differential expression

This article was originally published in a journal published byElsevier, and the attached copy is provided by Elsevier for the

author’s benefit and for the benefit of the author’s institution, fornon-commercial research and educational use including without

limitation use in instruction at your institution, sending it to specificcolleagues that you know, and providing a copy to your institution’s

administrator.

All other uses, reproduction and distribution, including withoutlimitation commercial reprints, selling or licensing copies or access,

or posting on open internet sites, your personal or institution’swebsite or repository, are prohibited. For exceptions, permission

may be sought for such use through Elsevier’s permissions site at:

http://www.elsevier.com/locate/permissionusematerial

http://www.elsevier.com/locate/permissionusematerial

Autho

r's

pers

onal

co

py

Differential expression of molecular motors in the motor cortex ofsporadic ALS

Maria Pantelidou,a Spyros E. Zographos,a,1 Carsten W. Lederer,a Theodore Kyriakides,b

Michael W. Pfaffl,c and Niovi Santamaa,⁎

aDepartment of Biological Sciences, University of Cyprus and Cyprus Institute of Neurology and Genetics, P.O. Box 20537, 1678 Nicosia, CyprusbThe Cyprus Institute of Neurology and Genetics, Nicosia, CypruscPhysiology-Weihenstephan, Center of Life and Food Sciences, Technical University of Munich, Germany

Received 18 October 2006; revised 31 January 2007; accepted 4 February 2007Available online 16 February 2007

The molecular mechanisms underlying the selective neurodegenerationof motor neurons in amyotrophic lateral sclerosis (ALS) are inade-quately understood. Recent breakthroughs have implicated impairedaxonal transport, mediated by molecular motors, as a key element fordisease onset and progression. The current work identifies theexpression of 15 kinesin-like motors in healthy human motor cortex,including three novel isoforms. Our comprehensive quantitative mRNAanalysis in control and sporadic ALS (SALS) motor cortex specimensdetects SALS-specific down-regulation of KIF1Bβ and novel KIF3Aβ,two isoforms we show to be enriched in the brain, and also of SOD1, akey enzyme linked to familial ALS. This is accompanied by a markedreduction of KIF3Aβ protein levels. In the motor cortex KIF3Aβlocalizes in cholinergic neurons, including upper motor neurons. Nomutations causing splicing defects or altering protein-coding sequenceswere identified in the genes of the three proteins. The present studyimplicates twomotor proteins as possible candidates in SALS pathology.© 2007 Elsevier Inc. All rights reserved.

Keywords: Real-time RT-PCR; Motor neuron disease; Kinesin-likeproteins; KIF3A; KIF1Bβ

Introduction

Amyotrophic lateral sclerosis (ALS) is an invariably fatal,adult-onset neurological disorder characterized by upper and lowermotor neuron degeneration in motor cortex, brainstem, and spinalcord, leading to progressive atrophy and paralysis of skeletalmuscles. Approximately 10% of ALS is familial (FALS) and about20% of FALS cases are associated with mutations in the copper/zinc superoxide dismutase 1 (SOD1) gene (Deng et al., 1993;Rosen et al., 1993). The remaining 90% of ALS are without familyhistory and therefore considered sporadic (SALS) (Al-Chalabi andLeigh, 1998). Although the molecular mechanisms of motorneuron death are still unclear, several models have been putforward to explain the pathogenesis of ALS. These includeoxidative stress (Ferrante et al., 1997), glutamate excitotoxicity(Lin et al., 1998; Howland et al., 2002 but also see Meyer et al.,1999), excitotoxic influx of Ca2+ through AMPA receptors(Kawahara et al., 2004; Kuner et al., 2005), neurofilamentaggregation (Williamson and Cleveland, 1999), mitochondrialdysfunction (Kirkinezos et al., 2005; Manfredi and Xu, 2005),apoptosis (Estevez et al., 1999; Martin, 2000), and others.

Recent breakthroughs have illustrated that the intracellulartransport machinery of neurons is a major cellular target for theinitiation or progression of neurodegeneration in motor neurondisease (reviewed by Goldstein, 2003; Levy and Holzbaur, 2006).Key players of the machinery that mediates cytoskeleton-based,bidirectional transport are a large number of “molecular motors”,otherwise known as motor proteins, together with adaptors,effectors and regulators of transport complexes as well as diverseinteracting proteins and cellular cargoes. Motor proteins have beenimplicated in the pathogenesis of motor neuron disorders by anumber of studies. Early evidence indicated that mutations in thefast axonal transport motor protein kinesin in Drosophila causeorganelle jams that disrupt fast axonal transport bidirectionally,leading to defective action potentials, dystrophic terminals,reduced transmitter secretion, and progressive distal paralysis

www.elsevier.com/locate/ynbdiNeurobiology of Disease 26 (2007) 577–589

Abbreviations: ALS, amyotrophic lateral sclerosis; CMT 2A, Charcot–Marie–Tooth motor neuronopathy type 2A; FALS, familial amyotrophiclateral sclerosis; KLPs, kinesin-like proteins; ORF, open reading frame;Real-time RT-PCR, real-time reverse transcription-polymerase chain reac-tion; RG, reference gene; SALS, sporadic amyotrophic lateral sclerosis;SNP, single nucleotide polymorphism; SOD1, superoxide dismutase 1⁎ Corresponding author. Fax: +357 22 350557.E-mail address: [email protected] (N. Santama).

1 Present address: Institute of Organic and Pharmaceutical Chemistry,National Hellenic Research Foundation, Athens, Greece.

Available online on ScienceDirect (www.sciencedirect.com).

0969-9961/$ - see front matter © 2007 Elsevier Inc. All rights reserved.doi:10.1016/j.nbd.2007.02.005

Autho

r's

pers

onal

co

py

(Hurd and Saxton, 1996), all typical manifestations of motorneuron disorders. More recent work revealed that a loss-of-function mutation in the motor protein KIF1Bβ, which transportssynaptic vesicle precursors along the axon, is a cause of the mostcommon inherited human peripheral neuropathy, Charcot–Marie–Tooth (CMT) disease type 2A (Zhao et al., 2001). Furthermore,hereditary spastic paraplegia (SPG10) was found to be associatedwith a missense mutation in the motor domain of the neuronalkinesin heavy chain KIF5A gene (Reid et al., 2002). Intriguingly,targeted disruption of the KIF5A heavy chain in transgenic micecauses abnormal neurofilament transport (Xia et al., 2003), ahallmark cytopathic manifestation of motor neuron disease. Alongthe same line, functional inhibition or mutations of the motordynein/dynactin complex, involved in retrograde transport, cause alate-onset progressive motor neuron degeneration in mouse modelsthat phenocopies human motor neuron disease (LaMonte et al.,2002; Puls et al., 2003; Hafezparast et al., 2003). Surprisingly, amutation in dynein in the SOD1 G93A mutant background of aFALS mouse model rescues axonal transport defects and prolongssurvival (Kieran et al., 2005). Therefore, independent and diversestudies have established a tantalizing link between defective motorproteins that compromise axonal transport and the pathogeneticmechanisms of different types of motor neuron disorders.

Mindful of the possible involvement of motor proteins in ALSpathology, we therefore set out to assess their expression in motorneurons of SALS patients. To achieve this goal we performed as aninitial step the first, to our knowledge, systematic identification ofkinesin-like motor proteins expressed in cultured rat motor neuronsand normal human motor cortex. We then conducted a comprehen-sive analysis of the mRNA expression of the human motors in post-mortem motor cortex specimens of SALS patients, using isoform-specific quantitative real-time RT-PCR. For a novel isoform ofKIF3A, KIF3Aβ, showing the highest significant mRNA down-regulation (<50% of normal values), this analysis was complemen-ted by quantitative protein expression analysis and immunolocaliza-tion in the motor cortex. We also quantified mRNA expression fordifferentially expressed candidate genes in many human tissues andassessed by genomic sequencing the possibility that mutations inthese genes might be involved in SALS pathology.

Materials and methods

Human samples

Precentral gyrus sections from post-mortem human brains wereused for genomic DNA and RNA extraction. All brain samplesused in this study were provided by the University of Miami Brainand Tissue Bank for Developmental Disorders in Florida US in2000 and 2002 through NICHD contract #NO1-HD-8-3284.Eleven brain samples were derived from clinically diagnosedsporadic ALS patients, and nine samples originated from non-ALSindividuals, who died from unrelated causes (Table 2A). Sampleswere fresh-frozen in 2-methylbutane and stored at −80 °C untiluse.

Peripheral blood, used for genomic DNA isolation, wascollected from thirteen Cypriot SALS patients fulfilling the ElEscorial criteria for definite ALS, and six healthy individuals(Table 2B). They were provided through the Cyprus Institute ofNeurology and Genetics, a national ALS referral center.

All samples used in this study were procured followinginstitutionally approved bioethical procedures pertaining to the

collection, transport and use of human tissue and protection ofpersonal data and with the informed written consent of donors.

Transgenic mice

Mouse strains B6SJL-TgN(SOD1-G93A)1Gur (an ALS mousemodel carrying the disease-causing G93A mutation in an SOD1human transgene) and B6SJL-TgN(SOD1)2Gur (a reference straincarrying a wild-type SOD1 human transgene), originally obtainedfrom the Jackson Laboratory, were verified by PCR genotyping oftail material and used for real-time RT-PCR. The animals usedwere 60 days old without ALS-like symptoms, 90 days olddisplaying the onset of neurodegenerative symptoms, and end-stage animals of 126 days of age.

RNA extraction and cDNA preparation

Total RNA was isolated from human brain tissue (∼100 mg),using the TRIzol reagent according to the manufacturer’sprotocol (Invitrogen) with the following modifications: after thephase separation step of the protocol, one volume of 70%ethanol was added and each sample transferred to a mini column(RNeasy mini kit, Qiagen). RNA extraction and DNase Itreatment were conducted according to the RNeasy mini kitinstructions. RNA from other human tissues (Fig. 2) waspurchased from Ambion Inc. First strand cDNA was synthesizedfrom 1 μg of RNA using the ProtoScript Synthesis kit (NewEngland Biolabs), according to the manufacturer’s protocol. Ratspinal motor neuron culture cDNA was kindly donated by Dr.Giampietro Schiavo (Molecular Neuropathology Laboratory,Cancer Research, UK).

Oligonucleotides

All oligonucleotides used in this work were synthesized byMWG Biotech (Germany) (Tables S1 and S2). They were designedin a manner that would allow the specific amplification ofindividual isoforms within each motor family, and their specificitywas confirmed prior to use.

Real-time RT-PCR

Real-time RT-PCR was performed on the LightCycler (RocheMolecular Biochemicals) using the primers listed in Table S1.Oligonucleotides were designed to amplify short diagnostic cDNAfragments, appropriate for real-time RT-PCR and specific for eachof the 13 cDNAs tested. Resulting amplicons were assessed byagarose electrophoresis, which revealed in each case a singleproduct of the predicted size (data not shown).

Each 20-μl reaction included 2 μl LightCycler FastStart DNAMaster SYBR Green I, 4 mM of MgCl2, 0.5 μM each of forwardand reverse primer, and 2 μl of 1:2.5 (v/v) diluted cDNA (madefrom 1 μg total RNA). In each experiment an internal standardcurve was used, which consisted of five serial dilutions of a cDNAmix in duplicate. The experiment was performed according to thefollowing protocol: (i) denaturation at 95 °C for 10 min, (ii)amplification (30–40 cycles) at 95 °C for 15 s, 55–65 °C for 10 s,depending on primer sequence (Table S1), and 72 °C for 10–28 swith a single fluorescence measurement (521 nm), depending onamplicon size (Table S1), (iii) melting at 60–75 °C for 15 s withcontinuous fluorescence measurement up to 95 °C, and (iv)

578 M. Pantelidou et al. / Neurobiology of Disease 26 (2007) 577–589

Autho

r's

pers

onal

co

py

cooling at 40 °C for 30 s. All experiments included two no-template controls and all samples were analyzed independentlybetween two and four times. Crossing Point (CP) determinationand quantification were achieved using the Second DerivativeMaximum Method of the LightCycler Software 3.5 (RocheMolecular Biochemicals). For data normalization, the expressionof the following reference genes (RGs) was also quantified: beta-2-microglobulin (B2M), ribosomal protein L19 (RPL19), andpumilio homolog 1 transcript variant 2 (PUM1). Melting curveanalysis was performed at the end of each experiment to determinethe specificity of the amplification.

Statistical analysis

The data for real-time RT-PCR were analyzed by REST-384©(November 2005), a Microsoft-Excel-based software which usespermutation analysis, specifically a Pair Wise Fixed ReallocationRandomization Test©, to compare the relative quantificationbetween two groups and to determine the significance of theresults (Pfaffl et al., 2002). The geometric mean of the CPs from allthree RGs employed in this study was used for gene expressiondata normalization, and 10,000 permutations were used forstatistical evaluation.

The multi-tissue quantitative analysis was based on thecalculated concentrations provided by the LightCycler Software3.5 using the Second Derivative Maximum Method. The calculatedconcentration of RPL19 reference gene in each type of tissue wasused for gene expression data normalization and the standarddeviation was calculated from three independent experiments foreach gene.

Antibodies and immunofluorescence with frozen sections

A mouse monoclonal antibody, raised to the C-terminal aminoacids (aa) 563–671 of human KIF3A, (611508, BD Biosciences)was used at 1:250 dilution in Western blotting and at 1:50 forimmunofluorescence for the detection of KIF3A isoforms. Anantibody to choline acetyl transferase (ChAT) (AB144P, BDBiosciences) was used at 1:25 for immunofluorescence. Alexa-Fluor 488-labeled donkey anti-mouse IgG (A11057, MolecularProbes) and Alexa Fluor 568-labeled donkey anti-goat IgG(A21202, Molecular Probes) secondary antibodies were used at1:2000 and 1:1500, respectively. For the detection of α-tubulin inWestern blotting, a mouse monoclonal antibody (T5168, Sigma)was used at 1:8000 dilution and HRP-conjugated secondaryantibody to mouse IgG (NA931, Amersham Pharmacia Biotech)was used at 1:10,000 dilution.

For immunofluorescence, frozen motor cortex slices werefixed in acetone at −20 °C and blocked with 5% BSA/0.5%Triton X-100 in PBS for 1 h. Incubation with both primaryantibodies in 0.05% Tween/PBS was performed overnight at4 °C followed by incubation with secondary antibodies at roomtemperature for 1 h. All washes were done with PBS. Toquench lipofuscin autofluorescence, slices were incubated inAutofluorescence Eliminator Reagent (Chemicon) for 20 minand washed with 70% ethanol. Nuclei were stained withHoechst33342 and mounted for microscopy in 10% Mowiol,1% 1,4-Diazabicyclo[2.2.2]octane, and 25% glycerol in 0.1 MTris buffer (pH 8). Images were acquired using an Axiovert200M microscope with the AxioVision 4.5 image acquisitionsoftware (Carl Zeiss Inc.).

SDS–PAGE, Western blotting and quantification

Following SDS–PAGE as per standard methods, Westernblotting was carried out with the semi-dry method (transfer bufferwas 10 mM NaHCO3, 3 mM NaCO3, 20% methanol, pH 9.9).Visualization of immunoreactive bands was performed by theenhanced chemiluminescence system (Amersham PharmaciaBiotech). Quantification of signals was carried out with the ScionImage software 4.0.3.2 (Scion Corporation). Statistical analysis ofprotein expression levels was performed using the individualaverages for each triplicate sample and comparing patient andcontrol samples by a heteroscedastic two-tailed t-test, i.e. a two-sample t-test not assuming equal variance.

Human genomic DNA extraction, amplification and sequencing

Extraction of human genomic DNA from peripheral blood(nineteen donors; H11–H16 and D12–D24) was carried outaccording to a modified protocol based on Miller et al. (1988)and from frozen motor cortex specimens (eleven donors; D1-D11)using the TRIzol reagent according to the manufacturer’s instruc-tions (Invitrogen).

In human, the KIF1B gene is located on chromosome 1 (genomiccontig NT086572) and consists of 47 exons, of which exons 1–11encode the motor domain of the protein. The gene for KIF3A islocated on chromosome 5 (genomic contig NT086668) and consistsof 17 exons, of which exons 2–8 encode the motor domain. The genefor SOD1 is on chromosome 21 (genomic contig NT011512) andconsists of 5 exons that encode its entire ORF. To screen formutations, PCR fragments corresponding to exons 1–11 of theKIF1B gene, exons 2–8 of the KIF3 gene and exons 1–5 of SOD1were amplified. The PCR primers were designed to bracket theexons at sufficient distance from their respective 5′ and 3′ ends toinclude intronic sequences containing the characteristic intron/exonand branch site consensus sequences (Table S2). Amplification ofselected sequences from all 30 samples was carried out by PCR,using 500 ng genomic DNA, 0.8 μMof each oligonucleotide and 1Uof Taq polymerase for 33 cycles. Purification and sequencing ofPCR products were carried out by MWG Biotech (Germany).

Other molecular biology techniques

All other molecular biology techniques used for KLPidentification in rat motor neurons and human motor cortex(standard PCR, restriction digests, ligations, blue/white bacterialcolony screening, plasmid purification, Southern blotting, etc.)were performed according to standard procedures. T/A cloningvector pCR2.1 was purchased from Invitrogen and used accordingto the manufacturer’s instructions.

Results

Identification of kinesin-like motor proteins expressed in rat spinalmotor neurons and human motor cortex

Despite the wealth of information about the identification andexpression of kinesin-like motors (KLPs) in many human tissues,little is known about their specific expression in motor neurons inhuman or other mammals, a shortcoming addressed in the presentstudy. The obvious starting point of our analysis was the iden-tification of distinct KLP motors in motor neurons. In the absence of

579M. Pantelidou et al. / Neurobiology of Disease 26 (2007) 577–589

Autho

r's

pers

onal

co

py

a human equivalent for routine analyses, we used rat spinal motorneuron cultures as the source of initial identification. The results ofthe analysis in rat were then verified, expanded, and refined with theuse of human motor cortex tissue, enriched in motor neurons.

For the purpose of KLP identification in rat motor neurons, a pairof degenerate oligonucleotide primers was designed (primer pair #1,Table S1) (Santama, 2001) specific for the extremely well-conserved“signature motifs” shared by the motor domains of all known KLPs(Hirokawa et al., 1998). A PCR using this degenerate primer pairwith cDNA synthesized from a homogeneous primary culture of ratspinal motor neurons as template resulted in a mixed product of500 bp that was subcloned in the T/A vector pCR2.1. A total of 372recombinant clones, carrying partial motor protein cDNAs, wereanalyzed by a combination of restriction digest mapping, Southernblotting and nucleotide sequencing. This revealed the presence ofseven distinct types of KLP cDNAs: KIF1A, KIF1Bβ3, KIF2,KIF2C, KIF3A, KIFC3, and KIFC5A (Table 1). Additionally, wespecifically tested for the expression of KLPs that were not pickedup in our screen but were known to be important in the nervoussystem, namely KIF5A (neuronal kinesin heavy chain) (Niclas et al.,1994), KIF5B (conventional kinesin heavy chain) (Navone et al.,1992), the motor neuron-enriched KIF5C (Kanai et al., 2000) andtheKIFC5B isoform (Navolanic and Sperry, 2000). Of those, KIF5Aand KIF5C were found to be abundantly expressed, while KIF5Band KIFC5B were barely detectable. In conclusion, the combination

of these approaches led to the identification of nine KLPs in ratspinal motor neurons (Table 1).

Having identified KLPs in rat spinal motor neurons, oligonu-cleotide primers for their human orthologues, their variants and forother human KLPs possibly relevant to ALS pathology weredesigned and used to identify KLPs in a human motor cortexspecimen (healthy donor H8) by RT-PCR. Systematic PCRanalysis of partial cDNAs, and eventual full-length cloning ofopen reading frames (ORFs) combined with cDNA sequencing,confirmed the expression of fifteen kinesin-like motors in healthyhuman motor cortex (Table 1). These KLPs included members ofthe kinesin-1 family (KIF5A and KIF5C/KIAA531), the kinesin-2family (KIF3A, KIF3Aβ, KIF3B and KIF3C), the kinesin-3 family(KIF1A, KIF1Bα, KIF1Bβ, KIAA591/KIF1Bβ and KIF1C), thekinesin-14 family (C-terminal kinesin KIFC3), and also KLPsKIF13A2, KIF21A variant, and KIF21B variant (Table 1). Amongthese were novel KLP isoforms, namely KIF3Aβ, KIF21A variant,and KIF21B variant (deposited to the EMBL database withaccession numbers AM177178–80, respectively).

Analysis by real-time RT-PCR in healthy and SALS motor cortexshows significant down-regulation of mRNAs for KLPs KIF1Bβ,KIF3Aα, and for SOD1

Of 15 KLP motors detected in the human motor cortex (Table 1),13 cDNAs could be reliably quantified in post-mortem specimensof fresh-frozen motor cortex tissue of eleven clinically diagnosedSALS patients and nine age- and sex-matched control subjects(Table 2A) by real-time RT-PCR. Our analysis also included SOD1,whose mutations are a causative event in 20% of FALS and in asmall proportion of SALS. Two of the KLPs expressed in motorneurons were not included in the real-time RT-PCR analysis, eitherbecause of insufficient specificity of multiple primer sets tested(KIF1Bβ NM015074) or because the abundance of the ampliconwas consistently low, indicating low-level expression of theparticular cDNA in the motor cortex (KIF21B variant). To allowthe normalization of candidate gene expression (Bustin, 2002;Dheda et al., 2005) we also performed the quantification of multiplereference genes, whose expression does not vary across theexperimental conditions, specifically B2M, RPL19, and PUM1(Vandesompele et al., 2002; Pfaffl et al., 2004).

The results of relative quantification and statistical analysis arelisted in Table 3 and expressed as the fold-change of normalizedexpression levels between diseased and healthy samples. Of the 13KLP cDNAs tested, two motors were statistically significantlydown-regulated (p<0.05) in human SALS motor cortex tissue (Fig.1). Specifically, KIAA0591/KIF1Bβ was down-regulated by afactor of 0.686 (or 68.60± 1.56% of normalized average controlvalue) and the novel isoform KIF3Aβ by a factor of 0.425 (42.49±2.35%) (Fig. 1). The expression levels of the remaining KLPsshowed no statistically significant difference, although all membersof the kinesin-2 (KIF3) family showed marked down-regulation.Intriguingly, the expression of SOD1 in SALS patients was alsosignificantly down-regulated by a factor of 0.713 (71.31±2.80%)(p<0.05) (Table 3, Fig. 1).

KIF3Aβ and KIAA0591/KIF1Bβ are isoforms particularly enrichedin the human central nervous system including the motor cortex

We next wanted to test the gene expression pattern of the twomotors that appeared to be significantly down-regulated in human

Table 1Identification of KLPs in rat spinal motor neurons in culture and in thehealthy human motor cortex

KLP expression EMBL accession no.

Rat motor neuronsKIF1A AM180765KIF1Bβ3 AB120429KIF2 XM345150KIF2C XM579644KIF3A AM180763KIF5A AY535015KIF5C XM241981KIFC3 AM180764KIFC5A NM001005878

Human motor cortexKIF1A NM004321KIF1Bα (KIAA1448) NM183416KIF1Bβ NM015074KIAA0591/KIF1Bβ AB017133KIF1C NM006612KIF3A NM007054KIF3Aβ AM177178KIF3B NM004798KIF3C NM002254KIFC3 NM005550KIF5A NM004984KIF5C AB011103KIF13A2 AJ291579KIF21A variant AM177179KIF21B variant AM177180

Rat cDNA sequences KIF1A, KIF3A and KIFC3, previously included in theEMBL database as “predicted" species, were submitted as new, experimen-tally-confirmed entries with new accession nos., as shown. Human isoformsKIF3Aβ, KIF21Avariant and KIF21B variant are novel and deposited in theEMBL database with the accession nos. shown.


Autho

r's

pers

onal

co

py

SALS samples and in particular assess their relative levels ofexpression in the motor cortex. A quantitative RT-PCR analysis ofmRNA levels with ten non-brain adult human tissues and five brainareas, including normal motor cortex, occipital (visual), temporaland parietal (somatosensory) cortex, and cerebellum specimens,revealed that KIF3Aβ is particularly enriched in all brain areasexamined, including motor cortex, with other tissues having barelydetectable or little expression of this isoform (Fig. 2). Our analysisalso showed that KIAA0591/KIF1Bβ mRNA was also highlyenriched specifically in the brain, including the motor cortex, with

testis being the only other tissue of those tested with substantialexpression levels (Fig. 2).

The pattern of expression was different for KIF3A, the isoformvery closely related to KIF3Aβ but not significantly downregulatedin SALS. KIF3A appeared mostly ubiquitous and abundant acrosstissues and brain areas (with the notable exceptions of the heart,ovary, and striated muscle) (Fig. 2). KIF3A and KIF3Aβ areidentical at the nucleotide sequence level, except for an additionalnucleotide stretch of 81 bp, uniquely present in KIF3Aβ (EMBLaccession no. AM177178). This encodes a 27 aa peptide of about2.5 kDa (Fig. 3). Interestingly, this peptide sequence (aa 409–436)lies within a protein region that interrupts the predicted coiled-coildomain (known as rod) of the motor protein (Fig. 3). The presence ofthe 27-aa peptide appears to extend this coil-breaking region, ascompared with the KIF3A isoform (Fig. 3) and thus seems importantfor the protein structure and possibly its protein–protein interactions.

Protein expression of KIF3Aβ is markedly down-regulated in theSALS motor cortex

To analyze KIF3Aβ expression in the motor cortex at theprotein level we performed SDS–PAGE followed by quantitativeWestern immunoblotting on protein extracts of six SALS and threecontrol motor cortex specimens.

For the detection of KIF3Aβ we used a mouse monoclonalantibody raised against a C-terminal segment of KIF3A (aa 563–671) that would recognize both KIF3A isoforms. Indeed, accordingto the product documentation and our own observations it detectsat least two proteins between 80 and 85 kDa, which wouldcorrespond to KIF3A and KIF3Aβ and can be resolved as separatesignals (Fig. 4A, arrows). Quantification of total KIF3A-likeimmunoreactivity revealed that in SALS motor cortex the averagenormalized protein levels reach only 56.43±15.16% (SD) ofcontrol levels (Fig. 4B). This difference in protein levels was foundto be statistically significant (p=0.008).

Table 2SALS and control populations

Diseased Controls

A. Brain sample groupsNumber of individuals 11 9Sample codes D1–D11 H1–H4, H6–H10Male/Female 7/4 4/5Age of death (mean±SD) [years] 68±7.89 66.44±10.16Post-mortem interval (mean±SD)

[hours]18.57±4.43 13.19±6.07

B. Blood sample groupsNumber of individuals 13 6Sample codes D12–D24 H11–H16Male/Female 5/8 3/3Age of ALS onset (mean±SD) [years] 51.62±11.15 N/AAge of blood sampling

(mean±SD) [years]58.54±11.53 45.67±23.48

A. Details of the SALS and control populations, used for the quantitativeanalysis of motor protein gene expression in the motor cortex (as shown inTable 3).B. Details of the SALS and control populations, used for the DNA extractionfrom peripheral blood. In this study a total number of 24 SALS cases(D1–11 included in A and D12–24 in B) and six control individuals(H11–16, B) were used for genomic DNA extraction and sequence analysis.

Table 3Quantitative results of real-time RT-PCR for human subjects

Expression of 13 KLPs and of SOD1 was assessed in the human motorcortex of SALS individuals. The results from two or three independentexperiments listed in this table are expressed as normalized fold change ofdiseased versus control (geometric average of all measurements perspecimen±standard deviation). Highlighted cDNA species were found tobe significantly down-regulated (p<0.05) in SALS patients. A more detailedpresentation of data can be found in Table S3.

Fig. 1. mRNA levels of motor proteins KIF1Bβ and KIF3Aβ and SOD1 arestatistically significantly down-regulated in SALS motor cortex (p<0.05).The quantitative results of real-time RT-PCR for SALS specimens areexpressed as a percentage of the normalized average control values percDNA. Asterisk denotes statistical significance. Error bars correspond tostandard deviation of three independent experiments.


Autho

r's

pers

onal

co

pyCellular localization of KIF3Aβ in the human motor cortex

Having established that both mRNA and protein levels ofKIF3Aβ are statistically significantly down-regulated in SALSpatients, we next examined its cellular expression in the motor

cortex. We therefore conducted double immunofluorescenceexperiments on frozen sections representing layers I–VI of themotor cortex (Fig. 5A, panel A) using antibodies to ChAT (amarker of presynaptic cholinergic neurons) and the anti-KIF3A/KIF3Aβ antibody. Cell nuclei were marked with the fluorescent

Fig. 2. KIAA0591/KIF1Bβ and KIF3AβmRNAs are highly enriched in the human central nervous system, including the motor cortex, among the tissues tested,whereas KIF3A is ubiquitously expressed across most of the fifteen tissues or brain areas tested. Quantitative analysis by real-time RT-PCR of the levels ofKIF3A, KIF3Aβ, and KIAA0591/KIF1Bβ cDNAs was carried out using identical amounts of RNA isolated from the human tissues shown. For the motor cortexa sample from a healthy individual (H3) was used for the analysis. Error bars correspond to standard deviation of three independent experiments. AU: arbitraryunits.

Fig. 3. Predicted probability of coiled-coil formation and protein domains in KIF3A and KIF3Aβ. Prediction of coils (KIF3A in blue and KIF3Aβ in black) wasbased on the algorithm of Lupas et al. (1991) and domain analysis was performed using the Predict Protein Server (Rost and Liu, 2003). Sketches of the proteinoverall organization with the 27-aa sequence uniquely present in KIF3Aβ are shown at the bottom of the panel.


Autho

r's

pers

onal

co

py

dye Hoechst. We discovered that in both healthy and diseasedsamples and throughout the layers of the motor cortex KIF3A/KIF3Aβ immunoreactivity co-localized in a mixed population ofcholinergic cells including pyramidal neurons and non-pyramidalcells (Fig. 5A, panels B1 and B2 and C1 and C2). In layer V inparticular the Betz cells, giant upper motor neurons, and in layersIII and V smaller pyramidal and non-pyramidal neurons wereimmunoreactive for KIF3A/KIF3Aβ (Fig. 5B). The co-localizationof KIF3A/KIF3Aβ with the cholinergic marker ChAT was over95% throughout the motor cortex layers.

These results revealed that KIF3Aβ is abundantly expressed inthe motor cortex, including layer V upper motor neurons.

Expression levels of KIF1Bβ and KIF3Aβ mRNAs are not alteredthroughout disease progression in the motor cortex of SOD1 G93Amice, a model of FALS

Following the identification of KIAA0591/KIF1Bβ andKIF3Aβ as two motor proteins specifically down-regulated insporadic cases of ALS in human, we were interested in testingwhether these changes in expression levels would also bemanifested in a widely used mouse model of SOD1-inducedALS, B6SJL-TgN(SOD1-G93A)1Gur. We first confirmed theexistence of the KIF3Aβ isoform in mouse by cloning fromhealthy adult mouse motor cortex and sequencing a partialmouse cDNA spanning the sequences unique to the humanKIF3Aβ ortholog (Table S1, AM231689). We then carried outreal-time RT-PCR with oligonucleotides specific for mouseKIF3Aβ, analyzing the mutant mice against age-matchedB6SJL-TgN(SOD1)2Gur controls. Three groups of animals wereused: animals 60 days old, a stage when no disease symptomscan be detected, 90-day-old mice, at an age typically within theonset period of the first ALS-like symptoms, and in 126-day-oldmice, corresponding to end-stage, close to death. In all three agegroups and for both cDNAs the fold changes detected werepractically unchanged in reference to the healthy, controltransgenic strain and any small changes were not statisticallysignificant (p>0.05) (Fig. 6), indicating that, throughout the

progression of the disease, the molecular physiology of mutantSOD1-induced ALS is different from that of the human sporadiccases under study.

Genomic sequencing of the human KIF1B and KIF3A motordomain, and of SOD1 in SALS and healthy individuals

We next wanted to test whether the reduced levels of expressionfor the motors KIAA0591/KIF1Bβ and KIF3Aβ, in the particularindividuals analyzed in our study, were a result of or wereaccompanied by mutations in the corresponding genes. This waspertinent given that the CMT 2A motor neuronopathy has beenassociated with a specific mutation of the ATP-binding site in theKIF1Bβ motor domain (Zhao et al., 2001).

Genomic DNA from 24 sporadic ALS cases and six age-matched non-ALS individuals was subjected to nucleotidesequencing (Table 2B). The 24 ALS cases comprised the elevenpatients also analyzed by real-time RT-PCR and an additional 13SALS individuals from Cyprus (Table 2B).

Sequencing of the KIF1B gene motor domain revealed fourknown single nucleotide polymorphisms (SNPs) and a novelgenotype in the 30 individuals examined (Table 4). The knownSNPs were present in disease and control samples andcorresponded to rs12402052 (in ten disease and one controlsamples), rs484609 (in twelve disease and two control samples),rs12141246 (in twelve disease and one control samples), andrs1339458 (in nine disease and two control samples) of thedbSNP (NCBI). The first SNP is a synonymous substitution inexon 2 (A95), whereas the other three are intronic, not affectingthe consensus splice or branch sites. The novel genotype wasdetected in three patients (two heterozygotes and one homo-zygote) and corresponded to a T/G intronic substitution, 110 bpupstream of the 3′ splice site of exon 8 (ss52050753, ss52085996;Table 4).

In the KIF3A gene, seven patients and one healthy individualcarried the known intronic rs9784600 SNP (Table 4).

In the SOD1 gene, we detected known intronic SNPsrs16988404 and rs2234694 in a number of patients, and a novel

Fig. 4. Protein levels of KIF3A/Aβ are reduced in SALS motor cortex. (A) Western immunoblotting of equivalent protein extracts of 3 healthy (H series) and 6SALS (D series) specimens probed with an antibody recognizing KIF3A and KIF3Aβ (top panel, 50 μg of protein; arrows indicate the expected positions of thetwo proteins) or with an antibody for α-tubulin, serving as a normalizing internal standard (bottom panel). (B) Quantification of results showing normalizedaverage values of healthy specimens (set to 100±13.09%) and SALS specimens (56.43±15.16%) reveals statistically significant reduction of KIF3A-likeimmunoreactivity (p=0.008). Error bars correspond to standard deviation of three independent experiments.


Autho

r's

pers

onal

co

py


Autho

r's

pers

onal

co

pyexonic genotype in a heterozygous patient, corresponding to a T/Gsubstitution in the 3′ UTR (ss65640445; Table 4).

In conclusion, genomic sequence variations in the motordomains of KIF1B and KIF3A or the entire ORF of SOD1 wereeither not patient-specific or did not appear to alter RNA splicingor amino acid composition of the corresponding protein.

Discussion

This work sought to contribute to the understanding of the linkbetween disruption in molecular motors, mediating axonaltransport in motor neurons, and the pathophysiology of ALS. Itwas prompted by the finding that neurodegenerative diseases,such as CMT 2A, hereditary spastic paraplegia, and forms ofmotor neuron disease, are in some cases associated with mutationsin specific motor protein genes (see Introduction). Furthermore,there are indications that axonal transport impairment may beinvolved in an even wider spectrum of human neurodegenerativediseases, including Huntington’s disease (Gunawardena et al.,2003) and early axonopathy in Alzheimer’s disease (Stokin et al.,2005).

We chose to conduct this analysis in the motor cortex, whereupper motor neurons are located, rather than the spinal cord withlower motor neurons. Despite the general depletion of upper motorneurons in ALS (Kaufmann et al., 2004), most investigations ofALS focus on lower motor neurons, possibly owing to earlierclinical manifestations of defects in lower motor neurons and theireasier accessibility. It is in line with this general trend that ALS-related defects of axonal transport are under-investigated in uppermotor neurons, despite the reduction of corticospinal tract volumein ALS patients, indicating axonal degeneration of upper motorneurons (Wang et al., 2006). Naturally, signaling of upper motorneurons via their long axons in the corticospinal tract is requiredfor lower motor neuron function (Purves et al., 2001) and subject tothe same maintenance constraints acting on lower motor neurons.The differential effect of ALS pathology on upper and lower motorneurons is of high clinical relevance as it creates problems in thedefinition and diagnosis of ALS types (Ince et al., 2003), and inthis respect a further characterization of ALS motor cortex is ofgeneral interest. Importantly, patients with SALS show alterationsin the motor cortex, such as increased excitability and reducedinhibitory activity, which are not readily detectable in (SOD1-linked) FALS patients, thus stressing the general relevance ofinvestigating upper motor neurons in SALS (Turner et al., 2005).

Investigating global motor protein expression in the humanmotor cortex of SALS patients, we first determined which kinesin-like motor proteins (KLPs) are expressed in healthy motor cortex.The 15 KLP motors detected include KIF5A and KIF5C/KIAA531 (kinesin-1 family), KIF3A, KIF3Aβ, KIF3B and KIF3C(kinesin-2), KIF1A, KIF1Bα, KIF1Bβ, KIAA591/KIF1Bβ andKIF1C (kinesin-3), C-terminal kinesin KIFC3 (kinesin-14), andalso KLPs KIF13A2, KIF21A variant, and KIF21B variant (Table1). Three of these motors, KIF3Aβ, KIF21A variant and KIF21Bvariant, are novel isoforms of the known KIF3A, KIF21A, andKIF21B proteins, respectively. We then compared, by quantitativeanalysis, combining real-time RT-PCR and statistical evaluationusing three reference genes, the mRNA levels of all specificallydetectable KLPs and of SOD1 in nine healthy individuals and elevenSALS patients. This analysis did not show a generalized disruptionof the motor protein expression pattern, but instead revealed aspecific down-regulation of motors KIAA591/KIF1Bβ (69% ofhealthy levels), KIF3Aβ (43% of healthy levels), and, interestingly,of SOD1 (71% of healthy levels). Although none of the fold-changesobserved was dramatic as an average across samples, all weresignificant (p<0.05) and reached as low as 19% of the controlaverage in some patients. Importantly, quantitative analysis ofprotein expression for KIF3Aβ corroborated these findings byshowing reproducible and statistically significant reduction ofcorresponding protein levels. We are confident that the down-regulation observed was not a generic effect, due to the loss of

Fig. 5. Cellular localization of KIF3Aβ in the human motor cortex. A. (A) Experiments were carried out with representative parallel sections of motor cortex(precentral gyrus, area 4) including all layers, here shown after Nissl staining. This section was from SALS patient D1, and some loss of giant pyramidal cells inareas of layer V is visible (w.m. is white matter). (B1 and B2) Frozen sections from the same patient D1 (panel B1) and control individual H6 (panel B2) wereprocessed for double immunofluorescence with an antibody against ChAT (red) and an antibody recognizing KIF3A and KIF3Aβ (green). DNAwas visualizedwith Hoechst (blue). KIF3A/KIF3Aβ immunoreactivity co-localized in a mixed population of cholinergic cells (overlay of red and green fluorescence givesyellow) including small and large pyramidal neurons and smaller non-pyramidal cells (open arrowheads). Labeled processes (arrows) were regularly visible [(d)is diseased; (h) is healthy]. Scale bar 50 μm. (C1 and C2) A detail of localization from patient D1 at higher magnification is shown. Scale bar 20 μm. B. KIF3Aβis expressed in cholinergic cells including layer V upper motor neurons in the human motor cortex. Frozen sections from control samples H4 and H6 wereprocessed with an antibody against ChAT (red), an antibody recognizing KIF3A and KIF3Aβ (green), and DNA was visualized with Hoechst (blue).Representative examples from layer V, showing the giant Betz motor neurons (marked with asterisks) with the characteristic lipofuscin granules (see panel C1)and perisomatic dendrites (see panel A2 bottom), are shown in pairs. Scale bars 20 μm.

Fig. 6. Quantitative results of real-time RT-PCR for FALS SOD1 G93Amice. Expression of mouse KIF1Bβ and KIF3Aβ was assessed in the motorcortex of the SOD1-G93A transgenic strain B6SJL-TgN(SOD1-G93A)1GurFALS model and reference B6SJL-TgN(SOD1)2Gur strain (60, 90 and 126days old). The quantitative results of real-time RT-PCR for FALS mice areexpressed as a percentage of the normalized average control values percDNA and indicate the absence of differential expression for both isoformstested (p>0.05). Error bars correspond to standard deviation of threeindependent experiments.


Autho

r's

pers

onal

co

py

motor neurons in SALS motor cortex, because we showed that theexpression of motor neuron markers such as KIF5C, an isoformparticularly enriched or even unique in motor neurons (Kanai et al.,2000), remained relatively unaffected (Table 3).

The two down-regulated motors, KIAA591/KIF1Bβ andKIF3Aβ, are very highly expressed in the central nervous system,including all four cortex areas of the cerebral cortex (motor, visual,temporal, and somatosensory) and the cerebellum, while theirexpression in other adult tissues in general is very low incomparison. In the motor cortex (area 4) the cellular localization ofKIF3Aβ is completely overlapping with pyramidal and non-pyramidal cholinergic neurons and is marked in layers III and V,rich in pyramidal small and giant upper motor neurons. Thepresence of KIF3Aβ in cholinergic neurons, where it co-localizeswith ChAT, in human is consistent with the localization of itshomolog in cholinergic neurons in the central and peripheralnervous system of Drosophila, where it functions as a transporterof ChAT (Ray et al., 1999). We have not investigated whether co-localization in cholinergic presynaptic cells is a consistent featurethroughout the CNS, but given the absence of cholinergic cellbodies in the cerebellum where KIF3Aβ is present, it is very likelythat KIF3Aβ is not restricted to cholinergic systems.

The changes that were observed in SALS patients were notfound in the FALS SOD1 G93A mouse model, neither pre-symptomatically nor at the onset of symptoms or at end-stage, thestage which would be equivalent to the post-mortem SALS casesexamined in our study. This finding provides additional, indirectevidence that the observed quantitative changes in human SALS,specifically affecting molecular motors that are particularlyabundant in motor neurons of the motor cortex (i.e. KIF3Aβ),are not a mere consequence of the destruction of motor neuronsand their relative elimination from the motor cortex tissue sincethese conditions would apply also to day-126 transgenic SOD1G93A mice. In conclusion, therefore, the down-regulation ofmotors KIAA591/KIF1Bβ and KIF3Aβ, that we observe in SALSsamples, suggests that the qualitative characteristics of the axonaltransport impairment in SALS and of the particular FALS mousemodel differ. The lack of quantitative changes in KIF1Bβexpression in SOD1 G93A mice is consistent with a previousreport analyzing KIF1Bβ levels in the spinal cord of early-disease-stage mice (Conforti et al., 2003). Similarly, work on the SOD1G86R FALS strain reported KAP3 (a kinesin II-binding partner)up-regulation as an early event arising before axonal neurodegen-

eration, without change, however, of the other components of thekinesin II complex (Dupuis et al., 2000). This latter work alsoreported the up-regulation of the ubiquitously expressed KIF1A inmouse spinal cord, a finding that we could not confirm in the motorcortex of SALS patients.

Our finding of SOD1 down-regulation in SALS is intriguing as,at least in the specimens under study, it is not caused by genomicmutations in the protein ORF or splicing junctions as genomicsequencing has confirmed. While it is known that the reduction ofSOD1 activity or half-life does not correlate well with ALSpathogenesis or disease phenotypes (Al-Chalabi and Leigh, 1998),it cannot be excluded as a contributing or aggravating factor inneuronal degeneration.

Motor proteins KIF1Bβ and the novel KIF3Aβ, that we showto be highly expressed in the motor cortex, are particularlyinteresting targets as candidates involved in other neuro-degenerative diseases.

In human, KIF1B is expressed in two types of isoforms,KIF1Bα, originally reported as a mitochondria transporter(Nangaku et al., 1994) but also shown to interact with post-synaptic density and synaptic scaffold proteins (Mok et al.,2002), and at least three currently known KIF1Bβ-like forms,one of which, KIF1Bβ (NM015074), is a transporter of synapticvesicle precursors and implicated in CMT 2A motor neurondisease (Zhao et al., 2001). In the motor cortex we havedetermined the expression of KIF1Bα, KIF1Bβ (NM015074),and KIAA0591/KIF1Bβ (AB017133) (Table 1). We have shownthat KIAA0591/KIF1Bβ was specifically down-regulated inSALS patients. Although it is not clear whether the individualKIF1Bβ-like isoforms all share identical cellular cargoes, it islikely that they all serve as transporters of synaptic materialsbecause of the high sequence similarity between them.KIAA0591/KIF1Bβ (AB017133) is an alternatively spliced formof KIF1Bβ (NM015074), with an additional 40-aa sequencewithin its ORF and a different extreme carboxy-terminus. Thedown-regulation of KIF1Bβ motors (or loss of function in thecase of CMT 2A) and consequent impairment of the transport ofsynaptic material would debilitate synaptic function and henceaxonal integrity.

The kinesin II complex is a multi-task motor formed byinteraction of KIF3Awith either KIF3B or KIF3C. It is involved inintraflagellar transport, renal ciliogenesis, the establishment of theleft–right axis in the embryogenesis of mammals (Nonaka et al.,

Table 4Mutations detected by genomic sequencing

Summary of the sequencing results of the motor domain-encoding exons of KIF1B and KIF3A and of exons encoding the complete ORF of SOD1. Intronicmutations did not affect conserved splice or branch sites. Novel SNPs, deposited to the dbSNP125 of NCBI, are highlighted.


Autho

r's

pers

onal

co

py

1998), signal transduction (Jimbo et al., 2002; Huangfu et al.,2003), transport of late endosomes and lysosomes (Bananis et al.,2004; Brown et al., 2005), and even mitosis (Haraguchi et al.,2006). Mutations in KIF3A have been implicated in situs inversus(Marszalek et al., 1999) and polycystic kidney disease (Lin et al.,2003). Critically, kinesin II is also important for neuronaldevelopment and function. Although KIF3C appears to bedispensable for normal neural development and behavior in mouse(Yang et al., 2001), kinesin II is critical, both for neurite building asa carrier of fodrin-associating vesicles (Takeda et al., 2000) and forthe establishment of neuronal polarity via the targeting of PAR3 tothe distal tip of the axon (Nishimura et al., 2004; Shi et al., 2004).In addition, in neurons KIF3A is detected within and probablytransports tau mRNA-containing RNPs, locally translated in axons(Aronov et al., 2002). Importantly, reduction of kinesin II has beenlinked to neurodegeneration in mammals: kinesin II is a selectivetransporter of opsin and arrestin, but not α-transducin, inmammalian photoreceptors (Marszalek et al., 2000), and removalof KIF3A by photoreceptor-specific conditional mutagenesiscauses a rapid photoreceptor cell degeneration (Marszalek et al.,2000; Pazour et al., 2002). In our work, it is of interest to note thatall members of the kinesin II subfamily, KIF3A, KIF3Aβ, KIF3B,and KIF3C, showed down-regulation in the motor cortex of SALSpatients, although only the motor-neuron-enriched KIF3Aβ iso-form is down-regulated in a statistically significant manner. Theabsence of genomic mutations in the motor-domain-encodingexons and splicing junctions in KIF3A, sequenced by us, indicatesfor the SALS population under study that this down-regulation iscaused at the transcriptional level and possibly induced underconditions that are conducive to the culmination of the neurode-generative pathogenesis. To what extent this contributes to axonaldegeneration over time remains to be addressed. In mouse,heterozygous KIF3A animals did not show retinal degeneration(Jimeno et al., 2006), contrary to the dramatic effects observedwith the complete removal of functional KIF3A (see above).However, the two paradigms and cell systems may not be directlycomparable, and isoform-specific effects, such as the down-regulation of the motor neuronally enriched KIF3Aβ, may have asignificant impact on the multi-factorial interplay that underliesSALS pathology.

Indeed, the investigation of individual KLP isoforms, some ofthem as yet unidentified, may lead to a definite link between animpairment of axonal transport and motor neuron degeneration inALS. The present study indicates KIAA0591/KIF1Bβ (AB017133)and KIF3Aβ as promising candidates for further functional in vivoassays and highlights the importance of the isoform specificity ofanalyses in the definition of ALS pathogenesis.

Acknowledgments

This work was funded in parts by the American MuscularDystrophy Association (MDA USA), the RTN Network grant IHP-RTN-99-1 (EU), the Telethon Foundation (Cyprus) and theResearch Promotion Foundation of Cyprus grant YGEIA 0603/03.

We would like to thank Giampietro Schiavo at the MolecularNeuropathology Laboratory (Cancer Research, UK) for donatingrat spinal motor neuron culture cDNA and, at the Cyprus Instituteof Neurology and Genetics, Eleni Papanicolaou for help withproviding peripheral blood samples from SALS and referencepatients, Stavros Malas for providing transgenic mice for analysisand Kleopas Kleopa for helpful discussions.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.nbd.2007.02.005.

References

Al-Chalabi, A., Leigh, P.N., 1998. Familial amyotrophic lateral sclerosis andrelated motor neurone disorders. In: Emery, A.E.H. (Ed.), Neuromus-cular Disorders: Clinical and Molecular Genetics. J. Wiley and SonsLtd., New York, pp. 433–467.

Aronov, S., Aranda, G., Behar, L., Ginzburg, I., 2002. Visualization oftranslated tau protein in the axons of neuronal P19 cells andcharacterization of tau RNP granules. J. Cell Sci. 115, 3817–3827.

Bananis, E., Nath, S., Gordon, K., Satir, P., Stockert, R.J., Murray, J.W.,Wolkoff, A.W., 2004. Microtubule-dependent movement of lateendocytic vesicles in vitro: requirements for Dynein and Kinesin. Mol.Biol. Cell 15, 3688–3697.

Brown, C.L., Maier, K.C., Stauber, T., Ginkel, L.M., Wordeman, L., Vernos,I., Schroer, T.A., 2005. Kinesin-2 is a motor for late endosomes andlysosomes. Traffic 6, 1114–1124.

Bustin, S.A., 2002. Quantification of mRNA using real-time reversetranscription PCR (RT-PCR): trends and problems. J. Mol. Endocrinol.29, 23–39.

Conforti, L., Dell’Agnello, C., Calvaresi, N., Tortarolo, M., Giorgini, A.,Coleman, M.P., Bendotti, C., 2003. Kif1Bbeta isoform is enriched inmotor neurons but does not change in a mouse model of amyotrophiclateral sclerosis. J. Neurosci. Res. 71, 732–739.

Deng, H.X., Hentati, A., Tainer, J.A., Iqbal, Z., Cayabyab, A., Hung, W.Y.,Getzoff, E.D., Hu, P., Herzfeldt, B., Roos, R.P., et al., 1993.Amyotrophic lateral sclerosis and structural defects in Cu/Zn superoxidedismutase. Science 261, 1047–1051.

Dheda, K., Huggett, J.F., Chang, J.S., Kim, L.U., Bustin, S.A., Johnson,M.A., Rook, G.A., Zumla, A., 2005. The implications of using aninappropriate reference gene for real-time reverse transcription PCRdata normalization. Anal. Biochem. 344, 141–143.

Dupuis, L., de Tapia, M., Rene, F., Lutz-Bucher, B., Gordon, J.W., Mercken,L., Pradier, L., Loeffler, J.P., 2000. Differential screening of mutatedSOD1 transgenic mice reveals early up-regulation of a fast axonaltransport component in spinal cord motor neurons. Neurobiol. Dis. 7,274–285.

Estevez, A.G., Crow, J.P., Sampson, J.B., Reiter, C., Zhuang, Y.,Richardson, G.J., Tarpey, M.M., Barbeito, L., Beckman, J.S., 1999.Induction of nitric oxide-dependent apoptosis in motor neurons by zinc-deficient superoxide dismutase. Science 286, 2498–2500.

Ferrante, R.J., Browne, S.E., Shinobu, L.A., Bowling, A.C., Baik, M.J.,MacGarvey, U., Kowall, N.W., Brown Jr., R.H., Beal, M.F., 1997.Evidence of increased oxidative damage in both sporadic and familialamyotrophic lateral sclerosis. J. Neurochem. 69, 2064–2074.

Goldstein, L.S., 2003. Do disorders of movement cause movement disordersand dementia? Neuron 40, 415–425.

Gunawardena, S., Her, L.S., Brusch, R.G., Laymon, R.A., Niesman, I.R.,Gordesky-Gold, B., Sintasath, L., Bonini, N.M., Goldstein, L.S., 2003.Disruption of axonal transport by loss of huntingtin or expression ofpathogenic polyQ proteins in Drosophila. Neuron 40, 25–40.

Hafezparast, M., Klocke, R., Ruhrberg, C., Marquardt, A., et al., 2003.Mutations in dynein link motor neuron degeneration to defects inretrograde transport. Science 300, 808–812.

Haraguchi, K., Hayashi, T., Jimbo, T., Yamamoto, T., Akiyama, T., 2006.Role of the kinesin-2 family protein, KIF3, during mitosis. J. Biol.Chem. 281, 4094–4099.

Hirokawa, N., Noda, Y., Okada, Y., 1998. Kinesin and dynein superfamilyproteins in organelle transport and cell division. Curr. Opin. Cell Biol.10, 60–73.

Howland, D.S., Liu, J., She, Y., Goad, B., Maragakis, N.J., Kim, B.,Erickson, J., Kulik, J., DeVito, L., Psaltis, G., DeGennaro, L.J.,


Autho

r's

pers

onal

co

py

Cleveland, D.W., Rothstein, J.D., 2002. Focal loss of the glutamatetransporter EAAT2 in a transgenic rat model of SOD1 mutant-mediatedamyotrophic lateral sclerosis (ALS). Proc. Natl. Acad. Sci. U. S. A. 99,1604–1609.

Huangfu, D., Liu, A., Rakeman, A.S., Murcia, N.S., Niswander, L.,Anderson, K.V., 2003. Hedgehog signalling in the mouse requiresintraflagellar transport proteins. Nature 426, 83–87.

Hurd, D.D., Saxton, W.M., 1996. Kinesin mutations cause motor neurondisease phenotypes by disrupting fast axonal transport in Drosophila.Genetics 144, 1075–1085.

Ince, P.G., Evans, J., Knopp, M., Forster, G., Hamdalla, H.H.M., Wharton,S.B., et al., 2003. Corticospinal tract degeneration in the progressivemuscular atrophy variant of ALS. Neurology 60, 1252–1258.

Jimbo, T., Kawasaki, Y., Koyama, R., Sato, R., Takada, S., Haraguchi, K.,Akiyama, T., 2002. Identification of a link between the tumoursuppressor APC and the kinesin superfamily. Nat. Cell Biol. 4, 323–327.

Jimeno, D., Lillo, C., Roberts, E.A., Goldstein, L.S., Williams, D.S., 2006.Kinesin-2 and photoreceptor cell death: requirement of motor subunits.Exp. Eye Res. 82, 351–353.

Kanai, Y., Okada, Y., Tanaka, Y., Harada, A., Terada, S., Hirokawa, N.,2000. KIF5C, a novel neuronal kinesin enriched in motor neurons.J. Neurosci. 20, 6374–6384.

Kaufmann, P., Pullman, S.L., Shungu, D.C., Chan, S., Hays, A.P., Del Bene,M.L., et al., 2004. Objective tests for upper motor neuron involvement inamyotrophic lateral sclerosis (ALS). Neurology 62, 1753–1757.

Kawahara, Y., Ito, K., Sun, H., Aizawa, H., Kanazawa, I., Kwak, S., 2004.Glutamate receptors: RNA editing and death of motor neurons. Nature427, 801.

Kieran, D., Hafezparast, M., Bohnert, S., Dick, J.R., Martin, J., Schiavo, G.,Fisher, E.M., Greensmith, L., 2005. A mutation in dynein rescues axonaltransport defects and extends the life span of ALSmice. J. Cell Biol. 169,561–567.

Kirkinezos, I.G., Bacman, S.R., Hernandez, D., Oca-Cossio, J., Arias, L.J.,Perez-Pinzon, M.A., Bradley, W.G., Moraes, C.T., 2005. Cytochrome cassociation with the inner mitochondrial membrane is impaired in theCNS of G93A-SOD1 mice. J. Neurosci. 25, 164–172.

Kuner, R., Groom, A.J., Bresink, I., Kornau, H.C., Stefovska, V., Muller, G.,Hartmann, B., Tschauner, K., Waibel, S., Ludolph, A.C., Ikonomidou,C., Seeburg, P.H., Turski, L., 2005. Late-onset motoneuron diseasecaused by a functionally modified AMPA receptor subunit. Proc. Natl.Acad. Sci. U. S. A. 102, 5826–5831.

LaMonte, B.H., Wallace, K.E., Holloway, B.A., Shelly, S.S., Ascano, J.,Tokito, M., Van Winkle, T., Howland, D.S., Holzbaur, E.L.F., 2002.Disruption of dynein/dynactin inhibits axonal transport in motor neuronscausing late-onset progressive degeneration. Neuron 34, 715–727.

Levy, J.R., Holzbaur, E.L., 2006. Cytoplasmic dynein/dynactin function anddysfunction in motor neurons. Int. J. Dev. Neurosci. 24, 103–111.

Lin, C.L.G., Bristol, A., Jin, L., Dykes-Hoberg, M., Crawford, T., Clawson,L., Rothstein, J.D., 1998. Aberrant RNA processing in a neurodegen-erative disease: the cause for absent EAAT2, a glutamate transporter, inamyotrophic lateral sclerosis. Neuron 20, 589–602.

Lin, F., Hiesberg, T., Cordes, K., Sinclair, A.M., Goldstein, L.S.B., Somlo,S., Igarashi, P., 2003. Kidney-specific inactivation of the KIF3A subunitof kinesin-II inhibits renal ciliogenesis and produces polycystic kidneydisease. Proc. Natl. Acad. Sci. U. S. A. 100, 5286–5291.

Lupas, A., Van Dyke, M., Stock, J., 1991. Predicting coiled coils fromprotein sequences. Science 252, 1162–1164.

Manfredi, G., Xu, Z., 2005. Mitochondrial dysfunction and its role in motorneuron degeneration in ALS. Mitochondrion 5, 77–87.

Marszalek, J.R., Ruiz-Lozano, P., Roberts, E., Chien, K.R., Goldstein,L.S.B., 1999. Situs inversus and embryonic ciliary morphogenesisdefects in mouse mutants lacking the KIF3A subunit of kinesin II. Proc.Natl. Acad. Sci. U. S. A. 96, 5043–5048.

Marszalek, J.R., Liu, X., Roberts, E.A., Chui, D., Marth, J.D., Williams,D.S., Goldstein, L.S., 2000. Genetic evidence for selective transport ofopsin and arrestin by kinesin-II in mammalian photoreceptors. Cell102, 175–187.

Martin, L.J., 2000. p53 is abnormally elevated and active in the CNS ofpatients with amyotrophic lateral sclerosis. Neurobiol. Dis. 7, 613–622.

Meyer, T., Fromm, A., Munch, C., Schwalenstocker, B., Fray, A.E., Ince,P.G., Stamm, S., Gron, G., Ludolph, A.C., Shaw, P.J., 1999. TheRNA of the glutamate transporter EAAT2 is variably spliced inamyotrophic lateral sclerosis and normal individuals. J. Neurol. Sci.170, 45–50.

Miller, S.A., Dykes, D.D., Polesky, H.F., 1988. A simple salting outprocedure for extracting DNA from human nucleated cells. NucleicAcids Res. 16, 1215.

Mok, H., Shin, H., Kim, S., Lee, J.-R., Yoon, J., Kim, E., 2002. Associationof the kinesin superfamily motor protein KIF1B with postsynapticdensity-95 (PSD-95), synapse-associated protein-97, and synapticscaffolding molecule PSD-95/discs large/zona occludens-1 proteins.J. Neurosci. 22, 5253–5258.

Nangaku, M., Sato-Yoshitake, R., Okada, Y., Noda, Y., Takemura, R.,Yamazaki, H., Hirokawa, N., 1994. KIF1B, a novel microtubule plusend-directed monomeric motor protein for transport of mitochondria.Cell 79, 1209–1220.

Navolanic, P.M., Sperry, A.O., 2000. Identification of isoforms of a mitoticmotor in mammalian spermatogenesis. Biol. Reprod. 62, 1360–1369.

Navone, F., Niclas, J., Hom-Booher, N., Sparks, L., Bernstein, H.D.,McCaffrey, G., Vale, R.D., 1992. Cloning and expression of a humankinesin heavy chain gene: interaction of the COOH-terminal domainwith cytoplasmic microtubules in transfected CV-1 cells. J. Cell Biol.117, 1263–1275.

Niclas, J., Navone, F., Hom-Booher, N., Vale, R.D., 1994. Cloning andlocalization of a conventional kinesin motor expressed exclusively inneurons. Neuron 12, 1059–1072.

Nishimura, T., Kato, K., Yamaguchi, T., Fukata, Y., Ohno, S., Kaibuchi, K.,2004. Role of the PAR-3–KIF3 complex in the establishment ofneuronal polarity. Nat. Cell Biol. 6, 328–334.

Nonaka, S., Tanaka, Y., Okada, Y., Takeda, S., Harada, A., Kanai, Y., Kido,M., Hirokawa, N., 1998. Randomization of right–left asymmetry due toloss of nodal cilia generating leftward flow of extraembryonic fluid inmice lacking KIF3B motor protein. Cell 95, 829–837.

Pazour, G.J., Baker, S.A., Deane, J.A., Cole, D.G., Dickert, B.L.,Rosenbaum, J.L., Witman, G.B., Besharse, J.C., 2002. The intraflagellartransport protein, IFT88, is essential for vertebrate photoreceptorassembly and maintenance. J. Cell Biol. 157, 103–113.

Pfaffl, M.W., Horgan, G.W., Dempfle, L., 2002. Relative expressionsoftware tool (REST) for group-wise comparison and statistical analysisof relative expression results in real-time PCR. Nucleic Acids Res. 30,e36.

Pfaffl, M.W., Tichopad, A., Prgomet, C., Neuvians, T.P., 2004. Determina-tion of stable housekeeping genes, differentially regulated target genesand sample integrity: BestKeeper-Excel-based tool using pair-wisecorrelations. Biotechnol. Lett. 26, 509–515.

Puls, I., Jonnakuty, C., LaMonte, B.H., Holzbaur, E.L., Tokito, M., Mann,E., Floeter, M.K., Bidus, K., Drayna, D., Oh, S.J., Brown Jr., R.H.,Ludlow, C.L., Fischbeck, K.H., 2003. Mutant dynactin in motor neurondisease. Nat. Genet. 33, 455–456.

Purves, D., Augustine, G.J., Fitzpatrick, D., Katz, L.C., LaMantia, A.-S.,McNamara, J.O., et al., 2001. In: Sunderland, M.A. (Ed.), III. Movementand its central control. Neuroscience Sinauer Associates, Inc.

Ray, K., Perez, S.E., Yang, Z., Xu, J., Ritchings, B.W., Steller, H., Goldstein,L.S.B., 1999. Kinesin-II is required for axonal transport of cholineacetyltransferase in Drosophila. J. Cell Biol. 147, 507–517.

Reid, E., Kloos, M., Ashley-Koch, A., Hughes, L., Bevan, S., Svenson, K.I.,Graham Lennon, F., Gaskell, P.C., Dearlove, A., Pericak-Vance, M.A.,Rubinstein, D.C., Marchuk, D.A., 2002. A kinesin heavy chain (KIF5A)mutation in hereditary spastic paraplegia (SPG10). Am. J. Hum. Genet.71, 1189–1194.

Rosen, D.R., Siddique, T., Patterson, D., Figlewicz, D.A., Sapp, P., Hentati,A., Donaldson, D., Goto, J., O’Regan, J.P., Deng, H.X., et al., 1993.Mutations in Cu/Zn superoxide dismutase gene are associated withfamilial amyotrophic lateral sclerosis. Nature 362, 59–62.


Autho

r's

pers

onal

co

py

Rost, B., Liu, J., 2003. The PredictProtein server. Nucleic Acids Res. 31,3300–3304.

Santama, N., 2001. RT-PCR for the identification of developmentallyregulated novel members of the kinesin-like superfamily. Methods Mol.Biol. 164, 9–20.

Shi, S.H., Cheng, T., Jan, L.Y., Jan, Y.N., 2004. APC and GSK-3beta areinvolved in mPar3 targeting to the nascent axon and establishment ofneuronal polarity. Curr. Biol. 14, 2025–2032.

Stokin, G.B., Lillo, C., Falzone, T.L., Brusch, R.G., Rockenstein, E., Mount,S.L., Raman, R., Davies, P., Masliah, E., Williams, D.S., Goldstein, L.S.,2005. Axonopathy and transport deficits early in the pathogenesis ofAlzheimer’s disease. Science 307, 1282–1288.

Takeda, S., Yamazaki, H., Seog, D.H., Kanai, Y., Terada, S., Hirokawa, N.,2000. Kinesin superfamily protein 3 (KIF3) motor transports fodrin-associating vesicles important for neurite building. J. Cell Biol. 148,1255–1265.

Turner, M.R., Osei-Lah, A.D., Hammers, A., Al-Chalabi, A., Shaw, C.E.,Andersen, P.M., et al., 2005. Abnormal cortical excitability in sporadicbut not homozygous D90A SOD1 ALS. J. Neurol., Neurosurg.Psychiatry 76, 1279–1285.

Vandesompele, J., De Preter, K., Pattyn, F., Poppe, B., Van Roy, N.,

De Paepe, A., Speleman, F., 2002. Accurate normalization of real-timequantitative RT-PCR data by geometric averaging of multiple internalcontrol genes. Genome Biol. 3 (7) RESEARCH0034.

Wang, S., Poptani, H., Bilello, M., Wu, X., Woo, J.H., Elman, L.B., et al.,2006. Diffusion tensor imaging in amyotrophic lateral sclerosis:volumetric analysis of the corticospinal tract. Am. J. Neuroradiol. 27,1234–1238.

Williamson, T.L., Cleveland, D.W., 1999. Slowing of axonal transport is avery early event in the toxicity of ALS-linked SOD1 mutants to motorneurons. Nat. Neurosci. 2, 50–56.

Xia, C.-H., Roberts, E.A., Her, L.-S., Liu, X., Williams, D.S., Cleveland,C.W., Goldstein, L.S.B., 2003. Abnormal neurofilament transportcaused by targeted disruption of neuronal kinesin heavy chain KIF5A.J. Cell Biol. 161, 55–66.

Yang, Z., Roberts, E.A., Goldstein, L.S., 2001. Functional analysis of mousekinesin motor Kif3C. Mol. Cell. Biol. 21, 5306–5311.

Zhao, C., Takita, J., Tanaka, Y., Setou, M., Nakagawa, T., Takeda, S.,Yang, H.M., Terada, S., Nakata, T., Takei, Y., Saito, M., Tsuji, S.,Hayashi, Y., Hirokawa, N., 2001. Charcot–Marie–Tooth disease type2A caused by a mutation in a microtubule motor KIF1Bβ. Cell 105,587–597.


differential expression

Documents