Embed Size (px)
Citation preview
Neurobiology of Disease 60 (2013) 11–17
Contents lists available at ScienceDirect
Neurobiology of Disease
j ourna l homepage: www.e lsev ie r .com/ locate /ynbd i
Genetic ablation of phospholipase C delta 1 increases survival inSOD1G93A mice
Kim A. Staats a,b,c, Lawrence Van Helleputte a,b,c, Ashley R. Jones d, André Bento-Abreu a,b,c,Annelies Van Hoecke a,b,c, Aleksey Shatunov d, Claire L. Simpson e, Robin Lemmens a,b,c,f, Tom Jaspers a,b,c,Kiyoko Fukami g, Yoshikazu Nakamura g, Robert H. Brown Jr. h, Philip Van Damme a,b,c,f, Adrian Liston i,j,Wim Robberecht a,b,c,f, Ammar Al-Chalabi d, Ludo Van Den Bosch a,b,c,⁎a Laboratory of Neurobiology, Belgiumb Leuven Research Institute of Neuroscience and Disease (LIND), KU Leuven, Belgiumc Vesalius Research Center, VIB, Belgiumd King's College London, Institute of Psychiatry, Department of Clinical Neuroscience, UKe Inherited Disease Research Branch, National Human Genome Research Institute, US National Institutes of Health, Baltimore, USAf Department of Neurology, University Hospital, Belgiumg Laboratory of Genome and Biosignal, Tokyo University of Pharmacy and Life Sciences, Hachioji-shi, Japanh Department of Neurology, University of Massachusetts Medical School, Worcester, USAi Autoimmune Genetics Laboratory, VIB, Leuven, Belgiumj Department of Microbiology and Immunology, KU Leuven, Belgium
Abbreviations: ALS, amyotrophic lateral sclerosis; Adipeptidyl-peptidase 6; ELP3, elongator protein 3; ER, end1,4,5-trisphosphate receptor 2; IP3, inositol 1,4,5-trisphospPLCD1, phosphoinositide phospholipase C; SDS-PAGE, sodTARDBP, TAR-DNA binding protein 43; UBQLN2, ubiquilin 2⁎ Corresponding author at: Campus Gasthuisberg, O&N
E-mail addresses: [email protected] (K.A. [email protected] (A. Bento-Abreu), [email protected] (R. Lemmens), [email protected] (R.H. Brown), philip.vandam(W. Robberecht), [email protected] (A. Al-Chala
Available online on ScienceDirect (www.sciencedire
0969-9961/$ – see front matter © 2013 Elsevier Inc. All rihttp://dx.doi.org/10.1016/j.nbd.2013.08.006
a b s t r a c t
a r t i c l e i n f oArticle history:Received 13 June 2013Revised 31 July 2013Accepted 7 August 2013Available online 19 August 2013
Keywords:Amyotrophic lateral sclerosisPhospholipase C delta 1NeurogeneticsSOD1-G93A miceExcitotoxicityMotor neuron diseaseNuclear shrinkage
Amyotrophic Lateral Sclerosis (ALS) is a devastating progressive neurodegenerative disease, resulting in selectivemotor neuron degeneration and paralysis. Patients die approximately 3–5 years after diagnosis. Disease pathophys-iology ismultifactorial, including excitotoxicity, but is not yet fully understood. Genetic analysis has proven fruitful inthe past to further understand genes modulating the disease and increase knowledge of diseasemechanisms. Here,we revisit a previously performedmicrosatellite analysis in ALS and focus on another hit, PLCD1, encoding phospho-lipase C delta 1 (PLCδ1), to investigate its role in ALS. PLCδ1 may contribute to excitotoxicity as it increases inositol1,4,5-trisphosphate (IP3) formation, which releases calcium from the endoplasmic reticulum through IP3 receptors.We find that expression of PLCδ1 is increased in ALSmouse spinal cord and in neurons fromALSmice. Furthermore,genetic ablation of this protein in ALS mice significantly increases survival, but does not affect astrogliosis,microgliosis, aggregation or the amount of motor neurons at end stage compared to ALSmice with PLCδ1. Interest-ingly, genetic ablation of PLCδ1 prevents nuclear shrinkage ofmotor neurons in ALSmice at end stage. These resultsindicate that PLCD1 contributes to ALS and that PLCδ1 may be a new target for future studies.
© 2013 Elsevier Inc. All rights reserved.
Introduction
Amyotrophic Lateral Sclerosis (ALS) is a devastating progressiveneurodegenerative disease, which involves the loss of motor neuronsand denervation of muscle fibres, resulting in muscle weakness and
MPA, α-amino-3-hydroxy-5-methyoplasmic reticulum; FUS/TLS, fused ihate; POAG, primary open-angle glaium dodecyl sulfate polyacrylamide g; UNC13a, unc-13 homolog A.4, PB 912, Herestraat 49, B-3000 Leuaats), Lawrence.vanhelleputte@[email protected] (A. Van [email protected] (T. Jaspers), [email protected] (P. Van Damm
bi), [email protected]).
ghts reserved.
paralysis. The disease has an annual incidence of 2.7 cases per 100,000people in Europe (Logroscino et al., 2010) and most patients succumbto the disease within 3 to 5 years of diagnosis. A family history of ALSis present in approximately 10% of cases. Family-based studies haveled to the identification of disease-causing mutations in the genes
l-4-isoxazolepropionic acid; C9orf72, chromosome 9 open reading frame 72; DPP6,n sarcoma/translocated in liposarcoma; GFAP, glial fibrillary acidic protein; ITPR2, inositolucoma; PBS, phosphate buffered saline; PCR, polymerase chain reaction; PFN1, profilin 1;el electrophoresis; SNP, single nucleotide polymorphism; SOD1, superoxide dismutase 1;
ven, Belgium. Fax: +32 16 330770.uleuven.be (L. Van Helleputte), [email protected] (A.R. Jones),cke), [email protected] (A. Shatunov), [email protected] (C.L. Simpson),[email protected] (K. Fukami), [email protected] (Y. Nakamura),e), [email protected] (A. Liston), [email protected] (L. Van Den Bosch).
12 K.A. Staats et al. / Neurobiology of Disease 60 (2013) 11–17
encoding superoxide dismutase 1 (SOD1) (Rosen et al., 1993), TAR-DNAbinding protein 43 (TARDBP) (Gitcho et al., 2008; Kabashi et al., 2008;Sreedharan et al., 2008), fused in sarcoma/translocated in liposarcoma(FUS/TLS) (Kwiatkowski et al., 2009; Vance et al., 2009), chromosome9 open reading frame 72 (C9orf72) (Dejesus-Hernandez et al., 2011;Renton et al., 2011), ubiquilin 2 (UBQLN2) and profilin 1 (PFN1)(reviewed in Andersen and Al-Chalabi, 2011). Disease-associated genevariants have also been identified in genes encoding elongator protein3 (ELP3) (Simpson et al., 2009), inositol 1,4,5-trisphosphate receptor 2(ITPR2) (van Es et al., 2007), dipeptidyl-peptidase 6 (DPP6) (van Eset al., 2008) and unc-13 homolog A (UNC13a) (van Es et al., 2009), al-though the exact role of these genes in ALS is not fully clear.
As disease is indistinguishable between familial and sporadic ALS, itis likely that there are common disease mechanisms involved. Suchmechanisms include aggregation, inflammation andmitochondrial dys-function (reviewed in Bruijn et al., 2004). Additionally, there is evidencefor the involvement of excitotoxicity as illustrated by the therapeutic ef-fect of riluzole, currently the only disease-modifying drug available(Miller et al., 2012). Excitotoxicity is the glutamatergic overstimulationof motor neurons leading to neurodegeneration by excessive cytosoliccalcium. Calcium influx can originate from the extracellular spacethrough stimulation of calcium-permeable α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors (Cid et al., 2003;Van Damme et al., 2003a). Additionally, calcium can also originatefrom intracellular stores such as the endoplasmic reticulum (ER)through activation of inositol 1,4,5-trisphosphate (IP3) receptors locat-ed in the ER membrane and that allow calcium to flow from the ERlumen into the cytosol after binding of IP3. This process may be detri-mental in ALS (Staats et al., 2012a).
We have previously reported the results of a microsatellite basedgenome-wide association study using 1884 markers, in which twopairs of markers, each about 1 Mb apart, were considered appropriatefor follow-up. At least one marker of each pair was ranked in the topfour results, and all markers were in the top 12 (Simpson et al., 2009).In conjunction with a parallel mutagenesis study in Drosophila thestudy of the first set of markers identified ELP3 variants on chromosome8 as associated with ALS (Simpson et al., 2009).
We have now studied the second pair of markers, which suggestPLCD1 as a candidate gene that modulates ALS. In addition we haveassessed the potential role of its gene product in ALS mice by geneticablation.
Material and methods
Genetic association
Details of the microsatellite association studymethods, patient pop-ulations and results are given elsewhere (Simpson et al., 2009). In brief,1884 polymorphic microsatellite markers were typed in three popula-tions comprising 1483 individuals using standard PCR of pooled DNAusing fluorescently labelled oligonucleotides followed by analysis offluorescence intensity patterns after electrophoresis using an AppliedBiosystems ABI3100 or 3130XL Genotyper (UK and Belgium) or LiCorGenotyping system (USA) and associated software. Allele image pat-ternswere converted to estimates of allele frequency using an automat-ed method (Simpson et al., 2009). Analysis of frequency estimates wasbymeta-regression, followed bymicrosatellite genotyping of individualDNA samples for confirmation of findings, prioritised for adjacentmarkers showing association in the top 1% of results. SNPs in andaround candidate genes were genotyped as part of genome-wide asso-ciation studies (Shatunov et al., 2010; van Es et al., 2007, 2008, 2009;Wain et al., 2009) and non-genotyped SNP alleles were imputed usingMaCH and minimac (Scott et al., 2007) with 1000 genomesV3.20101123. Build 37 (Hg19) of the human genome was used formapping.
Human material
All experiments on human DNA were approved by the Ethical Com-mittee of University Hospital Leuven and the London–Camberwell St.Giles Research Ethics Committee. We collected samples after weobtained informed consent from all human subjects.
Animal generation and housing
Mice overexpressing human SOD1WT and SOD1G93Awere purchasedfrom The Jackson Laboratories (Bar Harbor, USA) and maintained on aC57BL/6 background. PLCδ1 knockout mice were obtained on a mixed129/Ola × C57BL/6 background and subsequently backcrossed for an-other 3 generations prior to intercrossing females with SOD1G93A
males on a C57BL/6 background to obtain experimentalmice. Heterozy-gous PLCδ1+/− mice do not phenotypically differ from wild-type(PLCδ1+/+) mice and in line with a previous study (Hirata et al.,2011) we used these mice as controls. Chow and water were providedad libitum and mice were housed in the conventional animal facility ofKU Leuven under standard conditions according to the guidelines ofthe KU Leuven. All animal experiments were performed with the ap-proval of the Animal Ethical Committee of KU Leuven (020/2010).
Behavioural testing
The hanging wire test was used to determine disease onset byassessing the ability of the mice to hold their own weight for 60 s. Inbrief, the mice are placed on a wire grid and turned over while hangingupside–down.When amouse fails (drops from the grid before 60 s) andin consecutive trials cannot hold its ownweight for 60 s, it is defined assymptomatic. Disease onset determined by the hanging wire test wasused to calculate disease duration. Additionally, mice were weighedevery 5 days and relativeweightwas obtained by normalising the abso-lute weight to the average weight of each mouse between day 90 and105. For weight curves, mice no longer surviving were assessed as 0 g.End stage was defined as the age at which mice could no longer rightthemselves within 30 s when placed on their back. End stage is usedas a measurement of survival and is the moment when mice are eutha-nized to prevent further suffering.
Laser capture microscopy
Mice spinal cords were snap-frozen in Tissue-Tec (Sakura FinetekEurope, Alphen aan de Rijn, The Netherlands) to make cryostat sectionsof 20-μm thickness. Then, cresyl violet–stained motor neurons, locatedin the ventral horn of the lumbar spinal cord, were collected on mem-brane slides 1.0 PEN (Carl Zeiss AG, Oberkochen, Germany), using dis-section by a laser-capture microscope (Carl Zeiss AG) and capturing inAdhesive Cap 500 opaque (Carl Zeiss AG). Only motor neurons inwhich the nucleuswas visible andwith soma area N 250 μm2, were col-lected. We dissected at least 1500 motor neurons for each animal.
Quantitative PCR
Isolation ofmRNAoccurred by the TriPure (Roche, Basel, Switzerland)method and the RNeasy kit (Qiagen, Venlo, The Netherlands). Reversetranscriptase PCR used random hexamers (Life Technologies, Carlsbad,USA) and Moloney Murine Leukaemia Virus Reverse Transcriptase(MMLV RT; Invitrogen, Carlsbad, USA). Quantitative PCR was performedwith the StepOnePlus (Life Technologies) and TaqMan Universal PCRMaster Mix (Life Technologies). Gene expression assays were purchasedfrom Life Technologies and IDT DNA (Coralville, USA).
Table 1Highest ranked microsatellite markers associated to ALS by meta-regression analysis.
Rank Marker Chromosome Position Within gene
1 D3s1298 3 38,048,763 PLCD12 D3s1260 3 38,411,708 XYLB3 D10s564 10 92,599,673 HTR74 D8s1048 8 26,811,4875 D6s271 6 43,500,818 XPO56 CHLC.GATA143D05 16 50,594,541 NKD17 D12s800 12 104,410,367 GLT8D28 D3s3665 3 114,203,392 ZBTB209 D2s172 2 231,172,538 SP14010 D20s198 20 2,641,579 IDH3B11 D15s153 15 66,559,86312 D8s1820 8 27,997,347 ELP3
13K.A. Staats et al. / Neurobiology of Disease 60 (2013) 11–17
Nissl staining
To visualise neurons, Nissl staining was performed on 4% formalde-hyde fixed spinal cords sections. Sections were briefly immersed in acresyl violet solution and subsequently in a 70% ethanol with 10% aceticacid. Slides were dehydrated by an increased ethanol concentration se-ries and mounted with PerTex® (Histolab AB, Goteborg, Sweden). Im-ages were collected by Zeiss Axio Imager M1 microscope (Carl ZeissAG) with AxioCam Mrc5 camera (Carl Zeiss AG). The amount of(motor) neurons was quantified by measurement of the soma area asvisualised by cresyl violet staining in ImageJ (National Institute forHealth) onmultiple 40 μm thick sections in the ventral horn of the lum-bar spinal cord. Characterisation of motor neurons occurred as previ-ously described (Fischer et al., 2004).
Immunohistochemistry
Mice were transcardially perfused with phosphate buffered saline(PBS) and subsequently with 4% formaldehyde. Spinal cords werepost-fixed with 4% formaldehyde overnight at 4 °C and transferred to30% sucrose for an additional night. After snap freezing, tissue was sec-tioned by cryostat at 40 μm thickness and stained with SMI-32(Covance, Princeton, USA) for motor neurons and with a polyclonal an-tibody directed against ubiquitin (Dako, Glostrup, Denmark). Additionalstainings were performed with antibodies against glial fibrillary acidicprotein (GFAP, Santa Cruz Biotechnology, Santa Cruz, USA) and Iba1(Wako, Japan). Furthermore, to visualise the nucleus antibodies againstacetylated Histone 3 (Cell Signalling, Danvers, USA) were used. Second-ary antibodies for immunofluorescence include Alexa-555 and Alexa-488 (Invitrogen). Vectashield with DAPI (Vector, Burlingame, CA) wasused for mounting spinal cord sections. Images were collected by ZeissAxio ImagerM1microscope (Carl Zeiss AG)with AxioCamMrc5 camera(Carl Zeiss AG). Quantification of astrogliosis occurred by assessing therelativefluorescence of the ventral horn of GFAP staining by ImageJ (Na-tional Institute for Health, Bathesda, USA). Values were normalised tothe non-transgenic control group. Microgliosis was determined bycounting the amount of IbaI immunopositive cells per ventral hornwith the help of ImageJ. Additionally, ubiquitin immunopositive aggre-gates were also counted per ventral horn using ImageJ.
Western blot
Murine spinal cord homogenates were size separated through dena-turing sodiumdodecyl sulfate polyacrylamide gel electrophoresis (SDS–PAGE). An equal amount of protein, determined by Pierce BCA ProteinAssay (Pierce Biotechnology Inc., Rockford, USA), for each homogenatesample was heated at 100 °C for 5 min with an equivalent volume ofsample buffer (containing 8% sodium dodecyl sulfate (SDS) and 2%mercaptoethanol) and loaded onto a polyacrylamide gel. Protein waselectro-transferred to a nitrocellulosemembrane in Tris–glycine–meth-anol buffer. Themembrane was blocked for 1 h at room temperature ina blocking solution of 5% non-fat dry milk, 0.1% Tween 20, and TBS. Themembrane was incubated for 2 h at room temperature with primaryantibody. The following antibodies were used: anti-PLCδ1 (H140,Santa Cruz Technologies), anti-actin (A5441, Sigma). The membranewas incubated for 1 h at room temperature in peroxidase-labelledsecondary antibody and processed for analysis using a SuperSignalChemiLuminiscence detection kit (Pierce Biotechnology Inc.) asdescribed by the manufacturer.
Statistical analysis
Genetic analyses were described previously (Simpson et al., 2009).In brief, we used a modified meta-regression analysis in STATA of esti-mates of allele frequencies derived from DNA pools, with age and sexas covariates. We modelled a measurement error term in Mx, to allow
for stutter and size-based differential amplification of alleles which aregenotyping artefacts thatwould impact on the allele frequency estimateof pooled samples. The error termwas derived from data for 400micro-satellite markers typed in 16 individuals, allowing us to model the be-haviour of different repeat types. This was incorporated into the meta-regression by weighting each pool result by the inverse of the samplingandmeasurement error. Results of microsatellite genotyping of individ-ual as opposed to pooled DNA samples were analysed using the pro-gramme CLUMP (Sham and Curtis, 1995). Association of SNP alleles inand around candidate genes was analysed using the programmePLINK (Purcell et al., 2007). Non-genetic analyses were performedwith the statistical software package Prism Origin (GraphPad Software,La Jolla, USA). Survival and disease onset was analysed by Log–Ranktesting. Differences over 2 groups with comparable variance wereanalysed by Student's t-test. Significance was assumed at p b 0.05.Data are presented as mean ± standard error of the mean.
Results
PLCD1 is genetically associated with ALS
As reported previously (Simpson et al., 2009), we performed a mi-crosatellite analysis over 781 ALS cases and 702 controls. The highestranked microsatellite markers for association with ALS by meta-regression analysis were D3s1298 and D3s1260, less than 400 Kbapart on chromosome 3 (Table 1). Marker D3s1298 lies within thegene PLCD1, coding for phospholipase C delta1, and close to the geneVILL, coding for villin.Marker D3s1260 lies within the gene XYLB, codingfor xylulokinase homolog B. Because these markers fulfilled the criteriafor follow-up, we performed genotyping of individual samples(Simpson et al., 2009). Alleles of marker D3s1298 were associatedwith ALS (meta-analysis P = 5 × 10−4), although the association wasnot observed in the US population. There was no allele consistentlyover- or under-represented in cases rather than controls across thethree countries. In the 400 Kb region between the two markers are 8genes, PLCD1, DLEC1, ACAA1, MYD88, OXSR1, SLC22A13, SLC22A14,andXYLB. Because the top rankedmarker lieswithin PLCD1 and becausePLCD1 is a strong candidate for involvement in ALS a priori for its in-volvement in excitotoxicity, we tested the hypothesis that PLCδ1 isfunctionally important in motor neuron degeneration by studying anALS mouse model and investigating levels of PLCδ1 in ALS mice.
PLCδ1 expression is increased in SOD1G93A mice
PLCδ1 expression is reportedly increased upon excitotoxic insultsin vitro (Shimohama et al., 1995a) and in brain samples of Alzheimer pa-tients (Shimohama et al., 1995b). To investigate whether PLCδ1 is alsoupregulated in ALS mice, we investigated the expression of PLCδ1 inSOD1G93A mice. Ventral spinal cords analysed by qPCR show an in-creased relative gene expression of plcd1 at end stage in SOD1G93A
mice (Fig. 1A). Western blot analysis of spinal cords of end stage
14 K.A. Staats et al. / Neurobiology of Disease 60 (2013) 11–17
SOD1G93A and non-transgenic control mice confirmed this upregulationat the protein level (Figs. 1B and C). Additionally, we show that plcd1gene expression is increased by neurons in SOD1G93A mice at 60 daysof age and at 130 days of age compared to neurons from age-matchedSOD1WT mice (Fig. 1D). This implies that the upregulation detected inspinal cord homogenate of SOD1G93A mice is at least in part due to in-creased upregulation of PLCδ1 by neurons.
PLCδ1 knockout is protective in ALS mice
PLCδ1−/− mice have been previously characterised and show vary-ing degrees of hairlessness (Nakamura et al., 2003). To assess the effectof PLCδ1 in ALS mice, we interbred PLCδ1–/– mice with SOD1G93A mice.PLCδ1−/− SOD1G93Amice showa trend for a delayed onset of symptomsas compared to PLCδ1+/− SOD1G93Amice (Fig. 2A). This delay is also ob-served in the start of weight loss of PLCδ1−/− SOD1G93A compared toPLCδ1+/− SOD1G93Amice (Fig. 2B) and the disease duration is very sim-ilar for both genotypes (36.3 ± 3.2 versus 34.3 ± 2.8 days; Fig. 2C).Moreover, PLCδ1–/– SOD1G93A mice live on average 7.3 days longercompared to PLCδ1+/– SOD1G93A mice (p = 0.02; Fig. 2D), implying aprotective role of the genetic ablation of PLCδ1 in ALS.
To verify that PLCδ1−/− SOD1G93Amice reach a similar terminal stageof the disease, end stage pathology of these mice was assessed. A signif-icant decrease of neurons was detected in all end stage mice indepen-dent of genotype (Supplementary Material Fig. 1C). In addition, adultPLCδ1–/– mice have similar amounts of neurons compared to adultcontrol mice (Supplementary Material Fig. 1C). Comparably, the num-ber of ubiquitin-positive aggregates per ventral horn (SupplementaryMaterial Fig. 1 F), astrogliosis (Supplementary Material Fig. 1I) andmicrogliosis (Supplementary Material Fig. 1 L) are all increased at endstage and there is no difference between genotypes. These results indi-cate that end stage pathology is similar across genotypes and that themice became end stage due to similar disease conditions.
Lower weight and compromisedweight gain have been reported forPLCδ1−/− mice (Hirata et al., 2011) and as the effect of low weight andcompromised weight gain in ALS is currently unknown, we closely
Fig. 1.Expression of plcd1/PLCδ1 inALSmouse spinal cord. Relative expression of plcd1 in ventramice (n = 5; A). PLCδ1 protein as determined by Western blot relative to β-actin (B) in spinaltification of Western blot data from sub-image B (C). Relative plcd1 gene expression determineand SOD1G93A mice of 60 days old (n = 2 and n = 3, respectively) and of SOD1WT and SOD1G
monitored the weight of the mice. No difference in absolute weight atpresymptomatic stagewas detected for female (SupplementaryMateri-al Fig. 2A) or male PLCδ1+/– SOD1G93A and PLCδ1−/−SOD1G93A mice(Supplementary Material Fig. 2B). This indicates that a difference inweight does not play a role in the current experiment.
PLCδ1 knockout prevents nuclear shrinkage in ALS mice
PLCδ1 knockout is reported to be protective in excitotoxic paradigmsin vitro by preventing nuclear shrinkage (Okada et al., 2010). To assesswhether PLCδ1 induces nuclear shrinkage of motor neurons in ALS, wemeasured the size of the nucleus of SMI-32 positive neurons in the ven-tral horn. The soma size of end stage mice is lower than in the controlcondition (Fig. 3C), which is expected as large motor neurons aremost vulnerable to ALS-induced degeneration (Pun et al., 2006) andhave most likely degenerated at the moment of the analysis (endstage). As the area of the soma of SMI-32 positive neurons ofSOD1G93A and PLCδ1−/−SOD1G93A mice at end stage are not different,the difference of nuclear shrinkage is due to the absolute area of the nu-cleus, where the control SOD1G93A mice have smaller nuclei in neuronsthan PLCδ1−/−SOD1G93A mice at end stage (Fig. 3D). This is confirmedby the analysis of the relative nuclear area between the control miceand SOD1G93A at end stage, whereas the decrease of relative nucleararea is fully rescued in PLCδ1−/−SOD1G93A mice (Fig. 3B). To minimiseany potential bias by assessing larger neurons in mice at baseline orfor also assessing neurons b250 μm2, linear regression was performedfor all neurons assessed per genotype (Supplementary Material Fig. 3).This implies that of the conditions assessed nuclear shrinkage only oc-curs in SOD1G93A end stage mice independent of the size of the neuron.
Discussion
We previously reported the results of a microsatellite-basedgenome-wide association study in ALS documenting the potential roleof ELP3 in ALS (Simpson et al., 2009). This association of ELP3 and ALShas been recently found in a small cohort of US veterans with ALS
l spinal cords assessed byqPCRof non-transgenic (non-tg, n = 6) and end stage SOD1G93A
cord homogenate of non-transgenic (non-tg, n = 4) and SOD1G93A mice (n = 4). Quan-d by qPCR in neurons dissected by laser capture microscopy from spinal cords of SOD1WT
93A mice of 130 days old (n = 3 and n = 3, respectively; D). *: p b 0.05, **: p b 0.01.
Fig. 2. Genetic removal of PLCδ1 is protective in ALS mice. Onset of symptoms as measured by the hanging wire test in PLCδ1+/− SOD1G93A mice (117.4 ± 1.6 days, n = 30) and PLCδ1−/−
SOD1G93A mice (122.7 ± 1.9 days, n = 22, p = 0.08; A). Weight measurements of PLCδ1−/− SOD1G93A mice and PLCδ1+/− SOD1G93A mice (p b 0.0001; B). Disease duration of PLCδ1+/−
SOD1G93A mice and of PLCδ1−/− SOD1G93A mice (C). Survival curve of PLCδ1+/− SOD1G93A mice (151.7 ± 2.1 days, n = 30) and PLCδ1−/− SOD1G93A mice (159.0 ± 2.8 days, n = 22; p =0.024, D).
15K.A. Staats et al. / Neurobiology of Disease 60 (2013) 11–17
(Kwee et al., 2012). In this study, we report a second finding from themicrosatellite study which associates a microsatellite marker in PLCD1with ALS. In addition,we observed anupregulation of PLCδ1, the proteinproduct of plcd1, and of plcd1 gene expression in neurons of ALS mice.Moreover, the removal of PLCδ1 in ALS mice improves survival. All
Fig. 3. Nuclear shrinkage of motor neurons in SOD1G93A mice is counteracted by PLCδ1 knockoacetylated histone immunoreactive nucleus (green) of adult control, PLCδ1−/− , end stage (ES) Simmunoreactive neurons from adult control (n = 91 neurons from 5mice) and PLCδ1−/−miceand PLCδ1−/− SOD1G93A mice (n = 29 neurons from 4mice) at end stage (B). Absolute soma aSOD1G93A and PLCδ1−/− SOD1G93A mice at end stage. Scale bar = 25 μm. *: p b 0.05, **: p b 0
together, these data indicate that PLCδ1 plays a detrimental role in theselective motor neuron death during ALS.
Themicrosatellite association was not replicated in all three popula-tions (Simpson et al., 2009), and the signal came fromdifferent alleles ineach case, with allele 15 associated with increased risk in the UK
ut. SMI-32 immunoreactive neurons (red) in the ventral horn of the spinal cord and theirOD1G93A and end stage (ES) PLCδ1−/− SOD1G93Amice (A). Relative nuclear area of SMI-32(n = 36 neurons from 4mice) and from control SOD1G93A (n = 30 neurons from4mice)rea (C) and nuclear area (D) of neurons from control and PLCδ1−/− mice and from control.01, ***: p b 0.001.
16 K.A. Staats et al. / Neurobiology of Disease 60 (2013) 11–17
population and reduced risk in the Belgian population, and allele 11 as-sociated with increased risk in the Belgian population and being essen-tially of equal frequency in cases and controls in the UK population. Themost plausible explanation for this observation is that themicrosatellitemarker is highly polymorphic, with the alleles tagging a functional var-iant nearby, but the tagging allele is different in each country.While thisimproves power for detection of association over themarker alleles as awhole, it means that the total counts are reduced for any one allele, anddifferences between cases and controls are more prone to samplingvariance.
PLCD1 expression is increased in the spinal cord of ALS mice. This issimilar to the increase of PLCδ1 in brain homogenate of Alzheimer dis-ease where it also co-localises with neurofibrillary tangles (Shimohamaet al., 1995b), implying a role of PLCD1 in other neurodegenerative dis-eases. A possible role may be the upregulation of PLCδ1 duringexcitotoxicity, as PLCδ1 is increased dose-dependently in vitro in embry-onic rat cortical neurons after exposing them to increased concentrationsof glutamate (Shimohama et al., 1995a). Of the many isoforms of PLC,PLC delta 1 is one of the most sensitive to Ca2+ and is expressedamong others in hair follicles, all regions of the brain and in testis (Leeet al., 1999; Nakamura et al., 2005). Phospholipase C hydrolyses phos-phatidylinositol 4,5-bisphosphate (PIP2) to diacylglycerol (DAG) andIP3. Increase of IP3 formation results in calcium release from the ERthrough IP3 receptors. This additionally implies a role for ER calcium inALS, as deletion of PLCδ1 in keratinocytes decreases cytosolic calciumlevels (Nakamura et al., 2003) and thatwe additionally show that geneticablation of PLCδ1 improves survival of SOD1G93Amice. This is in linewithour recent results implying that increasing ER calciumrelease in neurons,by overexpressing IP3 receptor 2, is detrimental in ALSmice (Staats et al.,2012a). Interestingly, inhibiting ER calcium release by other ER calciumchannels, the ryanodine receptors, does not alter the survival of ALSmice when treated with dantrolene (Staats et al., 2012b). This addition-ally shows that targeting ER calcium release is less effective at increasingsurvival of ALS mice (7.3 days in this study) than pharmacologicallytargeting extracellular calcium influx with AMPA receptor antagonists(approximate increase of 14–20 days of survival (Canton et al., 2001;Tortarolo et al., 2006; Van Damme et al., 2003b)).
We additionally show that genetic ablation of PLCδ1 fully rescues thenuclear shrinkage of motor neurons in ALS mice. Whether nuclearshrinkage is a detrimentalmechanismor a bystander effect in ALS is un-known. Nuclear shrinkage occurs during apoptosis of neurons(Takadera and Ohyashiki, 2007) and the absence of such shrinkage inPLCδ1–/–SOD1G93A mice could imply that there is less apoptosis inthesemice. PLCδ1 hydrolyses PIP2 and thus PLCδ1−/−micemay have in-creased levels of PIP2. PIP2 exerts an anti-apoptotic function byinhibiting caspase 3 (Azuma et al., 2000), a key molecule in the induc-tion of apoptosis. However, the extent of apoptosis in motor neuronsduring ALS is unclear (Yamazaki et al., 2005), although increases in sur-vival are observed in ALS mice when the apoptotic pathway is targetedgenetically (Gould et al., 2006; Kostic et al., 1997).
A role for PLCD1 is also associated to other disorders includingleukonychia (porcelain nails) by decreased function (Kiuru et al.,2011) and coronary vasospasm by increased enzymatic activity(Shibutani et al., 2012). Microsatellite analysis also implies that PLCD1may be associated to primary open-angle glaucoma (POAG) (Bairdet al., 2005). Interestingly, mutations in optineurin, another ALS gene(Maruyama et al., 2010), also cause POAG (Rezaie et al., 2002), indicat-ing a possible overlap of genes associated to these diseases.
In conclusion,microsatellite analysis in ALS patients and controls hasimplied an association between genetic variation in PLCD1 and ALS. Ad-ditionally, we show that plcd1 expression is increased in SOD1G93A
mouse spinal cord and motor neurons. The potential relevance of thisis shown by the increased survival of ALS mice in the absence of plcd1,which correlates to a decreased nuclear shrinkage, and may be of inter-est for developing novel therapeutic strategies for ALS patients by phar-macologically inhibiting PLCδ1.
Acknowledgments
AAC thanks the ALS Association andMotor Neurone Disease Associa-tion of England,Wales and Northern Ireland for support. He receives sal-ary support from the NIHR Dementia Biomedical Research Unit at SouthLondon andMaudsley NHS Foundation Trust and King's College London.This work was supported by grants from the “Fund for Scientific Re-search Flanders” (FWO-Vlaanderen), the University of Leuven (KU Leu-ven), the Belgian Government (Interuniversity Attraction Poles,programme P7/16 of the Belgian Federal Science Policy Office), the ALSTherapy Alliance, the Angel Fund and the European Community's HealthSeventh Framework Programme (FP7/2007–2013 under grant agree-ment 259867). PVD and RL hold a clinical investigatorship of the FWO-Vlaanderen.WR is supported through the E von Behring Chair for Neuro-muscular and Neurodegenerative Disorders. CLS is funded by the Intra-mural Research Program of the National Human Genome ResearchInstitute, National Institutes of Health. The views expressed are thoseof the authors and not necessarily those of the National Health Service,the NIHR, or the Department of Health.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.nbd.2013.08.006.
References
Andersen, P.M., Al-Chalabi, A., 2011. Clinical genetics of amyotrophic lateral sclerosis:what do we really know? Nat. Rev. Neurol. 7, 603–615.
Azuma, T., et al., 2000. Gelsolin in complex with phosphatidylinositol 4,5-bisphosphate in-hibits caspase-3 and -9 to retard apoptotic progression. J. Biol. Chem. 275, 3761–3766.
Baird, P.N., et al., 2005. Evidence for a novel glaucoma locus at chromosome 3p21–22.Hum. Genet. 117, 249–257.
Bruijn, L.I., et al., 2004. Unraveling the mechanisms involved in motor neuron degenera-tion in ALS. Annu. Rev. Neurosci. 27, 723–749.
Canton, T., et al., 2001. RPR 119990, a novel α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid antagonist: synthesis, pharmacological properties, and activ-ity in an animal model of amyotrophic lateral sclerosis. J. Pharmacol. Exp. Ther. 299,314–322.
Cid, C., et al., 2003. Low concentrations of glutamate induce apoptosis in cultured neu-rons: implications for amyotrophic lateral sclerosis. J. Neurol. Sci. 206, 91–95.
Dejesus-Hernandez, M., et al., 2011. Expanded GGGGCC hexanucleotide repeat in noncodingregion of C9ORF72 causes chromosome 9p-linked FTD and ALS. Neuron 72, 245–256.
Fischer, L.R., et al., 2004. Amyotrophic lateral sclerosis is a distal axonopathy: evidence inmice and man. Exp. Neurol. 185, 232–240.
Gitcho, M.A., et al., 2008. TDP-43 A315T mutation in familial motor neuron disease. Ann.Neurol. 63, 535–538.
Gould, T.W., et al., 2006. Complete dissociation of motor neuron death from motor dys-function by Bax deletion in a mouse model of ALS. J. Neurosci. 26, 8774–8786.
Hirata, M., et al., 2011. Genetic defect in phospholipase Cdelta1 protects mice from obesi-ty by regulating thermogenesis and adipogenesis. Diabetes 60, 1926–1937.
Kabashi, E., et al., 2008. TARDBP mutations in individuals with sporadic and familialamyotrophic lateral sclerosis. Nat. Genet. 40, 572–574.
Kiuru,M., et al., 2011. Hereditary leukonychia, or porcelain nails, resulting frommutationsin PLCD1. Am. J. Hum. Genet. 88, 839–844.
Kostic, V., et al., 1997. Bcl-2: prolonging life in a transgenic mouse model of familialamyotrophic lateral sclerosis. Science 277, 559–562.
Kwee, L.C., et al., 2012. A high-density genome-wide association screen of sporadic ALS inUS veterans. PLoS One 7, e32768.
Kwiatkowski Jr., T.J., et al., 2009. Mutations in the FUS/TLS gene on chromosome 16 causefamilial amyotrophic lateral sclerosis. Science 323, 1205–1208.
Lee, W.K., et al., 1999. Molecular cloning and expression analysis of a mouse phospholi-pase C-delta1. Biochem. Biophys. Res. Commun. 261, 393–399.
Logroscino, G., et al., 2010. Incidence of amyotrophic lateral sclerosis in Europe. J. Neurol.Neurosurg. Psychiatry 81, 385–390.
Maruyama, H., et al., 2010. Mutations of optineurin in amyotrophic lateral sclerosis. Na-ture 465, 223–226.
Miller, R.G., et al., 2012. Riluzole for amyotrophic lateral sclerosis (ALS)/motor neuron dis-ease (MND). Cochrane Database Syst. Rev. 3 (CD001447).
Nakamura, Y., et al., 2003. Phospholipase Cdelta1 is required for skin stem cell lineagecommitment. EMBO J. 22, 2981–2991.
Nakamura, Y., et al., 2005. Phospholipase C-delta1 and -delta3 are essential in the tropho-blast for placental development. Mol. Cell. Biol. 25, 10979–10988.
Okada, M., et al., 2010. Calcium fluxes cause nuclear shrinkage and the translocation ofphospholipase C-delta1 into the nucleus. Neurosci. Lett. 472, 188–193.
Pun, S., et al., 2006. Selective vulnerability and pruning of phasic motoneuron axons inmotoneuron disease alleviated by CNTF. Nat. Neurosci. 9, 408–419.
17K.A. Staats et al. / Neurobiology of Disease 60 (2013) 11–17
Purcell, S., et al., 2007. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am. J. Hum. Genet. 81, 559–575.
Renton, A.E., et al., 2011. A hexanucleotide repeat expansion in C9ORF72 is the cause ofchromosome 9p21-linked ALS-FTD. Neuron 72, 257–268.
Rezaie, T., et al., 2002. Adult-onset primary open-angle glaucoma caused by mutations inoptineurin. Science 295, 1077–1079.
Rosen, D.R., et al., 1993. Mutations in Cu/Zn superoxide dismutase gene are associatedwith familial amyotrophic lateral sclerosis. Nature 362, 59–62.
Scott, L.J., et al., 2007. A genome-wide association study of type 2 diabetes in Finns detectsmultiple susceptibility variants. Science 316, 1341–1345.
Sham, P.C., Curtis, D., 1995. Monte Carlo tests for associations between disease and allelesat highly polymorphic loci. Ann. Hum. Genet. 59, 97–105.
Shatunov, A., et al., 2010. Chromosome 9p21 in sporadic amyotrophic lateral sclerosis in theUK and seven other countries: a genome-wide association study. Lancet Neurol. 9,986–994.
Shibutani, S., et al., 2012. Coronary vasospasm induced in transgenic mouse with in-creased phospholipase C-delta1 activity. Circulation 125, 1027–1036.
Shimohama, S., et al., 1995a. Glutamate-induced antigenic changes of phospholipase C-delta in cultured cortical neurons. J. Neurosci. Res. 41, 418–426.
Shimohama, S., et al., 1995b. Differential involvement of phospholipase C isozymes inAlzheimer's disease. Gerontology 41 (Suppl. 1), 13–19.
Simpson, C.L., et al., 2009. Variants of the elongator protein 3 (ELP3) gene are associatedwith motor neuron degeneration. Hum. Mol. Genet. 18, 472–481.
Sreedharan, J., et al., 2008. TDP-43 mutations in familial and sporadic amyotrophic lateralsclerosis. Science 319, 1668–1672.
Staats, K.A., et al., 2012a. Neuronal overexpression of IP(3) receptor 2 is detrimental inmutant SOD1 mice. Biochem. Biophys. Res. Commun. 429, 210–213.
Staats, K.A., et al., 2012b. Dantrolene is neuroprotective in vitro, but does not affect surviv-al in SOD1G93A mice. Neuroscience 220, 26–31.
Takadera, T., Ohyashiki, T., 2007. Caspase-dependent apoptosis induced by calcineurin in-hibitors was prevented by glycogen synthase kinase-3 inhibitors in cultured rat cor-tical cells. Brain Res. 1133, 20–26.
Tortarolo, M., et al., 2006. Glutamate AMPA receptors change in motor neurons ofSOD1(G93A) transgenic mice and their inhibition by a noncompetitive antagonistameliorates the progression of amytrophic lateral sclerosis-like disease. J. Neurosci.Res. 83, 134–146.
Van Damme, P., et al., 2003a. Chloride influx aggravates Ca2+-dependent AMPA receptor-mediated motoneuron death. J. Neurosci. 23, 4942–4950.
Van Damme, P., et al., 2003b. The AMPA receptor antagonist NBQX prolongs survival in atransgenic mouse model of amyotrophic lateral sclerosis. Neurosci. Lett. 343, 81–84.
van Es, M.A., et al., 2007. ITPR2 as a susceptibility gene in sporadic amyotrophic lateralsclerosis: a genome-wide association study. Lancet Neurol. 6, 869–877.
van Es, M.A., et al., 2008. Genetic variation in DPP6 is associated with susceptibility toamyotrophic lateral sclerosis. Nat. Genet. 40, 29–31.
van Es, M.A., et al., 2009. Genome-wide association study identifies 19p13.3 (UNC13A)and 9p21.2 as susceptibility loci for sporadic amyotrophic lateral sclerosis. Nat.Genet. 41, 1083–1087.
Vance, C., et al., 2009. Mutations in FUS, an RNA processing protein, cause familialamyotrophic lateral sclerosis type 6. Science 323, 1208–1211.
Wain, L.V., et al., 2009. The role of copy number variation in susceptibility to amyotrophiclateral sclerosis: genome-wide association study and comparisonwith published loci.PLoS One 4, e8175.
Yamazaki, M., et al., 2005. Is motoneuronal cell death in amyotrophic lateral sclerosis ap-optosis? Neuropathology 25, 381–387.
