B R A I N R E S E A R C H 1 1 8 8 ( 2 0 0 8 ) 6 1 – 6 8

ava i l ab l e a t www.sc i enced i rec t . com

www.e l sev i e r. com/ l oca te /b ra in res

Research Report

Atrophy and degeneration in sciatic nerve of presymptomaticmice carrying the Huntington's disease mutation

Anna Wade1, Peter Jacobs1, A. Jennifer Morton⁎

Department of Pharmacology, University of Cambridge, Tennis Court Road, Cambridge CB2 1PD, UK

A R T I C L E I N F O

⁎ Corresponding author. Fax: +44 1223 334100E-mail address: ajm41@cam.ac.uk (A.J. M

1 These two authors contributed equally to

0006-8993/$ – see front matter © 2007 Elsevidoi:10.1016/j.brainres.2007.06.059

A B S T R A C T

Article history:Accepted 26 June 2007Available online 14 July 2007

Huntington's disease (HD) is a progressive neurological disorder characterised by motorimpairments caused by degeneration in the striatum. The mechanism by which the HDmutation leads to the neurodegenerative pathology of HD is still unknown. Recently it wasshown that, in HD patients, early pathological changes in whitematter precede selective celldeath in the striatum. We wondered whether axonal pathology is also an early pathologicalfeature in a transgenic mouse model carrying the HD mutation (R/2 line). R6/2 mice showbrain atrophy, a progressive neurological deterioration and skeletal muscle atrophy thatresemble those seen in HD patients. However, there is very little neuronal cell loss seen inthese animals, even when they show severe symptoms. Here we used sciatic nerve to lookfor evidence of early neurodegenerative changes in axons of the R6/2 mouse at anultrastructural level. We observed ultrastructural changes that preferentially affected largemyelinated fibres of the sciatic nerve in 10-week-old asymptomatic R6/2 mice. The changesincluded a significant decrease in the axoplasm diameter of myelinated neurons and anincrease in the number of degenerating myelinated fibres compared to age-matched wildtype littermates. Myelin thickness and unmyelinated fibre diameter were not affected. Theabnormalities described here precede the appearance of overt motor symptoms in the R6/2mouse and occur in parallel with pathophysiological changes at the neuromuscularjunction. We suggest that degenerative changes in axons are likely to contribute to the earlypathological phenotype in HD, even in the absence of frank neuronal cell loss.

© 2007 Elsevier B.V. All rights reserved.

Keywords:AxonSkeletal muscleR6/2 miceWallerian degenerationPolyglutamine repeat

1. Introduction

Huntington disease (HD) is an autosomal dominant neurode-generative disorder characterised by a progressive decline inmotor and cognitive function and emotional disturbance (Bateset al., 2002). The genetic mutation responsible for HD is anexpansionof theCAGrepeat in the coding regionof theHDgene.A causal link between genemutation anddisease pathologyhasyet to be established. However, it is clear that, in HD patients,symptoms are associated with atrophy and neurodegeneration

.orton).this work.

er B.V. All rights reserved

in the brain, first in the caudate andputamen (neostriatum) andthen in the cortex, white matter and thalamus (for references,see Bates et al., 2002).

The widespread cortical pathology associated with thedisease is particularly interesting, given that both cognitiveand psychiatric symptoms are common in HD. However, theonset of these symptoms in HD is complex and insidious, andalthough there is a correlation between neuronal loss andsymptomology in striatal regions (Tippett et al., 2007), this hasnot yet been shown in the cortex. However, there is little doubt

.

62 B R A I N R E S E A R C H 1 1 8 8 ( 2 0 0 8 ) 6 1 – 6 8

that cortical abnormalities contribute to the onset andprogression of HD, and considerable effort is being made tounderstand early changes there. Focus on cortical changes hasnot been restricted to grey matter. There is an associationbetween early neurological dysfunction and pathologicalchanges in white matter in human brain (Sapp et al., 1997,1999; Rosas et al., 2002). More recently, presymptomaticchanges in white matter have also been observed throughneuroimaging of HD patients and it has been suggested thatthis could lead to axonal dysfunctionand cell death (Thiebenetal., 2002; Fennema-Notestine et al., 2004; Reading et al., 2004;Beglinger et al., 2005; Ciarmiello et al., 2006). There is otherevidence for disruption of neurocytological architecture atearly stages in HD. Medium spiny neurons have an abnormaldendriticmorphology in bothHDpatient brains (Ferrante et al.,1991; DiFiglia et al., 1997) and the R6/2 mouse model of HD(Klapstein et al., 2001) and there is evidence for abnormaldendritic morphology in a full-length transgenic mouse in theabsence of neurodegeneration (Guidetti et al., 2001). There isalso evidence of neuritic inclusion pathology in R6/2 mice(Morton et al., 2000) and neuritic and axonal inclusion pa-thology in the brain of a full-length knock-in transgenicmousemodel of HD (Li et al., 1999, 2001). Together these data areconsistent with the idea that early pathological changesassociated with HD may occur in the processes of neurons(Charrin et al., 2005; Cowan and Raymond, 2006) and that theyprecede neurodegeneration.

Transgenic mouse models provide invaluable systems forstudying progressive pathological changes inHD. The R6/2 lineis a transgenic mouse model of HD that carries a fragment ofthe HD gene including a pathological length CAG repeat(Mangiarini et al., 1997). R6/2 mice show a progressiveneurological phenotype characterised by progressive deterio-ration ofmotor and cognitive function (Carter et al., 1999; Lioneet al., 1999). However, while R6/2 mice show a significant andearly brain atrophy, evidence for neurodegeneration has beenmixed. Some studies report neuronal degeneration (Stack etal., 2005). However, most studies have either found noevidence of degeneration in R6/2 mice (Davies et al., 1997;Mangiarini et al., 1997) or have found only minimal evidencefor degeneration in the brain at late stages of the disease(Turmaine et al., 2000). It is generally accepted that frankneurodegeneration is not a major feature of the R6/2 mouse(Tobin and Signer, 2000). Importantly this raises the questionof the role that neuronal cell death plays in the progression ofthe disease. Absence of specific neurodegeneration in R6/2mice, despite the presence of profound progressive neurolog-ical signs, supports the idea that initially, neurologicalsymptoms may be caused by neuronal dysfunction ratherthen cell death (Morton et al., 2000; for further references seeBossy-Wetzel et al., 2004; Smith et al., 2006), although theimportance of neurodegeneration in the progression of HD isundisputed.

The aim of this study was to determine whether or notthere were degenerative changes in myelinated axons in theR6/2 mouse. We used sciatic nerve for our studies. The sciaticnerve is a long fasciculated nerve that permits ultrastructuralexamination of both myelinated motor and unmyelinatedsensory axons that arise from neurons with cell bodies in thespinal cord and dorsal root ganglia. We used the sciatic nerve

rather than central nervous system (CNS) white matter forthree reasons. First, the sciatic nerve has a well-describedultrastructural morphology and there is a large literature thatallows comparison of ultrastructure in both normal anddegenerating nerves. Second, we know there are skeletalmuscle and neuromuscular junction abnormalities in the R6/2mouse in the absence of neuronal loss in skeletal musclesinnervated by the sciatic nerve (Ribchester et al., 2004). Third,the axons in the sciatic nerve are easily accessible and well-defined. Finally, we hoped that by studying the sciatic nervewe would be able to shed some light on the neuropathologicalorigin of some of the abnormalities seen in the neuromuscu-lar junction (Ribchester et al., 2004).

2. Results

2.1. Myelinated axons showed significant atrophy

There was no significant difference between the whole nervecross-sectional areas measured in WT and R6/2 mice(P=0.5252). Total axon diameter showed a trend towardsbeing smaller in R6/2 mice compared to the WT mice, but thisdid not reach statistical significance (P=0.0578). When weanalysed axoplasmdiameter andmyelin thickness separately,we found a significant difference in axoplasm diameter ofsciatic nerves from WT and R6/2 mice (Fig. 1a; Pb0.01).However, myelin thickness was not significantly differentbetween WT and R6/2 mice (P=0.299).

The frequency distribution of axon diameters inWT and R6/2 was significantly different (Pb0.05); with larger (N0.5 μm)calibre axons seen in WT mice compared with their R6/2littermates (71±1% in R6/2 nerves compared to 76±2% in WTnerves). However, there was no difference in the myelinatedaxon packing density between WT and R6/2 mice (P=0.3432).This is consistent with previous findings that there is nodifference in the neuronal packing density in the CNS of WTand R6/2mice brains (Mangiarini et al., 1997; Davies et al., 1997).

2.2. Evidence of degeneratingmyelinated axons in R6/2mice

Of the 4 axonal phenotypes separately categorised (for details,see Experimental procedures), only the number of degeneratingneurons was significantly different between R6/2 andWTmice.The number of degenerating myelinated axons per nerve wassignificantly higher in R6/2 mice than in their WT littermates(Fig. 1c; Pb0.01). There was no significant difference in thepercentage of axons classed as ‘mildly degenerating’ betweenR6/2 and WT nerves (P=0.111), although the number of degen-erating axons remained significant when the axons that wereclassified as mildly degenerating were included in the totalcount (Pb0.01). This is the first direct evidence showing frankneurodegenerative changes in axons in the R6/2 mouse.

Therewas no difference seen betweenWT and R6/2mice inthe percentage of split axons per nerve (P=0.5065) or abnormalaxons per nerve (P=0.8981). Similarly there was no differencebetween genotypes in the number of axons that showed acombination of split and abnormal morphology (P=0.7643). Itseems likely that the ‘split’ and ‘abnormal’ phenotypes weredue to artefacts arising from the preparation of the tissue.

Fig. 1 – Quantification of morphological changes in R6/2axons. (a) Mean diameters and myelin thickness ofmyelinated axons from WT (open columns) and R6/2 mice(closed columns); (b) the frequency distribution of axoplasmdiameters in R6/2 (squares) and WT (empty circles); (c) thenumber of morphological abnormalities seen in myelinatedaxons inWT (open columns) and R6/2mice (closed columns).* indicates P<0.01.

63B R A I N R E S E A R C H 1 1 8 8 ( 2 0 0 8 ) 6 1 – 6 8

2.3. No difference in unmyelinated axon parameters

The total numbers of unmyelinated fibres (see Figs. 2d, e) in thenerves were similar in R6/2 andWTmice. The mean number ofunmyelinated fibres seen in R6/2 mice was 7172±739 per nervecompared with 8153±631 per nerve in WT mice. Although the

diameters of unmyelinated axons showed a trend towards areduced axoplasm calibre in R6/2mice comparedwith theirWTlittermates, this did not reach statistical significance (P=0.0769).We did not find any evidence of degeneration in unmyelinatedaxons, although therewasa significantdifference in thepackingdensity between the WT and R6/2 nerves (Pb0.05). Axons fromR6/2 mice showed a significantly lower packing density thantheir WT littermates (40246±2885 axons/mm2 in R6/2 micecompared to 45937±3474 axons/mm2 in WT mice; Pb0.05).However, if the cross-sectional nerve area was taken intoaccount and total number of axons per nerve calculated, therewasnosignificantdifferencebetweengenotypes (meannumberof axons=8153±632 per nerve in WT mice and 7172±739 pernerve for R6/2 mice; P=N0.05).

3. Discussion

Theprincipal finding from this study is that there are significantdegenerative abnormalities in the axons of peripheral nervestaken from asymptomatic R6/2 mice. Although neuropatholog-ical features (such as NIIs and dystrophic neurites) have beenreported previously, this is the first direct evidence of presymp-tomatic neurodegenerative changes in R6/2 mice.

There was a significant atrophy of axoplasm in sciatic nerveaxons in R6/2 mice, with no evidence of changes in myelinsheath. This parallels thewidespread atrophy seen inR6/2 brain(Davies et al., 1997; Mangiarini et al., 1997). At 12 weeks of age, a20% reduction in brain volume with no difference in neuronaldensity was reported (Davies et al., 1997). Here we found that at10 weeks of age the axoplasm of myelinated axons showed a15.5%decrease in diameter. Interestingly, post hoc examinationof sample photomicrographs suggests that although there is asignificant atrophy of axoplasm, axons with degeneratingphenotypes appeared to be swollen (see Fig. 2g). Swollenaxons are characteristic hallmarks of neurodegeneration (Cole-man et al., 2005; Smith and Jeffery, 2006). We did not measurethis separately, because we measured all axon diameters blindand irrespective of morphology. However, in future studies, itwould be useful to quantify the diameter of degenerating axonsseparately. It should be noted that if axonal swelling issignificant, we are likely to have underestimated the degree ofatrophy seen, since we included all degenerating axons in ourcalculation of axon calibre. It is notable that axonal atrophy anddegenerative changeswereonly seen inmyelinatedaxons. Thusit seems that, at 10weeks of age, abnormalities inR6/2 axonsarerestricted to the motor neurons and that the sensory neuronsare not affected. This is consistent with human data whereprincipal features of the disease are motor impairments ratherthan sensory deficits (for references, see Bates et al., 2002).

The results from the current study are consistent with thepathophysiology observed in a previous studywepublished onR6/2mouse skeletalmuscle function (Ribchester et al., 2004). Inthat study we found distinctive abnormalities at motor end-plates that became apparent at 12 weeks in R6/2 mice. We didnot quantify those abnormalities systematically, because theirincidence was very low-estimated at fewer than 5% of end-plates. However, when recording from motor units, we foundthat about 5% ofmuscle fibres were consistently unresponsiveto repeated supramaximal nerve stimulation at 12 weeks. We

Fig. 2 – Morphology of sciatic nerve from R6/2 and WT mice. The cartoon shows the dissection position of the sciatic nervesamples (a). Electronmicrographs (b–g) show cross-sections of sciatic nerves fromWT (b, f) and R6/2 (c–e, g)mice. In panels b andc, all axons are myelinated. In panels d and e, both myelinated and unmyelinated (small arrows) axons are visible. Lines inpanel b indicate typical measurements for axoplasm diameter (A) andmyelin thickness (M). Degenerating axons (D in panels c,e, arrowheads in panel g) were visible in all R6/2 nerves. Axons with separated myelin (S) were observed in both R6/2 and WTtissue (d). Note that this is likely to be an artefact of fixation; this section has been chosen deliberately to illustrate thismorphology. Schwann cell nuclei (* in panel d) were observed in all nerves. No other nuclei were seen. Scale bar in panel grepresents 5 μm for panels b and c, 50 μm in panels d and e and 100 μm for panels f and g.

64 B R A I N R E S E A R C H 1 1 8 8 ( 2 0 0 8 ) 6 1 – 6 8

speculated that neurodegenerative changes in individualaxons could cause the failures seen in the motor units. Thepercentage of abnormalities in endplates and motor unitresponsiveness (5% at 12 weeks of age) corresponds very wellto the percentage of degeneratingmyelinated axonswe saw in

the sciatic nerve (4% at 10 weeks of age). In our previous study,we found that the number of unresponsive motor unitsincreased as the disease progressed to ≈20% by 18 weeks. Ifthe non-responsiveness of the motor units is caused bydegeneration in single axons, we would predict that, as the

65B R A I N R E S E A R C H 1 1 8 8 ( 2 0 0 8 ) 6 1 – 6 8

disease progresses, the percentage of degenerating fibreswould increase, since by 18 weeks about 20% of motor unitsfailed. This would contribute significantly to motor dysfunc-tion that is seen in late stage animals.

Non-apoptotic degeneration identified at an electronmicro-scopic level has been reported in symptomatic (N14weeks) R6/2mice in the anterior cingulate cortex, dorsal striatum andPurkinjecells of thecerebellum(Turmaineetal., 2000), but toourknowledge there is no direct ultrastructural evidence forneurodegeneration of the presymptomatic R6/2mouse nervoussystem. Interestingly, the absence of cell death has previouslybeen reported as a problem for this mouse model (Tobin andSigner, 2000). However, since many of the signs present inhuman patients are found in the R6/2 model, it is clear that celldeath is not a prerequisite for signs (Levine et al., 2004). Theresults from the present study are consistent with a functionaldeficit causing the signs in theR6/2mouse that doesnot dependon death of the cells.

It is interesting to note that although the axons showeddegenerating morphology, they were still intact, and there wasno change in packing density of the myelinated neurons. Thisimplies that the cell body is not dead. However, even if theneuron is not dead, there may be pathological changes asso-ciated with axonal degeneration. For example, axonal damagehas been reported to cause' reflex' increases in the number anddensity of nuclear membrane pores (Liebermann, 1971). Inter-estingly, Davies and colleagues observed an increase in nuclearpores in the absence of any other degenerative changes in R6/2brain (Davies et al., 1997). It is interesting to speculate that thismay be caused by degeneration of the axons.

Our study confirms and extends important previous studies(Li et al., 1999, 2001) where ultrastructural changes in whitematter suggestive of degenerating axons were observed. The2001 Li study was conducted in a HD-repeat knock-in mousemodel and they predicted that axon degeneration would befound in other HD mouse models. Our study confirms theirprediction. One of the difficulties that Li et al. encountered isthat, in the brain, it is not possible to identify the origin ofmyelinatedprocessesatanelectronmicroscopic level. Indeed, itis very difficult to distinguish a myelinated neurite from anaxon. In Li's study they deduced from circumstantial evidencethat the changes seen in the striatumwere axonal. However, inour study, axon degeneration was observed directly andunequivocally (since the sciatic nerve is predominantly axontracts and the only the cell bodies present are those of Schwanncells). We suggest that changes in the myelinated axons of thesciatic nerve we describe here are predictive of similar changesin the central nervous system. It will be interesting to see ifsimilar changes are seen in othermousemodels of HDwhere todate no changes in white matter have been reported.

The mechanisms mediating the changes we observe areunknown, as is the mechanism by which the HD mutationcauses HD pathology. Several interesting studies implicatedysregulation of neurotrophic factors in the pathology of HD(for references, see Zuccato andCattaneo, 2007). However, thereis no evidence for nerve growth factor abnormalities in the R6/2mouse (Kuhn et al., 2007).

Both Li's study and ours provide evidence that axonalpathology is present in HD transgenic mice, and support theidea that axonal pathology develops early in the course of HD. Li

found that the neuritic aggregates could block axon transport,and they suggested that there is a causal relationship betweendeficits in axon transport and axon degeneration, which in turncan lead to neuron cell death. There is considerable support forthis idea. Both Caenorhabditis elegans (Parker et al., 2001) andDrosophila (Gunawardena et al., 2003; Lee et al., 2004; Gunawar-dena and Goldstein, 2005) models that express pathogenicforms of human huntingtin show a progressive loss of motorcoordination that correlates with the formation of cytoplasmicand nuclear aggregates in axons. A deficit of axonal transporthas also been seen in vitro in a mammalian model (Trushinaet al., 2004). Ribchester et al. (2004) found evidence that motorendplate abnormalities showed similaritieswithmousemodelsof dying-back neuropathy and motor neuron disease. At leasttwo studies show that disruption of axon transport alone issufficient to cause axon degeneration and cell death (LaMonteet al., 2002; Hafezparast et al., 2003). Interestingly, the axondegeneration in the study by LaMonte et al. (2002) showedultrastructural abnormalities similar to those we found in theR6/2 sciatic nerve. Theydescribe evidence of degeneratingaxonswith axonal inclusions, myelin defects and observed invagina-tions ofmyelin sheath into the axons. They attributed this latterunusual characteristic to axonal atrophy and a deficient cyto-skeleton. Although we did not quantify it, we also found evi-dence of invaginations of myelin. It would be useful to quantifythese myelin invaginations in the degenerating axons in thefuture.

As well as being the first direct demonstration of axonalpathology in R6/2 mice, our study highlights the fact that themotor impairments in HD may be due, at least in part, toperipheral pathology. We have shown previously that approx-imately 20% of motor units from late stage R6/2 mice (agedbetween 15 and 18 weeks) were unresponsive to electricalstimulation measured by excitatory postsynaptic potentialresponse (Ribchester et al., 2004). If the primary cause of HDsymptoms is related to axonal dysfunction rather than celldegeneration then future therapies should focus on this as atarget as well as current strategies that aim to prevent celldeath (Coleman and Perry, 2002; Raff et al., 2002; Luo andO'Leary, 2005). It would be interesting to cross the R6/2 linewith the Wlds mouse line as has been done with pmn(progressive motor neuronopathy) and SOD (ALS) mice (Ferriet al., 2003; Fischer et al., 2005). If the onset of symptoms isdelayed and their life span increased, it would support theproposal that the R6/2 mouse, axon degeneration is animportant part of the pathological process.

In summary, our data are consistent with the idea that inHD, some neuropathology originates in the axons and thataxon degeneration and neuronal dysfunction (rather than celldeath) are primary causes of early HD symptoms. Our studyalso provides an explanation for the selective cell death ofprojection neurons in HD brains since the longest axons arethe most susceptible to axonal transport disruption. Finally,our data support the idea that presymptomatic changes inwhitematter in HD are due to axonal degeneration that occursbefore neurodegeneration. The clinical relevance of our studyremains speculative. If evidence of axonal degeneration isfound in human HD patients then therapeutic interventionsshould aim to prevent axonal dysfunction and/or degenera-tion. The R6/2 mouse should prove to be a useful model in

66 B R A I N R E S E A R C H 1 1 8 8 ( 2 0 0 8 ) 6 1 – 6 8

which to test therapeutics aimed at preventing or slowingaxonal degeneration.

4. Experimental procedures

All studies were carried out in accordance with the UK Animals(Scientific Procedures) Act 1986. Mice were taken from a colonyestablished in the University of Cambridge and maintained bybackcrossing onto CBA×C57/BL6 F1 mice as described by Carteret al. (1999). Mice of mixed genotype were housed in polypro-pylene cages under a 12:12 h light–dark cycle in a temperature-controlled room. For all mice, chow soaked for 1 h in water(mash) was delivered to the floor of the home cage once a day(at 08:00 h) from weaning. Genotyping was confirmed bypolymerase chain reaction and mean repeat lengths of R6/2micewere 259±7 (measured by Laragen, USA). None of themiceused in this study was diabetic. None of the mice showedchanges in body weight compared to the WT mice (data notshown). Furthermore, none showed overt signs of disease, suchas hindlimb grooming, gait abnormalities or piloerection.

Tissuewas taken from6WTand 6 R6/2mice aged 10weeks.Mice were terminally anaesthetised with Sagittal. They wereperfuse-fixed with 3% glutaraldehyde and 1% formaldehyde in0.1MPIPES buffer pH 7.4with 2mMol/L CaCl2. The sciatic nervewas exposedanda lengthof nerve 5mmlongwas removedat apoint 10 mm distal to the hip joint (Fig. 2a). Nerves from bothsides of each animal were used (N=12 for each genotype). Thenerves were osmicated and dehydrated through a gradedalcohol series and embedded in Spurr's low viscosity epoxyresin.

4.1. Electron microscopy

Thin (60 nm) transverse sections of all nerves were cut andviewed by transmission electron microscopy using an FEI-Phillips CM100 (Eindhoven, Netherlands) at a magnification of6200 (for measurement of myelinated fibre parameters) or at11,000 (for measurement of unmyelinated fibre parameters).

Myelinated and unmyelinated axons were counted in non-overlapping fields selected at random using a two-way axialstrip through each of the nerve bundles. All axons countedwere counted in at least 7 fields for myelinated fibres and atleast 31 fields for unmyelinated fibres. Axon counts within afield of view were determined using the forbidden line rule(Gundersen, 1986). The packing density was calculated as totalaxons counted per nerve divided by the total area sampled pernerve. Total area was calculated as number of fields of viewmultiplied by field of view area. Axon diameter and myelinthickness (Fig. 2b) weremeasured using Leica Quantimet Q500(Leica Microsystems, Heidleberg, Germany). Measurementsmade on line were calibrated against a 2060 lines/mm linegrating (Agar Scientific Ltd.). Axoplasm (A, Fig. 2b) wasmeasured across its shortest axis to minimise distortionsdue to the transection process. Myelin thickness (M, Fig. 2b)was estimated once per axon at the most regular point on thecircumference. For myelinated axons, the total diameter wascalculated as 2M+A. Unmyelinated axon diameters weremeasured from the inner boundary of the membrane acrossthe shortest axis.

4.2. Confocal microscopy

The cross-sectional area of each axon bundle was measuredfrom photomicrographs taken on a Leica DM IRBE confocalmicroscope from a 1 μm thick, methylene blue-stained sectioncut immediately following the ultrathin sections. Images wereanalysed using Leica Confocal Software (Leica Microsystems,Heidleberg, Germany). The total number of axons in the nervewasestimatedbymultiplyingpackingdensity (seeabove) by thecross-sectional area of the nerve.

4.3. Ultrastructural analysis

All experiments were conducted with the experimenter doubleblind to the genotype of the mice for the entire data collectionperiod.Whenall of the data had been collected and collated, thecodewas broken, the genotype of themice was established andthe mean data from WT and R6/2 mice were calculated.

Myelinated axons were classified by their morphology intoone of several different categories as determined by twoobservers. Axons were assigned by their morphology to one offour groups; ‘degenerating’, ‘mildly degenerating’, ‘abnormal’and ‘split myelin’. Degenerating axons (D in Figs. 2c and e;arrows in Fig. 2g) were characterised as having (1) a distortedshape, (2) electron dense axoplasm and (3) disorganised/non-uniform myelin structural abnormalities such as myelinfragmentation into the axoplasm. ‘Mildly degenerating’ axonsshowed only one or two of these characteristics. Axonscharacterised as ‘abnormal’ showed unusual myelin staining,for example where the outer and inner edges of the myelinappeared darker than the intervening wraps. ‘Split myelin’axons were characterised as having a myelin sheath partiallyseparated from the axoplasm (S, Fig. 2d). Axons were alsoassigned to categories thatwere combinationsof these features.The number of axons assigned to each category was convertedto a percentage of axons counted from a particular nerve as(number of axons with morphology classification)/(number ofaxons counted in nerve (n)×100).

Unmyelinated axons (arrows, Figs. 2d and e) did not varyqualitatively, therefore only total axon numbers and total axondiameter were quantified. Axon diameter was taken at thenarrowest point of the axon and included only axoplasm.

4.4. Statistical analysis

All data were tested using D'Agostino and Pearson's omnibusnormality test and variances were tested using the F test. Datawith a normal distribution were analysed using eitherStudent's unpaired t-test or a one-way analysis of variance(ANOVA) and the Neuman–Keuls post hoc test. If data failedthe normality test a Mann–Whitney test was used forcomparison. All data are presented as mean±SEM. A criticalvalue of Pb0.05 was used throughout this study.

Acknowledgments

We wish to thank Dr. Jeremy Skepper for helpful advice, andCheney Drew, Janet Powell and Lyn Carter for technical assis-tance.Thisworkwassupported inpart by theHighQFoundation.

67B R A I N R E S E A R C H 1 1 8 8 ( 2 0 0 8 ) 6 1 – 6 8

R E F E R E N C E S

Bates, G.P., Harper, P.S., Jones, L., 2002. Huntington's Disease, 3rdEdition. Oxford University Press Ltd, Oxford.

Beglinger, L.J., Nopoulos, P.C., Jorge, R.E., Langbehn, D.R., Mikos, A.E.,Moser, D.J., Duff, K., Robinson, R.G., Paulsen, J.S., 2005. Whitematter volume and cognitive dysfunction in early Huntington'sdisease. Cogn. Behav. Neurol. 18, 102–107.

Bossy-Wetzel, E., Schwarzenbacher, R., Lipton, S.A., 2004. Molecularpathways to neurodegeneration. Nat. Med. 10, S2–S9 (Suppl).

Carter, R.J., Lione, L.A., Humby, T., Mangiarini, L., Mahal, A., Bates,G.P., Dunnett, S.B., Morton, A.J., 1999. Characterization ofprogressive motor deficits in mice transgenic for the humanHuntington's disease mutation. J. Neurosci. 19, 3248–3257.

Charrin, B.C., Saudou, F., Humbert, S., 2005. Axonal transport failurein neurodegenerative disorders: the case of Huntington'sdisease. Pathol. Biol. (Paris) 53, 189–192.

Ciarmiello, A., Cannella, M., Lastoria, S., Simonelli, M., Frati, L.,Rubinsztein, D.C., Squitieri, F., 2006. Brain white-mattervolume loss and glucose hypometabolism precede the clinicalsymptoms of Huntington's disease. J. Nucl. Med. 47, 215–222.

Coleman, M.P., Perry, V.H., 2002. Axon pathology in neurologicaldisease: a neglected therapeutic target. Trends Neurosci. 25,532–537.

Coleman, M.P., Adalbert, R., Beirowski, B., 2005. Neuroprotectivestrategies inMS: lessons fromC57BL/Wld(S) mice. J. Neurol. Sci.233, 133–138.

Cowan, C.M., Raymond, L.A., 2006. Selective neuronal degenerationin Huntington's disease. Curr. Top. Dev. Biol. 75, 25–71.

Davies, S.W., Turmaine, M., Cozens, B.A., DiFiglia, M., Sharp, A.H.,Ross, C.A., Scherzinger, E.,Wanker, E.E.,Mangiarini, L., Bates,G.P.,1997. Formation of neuronal intranuclear inclusions underliesthe neurological dysfunction in mice transgenic for the HDmutation. Cell 90, 537–548.

DiFiglia, M., Sapp, E., Chase, K.O., Davies, S.W., Bates, G.P.,Vonsattel, J.P., Aronin, N., 1997. Aggregation of huntingtin inneuronal intranuclear inclusions and dystrophic neurites inbrain. Science 277, 1990–1993.

Fennema-Notestine, C., Archibald, S.L., Jacobson,M.W., Corey-Bloom,J., Paulsen, J.S., Peavy, G.M., Gamst, A.C., Hamilton, J.M., Salmon,D.P., Jernigan, T.L., 2004. In vivo evidence of cerebellar atrophyand cerebral whitematter loss inHuntington disease. Neurology63, 989–995.

Ferrante,R.J., Kowall,N.W.,Richardson Jr., E.P., 1991.Proliferativeanddegenerative changes in striatal spiny neurons in Huntington'sdisease: a combined study using the section-Golgi method andcalbindin D28k immunocytochemistry. J. Neurosci. 11,3877–3887.

Ferri, A., Sanes, J.R., Coleman, M.P., Cunningham, J.M., Kato, A.C.,2003. Inhibiting axon degeneration and synapse loss attenuatesapoptosis and disease progression in a mouse model ofmotoneuron disease. Curr. Biol. 13, 669–673.

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 theSOD1G93A fALS mouse. Neurobiol. Dis. 19, 293–300.

Guidetti, P., Charles, V., Chen, E.Y., Reddy, P.H., Kordower, J.H.,Whetsell Jr., W.O., Schwarcz, R., Tagle, D.A., 2001. Earlydegenerative changes in transgenic mice expressing mutanthuntingtin involve dendritic abnormalities but noimpairment of mitochondrial energy production. Exp. Neurol.169, 340–350.

Gunawardena, S., Goldstein, L.S., 2005. Polyglutamine diseasesand transport problems: deadly traffic jams on neuronalhighways. 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 orexpression of pathogenic polyQ proteins in Drosophila. Neuron40, 25–40.

Gundersen, H.J., 1986. Stereology of arbitrary particles. A review ofunbiased number and size estimators and the presentationof some new ones, in memory of William R. Thompson.J. Microsc. 143, 3–45.

Hafezparast, M., Klocke, R., Ruhrberg, C., Marquardt, A.,Ahmad-Annuar, A., Bowen, S., Lalli, G., Witherden, A.S.,Hummerich, H., Nicholson, S., Morgan, P.J., Oozageer, R.,Priestley, J.V., Averill, S., King, V.R., Ball, S., Peters, J., Toda, T.,Yamamoto, A., Hiraoka, Y., Augustin, M., Korthaus, D., Wattler,S., Wabnitz, P., Dickneite, C., Lampel, S., Boehme, F., Peraus, G.,Popp, A., Rudelius, M., Schlegel, J., Fuchs, H., Hrabe de Angelis,M., Schiavo, G., Shima, D.T., Russ, A.P., Stumm, G., Martin, J.E.,Fisher, E.M., 2003. Mutations in dynein link motor neurondegeneration to defects in retrograde transport. Science 300,808–812.

Klapstein, G.J., Fisher, R.S., Zanjani, H., Cepeda, C., Jokel, E.S.,Chesselet, M.F., Levine, M.S., 2001. Electrophysiological andmorphological changes in striatal spiny neurons in R6/2Huntington's disease transgenic mice. J. Neurophysiol. 86,2667–2677.

Kuhn, A., Goldstein, D.R., Hodges, A., Strand, A.D., Sengstag, T.,Kooperberg, C., Becanovic, K., Pouladi, M.A., Sathasivam, K.,Cha, J.H., Hannan, A.J., Hayden, M.R., Leavitt, B.R., Dunnett,S.B., Ferrante, R.J., Albin, R., Shelbourne, P., Delorenzi, M.,Augood, S.J., Faull, R.L., Olson, J.M., Bates, G.P., Jones, L.,Luthi-Carter, R., 2007. Mutant huntingtin's effects on striatalgene expression in mice recapitulate changes observed inhuman Huntington's disease brain and do not differ withmutant huntingtin length or wild-type huntingtin dosage.Hum. Mol. Genet. 16, 1845–1861.

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 inmotor neurons causing late-onset progressive degeneration.Neuron 34, 715–727.

Lee,W.C., Yoshihara,M., Littleton, J.T., 2004. Cytoplasmic aggregatestrap polyglutamine-containing proteins and block axonaltransport in a Drosophilamodel of Huntington's disease. Proc.Natl. Acad. Sci. U. S. A. 101, 3224–3229.

Levine, M.S., Cepeda, C., Hickey, M.A., Fleming, S.M., Chesselet, M.F.,2004. Genetic mouse models of Huntington's and Parkinson'sdiseases: illuminating but imperfect. Trends Neurosci. 27,691–697.

Li, H., Li, S.H., Cheng, A.L., Mangiarini, L., Bates, G.P., Li, X.J., 1999.Ultrastructural localization and progressive formation ofneuropil aggregates in Huntington's disease transgenic mice.Hum. Mol. Genet. 8, 1227–1236.

Li, H., Li, S.H., Yu, Z.X., Shelbourne, P., Li, X.J., 2001. Huntingtinaggregate-associated axonal degeneration is an earlypathological event in Huntington's disease mice. J. Neurosci.21, 8473–8481.

Liebermann, A.R., 1971. A review of the principal features ofperikaryal responses to axon injury. Int. Rev. Neurobiol. 14,49–124.

Lione, L.A., Carter, R.J., Hunt, M.J., Bates, G.P., Morton, A.J., Dunnett,S.B., 1999. Selective discrimination learning impairments inmice expressing the human Huntington's disease mutation.J. Neurosci. 19, 10428–10437.

Luo, L., O'Leary, D.D., 2005. Axon retraction and degenerationin development and disease. Annu. Rev. Neurosci. 28,127–156.

Mangiarini, L., Sathasivam, K., Mahal, A., Mott, R., Seller, M., Bates,G.P., 1997. Instability of highly expanded CAG repeats in micetransgenic for the Huntington's disease mutation. Nat. Genet.15, 197–200.

68 B R A I N R E S E A R C H 1 1 8 8 ( 2 0 0 8 ) 6 1 – 6 8

Morton, A.J., Lagan, M.A., Skepper, J.N., Dunnett, S.B., 2000.Progressive formation of inclusions in the striatum andhippocampus of mice transgenic for the human Huntington'sdisease mutation. J. Neurocytol. 29, 679–702.

Parker, J.A., Connolly, J.B., Wellington, C., Hayden, M., Dausset, J.,Neri, C., 2001. Expanded polyglutamines in Caenorhabditiselegans cause axonal abnormalities and severe dysfunction ofPLM mechanosensory neurons without cell death. Proc. Natl.Acad. Sci. U. S. A. 98, 13318–13323.

Raff, M.C., Whitmore, A.V., Finn, J.T., 2002. Axonal self-destructionand neurodegeneration. Science 296, 868–871.

Reading, S.A., Dziorny, A.C., Peroutka, L.A., Schreiber,M., Gourley,L.M., Yallapragada, V., Rosenblatt, A., Margolis, R.L., Pekar, J.J.,Pearlson, G.D., Aylward, E., Brandt, J., Bassett, S.S., Ross, C.A.,2004. Functional brain changes in presymptomatic Huntington'sdisease. Ann. Neurol. 55, 879–883.

Ribchester, R.R., Thomson, D., Wood, N.I., Hinks, T.,Gillingwater, T.H., Wishart, T.M., Court, F.A., Morton, A.J.,2004. Progressive abnormalities in skeletal muscle andneuromuscular junctions of transgenic mice expressing theHuntington's disease mutation. Eur. J. Neurosci. 20,3092–3114.

Rosas, H.D., Liu, A.K., Hersch, S., Glessner, M., Ferrante, R.J., Salat,D.H., van der Kouwe, A., Jenkins, B.G., Dale, A.M., Fischl, B.,2002. Regional and progressive thinning of the cortical ribbonin Huntington's disease. Neurology 58, 695–701.

Sapp, E., Schwarz, C., Chase, K., Bhide, P.G., Young, A.B., Penney, J.,Vonsattel, J.P., Aronin, N., DiFiglia, M., 1997. Huntingtinlocalization in brains of normal and Huntington's diseasepatients. Ann. Neurol. 42, 604–612.

Sapp, E., Penney, J., Young, A., Aronin, N., Vonsattel, J.P., DiFiglia,M., 1999. Axonal transport of N-terminal huntingtin suggestsearly pathology of corticostriatal projections in Huntingtondisease. J. Neuropathol. Exp. Neurol. 58, 165–173.

Smith, P.M., Jeffery, N.D., 2006. Histological and ultrastructuralanalysis of white matter damage after naturally-occurringspinal cord injury. Brain Pathol. 16, 99–109.

Smith, R., Chung, H., Rundquist, S., Maat-Schieman, M.L., Colgan, L.,Englund, E., Liu, Y.J., Roos, R.A., Faull, R.L., Brundin, P., Li, J.Y., 2006.Cholinergic neuronal defect without cell loss in Huntington'sdisease. Hum. Mol. Genet. 15, 3119–3131.

Stack, E.C., Kubilus, J.K., Smith, K., Cormier, K., Del Signore, S.J.,Guelin, E., Ryu, H., Hersch, S.M., Ferrante, R.J., 2005. Chronologyof behavioral symptoms and neuropathological sequela in R6/2Huntington's disease transgenic mice. J. Comp. Neurol. 490,354–370.

Thieben, M.J., Duggins, A.J., Good, C.D., Gomes, L., Mahant, N.,Richards, F.,McCusker, E., Frackowiak, R.S., 2002. Thedistributionof structuralneuropathology inpre-clinicalHuntington'sdisease.Brain 125, 1815–1828.

Tippett, L.J.,Waldvogel, H.J., Thomas, S.J., Hogg, V.M., van Roon-Mom,W., Synek, B.J., Graybiel, A.M., Faull, R.L., 2007. Striosomes andmood dysfunction in Huntington's disease. Brain 130, 206–221.

Tobin, A.J., Signer, E.R., 2000. Huntington's disease: the challengefor cell biologists. Trends Cell Biol. 10, 531–536.

Trushina, E., Dyer, R.B., Badger II, J.D., Ure, D., Eide, L., Tran, D.D.,Vrieze, B.T., Legendre-Guillemin, V., McPherson, P.S., Manda-villi, B.S., Van Houten, B., Zeitlin, S., McNiven, M., Aebersold, R.,Hayden, M., Parisi, J.E., Seeberg, E., Dragatsis, I., Doyle, K.,Bender, A., Chacko, C., McMurray, C.T., 2004. Mutant huntingtinimpairs axonal trafficking in mammalian neurons in vivo andin vitro. Mol. Cell. Biol. 24, 8195–8209.

Turmaine, M., Raza, A., Mahal, A., Mangiarini, L., Bates, G.P.,Davies, S.W., 2000. Nonapoptotic neurodegeneration in atransgenic mouse model of Huntington's disease. Proc. Natl.Acad. Sci. U. S. A. 97, 8093–8097.

Zuccato, C., Cattaneo, E., 2007. Role of brain-derived neurotrophicfactor in Huntington's disease. Prog. Neurobiol. 81, 294–330.