Cell biology

Alternative energy for neuronal motors

Article metrics

Neurons use molecular motors to power the transport of cargoes along their axonal extensions. Fresh evidence challenges the view that cellular organelles called mitochondria are the main energy providers for this process.

Intracellular transport requires energy, and nowhere is this process more energy-demanding than in the axon of a neuron. The length of an axon can range from millimetres in flies to metres in large mammals — a prodigious distance for the molecular motors that travel back and forth conveying cellular cargoes between the neuronal cell body and the nerve endings. These motors move along microtubules in discrete steps of a few nanometres each, and every step requires energy that the motors derive from the hydrolysis of ATP molecules. Consequently, axonal transport consumes large amounts of ATP, and organelles called mitochondria are widely assumed to be the principal source of this energy. But in a paper published in Cell, Zala et al.1 challenge this assumption, showing that a sugar-hydrolysing enzyme bound to the surface of transport vesicles provides the energy for this process. In other words, motor complexes in axons seem to use on-board 'generators' to ensure a steady energy supply en route to their destination.

Mitochondria regulate their mobility according to the intracellular levels of calcium ions that accumulate in the axon at specialized energy-requiring sites rich in ion channels2. The distance between adjacent clusters of mitochondria can be greater than known ranges of mitochondrial-ATP gradients3. Zala et al. therefore set out to identify the energy source for transport in axonal regions that lack mitochondria. Surprisingly, inhibition of mitochondrial ATP production did not significantly affect the axonal transport of vesicles carrying the growth factor BDNF, regardless of the apparent ATP levels in the axon.

Cells can generate energy independently of mitochondria by breaking down sugars using a process called glycolysis. Unexpectedly, Zala and colleagues found that chemical inhibition of the key glycolytic enzyme glyceraldehyde 3-phosphate dehydrogenase (GAPDH), or reduction in its expression, blocked the transport of axonal vesicles. GAPDH was present on the surface of these transport vesicles (Fig. 1), where it was bound by huntingtin — a motor-associated protein that is mutated in the neurodegenerative disorder Huntington's disease4. These findings have strong parallels with earlier work5 demonstrating a role for GAPDH, but not mitochondria, in the bioenergetics of the uptake of neurotransmitter molecules into synaptic vesicles at nerve endings. The two studies add localized energy provision to the growing list of diverse functions of GAPDH in different subcellular compartments6.

Figure 1: On-board energy generators.

Neurons have long axonal processes that require a constant supply of energy in the form of ATP molecules to transport cargo-carrying vesicles along microtubules from the cell body to axonal terminals (red arrow) and back (blue arrow). The prevalent assumption is that mitochondria (not shown) are the source of ATP for this purpose. Zala et al.1, however, find that the glycolytic enzyme GAPDH, which is bound to transport vesicles through the huntingtin protein and possibly other factors (question marks), powers the motor proteins cytoplasmic dynein and kinesin on microtubules by locally generating ATP through glycolysis.

Zala and co-authors' work might also have therapeutic implications. Their results indicate that mitochondrial impairment is unlikely to be the main cause of the alterations in fast axonal transport seen in neurodegenerative disorders such as Huntington's disease and amyotrophic lateral sclerosis (ALS). This could explain the failure of previous efforts7 to ameliorate disease in a mouse model of ALS by increasing mitochondrial mobility. Furthermore, the affinity of GAPDH for aggregating proteins found in various neurodegenerative disorders8 indicates that sequestration of this enzyme away from transport vesicles might contribute to disease development and progression. Nonetheless, therapeutic targeting of GAPDH will probably require the development of drugs that specifically modulate only a subset of its functions6, as indicated by the failure of a GAPDH-directed molecule in a clinical trial for treating ALS9.

The uncoupling of vesicular axonal transport from mitochondrial ATP production is the most intriguing yet puzzling aspect of this work. For instance, Zala et al. describe an alternative energy source for fast transport across axonal areas devoid of mitochondria, yet they did not detect any overt variation in ATP levels in the axons they analysed. Another mystery arises from the authors' observations that mitochondrial ATP plays no part in fast axonal transport, and that motors associated with vesicular cargoes cannot access this source of ATP. This finding cannot be explained by intrinsic properties of the motor complexes, because the same types of molecular motor can hydrolyse mitochondrial ATP for mitochondrial transport, which was not affected by inhibition or reduced expression of GAPDH1.

Similarly, a direct link between GAPDH and motor complexes is unlikely. The authors report that GAPDH association with transport vesicles is necessary for axonal transport, although its binding to molecular motors is not: in huntingtin-depleted cells, GAPDH targeting to vesicles by other means was sufficient to recover transport. A direct link between GAPDH and motor complexes is also unlikely because GAPDH-mediated axonal transport was shown to occur with different motors moving in different directions (Fig. 1). To resolve this conundrum, in vitro studies are required that examine the effects of GAPDH on the biophysical properties of isolated motor complexes moving along microtubules. Such studies might reveal additional functions for GAPDH in regulating these molecular nanomachines. For example, it could be that other enzymatic activities of GAPDH, such as S-nitrosylation and mono-ADP-ribosylation6,8, are also necessary for transport.

Whether GAPDH is required for the axonal transport of non-vesicular complexes, such as RNA granules, is also unknown. And does a similar GAPDH-dependent mechanism apply to other cellular ATP-hydrolysing enzymes, such as ion pumps and remodelling enzymes8? Whatever the answers, Zala and co-workers' findings place alternative energy sources in the axonal domain, calling for a re-examination of fundamental assumptions and pointing the field in intriguing and unexplored directions.


  1. 1

    Zala, D. et al. Cell 152, 479–491 (2013).

  2. 2

    MacAskill, A. F. & Kittler, J. T. Trends Cell Biol. 20, 102–112 (2010).

  3. 3

    Niethammer, P., Kueh, H. Y. & Mitchison, T. J. Curr. Biol. 18, 586–591 (2008).

  4. 4

    Caviston, J. P. & Holzbaur, E. L. Trends Cell Biol. 19, 147–155 (2009).

  5. 5

    Ikemoto, A., Bole, D. G. & Ueda, T. J. Biol. Chem. 278, 5929–5940 (2003).

  6. 6

    Tristan, C., Shahani, N., Sedlak, T. W. & Sawa, A. Cell Signal. 23, 317–323 (2011).

  7. 7

    Zhu, Y.-B. & Sheng, Z.-H. J. Biol. Chem. 286, 23432–23440 (2011).

  8. 8

    Seidler, N. W. Adv. Exp. Med. Biol. 985, 249–267 (2013).

  9. 9

    Miller, R. et al. Neurology 69, 776–784 (2007).

Download references

Author information

Correspondence to Giampietro Schiavo or Mike Fainzilber.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Schiavo, G., Fainzilber, M. Alternative energy for neuronal motors. Nature 495, 178–179 (2013) doi:10.1038/495178a

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.