Developmental biology

A cellular view of regeneration

How the salamander regrows an entire limb after injury has flummoxed the wisest of scientists. A closer look at the cells involved in limb regeneration shows that remembering past origins may be crucial for this feat.

When a salamander loses an appendage, such as a limb, a remarkable series of events unfolds: a clump of cells forms at the site of the injury, and this deceptively simple structure, known as a blastema, regenerates the missing body parts. Skin, muscle, bone, blood vessels and neurons all arise from this collection of nondescript cells through patterning and self-assembly. So complete is the repair that it is difficult — if not impossible — to tell that the animal has been injured. It has long been believed that the cells responsible for this repair process are pluripotent, capable of giving rise to all tissue types. On page 60 of this issue, Kragl et al.1 describe a novel cell-tracking technique that reveals some surprising findings about the origins and fates of cells involved in salamander limb regeneration.

Far from being a mere curiosity confined to a few species, regeneration is a phenomenon that is broadly but unevenly distributed among the animal and plant kingdoms. Body-part regeneration is found in most animal phyla thus far examined, indicating that this biological attribute may actually be an ancient evolutionary invention2. Yet how new body parts arise from old ones has been a riddle within a mystery dating back to antiquity. Regeneration in animals and plants has astounded successive generations of great thinkers, including Aristotle, Lazzaro Spallanzani, Voltaire, Charles Darwin and the biologist Thomas Hunt Morgan.

Persistent efforts by current researchers in the field3,4,5,6,7 have begun to reveal crucial insights into this biological phenomenon. Nevertheless, much remains to be understood. Key to elucidating the mechanism of regeneration is defining the origins, lineages and fates of the cells that mount a regenerative response after injury. Because the cells that collect in the blastema look identical, it has long been thought that they have dedifferentiated from tissues near to the plane of amputation into a single population of pluripotent cells5. However, it is quite possible that blastema cells involved in regeneration have instead entered a migratory or proliferative state without changing their tissue identity or their lineage potential. Without robust methods to track the fate of cells from the time of amputation to their differentiation during regeneration, definitive conclusions about the nature of the regenerative cells are difficult to reach.

Kragl et al.1 investigate limb regeneration in the axolotl (Ambystoma mexicanum), a salamander endemic to Mexico and a favourite model system for studying vertebrate development. The authors use transgenic technology to introduce DNA coding for a green fluorescent protein molecule into the genome of axolotl cells. Once marked out by its permanent, glowing, green fluorescent 'tattoo', a specific tissue type (for instance, muscle or skin) can be transplanted into a host animal and followed in time and space after the host tissues have been amputated (Fig. 1).

Figure 1: Cell-tracking techniques used by Kragl and colleagues1.

Tissue from axolotls expressing a transgene encoding the green fluorescent protein (GFP) molecule is transplanted into juvenile non-transgenic axolotls (upper panel). The transplanted tissue is included in the amputation site, and the fluorescently labelled cells are followed during limb regeneration. Alternatively, specific embryonic tissue is transplanted from GFP-labelled axolotl embryos into non-transgenic embryos (lower panel), which mature into juvenile axolotls with a tissue-specific fluorescent marker. These glowing cells can also be tracked after amputation.

Kragl et al. find that the cells that regenerate the amputated body parts in the axolotl are not pluripotent. Instead, they respect their developmental origins and restrict their differentiation potential accordingly (Fig. 2). For instance, labelled muscle cells at the site of amputation differentiate only into muscle and do not differentiate into other tissue types in the regenerating limb. The authors observed similar lineage restriction for epidermal cells, cartilage cells and Schwann cells (a type of cell that ensheaths nerve axons). Dermal cells are the sole exception — they contribute cells to both the dermis and the skeleton of the regenerating limb. These data strongly suggest that cells tasked with regeneration retain a memory of their previous identity and, in some cases, of their position in the limb.

Figure 2: Lineages are respected during axolotl limb regeneration.

Kragl et al.1 find that, with the exception of cells in the dermis, the dedifferentiated cells contributing to limb regeneration largely remain lineage restricted. Because the experiments were carried out in juveniles, in which the bones have not fully ossified, the fate of bone cells after amputation is unknown. Similarly, owing to the paucity of markers, the fate of connective-tissue fibroblasts contributing to limb regeneration remains undefined. The blue arrow indicates that cartilage cells can sometimes contribute to the dermis.

The use of transgenic methodologies to map cell behaviour and population dynamics during vertebrate regeneration1 is without doubt a significant technical accomplishment that adds a new dimension to our understanding of regeneration. Like all important work, it also leaves us with questions that will probably occupy researchers for the next few years. One question concerns the contribution of connective-tissue fibroblast cells to the regenerating limb. Fibroblasts are relatively abundant in the limb and are known to contribute significantly to regenerative activities in salamanders8. Are these cells lineage restricted, or can they give rise to other types of tissue? Because there are no reliable markers for identifying connective-tissue fibroblasts yet, Kragl and colleagues could not definitively answer this particularly interesting question.

Is lineage restriction in regenerating cells unique to the axolotl, or can these principles be applied to other commonly studied species of salamander? The animals used for these studies were juveniles, with a skeletal system made up of cartilage rather than bone. Most other studies in salamanders have been done in adult newts, which have ossified skeletons. Also, there are well-documented differences between the regenerative properties of newts and axolotls. For instance, whereas newts can regenerate the lens of the eye after its removal, the axolotl cannot9. These and other issues will have to be resolved before the biological significance of Kragl and colleagues' findings1 can be fully appreciated.


  1. 1

    Kragl, M. et al. Nature 460, 60–65 (2009).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Birnbaum, K. D. & Sánchez Alvarado, A. Cell 132, 697–710 (2008).

    CAS  Article  Google Scholar 

  3. 3

    Grogg, M. W. et al. Nature 438, 858–862 (2005).

    ADS  CAS  Article  Google Scholar 

  4. 4

    Gurley, K. A., Rink, J. C. & Sánchez Alvarado, A. Science 319, 323–327 (2008).

    ADS  CAS  Article  Google Scholar 

  5. 5

    Kumar, A., Godwin, J. W., Gates, P. B., Garza-Garcia, A. A. & Brockes, J. P. Science 318, 772–777 (2007).

    ADS  CAS  Article  Google Scholar 

  6. 6

    Morrison, J. I., Lööf, S., He, P. & Simon, A. J. Cell Biol. 172, 433–440 (2006).

    CAS  Article  Google Scholar 

  7. 7

    Reddien, P. W. et al. Dev. Cell 8, 635–649 (2005).

    CAS  Article  Google Scholar 

  8. 8

    Bryant, S. V., Endo, T. & Gardiner, D. M. Int. J. Dev. Biol. 46, 887–896 (2002).

    PubMed  Google Scholar 

  9. 9

    Tsonis, P. A., Madhavan, M., Tancous, E. E. & Del Rio-Tsonis, K. Int. J. Dev. Biol. 48, 975–980 (2004).

    Article  Google Scholar 

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Sánchez Alvarado, A. A cellular view of regeneration. Nature 460, 39–40 (2009).

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.