Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Hepatic stellate cells suppress NK cell-sustained breast cancer dormancy

Nature volume 594, pages 566–571 (2021)Cite this article

Subjects

An Author Correction to this article was published on 11 November 2021

This article has been updated

Abstract

The persistence of undetectable disseminated tumour cells (DTCs) after primary tumour resection poses a major challenge to effective cancer treatment1,2,3. These enduring dormant DTCs are seeds of future metastases, and the mechanisms that switch them from dormancy to outgrowth require definition. Because cancer dormancy provides a unique therapeutic window for preventing metastatic disease, a comprehensive understanding of the distribution, composition and dynamics of reservoirs of dormant DTCs is imperative. Here we show that different tissue-specific microenvironments restrain or allow the progression of breast cancer in the liver—a frequent site of metastasis4 that is often associated with a poor prognosis5. Using mouse models, we show that there is a selective increase in natural killer (NK) cells in the dormant milieu. Adjuvant interleukin-15-based immunotherapy ensures an abundant pool of NK cells that sustains dormancy through interferon-γ signalling, thereby preventing hepatic metastases and prolonging survival. Exit from dormancy follows a marked contraction of the NK cell compartment and the concurrent accumulation of activated hepatic stellate cells (aHSCs). Our proteomics studies on liver co-cultures implicate the aHSC-secreted chemokine CXCL12 in the induction of NK cell quiescence through its cognate receptor CXCR4. CXCL12 expression and aHSC abundance are closely correlated in patients with liver metastases. Our data identify the interplay between NK cells and aHSCs as a master switch of cancer dormancy, and suggest that therapies aimed at normalizing the NK cell pool might succeed in preventing metastatic outgrowth.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on Springer Link
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: NK cells sustain breast cancer dormancy in the liver.
Fig. 2: NK cells sustain dormancy through IFNγ.
Fig. 3: aHSCs steer NK cell depletion and promote liver metastasis.
Fig. 4: CXCL12 mediates hepatic stellate cell-induced quiescence in NK cells.

Similar content being viewed by others

Data availability

All mass spectrometry raw data files have been deposited to the ProteomeXchange Consortium via the PRIDE60 partner repository with the data set identifier PXD015426. The mRNA sequencing data have been deposited in the Sequence Read Archive (SRA) database under BioProject accession number PRJNA576660. Source data are provided with this paper.

Code availability

The source code to replicate genomics and image analysis presented in this study is available from Zenodo at https://doi.org/10.5281/zenodo.4570079.

Change history

References

  1. Chambers, A. F., Groom, A. C. & MacDonald, I. C. Dissemination and growth of cancer cells in metastatic sites. Nat. Rev. Cancer 2, 563–572 (2002).

    Article  CAS  PubMed  Google Scholar 

  2. Polzer, B. & Klein, C. A. Metastasis awakening: the challenges of targeting minimal residual cancer. Nat. Med. 19, 274–275 (2013).

    Article  CAS  PubMed  Google Scholar 

  3. Sosa, M. S., Bragado, P. & Aguirre-Ghiso, J. A. Mechanisms of disseminated cancer cell dormancy: an awakening field. Nat. Rev. Cancer 14, 611–622 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  4. Disibio, G. & French, S. W. Metastatic patterns of cancers: results from a large autopsy study. Arch. Pathol. Lab. Med. 132, 931–939 (2008).

    Article  PubMed  Google Scholar 

  5. Diamond, J. R., Finlayson, C. A. & Borges, V. F. Hepatic complications of breast cancer. Lancet Oncol. 10, 615–621 (2009).

    Article  PubMed  Google Scholar 

  6. Cote, R. J. et al. Monoclonal antibodies detect occult breast carcinoma metastases in the bone marrow of patients with early stage disease. Am. J. Surg. Pathol. 12, 333–340 (1988).

    Article  CAS  PubMed  Google Scholar 

  7. Schlimok, G. et al. Micrometastatic cancer cells in bone marrow: in vitro detection with anti-cytokeratin and in vivo labeling with anti-17-1A monoclonal antibodies. Proc. Natl Acad. Sci. USA 84, 8672–8676 (1987).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  8. Braun, S. et al. A pooled analysis of bone marrow micrometastasis in breast cancer. N. Engl. J. Med. 353, 793–802 (2005).

    Article  CAS  PubMed  Google Scholar 

  9. Cote, R. J., Rosen, P. P., Lesser, M. L., Old, L. J. & Osborne, M. P. Prediction of early relapse in patients with operable breast cancer by detection of occult bone marrow micrometastases. J. Clin. Oncol. 9, 1749–1756 (1991).

    Article  CAS  PubMed  Google Scholar 

  10. Janni, W. et al. Persistence of disseminated tumor cells in the bone marrow of breast cancer patients predicts increased risk for relapse—a European pooled analysis. Clin. Cancer Res. 17, 2967–2976 (2011).

    Article  PubMed  Google Scholar 

  11. Klein, C. A. Cancer progression and the invisible phase of metastatic colonization. Nat. Rev. Cancer 20, 681–694 (2020).

    Article  CAS  PubMed  Google Scholar 

  12. Risson, E., Nobre, A. R., Maguer-Satta, V. & Aguirre-Ghiso, J. A. The current paradigm and challenges ahead for the dormancy of disseminated tumor cells. Nat. Cancer 1, 672–680 (2020).

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  13. Correia, A. L. & Bissell, M. J. The tumor microenvironment is a dominant force in multidrug resistance. Drug Resist. Updat. 15, 39–49 (2012).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Ghajar, C. M. et al. The perivascular niche regulates breast tumour dormancy. Nat. Cell Biol. 15, 807–817 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Oskarsson, T. et al. Breast cancer cells produce tenascin C as a metastatic niche component to colonize the lungs. Nat. Med. 17, 867–874 (2011).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Oki, T. et al. A novel cell-cycle-indicator, mVenus-p27K−, identifies quiescent cells and visualizes G0–G1 transition. Sci. Rep. 4, 4012 (2014).

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  17. Cabezas-Wallscheid, N. et al. Vitamin A–retinoic acid signaling regulates hematopoietic stem cell dormancy. Cell 169, 807–823 (2017).

    Article  CAS  PubMed  Google Scholar 

  18. Ren, D. et al. Wnt5a induces and maintains prostate cancer cells dormancy in bone. J. Exp. Med. 216, 428–449 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Sosa, M. S. et al. NR2F1 controls tumour cell dormancy via SOX9- and RARβ-driven quiescence programmes. Nat. Commun. 6, 6170 (2015).

    Article  ADS  CAS  PubMed  Google Scholar 

  20. Albrengues, J. et al. Neutrophil extracellular traps produced during inflammation awaken dormant cancer cells in mice. Science 361, eaao4227 (2018).

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  21. Coffelt, S. B. et al. IL-17-producing γδ T cells and neutrophils conspire to promote breast cancer metastasis. Nature 522, 345–348 (2015).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  22. Wculek, S. K. & Malanchi, I. Neutrophils support lung colonization of metastasis-initiating breast cancer cells. Nature 528, 413–417 (2015).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  23. Lee, J. W. et al. Hepatocytes direct the formation of a pro-metastatic niche in the liver. Nature 567, 249–252 (2019).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  24. Nielsen, S. R. et al. Macrophage-secreted granulin supports pancreatic cancer metastasis by inducing liver fibrosis. Nat. Cell Biol. 18, 549–560 (2016).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Limberis, M. P., Bell, C. L. & Wilson, J. M. Identification of the murine firefly luciferase-specific CD8 T-cell epitopes. Gene Ther. 16, 441–447 (2009).

    Article  CAS  PubMed  Google Scholar 

  26. Stripecke, R. et al. Immune response to green fluorescent protein: implications for gene therapy. Gene Ther. 6, 1305–1312 (1999).

    Article  CAS  PubMed  Google Scholar 

  27. Waldmann, T. A. The biology of interleukin-2 and interleukin-15: implications for cancer therapy and vaccine design. Nat. Rev. Immunol. 6, 595–601 (2006).

    Article  CAS  PubMed  Google Scholar 

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

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Barrow, A. D. et al. Natural killer cells control tumor growth by sensing a growth factor. Cell 172, 534–548 (2018).

    Article  CAS  PubMed  Google Scholar 

  30. Tsuchida, T. & Friedman, S. L. Mechanisms of hepatic stellate cell activation. Nat. Rev. Gastroenterol. Hepatol. 14, 397–411 (2017).

    Article  CAS  PubMed  Google Scholar 

  31. Brown, Z. J., Heinrich, B. & Greten, T. F. Mouse models of hepatocellular carcinoma: an overview and highlights for immunotherapy research. Nat. Rev. Gastroenterol. Hepatol. 15, 536–554 (2018).

    Article  CAS  PubMed  Google Scholar 

  32. Taub, R. Liver regeneration: from myth to mechanism. Nat. Rev. Mol. Cell Biol. 5, 836–847 (2004).

    Article  CAS  PubMed  Google Scholar 

  33. Bleul, C. C., Fuhlbrigge, R. C., Casasnovas, J. M., Aiuti, A. & Springer, T. A. A highly efficacious lymphocyte chemoattractant, stromal cell-derived factor 1 (SDF-1). J. Exp. Med. 184, 1101–1109 (1996).

    Article  CAS  PubMed  Google Scholar 

  34. Müller, A. et al. Involvement of chemokine receptors in breast cancer metastasis. Nature 410, 50–56 (2001).

    Article  ADS  PubMed  Google Scholar 

  35. Sugiyama, T., Kohara, H., Noda, M. & Nagasawa, T. Maintenance of the hematopoietic stem cell pool by CXCL12–CXCR4 chemokine signaling in bone marrow stromal cell niches. Immunity 25, 977–988 (2006).

    Article  CAS  PubMed  Google Scholar 

  36. Price, T. T. et al. Dormant breast cancer micrometastases reside in specific bone marrow niches that regulate their transit to and from bone. Sci. Transl. Med. 8, 340ra73 (2016).

    Article  PubMed  PubMed Central  Google Scholar 

  37. Orimo, A. et al. Stromal fibroblasts present in invasive human breast carcinomas promote tumor growth and angiogenesis through elevated SDF-1/CXCL12 secretion. Cell 121, 335–348 (2005).

    Article  CAS  PubMed  Google Scholar 

  38. Stange, D. E. et al. Expression of an ASCL2 related stem cell signature and IGF2 in colorectal cancer liver metastases with 11p15.5 gain. Gut 59, 1236–1244 (2010).

    Article  CAS  PubMed  Google Scholar 

  39. Robinson, D. R. et al. Integrative clinical genomics of metastatic cancer. Nature 548, 297–303 (2017).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  40. López-Soto, A., Gonzalez, S., Smyth, M. J. & Galluzzi, L. Control of metastasis by NK cells. Cancer Cell 32, 135–154 (2017).

    Article  CAS  PubMed  Google Scholar 

  41. Schreiber, R. D., Old, L. J. & Smyth, M. J. Cancer immunoediting: integrating immunity’s roles in cancer suppression and promotion. Science 331, 1565–1570 (2011).

    Article  ADS  CAS  PubMed  Google Scholar 

  42. Koebel, C. M. et al. Adaptive immunity maintains occult cancer in an equilibrium state. Nature 450, 903–907 (2007).

    Article  ADS  CAS  PubMed  Google Scholar 

  43. Khoo, W. H. et al. A niche-dependent myeloid transcriptome signature defines dormant myeloma cells. Blood 134, 30–43 (2019).

    Article  CAS  PubMed  Google Scholar 

  44. Alspach, E., Lussier, D. M. & Schreiber, R. D. Interferon γ and its important roles in promoting and inhibiting spontaneous and therapeutic cancer immunity. Cold Spring Harb. Perspect. Biol. 11, a028480 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  45. Burke, J. D. & Young, H. A. IFN-γ: a cytokine at the right time, is in the right place. Semin. Immunol. 43, 101280 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  46. Sahai, E. et al. A framework for advancing our understanding of cancer-associated fibroblasts. Nat. Rev. Cancer 20, 174–186 (2020).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  47. Barkan, D. et al. Metastatic growth from dormant cells induced by a Col-I-enriched fibrotic environment. Cancer Res. 70, 5706–5716 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  48. Fearon, D. T. The carcinoma-associated fibroblast expressing fibroblast activation protein and escape from immune surveillance. Cancer Immunol. Res. 2, 187–193 (2014).

    Article  CAS  PubMed  Google Scholar 

  49. Feig, C. et al. Targeting CXCL12 from FAP-expressing carcinoma-associated fibroblasts synergizes with anti-PD-L1 immunotherapy in pancreatic cancer. Proc. Natl Acad. Sci. USA 110, 20212–20217 (2013).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  50. Joyce, J. A. & Fearon, D. T. T cell exclusion, immune privilege, and the tumor microenvironment. Science 348, 74–80 (2015).

    Article  ADS  CAS  PubMed  Google Scholar 

  51. Dull, T. et al. A third-generation lentivirus vector with a conditional packaging system. J. Virol. 72, 8463–8471 (1998).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  52. Liu, H. et al. Cancer stem cells from human breast tumors are involved in spontaneous metastases in orthotopic mouse models. Proc. Natl Acad. Sci. USA 107, 18115–18120 (2010).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  53. Hoffmann, S. et al. Fast mapping of short sequences with mismatches, insertions and deletions using index structures. PLOS Comput. Biol. 5, e1000502 (2009).

    Article  MathSciNet  PubMed  PubMed Central  CAS  Google Scholar 

  54. Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014).

    PubMed  PubMed Central  Google Scholar 

  55. MacParland, S. A. et al. Single cell RNA sequencing of human liver reveals distinct intrahepatic macrophage populations. Nat. Commun. 9, 4383 (2018).

    Article  ADS  PubMed  PubMed Central  CAS  Google Scholar 

  56. Zhou, J. et al. Liver-resident NK cells control antiviral activity of hepatic T cells via the PD-1–PD-L1 axis. Immunity 50, 403–417 (2019).

    Article  CAS  PubMed  Google Scholar 

  57. Subramanian, A. et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc. Natl Acad. Sci. USA 102, 15545–15550 (2005).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  58. Ahrné, E. et al. Evaluation and improvement of quantification accuracy in isobaric mass tag-based protein quantification experiments. J. Proteome Res. 15, 2537–2547 (2016).

    Article  CAS  PubMed  Google Scholar 

  59. Glatter, T. et al. Large-scale quantitative assessment of different in-solution protein digestion protocols reveals superior cleavage efficiency of tandem Lys-C/trypsin proteolysis over trypsin digestion. J. Proteome Res. 11, 5145–5156 (2012).

    Article  CAS  PubMed  Google Scholar 

  60. Vizcaíno, J. A. et al. 2016 update of the PRIDE database and its related tools. Nucleic Acids Res. 44, D447–D456 (2016).

    Article  PubMed  Google Scholar 

Download references

Acknowledgements

We thank the following: T. Oki and T. Kitamura for providing the pMXs-IRES-puro/mVenus-p27K− vector; A. Bottos and N. E. Hynes for providing the pFU-Luc2-tdTomato vector; H. Kohler, T. Lopes, L. Raeli and E. Traunecker for assistance with FACS; T. Roloff, K. Hirschfeld, P. Demougin and C. Beisel for genomic library preparation and mRNA sequencing; the sciCORE team for their maintenance of the High Performance Computing facility at the University of Basel; S. Eppenberger-Castori for support on selecting the paired liver biopsies; S. Bichet, A. Bogucki, D. Calabrese and M-M. Coissieux for assistance with immunohistochemical staining; M. Ritter, S. D. Soysal and B.T. Preca for writing ethical permits; members of the Bentires-Alj laboratory for advice and discussion; and N. E. Hynes and S. Eppenberger-Castori for critically reading the manuscript. We also thank all the patients who donated tissues and blood to this study. A.L.C. was supported by an EMBO long-term postdoctoral fellowship (ALTF 1107-2014) and the research fund for excellent junior researchers from the University of Basel. Part of J.C.G.’s work was performed while he was in the laboratory of Mihaela Zavolan at the University of Basel supported by a SystemsX.ch Transitional Postdoctoral Fellowship (grant 51FSP0_157344) to J.C.G., a Novartis University of Basel Excellence Scholarship for Life Sciences to J.C.G., and the Swiss National Science Foundation grant 51NF40_141735 (National Center for Competence in Research ‘RNA & Disease’; co-PI Mihaela Zavolan). Research in the Bentires-Alj laboratory is supported by the Swiss Initiative for Systems Biology-SystemsX, the European Research Council (ERC advanced grant 694033 STEM-BCPC), the Swiss National Science Foundation, Novartis, the Krebsliga Beider Basel, the Swiss Cancer League, the Swiss Personalized Health Network (Swiss Personalized Oncology driver project) and the Department of Surgery of the University Hospital Basel.

Author information

Author notes

  1. Joao C. Guimaraes & Duvini De Silva

    Present address: F. Hoffmann-La Roche AG, Basel, Switzerland

  2. Ryoko Okamoto

    Present address: Department of Biosystems Science and Engineering, ETH Zürich, Basel, Switzerland

  3. Sandro Bruno

    Present address: Novartis Institutes for BioMedical Research, Basel, Switzerland

Authors and Affiliations

  1. Department of Biomedicine, University of Basel, Basel, Switzerland

    Ana Luísa Correia, Priska Auf der Maur, Duvini De Silva, Marcel P. Trefny, Ryoko Okamoto, Katrin Volkmann, Alfred Zippelius & Mohamed Bentires-Alj

  2. Friedrich Miescher Institute for Biomedical Research, Basel, Switzerland

    Ana Luísa Correia, Duvini De Silva, Ryoko Okamoto, Sandro Bruno & Mohamed Bentires-Alj

  3. Department of Surgery, University Hospital Basel, Basel, Switzerland

    Ana Luísa Correia, Priska Auf der Maur, Ryoko Okamoto, Katrin Volkmann, Walter Paul Weber & Mohamed Bentires-Alj

  4. Computational and Systems Biology, Biozentrum, University of Basel, Basel, Switzerland

    Joao C. Guimaraes

  5. Proteomics Core Facility, Biozentrum, University of Basel, Basel, Switzerland

    Alexander Schmidt

  6. Institute of Pathology Liestal, Cantonal Hospital Basel-Land, Liestal, Switzerland

    Kirsten Mertz

  7. Institute of Pathology, University Hospital Basel, Basel, Switzerland

    Luigi Terracciano

  8. Department of Medical Oncology, University Hospital Basel, Basel, Switzerland

    Alfred Zippelius & Marcus Vetter

  9. Gynecologic Cancer Center, University Hospital Basel, Basel, Switzerland

    Marcus Vetter & Christian Kurzeder

  10. Breast Center, University of Basel and University Hospital Basel, Basel, Switzerland

    Marcus Vetter, Christian Kurzeder & Walter Paul Weber

Authors
  1. Ana Luísa Correia
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Joao C. Guimaraes
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Priska Auf der Maur
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Duvini De Silva
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Marcel P. Trefny
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Ryoko Okamoto
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Sandro Bruno
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Alexander Schmidt
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Kirsten Mertz
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Katrin Volkmann
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Luigi Terracciano
    View author publications

    You can also search for this author in PubMed Google Scholar

  12. Alfred Zippelius
    View author publications

    You can also search for this author in PubMed Google Scholar

  13. Marcus Vetter
    View author publications

    You can also search for this author in PubMed Google Scholar

  14. Christian Kurzeder
    View author publications

    You can also search for this author in PubMed Google Scholar

  15. Walter Paul Weber
    View author publications

    You can also search for this author in PubMed Google Scholar

  16. Mohamed Bentires-Alj
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

A.L.C. conceived the study, conducted experiments, analysed and interpreted data and wrote the manuscript. J.C.G. contributed to experimental design, and conducted all analyses and data interpretation related to mRNA sequencing. P.A.d.M. designed and performed experiments related to CRISPR-mediated knockout of Cxcr4, and assisted with animal experiments. D.D.S. performed many experiments involving flow cytometry, and analysed and interpreted the resulting data. M.P.T. helped design and perform experiments related to NK cell-mediated cytotoxicity. R.O. assisted with animal experiments. S.B. performed experiments to harmonize nutritional requirements and assemble different cell types in the liver co-cultures. A.S. conducted proteomics experiments, and analysed and interpreted the resulting data. K.M. performed immunohistochemistry and analysed NK cell frequency on liver biopsies. K.V. conducted image acquisitions on the CQ1 Benchtop High-Content Analysis System. L.T. provided patient materials and assisted in analysing HSC frequency on liver biopsies. A.Z., M.V., W.P.W. and C.K. provided patient material. M.B.-A. conceived the study. All authors read and provided feedback on the manuscript.

Corresponding authors

Correspondence to Ana Luísa Correia or Mohamed Bentires-Alj.

Ethics declarations

Competing interests

A.L.C., P.A.d.M., M.P.T., R.O., A.S., K.M., K.V., L.T., M.V. and C.K. declare no competing interests. J.C.G. and D.D.S. are employees of F. Hoffmann–La Roche. S.B. is an employee of Novartis. A.Z. received honoraria from Bristol-Myers Squibb, Merck Sharp & Dohme, Hoffmann–La Roche, NBE Therapeutics, Secarna, ACM Pharma and Hookipa. A.Z. maintains non-commercial research agreements with Hoffmann–La Roche. A.Z. maintains further non-commercial research agreements with NBE Therapeutics, Secarna, ACM Pharma, Hookipa and BeyondSpring. M.B.-A. owns equities in and receives laboratory support and compensation from Novartis, and serves as consultant for Basilea. Outside of the submitted work, W.P.W. received research support from Takeda Pharmaceuticals International paid to the Swiss Group for Clinical Cancer Research (SAKK) and personal honoraria from Genomic Health. Support for meetings was paid to his institution from Sandoz, Genomic Health, Medtronic, Novartis Oncology, Pfizer and Eli Lilly.

Additional information

Peer review information Nature thanks Eric Vivier and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

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 Expression profiling of breast DTCs and stroma from dormant and metastatic milieus reveals the determinants of progression of breast cancer in the liver.

a, Principal component analysis (PCA) of cycling and quiescent DTCs in the MDA-MB-231 model. Transcriptional profiles cluster on the basis of cell cycle state. The dots in the plot represent DTCs isolated from different liver parts (n = 11 cycling, n = 13 quiescent; data combine three independent experiments). b, Scatter plot of mRNA expression levels (library-normalized mRNA counts) in cycling and quiescent DTCs. Shown are mean expression values for each transcript in each cell cycle state (n = 11 cycling, n = 13 quiescent). mRNAs significantly upregulated or downregulated (that is, log2(mRNA counts in cycling DTCs/mRNA counts in quiescent DTCs) > 1 and FDR < 0.01) in cycling DTCs are shown in red or blue, respectively. The dashed line indicates equal abundances in the two different conditions. c, d, GSEA comparing gene-expression data from quiescent (c) and cycling (d) DTCs. e, PCA of dormancy and metastasis stroma. Transcriptional profiles cluster on the basis of disease stage. The dots in the plot represent stroma isolated from different liver parts (n = 17 dormancy, n = 12 metastasis). f, Scatter plot of mRNA expression levels (library-normalized mRNA counts) in metastasis and dormancy stroma. Shown are mean expression values for each transcript in each stroma (n = 12 metastasis, n = 17 dormancy). mRNAs significantly upregulated or downregulated (log2(mRNA counts in metastasis stroma/mRNA counts in dormancy stroma) > 1 and FDR < 0.01) in metastasis stroma are shown in red or blue, respectively. The dashed line indicates equal abundances in the two different conditions. g, GSEA comparing gene-expression data from metastasis and dormancy liver stroma. h, Heat map depicting the hierarchical clustering of standard-score-normalized (z-score) expression level of NK cell markers55 across stroma (n = 12 metastasis, n = 17 dormancy). i, Mean ± s.e.m. mRNA fold change (log2-transformed) of NK cell markers in metastasis (n = 12) compared to dormancy (n = 17) stroma samples. Multiple-test-corrected P values for two-tailed Wald tests comparing the fold change between metastasis and dormancy samples are depicted above each dot (*P < 0.05, ***P < 0.001). j, Violin plot showing the distribution of the median standard-score normalized (z-score) expression level of NK cell markers across metastasis and dormancy stroma (n = 12 metastasis, n = 17 dormancy). Solid and dashed horizontal lines depict the median and the upper and lower quartiles, respectively. Shown is the P value for the two-tailed nonparametric Mann–Whitney U test. In c, d, g, P values were calculated by one-tailed comparisons of the empirical ES of a gene set to a null distribution of ESs derived from permuting the gene set, and then adjusted for multiple-hypotheses testing (that is, FDR).

Source data

Extended Data Fig. 2 NK cells are specifically enriched in liver dormancy milieus.

a, Flow cytometry quantification of the frequency (top) and number (bottom) of different immune cell subsets in liver parts isolated from the MDA-MB-231 model (n = 11 no tumour, n = 17 dormancy, n = 20 metastasis; data combine three independent experiments). b, c, Histological characterization of the dormant 4T07 (b) and metastatic 4T1 (c) models. Left, representative H&E-stained liver lobes. Scale bars, 2 mm. Right, examples (corresponding to i–vi from the left images) of scattered Ki67− quiescent DTCs (indicated by arrowheads), and liver metastases (surrounded by a dashed coloured line). Scale bars, 30 μm. d, Quantification of metastatic foci in livers of 4T07 and 4T1 models, normalized to the liver lobe area analysed (n = 10 4T07, n = 10 4T1; mean ± s.d.; two-tailed nonparametric Mann–Whitney U test). e, Flow cytometry quantification of the frequency (top) and number (bottom) of different immune cell populations in livers from dormant 4T07 and metastatic 4T1 models (n = 10 no tumour, n = 10 4T07, n = 10 4T1; data combine two independent experiments). f, Flow cytometry quantification of the frequency of NK and T cells, as well as T cell-activated populations, in liver sub-microenvironments from the metastatic 4T1 model (n = 10 no tumour, n = 10 dormancy, n = 10 metastasis; data combine two independent experiments). In a, e, f, mean ± s.d.; two-tailed nonparametric Kruskal–Wallis test with Dunn’s multiple comparison post-hoc test.

Source data

Extended Data Fig. 3 Both conventional and liver-resident NK cells decrease during metastatic progression.

a, The gene signature of NK cells alone—but not that of conventional NK (cNK) cells or liver-resident NK (LrNK) cells—can reliably distinguish dormancy and metastasis in the liver. Violin plots show the distribution of the median standard-score normalized (z-score) expression level of NK cell (left), cNK cell (middle) and LrNK cell (right) markers across stroma (n = 12 metastasis, n = 17 dormancy). Solid and dashed horizontal lines depict the median and the upper and lower quartiles, respectively. Shown is the P value for the two-tailed nonparametric Mann–Whitney U test. b, c, cNK and LrNK cells are similarly represented within the NK compartment across different hepatic milieus. Flow cytometry quantification of the number per gram of liver (b) or the frequency within the NK cell compartment (c) of cNK cells (CD49b+CD49a−TRAIL−) and LrNK cells (CD49b−CD49a+TRAIL+) in liver parts isolated from the MDA-MB-231 model (n = 6 no tumour, n = 12 dormancy, n = 9 metastasis; data combine two independent experiments; mean ± s.d.; two-tailed nonparametric Kruskal–Wallis test with Dunn’s multiple comparison post-hoc test).

Source data

Extended Data Fig. 4 Normalizing the NK cell pool prevents hepatic metastases.

a, Flow cytometry quantification of NK cell frequency in mice treated with IgG, anti-GM1, PBS or IL-15. Left, MDA-MB-231 model (n = 8 IgG, n = 10 anti-GM1, n = 8 PBS, n = 10 IL-15; data combine two independent experiments). Right, 4T1 model (n = 4 IgG, n = 5 anti-GM1, n = 5 PBS, n = 5 IL-15). b, Bioluminescence imaging 10 weeks after MDA-MB-231 tumour resection. c, d, Quantification of metastatic foci (c) and metastatic area (d) in livers of mice treated with IgG, anti-GM1, PBS or IL-15, normalized to the liver lobe area analysed. Left, MDA-MB-231 model (n = 8 IgG, n = 10 anti-GM1, n = 8 PBS, n = 10 IL-15; data combine two independent experiments). Right, 4T1 model (n = 5 IgG, n = 10 anti-GM1, n = 6 PBS, n = 10 IL-15). e, Experimental design for examining the effects of NK cell depletion on the 4T07 model. f, Sustained NK cell depletion reactivates dormant 4T07 DTCs in the liver. Arrowheads indicate single Ki67− quiescent DTCs. Coloured line delineates a metastasis. Scale bars, 30 μm. g, Quantification of liver metastatic foci after NK cell depletion in the 4T07 model, normalized to the liver lobe area analysed (n = 10 IgG, n = 10 anti-GM1). h, Quantification of scattered quiescent DTCs in livers of mice treated with IgG, anti-GM1, PBS or IL-15, normalized to the liver lobe area analysed (n = 8 IgG, n = 10 anti-GM1, n = 8 PBS, n = 10 IL-15; data combine two independent experiments). i, Expression of IL-15Rα in MDA-MB-231, 4T07 and 4T1 cells assessed by western blotting. ERK2 was used as a loading control. For gel source data, see Supplementary Fig. 1 (n = 3 experiments). j, Histogram of IL-15Rα measured by antibody-based staining and flow cytometry in MDA-MB-231, 4T07 and 4T1 cells. k, Quantification of the relative percentages of quiescent (Tomato+mVenus+) and cycling (Tomato+mVenus−) cancer cells after 24 h of treatment with IL-15 shows no effect on cell population ratios (n = 3 independent experiments). l–n, Flow cytometry quantification of T cell frequency (l) and activation (m, n) in livers from 4T1-injected mice treated with PBS or IL-15 (n = 5 PBS, n = 5 IL-15). In a, c, d, g, h, k–n, mean ± s.d.; two-tailed nonparametric Mann–Whitney U test.

Source data

Extended Data Fig. 5 Quiescent DTCs are not intrinsically resistant to recognition and killing by NK cells.

a, b, Mean ± s.e.m. mRNA fold change (log2-transformed) of NK cell activating (a) and inhibitory (b) ligands in cycling (n = 11) compared to quiescent DTCs (n = 13). Multiple-test-corrected P values for two-tailed Wald tests comparing the fold change between cycling and quiescent DTCs are depicted above each dot (*P < 0.05, **P < 0.01, ***P < 0.001). c, Schematic of experiment to test the sensitivity of cycling and quiescent DTCs to NK cell-mediated cytotoxicity. Human MDA-MB-231 or mouse 4T07 and 4T1 cancer cells co-expressing Tomato and mVenus-p27K− were co-cultured with NK cells derived from human blood or mouse livers, and assayed for cytolysis. d, NK cells kill DTCs regardless of their cell-cycle stage. The percentage of specifically killed cycling and quiescent cancer cells was calculated for different effector:target (E:T) ratios. For 4T07 and 4T1, n = 3 pooled mice per experiment, data combine three independent experiments; for MDA-MB-231, n = 4 healthy donors; mean ± s.d.; two-tailed nonparametric Mann–Whitney U test.

Source data

Extended Data Fig. 6 Transcriptional landscape of NK cells from dormant, metastatic and tumour-free liver milieus.

a, b, GSEA comparing gene expression data from dormancy (a) and tumour-free (b) liver NK cells (n = 10 no tumour, n = 17 dormancy). c, d, GSEA comparing gene-expression data from metastasis (c) and tumour-free (d) liver NK cells (n = 10 no tumour, n = 7 metastasis). e, GSEA comparing gene-expression data from metastasis and dormancy liver NK cells (n = 17 dormancy, n = 7 metastasis). In a–e, one-tailed comparisons of the ES of a gene set to a null distribution of ESs derived from permuting the gene set, and then adjusted for multiple-hypotheses testing (that is, FDR). f, Flow cytometry quantification of liver TNF+ NK cells. Left, liver parts from the MDA-MB-231 model (n = 6 no tumour, n = 12 dormancy, n = 9 metastasis; data combine two independent experiments). Right: livers from dormant 4T07 and metastatic 4T1 models (n = 12 no tumour, n = 12 4T07, n = 12 4T1; data combine two independent experiments; mean ± s.d.; two-tailed nonparametric Mann–Whitney U test.). g, GSEA of the Hallmark ‘IFNγ response’ pathway in DTCs (n = 11 cycling, n = 13 quiescent). NES, normalized enrichment score. h, Mean ± s.e.m. mRNA fold change (log2-transformed) of members of the IFNγ signalling pathway in cycling (n = 11) compared to quiescent DTCs (n = 13). Multiple-test-corrected P values for two-tailed Wald tests comparing the fold change between cycling and quiescent DTCs are depicted above each dot (*P < 0.05, **P < 0.01, ***P < 0.001).

Source data

Extended Data Fig. 7 aHSCs mediate NK cell depletion and promote liver metastasis.

a, Mean ± s.e.m. mRNA fold change (log2-transformed) of aHSC markers in metastasis (n = 12) compared to dormancy (n = 17). Multiple-test-corrected P values for two-tailed Wald tests comparing the fold change between metastasis and dormancy are depicted above each dot (*P < 0.05, ***P < 0.001). b, Violin plot showing the distribution of the median z-score expression level of aHSC markers across liver stroma (n = 12 metastasis, n = 17 dormancy). Solid and dashed horizontal lines depict the median and the upper and lower quartiles, respectively; two-tailed nonparametric Mann–Whitney U test. c, Quantification of α-SMA+ aHSCs after NK cell modulation. Left, MDA-MB-231 model (n = 8 IgG, n = 10 anti-GM1, n = 8 PBS, n = 10 IL-15; data combine two independent experiments). Right, 4T1 model (n = 5 IgG, n = 10 anti-GM1, n = 6 PBS, n = 10 IL-15). d, Bioluminescence imaging six weeks after tumour resection. e, Quantification of bioluminescence shows no changes in lung metastatic burden (n = 10 oil, n = 16 CCl4; data combine two independent experiments). In c, e, mean ± s.d.; two-tailed nonparametric Mann–Whitney U test.

Source data

Extended Data Fig. 8 Activation of HSCs shrinks the NK cell compartment even in the absence of tumour cells.

a, Representative micrographs of α-SMA+ aHSCs and collagen deposition in livers from non-tumour-bearing NOD-SCID mice that were treated with oil or CCl4 for one or six weeks. Scale bars, 30 μm. b, c, Flow cytometry quantification of NK cell frequency (b) and proliferation (c) in livers from non-tumour-bearing NOD-SCID mice that were treated with oil or CCl4 for one or six weeks (for each time point, n = 10 oil, n = 10 CCl4; data combine two independent experiments; mean ± s.d.; two-tailed nonparametric Mann–Whitney U test).

Source data

Extended Data Fig. 9 CXCL12 limits the proliferation of NK cells from healthy donors and patients with breast cancer with liver metastases.

a, t-distributed stochastic neighbour embedding (t-SNE) plot showing the relative expression of CXCR4 on different liver cell types based on 8,444 human liver cells previously sequenced55. Each dot represents a single cell, and cells are coloured from lowest (yellow) to highest (purple) expression. b, Histogram of CXCR4 measured by flow cytometry on human NK-92 cells. c, Exogenous CXCL12 increases the number of G0–G1 resting NK-92 cells, but it has no effect on NK cell viability (n = 5 independent experiments; mean ± s.d.; two-tailed nonparametric Kruskal–Wallis test with Dunn’s multiple comparison post-hoc test). d, Schematic of experiments to test the effect of CXCL12 on blood-derived NK cells purified from healthy donors and patients with breast cancer (BC) with liver metastases. NK cells labelled with CellTrace Violet (CTV) were primed with IL-2 and IL-15, and then expanded with IL-2 in the presence of CXCL12 alone or combined with IL-15 until assessed for division profile. e, Representative histogram of the NK cell division profile of a healthy donor. f, Quantification of the division index (that is, the average number of cell divisions a cell has undergone) of blood-derived NK cells from healthy donors (left, n = 6) and patients with breast cancer with liver metastases (right, n = 6) after treatment with CXCL12 alone or combined with IL-15. C1–C3 correspond to different concentrations of recombinant CXCL12 (C1 = 0.02 μg ml−1, C2 = 0.2 μg ml−1 and C3 = 2 μg ml−1). Mean ± s.d.; two-tailed nonparametric Kruskal–Wallis test with Dunn’s multiple comparison post-hoc test. g, Experimental schematic to assess the effects of aHSC-secreted CXCL12 on liver NK cells. Mouse NK cells were treated with conditioned medium (CM) from liver-derived aHSCs in the presence of a function-blocking antibody against CXCL12 or a control IgG, and G0–G1 resting cells were quantified after EdU incorporation. h, Flow cytometry quantification of quiescent Ki67− NK cells in mouse liver milieus (n = 6 no tumour, n = 12 dormancy, n = 9 metastasis; data combine two independent experiments; mean ± s.d.; nonparametric two-tailed Kruskal–Wallis test with Dunn’s multiple comparison post-hoc test). i, Proliferation of CXCR4+ NK cells from metastatic milieus (n = 9 metastasis; mean ± s.d.; nonparametric Mann–Whitney U test).

Source data

Extended Data Fig. 10 CXCR4 expression confers DTCs with a proliferative advantage but is not required for outgrowth.

a, Experimental design for testing the influence of aHSC-secreted CXCL12 on cancer cell proliferation. Co-cultures of hepatocytes and sparsely seeded cancer cells were exposed to recombinant CXCL12 protein or conditioned medium (CM) from aHSCs alone or in combination with anti-CXCL12, anti-CXCR4, control IgG or a CXCR4 inhibitor, and the number of cancer cells was analysed by flow cytometry. b, Quantification of the number of cancer cells in different liver-like milieus shows that CXCL12–CXCR4 signalling induces cancer cell proliferation (for each cell line, n = 5 independent experiments). c, Scheme of Cxcr4 sites targeted by single-guide RNAs to generate 4T1 Cxcr4-KO cells. d, Genotyping of clonally derived cells obtained through CRISPR–Cas9 targeting of Cxcr4. Coloured lanes represent clones selected and pooled as 4T1 Cxcr4 wild type (Cxcr4-WT) and 4T1 Cxcr4-KO lines (n = 1 PCR per clone; selected clones were also confirmed by sequencing). bp, base pair. e, Experimental design for assessing the requirement of CXCR4 for liver metastasis. f, Representative H&E-stained livers from 4T1 Cxcr4-WT and 4T1 Cxcr4-KO lines injected in BALB/c immunocompetent mice. Arrowheads and coloured lines indicate metastases. Scale bars, 2 mm. g, Quantification of liver metastatic foci in livers of oil- and CCl4-treated mice normalized to the liver lobe area analysed (n = 6 WT oil, n = 9 WT CCl4, n = 8 KO oil, n = 11 KO CCl4). h, Quantification of metastatic area in livers of oil- and CCl4-treated mice, normalized to the liver lobe area analysed (n = 6 WT oil, n = 9 WT CCl4, n = 8 KO oil, n = 11 KO CCl4). In b, g, h, mean ± s.d.; two-tailed nonparametric Mann–Whitney U test.

Source data

Extended Data Fig. 11 Activated hepatic stellate cells and CXCL12 accumulate in patients with liver metastases.

a, Staining of NK cells (CD3−CD57+) and aHSCs (α-SMA+) in paired metastases and normal adjacent tissues in liver biopsies from patients with breast cancer. Arrowheads indicate HSCs (top) and NK cells (bottom). Scale bars, 30 μm. b, Correlation between aHSCs and NK cells in paired metastases and normal adjacent tissues in liver biopsies from patients with breast cancer (n = 34 paired biopsies; Fisher’s exact test). c, Staining of NK cells (CD3−CD57+) and aHSCs (α-SMA+) in liver biopsies from patients with breast cancer with chronic liver disease but no metastases. Arrowheads indicate HSCs (top) and NK cells (bottom). Scale bars, 30 μm. d, Correlation between aHSCs and NK cells in liver biopsies from patients with breast cancer with chronic liver disease but no metastases (n = 35 biopsies; Fisher’s exact test). e, Heat map depicting the hierarchical clustering of standard-score-normalized (z-score) expression level of aHSC markers55 across normal and metastatic liver samples from patients with colon cancer38 (n = 5 normal livers, n = 18 liver metastases). f, Violin plot showing the distribution of the z-score expression level of aHSC markers across human healthy livers (n = 5) and liver metastases (n = 18). Solid and dashed horizontal lines depict the median and the upper and lower quartiles, respectively. Shown is the P value for the two-sided nonparametric Mann–Whitney U test. g, Heat map depicting the hierarchical clustering of z-score expression level of NK cell markers55 across healthy livers (n = 5) and liver metastases (n = 18) from patients with colon cancer38. h, Violin plot showing the distribution of the z-score expression level of NK markers across human healthy livers (n = 5) and liver metastases (n = 18). Solid and dashed horizontal lines depict the median and the upper and lower quartiles, respectively. Two-sided nonparametric Mann–Whitney U test. i, Scatter plot of median standard-score-normalized (z-score) expression level of HSC markers and CXCL12 expression across human liver metastases (n = 134). The Pearson correlation coefficient (R) and respective P value are also shown. The dashed line indicates the linear regression between the two estimates. FPKM, fragments per kilobase per million mapped reads.

Source data

Supplementary information

Supplementary Figures

This file contains Supplementary Figures 1-2, which show the uncropped blots and the gating strategy used to quantify liver immune cell subsets in main and Extended Data figures.

Reporting Summary

Supplementary Table 1

Differential gene expression analysis between metastasis and dormancy liver stroma. mRNA expression levels (library normalized mRNA counts) in metastasis compared to dormancy stroma (n = 12 metastases, n = 17 dormancy). Multiple test corrected P-values for two-tailed Wald tests comparing fold-changes between metastasis and dormancy.

Supplementary Table 2

Differential gene expression analysis between metastasis and dormancy liver NK cells. mRNA expression levels (library normalized mRNA counts) in metastasis compared to dormancy NK cells (n = 9 metastases, n = 12 dormancy). Multiple test corrected P-values for two-tailed Wald tests comparing fold-changes between metastasis and dormancy.

Supplementary Table 3

Proteomic analysis of aHSCs and hepatocytes secretome. Proteomic analysis of aHSCs and hepatocytes (Heps) secretome (n = 3 CM_aHSCs, n = 3 CM_Heps, normalized by n = 3 control growth medium; Bayes-moderated t-statistics, P values corrected for multiple testing using the Benjamini-Hochberg method).

Supplementary Table 4

Details on antibodies, cytokines and inhibitors used in this study.

Supplementary Table 5

Details on crRNA and genotyping primers used for CRISPR mediated knockout of CXCR4.

Source data

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Correia, A.L., Guimaraes, J.C., Auf der Maur, P. et al. Hepatic stellate cells suppress NK cell-sustained breast cancer dormancy. Nature 594, 566–571 (2021). https://doi.org/10.1038/s41586-021-03614-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41586-021-03614-z

This article is cited by

Comments

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.

Search

Advanced search

Quick links

Nature Briefing: Cancer

Sign up for the Nature Briefing: Cancer newsletter — what matters in cancer research, free to your inbox weekly.

Get what matters in cancer research, free to your inbox weekly. Sign up for Nature Briefing: Cancer