glyco-protein (Pgp), breast cancer resistance protein, etc.5 Micro-RNAs (miRNAs) are short noncoding RNA molecules, whichposttranscriptionally regulate genes expression and play crucialroles in diverse biological processes, such as development, dif-ferentiation, apoptosis and proliferation.6–9 For instance, Si etal. found that miR-21 may play a role in antiapoptosis eitherin vitro or in the xenograft mouse model.10

Currently, extensive studies have indicated that the acqui-sition of drug resistance by cancer cells may also be modu-lated via the changes in miRNA levels,11–15 for instance,the miRNA profile in a panel of paclitaxel- and cisplatin(CDDP)-resistant ovarian cancer cells showed that 6 miRNAs(let-7e, miR-30c, miR-125b, miR-130a and miR-335) werealways diversely expressed in all the resistant cell lines,12

which suggest that these 6 miRNAs may be related with drugresistance and alter the levels of the drug resistance–relatedmiRNAs, which maybe an effective way to modulate drugresistance of cancer cells. Exactly, upregulating miRNA-451sensitized doxorubicin-resistant MCF-7 cells to doxorubicinby downregulating its target protein Pgp.13 Moreover, inmultidrug-resistant gastric cancer cell line SGC7901/VCR,miR-15b and miR-16 were downregulated, compared with itsparental SGC7901 cell line. Upregulating miR-15b and miR-

Key words: microRNA, miR-181b, human cancer, multidrug

resistance, BCL2

Abbreviations: 5-FU: 5-fluorouracil; ADR: adriamycin; CDDP:

cisplatin; MDR: multidrug resistance; miRNAs: microRNAs; MMC:

mitomycin C; VCR: vincristine; VP-16: etoposide

Additional Supporting Information may be found in the online

version of this article

Wei Zhu and Xia Shan contributed equally to this work

Grant sponsor: National Natural Science Foundation of China;

Grant number: 30840095; Grant sponsor: Cancer Center of

Nanjing Medical University; Grant number: 08ZLKF02

DOI: 10.1002/ijc.25260

History: Received 11 Dec 2009; Accepted 2 Feb 2010; Online 16 Feb

2010

Correspondence to: Ping Liu, Department of Oncology, First

Affiliated Hospital of Nanjing Medical University, 300 Guangzhou

Road, Nanjing 210029, China, Tel.: þ86-25-83718836-6080, Fax:

þ86-25-83780826, E-mail: liu-ping@csco.org.cn (or) belliwether@

163.com

Can

cerCellBiology

Int. J. Cancer: 127, 2520–2529 (2010) VC 2010 UICC

International Journal of Cancer

IJC

16 could sensitize SGC7901/VCR cells to VCR-inducedapoptosis via targeting BCL2.14 Recently, let-7a was foundoverexpressed in a doxorubicin-resistant human squamouscarcinoma A431 cells subline, and downregulation of let-7aincreased the doxorubicin-induced apoptosis through target-ing caspase-3.15 Except for above 4 studies, more and moreevidence verified that miRNAs may play a role in chemore-sistance,16 and modulation of some drug resistance–relatedmiRNAs may increase the sensitivity of cancer cells to chem-otherapy drugs.

In this study, we reported that miR-181b was downregu-lated in both multidrug-resistant human gastric cancer cellline SGC7901/VCR and multidrug-resistant human lung can-cer cell line A549/CDDP, compared with the parentalSGC7901 and A549 cell lines, respectively. We demonstratedthat miR-181b may play a role in the development of MDRin human cancer cell lines by targeting the antiapoptoticgene BCL2.

Material and MethodsCell culture

Human gastric adenocarcinoma cell line SGC7901 (obtainedfrom Academy of Military Medical Science, Beijing, China)and its multidrug-resistant variant SGC7901/VCR (obtainedfrom the State Key Laboratory of Cancer Biology and Insti-tute of Digestive Diseases, Xijing Hospital, Fourth MilitaryMedical University, Xian, China), human lung cancer cellline A549 and its multidrug-resistant variant A549/CDDP(both obtained from Biosis Biotechnology Company, Shang-hai, China) were all cultured in RPMI-1640 medium supple-mented with 10% fetal calf serum (Gibco BRL, Grand Island,NY) in a humidified atmosphere containing 5% CO2 at37�C.

To maintain the MDR phenotype, vincristine (VCR, withfinal concentration of 1 lg/ml) and CDDP (with final con-centration of 4 lg/ml) were added to the culture media forSGC7901/VCR and A549/CDDP cells, respectively.

miRNA microarray analysis

Before experimentation, SGC7901/VCR cells were cultured 1week without VCR. Total RNA from SGC7901 andSGC7901/VCR cell lines was isolated with Trizol reagent(Invitrogen, Carlsbad, CA), and miRNA fraction was furtherpurified using a mirVanaTM miRNA isolation kit (Ambion,Austin, TX). The isolated miRNAs from the 2 cell lines werethen labeled with Hy3 using the miRCURYTM Array Labelingkit (Exiqon, Vedbaek, Denmark) and hybridized, respectively,on a miRCURY TM LNA microRNA Array (v 8.0, Exiqon)as described.17 Microarray images were acquired using aGenepix 4000B scanner (Axon Instruments, Union City, CA),processed and analyzed with Genepix Pro 6.0 software (AxonInstruments).

Real-time quantification of miRNAs by stem-loop RT-PCR

The RNA preparation was as described above. The concen-tration and purity of the RNA samples were determinedspectroscopically. For the TaqMan-based real-time reversetranscription-polymerase chain reaction (RT-PCR) assays, theABI 7300 HT Sequence Detection system (Applied Biosys-tem, Foster City, CA) was used. All the primers and probesof hsa-miR-181b (P/N: 4373116) and RNU6B endogenouscontrols (P/N: 4373381) for TaqMan miRNA assays werepurchased from Applied Biosystems. Real-time PCR was per-formed as described.18 The relative amount of each miRNAwas normalized to U6 snRNA. The fold change for miRNAfrom SGC7901/VCR cells and A549/CDDP cells relative toeach control SGC7901 and A549 cells was calculated usingthe 2�DDCt Method,19 where DDCt ¼ DCt SGC7901/VCR� DCt SGC7901 or DDCt ¼ DCt A549/CDDP � DCt A549and DCt ¼ Ct miRNA � Ct U6 snRNA. PCR was performedin triplicate.

Exogenous overexpression of the mature miR-181s

through transient transfection of the mature

miR-181s mimic

The mature miR-181s (miR-181a, miR-181b, miR-181c andmiR-181d) mimic and control miRNA mimic were chemi-cally synthesized by Shanghai GenePharma Company(Shanghai, China). The sequence of the mature miR-181smimic and control miRNA mimic is shown in Table 1.SGC7901/VCR and A549/CDDP cells were plated in 6-wellplates (6 � 105 cells/well) and transfected with 100 nM ofthe mature miR-181s mimic or 100 nM control miRNAmimic using Lipofectamine 2000 (Invitrogen, Long Island,NY) according to the manufacturer’s protocol.

In vitro drug sensitivity assay

All 4 kinds of cells (SGC7901, SGC7901/VCR, A549 andA549/CDDP) were seeded into 96-well plates (5 � 103 cells/well) and allowed to attach overnight. In addition, as toSGC7901/VCR and A549/CDDP cells, 24 hr after transfec-tion of the mature miR-181s mimic or control miRNAmimic, cells were also seeded into 96-well plates for next stepexperimentation. After cellular adhesion, freshly preparedanticancer drugs including VCR, adriamycin (ADR), 5-fluo-rouracil (5-FU), CDDP, mitomycin C (MMC) and etoposide(VP-16) were added with the final concentration being 0.01,0.1, 1 and 10 times of the human peak plasma concentrationfor each drug as previously described.14 The peak serum con-centrations of various anticancer drugs are 0.4 lg/ml forADR, 10lg/ml for 5-FU, 2.0 lg/ml for CDDP, 1.0 lg/ml forMMC, 0.5 lg/ml for VCR and 10 lg/ml for VP-16.20,21

Forty-eight hours after addition of drugs, cell viability wasassessed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazo-lium bromide (MTT) assay. The absorbance at 490 nm(A490) of each well was read on a spectrophotometer. Theconcentration at which each drug produced 50% inhibition

Can

cerCellBiology

Int. J. Cancer: 127, 2520–2529 (2010) VC 2010 UICC

Zhu et al. 2521

of growth (IC50) was estimated by the relative survivalcurve. Three independent experiments were performed inquadruplicate.

Dual luciferase activity assay

The 30UTR of human BCL2 cDNA containing the putativetarget site for the mature miR-181s (sequence shown in Sup-porting Information 1) was chemically synthesized andinserted at the XbaI site, immediately downstream of the lu-ciferase gene in the pGL3-control vector (Promega, Madison,WI) by Biomics Biotechnologies (Nantong, China). Twenty-four hours before transfection, cells were plated at 1.5 � 105cells/well in 24-well plates. Two hundred nanograms ofpGL3-BCL2-30UTR plus 80 ng pRL-TK (Promega) weretransfected in combination with 60 pmol of the mature miR-181s mimic or the control miRNA mimic using Lipofect-amine 2000 (Invitrogen) according to the manufacturer’s pro-tocol as described.14 Luciferase activity was measured 24 hrafter transfection using the Dual Luciferase Reporter AssaySystem (Promega). Firefly luciferase activity was normalizedto renilla luciferase activity for each transfected well. Threeindependent experiments were performed in triplicate.

Western blot analysis

Cells were harvested and homogenized with lysis buffer 24,48 and 72 hr after the transfection. Total protein was sepa-rated by denaturing 15% SDS-polyacrylamide gel electropho-resis. Western analysis was performed as described.22 Theprimary antibodies for BCL2, b-actin and a-Tubulin werepurchased from Bioworld Technology and Santa Cruz Bio-technology, respectively. Protein levels were normalized tob-actin or a-Tubulin. Fold changes were determined.

Apoptosis assay

Twenty-four hours after the transfection as described above,SGC7901/VCR and A549/CDDP cells were treated by VCRand CDDP, with final concentration of 5 and 20 lg/ml,respectively. Forty-eight hours after the treatment of VCRand CDDP, flow cytometry was used to detect apoptosis of

the transfected SGC7901/VCR and A549/CDDP cells bydetermining the relative amount of AnnexinV-FITC-positive,PI-negative cells as previously described,23 respectively.

Statistical analysis

Each experiment was repeated at least 3 times. Numericaldata were presented as mean 6 SD. The difference betweenmeans was analyzed with Student’s t-test. All statistical analy-ses were performed using SPSS11.0 software (Chicago, IL).Differences were considered significant when p < 0.01.

ResultsmiR-181b is downregulated in both SGC7901/VCR

and A549/CDDP cells, compared with SGC7901

and A549 cells, respectively

miRNA microarray analysis of SGC7901 and SGC7901/VCRcells showed that there were 24 miRNAs downregulatedmore than 2-fold and 11 miRNAs upregulated more than2-fold in SGC7901/VCR cells, compared with SGC7901 cells(Table 2). Among the downregulated miRNAs were themature miR-181s. Quantitative RT-PCR for miR-181b furtherverified that miR-181b was downregulated 4.03-fold inSGC7901/VCR and 2.88-fold in A549/CDDP cells, comparedwith SGC7901 and A549 cells, respectively (Fig. 1).

miR-181b modulates MDR of SGC7901/VCR

and A549/CDDP cell lines

In vitro drug sensitivity assay verified the MDR phenotype ofSGC7901/VCR and A549/CDDP cell lines, compared withthe parental SGC7901 and A549 cell lines, respectively (Sup-porting Information 2).

To investigate whether miR-181b has a direct function inMDR development or is just differentially modulated inMDR cancer cells, we transfected SGC7901/VCR and A549/CDDP cells with the mature miR-181s mimic and the controlmiRNA mimic, respectively, to observe the effects on MDRphenotype thereafter. In SGC7901/VCR cells, MTT assayrevealed that only those transfected with miR-181b mimicexhibited greatly enhanced sensitivity to VCR, 5-Fu, CDDP,

Table 1. The sequence of the control miRNA mimic and the mature miR-181s mimic

miRNA mimics Sequence

has-miR control mimic 50 to 30 Sense UUCUCCGAACGUGUCACGUTT

Antisense ACGUGACACGUUCGGAGAATT

has-miR-181a mimic 50 to 30 Sense AACAUUCAACGCUGUCGGUGAGU

Antisense UCACCGACAGCGUUGAAUGUUUU

has-miR-181b mimic 50 to 30 Sense AACAUUCAUUGCUGUCGGUGGGU

Antisense CCACCGACAGCAAUGAAUGUUUU

has-miR-181c mimic 50 to 30 Sense AACAUUCAACCUGUCGGUGAGU

Antisense UCACCGACAGGUUGAAUGUUUU

has-miR-181d mimic 50 to 30 Sense AACAUUCAUUGUUGUCGGUGGGU

Antisense CCACCGACAACAAUGAAUGUUUU

Can

cerCellBiology

2522 Role of microRNAs in multidrug resistance

Int. J. Cancer: 127, 2520–2529 (2010) VC 2010 UICC

VP-16 and ADR, but not to MMC, compared with thosetransfected with the control miRNA mimic, as indicated bysignificantly decreased IC50 values (Fig. 2a).

In A549/CDDP cells, MTT assay revealed that all A549/CDDP cells transfected with the mature miR-181s mimicexhibited greatly enhanced sensitivity to VCR, 5-Fu, CDDP,VP-16 and ADR, but not to MMC, compared with thosetransfected with the control miRNA mimic, respectively(Fig. 2b).

BCL2 is a target gene of the mature miR-181s

TargetScanHuman 5.1 (http://www.targetscan.org) was usedfor the prediction of the mature miR-181s target genes. Thesequence alignment of the mature miR-181s with BCL230UTR of different species is conserved, which indicates thatBCL2 is one of the potential targets of the mature miR-181s.More interestingly, there are 2 conserved target sites of themature miR-181s in the 30UTR of BCL2 by TargetScan pre-diction (Supporting Information 3). The study by Neilson JRet al. shows that the position 1744–1750 of BCL2 30UTR is a

target site by mmu-miR-181a, which was verified by lucifer-ase activity assay.24

To explore whether the other predicted position 2888–2894 of BCL2 30UTR is also a target site by the mature miR-181s, we constructed a luciferase reporter vector with the pu-tative BCL2 30UTR target site for the mature miR-181sdownstream of the luciferase gene (pGL3-BCL2-30UTR). Lu-ciferase reporter vector together with the mature miR-181smimic or the control miRNA mimic were transfected intoSGC7901/VCR and A549/CDDP cells, respectively.

In SGC7901/VCR and A549/CDDP cells, a significantdecrease in relative luciferase activity was noted when pGL3-BCL2-30UTR was cotransfected with the mature miR-181smimic but not with the control miRNA mimic, respectively(Figs. 3a and 3b), which suggests that the position 2888–2894of BCL2 30UTR is also a target site by the mature miR-181s.

miR-181b modulates MDR by repressing the BCL2

protein expression

Worth of note, in our study, the decreased expression ofmiR-181b in SGC7901/VCR and A549/CDDP cells was con-current with the overexpression of BCL2 protein, compared

Table 2. miRNAs differentially expressed in SGC7901/VCR cell lineand SGC7901 cell line

miRNA decreasedFoldchange miRNA increased

Foldchange

hsa-let-7d �3.97 hsa-miR-129-5p 2.07

hsa-let-7f �7.58 hsa-miR-182-3p 2.01

hsa-let-7g �9.28 hsa-miR-494 2.77

hsa-miR-101 �9.22 hsa-miR-503 2.65

hsa-miR-105 �4.59 hsa-miR-565 2.55

hsa-miR-10a �12.77 hsa-miR-602 2.14

hsa-miR-125b �4.40 hsa-miR-663 3.39

hsa-miR-181a �14.31 hsa-miR-668 2.30

hsa-miR-181b �2.16 hsa-miR-801 3.09

hsa-miR-181c �6.70 hsa-miR-886-3p 3.21

hsa-miR-181d �26.56 hsa-miR-886-5p 3.62

hsa-miR-15a-3p �5.41

hsa-miR-15b �34.04

hsa-miR-16 �7.99

hsa-miR-216a �22.43

hsa-miR-220 �32.49

hsa-miR-33a �35.58

hsa-miR-451 �9.80

hsa-miR-497 �7.86

hsa-miR-507 �16.05

hsa-miR-543 �4.90

hsa-miR-612 �43.06

hsa-miR-758 �6.96

hsa-miR-96 �8.28

Figure 1. Real-time quantification of miR-181b by stem-loop RT-PCR

showed that miR-181b was downregulated in both SGC7901/VCR

and A549/CDDP cell lines, compared with SGC7901 and A549 cell

lines, respectively. Triplicate assays were performed for each RNA

sample, and the relative amount of miR-181b was normalized to

U6 snRNA. Data were shown as fold changes of miR-181b levels in

SGC7901/VCR and A549/CDDP cell lines relative to SGC7901 and

A549 cell lines, respectively, which were set as 1 (mean 6 SD).

*p < 0.01.

Can

cerCellBiology

Int. J. Cancer: 127, 2520–2529 (2010) VC 2010 UICC

Zhu et al. 2523

with the parental SGC7901 and A549 cells, respectively(Fig. 4). As BCL2 is an antiapoptotic protein and a target ofthe mature miR-181s, we hypothesized that the miR-181bmight modulate MDR of cancer cells by repressing the BCL2protein expression.

To ascertain our hypothesis, we transfected the maturemiR-181s mimic and the control miRNA mimic intoSGC7901/VCR and A549/CDDP cells to detect the BCL2expression level changes, respectively.

No significant changes in BCL2 expression level werefound neither in SGC7901/VCR nor in A549/CDDP cells 24hr after the transfection of the mature miR-181s mimic, com-pared with the transfection of the control miRNA mimic,

respectively. However, 48 and 72 hr after the transfection,significant changes were found.

In SGC7901/VCR cells, 48 hr after the transfection, West-ern Blot demonstrated no significantly decreased BCL2 pro-tein level in all the mature miR-181s mimic transfected cells,compared with the control miRNA mimic transfected cells(Supporting Information 4a), whereas 72 hr after the trans-fection, Western Blot demonstrated significantly decreasedBCL2 protein level only in miR-181b mimic transfected cellscompared with the control miRNA mimic transfected cells(Fig. 5a). In A549/CDDP cells, 48 hr after the transfection,Western Blot demonstrated significantly decreased BCL2protein level in miR-181b mimic transfected cells, compared

Figure 2. miR-181b modulates multidrug resistance of SGC7901/VCR and A549/CDDP cell lines. (a) In SGC7901/VCR cells, only those

transfected with miR-181b mimic exhibited greatly enhanced sensitivity to VCR, 5-Fu, CDDP, VP-16 and ADR, but not to MMC, compared

with those transfected with the control miRNA mimic. (b) In A549/CDDP cells, all those transfected with the mature miR-181s mimic

exhibited greatly enhanced sensitivity to VCR, 5-Fu, CDDP, VP-16 and ADR, but not to MMC, compared with those transfected with the

control miRNA mimic, respectively. The data shown represent the mean 6 SD from 3 independent experiments. *p < 0.01.

Can

cerCellBiology

Int. J. Cancer: 127, 2520–2529 (2010) VC 2010 UICC

2524 Role of microRNAs in multidrug resistance

with the control miRNA mimic transfected cells (SupportingInformation 4b), whereas 72 hr after the transfection, West-ern Blot demonstrated significantly decreased BCL2 proteinlevel in all the mature miR-181s mimic transfected cells,compared with the control miRNA mimic transfected cells(Fig. 5b).

These results demonstrated that miR-181b may modulateMDR of cancer cell lines, at least in part, by repressing theBCL2 protein expression.

miR-181b sensitizes SGC7901/VCR and A549/CDDP cells

to VCR- and CDDP-induced apoptosis, respectively

The development of drug resistance in various cancer cellshas been linked to a reduced susceptibility to drug-inducedapoptosis, which was shown to be a consequence, at least insome cases, of overexpression of antiapoptotic proteins, suchas BCL2 and BCL-XL.25–27

As miR-181b may modulate MDR of cancer cell lines byrepressing the BCL2 protein expression, considering the well-characterized role of BCL2 in apoptosis and drug resistance,we suggest a hypothesis that miR-181b may play a role inthe development of MDR, at least in part, by modulation ofapoptosis via targeting BCL2.

To confirm this hypothesis, we evaluated VCR- andCDDP-induced apoptosis after transfection SGC7901/VCRand A549/CDDP cells with the mature miR-181s mimic andthe control miRNA mimic, respectively.

In SGC7901/VCR cells, a marked increase in apoptosis, asassessed by flow cytometry, was observed in only miR-181bmimic transfected cells after VCR treatment, compared withthe control miRNA mimic transfected cells (Figs. 6a and 6c),whereas in A549/CDDP cells, a significant increase in apo-ptosis was found in all the miR-181s mimic transfected cellsafter CDDP treatment, compared with the control miRNAmimic transfected cells (Figs. 6b and 6c).

Taken together, our findings suggest that miR-181b couldplay a role in the development of MDR in both gastric andlung cancer cell lines, at least in part, by modulation of apo-ptosis via targeting BCL2.

DiscussionOne major mechanism of drug resistance in cancer cells isthe defective apoptosis pathway.2–4 Recently, more and morefindings have established that miRNAs modulate drugresistance of cancer cells, at least in part, through thismechanism.11–15 In our study, we found that the antiapop-totic protein BCL2 is upregulated, whereas miR-181b isdownregulated in both SGC7901/VCR and A549/CDDP cells,compared with SGC7901 and A549 cells, respectively. Themechanistic connection of miR-181b dysregulation with theestablishment of MDR in SGC7901/VCR and A549/CDDPcells was evidenced by the correlation between exogenousoverexpression of miR-181b and corresponding changes inthe protein levels of its target BCL2, which has a documented

Figure 3. Dual luciferase assay performed in SGC7901/VCR and A549/CDDP cells suggests that the position 2888–2894 of BCL2 30UTR is

also a target site by the mature miR-181s. Luciferase reporter vector together with the mature miR-181s mimic or the control miRNA mimic

were cotransfected into SGC7901/VCR and A549/CDDP cells, respectively. (a) In SGC7901/VCR cells, a significant decrease in relative

luciferase activity was noted when pGL3-BCL2-30UTR was cotransfected with the mature miR-181s mimic but not with the control miRNA

mimic. (b) In A549/CDDP cells, the same effect was found. The results shown represent the mean 6 SD from 3 independent experiments.

*p < 0.01.

Can

cerCellBiology

Int. J. Cancer: 127, 2520–2529 (2010) VC 2010 UICC

Zhu et al. 2525

importance in the development of cancer cells drugresistance.

Function research of miR-181s was first focused on hema-topoietic lineage differentiation in mouse, and mmu-miR-181s was reported to show an obviously high expression atthe adult stage, compared with embryonic and early postnatalstages in hematopoietic lineage differentiation.28 Recent stud-ies by Fanini et al. and Shi et al. showed that miR-181a andmiR-181b may serve as tumor suppressors in human acutemonocytic leukemia (AML) and human glioma cells, respec-tively. Agents that increase miR-181a expression inducedapoptosis of AML blasts, while exogenous overexpression ofmiR-181a and miR-181b also induced apoptosis of humanglioma cells29,30; however, study by Ji et al. showed that hsa-miR-181s was highly expressed in EpCAM-positive hepaticcancer stem cells, which was more aggressive, compared withalpha-fetoprotein-positive hepatic cancer cells, and inhibitionof miR-181 led to a reduction in EpCAM-positive hepaticcancer stem cell quantity and tumor-initiating ability; further-more, hsa-miR-181s was also highly expressed in embryoniclivers and in isolated hepatic stem cells.31 The above 3 studiessuggest that miR-181s may play a totally different role in dif-ferent types of cancer cells. Our study was in concordancewith the studies by Fanini et al. and Shi et al.,29,30 and we

also found that exogenous overexpression of miR-181b sensi-tizes SGC7901/VCR and A549/CDDP cells to VCR- andCDDP-induced apoptosis, respectively, and except for VCRand CDDP, exogenous overexpression of miR-181b alsoaltered sensitivity of both SGC7901/VCR and A549/CDDPcells to ADR, VP-16, 5-Fu, but not to MMC, a possibleexplanation for this phenomenon could be that MMC trig-gers apoptosis of cancer cells via a pathway where BCL2 isnot necessarily involved.32–34

Although a common function of miR-181b in modulatingMDR of cancer cell lines was found in our study, the func-tion of other miR-181s seems to be inconsistent between gas-tric and lung cancer cell lines. In SGC7901/VCR cells, exoge-nous overexpression of other mature miR-181s (miR-181a,miR-181c and miR-181d) has no significant effect on MDR,whereas in A549/CDDP cells, exogenous overexpression of

Figure 5. The mature miR-181s mimic modulates BCL2 protein

level in SGC7901/VCR and A549/CDDP cells, respectively. (a) In

SGC7901/VCR cells, quantified Western analysis of BCL2

expression level showed that 72 hr after the transfection,

significantly decreased BCL2 protein level in miR-181b mimic

transfected cells was found, compared with the control miRNA

mimic transfected cells. (b) In A549/CDDP cells, 72 hr after the

transfection, significantly decreased BCL2 protein level was found

in all the mature miR-181s mimic transfected cells, compared with

the control miRNA mimic transfected cells. Representative Western

blots were attached beside the graphs. The results shown

represent the mean 6 SD from 3 independent experiments.

*p < 0.01.

Figure 4. Quantified Western analysis of BCL2 expression level

showed significant overexpression of BCL2 protein in SGC7901/

VCR and A549/CDDP cell lines, compared with the parental

SGC7901 and A549 cell lines, respectively. Representative Western

blots were attached below the graphs. The results shown represent

the mean 6 SD from 3 independent experiments. *p < 0.01.

Can

cerCellBiology

Int. J. Cancer: 127, 2520–2529 (2010) VC 2010 UICC

2526 Role of microRNAs in multidrug resistance

other miR-181s (miR-181a, miR-181c and miR-181d) has thesame significant effect on MDR as miR-181b. It is wellknown that the 4 miR-181 family members are evolutionarilyconserved among the vertebrate lineage with high homology,implicating their functional redundancy,31 just like what wefound in lung cancer cell line; however, in gastric cancer cellline, inconsistence was found; one possible explanation forthis is that the function of miRNAs may exist cell-type speci-ficity just as mentioned above. In gastric cancer cell line, theother miR-181 family members may play a weaker functionthan miR-181b. Although BCL2 is a target gene by themature miR-181s, the silencing effect between each member

of the mature miR-181s is different. For instance, in thestudy by Shi et al.,30 the apoptosis induction effect of hsa-miR-181b in glioma cells was more apparent than the effectof hsa-miR-181a. The molecular mechanism underlying thisdifferent silencing effect between each member of the maturemiR-181s may be explained by the different length of the‘‘seed region’’ of the target gene. One of the ‘‘seed region’’ ofBCL2 30UTR for the mature miR-181s verified by our experi-ment showed that the ‘‘seed region’’ for miR-181b is 5 baseslonger than that for the other members of the mature miR-181s. However, as to the intrinsic mechanism, more researchis needed.

Figure 6. miR-181b sensitizes SGC7901/VCR and A549/CDDP cells to VCR- and CDDP-induced apoptosis, respectively. (a) In SGC7901/VCR

cells, apoptosis evaluated by flow cytometry showed that a marked increase in apoptosis only in miR-181b mimic transfected cells after

VCR treatment. (b) In A549/CDDP cells, apoptosis evaluated by flow cytometry showed that a marked increase in apoptosis in all the

mature miR-181s mimic transfected cells after CDDP treatment. The results shown represent the mean 6 SD from 3 independent

experiments. *p < 0.01. (c) Representative flow cytometry report was attached under the graphs. [Color figure can be viewed in the online

issue, which is available at wileyonlinelibrary.com.]

Can

cerCellBiology

Int. J. Cancer: 127, 2520–2529 (2010) VC 2010 UICC

Zhu et al. 2527

In our study, we overexpressed the mature miR-181sthrough transient transfection of the mature miR-181s mimic.It is reported that transfection of small RNAs (such as smallinterfering RNAs and miRNAs) into cells exhibited concentra-tion and temporal dependence.35 Therefore, we checked theBCL2 expression levels 24, 48 and 72 hr after the transfection;indeed, the most significant effect of small RNAs happened 72hr after the transfection, which was consistent with the studyby Xia L et al.14 However, the concentration dependence oftransfection of small RNAs was not done in our study, whichwas a limitation of our experiment. In addition, among the dif-ferentially expressed miRNAs in SGC7901/VCR cells comparedwith SGC7901 cells are some with well-characterized drug re-sistance association such as miR-15b, miR-16 and miR-451,12–14 which indicated that MDR of cancer cell lines may be amiRNA net regulated biological process.

In summary, the findings we reported here presented thefirst evidence that miR-181b may be involved in the devel-opment of MDR in gastric and lung cancer cell lines. miR-181b could modulate the sensitivity of gastric and lung can-cer cell lines to some anticancer drugs, at least in part,through target BCL2 expression. Our study may have impli-cations for cancer chemotherapy whose efficiency is oftenimpeded by the development of MDR. Therapeutic strat-egies targeting the MDR-related miRNAs, such as miR-181b, may be another promising way to enhance therapeu-tic effect. However, it should be noted that our data arederived from cell lines, which have been removed fromtheir in vivo context and cannot be considered accurate sur-rogates for clinical tumors. Thus, future studies to assessthe roles of miR-181b in vivo and in clinical context arewarranted.

References

1. Fan D, Zhang X, Chen X, Mou Z, Hu J,Zhou S, Ding J, Wu K. Bird’s-eye view ongastric cancer research of the past 25 years.J Gastroenterol Hepatol 2005;20:360–5.

2. Rabik CA, Dolan ME. Molecularmechanisms of resistance and toxicityassociated with platinating agents. CancerTreat Rev 2007;33:9–23.

3. Johnstone RW, Ruefli AA, Lowe SW.Apoptosis: a link between cancergenetics and chemotherapy. Cell 2002;108:153–64.

4. Zhang K, Mack P, Wong KP. Glutathione-related mechanisms in cellular resistance toanticancer drugs. Int J Oncol 1998;12:871–82.

5. Szakacs G, Paterson JK, Booth-Genthe C,Gottesman MM. Targeting multidrugresistance in cancer. Nat Rev Drug Disc2006;5:219–34.

6. Osada H, Takahashi T. MicroRNAs inbiological processes and carcinogenesis.Carcinogenesis 2007;28:2–12.

7. Harfe BD. MicroRNAs in vertebratedevelopment. Curr Opin Genet 2005;15:410–15.

8. Hwang HW, Mendell JT. MicroRNAs incell proliferation, cell death, andtumorigenesis. Br J Cancer 2006;94:776–80.

9. Chen CZ, Li L, Lodish HF, Bartel DP.MicroRNAs modulate hematopoieticlineage differentiation. Science 2004;303:83–6.

10. Si ML, Zhu S, Wu H, Lu Z, Wu F, Mo YY.miR-21-mediated tumor growth. Oncogene2007;26:2799–803.

11. Blower PE, Verducci JS, Lin S, Zhou J,Chung JH, Dai Z, Liu CG, Reinhold W,Lorenzi PL, Kaldjian EP, Croce CM,Weinstein JN, et al. MicroRNA expressionprofiles for the NCI-60 cancer cell panel.Mol Cancer Ther 2007;6:1483–91.

12. Sorrentino A, Liu CG, Addario A, PeschleC, Scambia G, Ferlini C. Role ofmicroRNAs in drug-resistant ovariancancer cells. Gynecol Oncol 2008;111:478–86.

13. Kovalchuk O, Filkowski J, Meservy J,Ilnytskyy Y, Tryndyak VP, Chekhun VF,Pogribny IP. Involvement of microRNA-451 in resistance of the MCF-7 breastcancer cells to chemotherapeutic drugdoxorubicin. Mol Cancer Ther 2008;7:2152–9.

14. Xia L, Zhang D, Du R, Pan Y, Zhao L, SunS, Hong L, Liu J, Fan D. miR-15b andmiR-16 modulate multidrug resistance bytargeting BCL2 in human gastriccancer cells. Int J Cancer 2008;123:372–9.

15. Tsang WP, Kwok TT. Let-7a microRNAsuppresses therapeutics-induced cancer celldeath by targeting caspase-3. Apoptosis2008;13:1215–22.

16. Zheng T, Wang J, Chen X, Liu L. Role ofmicroRNA in anticancer drug resistance.Int J Cancer 2010;126:2–10.

17. Castoldi M, Schmidt S, Benes V, NoerholmM, Kulozik AE, Hentze MW,Muckenthaler MU. A sensitive array formicroRNA expression profiling (miChip)based on locked nucleic acids (LNA). RNA2006;12:913–20.

18. Chen C, Ridzon DA, Broomer AJ, Zhou Z,Lee DH, Nguyen JT, Barbisin M, Xu NL,Mahuvakar VR, Andersen MR, Lao KQ,Livak KJ, et al. Real-time quantification ofmicroRNAs by stem-loop RT-PCR. NucleicAcids Res 2005;33:e179.

19. Livak KJ, Schmittgen TD. Analysis ofrelative gene expression data using real-time quantitative PCR and the 2(-DeltaDelta C(T)) method. Methods 2001;25:402–8.

20. Yamaue H, Tanimura H, Noguchi K, HottaT, Tani M, Tsunoda T, Iwahashi M, TamaiM, Iwakura S. Chemosensitivity testing offresh human gastric cancer with highlypurified tumor cells using the MTT assay.Br J Cancer 1992;66:794–9.

21. Yamaue H, Tani M, Onishi H, KinoshitaH, Nakamori M, Yokoyama S, Iwahashi M,Uchiyama K. Locoregional chemotherapyfor patients with pancreatic cancer intra-arterial adjuvant chemotherapy afterpancreatectomy with portal vein resection.Pancreas 2002;25:366–72.

22. Zeng Y, Zhang X, Huang Z, Cheng L, YaoS, Qin D, Chen X, Tang Q, Lv Z, Zhang L,Lu C. Intracellular Tat of humanimmunodeficiency virus type 1 activateslytic cycle replication of Kaposi’s sarcoma-associated herpesvirus: role of JAK/STATsignaling. J Virol 2007;81:2401–17.

23. Malugin A, Kopeckova P, Kopecek J.HPMA copolymer-bound doxorubicininduces apoptosis in human ovariancarcinoma cells by a Fas-independentpathway. Mol Pharm 2004;1:174–82.

24. Neilson JR, Zheng GX, Burge CB, SharpPA. Dynamic regulation of miRNAexpression in ordered stages of cellulardevelopment. Genes Dev 2007;21:578–89.

25. Reed JC. Regulation of apoptosis by bcl-2family proteins and its role in cancer andchemoresistance. Curr Opin Oncol 1995;7:541–6.

26. Wang S, Yang D, Lippman ME. TargetingBcl-2 and Bcl-XL with nonpeptidic small-molecule antagonists. Semin Oncol 2003;30:133–42.

27. Reed JC. Drug insight: cancer therapystrategies based on restoration ofendogenous cell death mechanisms. NatClin Pract Oncol 2006;3:388–98.

Can

cerCellBiology

Int. J. Cancer: 127, 2520–2529 (2010) VC 2010 UICC

2528 Role of microRNAs in multidrug resistance

28. Chen CZ, Li L, Lodish HF, Bartel DP.MicroRNAs modulate hematopoieticlineage differentiation. Science 2004;303:83–6.

29. Fanini F, Vannini I, Faabbri M.MicroRNAs: tiny players with a big role inthe pathogenesis of leukemias andlymphomas. Hematol Rev 2009;1:40–5.

30. Shi L, Cheng Z, Zhang J, Li R, Zhao P, FuZ, You Y. hsa-mir-181a and hsa-mir-181bfunction as tumor suppressors in humanglioma cells. Brain Res 2008;1236:185–93.

31. Ji J, Yamashita T, Budhu A, Forgues M, JiaHL, Li C, Deng C, Wauthier E, Reid LM,Ye QH, Qin LX, Yang W, et al.

Identification of microRNA-181 bygenome-wide screening as a critical playerin EpCAM-positive hepatic cancerstem cells. Hepatology 2009;50:472–80.

32. Park IC, Park MJ, Hwang CS, Rhee CH,Whang DY, Jang JJ, Choe TB, Hong SI,Lee SH. Mitomycin C induces apoptosis ina caspasesdependent and Fas/CD95-independent manner in human gastricadenocarcinoma cells. Cancer Lett 2000;158:125–32.

33. Kang YH, Lee KA, Ryu CJ, Lee HG, LimJS, Park SN, Paik SG, Yoon DY.Mitomycin C induces apoptosis via Fas/

FasL dependent pathway and suppressionof IL-18 in cervical carcinoma cells. CancerLett 2006;237:33–44.

34. Liedtke C, Lambertz D, Schnepel N,Trautwein C. Molecular mechanismof Mitomycin C-dependentcaspase-8 regulation: implications forapoptosis and synergism with interferon-alpha signalling. Apoptosis 2007;12:2259–70.

35. Khan AA, Betel D, Miller ML, Sander C,Leslie CS, Marks DS. Transfection of smallRNAs globally perturbs gene regulation byendogenous microRNAs. Nat Biotechnol2009;27:549–55.

Can

cerCellBiology

Int. J. Cancer: 127, 2520–2529 (2010) VC 2010 UICC

Zhu et al. 2529