SET Domain Containing Protein 4 Epigenetically Controls ......Tumor Biology and Immunology SET Domain–Containing Protein 4 Epigenetically Controls Breast Cancer Stem Cell Quiescence

Tumor Biology and Immunology

SET Domain–Containing Protein 4 EpigeneticallyControls Breast Cancer Stem Cell QuiescenceSen Ye1, Yan-Fu Ding1,Wen-Huan Jia2, Xiao-Li Liu2, Jing-Yi Feng2, Qian Zhu2,Sun-Li Cai2, Yao-Shun Yang2, Qian-Yun Lu2, Xue-Ting Huang2, Jin-Shu Yang2,Sheng-Nan Jia3, Guo-Ping Ding3, Yue-Hong Wang4, Jiao-Jiao Zhou5,Yi-Ding Chen5, and Wei-Jun Yang1,2

Abstract

Quiescent cancer stem cells (CSC) play important roles intumorigenesis, relapse, and resistance to chemoradiotherapy.However, the determinants of CSC quiescence and how theysustain themselves to generate tumors and relapse beyondresistance to chemoradiotherapy remains unclear. Here, wefound that SET domain–containing protein 4 (SETD4) epige-netically controls breast CSC (BCSC) quiescence by facilitatingheterochromatin formation via H4K20me3 catalysis.H4K20me3 localized to the promoter regions and regulatedthe expression of a set of genes in quiescent BCSCs (qBCSC).SETD4-defined qBCSCs were resistant to chemoradiotherapyand promoted tumor relapse in a mouse model. Upon acti-vation, a SETD4-defined qBCSC sustained itself in a quiescentstate by asymmetric division and concurrently produced anactive daughter cell that proliferated to produce a cancer cellpopulation. Single-cell sequence analysis indicated thatSETD4þ qBCSCs clustered together as a distinct cell type

within the heterogeneous BCSC population. SETD4-definedquiescent CSCs were present in multiple cancer types includ-ing gastric, cervical, ovarian, liver, and lung cancers and wereresistant to chemotherapy. SETD4-defined qBCSCs had a hightumorigenesis potential and correlated with malignancy andchemotherapy resistance in clinical breast cancer patients.Taken together, the results fromour previous study and currentstudy on six cancer types reveal an evolutionarily conservedmechanism of cellular quiescence epigenetically controlled bySETD4. Our findings provide insights into the mechanism oftumorigenesis and relapse promoted by SETD4-defined qui-escentCSCs andhave broad implications for clinical therapies.

Significance: These findings advance our knowledge on theepigenetic determinants of quiescence in cancer stem cellpopulations and pave theway for future pharmacologic devel-opments aimed at targeting drug-resistant quiescent stemcells.

IntroductionAs a major global health problem, cancer is one of the leading

causes of morbidity worldwide. Because of the heterogeneity oftumor cells, the efficacy of chemoradiotherapy treatment is oftensuboptimal, as indicated by the high death rate of patients withcancer (1, 2). Current therapies are particularly limited by the

emergence of therapy-resistant cancer cells (3–5). Increasingevidence has revealed that a small fraction of cancer cells, termedcancer stem cells (CSC), are responsible for therapeutic resistance,where the quiescence inCSCs is a crucialmechanism for resistanceand survival (6–9). Standard therapies mainly target the tumorbulk, but they fail to eradicate the resistant CSCs that may causerelapse in patients after clinical treatments are finished (2, 7, 10).

Cellular quiescence is a reversible and nondividing state, thecounterpart to proliferation (11). Previous studies identifiedthat quiescent CSCs were more resistant to chemotherapy andcould retain the capacity to proliferate after chemotherapywithdrawal (12–14). Although several molecular players in theregulation of cellular quiescence have been reported, we knowremarkably little about the determinants of quiescence and themechanisms of the transition between active and quiescent statesin CSCs (7, 10, 15). Epigenetic studies have shown that hetero-chromatin is involved in maintaining the reversibility of cellularquiescence, in which the methylation of histones contributes toheterochromatin formation (16, 17). Heterochromatin exists intwo varieties, constitutive heterochromatin (cHC) and facultativeheterochromatin (fHC), that silence gene expression by virtueof their highly condensed structure (18, 19). The two varietiesare distinguished by their distinct epigenetic signatures, fHCinvolving high levels of trimethylation of lysine 9 of histone 3(H3K9me3) and lysine 27 of histone 3 (H3K27me3), and cHCexhibiting high levels of trimethylation of lysine 20 of histone 4

1MOE Laboratory of Biosystem Homeostasis and Protection, College of LifeSciences, Zhejiang University, Hangzhou, China. 2Institute of Cell and Develop-mental Biology, College of Life Sciences, Zhejiang University, Hangzhou, China.3Department of General Surgery, Sir Run Run Shaw Hospital, College of Med-icine, Zhejiang University, Hangzhou, China. 4Department of Respiratory Med-icine, The First Affiliated Hospital, College of Medicine, Zhejiang University,Hangzhou, China. 5Department of Surgical Oncology, The Second AffiliatedHospital, College of Medicine, Zhejiang University, Hangzhou, China.

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

S. Ye and Y.-F. Ding contributed equally to this article.

Corresponding Author: Wei-Jun Yang, Zhejiang University, 866 YuhangtangRd, Hangzhou 310058, Zhejiang Province, China. Phone:/Fax: 86-571-88273176;E-mail: [email protected]

Cancer Res 2019;79:4729–43

doi: 10.1158/0008-5472.CAN-19-1084

�2019 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 4729

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

Published OnlineFirst July 15, 2019; DOI: 10.1158/0008-5472.CAN-19-1084

http://crossmark.crossref.org/dialog/?doi=10.1158/0008-5472.CAN-19-1084&domain=pdf&date_stamp=2019-9-3

http://cancerres.aacrjournals.org/

(H4K20me3). In contrast, euchromatin displays a high level ofacetylation of lysine 9 of histone 3 (H3K9ac; refs. 16, 18, 20, 21).

The family of SET domain–containing proteins (SETD), his-tone lysine methyltransferases, has been reported to play a role inthe regulation of chromatin structure, gene expression by cata-lyzing the methylation of histone proteins and cell proliferationin several cell lines (22–25). To study cellular quiescence regula-tion in CSCs, we used Artemia, the brine shrimp, as a modelsystem. This primitive crustacean undergoes cellular quiescencefor prolonged periods during embryonic diapause, a state ofobligate dormancy to cope with environmental stresses (26).Previously, we reported that SETD4 regulates cellular quiescenceby catalyzing H4K20me3 during Artemia diapause entry (27).

Here, we show that SETD4 facilitates heterochromatin forma-tion via H4K20me3 catalysis in BCSCs that are located at certainpromoter regions and that it regulates the expression of a set ofgenes in the quiescent regulation of BCSCs. Indeed, the quiescentBCSCs play critical functions in resistance to chemoradiotherapyand in relapse and correlate with malignancy in clinical patients.We demonstrate an evolutionarily conserved mechanism of CSCquiescence and establish a new cellular narrative for tumorigen-esis and relapse.

Materials and MethodsMice

NOD/SCID female mice were purchased from the ShanghaiLaboratory Animal Center (SLAC) of China. They were housedunder a 12-hour light/dark cycle (light between 06:00 and 18:00)in a temperature-controlled room (22 � 1�C) with free access towater and food. Allmiceweremaintainedwith the approval of theAnimal Ethics Committee of the Zhejiang University (Hangzhou,China) and in accordancewith the university's animal experimentguidelines.

Cell lines and cell cultureHEK293T, MKN45, MCF7, T47D, and HCC1937 were pur-

chased from the Tumor Cell Bank of Chinese Academy of Sciences(Shanghai, China). All cells were obtained directly from cell bankand passaged in the laboratory for fewer than 6 months afterreceipt. All cells were authenticated using short tandem repeatprofiling.Mycoplasma detection was performed using aMycoplas-ma Detection Set (TaKaRa; 6601) for all the cell lines. They werecultured according to the vendor's instructions at 37�C in ahumidified atmosphere with 5% CO2.

AntibodiesThe following antibodies were used: H3K4me3 (Millipore;

2207275), H3K9me3 (Abcam; ab1773), H3K27me3 (Abcam;ab6174), H3K36me3 (Abcam; ab9050), H3K79me3 (Abcam;ab2621), H4K20me1 (Santa Cruz Biotechnology; sc-134221),H4K20me2 (GeneTeX; GTX630545), H4K20me3 for WB (CellSignaling Technology; 5737s), H4K20me3 for IF (Abcam;ab9053), GAPDH (Cell Signaling Technology; 2118), H3K9ac(Abcam; ab10812), HP1-a (Santa Cruz Biotechnology; sc-130446), SUV4-20h2 (Santa Cruz Biotechnology; sc-366867),H3 (Abcam; ab1791), H4 (Abcam;ab10158), H3S10ph (CellSignaling Technology; 53348), RbS807/811ph (Cell SignalingTechnology; 9308s), Ki67 (Abcam; ab16667), SETD4-(R) (Sig-ma-Aldrich; HPA035405), SETD4 (Sigma-Aldrich; HPA024073),SETD4-(m) (Santa Cruz Biotechnology; sc-514060), CD44-FITC

(eBioscience; 11-0441-81), CD44 (Cell Signaling Technology;3570), CD24-PE (eBioscience; 12-0242), CD24 (eBioscience;14-0242-81), CD24-647 (BioLegend; 311110), CD133(HuaAn-Biotec; EM1701-28), ALDH-1 (Abcam; ab52492), Sox2(Abcam; ab97959), Oct4 (Abcam; ab18976), Nanog (Abcam;ab109250), PCNA (Abcam; ab29), LC3B (Sigma-Aldrich;L7543), Active-caspase-3 (Abcam; ab32042), and H2AS139ph(g-H2A; Novus; NB100-384).

Discrimination of BCSCs and qBCSCsMCF7 or HCC1937 cells were digested by accutase (Gibco;

A1110501) and solid tumors fromMCF7-CDXs,HCC1937-CDXsor clinical patients were cut up into small pieces and then digestedwith ultrapure collagenase III (LS004180) in DMEM at 37�C for 3to 4 hours. Single cells were filtered through a 45-mmnylonmeshand then resuspended in 100 mL (per 106 cells) HBSS containing2% FBS for FACS. Antibodies of CD24 and CD44were added andincubated for 20 minutes on ice. Flow cytometry was performedusing a FACS Vantage (BD). Cells were routinely sorted twice andreanalyzed for purity of 90%. The population of CD44high/CD24lowwas used as the BCSCs in this study. For PKH26 staining,BCSCswere labeledwith PKH26 (Sigma-Aldrich; PKH26GL-1KT)dye according to themanufacturer's instructions. The labeled cellswere cultured under tumorsphere formation conditions for2 weeks; then, the tumorspheres were dissociated by accutaseand subjected to FACS.

Tumorsphere formation assayCells were plated at a density of 4,000 to 8,000 cells per well in

the 6-well ultralow attachment plates and cultured in tumor-sphere formation conditions [DMEM/F12 (Corning; 10-092-cv)supplemented with 10% serum replacement (SR; Thermo FisherScientific; 10828028), 20 ng/mL EGF, 5 ng/mL heparin sodium(MedChemExpress; 9041-08-1), 20 ng/mL bFGF (PeproTech; 96-100-18B-500)] at 37�C in a humidified 5% CO2 incubator.

Activation of qBCSCsFACS-sorted qBCSCs or chemoradioresistance qBCSCs were

cultured in tumorsphere formation conditions (described above)plus 50 ng/mL exosomes that has been isolated from the culturedmediumofMCF7 andHCC1937 cell lines. The exosome-depletedFBS was used in the culture of the MCF7 and HCC1937 cell lines.The total exosome was isolated from the cultured medium usingCellMediumExosome IsolationKit (Life Technologies; 4478359)according to themanufacturer's instructions and after 20 hours ofculture, the time just before one cell begins dividing into two.These cells were used as the A-qBCSCs in this study.

Tumorigenesis in miceFor mammary fat pad orthotopic xenograft experiments, 6 to

8 weeks of age (at the time of injections) female NOD/SCID micewere used. Cells were resuspended in HBSS/Matrigel (Corning;354234) and injected into the lower right and left mammary padsof each mouse. Animals were euthanized when the tumors wereapproximately 0.5 to1.2 cm in the largest diameter, to avoid tumornecrosis. The tumor width (w) and length (l) was recorded with acaliper and tumor sizewas calculated using the formula (l�w2/2).

Immunofluorescence and hematoxylin and eosin stainingSection samples were fixed with 4% paraformaldehyde and

embedded in OCT (Sakura; 4583). Cell samples were fixed with

Ye et al.

Cancer Res; 79(18) September 15, 2019 Cancer Research4730




4% paraformaldehyde. The samples were incubated with theappropriate antibodies. Tumor tissue sections were stained withhematoxylin and eosin (Beyotime, c0105) according to the man-ufacturer's instructions. Briefly, samples were stained in hema-toxylin staining solution for 10 minutes and stained in eosinstaining solution for 1 minute then dehydrated and sealed with aneutral gum and detected on the microscope.

Western blot analysis and qRT-qPCRTotal proteins were extracted by RIPA lysis buffer (Beyotime,

P0013B) containing protease inhibitor cocktail (MedChemEx-press; HY-K0010). Each protein sample (25 mg) was subjected toSDS-PAGE and then transferred to a nitrocellulose membrane forWestern blot analysis using Bio-Rad system, according to themanufacturer's instructions.

qRT-PCR reactions were performed on the Bio-Rad MiniOp-ticon system using SYBR Premix Ex Taq (TaKaRa; RR420A).Gene-specific primers were used (Supplementary Table S1). Therelative amounts of mRNAs were analyzed using the compar-ative CT method, as described previously (28).

Overexpression and RNAi of SETD4On the basis of the sequence of the human SETD4 gene

(NM_017438.4) in GenBank, the pLent-EF1a-SETD4-P2A-GFP-CMV-Purooverexpression plasmid (Vigene biosciences; LT88002)was synthesized and transfected with a viral packaging plasmid of10 mg of plasmid containing the vector of SETD4, 10 mg of pMD2.G (Addgene; 12259) and 15 mg of psPAX2 (Addgene; 12260) byLipo 3000 (Invitrogen; L3000015) into HEK293T cells overnight,and the viral supernatant was collected 48 hours later. The viralsupernatant was filtered through a 0.45-mm filter, and the freshlysorted BCSCs were used for infection in the presence of 5 mg/mLPolybrene (Sigma-Aldrich; 107689-10G). RNAi was performed toknockdown the overexpressed SETD4 gene at posttransfection day3, and images were obtained after further incubation with Scram-ble or SETD4 siRNA (Santa Cruz Biotechnology; sc-91446) for4 days, respectively. The SETD4 and control siRNAs (100 pmol/Lper 10,000 cells) were transfected into overexpressed SETD4BCSCs using the siRNA transfection system (Santa Cruz Biotech-nology; sc-45064), according to the manufacturer's instructions.

EdU incorporation assaySETD4þ qBCSCs asymmetric division was determined by the

incorporationofEdU(Ribobio;C00054). The SETD4þqBCSCs (n¼ 20) were incubated in tumorsphere formation medium con-taining 50mmol/L EdU for 40and60hours. All sampleswerefixedwith 4% paraformaldehyde for immunofluorescence analysis ofSETD4 or Ki67 by detection with second antibody conjugatedwith Alexa Fluor 594, and EdU was incubated with Alexa 488–conjugated Apollo staining reaction solution (Ribobio; C10310-3) for 30 minutes.

In vitro histone methylation transferase assayConstruction of GST-SETD4 and SETD4 mutants: The open

reading frame of GST-SETD4 and four GST-SETD4 mutants werecloned using specific primers (Supplementary Table S1). Thein vitro histone methylation transferase reactions were modifiedversions of protocols described previously (29) and were per-formed in 50 mL of methylase activity buffer containing 10 mg ofcore histones as substrates, 10 mmol/L AdoMet (Sangon; A6555-5g) as amethyl donor, andGST,GST-SETD4, dose-enhancedGST-

SETD4 and four mutant types of SETD4 as catalyzers. Afterincubation for 60 minutes at 30�C, the reaction products wereexamined by Western blotting or mass spectrometry analyses.

Transmission electron microscope analysisSamples were fixed with 2.5% glutaraldehyde overnight and

dehydrated with ethanol, then infiltrated with the mixture ofabsolute acetone and Spurr resin at room temperature. Afterinfiltration, samples were sectioned in ultratome (Leica EMUC7)and then observed using a transmission electron microscope(TEM; JEM-1230, JEOL Inc., Japan) at 80 kV. Images of the cellswere captured using a digital camera.

Chemical drug and radiation treatmentsCells were seeded into ultralow attachment 6-well plates at a

density of 8,000 cells per well. The cells were incubated with 100nmol/L Taxol (Sangon; A601183-0100) and 1 mmol/L 5-FU(Sangon; A100597-0001) for 10 days and/or exposed to X-raywith 30 Gy and cultured for the further 4 days. The medium waschanged every 2 days. After treatments, trypan blue (Beyotime;C0011) analysis was performed according to the manufacturer'sinstructions. The surviving cells were harvested using a dead cellremoval kit (Miltenyi Biotec; 130-090-101).

Single-cell RNA sequencingCellular suspensions were loaded on a single-cell instrument

(10x Genomics) to generate single-cell GEMs. Sequencing librar-ies were loaded on an Illumina Hiseq PE150 a 150 bp paired-endmodule. TheCell Ranger SingleCell Software Suite 1.3was used toperform sample demultiplexing, barcode processing and UMIcounting. All barcodes, the total UMI counts of which exceedm/10, were considered as cells. For visualizing data in 2-d space,Cell Ranger passes the PCA-reduced data into t-stochastic neigh-bor embedding (t-SNE). The graph-based clustering algorithmconsists of building a sparse nearest-neighbor graph, followed byLouvain modularity optimization (30). The value of k, the num-ber of nearest neighbors, is set to scale logarithmically with thenumber of cells.

Bulk-cell RNA sequencingThe RNAs of bulk cells were extracted using TRIzol, reverse-

transcribed, and included in the cDNA library. The library of bulkcells was sequenced on an Illumina Hiseq X Ten platform with a150bppaired-endmodule. Cuffdiff (v2.2.1)was used to calculateFPKMs for the coding genes of each sample. Genes with correctedP values less than 0.05 and the absolute value of |log2 (foldchange)| >2were assigned as significantly differentially expressed.The significantly differentially expressed genes were selected (alsoreported in the previous literature related to stem cell quiescenceregulations) and showed in the heatmap.GOenrichment analysisof differentially expressed genes was implemented with Perlmodule (GO::TermFinder). R functions (q value) were used totest for statistical enrichment of differentially expressed genesamong the Kyoto Encyclopedia of Genes and Genomes (KEGG)pathways. Gene set enrichment analysis (GSEA) was performedaccording to the instructions provided on the GSEA website(http://software.broadinstitute.org/gsea/).

Chromatin immunoprecipitation sequencingChromatin immunoprecipitation sequencing (ChIP-seq) was

performed using Anti-H4K20me3 according to the manufacturer's

SETD4-Defined Quiescent CSCs Originate Tumor and Relapse

www.aacrjournals.org Cancer Res; 79(18) September 15, 2019 4731



http://software.broadinstitute.org/gsea/


instructionsofMilliporeEZChIPKit (Millipore; 17-371). TheDNAwere constructed for the library and sequenced on an IlluminaHiseq 2500 platform. Peaks were called for aligned reads usingMACS2. Differentiated enriched peaks were analyzed using differ-ential peaks and notated using notate peaks.pl in the homersoftware. Heatmaps were created to present the differentiatedenrichedpeaks, according to thepeakenrichment value. Illustrativeread coverage graphs of H4K20me3 patterns across candidategenes were analyzed by the Integrative Genomics Viewer.

Assay for transposase-accessible chromatin with high-throughput sequencing

Assay for transposase-accessible chromatin with high-throughput sequencing (ATAC-seq) was performed as reportedpreviously (31). Briefly, nuclei were extracted from BCSCsSETD4

and BCSCsGFP, and the nuclei pellet was resuspended in the Tn5transposase reaction mix. The transposition reaction was incu-bated at 37�C for 30 minutes. Libraries were purified usingAMPure beads and then sequenced on an Illumina Hiseq X Tenplatform. The data analysis methods were described in the ChIP-seq sections above.

Quantification and statistical analysisFor quantification, at least three experiments were analyzed

using ImageJ software. All statistical analyses of the data wereperformed using means � SD. For statistical comparison, weperformed a one-tailed Student t test. The value of P < 0.05 wasconsidered significant [P > 0.05 considered not significant (NS)]and the exact P value is stated in the figures.

Data availabilityAll deep sequencing data that support the findings of this study

have been deposited in the Gene Expression Omnibus (GEO).The accession number for the RNA-seq data of BCSCs, qBCSCs,and A-qBCSCs is GSE123810. The RNA-seq data of BCSCsSETD4

and BCSCsGFP have been deposited under the accession codeGSE123842. The single cell RNA-seq data of BCSCs from solidtumors taken from mice and BCSCs from MCF7 cell lines areavailable under accession codes GSE124888 and GSE124887,respectively. ChIP-seq data have been deposited under accessioncode GSE123842. ATAC-seq data are available under accessioncodes GSE131586.

ResultsIdentification and characterization of quiescent BCSCs

On the basis of previous studies (32, 33), we obtained BCSCsby isolating a population of CD44high/CD24low cells using FACSfrom luminal type (MCF7) and basal type (HCC1937) humanbreast cancer cell lines (left FACS plots in Fig. 1A), whichexpressed high levels of pluripotency markers, such as ALDH-1,Sox2, Oct4, Nanog and displayed abilities of tumorsphere for-mation and tumorigenesis in NOD/SCID mice (SupplementaryFig. S1A–S1D). PKH26 has been used formarking nondividing orquiescent cells in previous reports (10, 12, 14) and in the currentstudy. A few cells with PKH26 label retention (PKH26þ) wereidentified in the BCSC formed tumorspheres (Fig. 1A), indicatingthat they were in a nondividing or quiescent state and named asqBCSCs. The PKH26þ qBCSCs isolated from the tumorspheres(middle FACS plots in Fig. 1A) could be activated in a tumor-sphere formation medium, named as activated qBCSCs (A-

qBCSCs), and it was these that subsequently formed the tumor-spheres (Fig. 1A). However, we did not observe tumorsphereformation or tumorigenesis in the PKH26� population dissoci-ated from the tumorspheres. We found that both the FACS-sortedqBCSCs and the qBCSCs in tumorspheres had very low expressionlevels of the proliferation markers Ki67 and PCNA (Fig. 1B–D),and very low phosphorylation levels of H3S10 (H3S10ph) andRbS807/S811 (RbS807/S811ph), in contrast to BCSCs and A-qBCSCs (Fig. 1E). Furthermore, FACS-sorted qBCSCs, BCSCs, andA-qBCSCs all exhibited CD44high/CD24low, ALDH-1high, andsimilar levels of the pluripotency markers (SupplementaryFig. S1E–S1G). Importantly, A-qBCSCs were more capable oftumorsphere formation and tumorigenesis than BCSCs (Fig. 1Fand G). However, no tumor formation was observed followinginjection of the same numbers of qBCSCs from the tumorspheres.In this study, we found that qBCSCs could be activated underconditions of tumorsphere formation medium and the additionof exosomes from the culture medium of MCF7 and HCC1937accelerated the activation of qBCSCs, especially for tumorigenesisof A-qBCSCs in mice. These results indicate that PKH26þ cells intumorspheres are qBCSCs and have a high tumorigenesis poten-tial upon activation.

Molecular signatures and specific expression of SETD4 inqBCSCs

To characterize themolecular signatures of qBCSCs, RNA-seqofBCSCs, qBCSCs, and A-qBCSCs from the MCF7 cell lines wereperformed. The differentially expressed genes in qBCSCs areshowed in the heatmap (Fig. 2A; Supplementary Fig. S2A). GSEAshowed downregulated expression of genes involved in cellactivation, proliferation, and signaling pathways of Wnt, TGFb,Notch, and JAK-STAT3, and upregulated expression of genes inBMP, p53, BMI1, hedgehog, Brac1, andHES1 in qBCSCpathways,as compared with A-qBCSCs (Fig. 2B; Supplementary Fig. S2B).This pattern has previously been reported as a signature ofquiescent stem cells (7, 8, 34–40). The differential expressionof related genes in qBCSCs was also validated by qRT-PCR(Supplementary Fig. S2C). GeneOntology (GO) analysis showedthat the expression of specific genes up- or downregulated inqBCSCs were significantly enriched for GO terms linked to theregulation of chromatin stability, proliferation, differentiation,metabolism, and related signaling pathways (SupplementaryFig. S2D).

In our previous report, we found that SETD4 was expressedabundantly in the quiescent cells of Artemia diapause embry-os (27). Similarly, we also found high expression levels of SETD4in qBCSCs (PKH26þ) as compared with PKH26� cells in tumor-spheres (Fig. 2C). To further confirm the specificity of SETD4expression in qBCSCs, we analyzed SETD4 expression in variouscells, including FACS-sorted BCSCs, qBCSCs, A-qBCSCs, and fourcancer cell lines (T47D, MKN45, MCF7, HCC1937). The resultsshowed that SETD4 was abundantly expressed only in qBCSCs,while its expressionwas very low inBCSCs andA-qBCSCs (Fig. 2Dand E) and remained undetected in any of the four cancer celllines or in the arrest state as triggered by starvation treatment(Supplementary Fig. S3A).

Because SETD4 was detected at a high level in qBCSCs, but at alow level in BCSCs, and to explore the function of SETD4 in theregulation of BCSC quiescence, GFP-fused SETD4 (GFP-SETD4)was overexpressed in BCSCs (BCSCsSETD4; SupplementaryFig. S3B). We found that the capability of tumorsphere formation

Ye et al.





and the expression levels of Ki67, H3S10ph, and RbS807/S811phin BCSCs were all inhibited by the overexpression of GFP-SETD4,but not inhibited in BCSCs that were overexpressing GFP(BCSCsGFP; Fig. 2F and G; Supplementary Fig. S3C). To validatethe function of overexpressed SETD4 in the regulation of BCSCquiescence, RNAi was performed to knock down the overex-

pressed SETD4gene. The BCSCsSETD4 treatedwith scramble siRNAweremaintained in a quiescent state and could not divide to formtumorspheres. In contrast, those treated with siRNA of SETD4produced tumorspheres in a similar manner to BCSCsGFP

(Fig. 2H). In addition, we found very low levels of H3S10ph andRbS807/S811ph and distinct lack of any Ki67 signal in

Figure 1.

Identification and isolation of BCSCs, qBCSCs, and A-qBCSCs fromMCF7 and HCC1937 cell lines. A, Discrimination of BCSCs, PKH26 labeling of qBCSCs, andactivation of the qBCSCs. Red, PKH26 retained cells. Scale bar, 50 mm. B and C, Immunofluorescent analysis of Ki67 (B) and PCNA (C) in BCSCs, qBCSCs, andA-qBCSCs, n¼ 3. Scale bar, 10 mm. D, Fluorescence microscopy of tumorsphere with PKH26 retention, immunostained for PCNA and Ki67. n¼ 10. Scale bar,20 mm. E,Western blot analysis of H3S10ph and RbS807/S811ph in BCSCs, qBCSCs, and A-qBCSCs. n¼ 3. F, Tumorsphere formation assay of BCSCs andA-qBCSCs. n¼ 3. Scale bar, 100 mm.G, Tumorigenesis rates of BCSCs, qBCSCs, and A-qBCSCs fromMCF7 cell line in NOD/SCIDmice.






Figure 2.

Molecular signatures of qBCSCs and specific expression of SETD4 in qBCSCs. A, Heatmap of differentially expressed genes in qBCSCs. The color bar representsthe log10(FPKMþ1). n¼ 3. B, GSEA of qBCSCs compared with A-qBCSCs. The green line shows the enrichment profile. C and D, Immunofluorescent analysis ofSETD4 in tumorsphere cells with PKH26 retention (C) and in FACS sorted BCSCs, qBCSCs, and A-qBCSCs (D). n¼ 3. Scale bar, 10 mm. E, SETD4 levels in BCSCs,qBCSCs, and A-qBCSCs, as determined byWestern blot analysis. n¼ 3. F, Representative images and rates of tumorspheres formed by BCSCsSETD4 andBCSCsGFP. n¼ 3. Scale bar, 100 mm. G, Immunofluorescent analysis of Ki67 in BCSCsGFP and BCSCsSETD4. n¼ 3. Scale bar, 20 mm. H, Representative images andrates of tumorspheres formed by siRNA (Si Scramble or Si SETD4)-treated BCSCsSETD4 and BCSCsGFP. Scale bar, 100 mm. n¼ 3.

Ye et al.





BCSCsSETD4 in contrast to the BCSCsGFP (Supplementary Fig. S3Dand S3E). These results indicate that SETD4 is required formaintenance of BCSCs quiescence.

SETD4 catalyzed H4K20me3 in heterochromatin formation inqBCSCs

We previously established that SETD4 catalyzes H4K20me3during diapause formation and regulates cell quiescence inArtemia (27). Here, we found a specific enrichment of H4K20me3in qBCSCs (PKH26þ) in tumorsphere and in FACS-sortedqBCSCs, but not in BCSCs or A-qBCSCs (Fig. 3A–C). Anin vitro histone methylation transferase assay showed that thelevel of H4K20me3 was enhanced upon the supplementation ofGST-SETD4 (Supplementary Fig. S4A–S4C). We repeated theassay using four SETD4 mutations, none of the four mutants ofSETD4 showed any methyltransferase activity on H4K20me3(Supplementary Fig. S4D). Moreover, we observed thatH4K20me3 was increased in BCSCsSETD4 but not in BCSCsGFP

(Fig. 3D; Supplementary Fig. S4E), and this effect was eliminatedby knockdown of the overexpressed GFP-SETD4 (Fig. 3E; Sup-plementary Fig. S4F). Likewise, SETD4 overexpression alsoinduced the increase of H4K20me3 in MCF7 and HCC1937 celllines (Supplementary Fig. S4G). These results indicate that SETD4functions in the specific catalysis of H4K20me3 in the qBCSCs.

Analysis of TEM revealed a striking increase of condensedheterochromatin in the nuclei of qBCSCs (Fig. 3F) andBCSCsSETD4 (Fig. 3G), in contrast to that in BCSCs, A-qBCSCsand BCSCsGFP. In addition, we observed increased H4K20me3 (amarker for cHC) and low levels of H3K9ac (a marker for euchro-matin) in qBCSCs (Fig. 3C; Supplementary Fig. S5A) andBCSCsSETD4 (Supplementary Fig. S5B–S5D), in contrast to thosein BCSCs, A-qBCSCs, and BCSCsGFP. Here, we found that HP1-a,which plays an essential role in heterochromatin formation (41),was also enriched in qBCSCs and BCSCsSETD4 (Fig. 3C; Supple-mentary Fig. S5A, S5B and S5E). However, we did not observe anysignificant differences in levels of the fHC markers of H3K9me3and H3K27me3 between BCSCs, qBCSCs, and A-qBCSCs orbetween BCSCsSETD4 and BCSCsGFP (Supplementary Fig. S5A,S5B, and S5F–S5I). These results indicate that qBCSCs andBCSCsSETD4 containmore cHC and less euchromatin than BCSCs,A-qBCSCs, and BCSCsGFP. Thus, we conclude that SETD4 controlsBCSCs quiescence by cHC formation via H4K20me3 catalysis.

H4K20me3 enhanced by SETD4 is located at certain promoterregions and regulates the expression of a set of genes in thequiescent BCSCs

In epigenetic regulation, local gene expression is influencedthrough modifications of chromatin that recruit transcriptionfactors that can either activate or repress gene transcription (42).To explore epigenetic regulation by H4K20me3 in BCSCs quies-cence, we performed ChIP-seq in BCSCsSETD4 and BCSCsGFP fromtheMCF7 cell line.We found that in BCSCsSETD4, H4K20me3wasdistributed on all 23 chromosomes (Supplementary Fig. S6A).Interestingly, quiescent BCSCsSETD4 showed marked enrichmentof H4K20me3 modifications relative to BCSCsGFP, a result com-patiblewithwidespread repressionof gene expression. Theoverallpattern of H4K20me3 modifications for the unique signaturegenes of BCSCsSETD4 (vs. BCSCsGFP) is shown as a heatmap(Supplementary Fig. S6B).

Illustrative read coverage graphs of H4K20me3 patterns acrosscandidate genes showed that in BCSCsSETD4, H4K20me3 was

typically enriched at the promoter regions and negatively corre-lated with the expression of the genes of MYC, WNT1, EEF1A1,IGF1, SMAD4, but was decreased at the promoter region andupregulated the expression of TP53 gene (Fig. 3H). In addition,ATAC-seq was performed. We found lower open chromatinenriched peaks in BCSCsSETD4 than that in BCSCsGFP, in whichBCSCsSETD4 showedmore widespread repressions of gene expres-sions relative to BCSCsGFP (Supplementary Fig. S6A and S6B).Notably, we found weak ATAC-seq signals at the MYC, WNT1,EEF1A1, IGF1, and SMAD4promoters inBCSCsSETD4,which likelyexplains their repressed gene expression status (Fig. 3I); however,the TP53 promoter exhibited stronger ATAC-seq signals inBCSCsSETD4 than BCSCsGFP, suggesting the upregulated geneexpression status of TP53 in BCSCsSETD4. The results of ATAC-seq indicated a lower chromatin accessibility and widespreadrepression of gene expression in BCSCsSETD4, which were alsoconsistent with the results of our prior H4K20me3 ChIP-seqanalysis (Fig. 3H).

We also compared the gene expression profiles of BCSCsSETD4

and BCSCsGFP using bulk RNA sequencing analysis (Supplemen-tary Fig. S6C). Consistent with our analysis of GO terms and qRT-PCR results, KEGG pathway analysis revealed that genes corre-latedwith cell activation and proliferationwere downregulated inBCSCsSETD4 (Supplementary Fig. S6D and S6E). Our data showthat SETD4-overexpressed BCSCs (BCSCsSETD4) are similar toqBCSCs in terms of its global gene expression pattern based onthe analysis of transcriptome (Fig. 2A and B; SupplementaryFig. S2A–S2D; Supplementary Fig. S6C–S6E). Thus, SETD4 pro-motes cHC formation in qBCSCs and epigenetically regulates theexpressionof a set of genes by catalyzing theH4K20me3 located atthe promoter regions.

SETD4-definedqBCSCs are resistant to chemoradiotherapy andcause tumor relapse

To investigate BCSCs' resistance to chemoradiotherapy,FACS-sorted BCSCs from MCF7 and HCC1937 cell lines weretreated with drugs and radiation. Interestingly, a few cells(average 3.28% and 2.92% of MCF7-BCSCs, average 3.64%and 3.38% of HCC1937-BCSCs) were survived after the drugand radiation treatments, respectively, as determined by trypanblue staining (Fig. 4A). Importantly, we found that all testedsurviving cells were SETD4 positive and Ki67 negative (Fig. 4Aand B), indicating that these surviving BCSCs were in quiescentstate or SETD4-defined qBCSCs. Analysis of TEM revealed thatthey contained more condensed heterochromatin (Fig. 4C). Inaddition, Western blot analysis revealed that these SETD4-defined qBCSCs had also abundant H4K20me3, HP1-a,and low amounts of H3K9ac, indicating that they contain morecHC and less euchromatin than did the BCSCs before treat-ments (Supplementary Fig. S7A). To confirm the resistanceof SETD4-defined qBCSCs to chemoradiotherapy, we per-formed drug and radiation treatments on FACS-sorted BCSCs,qBCSCs, and A-qBCSCs. As expected, FACS-sorted qBCSCs hadhigh survival rates (55.56% and 67.02%) after drug and radi-ation treatments compared with BCSCs (2.82% and 5.52%)and A-qBCSCs (0.42% and 1.03%; Fig. 4D). Moreover, over-expressed SETD4 enabled BCSCs to survive with resistance toboth treatments of chemical drugs and radiation, whereasBCSCsGFP were all sensitive to the treatments and exhibitedwidespread cell death (Fig. 4E). However, BCSCsSETD4 treatedwith siRNA to SETD4 lost chemoradiotherapy resistance and






Figure 3.

SETD4 catalyzed H4K20me3 in heterochromatin formation and epigenetic regulation in qBCSCs. A, Fluorescence microscopy and levels of immunostaining forH4K20me3 in the tumorsphere. n¼ 10. Scale bar, 10 mm. B, Levels of H4K20me3 (indicated by a box) and other methylation of histones in BCSCs, qBCSCs, andA-qBCSCs, as determined byWestern blot analysis. n¼ 3. C, Fluorescence microscopy and levels of BCSCs, qBCSCs, and A-qBCSCs immunostained forH4K20me3, H3K9ac, and HP1-a. n¼ 3. Scale bar, 10 mm. D and E, Levels of H4K20me3 (indicated by a box) in BCSCsGFP and BCSCsSETD4 (D) and in siRNA(Si Scramble and Si SETD4)-treated BCSCsSETD4 and BCSCsGFP (E) as determined byWestern blot analysis. n¼ 3. F and G, TEM images of condensedheterochromatin and percentages of heterochromatin in the nuclei of BCSCs, qBCSCs, and A-qBCSCs (F), and in BCSCsSETD4 and BCSCsGFP (G). n¼ 10. Scale bar,2 mm. H and I, Representative gene read coverage graphs of enriched H4K20me3 distribution by ChIP-seq (H) and the chromatin accessibility around thecandidate gene by ATAC-seq (I). Dashed boxes indicate the promoter regions of the genes.

Ye et al.





Figure 4.

SETD4-defined qBCSCs are resistant to chemoradiotherapy and cause tumor relapse. A, Representative images of BCSCs after treatments with drugs orradiation. Blue cells stained by trypan blue were dead cells. Red arrows, surviving cells. Immunofluorescence analysis of SETD4 and Ki67 in the surviving BCSCs.Scale bar, 20 mm. n¼ 3. B,Western blot analysis of SETD4 and Ki67 in BCSCs before treatment (pretreatment), surviving BCSCs after drug and radiationtreatments. n¼ 3. C, Representative TEM images and levels of heterochromatin in the nuclei of BCSCs before treatment (pretreatment), surviving cells after drugand radiation treatments. Scale bar, 2 mm. n¼ 10.D and E, Representative images and survival rates of BCSCs, qBCSCs, and A-qBCSCs (D), BCSCsSETD4 andBCSCsGFP (E) after treatments with drugs or radiation. Scale bar, 20 mm. n¼ 3. F, Chemotherapy and tumor relapse of MCF7-CDXs and HCC1937-CDXs.Representative tumor size of pre-chemo, post-chemo, and relapse in NOD/SCID mice. Scale bar, 0.5 cm. n¼ 3. G, Hematoxylin and eosin staining of pre-chemo,post-chemo, and relapse tumors. n¼ 3. Scale bar, 50 mm. H and I, Immunofluorescent analysis (H) andWestern blot analysis (I) of SETD4 in pre-chemo, post-chemo, and relapsed tumors. n¼ 3. Scale bar, 50 mm. J and K, Levels of CD44 and CD24 (J), and Ki67 (K) in pre-chemo, post-chemo, and relapsed tumors. n¼ 3.Scale bar, 50 mm.






died, but BCSCsSETD4 treated with scrambled siRNAmaintainedtheir quiescent state and retained resistance to the treatments(Supplementary Fig. S7B). These results indicated that SETD4-defined qBCSCs were resistant to chemoradiotherapy.

To reveal the cellular response of SETD4-defined qBCSCs tochemoradiotherapy, we also analyzed the activities of autophagy,apoptosis, and DNA damage after chemoradiotherapy. As shownin Supplementary Fig. S7C–S7E, no signals were detected forLC3B, activated caspase-3, and g-H2A in SETD4-defined qBCSCs,while high levels were observed in BCSCs and A-qBCSCs afterdrug and radiation treatment. The results indicate that SETD4-defined qBCSCs are resistant to chemoradiotherapy and displayno evident cellular damage beyond it, in contrast to BCSCs andA-qBCSCs.

We then investigated whether SETD4-defined qBCSCs wereable to cause tumor relapse in MCF7 and HCC1937 cell derivedxenografts (CDX) after development of resistance to chemother-apy. Tumors were generated in NOD/SCIDmice and had reachedtheir minimum size after chemotherapeutic treatment. Tumorshad then relapsed after chemotherapy's completion (Fig. 4F andG). Upon the completion of chemotherapy treatment, highpercentage of surviving tumor cells (34.82% in MCF7-CDXs and32.75% in HCC1937-CDXs) were SETD4þ in contrast to thesituation of prechemotherapy treatment tumors (1.08% inMCF7-CDXs and 1.32% in HCC1937-CDXs) and the relapsedtumors 2 weeks beyond treatment (2.71% in MCF7-CDXs and2.19% inHCC1937-CDXs; Fig. 4H). Analysis ofWestern blot alsoshowed significantly high expression level of SETD4 in the solidtumors after chemotherapy (Fig. 4I). Using immunofluorescenceanalysis, we found that SETD4þ cells in the tumors wereCD44high/CD24low (Fig. 4J; Supplementary Fig. S8A) and had alow level of Ki67 expression (Fig. 4K; Supplementary Fig. S8B).These results suggest that SETD4-defined qBCSCs in tumor areable to cause relapse beyond resistance to chemoradiotherapy. Inaddition, the chemotherapy-resistant SETD4þ qBCSCs in post-chemo treatment tumors had a high level of HP1-a expressionand the enrichment of H4K30me3, but a low abundance ofH3K9ac, indicating that they contained more cHC, but lesseuchromatin than that in pre-chemo treatment tumor cells orrelapsed tumor cells (Supplementary Fig. S8C–S8E).

SETD4-defined qBCSCs sustain themselves by asymmetricdivision

Interestingly, we found that these SETD4-defined qBCSCscould survive for more than 2months in the presence of chemicaldrugs, and sustained themselves and formed typical tumor-spheres after the drugs were removed in vitro (Fig. 5A). To addresshow SETD4-defined qBCSCs balance self-renewal during tumor-igenesis and relapse, two independent experiments were per-formed (Fig. 5B). In the first experiment, a SETD4-defined qBCSCdivided into two cells after approximately 40 hours activation.One cell was SETD4þ/EdU�/Ki67� and the other was SETD4�/EdUþ/Ki67þ, indicating that the SETD4þ qBCSC sustained itselfin a quiescent state with two original DNA strands by an asym-metric division, while producing a daughter cell. The daughter cellsubsequently divided symmetrically into two SETD4�/EdUþ/Ki67þ cells under conditions without EdU, indicating that thedaughter cell contains two newly synthesized DNA strands. In thesecond experiment, EdUwas added only after the SETD4þ qBCSChad alreadydivided into three cells. This resulted in the generationof one SETD4þ/EdU�/Ki67� cell and four SETD4�/EdUþ/Ki67þ

cells that could proliferate to produce a population of cells,indicating that the SETD4þ qBCSC is always maintained in aquiescent state during the proliferation process. We conclude thatSETD4-defined qBCSCs are able to sustain themselves by asym-metric division and concurrently produce a daughter cell, whichthen proliferates into a population of cancer cells by symmetricdivisions.

Identification of a distinct cluster of SETD4-defined qBCSCswithin the heterogeneous BCSCs population

Recent advances in single-cell gene expression analysis offers anopportunity to greatly improve the identification and classifica-tion of different cell types within a heterogeneous cell popula-tion (43, 44). We performed single-cell profiling of 3,765 BCSCsfrom tumors of MCF7-CDXs and identified clusters using t-SNEanalysis. BCSCs were distributed in seven clearly delineatedclusters, in which all clusters showed characteristic CD44high/CD24low expression (Fig. 6A). Importantly, theBCSCspopulationcontained 5.42% SETD4þ qBCSCs (204 cells) in the BCSC pop-ulation (3,765 cells) that had partitioned into a cluster (cluster 7)as a distinct cell type with high cellular component homogeneity.In addition, sequence data from 3,037 single BCSCs FACS-sortedfrom tumors of MCF7-CDXs showed that, after drug treatment,93.45% of the surviving BCSCs were in the defined cluster ofSETD4þ qBCSCs, and 6.55% were SETD4� and spread into otherclusters of BCSCs (Fig. 6B).

Differential gene expression analysis identified molecular sig-natures for each cell type and provided a comprehensive geneticmodule repertoire for the BCSCs population (Fig. 6C). Our t-SNEanalysis showed that in expression of related genes, some becomeenhanced (such asHES1, TP53, andBMP2) and others suppressed(such as MKI67, TGFBR3, and WNT10A) in the SETD4þ qBCSCscluster (Fig. 6D). Thismatched the expression known in quiescentcell types (7, 8, 34–40). We also identified new consensus mar-kers, including SETD4, ANGPTL4 (an inhibitor of tumor angio-genesis; ref. 45), and CA9 (a transmembrane protein of carbonicanhydrase; Fig. 6E; ref. 46). Similar results were also observedusing t-SNE analysis of single-cell sequences on 3,575 BCSCs and1,175 drug resistant qBCSCs derived from the MCF7 cell line(Supplementary Fig. S9A–S9E). Taken together, our results showthat SETD4-defined qBCSCs present as a small population(approximately 5%) in BCSCs, representing a distinct cell typewithin heterogeneous BCSCs and play critical functions in resis-tance to chemoradiotherapy and relapse.

SETD4-defined qBCSCs correlate with malignancy andchemotherapy resistance in clinic breast cancer patients and areidentified in multiple types of cancer

We next focused on the role of SETD4-defined qBCSCs intumorigenesis, chemotherapy resistant and relapse in clinicalbreast cancer patients. We obtained solid tumors that had beenremoved from patients with breast cancer, dissociated cancerouscells from them, and subjected these cells to treatment withchemotherapy drugs and radiation. We found that all survivingcells were SETD4þ, CD44high/CD24low and Ki67� (Fig. 7A), sug-gesting that they were SETD4-defined qBCSCs. Furthermore, thehigh levels of H4K20me3 and HP1-a and low levels of H3K9acindicated that these SETD4-defined qBCSCs carried higher con-tents of cHC and lower contents of euchromatin (SupplementaryFig. S10A). Subsequently, these SETD4-defined qBCSCs could beactivated in tumorsphere formation medium for 20 hours and

Ye et al.





then transplanted into NOD/SCID mice. Interestingly, 8 weeksafter injection of only 10 of these cells, tumors had all occurred inall such NOD/SCID mice (Fig. 7B). However, we did not observetumorigenesis when the same numbers of BCSCs from the sametumors of patients with breast cancer were injected. This indicates

that SETD4-defined qBCSCs have roles as the originators of tumorand relapse.

Analysis of clinical samples showed that the ratio of SETD4-defined qBCSCs in solid tumors from advanced stage (stage III)patients was more than 3-fold higher than in tumors obtained

Figure 5.

SETD4-defined qBCSCs sustain themselves by asymmetric division upon activation. A, Tumorsphere formation of chemotherapy-resistant SETD4-definedqBCSCs after withdrawing chemical drugs treatment during twomonths. n¼ 10. Scale bar, 50 mm. B,Asymmetric division of SETD4-defined qBCSCs.Fluorescence microscopy and levels of SETD4-defined qBCSCs during proliferation, immunostained for EdU, SETD4, and Ki67. Arrows, sustained quiescentBCSCs (EdU�/SETD4þ/Ki67�). n¼ 20. Scale bar, 10 mm.






from early stage (stages I and II) patients (Fig. 7C; SupplementaryFig. S10B). Moreover, the ratio of SETD4-defined qBCSCs wasmore than 3-fold higher in solid tumors obtained from patientswho had received chemotherapy treatment than in tumors frompatients who had not received treatment prior to surgery (Fig. 7D;Supplementary Fig. S10C). These results suggest that the presenceof SETD4-defined qBCSCs may correlate with malignancy andchemotherapy resistance in clinical breast cancer patients.Wenextassessed whether SETD4-defined quiescent CSCs were present in

other types of cancer. We obtained solid tumors that had beenremoved from patients with gastric, cervical, ovarian, liver, andlung cancers (Supplementary Fig. S10D), disaggregated the cells,and subjected them to the chemical drug treatments. We foundthat all chemotherapy-resistant cells tested from each of the fivesolid tumors were SETD4þ, Ki67� and high levels of the cancerstem cell marker (Fig. 7E), indicating that SETD4-defined quies-cent CSCs were also present in all examined cancer types. Ourresults are summarized in Fig. 7F.

Figure 6.

Identification of distinct clusters of SETD4-defined qBCSCs within BCSC population fromMCF7-CDXs.A, t-SNEmaps of BCSCs FACS sorted fromMCF7-CDXs,colored by cluster assignment. The SETD4þ qBCSCs cluster (blue) is outlined. t-SNEmaps of BCSCs, with cell color based on the expression of CD44 and CD24.Gene expression levels (log2 mean UMI) are indicated by shades of red. B, Classification of surviving BCSCs in BCSC population after drug treatment isshown in the t-SNEmaps. C, Heatmap of differential gene expression for each cell cluster. Color bar represents the log2-fold change.D and E, t-SNEmaps ofBCSCs, with cell color based on the expression of marker genes for quiescent cells (D) and newly identified consensus marker genes (E) for SETD4-definedqBCSCs.

Ye et al.





DiscussionTaken together, the results from our previous study on

diapause cysts of Artemia (27) and our current work on sixtypes of cancer reveal an evolutionarily conserved mechanismof cellular quiescence epigenetically controlled by SETD4.Although Suv4-20h2 has previously been reported as respon-sible for catalyzing H4K20me3 in mouse and human fibro-blasts (47, 48), we did not observe any significant differences inSuv4-20h2 expression in response to H4K20me3 enrichment inqBCSCs (Supplementary Fig. S11A). We also did not observeSETD4 expression in response to the increase in H4K20me3 inthe quiescence of mouse embryonic fibroblasts induced bycontact-inhibition, in which Suv4-20h2 catalyzes H4K20me3(Supplementary Fig. S11B). It seems that both Suv4-20h2 andSETD4 are able to catalyze H4K20me3, but they function in

different types of cell, although SETD4 has been studied in theregulation of gene expression and cell proliferation in severalcell lines (24, 49). We propose that SETD4 is a determinant ofquiescence specifically in CSCs, but not in cancer cells. Thus,SETD4 can be applied to define qCSCs occurring within thelarge heterogeneity of tumor cells and even, more specificallywithin the wider CSCs population.

In this study, we did not find any tumor occurrence by injectionof qBCSCsduring theperiodof 6months.Our results indicate thatthe activation of qBCSCs is required for tumor occurrence. On thebasis of evidence of a strong correlation between SETD4-definedqBCSCs and malignancy and chemoresistance in patients withbreast cancer, it may be possible to use the SETD4 and/or SETD4-defined qBCSCs as important indicators for assessing the grade ofmalignancy and likelihood of tumor recurrence in a clinical

Figure 7.

SETD4-defined qBCSCs correlate with malignancy and chemotherapy resistance in clinic breast cancer patients and are identified in multiple types of cancer. A,Fluorescence microscopy and levels of SETD4, Ki67, CD44, and CD24 in the surviving cells isolated from a solid tumor of breast cancer patient. n¼ 3. Scale bar,10 mm. B, Ten of A-qBCSCs were used to generate tumors in NOD/SCID mice. C and D, Analysis of SETD4-defined qBCSCs in tumors from different stages ofbreast cancer patients (C) and in tumors from breast cancer patients who had or had not received chemotherapy (D). Each dot represents data of a patient.n¼ 8 patients. E, Fluorescence microscopy of SETD4-defined quiescent CSCs in multiple types of cancer. n¼ 3. Scale bar, 10 mm. F, Summary of this study.






setting. In the current study, we found that the ability of chemor-adiotherapy resistance disappeared completely after SETD4-defined qBCSCs were activated, and therefore activating BCSCsmay enable their eradication by subsequent treatments withstandard chemoradiotherapy. Our findings suggest that SETD4and/or SETD4-defined qCSCs could be also used as key targets inclinical treatment for a wide range of cancers.

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

Authors' ContributionsConception and design: S. Ye, W.-J. YangDevelopment of methodology: S. Ye, Y.-F. Ding, W.-H. Jia, X.-L. Liu, J.-Y. FengAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): Q. Zhu, S.-L. Cai, Y.-S. Yang, Q.-Y. Lu, X.-T. Huang,Y.-H. WangAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): S. Ye, Y.-F. Ding, J.-S. YangWriting, review, and/or revision of the manuscript: S. Ye, W.-J. Yang

Administrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): S.-N. Jia, G.-P. Ding, J.-J. Zhou, Y.-D. ChenStudy supervision: W.-J. Yang

AcknowledgmentsWe thank S. Zhang for the help with the laser microscopy, X. Song for flow

cytometry analysis, J. Li for transmission electron microscopy, Y. Xu for massspectrometer analysis, and X. Xu for mouse husbandry support. We would liketo express our sincere gratitude to Mr. C. Wood for critical reading of themanuscript. This work was supported by the National Major Research andDevelopment Project (2016YFA0101201) and the National Natural ScienceFoundation of China (31730084).

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 April 4, 2019; revised June 6, 2019; accepted July 10, 2019;published first July 15, 2019.

References1. MeachamCE,Morrison SJ. Tumour heterogeneity and cancer cell plasticity.

Nature 2013;501:328–37.2. Dagogo-Jack I, Shaw AT. Tumour heterogeneity and resistance to cancer

therapies. Nat Rev Clin Oncol 2018;15:81–94.3. Early Breast Cancer Trialists' Collaborative G. Long-term outcomes for

neoadjuvant versus adjuvant chemotherapy in early breast cancer: meta-analysis of individual patient data from ten randomised trials. LancetOncol 2018;19:27–39.

4. Holohan C, Van Schaeybroeck S, Longley DB, Johnston PG. Cancer drugresistance: an evolving paradigm. Nat Rev Cancer 2013;13:714–26.

5. Klement GL. Eco-evolution of cancer resistance. Sci Transl Med 2016;8:327fs5

6. Batlle E, Clevers H. Cancer stem cells revisited. Nat Med 2017;23:1124–34.

7. Chen W, Dong J, Haiech J, Kilhoffer MC, Zeniou M. Cancer stem cellquiescence and plasticity as major challenges in cancer therapy. Stem CellsInt 2016;2016:1740936.

8. Brown JA, Yonekubo Y, Hanson N, Sastre-Perona A, Basin A, Rytlewski JA,et al. TGF-beta-induced quiescence mediates chemoresistance of tumor-propagating cells in squamous cell carcinoma. Cell Stem Cell 2017;21:650–64.

9. Chen J, Li Y, Yu TS, McKay RM, Burns DK, Kernie SG, et al. A restricted cellpopulation propagates glioblastoma growth after chemotherapy. Nature2012;488:522–6.

10. Clevers H. The cancer stem cell: premises, promises and challenges.Nat Med 2011;17:313–9.

11. Cheung TH, Rando TA. Molecular regulation of stem cell quiescence.Nat Rev Mol Cell Biol 2013;14:329–40.

12. Pece S, Tosoni D, Confalonieri S, Mazzarol G, Vecchi M, Ronzoni S, et al.Biological and molecular heterogeneity of breast cancers correlates withtheir cancer stem cell content. Cell 2010;140:62–73.

13. Cicalese A, Bonizzi G, Pasi CE, Faretta M, Ronzoni S, Giulini B, et al. Thetumor suppressor p53 regulates polarity of self-renewing divisions inmammary stem cells. Cell 2009;138:1083–95.

14. Kusumbe AP, Bapat SA. Cancer stem cells and aneuploid populationswithin developing tumors are themajor determinants of tumor dormancy.Cancer Res 2009;69:9245–53.

15. Li L, Clevers H. Coexistence of quiescent and active adult stem cells inmammals. Science 2010;327:542–5.

16. BoonsanayV, Zhang T,Georgieva A, Kostin S,QiH, YuanX, et al. Regulationof skeletal muscle stem cell quiescence by Suv4-20h1-dependent facultativeheterochromatin formation. Cell Stem Cell 2016;18:229–42.

17. Liu L, Cheung TH, Charville GW, Hurgo BMC, Leavitt T, Shih J, et al.Chromatin modifications as determinants of muscle stem cell quiescenceand chronological aging. Cell Rep 2013;4:189–204.

18. Trojer P, Reinberg D. Facultative heterochromatin: Is there a distinctivemolecular signature? Mol Cell 2007;28:1–13.

19. Schotta G, Lachner M, Sarma K, Ebert A, Sengupta R, Reuter G, et al. Asilencing pathway to induce H3-K9 and H4-K20 trimethylation at consti-tutive heterochromatin. Gene Dev 2004;18:1251–62.

20. Peters AHFM, Mermoud JE, O'Carroll D, Pagani M, Schweizer D, Brock-dorff N, et al. Histone H3 lysine 9 methylation is an epigenetic imprint offacultative heterochromatin. Nat Genet 2002;30:77–80.

21. Evertts AG, Manning AL, Wang X, Dyson NJ, Garcia BA, Coller HA. H4K20methylation regulates quiescence and chromatin compaction. Mol BiolCell 2013;24:3025–37.

22. Qian C, Zhou MM. SET domain protein lysine methyltransferases: struc-ture, specificity and catalysis. Cell Mol Life Sci 2006;63:2755–63.

23. Herz HM, Garruss A, Shilatifard A. SET for life biochemical activities andbiological functions of SET domain-containing proteins. Trends BiochemSci 2013;38:621–39.

24. Faria JA, Correa NC, de Andrade C, de Angelis Campos AC, Dos SantosSamuel deAlmeidaR, Rodrigues TS, et al. SETdomain-containing protein 4(SETD4) is a newly identified cytosolic and nuclear lysine methyltransfer-ase involved in breast cancer cell proliferation. J Cancer Sci Ther 2013;5:58–65.

25. O'Neill DJ, Williamson SC, Alkharaif D, Monteiro IC, Goudreault M,Gaughan L, et al. SETD6 controls the expression of estrogen-responsivegenes and proliferation of breast carcinoma cells. Epigenetics 2014;9:942–50.

26. Clegg JS, Trotman CAN. Physiological and biochemical aspects of Artemiaecology, p129–170. In: Abatzopoulos THJ, Beardmore JA, Clegg JS, Sorge-loos P. Artemia, basic and applied biology. Dordrecht, the Netherlands:Kluwer Academic Publishers; 2002.

27. Dai L, Ye S, Li HW, Chen DF, Wang HL, Jia SN, et al. SETD4regulates cell quiescence and catalyzes the trimethylation ofH4K20 during diapause formation in Artemia. Mol Cell Biol2017;37:e00453–16.

28. Schmittgen TD, Livak KJ. Analyzing real-time PCR data by the comparativeC(T) method. Nat Protoc 2008;3:1101–8.

29. Rea S, Eisenhaber F, O'Carroll D, Strahl BD, Sun ZW, Schmid M, et al.Regulation of chromatin structure by site-specific histone H3methyltrans-ferases. Nature 2000;406:593–9.

30. Blondel VD, Guillaume JL, Lambiotte R, Lefebvre E. Fast unfoldingof communities in large networks. J Stat Mech-Theory E 2008;10.doi: 10.1088/1742-5468/2008/10/P10008.

31. Buenrostro JD,Giresi PG, Zaba LC, ChangHY,GreenleafWJ. Transpositionof native chromatin for fast and sensitive epigenomic profiling of openchromatin, DNA-binding proteins and nucleosome position. NatMethods2013;10:1213–8.

Ye et al.





32. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF.Prospective identification of tumorigenic breast cancer cells. Proc NatlAcad Sci U S A 2003;100:3983–8.

33. Ponti D, Costa A, Zaffaroni N, Pratesi G, Petrangolini G, Coradini D, et al.Isolation and in vitro propagation of tumorigenic breast cancer cells withstem/progenitor cell properties. Cancer Res 2005;65:5506–11.

34. Wang C, Guo X, Xi R. EGFR and Notch signaling respectively regulateproliferative activity andmultiple cell lineage differentiation ofDrosophilagastric stem cells. Cell Res 2014;24:610–27.

35. Peng T, FrankDB, Kadzik RS,MorleyMP, Rathi KS,Wang T, et al. Hedgehogactively maintains adult lung quiescence and regulates repair and regen-eration. Nature 2015;526:578–82.

36. Chen D, Wu M, Li Y, Chang I, Yuan Q, Ekimyan-Salvo M, et al. TargetingBMI1(þ) cancer stem cells overcomes chemoresistance and inhibitsmetas-tases in squamous cell carcinoma. Cell Stem Cell 2017;20:621–34.

37. Liu Y, Elf SE, Miyata Y, Sashida G, Liu Y, Huang G, et al. p53 regulateshematopoietic stem cell quiescence. Cell Stem Cell 2009;4:37–48.

38. Sang L, Coller HA, Roberts JM. Control of the reversibility of cellular quies-cence by the transcriptional repressor HES1. Science 2008;321:1095–100.

39. Wu MG, Fernandez N, Celebre A, Srikanth M, Kessler J, Karamchandani J,et al. Bmp signaling regulates a quiescent cancer stem cell phenotyperesponsible for treatment resistance in glioblastoma. Neuro Oncol2014;16(suppl 5)v204.

40. Mendelson A, Frenette PS. Hematopoietic stem cell niche maintenanceduring homeostasis and regeneration. Nat Med 2014;20:833–46.

41. Boros J, Arnoult N, Stroobant V, Collet JF, Decottignies A. Polycombrepressive complex 2 and H3K27me3 cooperate with H3K9 methylation

to maintain heterochromatin protein 1 alpha at chromatin. Mol Cell Biol2014;34:3662–74.

42. Berger SL. The complex language of chromatin regulation during tran-scription. Nature 2007;447:407–12.

43. Han X, Wang R, Zhou Y, Fei L, Sun H, Lai S, et al. Mapping the Mouse CellAtlas by Microwell-Seq. Cell 2018;173:1307.

44. Plasschaert LW, Zilionis R, Choo-Wing R, Savova V, Knehr J, Roma G, et al.A single-cell atlas of the airway epithelium reveals the CFTR-rich pulmo-nary ionocyte. Nature 2018;560:377–81.

45. Okochi-TakadaE,HattoriN, TsukamotoT,MiyamotoK,AndoT, Ito S, et al.ANGPTL4 is a secreted tumor suppressor that inhibits angiogenesis.Oncogene 2014;33:2273–8.

46. Hussain SA, Ganesan R, Reynolds G, Gross L, Stevens A, Pastorek J, et al.Hypoxia-regulated carbonic anhydrase IX expression is associated withpoor survival in patients with invasive breast cancer. Brit J Cancer 2007;96:104–9.

47. Schotta G, Sengupta R, Kubicek S, Malin S, Kauer M, Callen E, et al. Achromatin-wide transition to H4K20 monomethylation impairs genomeintegrity and programmed DNA rearrangements in the mouse. Genes Dev2008;22:2048–61.

48. Yang HB, Pesavento JJ, Starnes TW, Cryderman DE, Wallrath LL, KelleherNL, et al. Preferential dimethylation of histone H4 lysine 20 by Suv4-20.J Biol Chem 2008;283:12085–92.

49. Li GM, Wang YG, Pan Q, Wang J, Fan JG, Sun C. RNAi screening withshRNAs against histone methylation-related genes reveals determinants ofsorafenib sensitivity in hepatocellular carcinoma cells. Int J Clin Exp Pathol2014;7:1085–92.






2019;79:4729-4743. Published OnlineFirst July 15, 2019.Cancer Res Sen Ye, Yan-Fu Ding, Wen-Huan Jia, et al. Cancer Stem Cell Quiescence

Containing Protein 4 Epigenetically Controls Breast−SET Domain

Updated version

10.1158/0008-5472.CAN-19-1084doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2019/07/13/0008-5472.CAN-19-1084.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/79/18/4729.full#ref-list-1

This article cites 47 articles, 11 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://cancerres.aacrjournals.org/content/79/18/4729To 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-19-1084

http://cancerres.aacrjournals.org/content/suppl/2019/07/13/0008-5472.CAN-19-1084.DC1

http://cancerres.aacrjournals.org/content/79/18/4729.full#ref-list-1

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/79/18/4729


Documents

SET Domain Containing Protein 4 Epigenetically Controls ......Tumor Biology and Immunology SET Domain–Containing Protein 4 Epigenetically Controls Breast Cancer Stem Cell Quiescence