Cut the forelimb off an axolotl and it will grow another one, complete with bone, muscle, nerve and skin, in roughly six to eight weeks. That much has been known for well over a century. What the past decade of work has clarified is stranger, and it has little to do with healing as we usually picture it. The salamander does not patch the wound. It reaches back into the genetic instructions it used to build the limb as an embryo, reopens part of that programme in adult cells, and runs it again.
What the amputation actually triggers
Within hours of an amputation, skin cells migrate across the cut to form a wound epidermis, a signalling layer rather than a scar. Beneath it, cells from the stump gather into a mound of progenitor cells called the blastema. This is the structure that does the rebuilding, and most of it comes from the limb’s connective tissue: the fibroblasts, cartilage and related cells that make up the bulk of the stump.
What matters is what those cells do to get there. Rather than a reserve of dormant stem cells waking up, mature adult cells appear to step backwards, shedding their specialised roles and returning to a less committed state. Biologists call this dedifferentiation.
What the 2018 single-cell work showed
The clearest picture of this comes from a 2018 paper in Science by Tobias Gerber, Elly Tanaka and colleagues, a collaboration between the Max Planck Institute for Evolutionary Anthropology and the Research Institute of Molecular Pathology in Vienna, Single-cell analysis uncovers convergence of cell identities during axolotl limb regeneration. Using genetic labelling and single-cell RNA sequencing, they tracked connective-tissue cells through regeneration one cell at a time. The varied, mature cell types of the adult limb converged on a single, relatively uniform progenitor state. That state, in the authors’ reading, resembled an embryonic limb bud: the same tissue that forms a limb the first time around, in the developing embryo.
They found no pre-existing pool of blastema-seeding stem cells sitting in the adult limb. The progenitor population is assembled on demand, out of ordinary adult cells, and the limb’s full variety is then rebuilt from that common starting point.
The blueprint is not stored anywhere in the adult limb. It gets re-derived on the spot.
Why it is not a simple replay of the embryo
The obvious question is whether the stump just reverts to being an embryo. It does not, and the reason is the more interesting part of the story. The cells reopen an embryonic-like programme, but they do not forget where they are.
Earlier work from the same lab, published in Nature in 2009 by Martin Kragl and colleagues, found that blastema cells keep a memory of their tissue origin: cartilage cells tend to make cartilage, skin cells make skin. The limb also remembers its position. A blastema formed at the wrist rebuilds a hand, not an entire arm. That positional sense is governed in part by retinoic acid, a vitamin A derivative that forms a gradient along the limb and effectively tells cells how much of the structure to replace. A 2025 paper in Nature Communications, Retinoic acid breakdown is required for proximodistal positional identity during axolotl limb regeneration, traced how precisely that signal has to be degraded for the right segment to form. A separate 2025 paper in Nature, again from Tanaka’s group, Molecular basis of positional memory in limb regeneration, described the genetic circuit that holds a cell’s front-to-back identity in place.
So it is a controlled reopening, not a reset. Adult cells regain access to the embryonic construction programme while keeping an adult sense of place. That pairing is what lets the animal rebuild exactly what was lost and stop at the right point.
The part that gets oversold
Every few months a headline promises that this line of work will soon let humans regrow limbs. The gap is wide. Mammals, including humans, carry many of the same genes; what we appear to lack is the ability to marshal them into a blastema rather than a scar. Whether that gap can be closed is an open research question, not an imminent therapy. The axolotl genome, sequenced in 2018, is enormous, roughly ten times the size of the human genome, which has made the molecular work correspondingly slow. Individual genes identified in salamanders may inform wound-healing research over time. A salamander limb regrowing on a person is not on any credible near-term timeline.
What is still open
Much of the detail remains unsettled. Researchers are still working out how a cell decides to dedifferentiate in the first place, how the blastema measures how much limb to make, and why mammalian tissue forms a scar where salamander tissue forms a blastema. The 2018 single-cell study is one detailed line of evidence, built on one animal’s connective tissue, and not a finished account of regeneration in general.
What is reasonably clear is the shape of the trick. An axolotl keeps no spare limb in reserve, and it does not heal the way we do. It takes ordinary adult cells, walks them back toward the state they occupied before the limb first formed, and builds again from there, guided the whole way by a memory of where each part belongs.
