In the case of axonal transport, it may not be about who wins the race, but what’s carried to the finish line. Retrograde axonal transport slows down in motor neurons in animal models of amyotrophic lateral sclerosis (ALS), depriving the cell body of neurotrophic factors. But the real problem, according to a paper in the August 5 Journal of Neuroscience, is that when the dynein-powered retrograde transporters finally get to the cell body, it is not those feel-good neurotrophins they deliver, but stress signals that cause apoptosis. First author Eran Perlson and principal investigator Erika Holzbaur, both of the University of Pennsylvania in Philadelphia, and colleagues suggest that the cargo itself is key to the rapid decline of ALS. In other ALS news, two recent papers in the Journal of Biological Chemistry may help to explain the mechanism of protein aggregation that underlies some forms of the disease.

Mice expressing mutant human superoxide dismutase 1 (SOD1), an enzyme sometimes mutated in people with familial ALS, are a common model for the condition. In these animals, retrograde transport is inhibited, and animals die well before reaching a year of age. Yet there are other animal models with sluggish retrograde transport, and they suffer milder neurodegeneration than mSOD1 animals. The Loa mouse, with a point mutation in dynein, and a mouse that overexpresses the dynein inhibitor dynamitin, both exhibit transport delays without such severe disease (see ARF related news story on Hafezparast et al., 2003 and ARF related news story on LaMonte et al., 2002). Perlson and colleagues sought to determine what makes disease in mSOD1 animals so much worse.

First, the researchers analyzed transport in mSOD1-G93A animals using sciatic nerve ligation assays. Retrograde travel of dynein subunits was reduced by approximately 30 percent compared to that in mice expressing wild-type SOD1. In previous studies, retrograde transport in Loa and dynamitin-overexpressing mice was reduced by a similar amount. Perlson and colleagues also used live imaging to monitor vesicle movement in embryonic rat motor neuron cultures. In cells expressing mSOD1-G85R, retrograde transport slowed to 0.41 microns per second, compared to 0.76 microns per second in cells expressing wild-type SOD1. Transport of specific neurotrophic cargoes, such as the nerve growth factor NRG, was also inhibited. The Loa and dynamitin mice showed similar deficiencies in growth factor trafficking.

The scientists reasoned that if the transport speed is not the difference, perhaps cargo is the issue. To assess what dynein was carrying, Perlson and colleagues performed dynein pulldowns on axoplasm-enriched fractions from 85-day-old mSOD1 and control non-transgenic mice. They used an antibody microarray to identify protein levels that differed between the two. Of the 102 proteins that linked less often with dynein in mSOD1 animals, 46 percent were associated with promoting cell survival. Conversely, among the 52 proteins that associated more frequently with dynein in the mSOD1 mice, 44 percent were associated with cellular stress. “Right away, the answer fell out of the analysis,” Holzbaur said. “Axonal transport is clearly going from a survival signaling pattern to a stress signaling pattern.”

Using directed pulldown assays and immunoblotting to confirm the results from the microarrays, the researchers determined that pro-survival molecules such as P-Trk, P-Erk1/2, and P-Erk5 were reduced by approximately 60 percent in sciatic nerve extracts from mSOD1 animals, compared to the levels in non-transgenic mice. Activated caspase-8, P-JNK, and p75-NTR cleavage products were increased in the mSOD1 dynein pulldowns, suggesting the motor protein toted these cell stress signals to the cell body. Neurodegeneration is likely a result of all of these changes, Perlson said: “There is not one magic pathway.”

If stress signals moving up the axon are the cause of cell damage, then inhibiting those signals should save the cell. To test this idea, the researchers used a compartmentalized cell culture system, with motor neurons growing in sections adjacent to glial cells. Axons could extend across the barriers. After six days in culture, half as many mSOD1-expressing cells survived as those expressing wild-type SOD1 or no transgene. The researchers then added specific inhibitors to JNK, caspase-8, and p75-NTR. Under these conditions, survival of mSOD1 cells rose to 86 percent of control levels. Overexpression of dynamitin, to slow transport, was not as effective protection as the inhibitors, suggesting that when transport was slowed, some stress signals still made it to the cell body. This supports the hypothesis that cell stress signaling, not slowed transport, is the real killer. Other experiments indicated that the stressful communiqués come from glia, as mSOD1 neurons cultured alone did not experience the uptick in P-Trk levels associated with cell stress and death. This fits with the popular theory that motor neuron death in ALS is non-cell-autonomous (reviewed in Boillée et al., 2006).

The work suggests that fixing the transport problem alone would not be a valid ALS treatment, nor would providing motor neurons with additional neurotrophic factors. A therapeutic would also have to block the stress signaling, Perlson suggested.

“It is a whole new hypothesis as to what is going on,” said Julie Andersen of the Buck Institute for Research in Aging in Novato, California. Yet the research says little about how axonal dynein comes by the additional stress factors. “To me, that is kind of the million-dollar question,” she said. The mechanism is key, agreed Scott Brady of the University of Illinois in Chicago. “What remains unclear is how mutant SOD1 affects the interaction of dynein with such a wide variety of transported cargoes,” he wrote in an e-mail to ARF. “Further, we need to understand what mutant proteins like SOD1 do to affect motor function.” Neither Andersen nor Brady was involved in the current study.

What mSOD1 does is a question many are addressing with biochemistry and structural biology studies, but the puzzle is compounded by the fact that more than 100 different SOD1 mutations can cause ALS. “We’re all looking for the one gain of function, the one thing that all the mutants have in common,” said Jeffrey Agar of Brandeis University in Waltham, Massachusetts. Agar and colleagues published a paper in the Journal of Biological Chemistry July 27 suggesting that a freely flapping electrostatic loop could be that common factor. The loosened structure could make the protein more likely to form intermolecular bonds leading to dangerous aggregates. Agar, first author Kathleen Molnar of the ExSAR Corporation in Monmouth Junction, New Jersey, and colleagues analyzed 13 different SOD1 mutants. “We were not expecting this when we started, believe me, but in all of them that electrostatic loop is floppier or more dynamic,” Agar said.

Another paper in the Journal of Biological Chemistry August 3 confirms the importance of SOD1’s metal cofactors in maintaining the protein’s stability. Principal investigator Lawrence Hayward of the University of Massachusetts Medical School in Worcester and first author Ashutosh Tiwari, who recently moved from Hayward’s lab to start his own at the Michigan Technological University in Houghton, led this study and were also coauthors on Agar’s July 27 paper. By analyzing the exposure of hydrophobic residues in different forms of the enzyme, the authors found that both zinc and copper ions were necessary to fully stabilize SOD1, making it less aggregation-prone.—Amber Dance


  1. I think the exciting finding is that while axonal transport is disrupted in several models of motor neuron disease (MND)/amyotrophic lateral sclerosis, this is not the underlying cause of the neurodegeneration. Rather, there is a switch in what cargo is carried. Unfortunately, the death signaling molecules manage to get through, but not those involved in signaling survival. This is a major conceptual advance for the MND field, but it may well have implications for Alzheimer disease (AD) where axonal transport is also disrupted with the development of neurofibrillary tangles. It will be very interesting to see whether there is a similar switch in which molecules get trafficked down axons in AD, and whether this is an explanation for how the pathology (plaques and tangles) result in neurodegeneration.

    View all comments by Elizabeth J. Coulson
  2. The paper by Perlson et al. brings to our attention the important issue of the contribution of axonal transport to neuronal pathology, by pointing to a neglected form of axonal transport deficiency. With a mouse model for amyotrophic lateral sclerosis (ALS), Erika Holzbaur’s team shows that, although the axonal transport in these mice is globally perturbed, it is the change in the spectrum of proteins that are being transported that actually inflicts the neuronal pathology.

    The axonal transport powered by kinesins and cytoplasmic dynein ensures the accurate delivery of the functional proteins, including those with a role in signaling, to their specific sites of action. Although the axonal transport in neurons occurs without interruption, the composition of the transported cargo vesicles varies in time according to the changes in the environment and physiological needs. While the overall transport does not normally change (with the transport machinery functioning normally all the time), the delivery of individual signaling proteins, unlike that of housekeeping proteins, fluctuates continuously. Thus, in cases of injury, or some other form of stress, the spectrum of proteins transported in and out of the neurites could change dramatically.

    The paper by Perlson et al. shows that this likely is the case with mouse models of ALS—and probably the human disease, too, where the changes in global transport may be less relevant than the changes in the delivered signals. Although anticipated, this result is hugely important, because it reminds us that the deficiencies in axonal transport of specific proteins, even in the absence of global perturbation of transport, could cause neuronal pathology and neurodegeneration.

    We would like to point at two important things. First, such subtle changes in the transport of a few proteins, on a normal background, may be difficult to detect. Second, this form of modification in axonal transport is not really a deficiency in transport (although it is carried out via transport). Rather, it is a problem of change in protein expression, protein degradation, or—most importantly—protein sorting. Studies will now have to focus more on such aspects, especially on the two major sorting stations for anterograde and retrograde axonal (and dendritic) transport, the trans-Golgi network and the recycling endosome. Thus, analysis of axonal transport in neurodegenerative diseases should also take into consideration changes that occur prior to the transport, at the sorting stations.

    With regard to Alzheimer disease, many studies have focused on global perturbation of axonal transport (e.g., 1,2). However, even in this case, disease-related conditions could change the post-translational processing (e.g., phosphorylation, proteolytic cleavage) of relevant cargo proteins (e.g., the amyloid-β precursor protein). As previously shown, such post-translational modifications could affect sorting into transport vesicles (3), as well as motor recruitment (4). As a consequence, the pattern of APP transport will change. As Erika Holzbaur’s group now clearly shows, axonal transport is a sum of multiple cellular processes that affect the transported proteins in various degrees, under different circumstances.


    . Disruption of fast axonal transport is a pathogenic mechanism for intraneuronal amyloid beta. Proc Natl Acad Sci U S A. 2009 Apr 7;106(14):5907-12. PubMed.

    . Axonopathy and transport deficits early in the pathogenesis of Alzheimer's disease. Science. 2005 Feb 25;307(5713):1282-8. PubMed.

    . The cleavage products of amyloid-beta precursor protein are sorted to distinct carrier vesicles that are independently transported within neurites. J Neurosci. 2009 Mar 18;29(11):3565-78. PubMed.

    . Coordinated transport of phosphorylated amyloid-beta precursor protein and c-Jun NH2-terminal kinase-interacting protein-1. J Cell Biol. 2005 Nov 21;171(4):615-25. PubMed.

    View all comments by Virgil Muresan
  3. Neurons are unique in that they are highly polarized cells with long projections. Motor neurons have axons that extend from the spinal cord out to the periphery to synapse with muscles; in the case of humans, these axons may extend for over a meter away from the cell body. Active transportation of proteins and organelles along the axon, in both directions between the cell body and the neuron synapse is essential for neuronal survival and communication. Anterograde axonal transport, from cell body to synapse, is undertaken by kinesins and other motor proteins. Retrograde axonal transport, from synapse to the cell body, is driven by the dynein motor within the dynein-dynactin complex.

    Defects in axonal transport have been shown to be present in mouse models of several neurodegenerative diseases, including Huntington disease, Alzheimer disease, and amyotrophic lateral sclerosis (ALS), and pathological findings such as axonal swellings that may be indicative of axonal transport defects have been found in patients with ALS. While mutations in the components of the motor complexes involved in axonal transport can cause neuropathies in humans—for example, defects in p150 dynactin are causal for a form of distal hereditary neuropathy—it has been unclear how much transport deficits contribute to ALS, and in particular we do not know about the effects of disruption of transport of specific cargos in ALS.

    In this landmark paper, Eran Perlson, Erika Holzbaur, and colleagues have assessed retrograde transport and cargoes, and looked at interactions between neurons and glial cells, in mouse models with motor neuron degeneration. The SOD1G93A transgenic mouse carries a human transgene array with a defect in the SOD1 gene that is known to be causative for ALS in humans; this mouse model of human ALS has an aggressive disease profile leading to paralysis and death usually by about three months of age. The Dync1h1Loa (Loa/+) mouse has a defect in the dynein heavy chain, which results in a mild progressive loss of motor neurons (and fairly profound defects in proprioceptive neurons), these mice have a normal lifespan. The Tgdynamitin mice overexpress dynamitin, which results in disruption of the dynein-dynactin complex and slowly progressive motor neuron death.

    Perlson and colleagues have shown that the retrograde transport inhibition in SOD1G93A mice is similar to that observed in both Loa/+ and Tgdynamitin mice, although the dynein motor complex appears to function normally in the ALS model, unlike the latter two mouse strains. In perhaps the most important finding of this exciting paper, the authors have also shown that the nature of the neurotrophin cargoes being transported in SOD1G93A mice is different from those found in wild-type animals. This difference could account for the aggressive motor neuron death seen in the ALS model and the relatively benign deficit seen in Loa/+ mice.

    Perlson, Holzbaur, and colleagues show wild-type mice have retrograde transport of neurotrophic survival factors within motor neurons, whereas in SOD1G93A mice stress/death factors are retrogradely transported in these cells. Importantly, this transport of stress/death factors is already occurring pre-symptomatically (50 days of age) in this ALS model. Another important finding of this paper is that non-neuronal cells such as glia are likely to help mediate this change in signaling. Furthermore, the authors have shown that inhibition of the retrograde stress signaling in motor neuron cultures from SOD1G93A mice was sufficient to inhibit/slow neuronal cell death.

    The results presented in this paper do not identify the primary cause of motor neuron death arising from mutant SOD1. Many changes in cellular homeostasis are observed in SOD1G93A mice, including protein misfolding, aberrant RNA processing, mitochondrial dysfunction, and oxidative stress. Any or many of these changes could contribute to the difference in retrograde transport and signaling in mutant SOD1 mice compared to wild-type animals. Further investigation is required to determine when and what causes retrograde signaling to switch primarily from survival to stress/death signaling, as this may help elucidate what aberrant cellular process(es) kill motor neurons. In addition, work with mouse models always needs to be validated in the human condition, as far as possible. Other interesting questions remain, including whether this profound change in intracellular signaling is specific to SOD1 ALS or is also present in other genetic forms of ALS including that caused by TDP43 and FUS mutations.

    However, this is an extremely important paper for research into motor neuron death, and potentially other neurodegenerative disease, because it highlights the importance of the retrograde signaling cargoes and thus potentially provides a target for therapy. In addition, this paper provides further support to the non-cell-autonomous theory of ALS, whereby changes within motor neurons alone have the ability to cause cell death, but this death is accelerated by interactions with surrounding non-neuronal cells which are also experiencing cellular stress.

    View all comments by Anny Devoy

Make a Comment

To make a comment you must login or register.


News Citations

  1. Role of the Motor in Motor Neuron Diseases
  2. Dynamitin in Motor Neurons: Dynamite for ALS Research?

Paper Citations

  1. . Mutations in dynein link motor neuron degeneration to defects in retrograde transport. Science. 2003 May 2;300(5620):808-12. PubMed.
  2. . Disruption of dynein/dynactin inhibits axonal transport in motor neurons causing late-onset progressive degeneration. Neuron. 2002 May 30;34(5):715-27. PubMed.
  3. . ALS: a disease of motor neurons and their nonneuronal neighbors. Neuron. 2006 Oct 5;52(1):39-59. PubMed.

Further Reading


  1. . Role of axonal transport in neurodegenerative diseases. Annu Rev Neurosci. 2008;31:151-73. PubMed.
  2. . Metal-free superoxide dismutase forms soluble oligomers under physiological conditions: a possible general mechanism for familial ALS. Proc Natl Acad Sci U S A. 2007 Jul 3;104(27):11263-7. PubMed.
  3. . Posttranslational modifications in Cu,Zn-superoxide dismutase and mutations associated with amyotrophic lateral sclerosis. Antioxid Redox Signal. 2006 May-Jun;8(5-6):847-67. PubMed.
  4. . Astrocytes as determinants of disease progression in inherited amyotrophic lateral sclerosis. Nat Neurosci. 2008 Mar;11(3):251-3. PubMed.
  5. . Onset and progression in inherited ALS determined by motor neurons and microglia. Science. 2006 Jun 2;312(5778):1389-92. PubMed.
  6. . Do disorders of movement cause movement disorders and dementia?. Neuron. 2003 Oct 9;40(2):415-25. PubMed.

Primary Papers

  1. . A switch in retrograde signaling from survival to stress in rapid-onset neurodegeneration. J Neurosci. 2009 Aug 5;29(31):9903-17. PubMed.
  2. . Metal deficiency increases aberrant hydrophobicity of mutant superoxide dismutases that cause amyotrophic lateral sclerosis. J Biol Chem. 2009 Aug 3; PubMed.
  3. . A common property of amyotrophic lateral sclerosis-associated variants: destabilization of the copper/zinc superoxide dismutase electrostatic loop. J Biol Chem. 2009 Nov 6;284(45):30965-73. PubMed.