Stem cells

Chasing blood

Many experiments have probed the mechanisms by which transplanted stem cells give rise to all the cell types of the blood, but it emerges that the process is different in unperturbed conditions. See Letter p.542

Blood is one of the most dynamic tissues in the human body, with millions of cells being produced each second. But how blood-cell production occurs under unperturbed conditions, and which stem and progenitor cell types are responsible for stable maintenance of the blood, has been unclear. In this issue, Busch et al.1 (page 542) use a genetic labelling strategy to gain insight into blood-cell production under normal conditions. They reveal that an unexpected subset of blood stem cells is the major player in day-to-day blood-cell production.

Our understanding of haematopoiesis (the process through which all the cells of the blood are generated) is built on experiments in which blood cells are transplanted into recipients whose bone marrow (the source of blood-cell production) has been depleted through a process called myeloablation. The cells' progeny are then tracked over time2. Such work has helped to define key molecular and functional properties of blood stem cells, and to segregate the cells into different subsets, or compartments, on the basis of the time over which they contribute to haematopoiesis and on the lineages of cells that they can generate. At the top of the hierarchy are long-term haematopoietic stem cells (LT-HSCs), which give rise to short-term haematopoietic stem cells (ST-HSCs), both of which are named in accordance with the time over which they contribute to post-transplantation haematopoiesis. Both LT-HSCs and ST-HSCs exhibit self-renewal potential and typically give rise to all blood-cell lineages.

Only recently have we had the tools to extend our understanding of post-transplantation haematopoiesis to unperturbed conditions. Busch and colleagues' technique for studying unperturbed haematopoiesis is the cellular equivalent of classic 'pulse–chase' experiments, in which intracellular molecular components are traced by transient labelling of molecules to discover biosynthetic pathways. In the current study, the authors engineered mice such that LT-HSCs expressing the gene Tie2 could be genetically labelled with a fluorescent protein. Those cells and all their descendants will then fluoresce regardless of whether or not they express Tie2, allowing tracing of the cell lineages arising from the labelled stem cells. Following labelling, the authors performed a battery of assays at different times, and used mathematical modelling to analyse the resulting data (which were supported by in vivo stem-cell-transplantation experiments). This enabled them to define the rates of transition between blood-cell compartments, the length of time that cells spent in each compartment and the relative cell numbers in the different compartments (Fig. 1).

Figure 1: A balance in the blood.

During blood-cell production (haematopoiesis), long-term haematopoietic stems cells (LT-HSCs) give rise to short-term HSCs (ST-HSCs), which eventually produce the cells of the peripheral blood system. a, Busch et al.1 investigated the numbers of each cell type over time. Numbers in each compartment rise during embryonic development to reach steady-state levels, which are maintained throughout adulthood. The death of peripheral blood cells owing to the introduction of a cytotoxic agent does not affect the numbers in either stem-cell compartment. If LT-HSCs and ST-HSCs are destroyed by myeloablation (depletion of bone-marrow cells), the numbers of peripheral blood cells dip, but quickly return to normal. LT-HSCs are derived from many clones under normal conditions, but fewer clones typically contribute to post-transplantation haematopoiesis. b, The authors also investigated the rate at which LT-HSCs transition to ST-HSCs, and at which ST-HSCs transition to peripheral blood cells in unperturbed and post-transplantation conditions (development was not studied). The rate of transition out of the ST-HSC compartment is consistently higher than the rate of transition from the LT-HSC compartment, implying that ST-HSCs are the major players in unperturbed haematopoiesis.

One of Busch and co-workers' central findings is that large numbers of ST-HSCs are responsible for most blood-cell production throughout the lifetime of the animal, with LT-HSCs participating to only a limited extent. This is perhaps not surprising, given that ST-HSCs proliferate faster than LT-HSCs. The authors also infer that at least 30% of LT-HSCs (around 5,000 cells in an adult mouse) go on to give rise to differentiated blood-cell lineages. This suggests that, in unperturbed conditions, the composition of the blood is highly polyclonal — that is, it is derived from many different stem cells.

The high level of polyclonality inferred by these experiments is in contrast to the findings of many transplant studies, in which a handful of individual LT-HSCs typically repopulate the entire blood system. However, to some extent, this dogma may be associated with the fact that often only a few transplanted cells successfully engraft in the bone marrow under normal experimental conditions. Busch et al. provide evidence suggesting that, unlike the case in unperturbed haematopoiesis, LT-HSCs are the more important contributors to the haematopoiesis that occurs during embryonic development or after treatment with a cytotoxic agent that kills blood cells. This finding suggests that feedback signalling from mature blood and progenitor cells3 to the LT-HSC compartment is key to controlling transition rates between compartments.

A paper published last year4 described the use of another genetic technique to track the contribution of a few thousand blood-stem-cell clones (each tagged with a different molecular signature) to the peripheral blood under normal conditions. A key finding of this work was that two blood-cell lineages, myeloid and lymphoid, were populated with cells harbouring different tags. This result agrees with Busch and colleagues' findings, suggesting that ST-HSCs that are beginning to become restricted to one lineage are the predominant source of haematopoiesis under unperturbed conditions. Also in agreement with Busch et al., the paper reported that the peripheral blood was highly polyclonal.

Busch and colleagues' work, along with related papers4,5 examining unperturbed haematopoiesis in vivo, opens up several avenues for future research. A study6 involving transplantation of blood stem cells revealed a high degree of variability between the clonal populations in the post-transplantation blood, in both their size and their ability to give rise to different lineages. This variability can be attributed to various mechanisms, both intrinsic (different developmental potencies for the stem cells) and extrinsic (the surrounding microenvironment). Because of our limited understanding of clone sizes and lineage-contribution bias in unperturbed conditions, it is unclear how this post-transplantation variability relates to Busch and colleagues' observations under normal conditions. It could be that large numbers of contributing ST-HSCs are needed to ensure that robust polyclonality is maintained during normal haematopoiesis — something that is compromised when only limiting numbers of stem cells contribute to post-transplantation haematopoiesis. Alternatively, the reported post-transplantation clonal variability could be a stem-cell-intrinsic phenomenon specific to the post-transplantation environment. The ability to distinguish between LT- and ST-HSCs in vivo will be required to address this.

Another key area for investigation is why ST-HSCs can contribute to long-term haematopoiesis in unperturbed conditions, but contribute only transiently to post-transplantation haematopoiesis. Whether this reflects an effect of the post-transplant environment on these cells, or whether it is related to the increased proliferation in ST-HSCs during repopulation, is not clear. Finally, as Busch et al. mention, if these results extend to humans, efforts to capture the potential of ST-HSCs for clinical transplants could be valuable.

These areas of uncertainty cannot be addressed without the development of experimental techniques to mark prospective ST- and LT-HSCs in vivo, to quantitatively analyse the resulting lineages and model the data statistically7. Only then will we be able to fully interpret this complex and dynamic process.Footnote 1


  1. 1.

    See all news & views


  1. 1

    Busch, K. et al. Nature 518, 542–546 (2015).

    ADS  CAS  Article  Google Scholar 

  2. 2

    McCulloch, E. A., Siminovitch, L. & Till, J. E. Science 144, 844–846 (1964).

    ADS  CAS  Article  Google Scholar 

  3. 3

    Qiao, W. et al. Mol. Syst. Biol. 10, 741 (2014).

    Article  Google Scholar 

  4. 4

    Sun, J. et al. Nature 514, 322–327 (2014).

    ADS  CAS  Article  Google Scholar 

  5. 5

    Harrison, D. E., Lerner, C., Hoppe, P. C., Carlson, G. A. & Alling, D. Blood 69, 773–777 (1987).

    CAS  PubMed  Google Scholar 

  6. 6

    Dykstra, B. et al. Cell Stem Cell 1, 218–229 (2007).

    CAS  Article  Google Scholar 

  7. 7

    Blanpain, C. & Simons, B. D. Nature Rev. Mol. Cell Biol. 14, 489–502 (2013).

    CAS  Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Peter W. Zandstra.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Goyal, S., Zandstra, P. Chasing blood. Nature 518, 488–490 (2015).

Download citation


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

Sign up for the Nature Briefing newsletter for a daily update on COVID-19 science.
Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing