Tumor and Stem Cell Biology

PML/RARa-RegulatedmiR-181a/b Cluster Targetsthe Tumor Suppressor RASSF1A in AcutePromyelocytic LeukemiaDaniela Br€auer-Hartmann1, Jens-UweHartmann1, Alexander ArthurWurm1, Dennis Gerloff1,Christiane Katzerke1, Maria Vittoria Verga Falzacappa2, Pier Giuseppe Pelicci2,Carsten M€uller-Tidow3, Daniel G. Tenen4,5, Dietger Niederwieser1, and Gerhard Behre1

Abstract

In acute promyelocytic leukemia (APL), all-trans retinoic acid(ATRA) treatment induces granulocytic maturation and com-plete remission of leukemia. microRNAs are known to becritical players in the formation of the leukemic phenotype.In this study, we report downregulation of the miR-181a/b genecluster in APL blasts and NB4 leukemia cells upon ATRAtreatment as a key event in the drug response. We found thatmiR-181a/b expression was activated by the PML/RARa onco-gene in cells and transgenic knock-in mice, an observationconfirmed and extended by evidence of enhanced expressionof miR-181a/b in APL patient specimens. RNA interference(RNAi)-mediated attenuation of miR-181a/b expression inNB4 cells was sufficient to reduce colony-forming capacity,

proliferation, and survival. Mechanistic investigations revealedthat miR-181a/b targets the ATRA-regulated tumor suppressorgene RASSF1A by direct binding to its 30-untranslated region.Enforced expression of miR-181a/b or RNAi-mediated attenu-ation of RASSF1A inhibited ATRA-induced granulocytic differ-entiation via regulation of the cell-cycle regulator cyclin D1.Conversely, RASSF1A overexpression enhanced apoptosis.Finally, RASSF1A levels were reduced in PML/RARa knock-inmice and APL patient samples. Taken together, our resultsdefine miR-181a and miR-181b as oncomiRs in PML/RARa-associated APL, and they reveal RASSF1A as a pivotal element inthe granulocytic differentiation program induced by ATRA inAPL. Cancer Res; 75(16); 3411–24. �2015 AACR.

IntroductionAcute promyelocytic leukemia (APL) is characterizedby specific

chromosomal translocations involving the retinoic acid receptora (RARa; refs. 1, 2). The most frequent translocation fuses theRARa with the promyelocytic leukemia protein (PML) gene (3).At physiological levels of retinoids, the PML/RARa fusion proteincauses block of differentiation and neoplastic transformation bydisrupting the function of PML and repressing transcription ofgenes regulated by RARa (2, 4, 5). Pharmalogic doses of retinoidscan overcome this block, lead to the expression of granulocyticspecific transcription factors like C/EBPb (6), and thereby induceterminal differentiation of APL blasts in vitro and in vivo (1, 2).

Recent studies identified a group of small molecules that areinvolved in posttranscriptional regulation of gene expression.microRNAs (miRNA) are endogenous, non–protein-coding

small RNAs that play critical roles in the posttranscriptionalregulationof target genesbydirect targetingofmRNAs for cleavage,translational repression, or destabilization (7). A selectednumber of miRNAs has been shown to play key roles in hemato-poietic differentiation (8) as well as in the formation and main-tenance of leukemia (9). We and others already showed that miR-223, miR-34a, and miR-30c are important factors in myeloiddifferentiation (10–13). While some miRNAs like miR-223 havebeen implied in APL differentiation (14) and tumorigenesis, thereis still a lack of knowledge about the expression and function ofother miRNAs.

In this study, we showed that the genomic clustered miR-181aand miR-181b (miR-181a/b) are highly expressed in APL anddownregulated during all-trans retinoic acid (ATRA)-induceddifferentiation (14–16). By analyzing APL and acute myeloidleukemia (AML) patient samples as well as PML/RARa knock-inmice, we demonstrated that miR-181a and miR-181b display avery specific PML/RARa dependency in vivo. Furthermore, werevealed that miR-181a and miR-181b are involved in the for-mation of the PML/RARa caused oncogenic phenotype. Weshowed that the miR-181a/b function is determined by the directbinding to their target RASSF1A. Finally, we firstly describe thetumor suppressor RASSF1A as a new and essential member of theretinoic acid–induced differentiation network in APL.

Materials and MethodsHuman cell samples from AML patients and healthy donors

AML patient samples were obtained as RNA and as frozen bonemarrow samples from the University of M€unster (M€unster,

1Division of Hematology and Oncology, Leipzig University Hospital,Leipzig, Germany. 2Department of Experimental Oncology, EuropeanInstitute ofOncology,Milan, Italy. 3Department of InternalMedicine IV,Hematology/Oncology, University Hospital Halle, Halle, Germany.4Harvard Stem Cell Institute, Harvard Medical School, Boston, Massa-chusetts. 5Cancer Science Institute, National University of Singapore,Singapore.

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

Corresponding Author: Gerhard Behre, Division of Hematology and Oncology,UniversityHospitalLeipzig,Johannisallee32A,04103Leipzig,Germany.Phone:49-341-9713846; Fax: 49-341-9713059; E-mail: gerhard.behre@medizin.uni-leipzig.de

doi: 10.1158/0008-5472.CAN-14-3521

�2015 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 3411

on July 18, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 3, 2015; DOI: 10.1158/0008-5472.CAN-14-3521

Germany) and from the OSHO patient sample collection (Leip-zig, Germany). Blood cells samples from patients without anyhematopoietic disease were obtained from the University Hos-pital Halle (Halle, Germany). Ethics committee approval wasobtained and all patients provided informed consent. All sampleswere karyotyped and molecular genetic analysis was performedpreviously.

Transgenic mouse modelBonemarrow cells from 5 PML/RARa knock-inmice (C57BI/6-

mCGþ/PR), in which PML/RARa is expressed under the control ofthe murine cathepsin G gene (C57BI/6-mGCþ/PR), and 5 wild-type mice (C57BI/6-WT) were obtained from Pier GiuseppePelicci (IFOM-EIO, Milan, Italy; ref. 17).

Cells, reagents, and transfectionsNB4, HL60 and U937 cells were cultured under standard con-

ditions. For differentiation, cells were induced with 10�6 mol/LATRA (Sigma-Aldrich) and as controlDMSO. For differentiation ofmiRNA mimic transfected and RASSF1A shRNA-expressing NB4cells, 10�7 mol/L ATRAwere used. Arsenic trioxide (As2O3; Sigma-Aldrich) were used at 1 mmol/L. As control 1% HCl solution wasused. Doxorubicin (Selleckchem), solved inDMSO,was applied at0.01 mg/mL and cytarabine at 0.05 mmol/L. Transfection ofmiRNAmimics (Dharmacon) and pcDNA3.1 vectors was done by usingAmaxa Technology (Lonza) according to the manufacturer'sinstructions. The transfection efficiency was 30% to 40%. U937-PR9 cells, which carry the PML/RARa cDNA under the control ofmetallothionine promoter, and the control cells U937-PC-18,which carry the empty vector, were cultured, and the expressionof the PML/RARa fusion protein was induced as previouslydescribed (18).

DNA constructs, cloning, and mutagenesisFor luciferase assay, RASSF1A 30untranslated region (UTR) was

inserted in the pGL3-luciferase reporter control vector down-stream of the luciferase encoding region (Promega). MiR-181–binding site positions in the 30UTR were taken from the miRNAtarget data base targetscan (www.targetscan.org). The RASSF1A30UTR (Acc. No.[NM_007182.4]) was amplified from cDNA ofNB4 cells treated with 1 mmol/L ATRA for 48 hours using primerpairs, which generate a XbaI restriction enzyme recognition site atthe 30- and 50-ends of the amplified DNA product. The followingprimer pairs were used: RASSF1A 30UTR: forward, 50-GTCTA-GACCTCTTGTACCCCCAGGTGG-30; reverse, 50-GTCTAGAGAG-GATCTTGAAATCTTTATTGAG-30. The purified DNA fragmentand the pGL3 luciferase reporter control vector were digestedwith XbaI and fused in a T4 ligase reaction (Invitrogen). Muta-genesis of miR-181–binding sites was done with the QuikChangeSite-DirectedMutagenesis Kit (Agilent Technologies) according tothe manufacturer's instructions. All sequences were verified bysequencing. pcDNA3.1/RASSF1A expression vectorwas a kind giftfrom Reinhard Dammann (Justus-Liebig University, Giessen,Germany).

Luciferase reporter assayTo prove the direct binding of miR-181a/b to the 30UTR of

RASSF1A mRNA, U937 cells were transiently cotransfected with0.5 mg of each reporter construct (pGL3 control vector, pGL3/30UTR-RASSF1A, and pGL3/30UTR-RASSF1A mutated), 0.1 mg of

Renilla construct (pRL) and 1 mmol/L miR-181a, miR-181bmimics, or control mimics using Amaxa Technology (Lonza).Luciferase activities were determined 24 hours after transfectionusing the Dual-Luciferase Reporter Assay System (Promega).Values were normalized using Renilla luciferase.

Lentivirus production and transductionpmiR-ZIP-lentivirus vectors were purchased from System Bio-

sciences, and p-RFP-CB-shLenti vectors were purchased fromOrigene. miR-ZIP-lentiviral particles were produced according tothemanufacturer's instructions. NB4 cells were infected two timeswithin 48 hours with 10 mL of PEG-it–concentrated viral particlesfollowed by Puromycin selection of transduced cells for addi-tional 7 days followed by FACS-GFP sorting to create stable miR-ZIPNB4 cell lines. For generating aRASSF1A-knockdownNB4 cellline, viral particles were produced by tranfecting 293NT cells witha set of four different shRNA vectors and the control shRNA vectoraccording to the manufacturer's instructions. Selection of shRNA-expressing clones was performed by using blasticidin for 2 weeks.

Cell growth assayNB4 cells stably transduced with pmiR-ZIP-181a, pmiR-ZIP-

181b, or control vectorwere plated in a density of 1� 104 cells/mL.Proliferation rate was ascertained by cell counting in a Neu-bauer camber using Trypan blue staining for excluding deathcells.

Clonogenic assayNB4 cells stably expressing anti-miR-181a, anti-miR-181b, or

an unspecific control sequence were seeded in a density of 1� 103

cells/mL in methylcellulose-based media (MethoCult H4230;StemCell Technologies) as triplicates according to the manufac-turer's instructions. Replating was done after 6 days and repeatedthree times. Colony numbers were evaluated after each plating bystandard criteria. For colony size measurement, pictures weretaken randomly from each condition and a total number of135 colonies were measured by using ImageJ software.

miRNA and mRNA detection by quantitative real-time PCRTotal RNA was extracted using TRIzol. miRNA quantification

was performed as previously described by using hsa-miR-181aandhsa-miR-181b primer sets ormmu-miR-181a andmmu-miR-181b-1 primer sets (Applied Biosystems Inc.). Normalizationwasdone by measuring RNU6B (U6) expression and small nuclearRNA135 (snoR135) expression. mRNA amplification was per-formed as previously described by using GAPDH expression fornormalization (11). Primer sequences are provided in Supple-mentary Table S3. PCR reactionswere performed in aMyiQ2Two-Color Real-Time PCR Detection System (Bio-Rad).

Immunoblot analysisImmunoblot analysis was performed as previously described

(11). Following antibodies were used: mouse monoclonal andpolyclonal antibody anti-RASSF1A for analyzing human cell linesandprimary patient cell samples, rabbit polyclonal antibody anti-RASSF1 for analyzing mouse bone marrow samples (bothAbcam) and rabbit polyclonal antibody anti-RARa, rabbit poly-clonal antibody anti-cyclin D1 and rabbit polyclonal antibodyanti-p53 (Santa Cruz Biotechnology). Rabbit polyclonal antibo-dies anti-b-tubulin and anti-GAPDHwere used for normalization

Br€auer-Hartmann et al.

Cancer Res; 75(16) August 15, 2015 Cancer Research3412

on July 18, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 3, 2015; DOI: 10.1158/0008-5472.CAN-14-3521

(both Santa Cruz Biotechnology). Immunodetection was per-formed with WesternSure Chemiluminescent Substrate (LI-CORBiosciences). Band intensities were quantified using ImageJ soft-ware (NIH, Bethesda, MD).

Flow cytometryCell differentiation was evaluated by direct immunofluores-

cent staining using phycoerythrin (PE)-conjugated mouse anti-human CD11b/Mac-1, allophycocyanin (APC)-conjugatedmouse anti-human CD11b/Mac-1 (BD Biosciences), andAPC-conjugated mouse anti-CD114 (GCSF-R; Biolegend) cellsurface myeloid-specific antigens. Apoptosis was measuredwith a PE Annexin V Apoptosis Detection Kit I (BD Biosciences)according to the manufacturer's instructions. Cell cycle wasmeasured by performing ethanol fixation of cells followed byRNAse A digestion and propidium iodide staining of DNA. Aminimum of 10,000 events were collected for each sample by aFACScan flow cytometer (Becton Dickinson) using CellQuestsoftware for data acquisition and Cyflogic software for dataanalysis.

Statistical analysisWeused the Student t test to determine statistical significance of

experimental results. A P value of 0.05 or less was consideredsignificant (�) and a P value of 0.01 or less were considered ashighly significant (��). The results were represented as themedian� SD from three independent experiments for cell line experi-ments and from two independent reverse transcription and quan-titative PCR (qPCR) analysis for primary cell samples.

ResultsATRA treatment represses miR-181 family member expressionin APL cell line and in APL patients

Several studies show the modulation of miRNA pattern uponATRA treatment (10, 14, 19). We analyzed miR-181 familymember expression (miR-181a-d) 24 hours after ATRA treat-ment and observed a significant downregulation of all miR-181family members (Fig. 1A). The genomically clustered miR-181aand miR-181b show similar expression levels, whereas miR-181c and miR-181d, also organized in a genomic cluster,were differentially expressed (20). miR-223 expression wasused as experimental control and showed a 2-fold upregulation(Fig. 1B).

Furthermore, we induced granulocytic differentiation by ATRAin NB4 cells and the non-APL cell lines U937 and HL60. Differ-entiation was confirmed by CD11bmeasurement (Fig. 1C, E, andG). NB4 cells showed a significant reduction of miR-181a/bexpression over time (Fig. 1D). No significant change of miR-181a/b expression could be observed in U937 (Fig. 1F) andHL60cells (Fig. 1H).

In clinical APL therapy, ATRA is used in combination withanthracyclines, cytarabines, and arsenic trioxide. To test theireffects on miR-181a/b expression, NB4 cells were treated withanthracycline (doxorubicin) with or without cytarabine (ara-c)or arsenic trioxide (As2O3) and the appreciate controls for 24hours. The results showed no repression of miR-181a/b expres-sion after treatment with cytostatics or arsenic trioxide in contrastto ATRA, which strongly reduced the miRNA expression (Fig. 1Iand J). 7-Aminoactinomycin D (AAD)/Annexin V measurementrevealed strong induction of apoptosis by cytostatics and As2O3.

CD11b cell surface marker and C/EBPb mRNA were slightlyand only temporary induced by As2O3 or cytostatics comparedwith ATRA, which strongly induced granulopoiesis (Supplemen-tary Fig. S1). Furthermore, we used samples from patientswith APL who received a combination of ATRA with chemo-therapy to investigate the expression of miR-181a/b. Analysiswas performed at time point of diagnosis and at one timepoint during therapy (Supplementary Table S1). In all sixanalyzed patients, we observed a highly significant repressionof miR-181a/b expression in consequence to ATRA-based ther-apy (Fig. 1K and L).

miRNA-181a/b are induced by the oncogenic fusion proteinPML/RARa in vivo and in vitro

The oncogenic fusion protein PML/RARa is known to reg-ulate a huge number of different genes (16, 18, 21). To showthe regulatory influence of PML/RARa on miR-181a/b expres-sion, we induced PML/RARa protein in U937-PR9 cells withZnSO4. miR-181a/b expression was immediately upregulatedupon PML/RARa induction (Fig. 2A). The control cell lineU937-PC18 showed no significant change in miRNA expres-sion after ZnSO4 application (Fig. 2B). In addition, we ana-lyzed bone marrow samples from PML/RARa knock-in miceand wild-type animals (C57BI/6-WT; ref. 17). We observed asignificant enhanced expression of murine miR-181a/b inPML/RARa knock-in mouse samples in comparison to wild-type samples (Fig. 2C). Furthermore, we analyzed bone mar-row samples from patients with different AML subtypesand blood cell samples from healthy donors. The resultsshowed significantly higher miR-181a/b expression values inAPL patient samples than in the samples with normal kar-yotype, whereas all other analyzed samples showed no sig-nificant miRNA expression change (Supplementary Table S2and Fig. 2D).

The miR-181a/b cluster is necessary for proliferation, inducesapoptosis, and inhibits granulocytic differentiation of APL cells

To address the role of miR-181a/b in APL in detail, westably knocked down both miRNAs in NB4 cells by usingmiR-ZIP-lentiviral particles. Knockdown efficiency was verifiedby qPCR (Fig. 3A). The PML/RARa fusion protein promotes cellsurvival (4). To evaluate a potential role for miR-181a and miR-181b in this process, we examined an apoptosis assay. Knock-downofmiR-181a andmiR-181b results in significantly increasedapoptosis (Fig. 3B). According to this, p53 protein was induced inthe miR-ZIP-181a and -181b expressing cells compared with thecontrol cells (Supplementary Fig. S2). Furthermore, in a replatingassay, we could show strongly reduced colony size and colony-forming capacity of NB4 cells after knockdown of miR-181a andmiR-181b (Fig. 3C andD). According to this, the proliferation rateof miR-ZIP-181a- or miR-ZIP-181b–expressing cells was signifi-cantly reduced (Fig. 3E).

To investigate the influence of miR-181a/b on ATRA-induceddifferentiation of APL cells, we transiently transfected NB4 cellswith miR-181a- and miR-181b–specific mimics and controlmimics. Granulocytic differentiation was induced by ATRA 24hours after transfection. CD11b expression was significantlydecreased after overexpression of one of the miRNAs in compar-ison to the control 48 hours after transfection. The effect wasslightly increased by simultaneous transfection of miR-181a and

miR-181a/b Targets RASSF1A in APL

www.aacrjournals.org Cancer Res; 75(16) August 15, 2015 3413

on July 18, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 3, 2015; DOI: 10.1158/0008-5472.CAN-14-3521

Figure 1.In APL, miR-181 family member expression is repressed by ATRAin vitro and in vivo. A and B, qPCR for miR-181 family member andmiR-223 expression 24 hours after ATRA treatment ofNB4 cells. C, E, and G, FACS for CD11b expression in NB4 (C),U937 (E), and HL60 cells (G) 72 hours after ATRA application.D, F, and H, qPCR for miR-181a/b in NB4 (D), U937 (F), and HL60cells (H) after ATRA application at indicated time points.I and J, qPCR for miR-181a/b in NB4 cells treated with As2O3,doxorubicin (doxo) w/o cytarabine or ATRA for 24 hours. K andL, qPCR formiR-181a/b in bonemarrow samples fromAPLpatientat diagnosis time point (newly diagnosed APL) and during ATRAtreatment (APL during ATRA-based therapy). n.s., notsignificant. � , P � 0.05; �� , P � 0.01.

Br€auer-Hartmann et al.

Cancer Res; 75(16) August 15, 2015 Cancer Research3414

on July 18, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 3, 2015; DOI: 10.1158/0008-5472.CAN-14-3521

Figure 2.PML/RARa induces miR-181a and miR-181b expression in vitro and in vivo. A and B, U937-PR9 and the control cell line U937-PC18 were treated with ZnSO4

or H2O (vehicle). qPCR for miR-181a/b was performed at indicated time points. Western blotting for RARa and PML/RARa protein upon ZnSO4 application in PR9and PC18 cells (left). C, qPCR for murine miR-181a and miR-181b (mmu-miR-181b-1) in 5 PML/RARa knock-in mice (C57BI/6-mCGþ/PR) and 5 wild-type mice(C57BI/6-WT). Data represents the dCt values. D, qPCR for miR-181a/b expression in bone marrow samples from patients with AML, with indicated subtypesand healthy donors. n.s., not significant. � , P � 0.05; �� , P � 0.01.

miR-181a/b Targets RASSF1A in APL

www.aacrjournals.org Cancer Res; 75(16) August 15, 2015 3415

on July 18, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 3, 2015; DOI: 10.1158/0008-5472.CAN-14-3521

Br€auer-Hartmann et al.

Cancer Res; 75(16) August 15, 2015 Cancer Research3416

on July 18, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 3, 2015; DOI: 10.1158/0008-5472.CAN-14-3521

miR-181b mimics (Fig. 3F). In addition, qPCR showed a strongdecrease of ATRA induced C/EBPb and granulocyte colony-stim-ulating factor (GCSF) receptor mRNA expression 48 hours aftermiR-181a andmiR-181bmimic transfection in comparison to thecontrol (Fig. 3G).

The tumor suppressor RASSF1A is a direct target of themiR-181a/b cluster

We hypothesized that miR-181a and miR-181b are involvedin formation of the PML/RARa-induced oncogenic transfor-mation by targeting differentiation-required genes. By compu-tational analysis using miRNA target prediction programssuch as Target Scan (http://www.targetscan.org), we identifiedthe tumor suppressor Ras association domain family member1 isoform A (RASSF1A) as a putative target of the miR-181family (Fig. 4D). RASSF1A is a well-characterized tumor sup-pressor that is epigenetically suppressed by promoter hyper-methylation in a wide range of tumors (22). In APL, no case isreported until now (23). We evaluated RASSF1A protein inNB4 cells by Western blotting. The data showed a significantincrease of RASSF1A protein upon ATRA treatment over time(Fig. 4A). Interestingly, RASSF1A protein showed no upregula-tion in NB4 cells upon induction of apoptosis by arsenictrioxide (Fig. 4B and Supplementary Fig. S1A). The 30UTRof RASSF1A harbors three potential miR-181–binding sites(Fig. 4C). To analyze direct binding of miR-181a and miR-181b to the 30UTR of RASSF1A, we generated a luciferaseconstruct containing the complete 30UTR of RASSF1A andmutated the binding sites (Fig. 4D and E). Reporter assayshowed repression of luciferase activity after miR-181a andmiR-181b mimic transfection in comparison to the control.Mutation of the miR-181–binding sites resulted in the recoveryof luciferase activity and revealed direct binding of miR-181a/bto the 30UTR of RASSF1A (Fig. 4F).

To validate our finding that RASSF1A is a direct target of miR-181a/b, we analyzed RASSF1A protein after miR-181a and miR-181b mimic transfection in U937 cells and observed repressionof RASSF1A protein 24 hours and much stronger 48 hours aftertransfection (Fig. 4G). In addition, RASSF1A protein wasincreased after miR-ZIP–mediated knockdown of miR-181aand miR-181b in NB4 cells compared with the control cells(Fig. 4H).

RASSF1A protein is specifically suppressed in APLRASSF1A is shown to be an important regulator of cell

differentiation in a wide range of cell types (22) by exercisingits functions as a modulator of two pathways commonlyderegulated in cancer, apoptosis, and cell cycle (24, 25). There-fore, we hypothesize that RASSF1A could function as tumorsuppressor in APL, where PML/RARa causes oncogenic trans-formation by deregulating cell cycle and apoptosis (4).

Western blot analysis of bone marrow samples from patientswith different AML subtypes and blood cell samples from healthydonors showed significantly lower amounts of RASSF1A proteinin patients with APL compared with patients with AML withnormal karyotype (Supplementary Table S2 and Fig. 5A). Nosignificant change in RASSF1A protein levels could be observed inother AML subtypes and in non-AML samples. Interestingly,RASSF1A mRNA exhibited a heterogeneous distribution andespecially no reduction in t(15;17) (Fig. 5B). Furthermore,we observed significantly lower RASSF1A protein levels inPML/RARa knock-in mice (C57BI/6-mCGþ/PR) compared withwild-type animals (C57BI/6-WT) (Fig. 5C).

RASSF1A is essential for ATRA-induced granulocyticdifferentiation and induces apoptosis in APL

To verify the hypothesized differentiation-associated functionof RASSF1A, we used a set of four constructs encoding differentshRNA sequences specific for RASSF1AmRNA to generate a stableNB4 RASSF1A-knockdown cell line. The knockdown efficiencywas confirmed by qPCR and showed a significant reduction ofRASSF1A mRNA to 0.2-fold in the RASSF1A shRNAs expressingcells in comparison to the cells expressing an unspecific shRNA(Fig. 6A). ATRA-induced CD11b expression and GCSF-R expres-sion were strongly repressed in RASSF1A shRNA-expressing cellsin comparison to the control cells (Fig. 6B). Transient overexpres-sion of RASSF1A was performed to point out the tumor-suppres-sive function of RASSF1A in APL. Apoptosis assay displayed astrong increase in the Annexin V–positive cell population 24hours after pcDNA3.1-RASSF1A transfection in comparison tothe empty vector transfection (Fig. 6C).

miR-181a/b and RASSF1A modulate differentiation in APL viaregulation of cell cycle

ATRA-induced cell growth arrest and terminal differentiation ofAPL blasts involves downregulation of cyclin D1 (26). To confirmtheATRA-dependent downregulation of cyclinD1 in APL cells, weperformed Western blotting 48 hours after ATRA stimulation ofNB4 cells and observed a decrease in cyclin D1 protein (Fig. 7A).RASSF1A is also able to induce cell-cycle arrest by inhibition ofcyclin D1 accumulation (24). Overexpression of RASSSF1A bypcDNA3.1/RASSF1A resulted in a complete repression of cyclinD1 protein in NB4 cells after 24 hours in comparison to thecontrol vector transfection (Fig. 7B). We also performed cell-cycleanalysis 24 hours after ATRA treatment of RASSF1A shRNA-expressing NB4 cells and observed that shRNA-mediated knock-down of RASSF1A leads to a marked reduction in G1–G0 phaseand an increase in the S- and G2 phase in comparison to thecontrol shRNA-expressing cells (Fig. 7C). In addition,we analyzedcell cycle in miR-ZIP-181a- and miR-ZIP-181b–expressing NB4cells. The results showed that knockdown of miR-181a and miR-181b reduces S- and G2 phase and increases G1–G0 phase as well

Figure 3.In APL, themiR-181a/b cluster is necessary for cell proliferation, induces apoptosis, and inhibits granulocytic differentiation. A, qPCR formiR-181a/b inNB4 cells stablyexpressing miR-ZIP-181a, miR-ZIP-181b, or miR-ZIP control sequences. B, apoptosis assay in miR-ZIP-181a, miR-ZIP-181b, and miR-ZIP control expressing NB4 cells.The diagram represents the amount of Annexin Vþ/AAD� cells in each condition. C, replating assay of miR-ZIP-181a/b or miR-ZIP control expressingNB4 cells. Pictures represent one well of a 12-well plate as an example of each condition at the first and the third plating. D, colony size measurement of NB4cells stably expressing miR-ZIP-181a/b or miR-ZIP control sequences. Pictures represent one example for each condition. Bars, 300 mm. E, cell growth curveof miR-ZIP-181a/b or miR-ZIP control expressing NB4 cells. F, FACS for ATRA-induced CD11b expression 48 hours after miR-181a mimic, miR-181b mimic, orcontrol mimic transfection of NB4 cells. The values indicate the amount of CD11b-positive cells (%) for each transfection condition. G, qPCR for C/EBPb andGCSF-R mRNA in miR-181a, miR-181b, or control mimic transfected NB4 cells 48 hours after transfection. � , P � 0.05; ��, P � 0.01.

miR-181a/b Targets RASSF1A in APL

www.aacrjournals.org Cancer Res; 75(16) August 15, 2015 3417

on July 18, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 3, 2015; DOI: 10.1158/0008-5472.CAN-14-3521

Br€auer-Hartmann et al.

Cancer Res; 75(16) August 15, 2015 Cancer Research3418

on July 18, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 3, 2015; DOI: 10.1158/0008-5472.CAN-14-3521

as sub-G1 phase (Fig. 7D). To show the regulatory impact of miR-181a andmiR-181b on cyclin D1, we transiently transfectedmiR-181a and miR-181b mimics in U937 cells. Western blotting

showed an elevated protein level of cyclin D1 48 hours aftersingle or combined miR-181a and miR-181b mimic transfectioncompared with the control (Fig. 7E). To prove that cyclin D1

Figure 5.RASSF1A protein is specificallysuppressed by PML/RARa. A, Westernblotting for RASSF1A protein in bonemarrow samples of patients withdifferent AML subtypes and cellsamples from nonleukemic patients(healthy donors). Bars in the diagramshow the median of the normalizationratio for each patient group. Westernblot (right) analysis of RASSF1A proteinin each analyzed patient sample, withthe corresponding normalization ratiobelow. B, qPCR for RASSF1A mRNAexpression in AML patient samples.C, Western blotting for RASSF1Aprotein in bone marrow samplesfrom PML/RARa knock-in mice(C57BI/6-mCGþ/PR) and wild-typemice (C57BI/6-WT; right). The numberofmice (n) or patients (n) in each groupis shown below the bars. �, P � 0.05.

Figure 4.The tumor suppressor RASSF1A is a direct target of the miR-181a/b-cluster. A, Western blotting for RASSF1A protein in ATRA and DMSO treatment of NB4 cells atindicated time points. B, Western blotting for RASSF1A protein in As2O3 or with the control-treated NB4 cells at indicated time points. C, schematic representationof the RASSF1A 30UTR, including the three predicted miR-181–binding sites. D, schematic representation of the pGL3 constructs for the wild-type 30UTR ofRASSF1A and the mutated 30UTR of RASSF1A. E, schematic representation of the predicted and mutated miR-181–binding sites in the RASSF1A 30UTR. Numbersbehind the sequence represent the position of the nucleotides relative to the termination codon of human RASSF1A. F, luciferase assay was performed inU937 cells 24 hours after cotransfection of pGL3-30UTR RASSF1A (wild-type or mutated) or pGL3 control vector and miR-181a, miR-181b, or control mimics. Barsrepresent the luciferase activity for the corresponding vectors. Normalization was done by Renilla luciferase. G, Western blotting for RASSF1A protein48 hours after miR-181a, miR-181b, or control (nc) mimic transfection in U937 cells. H, Western blotting for RASSF1A protein in NB4 cells stably expressingmiR-ZIP-181a, miR-ZIP-181b, or miR-ZIP–control sequences. � , P � 0.05.

www.aacrjournals.org Cancer Res; 75(16) August 15, 2015 3419

miR-181a/b Targets RASSF1A in APL

on July 18, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 3, 2015; DOI: 10.1158/0008-5472.CAN-14-3521

protein increment upon miR-181a and miR-181b mimic trans-fection is mediated by repression of RASSF1A, we cotransfectedmiR-181a and miR-181b mimics and a pcDNA3.1/RASSF1Avector or the corresponding pcDNA3.1/control vector in NB4cells. Western blotting 48 hours after transfection revealed asignificant repression of cyclin D1 protein when RASSF1A isexpressed lacking a 30UTR in the presence of miRNA mimics(Fig. 7F).

DiscussionAn increasing number of studies have shown the importance of

miRNAs in the formation and maintenance of leukemia. In thisreport, we demonstrate that ATRA is able to significantly down-regulate the expression of the whole miR-181 family in APL (Fig.1A). We show the constant downregulation of the miR-181a/bcluster upon ATRA treatment over time in APL in vitro and in vivo(Fig. 1D, K, and L). Furthermore, we demonstrate that cytostaticsand arsenic trioxide, which are typically used in APL therapy andare predominately inducers of apoptosis, donot affectmiR-181a/bexpression (Fig. 1I and J). These results expand and confirmprevious observations (10, 14–16) and suggest a specific role for

the miR-181 family in the response to ATRA in APL. Diversepublications illustrate the expression pattern and define multiplefunctions for miR-181a and miR-181b in hematopoiesis andleukemia, whereas miR-181c and miR-181d are less described(8, 27–31). The fact that ATRA leads to the degradation of PML/RARa and thereby changes gene expression, assuming that miR-181a/b expression is dependent on PML/RARa (1). We followedmiR-181a/b expression upon ATRA treatment of the non-APL celllines U937 and HL60. Both cell lines respond to ATRA but showno significant change inmiR-181a/b expression (Fig. 1E–H). Thisobservation substantiates the proposed PML/RARa dependencyof miR-181a/b expression.

The miR-181a/b cluster has been shown to be upregulated inpatients with AML with C/EBPa mutations who have a favor-able prognosis and have also shown to be associated withfavorable outcome in patients with cytogenetically normal AMLand cytogenetically abnormal AML (32–34). Combining thesedata, high expression of miR-181a and miR-181b occurs incombination with a favorable outcome of AML. In APL, acombination of ATRA and arsenic trioxide therapy generatesa complete remission rate (CR) of more than 90% (35). Ourobservation that the miR-181a/b cluster is highly expressed in

Figure 6.RASSF1A is essential for ATRA-induced granulocytic differentiation and induces apoptosis in APL. A, qPCR for RASSF1A mRNA in NB4 cells stably expressingRASSF1A shRNA sequences or an unspecific control. B, FACS for CD11b expression 48 hours and GCSF-R expression 72 hours after ATRA application. Graphson right show the amount of CD11b- or GCSF-R–positive cells (%) for each condition. C, Annexin V (PE)/7-AAD staining of NB4 cells 24 hours after overexpressionof RASSF1A. The graph shows the Annexin Vþ/7AAD� fraction of pcDNA3.1-RASSF1A and pcDNA3.1-nc (control vector)–transfected cells. � , P � 0.05;�� , P � 0.01.

Cancer Res; 75(16) August 15, 2015 Cancer Research3420

Br€auer-Hartmann et al.

on July 18, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 3, 2015; DOI: 10.1158/0008-5472.CAN-14-3521

APL and significantly downregulated upon ATRA treatmentin vitro and in vivo points to a role for the miRNA cluster asprognostic marker in t(15;17).

Besides its function as transcriptional repressor (2), PML/RARais also able to induce transcription, whereas this effect seems to beindirect due the sequestration of corepressors (36). In this study,

we demonstrate the PML/RARa-dependent upregulation of miR-181a/b in PR9 cells and in PML/RARa knock-inmice (Fig. 2A–C).In addition, we show significantly higher expression of the miR-181a/b cluster in bone marrow samples from patients with APL(Fig. 2D). These results are reinforced by data from Li andcolleagues (31) and Jongen-Lavrencicand colleagues (32). Taken

Figure 7.miR-181a/b and RASSF1A modulate differentiation and apoptosis in APL via regulation of cyclin D1. A and B, Western blotting for cyclin D1 protein in NB4 cells 48hours after ATRA application (A) and 24 hours after overexpression of pcDNA3.1-RASSF1A or pcDNA3.1-nc (B). C, cell-cycle analysis of NB4 cells stablyexpressing RASSF1A-specific shRNA sequences 24 hours after ATRA application. D, cell-cycle analysis of miR-ZIP-181a, miR-ZIP-181b, or miR-ZIP control expressingNB4 cells. E, Western blot analysis for cyclin D1 in U937 cells 48 hours after miR-181a, miR-181b, or control mimic transfection. F, Western blot analysis forcyclin D1 protein 48 hours after cotransfection of miR-181a, miR-181b, or control mimics and pcDNA3.1-RASSF1A or the pcDNA3.1-nc-vector. G, schematicrepresentation of ATRA- and PML/RARa-dependent regulation of the miR-181a/b cluster and RASSF1A. The oncogenic fusion protein PML/RARa leads to thetranscriptional induction of miR-181a and miR-181b. Both miRNAs block RASSF1A protein synthesis, which leads to cell-cycle progression and proliferation viaaccumulation of cyclin D1. Treatment of APL blasts with pharmacologic doses of ATRA leads to the destruction of the PML/RARa fusion protein followed byrepression ofmiR-181a/b transcription in vitro and in vivo. This results in the recovery of RASSF1Aprotein synthesis, which leads to cell-cycle arrest anddifferentiationas well as apoptosis through inhibition of cyclin D1 accumulation. n.s., not significant. � , P � 0.05; �� , P � 0.01.

miR-181a/b Targets RASSF1A in APL

www.aacrjournals.org Cancer Res; 75(16) August 15, 2015 3421

on July 18, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 3, 2015; DOI: 10.1158/0008-5472.CAN-14-3521

together, to the best of our knowledge, we are the first to showPML/RARa-dependent upregulation of the miR-181a/b cluster inAML. Because PML/RARa has no direct binding site in the pro-moter region of the miR-181a/b cluster (16), the transcriptionalinduction has to occur indirectly. The exact mechanism howthe miRNA cluster is regulated is still unknown and has to beinvestigated in further experiments.

Diverse publications describe miR-181 family members aseither oncogenes or tumor suppressors in various cancers depend-ing on tissue type (37–40). The fact thatmiR-181a/b expression issignificantly high in t(15;17), assuming that both miRNAs areinvolved in the formation of the oncogenic phenotype caused byPML/RARa. Our functional studies show that ectopic expressionof miR-181a and miR-181b effectively blocks ATRA-inducedgranulocytic differentiation (Fig. 3F and G) and that inhibitionof the miR-181a/b cluster effectively represses cell proliferationand induces apoptosis in APL cells (Fig. 3A–E). In contrast todata from Hickey and colleagues (33) and Li and colleagues(34) who assign miR-181a as an anti-leukemic miRNA in AML,our results show an oncogenic function for the miR-181a/bcluster in APL. This is supported by recently published datafrom Su and colleagues, which showed that miR-181a blocksmyeloid differentiation of HL60 and CD34þ hematopoieticstem/progenitor cells (41).

In cancer, oncogenic miRNAs exercise their function bytargeting tumor suppressors (42, 43). In our study, we identifythe known tumor suppressor RASSF1A (Ras associationdomain family member 1 isoform A) as a direct target of themiR-181a/b cluster in APL. miR-181 family members are theonly miRNAs that have three conserved binding sites in the30UTR of RASSF1A (Fig. 4A and E). RASSF1A has been found tobe epigenetically inactivated in a variety of cancers by promoterhypermethylation, and reintroduction of RASSF1A in RASSF1A-deficient tumor cells leads to the reduction of tumorigenicity(22, 44). Because there is no RASSF1A promoter hypermethyla-tion in APL, there must be other mechanisms how the tumorsuppressor is inactivated (23, 45). In this study, we show theupregulation of RASSF1A protein upon ATRA treatment in NB4cells while miR-181a/b expression is decreasing (Figs. 1A, D, K,and L and 4A). In addition, we demonstrate that RASSF1Aprotein is not regulated by arsenic trioxide, which also does notaffect miR-181a/b expression (Fig. 4B, Fig. 1I and J). Finally, weprove by luciferase assay that repression of RASSF1A proteinoccurs via direct binding of miR-181a and miR-181b to its30UTR (Fig. 4F), which has also been shown in hepatocellularcancer stem cells by Meng and colleagues (46). Our luciferaseassay data in combination with our data from knockdown andoverexpression experiments of miR-181a and miR-181b firstlyshow the direct repression of RASSF1A translation by the miR-181a/b cluster in the background of APL (Fig. 4F–H). Further-more, we observed the APL-specific inverse correlation of miR-181a/b expression and RASSF1A protein, which could not beseen in the other analyzed AML subgroups and healthydonors. The proposed specificity of RASSF1A suppression by aPML/RARa-dependent mechanism is supported by proteinexpression data in AML patient samples and in PML/RARaknock-in mice, which show reduced RASSF1A protein levelswhen the miR-181a/b cluster is highly expressed (Figs. 2C andD and 5). In addition, mRNA expression analysis of RASSF1A inAML patient samples and healthy donors substantiates thesuggested regulation mechanism via miR-181a/b (Fig. 5B).

These findings are supported by a recently published work byZare-Abdollahi and colleagues (47).

Until now, no function for RASSF1A in APL or granulocyticdifferentiation has been shown. In this study, we describeRASSF1A as an essential component of the ATRA-induced gran-ulocytic differentiation network in APL. Enforced expression ofRASSF1A leads to enhanced apoptosis of NB4 cells, which con-firms the tumor-suppressive function of RASSF1A in APL (Fig.6C). In addition, reduced granulocytic differentiation ofNB4 cellsin consequence to RASSF1A knockdown supports the proposeddifferentiation associated function of RASSF1A (Fig. 6B). It wasshown that RASSF1A is able to effectively prevent G1–S phasetransition by blocking cyclinD1 accumulation (24) and to induceapoptosis (48). In APL, ATRA induces APL cell differentiation intomature granulocytes and results in cell apoptosis (26). Thisprocess involves the sequential regulation of cell-cycle regulatoryproteins, such as cyclin D1, which promotes G1–S progression(49). In our study, we confirm the ATRA-induced repression ofcyclin D1 in APL cells (Fig. 7A). We also show that overexpressionof RASSF1A leads to a dramatic repression of cyclin D1 proteinand that knockdown of RASSF1A promotes cell-cycle progressionin APL (Fig. 7B and C). Furthermore, we could demonstrate thatRASSF1A is the key mediator for miR-181a- and miR-181b–mediated induction of cyclin D1 accompanied by cell-cycle pro-gression in APL (Fig. 7E and F). Finally, in contrast to ATRA,arsenic trioxide, which induces apoptosis in APL, does not affectmiR-181a/b expression andRASSF1Aprotein. This correlateswiththe finding that arsenic trioxide does not downregulate cyclin D1protein (50). On the basis of these findings, we claim RASSF1A asan important factor in the granulocytic differentiation, whichprevents cyclin D1 accumulation, cell-cycle progression, andpromotes differentiation upon ATRA treatment in APL blasts.

In summary, our study highlights the clustered miR-181a andmiR-181b as important factors in the PML/RARa-associated APL.To the best of our knowledge, we are the first describing the miR-181a/b target RASSF1A as an essential member of the ATRA-induced granulocytic differentiation network in APL (Fig. 7G).Both miRNAs, transcriptional induced by PML/RARa, lead totranslational repression of the tumor suppressor RASSF1A. Itsfunction is restored trough repression of miR-181a/b expressionby ATRA-induced degradation of PML/RARa, which preventsaccumulation of cyclin D1 and induces cell-cycle arrest, whichresults in granulocytic differentiation of APL blasts. Our datareveal a mechanism of tumor suppressor inhibition by a micro-RNA cluster that seems to be highly specific for APL. Finally,manipulation of miR-181a/b could offer novel treatment strate-gies in PML/RARa-associated APL.

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

Authors' ContributionsConception and design: D. Br€auer-Hartmann, G. BehreDevelopment ofmethodology:D. Br€auer-Hartmann, J.-U. Hartmann, G. BehreAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): D. Br€auer-Hartmann, J.-U. Hartmann, M.V. VergaFalzacappa, P.G. Pelicci, G. BehreAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): D. Br€auer-Hartmann, J.-U. Hartmann, A.A. Wurm,C. Katzerke, D. Niederwieser, G. BehreWriting, review, and/or revision of the manuscript: D. Br€auer-Hartmann,D. Gerloff, C. Katzerke, C. M€uller-Tidow, D.G. Tenen, D. Niederwieser, G. Behre

Br€auer-Hartmann et al.

Cancer Res; 75(16) August 15, 2015 Cancer Research3422

on July 18, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 3, 2015; DOI: 10.1158/0008-5472.CAN-14-3521

Administrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): D. Br€auer-Hartmann, J.-U. Hartmann,A.A. Wurm, D. Gerloff, C. Katzerke, D.G. Tenen, D. Niederwieser, G. BehreStudy supervision: D. Niederwieser, G. Behre

AcknowledgmentsThe authors thank R. Dammann for providing pGL3.1/RASSF1A expression

construct.

Grant SupportThis study was supported by grants from DFG (German Research Foun-

dation, BE 2042/7-1, BE 2042/12-1), Deutsche José Carreras Stiftung E.V.

(R11/17, R12/31), Deutsche Krebshilfe, Wilhelm Sander Stiftung (Nr.2013.153.1), and Translational Centre for Regenerative Medicine Leipzig toG. Behre and Deutsche José Carreras Stiftung e. V. to D. Bräuer-Hartmann andthe National Institute of Health (CA66996 and HL112719) to D.G. Tenen.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received December 1, 2014; revised April 27, 2015; accepted May 12, 2015;published OnlineFirst June 3, 2015.

References1. Melnick A, Licht JD. Deconstructing a disease: RARalpha, its fusion part-

ners, and their roles in the pathogenesis of acute promyelocytic leukemia.Blood 1999;93:3167–321.

2. Di Croce L, Raker VA, Corsaro M, Fazi F, Fanelli M, Faretta M, et al.Methyltransferase recruitment and DNA hypermethylation of target pro-moters by an oncogenic transcription factor. Science 2002;295:1079–82.

3. De The H, Lavau C, Marchio A, Chomienne C, Degos L, Dejean A. ThePMLRARalpha fusion mRNA generated by the t(15;17) translocation inacute promyelocytic leukemia encodes a functionally altered RAR.Cell 1991;66:675–84.

4. Grignani F, Ferrucci PF, Testa U, Talamo G, Fagioli M, Alcalay M, et al. Theacute promyelocytic leukemia-specific PML-RAR alpha fusion proteininhibits differentiation and promotes survival of myeloid precursor cells.Cell 1993;74:423–31.

5. Wang ZG, Ruggero D, Ronchetti S, Zhong S, Gaboli M, Rivi R, et al. PML isessential for multiple apoptotic pathways. Nat Genet 1998;20:266–72.

6. Duprez E, Wagner K, Koch H, Tenen DG. C/EBPbeta: a major PML-RARAresponsive gene in retinoic acid-induced differentiation of APL cells. EMBOJ 2003;22:5806–16.

7. Bartel DP. MicroRNAs: target recognition and regulatory functions.Cell 2009;136:215–33.

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

9. Havelange V, Garzon R, Croce CM. MicroRNAs: new players in acutemyeloid leukaemia. Br J Cancer 2009;101:743–8.

10. Fazi F, Rosa A, Fatica A, Gelmetti V, De Marchis ML, Nervi C, et al. Aminicircuitry comprised of microRNA-223 and transcription factorsNFI-A and C/EBPalpha regulates human granulopoiesis. Cell 2005;123:819–31.

11. Pulikkan JA, Dengler V, Peramangalam PS, Peer Zada AA, M€uller-Tidow C,Bohlander SK, et al. Cell-cycle regulator E2F1 andmicroRNA-223 comprisean autoregulatory negative feedbackloop in acute myeloid leukemia.Blood 2010;115:1768–78.

12. Pulikkan JA, Peramangalam PS, Dengler V, Ho PA, Preudhomme C,Meshinchi S, et al. C/EBPa regulated microRNA-34a targets E2F3 duringgranulopoiesis and is down-regulated in AML with CEBPA mutations.Blood 2010;116:5638–49.

13. Katzerke C,Madan V, Gerloff D, Br€auer-HartmannD,Hartmann JU,WurmAA, et al. Transcription factor C/EBPa-induced microRNA-30c inactivatesNotch1 during granulopoiesis and is downregulated in acute myeloidleukemia. Blood 2013;122:2433–42.

14. GarzonR, Pichiorri F, Palumbo T, VisentiniM, Aqeilan R, CimminoA, et al.MicroRNA gene expression during retinoic acid-induced differentiation ofhuman acute promyelocytic leukemia. Oncogene 2007;26:4148–57.

15. Careccia S, Mainardi S, Pelosi A, Gurtner A, Diverio D, Riccioni R, et al. Arestricted signature of miRNAs distinguishes APL blasts from normalpromyelocytes. Oncogene 2009;28:4034–40.

16. Saumet A, Vetter G, Bouttier M, Portales-Casamar E, Wasserman WW,Maurin T, et al. Transcriptional repression of microRNA genes by PML-RARA increases expression of key cancer proteins in acute promyelocyticleukemia. Blood 2009;113:412–21.

17. Westervelt P, Lane AA, Pollock JL, Oldfather K,HoltMS, Zimonjic DB, et al.High-penetrance mouse model of acute promyelocytic leukemia with verylow levels of PML-RARalpha expression. Blood 2003;102:1857–65.

18. Zada AA, Geletu MH, Pulikkan JA, M€uller-Tidow C, Reddy VA, ChristopeitM, et al. Proteomic analysis of acute promyelocytic leukemia: PML-RAR-alpha leads to decreased phosphorylation ofOP18 at serine 63. Proteomics2006;6:5705–19.

19. Fatica A, Rosa A, Ballarino M, De Marchis ML, Rasmussen KD, Bozzoni I.Role of microRNAs in myeloid differentiation. BiochemSoc Trans 2008;36:1201–5.

20. Yu J, Wang F, Yang GH,Wang FL, Ma YN, Du ZW , et al. HumanmicroRNAclusters: genomic organization and expression profile in leukemia celllines. BiochemBiophys Res Commun 2006;349:59–68.

21. Hoemme C, Peerzada A, Behre G, Wang Y, McClelland M, Nieselt K, et al.Chromatin modifications induced by PML-RARalpha repress critical tar-gets in leukemogenesis as analyzed by ChIP-Chip. Blood 2008;111:2887–95.

22. Agathanggelou A, Cooper WN, Latif F. Role of the Ras-association domainfamily 1 tumor suppressor gene in human cancers. Cancer Res2005;65:3497–508. Review. Erratum in: Cancer Res 2005;65:5480.

23. Johan MF, Bowen DT, Frew ME, Goodeve AC, Reilly JT. Aberrant meth-ylation of the negative regulators RASSFIA, SHP-1 and SOCS-1 in myelo-dysplastic syndromes and acute myeloid leukaemia. Br J Haematol 2005;129:60–5.

24. Shivakumar L, Minna J, Sakamaki T, Pestell R, White MA. The RASSF1Atumor suppressor blocks cell cycle progression and inhibits Cyclin D1accumulation. Mol Cell Biol 2002;22:4309–18.

25. Vos MD, Ellis CA, Bell A, Birrer MJ, Clark GJ. Ras uses the novel tumorsuppressor RASSF1 as an effector to mediate apoptosis. J BiolChem2000;275:35669–72.

26. Breitman TR, Collins SJ, Keene BR. Terminal differentiation of humanpromyelocytic leukemic cells in primary culture in response to retinoicacid. Blood 1981;57:1000–4.

27. Li QJ, Chau J, Ebert PJ, Sylvester G, Min H, Liu G, et al. miR-181a is anintrinsic modulator of T cell sensitivity and selection. Cell 2007;129:147–61.

28. Debernardi S, Skoulakis S, Molloy G, Chaplin T, Dixon-McIver A, YoungBD. MicroRNAmiR-181a correlates with morphological sub-class of acutemyeloid leukaemia and the expression of its target genes in global genome-wide analysis. Leukemia 2007;21:912–6.

29. Georgantas RW III, Hildreth R,Morisot S, Alder J, Liu CG, Heimfeld S, et al.CD34þ hematopoietic stem-progenitor cell microRNA expression andfunction: a circuit diagram of differentiation control. Proc Natl Acad SciU S A 2007;104:2750–5.

30. Visone R, Veronese A, Rassenti LZ, Balatti V, Pearl DK, AcunzoM, et al.miR-181b is a biomarker of disease progression in chronic lymphocytic leuke-mia. Blood 2011;118:3072–9.

31. Li Z, Lu J, Sun M, Mi S, Zhang H, Luo RT, et al. Distinct microRNAexpression profiles in acute myeloid leukemia with common transloca-tions. Proc Natl Acad Sci U S A 2008;105:15535–40.

32. Jongen-Lavrencic M, Sun SM, Dijkstra MK, Valk PJ, L€owenberg B. Micro-RNA expression profiling in relation to the genetic heterogeneity of acutemyeloid leukemia. Blood 2008;111:5078–85.

33. Hickey CJ, Schwind S, Radomska HS, Dorrance AM, Santhanam R, MishraA, et al. Lenalidomide-mediated enhanced translation of C/EBPa-p30protein up-regulates expression of the antileukemic microRNA-181a inacute myeloid leukemia. Blood 2013;121:159–69.

miR-181a/b Targets RASSF1A in APL

www.aacrjournals.org Cancer Res; 75(16) August 15, 2015 3423

on July 18, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 3, 2015; DOI: 10.1158/0008-5472.CAN-14-3521

34. Li Z, Huang H, Li Y, Jiang X, Chen P, Arnovitz S, et al. Up-regulation of aHOXA-PBX3 homeobox-gene signature following down-regulation ofmiR-181 is associated with adverse prognosis in patients with cytogeneti-cally abnormal AML. Blood 2012;119:2314–24.

35. Wang ZY, Chen Z. Acute promyelocytic leukemia: from highly fatal tohighly curable. Blood 2008;111:2505–15.

36. Licht JD. Reconstructing a disease: What essential features of the retinoicacid receptor fusion oncoproteins generate acute promyelocytic leukemia?Cancer Cell 2006;9:73–4.

37. Pekarsky Y, SantanamU, Cimmino A, Palamarchuk A, Efanov A, MaximovV, et al. Tcl1 expression in chronic lymphocytic leukemia is regulated bymiR-29 and miR-181. Cancer Res 2006;66:11590–3.

38. IliopoulosD, Jaeger SA,HirschHA, BulykML, Struhl K. STAT3 activation ofmiR-21 and miR-181b-1 via PTEN and CYLD are part of the epigeneticswitch linking inflammation to cancer. Mol Cell 2010;39:493–506.

39. Wang B, Hsu SH,Majumder S, KutayH, HuangW, Jacob ST, et al. TGFbeta-mediated upregulation of hepatic miR-181b promotes hepatocarcinogen-esis by targeting TIMP3. Oncogene 2010;29:1787–97.

40. Zhu DX, Zhu W, Fang C, Fan L, Zou ZJ, Wang YH, et al. miR-181a/bsignificantly enhances drug sensitivity in chronic lymphocytic leukemiacells via targeting multiple anti-apoptosis genes. Carcinogenesis2012;33:1294–301. Erratum in: Carcinogenesis 2013 Sep;34:2192.

41. Su R, Lin HS, Zhang XH, Yin XL, Ning HM, Liu B, et al. MiR-181 family:regulators of myeloid differentiation and acute myeloid leukemia as wellas potential therapeutic targets. Oncogene 2015;34:3226–39.

42. Lujambio A, Lowe SW. The microcosmos of cancer. Nature 2012;482:347–55.

43. Gerloff D, Grundler R, Wurm AA, Br€auer-Hartmann D, Katzerke C,Hartmann JU, et al. NF-kB/STAT5/miR-155 network targets PU.1 inFLT3-ITD-driven acute myeloid leukemia. Leukemia 2015;29:535–47.

44. Donninger H, VosMD, Clark GJ. The RASSF1A tumor suppressor. J Cell Sci2007;120:3163–72.

45. Griffiths EA, Gore SD, Hooker C, McDevitt MA, Karp JE, SmithBDet al. Acute myeloid leukemia is characterized by Wnt pathwayinhibitor promoter hypermethylation. Leuk Lymphoma 2010;51:1711–9.

46. Meng F, Glaser SS, Francis H, DeMorrow S, Han Y, Passarini JD, et al.Functional analysis of microRNAs in human hepatocellular cancer stemcells. J Cell Mol Med 2012;16:160–73.

47. Zare-Abdollahi D, Safari S, Movafagh A, Ghadiani M, Riazi-Isfahani S,Omrani MD. Intact expression status of RASSF1A in acute myeloid leu-kemia. Med Oncol 2014;31:770.

48. Yi M, Yang J, Chen X, Li J, Li X, Wang L, et al. RASSF1A suppressesmelanoma development by modulating apoptosis and cell-cycle progres-sion. J Cell Physiol 2011;226:2360–9.

49. Wang J, Barsky LW, Shum CH, Jong A, Weinberg KI, Collins SJ, et al.Retinoid induced G1 arrest and differentiation activation are associatedwith a switch to cyclin-dependent kinase-activating kinase hypopho-sphorylation of retinoic acid receptor alpha. J Biol Chem 2002;277:43369–76.

50. Liu Y, Hock JM, Sullivan C, FangG, Cox AJ, Davis KT, et al. Activation of thep38 MAPK/Akt/ERK1/2 signal pathways is required for the protein stabi-lization of CDC6 and cyclin D1 in low-dose arsenite-induced cell prolif-eration. Cell Biochem 2010;111:1546–55.

Cancer Res; 75(16) August 15, 2015 Cancer Research3424

Br€auer-Hartmann et al.

on July 18, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 3, 2015; DOI: 10.1158/0008-5472.CAN-14-3521

2015;75:3411-3424. Published OnlineFirst June 3, 2015.Cancer Res   Daniela Bräuer-Hartmann, Jens-Uwe Hartmann, Alexander Arthur Wurm, et al.   Suppressor RASSF1A in Acute Promyelocytic Leukemia

-Regulated miR-181a/b Cluster Targets the TumorαPML/RAR

  Updated version

  10.1158/0008-5472.CAN-14-3521doi:

Access the most recent version of this article at:

  Material

Supplementary

  http://cancerres.aacrjournals.org/content/suppl/2015/06/03/0008-5472.CAN-14-3521.DC1

Access the most recent supplemental material at:

   

   

  Cited articles

  http://cancerres.aacrjournals.org/content/75/16/3411.full#ref-list-1

This article cites 50 articles, 24 of which you can access for free at:

  Citing articles

  http://cancerres.aacrjournals.org/content/75/16/3411.full#related-urls

This article has been cited by 2 HighWire-hosted articles. Access the articles at:

   

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

  Subscriptions

Reprints and

  .pubs@aacr.org

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

  Permissions

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

.http://cancerres.aacrjournals.org/content/75/16/3411To request permission to re-use all or part of this article, use this link

on July 18, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 3, 2015; DOI: 10.1158/0008-5472.CAN-14-3521