Epigenetic Regulation of the PTEN AKT RAC1 Axis by G9a Is ...Molecular Cell Biology Epigenetic Regulation of the PTEN–AKT–RAC1 Axis by G9a Is Critical for Tumor Growth in Alveolar

Molecular Cell Biology

Epigenetic Regulation of the PTEN–AKT–RAC1Axis by G9a Is Critical for Tumor Growth inAlveolar RhabdomyosarcomaAkshay V. Bhat1, Monica Palanichamy Kala1, Vinay Kumar Rao1, Luca Pignata2,HueyJinLim3, SudhaSuriyamurthy1, KennethT.Chang4,VictorK. Lee3, ErnestoGuccione2,and Reshma Taneja1

Abstract

Alveolar rhabdomyosarcoma (ARMS) is an aggressivepediatric cancer with poor prognosis. As transient and stablemodifications to chromatin have emerged as criticalmechanisms in oncogenic signaling, efforts to target epige-netic modifiers as a therapeutic strategy have accelerated inrecent years. To identify chromatin modifiers that sustaintumor growth, we performed an epigenetic screen andfound that inhibition of lysine methyltransferase G9a sig-nificantly affected the viability of ARMS cell lines. Targetingexpression or activity of G9a reduced cellular proliferationand motility in vitro and tumor growth in vivo. Transcrip-tome and chromatin immunoprecipitation–sequencinganalysis provided mechanistic evidence that the tumor-

suppressor PTEN was a direct target gene of G9a. G9arepressed PTEN expression in a methyltransferase activity–dependent manner, resulting in increased AKT and RAC1activity. Re-expression of constitutively active RAC1 in G9a-deficient tumor cells restored oncogenic phenotypes, demon-strating its critical functions downstream of G9a. Collectively,our study provides evidence for a G9a-dependent epigeneticprogram that regulates tumor growth and suggests targetingG9a as a therapeutic strategy in ARMS.

Significance: These findings demonstrate that RAC1 is aneffector of G9a oncogenic functions and highlight the poten-tial of G9a inhibitors in the treatment of ARMS.

IntroductionRhabdomyosarcoma (RMS) is a highly malignant soft-tissue

sarcoma of childhood, which accounts for approximately 5% to8% of pediatric cancers, occurring mostly in children below 10years of age. Despite expression of the myogenic proteins MyoDandmyogenin, RMS cells fail to undergomyogenic differentiationand are morphologically similar to premature mesenchymalcells (1). The most common forms of RMS are embryonal rhab-domyosarcoma (ERMS) and alveolar rhabdomyosarcoma(ARMS). ARMS, the most aggressive subtype, is associated withfrequent metastasis at the time of diagnosis and exhibits limitedresponse to treatment, resulting in poor survival rates. ARMS isdistinguished from the other subtypes of RMS by frequent trans-locations t(2; 13) (q35; q14) and t(1; 13) (p36; q14), resulting in

the expression of PAX3-FOXO1 and PAX7-FOXO1 fusion pro-teins in 60% and 20% cases, respectively (2, 3). PAX3-FOXO1 isbelieved to be critical for oncogenesis by control of genes requiredfor proliferation, survival, andmetastasis (4–8). ARMS expressingeither of these translocations is termed fusion positive, whereas asmall percentage of ARMS tumors (�20%) are devoid of thesechromosomal translocations and are termed fusion negative.Fusion-negative ARMS is similar to ERMS in terms of prognosis,event-free survival, frequency of metastases, and distribution ofsite signifying the importance of PAX3/PAX7-FOXO1 in fusion-positive ARMS tumors (9). In addition to the presence of fusiononcoproteins, comprehensive analysis of the genomic landscapein ARMS has shown deregulation of growth factor signalingincluding insulin-like growth factor (IGF)/insulin-like growthfactor receptor 1, mutations in PIK3CA, and amplification ofCDK4 (10–12). Elevated growth factor levels stimulate the PI3Kpathway, resulting in phosphorylation of phosphatidylinositol-4,5-bisphosphate (PIP2) to produce phosphatidylinositol-3,4,5-bisphosphate (PIP3). This facilitates recruitment of AKT to theplasma membrane. Subsequent phosphorylation of AKT at thre-onine 308 (Thr308) by phosphoinositide-dependent kinase(PDK1) and at serine 473 (Ser473) by the mTORC-2 leads to itsactivation and affects several cellular processes such as prolifer-ation, growth, survival, migration, and metabolism (13). Theguanine nucleotide exchange factors are also activated by PI3K/AKT signaling, resulting in increased Ras-related C3 botulinumtoxin substrate 1 (RAC1) activity (14). Elevated RAC1 activity isassociatedwith aggressive cancers (15) andhasmultiple effects ongrowth, proliferation, and cell invasion. The tumor-suppressorPTEN dephosphorylates PIP3 and thus negatively regulates the

1Department of Physiology, YongLooLin School ofMedicine, National Universityof Singapore, Singapore, Singapore. 2Institute of Molecular and Cell Biology(IMCB), Agency for Science, Technology and Research (A�STAR), Singapore,Singapore. 3Department of Pathology, Yong Loo Lin School of Medicine,National University of Singapore, Singapore, Singapore. 4Department of Pathol-ogy, KK Women and Children's Hospital, Singapore, Singapore.

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

Corresponding Author: Reshma Taneja, National University of Singapore, BlockMD9, 2 Medical Drive, Singapore 117593, Singapore. Phone: 65-6516-3236;Fax: 65-6778-8161; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-18-2676

�2019 American Association for Cancer Research.

CancerResearch

Cancer Res; 79(9) May 1, 20192232

on April 23, 2021. © 2019 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst March 4, 2019; DOI: 10.1158/0008-5472.CAN-18-2676

http://crossmark.crossref.org/dialog/?doi=10.1158/0008-5472.CAN-18-2676&domain=pdf&date_stamp=2019-4-19

http://cancerres.aacrjournals.org/

PI3K–AKT pathway. The downregulation of PTEN expression inARMS contributes to elevated PI3K/AKT/RAC1 signaling that islinked to tumor cell survival (16, 17). However, the molecularmechanisms that result in loss of PTEN expression are not under-stood and could provide a tool to suppress oncogenic signalingthrough the PI3K/AKT/RAC1 pathway.

Recent advances in chromatin biology have revealed deregula-tion of the epigenetic landscape in ARMS (18). The chromatinmodifiers CHD4, PCAF, and BRD4 have been shown to interactwith and affect PAX3-FOXO1 activity (19–21). Nevertheless, theepigenetic network that is central to ARMS tumorigenesis remainstobe identified. EHMT2/G9aandEHMT1/GLP are key SETdomaincontaining lysinemethyltransferases that catalyze dimethylation ofhistoneH3 lysine 9 (H3K9me2), resulting in transcriptional repres-sion of target genes (22). Deregulated expression and function ofG9a, and to a lesser extent GLP, have been reported in cancers ofdifferent origins (23–34). In addition to its canonical role in generepression, G9a can also activate genes through association withcoactivator complexes in a context-dependent manner (23). How-ever, the signaling pathways that G9a regulates and its downstreameffectors are poorly understood. In the context of skeletal myogen-esis, we and others have shown that G9a inhibits muscle differen-tiation and promotes proliferation of myoblasts (35–38). Wetherefore hypothesized that G9a expression may be deregulatedin ARMS and contribute to tumorigenesis.

Here, we provide evidence that the G9a pathway is central toARMS tumorigenesis. Using an epigenetic screen, we demonstratethat G9a inhibitors potently affect viability of ARMS cell lines.Downregulation of G9a expression or pharmacologic inhibitionof its activity not only reduces tumor cell proliferation andinvasion, but also results in myogenic differentiation. In vivo,reduction of G9a function reduces tumor growth in xenograftmouse models. Transcriptomic and chromatin immunoprecipi-tation (ChIP)-sequencing (ChIP-seq) analyses demonstratethat G9a directly inhibits PTEN expression in a methylation-dependent manner, resulting in the upregulation of phospho-AKT levels and RAC1 activity. Re-expression of constitutivelyactive RAC1 reverses the impact of G9a deficiency in tumorgrowth. Our data suggest that the PTEN–RAC1 axis downstreamof G9a is central to ARMS tumorigenesis.

Materials and MethodsCell culture and stable cell lines

Primary human skeletalmusclemyoblasts (HSMM; Lonza Inc.)were cultured in growth medium (SkGM-2 BulletKit). ARMS celllines RH30, RH4, and RH41 were provided by Peter Houghton(Nationwide Children's Hospital, Columbus, OH) and RosellaRota (Bambino Gesu Children's Hospital, Rome, Italy) andcultured in RPMI 1640 with L-glutamine (Thermo Fisher Scien-tific) with 10% FBS (Hyclone). ARMS cells lines were authenti-cated for PAX3-FOXO1, MyoD, and myogenin expression byWestern blot as described (21). Cell lines were tested at leastevery 6 months forMycoplasma contamination using the MycokitDetection Kit fromBiowest (K1000). After thawing, cell lineswerenot passaged more than 10 times. The cell lines were maintainedfor no more than 3 passages between experiments. Where indi-cated, cells were treated with 2.5 mmol/L UNC0642 (Sigma-Aldrich) or RAC1 inhibitor NSC23766 (Tocris Bioscience) at50 mmol/L (RH30) and 30 mmol/L (RH41) for 48 hours. Controlcells were treated with equivalent concentrations of DMSO (Sig-

ma-Aldrich). For transient knockdown experiments, cells weretransfected with 100 nmol/L of human G9a-specific siRNA (ON-TARGETplus EHMT2 siRNA SMARTpool, Dharmacon) contain-ing a pool of three to five 19–25 nucleotide siRNAs. Control cellswere transfected with 100 nmol/L scrambled siRNA (ON-TAR-GETplus, nontargeting pool, Dharmacon) using LipofectamineRNAiMax (Thermo Fisher Scientific). Cells were analyzed 48hours after transfection for all assays. For generating stable knock-down cell lines, 293FT cells were transfected with packagingplasmids plP1 (5 mg) and plP2 (5 mg), envelope plasmid plP/VSV-G (5 mg; ViraPower Lentiviral Packaging Mix, Thermo FisherScientific), and 5 mg lentiviral expression constructs shRNA(pLKO.1, Mission shRNA DNA clone, Sigma-Aldrich) or shG9a(pLKO.1) (#SHCLND-NM_025256 MISSION shRNA PlasmidDNA). Sixteen hours after transfection, the cell supernatant wasreplaced with basal DMEM. The cell supernatants were collected48 and 72hours after transfection, centrifuged, and the viral pelletwas resuspended in 200 mL of RPMI medium as stock solution.RH30 cells at 40% to 50% confluency were transduced withshRNA control virus (control), or shG9a virus and 2 mL polybrene(8mg/mL; Sigma-Aldrich) in RPMI1640basalmedium. Six hoursafter transduction, cell supernatants were replaced with RPMImedium (10% FBS) for 24 hours. Transduced cells were selectedwith 1 mg/mL puromycin (Sigma-Aldrich) for 3 days. For rescueexperiments, shcontrol and shG9a cells were transfected with 2 mgof Rac1-GFPplasmid (pcDNA3-EGFP-Rac1-Q61L, Addgene) usingLipofectamine 2000 with PLUS reagent (Thermo Fisher Scientific).

Drug sensitivity of ARMS cells to inhibitors ofmethyltransferases

RH30 and RH4 cells were treated for 8 days with the indicatedcompounds in 384 wells at three different concentrations (3, 1,and0.3mmol/L). Viability at days 4 and 8was scored byMTS assayand reported as the ratio over control-treated cells (with equiv-alent dilution ofDMSO; RED> control;WHITE¼ control; BLUE <control). The experiment was conducted in triplicate, and(þ)-JQ1 was used as a positive control.

Western blot analysisWhole-cell extracts were isolated using RIPA or SDS lysis buffer

supplemented with protease inhibitors (Complete Mini, Sigma-Aldrich). The following primary antibodies were used: anti-G9a(#3306S, 1:300, Cell Signaling Technology), anti-PTEN (138G6)(#9559, 1:1,000, Cell Signaling Technology), anti–phospho-AKT(Ser473) (D9E) (#4060, 1:500, Cell Signaling Technology),anti-AKT (#9272, 1:1,000, Cell Signaling Technology), anti-H3K9me2 (#9753S, 1:1,000, Cell Signaling Technology), anti-RAC1 (23A8) (#05-389, 1:500, Merck Millipore), anti–cyclinD1 (H295) (#sc753, 1:250, Santa Cruz Biotechnology), anti–ß-actin (#A2228, 1:10,000; Sigma-Aldrich), anti-H3 (#ab1791,1:10,000; Abcam), and anti-GAPDH (14C10) (#2118, 1:500,Santa Cruz Biotechnology). Appropriate secondary antibodies(IgG-Fc–Specific Peroxidase) of mouse or rabbit origin (Sigma-Aldrich) were used.

RAC1 activity assayRAC1 activity in control and shG9a cells and RH30 cells

untreated and treated with UNC0642 was detected using theActive RAC1 Detection Kit (Cell Signaling Technology). GST-PAK1-PBD fusion protein was used to bind the activated formof GTP-bound RAC1. The complex was immunoprecipitated with

G9a Regulates RAC1 Activity in ARMS

www.aacrjournals.org Cancer Res; 79(9) May 1, 2019 2233




glutathione resin through GST-linked binding protein. Unboundproteins were washed off during centrifugation, and glutathioneresin-bound GTPase was eluted with SDS buffer. Active RAC1levels were determined byWestern blot using anti-RAC1 antibody(Cell Signaling Technology).

Proliferation, migration, and invasion assaysCell proliferation was assessed by 5-bromo-20-deoxy-uridine

(BrdU) labeling (Roche). Briefly, cells grown on coverslips werepulsed with 10 mmol/L BrdU for 30 minutes, fixed, and stainedwith anti-BrdU antibody (1:100) for 30 minutes. The coverslipswere washed, incubated with anti-mouse Ig-fluorescein antibody(1:200), andmounted onto a glass slide using DAPI (Vectashield,Vector Laboratories). Images were captured using fluorescencemicroscope BX53 (Olympus Corporation).

For transwell migration assays, cells were seeded in a 6-welldish at a confluency of 1� 105 cells/well. Where appropriate, cellswere treated with UNC0642 (2.5 mmol/L) or siG9a (100 nmol/L)for 48 hours and serum starved for 24 hours. Note that 5 � 104

cells were seeded into extracellular matrix (Matrigel, collagen I)–coated polycarbonate membrane inserts (8.0 mm pore size) andplaced in a 24-well plate containing RPMI 1640 (10% FBS). After24 hours, cells that migrated to the bottom surface of the insertwerefixedwith 4%paraformaldehyde and stainedwith 6%crystalviolet for 20 minutes. Cells were then counted based on five fielddigital images taken at 10X magnification using EVOS XL coreimaging system AMEX1000 (Thermo Fisher Scientific).

For wound-healing assays, two wounds were created perpen-dicular to each other using a 200 mL yellow tip in the plates withcells at approximately 95% confluency. The wound was thenmonitored at 24 hours. The migratory capacity was calculated asa percentage of wound closure with respect to zero-time point.The cells were fixed and imaged as described above.

Immunofluorescence assaysFor differentiation assays, RH30 and RH41 cells were cultured

for 3 to 5 days in RPMI 1640 supplemented with 2% horseserum (Gibco) at 90% to 95% confluency. Cells were fixed with4% paraformaldehyde, and incubated with anti-Myosin HeavyChain (MHC; R&D Systems; 1:500, 1 hour at room temperature),followed by secondary goat anti-Mouse IgG (HþL) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor 568 (Thermo FisherScientific). Coverslips were mounted with DAPI (Vectashield,Vector Laboratories) and imaged using upright fluorescencemicroscope BX53 (Olympus Corporation).

Transcriptome analysis and quantitative real-time PCRMultiplex gene expression analysis was carried out using Pan-

Cancer pathway Panel (#XT-CSO-PATH1-12, Nanostring) usingRNA from two biological replicates of RH41 cells transfected withscrambled siRNA or siG9a. RNA clean-up was performed usingRNeasyMiniElute CleanupKit (Qiagen). The hybridized RNAwasscanned using nCounter digital analyzer. The RCC files generatedfrom nCounter were analyzed using nSolver 3.0 software. Rawcountswere normalized to the geometricmean of positive controland housekeeping genes. To identify genes and pathways thatare deregulated, PanCancer Pathway Advanced Analysis modulewas applied. The PanCancer pathway data are compliant withMinimum Information About aMicroarray Experiment (MIAME)guidelines and have been deposited in NCBI Gene ExpressionOmnibus (GEO)Database (GEOaccessionnumberGSE118777).

For qPCR, total RNA was extracted using TRIzol (Thermo FisherScientific) and quantified using Nanodrop. mRNA was convertedto a single-stranded cDNA using iScript cDNA Synthesis Kit (Bio-Rad). qPCR was performed using Lightcycler 480 SYBR Green 1Master Kit (Roche). PCR amplification was performed as follows:95�C for 5minutes, followedby 95�C for 10 seconds, annealing at60�C for 10 seconds, followed by 45 cycles at 72�C for 10 seconds.Melting curves were generated and tested for a single product afteramplification. Light Cycler 480 software (version 1.3.0.0705)wasused for analysis. CT values of samples were normalized tointernal control GAPDH to obtain delta CT (DCT). Relativeexpression was calculated by 2�DCT equation. qPCR was doneusing reaction triplicates, and at least two independent biologicalreplicates were done for each analysis. Representative data areshown. Error bars indicate the mean � SD. Primer sequences forPTEN, RAC1, and PI3KR1 have been described (39–41). Primersfor G9a are 50-TGGGCCATGCCACAAAGTC-30 and 50-CAGATG-GAGGTGATTTTCCCG-30. The transcriptome data are compliantwith MIAME guidelines and have been submitted to the GEOrepository (accession number GSE118777).

ChIPChIP-seq was performed with anti-G9a antibody (Abcam).

Twenty million RH41 cells were treated with protein cross-linkerdisuccinimidyl glutarate, followed by chromatin cross-linkingwith 1% formaldehyde. Cellswere lysedwith Farnham lysis buffer(5mmol/L PIPES, pH 8.0, 85mmol/L KCl, and 0.5%NP-40, withprotease inhibitors) and passed through 27-gauge needle. Crudenuclear preparation was collected by centrifugation and lysedwith RIPA buffer (1X PBS, 1%NP-40, 0.5% sodiumdeoxycholate,and 0.1% SDS with protease inhibitor). Chromatin shearing wasperformed using Bioruptor (Diagenode).Washes were performedusing buffers provided in the ChIP Kit (#17-295, Merck Milli-pore). Chromatin complex was eluted with elution buffer, reversecross-linked, and treated with proteinase K. DNA was extractedusing phenol-chloroform-isoamyl alcohol. Note that 30 ngChIPed DNA and corresponding input were used for ChIP-seqlibrary construction (NewEnglandBiolabs). Library fragment sizewas determined using the DNA 1000 Kit on the Agilent Bioana-lyzer (Agilent Technologies). Libraries were quantified by qPCRusing the KAPA Library Quantification Kit (KAPA Biosystems).Libraries were pooled in equimolar, and sequencing (150 bp pair-end) was performed on the Illumina MiSeq at the Duke-NUSGenome Biology Facility, according to the manufacturer's proto-col (Illumina). The total analyzable reads were computed afterfiltering reads with low mapping score (MAPQ < 10) and PCR-duplicated reads. To evaluate immunoprecipitation, the fractionof reads in peaks (FRiP) was computed using deeptools. Thelibrary of G9a ChIP-seq had an FRiP score of 23%, indicating thatthe library achieved successful global ChIP enrichment. Predic-tion of G9a-binding sites was done using MACS2 to capture bothbroad (q value < 0.10) and narrow peaks (q value < 0.05). Theprediction revealed 21,506 binding sites. The unique number ofpredicted binding sites from chr1-chrX was used in the analysis.The average fold enrichment of multiple peaks from the samebinding site was computed. To assess whether the distribution ofG9a bindings at promoters, gene body, and intergenic regions isnot obtained from random chance, we compared the total num-ber of G9a bindings overlapping with each genomic featureagainst a random distribution. Particularly, we shuffled the loca-tion of G9a-binding sites and computed the total number of

Bhat et al.

Cancer Res; 79(9) May 1, 2019 Cancer Research2234




random bindings overlapping with each genomic feature. Thisprocedure was repeated for 10,000 iterations. Empirical P valueswere then computed as the fraction of total iterations where thenumber of random bindings is greater than the observed numberof G9a bindings overlapping with each feature. We detected21,506 G9a-binding sites across the genome, of which 44%(n ¼ 9,560) localized at promoters, 31% (n ¼ 6,623) at genebodies, and 25% (n ¼ 5,323) at intergenic regions. Interestingly,the overlap of G9a-binding sites at promoters was approximately3x higher than the random bindings at promoters, and wasstatistically significant (empirical P value < 0.0001), highlightingthat thedistribution ofG9a binding at promoterwas not obtainedby random chance. On the other hand, the median occurrence ofrandombinding events at genebodies and intergenic regionswere9,020 and 9,425, respectively, indicating that the observed G9a-binding sites at these genomic features were underrepresentedcompared with random events. Top 5,000 predicted G9a-bindingsites showing high fold enrichment were used for the associationanalysis with biological processes. This was performed usingthe Genomic Regions Enrichment of Annotations Tool (GREAT)(http://great.stanford.edu/public/html/)with default parameters.To ensure consistency of the results, different top predictedbinding sites (n ¼ 2,500; n ¼ 10,000) were also tested.Sequencing reads mapped against hg19- and G9a-boundregions were identified by MACS2000-predicted binding sitesshowing high fold enrichment (binding signal to backgroundratio). The observation remained consistent even by using adifferent number of predicted sites (n ¼ 2,500; n ¼ 10,000),suggesting that the associations with biological processes arerobust and independent from the arbitrary number of sites. TheChIP-seq data are compliant with MIAME guidelines and havebeen deposited in the NCBI GEO database (accession numberGSE118666). ChIP-PCR was done as previously described (38).Briefly, 5 � 106 cells were cross-linked with 1% formaldehydefor 10 minutes at 37�C, sonicated, and ChIP was carried outaccording to the kit protocol (Merck Millipore). Immunopre-cipitates were reverse cross-linked, DNA extracted, and qPCRperformed as described above. Ten percent of the initial DNAwas used as input control. Relative enrichment was calculatedusing 2�DCT equation. The following antibodies were used forChIP assays: ChIP-grade anti-G9a (Abcam), anti-H3K9me2(Abcam), anti-H3K9ac (Abcam), anti-H3K27ac (Abcam). Pri-mers sequences for CSF2 and PGC1a have been described (42).G9a occupancy at the PTEN was analyzed using the primers:Forward: 50-GCAGGAAGGGTTGGGGTTCC-30 and Reverse: 50-GGATACACGGGCCACAGTCG-30 as described (42).

Mouse xenograft experiments and IHCAll animal procedures were approved by the Institutional

Animal Care and Use Committee. Note that 6 � 106 RH30 cellswere injected subcutaneously into the right flank of 6-week-oldCrTac:NCr-Foxn1 nude female mice (InVivos) as described (21).When tumor nodules were palpable, mice (n ¼ 10/group) wereinjected intraperitoneally with UNC0642 at 5mg/kg body weightevery alternate day. Control group was injected with DMSO.Tumor diameter was measured using digital vernier caliper onalternate days. Tumor volume was calculated using the followingformula: V¼ (L�W�W)/2, where V is the tumor volume,W isthe tumor width, and L is the tumor length. Body weight wasmeasured every alternate day. Once the tumors reached 1.5 cm indiameter in the control cohort,micewere euthanized and resected

tumors were fixed in paraformaldehyde. Paraffin sections werestained with hematoxylin and eosin or analyzed by IHC. In analternative approach, shcontrol and shG9a cells were injected inmice, and tumor growth was monitored as described above. ForIHC, slides were deparaffinized and rehydrated, followed byantigen retrieval with Na-citrate buffer (pH 7.0). Sections wereincubated overnight at 4�C with anti-G9a (1:200, Abcam),anti-H3K9me2 (Abcam), anti-Ki67 (1:100, Santa Cruz Biotech-nology), anti-active caspase 3 (Asp-175, 1:100; Cell SignalingTechnology), anti-RAC1 (23A8; Merck Millipore), anti-activeRAC1 (1:100, New East Biosciences) antibodies, followed bybiotinylated goat anti-rabbit/anti-mouse IgG (HþL) secondaryantibody (Vector Laboratories) for 1 hour at 37�C. Sections werewashed and incubated with Vectastain Avidin–Biotin Complex(Vector Laboratories) for 20 minutes at 37�C. After incubation,DAB substrate (Vector Laboratories) was added until colordevelopment. Sections were counterstained with hematoxylin(Sigma-Aldrich). Slides were dehydrated andmounted usingDPX(Sigma-Aldrich) and imaged using BX53 Olympus microscope.

IHC on 15 primary fusion-positive archival tumor tissuesobtained from the National University Hospital and KK Women'sand Children Hospital in Singapore was done essentially asdescribed (21). Sections were stained with anti-G9a antibody(1:50 dilution, Cell Signaling Technology). Three normal muscletissue sections were used as controls. Negative controls were per-formed using secondary antibody only. Images were captured withOlympus BX43 microscope. Approval for the study was obtainedfrom the ethics committee (Institutional Review Board) at NUS.

Statistical analysisError bars indicate mean � SD unless specified otherwise.

Significance was determined using the Student t test (two-sided),and P values less than 0.05were considered statistically significant(�, P < 0.05; ��, P < 0.01; and ��, P < 0.001).

Data availabilityThe ChIP-seq data have been deposited in GEO under the

accession numberGSE118666. The PanCancer pathwaydata havebeen deposited in GEO under the accession number GSE118777.

ResultsG9a is an overexpressed and druggable target in ARMS

We performed a small-molecule screen in order to identifydruggable methyltransferases involved in the tumorigenesis ofARMS. Three concentrations (3, 1, and 0.3 mmol/L) of 13 methyl-transferases inhibitors (provided by the Structural Genomic Con-sortium) were tested in two different cell lines, RH30 and RH4, attwo different time points (days 4–8). Viability was evaluated via anMTSassay (Fig. 1A).Blue and red in theheatmap indicate a lower orhigher number of cells, respectively, compared with the control. Inaddition to JQ1 that inhibits Brd4 activity and has recently beenshown to affect PAX3-FOXO1 in RMS (19), the screen identifiedfour drugs that show a strong effect on viability in both cell lines:MS023 (PRMT type I inhibitor), GSK591 (PRMT5 inhibitor),UNC0638, and UNC0642 (G9a inhibitors). We decided in thisstudy to focus on the oncogenic activity of G9a. The effect ofmodulating arginine methylation in ARMS will be described else-where. Consistent with the screen, treatment of patient-derivedRH30 and RH41 cells with UNC0642 (Supplementary Fig. S1Aand S1B), or downregulation of G9a expression using shRNA





http://great.stanford.edu/public/html/


A

CB

D

RH4Alveolar

(Pax3-Foxo1)

RH30Alveolar

(Pax3-Foxo1)DRUGS TARGETS

GSK343 EZH2, EZH1

SGC707 PRMT3

R)-PFI-2 (hydrochloride) SETD7

DMSO

BAY-598 SMYD2

A-366 G9a, EHMT1

SGC0946 Dot1L

UNC1999 EZH2, EZH1

A-196 SUV420H1/2

MS049 (HCl salt) PRMT4/6

MS023 (HCl salt) PRMT type I

UNC0642 G9a, EHMT1

UNC0638 G9a, EHMT1

GSK591 PRMT5

(+)-JQ1 BRD4

Day 4DRUGS TARGETS

BAY-598 SMYD2

DMSO

R)-PFI-2 (hydrochloride) SETD7

SGC0946 Dot1L

SGC707 PRMT3

GSK343 EZH2, EZH1

A-366 G9a, EHMT1

MS049 (HCl salt) PRMT4/6

A-196 SUV420H1/2

MS023 (HCl salt) PRMT type I

UNC0642 G9a, EHMT1

GSK591 PRMT5

UNC1999 EZH2, EZH1

UNC0638 G9a, EHMT1

(+)-JQ1 BRD4

Day 8

3 mmol/L 3 mmol/L 3 mmol/L0.3 mmol/L 3 mmol/L 0.3 mmol/L1 mmol/L 1 mmol/L 1 mmol/L0.3 mmol/L 0.3 mmol/L1 mmol/L

RH4Alveolar

(Pax3-Foxo1)

RH30Alveolar

(Pax3-Foxo1)

02468

1012

Rel

ativ

e ex

pres

sion

******

G9a - 160

- 42

Normal muscle

Negativecontrol ARMS patient tumors (PAX3-FOXO1 positive )

1

2

1

2

3 3

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

RH4RH30DRUGS

MS023 2,500 nmol/L300 nmol/L

UNC0642 2,000 nmol/L300 nmol/L

UNC0638 2,000 nmol/L2,000 nmol/L

GSK591 750 nmol/L<300 nmol/L

Color keyColor key

0 2 0 2ValueValue

b-Actin

Figure 1.

G9a is overexpressed in fusion-positive ARMS. A, Drug sensitivity of ARMS cell lines RH30 and RH4 to methyltransferase inhibitors was tested. Cell lines weretreated up to 8 dayswith the indicated compounds in 384 wells (3, 1, and 0.3 mmol/L). Viability at days 4 and 8 was scored by MTS assay and reported as the ratioover control-treated cells (with equivalent dilution of DMSO; RED > control; WHITE¼ control; BLUE < control). The experiment was conducted in triplicates, and(þ)-JQ1 was used as a positive control. The results showed that MS023, UNC0642, UNC638, and GSK591 had a strong effect on the viability of ARMS cell lines.The IC50 of MS023, UNC0642, UNC0638, and GSK591 in RH30 and RH4 after 8 days is shown. B and C, G9amRNA (B) and protein (C) levels were examined inprimary HSMM and ARMS cells lines (RH30 and RH41) by qPCR andWestern blotting, respectively. Error bars, mean of qPCR triplicates in each experiment� SD.The results are representative of two independent experiments. Numbers indicate the molecular weight of proteins. D,G9a expression in three normal skeletalmuscles (Normal 1–3) was compared with 15 archival ARMS patient tumor specimens (#1–15) by IHC using anti-G9a antibody. Higher nuclear G9a staining wasapparent in all tumors, albeit to varying levels. The upper right inset shows a zoomed-in image. Negative controls were done by immunostaining three tumorspecimens with secondary antibody alone. Scale bar, 50 mm. �� , P < 0.001.

Bhat et al.





(Supplementary Fig. S1C), led to a striking reduction in colonyformation (Supplementary Fig. S1D). These findings led us toexamine G9a expression in ARMS cell lines and tumor specimens.G9amRNAandprotein levelswereupregulated inRH30andRH41cell lines (Fig. 1B and C) relative to primary HSMM. In addition,IHC analysis of 16 archival PAX3-FOXO1 fusion-positive ARMStumor sections showed thatG9awasoverexpressed in all specimensin comparison with normal human muscle (Fig. 1D).

Knockdown of G9a expression or activity suppressesproliferation and migration, and enhances myogenicdifferentiation

To examine its function in ARMS, we used two approaches. Wedown regulated endogenous G9a expression either stably (Sup-

plementary Fig. S1C), or transiently, using siRNA in RH30 (Sup-plementary Fig. S2A) cell line. In addition, we inhibited itscatalytic activity using UNC0642 (UNC) for 48 hours, whichresulted in reduced H3K9me2, a signature of G9a activity (Sup-plementary Fig. S1A and S1B). A striking reduction in cellularproliferation in shG9a cells was apparent by BrdU assay (Fig. 2A)that was confirmed by growth curve assays (Fig. 2B). A similarimpact on cellular proliferation was seen in RH30 cells upontransient G9a knockdown (siG9a) and UNC treatment comparedwith their respective controls (Supplementary Fig. S2B). Interest-ingly, an increased propensity to undergo myogenic differentia-tionwas also apparent by immunostainingwithMHCantibody inshG9a cells (Fig. 2C). These results were validated in siG9a RH30,as well as in response to UNC treatment (Supplementary Fig.

0

20

40

60

Control shG9a

*

% B

rdU

pos

itivi

ty

MHCDAPI Merge

Con

trol

shG

9a

RH30

D5

Con

trol

shG

9a

0 hr 24 hr

RH30

A

C

D E

Con

trol

RH30

shG

9a

24 hr

Con

trol

shG

9a

RH30

DAPI MHC Merge

0

0.2

0.4

0.6

0.8

1

D0 D1 D2 D3 D4 D5

sh control shG9a

RH30 ***

*

*

Cel

l den

sity

*106

B

0

20

40

60

80

100*

% M

igra

tion

Control shG9a0

20

40

60

80

100***

% In

vasi

on

Control shG9a

Figure 2.

Stable knockdown of G9a alters proliferation, differentiation, and motility. A, BrdU assays were done with control and shG9a cells. The percentage ofproliferating cells was quantified. B, Growth curve assay was performed by seeding 0.1 million control and shG9a RH30 cells. Cells were trypsinized every 24hours, and cell numbers were counted for a period of 5 days. C, Control and shG9a cells were differentiated for 5 days in 2% horse serum-containing medium.Cells were immunostained with anti-MHC antibody (red). D and E,Motility of control and shG9a cells was analyzed by wound-healing assays to assess migration(D) and by Matrigel Boyden chamber assays (E). Migration of cells was monitored over a 24-hour period. Error bars, mean� SD. � , P < 0.05; �� , P < 0.001.






S2C). Consistently, expression of myogenic markers was elevatedin siG9a and UNC-treated RH30 cells (Supplementary Fig. S2D).Because G9a has been reported to influence migratory and inva-sive capacity of cancer cells (22–24), we tested its involvement inmigration by wound-healing assays. Wound closure was moni-tored for 24 hours in G9a knockdown cells or cells that had beenpretreated with UNC0642 for 48 hours. Defective migration wasobserved in shG9a cells (Fig. 2D), RH30 siG9a cells (Supplemen-tary Fig. S2E) as well as upon treatment withUNC0642 comparedwith respective control cells. In addition, significantly fewer G9aknockdown cells and UNC0642-treated cells invaded throughMatrigel relative to controls (Fig. 2E; Supplementary Fig. S2F).Wefurther validated the effects of G9a depletion in a second PAX3-FOXO1þ ARMS cell line RH41. Both siG9a (Supplementary Fig.S3A) andUNC-treated cells showed reduced proliferation relativeto their controls (Supplementary Fig. S3B). Myogenic differenti-ation was increased as seen by MHC staining and elevated Myh1expression (Supplementary Fig. S3C and S3D) in siG9a andUNC-treated cells, whereas cellular migration and invasion weredecreased (Supplementary Fig. S3E and S3F).

PI3K signaling pathway is differentially regulated in ARMS cellssilenced for G9a

To identify the molecular pathways and functional targetsdownstream of G9a, we performed Nanostring PanCancer path-way analysis using RNA from control and siG9a cells. Globalsignificance scores indicated that PI3K signaling was among themost differentially regulated pathways (Fig. 3A; SupplementaryTable S1A). Of the 55 genes that were significantly altered by lossof G9a expression, several genes in PI3K signaling pathway werefound to be deregulated (Fig. 3B; Supplementary Table S1B). Tovalidate these findings, we focused on PTEN, an upstreamregulator of the PI3K/AKT pathway. Consistent with the tran-scriptome analysis, PTEN mRNA was significantly upregulated inG9a knockdown cells, aswell as uponUNC0642 treatment in bothcell lines compared with respective controls (Fig. 3C and D),indicating that it is regulated by G9a in a methylation-dependentmanner. The expressionof additional differentially expressed geneswas validated by qPCR in siG9a cells (Supplementary Fig. S4).

G9a directly regulates PTEN expression and RAC1 activityBecause genome-wide direct targets of G9a in RMS are

unknown, we performed ChIP, followed by high-throughputsequencing with a G9a-specific antibody. G9a occupancy wasenriched at promoters, gene bodies, and intergenic regions(Fig. 3E), and 44% of the peaks were mapped to promotersat the transcription start site (TSS; Fig. 3F). Gene ontologyanalysis using GREAT showed that genes with G9a peaks weresignificantly enriched for biological processes that includeskeletal muscle development, regulation of stem cell, or mes-enchymal cell proliferation (Fig. 3G). G9a enrichment wasevident at the PTEN promoter region by ChIP-seq (Fig. 3H),and its occupancy was validated by ChIP-PCR in RH30 andRH41 cells (Fig. 3I; Supplementary Fig. S5A). We then analyzedactivation (H3K9ac) and repression (H3K9me2) marks at thePTEN promoter by ChIP-PCR in control and shG9a cells, aswell as upon UNC0642 treatment (Fig. 3J and K). Consistentwith an upregulation of its expression, H3K9ac was increased,whereas H3K9me2, a signature of G9a activity, was expectedlyreduced. A similar trend with both histone marks was seen atthe PTEN promoter in response to UNC0642 treatment. In line

with previous reports (42), we identified G9a occupancy at thePGC1a and CSF2 promoters by ChIP-seq that was validated byChIP-PCR (Supplementary Fig. S5B and S5C).

Given the repression of its expression by G9a, we examinedlevels of the PI3K–AKT pathway that is regulated by PTEN. UponUNC0642 treatment, as well as in siG9a cells, reduced phospho-AKT levels were apparent, indicating G9a regulates the PTEN–

pAKT axis in a methyltransferase activity–dependent manner(Fig. 4A). In addition, a clear downregulation of active RAC1levels was seen uponUNC0642 treatment aswell as in shG9a cells(Fig. 4B and C).

Re-expression of active RAC1 restores proliferative andmetastatic capacity of shG9a cells

Because RAC1 activity was down regulated by loss of G9a, wetestedwhether it is an effector ofG9a.Wefirst determinedwhetherinhibition of RAC1 activity mimics G9a deficiency in ARMS celllines. We used NSC23766, a small-molecule inhibitor of activeRAC1. Cells treated with NSC23766 showedmarked reduction ofBrdUþ cells (Supplementary Fig. S6A). In addition,migration andinvasion were significantly reduced in cells treated withNSC23766 compared with controls (Supplementary Fig. S6B andS6C). We then examined if re-expression of constitutively activeRAC1 restores proliferation and motility defects observed inshG9a cells. To this end, we transfected control and shG9a cellswith constitutively active RAC1 (RAC1Q61L) or empty vector(Fig. 4D). The proliferative capacity of shG9a cells was signifi-cantly rescued upon expression of RAC1Q61L-GFP comparedwith empty vector control (Fig. 4E). In addition, there wassignificant rescue of motility in shG9a cells as seen by wound-healing and invasion assays (Fig. 4F and G).

G9a knockdown inhibits tumorigenicity in vivoTo validate the role of G9a in tumorigenesis in vivo, we gener-

ated xenograft models by injecting RH30 cells subcutaneously innudemice.Weused two approaches. In thefirst,micewere treatedwith UNC0642 once tumors were palpable. The control groupwas treated with DMSO. Tumor volumes and body weights weremeasured. In the UNC0642-treated cohort, tumor volume wassignificantly lower compared with the matched DMSO controls(Fig. 5A–C) with no significant change in body weight (Fig. 5D).Tumor sections from DMSO control and UNC0642-treatedgroups were analyzed histologically and by IHC. A significantreduction in Ki67 stainingwas seen, indicating reduced tumor cellproliferation. In addition, H3K9me2 staining was expectedlyreduced in the UNC0642-treated mice compared with controls(Fig. 5E). No significant changes were seen in the staining for G9aand cleaved caspase 3. Interestingly, active RAC1 levels werestrongly down regulated correlating with the in vitro data. More-over, a significant increase in PTENexpressionwas observed in thetumors (Fig. 5F). In the second approach, we subcutaneouslyinjected control and shG9a cells in nudemice. Tumor volumewassignificantly lesser in mice injected with shG9a cells comparedwith controls (Fig. 5G–I), with no overt change in body weight(Fig. 5J). Ki67 and active RAC1 levels were strikingly reduced,whereas PTEN expression was elevated in the tumors (Fig. 5L).

DiscussionG9a has been reported to be overexpressed in a variety of

cancers (24–34), yet the signaling pathways that it controls and

Bhat et al.





0

0.01

0.02

0.03

0.04

H3K9ac**

% In

put

00.05

0.10.15

0.20.25

0.3

H3K9me2***

% In

put

0

0.1

0.2

0.3**

H3K9me2

% In

put

0

0.01

0.02

0.03

0.04*

H3K9ac

% In

put

BA

C

E

RH41 (PTEN mRNA)

00.5

11.5

22.5

Rel

ativ

e ex

pres

sion *

012345

Rel

ativ

e ex

pres

sion *

RH30 (PTEN mRNA) D

G

IH

Rel

ativ

e ex

pres

sion *

0

1

2

3

4R

elat

ive

expr

essi

on *

0

1

2

3

F

J RH30 (PTEN promoter) K RH30 (PTEN promoter)

*RH30 (PTEN promoter)

% In

put

00.010.020.030.040.050.06

-10 × log10 (P value)

log10

P value (Binomial test)

Num

ber o

f diff

eren

tially

expr

esse

d ge

nes

0

2

4

6

8

3024181260GNG4MMP9AKT3

PIK3R4RELA

FASAKT2

PIK3R1PTENRAC1

Control

ControlControl

Control

100

80

60

40

20

0PromoterGene bodyIntergenic -20 Kb +20 KbTSS

G9a

CH

IP-s

eqre

ad d

ensi

ty

5 10 15 20 25 30

25%44%

31%

5.3239.560

6.623

50 Kb

(q-value = 0.07)

70

ATAD1KLLNPTEN

CFL1P1

Biological processes

Skeletal muscle tissue developmentSkeletal muscle organ development

Positive regulation of stem cell proliferation

Positive regulation of epithelial cell proliferation involved in lung morphogenesis

Regulation of stem cell proliferationRegulation of mesenchymal cell proliferation

Positive regulation of mesenchymal cell proliferationRegulation of epithelial cell proliferation involved in lung morphogenesis

Positive regulation of multicellular organism growthNeuron recognition

Figure 3.

PTEN is a direct G9a target gene. A,Nanostring PanCancer Pathway analysis was performed with two biological replicates of control and siG9a cells. Gene setenrichment analysis revealed that PI3K signaling was among the most significantly altered pathways with deregulated expression of a large number of genes. B,A list of genes in the PI3K–AKT pathway that were significantly altered in siG9a cells relative to controls. The significance levels are shown. C and D, PTENmRNAlevels in control and siG9a cells, as well as upon UNC treatment, were analyzed by qPCR in RH30 and RH41 cells. The results are representative of twoindependent experiments. Error bars, mean of qPCR triplicates in each experiment� SD. E, Genome-wide G9a occupancy was analyzed by ChIP-seq. A pie chartshowing genomic distribution of G9a peaks at promoters, gene body, and intergenic regions. Highest G9a occupancy was observed at the promoter regions,followed by gene body and intergenic regions. F, Histogram showing G9a distribution at the TSS� 20 kb. G,MsigDB analysis using GREAT showed biologicalprocesses that are significantly regulated by G9a. H,Genomic snap shot using UCSC Genome Browser showing G9a enrichment at the PTEN promoter. I, ChIP-PCR validation showed G9a occupancy at the PTEN promoter. The results are representative of two independent experiments. Error bars, mean of qPCRtriplicates in each experiment� SD. J and K, ChIP-PCR analysis in control and shG9a cells for H3K9ac and H3K9me2 marks at the PTEN promoter. In addition,ChIP-PCR analysis was done in DMSO control and UNC0642-treated cells for H3K9ac and H3K9me2 enrichment. Error bars, mean of qPCR triplicates in eachexperiment� SD. The results are representative of two independent experiments. � , P < 0.05; �� , P < 0.01; �� , P < 0.001.






its transcriptional targets remain to be identified in differenttumor types. In this study, we report an unprecedented role ofG9a in epigenetic regulation of PI3K pathway in ARMS. Throughtranscriptome and ChIP-seq analysis, we identified PTEN to be a

direct G9a target gene. We show G9a occupancy and its signatureH3K9me2 repressive marks at the PTEN promoter. Loss of G9aexpression or pharmacologic inhibition of its methyltransferaseactivity relieved repression of PTEN expression, resulting in

A C

G

01530456075 ** ***

**

% B

rdU

+ce

lls

0

50

100

150

200

250

% In

vasi

on

***

*

D E

B

0

50

100

150

200

250***

***

% M

igra

tion

DA

PIB

rdU

ControlControl+RAC1 shG9a

shG9a+RAC1

RH30

Control Control + RAC1

0 hr

24 h

r

shG9a shG9a + RAC1

RH30

FControl Control+RAC1

shG9a shG9a+RAC1

RH30

24 hr

β-Actin

PTEN

G9a

p-AKT(Ser473)

RH41

-160

- 54

- 42

- 60

DMSO UNCRH41

G9a

ActiveRAC1Total

RAC1

GAPDH

-160

- 21

-21

- 36

shG9aControl

GAPDH

ActiveRAC1

TotalRAC1

RH30

- 21

- 21

- 36

G9a

GFP

GAPDH

Rac1-GFPControl shG9a

_ _ + + _ _ + +

RH30

- 160

- 27

- 36

Control

Control

Control

Control

Control

Control

Control

Figure 4.

RAC1 is an effector of G9a-dependent oncogenic phenotypes. A, PTEN expression and p-AKT (Ser473) levels were analyzed byWestern blot upon treatmentwith 2.5 mmol/L UNC0642 for 48 hours, as well as in control and siG9a cells. The results are representative of two independent experiments. B, RAC1 activity wasanalyzed 48 hours after UNC0642 treatment by immunoprecipitation with anti-active RAC1 antibody. The blots were immunoblotted with total RAC1 antibody.The results are representative of two independent experiments. C, RAC1 activity was examined in control and shG9a cells. Total RAC1 antibody was used as theloading control. The results are representative of two independent experiments. D–G, Control and shG9a cells were transfected with constitutively activeRAC1-GFP plasmid. Expression of G9a and RAC1 was analyzed byWestern blot using anti-G9a and anti-GFP antibodies (D). BrdU assay was done to measureproliferation (E), Matrigel transwell assay was done for invasion (F), and wound-healing assay for migration (G). Quantification of data in each assay is shown inthe respective bar graphs. The results are representative of at least two independent experiments. Error bars, mean� SD. � , P < 0.05; �� , P < 0.01; �� , P < 0.001.

Bhat et al.





A

UN

C

DM

SO

M1 M2 M3

B

Con

trol

shG

9a

M1 M2 M3

Con

trol

shG

9a

Bod

y w

t. (g

)

01020304050

D12D10D8D6D4D2D0

Control shG9a

010203040

D10D8D6D4D2D0

DMSO UNC0642 (5 mg/kg)

Bod

y w

t. (g

)

C

D

E

F

G H

J K

L

UN

C

DM

SO

I

RAC1

PTEN

β-Actin

DMSOUNCDMSO UNC

M1 M2

- 54

- 21

- 42

RAC1

PTEN

β-Actin

G9a

M1 M2

- 54

- 21

- 42

- 160

Rel

ativ

e tu

mor

vo

lum

e

01,0002,0003,0004,0005,000

D10D8D6D4D2D0

DMSO

UNC0642(5 mg/kg)

**

***

Rel

ativ

e tu

mor

vo

lum

e

0

2,000

4,000

6,000

8,000

D12D10D8D6D4D2D0

Control

shG9a*

*

*

Cas

p3

DMSO Tumor 1

UNCTumor 2 Tumor 1 Tumor 2

H&

EK

i67

H3K

9m

e2G

9aA

ctiv

eR

AC

1To

tal

RA

C1

H&

EK

i67

Tota

lR

AC

1G

9aA

ctiv

eR

AC

1

shG9aTumor 1 Tumor 2

Control

Control

Control

Tumor 1 Tumor 2

50 mmol/L

50 mmol/L

Figure 5.

G9a inhibition impairs tumor growth in vivo. A and B, Nude mice (n¼ 10/group) were injected with RH30 cells. When tumors were palpable, mice were treatedwith either vehicle DMSO or UNC0642 (5 mg/kg body weight). Three representative control and UNC0642-treated mice (M1, M2, and M3) and resected tumorsare shown inA and B, respectively. C, Relative tumor volume in control and UNC0642-treated mice was measured every 2 days after DMSO or UNC0642treatment. Day 0 (D0) refers to the day injections with DMSO or UNC0642 were started. D, Body weight of control and UNC0642-treated mice was measuredevery 2 days after treatment. E, Tumor sections from two independent DMSO and UNC0642-treated mice were stained with hematoxylin and eosin (H&E).Sections from the tumors were immunostained with anti-Ki67, anti-caspase 3 (Casp3), anti-G9a, anti-H3K9me2, and anti-active RAC1 and total RAC1 antibodies.F, Tumors were analyzed for PTEN and total RAC1 levels byWestern blot.G and H, Control and shG9a RH30 cells were injected in nude mice (n¼ 5/group). Micewere sacrificed at D12 after injection of cells. Representative images of mice and resected tumors are shown in G and H, respectively. I and J, Relative tumorvolume and body weight of control and shG9a mice were measured every 2 days. K, Tumor sections from two independent DMSO- and UNC0642-treated micewere stained with hematoxylin and eosin. Sections were immunostained with anti-Ki67, anti-G9a, anti-active RAC1, and anti-RAC1 antibodies. L, Tumor lysatesfrom control and shG9amice were analyzed byWestern blot with anti-PTEN, anti-RAC1, and anti–b-actin antibodies.






reduced AKT and RAC1 activity, and consequently impairedproliferation and migration of tumor cells. Interestingly, inhibi-tion of G9a also resulted in an increased propensity of tumor cellsto undergo differentiation. Our results demonstrate that G9a notonly sustains oncogenic signaling in the tumors, but also blockstheir ability to undergo myogenic differentiation. Our data areconsistent with a recent study that showed G9a-mediated repres-sion of PTEN expression in non–small cell lung cancer (43).

Previous studies have shown that the PI3K/AKT signaling iselevated in ARMS. This increase stems from both elevated growthfactors such as IGF-II and FGFR4, which are targets of PAX3-FOXO1, and decreased PTEN levels (16, 17). In addition tofunctioning as an antagonist of the PI3K/AKT pathway, PTENalso dephosphorylates other substrates such as p125FAK (44).However, increased motility of PTEN�/�

fibroblasts occurs dueto increased AKT and RAC1 activity without changes in p125FAK

levels, indicating that RAC1 mediates the effects PTEN in cellmigration (45). Our data are in line with these findings. PTENexpression is derepressed by UNC0642, which correlates withreduced RAC1 activity. Several positive and negative regulators ofPTEN expression have been described (46). Among these, the zincfinger transcription factors Snail and Slug have been shown torepress PTEN expression. Interestingly, G9a has been found tointeract with Snail (25) to mediate epithelial-to-mesenchymaltransition in breast cancer. These studies together with our datasuggest that Snail may recruit G9a to the PTEN promoter. We notethat RAC1 mRNA is reduced in G9a knockdown cells, suggestingthat G9amay affect RAC1 activity through additionalmechanisms.

Similar to G9a, RAC1 is a key regulator of cell migration,impairs cell-cycle exit, and inhibits myogenic differentiation (35–38, 47, 48). In addition, active RAC1 levels are elevated inARMS (48). We show that an RAC1 inhibitor mimics the effectsof aG9a inhibitor, suggesting thatG9a andRAC1downregulationhas similar phenotypic outcomes. Because re-expression of con-stitutively active RAC1 rescues the effects of G9a deficiency, ourdata demonstrate that RAC1 is a downstream effector of G9arather than functioning in a parallel pathway to regulatetumorigenesis.

The PI3K/AKT signaling is hyperactivated in many cancers andcontributes to survival and growth of tumor cells. Therefore,targeting this signaling axis provides an attractive opportunity

for therapeutic intervention (49, 50). Although a few tyrosinekinase inhibitors have now been approved by the FDA, develop-ment of drug resistance remains a major challenge with PI3Kinhibitors. AKT inhibitors have also met with limited clinicalsuccess. Similarly, no clinically effective drugs targeting RAC1activity are available to date (51). Recent bioinformatics analysesof cancer genome databases implicate several methyltransferasesand acetyltransferases in various human cancers includingRMS (18, 52). In particular, misregulation of lysine methylationhas been reported in various cancers that make proteins thatgovern this modification attractive drug targets. Our studiessuggest that the use of selective G9a inhibitors to increase PTENexpression and dampen AKT/RAC1 activity could provide a noveltherapeutic approach in ARMS.

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

Authors' ContributionsConception and design: A.V. Bhat, R. TanejaDevelopment of methodology: A.V. Bhat, M.P. Kala, L. Pignata, H.J. LimAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): V.K. Rao, S. Suriyamurthy, K.T. Chang, V.K. LeeAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): V.K. Rao, S. Suriyamurthy, R. TanejaWriting, review, and/or revision of the manuscript: R. TanejaAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): A.V. Bhat, V.K. Rao, K.T. ChangStudy supervision: E. Guccione, R. Taneja

AcknowledgmentsWe thankPeterHoughton andRosella Rota for ARMS cell lines, andOoiWen

Fong, Genome Institute of Singapore, for bioinformatics analysis.This work was supported by a National Medical Research Council grant

(NMRC/CBRG/0063/2014) toR. Taneja andE.Guccione. A.V. Bhat is supportedby the President's Graduate Scholarship at theNational University of Singapore.

The costs of publication of this articlewere defrayed inpart by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received August 25, 2018; revised December 17, 2018; accepted February 26,2019; published first March 4, 2019.

References1. Keller C, Guttridge DC. Mechanisms of impaired differentiation in rhab-

domyosarcoma. FEBS J 2013;280:4323–34.2. Linardic CM. PAX3-FOXO1 fusion gene in rhabdomyosarcoma.

Cancer Lett 2008;270:10–8.3. Olanich ME, Barr FG. A call to ARMS: targeting the PAX3-FOXO1 gene in

alveolar rhabdomyosarcoma. Expert Opin Ther Targets 2013;17:607–23.4. Wachtel M, Sch€afer BW. PAX3-FOXO1: zooming in on an "undruggable"

target. Semin Cancer Biol 2018;50:115–23.5. Cao L, Yu Y, Bilke S, Walker RL, Mayeenuddin LH, Azorsa DO, et al.

Genome-wide identification of PAX3-FKHR binding sites in rhabdomyo-sarcoma reveals candidate target genes important for development andcancer. Cancer Res 2010;70:6497–508.

6. Davicioni E, Finckenstein FG, Shahbazian V, Buckley JD, Triche TJ, Ander-son MJ. Identification of a PAX-FKHR gene expression signature thatdefines molecular classes and determines the prognosis of alveolar rhab-domyosarcomas. Cancer Res 2006;66:6936–46.

7. Mercado G E, Xia SJ, Zhang C, Ahn EH, Gustafson DM, La�e M, et al.Identification of PAX3-FKHR-regulated genes differentially expressedbetween alveolar and embryonal rhabdomyosarcoma: focus on MYCN as

a biologically relevant target. Genes Chromosomes Cancer 2008;47:510–20.

8. Bernasconi M, Remppis A, Fredericks WJ, Rauscher FJ, Sch€afer BW. Induc-tion of apoptosis in rhabdomyosarcoma cells through down-regulation ofPAX proteins. Proc Natl Acad Sci U S A 1996;93:13164–9.

9. Williamson D, Missiaglia E, de Reyni�es A, Pierron G, Thuille B, PalenzuelaG, et al. Fusion gene-negative alveolar rhabdomyosarcoma is clinically andmolecularly indistinguishable from embryonal rhabdomyosarcoma. J ClinOncol 2010;28:2151–8.

10. Shern JF, Chen L, Chmielecki J, Wei JS, Patidar R, Rosenberg M, et al.Comprehensive genomic analysis of rhabdomyosarcoma reveals a land-scape of alterations affecting a common genetic axis in fusion-positive andfusion-negative tumors. Cancer Discov 2014;4:216–31.

11. Seki M, Nishimura R, Yoshida K, Shimamura T, Shiraishi Y, Sato Y, et al.Integrated genetic and epigenetic analysis defines novel molecular sub-groups in rhabdomyosarcoma. Nat Commun 2015;6:7557.

12. Anderson J, Gordon A, Pritchard-Jones K, Shipley J. Genes, chromo-somes, and rhabdomyosarcoma. Genes Chromosomes Cancer 1999;26:275–85.

Bhat et al.





13. Hemmings BA, Restuccia DF. PI3K-PKB/Akt pathway. Cold Spring HarbPerspect Biol 2012;4:a011189.

14. Han J, Luby-Phelps K, Das B, Shu X, Xia Y, Mosteller RD, et al. Role ofsubstrates and products of PI 3-kinase in regulating activation of Rac-related guanosine triphosphatases by Vav. Science 1998;279:558–60.

15. Kazanietz MG, Caloca MJ. The Rac GTPase in cancer: from old concepts tonew paradigms. Cancer Res 2017;77:5445–51.

16. Cao L, YuY,Darko I,CurrierD,Mayeenuddin LH,WanX, et al. Addiction toelevated insulin-like growth factor I receptor and initial modulation of theAKT pathway define the responsiveness of rhabdomyosarcoma to thetargeting antibody. Cancer Res 2008;68:8039–48.

17. Wan X, Helman LJ. Levels of PTEN protein modulate Akt phosphorylationon serine 473, but not on threonine 308, in IGF-II-overexpressing rhab-domyosarcomas cells. Oncogene 2003;22:8205–11.

18. Shern JF, Yohe ME, Khan J. Pediatric rhabdomyosarcoma. Crit Rev Oncog2015;20:227–43.

19. Gryder BE, YoheME,ChouHC,ZhangX,Marques J,WachtelM, et al. PAX3-FOXO1 establishes myogenic super enhancers and confers BET bromo-domain vulnerability. Cancer Discov 2017;7:884–99.

20. B€ohm M, Wachtel M, Marques JG, Streiff N, Laubscher D, Nanni P, et al.Helicase CHD4 is an epigenetic coregulator of PAX3-FOXO1 in alveolarrhabdomyosarcoma. J Clin Invest 2016;126:4237–49.

21. Bharathy N, Suriyamurthy S, Rao VK, Ow JR, Lim HJ, Chakraborty P, et al.P/CAF mediates PAX3-FOXO1-dependent oncogenesis in alveolar rhab-domyosarcoma. J Pathol 2016;240:269–81.

22. Shinkai Y, Tachibana M. H3K9 methyltransferase G9a and the relatedmolecule GLP. Genes Dev 2011;25:781–8.

23. Shankar SR, Bahirvani AG, Rao VK, Bharathy N, Ow JR, Taneja R. G9a, amultipotent regulator of gene expression. Epigenetics 2013;8:16–22.

24. Chen MW, Hua KT, Kao HJ, Chi CC, Wei LH, Johansson G, et al. H3K9histone methyltransferase G9a promotes lung cancer invasion and metas-tasis by silencing the cell adhesionmolecule Ep-CAM. Cancer Res 2010;70:7830–40.

25. Dong C, Wu Y, Yao J, Wang Y, Yu Y, Rychahou PG, et al. G9a interacts withSnail and is critical for Snail-mediated E-cadherin repression in humanbreast cancer. J Clin Invest 2012;122:1469–86.

26. Lee SH, Kim J, Kim WH, Lee YM. Hypoxic silencing of tumor suppressorRUNX3 by histonemodification in gastric cancer cells. Oncogene 2009;28:184–94.

27. Lehnertz B, Pabst C, Su L, Miller M, Liu F, Yi L, et al. The methyltransferaseG9a regulates HoxA9-dependent transcription in AML. Genes Dev 2014;28:317–27.

28. Son HJ, Kim JY, Hahn Y, Seo SB. Negative regulation of JAK2 by H3K9methyltransferase G9a in leukemia. Mol Cell Biol 2012;32:3681–94.

29. Ding J, Li T, Wang X, Zhao E, Choi JH, Yang L, et al. The histone H3methyltransferase G9A epigenetically activates the serine-glycine synthesispathway to sustain cancer cell survival and proliferation. Cell Metab 2013;18:896–907.

30. Kang J, Shin SH, Yoon H, Huh J, Shin HW, Chun YS, et al. FIH is an oxygensensor in ovarian cancer for G9a/GLP-driven epigenetic regulation ofmetastasis-related genes. Cancer Res 2018;78:1184–99.

31. Bao L, Chen Y, Lai HT, Wu SY, Wang JE, Hatanpaa KJ, et al. Methylationof hypoxia-inducible factor (HIF)-1a by G9a/GLP inhibits HIF-1transcriptional activity and cell migration. Nucleic Acids Res 2018;6576–91.

32. Wei L, Chiu DK, Tsang FH, Law CT, Cheng CL, Au SL, et al. Histonemethyltransferase G9a promotes liver cancer development by epigeneticsilencing of tumor suppressor gene RARRES3. J Hepatol 2017;67:758–69.

33. Wang YF, Zhang J, Su Y, Shen YY, Jiang DX, Hou YY, et al. G9a regulatesbreast cancer growth by modulating iron homeostasis through the repres-sion of ferroxidase hephaestin. Nat Commun 2017;8:274.

34. Casciello F, Al-Ejeh F, Kelly G, Brennan DJ, Ngiow SF, Young A, et al. G9adrives hypoxia-mediated gene repression for breast cancer cell survival andtumorigenesis. Proc Natl Acad Sci U S A 2017;114:7077–82.

35. Wang J, Abate-Shen C. The MSX1 homeoprotein recruits G9amethyltrans-ferase to repressed target genes inmyoblast cells. PLoSOne2012;7:e37647.

36. Ling BM, Bharathy N, Chung TK, Kok WK, Li S, Tan YH, et al. Lysinemethyltransferase G9a methylates the transcription factor MyoD andregulates skeletal muscle differentiation. Proc Natl Acad Sci U S A 2012;109:841–6.

37. Battisti V, Pontis J, Boyarchuk E, Fritsch L, Robin P, Ait-Si-Ali S, et al.Unexpected distinct roles of the related histone H3 lysine 9 methyltrans-ferases G9a and G9a-like protein in myoblasts. J Mol Biol 2016;428:2329–43.

38. Rao VK, Ow JR, Shankar SR, Bharathy N, Manikandan J, Wang Y, et al. G9apromotes proliferation and inhibits cell cycle exit during myogenic differ-entiation. Nucleic Acids Res 2016;44:8129–43.

39. Clermont PL, Crea F, Chiang YT, Lin D, Zhang A, Wang JZ, et al. Identi-fication of the epigenetic reader CBX2 as a potential drug target in advancedprostate cancer. Clin Epigenetics 2016;8:16.

40. Wang W, Chen Y, Wang S, Hu N, Cao Z, Wang W, et al. PIASxa ligaseenhances SUMO1 modification of PTEN protein as a SUMO E3 ligase.J Biol Chem 2014;28:3217–30.

41. Cheng HW, Chen YF, Wong JM, Weng CW, Chen HY, Yu SL, et al. Cancercells increase endothelial cell tube formation and survival by activating thePI3K/Akt signaling pathway. J Exp Clin Cancer Res 2017;36:27.

42. Purcell DJ, Khalid O, Ou CY, Little GH, Frenkel B, Baniwal SK, et al.Recruitment of coregulator G9a by Runx2 for selective enhancement orsuppression of transcription. J Cell Biochem 2012;113:2406–14.

43. Wang L, Dong X, Ren Y, Luo J, Liu P, Su D, et al. Targeting EHMT2 reversesEGFR-TKI resistance in NSCLC by epigenetically regulating the PTEN/AKTsignaling pathway. Cell Death Dis 2018;9:129.

44. Tamura M, Gu J, Matsumoto K, Aota S, Parsons R, Yamada KM. Inhibitionof cell migration, spreading, and focal adhesions by tumor suppressorPTEN. Science 1998;280:1614–7.

45. Liliental J, Moon SY, Lesche R, Mamillapalli R, Li D, Zheng Y, et al. Geneticdeletion of the Pten tumor suppressor gene promotes cell motility byactivation of Rac1 and Cdc42 GTPases. Curr Biol 2000;10:401–4.

46. Haddadi N, Lin Y, Travis G, Simpson AM,Nassif NT,McGowan EM. PTEN/PTENP1: `regulating the regulator of RTK-dependent PI3K/Akt signaling',new targets for cancer therapy. Mol Cancer 2018;17:37.

47. Heller H, Gredinger E, Bengal E. Rac1 inhibits myogenic differentiation bypreventing the complete withdrawal ofmyoblasts from the cell cycle. J BiolChem 2001;276:37307–16.

48. Meriane M, Charrasse S, Comunale F, M�ery A, Fort P, Roux P, et al.Participation of small GTPases Rac1 and Cdc42Hs in myoblast transfor-mation. Oncogene 2002;21:2901–7.

49. Hennessy BT, Smith DL, Ram PT, Lu Y, Mills GB. Exploiting the PI3K/AKTpathway for cancer drug discovery. Nat RevDrugDiscov 2005;4:988–1004.

50. Janku F, Yap TA, Meric-Bernstam F. Targeting the PI3K pathway in cancer:are we making headway? Nat Rev Clin Oncol 2018;15:273–91.

51. Marei H, Malliri A. Rac1 in human diseases: the therapeutic potential oftargeting Rac1 signaling regulatory mechanisms. Small GTPases 2017;8:139–63.

52. Helin K, Dhanak D. Chromatin proteins andmodifications as drug targets.Nature 2013;502:480–8.






2019;79:2232-2243. Published OnlineFirst March 4, 2019.Cancer Res Akshay V. Bhat, Monica Palanichamy Kala, Vinay Kumar Rao, et al. Critical for Tumor Growth in Alveolar Rhabdomyosarcoma

RAC1 Axis by G9a Is−AKT−Epigenetic Regulation of the PTEN

Updated version

10.1158/0008-5472.CAN-18-2676doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2019/03/02/0008-5472.CAN-18-2676.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/79/9/2232.full#ref-list-1

This article cites 51 articles, 19 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/79/9/2232.full#related-urls

This article has been cited by 3 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

[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://cancerres.aacrjournals.org/content/79/9/2232To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-18-2676

http://cancerres.aacrjournals.org/content/suppl/2019/03/02/0008-5472.CAN-18-2676.DC1

http://cancerres.aacrjournals.org/content/79/9/2232.full#ref-list-1

http://cancerres.aacrjournals.org/content/79/9/2232.full#related-urls

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/79/9/2232


Documents

Epigenetic Regulation of the PTEN AKT RAC1 Axis by G9a Is ...Molecular Cell Biology Epigenetic Regulation of the PTEN–AKT–RAC1 Axis by G9a Is Critical for Tumor Growth in Alveolar