Download Mechanisms of axon degeneration: From development to disease

+ Models
PRONEU-845; No of Pages 18
Progress in Neurobiology xxx (2007) xxx–xxx
www.elsevier.com/locate/pneurobio
Mechanisms of axon degeneration: From development to disease
Smita Saxena, Pico Caroni *
Friedrich Miescher Institute, Maulbeerstrasse 66, CH-4058 Basel, Switzerland
Received 10 January 2007; received in revised form 31 March 2007; accepted 20 July 2007
Abstract
Axon degeneration is an active, tightly controlled and versatile process of axon segment self-destruction. Although not involving cell death, it
resembles apoptosis in its logics. It involves three distinct steps: induction of competence in specific neurons, triggering of degeneration at defined
axon segments of competent neurons, and rapid fragmentation and removal of the segments. The mechanisms that initiate degeneration are specific
to individual settings, but the final pathway of pruning is shared; it involves microtubule disassembly, axon swellings, axon fragmentation, and
removal of the remnants by locally recruited phagocytes. The tight regulatory properties of axon degeneration distinguish it from passive loss
phenomena, and confer significance to processes that involve it. Axon degeneration has prominent roles in development, upon lesions and in
disease. In development, it couples the progressive specification of neurons and circuits to the removal of defined axon branches. Competence
might involve transcriptional switches, and local triggering can involve axon guidance molecules and synaptic activity patterns. Lesion-induced
Wallerian degeneration is inhibited in the presence of WldS fusion protein in neurons; it involves early local, and later, distal degeneration. It has
recently become clear that like in other settings, axon degeneration in disease is a rapid and specific process, which should not be confused with a
variety of disease-related pathologies. Elucidating the specific mechanisms that initiate axon degeneration should open up new avenues to
investigate principles of circuit assembly and plasticity, to uncover mechanisms of disease progression, and to identify ways of protecting synapses
and axons in disease.
# 2007 Elsevier Ltd. All rights reserved.
Keywords: Axon degeneration; Development; Neurodegenerative disease; Wallerian degeneration; Proteasome; Synaptic activity; SOD1
Contents
1.
2.
3.
4.
5.
6.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Axon degeneration versus axon retraction . . . . . . . . . . . . . . . . . . . . .
2.1. Definitions and regulation . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2. Differential recruitment, and comparison to synapse elimination
The diverse settings involving axon degeneration. . . . . . . . . . . . . . . .
3.1. Developmental axon degeneration . . . . . . . . . . . . . . . . . . . . .
3.2. Lesions: Wallerian degeneration . . . . . . . . . . . . . . . . . . . . . . .
3.3. Axon degeneration in disease. . . . . . . . . . . . . . . . . . . . . . . . .
Shared sequence of events in axon degeneration . . . . . . . . . . . . . . . .
Mechanisms of axon degeneration . . . . . . . . . . . . . . . . . . . . . . . . . .
5.1. Wallerian degeneration . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2. Developmental axon degeneration . . . . . . . . . . . . . . . . . . . . .
5.3. Mechanisms of axon degeneration in FALS . . . . . . . . . . . . . . .
Shared mechanisms and outstanding issues . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
000
000
000
000
000
000
000
000
000
000
000
000
000
000
Abbreviations: AAD, acute axonal degeneration; APP, Alzheimer precursor protein; CNTF, ciliary neurotrophic factor; FALS, familial amyotrophic lateral
sclerosis; FF motoneuron, fast-fatiguable motoneuron; FR motoneuron, fast fatigue-resistant motoneuron; MB, mushroom body; Nmnat1, nicotinamide mononucleotide adenylytransferase 1; P15, postnatal day 15; S motoneuron, slow motoneuron; SOD1, superoxide dismutase 1; UPS, ubiquitin-proteasome system; VCP,
valosin containing protein; WldS, Wallerian degeneration slow.
* Corresponding author. Tel.: +41 61 6973727; fax: +41 61 6973976.
E-mail address: [email protected] (P. Caroni).
0301-0082/$ – see front matter # 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.pneurobio.2007.07.007
Please cite this article in press as: Saxena, S., Caroni, P., Mechanisms of axon degeneration: From development to disease, Prog. Neurobiol.
(2007), doi:10.1016/j.pneurobio.2007.07.007
+ Models
PRONEU-845; No of Pages 18
2
S. Saxena, P. Caroni / Progress in Neurobiology xxx (2007) xxx–xxx
7.
8.
Pathology versus axon degeneration in neurodegenerative diseases
7.1. ‘‘Dying-back’’ diseases . . . . . . . . . . . . . . . . . . . . . . . . . .
7.2. Axonal swellings and spheroids . . . . . . . . . . . . . . . . . . . .
Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction
Recent mechanistic studies of axon degeneration have
highlighted how this is an active process of controlled axon
self-destruction similar in many ways to the active selfdestruction of cells during apoptosis (Coleman, 2005; Low and
Cheng, 2005; Luo and O’Leary, 2005; Raff et al., 2002).
Although the molecular mechanisms involved are different (but
see Williams et al., 2006; Kuo et al., 2006), the principles
underlying the triggering and execution of these two selfdestruction processes do exhibit similarities. Thus, apoptosis
and axon degeneration both seem to involve tight and versatile
control pathways culminating into the disinhibition of
ubiquitous intrinsic destruction machinery, and the efficient
recruitment of local phagocytes to ensure removal of any trace
that might activate autoimmune responses. The available
evidence further suggests that the local self-destruction of
axons and dendrites underlie related mechanisms (Luo and
O’Leary, 2005; Williams and Truman, 2005), suggesting that
the degeneration process reflects a general intrinsic mechanism
to actively remove, and sometime replace, defined parts of a
neuron’s processes.
In view of the well-defined properties and significant
explanatory value of these active degeneration processes, it
seems to us important to reserve the term axon (or dendrite)
degeneration for self-destruction processes of entire neurite
segments, distinguishing them from local neuritic and organelle
pathology. According to this definition, diseases of the nervous
system can thus involve axonal and/or dendritic degeneration
processes, but these likely only represent a rather small fraction
of the pathology that can be associated with disease and/or age.
As will become clear in the last part of this review (Sections 7.1
and 7.2), such a distinction is particularly important in order to
avoid confusions that arise from our limited understanding of
the causal relationships between neuronal pathology, neuronal
dysfunction and disease progression. In contrast, the tightly
regulated and active aspects of the degeneration processes,
together with their dramatic outcome make it likely that, where
they can be unambiguously identified, axon and/or dendrite
degeneration processes will serve as valuable markers of
disease onset and progression. As discussed in the final part of
this review (Sections 7.1 and 7.2), this sets axon degeneration
processes apart from less well-delineated local pathology and
dysfunction phenomena.
This review elaborates on some of the major recent advances
in understanding axon degeneration and its regulation. We will
first define the process, comparing and contrasting it to axon
retraction. We will then discuss examples involving axon
degeneration during development, upon lesions, and in disease.
These distinct scenarios share a set of characteristic features
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
000
000
000
000
000
that define the degeneration process and provide first insights
into its mechanisms. We will then discuss what is currently
known about the specific mechanisms that control axon
degeneration during development, upon lesions and in disease.
A comparison of these specific mechanisms suggests a tentative
model of how axon degeneration induced in these distinct
settings might involve dedicated pathways of degeneration derepression. Finally, we will discuss the evidence for the
presence of axon degeneration processes in major nervous
system disease settings. Based on this analysis, we identify a
number of issues that, in our opinion, are currently unresolved,
and make suggestions for experimental systems and approaches
that will likely advance our understanding of the mechanisms of
axon loss in neurodegenerative diseases.
2. Axon degeneration versus axon retraction
2.1. Definitions and regulation
There are two opposite mechanisms through which
sections of axons are removed: retraction and degeneration
(Low and Cheng, 2005; Luo and O’Leary, 2005). In retraction,
the axonal process is gradually pulled backward, and the
corresponding axonal material is transported back to more
proximal sections of the axon. In contrast, degeneration
involves rapid blebbing, and fragmentation of an entire axonal
stretch into short segments, which are then removed by
phagocytic cells such as locally activated glia or macrophages.
Both processes are efficient and tightly regulated, and both can
be followed by regrowth of the axon from its proximal end.
However, although their ultimate outcome is similar, the two
processes differ in that all cellular material, including
intracellular signaling organelles (e.g. signaling endosomes),
is conserved inside the neuron in retraction, whereas
degeneration involves shedding of the axonal material, and
its uptake by local phagocytes. Axon retraction has only been
documented for modest local axon shortening events during
axon outgrowth, whereas larger removal events all appear to
involve axon degeneration.
In keeping with the recycling of distal axonal material,
axonal microtubules are not disrupted during retraction (Luo
and O’Leary, 2005), where they continue to carry out their
essential role as the predominant transport track for intracellular motor machinery (Gunawardena and Goldstein, 2004).
Instead, actin filaments are the cytoskeletal system that
mediates the retraction process. This mainly involves activation
of the small GTPase RhoA, followed by activation of the major
microfilament contracting component myosin light chain
through phosphorylation by the Rho effector ROCK (Bito
et al., 2000; Kozma et al., 1997; Luo and O’Leary, 2005).
Please cite this article in press as: Saxena, S., Caroni, P., Mechanisms of axon degeneration: From development to disease, Prog. Neurobiol.
(2007), doi:10.1016/j.pneurobio.2007.07.007
+ Models
PRONEU-845; No of Pages 18
S. Saxena, P. Caroni / Progress in Neurobiology xxx (2007) xxx–xxx
Interestingly, activation of RhoA to initiate retraction involves a
de-repression mechanism: in several cases, inactivation of the
predominant inhibitor of Rho p190RhoGAP initiates retraction
(Billuart et al., 2001; Lamprecht et al., 2002). As will become
apparent below, this mode of activation through de-repression
of pre-existing machinery appears to be a common theme in the
regulation of regressive processes in neurons, including axon
retraction, degeneration and of course apoptosis. One obvious
advantage of this mode of regulation is that in de-repression the
regressive machinery is already in place and poised to function
throughout the axonal (and dendritic) tree, allowing for rapid
and local triggering of the retraction (or degeneration) process
(Luo and O’Leary, 2005; Portera-Cailliau et al., 2005). A
further advantage is that systems operating based on derepression mechanisms allow for particularly tight and versatile
regulation of the triggering process.
In contrast to axon retraction, and as will be discussed in
more detail below (Section 5), axon degeneration involves
disassembly of microtubule filaments as an early, and possibly
critical, process (Luo and O’Leary, 2005). Although the issue
has not been addressed directly yet, it is therefore possible that,
like in axon retraction, the regulation of axon degeneration
might impinge onto the regulation of a major cytoskeletal
element in neurons (Yamada et al., 1970). Accordingly,
mechanisms regulating microtubule stability may not only
play important roles in axon guidance, but also in axon
degeneration. Candidate microtubule regulatory components of
established neuronal processes include Microtubule Associated
Proteins (MAPs) such as tau and MAP2, tubulin acetylating and
de-acetylating proteins such as HDAC6, and filament severing
proteins such as Katanin (Baas and Qiang, 2005; King et al.,
2006; Qiang et al., 2006; Yu et al., 2005).
2.2. Differential recruitment, and comparison to synapse
elimination
When are axon segments eliminated by retraction and when
by degeneration? As a simple rule, retraction seems to involve the
elimination of shorter branches, whereas degeneration tends to
affect longer axonal segments (Low and Cheng, 2005; Luo and
O’Leary, 2005). That might in part reflect the fact that
degeneration seems to be the more economical and mechanistically simpler solution to the problem of rapidly eliminating
larger sections of a process. However, there are exceptions to this
simple size rule, and the issue is likely more interesting and
complex. Thus, retraction can involve comparatively long axonal
branches such as collaterals of hippocampal mossy fibers that
initially form extended infrapyramidal projections in CA3, and
are pruned through an Ephrin-dependent retraction process
during hippocampal development (Bagri et al., 2003; Gao et al.,
1999). Likewise, although the removal of short axonal sidebranches during synapse elimination in the development of
motor nerve-muscle innervation does not proceed through a
canonical degeneration process, it is clearly more reminiscent of
degeneration than retraction (Bernstein and Lichtman, 1999;
Bishop et al., 2004). As will be discussed (Section 5),
developmental axon degeneration (i.e. axon degeneration in
3
the absence of lesions) seems to require transcriptional processes
in neurons that confer intrinsic competence for degeneration. In
contrast, although intrinsic factors could also be involved,
retraction might predominantly involve local regulation.
Accordingly, retraction could be viewed as a more contingent,
dynamically regulated and potentially reversible local phenomenon, whereas degeneration processes likely reflect more global
(e.g. transcriptional) transition states in neurons.
Developmental synapse elimination is a local process
involving the activity-dependent and competitive elimination
of inputs by most axons, and the partial reoccupation of vacated
synaptic sites by side-branches of the remaining axons
(Bernstein and Lichtman, 1999; Buffelli et al., 2003; Kasthuri
and Lichtman, 2003; Lichtman and Colman, 2000; Lohof et al.,
1996; Penn et al., 1998; Personius and Balice-Gordon, 2001).
Synapse elimination occurs subsequent to naturally occurring
neuronal cell death; it does not involve losses of neurons, but
instead shrinkage and focusing of their local arborization
territories. The process has been analyzed in a series of
groundbreaking in vivo imaging studies of postnatal neuromuscular junctions by the laboratory of Jeff Lichtman (BaliceGordon and Lichtman, 1993; Buffelli et al., 2003; Kasthuri and
Lichtman, 2003, 2004; Walsh and Lichtman, 2003). These
authors have recently shown that short axonal side-branches to
individual neuromuscular junctions are eliminated through the
formation of swellings (for which they introduced the term
axosomes), the thinning and disappearance of the intervening
process connecting the axosomes, and the removal of the
axosomes by phagocytosis through Schwann cells (Bishop
et al., 2004). Axosomes can then be detected as doublemembrane systems within Schwann cell cytosol, and some of
the axosome content eventually gets access to Schwann cell
cytosol. Interestingly, and in further analogy to axon
degeneration, swellings and axosomes contain very few intact
microtubules. Similar double-membrane engulfment systems
have been reported during the synapse elimination process of
climbing fibers in the cerebellum (Eckenhoff and Pysh, 1979),
where entire climbing fiber inputs are eliminated from
individual Purkinje cells, to produce the mature innervation
pattern of one climbing fiber per Purkinje cell (Lohof et al.,
1996; Mason and Gregory, 1984). Although no corresponding
in vivo imaging data have been reported yet for this synapse
elimination process on a larger terminal arborization scale than
that of short side-branches to individual neuromuscular
junctions, it seems likely that it will again involve an axosome
shedding process. It will be of particular interest to determine
the extent to which synapse elimination is mechanistically
similar to developmental axon degeneration. The two processes
might well be closely related, and their comparison might
provide insights into the regulatory mechanisms that initiate
and regulate synapse elimination, a process about which
comparatively little is known so far; a comparison to
developmental axon degeneration and synapse elimination
might also provide mechanistic insights into side-branch and
synapse elimination processes mediating structural plasticity in
the adult (De Paola et al., 2006; Galimberti et al., 2006;
Lichtman and Colman, 2000).
Please cite this article in press as: Saxena, S., Caroni, P., Mechanisms of axon degeneration: From development to disease, Prog. Neurobiol.
(2007), doi:10.1016/j.pneurobio.2007.07.007
+ Models
PRONEU-845; No of Pages 18
4
S. Saxena, P. Caroni / Progress in Neurobiology xxx (2007) xxx–xxx
3. The diverse settings involving axon degeneration
3.1. Developmental axon degeneration
All studies of neuronal circuit assembly during development
have highlighted how axon guidance and target innervation
proceed with high specificity and accuracy, producing a very low
degree of errors (Katz and Shatz, 1996). Why then should axon
degeneration be a significant feature in neural development? The
answers to this simple-minded question are interesting, and often
provide unexpected insights into the logics underlying the
evolution and development of nervous systems. Not surprisingly,
and like synapse elimination, axon degeneration does not
primarily serve the purpose of improving accuracy during circuit
assembly (Low and Cheng, 2005; Luo and O’Leary, 2005).
Instead, it seems to reflect processes of progressive neuronal
specification, and to provide flexibility for circuit assembly
mechanisms among individuals and across species (Luo and
O’Leary, 2005). Due to the fact that it represents a natural process
uninfluenced by artificial manipulations or pathology and due to
its high predictability in time and space, developmental axon
degeneration is particularly well suited to investigate the
mechanisms underlying axon degeneration.
Perhaps the best-known examples for developmental processes involving axon degeneration can be found in the
development of cortical layer 5 projection neurons, a system
that has been thoroughly investigated by the laboratory of Dennis
O’Leary (Luo and O’Leary, 2005; O’Leary, 1992; O’Leary and
Koester, 1993). These neurons project to multiple specific targets
as a function of their cortical identity. Thus, for example, motor
cortex neurons project, among other targets, to the spinal cord,
whereas visual cortex neurons project to the superior colliculus.
Remarkably, however, these diverse cortical neurons initially
send off axon branches to most or even all potential targets of
layer 5 neurons, and subsequently prune those branches that are
inappropriate for their identity (Bastmeyer and O’Leary, 1996;
O’Leary and Stanfield, 1989; O’Leary and Terashima, 1988;
Stanfield and O’Leary, 1985b; Stanfield et al., 1982). This
pruning process takes place at a well-defined time in
development, and exhibits typical features of axon degeneration,
including widespread blebbing and rapid removal of axonal
remnants (Hoopfer et al., 2006; Luo and O’Leary, 2005).
Transplantation experiments early during the development of the
neocortex have established that layer 5 projections degenerate
according to the identity of the host region. This suggests that
specification processes involving interactions with the local
environment play an important role in determining the patterns of
branch degeneration of cortical projection neurons (Stanfield and
O’Leary, 1985a). The above finding highlights how this late
process of axonal specification might provide a high degree of
flexibility in matching requirements for early outgrowth and
target invasion by cortical projection neurons to protracted
processes of neuronal identity plasticity in the development of
the neocortex (Luo and O’Leary, 2005). This dissociation
between early generic processes of axon guidance, and
protracted processes of neuronal specification followed by
corresponding axonal refinements might also provide the
developing neocortex with the flexibility required to modify
neocortical neuron projection patterns among closely related
subtypes of cortical projection neurons, and among the same type
of neuron in closely related species (Innocenti et al., 1999; Luo
and O’Leary, 2005; Zufferey et al., 1999). Similar processes of
extensive degeneration have been described for the callosal
projections of cortical neurons (Innocenti, 1981, 1995; Innocenti
et al., 1983, 1999; Zufferey et al., 1999), and for the fornix
projections of hippocampal subiculum neurons beyond the
mamillary bodies of the hypothalamus (Stanfield et al., 1987;
Stanfield and O’Leary, 1985a).
In addition to pruning their transient projections to subsets of
targets that have become inappropriate, neurons can prune
aspects of their arborization within their target region (Low and
Cheng, 2005; Luo and O’Leary, 2005; O’Rourke et al., 1994).
This pruning again involves specific axon degeneration; where it
has been described so far, it serves the purpose of achieving
appropriate topographic termination patterns (Luo and O’Leary,
2005). Degeneration in the target region has been studied most
extensively in the retinotopic projection of ganglion cells from
distinct positions in the retina to the superior colliculus
(mammals) or optic tectum (birds) (Luo and O’Leary, 2005).
Ganglion cell axons initially grow throughout the longitudinal
extension of their target, extend transverse collaterals from
multiple points along the longitudinal projection, and then local
arborizations at their topographically correct terminations. This
process is followed by degeneration of the longitudinal
projection distal to the appropriate collateral, and of all
inappropriate proximal collaterals (Luo and O’Leary, 2005).
The molecular information for appropriate retinotopic connectivity is based on the graded expression of ephrins and their
receptors in the retina and the superior colliculus, suggesting that
this repulsive signaling mechanism might be involved in
directing the degeneration of inappropriate processes in the
colliculus (Hindges et al., 2002; Yates et al., 2001). Degeneration, however, further depends on the expression of correlated
synaptic activity among groups of retinal ganglion cells during a
critical time window of development (Bansal et al., 2000; Debski
and Cline, 2002; Feller, 2002; McLaughlin et al., 2003a,b;
Meister et al., 1991; O’Rourke et al., 1994; Tian and
Copenhagen, 2003; Wong et al., 1993).
In addition to projecting to the superior colliculus, a subset
of retinal ganglion cells sends collaterals to the dorsal lateral
geniculate nucleus, where they ultimately establish projections
in an eye-specific local segregation pattern (Chen and Regehr,
2000). Again, the segregated pattern emerges from local
pruning of terminal branches in a process that depends on
correlated synaptic activity among ganglion cells (Bansal et al.,
2000; Chen and Regehr, 2000; Feller, 2002; Penn et al., 1998;
Ruthazer et al., 2003; Stellwagen and Shatz, 2002; Tian and
Copenhagen, 2003; Wong et al., 1993). Interestingly, this
smaller-scale pruning process can be induced during a longer
period of late development than the degeneration of retinotopically inappropriate projections in the superior colliculus (Luo
and O’Leary, 2005).
At least for certain types of neurons, the local expression
levels of target-derived neurotrophic factors such as NGF,
Please cite this article in press as: Saxena, S., Caroni, P., Mechanisms of axon degeneration: From development to disease, Prog. Neurobiol.
(2007), doi:10.1016/j.pneurobio.2007.07.007
+ Models
PRONEU-845; No of Pages 18
S. Saxena, P. Caroni / Progress in Neurobiology xxx (2007) xxx–xxx
BDNF, NT3 or GDNF can influence the extent of terminal
arborization within the target region (Alsina et al., 2001;
Markus et al., 2002; Patel et al., 2003; Petersson et al., 2003). In
experiments that have established this principle, transgenic
overexpression of neurotrophins in natural or ectopic targets of
sensory neurons expressing corresponding receptors led to
higher innervation densities or the establishment of ectopic
innervation by these neurons in the adult (Causing et al., 1997;
LeMaster et al., 1999). In complementary experiments, sensory
neurons lacking neurotrophin receptors, but prevented from
dying by deletion of the pro-apoptotic gene Bax, exhibited
greatly reduced target arborization sizes (Markus et al., 2002;
Patel et al., 2003). The relationship between target-derived
neurotrophic factors influencing target innervation densities
and axonal degeneration processes has not yet been investigated
in much detail. However, the principles revealed by neurotrophic factor-controlled axon growth and innervation density
have led to a powerful experimental paradigm to investigate
axon degeneration in vitro. Thus, elegant studies by Robert
Campenot using a two-compartment in vitro assay, which
allows separate access and treatment of the cell body and
axonal compartments of sensory neurons, have established that,
at least in vitro, neurotrophins acting locally on axons have
specific roles in preventing axon degeneration (Campenot,
1982). Neurotrophin depletion in the axonal but not the cell
body compartment led to a rapid degeneration of the section of
the axon depleted of NGF, but spared the axon stump inside the
cell body compartment (MacInnis and Campenot, 2005). These
findings have provided compelling evidence that local signaling
can direct the patterns of axon degeneration. In addition, they
have provided an attractive experimental system to investigate
the regulation of axon degeneration in the absence of lesions
(see also below).
Vertebrate systems have provided a fascinating range of
experimental paradigms to investigate the roles of axon
degeneration in the assembly of neuronal circuits, but possibly
the most powerful experimental systems to investigate
mechanisms of axon degeneration in development and upon
lesions have been established in Drosophila, mainly in the
laboratory of Liqun Luo (Luo and O’Leary, 2005). While many
fly neurons function specifically either during larval or during
adult stage, some neurons function during both stages and
reorganize their projections during metamorphosis to accommodate new connectivity requirements in the adult (Levine
et al., 1995; Technau and Heisenberg, 1982; Tissot and Stocker,
2000). This is thus again an example of axonal projections that
have become inappropriate upon a switch in neuronal and
circuit specification, and are, as a consequence, eliminated
through axon degeneration. Mushroom body (MB) g neuron
axons prune their larval collaterals and termination regions by a
typical degeneration process at the onset of metamorphosis, but
retain the initial segment of the axon, which then re-grows to
form the adult projection (Watts et al., 2003). Interestingly,
unlike g neurons, MB a and b neurons do not remove parts of
their projection through retraction, but inhibition of p190RhoGAP does cause retraction in these neurons, underscoring the
exquisite specificity of degeneration- and retraction-mediated
5
regressive events (Billuart et al., 2001; Luo and O’Leary, 2005).
The degeneration process of MB g neuron axons depends on the
onset of ecdysone hormone signaling (a system related to
retinoic acid signaling in vertebrates) at metamorphosis
(Schubiger et al., 1998; Technau and Heisenberg, 1982), and
on the cell autonomous expression and function of corresponding nuclear hormone receptor pairs in MB g neurons (Lee et al.,
2000; Schubiger et al., 1998). The degeneration process
proceeds in a temporally well-defined manner, so that its stages
could be characterized in great detail (Luo and O’Leary, 2005;
Watts et al., 2003, 2004; Zhai et al., 2003). The analysis
revealed that axonal degeneration first involves microtubule
depolymerization, followed by axonal blebbing, axon fragmentation, neurofilament degradation, and removal of the
axonal swellings (Watts et al., 2003; Zhai et al., 2003).
Interestingly, microtubule depolymerization and axonal blebbing are not restricted to the sections of distal axons that would
degenerate, but also extend into the proximal section of the
axon peduncle (i.e. the section that persists, and will grow again
after the degeneration of the distal parts). In contrast,
fragmentation is restricted to the axonal sections destined for
removal (Watts et al., 2003). It might be significant that the
proximal axon peduncle, which does not fragment, does thin
out substantially before starting to grow (Watts et al., 2003).
These results suggest the existence of separate control
mechanisms to initiate degeneration in specific neurons, and
to spatially restrict the process to well-defined sections of their
axonal arbor. As discussed below (Section 6), the mechanisms
that assign defined axon segments for degeneration, ensuring
that degeneration is restricted spatially to those defined axonal
sections remain a major open issue in understanding axonal
degeneration.
3.2. Lesions: Wallerian degeneration
One setting where defining the segment of an axon destined
for degeneration does not seem to be a major challenge is the
degeneration of the portion of an axon that lies distally from a
physical lesion. This lesion-induced degeneration process had
already been described by Waller in 1850 for peripheral nerves
(Waller, 1850), and is since known as Wallerian degeneration
(Coleman, 2005; Raff et al., 2002). The interest in the
mechanisms underlying Wallerian degeneration has grown
dramatically since the discovery of a mutant mouse strain (WldS
mice, for Wallerian degeneration Slow), where Wallerian
degeneration is greatly delayed, and possibly suppressed
(Coleman, 2005; Glass et al., 1993; Mack et al., 2001).
Transplantation experiments established that the alteration was
intrinsic to the transected axons (Glass et al., 1993), thus ruling
out the possibility that Wallerian degeneration might involve a
passive process of starvation caused by the interruption of
material delivery from the neuronal cell body to the distal part of
the axon. Subsequent cloning of the mutation revealed the
presence of a genomic triplication, giving rise to a unique fusion
protein (the WldS protein), consisting of the first 70 residues
of UFD2a/UBE4B, a polyubiquitination enzyme, full-length
nicotinamide mononucleotide adenylytransferase (Nmnat1), an
Please cite this article in press as: Saxena, S., Caroni, P., Mechanisms of axon degeneration: From development to disease, Prog. Neurobiol.
(2007), doi:10.1016/j.pneurobio.2007.07.007
+ Models
PRONEU-845; No of Pages 18
6
S. Saxena, P. Caroni / Progress in Neurobiology xxx (2007) xxx–xxx
enzyme involved in NAD synthesis, and a brief linker sequence
(Conforti et al., 2000; Mack et al., 2001). The exact mechanism
through which WldS suppresses Wallerian degeneration is not
entirely clear yet, and, as discussed below (Section 6), is a major
challenge for future studies. Nevertheless, by providing the field
with a molecular tool to influence axonal degeneration, the
discovery, characterization and exploitation of WldS in gain-offunction studies, mainly by Michael Coleman and his
collaborators, has been a critical breakthrough in studying the
mechanisms of axonal degeneration.
Wallerian degeneration can be investigated in the PNS and
CNS in vivo, but also in in vitro models of lesion-induced
axonal degeneration (Coleman, 2005; Hoopfer et al., 2006;
MacInnis and Campenot, 2005). As an experimental system, it
has the obvious advantage that the experimenter determines the
precise site and initiation time of the process. Its scientific and
clinical interest is further enhanced by the fact that, through its
removal of axon remnants, debris and inhibitory factors,
Wallerian degeneration likely facilitates the regeneration of the
proximal stump of the lesioned axon into and along the distal
nerve. In adult mice, the distal sections of physically lesioned
nerves (crush or transection) do not exhibit obvious structural
changes during the first 36 h after the lesion. This period of
apparent preservation is followed by a rapid process of axonal
degeneration, which is completed within about an hour
(Beirowski et al., 2004; Coleman, 2005). In close similarity
to developmental axon degeneration, Wallerian degeneration
involves microtubule disassembly, blebbing of the distal axon,
followed by fragmentation and removal of the remnants by
phagocytes (mainly macrophages upon peripheral nerve
lesions) (Coleman, 2005; Luo and O’Leary, 2005). Peripheral
synapses are removed slightly prior to distal axon segments, but
the difference in time is not such that it by itself makes a
compelling case for the existence of a separate mechanism of
synapse degeneration (Parson et al., 1997; but see also Miledi
and Slater, 1970). While very similar processes of Wallerian
degeneration have been observed in several organisms
including Drosophila, and in animals of different ages
including those in which lesions were made during development, the time it takes from the lesion to the first signs of
degeneration can vary from 5 h (Drosophila) to up to 2 days
(Luo and O’Leary, 2005). Finally, some chemically induced
acute lesions such as local lesions of microtubules with
colchicine or vincristine induce axon degeneration closely
resembling Wallerian degeneration, and should probably be
considered as equivalent to physical lesions (Coleman, 2005).
A recent in vivo imaging study of lesioned DRG axons in
dorsal spinal cord has provided novel and unanticipated
insights into processes of Wallerian degeneration (Kerschensteiner et al., 2005). Using Thy1 transgenic mice in which a few
DRG neurons and their axons express the Green Fluorescent
Protein GFP, and can thus be repeatedly visualized in vivo, Jeff
Lichtman and his colleagues have discovered that nerve lesions
in vivo are followed by a further axon degeneration process that
takes place long before the onset of classical Wallerian
degeneration (Kerschensteiner et al., 2005). Thus, about 20 min
after the lesion, axonal swellings appear near the lesion site, but
at both the proximal and distal stump. These swellings are
rapidly followed by fragmentation and removal of about
150 mm of axon from the proximal and from the distal stump;
the fragmentation and removal process only takes about 5 min
from onset to completion. This striking degeneration process is
again suppressed in mice expressing the WldS protein,
providing strong evidence that it closely resembles classical
Wallerian degeneration (Kerschensteiner et al., 2005). Based on
its rapid onset, rapid completion, and restriction to axon
segments near the lesion site, Lichtman and colleagues termed
the process acute axonal degeneration (AAD). The rapid onset
and limited spatial extent of the process might reflect local
access by extracellular solutes, including calcium, to axonal
cytosol in the proximal and distal stump through the lesion site.
The functional consequences of this acute degeneration process
are important, and include disconnection from the cell body of
any side-branch within 150–200 mm from the lesion site, and
the formation of a gap that might significantly hinder
regeneration. The in vivo imaging studies also provided novel
insights about the Wallerian degeneration process itself (i.e.
distally from the lesion site), including the fact that, following a
delay time, axonal swellings develop rapidly (minutes to
seconds), that fragmentation rapidly follows the formation of
the swellings, and that WldS prevents all of these events
(Kerschensteiner et al., 2005). Several of these features had not
been appreciated in previous studies, due to the fact that the
degeneration process does not proceed synchronously in
individual lesioned nerves. This elegant study demonstrates
vividly how it will be important and rewarding to monitor axon
degeneration processes on-line, in their in vivo setting, and at
the level of identified axons.
3.3. Axon degeneration in disease
It has long been noted that some neurodegenerative diseases
involve protracted gradual ‘‘dying-back’’ processes of distal
synapses and axons that can precede the loss of neuronal cell
bodies by months (Cavanagh, 1964; Griffin et al., 1995). More
recently, it has become clear that most, and possibly all,
neurodegenerative diseases involve major synapse and axon
losses before the appearance of symptoms, and long before the
loss of neurons (Bjartmar et al., 2003; Griffin et al., 1995; Bruck,
2005; Fischer et al., 2004; Frey et al., 2000; Raff et al., 2002;
Sagot et al., 1996; Schmalbruch et al., 1991). These diseases
include Alzheimer’s, Parkinson’s, Huntington’s, motoneuron
diseases, and Prion diseases (Coleman and Yao, 2003; Fischer
et al., 2004; Frey et al., 2000; Gunawardena and Goldstein, 2005;
Li et al., 2001; Luo and O’Leary, 2005; Sagot et al., 1996;
Schmalbruch et al., 1991; Stokin et al., 2005). The long
prevailing focus of neurodegenerative research on neuronal cell
death and apoptosis pathways has tended to distract from these
important early observations (Raff et al., 2002). However, recent
unambiguous experimental demonstrations that, in at least some
neurodegenerative disease models, protecting cell bodies from
death has no impact on disease progression and life-span,
whereas protecting axons and synapses does, have helped to
refocus research on the ‘‘dying-back’’ phenomenon (Gould et al.,
Please cite this article in press as: Saxena, S., Caroni, P., Mechanisms of axon degeneration: From development to disease, Prog. Neurobiol.
(2007), doi:10.1016/j.pneurobio.2007.07.007
+ Models
PRONEU-845; No of Pages 18
S. Saxena, P. Caroni / Progress in Neurobiology xxx (2007) xxx–xxx
2006; Sagot et al., 1995, 1996, 1998; Ferri et al., 2003; see also
Libby et al., 2005). Clearly, if a neuron is chronically
disconnected from its targets through peripheral axon losses,
it will probably make little difference to the organism whether
that neuron stays alive or dies.
But is it so, that most neurodegenerative diseases involve
‘‘dying-back’’ processes? And, more importantly, do the
‘‘dying-back’’ processes exhibit any relationship to welldefined axon retraction or degeneration processes? These
questions are much more difficult to address experimentally
than it would seem. The main difficulties arise from the
apparently heterogeneous and non-synchronous nature of the
pathology processes in neurodegenerative diseases and their
animal models. One criterion that has been used in several
recent animal model studies has been to determine whether
pathology and/or disease progression were slowed down in the
presence of WldS protein. These attempts have met with
success in some cases, and with no success in others, but the
conclusions are not as straightforward as one might have
expected (see discussion in Sections 7.1 and 7.2). A
fundamentally different approach consists in investigating
the progression of ‘‘dying-back’’ pathology in well-defined
animal models of neurodegenerative disease as if it was a
developmental process, investigating disease progression at the
level of identified axons, and attempting to establish a detailed
and predictive map of the disease process that would serve as a
basis to investigate the properties and mechanisms of the
‘‘dying-back’’ phenomenon. This approach was recently
applied successfully to two mouse models of human familial
amyotrophic lateral sclerosis (FALS) (Pun et al., 2006), and the
main findings are summarized below.
ALS is a deadly late-onset degenerative motoneuron disease
of high prevalence (life-time risk of ca. 1:1000) (Boillee et al.,
2006a). The familial forms of ALS represent about 10% of total
cases, and about a fifth of them are due to point mutations in the
ubiquitous and abundant cytosolic enzyme superoxide dismutase 1 (SOD1) (Boillee et al., 2006a). The mode of
inheritance of mutant SOD1-linked FALS is autosomal
dominant. Why inherited mutations in a ubiquitous protein
cause late-onset degeneration restricted to specific types of
neurons is a conundrum that is not unique to FALS, and is in
fact a main mystery in neurodegenerative diseases (Monani,
2005). Disease-linked human SOD1 mutants have been
expressed as transgenes in mice, using a human minigene to
drive ubiquitous expression resembling that of the protein in
humans (Gurney et al., 1994). These studies have established
that SOD1 mutants act through toxic gain-of-function
mechanisms to cause FALS in humans and mice (Bruijn
et al., 1997; Wong et al., 1995). Subsequent detailed studies,
mainly from the laboratories of Don Cleveland and Jean-Pierre
Julien, have established that axonal transport is a major early
target of disease in FALS (Collard et al., 1995; Lobsiger et al.,
2005; Xia et al., 2003; Zhang et al., 1997), although the causal
relationships between early presymptomatic defects in axonal
transport and disease onset and progression are not entirely
clear yet (Boillee et al., 2006a). Finally, in a striking and
important new development, studies using cell type specific
7
transgenics (Lino et al., 2002), conditional knockouts of mutant
transgene (Boillee et al., 2006b), and elegant analysis of
chimeric mice (Clement et al., 2003) have established that the
disease process in FALS is not cell autonomous to motoneurons
(Boillee et al., 2006a). Instead, mutant SOD1 expression in
motoneurons is necessary but not sufficient for disease
initiation and progression, and expression in microglia drives
disease progression; it is not excluded that mutant expression in
further cell types also influences the disease process in these
FALS models (Boillee et al., 2006a,b).
A line of human-minigene transgenic mice, which express
high levels of human SOD1(G93A), exhibit first clinical signs
of disease at P80–90, first motoneuron losses at P90–100, and
die at postnatal day (P) 135 4 (Gurney et al., 1994). The
consistency in the timing of disease onset and progression is a
striking feature of several animal models of neurodegenerative
diseases, and its investigation will likely lead to interesting
insights about the relationships between disease and aging. In
SOD1(G93A) mice, this tight temporal predictability has made
it possible to analyze motoneuron axons and synapses in
separate mice of different ages, and to treat the data as if they
were derived from a longitudinal study (i.e. a study of disease
progression within the same individual) (Pun et al., 2006). A
detailed comparative analysis of neuromuscular innervation
patterns in the hindlimb of wildtype and SOD1(G93A) mice
revealed that reproducible numbers of axons innervating
topographically defined regions of several hindlimb muscles
were lost abruptly between P48 and P50 in the mutant mice
(Fischer et al., 2004; Frey et al., 2000; Pun et al., 2006). A
second group of axons were lost in many muscles at P80–90 and
the remaining motoneuron axons still innervated their muscles
when the mice died. Further analysis revealed that the axons
that were lost at P48–50 invariably belonged to fast-fatiguable
(FF) motoneurons, the intermediate axons (lost at P80–90)
belonged to fast fatigue-resistant (FR) motoneurons, and the
most resistant axons belonged to slow (S) motoneurons (Pun
et al., 2006). These motoneuron subtypes exhibit distinct
physiological properties, and are recruited under distinct and
specific circumstances to produce distinct muscle contraction
forces (Burke, 1994). Since individual motor pools (the set of
all alpha-motoneurons in the spinal cord innervating a given
skeletal muscle) are made of distinct and characteristic
combinations of physiological subtypes of motoneurons
(Burke, 1994), this selective vulnerability of FF, and then
FR motoneuron axons means that in SOD1(G93A) mice certain
muscles lose most of their innervation at P48–50, whereas other
muscles still maintain a large part of their innervation when the
mice die. Importantly, because the distribution of the
innervation in individual muscles is highly organized and
reproducible among the individuals of a species (Pun et al.,
2006), anatomically defined subcompartments of large muscles
exclusively innervated by FF motoneurons become completely
and reproducibly denervated at P48–50; furthermore, subcompartments innervated by FF and FR motoneurons lose their
FF innervation at P48–50, and all remaining innervation at
P80–90 (Pun et al., 2006). These results have made it possible
to investigate the process of disease-related axon loss with
Please cite this article in press as: Saxena, S., Caroni, P., Mechanisms of axon degeneration: From development to disease, Prog. Neurobiol.
(2007), doi:10.1016/j.pneurobio.2007.07.007
+ Models
PRONEU-845; No of Pages 18
8
S. Saxena, P. Caroni / Progress in Neurobiology xxx (2007) xxx–xxx
unprecedented precision and predictability. By focusing on the
loss of axons innervating purely FF subcompartments, it was
possible to not only monitor the process of axon loss, but also to
identify earlier alterations in neuromuscular junctions and
axons that were predictive of subsequent axon loss. The
analysis established that in SOD1 models of FALS, the
peripheral intramuscular part of FF motoneuron axons, with all
their peripheral synapses, is lost abruptly (i.e. pruned, see
below) between P48–50, and that the pruning process exhibits
the characteristic features of axon degeneration, including rapid
appearance of swellings, axon fragmentation and rapid removal
of the debris. FR axons are pruned at P80–90, through a
comparable process (Pun et al., 2006; Schaefer et al., 2005).
Starting about 8–10 days prior to axon degeneration, vulnerable
FF axons exhibit local accumulations of synaptic vesicles and a
loss of synaptic vesicles from neuromuscular junctions,
suggesting the presence of a defect in anterograde transport
in these vulnerable axons (Pun et al., 2006). Local daily
applications to muscle of CNTF, a trophic factor known to
protect motoneuron axons in motoneuron diseases (Linker
et al., 2005), prevents synaptic vesicle depletion from
neuromuscular junctions and FF axon degeneration. Interestingly, applications of CNTF after P45 fails to protect FF
motoneuron axons from degeneration, suggesting that CNTF
might counteract pathological processes leading up to axon
degeneration, but might not inhibit axon degeneration itself
(Pun et al., 2006). Very similar, but time-shifted observations
were made in SOD1(G85R) mice, a different SOD1 model of
FALS in which mice exhibit clinical signs of disease around 7
months, and die at 8–9 months (Bruijn et al., 1997; Pun et al.,
2006). These results suggested that selective vulnerability and
axonal degeneration of first FF and then FR motoneuron axons
is a central feature of disease progression in mouse models of
FALS (Pun et al., 2006). More generally, these FALS models in
mice provided evidence that regulated, selective and efficient
axonal degeneration does take place during the course of at
least one neurodegenerative disease (Pun et al., 2006; Schaefer
et al., 2005). Together with the detailed anatomical and
temporal information about the degeneration process, these
models provide an experimental system to investigate the
mechanisms that can trigger axon degeneration in the adult, in
the absence of acute physical or chemical lesions to axons.
4. Shared sequence of events in axon degeneration
As we have seen, the settings in which axon degeneration
has been documented are very different. Axon degeneration
during development is a tightly regulated and confined selfdestruction process triggered in the absence of lesions or
pathology; it depends on mechanisms that define the onset, and
the exact location of the destruction. These mechanisms can be
very different, ranging from differentiation processes to
patterns of synaptic activity during critical periods. The scale
of ‘‘programmed’’ axon destruction can vary from most of a
neuron’s processes in Drosophila MB g neurons, to the
elimination of innervation to entire targets (e.g. layer 5 cortical
neurons), and on to the local elimination of collaterals in the
target region (e.g. retinal ganglion cell innervation in superior
colliculus). Elimination on an even smaller scale might be
involved in the segregation of eye inputs in dorsal lateral
geniculate nucleus, and in synapse elimination, although these
local events based on activity-dependent competition might
underlie qualitatively different mechanisms. When comparing
the different developmental settings (Fig. 1), it seems clear that
regulation type or pruning scale do not make a significant
difference with respect to the characteristic temporal and
Fig. 1. Schematic representation of the shared (black) and setting-specific (colours) processes involved in axon degeneration upon lesions (Wallerian degeneration)
and in development. Although it likely reflects a general feature of axon degeneration, a requirement for phagocyte recruitment (glia) has been demonstrated for
developmental degeneration in Drosophila MB g neurons, but not yet in the other settings.
Please cite this article in press as: Saxena, S., Caroni, P., Mechanisms of axon degeneration: From development to disease, Prog. Neurobiol.
(2007), doi:10.1016/j.pneurobio.2007.07.007
+ Models
PRONEU-845; No of Pages 18
S. Saxena, P. Caroni / Progress in Neurobiology xxx (2007) xxx–xxx
morphological features of developmental axon degeneration.
In each case, the entire axonal section to be removed is affected
in a nearly synchronous manner, and degeneration involves
microtubule disassembly, followed by axonal swellings,
neurite and neurofilament fragmentation, and removal of the
remnants by recruited phagocytes. Furthermore, for any given
axon, the process typically only takes a few hours from onset to
completion. Significantly, although Wallerian degeneration is
brought about by a lesion, which the axons involved sustain
passively, the characteristic sequence of events, in vivo or in
vitro, does not differ significantly from that in developmental
axon degeneration (Coleman, 2005; Luo and O’Leary, 2005)
(Fig. 1). Finally, although the information is currently less
complete, axonal degeneration associated with disease, i.e. a
process that is neither programmed nor induced by an acute
lesion, does not seem to be significantly different from
degeneration in development or upon lesions. This convergence onto an apparently common pattern of axon selfdestruction suggests that it might be useful to subdivide studies
of axon degeneration into three main aspects: the specific
mechanisms that activate the pathway of self-destruction, the
particular mechanisms that define the exact section of an axon
to be destroyed, and the common mechanism of axon
destruction.
5. Mechanisms of axon degeneration
5.1. Wallerian degeneration
Wallerian degeneration (i.e. lesion-induced axon degeneration) is efficiently blocked in the presence of mouse WldS
protein, irrespective of whether it is induced in vertebrates or
Drosophila (Mack et al., 2001; MacDonald et al., 2006). In the
presence of WldS, lesioned vertebrate axons fail to degenerate
and continue to conduct action potentials for 2–3 weeks; the
slow degradation that eventually ensues differs from the
degeneration process, suggesting that the presence of WldS
completely negates Wallerian degeneration (Coleman, 2005).
As we will see below (Sections 5.2 and 5.3), this sets Wallerian
degeneration apart from developmental axon degeneration
(WldS insensitive, Hoopfer et al., 2006), and also, at least in
some cases, from disease-related degeneration. WldS protein
thus likely affects a reaction specific to lesion-induced
degeneration. That reaction must lie at the very onset of the
degeneration process or at some point during the delay period
that precedes axon degeneration, since WldS protein blocks all
steps of the degeneration process, including microtubule
disassembly. Since WldS is not detected in the axon, and is
detected essentially only in the nucleus, the reaction might
involve gene expression processes (Coleman, 2005). That
suggests that WldS might affect the expression of proteins
important to initiate Wallerian degeneration; indeed, WldS does
alter the expression of a specific set of genes in neurons
(Gillingwater et al., 2006). Based on these observations, a
tentative model for the initiation of Wallerian degeneration
would include the constitutive presence (or absence) throughout axons of WldS-sensitive proteins essential to initiate lesion-
9
induced degeneration (Coleman, 2005). In the presence of
signals initiated by the lesion process, these targets of WldS
would be essential to disinhibit, and thus activate the
degeneration machinery (Coleman, 2005). The fact that
activation of local degeneration in AAD is dramatically more
rapid (about 20 min) than the 36 h needed for the onset of the
distal degeneration process in the very same axons, suggests
that degeneration might be triggered by a leak process
associated with physical disruption, e.g. a leak of ions such
as calcium and/or sodium, possibly followed by reverse
operation of the sodium-calcium exchanger and the accumulation of calcium, which have been shown to contribute to
Wallerian degeneration (Coleman, 2005). The exact mechanism affected by WldS is not clear yet, but it specifically affects
axon degeneration and not apoptosis (Finn et al., 2000), does
not seem to affect ubiquitination, and also does not seem to
represent simple overexpression of Nmnat1 protein (Coleman,
2005; Conforti et al., 2007). One intriguing suggestion is that
WldS might interact with the chromatin-deacetylation protein
SIRT1, an effector of NAD that has been implicated in the
regulation of cellular aging (Araki et al., 2004). A further lead is
that its N-terminal end interacts in the nucleus with the key
ubiquitin-proteasome system (UPS) component valosin-contaning protein (VCP) (Coleman, 2005; DiAntonio et al., 2001).
In addition to WldS protein, Wallerian degeneration depends
on the activity of the UPS (Zhai et al., 2003). Blockade of UPS
with specific inhibitors, or upon overexpression of interfering
mutants, prevents all steps of the degeneration process,
including microtubule disassembly. Significantly, and unlike
WldS, UPS inhibitors prevent axonal degeneration in all
settings where it has been tested (see Sections 5.2 and 5.3)
(Hoopfer et al., 2006). UPS activity is thus a general
requirement to initiate axon degeneration. That requirement
must specifically apply to an upstream process in the initiation
of the degeneration pathway because application of UPS
inhibitors just prior or subsequent to lesions did not interfere
with Wallerian degeneration (Zhai et al., 2003; Coleman, 2005;
Hoopfer et al., 2006). Although uncertainty about the exact
time when the inhibitors reach sufficient concentrations inside
axons precludes a definitive conclusion, these findings suggest
that, although distal degeneration might only start 36 h after a
lesion, critical reactions are triggered around the time of the
actual lesion. A possible and testable model could thus envision
UPS-dependent activation of degeneration competence at the
time of the lesion or shortly thereafter, and variable initiation of
the degeneration process, either 20 min (local AAD; Kerschensteiner et al., 2005) or 5–36 h after the lesion (distal
degeneration). In this model, WldS protein might lead to the
absence of an axonal component essential to trigger lesioninduced competence (or the presence of an axonal protein
preventing this triggering) (Coleman, 2005). Importantly, the
model would dissociate the induction of degeneration
competence from the initiation of the degeneration process
itself, both temporally and spatially. Although speculative at
this point, this dissociation would tie in well with the regulatory
properties of developmental, and possibly also disease-related
axon degeneration (see below).
Please cite this article in press as: Saxena, S., Caroni, P., Mechanisms of axon degeneration: From development to disease, Prog. Neurobiol.
(2007), doi:10.1016/j.pneurobio.2007.07.007
+ Models
PRONEU-845; No of Pages 18
10
S. Saxena, P. Caroni / Progress in Neurobiology xxx (2007) xxx–xxx
Research on Wallerian degeneration further uncovered a
requirement for the recruitment of phagocytes (e.g. activated
glia) in the degeneration process (MacDonald et al., 2006;
Awasaki et al., 2006). This recruitment was prevented in the
presence of WldS protein in neurons, or of UPS inhibitors,
suggesting that the recruitment of phagocytes by lesioned axons
depends on the axonal degeneration pathway. Significantly, this
pathway depends on the expression and accumulation of Draper
in the prospective phagocytes, a receptor also required to recruit
phagocytes to rapidly clear cellular remnants in apoptosis
(Zhou et al., 2001; Hoopfer et al., 2006; MacDonald et al.,
2006). These findings in Wallerian and developmental
degeneration (see below) thus establish a striking mechanistic
relationship between the late phases of active cell and axonal
self-destruction processes.
5.2. Developmental axon degeneration
The most extensive mechanistic insights about developmental axonal degeneration derive from genetic analyses of
MB g neuron axon pruning during Drosophila metamorphosis
(Luo and O’Leary, 2005). These studies have established that
competence for axonal (and dendritic) degeneration involves
transcriptional processes initiated by ecdysone/ecdysone
receptor-mediated signaling. While the ecdysone is secreted
into the environment of neurons, ecdysone receptor (a nuclear
receptor that functions together with retinoic receptor protein)
must be expressed in the neurons that will prune their axons
(Lee et al., 2000). Ecdysone/ecdysone receptor signaling
appears to mediate competence for axonal degeneration in all
neurons known to prune their axons during metamorphosis in
Drosophila (Luo and O’Leary, 2005; Schubiger et al., 1998).
The relevant neuronal products of ecdysone signaling in axon
degeneration have not been identified yet, and it is not yet clear
how they are eventually required for the initiation of
microtubule disassembly in pruning axons. Strikingly, overexpression of dominant-negative ecdysone receptor, which
prevents developmental pruning during metamorphosis, did not
influence lesion-induced axon degeneration of the very same
axons in developing Drosophila (Hoopfer et al., 2006). Even
more strikingly, overexpression of WldS protein, which
suppresses lesion-induced axon degeneration in Drosophila,
had no detectable effect on developmental axon pruning during
Drosophila metamorphosis (Hoopfer et al., 2006; Luo and
O’Leary, 2005). Furthermore, developmental axon degeneration in vertebrates was also not affected by overexpression of
WldS protein (Hoopfer et al., 2006; Luo and O’Leary, 2005).
Taken together, these results provide compelling experimental
evidence that axonal degeneration induced during development
and upon lesions is brought about by distinct mechanisms, and
that WldS specifically inhibits Wallerian, but not developmental
axon degeneration (Hoopfer et al., 2006; Luo and O’Leary,
2005). Interestingly, competence for developmental degeneration of cortical axon collaterals in vertebrates depends on the
expression and nuclear translocation of the transcription factor
Otx1 in these neurons (Weimann et al., 1999; Zhang et al.,
2002) suggesting that induction of competence through
transcriptional processes might be a general feature in
developmental axon degeneration. One possibility could be
that transcriptional switches during defined periods of
development provide a prerequisite to induce competence
for axon degeneration, and that additional signals, e.g. through
axon guidance molecules, trophic factor deprivation, or
signaling initiated in the absence of synchronized synaptic
activity might confer competence for degeneration. But this
still leaves a need to identify the mechanisms that initiate the
degeneration of specific axon branches. In an alternative, and
perhaps more parsimonious scenario, neurons might go through
prolonged periods of axon degeneration competence induced
by transcriptional switches during defined periods of development, and the additional more contingent signals would initiate
the pruning at specific locations along the axonal tree.
The analysis in Drosophila has further established that, like
in Wallerian degeneration, UPS activity is required for
developmental axon degeneration (Watts et al., 2003). Genetic
inactivation of ubiquitin activating enzyme (Uba1), or of the
proteasome itself prevented microtubule fragmentation, axon
blebbing and pruning, leading to a persistence of larval MB g
axons into adulthood (Watts et al., 2003, 2004). A requirement
for proteasome activity was also demonstrated for axon
degeneration induced upon NGF deprivation (MacInnis and
Campenot, 2005). These experiments thus suggest that UPSdependent degradation is a general prerequisite to confer
competence for (or initiate) axon degeneration in development
and upon lesions. Proteasome-mediated degradation might for
example remove a critical inhibitor of the axon degeneration
machinery. The fact that only a restricted set of ubiquitin ligases
was critically required to induce axon pruning in Drosophila
gives reasons to hope that the specific ubiquitination
substrate(s) involved in axon degeneration will be identified
soon, thus providing first insights about the axonal components
that control the degeneration process (Luo and O’Leary, 2005;
Watts et al., 2003).
In addition to implicating UPS activity for axon
degeneration, the analysis of developmental dendrite degeneration in Drosophila has revealed a specific requirement for
local caspase activity in this process (Kuo et al., 2006;
Williams et al., 2006). This discovery suggests the exciting
possibility that aspects of the molecular machinery involved
in cellular apoptosis might also be involved in local neurite
degeneration.
Finally, the analysis in Drosophila has established how glial
recruitment through Draper and ced-6 has a critical role in the
execution of developmental axon pruning (Hoopfer et al., 2006;
Watts et al., 2004; Awasaki and Ito, 2004; Awasaki et al., 2006).
The analysis demonstrated that in addition to its role in neurons,
ecdysone also signals through glial ecdysone receptors, and that
this signaling is essential for glial recruitment to blebbing axons
(Awasaki et al., 2006). In addition, however, glial recruitment
depended on a signal from the axons, which was suppressed in
the absence of UPS activity (Awasaki et al., 2006; Luo and
O’Leary, 2005). This signal seemed to only act over very short
ranges, and might be mediated through direct contact between
degenerating axons and glial processes; that might involve
Please cite this article in press as: Saxena, S., Caroni, P., Mechanisms of axon degeneration: From development to disease, Prog. Neurobiol.
(2007), doi:10.1016/j.pneurobio.2007.07.007
+ Models
PRONEU-845; No of Pages 18
S. Saxena, P. Caroni / Progress in Neurobiology xxx (2007) xxx–xxx
mechanisms analogous to ‘‘find me’’ and ‘‘eat me’’ signals in
apoptosis (Zhou et al., 2001; Awasaki et al., 2006; Luo and
O’Leary, 2005). The dual signaling to glia, from ecdysone and
from blebbing axons, is reminiscent of the separate signals
involved in inducing competence and initiating the actual
execution process in axon self-destruction. The signals initiate
morphological responses in glia reminiscent of a phagocytic
phenotype; in addition, they induce plasmalemmal accumulation of Draper in glial processes. In the absence of glial
recruitment, axons still exhibited disrupted microtubules and
blebbing, but the fragments failed to clear efficiently. As a
consequence, flies with disrupted Draper function in glia
exhibited a delay in axon clearance, with some axons being
maintained into adulthood (Awasaki et al., 2006; Hoopfer et al.,
2006). Glial recruitment is thus required to disrupt and
remove degenerating axons, but not to initiate the axon
degeneration process. The local signals through which
degenerating axons recruit glia have not been identified yet,
and their elucidation should be of general interest to understand
axon degeneration.
5.3. Mechanisms of axon degeneration in FALS
Much less is known about the mechanisms through which
axon degeneration is induced in SOD1 mouse models of FALS,
or in any other disease setting. In principle, one could imagine
that axons get highly damaged in disease, e.g. through near to
complete blockades of axonal transport, and that the induction
of degeneration ultimately resembles that induced by a
chemical lesion of microtubules (a WldS-sensitive process).
Disease would then resemble Wallerian degeneration, and
axonal pruning would presumably be sensitive to WldS protein.
However, although SOD1-linked models of FALS clearly
involve axonal degeneration as revealed by the abrupt and
characteristic fractionation and removal of particular axons at
defined times in disease (Fischer et al., 2004; Pun et al., 2006),
disease onset and disease progression in the FALS models are
essentially insensitive to WldS protein (Fischer et al., 2005;
Vande Velde et al., 2004). It seems unlikely that this failure by
WldS to protect in disease would be due to an age-related
decline in WldS effectiveness, because the protein does confer
strong axonal protection in lesioned axons at ages beyond 3
months, when SOD1(G93A) mice exhibit clinical signs of
disease. We will get back to this issue because WldS protein
does seem to confer some protection in other disease models,
but the conclusion is that degeneration-inducing pathways
different from those operating in Wallerian degeneration must
be operating in FALS. One possible scenario could be that
disease induces a switch in affected neurons comparable to the
switches leading to competence for axonal degeneration in
development, and that subsequent local signals lead to
peripheral synapse and axonal pruning. An alternative scenario
could be that there are additional, yet unidentified, ‘‘safety’’
pathways to locally disinhibit the axon degeneration machinery
in a pathological setting, and that these ‘‘safety’’ pathways (or
at least some of them) are not sensitive to the effects of WldS
protein.
11
But to what extent does axon degeneration in FALS
resemble its developmental and Wallerian counterparts? Could
it for example be that degeneration in FALS simply bypasses
the initial competence checkpoints, which are sensitive to
WldS? One way of addressing these questions is to determine
whether the degeneration process in FALS is sensitive to UPS
inhibitors. As discussed in Section 3.3 of this review, the
appearance of local axonal swellings and the early depletion of
synaptic vesicles from neuromuscular junctions in FALS (from
P38 to P45 in FF motoneurons of SOD1(G93A) mice) could
be prevented by local applications of CNTF into muscle
(Pun et al., 2006). In addition, the axonal pathology in FF
motoneurons was accompanied by upregulation of Bcl2A1
mRNA and protein, a member of the Bcl2 family of
antiapoptotic proteins, in corresponding motoneurons from
P40–42 on, and this upregulation was again suppressed by local
applications of CNTF to muscle (Pun et al., 2006). The CNTF
treatment was, however, only effective if started not later than
P42–43, and applications of CNTF from P45 on, i.e. 3–5 days
before pruning of FF motoneurons, did not reverse the synaptic
vesicle phenotype, or affect the time course of axon pruning
(Pun et al., 2006). In marked contrast, daily local applications
of proteasome inhibitor into muscle starting at P45 or P46
prevented FF axon pruning up to at least P60 (Pun et al., 2006).
These findings provided evidence that axon degeneration in
FALS depends on UPS activity, and demonstrated that UPS
blockade can prevent pruning at a time when CNTF is not
effective anymore. Significantly, and unlike local applications
of CNTF, local blockade of UPS from P38 on failed to influence
the appearance of axon pathology, the depletion of synaptic
vesicles from neuromuscular junctions, or the upregulation of
Bcl2A1 in FF motoneurons (Pun et al., 2006). This was not due
to some toxic effect of proteasome inhibitor, because the same
treatment did prevent FF axon pruning up to at least P54 (Pun
et al., 2006). These results have a number of interesting
implications (Fig. 2): The fact that blocking axonal pathology
and Bcl2A1 upregulation with CNTF also prevented later
pruning of FF motoneuron axons, suggests that an early
pathological process specifically in FF motoneurons and their
axons sets up the stage for UPS-sensitive triggering of axon
degeneration in FF motoneurons. Because UPS activity in axon
degeneration appears to be required at the time of competence
induction, these findings suggest that the process of axon
degeneration in FF motoneurons of SOD1(G93A) mice is
initiated between P46 and P48, and not before P46. Perhaps
most importantly, the results further suggest that CNTFsensitive axonal pathology and Bcl2A1 upregulation are not
part of the axon degeneration process (because they are not
affected by UPS inhibitors). Taken together, these results make
two important points about axon degeneration in FALS mice:
first, and like in developmental and Wallerian degeneration, the
axon degeneration process in this chronic neurodegenerative
disease model is triggered and completed within 2–3 days at
most, and second, the prominent disease-related axonal
pathology and the related cell body reactions likely lead up
to the triggering of axon degeneration, but they are not part of
the axon degeneration process itself.
Please cite this article in press as: Saxena, S., Caroni, P., Mechanisms of axon degeneration: From development to disease, Prog. Neurobiol.
(2007), doi:10.1016/j.pneurobio.2007.07.007
+ Models
PRONEU-845; No of Pages 18
12
S. Saxena, P. Caroni / Progress in Neurobiology xxx (2007) xxx–xxx
Fig. 2. Schematic representation of the processes leading up to the early peripheral degeneration of FF motoneuron axons in SOD1(G93A) FALS mice. Briefly, local
applications of CNTF to muscle from P38 on prevent axon pathology (synaptic vesicle stalling in axons, and synaptic vesicle depletion from neuromuscular
junctions), upregulation of Bcl2A1 in motoneurons and axon degeneration in FF motoneurons. In contrast, when the CNTF treatment is initiated at P45, it fails to
protect axons from pathology and degeneration, although it does prevent Bcl2A1 upregulation. Proteasome inhibitor (Lactacystin) applied from P38 or P45 on
prevents axon degeneration without alleviation of axon pathology or Bcl2A1 upregulation. These findings suggest that CNTF-sensitive axon pathology leads up to
axon degeneration competence at P46, but that the axon pathology is not part of the (Lactacystin-sensitive) degeneration process. Green: synaptic vesicles; red:
Bcl2A1.
6. Shared mechanisms and outstanding issues
The comparative analysis of axon degeneration mechanisms
in development, upon lesions and in a neurodegenerative
disease model has not only revealed several shared features, but
also a number of aspects unique to the different settings (Fig. 1).
The unique aspects involve mechanisms that confer competence for self-destruction, and those that initiate the selfdestruction of defined sections of axons. Shared aspects include
the central role of an early competence conferring process, the
role of UPS, the unique sequence of structural and anatomical
events, and the role of glia (although the latter has not yet been
addressed in neurodegeneration) (Figs. 1 and 2). We summarize
these comparisons by proposing a model of degeneration
regulation that takes into account the findings described in the
previous sections of this review (Fig. 3). Although it is
speculative at this point, the model might be helpful in devising
future experiments. The model has two main features: First,
distinct regulatory pathways together with the shared requirement for UPS activity lead to axonal competence to initiate
Fig. 3. Proposed model of how the induction of axon degeneration might involve setting-specific regulatory processes, and a shared set of execution reactions. The
induction of a neuronal competence state for axon degeneration, and the local triggering of degeneration at defined axon segments would involve context-specific
regulation. In contrast, neuronal competence itself might reflect a more general de-repression state that could subsequently lead to local axon degeneration through
distinct local mechanisms. Finally, upon local triggering, the degeneration process itself involves a common pathway of microtubule disassembly, axonal swellings,
axon fragmentation and removal of the remnants. UPS activity is a general requirement to induce competence for axon degeneration. In contrast, WldS protein
suppresses competence induction in Wallerian, but not developmental degeneration. Degeneration induced in disease can be sensitive (e.g. pmn/pmn mice) or
insensitive (e.g. FALS mice) to inhibition by WldS.
Please cite this article in press as: Saxena, S., Caroni, P., Mechanisms of axon degeneration: From development to disease, Prog. Neurobiol.
(2007), doi:10.1016/j.pneurobio.2007.07.007
+ Models
PRONEU-845; No of Pages 18
S. Saxena, P. Caroni / Progress in Neurobiology xxx (2007) xxx–xxx
self-destruction. Second, separate processes subsequently
initiate self-destruction of defined axonal segments. The selfdestruction process itself then involves the shared sequence of
microtubule disassembly, axon blebbing, axon fragmentation,
and removal of the fragments by recruited phagocytes. In this
model, WldS protein specifically interferes with competence
induction upon lesions. The various developmental switches
could, for example, lead to changes in the expression of critical
components that control the inhibition of degeneration in the
axon. The pathways leading up to competence might converge
onto a UPS-sensitive step, but that step might also operate in
parallel to those pathways. Subsequent to the acquisition of
competence for degeneration, the model suggests that the
various local triggering pathways that operate in the absence of
lesions (e.g. activity patterns, neurotrophin deprivation, axon
guidance molecules) might have a role comparable to ion
leakage in Wallerian degeneration. In this way, the various local
signaling pathways might define the axon segments that initiate
self-destruction; these pathways could, for example, all
converge onto elevated levels of intra-axonal calcium. Finally,
the model suggests that the different competence and triggering
pathways might be interchangeable, meaning that competence
induced through any pathway might be compatible with any
type of triggering mechanism. This seems to us a parsimonious
assumption, but it might well be that axons have evolved
distinct competence-triggering pathways to ensure more tight
and specific regulation of the axon degeneration process.
Although research into axon degeneration has progressed
dramatically during the last 3–4 years, many important aspects of
the axon degeneration process remain unclear. Some of the more
fascinating questions are intrinsic to developmental neuroscience or to basic studies of neurodegenerative diseases.
However, addressing those questions will also depend on a better
mechanistic understanding of the axon degeneration process
itself. One major unresolved issue involves the molecular
mechanisms that control the degeneration process in the axon.
Since this likely involves complex signal transduction pathways,
including multiple pathways to confer competence, and
checkpoints to initiate the process, these questions might be
best addressed in a genetic model system, e.g. Drosophila (Lee
and Luo, 1999; Luo and O’Leary, 2005). The substrates of the
critical ubiquitin ligases that must be degraded by the proteasome
machinery in order to confer competence for axon degeneration
could be one specific focus for investigations. Investigating the
mechanisms through which WldS protein prevents lesioninduced degeneration will likely lead to important complementary information. Since those studies would eventually have to
identify critical downstream targets of WldS regulation in the
axon, they might again profit from the powerful forward ands
reverse genetics approaches available in Drosophila (Lee and
Luo, 1999). The identification of any positive upstream regulator
of the degeneration pathway in the axon will hopefully provide
much needed tools to dissociate competence and initiation of the
degeneration process experimentally. Such dissociations will be
important in order to investigate the mechanisms that trigger the
degeneration process in development and in disease models. In
both settings, one major open question is how defined sections of
13
the axon are targeted for degeneration. Answers to this question
should open up new avenues to investigate principles of
developmental pruning, and to identify ways of protecting
synapses and axons in disease.
7. Pathology versus axon degeneration in
neurodegenerative diseases
7.1. ‘‘Dying-back’’ diseases
If one defines ‘‘dying-back’’ diseases as those, where
progressive and substantial losses of synapses and axons precede
neuronal losses, then there is evidence that many neurodegenerative diseases exhibit ‘‘dying-back’’ patterns of losses. Welldocumented cases include motoneuron diseases, peripheral
neuropathies, spinocerebellar ataxias, nutritional disorders and
AIDS (Berger and Schaumburg, 1995; Bjartmar et al., 2003;
Griffin et al., 1995; Martin et al., 2002; Pun et al., 2006). In
addition, early losses of synapses have been reported for
Alzheimer’s, Parkinson’s, Huntington’s and Prion’s diseases
(Bommel et al., 2002; Coleman and Yao, 2003; Gunawardena
and Goldstein, 2005; Gunawardena et al., 2003; Li et al., 2001;
Tsai et al., 2004). The extent to which ‘‘dying-back’’ processes
are the predominant cause of functional losses in those diseases is
in most cases less clear. Even less clear is the extent to which
‘‘dying-back’’ reflects axon degeneration processes. Although it
does not seem unlikely that axon degeneration is involved in
several cases, the possibility exists that just as cells can die
through apoptosis or necrosis, axons might be lost through
degeneration or through more passive, and potentially more
toxic, forms of increasing pathology and dystrophy. For example,
dystrophic neurites associated with extracellular amyloid
plaques in Alzheimer’s disease definitely seem to have escaped
degeneration. On the other hand, there are several neurodegenerative diseases in which there is a strong case for the existence of
axon degeneration processes (Bommel et al., 2002; Martin et al.,
2002; Pun et al., 2006). Current criteria are of two kinds: rare
direct evidence of axon degeneration processes such as in FALS
(Pun et al., 2006) and in forms of early-onset motoneuron disease
(e.g. pmn/pmn mice, Sagot et al., 1995), and evidence that
synapse and/or axon losses are reduced in the presence of WldS
protein (Hasbani and O’Malley, 2006; Sagot et al., 1996; Sajadi
et al., 2004). A comparison between pmn/pmn mice and the
FALS models illustrates how alleviation of pathology by WldS
protein does not provide a straightforward discrimination
between diseases involving axon degeneration and those that
do not. Thus, although in pmn/pmn mice the patterns of axon loss
have not been investigated in detail comparable to that in FALS
mice, these mice do exhibit abrupt synapse and axon losses
suggestive of axon degeneration (Sagot et al., 1995). However,
while FALS mice are not sensitive to WldS (Fischer et al., 2005;
Vande Velde et al., 2004), pmn/pmn mice are (Ferri et al., 2003).
The reasons for this discrepancy are not clear at the moment, but
one possibility is that the molecular pathways leading to axon
degeneration in the two models are different. A detailed
investigation of ‘‘dying-back’’ diseases with respect to patterns
of axon degeneration and sensitivity to WldS protein should
Please cite this article in press as: Saxena, S., Caroni, P., Mechanisms of axon degeneration: From development to disease, Prog. Neurobiol.
(2007), doi:10.1016/j.pneurobio.2007.07.007
+ Models
PRONEU-845; No of Pages 18
14
S. Saxena, P. Caroni / Progress in Neurobiology xxx (2007) xxx–xxx
provide important insights into the mechanisms leading to axon
losses in these diseases. The value of these studies will likely be
enhanced by a detailed anatomical analysis of the synapse and
axon loss processes, and by progress in understanding the
mechanisms through which WldS protein prevents Wallerian and
other forms of axonal degeneration.
7.2. Axonal swellings and spheroids
A discussion of the axonal pathologies associated with
neurodegenerative diseases is clearly beyond the scope of this
review. However, in view of the possible relationships between
defects in axonal transport, axonal swellings, and reported effects
of WldS protein in several neurodegenerative disease models, a
brief discussion of how those specific issues might relate to
axonal degeneration processes seems important. A compelling
body of experimental evidence supports the notion that defects in
axonal transport are an early manifestation of pathology in most
neurodegenerative diseases (Collard et al., 1995; Gunawardena
and Goldstein, 2001, 2004, 2005; Gunawardena et al., 2003;
Sagot et al., 1998; Stokin and Goldstein, 2006; Stokin et al.,
2005; Williamson and Cleveland, 1999; Zhang et al., 1997).
Although the causal relationships between these early transport
defects and disease onset and progression are not clear yet, it
seems likely that these defects will turn out to have causative
roles in disease. To a large extent, this expectation is based on the
fact that in several cases loss-of-function mutations in
intracellular transport or membrane trafficking components
have now been related to familial forms of neurodegenerative
diseases (Bommel et al., 2002; Eaton et al., 2002; Garcia-Fresco
et al., 2006; Gotz et al., 2006; Gunawardena and Goldstein, 2004;
LaMonte et al., 2002; Martin et al., 2002; Stokin and Goldstein,
2006; Xia et al., 2003). There are, furthermore, reasons to believe
that the appearance of swellings and spheroids in axons could be
linked to defects in axonal transport, possibly as a macroscopic
manifestation of ‘‘roadblocks’’ (the swellings typically accumulate the fast axonal transport component amyloid precursor
protein (APP)), or perhaps even as a mechanism to alleviate
‘‘roadblocks’’ (Gunawardena and Goldstein, 2001, 2004, 2005;
Gunawardena et al., 2003). However, the main issue here is
whether and when axonal pathologies such as swellings and
larger spheroids are an aspect of the axonal degeneration process
in disease, whether they represent local pathologies that can
trigger axon degeneration, or whether most of these phenomena
are not causally related to axon degeneration in disease (Delio
et al., 1992; Tsai et al., 2004). These issues can be confusing
because axon degeneration does involve axonal swellings, and
because WldS protein has been shown to reduce swellings and
spheroids in some disease models (Mi et al., 2005). However, and
as discussed in a previous section of this review, although axonal
swellings in FALS mice might be part of the processes that lead
up to axon degeneration, they are clearly not a manifestation of
the degeneration process itself (Pun et al., 2006). Most likely,
therefore, there are several distinct forms of axonal swellings,
and only a certain type is diagnostic of axonal degeneration.
It seems to us reasonable to assume that, wherever it is
associated with disease, axonal degeneration will invariably
exhibit some of its characteristic hallmarks, including a rapid
onset (within a few days at most), and a rapid elimination of
axonal remnants. That clearly distinguishes axon degeneration
from the gradual accumulation of various forms of axonal
pathologies, swellings and spheroids, which can last for months
and even years (Gotz et al., 2006; Mi et al., 2005; Pirozzi et al.,
2006). This distinction is further underlined by the fact that in
some disease models distinct axonal pathologies can progress
for many months prior to the first evidence of synapse or axon
losses. Taking these considerations into account it is therefore
particularly puzzling how WldS protein nevertheless can
attenuate the development of these pathologies (Mi et al.,
2005; Samsam et al., 2003). One possibility, which seems
unlikely to us, is that axon degeneration associated with some
neurodegenerative diseases is an unusually protracted process.
An alternative possibility is that the fusion protein WldS has
additional protective effects not specifically related to axon
degeneration.
8. Outlook
Main recent developments in studying axon degeneration
include the growing appreciation that this is a widespread and
tightly regulated axon (and dendrite) self-destruction process
reminiscent in its logics and significance to apoptosis, and the
innovations in genetic tools and molecular live-imaging
techniques that have allowed first glimpses of how these
processes happen in vivo. A number of basic issues about
mechanisms regulating axon degeneration remain open and
provide exciting challenges for future research. These include
the specific mechanisms leading up to competence for axon
degeneration in its different settings, and the molecular
mechanisms of this regulation in neuronal cell bodies and at
axons. A second set of challenges involves elucidating the
mechanisms that determine which specific segments of axons
(and dendrites) are targeted for degeneration in the different
settings. Progress on these mechanistic issues will likely fuel
significant advances in developmental neuroscience, in studies
of plasticity in the adult, and in studies of neurodegenerative
diseases. Equally exciting advances can be expected from
detailed longitudinal live-imaging studies of axon degeneration
processes during development and in disease. The few initial in
vivo imaging studies of Wallerian degeneration, motoneuron
disease and Alzheimer’s disease have given us a taste of things
to come (Kerschensteiner et al., 2005; Schaefer et al., 2005;
Tsai et al., 2004). One can safely assume that combinations of
high-resolution longitudinal analyses in vivo and mechanistic
insights of degeneration processes at the molecular level will
have a major impact on studies of development, plasticity and
disease in the nervous system.
References
Alsina, B., Vu, T., Cohen-Cory, S., 2001. Visualizing synapse formation in
arborizing optic axons in vivo: dynamics and modulation by BDNF. Nat.
Neurosci. 4, 1093–1101.
Please cite this article in press as: Saxena, S., Caroni, P., Mechanisms of axon degeneration: From development to disease, Prog. Neurobiol.
(2007), doi:10.1016/j.pneurobio.2007.07.007
+ Models
PRONEU-845; No of Pages 18
S. Saxena, P. Caroni / Progress in Neurobiology xxx (2007) xxx–xxx
Araki, T., Sasaki, Y., Milbrandt, J., 2004. Increased nuclear NAD biosynthesis
and SIRT1 activation prevent axonal degeneration. Science 305, 1010–
1013.
Awasaki, T., Ito, K., 2004. Engulfing action of glial cells is required for
programmed axon pruning during Drosophila metamorphosis. Curr. Biol.
14, 668–677.
Awasaki, T., Tatsumi, R., Takahashi, K., Arai, K., Nakanishi, Y., Ueda, R., Ito,
K., 2006. Essential role of the apoptotic cell engulfment genes draper and
ced-6 in programmed axon pruning during Drosophila metamorphosis.
Neuron 50, 855–867.
Baas, P.W., Qiang, L., 2005. Neuronal microtubules: when the MAP is the
roadblock. Trends Cell. Biol. 15, 183–187.
Bagri, A., Cheng, H.J., Yaron, A., Pleasure, S.J., Tessier-Lavigne, M., 2003.
Stereotyped pruning of long hippocampal axon branches triggered by
retraction inducers of the semaphorin family. Cell 113, 285–299.
Balice-Gordon, R.J., Lichtman, J.W., 1993. In vivo observations of preand postsynaptic changes during the transition from multiple to single
innervation at developing neuromuscular junctions. J. Neurosci. 13, 834–
855.
Bansal, A., Singer, J.H., Hwang, B.J., Xu, W., Beaudet, A., Feller, M.B., 2000.
Mice lacking specific nicotinic acetylcholine receptor subunits exhibit
dramatically altered spontaneous activity patterns and reveal a limited role
for retinal waves in forming ON and OFF circuits in the inner retina. J.
Neurosci. 20, 7672–7681.
Bastmeyer, M., O’Leary, D.D., 1996. Dynamics of target recognition by
interstitial axon branching along developing cortical axons. J. Neurosci.
16, 1450–1459.
Beirowski, B., Berek, L., Adalbert, R., Wagner, D., Grumme, D.S., Addicks, K.,
Ribchester, R.R., Coleman, M.P., 2004. Quantitative and qualitative analysis of Wallerian degeneration using restricted axonal labelling in YFP-H
mice. J. Neurosci. Methods 134, 23–35.
Berger, A.R., Schaumburg, H.H., 1995. Human peripheral nerve disease
(peripheral neuropathies). In: Waxman, S.G., Jeffery, D.K., Stys, P.K.
(Eds.), The Axon: Structure, Function and Pathophysiolgy. Oxford University Press, New York, pp. 648–660.
Bernstein, M., Lichtman, J.W., 1999. Axonal atrophy: the retraction reaction.
Curr. Opin. Neurobiol. 9, 364–370.
Billuart, P., Winter, C.G., Maresh, A., Zhao, X., Luo, L., 2001. Regulating axon
branch stability: the role of p190 RhoGAP in repressing a retraction
signaling pathway. Cell 107, 195–207.
Bishop, D.L., Misgeld, T., Walsh, M.K., Gan, W.B., Lichtman, J.W., 2004.
Axon branch removal at developing synapses by axosome shedding. Neuron
44, 651–661.
Bito, H., Furuyashiki, T., Ishihara, H., Shibasaki, Y., Ohashi, K., Mizuno, K.,
Maekawa, M., Ishizaki, T., Narumiya, S., 2000. A critical role for a Rhoassociated kinase, p160ROCK, in determining axon outgrowth in mammalian CNS neurons. Neuron 26, 431–441.
Bjartmar, C., Wujek, J.R., Trapp, B.D., 2003. Axonal loss in the pathology of
MS: consequences for understanding the progressive phase of the disease. J.
Neurol. Sci. 206, 165–171.
Boillee, S., Vande Velde, C., Cleveland, D.W., 2006a. ALS: a disease of motor
neurons and their nonneuronal neighbors. Neuron 52, 39–59.
Boillee, S., Yamanaka, K., Lobsiger, C.S., Copeland, N.G., Jenkins, N.A.,
Kassiotis, G., Kollias, G., Cleveland, D.W., 2006b. Onset and progression in
inherited ALS determined by motor neurons and microglia. Science 312,
1389–1392.
Bommel, H., Xie, G., Rossoll, W., Wiese, S., Jablonka, S., Boehm, T., Sendtner,
M., 2002. Missense mutation in the tubulin-specific chaperone E (Tbce)
gene in the mouse mutant progressive motor neuronopathy, a model of
human motoneuron disease. J. Cell Biol. 159, 563–569.
Bruck, W., 2005. The pathology of multiple sclerosis is the result of focal
inflammatory demyelination with axonal damage. J. Neurol. 252 (Suppl 5),
v3–v9.
Bruijn, L.I., Becher, M.W., Lee, M.K., Anderson, K.L., Jenkins, N.A., Copeland, N.G., Sisodia, S.S., Rothstein, J.D., Borchelt, D.R., Price, D.L.,
Cleveland, D.W., 1997. ALS-linked SOD1 mutant G85R mediates damage
to astrocytes and promotes rapidly progressive disease with SOD1-containing inclusions. Neuron 18, 327–338.
15
Buffelli, M., Burgess, R.W., Feng, G., Lobe, C.G., Lichtman, J.W., Sanes, J.R.,
2003. Genetic evidence that relative synaptic efficacy biases the outcome of
synaptic competition. Nature 424, 430–434.
Burke, R.E., 1994. Physiology of motor units. In: Engel, A.G., FranziniArmstrong, C. (Eds.), Myology. McGraw-Hill, New York, pp. 464–484.
Campenot, R.B., 1982. Development of sympathetic neurons in compartmentalized cultures. II. Local control of neurite survival by nerve growth factor.
Dev. Biol. 93, 13–21.
Causing, C.G., Gloster, A., Aloyz, R., Bamji, S.X., Chang, E., Fawcett, J.,
Kuchel, G., Miller, F.D., 1997. Synaptic innervation density is regulated by
neuron-derived BDNF. Neuron 18, 257–267.
Cavanagh, J.B., 1964. The significance of the ‘‘dying back’’ process in
experimental and human neurological disease. Int. Rev. Exp. Pathol. 3,
219–267.
Chen, C., Regehr, W.G., 2000. Developmental remodeling of the retinogeniculate synapse. Neuron 28, 955–966.
Clement, A.M., Nguyen, M.D., Roberts, E.A., Garcia, M.L., Boillee, S., Rule,
M., McMahon, A.P., Doucette, W., Siwek, D., Ferrante, R.J., Brown Jr.,
R.H., Julien, J.P., Goldstein, L.S., Cleveland, D.W., 2003. Wild-type nonneuronal cells extend survival of SOD1 mutant motor neurons in ALS mice.
Science 302, 113–117.
Coleman, M., 2005. Axon degeneration mechanisms: commonality amid
diversity. Nat. Rev. Neurosci. 6, 889–898.
Coleman, P.D., Yao, P.J., 2003. Synaptic slaughter in Alzheimer’s disease.
Neurobiol. Aging 24, 1023–1027.
Collard, J.F., Cote, F., Julien, J.P., 1995. Defective axonal transport in a
transgenic mouse model of amyotrophic lateral sclerosis. Nature 375,
61–64.
Conforti, L., Tarlton, A., Mack, T.G., Mi, W., Buckmaster, E.A., Wagner, D.,
Perry, V.H., Coleman, M.P., 2000. A Ufd2/D4Cole1e chimeric protein and
overexpression of Rbp7 in the slow Wallerian degeneration (WldS) mouse.
Proc. Natl. Acad. Sci. U.S.A. 97, 11377–11382.
Conforti, L., Fang, G., Beirowski, B., Wang, M.S., Sorci, L., Asress, S.,
Adalbert, R., Silva, A., Bridge, K., Huang, X.P., Magni, G., Glass, J.D.,
Coleman, M.P., 2007. NAD(+) and axon degeneration revisited: Nmnat1
cannot substitute for Wld(S) to delay Wallerian degeneration. Cell Death
Differ. 14, 116–127.
De Paola, V., Holtmaat, A., Knott, G., Song, S., Wilbrecht, L., Caroni, P.,
Svoboda, K., 2006. Cell type-specific structural plasticity of axonal
branches and boutons in the adult neocortex. Neuron 49, 861–875.
Debski, E.A., Cline, H.T., 2002. Activity-dependent mapping in the retinotectal
projection. Curr. Opin. Neurobiol. 12, 93–99.
Delio, D.A., Fiori, M.G., Lowndes, H.E., 1992. Motor unit function during
evolution of proximal axonal swellings. J. Neurol. Sci. 109, 30–40.
DiAntonio, A., Haghighi, A.P., Portman, S.L., Lee, J.D., Amaranto, A.M.,
Goodman, C.S., 2001. Ubiquitination-dependent mechanisms regulate
synaptic growth and function. Nature 412, 449–452.
Eaton, B.A., Fetter, R.D., Davis, G.W., 2002. Dynactin is necessary for synapse
stabilization. Neuron 34, 729–741.
Eckenhoff, M.F., Pysh, J.J., 1979. Double-walled coated vesicle formation:
evidence for massive and transient conjugate internalization of plasma
membranes during cerebellar development. J. Neurocytol. 8, 623–638.
Feller, M.B., 2002. The role of nAChR-mediated spontaneous retinal activity in
visual system development. J. Neurobiol. 53, 556–567.
Ferri, A., Sanes, J.R., Coleman, M.P., Cunningham, J.M., Kato, A.C., 2003.
Inhibiting axon degeneration and synapse loss attenuates apoptosis and
disease progression in a mouse model of motoneuron disease. Curr. Biol. 13,
669–673.
Finn, J.T., Weil, M., Archer, F., Siman, R., Srinivasan, A., Raff, M.C., 2000.
Evidence that Wallerian degeneration and localized axon degeneration
induced by local neurotrophin deprivation do not involve caspases. J.
Neurosci. 20, 1333–1341.
Fischer, L.R., Culver, D.G., Davis, A.A., Tennant, P., Wang, M., Coleman, M.,
Asress, S., Adalbert, R., Alexander, G.M., Glass, J.D., 2005. The WldS gene
modestly prolongs survival in the SOD1G93A fALS mouse. Neurobiol. Dis.
19, 293–300.
Fischer, L.R., Culver, D.G., Tennant, P., Davis, A.A., Wang, M., CastellanoSanchez, A., Khan, J., Polak, M.A., Glass, J.D., 2004. Amyotrophic lateral
Please cite this article in press as: Saxena, S., Caroni, P., Mechanisms of axon degeneration: From development to disease, Prog. Neurobiol.
(2007), doi:10.1016/j.pneurobio.2007.07.007
+ Models
PRONEU-845; No of Pages 18
16
S. Saxena, P. Caroni / Progress in Neurobiology xxx (2007) xxx–xxx
sclerosis is a distal axonopathy: evidence in mice and man. Exp. Neurol.
185, 232–240.
Frey, D., Schneider, C., Xu, L., Borg, J., Spooren, W., Caroni, P., 2000. Early
and selective loss of neuromuscular synapse subtypes with low sprouting
competence in motoneuron diseases. J. Neurosci. 20, 2534–2542.
Galimberti, I., Gogolla, N., Alberi, S., Santos, A.F., Muller, D., Caroni, P., 2006.
Long-term rearrangements of hippocampal mossy fiber terminal connectivity in the adult regulated by experience. Neuron 50, 749–763.
Gao, P.P., Yue, Y., Cerretti, D.P., Dreyfus, C., Zhou, R., 1999. Ephrin-dependent
growth and pruning of hippocampal axons. Proc. Natl. Acad. Sci. U.S.A. 96,
4073–4077.
Garcia-Fresco, G.P., Sousa, A.D., Pillai, A.M., Moy, S.S., Crawley, J.N.,
Tessarollo, L., Dupree, J.L., Bhat, M.A., 2006. Disruption of axo-glial
junctions causes cytoskeletal disorganization and degeneration of Purkinje
neuron axons. Proc. Natl. Acad. Sci. U.S.A. 103, 5137–5142.
Gillingwater, T.H., Wishart, T.M., Chen, P.E., Haley, J.E., Robertson, K.,
MacDonald, S.H., Middleton, S., Wawrowski, K., Shipston, M.J., Melmed,
S., Wyllie, D.J., Skehel, P.A., Coleman, M.P., Ribchester, R.R., 2006. The
neuroprotective WldS gene regulates expression of PTTG1 and erythroid
differentiation regulator 1-like gene in mice and human cells. Hum. Mol.
Genet. 15, 625–635.
Glass, J.D., Brushart, T.M., George, E.B., Griffin, J.W., 1993. Prolonged
survival of transected nerve fibres in C57BL/Ola mice is an intrinsic
characteristic of the axon. J. Neurocytol. 22, 311–321.
Gotz, J., Ittner, L.M., Kins, S., 2006. Do axonal defects in tau and amyloid
precursor protein transgenic animals model axonopathy in Alzheimer’s
disease? J. Neurochem. 98, 993–1006.
Gould, T.W., Buss, R.R., Vinsant, S., Prevette, D., Sun, W., Knudson, C.M.,
Milligan, C.E., Oppenheim, R.W., 2006. Complete dissociation of motor
neuron death from motor dysfunction by Bax deletion in a mouse model of
ALS. J. Neurosci. 26, 8774–8786.
Griffin, J.W., Georges, E.B., Hsieh, S.T., Glass, J.D., 1995. Axonal degeneration and disorders of the axonal cytoskeleton. In: Waxman, S.G., Jeffery,
D.K., Stys, P.K. (Eds.), The Axon: Structure, Function and Pathophysiology. Oxford University Press, New York, pp. 375–390.
Gunawardena, S., Goldstein, L.S., 2001. Disruption of axonal transport and
neuronal viability by amyloid precursor protein mutations in Drosophila.
Neuron 32, 389–401.
Gunawardena, S., Goldstein, L.S., 2004. Cargo-carrying motor vehicles on the
neuronal highway: transport pathways and neurodegenerative disease. J.
Neurobiol. 58, 258–271.
Gunawardena, S., Goldstein, L.S., 2005. Polyglutamine diseases and transport
problems: deadly traffic jams on neuronal highways. Arch. Neurol. 62, 46–
51.
Gunawardena, S., Her, L.S., Brusch, R.G., Laymon, R.A., Niesman, I.R.,
Gordesky-Gold, B., Sintasath, L., Bonini, N.M., Goldstein, L.S., 2003.
Disruption of axonal transport by loss of huntingtin or expression of
pathogenic polyQ proteins in Drosophila. Neuron 40, 25–40.
Gurney, M.E., Pu, H., Chiu, A.Y., Dal Canto, M.C., Polchow, C.Y., Alexander,
D.D., Caliendo, J., Hentati, A., Kwon, Y.W., Deng, H.X., et al., 1994. Motor
neuron degeneration in mice that express a human Cu,Zn superoxide
dismutase mutation. Science 264, 1772–1775.
Hasbani, D.M., O’Malley, K.L., 2006. Wld(S) mice are protected against the
Parkinsonian mimetic MPTP. Exp. Neurol. 202, 93–99.
Hindges, R., McLaughlin, T., Genoud, N., Henkemeyer, M., O’Leary, D.D.,
2002. EphB forward signaling controls directional branch extension and
arborization required for dorsal-ventral retinotopic mapping. Neuron 35,
475–487.
Hoopfer, E.D., McLaughlin, T., Watts, R.J., Schuldiner, O., O’Leary, D.D., Luo,
L., 2006. Wlds protection distinguishes axon degeneration following injury
from naturally occurring developmental pruning. Neuron 50, 883–895.
Innocenti, G.M., 1981. Growth and reshaping of axons in the establishment of
visual callosal connections. Science 212, 824–827.
Innocenti, G.M., Clarke, S., Koppel, H., 1983. Transitory macrophages in the
white matter of the developing visual cortex. II. Development and relations
with axonal pathways. Brain Res. 313, 55–66.
Innocenti, G.M., 1995. Exuberant development of connections, and its possible
permissive role in cortical evolution. Trends Neurosci. 18, 397–402.
Innocenti, G.M., Kiper, D.C., Knyazeva, M.G., Deonna, T.W., 1999. On nature
and limits of cortical developmental plasticity after an early lesion, in a
child. Restor. Neurol. Neurosci. 15, 219–227.
Kasthuri, N., Lichtman, J.W., 2003. The role of neuronal identity in synaptic
competition. Nature 424, 426–430.
Kasthuri, N., Lichtman, J.W., 2004. Structural dynamics of synapses in living
animals. Curr. Opin. Neurobiol. 14, 105–111.
Katz, L.C., Shatz, C.J., 1996. Synaptic activity and the construction of cortical
circuits. Science 274, 1133–1138.
Kerschensteiner, M., Schwab, M.E., Lichtman, J.W., Misgeld, T., 2005. In vivo
imaging of axonal degeneration and regeneration in the injured spinal cord.
Nat. Med. 11, 572–577.
King, M.E., Kan, H.M., Baas, P.W., Erisir, A., Glabe, C.G., Bloom, G.S., 2006.
Tau-dependent microtubule disassembly initiated by prefibrillar {beta}amyloid. J. Cell Biol. 175, 541–546.
Kozma, R., Sarner, S., Ahmed, S., Lim, L., 1997. Rho family GTPases and
neuronal growth cone remodelling: relationship between increased complexity induced by Cdc42Hs, Rac1, and acetylcholine and collapse induced
by RhoA and lysophosphatidic acid. Mol. Cell. Biol. 17, 1201–1211.
Kuo, C.T., Zhu, S., Younger, S., Jan, L.Y., Jan, Y.N., 2006. Identification of E2/
E3 ubiquitinating enzymes and caspase activity regulating Drosophila
sensory neuron dendrite pruning. Neuron 51, 283–290.
LaMonte, B.H., Wallace, K.E., Holloway, B.A., Shelly, S.S., Ascano, J., Tokito,
M., Van Winkle, T., Howland, D.S., Holzbaur, E.L., 2002. Disruption of
dynein/dynactin inhibits axonal transport in motor neurons causing lateonset progressive degeneration. Neuron 34, 715–727.
Lamprecht, R., Farb, C.R., LeDoux, J.E., 2002. Fear memory formation
involves p190 RhoGAP and ROCK proteins through a GRB2-mediated
complex. Neuron 36, 727–738.
Lee, T., Luo, L., 1999. Mosaic analysis with a repressible cell marker for studies
of gene function in neuronal morphogenesis. Neuron 22, 451–461.
Lee, T., Marticke, S., Sung, C., Robinow, S., Luo, L., 2000. Cell-autonomous
requirement of the USP/EcR-B ecdysone receptor for mushroom body
neuronal remodeling in Drosophila. Neuron 28, 807–818.
LeMaster, A.M., Krimm, R.F., Davis, B.M., Noel, T., Forbes, M.E., Johnson,
J.E., Albers, K.M., 1999. Overexpression of brain-derived neurotrophic
factor enhances sensory innervation and selectively increases neuron number. J. Neurosci. 19, 5919–5931.
Levine, R.B., Morton, D.B., Restifo, L.L., 1995. Remodeling of the insect
nervous system. Curr. Opin. Neurobiol. 5, 28–35.
Li, H., Li, S.H., Yu, Z.X., Shelbourne, P., Li, X.J., 2001. Huntingtin aggregateassociated axonal degeneration is an early pathological event in Huntington’s disease mice. J. Neurosci. 21, 8473–8481.
Libby, R.T., Li, Y., Savinova, O.V., Barter, J., Smith, R.S., Nickells, R.W., John,
S.W., 2005. Susceptibility to neurodegeneration in a glaucoma is modified
by Bax gene dosage. PLoS Genet. 1, 17–26.
Lichtman, J.W., Colman, H., 2000. Synapse elimination and indelible memory.
Neuron 25, 269–278.
Linker, R.A., Sendtner, M., Gold, R., 2005. Mechanisms of axonal degeneration
in EAE–lessons from CNTF and MHC I knockout mice. J. Neurol. Sci. 233,
167–172.
Lino, M.M., Schneider, C., Caroni, P., 2002. Accumulation of SOD1 mutants in
postnatal motoneurons does not cause motoneuron pathology or motoneuron disease. J. Neurosci. 22, 4825–4832.
Lobsiger, C.S., Garcia, M.L., Ward, C.M., Cleveland, D.W., 2005. Altered
axonal architecture by removal of the heavily phosphorylated neurofilament
tail domains strongly slows superoxide dismutase 1 mutant-mediated ALS.
Proc. Natl. Acad. Sci. U.S.A. 102, 10351–10356.
Lohof, A.M., Delhaye-Bouchaud, N., Mariani, J., 1996. Synapse elimination in
the central nervous system: functional significance and cellular mechanisms. Rev. Neurosci. 7, 85–101.
Low, L.K., Cheng, H.J., 2005. A little nip and tuck: axon refinement during
development and axonal injury. Curr. Opin. Neurobiol. 15, 549–556.
Luo, L., O’Leary, D.D., 2005. Axon retraction and degeneration in development
and disease. Annu. Rev. Neurosci. 28, 127–156.
MacDonald, J.M., Beach, M.G., Porpiglia, E., Sheehan, A.E., Watts, R.J.,
Freeman, M.R., 2006. The Drosophila cell corpse engulfment receptor
Draper mediates glial clearance of severed axons. Neuron 50, 869–881.
Please cite this article in press as: Saxena, S., Caroni, P., Mechanisms of axon degeneration: From development to disease, Prog. Neurobiol.
(2007), doi:10.1016/j.pneurobio.2007.07.007
+ Models
PRONEU-845; No of Pages 18
S. Saxena, P. Caroni / Progress in Neurobiology xxx (2007) xxx–xxx
MacInnis, B.L., Campenot, R.B., 2005. Regulation of Wallerian degeneration
and nerve growth factor withdrawal-induced pruning of axons of sympathetic neurons by the proteasome and the MEK/Erk pathway. Mol. Cell.
Neurosci. 28, 430–439.
Mack, T.G., Reiner, M., Beirowski, B., Mi, W., Emanuelli, M., Wagner, D.,
Thomson, D., Gillingwater, T., Court, F., Conforti, L., Fernando, F.S., Tarlton,
A., Andressen, C., Addicks, K., Magni, G., Ribchester, R.R., Perry, V.H.,
Coleman, M.P., 2001. Wallerian degeneration of injured axons and synapses is
delayed by a Ube4b/Nmnat chimeric gene. Nat. Neurosci. 4, 1199–1206.
Markus, A., Patel, T.D., Snider, W.D., 2002. Neurotrophic factors and axonal
growth. Curr. Opin. Neurobiol. 12, 523–531.
Martin, N., Jaubert, J., Gounon, P., Salido, E., Haase, G., Szatanik, M., Guenet,
J.L., 2002. A missense mutation in Tbce causes progressive motor neuronopathy in mice. Nat. Genet. 32, 443–447.
Mason, C.A., Gregory, E., 1984. Postnatal maturation of cerebellar mossy and
climbing fibers: transient expression of dual features on single axons. J.
Neurosci. 4, 1715–1735.
McLaughlin, T., Hindges, R., O’Leary, D.D., 2003a. Regulation of axial
patterning of the retina and its topographic mapping in the brain. Curr.
Opin. Neurobiol. 13, 57–69.
McLaughlin, T., Torborg, C.L., Feller, M.B., O’Leary, D.D., 2003b. Retinotopic
map refinement requires spontaneous retinal waves during a brief critical
period of development. Neuron 40, 1147–1160.
Meister, M., Wong, R.O., Baylor, D.A., Shatz, C.J., 1991. Synchronous bursts of
action potentials in ganglion cells of the developing mammalian retina.
Science 252, 939–943.
Mi, W., Beirowski, B., Gillingwater, T.H., Adalbert, R., Wagner, D., Grumme,
D., Osaka, H., Conforti, L., Arnhold, S., Addicks, K., Wada, K., Ribchester,
R.R., Coleman, M.P., 2005. The slow Wallerian degeneration gene, WldS,
inhibits axonal spheroid pathology in gracile axonal dystrophy mice. Brain
128, 405–416.
Miledi, R., Slater, C.R., 1970. On the degeneration of rat neuromuscular
junctions after nerve section. J. Physiol. 207, 507–528.
Monani, U.R., 2005. Spinal muscular atrophy: a deficiency in a ubiquitous
protein; a motor neuron-specific disease. Neuron 48, 885–896.
O’Leary, D.D., 1992. Development of connectional diversity and specificity in
the mammalian brain by the pruning of collateral projections. Curr. Opin.
Neurobiol. 2, 70–77.
O’Leary, D.D., Koester, S.E., 1993. Development of projection neuron types,
axon pathways, and patterned connections of the mammalian cortex.
Neuron 10, 991–1006.
O’Leary, D.D., Stanfield, B.B., 1989. Selective elimination of axons extended
by developing cortical neurons is dependent on regional locale: experiments
utilizing fetal cortical transplants. J. Neurosci. 9, 2230–2246.
O’Leary, D.D., Terashima, T., 1988. Cortical axons branch to multiple subcortical targets by interstitial axon budding: implications for target recognition and ‘‘waiting periods’’. Neuron 1, 901–910.
O’Rourke, N.A., Cline, H.T., Fraser, S.E., 1994. Rapid remodeling of retinal
arbors in the tectum with and without blockade of synaptic transmission.
Neuron 12, 921–934.
Parson, S.H., Mackintosh, C.L., Ribchester, R.R., 1997. Elimination of motor
nerve terminals in neonatal mice expressing a gene for slow Wallerian
degeneration (C57Bl/Wlds). Eur. J. Neurosci. 9, 1586–1592.
Patel, T.D., Kramer, I., Kucera, J., Niederkofler, V., Jessell, T.M., Arber, S.,
Snider, W.D., 2003. Peripheral NT3 signaling is required for ETS protein
expression and central patterning of proprioceptive sensory afferents.
Neuron 38, 403–416.
Penn, A.A., Riquelme, P.A., Feller, M.B., Shatz, C.J., 1998. Competition in
retinogeniculate patterning driven by spontaneous activity. Science 279,
2108–2112.
Personius, K.E., Balice-Gordon, R.J., 2001. Loss of correlated motor neuron
activity during synaptic competition at developing neuromuscular synapses.
Neuron 31, 395–408.
Petersson, P., Waldenstrom, A., Fahraeus, C., Schouenborg, J., 2003. Spontaneous muscle twitches during sleep guide spinal self-organization. Nature
424, 72–75.
Pirozzi, M., Quattrini, A., Andolfi, G., Dina, G., Malaguti, M.C., Auricchio, A.,
Rugarli, E.I., 2006. Intramuscular viral delivery of paraplegin rescues
17
peripheral axonopathy in a model of hereditary spastic paraplegia. J. Clin.
Invest. 116, 202–208.
Portera-Cailliau, C., Weimer, R.M., De Paola, V., Caroni, P., Svoboda, K., 2005.
Diverse modes of axon elaboration in the developing neocortex. PLoS Biol.
3, e272.
Pun, S., Santos, A.F., Saxena, S., Xu, L., Caroni, P., 2006. Selective vulnerability and pruning of phasic motoneuron axons in motoneuron disease
alleviated by CNTF. Nat. Neurosci. 9, 408–419.
Qiang, L., Yu, W., Andreadis, A., Luo, M., Baas, P.W., 2006. Tau protects
microtubules in the axon from severing by katanin. J. Neurosci. 26, 3120–
3129.
Raff, M.C., Whitmore, A.V., Finn, J.T., 2002. Axonal self-destruction and
neurodegeneration. Science 296, 868–871.
Ruthazer, E.S., Akerman, C.J., Cline, H.T., 2003. Control of axon branch
dynamics by correlated activity in vivo. Science 301, 66–70.
Sagot, Y., Dubois-Dauphin, M., Tan, S.A., de Bilbao, F., Aebischer, P.,
Martinou, J.C., Kato, A.C., 1995. Bcl-2 overexpression prevents motoneuron cell body loss but not axonal degeneration in a mouse model of a
neurodegenerative disease. J. Neurosci. 15, 7727–7733.
Sagot, Y., Rosse, T., Vejsada, R., Perrelet, D., Kato, A.C., 1998. Differential
effects of neurotrophic factors on motoneuron retrograde labeling in a
murine model of motoneuron disease. J. Neurosci. 18, 1132–1141.
Sagot, Y., Tan, S.A., Hammang, J.P., Aebischer, P., Kato, A.C., 1996. GDNF
slows loss of motoneurons but not axonal degeneration or premature death
of pmn/pmn mice. J. Neurosci. 16, 2335–2341.
Sajadi, A., Schneider, B.L., Aebischer, P., 2004. Wlds-mediated protection of
dopaminergic fibers in an animal model of Parkinson disease. Curr. Biol. 14,
326–330.
Samsam, M., Mi, W., Wessig, C., Zielasek, J., Toyka, K.V., Coleman, M.P.,
Martini, R., 2003. The Wlds mutation delays robust loss of motor and
sensory axons in a genetic model for myelin-related axonopathy. J. Neurosci. 23, 2833–2839.
Schaefer, A.M., Sanes, J.R., Lichtman, J.W., 2005. A compensatory subpopulation of motor neurons in a mouse model of amyotrophic lateral sclerosis. J.
Comp. Neurol. 490, 209–219.
Schmalbruch, H., Jensen, H.J., Bjaerg, M., Kamieniecka, Z., Kurland, L., 1991.
A new mouse mutant with progressive motor neuronopathy. J. Neuropathol.
Exp. Neurol. 50, 192–204.
Schubiger, M., Wade, A.A., Carney, G.E., Truman, J.W., Bender, M., 1998.
Drosophila EcR-B ecdysone receptor isoforms are required for larval
molting and for neuron remodeling during metamorphosis. Development
125, 2053–2062.
Stanfield, B.B., Nahin, B.R., O’Leary, D.D., 1987. A transient postmamillary
component of the rat fornix during development: implications for interspecific differences in mature axonal projections. J. Neurosci. 7, 3350–
3361.
Stanfield, B.B., O’Leary, D.D., 1985a. Fetal occipital cortical neurones transplanted to the rostral cortex can extend and maintain a pyramidal tract axon.
Nature 313, 135–137.
Stanfield, B.B., O’Leary, D.D., 1985b. The transient corticospinal projection
from the occipital cortex during the postnatal development of the rat. J.
Comp. Neurol. 238, 236–248.
Stanfield, B.B., O’Leary, D.D., Fricks, C., 1982. Selective collateral elimination
in early postnatal development restricts cortical distribution of rat pyramidal
tract neurones. Nature 298, 371–373.
Stellwagen, D., Shatz, C.J., 2002. An instructive role for retinal waves in the
development of retinogeniculate connectivity. Neuron 33, 357–367.
Stokin, G.B., Goldstein, L.S., 2006. Axonal transport and Alzheimer’s disease.
Annu. Rev. Biochem. 75, 607–627.
Stokin, G.B., Lillo, C., Falzone, T.L., Brusch, R.G., Rockenstein, E., Mount,
S.L., Raman, R., Davies, P., Masliah, E., Williams, D.S., Goldstein, L.S.,
2005. Axonopathy and transport deficits early in the pathogenesis of
Alzheimer’s disease. Science 307, 1282–1288.
Technau, G., Heisenberg, M., 1982. Neural reorganization during metamorphosis of the corpora pedunculata in Drosophila melanogaster. Nature 295,
405–407.
Tian, N., Copenhagen, D.R., 2003. Visual stimulation is required for refinement
of ON and OFF pathways in postnatal retina. Neuron 39, 85–96.
Please cite this article in press as: Saxena, S., Caroni, P., Mechanisms of axon degeneration: From development to disease, Prog. Neurobiol.
(2007), doi:10.1016/j.pneurobio.2007.07.007
+ Models
PRONEU-845; No of Pages 18
18
S. Saxena, P. Caroni / Progress in Neurobiology xxx (2007) xxx–xxx
Tissot, M., Stocker, R.F., 2000. Metamorphosis in Drosophila and other insects:
the fate of neurons throughout the stages. Prog. Neurobiol. 62, 89–111.
Tsai, J., Grutzendler, J., Duff, K., Gan, W.B., 2004. Fibrillar amyloid deposition
leads to local synaptic abnormalities and breakage of neuronal branches.
Nat. Neurosci. 7, 1181–1183.
Vande Velde, C., Garcia, M.L., Yin, X., Trapp, B.D., Cleveland, D.W., 2004.
The neuroprotective factor Wlds does not attenuate mutant SOD1-mediated
motor neuron disease. Neuromol. Med. 5, 193–203.
Waller, A., 1850. Experiments on the sections of glossopharyngeal and hypoglossal nerves of the frog and observations of the alterations produced
thereby in the structure of their primitive fibers. Phil. Trans. R. Soc. Lond.
140, 423–429.
Walsh, M.K., Lichtman, J.W., 2003. In vivo time-lapse imaging of synaptic
takeover associated with naturally occurring synapse elimination. Neuron
37, 67–73.
Watts, R.J., Hoopfer, E.D., Luo, L., 2003. Axon pruning during Drosophila
metamorphosis: evidence for local degeneration and requirement of the
ubiquitin-proteasome system. Neuron 38, 871–885.
Watts, R.J., Schuldiner, O., Perrino, J., Larsen, C., Luo, L., 2004. Glia engulf
degenerating axons during developmental axon pruning. Curr. Biol. 14,
678–684.
Weimann, J.M., Zhang, Y.A., Levin, M.E., Devine, W.P., Brulet, P., McConnell,
S.K., 1999. Cortical neurons require Otx1 for the refinement of exuberant
axonal projections to subcortical targets. Neuron 24, 819–831.
Williams, D.W., Truman, J.W., 2005. Cellular mechanisms of dendrite pruning
in Drosophila: insights from in vivo time-lapse of remodeling dendritic
arborizing sensory neurons. Development 132, 3631–3642.
Williams, D.W., Kondo, S., Krzyzanowska, A., Hiromi, Y., Truman, J.W., 2006.
Local caspase activity directs engulfment of dendrites during pruning. Nat.
Neurosci. 9, 1234–1236.
Williamson, T.L., Cleveland, D.W., 1999. Slowing of axonal transport is a very
early event in the toxicity of ALS-linked SOD1 mutants to motor neurons.
Nat. Neurosci. 2, 50–56.
Wong, P.C., Pardo, C.A., Borchelt, D.R., Lee, M.K., Copeland, N.G., Jenkins,
N.A., Sisodia, S.S., Cleveland, D.W., Price, D.L., 1995. An adverse property
of a familial ALS-linked SOD1 mutation causes motor neuron disease
characterized by vacuolar degeneration of mitochondria. Neuron 14, 1105–
1116.
Wong, R.O., Meister, M., Shatz, C.J., 1993. Transient period of correlated
bursting activity during development of the mammalian retina. Neuron 11,
923–938.
Xia, C.H., Roberts, E.A., Her, L.S., Liu, X., Williams, D.S., Cleveland, D.W.,
Goldstein, L.S., 2003. Abnormal neurofilament transport caused by targeted disruption of neuronal kinesin heavy chain KIF5A. J. Cell Biol. 161,
55–66.
Yamada, K.M., Spooner, B.S., Wessells, N.K., 1970. Axon growth: roles of
microfilaments and microtubules. Proc. Natl. Acad. Sci. U.S.A. 66, 1206–
1212.
Yates, P.A., Roskies, A.L., McLaughlin, T., O’Leary, D.D., 2001. Topographicspecific axon branching controlled by ephrin-As is the critical event in
retinotectal map development. J. Neurosci. 21, 8548–8563.
Yu, W., Solowska, J.M., Qiang, L., Karabay, A., Baird, D., Baas, P.W., 2005.
Regulation of microtubule severing by katanin subunits during neuronal
development. J. Neurosci. 25, 5573–5583.
Zhai, Q., Wang, J., Kim, A., Liu, Q., Watts, R., Hoopfer, E., Mitchison, T., Luo,
L., He, Z., 2003. Involvement of the ubiquitin-proteasome system in the
early stages of Wallerian degeneration. Neuron 39, 217–225.
Zhang, B., Tu, P., Abtahian, F., Trojanowski, J.Q., Lee, V.M., 1997. Neurofilaments and orthograde transport are reduced in ventral root axons of
transgenic mice that express human SOD1 with a G93A mutation. J. Cell
Biol. 139, 1307–1315.
Zhang, Y.A., Okada, A., Lew, C.H., McConnell, S.K., 2002. Regulated nuclear
trafficking of the homeodomain protein otx1 in cortical neurons. Mol. Cell.
Neurosci. 19, 430–446.
Zhou, Z., Hartwieg, E., Horvitz, H.R., 2001. CED-1 is a transmembrane
receptor that mediates cell corpse engulfment in C. elegans. Cell 104,
43–56.
Zufferey, P.D., Jin, F., Nakamura, H., Tettoni, L., Innocenti, G.M., 1999. The
role of pattern vision in the development of cortico-cortical connections.
Eur. J. Neurosci. 11, 2669–2688.
Please cite this article in press as: Saxena, S., Caroni, P., Mechanisms of axon degeneration: From development to disease, Prog. Neurobiol.
(2007), doi:10.1016/j.pneurobio.2007.07.007