The nervous system keeps a tidy house. When axons get cut off from their cell bodies, they break down within days, their pieces swept away by the immune system in a process known as Wallerian degeneration. Now in the June 7 Science, researchers led by Marc Freeman at the University of Massachusetts Medical School, Worcester, report that it is possible to interrupt this cleanup by deleting a single gene. In both fly and mouse models lacking the gene, severed fibers stay intact for weeks. The findings make the case that axon death is an active process initiated by a molecular program, and not merely a side effect of injury, the authors note. The data also imply that this pathway could be a promising therapeutic target for disorders of the central and peripheral nervous systems. "For neurodegenerative disease in general, we think it is really exciting, because all these diseases exhibit profound axonal and synaptic loss," Freeman told Alzforum.

Jonathan Glass at Emory University, Atlanta, Georgia, expressed enthusiasm for the paper, noting, "Marc Freeman's group has basically found the gene responsible for Wallerian degeneration. It’s extraordinary work."

Scientists once thought that injured axons wasted away passively as a result of lack of nutrients from the cell body, but this view was challenged by the 1989 discovery of a mutant mouse (WldS) in which severed axons persisted for weeks (see Lunn et al., 1989 and Perry et al., 1990). The WldS gene, however, is a gain-of-function mutation that creates a chimeric fusion protein and may not relate to normal physiology. Its discovery implied, but could not prove, that an active axon death pathway exists.

To hunt for molecules that might be directly involved in axon death, first author Jeannette Osterloh performed a forward genetic screen in fruit flies, inducing mutations and then looking for lines in which nerve fibers remained intact one week after axotomy. Osterloh and colleagues found three lines in which severed axons survived up to 50 days, in other words, the fly’s entire lifespan. Synapses on the isolated nerves also persisted as long. All three lines turned out to arise from loss-of-function mutations in the same gene, Drosophila sterile alpha and armadillo motif (dSarm; also known as ect4). The authors confirmed the gene’s identity by expressing dSarm in the mutant flies, which restored Wallerian degeneration.

To see if the mouse ortholog, Sarm1 (also called MyD88-5), plays a similar role, the authors used knockout mice made by co-author Aihao Ding of Weill Medical College at Cornell University, New York City (see Kim et al., 2007). In ganglion cell cultures made from Sarm1 null animals, severed axons stayed intact about ten-fold longer than did axons in wild-type cultures. Axons enjoyed similar protection in vivo, with most sciatic nerve fibers staying whole two weeks after lesioning, compared to less than three days in wild-type animals. Neuromuscular junctions were preserved in the knockouts, and the macrophages that normally clean up damaged nerves stayed away. Significantly, the authors showed that dSarm/Sarm1 is not involved in apoptotic cell death or developmental axon pruning. "It’s a unique molecular program," Freeman said.

The dSarm/Sarm1 gene is something of a mystery. It seems to be expressed primarily in the nervous system in both flies and mice, where it is found in neuron cell bodies and neurites. It encodes an adaptor protein, a class of molecule that often regulates signal transduction pathways by linking other proteins together. This implies that dSarm plays a role in signal transduction, but the pathway is unknown. The protein also contains a Toll/Interleukin receptor 1 homology (TIR) domain, suggesting it can bind Toll receptors, but Freeman told Alzforum that dSarm does not seem to play a major role in immunity, as other Toll-binding proteins do. Instead, the molecule may act as a scaffold, bringing kinases together to activate a signaling cascade, Freeman suggested. "What those kinases are, we don’t know," he added. He is doing enhancer/suppressor screens in flies to find molecules in the pathway.

Preliminary evidence suggests other genes are involved in axon death, as well, Freeman said. In the worm, dSarm’s ortholog TIR-1 regulates odorant receptor expression in olfactory neurons (see Chuang and Bargmann, 2005), but Freeman’s group found no evidence that dSarm/Sarm1 signals in the same way. However, he finds it tantalizing that TIR-1 acts downstream of calcium channels and CaM kinase II, because calcium entry triggers Wallerian degeneration in severed axons (see George et al., 1995. The human ortholog of the mouse gene is also called Sarm1 and its function is unclear.

Could dSarm/Sarm1 protect nerve fibers from destruction in neurodegenerative diseases? Axons wither in many peripheral nervous system disorders, such as amyotrophic lateral sclerosis (ALS), Charcot-Marie-Tooth Disease, and diabetic neuropathy, as well as in central nervous system disorders like multiple sclerosis, stroke, Alzheimer’s, Parkinson’s, Huntington’s and frontotemporal diseases. This question can be addressed by crossing the Sarm1 knockout mouse with mouse models of the given neurodegenerative disease that feature neurodegeneration, and checking if axons are protected. Freeman is collaborating with Robert Brown at UMass to test an ALS mouse model, and with Neil Aronin at UMass to look at a Huntington’s model. He is interested in other collaborations, as well.

In previous ALS work, researchers have found several ways to keep motor cell bodies alive, such as expressing the anti-apoptotic protein Bcl-2 (see Sato et al., 2002), or the neurotrophic factor GDNF (see Suzuki et al., 2007), but the animals died anyway, Glass noted. Perhaps this was because their peripheral nerves died. A strategy might be to combine one of these methods with Sarm1 knockout, to see if protecting both the cell body and the fiber could prolong survival, Glass suggested. Freeman agreed, noting that those experiments are underway in his lab. Even if this approach works, however, translating it to human disease would be challenging. One encouraging sign is that the protective Sarm1 mutation causes loss of function, which opens the door to knocking down Sarm1 expression using interfering RNA methods, Freeman pointed out. On the downside, Sarm1 may not be amenable to pharmacological inhibition, as it is an adaptor molecule, not an enzyme and therefore doesn’t have an active site for drug developers to target. "All the more reason to try to identify more molecules in the pathway," Freeman said.—Madolyn Bowman Rogers


  1. I believe that the discovery of dSarm/Sarm1 by Osterloh and colleagues as an endogenous regulator of Wallerian degeneration is a really exciting finding and a major step forward—this has really been what many of us in the field have been hoping for since Marc Freeman and others started studying Wallerian degeneration in flies a few years ago. This, finally, puts the notion of an endogenously regulated axon destruction program on firm ground. As is often the case for a newly implicated gene, we do not yet know how, or even where, Sarm1 acts to promote Wallerian degeneration—and from the previous pioneering efforts of Michael Coleman, Mark Freeman, and others on the original WLD-S fusion protein, we know this will not be a trivial task. But it will be a highly exciting one. This also means that it is currently difficult to gauge how important this—or related—axon destruction pathways might be in neurodegenerative diseases. Obviously, axons are a site of prime vulnerability in neurodegeneration. Still, as Wallerian degeneration is kicked off by an “initial” event, classically a transection, it is not straightforward to predict how preventing Wallerian degeneration will affect outcome in a given disease. A lot of this uncertainty stems from the fact that for most neurodegenerative conditions, we actually do not understand what the primary lesion to a neuron is. Indeed, it remains unclear why Wallerian degeneration is actually “there for,” and why it has been conserved. It will be exciting to see whether the newly identified dSarm/Sarm1 will be able to shed some light on such questions.

    View all comments by Thomas Misgeld

Make a Comment

To make a comment you must login or register.


Paper Citations

  1. . Absence of Wallerian Degeneration does not Hinder Regeneration in Peripheral Nerve. Eur J Neurosci. 1989;1(1):27-33. PubMed.
  2. . A Toll-interleukin 1 repeat protein at the synapse specifies asymmetric odorant receptor expression via ASK1 MAPKKK signaling. Genes Dev. 2005 Jan 15;19(2):270-81. PubMed.
  3. . Axotomy-induced axonal degeneration is mediated by calcium influx through ion-specific channels. J Neurosci. 1995 Oct;15(10):6445-52. PubMed.
  4. . Bcl-2 rescues motoneurons from early cell death in the cervical spinal cord of the chicken embryo. J Neurobiol. 2002 Nov 15;53(3):381-90. PubMed.
  5. . GDNF secreting human neural progenitor cells protect dying motor neurons, but not their projection to muscle, in a rat model of familial ALS. PLoS One. 2007;2(8):e689. PubMed.

External Citations

  1. Perry et al., 1990
  2. ect4
  3. Sarm1

Further Reading

Primary Papers

  1. . dSarm/Sarm1 is required for activation of an injury-induced axon death pathway. Science. 2012 Jul 27;337(6093):481-4. PubMed.