This is the authors' abstract. We don't add key points for this paper.
Neurofilaments (NFs) are prominent constituents of mature axons. How the neuron establishes and maintains the NF array has been the subject of intense controversy1. NFs also represent the most highly phosphorylated proteins within the axonal cytoskeleton, yet how phosphorylated NF are segregated within axons, and how phosphorylation contributes to NF dynamics are also not entirely clear1. The recent demonstration that the microtubule motor protein kinesin participates in the intracellular distribution and axonal transport of NFs, and that phosphorylation regulates the association of NFs and kinesin2, 3, affords a new look at these unresolved aspects of NF dynamics. If one accepts that the anterograde motor kinesin either directly or indirectly mediates NF transport, then the inter-neuronal distribution of NFs may be viewed as by default arising from the inter-neuronal distribution of microtubules (MTs)4, 5. MTs within dendrites are non-uniformly oriented. That is, some MTs have their ‘+’ end facing away from the perikaryon, while others have it facing towards the perikaryon. Accordingly, the ‘+’-end-directed kinesin (and its cargo) would be equally likely to move into and out of dendrites. Conversely, axonal MTs are uniformly oriented with their ‘+’ ends toward the synapse/growth cone. Once entering an axon, kinesin (and its cargo) is committed to an anterograde procession along the length of the axon. The unique orientation of axonal MTs itself may therefore dictate how kinesin effects the accumulation of NFs within axons. The vast majority of NF subunits are polymerised within axons6. However, newly-expressed Triton-soluble subunits7, and monomeric or non-filamentous oligomeric subunits1, 8 also undergo axonal transport at rates equivalent to or exceeding those of cytoskeleton-associated NFs. Any proposed transport mechanism must encompass the ability to translocate subunits in multiple assembly states. This line of reasoning is substantiated by the intra-axonal association of kinesin with Triton-soluble as well as Triton-insoluble NF subunits3. Nevertheless, the vast majority of subunits assemble shortly after synthesis and remain as stable polymers once assembled6, 9. Therefore, at least following maturation, a NF motor protein that does not discriminate among assembly states, would translocate the vast majority of subunits in assembled form. Neurons may simply rely on the rapid kinetics of NF assembly to guarantee that the majority of subunits are transported as filaments, and need not invoke additional selection mechanisms to preclude transport of potentially inappropriate forms. The dynamics of establishing the NF array in growing axons may differ dramatically from those required for its maintenance following maturation. The observation of NF subunit immunoreactivity within axons in situ prior to the appearance of NFs themselves10 the initial predominance of non-filamentous oligomers within growing axons in culture, and the subsequent predominance of filaments2, suggests that axonal transport of non-filamentous subunits may facilitate establishment of the axonal cytoskeleton. Once fully established, continued transport of monomers, non-filamentous oligomers and short filaments themselves may provide for regional repair/remodelling and/or elongation of larger NFs, or instead may inadvertently be transported with no such physiological role. Hypotheses suggesting that monomers and/or small oligomers represent the major form of transported subunits seem intuitive in terms of energy conservation, since the neuron could replace portions of the cytoskeleton rather than continually translocate the entire axonal cytoskeleton11. However, each transport ‘unit’ by definition requires its own motor – monomers would require one motor per subunit, and translocation of small oligomers would require one motor for a relatively small number of subunits, a situation at odds with energy conservation. Further analyses will be required to resolve these considerations. Mitogen-activated protein kinase phosphorylates C-terminal side-arms of NFs12 and, in doing so, disrupts their association with kinesin3. Reversible phospho-dependent dissociation and re-association could account for the saltatory transport of individual NFs and oligomeric precursors2, 13: while the average transport rate of individual NFs falls within slow axonal transport, their transport often consists of combinations of very rapid and very slow or no motion. Spending only a portion of their time associated with kinesin further allows for NFs to display a net slow transport rate while kinesin itself travels within fast axonal transport14-16. Potentially compounding the influence of phospho-dependent dissociation of NFs from kinesin are phospho-dependent NF–NF interactions17, 18, which may compete with NF–kinesin interactions. NF–NF complexes may be too large to be translocated effectively. In addition, we have observed that faster-moving NFs and non-filamentous oligomers that surround a slower-translocating ‘bundle’ of closely-apposed NFs19. The latter bear a relatively higher concentration of a developmentally delayed C-terminal phospho-epitope; this same phospho-epitope is predominant on those NFs that are selectively not associated with kinesin3. NFs within bundles may be sterically inhibited from re-association with their transport vector ( Fig. 1). The combination of phospho-dependent dissociation from their transport vector, coupled with phospho-dependent NF–NF interactions, could generate both the observed slowing of NFs during their continued transport along the axon, as well as the broad range of rates under which NFs undergo axonal transport20-22. Working model for the influence of phosphorylation on NF transport by kinesin. Shown are older NFs (black), and more-recently transported NFs (grey). Hypophosphorylated neurofilaments (indicated by curved sidearm extensions) are associated with the transport motor (1) pending phosphorylation (2), some of which also promote neurofilament bundling. Neurofilaments within bundles may be sterically blocked from re-associating with motors, in part by the continued transport of additional neurofilaments (3), which would induce the characteristic broadening of the transport wave. In summary, the demonstration that kinesin is involved in the distribution of NFs has allowed us to offer simplified mechanisms as potentially underlying several aspects of NF biology. Put simply, kinesin-mediated transport in a phospho-dependent but not assembly-dependent manner via the neuron's polarised microtubule array can account for many aspects of NF dynamics. Further studies to determine which kinesin(s), KIFs23 or additional motors participate in NF transport, as well as site-specific phosphorylation events that disrupt NF-kinesin interactions and foster NF–NF interactions, will extend our understanding of NF dynamics in normal and pathological conditions.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Shea et al. (2000) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: