Review: Axon pathology in age-related neurodegenerative disorders.

'Dying back' axon degeneration is a prominent feature of many age-related neurodegenerative disorders and is widespread in normal ageing. Although the mechanisms of disease- and age-related losses may differ, both contribute to symptoms. Here, we review recent advances in understanding axon pathology in age-related neurodegenerative disorders such as Alzheimer's disease, Parkinson's disease, amyotrophic lateral sclerosis and glaucoma. In particular, we highlight the importance of axonal transport, autophagy, traumatic brain injury and mitochondrial quality control. We then place these disease mechanisms in the context of changes to axons and dendrites that occur during normal ageing. We discuss what makes ageing such an important risk factor for many neurodegenerative disorders and conclude that the processes of normal ageing and disease combine at the molecular, cellular or systems levels in a range of disorders to produce symptoms. Pathology identical to disease also occurs at the cellular level in most elderly individuals. Thus, normal ageing and age-related disease are inextricably linked and the term 'healthy ageing' downplays the important contributions of cellular pathology. For a full understanding of normal ageing or age-related disease we must study both processes.

NAD(+) and axon degeneration revisited: Nmnat1 cannot substitute for Wld(S) to delay Wallerian degeneration.

The slow Wallerian degeneration protein (Wld(S)), a fusion protein incorporating full-length nicotinamide mononucleotide adenylyltransferase 1 (Nmnat1), delays axon degeneration caused by injury, toxins and genetic mutation. Nmnat1 overexpression is reported to protect axons in vitro, but its effect in vivo and its potency remain unclear. We generated Nmnat1-overexpressing transgenic mice whose Nmnat activities closely match that of Wld(S) mice. Nmnat1 overexpression in five lines of transgenic mice failed to delay Wallerian degeneration in transected sciatic nerves in contrast to Wld(S) mice where nearly all axons were protected. Transected neurites in Nmnat1 transgenic dorsal root ganglion explant cultures also degenerated rapidly. The delay in vincristine-induced neurite degeneration following lentiviral overexpression of Nmnat1 was significantly less potent than for Wld(S), and lentiviral overexpressed enzyme-dead Wld(S) still displayed residual neurite protection. Thus, Nmnat1 is significantly weaker than Wld(S) at protecting axons against traumatic or toxic injury in vitro, and has no detectable effect in vivo. The full protective effect of Wld(S) requires more N-terminal sequences of the protein.

Calcium-containing endosomes at oculomotor terminals in animal models of ALS.

Altered calcium homeostasis has been demonstrated in human spinal cord motor axon terminals of ALS patients, in spinal motor neurons of mutant SOD transgenic mice and following injection of ALS immunoglobulins. In all three paradigms oculomotor neurons are relatively spared. To explore mechanisms of selective resistance, we applied similar calcium localization techniques to terminals of oculomotor neurons in the two animal models. In both cases large vacuoles, which connect with the extracellular space, accumulated the majority of intracellular calcium, while terminals of vulnerable neurons (e.g. innervating interosseus muscle), which possess no such vacuoles, displayed evenly distributed calcium. These relatively unique membrane enveloped structures may permit neurons to control their cytoplasmic Ca2+ concentration and contribute to selective resistance.

Intra-axonal calcium changes after axotomy in wild-type and slow Wallerian degeneration axons.

Calcium accumulation induces the breakdown of cytoskeleton and axonal fragmentation in the late stages of Wallerian degeneration. In the early stages there is no evidence for any long-lasting, extensive increase in intra-axonal calcium but there does appear to be some redistribution. We hypothesized that changes in calcium distribution could have an early regulatory role in axonal degeneration in addition to the late executionary role of calcium. Schmidt-Lanterman clefts (SLCs), which allow exchange of metabolites and ions between the periaxonal and extracellular space, are likely to have an increased role when axon segments are separated from the cell body, so we used the oxalate-pyroantimonate method to study calcium at SLCs in distal stumps of transected wild-type and slow Wallerian degeneration (Wld(S)) mutant sciatic nerves, in which Wallerian degeneration is greatly delayed. In wild-type nerves most SLCs show a step gradient of calcium distribution, which is lost at around 20% of SLCs within 3mm of the lesion site by 4-24h after nerve transection. To investigate further the association with Wallerian degeneration, we studied nerves from Wld(S) rats. The step gradient of calcium distribution in Wld(S) is absent in around 20% of the intact nerves beneath SLCs but 4-24h following injury, calcium distribution in transected axons remained similar to that in uninjured nerves. We then used calcium indicators to study influx and buffering of calcium in injured neurites in primary culture. Calcium penetration and the early calcium increase in this system were indistinguishable between Wld(S) and wild-type axons. However, a significant difference was observed during the following hours, when calcium increased in wild-type neurites but not in Wld(S) neurites. We conclude that there is little relationship between calcium distribution and the early stages of Wallerian degeneration at the time points studied in vivo or in vitro but that Wld(S) neurites fail to show a later calcium rise that could be a cause or consequence of the later stages of Wallerian degeneration.

Altered calcium homeostasis in spinal motoneurons but not in oculomotor neurons of SOD-1 knockout mice.

SOD-1-deficient mice demonstrate no loss of motoneurons but are still vulnerable to axotomy and ischemic insults. To investigate possible reasons for vulnerability of motoneuron populations, we studied changes in ultrastructural calcium distribution during maturation in spinal- and oculomotor neurons in SOD-1(-/-) mice. Between 3 and 11 months the cytoplasmic component of the intracellular calcium changed at a lower rate in spinal motoneurons and motor axon terminals in the interosseus muscle of SOD-1(-/-) animals compared to wild-type controls. No such dissimilarities were noted in the oculomotor system, or in mitochondrial calcium contents of either cell type. These data suggest that the lack of SOD-1 may be associated with vulnerability to insult by depletion of non-mitochondrial calcium stores selectively in motoneurons lacking parvalbumin and/or calbindin D28K.