Abstract
Mesenchymal plasticity has been extensively described in advanced epithelial cancers; however, its functional role in malignant progression is controversial1,2,3,4,5. The function of epithelial-to-mesenchymal transition (EMT) and cell plasticity in tumour heterogeneity and clonal evolution is poorly understood. Here we clarify the contribution of EMT to malignant progression in pancreatic cancer. We used somatic mosaic genome engineering technologies to trace and ablate malignant mesenchymal lineages along the EMT continuum. The experimental evidence clarifies the essential contribution of mesenchymal lineages to pancreatic cancer evolution. Spatial genomic analysis, single-cell transcriptomic and epigenomic profiling of EMT clarifies its contribution to the emergence of genomic instability, including events of chromothripsis. Genetic ablation of mesenchymal lineages robustly abolished these mutational processes and evolutionary patterns, as confirmed by cross-species analysis of pancreatic and other human solid tumours. Mechanistically, we identified that malignant cells with mesenchymal features display increased chromatin accessibility, particularly in the pericentromeric and centromeric regions, in turn resulting in delayed mitosis and catastrophic cell division. Thus, EMT favours the emergence of genomic-unstable, highly fit tumour cells, which strongly supports the concept of cell-state-restricted patterns of evolution, whereby cancer cell speciation is propagated to progeny within restricted functional compartments. Restraining the evolutionary routes through ablation of clones capable of mesenchymal plasticity, and extinction of the derived lineages, halts the malignant potential of one of the most aggressive forms of human cancer.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$32.99 / 30 days
cancel any time
Subscribe to this journal
Receive 51 print issues and online access
$199.00 per year
only $3.90 per issue
Buy this article
- Purchase on SpringerLink
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
All data supporting the findings of this study are available within the Article and its Supplementary Information. Mouse WGS, scRNA, scGET-seq and bulk ATAC–seq raw data have been deposited at the Sequence Read Archive (PRJNA1085291) and ArrayExpress (E-MTAB-13051 and E-MTAB-13052). Mouse and human genomic references (mm10, hg38) were downloaded from https://genome.ucsc.edu/. Human scRNA-seq data were downloaded from the Genome Sequence Archive project PRJNA1085291. Human clinicogenomic and transcriptomic information were downloaded from the TCGA (https://portal.gdc.cancer.gov/) and COMPASS (EGAD00001003584 and EGAD00001003585). Requests for resources and reagents can be directed to the lead contact G.G. Source data are provided with this paper.
Code availability
Codes used for this Article have been previously published and are referenced as appropriate. Methodological details on parameters used are provided in the Methods.
References
Zheng, X. et al. Epithelial-to-mesenchymal transition is dispensable for metastasis but induces chemoresistance in pancreatic cancer. Nature 527, 525–530 (2015).
Aiello, N. M. et al. Upholding a role for EMT in pancreatic cancer metastasis. Nature 547, E7–E8 (2017).
Ye, X. et al. Upholding a role for EMT in breast cancer metastasis. Nature 547, E1–E3 (2017).
Rhim, A. D. et al. EMT and dissemination precede pancreatic tumor formation. Cell 148, 349–361 (2012).
Kalluri, R. & Weinberg, R. A. The basics of epithelial-mesenchymal transition. J. Clin. Invest. 119, 1420–1428 (2009).
Turajlic, S., Sottoriva, A., Graham, T. & Swanton, C. Resolving genetic heterogeneity in cancer. Nat. Rev. Genet. 20, 404–416 (2019).
Greaves, M. & Maley, C. C. Clonal evolution in cancer. Nature 481, 306–313 (2012).
Aktipis, C. A., Boddy, A. M., Gatenby, R. A., Brown, J. S. & Maley, C. C. Life history trade-offs in cancer evolution. Nat. Rev. Cancer 13, 883–892 (2013).
Sottoriva, A. et al. A Big Bang model of human colorectal tumor growth. Nat. Genet. 47, 209–216 (2015).
Hanahan, D. Hallmarks of cancer: new dimensions. Cancer Discov. 12, 31–46 (2022).
Clarke, M. F. Clinical and therapeutic implications of cancer stem cells. N. Engl. J. Med. 380, 2237–2245 (2019).
Househam, J. et al. Phenotypic plasticity and genetic control in colorectal cancer evolution. Nature 611, 744–753 (2022).
Awad, M. M. et al. Acquired Resistance to KRASG12C inhibition in cancer. N. Engl. J. Med. 384, 2382–2393 (2021).
Dongre, A. & Weinberg, R. A. New insights into the mechanisms of epithelial–mesenchymal transition and implications for cancer. Nat. Rev. Mol. Cell Biol. 20, 69–84 (2019).
Thiery, J. P. Epithelial-mesenchymal transitions in tumour progression. Nat. Rev. Cancer 2, 442–454 (2002).
Fischer, K. R. et al. Epithelial-to-mesenchymal transition is not required for lung metastasis but contributes to chemoresistance. Nature 527, 472–476 (2015).
Padmanaban, V. et al. E-cadherin is required for metastasis in multiple models of breast cancer. Nature 573, 439–444 (2019).
Collisson, E. A. et al. Subtypes of pancreatic ductal adenocarcinoma and their differing responses to therapy. Nat. Med. 17, 500–503 (2011).
Moffitt, R. A. et al. Virtual microdissection identifies distinct tumor- and stroma-specific subtypes of pancreatic ductal adenocarcinoma. Nat. Genet. 47, 1168–1178 (2015).
Connor, A. A. & Gallinger, S. Pancreatic cancer evolution and heterogeneity: integrating omics and clinical data. Nat. Rev. Cancer 22, 131–142 (2022).
Genovese, G. et al. Synthetic vulnerabilities of mesenchymal subpopulations in pancreatic cancer. Nature 542, 362–366 (2017).
Mani, S. A. et al. The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell 133, 704–715 (2008).
Zhao, T. et al. Spatial genomics enables multi-modal study of clonal heterogeneity in tissues. Nature 601, 85–91 (2022).
Notta, F. et al. A renewed model of pancreatic cancer evolution based on genomic rearrangement patterns. Nature 538, 378–382 (2016).
Steele, C. D. et al. Signatures of copy number alterations in human cancer. Nature 606, 984–991 (2022).
Cortés-Ciriano, I. et al. Comprehensive analysis of chromothripsis in 2,658 human cancers using whole-genome sequencing. Nat. Genet. 52, 331–341 (2020).
Mueller, S. et al. Evolutionary routes and KRAS dosage define pancreatic cancer phenotypes. Nature 554, 62–68 (2018).
Barriga, F. M. et al. MACHETE identifies interferon-encompassing chromosome 9p21.3 deletions as mediators of immune evasion and metastasis. Nat. Cancer 3, 1367–1385 (2022).
Carstens, J. L. et al. Stabilized epithelial phenotype of cancer cells in primary tumors leads to increased colonization of liver metastasis in pancreatic cancer. Cell Rep. 35, 108990 (2021).
Krebs, A. M. et al. The EMT-activator Zeb1 is a key factor for cell plasticity and promotes metastasis in pancreatic cancer. Nat. Cell Biol. 19, 518–529 (2017).
Tedesco, M. et al. Chromatin velocity reveals epigenetic dynamics by single-cell profiling of heterochromatin and euchromatin. Nat. Biotechnol. 40, 235–244 (2022).
Klaasen, S. J. et al. Nuclear chromosome locations dictate segregation error frequencies. Nature 607, 604–609 (2022).
Umbreit, N. T. et al. Mechanisms generating cancer genome complexity from a single cell division error. Science 368, eaba0712 (2020).
McDonald, O. G. et al. Epigenomic reprogramming during pancreatic cancer progression links anabolic glucose metabolism to distant metastasis. Nat. Genet. 49, 367–376 (2017).
Crasta, K. et al. DNA breaks and chromosome pulverization from errors in mitosis. Nature 482, 53–58 (2012).
Cohen-Sharir, Y. et al. Aneuploidy renders cancer cells vulnerable to mitotic checkpoint inhibition. Nature 590, 486–491 (2021).
Karacosta, L. G. et al. Mapping lung cancer epithelial-mesenchymal transition states and trajectories with single-cell resolution. Nat. Commun. 10, 5587 (2019).
Comaills, V. et al. Genomic instability is induced by persistent proliferation of cells undergoing epithelial-to-mesenchymal transition. Cell Rep. 17, 2632–2647 (2016).
Peng, J. et al. Single-cell RNA-seq highlights intra-tumoral heterogeneity and malignant progression in pancreatic ductal adenocarcinoma. Cell Res. 29, 725–738 (2019).
Tan, T. Z. et al. Epithelial-mesenchymal transition spectrum quantification and its efficacy in deciphering survival and drug responses of cancer patients. EMBO Mol. Med. 6, 1279–1293 (2014).
Pastushenko, I. & Blanpain, C. EMT transition states during tumor progression and metastasis. Trends Cell Biol. 29, 212–226 (2019).
Bakouny, Z. et al. Integrative molecular characterization of sarcomatoid and rhabdoid renal cell carcinoma. Nat. Commun. 12, 808 (2021).
Gibbons, D. L. & Creighton, C. J. Pan-cancer survey of epithelial-mesenchymal transition markers across the Cancer Genome Atlas. Dev. Dyn. 247, 555–564 (2018).
O’Kane, G. M. et al. GATA6 expression distinguishes classical and basal-like subtypes in advanced pancreatic cancer. Clin. Cancer Res. 26, 4901–4910 (2020).
Serresi, M. et al. Polycomb repressive complex 2 is a barrier to KRAS-driven inflammation and epithelial-mesenchymal transition in non-small-cell lung cancer. Cancer Cell 29, 17–31 (2016).
Zhu, Z. et al. Mitotic bookmarking by SWI/SNF subunits. Nature 618, 180–187 (2023).
Schuhwerk, H. et al. The EMT transcription factor ZEB1 governs a fitness-promoting but vulnerable DNA replication stress response. Cell Rep. 41, 111819 (2022).
Alexandrov, L. B. et al. Signatures of mutational processes in human cancer. Nature 500, 415–421 (2013).
Li, Y. et al. Patterns of somatic structural variation in human cancer genomes. Nature 578, 112–121 (2020).
Chiou, S. H. et al. Pancreatic cancer modeling using retrograde viral vector delivery and in vivo CRISPR/Cas9-mediated somatic genome editing. Genes Dev. 29, 1576–1585 (2015).
Madisen, L. et al. A robust and high-throughput Cre reporting and characterization system for the whole mouse brain. Nat. Neurosci. 13, 133–140 (2010).
Madisen, L. et al. Transgenic mice for intersectional targeting of neural sensors and effectors with high specificity and performance. Neuron 85, 942–958 (2015).
Jackson, E. L. et al. Analysis of lung tumor initiation and progression using conditional expression of oncogenic K-ras. Genes Dev. 15, 3243–3248 (2001).
Rad, R. et al. A conditional piggyBac transposition system for genetic screening in mice identifies oncogenic networks in pancreatic cancer. Nat. Genet. 47, 47–56 (2015).
Hingorani, S. R. et al. Preinvasive and invasive ductal pancreatic cancer and its early detection in the mouse. Cancer Cell 4, 437–450 (2003).
Lakso, M. et al. Efficient in vivo manipulation of mouse genomic sequences at the zygote stage. Proc. Natl Acad. Sci. USA 93, 5860–5865 (1996).
Platt, R. J. et al. CRISPR-Cas9 knockin mice for genome editing and cancer modeling. Cell 159, 440–455 (2014).
Del Poggetto, E. et al. Epithelial memory of inflammation limits tissue damage while promoting pancreatic tumorigenesis. Science 373, eabj0486 (2021).
Sanjana, N. E., Shalem, O. & Zhang, F. Improved vectors and genome-wide libraries for CRISPR screening. Nat. Methods 11, 783–784 (2014).
Msaouel, P. et al. Comprehensive molecular characterization identifies distinct genomic and immune hallmarks of renal medullary carcinoma. Cancer Cell 37, 720–734 (2020).
Seth, S. et al. Pre-existing functional heterogeneity of tumorigenic compartment as the origin of chemoresistance in pancreatic tumors. Cell Rep. 26, 1518–1532 (2019).
Bolger, A. M., Lohse, M. & Usadel, B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120 (2014).
Clement, K. et al. CRISPResso2 provides accurate and rapid genome editing sequence analysis. Nat. Biotechnol. 37, 224–226 (2019).
Van der Auwera, G. A. et al. From FastQ data to high confidence variant calls: the Genome Analysis Toolkit best practices pipeline. Curr. Protoc. Bioinform. 43, 11.10.11–11.10.33 (2013).
Islam, S. M. A. et al. Uncovering novel mutational signatures by de novo extraction with SigProfilerExtractor. Cell Genom. 2, 100179 (2022).
Favero, F. et al. Sequenza: allele-specific copy number and mutation profiles from tumor sequencing data. Ann. Oncol. 26, 64–70 (2015).
Chen, X. et al. Manta: rapid detection of structural variants and indels for germline and cancer sequencing applications. Bioinformatics 32, 1220–1222 (2016).
Hu, Y. et al. OmicCircos: a simple-to-use R package for the circular visualization of multidimensional omics data. Cancer Inform. 13, 13–20 (2014).
Stickels, R. R. et al. Highly sensitive spatial transcriptomics at near-cellular resolution with Slide-seqV2. Nat. Biotechnol. 39, 313–319 (2021).
Draetta, E. L., Lazarević, D., Provero, P. & Cittaro, D. The frequency of somatic mutations in cancer predicts the phenotypic relevance of germline mutations. Front. Genet. 13, 1045301 (2022).
Smith, T., Heger, A. & Sudbery, I. UMI-tools: modeling sequencing errors in Unique Molecular Identifiers to improve quantification accuracy. Genome Res. 27, 491–499 (2017).
Benjamin, K., Dinar, Y. & Alexander, D. STARsolo: accurate, fast and versatile mapping/quantification of single-cell and single-nucleus RNA-seq data. Preprint at bioRxiv https://doi.org/10.1101/2021.05.05.442755 (2021).
Wolf, F. A., Angerer, P. & Theis, F. J. SCANPY: large-scale single-cell gene expression data analysis. Genome Biol 19, 15 (2018).
Wolock, S. L., Lopez, R. & Klein, A. M. Scrublet: computational identification of cell doublets in single-cell transcriptomic data. Cell Syst. 8, 281–291 (2019).
Polański, K. et al. BBKNN: fast batch alignment of single cell transcriptomes. Bioinformatics 36, 964–965 (2020).
Morelli, L., Giansanti, V. & Cittaro, D. Nested stochastic block models applied to the analysis of single cell data. BMC Bioinformatics 22, 576 (2021).
Gulati, G. S. et al. Single-cell transcriptional diversity is a hallmark of developmental potential. Science 367, 405–411 (2020).
Lange, J. T. et al. The evolutionary dynamics of extrachromosomal DNA in human cancers. Nat. Genet. 54, 1527–1533 (2022).
Thiery, J. P., Acloque, H., Huang, R. Y. & Nieto, M. A. Epithelial-mesenchymal transitions in development and disease. Cell 139, 871–890 (2009).
Badia-i-Mompel, P. et al. decoupleR: ensemble of computational methods to infer biological activities from omics data. Bioinform. Adv. 2, vbac016 (2022).
Molineris, I., Grassi, E., Ala, U., Di Cunto, F. & Provero, P. Evolution of promoter affinity for transcription factors in the human lineage. Mol. Biol. Evol. 28, 2173–2183 (2011).
Kulakovskiy, I. V. et al. HOCOMOCO: towards a complete collection of transcription factor binding models for human and mouse via large-scale ChIP-seq analysis. Nucleic Acids Res. 46, D252–D259 (2018).
Rausch, T. et al. DELLY: structural variant discovery by integrated paired-end and split-read analysis. Bioinformatics 28, i333–i339 (2012).
Giansanti, V. et al. Scalable integration of multiomic single-cell data using generative adversarial networks. Bioinformatics 40, btae300 (2024).
Korsunsky, I. et al. Fast, sensitive and accurate integration of single-cell data with Harmony. Nat. Methods 16, 1289–1296 (2019).
Zheng, G. X. et al. Massively parallel digital transcriptional profiling of single cells. Nat. Commun. 8, 14049 (2017).
Hao, Y. et al. Integrated analysis of multimodal single-cell data. Cell 184, 3573–3587 (2021).
Germain, P., Lun, A., Garcia Meixide, C., Macnair, W. & Robinson, M. Doublet identification in single-cell sequencing data using scDblFinder. F1000Res. https://doi.org/10.12688/f1000research.73600.2 (2022).
Chen, H. et al. Single-cell trajectories reconstruction, exploration and mapping of omics data with STREAM. Nat. Commun. 10, 1903 (2019).
Raghavan, S. et al. Microenvironment drives cell state, plasticity, and drug response in pancreatic cancer. Cell 184, 6119–6137 (2021).
Corces, M. R. et al. An improved ATAC-seq protocol reduces background and enables interrogation of frozen tissues. Nat. Methods 14, 959–962 (2017).
Marin, D. et al. Safety, efficacy and determinants of response of allogeneic CD19-specific CAR-NK cells in CD19+ B cell tumors: a phase 1/2 trial. Nat. Med. 30, 772–784 (2024).
Zhang, Y. et al. Model-Based Analysis of ChIP-Seq (MACS). Genome Biol. 9, R137 (2008).
Heinz, S. et al. Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol. Cell 38, 576–589 (2010).
Acknowledgements
We thank all of the members of the Draetta and Viale laboratories for discussions and reagents; S. Jiang and the staff at the MDACC Department of Veterinary Medicine for support in animal handling; C. Kingsley, J. Delacerda and the staff at the MDACC Small Animals Imaging Facility for their constant willingness; the staff at Genscript and VectorBiolab for support and service; the members of the Advanced Technology Genomics Core (E. Thompson, H. Tang and D. Pollock) at MDACC (CA016672 (ATGC) Core Grant); C. R. Jeter for confocal studies; A. K. Jain and the staff at the Epigenomics Profiling Core (EpiCore) for help with bulk ATAC–seq assays; the staff at the High-Performance Computing for Research facility at the University of Texas MD Anderson Cancer Center for providing computational resources that have contributed to the research results reported in this paper. L.P. was supported by the Ermenegildo Zegna Founder’s Scholarhip 2019/2020; F. Citron by AIRC and the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement no. 800924; G.F.D. by the Sheikh Ahmed Bin Zayed Al Nahyan Center for Pancreatic Cancer Grant and the Pancreatic Cancer Action Network Translational Research Grant; L.W. by the NIH/NCI grants R01CA266280, the Cancer Prevention and Research Institute of Texas (CPRIT) awards RP200385, The Break Through Cancer (BTC) Foundation, the start-up research fund and the University Cancer Foundation via the Institutional Research Grant Program at the University of Texas MD Anderson Cancer Center. L.W. is also an Andrew Sabin Family Foundation Fellows at MD Anderson Cancer Center. G. Tortora was supported by AIRC IG 26330, Ministry of Health (CO 2019-12369662), FIMP (J38D19000690001), Ministry for Universities and Research (MUR): PRIN 2022 Prot. 2022P79F9N, PRIN 2022 PNRR Prot. P2022LN3KS; A.V. was supported by Cancer Research and Prevention Institute of Texas (CPRIT) grant RP230373, the V Foundation (V2020-018), NIH/NCI grant R01CA258917; F. Chen acknowledges support from the NHGRI (R01, R01HG010647), the NIH SMAHT Consortium (1UG3NS132135), the Burroughs Wellcome Fund CASI award, the Searle Scholars Foundation, the Harvard Stem Cell Institute and the Merkin Institute; this research was supported by the NYSCF; F. Chen is a New York Stem Cell Foundation - Robertson Investigator; and G.G. was supported by the Barbara Massie Memorial Fund, the MDACC Moonshot FIT Program, the Bruce Krier Endowment Fund and the Lyda Hill Foundation. The Sid W. Richardson Foundation-supported University Distinguished Chair supported whole-genome sequencing conducted in this study. The UT MD Anderson Cancer Center provided permissions to use illustrations generated via BioRender and included in this work.
Author information
Authors and Affiliations
Contributions
Conceptualization: G.G., L.P., F. Chen, L.W., T.P.H., A.V., A.C., V.G., A.F., G. Tonon, M.P.K., G.F.D., G. Tortora, A.S. and R.K. Methodology for mouse models experiments: L.P., K.K.M., R.K., A.V., T.P.H. and G.G. Methodology for spatial DNA-seq: S. Mangiameli, A.J.C.R. and F. Chen. Methodology for single-cell analysis: L.Z., S. Mangiameli, F.P., L.W. and D. Cittaro. Bioinformatic formal analysis: L.Z., S. Mangiameli, F.P. and C.L. Analysis of mouse model experiments: L.P., L.Z., S. Marisetty, F.G., K.K.M., F.P., F. Carbone, C.L., H.K., F. Citron, M.S., T.N.A.L., S.L., C.Z., D. Catania, E.G., N.F., J.J.A. and A.M.T.M. Visualization: L.P., L.Z. and S. Marisetty. Writing—original draft: L.P. and G.G. Writing—review and editing: L.P., G.G., F. Chen, L.W., A.V., A.C., G.F.D. and A.S.
Corresponding authors
Ethics declarations
Competing interests
D. Cittaro, F.G. and G. Tonon are recipients of a patent application covering TnH (EP4288534A1) used for single-cell GET-seq experiments. F. Chen is an academic founder of Curio Biosciences and Doppler Biosciences, and scientific advisor for Amber Bio. F. Chen’s interests were reviewed and managed by the Broad Institute in accordance with their conflict of interest policies. The other authors declare no competing interests.
Peer review
Peer review information
Nature thanks Direna Alonso-Curbelo, Moritz Gerstung and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data figures and tables
Extended Data Fig. 1 Clinico-pathological characterization of the PCΦ SM-GEMM.
a) Representative macroscopic pictures of primary tumours (PT) and liver metastasis (Liver Met) of PCΦ SM-GEMM engineered with different combinations of driver events. b) Survival curves of PCΦ SM-GEMM engineered with different combinations of driver events (KPCC: N = 76; KCC: N = 36; KP: N = 24). P = 1.29752e-07, 6e-15, 3.70032e-08. c) Bar plot showing percentage of mice engineered with different combinations of driver events with either normal pancreas (Normal) or pancreatic tumour (PDAC) at endpoint (KPCC: N = 76; KCC: N = 36; KP: N = 24). P = 2.19655E-06, 5.161e-12. d) Scatter plot showing number of metastasis of PCΦ SM-GEMM engineered with different combinations of driver events (KPCC: N = 76; KCC: N = 36; KP: N = 24). e) Scatter plot showing number of TdT+ and TdT- metastasis of PCΦ SM-GEMM engineered with different combinations of driver events (KPCC: N = 76; KCC: N = 36; KP: N = 24). P = 0.0011, 1e-15, 1e-15. f) Scatter plot showing number of TdT + /GFP- and TdT + /GFP+ metastasis of PCΦ SM-GEMM engineered with different combinations of driver events (KPCC: N = 76; KCC: N = 36; KP: N = 24). P = 0.0383. g) Representative primary tumours histopathology pictures of PCΦ SM-GEMM engineered with different combinations of driver events (H&E: hematoxilyn and eosin; MTS: Masson thrichromic staining). h) Bar plot showing quantification of tumour grade of primary tumours of PCΦ SM-GEMM engineered with different combinations of driver events (KPCC and KCC, N = 10 different primary lesions; KP, N = 9 different primary lesions). i) Bar plot showing quantification of aniline blue positive and negative areas of primary tumours of PCΦ SM-GEMM engineered with different combinations of driver events (KPCC and KCC, N = 10 different primary lesions; KP, N = 9 different primary lesions). j) Bar plot showing quantification of TdT + /GFP- and TdT + /GFP+ tumours cells of primary tumours of PCΦ SM-GEMM engineered with different combinations of driver events (KPCC and KCC, N = 10 different primary lesions; KP, N = 9 different primary lesions). Data are shown as mean with standard deviation. Scale bars = 100 μm if not otherwise specified. P values are calculated as follows: b) Logrank Mentel-Cox test; c) Fisher’s exact chi square test; d-f) two-sided paired student t-test; h-j) two-way ANOVA.
Extended Data Fig. 2 Pancreatic tumours with mesenchymal competency display aggressive clinical behaviour.
a) Histopathological characterization of primary tumour site and metastatic sites of PCΦ SM-GEMM showing distribution of TdT + /GFPHIGH and TdT + /GFPLOW cells in tumoral areas. Images are representative of N = 10 independent lesions. b) Histopathological characterization of PCΦ SM-GEMM at different clinical stages showing concomitant expression of TdT, GFP and KRT19 tumour cells markers (N = 5 different lesions/group). P = 0.001, 1e-15. c-e) Schematic and representative pictures of the isolation of TdT+ and TdT- cells from PCΦ SM-GEMM-derived organoids (c-d) and representative scatter plot of gating strategies (e). Scatter plot is representative of N = 2 sorting experiments. f-g) Survival curves and metastasis (f) quantification of metastasis (g) of PCΦ SM-GEMM transplanted with either TdT+ or TdT- organoid-derived tumour cells showing reduced survival and metastatic burden in mice bearing TdT+ derived lesions (N = 10 mice/group). P = 4.495e-6, 0.019. h) Representative histopathological pictures of tumours derived by TdT+ or TdT- cells: a tendency to sarcomatoid transformation can be appreciated in lesions derived from TdT+ cells; noticeably, TdT- cells derived tumours evolve to TdT+ lesions. Images are representative of N = 2 experiments. i-j) Representative microscopic pictures of organoids derived from PCΨ Vim+/+ (right) and VimFlpo/+ (left) SM-GEMM (i) and survival curves of these organoids transplanted into the pancreata of recipient mice (j). N = 20 mice per group; data are representative of N = 2 experiments. P = 1.2124E-11. k) Representative immunofluorescence (left) and grading (right) of tumours arose from transplanted organoids derived from PCΨ Vim+/+ (top) and VimFlpo/+ (bottom) SM-GEMM. N = 25 tumour areas. P = 5.32527e-08. l) Histopathological quantification of cells co-expressing VIM and TdT in tumours arose from transplanted organoids derived from PCΨ Vim+/+ and VimFlpo/+SM-GEMM. N = 25 tumour areas. P = 1.74523e-10. Scale bars = 100 μm if not otherwise specified. P values are calculated as follows: b) one-way ANOVA with multiple comparisons (Tukey’s multiple comparisons test); f,j) Logrank Mentel-Cox test; g) paired student t-test k) Fisher’s exact chi square test; l) two-way ANOVA.
Extended Data Fig. 3 Clinico-pathological characterization of the PCW SM-GEMM.
a) Macroscopic pictures of primary tumours and metastatic lesions of a PCΩ SM-GEMM. Pictures representative of N = 20 mice. b) Box plot showing primary tumour dimensions for Vehicle, GCV and GCV-rec cohorts (top, KCC; bottom KP. N = 10 mice/cohort for KCC; N = 5 mice/cohort for KP). KCC, P = 1e-06; KP, P = 0.004, 0.0199, 1e-06. c) Box plot showing number of metastasis for Vehicle, GCV and GCV-rec cohorts (top, KCC; bottom KP. N = 10 mice/cohort for KCC; N = 5 mice/cohort for KP). KCC, P = 0.003, 0.001; KP, P = 0.0089. d) Survival curves of the Vehicle and GCV-rec cohorts, start and end of treatments are highlighted by green and red arrows respectively. (top, KCC; bottom KP. N = 10 mice/cohort for KCC; N = 5 mice/cohort for KP) KCC, P = 0.00096; KP, P = 0.0018. e) Representative histopathological analysis of the different cohorts of the KPCC PCΩ SM-GEMM; pictures are representative of N = 5 mice. f) Bar plot showing quantification of EPI-L and MS-L (TdT + /GFP- and TdT + /GFP+ respectively) of primary tumours for the different cohorts of KPCC PCΩ SM-GEMM (N = 25 lesions/cohort). P = 1e-15, 2e-15, 1e-15, 1e-15. g) Bar plot showing quantification of EPI-L and MS-L (TdT + /GFP- and TdT + /GFP+ respectively) of primary tumours for the different cohorts of KCC PCΩ SM-GEMM (left) and KP PCΩ SM-GEMM (right) (N = 25 lesions cohort). KCC, P = 0.0012, 0.026, 3.10031e-06; KP, P = 1.98129e-06, 0.0406, 0.00045. Scale bars = 100 μm if not otherwise specified. Error bars represent SD. Box plots are centred at the 50th percentile and bounds of box are fixed on the 25th and 75th percentile, min and max include all data points. P values are calculated as follows: b,c) one-way ANOVA with multiple comparisons (Tukey’s multiple comparison test); d) Logrank Mentel-Cox test; f-g) two-way ANOVA.
Extended Data Fig. 4 Cells with mesenchymal features are required for the expansion of tumour engraftments.
a) Representative pictures of pancreatic tumour-derived 3D cultures generated from PCΩ SM-GEMM. Images representative of N = 3 independent primary cell lines. b-c) Scatter plot with individual points (left) and box and whiskers plot showing area and number of pancreatic spheres upon GCV treatment. N = 32 fields per group (b) and N = 10 fields per group. P = 1.4094e-06. d) Schematic showing transplantation experiments and treatment schedule of 3D cultures generated from PCΩ SM-GEMM. e) Representative macroscopic pictures of primary tumours for the three groups. f) Scatter plots with individual values calculating tumour volume and number of metastasis for the same groups. Images are representative of 30 mice; N = 10 per group. P = 8.40064e-07, 1.19482E-08, 0.0071, 0.0002. g) Survival curves of mice bearing transplanted pancreatic tumour-derived 3D cultures upon orthotopic transplantation and GCV treatment. N = 20 independent mice over N = 2 independent experiments. P = 3.51436e-06. h-i) Representative macroscopic and microscopic pictures (left) and quantification of number of nodules (right) of pancreatic tumour-derived 3D cultures transplated via spleen injection and treated or not with GCV. Images are representative of N = 15 mice; N = 5 per group. P = 0.002 j) Survival curves of mice bearing pancreatic tumour-derived 3D cultures upon splenic transplantation and GCV treatment. N = 5 per group. P = 0.002 k-l) Representative histopathological pictures (left) and number of microscopic nodules (right) from mice transplated in the lateral vein with pancreatic tumour-derived 3D cultures. Images are representative of N = 15 mice; N = 5 mice per group. Scale bars = 100 μm if not otherwise specified. Error bars represent SD. Box plots are centred at the 50th percentile and bounds of box are fixed on the 25th and 75th percentile, min and max include all data points. P values are calculated as follows: b,c,f,i,l) two-sided student t-test; g, j) Logrank Mentel-Cox test.
Extended Data Fig. 5 Tumours with sarcomatoid features display increased aberrant mitotic events.
a) Schematic showing experimental workflow. b) Survival curves showing improved survival for mice in the DTR-tr rec and GCV-tr rec group. N = 10 mice/group over N = 2 independent experiments. P = 0.0004, 4.83711e-06, 0.0128. c-d) Box and wiskers with data point showing tumour dimension (c) and number of metastasis (d) in the different groups. N = 10 mice/group over N = 2 independent experiments. c, P = 7.17017e-06, 2.65958e-05, 7.38067E-05; d, P = 4.5863E-05, 0.00011, 8.12171e-05. e) Bar plot showing distribution of EPI-L and MS-L population among the different groups. N = 5 lesions/group. P = 2.72642e-05, 1.92443e-05, 0.0027. f) Representative histopathological pictures showing patient-derived pancreatic tumours pathologically annotated for sarcomatoid features and high magnification of normal and aberrant mitosis. g) Box and wiskers with data points showing number of mitosis in patient derived pancreatic tumours annotated with sarcomatoid features. h) Bar plot showing distribution of normal and aberrant mitosis in patient derived pancreatic tumours annotated with sarcomatoid features. Error bars represent SD. Box plots are centred at the 50th percentile and bounds of box are fixed on the 25th and 75th percentile, min and max include all data points. P values are calculated as follows: c,d, g) two sided paired student t-test; h) Fisher’s exact chi square test.
Extended Data Fig. 6 Spatial genomics of tumour cells with mesenchymal proficiency.
a-b) Slide-DNA-seq of two PCΦ SM-GEMM 3 weeks after AAV transduction showing TdT+ and TdT- regions of pancreatic tissue, differentially clustering according to genomic analysis. c-d) Normalized copy number scatter plots showing differential copy number aberrations in the three clusters as previously calculated with principal component analysis; particularly evident are gains of 6 and 15 and minor loss of chromosome 4. e) Heatmap showing genomic characteristics of tumours belonging to the EMT-on and EMT-off groups together with analysis of putative driver events as expected by genomic engineering. f) Bar plot showing percentage of distribution of single nucleotide variants in EMT-on and EMT-off groups. It can be readily appreciated that the two groups are characterized by similar mutational signatures belonging to clock-like types of mutagenesis (SBS5 and SBS1).
Extended Data Fig. 7 Structural variants and chromotripsis in KPCC EMT-on and EMT-off groups.
a) Box and whiskers with individual points showing number and type of SVs in the PCΨ Vim+/+ and VimFlpo/+ SM-GEMMs. N = 7 mice per group. b) Two representative scatter plots showing absolute copy number values of chromosome 4 for two different mice belonging to the EMT-on (top) and EMT-off (bottom groups). Poliploidy and focal copy number gains can be readily appreciated. c) ShatterSeek output plots for 4 different mice belonging to the EMT-on (left) and EMT-off (right) group. Error bars represent SD. Box plots are centred at the 50th percentile and bounds of box are fixed on the 25th and 75th percentile, min and max include all data points. Visual inspection, SVs count and number of oscillating copy number (Oscillating CN) were used to call high-confidence chromotripsis: the two cases on the left met the criteria (Methods).
Extended Data Fig. 8 Genomic features of KP and KCC EMT-on and EMT-off groups.
a) Representative circos plots of genomic features of mice belonging to the KP and KCC genotype in the EMT-on and EMT-off experimental groups. b) Bar plot with individual point showing fCNA values for tumours belonging to the KP (left) and the KCC (right) genomic backgrounds. N = 27 independent mice (KP: N = 10; KCC: N = 17) over N = 2 independent experiments. P = 0.015, 0.027. c) Bar plot with individual points showing SV load values for tumours belonging to the KP and the KCC genomic backgrounds. N = 27 independent mice (KP: N = 10; KCC: N = 17) over N = 2 independent experiments. P = 0.0063, 1e-05. d) Bar plot with individual points showing number of chromotripsis for tumours belonging to the KP and the KCC genomic backgrounds. P = 0.0354, 0.002. e) Representative ShatterSeek plots for tumours belonging to the KP and the KCC genomic backgrounds. N = 27 independent mice (KP: N = 10; KCC: N = 17) over N = 2 independent experiments. Data are presented as mean values +/− SD. P values are calculated as follows: b) two-sided Mann-Whitney test; c) two-sided Tukey’s multiple comparison test, d)two-sided paired student t-test.
Extended Data Fig. 9 Transcriptomic signatures of EMT correlate with genomic instability in human cancers.
a) Scatter plot showing values of EMT score per each sample among WGD status (0: no WGD; 1: one WGD event; 2: two WDG events). P = 0.0004. b) Three-dimensional scatter plot showing correlations between EMT score, estimated ploidy and tumour subclonal fraction (Sub. fr.). c) Scatter plot showing correlation between EMT score and chromotripsis is high purity samples from COMPASS cohort. N = 23. P = 0.0160. d) Box and whiskers with all datapoints showing no statistical difference in terms of EMT score in TP53 wild type (WT) and mutant (Mutant) samples. N = 23. e) Schematic showing the analytical worflow used for the TCGA Solid Tumours cohort (Methods). f-g) Scatter plot showing correlation between Chromotripsis score and a transcriptomic signatures of EMT (Methods) of TCGA Solid Tumours filtered by purity 0.75. P = 0.0056, 4.90308E-08. h) Bar plot showing distribution of samples with at least one event of chromotripsis based on TP53 genomic status of TCGA Solid Tumours filtered by purity 0.75. P = 3.61974e-05. i) Box plot with individual values showing Chromotripsis score based on TP53 genomic status of TCGA Solid Tumours filtered by purity 0.75. N = 217 independent tumours. j-k) Violin plots showing EMT scores based on TP53 genomic status of TCGA Solid Tumours filtered by purity 0.75. l) Scatter plot showing correlations between EMT scores and Subclonal fraction of TCGA Solid Tumours filtered by purity 0.75. P = 5.50454e-08. m) Box and whiskers with individual data points displaying Chromotripsis score values between groups divided by median value of EMT scores from the COMPASS dataset filtered by purity 0.75. N = 23 independent tumours. P = 0.0018. n) Scatter plot showing correlation between EMT score and ploidy from the COMPASS dataset filtered by purity 0.75. P = 0.014. o) Scatter plot showing correlation between fCNA and EMT score from the PDAC-PDX MDACC dataset. P = 0.025. Error bars represent SD. Box plots are centred at the 50th percentile and bounds of box are fixed on the 25th and 75th percentile, min and max include all data points. P values are calculated as follows: b-c,f-g,l-o) two-tailed Spearman correlation; h) Fisher’s exact chi square test; a,d,m) two sided Mann Whitney test; i-k) two sided paired student t test.
Extended Data Fig. 10 Histopathological and genomic characterization of the KPCY Zeb1WT/WT and Zeb1fl/fl GEMM.
a) Representative H&Es of primary tumours derived from KPCY Zeb1fl/fl and KPCY Zeb1WT/WT GEMMs. b) Bar plot showing quantification of grade in the two GEMMs (N = 10 areas per group). P = 0.0023. c) Circos plots showing genomic events in one KPCY Zeb1fl/fl mouse and KPCY Zeb1WT/WT mouse. d) Representative cases of chromotripsis as calculated via ShatterSeek. Scale bars = 100 μm if not otherwise specified. P values are a result of b) Fisher’s exact chi square test.
Extended Data Fig. 11 scGET and RNA seq of EPI-L and MS-L subpopulations.
a-b) UMAPs showing cluster distribution of MS-L and EPI-L tumour cells and top 10 significant pathways enriched in cluster n.1 according to scGET and RNA data integration. c) UMAPs showing the distribution of EPI-L and MS-L subpopulation in scRNA and GET seq data. d) UMAP showing clonotype distribution as calculated by copy number analysis of scGET seq data. e) Scatter plot displaying correlations between Pseudotime α-Diversity and β-diversity as calculated from scGET seq data. f) UMAP showing distribution of structural variants as calculated from scGET seq data after normalization of sequencing coverage. g) Violin plot showing average nuclear area per cell as calculated by histopathological analysis of PCΩ SM-GEMMs. P = 0.0099. h) Histopathological representative picture (left) and quantification (right) of PCΩ SM-GEMMs with the DNA damage marker γH2AX. N = 88 independent areas over N = 3 independent experiments. P = 6.26675e-10. Scale bars = 100 μm if not otherwise specified. Data are presented as mean values +/− SD P values are calculated as follows: g) two-sided student t-test; h) two-way ANOVA.
Extended Data Fig. 12 TGF-beta drives acquisition of nuclear abnormalities and chromatin remodelling.
a) Bar plot showing Mean Optical Density of MS-L and EPI-L cultures treated with either Reversine, MPS1i (MPI-0479605) or vehicle 14 days after plating. N = 3 biological replicates/group over N = 3 independent experiments. P = 7.36e-12, 1.13e-12, 6.63e-11, 4.803E-12. b) Representative pictures showing colonies forming under the effect of Mad2l1 knockdown in the EPI-L and MS-L subpopulations. c) Schemating showing experimental strategy to induce a TGF-beta driven EMT in the EPI-L subpopulation with analysis of nuclear and chromatin changes. d) Representative high magnification pictures showing normal and aberrant cell-cycle phases (anaphase, metaphase and interphase); red arrow indicates micronucleus in interphase. e) Bar plot with individual data points showing number of micronuclei per HPF in EPI-L subpopulations 48 h after incubation with either TGF-beta or vehicle (N = 20 HPF images per group). P = 0.015 f) Bar plot showing incidence of aberrant and normal mitosis per HPF in EPI-L subpopulations 48 h after incubation with either TGF-beta or vehicle (N = 50 mitosis per group). P = 3.32908E-05. g) Bar plot with individual data points showing number of peaks as extrapolated from ATAC-seq in EPI-L subpopulations 48 h after pulses of either TGF-beta or vehicle (N = 5 independent EPI-L models per group). P = 0.0085. h) Bar plot showing number and type of peaks in Centromeric genes as extrapolated from ATAC-seq in EPI-L subpopulations 48 h after incubation with either TGF-beta or vehicle (N = 5 models per group). P = 0.021. Scale bars = 100 μm if not otherwise specified. Data are presented as mean values +/− SD P values are a result of: e) two-sided student t test; f) Fisher’s exact chi-square test; g) two-sided Mann-Whitney test; a,h) two-sided two-way ANOVA.
Extended Data Fig. 13 Characterization of MS-L ‘diploid’ and ‘aneuploid’ clones.
a) Schematic showing experimental design of the isolation and transplantation of MS-L diploid and aneuploid clones from a stable MS-L ex vivo model. b) Representative pictures of metaphasic spreads of diploid and aneuploid MS-L (top) and quantification of chromosomes (bottom). N = 10 metaphases/group over N = 3 independent experiments. P = 1.82851E-08. c) Representative macroscopic pictures of liver metastasis derived from orthotopically transplanted MS-L diploid and aneuploid cells. d) Survival curves of tumour bearing mice transplanted with either MS-L diploid or aneuploid cells (N = 10 mice/group). P = 3.06476e-06. e) Scatter plot showing primary tumour size and number of metastasis of tumour bearing mice transplanted with either MS-L diploid or aneuploid cells (N = 10 mice/group). P = 0.047. f) Box and whiskers with individual data points showing number of mets in mice transplanted with either MS-L diploid or aneuploid cells (N = 10 mice/group). P = 0.033. g) Circle plots showing incidence of ascites in mice transplanted with either MS-L diploid or aneuploid cells (N = 10 mice/group). P = 0.019. h) Representative H&E of transplanted diploid and aneuploid tumours (top) and quantification of grade (bottom). i) DNA content histogram showing ploidy of MS-L bulk and clone#1 MS-L diploid and aneuploid derived from the bulk population. DNA profiles are similar in the isolated clones 10 passages after single cell seeding. j) Representative pictures of primary tumours derived from MS-L diploid and aneuploid clones transplanted orthotopically in the pancreas of recipient mice. k) Survival curves of mice transplanted with MS-L diploid and aneuploid clone #1 and #2. l-m) Number of metastasis and primary tumour volumes of mice transplanted with MS-L diploid and aneuploid clone #1 and #2. N = 5 mice/group over N = 2 independent experiments. Error bars represent SD. Box plots are centred at the 50th percentile and bounds of box are fixed on the 25th and 75th percentile, min and max include all data points. P values are a result of: d,k) Logrank Mentel-Cox; f,l-m) two-sided Tukey’s multiple comparison test; e) two sided pearson correlation; b,f) two sided paired student t test; g-h) chi square test.
Extended Data Fig. 14 EMT unlocks functional heterogeneity of pancreatic tumours.
a-b) Schematic and UMAP showing scRNA sequencing data processing of PCΩ tumours and their distribution across experimental groups and values of a transcriptomic signature of EMT. c) Scatter plot showing distribution of Simpson Index, Shannon Index values and branches as calculated by STREAM (Methods) in samples belonging to the EMT-on and EMT-off groups. Correlation and statistics are reported in figure. d) Representative stream plot showing inferred distribution of a tumour and metastasis derived from one GEMM belonging to the Vehicle and GCV group. A high number of putative transcriptomic branches can be appreciated. e) Representative H&E pictures of primary tumours belonging to Vehicle, GCV and GCV groups showing the phenotypic variability of pancreatic tumours (left) and quantification of observed phenotypes by pathological analysis in the EMT-on and EMT-off groups (N = 5 primary tumours/group). P = 0.0453. f) Schematic of the analytical workflow used for the analysis of patient-derived pancreatic tumour single cell RNA sequencing dataset retrieved from Peng et al. g) UMAP showing distribution of two groups characterized by high and low levels of EMT, as calculated by a transcriptomic signature of EMT (EMT-high and EMT-low respectively, Methods). h) Scatter plot showing distribution of Simpson Index, Shannon Index values and number of cells per each sample included in the analysis of the dataset published by Peng et al. Comparison and P values are reported in figure. Scale bars = 100 μm if not otherwise specified. P values are calculated as follows: c,h) two-tailed spearman correlation; e) chi-square test.
Supplementary information
Supplementary Information
Experimental model information. A document describing strategies for the generation of genetically engineered mouse models, genotyping, validation, as well as extended nomenclature for the models described in the main text.
Supplementary Note
Extended single-cell transcriptomic analysis. The details of scRNA-seq analysis of mouse models and patient data that were not included in the main text.
Supplementary Table 1
Genomic information of SM-GEMM.
Supplementary Table 2
Clinicogenomic and transcriptomic information of patient data obtained from TCGA, COMPASS and internal cohort of PDX-PDAC.
Rights and permissions
About this article
Cite this article
Perelli, L., Zhang, L., Mangiameli, S. et al. Evolutionary fingerprints of epithelial-to-mesenchymal transition. Nature 640, 1083–1092 (2025). https://doi.org/10.1038/s41586-025-08671-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41586-025-08671-2
This article is cited by
-
Spatially resolved analysis of TGF/BMP signalling in pancreatic ductal adenocarcinoma by digital pathology identifies patient subgroups with adverse outcome
BMC Cancer (2025)
-
Death-ision: the link between cellular resilience and cancer resistance to treatments
Molecular Cancer (2025)
-
Epithelial-to-mesenchymal transition drives cancer genomic instability
Journal of Experimental & Clinical Cancer Research (2025)
-
EMT and cancer: what clinicians should know
Nature Reviews Clinical Oncology (2025)
