Tumours build their niche


It emerges that tumour cells can give rise to non-dividing cells that form part of the supporting microenvironment known as the niche. These niche cells secrete proteins that drive tumour growth and progression. See Letters p.355 & p.360

The development of a tumour resembles, in some aspects, the development of a multicellular organism, except that the coordination of events such as cell division, cell-fate determination and tissue organization is dysregulated. The signalling pathways and cells that drive tumour growth and the formation of different cell types within a tumour (tumour heterogeneity) are poorly understood. In two papers, Tammela et al.1 (page 355) and Lim et al.2 (page 360) investigate lung tumours, and make a surprising discovery about how some tumour cells drive cancer development.

Secreted Wnt proteins function in a signalling pathway that controls myriad biological processes throughout development and adult life. Dysregulation of this signalling cascade underlies many diseases, from developmental disorders to cancer3. Tammela et al. investigated the role of Wnt signalling in a type of tumour called a lung adenocarcinoma. The authors found that mouse and human adenocarcinomas that had reached an advanced and aggressive stage were heterogeneous and contained at least two cell subpopulations. One of these responded to Wnt proteins, activated the downstream Wnt signalling pathway and expressed the protein Lgr5, a marker often used to identify stem-cell populations in the gut and other organs. The other cell population produced high levels of Wnt and expressed an enzyme called porcupine, which adds a lipid chain to Wnt during the process by which mature Wnt proteins are formed4.

Using a cell-lineage-tracing approach, Tammela and colleagues demonstrated in mouse tumours that the cell population that expresses porcupine (porcupine+ cells) derives from the tumour-cell population that expresses Lgr5 (Lgr5+ cells). When the authors treated a mouse model of lung adenocarcinoma with a molecule that inhibits porcupine, the expression of genes involved in the tumour's Wnt signalling pathway decreased, and the number of tumour cells was significantly reduced. This suggests that porcupine+ cells provide a source of Wnt in the tumour niche that drives Wnt signalling in the tumour cells (Fig. 1) .

Figure 1: Tumour cells and their niche.

a, Using a cell-lineage-tracing approach, Tammela et al.1 found that lung tumours called adenocarcinomas contain tumour cells that divide to produce two cell types: a tumour cell and, unexpectedly, a support cell that is not a tumour cell and that forms part of the tumour niche. b, The authors found that a niche cell (blue) derived from a tumour cell expresses the enzyme porcupine. In the endoplasmic reticulum (ER) organelle, this enzyme adds a lipid chain to the immature form of the Wnt signalling protein and enables the formation of mature Wnt that is secreted from the cell4. Wnt binding to the Wnt receptor on a tumour cell (red) activates the cell's Wnt signalling pathway, which drives tumour growth. Lim et al.2 investigated a different type of lung tumour (not shown), and report that niche cells can also support tumour growth through the secretion of a protein that activates the Notch signalling pathway in tumour cells.

The authors identified a subpopulation of human lung adenocarcinoma cells in close proximity to porcupine+ cells. This could indicate that the scenario involving Wnt-producing and Wnt-responding cells observed in the mouse tumour model might also occur in human lung tumours. The Wnt proteins Wnt5a and Wnt7b were both upregulated in human and mouse tumours relative to their expression in non-tumoral lung tissue, a finding consistent with the proposal that these proteins have a key role in tumour development and progression.

Evidence for the concept of a stem cell being 'fuelled' by a Wnt-producing niche cell has been seen in other systems. For example, in the gut, Wnt3a is produced by Paneth cells that derive from Lgr5+ stem cells, and these Paneth cells provide a niche that aids tissue homeostasis5. Tammela and colleagues' work now shows that this type of system can occur in tumour development, and their results answer the previously raised question3 of whether cancer stem-cell behaviours could be controlled by Wnt signalling.

The Notch receptor is another protein that, like Wnt, is intimately involved in the control of cell fates during development and tissue homeostasis. The mechanism of Notch signalling is apparently simple: after Notch binds its ligand molecule, the intracellular domain of the protein is cleaved and acts as a transcription factor to activate gene expression6. However, the role of Notch in cancer is complex, and Notch signalling has been implicated in both tumour promotion and tumour suppression7. Lim et al. now shed light on these opposing roles.

Small-cell lung cancers are derived from neuroendocrine cells8, which respond to Notch during normal embryonic development9,10. Investigating mouse small-cell lung cancers, Lim and colleagues identified a population of Notch-ligand-producing neuroendocrine tumour cells that signal to non-neuroendocrine tumour cells expressing the protein Hes1 and the receptors Notch1, Notch2 and Notch3. The authors found that Notch signalling imposed this non-neuroendocrine program of development on a subset of the tumour cells. The Hes1+ population formed slow-growing tumours, suggestive of a tumour-suppressive role for Notch signalling in that context. However, the authors also found a tumour-promoting role for Notch, because non-neuroendocrine tumour cells produce the protein midkine, a ligand for Notch2, which promotes the growth of neuroendocrine tumour cells in culture, thus maintaining the tumour.

Lim and colleagues' work was mainly carried out in vitro, and used either isolated human cancer cells or neuroendocrine cell lines. It therefore remains to be determined whether interactions between neuroendocrine and non-neuroendocrine cells in vivo might also be influenced by contact between cells and by the Notch-mediated phenomenon of lateral inhibition (in which the interaction between two adjacent cells results in the formation of two different cell types). Additional experiments, such as lineage tracing of the two cell types, will help to determine whether the proposed relationships between the cells exist in vivo.

Many questions remain. It has been proposed11 that, during homeostasis, neuroendocrine cells signal to their neighbours through Notch, possibly creating a specific stem-cell niche, but this remains a contentious point that should be reassessed in light of the latest results. Do populations of Wnt-responsive cells behave as stem cells in normal lung homeostasis? One could speculate that, in homeostasis, stem cells in the lung also generate their own niches (producing Wnt and/or Notch ligands) and, similar to the situation in tumours, these could also coexist in a dynamic equilibrium.

Another question is whether the roles of Wnt and Notch are specific to lung adenocarcinomas and small-cell lung cancer, respectively, or whether these signalling molecules have roles in other tumour types. If the latter, it would be interesting to discover whether the same type of niche cell produces both factors, or whether each is produced by specialized cell types. Perhaps the two pathways are intimately connected in lung-tumour formation, similarly to how Notch works with Wnt to control cell-fate transitions during development and in adult homeostasis6.

The observations by Tammela et al. and Lim et al. that lung tumour cells can produce their own niche has a parallel in homeostasis in the lung airways, in which stem cells signal to their immediate descendants in a Notch-dependent manner to mediate cell-fate decisions12. This potentially implicates the rewiring of normal homeostatic stem-cell mechanisms in tumour growth. The stage is now set to investigate such previously unsuspected aspects of lung-cancer biology.Footnote 1


  1. 1.

    See all news & views


  1. 1

    Tammela, T. et al. Nature 545, 355–359 (2017).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Lim, J. S. et al. Nature 545, 360–364 (2017).

    ADS  CAS  Article  Google Scholar 

  3. 3

    Clevers, H. & Nusse, R. Cell 149, 1192–1205 (2012).

    CAS  Article  Google Scholar 

  4. 4

    Kadowaki, T., Wilder, E., Klingensmith, J., Zachary, K. & Perrimon, N. Genes Dev. 10, 3116–3128 (1996).

    CAS  Article  Google Scholar 

  5. 5

    Sato, T. E. Nature 469, 415–418 (2011).

    ADS  CAS  Article  Google Scholar 

  6. 6

    Hayward, P., Kalmar, T. & Martinez Arias, A. Development 135, 411–424 (2008).

    CAS  Article  Google Scholar 

  7. 7

    Koch, U. & Radtke, F. Cell. Mol. Life Sci. 64, 2746–2762 (2007).

    CAS  Article  Google Scholar 

  8. 8

    Sutherland, K. D. et al. Cancer Cell 19, 754–764 (2011).

    CAS  Article  Google Scholar 

  9. 9

    Morimoto, M., Nishinakamura, R., Saga, Y. & Kopan, R. Development 139, 4365–4373 (2012).

    CAS  Article  Google Scholar 

  10. 10

    Noguchi, M., Sumiyama, K. & Morimoto, M. Cell Rep. 13, 2679–2686 (2015).

    CAS  Article  Google Scholar 

  11. 11

    Guha, A., Deshpande, A., Jain, A., Sebastiani, P. & Cardoso, W. V. Cell Rep. 19, 246–254 (2017).

    CAS  Article  Google Scholar 

  12. 12

    Pardo-Saganta, A. et al. Nature 523, 597–601 (2015).

    ADS  CAS  Article  Google Scholar 

Download references

Author information



Corresponding authors

Correspondence to Meritxell Huch or Emma L. Rawlins.

Related links

Related links

Related links in Nature Research

Cancer: Tumour stem-cell surprises

Cancer: Double trouble for tumours

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Huch, M., Rawlins, E. Tumours build their niche. Nature 545, 292–293 (2017).

Download citation

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.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing