RESEARCH ARTICLE

Sporadic amyotrophic lateral sclerosis (SALS) – skeletal muscleresponse to cerebrospinal fluid from SALS patients in a rat modelShruthi Shanmukha1, Gayathri Narayanappa2, Atchayaram Nalini3, Phalguni Anand Alladi1,* andTrichur R. Raju1,*

ABSTRACTSkeletal muscle atrophy is the most prominent feature of amyotrophiclateral sclerosis (ALS), an adult-onset neurodegenerative disease ofmotor neurons. However, the contribution of skeletal muscle todisease progression remains elusive. Our previous studies haveshown that intrathecal injection of cerebrospinal fluid from sporadicALS patients (ALS-CSF) induces several degenerative changes inmotor neurons and glia of neonatal rats. Here, we describe variouspathologic events in the rat extensor digitorum longus musclefollowing intrathecal injection of ALS-CSF. Adenosinetriphosphatase staining and electron microscopic (EM) analysisrevealed significant atrophy and grouping of type 2 fibres in ALS-CSF-injected rats. Profound neuromuscular junction (NMJ) damage,such as fragmentation accompanied by denervation, were revealedby α-bungarotoxin immunostaining. Altered expression of key NMJproteins, rapsyn and calpain, was also observed by immunoblotting.In addition, EM analysis showed sarcolemmal folding, Z-linestreaming, structural alterations of mitochondria and dilatedsarcoplasmic reticulum. The expression of trophic factors wasaffected, with significant downregulation of vascular endothelialgrowth factor (VEGF), marginal reduction in insulin-like growthfactor-1 (IGF-1), and upregulation of brain-derived neurotrophicfactor (BDNF) and glial-derived neurotrophic factor (GDNF).However, motor neurons might be unable to harness the enhancedlevels of BDNF and GDNF, owing to impaired NMJs. We propose thatALS-CSF triggers motor neuronal degeneration, resulting inpathological changes in the skeletal muscle. Muscle damagefurther aggravates the motor neuronal pathology, because of theinterdependency between them. This sets in a vicious cycle, leadingto rapid and progressive loss of motor neurons, which could explainthe relentless course of ALS.

This article has an associated First Person interview with the firstauthor of the paper.

KEY WORDS: Amyotrophic lateral sclerosis,Immunohistochemistry, Electron microscopy, NMJ, Muscle atrophy,Trophic factors

INTRODUCTIONAmyotrophic lateral sclerosis (ALS) is a motor neuron degenerativedisease that affects upper and lower motor neurons, leading toperceptibly severe muscle atrophy. There is a lack of a suitableanimal model for investigating the sporadic form of ALS (SALS),which accounts for 90% of cases. The complex interplay betweendifferent cell types, such as motor neurons, astrocytes, microglia,Schwann cells and skeletal muscles, in ALS pathogenesis needs tobe investigated.

The loss of muscle mass is one of the hallmark features of ALS.The degeneration of motor neurons and denervation leads to atrophyof muscle fibres in ALS (Baloh et al., 2007; Brooke and Engel,1969; Dobrowolny et al., 2008; Telerman-Toppet and Coërs, 1978;Wong and Martin, 2010). The large-calibre, fast-fatiguable motorneurons which innervate the type 2 muscle fibres are known to beselectively vulnerable in ALS (Hegedus et al., 2007; Pun et al., 2006).Following denervation, the neighbouring motor axons compensatefor the loss, resulting in re-innervation of the denervated fibres byaxonal sprouting, ultimately leading to grouping of muscle fibres(Baloh et al., 2007; Schaefer et al., 2005; Telerman-Toppet andCoërs, 1978). However, this compensatory mechanism fails tosupport regeneration, as the disease advances and results inprogressive decrease in muscle strength.

Motor neurons are dependent on skeletal muscles for thecontinuous supply of neurotrophic factors for their survival andfunctioning (Henderson et al., 1998; Oppenheim, 1996). The skeletalmuscle is a rich source of several neurotrophic factors. Brain-derivedneurotrophic factor (BDNF) prevents motor neuron cell death(Koliatsos et al., 1993; Oppenheim et al., 1992) and mediates anti-apoptotic effects (Almeida et al., 2005). Insulin-like growth factor-1(IGF-1) protects motor neurons during development and post-injuryrecovery (Neff et al., 1993), and aids in the restoration ofneuromuscular junction (NMJ) function (Messi and Delbono,2003). Glial-derived neurotrophic factor (GDNF) promotes thesurvival of motor neurons; overexpression of GDNF causeshyperinnervation of muscle (Henderson et al., 1994; Nguyen et al.,1998). Vascular endothelial growth factor (VEGF) has potentneurotrophic and mitogenic activity, and increases axonaloutgrowth and survival of neurons (Annex et al., 1998; Sondellet al., 1999). Interestingly, deletion of the hypoxia-response elementin its promoter region causes reduced VEGF expression in the spinalcord, resulting in motor neuron degeneration very similar to that seenin ALS (Oosthuyse et al., 2001). It also induces axonal regenerationafter ischemic injury and muscle innervation, acting through nervegrowth factor (NGF)/GDNF signalling (Shvartsman et al., 2014), andsignifying the influence of VEGF on other trophic factors.Received 12 September 2017; Accepted 5 March 2018

1Department of Neurophysiology, National Institute of Mental Health andNeurosciences, Hosur Road, Bangalore 560 029, India. 2Department ofNeuropathology, National Institute of Mental Health and Neurosciences, HosurRoad, Bangalore 560 029, India. 3Department of Neurology, National Institute ofMental Health and Neurosciences, Hosur Road, Bangalore 560 029, India.

*Authors for correspondence (trraju.nimhans@gmail.com;alladiphalguni@gmail.com)

S.S., 0000-0001-5659-2250; P.A.A., 0000-0002-2876-3478; T.R.R., 0000-0002-7742-4050

This is an Open Access article distributed under the terms of the Creative Commons AttributionLicense (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use,distribution and reproduction in any medium provided that the original work is properly attributed.

1

© 2018. Published by The Company of Biologists Ltd | Disease Models & Mechanisms (2018) 11, dmm031997. doi:10.1242/dmm.031997

Disea

seModels&Mechan

isms

Previous studies from our laboratory on a rat model and NSC34cell line have demonstrated the various degenerative changes uponexposure to cerebrospinal fluid (CSF) from SALS patients (ALS-CSF); these are listed in Table 1. Understanding the pathogenesis ofALS in the skeletal muscle, the end organ, appears more appropriateas it can be an efficient target for drug delivery. In this view, thecurrent study focuses on various aspects of skeletal muscle changesin rats following intrathecal injection of ALS-CSF.

RESULTSALS-CSF induces atrophy of type 2 fibres in extensordigitorum longus muscle of rat neonatesThe muscle fibres are broadly divided into type 1 and type 2 fibresbased on their adenosine triphosphatase (ATPase) staining. Pre-incubation with basic solution inhibits myosin ATPase in type 1fibres, causing them to be lightly stained, whereas type 2 fibres aredarkly stained at this pH, allowing their clear distinction. The normalpattern of type 1 and type 2 fibres and polygonal shape of myofibresare altered in ALS-CSF-injected animals compared with those incontrol groups. ALS-CSF caused significant atrophy, reflected byangulated myofibres and grouping of type 2 fibres in the extensordigitorum longus (EDL) muscle (as shown by ATPase staining at

pH 9.4) in ALS-CSF-injected (ALS) rats compared with normalcontrol (NC) rats (Fig. 1A,B). In addition, electron microscopic(EM) analysis of the muscle provides evidence of atrophy (Fig. 1C,D). Quantification of the cross-sectional area (CSA) of the musclefibres showed a significant reduction in the size of type 2 fibres,suggesting severe atrophy, whereas type 1 fibres were unaltered(Fig. 1E,F; **P<0.01, NC versus ALS).

ALS-CSF causes profound changes in NMJ structure in ratneonatesWe also assessed the structural integrity of NMJs. In NC, shamcontrol (SC) and non-ALS-CSF (NALS) rats, NMJs appear as nearpretzel-shaped structures (Fig. 2A-C); in contrast, NMJs in the ALSgroup were found to be fragmented and diffused (Fig. 2D). Therewas also a reduction in the total synaptic area in the ALS group(Fig. 2G; **P<0.01, NC versus ALS; $P<0.05, SC versus ALS).However, there was no difference in the expression of acetylcholinereceptor (AChR) protein between the groups (Fig. 2H).

In addition, double labelling with anti-200KD neurofilament fornerve terminals and Alexa Fluor 488-conjugated α-bungaratoxin forNMJs showed marked denervation in the ALS group compared withthe NC group (Fig. 2E,F). Further, the expression of two key proteinsinvolved in the clustering of AChR subunits, rapsyn and calpain, wasestimated. The expression of rapsyn, which helps in the clustering ofAChR subunits was decreased, whereas calpain, which counteractsthe rapsyn action, was upregulated, resulting in the disruption ofNMJ structure (Fig. 2I,J; *P<0.05, NC versus ALS).

Ultrastructural changesTo investigate the detailed pathological changes in the skeletalmuscle of the ALS animals, we carried out ultrastructural analysis ofmuscle tissue from NC, SC, NALS and ALS groups. The NC groupshowed a double-layered sarcolemma consisting of outer basementand inner plasma membranes, with a normal cytoskeletaldistribution, whereas the sarcolemma was folded in the ALSgroup, suggesting the loss of membrane integrity and confirmingatrophy (Fig. 3A,B). In addition, the normal striated appearance ofthe skeletal muscle was lost in the diseased group. The fibres in EDLmuscles from ALS rats showed clear alterations in the bandingpatterns. There was a distortion of filamentous pattern withstreaming of Z-band material or misalignment of sarcomeres and afocal loss of myofilaments (Fig. 3C,D). In the control groups(NC, SC and NALS), normal mitochondria with well-defined cristaeand membrane structure were seen in subsarcolemmmal andintermyofibrillar regions. However, in the ALS group, themitochondria had altered cristae, vacuolation and abnormal shape(Fig. 3E,F); additionally, many mitochondria had accumulated lipiddroplets. Marked variation in the mitochondrial morphologyconfirmed the damage caused by ALS-CSF. In addition tomitochondrial damage, the extensive network of longitudinallyoriented tubules of sarcoplasmic reticulum (SR) was damaged in theALS group compared with the control groups. Dilated SR at thesubsarcolemmmal and intermyofibrillar regions was observed,which might result in impaired calcium homeostasis (Fig. 3G,H).

Oxidative stressThe ultrastructural analysis showed significant mitochondrial damagein ALS-CSF-injected rat skeletal muscle, and it could be associatedwith oxidative stress, which was estimated by measuring the levels ofmalondialdehyde (MDA), a marker of lipid peroxidation in musclesamples. The results confirmed increased oxidative stress in theskeletal muscle as we found a 2.5-fold increase in nmol of MDA/mg

Table 1. Summary of ALS-CSF-mediated degenerative changes

Increased neurofilamentphosphorylation in rat ventral hornneurons

(Rao et al., 1995)

Neuroprotective effects of (-)-deprenyland cyclophosphamide; alleviationof ALS-CSF-induced pathologicalchanges in vivo

(Shahani et al., 2001, 2004)

Reduced level of glial glutamatetransporter 1 in vitro and in vivo

(Shobha et al., 2007)

Enhanced glial fibrillary acidic proteinand S100β expression in vitro

(Shobha et al., 2010)

Decreased Nav1.6 and Kv1.6 channelexpression in motor neurons in vitroand in vivo

(Gunasekaran et al., 2009)

Organelle dysfunction:Golgi fragmentation in spinal motorneurons in vivoIncreased endoplasmic reticulumstress in vitro and in vivoMitochondrial and lysosomaldysfunction in vivo

(Ramamohan et al., 2007; Sharmaet al., 2015; Vijayalakshmi et al.,2011)

Reduced choline acetyl transferaseexpression and increasedaggregation of phosphorylatedneurofilaments in vitro

(Vijayalakshmi et al., 2009)

Downregulated expression of trophicfactor genes in vivo

(Deepa et al., 2011)

Neuroprotective effects of VEGF andBDNF; attenuation of thedegenerative changes induced byALS-CSF

(Kulshreshtha et al., 2011;Shantanu et al., 2017; Shruthiet al., 2017; Vijayalakshmi et al.,2015)

Reduced muscle strength offorelimbs, impaired performance inrotarod test, and increased firingrate of the motor cortex neurons invivo

(Sankaranarayani et al., 2010;Sankaranarayani et al., 2014)

Differential regulation of micro RNA-206 in the skeletal muscle in vivo

(Sumitha et al., 2014)

Astrocyte- and microglia-mediatedneuroinflammation in vitro

(Mishra et al., 2016, 2017)

Chitotriosidase-1 is a putative toxicfactor present in ALS-CSF

(Varghese et al., 2013)

2

RESEARCH ARTICLE Disease Models & Mechanisms (2018) 11, dmm031997. doi:10.1242/dmm.031997

Disea

seModels&Mechan

isms

of protein in the ALS animals (Fig. 4A; ***P<0.001, NC versusALS; $$P<0.01, SC versus ALS; ###P<0.001, NALS versus ALS).We observed that muscle samples from ALS rats showed elevatedMDA content, compared with those from controls. To analysewhether increased oxidative stress in ALS is linked with theantioxidant function, we assayed the activities of the constitutiveantioxidant enzymes superoxide dismutase (SOD), glutathionereductase (GR) and thioredoxin reductase. There was anupregulation of thioredoxin reductase activity in the ALS group,probably as an adaptive response to increased oxidative stress(Fig. 4B; ****P<0.0001, NC, SC and NALS versus ALS). Further,

SOD activity was significantly downregulated in the ALS group and,surprisingly, upregulated in the NALS group (Fig. 4C; *P<0.05, NCversusNALS; **P<0.01, NC versusALS; $$P<0.01, SC versus ALS;###P<0.001, NALS versus ALS), while GR activity remainedunaltered (Fig. 4D).

Altered expression of BDNF and IGF-1 in ALSThe immunofluorescence analysis of BDNF showed minimalexpression in the NC group (Fig. 5A). The expression was increasedin the ALS group (Fig. 5D). BDNF expression was found to beuniformly distributed along the sarcolemmal region, and there was a

Fig. 1. ALS-CSF induces atrophy of type 2 fibres in EDL muscle of neonatal rats. (A,B) ATPase-stained cross-sections of EDL muscles from NC andALS-CSF-injected (ALS) rats. Type 1 fibres are lightly stained and type 2 fibres are darkly stained at pH 9.4. Note the altered pattern of type 1 and type 2 fibres inthe ALS group compared with the NC group. Rounding, angulation and grouping of type 2 fibres are also seen. Scale bars: 120 μm. (C,D) Electron micrographsshowing NC (C) and ALS (D) groups. Rounded (indicated by hash symbol) and angulated fibres (indicated by asterisk) can be seen in the ALS group,suggesting atrophy of the muscle. (E,F) Mean cross-sectional area (CSA) of muscle fibres. Note that there is no alteration in the CSA of type 1 fibres (E),whereas type 2 fibres showed a significant reduction in CSA, in the ALS group compared with the other control groups (F, **P<0.01, NC versus ALS). n=5 induplicates. Statistical significance was calculated using one-way ANOVA followed by Tukey’s post hoc test.

3

RESEARCH ARTICLE Disease Models & Mechanisms (2018) 11, dmm031997. doi:10.1242/dmm.031997

Disea

seModels&Mechan

isms

punctate staining pattern in the sarcoplasm. The co-labelled BDNFand IGF-1 for the NC (Fig. 5C) and ALS (Fig. 5F) groups areprovided, where the altered expression can be appreciated better.The quantification of BDNF expression confirmed the increasedexpression in the ALS group (Fig. 5G; *P<0.05, NC versus ALS,$P<0.05, SC versus ALS, #P<0.05, NALS versus ALS). Westernblot analysis also showed a significant increase in the expression ofBDNF (Fig. S1A,B). In contrast, immunoreactivity for IGF-1 incontrol muscles showed an intense sarcoplasmic and sarcolemmaldistribution (Fig. 5B). However, IGF-1 expression showed amarginal downregulation in the ALS-CSF-injected groupcompared with the control groups (Fig. 5E,H). Further, weconfirmed the trend of downregulation by immunoblotting forIGF-1 (Fig. S1A,C).

Upregulated GDNF expression and downregulated VEGFexpressionGDNF was primarily localised to sarcolemma and a minimalpunctuate staining was observed in the sarcoplasm (Fig. 6A).However, GDNF expression was significantly upregulated in theALS group compared with the control groups, with uniform

immunostaining even in the sarcoplasm (Fig. 6D). The co-labelledGDNF and VEGF for the NC (Fig. 6C) and ALS (Fig. 6F) groups areprovided, where the altered expression can be appreciated better. Thequantification of mean fluorescence intensity confirmed this finding(Fig. 6G; **P<0.01, NC versus ALS; $P<0.05, SC versus ALS,#P<0.05, NALS versusALS).Western blot analysis ofGDNF showedthe trend for upregulation, although not significant (Fig. S1A,D).Immunofluorescence analysis of VEGF protein in control samplesshowed immunoreactivity in the sarcolemmal region and moreprominently in the extracellular space (Fig. 6B). Further, thequantification showed a significantly decreased expression in theALS group compared with the other groups (Fig. 6E,H; **P<0.01,NC versus ALS, $$P<0.01, SC versus ALS, #P<0.05, NALS versusALS). Consistent with the immunohistochemical results, expressionof VEGF was found to be significantly downregulated in the westernblot analysis. (Fig. S1A,E).

DISCUSSIONIn the present study, we report neurogenic atrophy of muscle fibres,reduced structural complexity of NMJs along with denervation,altered levels of receptor-clustering proteins of NMJs, rapsyn and

Fig. 2. ALS-CSF causes profound changes in NMJ structure. (A-D) Z-stacked confocal images of Alexa Fluor 488-conjugated α-bungarotoxin-stained AChRsfrom NC, SC, NALS and ALS rats. Note the reduced complexity of NMJs in the ALS group compared with the control groups. Scale bars: 18.75 μm. (E,F) Nerveterminals stained with anti-200KD neurofilament (red) and postsynaptic sites with Alexa Fluor 488-conjugated α-bungarotoxin (green) to assess the status ofinnervation. Note the complete apposition of nerve terminals and postsynaptic site in the NC group (E), whereas the ALS group shows a lack of any juxtapositionbetween nerve terminals and AChR sites (F), suggesting denervation. (G,H) Quantitative analysis of total NMJ area and AChR expression in EDL muscle fromNC, SC, NALS and ALS groups. The total NMJ area is significantly reduced in the ALS group compared with the control groups (G, **P<0.01, NC versus ALS;$P<0.05 SC versus ALS), whereas the expression of AChR does not significantly differ among the groups (H). n=5 in duplicates. Data were analysed using one-way ANOVA followed by Tukey’s post hoc test. (I) Representative western blot of rapsyn and calpain proteins normalised to myosin loading in the EDL muscle ofthe NC and ALS groups. (J) Quantitative representation of rapsyn and calpain expression, indicated as densitometric ratio of calpain: myosin and rapsyn: myosin.Note the significant upregulation of calpain (*P<0.05, NC versus ALS; n=3 in duplicates) and downregulation of rapsyn in the ALS group (*P<0.05, NC versus ALS;n=3 in duplicates). n=5 in duplicates. Data were analysed using one-way ANOVA followed by Tukey’s post hoc test.

4

RESEARCH ARTICLE Disease Models & Mechanisms (2018) 11, dmm031997. doi:10.1242/dmm.031997

Disea

seModels&Mechan

isms

calpain, ultrastructural damage, increased oxidative stress andimpaired expression of trophic factors in EDL muscle of ALS-CSF-injected rats. Taken together, these acute pathological changessuggest that skeletal muscle is significantly involved in thepathogenesis of SALS.In mSOD1 mice, motor unit numbers are reduced, specifically in

the fast-twitch muscle; however, in slow-twitch muscle, thesechanges are shown to develop only at the later stage of the disease(Hegedus et al., 2007, 2008; Kennel et al., 1996). These fast-twitchmotor units innervate the type 2 muscle fibres, and hence in ALS,there is a selective vulnerability of these fibres (Atkin et al., 2005;Derave et al., 2003; Frey et al., 2000). Our finding of pathologicalchanges in type 2 fibres in ALS-CSF-injected animals proves theirvulnerability to ALS-CSF-induced toxicity. Further, grouping of

type 2 fibres suggests the sprouting of motor axon terminals, acompensatory response to denervation.

The early-occurring symptoms in ALS provide evidence fordistal axonopathy in this disease (Fischer et al., 2004; Frey et al.,2000; Narai et al., 2009; Rocha et al., 2013). In the present study, thestructural complexity of NMJs was significantly compromised inALS-CSF-treated animals. There was also increased denervation ofthe NMJs, suggesting the disruption of neuromuscular transmission.In addition, the significant decrease in rapsyn, a key moleculefor clustering of AChRs, and a concomitant increase in calpain, aCa2+-dependent protein, which acts against clustering, result indispersed AChR clusters and fragmented NMJs (Chen et al., 2007).We believe that the trigger for NMJ damage in our model could bethe denervation caused by the onset of degeneration in a smallnumber of motor neurons, which is perhaps further aggravated byother pathological changes (discussed below).

The ultrastructural studies showed distinct changes, such as markedsize variations in muscle fibres, fibre angulation, sarcolemmal folds andZ-line streaming, in the ALS group, confirming the muscle damage.Z-line streaming represents an altered expression of myofilaments,implying denervation (Massa et al., 1992). Interestingly, similarobservations were earlier reported in ALS patients as well as inanimal models (Iwasaki et al., 1991; Kristmundsdottir et al., 1990;Kumamoto et al., 1979; Visani et al., 2011).

Morphological abnormalities in mitochondria, such as giantmitochondria, paracrystalline inclusions, abnormal cristae andaggregation, have been reported previously (Chung and Suh, 2002;Dobrowolny et al., 2008). Our results are also in agreement with theabove findings. Further, the SR regulates the calcium levels in themuscle in close association with themitochondria (Rossi et al., 2009).Increased calcium signalling along with mitochondrial damage is alsoreported in the skeletal muscle of models of ALS (Kawamata andManfredi, 2010; Zhou et al., 2010). Thus, calcium dysregulation playsa key role in ALS pathogenesis (reviewed in Grosskreutz et al., 2010).In the present study, the skeletal muscle of ALS animals exhibiteddilated SR, damaged mitochondria and increased calpain expression,suggesting abnormal calcium homeostasis.

The ultrastructural finding of damaged mitochondria suggests animpaired redox status of the system. Oxidative stress leads tosignificant lipid peroxidation and we confirmed the same in theskeletal muscle of ALS-CSF-injected animals. Further, we observeddownregulated activity of SOD, which might exacerbate theoxidative stress in the muscle. However, enhancement of theactivity of thioredoxin reductase probably suggests a compensatoryresponse. Similar upregulation of antioxidant enzyme activity,probably in response to an enhanced oxidative stress, is reported inthe skeletal muscle of rodent models (Dobrowolny et al., 2008;Leclerc et al., 2001; Mahoney et al., 2006). The absence of anincrease in SOD activity in our model contrasts with the increasedSOD activity observed in a model of familial ALS (FALS) (DalCanto and Gurney, 1995). This increase in SOD activity might beassociated with the overexpression of 25 copies of the SOD gene inthe FALS model, resulting in toxic ‘gain of function’. Antioxidantenzymes are required for maintaining the structural integrity ofNMJs, and oxidative stress can impair neuromuscular transmission,as shown by G93A-SOD1 mice exhibiting a significant decrease inthe release of neurotransmitters at NMJs (Naumenko et al., 2011;Sakellariou et al., 2014). Thus, the above findings confirm thatoxidative stress is a major contributory factor to the NMJdegeneration seen in ALS (Pollari et al., 2014). Accordingly, inthe current study, we propose that increased oxidative stress couldbe accelerating NMJ damage.

Fig. 3. Ultrastructural pathology. (A-H) Representative EM image depictingthe loss of sarcolemmal integrity in the ALS group. The NC group (A) showsnormal sarcolemmal featureswith intact outer and inner layersmarked by blackand white arrows, respectively. However, in the ALS group, the sarcolemmawas folded (B, black arrows). The longitudinal section of the skeletal muscleshows the sarcomere with perfectly aligned Z lines along with alternate dark (Aband) and light (I band) bands in the NC group (C, asterisk). Note thestreaming of Z-band material (D) along with misalignment of sarcomeres in theALS group (D, asterisk). Normal mitochondria with intact membrane andcristae structure are observed in the NC group (E), whereasmitochondria havevacuolation with abnormal internal structure and ruptured membrane in theALS group (F, white arrow). Transverse sections show reticulum structure withdiscretely formed cisternae in the NC group (G), compared with dilatedsarcoplasmic reticulum in the ALS group (H, white arrows).

5

RESEARCH ARTICLE Disease Models & Mechanisms (2018) 11, dmm031997. doi:10.1242/dmm.031997

Disea

seModels&Mechan

isms

BDNF is differentially regulated in ALS as there are decreasedlevels of BDNF in the spinal cord and elevated levels in theskeletal muscle (Deepa et al., 2011; Küst et al., 2002; Nishio et al.,1998). The present study provides experimental evidence forelevated BDNF levels in the muscles of ALS-CSF-treated rats.This increase is either a compensatory response or a consequenceof degeneration of motor neurons, leading to neurotrophinaccumulation in the target skeletal muscle. Nevertheless, theincrease in BDNF expression is likely to be transient, in view of thegradual decrease in BDNF as the disease progresses (Küst et al.,2002). The motor neurons can differentially regulate growth factorexpression in skeletal muscle to promote regeneration of injuredperipheral nerves (Funakoshi et al., 1995; Gómez-Pinilla et al.,2001). Thus, upregulated BDNF can be an initial compensatorymechanism provided by the skeletal muscle to rescue thedegenerating motor neurons.IGF-1 maintains the integrity of muscles and enhances satellite

cell activity in mSOD1 mice (Dobrowolny et al., 2005). DecreasedIGF-1 levels are seen in the spinal cord of ALS patients as well as inALS-CSF-injected rats (Deepa et al., 2011; Wilczak et al., 2003). Inthe present study, IGF-1 expression was downregulated in theskeletal muscle of the ALS rats, similar to findings reported earlier inthe skeletal muscle of ALS patients (Lunetta et al., 2012).Inflammatory response occurring in the skeletal muscle, such asincreased expression of TNF-α, IL-6 and other cytokines, can inhibitIGF-1 expression (Frost et al., 2003; Street et al., 2006; Van Dykeet al., 2016; Wolf et al., 1996). Further, oxidative stress has thepropensity to impair IGF-1 mRNA expression in muscle culture(Sestili et al., 2009). Thus, reduced IGF-1 levels observed in thepresent study might be caused by oxidative stress in skeletal muscle.

Considering the significant role of IGF-1 in neuronal survival, thisreduction could affect the survival of motor neurons.

GDNF is a trophic factor mainly involved in NMJ formation(Wright and Snider, 1996). GDNF expression is increased indenervated skeletal muscle (Henderson et al., 1994; Lie and Weis,1998; Zhao et al., 2004). Elevated GDNF mRNA expression isobserved in the spinal cord (Yamamoto et al., 1996) as well as inskeletal muscle of ALS patients (Grundström et al., 1999; Lie andWeis, 1998; Yamamoto et al., 1999). It is a potential therapeuticagent, and adeno-associated virus-GDNF-treated ALS mice show adelayed disease onset and progression of motor dysfunction, alongwith prolonged life span (Wang et al., 2002). The significantincrease in GDNF expression in the skeletal muscle of the ALS-CSF-injected rats is perhaps a transient compensatory mechanism topromote re-innervation of motor neurons. However, it could also bea generalised response to denervation injury (Lie and Weis, 1998;Zhao et al., 2004). Increased expression of c-Ret, a GDNF receptor,in motor neurons in ALS suggests that skeletal muscle probablyattempts to compensate and prevent motor neuron degeneration byupregulating GDNF, but eventually it fails in its efforts (Mitsumaet al., 1999; Ryu et al., 2011). Further, the transient nature of GDNFupregulation is shown in the autopsied skeletal muscles of ALSpatients, where skeletal muscles with severe fibre depletion showreduced levels of GDNF mRNA (Yamamoto et al., 1999, 1996).

Altered VEGF levels in the serum, CSF and anterior horn cellsof ALS patients, and also in the spinal cord of mSOD1 mice, havebeen reported (Brockington et al., 2006; Gao et al., 2014; Guptaet al., 2011; Lunn et al., 2009). Moreover, therapeutic potential ofVEGF is shown using rodent models by retrograde delivery and byexogenous supplementation to NSC-34 motor neurons (Azzouz

Fig. 4. Oxidative stress in the skeletal muscle of the ALS-CSF-injected animals. (A-D) Assays of oxidative stress and antioxidant enzyme activity: thehistograms represent lipid peroxidation (***P<0.001, NC versus ALS; $$P<0.01, SC versus ALS; ###P<0.001, NALS versus ALS) (A), thioredoxin reductase(****P<0.0001, NC versus ALS; $$$$P<0.0001, Sham versus ALS; ####P<0.001, NALS versus ALS) (B), SOD (*P<0.05, NC versus NALS; **P<0.01, NC versusALS; $$P<0.01, SC versus ALS; ###P<0.001, NALS versus ALS) (C), and glutathione reductase (D) activities (n=5 in duplicates). Data were analysed using one-way ANOVA followed by Tukey’s post hoc test.

6

RESEARCH ARTICLE Disease Models & Mechanisms (2018) 11, dmm031997. doi:10.1242/dmm.031997

Disea

seModels&Mechan

isms

et al., 2004; Krakora et al., 2013; Kulshreshtha et al., 2011;Vijayalakshmi et al., 2015). Further, VEGF offers neuroprotectionby enhancing axonal outgrowth (Carmeliet and Ruiz deAlmodovar, 2013). Hence, decreased VEGF expression mightlead to significant muscle damage, thus accelerating motorneuronal loss.The present study attempts to give a comprehensive account of

changes seen in the skeletal muscle of neonatal rats injected withCSF from SALS patients (Fig. 7). These include loss of NMJ andatrophy of muscle. The skeletal muscle affected by exposure toALS-CSF attempts to confer protection to degenerating motorneurons by upregulating BDNF and GDNF, but this is countered bya loss of neuromuscular synapses and decreased levels of IGF-1 and

VEGF. All these changes might aggravate the degeneration ofsurviving motor neurons, thus initiating a vicious cycle, leading torapid progression of this disease.

MATERIALS AND METHODSCSF sample collectionCSF samples from clinically confirmed ALS patients [El Escorialcriteria (Brooks et al., 2000)], were collected through lumbar punctureby a neurologist. The Human Ethics Committee of the National Instituteof Mental Health and Neurosciences, Bengaluru, approved the useof human CSF samples for the study (Item no. III, SI no. 3.01, BasicSciences), and consent was obtained from all participants prior toCSF collection. Age- and gender-matched patients suffering from

Fig. 5. Altered expression of BDNF and IGF-1. (A-F) Representative photomicrographs of transverse sections of EDL muscle double labelled for BDNF(FITC) and IGF-1 (CY-3). Increased expression of BDNF and a trend towards reduction in IGF-1 levels can be observed in the ALS group. Scale bars:75 μm. (G,H) Quantification of immunofluorescence intensity (0-255) supported the qualitative observations. BDNF was significantly upregulated in theALS group compared with other groups (G, *P<0.05, NC versus ALS; $P<0.05, SC versus ALS; #P<0.05, NALS versus ALS). In contrast, IGF-1 expressionwas downregulated in the ALS group, although not significantly. n=5 in duplicates. Data were analysed using one-way ANOVA followed by Tukey’spost hoc test.

7

RESEARCH ARTICLE Disease Models & Mechanisms (2018) 11, dmm031997. doi:10.1242/dmm.031997

Disea

seModels&Mechan

isms

non-neurodegenerative, noninfectious neurological diseases, such asbenign intracranial hypertension and transverse myelitis, were includedas non-ALS controls. The CSF samples were snap frozen in liquidnitrogen and stored at −80°C until use.

Intrathecal injection procedureNeonatal Wistar rats required for the study were procured from the CentralAnimal Research Facility of the National Institute of Mental Health AndNeurosciences, Bengaluru, subsequent to approval by the InstitutionalAnimal Ethics Committee (IAEC) [AEC/52/324/NP, AEC/56/324(B)/NPand AEC/60/324(C)/NP]. The animals were handled in accordance withNational Institutes of Health (NIH) guidelines. Intrathecal injections werecarried out as described previously (Rao et al., 1995). Briefly, 3-day-oldWistar rat pups were deeply anesthetised with halothane and a dorsal

midline skin incision (1 mm) was made about 1 cm rostral to the base ofthe tail. Using a microinjector, 5 µl of CSF was intrathecally injected intothe subarachnoid space at a rate of 1 µl/2.5 min. The incision was sutured,cleaned and sprayed with Healex (Rallis, India), an anti-inflammatoryliquid. The injections were carried out on postnatal days 3, 9 and 14. Theanimals were sacrificed on postnatal day 16, and the whole extensordigitorum muscle was dissected out carefully and snap frozen inisopentane pre-cooled in liquid nitrogen, for enzyme histochemistry andimmunohistochemistry, and fixed in 3% glutaraldehyde for electronmicroscopy.

The animals were grouped as follows: (1) normal control (NC), animalsthat were not subjected to the injection procedure; (2) sham control (SC),animals subjected to the sham injection procedure; (3) non-ALS-CSF(NALS), animals injected with non-ALS-CSF samples; (4) ALS-CSF(ALS), animals injected with ALS-CSF samples.

Fig. 6. Upregulated GDNF and downregulated VEGF expression. (A-F) Representative photomicrographs of transverse muscle sections stained for GDNF(FITC) and VEGF (CY-3), showing altered expression. Increased and decreased expression of GDNF and VEGF, respectively, was seen in the muscle of ALSrats. Scale bars: 37.5 μm. (G,H) Quantification of immunofluorescence intensity supported the qualitative observations. GDNF showed a significant upregulation(G, **P<0.01, NC versus ALS; $P<0.05, SC versus ALS; #P<0.05, NALS versus ALS), whereas VEGF showed a significant downregulation (H, **P<0.01, NCversus ALS; $$P<0.01, SC versus ALS; #P<0.05, NALS versus ALS), in the ALS group compared with the other groups. n=5 in duplicates. Data were analysedusing one-way ANOVA followed by Tukey’s post hoc test.

8

RESEARCH ARTICLE Disease Models & Mechanisms (2018) 11, dmm031997. doi:10.1242/dmm.031997

Disea

seModels&Mechan

isms

ATPase stainingThe EDL muscle was flash frozen in isopentane, pre-cooled in liquidnitrogen, and serial cryosections (8 µM) were collected on glass slides. Thecryosections were incubated with pre-incubating solution (44 mg CaCl2,41 mg sodium barbiturate in 10 ml distilled water, pH 9.4) for 20 min at37°C. This was followed by exposure to the incubating solution (19.98 mgCaCl2, 20.6 mg sodium barbiturate, 2 mg ATP salt in 10 ml distilled water,pH 9.6) for 45 min. The slides were then washed twice with 1% CaCl2,followed by three washes in 2% CoCl2 and two to three washes in distilledwater. Later, the slides were developed in 1% yellow ammonium sulphidesolution (freshly prepared), washed in double-distilled water, air dried andmounted in glycerine jelly.

Electron microscopyThe whole EDL muscle was fixed in 3% buffered glutaraldehydeand post-fixed with 1% osmium tetroxide for 2 h at 4°C. The tissues

were then dehydrated through a graded series of ethanol washes,cleared in propylene oxide, embedded in resin and left undisturbed at60°C for 2 days to allow polymerisation. Ultrathin sections, contrastedwith uranyl acetate and lead citrate, were viewed using a transmissionelectron microscope (FEI, TECNAI G2 Spirit BioTWIN, TheNetherlands).

Immunostaining of NMJSerial longitudinal cryosections of 40 µM thickness were collected on glassslides and fixed in 4% paraformaldehyde (PFA) for 10 min. The sectionswere incubated in 1.5% bovine serum albumin (BSA) followed byα-bungarotoxin (1:200, Invitrogen, USA) for 3 h at room temperature,washed in 0.1 M phosphate buffered saline (PBS), pH 7.4 andmounted. Thesections were co-labelled with an antibody against 200 kD neurofilaments(1:1000, Abcam, USA) followed by a secondary antibody [anti-rabbitcyanine 3 (CY3), 1:1000] for easy detection of innervation-, denervation-and re-innervation-induced changes in NMJs.

Fig. 7. Schematic representation of the contribution ofskeletal muscle in SALS pathogenesis. The early insultto motor neurons caused by the intrathecal injection ofALS-CSF results in impaired neuromuscular transmission,causing the skeletal muscle to undergo alterations, such asmuscle atrophy, disintegration and disruption of NMJ. Theskeletal muscle also showed increased oxidative stress asa result of mitochondrial damage and impaired calciumhomeostasis, as suggested by dilated sarcoplasmicreticulum and increased expression of calpain. Expressionof IGF-1 and VEGF was also reduced, thus depleting themajor trophic support. Nevertheless, the muscle initiallyattempts to offer neuroprotection by upregulating the levelsof BDNF and GDNF but, owing to damaged NMJ, motorneurons are unable to capitalise on it. Deprivation of trophicsupport results in the residual surviving motor neurons,which escaped the initial insult, to undergo degeneration,thus resulting in the relentless progress of ALS.

9

RESEARCH ARTICLE Disease Models & Mechanisms (2018) 11, dmm031997. doi:10.1242/dmm.031997

Disea

seModels&Mechan

isms

Biochemical analysis of oxidative stressandantioxidant enzymeactivityPreparation of whole muscle protein extractsFrozen muscle tissue (50 mg) was thawed and minced in 10 volumes of 1×PBS containing protease inhibitors and homogenised. The samples weresonicated on ice in six cycles of 5 s each. The extract was centrifuged(14,000 g, 10 min) and the supernatant was subjected to protein estimationby the Bradford method.

Bradford methodBriefly, 10 μl of the suitably diluted protein sample was pipetted into a 96-well ELISA plate to which 200 µl 1× Bradford reagent [100 mg CoomassieBrilliant Blue G-250, 50 ml ethanol (95%) and 100 ml orthophosphoric acid(85%) in 200 ml distilled water] was added, thoroughly mixed andincubated for 10 min. The absorbance was read at 595 nm in an ELISAplate reader (TECAN, Austria). The concentration of protein in the unknownsample was calculated as compared to BSA standards (50-500 μg/ml). Allestimations were performed in triplicate.

Estimation of lipid peroxidationThe extracted protein samples (2000 μg) were added to TBA/TCA reagent(7.5 mg thiobarbituric acid, 300 μl 100% trichloroacetic acid, 256 μl 1 NHCl). The mixture was then heated for 20 min in boiling water, after whichthe samples were subjected to centrifugation at 955.89 g for 10 min. Thesupernatant was collected and absorbance measured at 532 nm. MDAconcentration was calculated using the molar extinction co-efficient (MEC)(241 mol/cm) and normalised per mg protein.

SOD assayFor SOD assay, 10 μl of the supernatant (100 μg protein) was mixed with30 mM Tris-HCl buffer (pH 9.1), 0.5 mM ethylenediaminetetraacetic acid(EDTA), 50 mM tetramethylethylenediamine (TEMED) and 0.05 mMquercetin. The rate of quercetin oxidation was monitored at 406 nm for10 min (1 U SOD activity=amount of enzyme/mg protein that inhibitsquercetin oxidation by 50%).

Thioredoxin reductase assayThe reaction mixture contained 0.2 M phosphate buffer, pH 7.4, 1 mMEDTA, 0.4 mM NADPH and sample (15 μl). The assay reaction wasinitiated by the addition of 10 μl of 2 mM 5,5′-dithio-bis-(2-nitrobenzoicacid) (DTNB) and 5-thio-2-nitrobenzoic acid formed was measured at412 nm. The activity was expressed as nM of DTNB reduced/min/mgprotein.

Glutathione reductase assayFor glutathione reductase assay, 10 μl of the supernatant (100 μg protein)was mixed with Tris-HCl buffer (0.1 M, pH 8.8), 0.1 mM EDTA andNADPH (0.2 mM). The reaction was initiated by adding 12.5 μl of G-S-S-G(oxidised glutathione) and the decrease in absorbance at 340 nm wasmonitored for 5 min. The enzyme activity was expressed as nM of NADPHoxidised/min/mg protein (MEC=6.220 M−1 cm−1).

ImmunoblottingTotal muscle protein was extracted from the muscle tissue (10 mg) with asolution containing 75 mM Tris-HCl, pH 6.8, 15% SDS, 20% glycerol, 5%dithiothreitol and 0.001% Bromophenol Blue, followed by sonication for30 s, boiling at 95°C for 5 min and centrifugation at 15,000 g for 5 min.The supernatant was then used for immunoblotting of rapsyn [mouse anti-rapsyn (1:500), Abcam, ab11423], calpain [rabbit anti-calpain II, largesubunit (1:500), Merck Millipore, USA, AB81023], BDNF [rabbit anti-BDNF (1:500), Santa Cruz Biotechnology, USA, sc-546], IGF-1 [goatanti-IGF-1 (1:500), Santa Cruz Biotechnology, sc-7144], GDNF [mouseanti-GDNF (1:200), Santa Cruz Biotechnology, sc-13147] and VEGF[rabbit anti-VEGF (1:500), Santa Cruz Biotechnology, sc-507] proteins.For the loading control, protein extracts were run on 6% SDS-PAGEfollowed by Coomassie Brilliant Blue staining.

Immunohistochemical labelling of trophic factorsCryosections (8 µM) of EDL muscles were equilibrated with 0.1 M PBS(pH 7.4) at room temperature and post-fixed with 2% PFA for 15 min,followed by blocking in 3% BSA for 2 h. Subsequently, the sections wereincubated with the first primary antibody [rabbit anti-BDNF (1:500), SantaCruz Biotechnology, sc-546 or rabbit anti-GDNF (1:500), Santa CruzBiotechnology, sc-328] for 48 h. The sections were washed in PBS andincubated with fluorochrome-conjugated secondary antibody [anti-rabbitfluorescein isothiocyanate (FITC) (1:1000) or anti-rabbit Cy3 (1:1000)] for4 h at room temperature. The sections were re-equilibrated with PBSfollowed by blocking in 3% BSA for 2 h at room temperature. Further, thesections were incubated with the second primary antibody [goat anti-IGF-1(1:500), Santa Cruz Biotechnology, sc-7144 or rabbit anti-VEGF (1:500),Santa Cruz Biotechnology, sc-507] for 48 h, washed in PBS and incubatedwith fluorochrome-conjugated secondary antibody [anti-goat Cy3 (1:1000)or anti-rabbit FITC (1:1000)] at room temperature for 4 h. Stained sectionswere mounted onto cover slips in 65% glycerol and processed for imagecapturing with a confocal laser microscope (DMIRE-TCS, Leica, Germany)(Vijayalakshmi et al., 2015).

Imaging and quantification of immunofluorescenceTotal NMJ area and AChR protein expression were measured using in-builtLeica software of the confocal microscope by demarcating the edges of theα-bungarotoxin-labelled NMJs using the poly-line profile of the program.The total area within the marked boundary was measured to obtain thecorresponding numerical values commensurate to the staining intensity ofα-bungarotoxin-labelled NMJs. Similarly, for the quantification of trophicfactors, each myofibre in the transverse section was demarcated and theimage was analysed. The quantification was performed on a scale of 0-255,where 0 depicts absence of staining and 255 represents the most intensestaining. The quantitative analysis was carried out on 10 sections of the EDLmuscle per animal, and data from five pairs of animals were considered foreach group.

Statistical analysisEach experimental group consisted of 10 rats. Five different CSF sampleswere used in duplicate for all experiments. Statistical analysis was carriedout using one-way analysis of variance (Naumenko et al., 2011) followed byTukey’s post hoc test. P<0.05 was considered significant. Data areexpressed as mean s.e.m.

AcknowledgementsWe thank G. S. Monica for contributing to immunostaining and quantification oftrophic factors in the immunofluorescence studies, and Sanjay Das and J. N.Jessiena Ponmalar for performing western blot experiments.

Competing interestsThe authors declare no competing or financial interests.

Author contributionsConceptualization: S.S., G.N., P.A.A., T.R.R.; Methodology: S.S., G.N., A.N., P.A.A.;Investigation: G.N., A.N.; Writing - original draft: S.S.; Writing - review & editing: S.S.,G.N., P.A.A., T.R.R.; Supervision: G.N., A.N., P.A.A., T.R.R.

FundingThis work was supported by the Council of Scientific and Industrial Research (09/490(0088)/2012-EMR-I to S.S.) and National Institute of Mental Health andNeurosciences.

Supplementary informationSupplementary information available online athttp://dmm.biologists.org/lookup/doi/10.1242/dmm.031997.supplemental

ReferencesAlmeida, R. D., Manadas, B. J., Melo, C. V., Gomes, J. R., Mendes, C. S., Graos,

M. M., Carvalho, R. F., Carvalho, A. P. and Duarte, C. B. (2005).Neuroprotection by BDNF against glutamate-induced apoptotic cell death ismediated by ERK and PI3-kinase pathways. Cell Death Differ. 12, 1329-1343.

Annex, B. H., Torgan, C. E., Lin, P., Taylor, D. A., Thompson, M. A., Peters, K. G.and Kraus, W. E. (1998). Induction and maintenance of increased VEGF protein

10

RESEARCH ARTICLE Disease Models & Mechanisms (2018) 11, dmm031997. doi:10.1242/dmm.031997

Disea

seModels&Mechan

isms

by chronic motor nerve stimulation in skeletal muscle. Am. J. Physiol. 274,H860-H867.

Atkin, J. D., Scott, R. L., West, J. M., Lopes, E., Quah, A. K. J. and Cheema, S. S.(2005). Properties of slow- and fast-twitch muscle fibres in a mouse model ofamyotrophic lateral sclerosis. Neuromuscul. Disord. 15, 377-388.

Azzouz, M., Ralph, G. S., Storkebaum, E., Walmsley, L. E., Mitrophanous, K. A.,Kingsman, S.M., Carmeliet, P. andMazarakis, N. D. (2004). VEGF delivery withretrogradely transported lentivector prolongs survival in a mouse ALS model.Nature 429, 413-417.

Baloh, R. H., Rakowicz, W., Gardner, R. and Pestronk, A. (2007). Frequentatrophic groups with mixed-type myofibers is distinctive to motor neuronsyndromes. Muscle Nerve 36, 107-110.

Brockington, A., Wharton, S. B., Fernando, M., Gelsthorpe, C. H., Baxter, L.,Ince, P. G., Lewis, C. E. and Shaw, P. J. (2006). Expression of vascularendothelial growth factor and its receptors in the central nervous system inamyotrophic lateral sclerosis. J. Neuropathol. Exp. Neurol. 65, 26-36.

Brooke, M. H. and Engel, W. K. (1969). The histographic analysis of humanmusclebiopsies with regard to fiber types. 2. Diseases of the upper and lower motorneuron. Neurology 19, 378-393.

Brooks, B. R., Miller, R. G., Swash, M. and Munsat, T. L. (2000). El Escorialrevisited: revised criteria for the diagnosis of amyotrophic lateral sclerosis.Amyotroph. Lateral Scler. 1, 293-299.

Carmeliet, P. and Ruiz de Almodovar, C. (2013). VEGF ligands and receptors:implications in neurodevelopment and neurodegeneration. Cell. Mol. Life Sci. 70,1763-1778.

Chen, F., Qian, L., Yang, Z.-H., Huang, Y., Ngo, S. T., Ruan, N.-J., Wang, J.,Schneider, C., Noakes, P. G., Ding, Y.-Q. et al. (2007). Rapsyn interaction withcalpain stabilizes AChR clusters at the neuromuscular junction. Neuron 55,247-260.

Chung, M. J. and Suh, Y.-L. (2002). Ultrastructural changes of mitochondria in theskeletal muscle of patients with amyotrophic lateral sclerosis. Ultrastruct. Pathol.26, 3-7.

Dal Canto, M. C. andGurney, M. E. (1995). Neuropathological changes in two linesof mice carrying a transgene for mutant human Cu,Zn SOD, and in miceoverexpressing wild type human SOD: a model of familial amyotrophic lateralsclerosis (FALS). Brain Res. 676, 25-40.

Deepa, P., Shahani, N., Alladi, P. A., Vijayalakshmi, K., Sathyaprabha, T. N.,Nalini, A., Ravi, V. and Raju, T. R. (2011). Down regulation of trophic factors inneonatal rat spinal cord after administration of cerebrospinal fluid from sporadicamyotrophic lateral sclerosis patients. J. Neural Transm. 118, 531-538.

Derave,W., Van DenBosch, L., Lemmens, G., Eijnde, B. O., Robberecht,W. andHespel, P. (2003). Skeletal muscle properties in a transgenic mouse model foramyotrophic lateral sclerosis: effects of creatine treatment. Neurobiol. Dis. 13,264-272.

Dobrowolny, G., Giacinti, C., Pelosi, L., Nicoletti, C., Winn, N., Barberi, L.,Molinaro, M., Rosenthal, N. andMusaro , A. (2005). Muscle expression of a localIgf-1 isoform protects motor neurons in an ALS mouse model. J. Cell Biol. 168,193-199.

Dobrowolny, G., Aucello, M., Rizzuto, E., Beccafico, S., Mammucari, C.,Bonconpagni, S., Belia, S., Wannenes, F., Nicoletti, C., Del Prete, Z. et al.(2008). Skeletal muscle is a primary target of SOD1G93A-mediated toxicity. CellMetab. 8, 425-436.

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.

Frost, R. A., Nystrom, G. J. and Lang, C. H. (2003). Tumor necrosis factor-alphadecreases insulin-like growth factor-I messenger ribonucleic acid expression inC2C12 myoblasts via a Jun N-terminal kinase pathway. Endocrinology 144,1770-1779.

Funakoshi, H., Belluardo, N., Arenas, E., Yamamoto, Y., Casabona, A.,Persson, H. and Ibanez, C. F. (1995). Muscle-derived neurotrophin-4 as anactivity-dependent trophic signal for adult motor neurons. Science 268,1495-1499.

Gao, L., Zhou, S., Cai, H., Gong, Z. and Zang, D. (2014). VEGF levels in CSF andserum in mild ALS patients. J. Neurol. Sci. 346, 216-220.

Gomez-Pinilla, F., Ying, Z., Opazo, P., Roy, R. R. and Edgerton, V. R. (2001).Differential regulation by exercise of BDNFandNT-3 in rat spinal cord and skeletalmuscle. Eur. J. Neurosci. 13, 1078-1084.

Grosskreutz, J., Van Den Bosch, L., Keller, B.U. (2010). Calcium dysregulation inamyotrophic lateral sclerosis. Cell Calcium 47, 165-174.

Grundstrom, E., Askmark, H., Lindeberg, J., Nygren, I., Ebendal, T. andAquilonius, S.-M. (1999). Increased expression of glial cell line-derivedneurotrophic factor mRNA in muscle biopsies from patients with amyotrophiclateral sclerosis. J. Neurol. Sci. 162, 169-173.

Gunasekaran, R., Narayani, R. S., Vijayalakshmi, K., Alladi, P. A., Shobha, K.,Nalini, A., Sathyaprabha, T. N. and Raju, T. R. (2009). Exposure to

cerebrospinal fluid of sporadic amyotrophic lateral sclerosis patients altersNav1.6 and Kv1.6 channel expression in rat spinal motor neurons. Brain Res.1255, 170-179.

Gupta, P. K., Prabhakar, S., Abburi, C., Sharma, N. K. and Anand, A. (2011).Vascular endothelial growth factor-A and chemokine ligand (CCL2) genes areupregulated in peripheral blood mononuclear cells in Indian amyotrophic lateralsclerosis patients. J. Neuroinflammation 8, 114.

Hegedus, J., Putman, C. T. and Gordon, T. (2007). Time course of preferentialmotor unit loss in the SOD1 G93A mouse model of amyotrophic lateral sclerosis.Neurobiol. Dis. 28, 154-164.

Hegedus, J., Putman, C. T., Tyreman, N. and Gordon, T. (2008). Preferentialmotor unit loss in the SOD1 G93A transgenic mouse model of amyotrophic lateralsclerosis. J. Physiol. 586, 3337-3351.

Henderson, C. E., Phillips, H. S., Pollock, R. A., Davies, A. M., Lemeulle, C.,Armanini, M., Simmons, L., Moffet, B., Vandlen, R. A., Simpson, L. C.(corrected to Simmons L.), et al. (1994). GDNF: a potent survival factor formotoneurons present in peripheral nerve and muscle. Science 266, 1062-1064.

Henderson, C. E., Yamamoto, Y., Livet, J., Arce, V., Garces, A. and deLapeyriere, O. (1998). Role of neurotrophic factors in motoneuron development.J. Physiol. 92, 279-281.

Iwasaki, Y., Sugimoto, H., Ikeda, K., Takamiya, K., Shiojima, T. and Kinoshita,M. (1991). Muscle morphometry in amyotrophic lateral sclerosis. Int. J. Neurosci.58, 165-170.

Kawamata, H. and Manfredi, G. (2010). Mitochondrial dysfunction and intracellularcalcium dysregulation in ALS. Mech. Ageing Dev. 131, 517-526.

Kennel, P. F., Finiels, F., Revah, F. and Mallet, J. (1996). Neuromuscular functionimpairment is not caused by motor neurone loss in FALS mice: anelectromyographic study. Neuroreport 7, 1427-1431.

Koliatsos, V. E., Clatterbuck, R. E., Winslow, J. W., Cayouette, M. H. and Price,D. L. (1993). Evidence that brain-derived neurotrophic factor is a trophic factor formotor neurons in vivo. Neuron 10, 359-367.

Krakora, D., Mulcrone, P., Meyer, M., Lewis, C., Bernau, K., Gowing, G.,Zimprich, C., Aebischer, P., Svendsen, C. N. and Suzuki, M. (2013).Synergistic effects of GDNF and VEGF on lifespan and disease progression ina familial ALS rat model. Mol. Ther. 21, 1602-1610.

Kristmundsdottir, F., Mahon, M., Froes, M. M. Q. and Cumming, W. J. K. (1990).Histomorphometric and histopathological study of the human cricopharyngeusmuscle: in health and in motor neuron disease. Neuropathol. Appl. Neurobiol. 16,461-475.

Kulshreshtha, D., Vijayalakshmi, K., Alladi, P. A., Sathyaprabha, T. N., Nalini, A.and Raju, T. R. (2011). Vascular endothelial growth factor attenuatesneurodegenerative changes in the NSC-34 motor neuron cell line induced bycerebrospinal fluid of sporadic amyotrophic lateral sclerosis patients.Neurodegener. Dis. 8, 322-330.

Kumamoto, T., Suematsu, C., Yata, Y. and Yase, Y. (1979). Fine structuralcorrelations between the muscle pathology of amyotrophic lateral sclerosis (ALS)patients and experimental Ca-Mg deficient rats. Experientia 35, 1604-1605.

Kust, B. M., Copray, J. C. V. M., Brouwer, N., Troost, D. andBoddeke, H.W. G. M.(2002). Elevated levels of neurotrophins in human biceps brachii tissue ofamyotrophic lateral sclerosis. Exp. Neurol. 177, 419-427.

Leclerc, N., Ribera, F., Zoll, J.,Warter, J.-M., Poindron, P., Lampert, E. andBorg,J. (2001). Selective changes in mitochondria respiratory properties in oxidative orglycolytic muscle fibers isolated from G93AhumanSOD1 transgenic mice.Neuromuscul. Disord. 11, 722-727.

Lie, D. C. andWeis, J. (1998). GDNF expression is increased in denervated humanskeletal muscle. Neurosci. Lett. 250, 87-90.

Lunetta, C., Serafini, M., Prelle, A., Magni, P., Dozio, E., Ruscica, M., Sassone,J., Colciago, C., Moggio, M., Corbo, M. et al. (2012). Impaired expression ofinsulin-like growth factor-1 system in skeletal muscle of amyotrophic lateralsclerosis patients. Muscle Nerve 45, 200-208.

Lunn, J. S., Sakowski, S. A., Kim, B., Rosenberg, A. A. and Feldman, E. L.(2009). Vascular endothelial growth factor prevents G93A-SOD1-induced motorneuron degeneration. Dev. Neurobiol. 69, 871-884.

Mahoney, D. J., Kaczor, J. J., Bourgeois, J., Yasuda, N. and Tarnopolsky, M. A.(2006). Oxidative stress and antioxidant enzyme upregulation in SOD1-G93Amouse skeletal muscle. Muscle Nerve 33, 809-816.

Massa, R., Carpenter, S., Holland, P. and Karpati, G. (1992). Loss and renewal ofthick myofilaments in glucocorticoid-treated rat soleus after denervation andreinnervation. Muscle Nerve 15, 1290-1298.

Messi, M. L. and Delbono, O. (2003). Target-derived trophic effect on skeletalmuscle innervation in senescent mice. J. Neurosci. 23, 1351-1359.

Mishra, P.-S., Dhull, D. K., Nalini, A., Vijayalakshmi, K., Sathyaprabha, T. N.,Alladi, P. A. and Raju, T. R. (2016). Astroglia acquires a toxic neuroinflammatoryrole in response to the cerebrospinal fluid from amyotrophic lateral sclerosispatients. J. Neuroinflammation 13, 212.

Mishra, P.-S., Vijayalakshmi, K., Nalini, A., Sathyaprabha, T. N., Kramer, B. W.,Alladi, P. A. and Raju, T. R. (2017). Etiogenic factors present in the cerebrospinalfluid from amyotrophic lateral sclerosis patients induce predominantly pro-inflammatory responses in microglia. J. Neuroinflammation 14, 251.

11

RESEARCH ARTICLE Disease Models & Mechanisms (2018) 11, dmm031997. doi:10.1242/dmm.031997

Disea

seModels&Mechan

isms

Mitsuma, N., Yamamoto, M., Li, M., Ito, Y., Mitsuma, T., Mutoh, T., Takahashi, M.and Sobue, G. (1999). Expression of GDNF receptor (RET and GDNFR-alpha)mRNAs in the spinal cord of patients with amyotrophic lateral sclerosis.Brain Res.820, 77-85.

Narai, H., Manabe, Y., Nagai, M., Nagano, I., Ohta, Y., Murakami, T., Takehisa, Y.,Kamiya, T. and Abe, K. (2009). Early detachment of neuromuscular junctionproteins in ALS mice with SODG93A mutation. Neurol. Int. 1, e16.

Naumenko, N., Pollari, E., Kurronen, A., Giniatullina, R., Shakirzyanova, A.,Magga, J., Koistinaho, J. and Giniatullin, R. (2011). Gender-specificmechanism of synaptic impairment and its prevention by GCSF in a mousemodel of ALS. Front. Cell. Neurosci. 5, 26.

Neff, N. T., Prevette, D., Houenou, L. J., Lewis, M. E., Glicksman, M. A., Yin, Q.-W. and Oppenheim, R. W. (1993). Insulin-like growth factors: putative muscle-derived trophic agents that promote motoneuron survival. J. Neurobiol. 24,1578-1588.

Nguyen, Q. T., Parsadanian, A. S., Snider, W. D. and Lichtman, J. W. (1998).Hyperinnervation of neuromuscular junctions caused by GDNF overexpression inmuscle. Science 279, 1725-1729.

Nishio, T., Sunohara, N. and Furukawa, S. (1998). Neutrophin switching in spinalmotoneurons of amyotrophic lateral sclerosis. Neuroreport 9, 1661-1665.

Oosthuyse, B., Moons, L., Storkebaum, E., Beck, H., Nuyens, D., Brusselmans,K., Van Dorpe, J., Hellings, P., Gorselink, M., Heymans, S. et al. (2001).Deletion of the hypoxia-response element in the vascular endothelial growthfactor promoter causes motor neuron degeneration. Nat. Genet. 28, 131-138.

Oppenheim, R. W. (1996). Neurotrophic survival molecules for motoneurons: anembarrassment of riches. Neuron 17, 195-197.

Oppenheim, R. W., Yin, Q.-W., Prevette, D. and Yan, Q. (1992). Brain-derivedneurotrophic factor rescues developing avian motoneurons from cell death.Nature 360, 755-757.

Pollari, E., Goldsteins, G., Bart, G., Koistinaho, J. and Giniatullin, R. (2014). Therole of oxidative stress in degeneration of the neuromuscular junction inamyotrophic lateral sclerosis. Front. Cell. Neurosci. 8, 131.

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.

Ramamohan, P. Y., Gourie-Devi, M., Nalini, A., Shobha, K., Ramamohan, Y.,Joshi, P. and Raju, T. R. (2007). Cerebrospinal fluid from amyotrophic lateralsclerosis patients causes fragmentation of the Golgi apparatus in the neonatal ratspinal cord. Amyotroph. Lateral Scler. 8, 79-82.

Rao, M. S., Devi, M. G., Nalini, A., Shahani, N. and Raju, T. R. (1995).Neurofilament phosphorylation is increased in ventral horn neurons of neonatal ratspinal cord exposed to cerebrospinal fluid from patients with amyotrophic lateralsclerosis. Neurodegeneration 4, 397-401.

Rocha, M. C., Pousinha, P. A., Correia, A. M., Sebastiao, 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.

Rossi, A. E., Boncompagni, S. and Dirksen, R. T. (2009). Sarcoplasmic reticulum-mitochondrial symbiosis: bidirectional signaling in skeletal muscle. Exerc. SportSci. Rev. 37, 29-35.

Ryu, H., Jeon, G. S., Cashman, N. R., Kowall, N.W. and Lee, J. (2011). Differentialexpression of c-Ret in motor neurons versus non-neuronal cells is linked to thepathogenesis of ALS. Lab. Invest. 91, 342-352.

Sakellariou, G. K., Davis, C. S., Shi, Y., Ivannikov, M. V., Zhang, Y., Vasilaki, A.,Macleod, G. T., Richardson, A., Van Remmen, H., Jackson, M. J. et al. (2014).Neuron-specific expression of CuZnSOD prevents the loss of muscle mass andfunction that occurs in homozygous CuZnSOD-knockout mice. FASEB J. 28,1666-1681.

Sankaranarayani, R., Nalini, A., Rao Laxmi, T. and Raju, T. R. (2010). Alteredneuronal activities in the motor cortex with impaired motor performance in adultrats observed after infusion of cerebrospinal fluid from amyotrophic lateralsclerosis patients. Behav. Brain Res. 206, 109-119.

Sankaranarayani, R., Raghavan, M., Nalini, A., Laxmi, T. R. and Raju, T. R.(2014). Reach task-associated excitatory overdrive of motor cortical neuronsfollowing infusion with ALS-CSF. J. Neural. Trans. (Vienna, Austria: 1996) 121,49-58.

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

Sestili, P., Barbieri, E., Martinelli, C., Battistelli, M., Guescini, M., Vallorani, L.,Casadei, L., D’Emilio, A., Falcieri, E., Piccoli, G. et al. (2009). Creatinesupplementation prevents the inhibition of myogenic differentiation in oxidativelyinjured C2C12 murine myoblasts. Mol. Nutr. Food Res. 53, 1187-1204.

Shahani, N., Gourie-Devi, M., Nalini, A. and Raju, T. R. (2001).Cyclophosphamide attenuates the degenerative changes induced by CSF frompatients with amyotrophic lateral sclerosis in the neonatal rat spinal cord.J. Neurol. Sci. 185, 109-118.

Shahani, N., Gourie-Devi, M., Nalini, A., Rammohan, P., Shobha, K., Harsha,H. N. and Raju, T. R. (2004). (-)-Deprenyl alleviates the degenerative changesinduced in the neonatal rat spinal cord by CSF from amyotrophic lateral sclerosispatients. Amyotroph. Lateral Scler. Motor Neuron Disord. 5, 172-179.

Shantanu, S., Vijayalakshmi, K., Shruthi, S., Sagar, B. K. C., Sathyaprabha,T. N., Nalini, A., Raju, T. R. and Alladi, P. A. (2017). VEGF alleviates ALS-CSFinduced cytoplasmic accumulations of TDP-43 and FUS/TLS in NSC-34 cells.J. Chem. Neuroanat. 81, 48-52.

Sharma, A., Varghese, A. M., Vijaylakshmi, K., Sumitha, R., Prasanna, V. K.,Shruthi, S., Chandrasekhar Sagar, B. K., Datta, K. K., Gowda, H., Nalini, A.et al. (2015). Cerebrospinal fluid from sporadic amyotrophic lateral sclerosispatients induces mitochondrial and lysosomal dysfunction. Neurochem. Res. 41,968-984.

Shobha, K., Vijayalakshmi, K., Alladi, P. A., Nalini, A., Sathyaprabha, T. N. andRaju, T. R. (2007). Altered in-vitro and in-vivo expression of glial glutamatetransporter-1 following exposure to cerebrospinal fluid of amyotrophic lateralsclerosis patients. J. Neurol. Sci. 254, 9-16.

Shobha, K., Alladi, P. A., Nalini, A., Sathyaprabha, T. N. and Raju, T. R. (2010).Exposure to CSF from sporadic amyotrophic lateral sclerosis patients inducesmorphological transformation of astroglia and enhances GFAP and S100betaexpression. Neurosci. Lett. 473, 56-61.

Shruthi, S., Sumitha, R., Varghese, A. M., Ashok, S., Chandrasekhar Sagar,B. K., Sathyaprabha, T. N., Nalini, A., Kramer, B. W., Raju, T. R.,Vijayalakshmi, K. et al. (2017). Brain-derived neurotrophic factor facilitatesfunctional recovery from ALS-cerebral spinal fluid-induced neurodegenerativechanges in the NSC-34 motor neuron cell line. Neurodegener. Dis. 17, 44-58.

Shvartsman, D., Storrie-White, H., Lee, K., Kearney, C., Brudno, Y., Ho, N.,Cezar, C., McCann, C., Anderson, E., Koullias, J. et al. (2014). Sustaineddelivery of VEGF maintains innervation and promotes reperfusion in ischemicskeletal muscles via NGF/GDNF signaling. Mol. Ther. 22, 1243-1253.

Sondell, M., Lundborg, G. and Kanje, M. (1999). Vascular endothelial growthfactor has neurotrophic activity and stimulates axonal outgrowth, enhancing cellsurvival and Schwann cell proliferation in the peripheral nervous system.J. Neurosci. 19, 5731-5740.

Street, M. E., Ziveri, M. A., Spaggiari, C., Viani, I., Volta, C., Grzincich, G. L.,Virdis, R. and Bernasconi, S. (2006). Inflammation is a modulator of the insulin-like growth factor (IGF)/IGF-binding protein system inducing reduced bioactivity ofIGFs in cystic fibrosis. Eur. J. Endocrinol. 154, 47-52.

Sumitha, R., Sidhu, R. J. S., Sathyaprabha, T. N., Nalini, A., Raju, T. R. andAlladi, P. A. (2014). Differential expression of microRNA-206 in thegastrocnemius and biceps brachii in response to CSF from sporadicamyotrophic lateral sclerosis patients. J. Neurol. Sci. 345, 254-256.

Telerman-Toppet, N. and Coers, C. (1978). Motor innervation and fiber typepattern in amyotrophic lateral sclerosis and in Charcot-Marie-Tooth disease.Muscle Nerve 1, 133-139.

Van Dyke, J. M., Smit-Oistad, I. M., Macrander, C., Krakora, D., Meyer, M. G. andSuzuki, M. (2016). Macrophage-mediated inflammation and glial response in theskeletal muscle of a rat model of familial amyotrophic lateral sclerosis (ALS). Exp.Neurol. 277, 275-282.

Varghese, A. M., Sharma, A., Mishra, P., Vijayalakshmi, K., Harsha, H. C.,Sathyaprabha, T. N., Bharath, S. M. M., Nalini, A., Alladi, P. A. and Raju, T. R.(2013). Chitotriosidase - a putative biomarker for sporadic amyotrophic lateralsclerosis. Clin. Proteomics 10, 19.

Vijayalakshmi, K., Alladi, P. A., Sathyaprabha, T. N., Subramaniam, J. R., Nalini,A. and Raju, T. R. (2009). Cerebrospinal fluid from sporadic amyotrophic lateralsclerosis patients induces degeneration of a cultured motor neuron cell line. BrainRes. 1263, 122-133.

Vijayalakshmi, K., Alladi, P. A., Ghosh, S., Prasanna, V. K., Sagar, B. C., Nalini,A., Sathyaprabha, T. N. and Raju, T. R. (2011). Evidence of endoplasmicreticular stress in the spinal motor neurons exposed to CSF from sporadicamyotrophic lateral sclerosis patients. Neurobiol. Dis. 41, 695-705.

Vijayalakshmi, K., Ostwal, P., Sumitha, R., Shruthi, S., Varghese, A. M., Mishra,P., Manohari, S. G., Sagar, B. C., Sathyaprabha, T. N., Nalini, A. et al. (2015).Role of VEGF and VEGFR2 receptor in reversal of ALS-CSF induceddegeneration of NSC-34 motor neuron cell line. Mol. Neurobiol. 51, 995-1007.

Visani, M., de Biase, D., Bartolomei, I., Plasmati, R., Morandi, L., Cenacchi, G.,Salvi, F. and Pession, A. (2011). A novel T137A SOD1 mutation in an Italianfamily with two subjects affected by amyotrophic lateral sclerosis. Amyotroph.Lateral Scler. 12, 385-388.

Wang, L. J., Lu, Y. Y., Muramatsu, S., Ikeguchi, K., Fujimoto, K., Okada, T.,Mizukami, H., Matsushita, T., Hanazono, Y., Kume, A. et al. (2002).Neuroprotective effects of glial cell line-derived neurotrophic factor mediated byan adeno-associated virus vector in a transgenic animal model of amyotrophiclateral sclerosis. J. Neurosci. 22, 6920-6928.

Wilczak, N., de Vos, R. A. I. and De Keyser, J. (2003). Free insulin-like growthfactor (IGF)-I and IGF binding proteins 2, 5, and 6 in spinal motor neurons inamyotrophic lateral sclerosis. Lancet 361, 1007-1011.

Wolf, M., Bohm, S., Brand, M. and Kreymann, G. (1996). Proinflammatorycytokines interleukin 1 beta and tumor necrosis factor alpha inhibit growthhormone stimulation of insulin-like growth factor I synthesis and growthhormone receptor mRNA levels in cultured rat liver cells. Eur. J. Endocrinol.135, 729-737.

12

RESEARCH ARTICLE Disease Models & Mechanisms (2018) 11, dmm031997. doi:10.1242/dmm.031997

Disea

seModels&Mechan

isms

Wong, M. and Martin, L. J. (2010). Skeletal muscle-restricted expression of humanSOD1 causes motor neuron degeneration in transgenic mice. Hum. Mol. Genet.19, 2284-2302.

Wright, D. E. and Snider, W. D. (1996). Focal expression of glial cell line-derivedneurotrophic factor in developing mouse limb bud. Cell Tissue Res. 286,209-217.

Yamamoto, M., Sobue, G., Yamamoto, K., Terao, S. and Mitsuma, T.(1996). Expression of glial cell line-derived growth factor mRNA in thespinal cord and muscle in amyotrophic lateral sclerosis. Neurosci. Lett. 204,117-120.

Yamamoto, M., Mitsuma, N., Inukai, A., Ito, Y., Li, M., Mitsuma, T. and Sobue, G.(1999). Expression of GDNF and GDNFR-alpha mRNAs in muscles of patientswith motor neuron diseases. Neurochem. Res. 24, 785-790.

Zhao, C., Veltri, K., Li, S., Bain, J. R. and Fahnestock, M. (2004). NGF, BDNF, NT-3, and GDNF mRNA expression in rat skeletal muscle following denervation andsensory protection. J. Neurotrauma 21, 1468-1478.

Zhou, J., Yi, J., Fu, R., Liu, E., Siddique, T., Rıos, E. and Deng, H.-X. (2010).Hyperactive intracellular calcium signaling associated with localizedmitochondrial defects in skeletal muscle of an animal model of amyotrophiclateral sclerosis. J. Biol. Chem. 285, 705-712.

13

RESEARCH ARTICLE Disease Models & Mechanisms (2018) 11, dmm031997. doi:10.1242/dmm.031997

Disea

seModels&Mechan

isms