Proliferation-driving mutations in haematopoietic stem cells often result in the loss of stem-cell properties. But at least one common oncogenic mutation seems to enhance both proliferation and stem-cell self-renewal. See Letter p.143
When a stem cell divides, it can either produce differentiated cells or self-renew to produce more stem cells. Because stem cells are thought to be the cells of origin for many types of cancer, understanding what controls this decision has become a central question in stem-cell and cancer research. During the formation of mature blood cells from haematopoietic stem cells (HSCs), these processes are often diametrically opposed: blood cells are produced through a hierarchical process of proliferation and differentiation, often at the expense of the stem-cell ability to self-renew. But how is this decision altered when a stem cell acquires a cancer-driving mutation? Previous studies have shown that mutations that increase the proliferation of HSCs tend to reduce the cells' potential for self-renewal. But on page 143 of this issue, Li et al.1 report that HSCs harbouring an activating mutation of the protein Nras show not only enhanced proliferation but also enhanced self-renewal.
Nras is a member of the Ras family of proteins, which transmit cellular proliferation and survival signals in many different contexts and which are frequently mutated to become constitutively active in cancer cells. Li and colleagues found in mice that expression of an activating mutant version of the Nras gene in HSCs led to an increased number of the cells entering the cell cycle. In line with previous observations2, the Nras-mutant HSCs outcompeted normal HSCs in their ability to reconstitute haematopoiesis when both cell types were transplanted into HSC-depleted mice. But surprisingly, the researchers also found that Nras-mutant HSCs could be serially transplanted in mice through more rounds of transplantation than normal cells, demonstrating enhanced self-renewal.
To determine how one signalling molecule, mutant Nras, could confer both enhanced proliferation and self-renewal potential on HSCs, Li et al. used Nras-expressing HSCs that expressed a fluorescent 'reporter' protein, so that they could monitor cell division by the dilution of fluorescence over time. Remarkably, they observed two distinct responses: mutant Nras reduced the division and increased the self-renewal potential of one subset of HSCs, but increased the division and reduced the self-renewal potential of another subset. These findings suggest that there is a bimodal response to Nras activation in HSCs (Fig. 1).
The authors then studied signalling pathways downstream of Nras in HSCs, and observed not only activation of the MEK–ERK kinase pathway as expected, but also activation of the STAT5 signalling pathway, which is not well known as an effector of activated Ras proteins. Remarkably, deletion of just one of the two copies of the gene encoding STAT5 in the Nras-mutant HSCs attenuated the increase in both proliferation and self-renewal, suggesting that STAT5 could be a therapeutic target that eradicates not only the rapidly proliferating subset of cells, but also those cells that have enhanced self-renewal and are therefore more quiescent. These findings are of particular interest because STAT5 has been previously implicated in Ras-driven haematopoietic malignancies3, and MEK inhibition alone had a variable effect on cycling HSCs in these studies. In general, MEK inactivation does not reliably eliminate mutant HSCs in mouse models of Ras-pathway activation4. But Li and colleagues' findings indicate that mutant Nras induces aberrant signalling in HSCs that could be exploited therapeutically.
There is an expanding body of literature suggesting heterogeneity and diversity of function in HSCs. Of the most primitive (least differentiated) HSCs, some are poised for proliferation and differentiation, whereas others are programmed for quiescence, and these cell populations exist in a dynamic equilibrium5. It is also known that individual HSCs do not have identical lineage potential6,7. The bimodal effect defined by Li et al. could be explained by this functional heterogeneity, with some cells responding to mutation by enhancing their self-renewal potential and quiescence, and others responding with enhanced proliferation and differentiation. This leads to the question of whether such bimodal behaviour is the result of a stochastic response or is determined by definable subsets of HSC responding in a predictable fashion. In light of recent data from normal HSCs5,6,7, we would predict the latter, but further work is required to test this.
Another intriguing question raised by this study is whether the bimodal effect of activating mutations is unique to Nras, or whether it is also a feature of other oncogenic mutations, such as the mutations in the Ras family member Kras that are seen in many solid tumours and in some haematopoietic cancers. In mouse models, both Kras8 and Nras9 mutations lead to excess production of cells from the bone marrow that seems to be initiated from primitive HSCs, although the type of leukaemia that arises differs (T-ALL and AML, respectively). It is unclear whether Kras activation increases HSC self-renewal potential in a similar manner to Nras activation, and this prompts the question of whether HSC responses to different Ras mutations could determine the type of leukaemia that develops. It will be important to study not only the similarities and differences in how these and other oncogenic mutations affect cell signalling and gene expression, but also the cellular contexts required for such responses.
It is becoming increasingly clear that cellular context can influence the response to an oncogenic mutation10,11. This suggests that it is no longer sufficient to know simply which mutations are present in a tumour; we must also consider the influence of where and when. Thus, as we learn more about the mutations that occur in tumour cells, we will need to assess these mutations in functional assays such as those described by Li et al., to obtain a more accurate picture of the effects of mutations in specific cell types and at specific points in development. Such experiments will further enhance our understanding of the complex cellular heterogeneity found in cancers.
Li, Q. et al. Nature 504, 143–147 (2013).
Wang, J. et al. Blood 121, 5203–5207 (2013).
Kotecha, N. et al. Cancer Cell 14, 335–343 (2008).
Chang, T. et al. J. Clin. Invest. 123, 335–339 (2013).
Wilson, A. et al. Cell 135, 1118–1129 (2008).
Dykstra, B. et al. Cell Stem Cell 1, 218–229 (2007).
Yamamoto, R. et al. Cell 154, 1112–1126 (2013).
Sabnis, A. J. et al. PLoS Biol. 7, e1000059 (2009).
Li, Q. et al. Blood 117, 2022–2032 (2011).
Wang, Y. et al. Science 327, 1650–1653 (2010).
Friedmann-Morvinski, D. et al. Science 338, 1080–1084 (2012).