CORRECTION

Correction: Vascular regression precedesmotor neuron loss in theFUS (1-359) ALS mouse modelMartin Crivello, Marion C. Hogg, Elisabeth Jirstrom, Luise Halang, Ina Woods, Megan Rayner,Karen S. Coughlan, Sebastian A. Lewandowski and Jochen H. M. Prehn

There was an error in Disease Models & Mechanisms (2019) 12, dmm040238 (doi:10.1242/dmm.040238).

The authors mistakenly used the wrong image for the FUS(1-359) vehicle panel in Fig. 6Awhen assembling the figure. The corrected andoriginal figure panels are shown below. The quantification of data shown in Fig. 6B is not affected by this error.

Both the online full-text and PDF versions have been updated and the authors apologise to readers for this error and any inconvenience itmay have caused.

Fig. 6A (original panel). huANG treatment did not affect expected motorneuron numbers. (A) Representative images of Nissl-stained motor neuronsin the ventral horn area of lumbar spinal cord tissue obtained at P70 fromvehicle (left)- and huANG (right)-treated WT (top) and FUS (1-359) (bottom)mice. Dashed circle indicates ventral horn region used for quantification.

Fig. 6A (corrected panel). huANG treatment did not affect expectedmotorneuron numbers. (A) Representative images of Nissl-stained motor neuronsin the ventral horn area of lumbar spinal cord tissue obtained at P70 fromvehicle (left)- and huANG (right)-treated WT (top) and FUS (1-359) (bottom)mice. Dashed circle indicates ventral horn region used for quantification.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution andreproduction in any medium provided that the original work is properly attributed.

1

© 2020. Published by The Company of Biologists Ltd | Disease Models & Mechanisms (2020) 13, dmm045310. doi:10.1242/dmm.045310

Disea

seModels&Mechan

isms

RESEARCH ARTICLE

Vascular regression precedes motor neuron loss in theFUS (1-359) ALS mouse modelMartin Crivello1,*, Marion C. Hogg1,2,*, Elisabeth Jirstrom1,2, Luise Halang1, Ina Woods1, Megan Rayner1,Karen S. Coughlan1, Sebastian A. Lewandowski3,4 and Jochen H. M. Prehn1,2,‡

ABSTRACTAmyotrophic lateral sclerosis (ALS) presents a poorly understoodpathogenesis. Evidence from patients and mutant SOD1 mousemodels suggests vascular damage may precede or aggravate motordysfunction in ALS. We have previously shown angiogenin (ANG)treatment enhances motor neuron survival, delays motor dysfunctionand prevents vascular regression in the SOD1G93A ALS model.However, the existence of vascular defects at different stages ofdisease progression remains to be established in other ALS models.Here, we assessed vascular integrity in vivo throughout differentdisease stages, and investigated whether ANG treatment reversesvascular regression and prolongs motor neuron survival in the FUS(1-359) mouse model of ALS. Lumbar spinal cord tissue wascollected from FUS (1-359) and non-transgenic control mice atpostnatal day (P)50, P90 and P120. We found a significant decreasein vascular network density in lumbar spinal cords from FUS (1-359)mice by day 90, at which point motor neuron numbers wereunaffected. ANG treatment did not affect survival or countervascular regression. Endogenous Ang1 and Vegf expression wereunchanged at P50 and P90; however, we found a significantdecrease in miRNA 126 at P50, indicating vascular integrity in FUSmice may be compromised via an alternative pathway. Our studydemonstrates that vascular regression occurs before motor neurondegeneration in FUS (1-359) mice, and highlights that heterogeneityin responses to novel ALS therapeutics can already be detected inpreclinical mouse models of ALS.

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

KEY WORDS: Amyotrophic lateral sclerosis, FUS, Motoneuron,Angiogenin, Vascular defects

INTRODUCTIONAmyotrophic lateral sclerosis (ALS) is a neurodegenerative diseasecharacterised by progressive motor neuron death, leading to muscleweakening and motor-function loss. Time to death after diagnosisaverages 3-5 years, usually as a result of respiratory failure(Robberecht and Philips, 2013). Its aetiology remains largelyenigmatic. Out of dozens of clinical trials, only two drugs havebecome medically approved, riluzole and recently the antioxidantedavarone (Petrov et al., 2017). Both are of limited effect, extendinglifespan by several months and enhancing motor-function outcomesin a subset of patients, respectively (Rothstein, 2017). This probablystems from the heterogeneous nature of the disease, in whichpatients can display wide differences in onset time and location, rateof decline and survival time (Chio et al., 2009; Pupillo et al., 2014;Turner et al., 2013).

There is a growing body of evidence that suggests ALS could betreated as a neurovascular disease (Garbuzova-Davis et al., 2011).Indeed, a recent review highlights that neurovascular changes occur inmany neurodegenerative diseases, indicating that this pathway couldbe functionally relevant across the spectrum of disorders (Sweeneyet al., 2018). SOD1G93A mutant mice have been shown to havecompromised endothelia prior to disease onset (Zhong et al., 2008;Winkler et al., 2014), and reduced vascular perfusion has beenreported in the central nervous system of ALS mice and patients(Zhong et al., 2008; Rule et al., 2010; Winkler et al., 2013). Loss-of-function mutations in angiogenin (ANG) have been shown tosegregate with ALS patients (van Es et al., 2009; Greenway et al.,2006; Wu et al., 2007). Angiogenin is a potent angiogenic factor thatis also capable of initiating stress responses (Yamasaki et al., 2009).Reduced Ang1 and vascular endothelial growth factor family (Vegf)gene expression has been reported in SOD1G93A mice (Lu et al.,2007). All this has stimulated research exploring the therapeuticpotential of VEGF by others (Keifer et al., 2014; Pronto-Laborinhoet al., 2014) and of ANG by our group (Kieran et al., 2008; Sebastiaet al., 2009). Most recently, our lab conducted a SOD1G93A mousemodel study following the guidelines set by Ludolph et al. for ALSpreclinical studies (Ludolph et al., 2010). Among the more salientresults, we found that systemic human (hu)ANG administration threetimes a week from symptom onset extended survival, delayed motordysfunction and protected against motor neuron loss and vascularregression (Crivello et al., 2018).

These preclinical guidelines further recommend the advancementof ALS models other than the SOD1G93A ‘gold standard’ (Ludolphet al., 2010). One such model is FUS transgenic mice. Here, weemployed FUS (1-359) transgenic mice, which carry a truncatedversion of FUS with no nuclear localisation signal (NLS)(Shelkovnikova et al., 2013). These mice present severe motordegeneration starting∼P107 (range 60-180) and displaya rapid rate ofprogression, reaching terminal disease stage within several days aftersymptom onset (Shelkovnikova et al., 2013; Hogg et al., 2018). WeReceived 25 April 2019; Accepted 12 July 2019

1Department of Physiology and Medical Physics, Centre for the Study ofNeurological Disorders, Royal College of Surgeons in Ireland, 123 St. Stephen’sGreen, Dublin 2, Ireland. 2FutureNeuro Research Centre, Royal College ofSurgeons in Ireland, St. Stephen’s Green, Dublin 2, Ireland. 3Tissue BiologyLaboratory, Department of Clinical Neuroscience, Center for Molecular Medicine,Karolinska Institute, Karolinska vagen 8A, 17164, Stockholm, Sweden.4Affinity Proteomics, SciLifeLab, School of Biotechnology, KTH-Royal Institute ofTechnology, Tomteboda vagen 23 A, Stockholm, Sweden.*These authors contributed equally to this work

‡Author for correspondence (prehn@rcsi.ie)

M.C., 0000-0001-9535-7595; M.C.H., 0000-0002-2815-8761; J.H.M.P., 0000-0003-3479-7794

This is an Open Access article distributed under the terms of the Creative Commons AttributionLicense (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use,distribution and reproduction in any medium provided that the original work is properly attributed.

1

© 2019. Published by The Company of Biologists Ltd | Disease Models & Mechanisms (2019) 12, dmm040238. doi:10.1242/dmm.040238

Disea

seModels&Mechan

isms

have characterised the evolution of vascular defects in relation tomotor neuron degeneration at different disease time points and testedthe potential of ANG treatment in the FUS (1-359) model.

RESULTSMotor neuron loss in the FUS (1-359) mouse modelWe first profiled motor neuron numbers in the lumbar spinal cordfrom a pre-symptomatic time point (P50), around onset (P90), to themid-to-late stages of disease (P120). As expected, P50 FUS (1-359)mice showed no motor neuron loss (Fig. 1). We also found nosignificant differences in motor neuron counts between transgenicmice and their wild-type (WT) counterparts at P90. In contrast, byP120 we found a significant decrease in motor neuron numbers inFUS (1-359) mice compared to WT littermates in the ventral horn ofthe lumbar spinal cord [Fig. 1; FUS (1-359): 3.61±0.36 (mean±s.e.m.)versus WT: 5.95±0.66; P<0.005].

Vascular regression precedes motor neuron loss in FUS(1-359) miceWe next analysed total vessel length per area of ventral horn greymatter as a metric of vascular integrity by immunostaining againstpodocalyxin, an endothelial cell marker in lumbar spinal cordsections. We observed no differences in the average total length ofthe lumbar ventral horn capillary network betweenWT and FUS (1-359) mice at P50 (Fig. 2). Interestingly, we found a significantdecrease in average vessel length per ventral horn in the transgenicmice by P90 [Fig. 2; FUS (1-359): 11.40±0.37 mm/mm2 versusWT: 12.49±0.25 mm/mm2; P<0.04].We have previously identified a significant increase in the average

diameter of alpha-smooth muscle actin positive (ASMA+) blood

vessels in the SOD1G93A mouse model (Crivello et al., 2018).Therefore we also immunostained the tissues with an antibody toASMA, a marker for the blood vessel-surrounding contractile cells(i.e. smooth muscle cells and pericytes). We speculated that increaseddiameter of vessels could be a compensatory mechanism to countervascular regression. However, we found no differences between WTand FUS (1-359) mice at either P50 or 90 when analysing ASMA+

vascular diameter (Fig. 3), highlighting significant heterogeneity invascular pathology among preclinical ALS disease models.

ANG treatment did not extend survival or delay symptomonset in FUS (1-359) miceFollowing the vascular results described above and our previousfinding in the SOD1 mouse model (Crivello et al., 2018), we nextexplored the effects of chronic huANG administration on survivaland disease onset. We adapted the pre-clinical study designpreviously reported by us, and treated n=22-25 FUS (1-359) micechronically three times aweek with 1 μg huANG or vehicle until eachmouse reached terminal disease stage and was terminallyanesthetised. Interestingly, log-rank analysis revealed no significantdifferences between sex- and litter-matched FUS (1-359) mice treatedchronically three times a week with 1 μg huANG versus vehicle interms of survival (Fig. 4A) or time of symptom onset (Fig. 4B).

ANG treatment did not prevent vascular regression or affectmotor neuron numbersWe also conducted a vascular analysis as described above in lumbarsections from sex- and litter-matched WT and FUS (1-359) micetreated chronically three times a week with 1 μg ANG from P50 toP70. We found FUS (1-359) mice to have a significantly reduced

Fig. 1. Nissl staining revealed significant motor neuron loss byP120 in FUS (1-359) mice. Nissl stain highlights nucleic acid, inparticular ribosomal RNA, which is abundant in motor neurons andresults in a dark purple stain of the cell body. Figure showsrepresentative images from the ventral horn area of spinal cordtissue fromWT and FUS (1-359) transgenic mice, accompanied bythe quantification scatter-plot graph, obtained at P50 [top; WT n=6,FUS (1-359) n=8], P90 [middle; WT n=8, FUS (1-359) n=10] andP120 [bottom; WT n=4, FUS (1-359) n=6]. Dashed circle indicatesventral horn region used for quantification. Each datapointrepresents mean motor neuron counts from 20 non-consecutivetissue slices along the lumbar spinal cord, dark grey lines representmean±s.e.m. *P=0.005 [t-test: WT versus FUS (1-359) at P120].

2

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm040238. doi:10.1242/dmm.040238

Disea

seModels&Mechan

isms

capillary network compared to their WT counterparts, which wasunaffected by treatment [Fig. 5; two-way ANOVA; interaction: notsignificant (ns); main effect treatment: ns; main effect genotype:P<0.04].We also Nissl stained tissues from this same group of spinal cord

samples treated with huANG until P70 (Fig. 6). We found nosignificant differences in the number of motor neurons at this timepoint between the two genotypes or between treatments (Fig. 6;two-way ANOVA; interaction: ns; main effect treatment: ns; maineffect genotype: ns).

mAng1 expression is not altered in FUS (1-359) miceQuantification of mouse (m)Ang1 (Ang) levels by qPCR on lumbarspinal cord RNA revealed no significant difference in endogenousmAng1 levels when comparingWT to FUS (1-359) littermates at P50or P90 (n=9-10 mice/group; Fig. 7A,B). Similarly, analysis of Vegf(Vegfa) mRNA levels by qPCR revealed that there was also no

significant difference when comparing WT to FUS (1-359)littermates at P50 or P90 (Fig. 7C,D). To further explore potentialmechanisms for the lack of beneficial effect of angiogenin weinvestigated levels of miRNA 126 in early disease stages, as it haspreviously been implicated in vascular disease via regulation ofangiogenesis (Jamaluddin et al., 2011). miRNA 126 is the mosthighly abundant miRNA in endothelial cells and knockdown ofmiRNA 126 in endothelial cells rendered them unresponsive toVEGF-mediated proliferation and wound repair (Fish et al., 2008).Here, we found that levels of miRNA 126 were significantlydecreased in spinal cord from FUS (1-359) mice compared to WTmice at P50 (Fig. 7E), but no significant difference was detected atP90 (Fig. 7F). This early downregulation of miRNA 126 may renderFUS (1-359) mice unable to respond to pro-angiogenic signallinginduced by angiogenin in the early stages of disease progression.

In conclusion, we find the FUS (1-359) transgenic mice showdepletion of miRNA 126 at a very early stage of disease (P50),

Fig. 3. No differences in diameter identified when analysing ASMA+ blood vessels. (A) Representative images from the ventral horn area of lumbarspinal cord tissue fromWT (left) and FUS (1-359) (right) transgenic mice obtained at P50 (top) and at P90 (bottom). Insets show magnified view of boxed areas;arrows indicate pericyte cell bodies; n indicates nuclei. (B) Scatter-plot graphs showing the quantified diameter of ASMA+ blood vessels per lumbar ventral horn atP50 [left; WT n=6, FUS (1-359) n=8] and P90 [right; WT n=7, FUS (1-359) n=9]. Each datapoint represents mean diameter from six non-consecutive tissue slicesalong the lumbar spinal cord, dark grey lines represent mean±s.e.m. t-test between FUS (1-359) and WT groups non-significant in all cases.

Fig. 2. Analysis of podocalyxinimmunostaining revealed vascular regressionby P90 in FUS (1-359) mice. (A) Representativeimages from the ventral horn area of lumbar spinalcord tissue from WT (left) and FUS (1-359) (right)transgenic mice obtained at P50 (top) and at P90(bottom). (B) Scatter-plot graphs showing thequantified total vessel length per lumbar ventralhorn area at P50 [left; WT n=6, FUS (1-359) n=8]and P90 [right; WT n=7, FUS (1-359) n=9]. Eachdatapoint represents mean vessel length from sixnon-consecutive tissue slices along the lumbarspinal cord, dark grey lines represent mean±s.e.m.*P<0.04 [t-test: WT versus FUS (1-359) at P90].

3

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm040238. doi:10.1242/dmm.040238

Disea

seModels&Mechan

isms

which appears to limit their ability to respond to endogenous orexogenous angiogenic factors.

DISCUSSIONSeveral genes have been linked directly or indirectly to ALS in recentyears (Chia et al., 2018; Picher-Martel et al., 2016), allowing for thegeneration of multiple new transgenic models and the need toproperly characterise them. Of particular interest here, mutations inthe highly conserved NLS region of the FUS gene, which encodes aprotein that is involved in RNA transcription and processing, havebeen described in ALS and in ALS with frontotemporal dementia(ALS/FTD) (Dormann and Haass, 2013; Hewitt et al., 2010; Nolanet al., 2016; Rademakers et al. 2010). FUS (1-359) mice express aversion of the FUS protein that lacks its entire NLS and RNA-bindingmotifs, but they share several phenotypic hallmarks with ALS-FUSpatients, who predominantly present an aggressive clinical course,severe neuronal loss in the spinal cord ventral horn (Ravits et al.,2013), and mislocalisation and aggregation of FUS protein in thecytoplasm (Shelkovnikova et al., 2013; Hewitt et al., 2010).In the present study, we characterised vascular integrity in the

FUS (1-359) mouse model at different time points in the diseasecourse. We found a significant reduction in vascular coverage per

ventral horn grey matter area in the lumbar spinal cord at P70(Fig. 5), and P90 (Fig. 2), but only by the next time point analysed(P120) did we observe a significant loss of motor neurons (Fig. 1).This shows that vascular regression precedes motor neuron loss inthis model, as has been previously described in mutant SOD1 miceas assessed by immunoglobulin leakage and haemosiderindeposition within the spinal cord (Zhong et al., 2008). However,our subsequent studies pointed at potential differences in vascularregression in SOD1 and FUS mutant mice. We previouslydemonstrated that early vasculature defects in SOD1G93A micecould be rescued by huANG administration from P90, whichprevented later motor neuron loss (analysed at P115, following25 days of treatment) (Crivello et al., 2018). Using a similartreatment protocol, huANG administration exerted no beneficialeffects on vascular regression, motor neuron loss or survival in FUS(1-359) mice. We also failed to detect an increase in the averagediameter of vascular mural cells in FUS (1-359) mice. We detectedsuch an increase in transgenic SOD1G93A mice at an early-to-midpoint in the disease (Crivello et al., 2018). We proposed this to be apossible compensatory mechanism in response to reduced vascularnetwork coverage. Collectively, the SOD1G93A and FUS (1-359)ALS mouse models show fundamental differences in terms of

Fig. 5. huANG treatment did not counteract vascularregression. (A) Representative images from the ventralhorn area of lumbar spinal cord tissue from vehicle- (left)and huANG- (right) treated WT mice (top) and FUS(1-359) mice (bottom) obtained at P70 following 20 daystreatment and stained with an antibody to podocalyxin tohighlight blood vessels. (B) Scatter-plot graph showingthe quantified total vessel length per lumbar ventral hornarea at P70 obtained from the four groups of mice. Eachdatapoint represents mean vessel length from six non-consecutive tissue slices along the lumbar spinal cord.WT-huANG n=10; WT-Veh n=8; FUS (1-359)-huANGn=8; FUS (1-359)-Veh n=7. Dark grey lines representmean±s.e.m. *P<0.04, main effect genotype; interactionand main effect treatment, not significant (two-wayANOVA).

Fig. 4. No differences in survival or time of diseaseonset in FUS (1-359) mice treated with huANG.(A) Kaplan–Meier curves showing survival rates fromanimals treated three times per week from P50 witheither 1 μg of huANG (grey line) or PBS vehicle(dashed black line). (B) Kaplan–Meier curves showingtime of disease onset measured in animals treatedthree times per week from P50 with either 1 μg ofhuANG (grey line) or PBS vehicle (dashed black line).Log-rank test showed no significant differencesbetween treated and control groups in both cases(n=22 huANG, n=25 PBS vehicle).

4

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm040238. doi:10.1242/dmm.040238

Disea

seModels&Mechan

isms

vascular pathophysiology, which may correlate to ANGresponsiveness. Of note, a previous study in the SOD1G93A ALSmouse model demonstrated that treatment of mice with an activated

protein C (APC) mutant (5A-APC) protected blood-spinal cordbarrier integrity and delayed onset of motor symptoms (Winkleret al., 2014), indicating there are multiple effective methods fortargeting vascular defects in ALS.

To explain these differences in vascular pathology and responsesto pro-angiogenic therapies in both mouse models, we highlight oneaspect of FUSmutant protein: loss of nuclear function in endothelia.FUS has been shown to be expressed in human endothelia frommultiple organs (Andersson et al., 2008). Furthermore, studies usingdifferent mammalian endothelial cells in culture have described howFUS levels are downregulated as the cells transition fromproliferation to quiescence (Alliegro and Alliegro, 1996), and theuse of antibodies or siRNA against FUS protein or mRNA inhibitedtheir proliferation (Alliegro and Alliegro, 2002; Yoshida et al.,2010). Given that FUS affects the transcription and/or processing ofmultiple RNAs in multiple cell types (Fujioka et al., 2013; Lagier-Tourenne et al., 2012) we should not ignore FUSopathic effects inneurons, glia or muscle; however, the evidence described suggeststhat dysfunctional FUS can send vascular homeostasis into disarray,even when ignoring the toxic potential of FUS cytoplasmicinclusions. We hypothesise that FUS (1-359) mice may belacking an important functionality directly related to endothelialcell proliferation, which would explain why huANG treatment didnot have the positive effects we have seen before in SOD1G93A

mice. To our knowledge, this is the first study into vasculaturedefects in the FUS (1-359) mouse model. Several other FUS ALSmouse models have been developed; however, vasculature analysisand blood-brain barrier (BBB) defects have not been explored inthese models to date (Devoy et al., 2017; Scekic-Zahirovic et al.,2017). BBB defects are implicated in multiple neurodegenerativediseases, which indicates that defects in this pathway may yieldtherapeutic targets of wide relevance (Sweeney et al., 2019).

The SOD1G93A mouse model displays downregulated Vegf mRNAearly in disease progression (Lu et al., 2007), and suffers a more gradualdisease progression than the FUS (1-359) mice (Hogg et al., 2018). Tofurther explore the role of FUS in the vasculature system we analysedlevels of pro-angiogenic factors Vegf and Ang1 by qPCR. We found nosignificant difference in expression of these factors between WT andFUS (1-359) mice at early time points when we observed vasculaturedefects. We also investigated miRNA 126, as it has previously beenimplicated in vascular disease via regulation of angiogenesis

Fig. 6. huANG treatment did not affect expectedmotor neuron numbers. (A) Representative imagesof Nissl-stained motor neurons in the ventral horn areaof lumbar spinal cord tissue obtained at P70 fromvehicle (left)- and huANG (right)-treated WT (top) andFUS (1-359) (bottom) mice. Dashed circle indicatesventral horn region used for quantification. (B) Scatter-plot graph showing the quantified number of motorneurons per lumbar ventral horn area at P70 obtainedfrom the four groups of mice. Each datapointrepresents mean number of motor neurons from 20non-consecutive tissue slices along the lumbar spinalcord. WT-huANG n=10; WT-Veh n=8; FUS (1-359)-huANG n=8; FUS (1-359)-Veh n=7. Dark grey linesrepresent mean±s.e.m. Interaction, main effecttreatment and main effect genotype, not significant(two-way ANOVA).

Fig. 7. miRNA 126 is significantly downregulated in FUS (1-359) mice atP50. (A,B) qPCR on Ang1 levels in spinal cord RNA from mice at P50 (A) andP90 (B) revealed no significant change in Ang1 levels (n=7-9 mice/group). (C,D) qPCR to quantify Vegf mRNA in spinal cord also showed no significantdifference in levels at P50 (C) and P90 (D). (E,F) Taqman assays were usedto quantify miRNA 126 levels in total RNA extracted from spinal cord at P50 (E)and P90 (F) (n=7-9 mice/group). At P50 miRNA 126 levels were significantlylower in transgenic (TG) FUS (1-359) mice compared to WT. *P<0.05(Mann–Whitney U test), whereas no significant differencewas detected at P90.Data are mean±s.e.m.

5

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm040238. doi:10.1242/dmm.040238

Disea

seModels&Mechan

isms

(Jamaluddin et al., 2011). In endothelial cells, miRNA 126 has beenshown to promote proliferation and enhance vascular repair followinginjury to the carotid artery (Schober et al., 2014). miRNA 126 has beenshown to negatively regulate levels of SPRED1 and PIK3R2, whichprevents downstream signalling in response to VEGF via the MAPkinase and the PI3 kinase pathways, respectively (Fish et al., 2008;Wang et al., 2008). Interestingly, we found that miRNA 126 wassignificantly downregulated in FUS (1-359) mice early in disease,indicating a profound early defect in angiogenesis signalling in thespinal cord. This defect could be related to the more aggressivepathology observed in FUS (1-359) mice. Interestingly, miRNA 126has been shown to be downregulated early in disease progression inmuscle from the SOD1G93Amousemodel (Maimon et al., 2018), andthe authors showed that overexpressing miRNA 126 could preventaxonal degeneration and loss of innervation at the neuromuscularjunction. This raises the possibility that increasing miRNA 126levels may delay onset of symptoms in the FUS (1-359) mousemodel. FUS has been associated with numerous stages of RNAprocessing involving both coding and non-coding RNAs (reviewedby Lagier-Tourenne et al., 2010). FUS interacts with argonaute 2 anddirectly binds to miRNAs (Zhang et al., 2018); however the FUS (1-359) fragment expressed in this mouse colony lacks the RNA-binding region required to interact directly with RNA, whichindicates that the observed decrease in miRNA 126 occurs via anindirect mechanism. In conclusion, our results highlight theimportance of incorporating multiple models in preclinical ALSstudies to more accurately approximate the full range of ALS clinicalmanifestations and effects of novel therapeutics (such as huANG),and thus provide a better basis for clinical trials.

MATERIALS AND METHODSAnimals and ethics approvalThis study was conducted in strict accordance with Directive 2010/63/EUon the protection of animals used for scientific purposes. FUS (1-359) micewere generated in the laboratory of Prof. Vladimir Buchman (CardiffUniversity; Shelkovnikova et al., 2013) and re-derived at the Institute ofMolecular Genetics of the ASCR, Prague, Czechia. Mice for this study wereage-, gender- and litter-matched according to ALS pre-clinical trialguidelines (Ludolph et al., 2010). The animals, congenic on the C57Bl/6background, were housed in cages of between 3 and 5 mice at a constanttemperature (22°C) on a 12 h light/dark cycle (07:00 h on, 19:00 h off ), withad libitum availability of food and water. Exact numbers of individuals usedare detailed in Table S1 and S2. Mice were genotyped by PCR. Ethicalapproval for the animal experiments conducted for this study was awardedby the Animal Research Ethics Committee of the Royal College of Surgeonsin Ireland (ethics reference number: REC1122b) and under license from theIrish Health Products Regulatory Authority (AE19127/P004).

Tissue collection and preparation for histological analysisFollowing ethical guidelines, mice at P50, P90 and P120 were terminallyanesthetised and transcardially perfused via the left ventricle with PBS at aslow rate (∼10 ml/min) until full exsanguination. Lumbar spinal cords(L1-L5) were collected and fixed with PFA 4% for 1 h before beingtransferred to a 30% sucrose solution for cryoprotection. After 1-5 days, thesucrose solution was completely removed, the tissues frozen in liquidnitrogen and stored at −80°C. The tissues were later sectioned with a cryostatset at 16 μm andmounted on Superfrost plus slides (MNJ-700-010N, VWR).

Vascular immunostaining and stereological analysisWe conducted immunohistochemistry against two markers that we hadanalysed previously to assess vascular integrity: podocalyxin (endothelialcell marker) and ASMA (expressed in contractile mural cells) (Crivelloet al., 2018). The tissue sections were washed in PBS and thenpermeabilised using 3% triton-PBS for 20 min. Blocking was conducted

with 5% donkey serum for 1 h before incubating the tissues overnight with1:200 anti-podocalyxin antibody (AF1556, R&D Systems) and 1:50 anti-ASMA (ab5694, Abcam). The next day the samples were washed andincubated with anti-goat Alexa-647 (A21447, Thermo Fisher Scientific) andanti-rabbit Alexa-488 (A10042, Thermo Fisher Scientific) secondaryantibodies 1:200 for 1 h at room temperature. Finally, the slides weremounted with DAPI-containing media (H1200, Vector Labs).

Pictures fitting greymatter from the left and right ventral hornwere capturedusing a Nikon TE2000s epifluorescence microscope at 20× magnification.The images were analysed using ImageJ software by a blinded observer asdescribed previously (Lewandowski et al., 2016). Briefly, a common thresholdselection was used to generate masks and the ‘skeletonisation’ function wasthen used for measuring blood vessel length. ASMA+ blood vessel diameterwas calculated by dividing the total mask area of each continuous ASMAsignal by their total length. Each data point graphed represents the averagefrom at least six non-adjacent tissue sections (>120 μm apart).

RNA extraction and qPCRMice were transcardially perfused as described above and tissues dissectedthen flash frozen in liquid nitrogen. Total RNA was extracted using theQiagen miRNeasy kit and protocol. For gene expression qPCR, 500 ng totalRNA was reverse transcribed with oligo dT primers and amplified withAng1 or Vegf primers (from Sebastia et al., 2009) using Quantitect SYBRgreen PCR kit (Qiagen). For miRNA analysis, total RNA was analysedusing the Taqman assay mmu-126-5p and normalised to U6 snRNA(Applied Biosystems). Fold change was calculated using the 2−DDCtmethod(Livak and Schmittgen, 2001).

Motor neuron counts with Nissl stainingMotor neuron survival was assessed from the same group of tissuesdescribed above. First, the samples were treated with 0.1% Cresyl Violetacetate solution for 20 min at 60°C before being rinsed with distilled waterand treated with increasing concentrations of ethanol (70%, 90% and100%). A 1 min treatment with HistoClear (HS200, National Diagnostics)was the final step, before mounting the slides with DPX mountingmedium. Nissl-stained motor neurons of the ventral horn region werecounted following previously published methods (cell bodies ≥30 μm insize, presence of a dark nucleolus and multi-polar structure) (Coughlanet al., 2015) in 20 non-consecutive sections per animal.

ANG treatmentFollowing results that we had obtained previously in SOD1G93A mice, wenext pursued a similar recombinant huANG (265AN250/CF, R&D Systems)treatment paradigm with FUS (1-359) mice (Crivello et al., 2018). FUS(1-359) mice and their WT counterparts were injected intraperitoneally with1 µg of huANG or vehicle (PBS) three times a week from P50 until P70 forhistological studies or from P50 until end stage of disease for survivalanalysis. Disease end stagewas determined by the extremeweakening or lossof hind limb function and inability of the mice to turn themselves within 30 sof being placed on their back (Ludolph et al., 2010).

Statistical analysisOutliers classed as ±2 s.d. from the mean were excluded from furtheranalysis. Two-tailed t-tests were used to analyse differences in the meansbetween groups of two. Two-way ANOVA was used for groups of fourtesting for factors of genotype and of treatment. Mantel-Cox test was used toanalyse differences in lifespan measurements. Normality was assessed withthe Shapiro-Wilk test and homoscedasticity with the Levene test.

AcknowledgementsWe thank Prof. Vladimir Buchman and Dr Natalia Ninkina for graciously donatingFUS (1-359) breeding pairs so we could establish a colony in our own facility for thisstudy.

Competing interestsJ.H.M.P. is a beneficiary of patents relating to the use of angiogenin for the treatmentof CNS diseases.

6

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm040238. doi:10.1242/dmm.040238

Disea

seModels&Mechan

isms

Author contributionsConceptualization: J.H.M.P.; Formal analysis: S.A.L.; Investigation: M.C., M.C.H.,E.J., L.H., I.W., M.R., K.S.C., S.A.L.; Writing - original draft: M.C., M.C.H., J.H.M.P.;Project administration: J.H.M.P.; Funding acquisition: J.H.M.P.

FundingJ.H.M.P. was supported by a grant from the Health Research Board (HRA_POR/2013/348). S.A.L. is supported by Stiftelsen Olle Engkvist Byggmastare (SLS-499431), Ulla-Carin Lindquists Foundation for ALS research, Åhlen Foundationand Neuroforbundet. This publication has emanated from research supported inpart by a research grant from Science Foundation Ireland (16/RC/3948) andco-funded under the European Regional Development Fund and by FutureNeuroindustry partners.

Supplementary informationSupplementary information available online athttp://dmm.biologists.org/lookup/doi/10.1242/dmm.040238.supplemental

ReferencesAlliegro, M. C. and Alliegro, M. A. (1996). A nuclear protein regulated during thetransition from active to quiescent phenotype in cultured endothelial cells. Dev.Biol. 174, 288-297. doi:10.1006/dbio.1996.0074

Alliegro, M. C. and Alliegro, M. A. (2002). Nuclear injection of anti-pigpenantibodies inhibits endothelial cell division. J. Biol. Chem. 277, 19037-19041.doi:10.1074/jbc.M200737200

Andersson, M. K., Ståhlberg, A., Arvidsson, Y., Olofsson, A., Semb, H.,Stenman, G., Nilsson, O. and Åman, P. (2008). The multifunctional FUS, EWSand TAF15 proto-oncoproteins show cell type-specific expression patterns andinvolvement in cell spreading and stress response. BMC Cell Biol. 9, 37. doi:10.1186/1471-2121-9-37

Chia, R., Chio , A. and Traynor, B. J. (2018). Novel genes associated withamyotrophic lateral sclerosis: diagnostic and clinical implications. Lancet Neurol.17, 94-102. doi:10.1016/S1474-4422(17)30401-5

Chio, A., Logroscino, G., Hardiman, O., Swingler, R., Mitchell, D., Beghi, E.,Traynor, B. G. and On Behalf Of The Eurals Consortium (2009). Prognosticfactors in ALS: a critical review. Amyotroph Lateral Scler. 10, 310-323. doi:10.3109/17482960802566824

Coughlan, K. S., Mitchem, M. R., Hogg, M. C. and Prehn, J. H. M. (2015).“Preconditioning” with latrepirdine, an adenosine 5’-monophosphate-activatedprotein kinase activator, delays amyotrophic lateral sclerosis progression inSOD1(G93A) mice. Neurobiol. Aging 36, 1140-1150. doi:10.1016/j.neurobiolaging.2014.09.022

Crivello, M., O’Riordan, S. L., Woods, I., Cannon, S., Halang, L., Coughlan,K. S., Lewandowski, S. A. and Prehn, J. H. M. (2018). Pleiotropic activity ofsystemically delivered angiogenin in the SOD1(G93A) mouse model.Neuropharmacology 133, 503-511. doi:10.1016/j.neuropharm.2018.02.022

Devoy, A., Kalmar, B., Stewart, M., Park, H., Burke, B., Noy, S. J., Redhead, Y.,Humphrey, J., Lo, K., Jaeger, J. et al. (2017). Humanized mutant FUS drivesprogressive motor neuron degeneration without aggregation in ‘FUSDelta14’knockin mice. Brain 140, 2797-2805. doi:10.1093/brain/awx248

Dormann, D. and Haass, C. (2013). Fused in sarcoma (FUS): an oncogene goesawry in neurodegeneration. Mol. Cell. Neurosci. 56, 475-486. doi:10.1016/j.mcn.2013.03.006

Fish, J. E., Santoro, M. M., Morton, S. U., Yu, S., Yeh, R. F., Wythe, J. D., Ivey,K. N., Bruneau, B. G., Stainier, D. Y. R., Srivastava, D. et al. (2008). miR-126regulates angiogenic signaling and vascular integrity. Dev. Cell 15, 272-284.doi:10.1016/j.devcel.2008.07.008

Fujioka, Y., Ishigaki, S., Masuda, A., Iguchi, Y., Udagawa, T., Watanabe, H.,Katsuno, M., Ohno, K., Sobue, and G. (2013). FUS-regulated region- and cell-type-specific transcriptome is associated with cell selectivity in ALS/FTLD. Sci.Rep. 3, 2388. doi:10.1038/srep02388

Garbuzova-Davis, S., Rodrigues, M. C. O., Hernandez-Ontiveros, D. G., Louis,M. K.,Willing, A. E. andBorlongan, C. V. (2011). Amyotrophic lateral sclerosis: aneurovascular disease. Brain Res. 1398, 113-125. doi:10.1016/j.brainres.2011.04.049

Greenway, M. J., Andersen, P. M., Russ, C., Ennis, S., Cashman, S., Donaghy,C., Patterson, V., Swingler, R., Kieran, D., Prehn, J. et al. (2006). ANGmutations segregatewith familial and ‘sporadic’ amyotrophic lateral sclerosis.Nat.Genet. 38, 411-413. doi:10.1038/ng1742

Hewitt, C., Kirby, J., Highley, J. R., Hartley, J. A., Hibberd, R., Hollinger, H. C.,Williams, T. L., Ince, P. G., Mcdermott, C. J., Shaw, P. J. et al. (2010). NovelFUS/TLS mutations and pathology in familial and sporadic amyotrophic lateralsclerosis. Arch. Neurol. 67, 455-461. doi:10.1001/archneurol.2010.52

Hogg, M. C., Halang, L., Woods, I., Coughlan, K. S. and Prehn, J. H. M. (2018).Riluzole does not improve lifespan or motor function in three ALS mouse models.Amyotroph Lateral Scler Frontotemporal Degener 19, 438-445. doi:10.1080/21678421.2017.1407796

Jamaluddin, M. S., Weakley, S. M., Zhang, L., Kougias, P., Lin, P. H., Yao, Q. andChen, C. (2011). miRNAs: roles and clinical applications in vascular disease.Expert Rev. Mol. Diagn. 11, 79-89. doi:10.1586/erm.10.103

Keifer, O. P., Jr., O’connor, D. M. and Boulis, N. M. (2014). Gene and proteintherapies utilizing VEGF for ALS. Pharmacol. Ther. 141, 261-271. doi:10.1016/j.pharmthera.2013.10.009

Kieran, D., Sebastia, J., Greenway, M. J., King, M. A., Connaughton, D.,Concannon, C. G., Fenner, B., Hardiman, O. and Prehn, J. H. M. (2008).Control of motoneuron survival by angiogenin. J. Neurosci. 28, 14056-14061.doi:10.1523/JNEUROSCI.3399-08.2008

Lagier-Tourenne, C., Polymenidou, M. and Cleveland, D. W. (2010). TDP-43 andFUS/TLS: emerging roles in RNA processing and neurodegeneration. Hum. Mol.Genet. 19, R46-R64. doi:10.1093/hmg/ddq137

Lagier-Tourenne, C., Polymenidou, M., Hutt, K. R., Vu, A. Q., Baughn, M.,Huelga, S. C., Clutario, K. M., Ling, S.-C., Liang, T. Y., Mazur, C. et al. (2012).Divergent roles of ALS-linked proteins FUS/TLS and TDP-43 intersect inprocessing long pre-mRNAs. Nat. Neurosci. 15, 1488-1497. doi:10.1038/nn.3230

Lewandowski, S. A., Nilsson, I., Fredriksson, L., Lonnerberg, P., Muhl, L.,Zeitelhofer, M., Adzemovic, M. Z., Nichterwitz, S., Lawrence, D. A., Hedlund,E. et al. (2016). Presymptomatic activation of the PDGF-CC pathway acceleratesonset of ALS neurodegeneration. Acta Neuropathol. 131, 453-464. doi:10.1007/s00401-015-1520-2

Livak, K. J. and Schmittgen, T. D. (2001). Analysis of relative gene expression datausing real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods25, 402-408. doi:10.1006/meth.2001.1262

Lu, L., Zheng, L., Viera, L., Suswam, E., Li, Y., Li, X., Estevez, A. G. and King,P. H. (2007). Mutant Cu/Zn-superoxide dismutase associated with amyotrophiclateral sclerosis destabilizes vascular endothelial growth factor mRNA anddownregulates its expression. J. Neurosci. 27, 7929-7938. doi:10.1523/JNEUROSCI.1877-07.2007

Ludolph, A. C., Bendotti, C., Blaugrund, E., Chio, A., Greensmith, L., Loeffler,J. P., Mead, R., Niessen, H. G., Petri, S., Pradat, P.-F. et al. (2010). Guidelinesfor preclinical animal research in ALS/MND: A consensus meeting. AmyotrophLateral Scler. 11, 38-45. doi:10.3109/17482960903545334

Maimon, R., Ionescu, A., Bonnie, A., Sweetat, S., Wald-Altman, S., Inbar, S.,Gradus, T., Trotti, D., Weil, M., Behar, O. et al. (2018). miR126-5pDownregulation Facilitates Axon Degeneration and NMJ Disruption via a Non-Cell-Autonomous Mechanism in ALS. J. Neurosci. 38, 5478-5494. doi:10.1523/JNEUROSCI.3037-17.2018

Nolan, M., Talbot, K. and Ansorge, O. (2016). Pathogenesis of FUS-associatedALS and FTD: insights from rodent models. Acta Neuropathol Commun. 4, 99.doi:10.1186/s40478-016-0358-8

Petrov, D., Mansfield, C., Moussy, A. and Hermine, O. (2017). ALS clinical trialsreview: 20 years of failure. arewe any closer to registering a new treatment? Front.Aging Neurosci. 9, 68. doi:10.3389/fnagi.2017.00068

Picher-Martel, V., Valdmanis, P. N., Gould, P. V., Julien, J. P. and Dupre, N.(2016). From animal models to human disease: a genetic approach forpersonalized medicine in ALS. Acta Neuropathol Commun. 4, 70. doi:10.1186/s40478-016-0340-5

Pronto-Laborinho, A. C., Pinto, S. and de Carvalho, M. (2014). Roles of vascularendothelial growth factor in amyotrophic lateral sclerosis. Biomed Res Int. 2014,947513. doi:10.1155/2014/947513

Pupillo, E., Messina, P., Logroscino, G., Beghi, E. and Group, S. (2014). Long-term survival in amyotrophic lateral sclerosis: a population-based study. Ann.Neurol. 75, 287-297. doi:10.1002/ana.24096

Rademakers, R., Stewart, H., Dejesus-Hernandez, M., Krieger, C., Graff-Radford, N., Fabros, M., Briemberg, H., Cashman, N., Eisen, A., Mackenzie,I. R. A. et al. (2010). Fus gene mutations in familial and sporadic amyotrophiclateral sclerosis. Muscle Nerve. 42, 170-176. doi:10.1002/mus.21665

Ravits, J., Appel, S., Baloh, R. H., Barohn, R., Brooks, B. R., Elman, L., Floeter,M. K., Henderson, C., Lomen-Hoerth, C., Macklis, J. D. et al. (2013).Deciphering amyotrophic lateral sclerosis: what phenotype, neuropathology andgenetics are telling us about pathogenesis. Amyotroph Lateral SclerFrontotemporal Degener 14, 5-18. doi:10.3109/21678421.2013.778548

Robberecht, W. and Philips, T. (2013). The changing scene of amyotrophic lateralsclerosis. Nat. Rev. Neurosci. 14, 248-264. doi:10.1038/nrn3430

Rothstein, J. D. (2017). Edaravone: a new drug approved for ALS. Cell 171, 725.doi:10.1016/j.cell.2017.10.011

Rule, R. R., Schuff, N., Miller, R. G. and Weiner, M. W. (2010). Gray matterperfusion correlates with disease severity in ALS. Neurology 74, 821-827. doi:10.1212/WNL.0b013e3181d3e2dd

Scekic-Zahirovic, J., Oussini, H. E., Mersmann, S., Drenner, K., Wagner, M.,Sun, Y., Allmeroth, K., Dieterle, S., Sinniger, J., Dirrig-Grosch, S. et al. (2017).Motor neuron intrinsic and extrinsic mechanisms contribute to the pathogenesis ofFUS-associated amyotrophic lateral sclerosis. Acta Neuropathol. 133, 887-906.doi:10.1007/s00401-017-1687-9

Schober, A., Nazari-Jahantigh, M., Wei, Y., Bidzhekov, K., Gremse, F.,Grommes, J., Megens, R. T. A., Heyll, K., Noels, H., Hristov, M. et al. (2014).MicroRNA-126-5p promotes endothelial proliferation and limits atherosclerosis bysuppressing Dlk1. Nat. Med. 20, 368-376. doi:10.1038/nm.3487

7

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm040238. doi:10.1242/dmm.040238

Disea

seModels&Mechan

isms

Sebastia, J., Kieran, D., Breen, B., King, M. A., Netteland, D. F., Joyce, D.,Fitzpatrick, S. F., Taylor, C. T. and Prehn, J. H. M. (2009). Angiogenin protectsmotoneurons against hypoxic injury. Cell Death Differ. 16, 1238-1247. doi:10.1038/cdd.2009.52

Shelkovnikova, T. A., Peters, O. M., Deykin, A. V., Connor-Robson, N.,Robinson, H., Ustyugov, A. A., Bachurin, S. O., Ermolkevich, T. G.,Goldman, I. L., Sadchikova, E. R. et al. (2013). Fused in sarcoma (FUS)protein lacking nuclear localization signal (NLS) and major RNA binding motifstriggers proteinopathy and severe motor phenotype in transgenic mice. J. Biol.Chem. 288, 25266-25274. doi:10.1074/jbc.M113.492017

Sweeney, M. D., Kisler, K., Montagne, A., Toga, A. W. and Zlokovic, B. V. (2018).The role of brain vasculature in neurodegenerative disorders. Nat. Neurosci. 21,1318-1331. doi:10.1038/s41593-018-0234-x

Sweeney,M. D., Zhao, Z., Montagne, A., Nelson, A. R. and Zlokovic, B. V. (2019).Blood-brain barrier: from physiology to disease and back. Physiol. Rev. 99, 21-78.doi:10.1152/physrev.00050.2017

Turner, M. R., Hardiman, O., Benatar, M., Brooks, B. R., Chio, A., de Carvalho,M., Ince, P. G., Lin, C., Miller, R. G., Mitsumoto, H. et al. (2013). Controversiesand priorities in amyotrophic lateral sclerosis. Lancet Neurol. 12, 310-322. doi:10.1016/S1474-4422(13)70036-X

Van Es, M. A., Diekstra, F. P., Veldink, J. H., Baas, F., Bourque, P. R., Schelhaas,H. J., Strengman, E., Hennekam, E. A. M., Lindhout, D., Ophoff, R. A. et al.(2009). A case of ALS-FTD in a large FALS pedigree with a K17I ANG mutation.Neurology 72, 287-288. doi:10.1212/01.wnl.0000339487.84908.00

Wang, S., Aurora, A. B., Johnson, B. A., Qi, X., Mcanally, J., Hill, J. A.,Richardson, J. A., Bassel-Duby, R. and Olson, E. (2008). The endothelial-

specific microRNA miR-126 governs vascular integrity and angiogenesis. Dev.Cell 15, 261-271. doi:10.1016/j.devcel.2008.07.002

Winkler, E. A., Sengillo, J. D., Sullivan, J. S., Henkel, J. S., Appel, S. H. andZlokovic, B. V. (2013). Blood-spinal cord barrier breakdown and pericytereductions in amyotrophic lateral sclerosis. Acta Neuropathol. 125, 111-120.doi:10.1007/s00401-012-1039-8

Winkler, E. A., Sengillo, J. D., Sagare, A. P., Zhao, Z., Ma, Q., Zuniga, E., Wang,Y., Zhong, Z., Sullivan, J. S., Griffin, J. H. et al. (2014). Blood-spinal cord barrierdisruption contributes to early motor-neuron degeneration in ALS-model mice.Proc. Natl. Acad. Sci. USA. 111, E1035-E1042. doi:10.1073/pnas.1401595111

Wu, D., Yu, W., Kishikawa, H., Folkerth, R. D., Iafrate, A. J., Shen, Y., Xin, W.,Sims, K., Hu, G. (2007). Angiogenin loss-of-function mutations in amyotrophiclateral sclerosis. Ann. Neurol. 62, 609-617. doi:10.1002/ana.21221

Yamasaki, S., Ivanov, P., Hu, G. F. and Anderson, and P. (2009). Angiogenincleaves tRNA and promotes stress-induced translational repression. J. Cell Biol.185, 35-42. doi:10.1083/jcb.200811106

Yoshida, T., Sato, Y., Morita, I. and Abe, M. (2010). Pigpen, a nuclear coiled bodycomponent protein, is involved in angiogenesis. Cancer Sci. 101, 1170-1176.doi:10.1111/j.1349-7006.2010.01494.x

Zhang, T., Wu, Y. C., Mullane, P., Ji, Y. J., Liu, H., He, L., Arora, A., Hwang, H.-Y.,Alessi, A. F., Niaki, A. G. et al. (2018). FUS regulates activity of microRNA-mediated gene silencing.Mol. Cell 69, 787-801.e8. doi:10.1016/j.molcel.2018.02.001

Zhong, Z., Deane, R., Ali, Z., Parisi, M., Shapovalov, Y., O’banion, M. K.,Stojanovic, K., Sagare, A., Boillee, S., Cleveland, D. W. et al. (2008). ALS-causing SOD1 mutants generate vascular changes prior to motor neurondegeneration. Nat. Neurosci. 11, 420-422. doi:10.1038/nn2073

8

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm040238. doi:10.1242/dmm.040238

Disea

seModels&Mechan

isms