runx2 suppression by mir-342 and mir-363 inhibits multiple ... · oncogenes and tumor suppressors...

Oncogenes and Tumor Suppressors

Runx2 Suppression by miR-342 and miR-363Inhibits Multiple Myeloma ProgressionPramod S. Gowda1, Benjamin J.Wildman2, Timothy N. Trotter1, Xiaoxuan Xu1,Xiaoxiao Hao1, Mohammad Q. Hassan2, and Yang Yang1,3

Abstract

In multiple myeloma, abnormal plasma cells accumulate andproliferate in the bonemarrow. Recently, we observed that Runx2,abone-specific transcription factor, is highly expressed inmultiplemyeloma cells and is a major driver of multiple myeloma pro-gression in bone. The primary goal of the present study was toidentify Runx2-targeting miRNAs that can reduce tumor growth.Expression analysis of a panel of miRNAs in multiple myelomapatient specimens, compared with healthy control specimens,revealed thatmetastaticmultiplemyeloma cells express low levelsofmiR-342 andmiR-363 but high levels of Runx2. Reconstitutingmultiple myeloma cells (CAG) with miR-342 and miR-363reduced the abundance of Runx2 and the expression of metasta-sis-promoting Runx2 target genes RANKL and DKK1, and sup-pressed Runx2 downstream signaling pathways Akt/b-catenin/survivin, which are required for multiple myeloma tumor pro-gression. Intravenous injection of multiple myeloma cells

(5TGM1), stably overexpressing miR-342 and miR-363 alone ortogether, into syngeneic C57Bl/KaLwRij mice resulted in a signif-icant suppression of 5TGM1 cell growth, decreased osteoclastsand increased osteoblasts, and increased antitumor immunity inthe bonemarrow, compared withmice injected with 5TGM1 cellsexpressing a miR-Scramble control. In summary, these resultsdemonstrate that enhanced expression of miR-342 and miR-363in multiple myeloma cells inhibits Runx2 expression and multi-ple myeloma growth, decreases osteolysis, and enhances antitu-mor immunity. Thus, restoring the function of Runx2-targeting bymiR-342 and miR-363 in multiple myeloma cells may afford atherapeutic benefit by preventingmultiple myeloma progression.

Implications: miR-342 and miR-363–mediated downregulationof Runx2 expression inmultiple myeloma cells prevents multiplemyeloma progression. Mol Cancer Res; 16(7); 1138–48. �2018 AACR.

IntroductionMultiple myeloma is a hematologic malignancy characterized

by high infiltration and accumulation of malignant plasma cellsin the bone marrow (1–3). Bone disease occurs in approximately90% of patients with multiple myeloma and is the main cause ofpatient mortality (4). The consequences of multiple myelomaprogression in bone can be devastating for patients and includeosteolytic bone lesions, hypercalcemia, renal insufficiency, andspinal cord and nerve-compression syndromes (5). However, theaggressive mechanisms driving multiple myeloma progression inbone remain unclear. Runx2 is a bone-specific transcription factorthat promotes osteoblastogenesis and bone formation (6, 7).Runx2 is also expressed in many cancer cells, including breastand prostate cancers, and has been shown to promote bonemetastasis (8, 9). Recently, our studies have demonstrated that

Runx2 expression is significantly higher in primary multiplemyeloma cells than in the plasma cells of bone marrow frompatients with monoclonal gammopathy of undetermined signif-icance or in normal plasma cells from healthy bone marrowdonors (10). Inhibition of Runx2 expression in multiple myelo-ma cell lines reduces tumor growth and prevents multiple mye-loma dissemination to bone, thus providing a basis for Runx2 as apotential therapeutic target against multiple myeloma progres-sion and dissemination (10). Unfortunately, transcription factorsare often considered to be non-druggable, as targeting theseintracellular proteins lacking enzymatic activity remains challeng-ing (11, 12). Nevertheless, a potential novel avenue to targetnuclear factors might be provided by miRNA technology.

MiRNAs are a class of small, non-coding RNAs that function bybinding to the 30 untranslated region (30UTR) of target mRNAsand repressing mRNA expression by interfering with the mRNAstability and/or by blocking protein translation (13). MiRNAshave critical roles in processes such as cell proliferation, differ-entiation, and survival and also during normal development,homeostasis, and disease (14, 15). Aberrant miRNA expressionis frequently observed in various human tumors, including mul-tiple myeloma, indicative of critical roles in tumorigenesis (16,17). Recently, it was demonstrated that miRNAs targeting oste-oclast function can inhibit bone metastatic disease (18, 19). Inaddition, in vivo delivery of miRNAs or miRNA antagonists canprevent cancer-induced bone diseases (20). Some reports indicatethat miRNAs can inhibit the growth of various tumors by directlysuppressing Runx2 function (21). However, miRNA-mediatedregulation of Runx2 in the context of multiple myeloma

1Department of Pathology, University of Alabama at Birmingham, Birmingham,Alabama. 2Department of Oral and Maxillofacial Surgery, School of Dentistry,University of Alabama at Birmingham, Birmingham, Alabama. 3ComprehensiveCancer Center, University of Alabama at Birmingham, Birmingham, Alabama.

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

Corresponding Author: Yang Yang, University of Alabama at Birmingham, WTI320A, 1824 6th Avenue South, Birmingham, AL 35294. Phone: 205-996-6228;Fax: 205-975-6615; E-mail: [email protected]

doi: 10.1158/1541-7786.MCR-17-0606

�2018 American Association for Cancer Research.

MolecularCancerResearch

Mol Cancer Res; 16(7) July 20181138

on September 11, 2020. © 2018 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst March 28, 2018; DOI: 10.1158/1541-7786.MCR-17-0606

http://crossmark.crossref.org/dialog/?doi=10.1158/1541-7786.MCR-17-0606&domain=pdf&date_stamp=2018-6-21

http://mcr.aacrjournals.org/

progression has never been studied. Thus, miRNAs could be usedas therapeutic agents, and identifying specific miRNAs that sup-press Runx2 would provide a novel approach for the treatment ofmultiple myeloma.

In this study, we show that reduced expression of miR-342 andmiR-363 is responsible for the upregulation of Runx2 in multiplemyelomaprogression.Wedemonstrate the effects ofmiR-342 andmiR-363 on multiple myeloma inhibition through direct down-regulation of Runx2 using in vitro and in vivo approaches. Impor-tantly, our data provides strong evidence that targeting Runx2 byan miRNA-based approach can be used to suppress multiplemyeloma progression.

Materials and MethodsCell culture

The mouse multiple myeloma 5TGM1 cell line expressingluciferase (Luc) was a kind gift from Dr. Fenghuang Zhan(University of Iowa Health Care, Iowa City, Iowa), the humanmultiple myeloma cell line CAG was from Dr. Sanderson'slaboratory, and the human multiple myeloma cell line U266was purchased from the ATCC. We have stably transfectedRunx2 shRNA into 5TGM1 cells and the methods were pub-lished previously (10). Cells were cultured in RPMI-1640 with10% FBS, 2 mmol/L L-glutamine, 1 U/mL penicillin, and10 mg/mL streptomycin at 37�C and 5% CO2. Cells wereestablished as free of Mycoplasma and bacteria followinginstructions of the ATCC. Cells were thawed from liquidnitrogen and media changed after 24 hours, followed byculturing for 48 to 96 hours before the cells were used foractual experiments. Cells were not passaged more than 2 timesafter initial thawing. Human multiple myeloma cells CAG andU266 and mouse multiple myeloma cell line 5TGM1 wereauthenticated routinely and before the in vivo experiments bymeasuring the levels of kappa, lambda and IgG2bk (solublemarkers of CAG, U266, and 5TGM1 cells, respectively) in theconditioned medium of cell culture by ELISA and the levels ofCD138 expression (surface marker of multiple myeloma cell)by flow cytometry.

Transfection and transductionDouble-stranded RNA oligos representing mature sequences

thatmimic endogenousmiR-342 andmiR-363 and a non-specific(NS) miRNA negative control (obtained from IDT, Coralville,Iowa) were electrotransfected at concentrations of 50 or 100 nminto CAG human myeloma cells using program EO-117 on the4D-Nucleofector system and the Amaxa SF cell line 4D-Nucleo-fector X Kit (Lonza). Cells were harvested 48 hours after trans-fection for protein and mRNA analysis. For miRNA lentiviralclone transfection, miRNA Lenti-miR vectors (System Bios-ciences) were used to produce vectors encoding miR-Scramblecontrol or mature miR-342, miR-363, or miR-342þ363 undercontrol of the CMV promoter with a GFP reporter to allowmonitoring ofmiRNA-expressing cells. The viruses were packagedin HEK293T cells (ATCC; ref. 22). 5TGM1 cells were infected at70% confluency for 48 hours with lentiviral particles and poly-brene (8 mg/mL). A second infection was repeated for another 48hours. Cells were selected by culturing for 14 days in puromycin(8 mg/mL). Protein and mRNA levels were determined in controland miRNA-overexpressing cells by Western blotting and reversetranscriptase quantitative PCR (RT-qPCR).

RNA isolation and real-time RT-qPCRTotal RNA was isolated using RNeasy Mini kits (Qiagen Inc.)

according to the manufacturer's specifications. cDNA was synthe-sized using MMLV reverse transcriptase (Clontech). For miRNAdetection, poly(A) tailing was performed using a poly(A) poly-merase (Pol) and reverse transcriptionwas carriedoutwith reversetranscriptase (System Biosciences) according to the manufac-turer's instructions. Gene expression was determined by real-timeRT-qPCR using Fast SYBRGreenMaster Mix (Applied Biosystems,Inc.) and gene-specific primers in an ABI StepOnePlus fast ther-mocycler. For each gene, expression levels were normalized to 28Sribosomal RNA (rRNA) or U6 RNA expression. Experiments wereperformed in triplicate, and results are given as mean values �SEM. Nucleotide sequences of primers are provided in Supple-mentary Table S1.

Western blot analysisEqual amounts of protein (50 mg) were subjected to sodium

dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)using 4% to 12% gradient gels (Bio-Rad) and transferred tonitrocellulose membranes (Schleicher and Schuell). Transferredproteins were probed with appropriate antibodies (Supplemen-tary Table S2) and detected using enhanced chemiluminescencereagents (Amersham Biosciences). Band intensity was quantifiedby NIH ImageJ software version 1.45 (rsb.info.nih.gov/ij).

Gene-expression profilesGene-expression profile (ID: GSE17498; ref. 23) was down-

loaded from Gene Expression Omnibus (GEO, http://www.ncbi.nlm.nih.gov/geo). GSE17498, based on the platform of GPL8227Agilent-019118 HumanmiRNAMicroarray 2.0 G4470B (miRNAID version), included a collection of 43 (normal: 3 and multiplemyeloma patients: 40) samples of purified plasma cells frombone marrow specimens for which miRNA profiles werecompared.

Cell proliferation assayThe number of viable cells after miRNA overexpression was

evaluated by a colorimetricMTT assay. CAGand5TGM1 cells (2�104 cells/well) were seeded separately in 96-well plates in tripli-cate. After 72 hours, cells were treated with 0.5 mg/mL MTT for 2hours. The optical density (OD) was then measured at 540 nm.Themeanvalue and standarddeviation (SD)were calculated fromtriplicate experiments.

Cell migration assayCAG cells were transfected with control miRNA or miR-342 or

miR-363mimics as described above; 24 hours later, the cells werecounted and resuspended inRPMI-1640 serum-freemedium.Cellmigration assays were conducted in 24-well, 15-mm internal-diameter multiwell notched tissue culture plates (Corning Inc.).Cells (2 � 105/500 mL) in RPMI-1640 serum-free medium wereadded into inserts (8-mm pore size), in triplicate for each group,and allowed to migrate toward complete medium (RPMI-1640with 10% FBS) in the bottom wells at 37�C and 5% CO2 for 24hours. Cellsmigrated into bottomwells were enumerated after 24hours, in triplicate, using a Z1 Dual Threshold Coulter Counter(Beckman Coulter).

Mouse modelsC57BL/KalwRij mice were purchased from Harlan Laborato-

ries, Inc. The 5TGM1 model is a syngeneic model of murine

miRs Target Runx2 and Limit Multiple Myeloma

www.aacrjournals.org Mol Cancer Res; 16(7) July 2018 1139



http://www.ncbi.nlm.nih.gov/geo

http://www.ncbi.nlm.nih.gov/geo


myeloma that faithfully replicates the human myeloma disease(24–26). 5TGM1-luc cells overexpressing miR-Scramble control,miR-342-, miR-363-, or miR-342þ363 were intravenouslyinjected into 6-week-old C57BL/KalwRij mice (males andfemales) via the tail vein. Serum was collected biweekly formeasurement of IgG2bk levels (a solublemarker of 5TGM1 cells),and bioluminescent imaging was performed weekly to monitortumor homing and growth. At the end of the experiment, femursand tibias were harvested, fixed, decalcified, paraffin-embedded,sectioned, and stainedwith hematoxylin and eosin (H&E) orwitha tartrate-resistant acid phosphatase (TRAP) staining kit. Allanimal studies were performed in accordance with UAB and NIHguidelines after institutional review and approval.

TRAP stainingTRAP staining was used to detect osteoclasts in bone sections

frommice intravenously injectedwith tumor cells. Paraffin-embed-ded murine bone sections were deparaffinized in xylene and thenrehydrated in a descending alcohol series (100%, 90%, 70%, and50% alcohol). Sections were then washed in distilled water. Bonesections were stained using a TRAP staining kit (1). Osteoclastnumber was determined by counting TRAP-positive osteoclastsadjacent to the bone surface as described previously (27).

ELISAThe levels of serum IgG2bk (a soluble marker of 5TGM1 cells)

inmiceweremeasured using amouse IgG2bk ELISAQuantitationkit (Bethyl Laboratories) according to the manufacturer's instruc-tions. All steps were performed at room temperature and thereaction productwas quantifiedusing a spectrophotometer at 450nm, with each sample measured in duplicate.

Flow cytometryBone marrow was flushed out from tibia and femurs of each

mouse in 1mL of PBS, followed by centrifugation of PBS to pelletbone marrow cells. Red blood cells from the harvested bonemarrow cells were lysed using ACK lysis buffer (Lonza), and bonemarrow cells without red blood cells were then stained for 30minutes at 4�C in the dark with anti-mouse B220-PECy7, IgM-FITC, andCD23-PE antibodies to detect regulatory B cells (Bregs);anti-mouse CD3-FITC, CD4-PerCP Cy5.5, CD127-BV510, andCD25-BV650 antibodies to detect regulatory T cells (Tregs); anti-mouse CD11c-AF700, CD11b-FITC, and B220-AF647 antibodiesto detect plasmacytoid dendritic cells (pDCs); anti-mouse CD3-FITC,CD8-BV605, andPD-1-BV421 todetect CD8þ specific PD-1;anti-mouse CD3-FITC, CD8-BV605, and Tim-3-PE to detectCD8þ specific Tim-3; and anti-mouse CD3-FITC and IFN-g-APCto detect CD3þ IFN-g cells. Cells were then washed twice withFACS buffer (PBS with 1% BSA and 0.1% sodium azide), resus-pended in 1X PBS, and analyzed by LSRII (BD Biosciences) andFlowJo software.

30UTR–reporter constructs and transfectionThe Runx2 30UTR (80–100 bp) fragment containing sites for

miR-342 and miR-363 binding (seed sequences) was amplifiedusing the In-fusion HD Cloning Kit (Clontech) with overhanging50 and 30 primers homologous to the target, pMIR REPORT. DNAfragmentswere inserted into the SpeI site of thepMIRREPORT-LucmRNA expression plasmid (Cat # AM5795, Thermo Fisher Sci-entific) using the In-fusion HD Cloning Kit manufacturer's pro-tocol to generate the Runx2 30UTR-Luc reporter plasmid. Also,

mutations were introduced into miR-342 and miR-363 bindingsites in the Runx2 30UTR and sub-cloned as described above intopMIR-REPORT-Luc plasmids to generate Runx2 30UTR-Lucreporter plasmids with mutated miR-342 and miR-363–bindingsites. Wild-type (WT) and mutant (MT) sequences of each 30UTRare shown in Supplementary Table S3. Luc assays were performedwith HEK293 cells co-transfected with the specified miRNAmimics (100 nmol/L each) or NS control miRNA (100 nmol/L)and the specified Runx2 30UTR-Luc reporter (200 ng) plasmidusing 6 mL of Fugene transfection reagent (Promega) in triplicatein six independent experiments. Transfection with Renilla Lucplasmid (Promega) was used to normalize the relative Luc values.Relative Luc activity (firefly Luc activity/Renilla Luc activity) wasexpressed in relative luminescence units and plotted.

ResultsmiR-342 and miR-363 expression negatively correlates withRunx2 expression in multiple myeloma cells

To seek the miRNAs that inhibit Runx2 expression in multiplemyeloma cells, a panel of miRNAs that were differentiallyexpressed between plasma cells from bone marrow specimensof patients with multiple myeloma and those of normal donorswas examined by downloading the miRNA microarray datasetGSE17498 from the Gene Expression Omnibus database. TwomiRNAs, miR-342 and miR-363, were found significantly down-regulated in multiple myeloma patient samples compared withnormal donor samples (Fig. 1A). Targetscan analysis revealedbinding sites for miR-342 and miR-363 at the 30UTR region ofRunx2 (Fig. 1B), suggesting Runx2 regulatory roles for miR-342andmiR-363. This inverse correlation betweenmiR-342/miR-363and Runx2 expression was also confirmed in CAG and U266human multiple myeloma cell lines (Fig. 1C and D).

miR-342 andmiR-363mimicsmarkedly suppress Runx2, alongwith cell proliferation and migration, in CAG multiplemyeloma cells

On the basis of the relatively high abundance of miR-342 andmiR-363 in the plasma cells of normal donors, and the strongdownregulation of these miRNAs in the plasma cells of patientswith multiple myeloma and in multiple myeloma cell lines (Fig.1A and D), we hypothesized that reconstitution of miR-342 andmiR-363 in multiple myeloma cells could reverse Runx2-enhancedmultiple myeloma progression. To test this hypothesis,we transiently transfected miR-342 and miR-363 mimics intoCAG multiple myeloma cells, both individually and in combi-nation, and examined the levels of Runx2 expression by RT-qPCRandWestern blot. The results showed a decrease in Runx2 expres-sion upon miRNA mimic transfection (Fig. 2A and B). Interest-ingly, the miR-342þ363 mimic was more effective than eitherindividualmimicwas in suppressing Runx2 expression.However,despite the strikingly decreased expression of Runx2 protein incells co-expressing miR-342 and miR-363 mimics, Runx2 mRNAexpression decreased by less than two-fold compared with that incontrols, suggesting that Runx2 protein is regulated through bothtranscription-dependent and alternate mechanisms, such as con-trol of protein translation.

Furthermore, to examine the specificity and efficacy ofmiR-342and miR-363 on Runx2 activity, we cloned an 80- to 100-bpfragment of the Runx2 30UTR containing binding sites for miR-342 and miR-363 into the 30UTR of a Luc reporter gene to create

Gowda et al.

Mol Cancer Res; 16(7) July 2018 Molecular Cancer Research1140




the Runx2 30UTR-Luc plasmid. Also, we mutated the miR-342–and miR-363–binding sites in the Runx2 30UTR and cloned thesequence into the 30UTRof the Luc reporter gene to create a Runx230UTR-Luc plasmid with mutated binding sites for miR-342 andmiR-363. InHEK293 cells, ectopic expression ofmiR-342 ormiR-363 mimics substantially repressed Luc activity when co-trans-fected with the WT Runx2 30UTR-Luc plasmid but not when co-transfected with the mutant Runx2 30UTR-Luc plasmid (Fig. 2CandD), suggesting thatmiR-342 andmiR-363 bind to and inhibitthe activity of the Runx2.

We then assessed the multiple myeloma cell proliferation andmigration abilities after transfection of miR-342 and miR-363mimics. MTT and Transwell migration assays showed miR-342,miR-363, and miR-342þ363 mimics markedly suppress bothproliferation andmigration of CAG cells (Fig. 2E and F). We thenassessed the expression of DKK1 and RANKL, which promotecancer cell migration and cause osteolytic bone lesions (28, 29),and founddecreased expression ofDKK1 andRANKL inCAG cellstreated with miR-342, miR-363, or miR-342þ363 mimics (Sup-plementary Fig. S1). Together, these data reveal that miR-342 andmiR-363 suppress Runx2 expression and function, the prolifera-tion of CAG cells, and the expression of molecules that promotecancer cell migration and thereby inhibit migration of CAG cells.Our results consistently demonstrate that co-transfection of miR-342 and miR-363 mimics had a more robust effect than theindividual miRNA mimics had.

miR-342 and miR-363 downregulate Akt/b-catenin/survivinsignaling in multiple myeloma cells

Wehave previously demonstrated thatmultiplemyeloma cell–derived Runx2 promotes tumor progression through upregula-tion of Akt/b-catenin/survivin signaling (10). Therefore, we exam-ined whether miR-342 and miR-363 downregulate these Runx2downstream signaling pathways inCAG cells.Western blot resultsshowed suppressed expressionof phosphorylated Akt (p-Akt) andsurvivin upon miR-342 and miR-363 mimic transfection (Fig.3A). Real-time PCR results showed b-catenin and GSK-3b to bedownregulated uponmiR-342þ363mimic transfection (Fig. 3B).These results demonstrate the effectiveness of targeting the abnor-mally and highly expressed transcription factor Runx2 by miR-342andmiR-363mimics. In combination,miR-342 andmiR-363had a synergistic effect on downregulating Runx2 and its targetgenes and signaling, which was likely the cause of reducedmultiple myeloma proliferation and migration.

Overexpression of miR-342 and miR-363 in 5TGM1 multiplemyeloma cells inhibits multiple myeloma cell growth in vitroand in vivo

To investigate the effects ofmiR-342 andmiR-363 in vitro and invivo, mouse multiple myeloma 5TGM1 cells were stably trans-duced with miR-Scramble control, miR-342, miR-363, or miR-342þ363 (Supplementary Fig. S2A). RT-qPCR showed miR-342,miR-363, andmiR-342þ363 overexpression in 5TGM1 cells, and

Figure 1.

Runx2 expression in patientswithmultiplemyelomaandmultiplemyeloma cell lines inversely correlateswith Runx2-targetingmiRNAs.A,Expression ofmiR-342 andmiR-363 is significantly downregulated in plasma cells of bone marrow specimens from patients with multiple myeloma. The data were derived from threenormal donors and 40 multiple myeloma patients at diagnosis [data provided by Lionetti and colleagues (ref. accession number GSE17498 to GEO datasets)]. Thebars and box represent the Tukey distribution from each group. B, TargetScan analysis revealed the binding sites of miR-342 and miR-363 on the Runx2 30UTR.Complementary miRNA seed sequence binding sites are shown in bold black (hsa: human; mmu: mouse). C, Relative expression of Runx2 in human multiplemyeloma cell lines U266 and CAGwasmeasured using RT-qPCR (left) andWestern blotting (right).D,Relative expression ofmiR-342 andmiR-363 in U266 and CAGcells as measured by RT-qPCR. The P values were obtained with two-tailed Student t test (�� , P < 0.005; �� , P < 0.0001).






Western blotting and RT-qPCR confirmed that Runx2 expressionin 5TGM1 cells was significantly reduced aftermiR-342,miR-363,or miR-342þmiR-363 transduction, compared with miR-Scram-ble control cells (Supplementary Fig. S2B and S2C). We alsoassessed 5TGM1 cell proliferation after stable transduction. MTTassays showed that overexpression of miR-342, miR-363, or miR-342þ363 markedly suppresses proliferation of 5TGM1 cells invitro (Supplementary Fig. S2D). Stably transduced cells wereintravenously injected into C57BL/KaLwRij mice via the tail vein(106 cells per mouse, n ¼ 10 mice/group). Bioluminescent imag-ing showed that mice injected with 5TGM1-miR-342 or 5TGM1-miR-363 cells had much smaller tumors in bone than miceinjected with 5TGM1-miR-Scramble control cells had, and mice

injected with 5TGM1-miR-342þ363 cells had the least tumorgrowth (Fig. 4A). Serum IgG2bk levels (a reliable marker for totaltumor burden) confirmed the bioluminescent imaging results(Fig. 4B). Bioluminescent imaging and ELISAs were performed 4weeks after multiple myeloma cell injection. In a separate 5TGM1intravenous injection experiment (n¼ 4mice/group), the survivalrates of mice bearing 5TGM1-miR-342, -miR-363, or -miR-342þ363 tumors were significantly higher than those of micebearing 5TGM1-Scramble control tumors (Fig. 4C). Furthermore,H&E staining demonstrated that all mice in the 5TGM1-miR-Scramble grouphad tumors in tibia/femur,whereas less than20%of the mice in the 5TGM1-miR-342, -miR-363, and -miR-342þ363 groups had detectable tumors in bone (Fig. 4D). These

Figure 2.

miR-342 and miR-363 suppress Runx2 expression and the proliferation and migration of multiple myeloma cells in vitro. A and B, Relative expression of Runx2in CAG cells transfected with NS-miR-control or miR-342, miR-363, or miR-342þ363 mimics was assessed by Western blotting (A) and RT-qPCR (B).C andD,Nonspecific (NS) miR-control or miR-342 (C) or miR-363 mimics (D) were cotransfected with eitherWT or mutant Runx2 30UTR-Luc reporter plasmids intoHEK293 cells (MT: mutant). Firefly Luc activity was normalized to cotransfected Renilla Luc activity and presented in relative luminescence units. E, Cellviability was measured by MTT assay in CAG cells transfected with miR-Scramble control or miR-342, miR-363, or miR-342þ363 mimics. F, Transwell migrationassays were conducted to assess cell migration in CAG cells transfected with miR-Scramble control or miR-342, miR-363, or miR-342þ363 mimics. The P valueswere obtained by one-way ANOVA followed by Tukey–Kramer post hoc test (ns: non-significant; � , P < 0.05; �� , P < 0.005).

Gowda et al.





results demonstrate that miRNA-mediated Runx2 knockdownsuppresses multiple myeloma cell growth in vitro and in vivo.

Overexpression of miR-342 and miR-363 in 5TGM1 cellsdecreases the number of multiple myeloma cell–inducedosteoclasts and increases the number of osteoblasts in vivo

Multiple myeloma is associated with osteolytic bone diseasemediated by increased osteoclastic bone resorption and impairedosteoblastic bone formation (30, 31). To understand the effects ofmiR-342 and miR-363 expression in multiple myeloma–relatedbone disease, the number of osteoblasts and osteoclasts wasassessed in animals injected with 5TGM1-miR-342 or 5TGM1-miR-363 cells versus 5TGM1-miR-Scramble control cells. Thenumber of osteoblasts was evaluated by immunostaining forosteocalcin (a marker for osteoblasts) and osteoclasts were visu-alized by TRAP staining as previously described (27). Enumera-tion of osteoblasts showed a trend toward increased osteoblast

numbers in mice injected with 5TGM1-miR-342 and 5TGM1-miR-363 cells compared with mice injected with 5TGM1-Scram-ble control cells, but the increase was statistically significant onlyin those injected with the 5TGM1-miR-342þ363 cells. Interest-ingly, osteoclast number was significantly decreased in miceinjected with 5TGM1 cells overexpressing any of the miRNAscompared with 5TGM1-miR-Scramble control cells (Fig. 5).Together, these data indicate that miR-342 and miR-363 expres-sion in multiple myeloma cells can impact the BM microenvi-ronment and bone destruction, possibly with a greater effect onosteoclasts than on osteoblasts.

miR-342 and miR-363 overexpression enhances antitumorimmunity in vivo

Cancer cells, including multiple myeloma cells, actively sup-press the immune system and recruit immunosuppressive cells toevade immune detection and killing (32, 33). To determine

Figure 3.

Akt/b-catenin/survivin signalingpathways downregulated by miR-342and miR-363. A, RepresentativeWestern blot showing the relativeexpression of Runx2, survivin, and p-Akt in CAG cells transfected with miR-Scramble control, miR-342, miR-363,ormiR-342þ363 (left). Survivin and p-Akt expression levels were quantifiedand normalized to b-actin levels usingImageJ software (right). B, Relativeexpression of the Wnt signalingpathway components b-catenin andGSK3b in CAG cells transfected withmiR-Scramble control or miR-342,miR-363, or miR-342þ363 mimics,measured by RT-qPCR. The P valueswere obtained by one-way ANOVAfollowed by Tukey–Kramerpost hoc test (ns: nonsignificant;� , P < 0.05; �� , P < 0.005;�� , P < 0.0001).






whether miR-342 and miR-363 overexpression in multiple mye-loma cells can enhance antitumor immunity, bone marrow cellswere harvested from hind limbs of mice injected with 5TGM1-miR-342, -miR-363, or -miR-342þ363 cells or 5TGM1-miR-Scramble control cells, stained with specific antibodies againstimmune cells, and analyzed byflow cytometry (n¼ 6mice in eachof the four groups). Flow-cytometry analysis revealed a significantdecrease in immunosuppressive Tregs and Bregs, and an increasein antigen-presenting pDCs in the bone marrow of mice injectedwith 5TGM1-miR-342þ363 cells compared with the bone mar-row of mice injected with 5TGM1-miR-Scramble control cells(Fig. 6A–C). Previous studies have demonstrated that blockade ofprogrammed cell death-1 (PD-1) and T-cell immunoglobulindomain and mucin domain-containing molecule-3 (Tim-3) sig-naling in CD8þ T cells enhances the cytotoxic response of CD8þ T

cells by production of effector cytokines such as IFN-g and TNF-a(34, 35). Our flow-cytometry analysis revealed that, although thenumber of cytotoxic CD3þ CD8þ T cells did not significantlydiffer, there was a significant decrease in CD8þ T cells expressingthe exhaustion markers PD-1 and Tim-3 in the bone marrow ofmice injected with 5TGM1-miR-342þ363 cells compared withmice injected with 5TGM1-miR-Scramble control cells (Supple-mentary Fig. S3A–S3C). Also, there was a significant increase inexpression of IFN-g fromCD3þT cells in the bonemarrowofmiceinjected with 5TGM1 miR-342þ363 cells compared with miceinjectedwith 5TGM1-miR-Scramble control cells (SupplementaryFig. S3D). Finally, we assessed the expression of PD-L1 inmultiplemyeloma cells after Runx2 knockdown by shRNA as well as in5TGM1-miR-342, -miR-363, or –miR-342þ363 cells. PD-L1expression was suppressed upon Runx2 knockdown by shRNA

Figure 4.

miRNA-mediated Runx2 knockdown impairs multiple myeloma progression in vivo. A, Representative bioluminescent images obtained 4 weeks after intravenousinjection of 5TGM1 cells stably transfected with miR-Scramble control or miR-342, miR-363, or miR-342þ363 mimics (n ¼ 10 mice/group). B, Serum IgG2bkconcentration measured by ELISA 4 weeks after intravenous injection of 5TGM1-miR-Scramble control, -miR-342, -miR-363-, or -miR-342þ363 cells(n¼ 10 mice/group). Significant differences between groups are indicated by P value (� , P < 0.05; ��, P < 0.005; �� , P < 0.0001). C, Kaplan–Meier survival curves inmice injected intravenously with 5TGM1-miR-Scramble control, -miR-342, -miR-363, or -miR-342þ363 cells. Survival curves were significantly different(�� , P < 0.005), according to the log-rank (Mantel–Cox) test. D, Representative micrographs of H&E-stained bone sections from mice injected intravenously with5TGM1-miR-Scramble control, -miR-342, -miR-363, or -miR-342þ363 cells. Tumors were present, with abundant myeloma cells, in mice injected with 5TGM1-miR-Scramble control cells, whereas very few tumors and myeloma cells were found in mice injected with 5TGM1-miR-342, -miR-363, or -miR-342þ363 cells.

Gowda et al.





or miRNA (Supplementary Fig. S4A–S4C). These results areconsistent with our hypothesis that multiple myeloma cells over-expressing Runx2 also upregulate PD-L1 and are responsible forCD8þ T-cell exhaustion by ligand–receptor interactions betweenPD-L1 and PD-1. Together, these results suggest that overexpres-sing miR-342 and miR-363 in 5TGM1 cells can enhance antitu-mor immunity in the bone marrow microenvironment.

DiscussionRunx2, as a transcription factor, is critical for regulating genes

that support bone formation (36, 37). However, recent studieshave demonstrated that Runx2 is aberrantly expressed in severalcancer types, including multiple myeloma (38, 39). Our earlier invivo data demonstrated that enhanced expression of Runx2 inmultiple myeloma cells is associated with a more aggressivephenotype of multiple myeloma, inhibiting that multiple mye-loma-Runx2 significantly decreases the progression of multiplemyeloma, and high Runx2 expression in multiple myeloma cellsis associatedwith a high-risk population of patients withmultiplemyeloma (10). These novel findings point to Runx2 as a newtarget for multiple myeloma therapy.

It is well known that miRNAs are deregulated in numerousdiseases and cancer (16, 17). Herein, we identified miR-342 andmiR-363 as suppressors of multiple myeloma progressionthrough direct targeting of Runx2 in multiple myeloma cells.Expression analysis revealeddecreased expressionofmiR-342 andmiR-363 in multiple myeloma cells that have high levels ofRunx2, and TargetScan analysis revealed multiple binding sitesofmiR-342 andmiR-363 at the 30UTR region of Runx2, indicatingthat miR-342 and miR-363 are negative regulators of Runx2.Indeed, upon reconstitution ofmiR-342 andmiR-363 inmultiplemyeloma cells, Runx2 expression was suppressed and, moreimportantly, the growth of multiple myeloma cells was impairedin vitro and in vivo. These data strongly suggest that miRNA

delivery, namely delivery of miR-342 and miR-363, is a viabletherapeutic strategy to inhibit progression and dissemination ofmultiple myeloma.

Multiple myeloma is characterized by the development ofprogressive and destructive osteolytic bone disease, yet despiteadvances in treatment strategies, bone destruction is amajor causeof morbidity in multiple myeloma patients (4, 40). Osteolyticbone disease is driven by increased osteoclast activity anddecreased osteoblast activity, which eventually destroys the bone.Current therapies to treat osteolytic bone disease are based onantiresorptives, which inhibit osteoclast activity via antibody-based blockade of RANKL (e.g., denosumab) or by inducingosteoclast apoptosis using bisphosphonates. Recent studies havedemonstrated that miRNA-mediated targeting of osteoclasts canbe used to treat osteolytic bone disease (18). RANKL is a keystimulator of osteoclast differentiation (41). Our prior workestablished that Runx2 in multiple myeloma cells enhances theexpression and secretion of RANKL by multiple myeloma cellsand promotes bone resorption (10). In the present study, wedemonstrate that the expression of RANKL in CAG multiplemyeloma cells is dramatically decreased by the induction ofmiR-342 and miR-363. Furthermore, our studies in mice bearingmiR-342-, miR-363-, or miR-Scramble–overexpressing multiplemyeloma tumors indicate that regulation of Runx2 by miR-342and miR-363 ultimately affects the bone marrow microenviron-ment by increasing osteoblast and decreasing osteoclast numbersin vivo. In addition to decreased RANKL production frommultiplemyeloma cells overexpressing miR-342 and miR-363, reducedtumor growth by miR-342 and miR-363 may contribute to thereduced bone resorption in thesemice. Moreover, increases in thenumbers of immunosuppressive Tregs andBregs and a decrease inthe number of pDCs in the tumor microenvironment predictsworse survival for patientswith various types of cancer (42, 43). Inour study, compared with bone marrow from mice injected with5TGM1-miR-Scramble cells, bone marrow from mice injected

Figure 5.

Overexpression of miR-342or miR-363 in 5TGM1 cells suppressesosteoclast formation and enhancesosteoblastogenesis in a C57BL/KalwRij mouse model system.Immunostaining for osteocalcin(osteoblasts) and TRAP staining forosteoclasts were performed on bonesharvested from C57BL/KalwRij miceinjected with 5TGM1-miR-Scramblecontrol, -miR-342, -miR-363, or-miR-342þ363 cells. Representativemicrographs are shown. Arrowsindicate osteoblasts and osteoclasts inthe respective panels. Osteocalcin-positive and TRAP-positive cells werephotographed and counted (bottom).The P values were obtained by one-way ANOVA followed by Tukey–Kramer post hoc test (Scr: Scramble;ns: nonsignificant; � , P < 0.05;�� , P < 0.005).






with 5TGM1 cells overexpressing miR-342 and miR-363 had alower percentage of Tregs, Bregs, and PD-1þ/Tim-3þ CD8þ cells,along with a higher percentage of pDCs and CD3þ IFN-g cells.This improved immune environment in bone marrow is anotherlikely reason for the decreased tumor growth inmice bearingmiR-342- and miR-363–overexpressing tumors.

Together, our studies open new avenues for the development ofmiRNA-based therapeutic strategies in patients with multiplemyeloma. For example, miRNA replacement therapy aimed atrestoring the levels of repressed miRNA is currently being devel-oped with miR-34 and has entered phase I clinical trials (44). Inaddition, formulated miRNA mimics are distinct from molecu-larly targeted drugs becausemiRNAs target a broad range of genes,rather than an individual gene product. Thus, our study demon-strates an opportunity to block multiple signaling pathways thatare crucial for multiple myeloma progression and dissemination,in addition to inhibiting Runx2. As therapeutics, miR-342 and

miR-363 are also clinically advantageous as they are less likely toinduce adverse effects, since most normal cells already expressthese miRNAs. Improved delivery strategies that target multiplemyeloma cells directly, such as using multiple myeloma-specificphage fusionproteins as targeting ligands for liposomal-mediateddelivery of miRNAs (45), will greatly enhance targeted deliveryandoffer promise formultiplemyeloma therapymoving forward.

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

Authors' ContributionsConception and design: P.S. Gowda, M.Q. Hassan, Y. YangDevelopment of methodology: P.S. Gowda, B.J. Wildman, T.N. Trotter, X. Xu,X. Hao, M.Q. Hassan, Y. YangAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): P.S. Gowda, B.J. Wildman, T.N. Trotter, X. Xu, Y. Yang

Figure 6.

Immune cell profile in bone marrow of C57BL/KalwRij mice injected with 5TGM1-miR-Scramble control, -miR-342, -miR-363, or -miR-342þ363 cells. Total bonemarrow cells were analyzed by flow cytometry. Numbers shown in each plot represent relative percentages of cells within the indicated gate. Total B cells,total T cells, and pDCs represent the respective population in all live cells. Bregs are shown as a population in total B cells, and Tregs as a population of CD4þ cells.A,left, Breg cells (IgMþ CD23þ) displayed in the gated area for each mouse group; A, right, quantification of the percentage of total B-lymphocytes (B220þ) and Bregcells in each group. B, left, Treg cells (CD4þ CD25hi CD127lo) displayed in the gated area for each group; B, right, quantification of the percentage of totalT-lymphocytes (CD3þ) and Treg cells in each group. C, left, pDCs (B220þ Cd11cþ) displayed in the gated area; C, right, quantification of the percentage ofpDCs in each group. Percentages are reported as the mean� SEM. The P values were obtained by one-way ANOVA followed by Tukey–Kramer post hoc test (Scr:Scramble; ns: non-significant; � , P < 0.05).

Gowda et al.





Analysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): P.S. Gowda, T.N. Trotter, X. Hao, M.Q. Hassan,Y. YangWriting, review, and/or revision of themanuscript: P.S. Gowda, B.J. Wildman,T.N. Trotter, M.Q. Hassan, Y. YangAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): M.Q. Hassan, Y. YangStudy supervision: M.Q. Hassan, Y. Yang

AcknowledgmentsThis work was supported by NIH grant R01CA151538, an International

Myeloma Foundation Senior Award, American Society of Hematology (ASH)Bridge Grant Award, and NIH Cancer Center Support GrantP30 CA13148(to Y. Yang).

The authors thank Dr. Ralph D. Sanderson (UAB, AL) for CAG multiplemyeloma cells, Dr. Fenghuang Zhan (UI, IA) for 5TGM1-luc multiple myelomacells. We also thank the UAB Animal Imaging Core for assistance with mousebioluminescence imaging, and the UAB Histomorphometry and MolecularAnalysis Core for tissue processing.

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 October 17, 2017; revised February 19, 2018; accepted March 16,2018; published first March 28, 2018.

References1. Yang Y, Macleod V, Bendre M, Huang Y, Theus AM, Miao HQ, et al.

Heparanase promotes the spontaneous metastasis of myeloma cells tobone. Blood 2005;105:1303–9.

2. Anderson KC, Carrasco RD. Pathogenesis of myeloma. Annu Rev Pathol2011;6:249–74.

3. Barlogie B, Mitchell A, van Rhee F, Epstein J, Morgan GJ, Crowley J. Curingmyeloma at last: defining criteria and providing the evidence. Blood2014;124:3043–51.

4. Jagannath S. Pathophysiological underpinnings of multiple myelomaprogression. J Manag Care Pharm 2008;14:7–11.

5. Roodman GD. Mechanisms of bone metastasis. N Engl J Med 2004;350:1655–64.

6. Kania MA, Bonner AS, Duffy JB, Gergen JP. The Drosophila segmentationgene runt encodes a novel nuclear regulatory protein that is also expressedin the developing nervous system. Genes Dev 1990;4:1701–13.

7. Mori K, Kitazawa R, Kondo T, Maeda S, Yamaguchi A, Kitazawa S. Mod-ulation of mouse RANKL gene expression by Runx2 and PKA pathway.J Cell Biochem 2006;98:1629–44.

8. van der Deen M, Akech J, Wang T, FitzGerald TJ, Altieri DC, Languino LR,et al. The cancer-related Runx2 protein enhances cell growth and responsesto androgen and TGFbeta in prostate cancer cells. J Cell Biochem 2010;109:828–37.

9. Das K, Leong DT, Gupta A, Shen L, Putti T, Stein GS, et al. Positiveassociation between nuclear Runx2 and oestrogen-progesterone receptorgene expression characterizes a biological subtype of breast cancer. Eur JCancer 2009;45:2239–48.

10. Trotter TN, Li M, Pan Q, Peker D, Rowan PD, Li J, et al. Myeloma cell-derived Runx2 promotes myeloma progression in bone. Blood 2015;125:3598–608.

11. Hopkins AL, Groom CR. The druggable genome. Nat Rev Drug Discov2002;1:727–30.

12. Overington JP, Al-Lazikani B, Hopkins AL. How many drug targets arethere? Nat Rev Drug Discov 2006;5:993–6.

13. Wang X, Guo B, Li Q, Peng J, Yang Z,Wang A, et al. miR-214 targets ATF4 toinhibit bone formation. Nat Med 2013;19:93–100.

14. Yoo AS, Greenwald I. LIN-12/Notch activation leads to microRNA-medi-ated down-regulation of Vav in C. elegans. Science 2005;310:1330–3.

15. Lu Y, Thomson JM, Wong HY, Hammond SM, Hogan BL. Transgenic over-expression of themicroRNAmiR- 17-92 cluster promotes proliferation andinhibits differentiation of lung epithelial progenitor cells. Dev Biol2007;310:442–53.

16. Macfarlane LA, Murphy PR. MicroRNA: biogenesis, function and role incancer. Curr Genomics 2010;11:537–61.

17. Wang JH, Zheng WW, Cheng ST, Liu BX, Liu FR, Song JQ. Correlationbetween microRNA21 and sprout homolog 2 gene expression in multiplemyeloma. Mol Med Rep 2015;11:4220–4.

18. Ell B,Mercatali L, IbrahimT,Campbell N, SchwarzenbachH, Pantel K, et al.Tumor-induced osteoclastmiRNA changes as regulators and biomarkers ofosteolytic bone metastasis. Cancer Cell 2013;24:542–56.

19. Krzeszinski JY, Wei W, Huynh H, Jin Z, Wang X, Chang TC, et al. miR-34ablocks osteoporosis and bone metastasis by inhibiting osteoclastogenesisand Tgif2. Nature 2014;512:431–5.

20. Takeshita F, Patrawala L, Osaki M, Takahashi RU, Yamamoto Y, Kosaka N,et al. Systemic delivery of synthetic microRNA-16 inhibits the growth of

metastatic prostate tumors via downregulation ofmultiple cell-cycle genes.Mol Ther 2010;18:181–7.

21. Colden M, Dar AA, Saini S, Dahiya PV, Shahryari V, Yamamura S, et al.MicroRNA-466 inhibits tumor growth and bone metastasis in prostatecancer by direct regulation of osteogenic transcription factor RUNX2. CellDeath Dis 2017;8:e2572.

22. Edwards CM, Edwards JR, Lwin ST, Esparza J, Oyajobi BO, McCluskey B,et al. Increasing Wnt signaling in the bone marrow microenvironmentinhibits the development of myeloma bone disease and reduces tumorburden in bone in vivo. Blood 2008;111:2833–42.

23. Lionetti M, Biasiolo M, Agnelli L, Todoerti K, Mosca L, Fabris S, et al.Identification of microRNA expression patterns and definition of a micro-RNA/mRNA regulatory network in distinct molecular groups of multiplemyeloma. Blood 2009;114:e20–6.

24. BakkusMH, Asosingh K, Vanderkerken K, Thielemans K, Hagemeijer A, DeRaeveH, et al. Myeloma isotype-switch variants in themurine 5Tmyelomamodel: evidence that myeloma IgM and IgA expressing subclones canoriginate from the IgG expressing tumour. Leukemia 2001;15:1127–32.

25. ZhuD, van Arkel C, KingCA,Meirvenne SV, deGreef C, Thielemans K, et al.Immunoglobulin VH gene sequence analysis of spontaneous murineimmunoglobulin-secreting B-cell tumours with clinical features of humandisease. Immunology 1998;93:162–70.

26. van den Akker TW, Radl J, Franken-Postma E, Hagemeijer A. Cytogeneticfindings in mouse multiple myeloma and Waldenstrom's macroglobuli-nemia. Cancer Genet Cytogenet 1996;86:156–61.

27. Kelly T, Suva LJ, Huang Y, Macleod V, Miao HQ, Walker RC, et al.Expression of heparanase by primary breast tumors promotes bone resorp-tion in the absence of detectable bone metastases. Cancer Res 2005;65:5778–84.

28. Schmiedel BJ, Scheible CA, Nuebling T, Kopp HG, Wirths S, Azuma M,et al. RANKL expression, function, and therapeutic targeting in multiplemyeloma and chronic lymphocytic leukemia. Cancer Res 2013;73:683–94.

29. Yaccoby S, Ling W, Zhan F, Walker R, Barlogie B, Shaughnessy JD Jr.Antibody-based inhibition of DKK1 suppresses tumor-induced boneresorption and multiple myeloma growth in vivo. Blood 2007;109:2106–11.

30. Roodman GD. Pathogenesis of myeloma bone disease. Leukemia 2009;23:435–41.

31. Heider U,Hofbauer LC, Zavrski I, KaiserM, JakobC, SezerO.Novel aspectsof osteoclast activation and osteoblast inhibition in myeloma bone dis-ease. Biochem Biophys Res Commun 2005;338:687–93.

32. Pratt G, Goodyear O, Moss P. Immunodeficiency and immunotherapy inmultiple myeloma. Br J Haematol 2007;138:563–79.

33. Ferrone C, Dranoff G. Dual roles for immunity in gastrointestinal cancers.J Clin Oncol 2010;28:4045–51.

34. Kamphorst AO, Wieland A, Nasti T, Yang S, Zhang R, Barber DL, et al.Rescue of exhausted CD8 T cells by PD-1-targeted therapies is CD28-dependent. Science 2017;355:1423–7.

35. Wu W, Shi Y, Li S, Zhang Y, Liu Y, Wu Y, et al. Blockade of Tim-3 signalingrestores the virus-specific CD8þ T-cell response in patients with chronichepatitis B. Eur J Immunol 2012;42:1180–91.

36. Karsenty G. Transcriptional control of skeletogenesis. Annu Rev GenomicsHum Genet 2008;9:183–96.






37. Marie PJ. Transcription factors controlling osteoblastogenesis. ArchBiochem Biophys 2008;473:98–105.

38. Pratap J, Imbalzano KM, Underwood JM, Cohet N, Gokul K, Akech J, et al.Ectopic runx2 expression in mammary epithelial cells disrupts formationof normal acini structure: implications for breast cancer progression.Cancer Res 2009;69:6807–14.

39. Onodera Y, Miki Y, Suzuki T, Takagi K, Akahira J, Sakyu T, et al. Runx2 inhuman breast carcinoma: its potential roles in cancer progression. CancerSci 2010;101:2670–5.

40. Giuliani N, Colla S, Rizzoli V. New insight in the mechanism of osteoclastactivation and formation in multiple myeloma: focus on the receptoractivator of NF-kappaB ligand (RANKL). Exp Hematol 2004;32:685–91.

41. O'Brien CA. Control of RANKL gene expression. Bone 2010;46:911–9.42. Zhou X, Su YX, Lao XM, Liang YJ, Liao GQ. CD19(þ)IL-10(þ) regulatory B

cells affect survival of tongue squamous cell carcinomapatients and induceresting CD4(þ) T cells to CD4(þ)Foxp3(þ) regulatory T cells. Oral Oncol2016;53:27–35.

43. Tanaka A, Sakaguchi S. Regulatory T cells in cancer immunotherapy. CellRes 2017;27:109–18.

44. Bouchie A. First microRNA mimic enters clinic. Nat Biotechnol 2013;31:577.

45. Bedi D, Gillespie JW, Petrenko VA, Ebner A, LeitnerM,Hinterdorfer P, et al.Targeted delivery of siRNA into breast cancer cells via phage fusionproteins. Mol Pharm 2013;10:551–9.


Gowda et al.




2018;16:1138-1148. Published OnlineFirst March 28, 2018.Mol Cancer Res Pramod S. Gowda, Benjamin J. Wildman, Timothy N. Trotter, et al. Myeloma ProgressionRunx2 Suppression by miR-342 and miR-363 Inhibits Multiple

Updated version

10.1158/1541-7786.MCR-17-0606doi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2018/03/28/1541-7786.MCR-17-0606.DC1

Access the most recent supplemental material at:

Cited articles

http://mcr.aacrjournals.org/content/16/7/1138.full#ref-list-1

This article cites 45 articles, 13 of which you can access for free at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://mcr.aacrjournals.org/content/16/7/1138To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-17-0606

http://mcr.aacrjournals.org/content/suppl/2018/03/28/1541-7786.MCR-17-0606.DC1

http://mcr.aacrjournals.org/content/16/7/1138.full#ref-list-1

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/16/7/1138


runx2 suppression by mir-342 and mir-363 inhibits multiple ... · oncogenes and tumor suppressors...

Documents