Thank you for visiting Nature. The browser version you are using has limited support for CSS. For the best experience, we recommend that you use a newer version of the browser (or turn off the compatibility mode in Internet Explorer). At the same time, to ensure continued support, we will display sites without styles and JavaScript.
Extrachromosomal DNA (ecDNA) is ubiquitous in human cancers and mediates the high expression of oncogenes through gene amplification and altered gene regulation1. Gene induction usually involves contacting and activating cis-regulatory elements of genes on the same chromosome2,3. Here, we show that the ecDNA center—a cluster of approximately 10-100 ecDNAs in the nucleus—makes intermolecular enhancer-gene interactions promote oncogene overexpression. The ecDNAs encoding multiple different oncogenes form centers in different cancer cell types and primary tumors. When spatially aggregated with other ecDNAs, each ecDNA is more likely to transcribe oncogenes. The center of the ecDNA is bound by the bromodomain and extra-terminal domain (BET) protein BRD4 in colorectal cancer cell lines amplified by MYC. The BET inhibitor JQ1 disperses the ecDNA center and preferentially inhibits the transcription of ecDNA-derived oncogenes. The PVT1 promoter bound by BRD4 is ectopicly fused with MYC and replicated in ecDNA, accepting mixed enhancer input to drive the effective expression of MYC. In addition, the PVT1 promoter on the exogenous episome is sufficient to mediate trans gene activation in a JQ1-sensitive manner through the ecDNA center. Silencing ecDNA enhancers by the CRISPR interference system revealed the intermolecular enhancer-gene activation between multiple oncogene loci amplified on different ecDNAs. Therefore, the ecDNA hub connected to the protein can perform intermolecular transcriptional regulation, and can be used as a unit of oncogene function and co-evolution, as well as a potential target for cancer treatment.
Get full access to the journal for 1 year
All prices are net prices. VAT will be added at checkout later. Tax calculation will be done at checkout.
Get time-limited or full article access on ReadCube.
All prices are net prices.
The ChIP-seq, HiChIP, Hi-C, RNA-seq and 10x Chromium Single Cell Multiome ATAC Gene Expression data generated in this study have been stored in GEO and are available under the accession number GSE159986. Nanopore sequencing data, WGS data, sgRNA sequencing data, targeted ecDNA sequencing data after CRISPR-Cas9 digestion, and PFGE generated in this study have been deposited in SRA and obtained under the deposit number PRJNA670737. The optical mapping data generated in this study has been deposited in GenBank, and the BioProject code is PRJNA731303. This study also used the following public data: TR14 H3K27ac ChIP–seq93 (GEO: GSE90683); COLO320-DM, COLO320-HSR and PC3 WGS1 (SRA: PRJNA506071); SNU16 WGS60 (SRA: PRJNA523380); and HK359 WGS6 (SRA: PRJNA338012). The microscope image file is available on figshare at https://doi.org/10.6084/m9.figshare.c.5624713.
The custom code used in this study is available at https://github.com/ChangLab/ecDNA-hub-code-2021.
Wu, S. et al. Round ecDNA promotes accessible chromatin and high oncogene expression. Nature 575, 699–703 (2019).
Gorkin, DU, Leung, D. & Ren, B. 3D genome in transcriptional regulation and pluripotency. Cell Stem Cell 14, 762–775 (2014).
CAS PubMed PubMed Central Google Scholar
Zheng, H. & Xie, W. The role of 3D genome organization in development and cell differentiation. Nat. Rev. Moore Cell Biology 20, 535–550 (2019).
Bailey, C., Shoura, MJ, Mischel, PS & Swanton, C. Extrachromosomal DNA—Remove genetic restrictions and accelerate tumor evolution. install. Tumor. 31, 884–893 (2020).
Kim, H. et al. Extrachromosomal DNA is associated with oncogene amplification and poor outcomes in many cancers. Nat. Gene. 52, 891–897 (2020).
CAS PubMed PubMed Central Google Scholar
Turner, KM, etc. Extrachromosomal oncogene amplification drives tumor evolution and genetic heterogeneity. Nature 543, 122–125 (2017).
CAS PubMed PubMed Central ADS Google Scholar
Verhaak, RGW, Bafna, V. & Mischel, PS Tumor pathogenesis and evolution of extrachromosomal oncogene amplification. Nat. Cancer Pastor 19, 283–288 (2019).
CAS PubMed PubMed Central Google Scholar
Cox, D., Yuncken, C. & Spriggs, AI Tiny chromatin bodies in childhood malignant tumors. The Lancet 286, 55–58 (1965).
van der Bliek, AM, Lincke, CR & Borst, P. 3T6R50 Double minichromosome circular DNA. Nucleic acid research. 16, 4841–4851 (1988).
PubMed PubMed Central Google Scholar
Hamkalo, BA, Farnham, PJ, Johnston, R. & Schimke, RT Ultrastructural characteristics of microchromosomes in the 3T3 cell line of methotrexate-resistant mice. Process National Academy of Sciences. science. 82, 1126–1130 (1985).
CAS PubMed PubMed Central ADS Google Scholar
Maurer, BJ, Lai, E., Hamkalo, BA, Hood, L. & Attardi, G. New submicroscopic extrachromosomal elements containing amplified genes in human cells. Nature 327, 434–437 (1987).
CAS PubMed ADS Google Scholar
VanDevanter, DR, Piaskowski, VD, Casper, JT, Douglass, EC & Von Hoff, DD The ability of circular extrachromosomal DNA molecules to carry the amplified MYCN proto-oncogene in human neuroblastoma in vivo. J Natl Cancer Institute. 82, 1815–1821 (1990).
Nathanson, DA etc. Targeted therapy resistance mediated by the dynamic regulation of extrachromosomal mutations in EGFR DNA. Science 343, 72–76 (2014).
CAS PubMed ADS Google Scholar
Ståhl, F., Wettergren, Y. and Levan, G. Amplicon structure in multidrug-resistant mouse cells: non-rearranged regions of genomic DNA corresponding to large circular DNA. Mole. cell. biology. 12, 1179–1187 (1992).
PubMed PubMed Central Google Scholar
Vicario, R. etc. The pattern of HER2 gene amplification and response to anti-HER2 therapy. PLoS ONE 10, e0129876 (2015).
PubMed PubMed Central Google Scholar
Carol, SM etc. Double minichromosomes can be produced from precursors derived from chromosomal deletions. Mole. cell. biology. 8, 1525–1533 (1988).
CAS PubMed PubMed Central Google Scholar
Kitajima, K., Haque, M., Nakamura, H., Hirano, T. & Utiyama, H. In the HL-60 subline, the irreversible loss of DMSO-induced granulocyte differentiation has uniformly stained areas. Biochemical. Biophysics. Reservoir community. 288, 1182–1187 (2001).
Quinn, LA, Moore, GE, Morgan, RT & Woods, LK is a cell line derived from human colon cancer with unusual cell products, two minutes and evenly stained areas. Cancer research. 39, 4914–4924 (1979).
Storazzi, CT etc. Gene amplification in two-minute or uniformly stained regions in solid tumors: origin and structure. Genome research. 20, 1198–1206 (2010).
CAS PubMed PubMed Central Google Scholar
Wahl, GM The importance of circular DNA in mammalian gene amplification. Cancer research. 49, 1333–1340 (1989).
Kumar, P. et al. ATAC-seq recognizes thousands of extrachromosomal circular DNAs in cancers and cell lines. science. Advanced 6. eaba2489 (2020).
CAS PubMed PubMed Central ADS Google Scholar
Morton, AR, etc. Functional enhancers shape the amplification of extrachromosomal oncogenes. Cell 179, 1330–1341 (2019).
CAS PubMed PubMed Central Google Scholar
Helmsauer, K. et al. Enhancer hijacking determines the extrachromosomal ring MYCN amplicon structure in neuroblastoma. Nat. Community. 11, 5823 (2020).
CAS PubMed PubMed Central ADS Google Scholar
Itoh, N. & Shimizu, N. DNA replication-dependent intranuclear relocation of double microchromatin. J. Cell Science. 111, 3275–3285 (1998).
Kanda, T., Sullivan, KF & Wahl, GM Histone-GFP fusion protein can perform sensitive analysis of chromosome dynamics in living mammalian cells. Curry creatures. 8, 377–385 (1998).
Oobatake, Y. & Shimizu, N. Double-strand breaks within two minutes outside the chromosomes trigger their aggregation, micronucleation and morphological changes in the nucleus. Genetic Chromosome Cancer 59, 133–143 (2020).
Beliveau, BJ etc. A versatile design and synthesis platform for visualizing the genome using Oligopaint FISH probes. Process National Academy of Sciences. science. United States 109, 21301–21306 (2012).
CAS PubMed PubMed Central ADS Google Scholar
Koche, RP, etc. Extrachromosomal circular DNA drives remodeling of the oncogenic genome in neuroblastoma. Nat. Genetics 52, 29–34 (2019).
Parker, SCJ, etc. The chromatin stretch enhancer state drives cell-specific gene regulation and hides human disease risk variants. Process National Academy of Sciences. science. United States 110, 17921–17926 (2013).
CAS PubMed PubMed Central ADS Google Scholar
White, WA etc. Master transcription factors and mediators establish super enhancers at key cell identity genes. Cell 153, 307–319 (2013).
CAS PubMed PubMed Central Google Scholar
Lowen, J. et al. Selective inhibition of tumor oncogenes by destroying super enhancers. Cell 153, 320–334 (2013).
PubMed PubMed Central Google Scholar
Filippakopoulos, P. etc. Selective inhibition of BET bromodomain. Nature 468, 1067–1073 (2010).
CAS PubMed PubMed Central ADS Google Scholar
Sabari, BR etc. The co-activator condensation in the super enhancer links phase separation with gene control. Science 361, eaar3958 (2018).
PubMed PubMed Central Google Scholar
Ren, C. etc. Spatially constrained tandem bromodomain inhibition supports sustained inhibition of BRD4 transcriptional activity in TNBC cell growth. Process National Academy of Sciences. science. United States 115, 7949–7954 (2018).
CAS PubMed PubMed Central Google Scholar
Deshpande, V. etc. Use AmpliconArchitect to explore the pattern of focus magnification in cancer. Nat. Community. 10, 392 (2019).
CAS PubMed PubMed Central ADS Google Scholar
Luebeck, J. et al. AmpliconReconstructor integrates NGS and optical mapping to resolve the complex structure of focus magnification. Nat. Community. 11, 4374 (2020).
CAS PubMed PubMed Central ADS Google Scholar
Schwab, M., Klempnauer, KH, Alitalo, K., Varmus, H. & Bishop, M. Amplification of c-myc 5'rearrangement in human COLO 320 cells is associated with abnormal transcription. Mole. cell. biology. 6, 2752–2755 (1986).
CAS PubMed PubMed Central Google Scholar
L'Abbate, A. etc. Genomic organization and evolution of two-minute/uniformly stained regions with MYC amplification in human cancers. Nucleic acid research. 42, 9131–9145 (2014).
PubMed PubMed Central Google Scholar
Hann, SR, King, MW, Bentley, DL, Anderson, CW & Eisenman, RN The initiation of non-AUG translation in c-myc exon 1 produces a unique N-terminal protein, which is synthesized in Burkitt’s lymph The tumor was destroyed. Cell 52, 185–195 (1988).
Carramusa, L. etc. The PVT-1 oncogene is the target of Myc protein and is overexpressed in transformed cells. J. Cells. physiological. 213, 511–518 (2007).
Cho, SW et al. The promoter of lncRNA gene PVT1 is a tumor suppressor DNA border element. Cell 173, 1398-1412 (2018).
CAS PubMed PubMed Central Google Scholar
Tolomeo, D., Agostini, A., Visci, G., Traversa, D. & Storlazzi, CT PVT1: a long non-coding RNA that repeatedly participates in tumor formation-related fusion transcripts. Gene 779, 145497 (2021).
Mumbach, MR, etc. HiChIP: Efficient and sensitive analysis of protein-oriented genome structure. Nat. Method 13, 919 (2016).
CAS PubMed PubMed Central Google Scholar
Furco, CP etc. Enhancer-promoter regulation activity-contact model from thousands of CRISPR perturbations. Nat. Gene. 51, 1664–1669 (2019).
CAS PubMed PubMed Central Google Scholar
Park, J. et al. CD44 and FGFR2 control the growth of gastric cancer cells through the mutual regulation circuit of c-myc. Oncotarget 7, 28670–28683 (2016).
PubMed PubMed Central Google Scholar
Furlong, EEM & Levine, M. Developmental enhancer and chromosome topology. Science 361, 1341–1345 (2018).
CAS PubMed PubMed Central ADS Google Scholar
Zhu, Y. etc. Oncogenic extrachromosomal DNA acts as a mobile enhancer to amplify chromosomal transcription globally. Cancer cell 39, 694–707 (2021).
Xue, KS, Hooper, KA, Ollodart, AR, Dingens, AS & Bloom, JD The cooperation between different virus variants promotes the growth of H3N2 influenza in cell culture. Elife 5, e13974 (2016).
PubMed PubMed Central Google Scholar
Vignuzzi, M., Stone, JK, Arnold, JJ, Cameron, CE & Andino, R. Quasispecies diversity determines pathogenesis through cooperative interactions in virus populations. Nature 439, 344–348 (2006).
CAS PubMed ADS Google Scholar
Henssen, A. etc. Inhibition of targeted MYCN-driven transcription by the BET bromodomain. clinical. Cancer research. 22, 2470–2481 (2016).
Xie, L. etc. 3D ATAC-PALM: accessible super-resolution imaging of the genome. Nat. Method 17,430–436 (2020).
CAS PubMed PubMed Central Google Scholar
Ambrose, PF etc. International consensus on the molecular diagnosis of neuroblastoma: a report from the International Neuroblastoma Risk Group (INRG) Biology Committee. Brother J. Cancer 100, 1471–1482 (2009).
CAS PubMed PubMed Central Google Scholar
Balaban-Malenbaum, G. & Gilbert, F. Double microchromosomes and uniformly stained regions in the chromosomes of human neuroblastoma cell lines. Science 198, 739–741 (1977).
CAS PubMed ADS Google Scholar
Marrano, P., Irwin, MS & Thorner, PS The heterogeneity of MYCN amplification in neuroblastoma during diagnosis, treatment, recurrence, and metastasis. Genetic Chromosome Cancer 56, 28–41 (2017).
Villamon, E. et al. MYCN expansion increases the genetic instability and intratumoral heterogeneity of neuroblastoma with 11q deletion. PLoS ONE 8, e53740 (2013).
PubMed PubMed Central ADS Google Scholar
Schindelin, J. et al. Fiji: An open source platform for biological image analysis. Nat. Method 9, 676–682 (2012).
CAS PubMed PubMed Central Google Scholar
Rajkumar, U. etc. EcSeg: Semantic segmentation of metaphase images containing extrachromosomal DNA. Iscience 21, 428–435 (2019).
CAS PubMed PubMed Central ADS Google Scholar
Wikki, SL, etc. The correlation function quantifies the super-resolution image and estimates the obvious clustering due to over-counting. PLoS ONE 7, e31457 (2012).
CAS PubMed PubMed Central ADS Google Scholar
Bolger, AM, Lohse, M. and Usadel, B. Trimmomatic: Flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114 (2014).
CAS PubMed PubMed Central Google Scholar
Gandhi, M. et al. The next generation characterization of the Encyclopedia of Cancer Cell Lines. Nature 569, 503–508 (2019).
CAS PubMed PubMed Central ADS Google Scholar
Normano, D. et al. Use TetR as a model searcher to explore the target search of DNA-binding proteins in mammalian cells. Nat. Community. 6, 7357 (2015).
Mirkin, EV, Chang, FS & Kleckner, N. Protein-mediated repetitive array chromosome pairing. J. Moore. biology. 426, 550–557 (2014).
Green, JB, etc. A general method for optimizing and functionalizing redshift rhodamine dyes. Nat. Method 17, 815–821 (2020).
CAS PubMed PubMed Central Google Scholar
Langmead, B. & Salzberg, SL and Bowtie 2 fast gap reading alignment. Nat. Method 9, 357–359 (2012).
CAS PubMed PubMed Central Google Scholar
Zhang, Y. et al. Model-based ChIP-seq (MACS) analysis. Genomic biology. 9. R137 (2008).
PubMed PubMed Central Google Scholar
Ramirez, F. et al. deepTools2: A next-generation web server for deep sequencing data analysis. Nucleic acid research. 44, W160–W165 (2016).
PubMed PubMed Central Google Scholar
Sedlazeck, FJ etc. Use single-molecule sequencing to accurately detect complex structural variations. Nat. Method 15, 461–468 (2018).
CAS PubMed PubMed Central Google Scholar
Overhauser, J. in Pulsed-Field Gel Electrophoresis, Methods in Molecular Biology Vol. 12 (eds. Burmeister, M. & Ulanovsky, L.) 129–134 (Humana Press, 1992).
Picelli, S. etc. Tn5 transposase and labeling program for large-scale sequencing projects. Genome research. 24, 2033-2040 (2014).
CAS PubMed PubMed Central Google Scholar
Corces, MR and others. The improved ATAC-seq protocol reduces background and enables interrogation of frozen tissues. Nat. Method 14, 959–962 (2017).
CAS PubMed PubMed Central Google Scholar
Talevich, E., Shain, AH, Botton, T. & Bastian, BC CNVkit: Whole genome copy number detection and visualization from targeted DNA sequencing. PLoS calculation. biology. 12, e1004873 (2016).
PubMed PubMed Central ADS Google Scholar
Raeisi Dehkordi, S., Luebeck, J. & Bafna, V. FaNDOM: Distance-based fast nested optical map seeding. Model 2, 100248 (2021).
PubMed PubMed Central Google Scholar
Haas, BJ etc. The accuracy of detection of fusion transcripts was evaluated by read mapping and de novo fusion transcript assembly methods. Genomic biology. 20, 213 (2019).
PubMed PubMed Central Google Scholar
Hahne, F. & Ivanek, R. in Statistical Genomics, Methods in Molecular Biology Vol. 1418 (eds. Mathé, E. & Davis, S.) 335–351 (Humana Press, 2016).
Butler, A., Hoffman, P., Smibert, P., Papalexi, E. & Satija, R. Integrate single-cell transcriptome data of different conditions, technologies and species. Nat. Biotechnology. 36, 411–420 (2018).
CAS PubMed PubMed Central Google Scholar
Granja, JM, etc. ArchR is an extensible software package for integrating single-cell chromatin accessibility analysis. Nat. Gene. 53, 403–411 (2021).
CAS PubMed PubMed Central Google Scholar
Satpathy, AT, etc. A massively parallel single-cell chromatin landscape of human immune cell development and T cell depletion within tumors. Nat. Biotechnology. 37, 925–936 (2019).
CAS PubMed PubMed Central Google Scholar
Mumbach, MR, etc. The enhancer junction group in primary human cells recognizes the target genes of disease-related DNA elements. Nat. Gene. 49, 1602–1612 (2017).
CAS PubMed PubMed Central Google Scholar
Mumbach, MR, etc. HiChIRP reveals the chromosomal conformation related to RNA. Nat. Method 16, 489–492 (2019).
CAS PubMed PubMed Central Google Scholar
Servant, N. etc. HiC-Pro: An optimized and flexible pipeline for Hi-C data processing. Genomic biology. 16, 259 (2015).
PubMed PubMed Central Google Scholar
Bhattacharyya, S., Chandra, V., Vijayanand, P. & Ay, F. FitHiChIP identifies important chromatin contacts from HiChIP data. Nat. Community. 10, 4221 (2019).
PubMed PubMed Central ADS Google Scholar
Rao, SSP, etc. The kilobase resolution 3D map of the human genome reveals the principle of chromatin cycling. Cell 159, 1665–1680 (2014).
CAS PubMed PubMed Central Google Scholar
Vidal, E. etc. OneD: Improve the repeatability of Hi-C samples with abnormal karyotypes. Nucleic acid research. 46, e49 (2018).
PubMed PubMed Central Google Scholar
Flynn, RA, etc. Discovery and functional inquiry of SARS-CoV-2 RNA-host protein interaction. Cell 184, 2394-2411 (2021).
CAS PubMed PubMed Central Google Scholar
Li, W. etc. MAGeCK can reliably identify essential genes from genome-scale CRISPR/Cas9 knockout screens. Genomic biology. 15, 554 (2014).
PubMed PubMed Central Google Scholar
Li, H. & Durbin, R. uses Burrows-Wheeler transformation for fast and accurate long-read comparison. Bioinformatics 26, 589–595 (2010).
PubMed PubMed Central Google Scholar
Scheinin, I. et al. Perform DNA copy number analysis on fresh and formalin-fixed samples by shallow whole-genome sequencing, and identify and eliminate problem areas in genome assembly. Genome research. 24, 2022-2032 (2014).
CAS PubMed PubMed Central Google Scholar
Hadi, K. et al. Different types of complex structural variations have been found in thousands of cancer genome maps. Cell 183, 197–210 (2020).
CAS PubMed PubMed Central Google Scholar
Blumrich, A. etc. The common fragile sites of FRA2C map to the boundaries of MYCN amplicons in neuroblastoma and are associated with overall chromosomal rearrangements in different cancers. Humph. Mole. Gene. 20, 1488–1501 (2011).
Gogolin, S. etc. In the context of doxorubicin-induced DNA damage, CDK4 inhibition can restore G1-S phase arrest in MYCN-expanded neuroblastoma cells. Cell Cycle 12, 1091–1104 (2013).
CAS PubMed PubMed Central Google Scholar
Durand, North Carolina, etc. Juicer provides a one-button system for analyzing loop resolution Hi-C experiments. Cell system. 3, 95–98 (2016).
CAS PubMed PubMed Central Google Scholar
Knight, PA & Ruiz, D. A fast algorithm for matrix balancing. IMA J. Numbers. anus. 33, 1029–1047 (2013).
Boeva, V. etc. Heterogeneity of neuroblastoma cell identities defined by transcription circuits. Nat. Gene. 49, 1408–1413 (2017).
Thanks to Chang, Liu, Mischel, and members of the Bafna laboratory for discussion; R. Zermeno, M. Weglarz and L. Nichols of the Stanford shared FACS facility for assisting in the cell sorting experiment; X. Ji, D. of the Stanford Functional Genomics Facility Wagh and J. Coller assisted in high-throughput sequencing; and A. Pang of Bionano Genomics assisted in optical mapping. HYC was supported by NIH R35-CA209919 and RM1-HG007735; KLH was supported by Stanford University Graduate Scholarship; and KEY was awarded the National Science Foundation Graduate Research Scholarship Program (NSF DGE-1656518), Stanford University Graduate Scholarship and NCI Postdoctoral Fellow Supported by the Fellow Transition Award (NIH F99CA253729). The cell sorting for this project was done on the instruments in the Stanford shared FACS facility. Sequencing was performed by the Stanford Functional Genomics Facility (supported by NIH grants S10OD018220 and 1S10OD021763). Perform microscopy on the instruments in the UCSD microscope core (supported by NINDS NS047101). AGH is supported by Deutsche Forschungsgemeinschaft (DFG; German Research Foundation) (398299703) and the European Research Council (ERC) under the European Union’s "Horizon 2020" research and innovation program (license agreement number 949172). ZL is the leader of Janelia's team, and HYC and RT are investigators from the Howard Hughes Medical Institute.
The contributions of these authors are the same: King L. Hung, Kathryn E. Yost, Zianqi Xie
Personal Dynamic Regulation Center, Stanford University School of Medicine, Stanford, California, USA
King L. Hung, Kathryn E. Yost, Shi Quanming, Natasha E. Weiser, Connor V. Duffy, Katerina Kraft, John C. Rose, M. Ryan Corces, Jeffrey M. Granja, Rui Li, and Howard Y. Chang
Howard Hughes Medical Institute Janelia Research Park, Ashburn, Virginia, USA
Xie Liangqi & Liu Zhe
Department of Molecular and Cell Biology, Li Ka-shing Center for Biomedicine and Health Sciences, CIRM Center of Excellence, University of California, Berkeley, California, USA
Xie Liangqi & Robert Tjian
Howard Hughes Medical Institute, Berkeley, California
Xie Liangqi & Robert Tjian
Department of Pediatric Oncology and Hematology, Charité-Universitätsmedizin Berlin, Berlin, Germany
Konstantin Helmsauer, Rocío Chamorro González, Celine Chen, and Anton G. Henssen
Bioinformatics and Systems Biology Graduate Program, University of California, San Diego, La Jolla, California, USA
Department of Computer Science and Engineering, University of California, San Diego, La Jolla, California, USA
Jens Luebeck, Siavash R. Dehkordi, Utkrisht Rajkumar and Vineet Bafna
Development and Disease Research Group, Max Planck Institute for Molecular Genetics, Berlin, Germany
Robert Schöpflin, Maria E. Valieva and Stefan Mundlos
Institute of Medicine and Human Genetics, Charité-Universitätsmedizin Berlin, Berlin, Germany
Robert Schöpflin, Maria E. Valieva and Stefan Mundlos
Department of Computational Molecular Biology, Max Planck Institute for Molecular Genetics, Berlin, Germany
Department of Cellular and Molecular Medicine, University of California, San Diego, La Jolla, California, USA
Chemistry-H, Stanford University, Stanford, California, USA
Joshua T. Lange, Ivy Tsz-Lo Wong, Jun Tang and Paul S. Mischel
Department of Pathology, Stanford University, Stanford, California, USA
Joshua T. Lange, Natasha E. Weiser, Ivy Tsz-Lo Wong, Jun Tang, Julia A. Belk, Ansuman T. Satpathy, and Paul S. Mischel
Institute of Children's Medical Center, University of Texas Southwestern Medical Center, Dallas, Texas, USA
Department of Computer Science, Stanford University, Stanford, California, USA
Tumor initiation and maintenance program, Sanford Burnham Prebys Medical Discovery Institute, La Jolla, CA, USA
Jordan Friedlein & Anindya Bagchi
Berlin-Brandenburg Regenerative Therapy Center (BCRT), Charité-Universitätsmedizin Berlin, Berlin, Germany
Experimental and Clinical Research Center (ECRC), Max Delbrück Center for Molecular Medicine and Charité-Universitätsmedizin Berlin, Berlin, Germany
German Cancer Federation (DKTK), partner site Berlin and German Cancer Research Center DKFZ, Heidelberg, Germany
Berlin Institute of Health, Berlin, Germany
Howard Hughes Medical Institute, Stanford University School of Medicine, California, USA
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
KLH, KEY and HYC conceived the project. KLH performed and analyzed CRISPRi and in vitro ecDNA digestion and PFGE experiments, as well as single-cell multi-omics, RNA-seq and ATAC-seq experiments. KEY performed and analyzed metaphase DNA FISH imaging, ChIP-seq, HiChIP, WGS, COLO320-DM nanopore sequencing and JQ1 perturbation experiments. LX performed and analyzed interphase DNA and RNA FISH imaging, TetO-eGFP cell line generation and live cell imaging, and PVT1p-nLuc imaging experiments. QS performed and analyzed all luciferase reporter gene experiments, except PVT1p-nLuc RNA and DNA FISH, and assisted with CRISPRi experiments. KH and RS analyzed TR14 Hi-C data and amplicon reconstruction. JL and SRD analyzed COLO320-DM WGS, nanopore sequencing, optical mapping data, and amplicon reconstruction. JTL, SW, CC, and JT performed and analyzed DNA FISH imaging. RCG generated TR14 Hi-C, DNA FISH, WGS and nanopore sequencing data. NEW After MS645 treatment, small molecule inhibitor experiments and DNA FISH imaging were analyzed and analyzed. MEV conducted Hi-C experiments and data analysis on TR14. IT-LW performed mid-term DNA FISH imaging. CVD performed and analyzed the ChIP-seq experiment. KK conducted the HiChIP experiment. JAB helped design and clone the CRISPRi experiment of the sgRNA pool. RL conducted RNA-seq experiments. UR analyzed interim DNA FISH data. JF generates COLO320-DM WGS data. MRC and JMG wrote the HiChIP data processing pipeline. MRC, JCR, AB, ATS, RT, SM, VB, AGH, PSM, ZL and HYC guide data analysis and provide feedback on experimental design. KLH, KEY and HYC wrote the manuscript with the participation of all authors.
Correspondence with Howard Y. Chang.
HYC is the co-founder of Accent Therapeutics, Boundless Bio and Cartography Biosciences, and an advisor to 10x Genomics, Arsenal Biosciences and Spring Discovery. PSM is the co-founder of Boundless Bio. He owns equity and serves as chairman of the scientific advisory board, and is paid for it. VB is the co-founder and consultant of Boundless Bio. ATS is the founder of Immunai and Cartography Biosciences. KEY is a consultant for Cartography Biosciences.
Peer review information Nature thanks Charles Lin and other anonymous reviewers for their contributions to the peer review of this work.
The publisher states that Springer Nature remains neutral on the jurisdiction claims of published maps and agency affiliates.
a, WGS tracks the position of the DNA FISH probe. For COLO320-DM and PC3, use 1.5 Mb MYC FISH probe (Figure 1a, b), 100 kb MYC FISH probe (Figure 1d-f), or 1.5 Mb chromosome 8 FISH probe. Commercial probes are used in SNU16 and HK359 cells. b. A representative DNA FISH image of a non-ecDNA amplified HCC1569 using a chromosome and a 1.5 Mb MYC probe, showing the expected pairing signal from the chromosome locus. c. Perform ecDNA clustering on a single COLO320-DM cell by autocorrelation g(r). d, Representative FISH image showing ecDNA clusters in primary neuroblastoma tumors (patients 11 and 17). e, ecDNA clustering of single primary tumor cells from all three patients, using autocorrelation g(r). f. Comparison of MYC copy number in COLO320-DM calculated based on WGS (n=7 genome box overlapping with DNA FISH probe), metaphase FISH (n=82 cells) and interphase FISH (n=47 cells). The P value is determined by Wilcoxon's test on both sides. g, a representative image of nascent MYC RNA FISH, showing the overlap of FISH probes for nascent RNA (intron) and total RNA (exon) in PC3 cells (repeated twice independently). h, Representative image of chromosomal DNA with nascent MYC RNA FISH in combined DNA FISH from MYC ecDNA (100 kb probe) and COLO320-DM cells (repeated independently four times). i, MYC transcription probability measured by freshman RNA FISH, compare single case ecDNA with the one found in the hub in COLO320-DM by FISH (box center line, median; box limit, upper and lower quartiles; box Must, 1.5 times the quartile range). In order to control the noise in the transcription probability of a small amount of ecDNA, we randomly resampled the RNA FISH data grouped by the center size and the calculated transcription probability. The violin graph represents the transcription probability of each ecDNA center sampled based on center size matching. The P value was determined by Wilcoxon's test on both sides.
a, based on TetO array knock-in and TetR-eGFP-labeled ecDNA imaging (left). Representative image of TetR-eGFP signal in TetO-eGFP COLO320-DM cells at specified time points in the time course (right; repeated twice independently). b, GFP signal in ecDNA-TetO COLO320-DM cells. The ecDNA centers labeled with TetR-eGFP and monomeric TetR-eGFP (A206K) seem to be smaller in living cells than in fixed-cell DNA FISH studies, possibly because the TetO array is not integrated into all ecDNA molecules, and there are The following factors can cause potential differences in DNA FISH and eGFP denaturation during dimerization. c, ecDNA center diameter in micrometers (box centerline, median; box boundaries, upper and lower quartiles; box whiskers, 1.5x interquartile range). The Tet-eGFP-labeled hub is slightly smaller than the TetR-eGFP(A206K)-labeled hub, which may be due to the eGFP dimerization effect (method). The P value was determined by Wilcoxon's test on both sides. d, the number of ecDNA centers per cell. The line represents the median. The P value was determined by Wilcoxon's test on both sides. e, chr8-chromosome-TetO (chr8: 116,860,000–118,680,000, left) and ecDNA-TetO (TetO-eGFP COLO320-DM, right) TetR-eGFP signal in COLO320-DM cells. f, the fluorescence intensity of chr8-chromosome-TetO and ecDNA-TetO lesions. g, h, based on the total fluorescence intensity relative to the chr8-chromosome, ecDNA-TetO labeled cells for each lesion (g; n = lesion/cell number) and each cell (h; n = cell number) inferred ecDNA copy number. TetO focus. The line represents the median. i, Representative image of the TetR-GFP signal in the parental COLO320-DM that does not integrate the TetO array, showing the smallest TetR-GFP foci. j, The average fluorescence intensity of a line drawn with the focus of ecDNA (TetO-eGFP) and BRD4 (HaloTag) across the center of the largest ecDNA (TetO-eGFP) signal. The data are the mean ± SEM of n=5 ecDNA lesions. k, Representative image of TetR-eGFP signal overlaps with BRD4-HaloTag signal in COLO320-DM cells without TetO array integration. The dashed line indicates the nuclear boundary. We noticed that there is no cytoplasmic TetR-eGFP signal in the subset of COLO320-DM cells integrated with the TetO array, but it is not co-localized with BRD4-HaloTag. 1. The MYC RNA measured by RT-qPCR on the parental COLO320-DM and BRD4-HaloTag COLO320-DM cells treated with DMSO or 500 nM JQ1 for 6 hours showed similar MYC transcription levels and BRD4 epitope labeling to inhibit JQ1 Sensitivity. The data are the mean ± SD between 3 biological replicates. The P value is determined by the student's t-test on both sides.
a. Representative metaphase FISH images and schematic diagrams show ecDNA in COLO320-DM and chromosomal HSR in COLO320-HSR (COLO320-DM is repeated twice independently, COLO320-HSR is not). b, Sequenced BRD4 ChIP-seq signal. Peaks in ecDNA or HSR amplification are highlighted and marked with the nearest gene. c, WGS at ATAC-seq, BRD4 ChIP-seq, H3K27ac ChIP-seq and MYC locus. d. After 6 hours of treatment with DMSO or 500 nM JQ1, the number of ecDNA positions (including ecDNA centers with >1 ecDNA and single cases of ecDNA) from individual COLO320-DM cells in interphase FISH imaging. N = number of cells quantified for each condition. The P value was determined by Wilcoxon's test on both sides. e. After 6 hours of treatment with DMSO or 500 nM JQ1, the ecDNA copy of each ecDNA position in the interphase FISH imaging in COLO320-DM (box center line, median; box limit, upper and lower quartiles; box to be , 1.5 times the interquartile range). N = the number of ecDNA positions quantified for each condition. The P value was determined by Wilcoxon's test on both sides. f. After treatment with DMSO or 500 nM JQ1 at the specified time point, a representative real-time image of TetR-eGFP-labeled ecDNA (top; repeated twice independently) and ecDNA center magnification (bottom). g, Representative images of combined DNA/RNA FISH in COLO320-DM cells treated with DMSO, 500 nM JQ1 or 1% 1,6-hexanediol for 6 hours. h, MYC transcription probability measured by double DNA/RNA FISH (box center line, median; Box boundaries, upper and lower quartiles; box whiskers, 1.5 times the interquartile range; n = number of cells). The P value is determined by Wilcoxon's test on both sides. i, Representative DNA FISH image of MYC ecDNA in the mesophase COLO320-DM, treated with 1% 1,6-hexanediol or 100 µg/mL α-ammantin for 6 hours. j, COLO320-DM treated with DMSO, 1% 1,6-hexanediol, or 100 µg/mL α-ammantin for 6 hours accumulates ecDNA in interphase cells through autocorrelation g(r). Data are mean ± SEM (n = 10 cells are quantified for each condition). k, the average BRD4 ChIP-seq signal and heat map of all BRD4 peaks of cells treated with DMSO or 500 nM JQ1 for 6 hours. l. After 48 hours of treatment with different JQ1 concentrations, the cell viability measured by ATP level (CellTiterGlo) is normalized to DMSO-treated cells. The data are the mean ± SD between 3 biological replicates. The P value is determined by the student's t-test on both sides. m, cell proliferation after 72 hours of treatment with different JQ1 concentrations. The data are the mean ± SD between 3 biological replicates. n, Cell doubling time after 72 hours of treatment with different JQ1 concentrations (top) or normalized to DMSO-treated cells (bottom). The data are the mean ± SD between 3 biological replicates. The P value is determined by the student's t-test on both sides. o, MYC RNA measured by RT-qPCR after 6 hours of treatment with the designated inhibitor (top; each dot represents a biological replicate, for DMSO and JQ1 treatment, n=6, for all other drug treatments, n= 3). The data are the mean ± SD. The P value is determined by the student's t-test on both sides. Details of the inhibitor panel, protein targets, significance of the effect on MYC transcription, and comparison of the effect on ecDNA and HSR transcription (bottom). p, q, representative DNA FISH image (p) and clustering of MYC ecDNA in COLO320-DM by autocorrelation g(r) (q), treated with DMSO or 500 nM MS645 for 6 hours. Data are mean ± SEM. The P value is determined by Wilcoxon's test on both sides with radius = 0.
a, AmpliconArchitect (AA) reconstructed structural variant (SV) view of MYC amplicon in COLO320-DM cells. b. Nanopore sequencing (left) and read length distribution of COLO320-DM cells. c, WGS for COLO320-DM, with connection points detected by WGS and nanopore sequencing. d, molecular length used for optical mapping and statistics. e, COLO320-DM ecDNA reconstructed after integrating WGS, optical mapping and in vitro ecDNA digestion. The original chromosome and the corresponding coordinates (hg19) are marked. Three internal circular tracks (light tan, slate, and brown; A, B, and C, respectively), representing the expected fragments cut using three different sgRNAs and Cas9 of expected sizes. The guide sequence is in Supplementary Table 2 (PFGE_guide_A-C). f, In vitro Cas9 digestion of COLO320-DM ecDNA, followed by PFGE (left). The fragment size is determined according to the H.wingei and S.cerevisiae steps. The uncropped gel image is in Supplementary Figure 1. The middle panel shows the short read sequencing of MYC ecDNA amplicons of all isolated fragments, sorted by fragment size. The image on the right shows the consistency of the expected fragment size obtained by optical mapping reconstruction, and the fragment size observed by in vitro Cas9 digestion (circled inconsistent fragments). Each sgRNA digestion is performed in a separate experiment. g, Mid-term FISH images show the co-localization of MYC, PCAT1, and PLUT, as predicted by optical mapping and in vitro digestion. N = 20 cells and 1,270 ecDNAs for MYC/PCAT1 DNA FISH, n = 15 cells and 678 ecDNAs for MYC/PLUT DNA FISH from an experiment. h. Measure the RNA expression of the designated transcript and designated sgRNA in COLO320-DM cells stably expressing dCas9-KRAB by RT-qPCR (n=2 biological replicates). Use primers MYC_exon1_fw and MYC_exon2_rv to amplify canonical MYC; fusion PVT1-MYC is amplified with PVT1_exon1_fw and MYC_exon2_rv; total MYC is amplified with total_MYC_exon2_fw and total_MYC_exon2_rv. All primer sequences are in Supplementary Table 1, and guide sequences are in Supplementary Table 2. i, Comparison of connection readings at PVT1-MYC breakpoints.
a. Combine single-cell RNA and ATAC-seq to simultaneously detect gene expression and chromatin accessibility, and to identify regulatory elements related to MYC expression. b. Unique ATAC-seq fragments and RNA characteristics of cells passing through the filter (both log2 conversion). c. Correlation between MYC accessibility score and standardized RNA expression. d, UMAP from RNA or ATAC-seq data (left). The log-normalized and scaled MYC RNA expression (top right) and MYC accessibility score (bottom right) are visualized on ATAC-seq UMAP, showing cells containing COLO320-ecDNA for MYC RNA-seq and ATAC-seq signals Horizontal heterogeneous DM. e, the gene expression score of MYC up-regulated genes (gene set M6506, molecular feature database; MSigDB) in all MYC RNA quantile boxes (calculated using Seurat in R). The horizontal line marks the median. The population differences of all individual cells are shown (top). The P value is determined by F-test on both sides. f, MYC expression levels of the top and bottom bins (left). Shows the normalized ATAC-seq coverage (right). g, the number of variable elements identified on COLO320-DM ecDNA compared to the chromosomal HSR in COLO320-HSR (left). 45 variable elements were uniquely observed on ecDNA. All variable elements on ecDNA are shown on the right (y-axis shows -log10(FDR), and the dot size represents log2 fold change. According to the relative position of kb with MYC TSS (negative, 5'; positive, 3').h , The correlation between the estimated MYC copy number of all individual cells and the normalized log2 converted MYC expression, showing a high level of copy number variability associated with increased expression, especially for COLO320-DM.i, separated by MYC RNA expression Estimated copy number of MYC amplicons for all cell bins. The amplification of ATAC-seq coverage of each of the five most significant variable elements identified in j, g (marked with a dashed box). k, similar distribution The enrichment of TSS in the high and low cell boxes indicates that the difference in the accessibility of variable elements is not an artifact of the difference in data quality. l, mean copy number regression, log normalization, scaled ATAC-seq coverage difference orange The rental peak of each cell box is relative to the average MYC RNA (lognormalized, mean centered, scaled). The same number of random non-difference peaks from the same amplicon interval are shown in gray. Error bands show the linear model 95% confidence interval. m, the cumulative probability of MYC amplicon copy number distribution (mean center, scaling) of single-cell ATAC-seq data and DNA FISH data, indicating that the copy number estimate of single-cell ATAC-seq data reflects the measured The heterogeneity of ecDNA copy number was determined by DNA FISH. The P value was determined by the Kolmogorov-Smirnov test (1,000 guided simulations).
a, from top to bottom: COLO320-DM H3K27ac HiChIP contact map (KR normalized read count, 10-kb resolution), reconstructed COLO320-DM amplicon, H3K27ac ChIP-seq signal, BRD4 ChIP-seq signal , WGS coverage, PVT1 (top, dark pink) and MYC (bottom, light pink) promoters, with a resolution of 10-kb, the FitHiChIP ring is shown below, and is colored according to the adjusted p value. The active elements recognized by scATAC and the overlapping H3K27ac HiChIP contact points are named after the genomic distance from the MYC start site: -1132E, -1087E, -679E, -655E, -401E, -328E, -85E. b, Comparison of the HiChIP matrix normalization method of COLO320-DM H3K27ac HiChIP at 10-kb resolution. The HiChIP signal is robust to different normalization methods. c, Quantification of NanoLuc luciferase signal from plasmids with PVT1p-, minp- or MYCp-driven NanoLuc reporter gene expression. The luciferase signal was calculated by normalizing NanoLuc readings to firefly readings. The bar graph shows the mean ± SEM. A two-sided Student's t-test (n=3 biological replicates) was used to calculate the P value. d. The violin chart shows the average fluorescence intensity and signal size of NanoLuc reporter RNA in cells transfected with PVT1p reporter gene and minp reporter gene. The P value is calculated using the two-sided Wilcoxon test. e, Schematic diagram of the luciferase reporter plasmid driven by the PVT1 promoter with cis-enhancer. The detailed information of the cis-enhancer is in the method. f. The bar graph shows the luciferase signal driven by PVT1p, MYCp, or constitutive TKp, with or without cis-enhancers (mean ± SEM). All values are normalized to the corresponding promoter-only construct, without the cis-enhancer. A two-sided Student's t-test (n=3 biological replicates) was used to calculate the P value. g, the dot plot shows the luciferase signal of JQ1 (Firefly standardized NanoLuc signal) in DMSO-treated COLO320-DM and COLO320-HSR cells after transfection with PVT1p or MYCp plasmids with or without cis-enhancers Multiple changes. A two-sided Student's t-test (n=3 biological replicates) was used to calculate the P value.
a. Representative DNA FISH image showing extrachromosomal single positive MYC and FGFR2 amplification (top left and top middle) and double positive MYC and FGFR2 amplification in the metaphase spread (top right) of parental SNU16 cells (top right). N = 42 cells and 8,222 ecDNA. Representative DNA FISH image showing different extrachromosomal MYC and FGFR2 amplification in metaphase spread in SNU16-dCas9-KRAB cells (bottom). N = 29 cells and 3,893 ecDNAs. b. A ranking chart showing the number of node reads that support each breakpoint in AmpliconArchitect. The color of the breakpoints depends on whether they span a region from the same amplicon (MYC/FGFR2) or from two different amplicons. c, HiChIP contact matrix with KR normalized 10-kb resolution of the parental SNU16 cell line (left) and SNU16-dCas9-KRAB cell line (right). Compared with SNU16-dCas9-KRAB cells, where the frequency of contact between chr8 and chr10 is greatly reduced, the contact matrix of the parental cells contains areas with increased cis-contact frequency between chr8 and chr10, as shown in the figure. The area where the focal interactions of the low-frequency structure rearrangement overlapped between chr8 and chr10 described in b increases is indicated by a box.
a, CRISPRi experiment disrupted candidate enhancers in SNU16-dCas9-KRAB cells. Single guide RNA (sgRNA) is designed to target candidate enhancers on FGFR2 and MYC ecDNA based on chromatin accessibility. b. Incorporating the experimental workflow for CRISPRI to inhibit putative enhancers. Generate stable SNU16-dCas9-KRAB cells from single cell clones. Cells were transduced with lentivirus library sgRNA, selected with antibiotics, and assessed oncogene RNA by flowFISH. Cells are sorted into six bins by fluorescence activated cell sorting (FACS) based on oncogene expression. sgRNAs are quantified as cells in each bin. c. FACS gate control strategy. d. Compared with the unclassified cells of the CRISPRI library targeting MYC or FGFR2 ecDNA, the log2 fold change of the sgRNA of each candidate enhancer element is changed, and then the cells are classified according to the expression level of MYC or FGFR2. Each point represents the average log2 fold change of 20 sgRNAs for candidate elements. Compared with the negative control sgRNA distribution in the same pool, elements negatively related to oncogene expression are marked in red. e, the histogram shows the importance of CRISPRI inhibition of the candidate enhancer element as shown in Figure 4e (top). As shown, the significant trans and cis enhancers are colored. SNU16-dCas9-KRAB H3K27ac HiChIP 1D signal tracking and FGFR2 and MYC promoter interaction profile, resolution is 10-kb, the cis-FitHiChIP loop is shown below. Violet indicates cis interaction, and orange indicates trans interaction. f, Spearman correlation of a single sgRNA targeting MYC TSS in a fluorescence box corresponding to MYC and FGFR2 expression. Shows the P value of comparing the target sgRNA with the negative control sgRNA (negcontrols) using the low-tail t-test. Each dot represents an independent sgRNA.
a, top: two-color DNA FISH on metaphase spreading to quantify the co-localization frequency of MYC gene and intermolecular enhancer shown in Figure 4e. Colocalization higher than random will indicate a fusion event. Bottom: Representative DNA FISH image. The DNA FISH probe targets the following hg19 genome coordinates: E1, chr10: 122,635,712–122,782,544 (RP11-95I16; n = 11 cells); E2, chr10: 122,973,293–123,129,601 (RP11-57H2; n = 12 cells); E3 /E4/E5, chr10: 123,300,005–123,474,433 (RP11-1024G22; n = 10 cells). b, top: the number of different and co-localized FISH signals. In order to estimate random colocalization, 100 simulated images were generated where the number of signals matched, and the average simulated frequency was compared with the observed colocalization. The P value is determined by a two-sided t test (Bonferroni adjustment). Bottom: The number of colocalization signals significantly higher than random chance. The colocalization higher than the simulated random distribution is the sum of the colocalized molecules that exceed the random mean in all FISH images, where the total localization is higher than the random mean plus the 95% confidence interval (100 simulated images per FISH image). c, In vitro Cas9 digestion of ecDNA containing MYC in SNU16-dCas9-KRAB, followed by PFGE (an independent experiment). The fragment size is determined according to the H.wingei and S.cerevisiae steps. The uncropped gel image is in Supplementary Figure 1. The MYC CDS guideline corresponds to guideline B in Supplementary Table 2. d. The enrichment of the enhancer DNA sequence in the isolated MYC ecDNAs band from background c (the range generated from undigested genomic DNA of the DNA isolated from a separate PFGE lane of the corresponding size) is based on normalized readings in the 5 kb window. Each dot represents DNA from a different gel band. Red indicates multiple changes higher than 4. e. The sequencing trace of gel-purified MYC ecDNA shows the enrichment of MYC amplicons and the depletion of FGFR2 amplicons containing enhancers E1-E5.
a, from top to bottom: long-read-based reconstruction of four different amplicons; genome map of long-read-based structural variants with a size greater than 10kb and greater than 20 supported reads, indicated by the red edge; The copy number variation and coverage of short-read whole-genome sequencing, and the location of the selected gene. b, Representative DNA FISH images of MYCN ecDNA in alternate TR14 cells (top) and ecDNA clusters compared with DAPI controls in the same cells assessed by autocorrelation g(r) (bottom). Data are mean ± SEM (n = 14 cells). c, Custom Hi-C image of reconstructed TR14 amplicon. MYCN/CDK4 amplicons and MYCN ecDNA share sequences, which prevents clear short-read mapping in these regions and appears as white areas. The trans-interaction between MYCN ecDNA and ODC1 amplicons is locally increased (indicated by arrows). The frequency of cis and trans contact is colored as indicated. d. Support for reads of structural variants identified by long-read sequencing of connected amplicons. Only one structural variation between the different amplicons (MYCN and MDM2 amplicons) was identified as having 3 supporting reads. e, Variant allele frequency of structural variants connected to amplicons. f, Trans-interaction mode between the enhancer on the MYCN amplicon fragment (vertical) and the ODC1 amplicon fragment (horizontal). Short read WGS coverage (grey), H3K27ac ChIP-seq trace shows the average multiple change entered in the 1kb box (yellow) and the Hi-C contact graph display (KR normalized count in the 5kb box). g, from top to bottom: three amplicons reconstruction, the enhancer-rich HPCAL1 locus on the ODC1 amplicon and the locus on other amplicons (red) and H3K27ac ChIP-seq (input multiple change; Yellow) virtual 4C interaction profile. h, the trans-interaction between different amplicons (KR normalized count in the 5kb bin) depends on the H3K27ac signal at the interaction site (left; box center line, median; box limit, upper and lower quartiles Number; frame must, 1.5x interquartile range). Trans-interactions (KR normalized counts in the 5kb bin) are separated by pairs of amplicons (right). H3K27ac high vs. low means that at least the average enrichment entered in the 5kb bin is less than 3 times. N = 114,636 H3K27ac low and low pair, n = 11,990 H3K27ac high and low pair, n = 296 H3K27ac high and high pair.
This file contains supplementary tables 1 and 2 and accompanying legends for supplementary tables 1-3.
Original image of agarose gel. Related to the extended data graph. 4f, 9c.
Please refer to Supplementary Table 3 for supplementary information in the legend.
Live cell imaging was performed using untreated TetO-eGFP COLO320-DM cells. A snapshot of the untreated cells is shown over the course of 30 minutes. GFP labeled TetO knocked into MYC ecDNA.
Live cell imaging was performed using TetO-eGFP COLO320-DM cells treated with DMSO. Control cells treated with DMSO were tracked within 1 hour. GFP labeled TetO knocked into MYC ecDNA.
Live cell imaging of TetO-GFP COLO320-DM cells after JQ1 treatment. Track cells treated with 500 nM JQ1 within 1 hour. GFP labeled TetO knocked into MYC ecDNA.
Hung, KL, Yost, KE, Xie, L. etc. The ecDNA center drives the coordinated expression of intermolecular oncogenes. Natural (2021). https://doi.org/10.1038/s41586-021-04116-8
DOI: https://doi.org/10.1038/s41586-021-04116-8
Anyone you share the following link with can read this content:
Sorry, there is currently no shareable link in this article.
Provided by Springer Nature SharedIt content sharing program
By submitting a comment, you agree to abide by our terms and community guidelines. If you find content that is abusive or does not comply with our terms or guidelines, please mark it as inappropriate.
Nature ISSN 1476-4687 (online) ISSN 0028-0836 (print)