Identifying the primary site of pathogenesis in ... · slow the progression of amyotrophic lateral sclerosis (ALS). ALS is a disorder with heterogeneous onset, which then leads to

© 2015. Published by The Company of Biologists Ltd | Disease Models & Mechanisms (2015) 8, 215-224 doi:10.1242/dmm.018606

215

ABSTRACTThere is a desperate need for targeted therapeutic interventions thatslow the progression of amyotrophic lateral sclerosis (ALS). ALS is adisorder with heterogeneous onset, which then leads to common finalpathways involving multiple neuronal compartments that span boththe central and peripheral nervous system. It is believed thatexcitotoxic mechanisms might play an important role in motor neurondeath in ALS. However, little is known about the mechanisms bywhich excitotoxicity might lead to the neuromuscular junctiondegeneration that characterizes ALS, or about the site at which thisexcitotoxic cascade is initiated. Using a novel compartmentalisedmodel of site-specific excitotoxin exposure in lower motor neurons invitro, we found that spinal motor neurons are vulnerable tosomatodendritic, but not axonal, excitotoxin exposure. Thus, wedeveloped a model of somatodendritic excitotoxicity in vivo usingosmotic mini pumps in Thy-1-YFP mice. We demonstrated that invivo cell body excitotoxin exposure leads to significant motor neurondeath and neuromuscular junction (NMJ) retraction. Using confocalreal-time live imaging of the gastrocnemius muscle, we found thatNMJ remodelling preceded excitotoxin-induced NMJ degeneration.These findings suggest that excitotoxicity in the spinal cord ofindividuals with ALS might result in a die-forward mechanism of motorneuron death from the cell body outward, leading to initial distalplasticity, followed by subsequent pathology and degeneration.

KEY WORDS: Motor neuron disease, Amyotrophic lateral sclerosis,Excitotoxicity, Lower motor neuron, Excitotoxin exposure

INTRODUCTIONAmyotrophic lateral sclerosis (ALS) is the predominant form of motorneuron disease (MND) and is one of the most devastatingneurodegenerative disorders, with no effective treatment or curecurrently available (Talbot, 2014). In ALS, the selective death ofupper and lower motor neurons of the cortico-motor-neuronal systemresults in progressive and terminal atrophy of muscle. As a result, themajority of affected individuals die from respiratory failure within 2-5 years of diagnosis (Rowland and Shneider, 2001). Although 10% ofALS cases have an inherited genetic mutation (Kiernan et al., 2011;

RESEARCH ARTICLE

1Menzies Research Institute Tasmania, University of Tasmania, Hobart, TAS 7000,Australia. 2Wicking Dementia Research and Education Centre, University ofTasmania, Hobart, TAS 7000, Australia.

*Author for correspondence ([email protected])

This is an Open Access article distributed under the terms of the Creative CommonsAttribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricteduse, distribution and reproduction in any medium provided that the original work is properlyattributed.

Received 1 October 2014; Accepted 20 January 2015

Ajroud-Driss and Siddique, 2014), the majority of cases are sporadic.ALS is a multifactorial disease involving RNA dysfunction, proteinmisfolding and aggregation, ER stress, axonal disruption andexcitotoxicity (Gibson and Bromberg, 2012).

Excitotoxicity is caused by over-stimulation of neuronal glutamatereceptors, leading to neuronal dysfunction and ultimately to cellulardeath (Arundine and Tymianski, 2003). Excitotoxicity has beensuggested to be a contributing factor in many neurodegenerativedisorders (Mehta et al., 2013), and there is accumulating evidence thatexcitotoxicity plays a central role in ALS (Talbot, 2014; Vucic et al.,2014). Motor neurons are vulnerable to excitotoxic insults (Bar-Peledet al., 1999; Sun et al., 2006; King et al., 2007; Mitra et al., 2013);individuals with ALS have raised levels of glutamate in theircerebrospinal fluid (Rothstein et al., 1990) and display hyper-excitability of motor neurons, which potentially starts before symptomonset (Pieri et al., 2003; Vucic and Kiernan, 2006; van Zundert et al.,2008; Bae et al., 2013). Adding to this, the astrocytes of affectedindividuals have an impaired glutamate-recycling ability (Rothstein etal., 1996; Sasaki et al., 2000; Banner et al., 2002). Furthermore, theonly drug approved for the treatment of ALS, riluzole, acts on theexcitotoxic pathway (reviewed in Morren and Galvez-Jimenez, 2012).However, it is not clear why the systemic application of riluzole haslimited therapeutic efficiency, increasing the life span of ALSindividuals for only a few months.

The cell bodies of lower motor neurons reside within the centralnervous system (CNS) in the spinal cord; their distal axons residewithin the peripheral nervous system, innervating skeletal muscle atthe neuromuscular junction (NMJ). The degeneration of NMJs isconsidered a hallmark feature of ALS onset and progression (Rochaet al., 2013). Previous studies have highlighted an under-appreciatedcapacity for NMJ plasticity prior to this NMJ retraction and synapticloss. Specifically, degeneration-resistant fibre types in the musclesof mouse models of ALS compensate for NMJ degeneration withcollateral sprouting and re-innervation (Frey et al., 2000; Schaeferet al., 2005; Pun et al., 2006). Hence, understanding the mechanismsof compensatory plasticity at the NMJ could aid in determiningstrategies to limit the denervation characteristic of ALS.

Despite the degeneration of these specialised NMJ structuresbeing a key early pathological feature of ALS, it is unknownwhether a peripheral excitotoxic insult or a cell body excitotoxicinsult results in NMJ retraction. To investigate the specific site ofvulnerability of lower motor neurons, we utilised sophisticated novelmodels of excitotoxin exposure targeted to either the cell body orthe NMJ. These studies identified that the cell bodies, but not theperipheral NMJs, of motor neurons are vulnerable to excitoxicity,suggesting that excitoxicity-directed therapeutics could be moreeffective if targeted to the lower motor neuron soma located in thespinal cord.

Identifying the primary site of pathogenesis in amyotrophic lateralsclerosis – vulnerability of lower motor neurons to proximalexcitotoxicityCatherine A. Blizzard1, Katherine A. Southam1, Edgar Dawkins1, Katherine E. Lewis1, Anna E. King2, Jayden A. Clark1 and Tracey C. Dickson1,*

Dis

ease

Mod

els

& M

echa

nism

s

216

RESULTSThe exposure of primary spinal motor neurons to kainicacid in vitro induces distal axon degenerationTo mimic the cellular interactions that occur in vivo, we usedmicrofluidic chambers to construct a compartmentalised co-culturemodel (Southam et al., 2013). The proximal compartment containedspinal glial cells and the cell bodies of primary spinal cord lower motorneurons from embryonic mice; these motor neuron axons extendedthrough axon channels to the distal compartment that containedskeletal muscle cells (Southam et al., 2013). Scanning electronmicroscopy (SEM) analyses showed that the cell bodies of the spinalmotor neurons were confined to the proximal chamber and that theiraxons extended from this compartment through interconnectingchannels to the distal compartment, which contained C2C12 musclecells that were confined to this chamber (Fig. 1A). We confirmed thecompartmentalisation with immunohistochemistry: spinal motorneurons that were immunoreactive for SMI32 (the neurofilamentprotein enriched in motor neurons) and MAP2 were present in theproximal compartment of the microfluidic chamber (Fig. 1B); SMI32-positive axons were present in the distal chamber (Fig. 1D).

To investigate the response of spinal motor neurons to site-specific excitotoxicity, 100 μm kainic acid was added to either thedistal or proximal compartment of the microfluidic chambers for24 hours, followed by fixation and analysis. Kainic acid is routinelyused as an experimental CNS excitotoxin (Kwak and Nakamura,1995; Nottingham and Springer, 2003; Mazzone et al., 2010; Mitraet al., 2013) and, unlike glutamate, is not taken up by astrocytes,ensuring a constant level of exposure of the motor neurons to thekainic acid excitotoxin. Application of kainic acid to the distal

compartment did not result in changes to either MAP2 or SMI32labelling in the proximal compartment (Fig. 1C), or SMI32 labellingin the distal compartment (Fig. 1F), relative to the vehicle controltreated cultures. However, axons located in the distal compartmentdemonstrated non-continuous immunolabelling of SMI32 followingproximal excitotoxin exposure (Fig. 1G). Quantitation of distal axondegeneration revealed a significant increase in the percentage ofdegenerating axons (displaying either a beaded or fragmentedmorphology) following proximal kainic acid exposure, incomparison with vehicle controls (degenerating axons – proximalkainic acid 55.80±2.58% versus vehicle control 43.40% ± 2.08%,P=0.0074) (Fig. 1H). Kainic acid application to the distalcompartment resulted in no significant distal axon degeneration(41.00±3.22% degenerating axons; P>0.05 versus vehicle controltreatment) (Fig. 1H). Investigation into the type of axonal damage(fragmentation of the axon versus axonal beading, Hosie et al.,2012) indicated that proximal kainic acid excitotoxicity initiated asignificant increase in axonal fragmentation (vehicle controltreatment 20.9±1.90%; distal kainic acid 23.30±2.58%; proximalkainic acid 31.20±1.60% fragmented axons, P=0.0375 versusvehicle control); however, there was no significant change in axonalbeading with either proximal or distal excitotoxin exposure (vehiclecontrol treatment 22.50±1.65%; distal kainic acid 22.60%±2.58;proximal kainic acid 25.60% ± 2.00% beaded axons; P>0.05).

The chronic exposure of lower motor neurons to a proximalexcitotoxin triggers dose- and time-dependent motor neurondegenerationTo investigate the effect of chronic excitotoxin exposure on lowermotor neuron function, we utilised osmotic mini pumps to providea continuous infusion of excitotoxin to the lumbar spinal cord. Toachieve targeted excitotoxin exposure to the yellow fluorescentprotein (YFP)-expressing lower motor neurons innervating thegastrocnemius muscle (Feng et al., 2000), we targeted the L4 regionof the spinal cord (Mohan et al., 2014); the mini pump catheter wasinserted caudal to the L4 region of the spinal cord (Fig. 2A). Toinvestigate the effect of excitotoxin exposure directly at the somataof lower motor neurons, 5 mM or 10 mM kainic acid was infusedinto the L3 to L4 region of the spinal cord over a 28-day infusiontime course. Fluoro-Ruby tracer was also included in all mini-pump-delivered infusions to ensure correct placement of the catheter in allmice. By 7 days of infusion, the subarachnoid space of the L3 to L4region had been clearly infused with Fluoro-Ruby (Fig. 2B),demonstrating a constant delivery from the mini pumps to thisregion. Immunohistochemical labelling of spinal cord sections forthe dephosphorylated neurofilament protein SMI32 showed thatYFP-expressing SMI32-positive cells in the ventral horn were alsopositive for Fluoro-Ruby (Fig. 2C-F), indicating that motor neuronshad been successfully exposed to the infusions.

After 7 days of infusion, there was no significant difference in thenumber of motor neurons in the ventral horn of the L3 to L4 spinalcord region between animals that had received the vehicle control(8.80±0.23 motor neurons) and those that had received the 10 mMkainic acid infusion (9.20±0.18 motor neurons, P>0.05 versusvehicle control) (Fig. 2K). However, after 14 days of infusion, therewas a significant decrease in the number of motor neurons inanimals that had received the 10 mM kainic acid infusion(5.40±1.35 motor neurons) in comparison to that of vehicle-treatedcontrol animals (10.00±0.78 motor neurons, P<0.0001 versus kainicacid infusion) (Fig. 2J). Interestingly, mice that had been treatedwith a 5 mM kainic acid infusion for 14 days showed no significantdifference in the number of motor neurons in comparison with that

RESEARCH ARTICLE Disease Models & Mechanisms (2015) doi:10.1242/dmm.018606

TRANSLATIONAL IMPACTClinical issueMotor neuron disease (MND) is a collective term for a devastating groupof disorders that result in muscle paralysis through the degeneration ofmotor neurons. Amyotrophic lateral sclerosis (ALS) is the most commonform of MND, in which affected individuals generally die from respiratoryfailure within 2-5 years of diagnosis. The degeneration of theneuromuscular junction is considered a hallmark of ALS onset andprogression. There is a desperate need to develop therapeutic strategiesfor this incurable disease. Previous research has provided extensiveevidence of a pivotal role for excitotoxicity in ALS pathophysiology.Despite this, little is known about the mechanisms by which excitotoxicitymight lead to the neuromuscular degeneration that characterizes ALS.In addition, the site at which this excitotoxic cascade is initiated has notbeen definitively identified.

ResultsIn this study, the authors established complementary in vitro and in vivomodels to investigate site-specific excitotoxic insult to lower motorneurons in the spinal cord. The results show that lower motor neuronsare vulnerable to somatodendritic, but not axonal, excitotoxin exposurein vitro. In vivo, cell body excitotoxin exposure leads to significant motorneuron death and neuromuscular junction retraction. Live imaging of theneuromuscular junctions demonstrated that, prior to degeneration,neuromuscular junctions have a capacity for compensatory plasticityfollowing chronic excitotoxin exposure.

Implications and future directionsThis research suggests that excitotoxicity in the spinal cord of individualswith ALS might result in a die-forward mechanism of motor neuron deathfrom the cell body outward, leading to initial distal plasticity that isfollowed by subsequent pathology and degeneration. Hence, targetedinterventions directed specifically to the cell body in the spinal cord mightbe efficacious in tackling this devastating disease. Additionally, the twomodels of motor neuron pathology established in the current study couldbe used to trial a suite of targeted therapeutics for ALS.

Dis

ease

Mod

els

& M

echa

nism

s

of vehicle control mice (5 mM kainic acid 9.58±0.70 motor neurons,P=0.7085 versus vehicle control), and they retained significantlymore motor neurons in the L3 to L4 region at 14 days of infusionthan those mice treated with 10 mM kainic acid (P<0.0001).

After 28 days of infusion, mice receiving either 5 mM or 10 mMkainic acid infusion showed significantly reduced numbers of motorneuron cell bodies in the L3 to L4 region in comparison with vehiclecontrols (motor neurons – vehicle control 9.20±0.38, 5 mM kainicacid 4.58±0.76, P=0.0003 versus vehicle control; 10 mM kainic acid2.2±0.42, P<0.0001 versus vehicle control) (Fig. 2K). Additionally,animals that had been infused with 5 mM or 10 mM kainic acid for28 days showed reduced numbers of motor neurons compared withthose that had received 5 mM or 10 mM kainic acid for 14 days(P<0.0015 and P<0.0001, respectively). At 28 days of infusion,

there was also a significant difference between animals that hadreceived 5 mM and 10 mM kainic acid (P=0.0342). Comparison ofthe effects of 5 mM and 10 mM kainic acid after 14 and 28 days ofinfusion suggests a dose-dependent and time-dependent loss ofmotor neurons in response to proximal excitotoxin exposure. High-power confocal microscopy analyses demonstrated that axonaldegeneration was also present in the corticospinal tract at 28 days ofinfusion (Fig. 2J). No pathology was present at 28 days of infusionin the cortex of treated mice (data not shown).

The chronic exposure of lower motor neurons to a proximalexcitotoxin triggers distal remodellingTo investigate the response of the NMJs in the gastrocnemiusmuscle to chronic, proximally delivered kainic acid exposure, the

217


Fig. 1. In vitro model of site-specific excitotoxicity. (A) Primary spinalmotor neurons were grown in microfluidic chambers. (A) SEM analysesdemonstrated primary motor neuron cell bodies (left) spatially separatedfrom C2C12 muscle cells (right) in the distal chamber withinterconnecting axons. (B-G) Immunocytochemistry for the cytoskeletalmarkers that are enriched in dendrites (MAP2, red) and in axons(SMI32, green). (B) Proximal chamber with vehicle control has neuronsand continuous neurites immune-positive for both MAP2 (red) andSMI32 (green). (C) The proximal compartment, following treatment ofthe distal compartment with kainic acid, has neurons and continuousneurites positive for both MAP2 (red) and SMI-32 (green). (D) Theproximal chamber, following treatment with kanic acid at the proximalend, demonstrated a loss of immunoreactivity for both MAP2 (red) andSMI32 (green). (E) The distal compartment following treatment with thevehicle control has continuous neurites that are immuno-positive forSMI32 (green). (F) The distal compartment following treatment withkainic acid at the distal end has continuous neurites that are immuno-positive for SMI32 (green). (G) The distal compartment, followingtreatment with kainic acid at the proximal end, demonstrated axonaldegeneration (green). (H) Quantitation of axon degeneration in the distalchamber after treatment with vehicle control, or application of kainic acid(KA) to the distal or proximal compartments. A significant increase intotal distal axon degeneration following treatment with kainic acid at theproximal end was seen. Furthermore, there was a significant (*P<0.05)increase in axon fragmentation in the distal compartment followingexposure of the proximal compartment to kainic acid. Error barsrepresent s.e.m. Scale bars 100 μm (A); 50 μm (B-G).

Dis

ease

Mod

els

& M

echa

nism

s

218

gastrocnemius muscle NMJs in mice that had been implanted withkainic-acid-containing osmotic mini pumps (as described above)were repeatedly imaged over an 8-week time course. Imaging wasperformed before pump insertion (at day −14), then after 14 and 28days of infusion (Fig. 3). In the animals that had been treated withthe vehicle control, the NMJs remained relatively stable (Fig. 3B-D,arrowheads indicate stable and persistent junctions). In the animalsthat received spinal cord infusion of 10 mM kainic acid, NMJsdemonstrated alterations in the morphology of the NMJ trees and inthe total junction area (Fig. 3F,G, arrowheads), and NMJs were alsolost (Fig. 3F,G, arrow). The number of branch points in the NMJtree structures was quantified. There were no significant differencesin the mean number of branch points between day −14 (3.11±0.12),

day 14 (3.02±0.18) and day 28 (3.10±0.08) for vehicle-treated mice.In the mice treated with 10 mM kainic acid, the mean branch pointarea was significantly increased at day 14 (4.11±0.33) and at day 28(4.34±0.08, P=0.0014) in comparison with that at day −14(3.36±0.12) (Fig. 3H). In mice that had been treated with the vehiclecontrol, there was no significant (P>0.05) difference in the meanNMJ area (normalised) after 14 (104.48±0.18) or 28 days ofinfusion (97.54±0.079). In mice treated with 10 mM kainic acid, themean NMJ area was significantly decreased at day 14 (64.60±0.33,P=0.073) and at day 28 (65.39±0.16, P=0.019) in comparison withthe starting NMJ area at day −14. There was no significantdifference in the mean NMJ area at day 14 and day 28 in micereceiving 10 mM kainic acid (Fig. 3I).

The chronic exposure of lower motor neurons to a proximalexcitotoxin triggers distal pathology and motor deficitsTo investigate the pathological changes occurring in distal motorneuron axons resulting from chronic excitotoxicity in the spinalcord, the NMJs in the gastrocnemius muscle were examined.Gastrocnemius muscle sections with YFP-labelled presynapticterminals (Fig. 4A,B, green) were counterstained with the post-synaptic stain α-bungarotoxin (Fig. 4A,B, red) so the completeNMJs could be investigated. The NMJs in vehicle-control animalswere intact and continuous throughout the endplate zone of thegastrocnemius muscle (Fig. 4A). By day 28, NMJs were frequentlyshrunken in the mice that had been treated with 10 mM kainic acid(Fig. 4B). To quantify the degree of degenerative pathologyoccurring over the 28-day time course, the numbers of intact(Fig. 4C) and degenerating (Fig. 4D) synapses were determined.There was a significant increase in the proportion of degeneratingsynapses relative to intact synapses at day 14 (30.85±0.07%,P=0.0161) and at day 28 (60.92±4.01%, P=0.0023) in kainic-acid-treated mice in comparison with the vehicle-treated control miceafter 28 days of infusion (2.20±0.59%) (Fig. 4E).

The motor function of vehicle-control mice and of mice treatedwith 10 mM kainic acid was investigated using the footprint test,measuring the ‘stride’, ‘sway’ and ‘stance’ length of both the mouseforelimbs and hindlimbs (Fig. 5A), as described previously(Girirajan et al., 2008). There was no significant (P>0.05) difference


Fig. 2. In vivo chronic exposure of the spinal cord to kainic acid –proximal effects. (A) Osmotic mini pumps were placed subcutaneously withthe catheter inserted into the subarachnoid space at L5 of YFP-expressingmice. (B) Fluoro-Ruby (red) labelling was present at L4 of the spinal cord ofYFP-expressing transgenic mice (green) after 7 days of infusion. (C-F) Fluoro-Ruby (C, red), YFP (D, green) with immunohistochemistry forSMI32 (E, blue) and merged (F), in the anterior ventral horn of the L4 region,demonstrated that kainic acid was being taken up by motor neurons of thespinal cord. (G,H) Toluidine Blue staining following treatment with the vehiclein the anterior ventral horn (H, inset of G) of the spinal cord at L4 on day 28of the infusion time course. (I) Toluidine Blue staining in the ventral horn ofmice treated with an infusion of 10 mM KA for 28 days (arrowheads)demonstrated cell loss. (J) Confocal microscopy demonstrated that there wasaxonal fragmentation (arrowheads) in the dorsal corticospinal tract after 28days of infusion. (K) Quantitation of Toluidine-Blue-stained motor neurons inthe anterior ventral horn of the spinal cord in the L3 to L4 region after 7, 14and 28 days of infusion (INF) demonstrated that there were no changesbetween animals that had been treated with vehicle control and 10 mM kainicacid (KA) for 7 days. There was a significant (*P<0.05) reduction in the meannumber of motor neurons at days 14 and 28 in the spinal cords that had beentreated with 10 mM kainic acid in comparison with those treated with vehiclecontrol. There was also a significant difference between animals treated with5 mM and 10 mM kainic acid after 14 and 28 days of infusion. Error barsrepresent s.e.m. Scale bars: 120 μm (B); 40 μm (C-F); 100 μm (G); 50 μm(H,I); 100 μm (J).

Dis

ease

Mod

els

& M

echa

nism

s

between the length of the vehicle-control forelimb stride(63.75±1.37 mm), stance (21.36±2.29 mm) or sway (38.48±1.02)and the length of the stride (61.38±2.17), stance (21.36±2.29) orsway (35.55±1.06) of animals that had been treated with kainic acid(Fig. 5B). In the hindlimbs, there was a significant difference instride length between the animals that had been treated with vehiclecontrol (63.72±1.17 mm) and animals that had received kainic acid(58.39±2.3 mm) (P=0.0038) (Fig. 5C). There was also a significantreduction in both sway length (37.19±1.19 mm) (P=0.0038) andstance (23.12±0.65 mm) (P=0.0004) in those animals that hadreceived the 28-day infusion of kainic acid to the spinal cordcompared with those of vehicle controls [43.21±1.02 mm (sway);26.72±0.55 mm (stance)] (Fig. 5C).

DISCUSSIONALS is a devastating disease that is characterised by the selectivedegeneration of the motor system neurons. However, the clinical

presentation and progression rate of ALS can vary (Talbot, 2014) –as such, the site of initiation and the direction of disease propagationhave not been conclusively identified. Hyperexcitability of the motorsystem is a ubiquitous clinical finding in ALS (Kiernan, 2009), withaffected individuals typically presenting with muscle fasciculationsand cramps (Kleine et al., 2008); this hyperexcitability suggests anexcitotoxic mechanism of motor neuron degeneration. Excitotoxicityis a well-recognised potential pathogenic mechanism present in ALS(Talbot, 2014; Vucic et al., 2014), yet it is unknown whether theprimary site of excitotoxic damage in ALS is at the cell body or atthe distal axon of upper or lower motor neurons. The literature onALS describes two potential phenomena of motor neurondegeneration: a dying-forward mechanism, where insult to the motorneuron cell body causes cell dysfunction and leads to neuromuscularjunction retraction (Vucic and Kiernan, 2006; Kiernan and Petri,2012); and a dying-back mechanism, where insult to the distal motorneuron axon causes axonal dysfunction and NMJ retraction,

219


Fig. 3. In vivo chronic exposure of the spinalcord to kainic acid – distal effects over a live-imaging time course. (A) Confocal microscopyimages of YFP-labelled NMJs in thegastrocnemius muscle. (B-D) NMJ trees ofvehicle-control mice were relatively stable(arrowheads) over the imaging time course atdays −14 (B), 14 (C) and 28 (D). (E-G) An NMJtree of mice treated with 10 mM kainic acid overthe imaging time course at days −14 (E), 14 (F) and 28 (G) underwent morphologicalalterations (F; arrowheads) and NMJ loss (G;arrows). (H) There was a significant increase inthe number of NMJ branch points at days 14 and28 for the kainic acid (KA)-treated mice.(I) Quantitation of the mean NMJ areademonstrated a significant (*P<0.05) reduction inthe mean NMJ area at both day 14 and day 28for the KA-treated mice compared with vehiclecontrol. There was no significant differencebetween the results at days 14 and 28. Errorbars represent s.e.m. INF, days of infusion.Scale bars: 500 μm (A); 50 μm (B-G).

Dis

ease

Mod

els

& M

echa

nism

s

220

potentially prior to cell loss (Fischer et al., 2004; reviewed inMoloney et al., 2014). Our present study is the first, to date, toinvestigate the effects of either proximal or distal excitotoxinexposure on spinal motor neuron health. In the present study, we usedboth in vitro microfluidic techniques with primary motor neurons andan in vivo model of chronic kainic acid exposure in order todemonstrate that exposure of the cell body alone to an excitotoxin(kainic acid) is sufficient to lead to substantial motor neuron deathand NMJ retraction. Additionally, we showed that neuromuscularjunction plasticity and remodelling preceded this excitotoxin-induceddegeneration. Our studies indicate that excitotoxin-exposed lowermotor neurons exhibit propagation of dysfunction in a dying-forwardmechanism, with exposure of the cell body to an excitotoxin causingNMJ retraction and degeneration, but distal excitotoxin exposurehaving little effect on the cell body.

Excitotoxin exposure at the cell body of lower motorneurons results in a dying-forward pathology with ALS-likefeaturesIn the current study, the in vitro localised application of kainic acidto either the distal axons of motor neurons or to the motor neuronsoma was achieved through culture in fluidically isolated chambers.Previous research has identified that cultured motor neurons arevulnerable to extrinsic excitotoxin exposure, resulting in axonaldysfunction and apoptotic neuronal death (Bar-Peled et al., 1999;King et al., 2007). In the present study, we expand on these findingsand show that in vitro exposure of lower motor neuron cell bodies,but not distal axons, to excitotoxic kainic acid results in substantialfragmentation and degeneration of the distal axons. Our resultssuggest that excitotoxicity triggers a dying-forward mechanism,where the insult to the lower motor neuron cell body results in distaldegeneration.

We have additionally shown that specific application of kainicacid to the spinal cord in vivo causes not only the loss of motor

neuron cell bodies but also neuromuscular junction remodelling anddegeneration. These experiments indicate that the exposure of lowermotor neuron cell bodies to an excitotoxic stimulus results in adying-forward mechanism of excitotoxic damage, characterised bythe development of ALS-like features: selective motor neuron loss,neuromuscular junction remodelling, and distal axon damage andneuromuscular junction retraction. Our results concur with those ofSun and colleagues (Sun at al., 2006) and of Mitra and colleagues(Mitra et al., 2013), who showed that in vivo chronic excitotoxinexposure can lead to neuronal cell loss in the rat. Furthermore, ourpresent study has characterised the distal NMJ degeneration arisingfrom excitotoxic signals at the proximal soma. Our data are inagreement with those from a previous study in mice, whichdemonstrate that exposure of retinal ganglion cell somas toglutamate results in a degeneration of the distal axon (Saggu et al.,2008). Thus, the dying-forward propagation of excitotoxin-inducedmotor neuron damage appears to be consistent between mice andrats.

Interestingly, the chronic in vivo application of kainic acid over a28-day time course did not result in complete motor neuron deathand functional loss. This is indicative of the presence of a potentiallyexcitotoxin-resistant population of motor neurons. Previous studieshave identified that the motor neurons that are resistant to ALSpathology are also resistant to excitotoxicity (Brockington et al.,2013; Saxena et al., 2013). Of note in the current investigation, theproportion of motor neuron cell bodies lost after 14 days of exposureto excitotoxin was greater than the proportion of neuromuscularjunctions lost at the same time point, compared to their respectivebaselines. Additionally, our live-imaging studies demonstratedremodelling of NMJ synaptic trees and an increase in synapticbranch points over the 28-day kainic acid infusion time course.Taken together, these data indicate that compensatory plasticitymight occur in response to excitotoxic damage at the cell body. It iswell documented that distal motor neuron segments have the ability


Fig. 4. In vivo chronic exposure of the spinalcord to kainic acid – distal effects. (A) After 28days of infusion (INF), NMJs treated with vehiclecontrol and labelled for YFP (green) and α-bungarotoxin (red) were intact throughout thegastrocnemius muscle. (B) At 28 days of infusionwith 10 mM kainic acid, NMJs labelled for YFP(green) and α-bungarotoxin (red) were shrunken.(C,D) NMJs were scored as either intact (C) ordegenerating (D). (E) Quantitation of thepercentage of degenerating synapsesdemonstrated a significant (*P<0.05) increase inthe percentage of degenerating synapses in themice that had been infused with 10 mM kainic acidfor 14 and 28 days, in comparison with vehiclecontrols. Error bars represent s.e.m. Scale bars:100 μm (A,B); 30 μm (C,D).

Dis

ease

Mod

els

& M

echa

nism

s

to remodel and make new junctions through collateral branching,and previous studies in mouse models of familial ALS havedemonstrated that subtypes of motor neurons demonstratecompensatory plasticity (Frey et al., 2000; Schaefer et al., 2005; Punet al., 2006). Our investigations suggest that this compensatoryplasticity might be occurring in the neurons that are potentiallyresistant to excitotoxicity. Understanding the mechanisms thatenable any excitotoxin-resistant motor neurons to resist apoptosisand undergo compensatory plasticity is an intriguing avenue forfuture investigations; confirming and harnessing these potentialmechanisms could present novel therapeutic targets for the treatmentof ALS.

It has been hypothesised that excitoxicity in ALS arises from twopotential mechanisms: (1) raised extracellular glutamate or (2)indirect excitotoxicity, occurring when extracellular levels ofglutamate are normal but neurons become more sensitised to normalglutamate levels. Indeed, motor neurons show an inherentsusceptibility to excitotoxicity due to their unique intrinsic handlingof Ca2+, which is reliant heavily upon mitochondria (Ince et al.,1993; Parone et al., 2013), and to their decreased expression of theCa2+-impermeable GluR2 glutamate receptor subunit (Van Dammeet al., 2002), allowing Ca2+ influx into motor neurons. A recentstudy in the SOD1 mouse model of ALS demonstrated that motorneurons are not intrinsically excitable prior to dysfunction, further

implying that the excitotoxicity-causing events must be extrinsic tothe motor neuron (Delestrée et al., 2014). Furthermore, levels of themain inhibitory neurotransmitter γ-aminobutyric acid (GABA) arereduced in individuals with ALS, indicating reduced interneuroninhibition at the cell body (Turner and Kiernan, 2012). Thus, it isclear that many possible sources of cell body excitotoxicity couldcontribute to ALS; our data in the present study certainly indicatethat excitotoxic stimulation in the murine spinal cord leads to ALS-like cell loss, NMJ remodelling, NMJ retraction and functionaldeficits.

Possible differences between upper and lower motor neuronaxons in the response to distal axon excitotoxicity In the present study, no distal degeneration was observed upon directexposure of lower motor neuron distal axons to the excitotoxicstimulus in vitro, indicating that distal axons from lower motorneurons are not susceptible to direct excitotoxicity. However, it isinteresting to note that we observed some corticospinal tractpathology after 28 days of infusion of kainic acid into the spinalcord. It is not clear whether this corticospinal tract pathology wasdue to a direct effect of kainic acid excitotoxin exposure on thedistal axons of upper motor neurons, or due to retraction of theupper motor neuron distal axons upon the loss of their targets. If thelatter were true, the corticospinal tract pathology could indicate that

221


Fig. 5. Forelimb and hindlimb motor functionfollowing 28 days infusion. (A) Motorperformance was investigated using analysis ofstride, sway and stance lengths for mice thathad received 28 days of infusion of kainic acidor vehicle control into the spinal cord. (B) Therewas no significant difference between the stride,stance and sway lengths of the forelimb stridestance and sway lengths in the mice that hadreceived kainic acid (KA) compared with thoseof animals receiving the vehicle control.(C) There was significantly (*P<0.05) reducedhindlimb stride, sway and stance length inkainic-acid-treated mice compared with those ofanimals treated with vehicle control. Error barsrepresent s.e.m.

Dis

ease

Mod

els

& M

echa

nism

s

222

upper motor neurons are affected by dying-back mechanisms ofdegeneration triggered by death of the lower motor neurons.However, if the former were true, this would indicate that the distalaxons of upper motor neurons, but not of lower motor neurons, aresusceptible to direct excitotoxicity. This remains an intriguingresearch question for further investigation, potentially by using liveimaging; however, limited access to the spinal cord due to theplacement of the osmotic pump prohibits this approach in thecurrent model.

The possibility of differential susceptibility to excitotoxicitybetween upper and lower motor neuron distal axons is intriguing,and might be related to the distribution of glutamate receptors on theupper and lower distal axons and their respective synapses. Theextra-synaptic expression of glutamate receptor subtypes,specifically along myelinated axons (Passafaro et al., 2001; Tovarand Westbrook, 2002; Kane-Jackson and Smith, 2003; van Zundertet al., 2004), suggests a wider role for glutamate than solely as aninter-neuronal excitatory transmitter (Araque and Perea, 2004).Excitotoxicity within the white matter occurs in a number ofdegenerative conditions – including glaucoma (Saggu et al., 2008)and multiple sclerosis (Pitt et al., 2003) – and it is also commonfollowing injury (reviewed in Lau and Tymianski, 2010). The whitematter tracts are distinctly vulnerable to degeneration in ALS,suggesting that myelinated axons in the white matter tracts arevulnerable to excitotoxicity owing to their expression of extra-synaptic glutamate receptor subtypes (Ozdinler et al., 2011; King etal., 2012). However, the majority of this regional vulnerabilityresearch has focused within the white matter of the CNS. Lowermotor neurons differ from upper motor neurons in both their spatiallocation, with their proximal cell bodies residing in the CNS (spinalcord) but their distal NMJs residing within the peripheral system (onskeletal muscle), and potentially the distribution of their extra-synaptic glutamate receptors.

In lower motor neurons, previous studies have identified post-synaptic ionotrophic glutamate receptors at the mouse NMJ (Mayset al., 2009) and at the pre-synapse in the lizard NMJ (Walder etal., 2013), and pre-synaptic ionotrophic glutamate receptororganisation occurs at the rat NMJ during re-innervation byglutamatergic neurons following transection injury (Brunelli et al.,2005), suggesting a potential role for glutamate at the NMJ.However, lower motor neurons have no known functional extra-synaptic glutamate receptor expression. This could explain the lackof direct excitotoxin-mediated damage at the lower motor neurondistal axons in our in vitro cell culture model in the current study,compared with the observation of corticospinal tract pathology inupper motor neurons after distal axon exposure to infusions ofkainic acid into the spinal cord. Understanding any differentialmechanisms of degeneration between upper and lower motorneurons might help to elucidate the key pathogenic mechanisms atwork in ALS.

ConclusionsIn summary, these investigations indicate that raised central levelsof excitotoxic stimuli could be a primary mechanism involved in thedegeneration of motor neurons, and therefore indicate thatexcitotoxicity could account for the selective motor neuron deathobserved in ALS. Understanding excitotoxic mechanisms ofneuronal cell loss and NMJ degeneration is a vital step in thepathway to developing potential new therapeutic interventions forALS. The data presented in this current study suggest that targetedinterventions, directed specifically at the cell body, might be moreefficacious in tackling this devastating disease.

MATERIALS AND METHODSIn vitro model of site-specific excitotoxin exposureMicrofluidic chambers (Xona Microfluidics) were prepared for co-culturesas previously described (Southam et al., 2013). Each chamber (microfluidicdevice with coverslip) was filled with 2.5 ng/ml of laminin in 0.001% poly-L-lysine on one side (proximal compartment) and 2.5 ng/ml laminin in0.01% collagen in the other side (distal compartment). Chambers wereincubated overnight at room temperature, substrates were then removed andthe chambers were filled with neuron initial medium (Neurobasalsupplemented with 10% heat-inactivated fetal calf serum, 2% B27 neuronalsupplement, 1% antibiotic-antimycotic, 0.5 μM L-glutamic acid and 2.5 nML-glutamine) for 2 hours before the addition of motor neurons.

Primary Mus musculus (mouse) motor neuron cultures were preparedfrom the spinal cords of embryonic day 13.5 (E13.5) C57Bl/6 embryos,purified on an Optiprep™ gradient (1.06 g/l, Sigma) and plated at 8.5×104

cells per prepared microfluidic chamber (Southam et al., 2013). Motorneurons were maintained in neuron initial medium until 7 days in vitro(DIV), and neuron subsequent medium (Neurobasal supplemented with 2%B27, 1% antibiotic-antimycotic, and 2.5 nM L-glutamine) thereafter. C2C12myoblast cells (American Type Culture Collection) were maintained inDulbecco’s modified Eagle’s medium supplemented with 10% non-heatinactivated fetal calf serum and 1% antibiotic-antimycotic) at less than 70%confluence. Cells for co-culture were maintained at passage 12 or below.C2C12 myoblasts were added to the distal compartments of microfluidicchambers when motor neurons were 4-days old and maintained in neuronsubsequent medium supplemented with 10% horse serum for 2 days,followed by neuron subsequent medium supplemented with 7% non-heatinactivated fetal calf serum for 3 days to promote differentiation. Distalcompartments were maintained in serum-free neuron subsequent mediumthereafter.

After motor neurons had been cultured for 9 DIV, co-cultures wereexposed to 100 μM kainic acid in either the proximal or distal compartmentfor 24 hours. Fluidic isolation of the treated compartment was achieved bymaintaining the fluid volume of the treated compartment at a level lowerthan that of the untreated compartment (Hosie et al., 2012). Forimmunocytochemistry analyses, primary antibodies (SMI32, 1:2000,Sternberger Monoclonals; MAP2, 1:1000, Millipore) were incubated in0.3% Triton X-100 (Sigma) in 0.1 M PBS overnight. Cells were thoroughlywashed in 0.1 M PBS and incubated with species-specific AlexaFluorsecondary antibodies (1:1000, Molecular Probes®) for 2 hours. Axonalpathology images of the distal compartment of each coverslip were capturedusing a Leica (DM LB2) fluorescence microscope fitted with a cooled CCDcamera. Axons were assessed within a 50-μm2 grid superimposed on eachimage (ImageJ) and counted as either whole, beaded (distinguishableswellings connected by sections of axon) or fragmented (disconnectedswellings). Overall degeneration was calculated as the percentage of axonsthat were beaded or fragmented.

Scanning electron microscopyCultures for SEM analysis were fixed in 4% paraformaldehyde with 2.5%glutaraldehyde. Microfluidic chambers were removed, and the cultures weredehydrated in an ethanol gradient and stored in acetone. Coverslips werecritical-point dried, sputter coated with platinum and examined using aHitachi SU-70 field emission scanning electron microscope.

In vivo model of site-specific excitotoxin exposureAll experimental procedures involving animals were approved by theAnimal Ethics Committee of the University of Tasmania and are inaccordance with the NHMRC Australian Code of Practice for the Care andUse of Animals for Scientific Purposes, 2013. Mus musculus (mice) werehoused in standard conditions (20°C, 12-12 hours light-dark cycle) withaccess to food and water ad libitum. Animals were monitored daily for signsof illness and stress. For targeting the spinal cord, osmotic mini pumps(Azlet® model 1004; infusion rate of 0.11 μl per hour for 28 days) fitted withpolyethylene catheter tubing were filled with 5 and 10 mM kainic acid(Sigma) in artificial cerebrospinal fluid (aCSF:125 mM NaCl, 5 mM KCl,10 mM D-glucose, 10 mM HEPES, 2 mM CaCl2, 2 mM MgSO4 pH 7.4; all


Dis

ease

Mod

els

& M

echa

nism

s

from Sigma). Vehicle control pumps were loaded with aCSF. To ensureappropriate placement of the catheters and focused delivery of all solutions,Fluoro-Ruby (0.01 μm; Fluorochrome, LLC) was added to both vehicle-control and excitotoxin pumps.

Previous investigations in the rat have demonstrated that a single 1 μginjection of kainic acid directly into the ventral horn resulted in excitotoxiccell loss (Mitra et al., 2013). Furthermore, using osmotic mini pumps in therat, 3.8 μg per 24 hours resulted in loss of hindlimb function and cell lossby 4 weeks (Sun et al., 2006). For proximal excitotoxin exposure in thecurrent study in mice, 10 mM and 5 mM kainic acid was applied to thesubarachnoid space of the L3 to L4 region of the spinal cord, which wasequivalent to 5 μg and 2.5 μg per 24 hours, respectively, this dosage waschosen based upon the previous studies in rats (Sun et al., 2006). Osmoticpumps were prepared 48 hours before surgery and equilibrated in 9% sterilesaline at 37°C, as previously described (Nakamura et al., 1994).

Thy-1-YFP mice [Jackson Laboratory B6;CBA-Tg(Thy1-YFP)GJrs/GfngJ] were used for this study, and adult male mice at ~12weeks were anaesthetised with a ketamine (120 mg/kg) and xylazine(10 mg/kg) cocktail, and then prepared for surgery. The L5 vertebra waspartially laminectomised and the catheter was inserted 3 mm rostral into thespinal cord to terminate at L4. The catheter was superficially fixed in placewith VetbondTM (3MTM), the osmotic pump was placed subcutaneouslyalong the spine and the incision site was sutured. For analgesia, mice weregiven Temgesic (0.01 mg/kg) for 72 hours post-surgery.

Immunohistochemistry and analysisAnimals were terminally anaesthetised (pentabarbitone sodium, 140 mg/kg),over a time course of infusion up to 28 days, and then perfused with 4%paraformaldehyde. The L3 to L5 region of the spinal column was dissectedand placed into a decalcification solution (5% nitric acid, 0.05% urea, bothfrom Sigma, in MilliQ®) overnight. The spinal columns were then washedwith 0.1 M phosphate buffered saline (PBS) and placed into a series ofsucrose solutions (4%, 18% and 30%, Sigma) in 0.1 M PBS. Spinal cordsections were serially cut into 20 μm sections and mounted ontosuperfrostTM plus slides (Thermo Scientific) for cell body quantitation, andwere cut into 60 μm sections for free floating immunohistochemistry. TheL5 region was used to check for correct catheter placement. For ToluidineBlue staining, sections were incubated in 0.05% Toluidine Blue in 0.1 MPBS and then thoroughly washed. Gastrocnemius muscles were dissected,placed into a series of sucrose solutions (4%, 18% and 30%, Sigma) in 0.1 M PBS for cryo-protection and sectioned at 80 μm.

For immunohistochemistry, sections were incubated with primaryantibodies (SMI32, 1:2000, Sternberger Monoclonals) diluted in 0.3% TritonX-100 (Sigma) in 0.1 M PBS and incubated overnight. Sections werethoroughly washed in 0.1 M PBS and incubated with species-specificAlexaFluor secondary antibodies (1:1000, Molecular Probes®) for 2 hours.Muscle sections were stained with α-bungarotoxin (1:100, MolecularProbes®) in 0.1 M PBS for synapse quantitation. Images were captured ona Zeiss LSM 510 confocal microscope using Zen software. Z-stacks wereflattened in ImageJ and processed in ImageJ and Photoshop (Adobe).

Repeated live imaging of gastrocnemius muscleMice were anaesthetised with the ketamine (120 mg/kg) and xylazine(10 mg/kg) cocktail described above, and imaging of the lateral gastrocnemiusmuscle was performed essentially as described by Wigston (Wigston, 1990)with modifications. Briefly, mice were placed on a custom-designed heatedstage that restrained the hindlimbs to avoid movement artefacts. The lateralsurface of the right hind leg was shaved and a small incision was made in theskin overlying the right lateral gastrocnemius muscle, and the muscle wasexposed. Images were captured on a Zeiss LSM 510 confocal microscopeusing a 488-nm confocal laser, 10× 0.3 NA objective with Zen software.Following imaging of NMJs, animals were removed from the microscope, theincisions were sutured and the animals allowed to recover. Fourteen mice(seven receiving vehicle control, seven receiving 10 mM kainic acid) wereimaged over a 6-week time course on days −14, 14 and 28, −14 correspondsto 14 days before the pump insertion, day 0 being the day that the pumps wereinserted. Z-stacks were analysed by using ImageJ.

Footprint testTo obtain footprints, the hind- and forefeet of the mice were coated with twodifferent nontoxic paints, respectively. The animals were then allowed towalk along a 50-cm-long, 10-cm-wide runway (with 10-cm-high walls) intoan enclosed box (Girirajan et al., 2008). All mice were given two trainingruns and one test run. After completion, mice were transcardially perfused.

Statistical analysisStudent’s t-tests and one-way and two-way analysis of variance with orwithout repeated measures (ANOVAs), performed in GraphPad PrismSoftware, were used where appropriate. Post-hoc comparisons were performedusing Bonferroni’s correction for multiple comparisons. P<0.05 wasconsidered significant. Data is reported ± standard error of the mean (s.e.m.).

AcknowledgementsWe acknowledge the expertise of Mr G.McCormack in preparing figureschematics.

Competing interestsThe authors declare no competing or financial interests.

Author contributionsC.B. performed all in vivo experiments and wrote the manuscript. K.S. performedthe majority of in vitro experiments. E.D. provided expertise on experiments andcontributed to the writing of the manuscript. K.L. contributed to data interpretationand writing of the manuscript. A.K. contributed to the design of experiments. J.C.assisted in experiments. T.C. designed experiments and contributed to the writingof the manuscript.

FundingThis work was supported by the National Health and Medical Research Council;Motor Neurone Disease Research Institute Australia; and the Tasmanian MasonicCentenary Medical Research Foundation.

ReferencesAjroud-Driss, S. and Siddique, T. (2014). Sporadic and hereditary amyotrophic lateral

sclerosis (ALS). Biochim. Biophys. Acta. Araque, A. and Perea, G. (2004). Glial modulation of synaptic transmission in culture.

Glia 47, 241-248. Arundine, M. and Tymianski, M. (2003). Molecular mechanisms of calcium-

dependent neurodegeneration in excitotoxicity. Cell Calcium 34, 325-337. Bae, J. S., Simon, N. G., Menon, P., Vucic, S. and Kiernan, M. C. (2013). The

puzzling case of hyperexcitability in amyotrophic lateral sclerosis. J Clin Neurol 9,65-74.

Banner, S. J., Fray, A. E., Ince, P. G., Steward, M., Cookson, M. R. and Shaw, P. J.(2002). The expression of the glutamate re-uptake transporter excitatory amino acidtransporter 1 (EAAT1) in the normal human CNS and in motor neurone disease: animmunohistochemical study. Neuroscience 109, 27-44.

Bar-Peled, O., O’Brien, R. J., Morrison, J. H. and Rothstein, J. D. (1999). Culturedmotor neurons possess calcium-permeable AMPA/kainate receptors. Neuroreport10, 855-859.

Brockington, A., Ning, K., Heath, P. R., Wood, E., Kirby, J., Fusi, N., Lawrence, N.,Wharton, S. B., Ince, P. G. and Shaw, P. J. (2013). Unravelling the enigma ofselective vulnerability in neurodegeneration: motor neurons resistant to degenerationin ALS show distinct gene expression characteristics and decreased susceptibility toexcitotoxicity. Acta Neuropathol. 125, 95-109.

Brunelli, G., Spano, P., Barlati, S., Guarneri, B., Barbon, A., Bresciani, R. andPizzi, M. (2005). Glutamatergic reinnervation through peripheral nerve graft dictatesassembly of glutamatergic synapses at rat skeletal muscle. Proc. Natl. Acad. Sci.USA 102, 8752-8757.

Delestrée, N., Manuel, M., Iglesias, C., Elbasiouny, S. M., Heckman, C. J. andZytnicki, D. (2014). Adult spinal motoneurones are not hyperexcitable in a mousemodel of inherited amyotrophic lateral sclerosis. J. Physiol. 592, 1687-1703.

Feng, G., Mellor, R. H., Bernstein, M., Keller-Peck, C., Nguyen, Q. T., Wallace, M.,Nerbonne, J. M., Lichtman, J. W. and Sanes, J. R. (2000). Imaging neuronalsubsets in transgenic mice expressing multiple spectral variants of GFP. Neuron 28,41-51.

Fischer, L. R., Culver, D. G., Tennant, P., Davis, A. A., Wang, M., Castellano-Sanchez, A., Khan, J., Polak, M. A. and Glass, J. D. (2004). Amyotrophic lateralsclerosis is a distal axonopathy: evidence in mice and man. Exp. Neurol. 185, 232-240.

Frey, D., Schneider, C., Xu, L., Borg, J., Spooren, W. and Caroni, P. (2000). Earlyand selective loss of neuromuscular synapse subtypes with low sproutingcompetence in motoneuron diseases. J. Neurosci. 20, 2534-2542.

Gibson, S. B. and Bromberg, M. B. (2012). Amyotrophic lateral sclerosis: drugtherapy from the bench to the bedside. Semin. Neurol. 32, 173-178.

Girirajan, S., Patel, N., Slager, R. E., Tokarz, M. E., Bucan, M., Wiley, J. L. andElsea, S. H. (2008). How much is too much? Phenotypic consequences of Rai1overexpression in mice. Eur. J. Hum. Genet. 16, 941-954.

223


Dis

ease

Mod

els

& M

echa

nism

s

224

Hosie, K. A., King, A. E., Blizzard, C. A., Vickers, J. C. and Dickson, T. C. (2012).Chronic excitotoxin-induced axon degeneration in a compartmented neuronal culturemodel. ASN Neuro 4, e00076.

Ince, P., Stout, N., Shaw, P., Slade, J., Hunziker, W., Heizmann, C. W. andBaimbridge, K. G. (1993). Parvalbumin and calbindin D-28k in the human motorsystem and in motor neuron disease. Neuropathol. Appl. Neurobiol. 19, 291-299.

Kane-Jackson, R. and Smith, Y. (2003). Pre-synaptic kainate receptors in GABAergicand glutamatergic axon terminals in the monkey globus pallidus. Neuroscience 122,285-289.

Kiernan, M. C. (2009). Hyperexcitability, persistent Na+ conductances andneurodegeneration in amyotrophic lateral sclerosis. Exp. Neurol. 218, 1-4.

Kiernan, M. C., Vucic, S., Cheah, B. C., Turner, M. R., Eisen, A., Hardiman, O.,Burrell, J. R. and Zoing, M. C. (2011). Amyotrophic lateral sclerosis. Lancet 377,942-955.

Kiernan, M. C. and Petri, S. (2012). Hyperexcitability and amyotrophic lateralsclerosis. Neurology 78, 1544-1545.

King, A. E., Dickson, T. C., Blizzard, C. A., Foster, S. S., Chung, R. S., West, A. K.,Chuah, M. I. and Vickers, J. C. (2007). Excitotoxicity mediated by non-NMDAreceptors causes distal axonopathy in long-term cultured spinal motor neurons. Eur.J. Neurosci. 26, 2151-2159.

King, A. E., Blizzard, C. A., Southam, K. A., Vickers, J. C. and Dickson, T. C.(2012). Degeneration of axons in spinal white matter in G93A mSOD1 mousecharacterized by NFL and α-internexin immunoreactivity. Brain Res. 1465, 90-100.

Kleine, B. U., Stegeman, D. F., Schelhaas, H. J. and Zwarts, M. J. (2008). Firingpattern of fasciculations in ALS: evidence for axonal and neuronal origin. Neurology70, 353-359.

Kwak, S. and Nakamura, R. (1995). Acute and late neurotoxicity in the rat spinal cordin vivo induced by glutamate receptor agonists. J. Neurol. Sci. 129 Suppl., 99-103.

Lau, A. and Tymianski, M. (2010). Glutamate receptors, neurotoxicity andneurodegeneration. Pflugers Arch. 460, 525-542.

Mays, T. A., Sanford, J. L., Hanada, T., Chishti, A. H. and Rafael-Fortney, J. A.(2009). Glutamate receptors localize postsynaptically at neuromuscular junctions inmice. Muscle Nerve 39, 343-349.

Mazzone, G. L., Margaryan, G., Kuzhandaivel, A., Nasrabady, S. E., Mladinic, M.and Nistri, A. (2010). Kainate-induced delayed onset of excitotoxicity with functionalloss unrelated to the extent of neuronal damage in the in vitro spinal cord.Neuroscience 168, 451-462.

Mehta, A., Prabhakar, M., Kumar, P., Deshmukh, R. and Sharma, P. L. (2013).Excitotoxicity: bridge to various triggers in neurodegenerative disorders. Eur. J.Pharmacol. 698, 6-18.

Mitra, N. K., Goh, T. E., Bala Krishnan, T., Nadarajah, V. D., Vasavaraj, A. K. andSoga, T. (2013). Effect of intra-cisternal application of kainic acid on the spinal cordand locomotor activity in rats. Int. J. Clin. Exp. Pathol. 6, 1505-1515.

Mohan, R., Tosolini, A. P., Morris, R. (2014). Targeting the motor end plates in themouse hindlimb gives access to a greater number of spinal cord motor neurons: anapproach to maximize retrograde transport. Neuroscience. 274, 318-330.

Moloney, E. B., de Winter, F. and Verhaagen, J. (2014). ALS as a distal axonopathy:molecular mechanisms affecting neuromuscular junction stability in thepresymptomatic stages of the disease. Front. Neurosci. 8, 252.

Morren, J. A. and Galvez-Jimenez, N. (2012). Current and prospective disease-modifying therapies for amyotrophic lateral sclerosis. Expert Opin. Investig. Drugs21, 297-320.

Nakamura, R., Kamakura, K. and Kwak, S. (1994). Late-onset selective neuronaldamage in the rat spinal cord induced by continuous intrathecal administration ofAMPA. Brain Res. 654, 279-285.

Nottingham, S. A. and Springer, J. E. (2003). Temporal and spatial distribution ofactivated caspase-3 after subdural kainic acid infusions in rat spinal cord. J. Comp.Neurol. 464, 463-471.

Ozdinler, P. H., Benn, S., Yamamoto, T. H., Güzel, M., Brown, R. H., Jr and Macklis,J. D. (2011). Corticospinal motor neurons and related subcerebral projectionneurons undergo early and specific neurodegeneration in hSOD1G93A transgenicALS mice. J. Neurosci. 31, 4166-4177.

Parone, P. A., Da Cruz, S., Han, J. S., McAlonis-Downes, M., Vetto, A. P., Lee, S.K., Tseng, E. and Cleveland, D. W. (2013). Enhancing mitochondrial calciumbuffering capacity reduces aggregation of misfolded SOD1 and motor neuron celldeath without extending survival in mouse models of inherited amyotrophic lateralsclerosis. J. Neurosci. 33, 4657-4671.

Passafaro, M., Piëch, V. and Sheng, M. (2001). Subunit-specific temporal and spatialpatterns of AMPA receptor exocytosis in hippocampal neurons. Nat. Neurosci. 4,917-926.

Pieri, M., Albo, F., Gaetti, C., Spalloni, A., Bengtson, C. P., Longone, P., Cavalcanti,S. and Zona, C. (2003). Altered excitability of motor neurons in a transgenic mousemodel of familial amyotrophic lateral sclerosis. Neurosci. Lett. 351, 153-156.

Pitt, D., Nagelmeier, I. E., Wilson, H. C. and Raine, C. S. (2003). Glutamate uptakeby oligodendrocytes: Implications for excitotoxicity in multiple sclerosis. Neurology61, 1113-1120.

Pun, S., Santos, A. F., Saxena, S., Xu, L. and Caroni, P. (2006). Selectivevulnerability and pruning of phasic motoneuron axons in motoneuron diseasealleviated by CNTF. Nat. Neurosci. 9, 408-419.

Rocha, M. C., Pousinha, P. A., Correia, A. M., Sebastião, A. M. and Ribeiro, J. A.(2013). Early changes of neuromuscular transmission in the SOD1(G93A) micemodel of ALS start long before motor symptoms onset. PLoS ONE 8, e73846.

Rothstein, J. D., Tsai, G., Kuncl, R. W., Clawson, L., Cornblath, D. R., Drachman,D. B., Pestronk, A., Stauch, B. L. and Coyle, J. T. (1990). Abnormal excitatoryamino acid metabolism in amyotrophic lateral sclerosis. Ann. Neurol. 28, 18-25.

Rothstein, J. D., Dykes-Hoberg, M., Pardo, C. A., Bristol, L. A., Jin, L., Kuncl, R.W., Kanai, Y., Hediger, M. A., Wang, Y., Schielke, J. P. et al. (1996). Knockout ofglutamate transporters reveals a major role for astroglial transport in excitotoxicityand clearance of glutamate. Neuron 16, 675-686.

Rowland, L. P. and Shneider, N. A. (2001). Amyotrophic lateral sclerosis. N. Engl. J.Med. 344, 1688-1700.

Saggu, S. K., Chotaliya, H. P., Cai, Z., Blumbergs, P. and Casson, R. J. (2008). Thespatiotemporal pattern of somal and axonal pathology after perikaryal excitotoxicinjury to retinal ganglion cells: a histological and morphometric study. Exp. Neurol.211, 52-58.

Sasaki, S., Komori, T. and Iwata, M. (2000). Excitatory amino acid transporter 1 and 2immunoreactivity in the spinal cord in amyotrophic lateral sclerosis. ActaNeuropathol. 100, 138-144.

Saxena, S., Roselli, F., Singh, K., Leptien, K., Julien, J. P., Gros-Louis, F. andCaroni, P. (2013). Neuroprotection through excitability and mTOR required in ALSmotoneurons to delay disease and extend survival. Neuron 80, 80-96.

Schaefer, A. M., Sanes, J. R. and Lichtman, J. W. (2005). A compensatorysubpopulation of motor neurons in a mouse model of amyotrophic lateral sclerosis.J. Comp. Neurol. 490, 209-219.

Southam, K. A., King, A. E., Blizzard, C. A., McCormack, G. H. and Dickson, T. C.(2013). Microfluidic primary culture model of the lower motor neuron-neuromuscularjunction circuit. J. Neurosci. Methods 218, 164-169.

Sun, H., Kawahara, Y., Ito, K., Kanazawa, I. and Kwak, S. (2006). Slow and selectivedeath of spinal motor neurons in vivo by intrathecal infusion of kainic acid: implicationsfor AMPA receptor-mediated excitotoxicity in ALS. J. Neurochem. 98, 782-791.

Talbot, K. (2014). Amyotrophic lateral sclerosis: cell vulnerability or systemvulnerability? J. Anat. 224, 45-51.

Tovar, K. R. and Westbrook, G. L. (2002). Mobile NMDA receptors at hippocampalsynapses. Neuron 34, 255-264.

Turner, M. R. and Kiernan, M. C. (2012). Does interneuronal dysfunction contribute toneurodegeneration in amyotrophic lateral sclerosis? Amyotroph. Lateral Scler. 13,245-250.

Van Damme, P., Van Den Bosch, L., Van Houtte, E., Callewaert, G. andRobberecht, W. (2002). GluR2-dependent properties of AMPA receptors determinethe selective vulnerability of motor neurons to excitotoxicity. J. Neurophysiol. 88,1279-1287.

van Zundert, B., Yoshii, A. and Constantine-Paton, M. (2004). Receptorcompartmentalization and trafficking at glutamate synapses: a developmentalproposal. Trends Neurosci. 27, 428-437.

van Zundert, B., Peuscher, M. H., Hynynen, M., Chen, A., Neve, R. L., Brown, R.H., Jr, Constantine-Paton, M. and Bellingham, M. C. (2008). Neonatal neuronalcircuitry shows hyperexcitable disturbance in a mouse model of the adult-onsetneurodegenerative disease amyotrophic lateral sclerosis. J. Neurosci. 28, 10864-10874.

Vucic, S. and Kiernan, M. C. (2006). Axonal excitability properties in amyotrophiclateral sclerosis. Clin. Neurophysiol. 117, 1458-1466.

Vucic, S., Rothstein, J.D., and Kiernan, M. C. (2014). Advances in treatingamyotrophic lateral sclerosis: insights from pathophysiological studies. TrendsNeurosci. 37, 433-442.

Walder, K. K., Ryan, S. B., Bzdega, T., Olszewski, R. T., Neale, J. H. and Lindgren,C. A. (2013). Immunohistological and electrophysiological evidence that N-acetylaspartylglutamate is a co-transmitter at the vertebrate neuromuscular junction.Eur. J. Neurosci. 37, 118-129.

Wigston, D. J. (1990). Repeated in vivo visualization of neuromuscular junctions inadult mouse lateral gastrocnemius. J. Neurosci. 10, 1753-1761.


Dis

ease

Mod

els

& M

echa

nism

s

Documents

Identifying the primary site of pathogenesis in ... · slow the progression of amyotrophic lateral sclerosis (ALS). ALS is a disorder with heterogeneous onset, which then leads to