• A Corrigendum to this article was published on 13 December 2017

This article has been updated


Metastasis is the leading cause of cancer-related deaths; metastatic lesions develop from disseminated cancer cells (DCCs) that can remain dormant1. Metastasis-initiating cells are thought to originate from a subpopulation present in progressed, invasive tumours2. However, DCCs detected in patients before the manifestation of breast-cancer metastasis contain fewer genetic abnormalities than primary tumours or than DCCs from patients with metastases3,4,5. These findings, and those in pancreatic cancer6 and melanoma7 models, indicate that dissemination might occur during the early stages of tumour evolution3,8,9. However, the mechanisms that might allow early disseminated cancer cells (eDCCs) to complete all steps of metastasis are unknown8. Here we show that, in early lesions in mice and before any apparent primary tumour masses are detected, there is a sub-population of Her2+p-p38lop-Atf2loTwist1hiE-cadlo early cancer cells that is invasive and can spread to target organs. Intra-vital imaging and organoid studies of early lesions showed that Her2+ eDCC precursors invaded locally, intravasated and lodged in target organs. Her2+ eDCCs activated a Wnt-dependent epithelial–mesenchymal transition (EMT)-like dissemination program but without complete loss of the epithelial phenotype, which was reversed by Her2 or Wnt inhibition. Notably, although the majority of eDCCs were Twist1hiE-cadlo and dormant, they eventually initiated metastasis. Our work identifies a mechanism for early dissemination in which Her2 aberrantly activates a program similar to mammary ductal branching that generates eDCCs that are capable of forming metastasis after a dormancy phase.

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Change history

  • 13 December 2017

    Please see accompanying Corrigendum (http://doi.org/10.1038/nature24666). In Extended Data Fig. 4g of this Letter, the rabbit IgG control image is duplicated from the source image in Extended Data Fig. 4f. See Supplementary Information to the Corrigendum for the revised Extended Data Fig. 4g and for the raw microscope output images.


  1. 1.

    , & Mechanisms of disseminated cancer cell dormancy: an awakening field. Nat. Rev. Cancer 14, 611–622 (2014)

  2. 2.

    & Hallmarks of cancer: the next generation. Cell 144, 646–674 (2011)

  3. 3.

    et al. Genomic analysis of single cytokeratin-positive cells from bone marrow reveals early mutational events in breast cancer. Cancer Cell 8, 227–239 (2005)

  4. 4.

    et al. Comparative genomic hybridization, loss of heterozygosity, and DNA sequence analysis of single cells. Proc. Natl Acad. Sci. USA 96, 4494–4499 (1999)

  5. 5.

    et al. From latent disseminated cells to overt metastasis: genetic analysis of systemic breast cancer progression. Proc. Natl Acad. Sci. USA 100, 7737–7742 (2003)

  6. 6.

    et al. EMT and dissemination precede pancreatic tumor formation. Cell 148, 349–361 (2012)

  7. 7.

    et al. Tumor cells disseminate early, but immunosurveillance limits metastatic outgrowth, in a mouse model of melanoma. J. Clin. Invest. 120, 2030–2039 (2010)

  8. 8.

    et al. Systemic spread is an early step in breast cancer. Cancer Cell 13, 58–68 (2008)

  9. 9.

    et al. Disseminated tumor cells in the bone marrow of patients with ductal carcinoma in situ. Int. J. Cancer 129, 2522–2526 (2011)

  10. 10.

    et al. p38α signaling induces anoikis and lumen formation during mammary morphogenesis. Sci. Signal. 4, ra34 (2011)

  11. 11.

    Validity of mouse mammary tumour models for human breast cancer: comparative pathology. Microsc. Res. Tech. 52, 224–230 (2001)

  12. 12.

    et al. 14-3-3ζ cooperates with ErbB2 to promote ductal carcinoma in situ progression to invasive breast cancer by inducing epithelial–mesenchymal transition. Cancer Cell 16, 195–207 (2009)

  13. 13.

    et al. p38 maintains E-cadherin expression by modulating TAK1-NF-κB during epithelial-to-mesenchymal transition. J. Cell Sci. 123, 4321–4331 (2010)

  14. 14.

    et al. Setup and use of a two-laser multiphoton microscope for multichannel intravital fluorescence imaging. Nat. Protocols 6, 1500–1520 (2011)

  15. 15.

    et al. Metastatic latency and immune evasion through autocrine inhibition of WNT. Cell 165, 45–60 (2016)

  16. 16.

    et al. Activation of AXIN2 expression by β-catenin-T cell factor. A feedback repressor pathway regulating Wnt signaling. J. Biol. Chem . 277, 21657–21665 (2002)

  17. 17.

    et al. Canonical and noncanonical Wnts use a common mechanism to activate completely unrelated coreceptors. Genes Dev. 24, 2517–2530 (2010)

  18. 18.

    , , & Emt: 2016. Cell 166, 21–45 (2016)

  19. 19.

    et al. TGF-β2 dictates disseminated tumour cell fate in target organs through TGF-β-RIII and p38α/β signalling. Nat. Cell Biol. 15, 1351–1361 (2013)

  20. 20.

    et al. Essential function of Wnt-4 in mammary gland development downstream of progesterone signaling. Genes Dev. 14, 650–654 (2000)

  21. 21.

    et al. Detection of circulating pancreas epithelial cells in patients with pancreatic cystic lesions. Gastroenterology 146, 647–651 (2014)

  22. 22.

    et al. Metastatic colonization requires the repression of the epithelial–mesenchymal transition inducer Prrx1. Cancer Cell 22, 709–724 (2012)

  23. 23.

    To differentiate or not – routes towards metastasis. Nat. Rev. Cancer 12, 425–436 (2012)

  24. 24.

    et al. Stem-cell-like properties and epithelial plasticity arise as stable traits after transient Twist1 activation. Cell Reports 10, 131–139 (2015)

  25. 25.

    et al. Epithelial-to-mesenchymal transition is not required for lung metastasis but contributes to chemoresistance. Nature 527, 472–476 (2015)

  26. 26.

    et al. p38α suppresses normal and cancer cell proliferation by antagonizing the JNK–c-Jun pathway. Nat. Genet. 39, 741–749 (2007)

  27. 27.

    et al. JNK suppresses tumor formation via a gene-expression program mediated by ATF2. Cell Reports 9, 1361–1374 (2014)

  28. 28.

    et al. Suppressor role of activating transcription factor 2 (ATF2) in skin cancer. Proc. Natl Acad. Sci. USA 105, 1674–1679 (2008)

  29. 29.

    et al. Cancers of unknown primary site: ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Annals Oncol. 26, v133–v138 (2015)

  30. 30.

    , , , & Breast cancer mortality after a diagnosis of ductal carcinoma in situ. JAMA Oncol. 1, 888–896 (2015)

  31. 31.

    et al. Early dissemination seeds metastasis in breast cancer. Nature (2016)

  32. 32.

    , & Morphogenesis and oncogenesis of MCF-10A mammary epithelial acini grown in three-dimensional basement membrane cultures. Methods 3, 256–268 (2013)

  33. 33.

    & Modelling glandular epithelial cancers in three-dimensional cultures. Nat. Rev. Cancer 5, 675–688 (2015)

  34. 34.

    , & NIH Image to ImageJ: 25 years of image analysis. Nat. Methods 9, 671–675 (2012)

  35. 35.

    , & A pyramid approach to subpixel registration based on intensity. IEEE Trans. Image Process. 7, 27–41 (1998)

Download references


We thank R. Parsons and P. Polulikakos for PI3K and AKT inhibitors, S. Aaronson and H.-C. Wen for WNT3A, SFRP1 and DKK1 reagents and expertise. Grant support:. HHMI (R.J.D.). SWCRF (J.A.A.-G. and E.F.F.), CA109182, CA196521 (J.A.A.-G.), CA163131 (J.A.A-G and J.C.), CA100324 (J.C), F31CA183185 (K.H.), BC132674 (J.A.A.-G and J.C.), BC112380 (M.S.S.). NIH 1S10RR024745. Microscopy CoRE at ISMMS. DFG KL 1233/10-1 and the ERC (322602) (C.A.K.).

Author information

Author notes

    • Kathryn L. Harper
    •  & Maria Soledad Sosa

    These authors contributed equally to this work.


  1. Division of Hematology and Oncology, Department of Medicine, Department of Otolaryngology, Department of Oncological Sciences, Tisch Cancer Institute, Black Family Stem Cell Institute, Icahn School of Medicine at Mount Sinai, 1 Gustave L. Levy Place, New York, New York 10029, USA

    • Kathryn L. Harper
    • , Maria Soledad Sosa
    • , Julie F. Cheung
    • , Rita Nobre
    • , Alvaro Avivar-Valderas
    • , Chandandaneep Nagi
    • , Eduardo F. Farias
    •  & Julio A. Aguirre-Ghiso
  2. Department of Anatomy and Structural Biology, Integrated Imaging Program, Gruss Lipper Biophotonics Center, Albert Einstein College of Medicine, 1300 Morris Park Ave, Bronx, New York, New York 10461, USA

    • David Entenberg
    •  & John Condeelis
  3. Experimental Medicine and Therapy Research, University of Regensburg, 93053 Regensburg, Germany

    • Hedayatollah Hosseini
    •  & Christoph A. Klein
  4. Howard Hughes Medical Institute, University of Massachusetts Medical School, Worcester, Massachusetts 01605, USA

    • Nomeda Girnius
    •  & Roger J. Davis
  5. Project group “Personalized Tumour Therapy”, Fraunhofer Institute for Toxicology und Experimental Medicine, 93053 Regensburg, Germany

    • Christoph A. Klein


  1. Search for Kathryn L. Harper in:

  2. Search for Maria Soledad Sosa in:

  3. Search for David Entenberg in:

  4. Search for Hedayatollah Hosseini in:

  5. Search for Julie F. Cheung in:

  6. Search for Rita Nobre in:

  7. Search for Alvaro Avivar-Valderas in:

  8. Search for Chandandaneep Nagi in:

  9. Search for Nomeda Girnius in:

  10. Search for Roger J. Davis in:

  11. Search for Eduardo F. Farias in:

  12. Search for John Condeelis in:

  13. Search for Christoph A. Klein in:

  14. Search for Julio A. Aguirre-Ghiso in:


K.L.H. designed, performed experiments, analysed data and co-wrote the manuscript; M.S.S. designed experimental approach, performed experiments, executed intravital imaging, provided oversight, analysed data and co-wrote the manuscript; D.E. designed and executed intravital imaging, analysed data and co-wrote the manuscript; H.H. provided materials and analysed data; A.A.V. performed experiments; C.N. provided materials and histopathological analysis; J.F.C. managed mouse colonies and performed experiments; R.N. performed experiments and analysed data; N.G. maintained the Mkk3/Mkk6 wild-type and knockout mice and provided materials; R.J.D. provided materials and co-wrote manuscript; C.A.K. provided input for the writing of the manuscript; J.C. designed intra-vital experiments, analysed data and co-wrote the manuscript; E.F.F. provided expertise and analysed data; J.A.A.-G. designed and optimized experimental approach, provided general oversight, collected microscopy data, analysed data and co-wrote the manuscript.

Competing interests

J.A.A.-G. receives funding from E. Lilly and co. Jo.C. is a consultant for, and has equity in MetaStat, Inc.; and is a consultant for Deciphera Pharmaceuticals.

Corresponding authors

Correspondence to Maria Soledad Sosa or Julio A. Aguirre-Ghiso.

Reviewer Information

Nature thanks M. Bissell, C. Ghajar and the other anonymous reviewer(s) for their contribution to the peer review of this work.

Extended data

Supplementary information

PDF files

  1. 1.

    Supplementary Figure 1

    This file contains the raw data for Extended Data Figures 2c, 3d, 6d.


  1. 1.

    Spinning-disc confocal imaging of MMTV-Her2 organoids reveals microinvasion

    Two independent MMTV-Her2 organoids cultured for 4 days in Matrigel were imaged for 2 hours and frames were collected every 20 minutes. Top row shows the differential interference contrast (DIC) images of the two organoids with cell movement and microinvading cells toward the bottom of the field of view. The bottom row shows a variance filter from Image J software that highlights the cell edges and allows having a bright view of the microinvading cells from the organoids into the Matrigel.

  2. 2.

    Two-photon imaging of MMTV-Her2-CFP mice at 10 weeks of age

    Intra-vital imaging of mammary glands in MMTV-Her2-CFP mice was performed using a specially designed mammary gland imaging window (see Experimental Procedures for details). CFP+/Her2+ cells can be seen in the cyan/light cyan signal and vasculature was delimited using i.v. delivered dextran-rhodamine (red). Note the apparent normal duct architecture at this time, which remained organized with some intra-ductal movement observed mainly in luminal spaces. Still image of this video is shown in Figure 2a left panel.

  3. 3.

    Two-photon imaging of MMTV-Her2-CFP mice at age 15 weeks

    Mammary gland imaging windows were used as in Video S2 in MMTV-Her2-CFP mice at age 15 weeks when early lesions are present (Extended Data 1a-c). Motile single CFP+ cells were found micro-invading from lesions and are outlined in purple and green in the video. CFP+/Her2+ cells can be seen in cyan and vasculature was stained with dextran-rhodamine (Red). Note that here a greater digital zoom was used to show in detail the motile cells micro-invading from the ductal structure that occupies the lower right quadrant of the video. Still image of video is shown in Figure 2a middle panel.

  4. 4.

    Two-photon imaging of MMTV-Her2-CFP mice at age 18 weeks

    Mammary gland imaging windows were used to image ducts of MMTV-Her2-CFP mice at age 18 weeks when early stage lesions are present. Single cells were found microinvading from the early stage ducts (upper right) as outlined in purple in the video. One single cell can be traced through the stroma between ducts. CFP+/Her2+ cells can be seen in cyan and vasculature was stained with dextran-rhodamine (red), which also revealed macrophages that capture the labeled dextran. Still images of video are shown in Figure 2a right panel.

  5. 5.

    Intra-vital imaging of multiple micro-invasion events in MMTV-Her2-CFP mice treated for two weeks with SB203580

    Imaging of MMTV-Her2-CFP mammary glands following treatment for two weeks with SB203580 (10mg/kg, i.p every 48hr). This video shows an early lesion and CFP+/Her2+ cells in cyan and vasculature stained with dextran-rhodamine (red). Note the luminal filling of ducts and increased movement of cells along the ductal walls into the stroma and into a low contrast vessel between the two lobules of the ductal structure. Still images are in Figure 2b.

  6. 6.

    Digital magnification of the region of Video S5 highlighted in a dotted line square inFigure 2b (inset b’)

    Note the amoeboid like movement of these CFP+/Her2+ cells micro-invading cells breaking the lesion boundaries into the stroma. Note also that the motility seem to occur at the single cell level and not collectively. Still image of video is shown in Figure 2b’.

  7. 7.

    3D rendering of the micro-invasion events in MMTV-Her2-CFP mice treated with the p38a/b inhibitor SB203580

    3D reconstruction of Video S6 revealed that CFP+/Her2+ cells were entering the lumen of the blood vessel (RED) as revealed by the yellow overlap signal and suggesting active intravasating capacity as supported by the video S8 and the detection of CTCs and DTCs in Extended data 4h, i, j. Entry into the vessel lumen is marked when CFP+ and Dextran-Rhodamine (Red) signals overlap shown in yellow. Still images of video are shown in Figure 2c.

  8. 8.

    Intravasation can be detected in MMTV-Her2-CFP mice treated with the p38a/b inhibitor SB203580

    Digital Magnification of an early lesion reveals a CFP+ cells entering the vasculature from the stroma in SB203580 treated mice. Left panel shows original imaging with a delineated vessel (yellow line) and the intravasating cell (purple line). Right panel shows a digital segmentation to isolate the vasculature (RED) and the intravasating eDCC precursor (CYAN) that turns yellow when the optical signals overlap. Still images of this sequence are shown in Figure 2d.

About this article

Publication history






Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.