News & Views | Published:


Stem cells on a stress-busting diet

Nature volume 494, pages 317318 (21 February 2013) | Download Citation

Knowing how an organism's tissues handle stress throughout life is key to understanding ageing and disease. Stems cells of the blood system seem to tackle metabolic stress by means of a process called autophagy. See Article p.323

Stem cells in adult tissues function to replace lost, damaged or diseased cells throughout an organism's life, thereby helping to maintain tissue health. But what protects the rare, long-lived stem cells from a lifetime of exposure to cellular and environmental stressors such as inflammation, radiation and metabolic alterations? On page 323 of this issue, Warr et al.1 propose that autophagy, a process of cellular self-cannibalization, is one mechanism that haematopoietic stem cells of the blood system use to protect themselves during times of metabolic stress, when nutrients are limited*.

When cells are starved of nutrients, the tissue's stem cells must choose whether to live or die. The options are committing suicide by apoptotic cell death, or self-preservation through autophagy, whereby cells recycle damaged or dispensable proteins and organelles into basic components to support cellular growth.

Autophagy is thought to be a major factor in ageing2. Loss of autophagy in tissues such as the brain, liver and heart leads to an increase in age-related disorders, including neurodegeneration, metabolic syndromes and cardiac dysfunction. Conversely, factors that stimulate autophagy abrogate these problems and have been linked to greater longevity. It has therefore been hypothesized that reduced autophagy contributes to the diminishing stem-cell function that occurs with age. No study, however, has investigated the direct role of autophagy in adult stem-cell function.

Warr et al. explore this question in both young mouse haematopoietic stem cells (HSCs) and in more-differentiated HSC progeny, including progenitor cells of the immune cells granulocytes and macrophages. The authors find that little or no autophagy occurs in freshly isolated young HSCs, but that this process can be rapidly induced when the cells are exposed to metabolic stress both in vitro and in vivo. Moreover, when autophagy is inhibited during such metabolic stress, young HSCs rapidly die through apoptosis, indicating that autophagy is crucial for their survival. By contrast, granulocyte–macrophage progenitor cells show higher baseline levels of autophagy, but no shift under starvation conditions.

Autophagy can be stimulated in several ways2, including through inhibition of the signalling molecule mTOR and activation of stress-induced transcription factors such as FoxO3 and p53. Warr and colleagues find that the primary driver in HSCs is FoxO3, with little contribution from mTOR or p53.

FoxO3 is a member of the FoxO family of transcription factors, which are involved in diverse processes, including stress resistance, apoptosis and metabolism. Loss of a single type of FoxO protein has been found to have little effect on HSCs, but simultaneous loss of three of them (FoxO1, FoxO3 and FoxO4) leads to dramatic changes in HSC proliferation and in levels of reactive oxygen species3. Warr et al. propose that a similar redundancy may explain the incomplete loss of autophagy seen when only FoxO3 is missing. It is also possible that sirtuin proteins act upstream of FoxO3, as they are known to promote autophagy and increase longevity — in part by activating FoxO transcription factors — and to affect the proliferation of aged HSCs and their response to metabolic stress2,4,5.

The implication of autophagy in ageing necessitates a better understanding of whether old stem cells use autophagy for similar purposes to their younger counterparts. Such knowledge should provide insight into the mechanisms underlying ageing and disease, and lead to improved strategies for enhancing health and longevity. Warr and co-authors show that, unlike young HSCs, freshly isolated old HSCs show some autophagy. Also, old cells could mount a similar response to their younger counterparts under starvation conditions. These findings contradict the currently held belief that autophagy diminishes with age.

Nonetheless, abrogated autophagy had different effects on the function of old and young HSCs. The authors report that blocking autophagy diminished the colony-forming potential of old HSCs to a much greater extent than in young cells. Colony formation is an in vitro metric used to determine the capacity of blood cells to form differentiated cell types. So although old and young cells show the same levels of autophagy, old cells seem to be more dependent on autophagy for their functioning. Confirmation of this would be to show that loss of autophagy more severely affects the capacity of old than young HSCs to repopulate the blood system of an individual whose immune system has been destroyed by radiation — the gold standard for assessing the function of HSCs.

The differences in autophagy between old and young HSCs did not seem to be due to changes in the activities of FoxO3, p53 or mTOR, and the transcriptional program triggered by FoxO3 was still active in old cells. Surprisingly, however, the old cells showed defects in the uptake of glucose, and so were in a state of metabolic stress even under normal conditions (Fig. 1). Feeding old cells a nutrient source that bypassed the need for glucose uptake restored the cells' colony-forming ability even when autophagy was blocked.

Figure 1: Stem cells, nutrient uptake and autophagy.
Figure 1

a, Warr et al.1 report that old haematopoietic stem cells (HSCs) cannot take up glucose normally and compensate for the resulting metabolic stress by using the process of autophagy to break down their own components as a source of nutrients. Unsurprisingly, therefore, blocking autophagy in these cells impairs their function. The surrounding environment of old HSCs contains low levels of cytokines, which are known to mediate glucose uptake by binding to cytokine receptors on the surface of the cells. b, Young HSCs, by contrast, can take up glucose normally, perhaps because cytokine levels in their environment are high. These cells can thus support their own metabolic needs, and so blocking autophagy under non-stress conditions does not notably affect their function.

A study in mice recently indicated6 that deregulation of cytokine proteins might be one explanation for the altered nutrient uptake in aged HSCs, as cytokines help to promote glucose uptake from the cellular environment. If such metabolic stress also occurs in humans, correcting it could help to improve stem-cell function in elderly people. Alternatively, experiencing prolonged autophagy may affect metabolism in old HSCs, as autophagy mediates the clearance of damaged mitochondria (organelles that act as cellular powerhouses), potentially resulting in lower energy output and slowing with age of essential cellular processes.

An intriguing question is whether the age-associated increase in autophagy is specific to HSCs or also occurs in stem cells of other tissues. So far, most of the other tissues studied for traits associated with age-related autophagy, such as the brain and heart, are those with a slow turnover of cells2. It could be that stem cells in highly proliferative tissues such as the blood have a different dependence on autophagy. For a definitive answer to this, researchers should examine stem cells in other tissues with a high cell turnover, such as the skin and gut.


  1. 1.

    *This article and the paper under discussion1 were published online on 6 February 2013.


  1. 1.

    et al. Nature 494, 323–327 (2013).

  2. 2.

    , & Cell 146, 682–695 (2011).

  3. 3.

    et al. Cell 128, 325–339 (2007).

  4. 4.

    , & Mol. Cell 40, 280–293 (2010).

  5. 5.

    et al. Blood 117, 440–450 (2011).

  6. 6.

    , & Blood 119, 2500–2509 (2012).

Download references

Author information


  1. Teresa V. Bowman and Leonard I. Zon are in the Stem Cell Program, Boston Children's Hospital, Boston, Massachusetts 02115, USA.

    • Teresa V. Bowman
    •  & Leonard I. Zon


  1. Search for Teresa V. Bowman in:

  2. Search for Leonard I. Zon in:

Corresponding authors

Correspondence to Teresa V. Bowman or Leonard I. Zon.

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.

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