new disease loci - biumedweb.md.biu.ac.il/research/tzipora-falik-zaccai/images/yifat_2012.pdf ·...

ORIGINAL ARTICLE

A founder mutation in Vps37A causes autosomalrecessive complex hereditary spastic paraparesis

Yifat Zivony-Elboum,1,2 Wendy Westbroek,3 Nehama Kfir,1 David Savitzki,4

Yishay Shoval,1 Assnat Bloom,4 Raya Rod,4 Morad Khayat,1 Bella Gross,5,6

Walid Samri,5 Hector Cohen,7 Vadim Sonkin,7 Tatiana Freidman,8 Dan Geiger,9

Aviva Fattal-Valevski,10 Yair Anikster,11 Aoife M Waters,12 Robert Kleta,12,13

Tzipora C Falik-Zaccai1,2,6

ABSTRACTBackground Members of two seemingly unrelatedkindreds of Arab Moslem origin presented withpronounced early onset spastic paraparesis of upper andlower limbs, mild intellectual disability, kyphosis, pectuscarinatum and hypertrichosis.Methods The authors performed neurological anddevelopmental examinations on the affected individuals.The authors conducted whole genome linkage andhaplotype analyses, followed by sequencing of candidategenes; RNA and protein expression studies; and finallyproof of principle investigations on knockdownmorpholino oligonucleotide injected zebrafish.Results The authors characterise a novel form ofautosomal recessive complex hereditary spasticparaparesis (CHSP). MRI studies of brain and spinal cordwere normal. Within a single significantly linked locus theauthors ultimately identified a homozygous missensemutation c.1146A>T (p.K382N) in the vacuolar proteinsorting 37A (Vps37A) gene, fully penetrant andsegregating with the disease in both families. Mobilitywas significantly reduced in Vps37A knockdownmorpholino oligonucleotide injected zebrafish, supportingthe causal relationship between mutations in this geneand the phenotype described in the patients of this study.Conclusions The authors provide evidence for theinvolvement of Vps37A, a member of the endosomalsorting complex required for transport (ESCRT) system, inupper motor neuron disease. The ESCRT system hasbeen shown to play a central role in intracellulartrafficking, in the maturation of multivesicular bodies andthe sorting of ubiquitinated membrane proteins intointernal luminal vesicles. Further investigation ofmechanisms by which dysfunction of this gene causesCHSP will contribute to the understanding of intracellulartrafficking of vesicles by the ESCRT machinery and itsrelevance to CHSP.

INTRODUCTIONFirst described in 1880,1 hereditary spastic para-plegia (HSP) comprises a heterogeneous group ofneurodegenerative disorders characterised byprogressive lower limb spasticity, retrogradedegeneration of the crossed cortico-spinal tractsand thinning of the posterior columns in the spinalcord.2 Complicated forms, also known as ‘complexhereditary spastic paraparesis (CHSP)’, are charac-

terised by the addition of such neurological featuresas spastic quadriparesis, seizures, dementia, amyo-trophy, extrapyramidal disturbance, cerebral orcerebellar atrophy, optic atrophy, and peripheralneuropathy, as well as by extra neurologicalmanifestations such as dysmorphism, albinism,retinitis pigmentosa, deafness, dementia, amyo-trophy and ichthyosis.2e4 CHSP forms are gener-ally inherited as autosomal recessive traits. InMediterranean countries, CHSP forms are morecommon, due to the increased frequency ofconsanguinity and its association with autosomalrecessive pathologies.4 5

Currently, more than 40 HSP loci and 21 causa-tive genes for pure and HSP forms have beenidentified.4 With some genes shown to causephenotypes associating with both forms, thehistorical classification of HSP types by pure andcomplex has become blurred.4 Alternatively, HSPtypes are classified by the biological function of theproteins encoded by the causative genes. Accord-ingly, a large group of HSP associated proteins havebeen found to be involved in membrane traffickingand protein sorting pathways, including microtu-bule-based transport.4 Some of these proteins areinvolved with the endosomal sorting complexrequired for transport (ESCRT) system, comprisingESCRT-0, -I, -II and -III. The ESCRT machinery isinvolved in the maturation of multivesicular bodies(MVB) and sorting of ubiquitinated membraneproteins into internal luminal vesicles.6 Spastin andspartin, which are mutated in two common formsof HSP (SPG4 and SPG20), have been previouslyshown to bind the ESCRT-III proteins CHMP1Band Ist1, respectively; and spartin plays a role incytokinesis.7 8 For a comprehensive discussion ofthe role of ESCRTs in neural functions in general,and in HSP in particular, see this recent review.9

We present two Arab Moslem consanguineouskindreds with multiple affected individualspresenting a unique phenotype of CHSP. Wedetermined the causative mutation in the genevacuolar protein sorting 37A (Vps37A), encoding fora subunit of the ESCRT-I complex.

METHODSThe ethics committee of Western Galilee Hospital,Nahariya, and the supreme Helsinki committee ofthe Israeli Ministry of Health approved the study.

< An additional table ispublished online only. To viewthis file please visit the journalonline (http://jmg.bmj.com/content/49/7.toc).

For numbered affiliations seeend of article.

Correspondence toDr Tzipora C Falik-Zaccai,Institute of Human Genetics,Western GalileeHospital-Naharia, Israel;[email protected]

Received 3 January 2012Revised 14 May 2012Accepted 17 May 2012Published Online First20 June 2012

462 J Med Genet 2012;49:462e472. doi:10.1136/jmedgenet-2012-100742

New disease loci

group.bmj.com on April 29, 2013 - Published by jmg.bmj.comDownloaded from

http://jmg.bmj.com/

http://group.bmj.com/

All study participants and parents of minors signed informedconsent.

Clinical examinationsComplete physical, neurological and developmental examina-tions were performed on available and consenting individualswith CHSP from the two kindreds. Family history, laboratorytest results, metabolic measurements, operations and medicalprocedures, including brain MRI, EEG, EMG, and muscle biopsywere accessed.

GenotypingTo localise the mutated gene we performed linkage analysis usinga genome-wide scan with 2000 microsatellite markers distributedthroughout the genome, at average intervals of 2 cM (DeCode,Iceland), as previously described.10 Additional markers onsuspected regions were selected from the UCSC Human (Homosapiens) Genome Browser Gateway and the ‘Linkage MappingSet v2.5 MD5 and MD10’ kits (Applied Biosystems, Carlsbad,California, USA). Data were analysed using the Superlink Onlinesoftware11 under an autosomal recessive mode of inheritance,with 99% penetrance and a disease allele frequency of 0.001,assuming uniform allele frequency distribution.

Candidate gene selection and mutation analysisInformation about the 24 genes and ORFs found in the genomicsegment tightly linked to the disease (supplementary table 1)was gathered from the public databases (Ensemble GenomeBrowser, UCSC Genome Browser, UCSC Human Gene Sorter,GeneCards and OMIM), and their relevance to HSP wasassessed. Relevant criteria included:< Genes already known to be involved in HSP or with

pathways involved in HSP< Genes expressed in the central or peripheral nervous system,

in neurons, glia and Schwann cells< Genes which have homology to known participants in HSP

pathology< Genes known to be involved in other neurodegenerative

diseases< Genes which have orthologues known to be involved in

neurodegenerative diseases in model animals< Genes involved in the development and differentiation of

neurons and dendritic cells, in particular paroxysmal axons ofthe motor and sensory systems.Coding regions and at least 50 bp of flanking intron regions of

eight genes, encompassing the top candidate genes in the linkedinterval, were sequenced and analysed using the ABI Prism 3100automated sequencer according to the manufacturers’ standardprotocol (Applied Biosystems) and nucleotide blast (NCBI).Known single nucleotide polymorphisms (SNPs) with no link tothe pathology were filtered.

Recently, two mutations in KIF1Awere reported to cause pureHSP, one of them in Palestinians.12 13 We have sequenced exon 8and exon 13 of this gene, as it is located on the tip of chromo-some 2 and might have generated a negative logarithm of odds(LOD) score due to the lack of markers next to it.

The mutation’s impact was assessed using Polyphen(Harvard), PANTHER (USC) and SIFT (JCVI) databases,PredictProtein package and the 1000 Genome Project and NCBI’sdbSNP databases were examined to make sure a mutation in thisposition was not reported before. A structure of VPS37A wasdeduced by homology to the saccharomyces cerevisiae ortho-logue of VPS37A (RCSB PDB 2p22_C), and the impact of theK>A mutation was analysed using Chimera software (UCSF).

The mutation’s impact on protein stability was analysed withMUpro (UCI) tool.

Disease mutation cosegregationThe presence of Vps37A c.1146A>T mutation was tested in allparticipating members of both kindreds, in 50 healthy controlsfrom the same village and in 214 ethnically matched healthycontrols (Arab Muslims, residents of the northern region of Israel),via restriction analysis with Hinf1. Genomic DNA was amplifiedby PCR, using primers flanking Vps37A exon 11: 59-ATTTTCTA-GATTTGCCACTG-39 forward primer and 59-TTCTA-CATTAGCAGCTAATG-39 reverse primer, followed by enzymaticdigestion with Hinf1 (New England Biolabs, Ipswich, Massachu-setts, USA). PCR products were separated on 8% acryl amid gel.The wild type allele presents two bands of 230 bp and 111 bp; andthe mutated, three bands of 230 bp, 72 bp and 39 bp.

RNA and cDNA analysesWe examined the nature of Vps37A expression in differenttissues from both healthy and affected individuals. Vps37AmRNA transcript (Human Multiple Tissue cDNA Panel, BDBiosciences, San Jose, California, USA) was studied in eightnormal tissue types: heart, brain, placenta, lung, liver, skeletalmuscle, kidney and pancreas.To compare the level of transcription of Vps37A in patients and

healthy controls, we purified total RNA from whole peripheralblood using the RiboPure-Blood Kit (Ambion, Life Technologies,Carlsbad, California, USA), and extracted total RNA from fibro-blasts using GenElute Mammalian Total RNA Miniprep Kit(Sigma-Aldrich, St. Louis, Missouri, USA). RNA samples werereverse transcribed to cDNA and PCR amplified to detect Vps37Aand b-actin. PCR conditions for Vps37A and b-actin were 30 s at948C, 30 s at 558C and 10 s at 728C for 34 cycles. Primer sequencesfor Vps37A were: 59-CACTTTCGAAAGGAAGATGCAA-39 and 59-CCCTCCAAGAAATCTTCTGC-39. Primer sequences for b-actinwere: 59-GGCATCGTGATGGACTCCG-39 and 59-GCTGGAA-GGTGGACAGCGA-39.

Western blottingWe compared the level of Vps37A protein expression in bloodlymphocytes and skin fibroblasts from affected individuals andhealthy control fibroblasts (Cascade Biologics, Portland, Oregon,USA). Actin was used as a protein loading control.

Coimmunoprecipitation assaySince Vps37A interacts with other ESCRT-I proteins, such asVps28 and Tsg101/vps23,6 we investigated whether Vps37A(K382N) is capable of interacting with members of the ESCRT-Iprotein complex. Protein extracts were made from control(Cascade Biologics) and mutant fibroblasts. Wild type andpatient fibroblasts were washed with PBS and lysed with 1%Nonidet P40, 50 mM TRIS pH 8, 150 mM NaCl, and a proteaseinhibitor mix (Roche Diagnostics, Indianapolis, Indiana, USA).Rabbit polyclonal Vps37A (Proteintech Group Inc, Chicago,Illinois, USA) was coupled to M-280 magnetic dynabeads(Invitrogen, Life Technologies, Carlsbad, California, USA) anda coimmunoprecipitation assay followed by western blottingwas performed. Equivalent amounts of total protein cellextracts, as determined by BCA assay (Pierce Biotechnology,Rockford, Illinois, USA), were mixed with the Vps37A coupleddynabeads, or dynabeads only (negative control), and incubatedovernight at 48C in lysis buffer. The beads were boiled inLaemmli buffer (Biorad, Hercules, California, USA) and eluateswere loaded onto a 4%e12% tris-glycine gel (Invitrogen, Life

J Med Genet 2012;49:462e472. doi:10.1136/jmedgenet-2012-100742 463

New disease loci


http://jmg.bmj.com/


Figure 1 Pedigrees of kindreds 1 (A) and 2 (B) showing haplotypes across the complex hereditary spastic paraparesis linked locus. Ten microsatellitemarkers spanning 6.8 cM on chromosome 8 were used for haplotype reconstruction. Roman capitals indicate generations. The code numbers of allfamily members appear below symbols. Filled symbols indicate affected members. Black dots indicate obligate carriers. Circles, female subjects;squares, male subjects; slant, deceased. The haplotype assumed to carry the disease allele is framed in square.


New disease loci


http://jmg.bmj.com/


Table1

Clinicalphenotypes

ofpatientswith

CHSP

Family/

patient

Sex

AO

Muscletone

and

tendon

refle

xes

lower

extrem

ities

Muscletone

and

tendon

refle

xes

upperextrem

ities

Deepsensation

vibrationandposition

GMFC

S(level)

Cognitive

and

language

developm

entaldelay

MRI/CTof

brain

Kyphosis/pectus

carinatum

Hypertrichosis

Smalljoint

hyper-fle

xibility

Miscellaneous

1/VI 7

F15

m+++

Clonus

Normalmuscletone,

increasedtendon

reflexes

Normal

IIIUseswalker*

Mild

NormalMRIof

brain

andspineat

3Y

++/�

�+

Normalhearing

andvision

1/VI 14

F36

m+++

Clonus

Increasedmuscle

tone

andtendon

reflexes

Diminishedvibration

sensation

IIIUseswalker*

Mild,speech

delay

NormalCTat

3Y,

norm

alMRIat

9Y,

mild

ventriculom

egaly

+/�

++

+Normalhearing

andvision

1/VI 15

F15

m+++

Clonus

Normalmuscletone,

increasedtendon

reflexes

Normal

IIIUseswalker

Normalintelligence,

mild

speech

difficulties

NormalMRIat

7Y

+/+

�+

Normalhearing

andvision

1/VI 16

M10

m+++

Spontaneous

clonus

Increasedmuscletone

andtendon

reflexes

Loss

ofvibration

sensation

IVUseswheel

chairandwalker*

Mild,speech

delay

At18

mmild

white

matterchangesd

left,

non-specific

+++/+

�++

Micro-m

eatus,

norm

alhearing

andvision

1/VI 17

M12

m++++

Severeclonus

Increasedmuscletone

andtendon

reflexes

Abnormalposition

sensation

IVUseswalker

Moderate

NA

++/+

++

Normalhearing

andvision

1/VII 1

F12

m++

Increasedmuscletone,

increasedtendon

reflexesd

severe

Normal

IIIUseswalker*

Severe,

nospeech

Suspected

cortical

dysplasia,

insularleft,

otherw

isenorm

alMRI

at4Y

+++/+

�+

Severehearing

impairm

ent,far

sightedneeds

glasses

1/VII 3

M12

m+++

Normalmuscletone,

increasedtendon

reflexesd

mild

Normal

IIIUseswheel

chairandwalker*

Mild,delayedspeech

NA

++/+

++

Severehearing

impairm

ent

2/II 8

F7m

+++

Clonus,

dystonia

Increasedmuscletone

andtendon

reflexes,

dystonia

Normal

IVUseswheel

chairandwalker

Mild,delayedspeech

NormalCTat

3Y

+++/+

�+

Farsighted

needsglasses

2/III12

F8m

+++

Dystonia

Increasedmuscletone

andtendon

reflexes,

dystonia

Normal

VModerate;

delayedspeech

Normalbrainand

spineMRIat

6Y

++/+

NA

NA

FTTmicrocephaly

norm

alhearing

andvision

*Botulinum

injections.

AO,ageat

onset;CHSP,

complex

hereditary

spastic

paraparesis;

FTT,

failure

tothrive;

GMFCS,Gross

Motor

FunctionClassificationSystem;m,months;

NA,notavailable;

Y,years.

+and

�indicatedclinicalmanifestation.

�¼

none;+

¼weak;

++

¼moderate;

+++

¼severe;NA¼

notavailable.


New disease loci


http://jmg.bmj.com/


Technologies, Carlsbad, California, USA). Equivalent amounts oftotal protein, as determined by BCA assay (Pierce Biotech-nology), from fibroblasts, lymphocytes or IP assays were loadedonto 4%e12% tris-glycine gels. After blotting, the nitrocellulosemembranes (Invitrogen, Life Technologies, Carlsbad, California,USA) were probed with primary antibodies against Vps37A,Tsg101 and Vps28. HRP-labelled antimouse antibody or anti-rabbit antibody (Amersham Biosciences, Piscataway, New Jersey,USA) was used as a secondary antibody. The antigeneantibodycomplexes were detected with an ECL kit (Amersham Biosci-ences). Antibodies used for western blotting were: rabbit poly-clonal anti-Vps37 (Proteintech Group Inc Novus Biologicals,Littleton, Colorado, USA), Tsg101 and Vps28 (Santa CruzBiotechnology Inc, Santa Cruz, California, USA).

Zebrafish and morpholino injectionTo investigate the in vivo function of Vps37A in vertebrates andto further establish the causative relationship between Vps37Aand the CHSP phenotype, we created a vps37a knockdownmorpholino oligonucleotide (MO) zebra fish. Wild type zebrafish(AB x Tup LF) were staged and housed as previously described.14

Groups of 25e30 stage-matched embryos were collected at 96 hpostfertilisation (hpf). For vps37a knockdown, antisense MO(GeneTools, Philomath, Oregon, USA) were designed against thevps37a gene (Accession number: NM 001145152), 25 base pairsupstream of transcript start codon (vps37a 59ATG: 59-GTCTGTATGAGTTACTGCCGGACAT-39) and the intron 1-exon 2 splice site (vps37a Spl: 59-CTCTGGTTTTCACTGCA-CAAGAAAA-39). Embryos were injected with MO (1 ng/embryo) at the 1- to 2-cell stage and allowed to develop at28.58C to desired stages. To evaluate functional consequences ofVps37A knockdown, and to detect reduced larval mobility, weapplied a touch response test at 96 hpf as published previously.15

For rescue experiments, full length human VSP37A was ampli-fied by PCR from a human cDNA clone (IMAGE clone: 5275060,Accession: BI550733.1, Open Biosystems) and cloned into pCS2+ (Invitrogen, Life Technologies, Carlsbad, California, USA).

Mutagenesis was undertaken using the QuickChange sitedirected mutagenesis kit (Stratagene, Agilent Technologies,Santa Clara, California, USA) according to the manufacturer ’sinstructions. The missense mutation was introduced forVSP37AK382N using the primer sets (Forward: 59-gagccaaggaa-gagaatcttcagcaggcgatag-39; Reverse: 59-ctatcgcctgctgaa-gattctcttccttggctc-39). VSP37A plasmids were linearised withNot1 and mRNA synthesised using Ambion mMessagemMachine SP6 kit. Either wild type or mutagenised mRNA(75 pg) were injected into the cytosol of one cell stage embryoswith vsp37a MO.

RESULTSFamily and patientsMedical history and physical examinations of seven individualsfrom one kindred (figure 1A) and two from a second (figure 1B)revealed common features (table 1), including normal pregnancyand delivery, developmental and motor delay from the first orsecond year of life, followed by unsteadiness in standing, anddifficulties in walking. All affected children presented withspasticity in the lower limbs that progressed to the upperextremities, requiring recurrent physiotherapy and ligamentlengthening operations. Several patients were treated withbotulinum injections with moderate improvement, regainingthe ability to walk with the aid of a walker. All patientspresented mild to moderate delays in cognition and speech.Marked kyphosis was noted in all patients. However, patients II8and III12, from kindred 2 (figure 1B), presented a more severephenotype (table 1). When assessed, laboratory test results,metabolic measurements, EEG, EMG, muscle biopsy andoxidative phosphorylation studies were normal (data notshown). Patients were not dysmorphic. Brain and spinal cordMRIs were essentially normal with non-specific findings in somepatients. Seven additional members (IV17, V9-14, figure 1A) of thefirst kindred and one additional of the second (II9, figure 1B)were suspected of being affected with an identical phenotype,but were not available for examination or genetic analysis.

Table 2 Fine mapping of kindred I, chromosome 8 linked region

Marker information

Position Recombination fraction

Genetic* (cM) Physicaly (Mb) 0 0.01 0.05 0.1 0.2 0.3 0.4

D8S550 21.33 10 918 926 �3.1 0.7 2.0 2.2 1.9 1.1 0.5

D8S552 26.43 12 752 448 4.7 4.8 4.8 4.3 3.2 1.9 0.8

D8S1754 27.4 12 999 960 3.1 3.3 3.3 3.0 2.1 1.2 0.5

D8S1790 27.4 13 076 501 3.3 3.4 3.3 3.0 2.0 1.1 0.4

D8S1827 30.49 14 828 720 1.2 4.0 4.1 3.8 2.7 1.6 0.6

D8S549 31.73 15 659 943 �2.6 3.3 3.6 3.3 2.4 1.5 0.7

D8S484 16 075 550 1.7 4.4 4.5 4.1 2.9 1.7 0.7

D8S254 34.76 16 618 430 �0.2 2.6 2.9 2.7 2.0 1.3 0.6

D8S261 37.04 17 836 463 6.7 6.5 5.9 5.1 3.6 2.1 0.9

D8S258 41.55 20 377 440 1.3 1.3 1.1 0.9 0.5 0.3 0.1

D8S280 41.55 20 437 063 2.9 2.8 2.5 2.7 1.4 0.8 0.3

D8S1053 20 819 082 1.8 1.8 1.6 1.4 0.9 0.5 0.2

D8S474 21 291 741 1.9 1.9 1.7 1.5 1.0 0.6 0.3

D8S1116 42.85 21 440 568 1.9 1.8 1.6 1.4 0.9 0.5 0.3

D8S282 42.85 21 459 090 3.2 3.6 3.6 3.1 2.1 1.1 0.3

D8S405 21 990 470 3.9 4.2 4.2 3.7 2.6 1.4 0.5

D8S1733 45.41 22 542 515 2.0 2.5 2.7 2.5 1.8 1.1 0.4

D8S1752 46.26 22 690 059 3.8 4.2 4.3 4.0 2.9 1.7 0.6

D8S1734 22 817 140 0.5 1.1 1.7 1.7 1.3 0.8 0.3

D8S131 27 395 748 0.6 0.6 0.5 0.4 0.2 0.1 0.0

Logarithm of odds scores for markers located across the region on chromosome 8p linked to complex hereditary spastic paraparesis in kindred 1. The area flanked by markers D8S550 andD8S131 produced a peak score of 6.7 at marker D8S261 (italic) at q¼0. Boldface numbers indicate scores $3.*Sex average marker locations according to Marshfield genetics map. cM, centiMorgan.yPhysical marker locations obtained from UCSC Human Genome Browser Gateway. Mb, megabase.


New disease loci


http://jmg.bmj.com/


Figure 2 (A) Genomic DNA sequence analysis of Vps37A reveals a homozygous A>T transition at position c.1146 (indicated by arrow) in exon 11,resulting in a lysine to asparagine substitution at position 382 of the Vps37A protein (p.K382N). The samples: wild type: unrelated control; healthyheterozygote: kindred 2, II10; affected homozygote: kindred 2, III12. (B) ConSeq results for PFAM multiple alignment of mod(r) domain demonstrate thatwithin this conserved domain, the mutated lysine (marked by red square) is highly conserved, and predicted to be solvent exposed and to havea functional role. (C) Evolutionary conservation of the Vps37A C-terminal part. NCBI-blastp protein sequence alignment of the Vps37A C-terminalencompassing the mutation site reveals high conservation of lysine at position 382 (in red) among various species.


New disease loci


http://jmg.bmj.com/


Mode of inheritanceThe healthy consanguineous parents of all affectedindividuals in the first kindred are descended from a singleancestor, indicating autosomal recessive inheritance andsuggesting a founder mutation. The pedigree of the secondkindred is also consistent with autosomal recessive inheritance.

Whole genome screen and linkage analysisGenome-wide screening of nine members of kindred 1 (figure1A), followed by fine mapping using 30 additional DNA samplesfrom members of this kindred, supported sole linkage to a 12 Mb(25 cM) region spanning markers D8S550 to D8S1734, witha maximum LOD score of 6.7, on marker D8S261 (table 2).Linkage to all other chromosomal regions was excluded bynegative LOD scores.

Haplotype reconstructionHaplotype reconstruction for kindred 1 revealed a uniformhomozygote region shared by all affected members, and cose-gregating with the disease, flanked by markers D8S261 andD8S405. Detection of the identical homozygous region betweenmarkers D8S280 and D8S405 in healthy individual VI-18narrowed the CHSP locus to the region between markersD8S254 and D8S280, an interval of 3.8Mb (6.8 cM, positionaccording to the March 2006 release of the human genomeassembly, hg18) (figure 1A).Haplotype reconstruction of kindred 2 using a slightly

different repertoire of markers within the same linked region,dictated by informatively of DNA markers, revealed a commonhaplotype cosegregating with the disease in affected individualII8 (figure 1B). However, affected individual III12 inherited thedisease linked allele from her father, and a slightly different allelefrom her mother (figure 1B), possibly due to expansion of therepeat number in marker D8S261 that occurred in one of thefounders of this family.

Candidate gene selection and mutation analysisThe common genomic segment encompasses 24 genes andpredicted transcripts (supplementary table 1). The relevance ofthese genes to HSP was assessed according to the criteria detailedin the Methods section.Out of the 24, eight genes were defined as good candidates to

have the disease-causing mutation: FGF20, NAT1, PSD3, LPL,SLC18A1, ATP6V1B2, LZTS1 and VPS37A. All eight candidategenes were sequenced, and 16 polymorphisms were found. Outof these, 15 were previously reported SNPs with no link to thepathology. VPS37A c.1146A>T was the only novel sequencevariant found.We identified a novel A>T point substitution in Vps37A

(OMIM 609927), in the coding region of exon 11, renderinga lysine to aspargine inframe substitution at position 382-p.K382N (figure 2A). Cross-species alignment of the amino acidsequences for Vps37A showed lysine in position 382 to be highlyconserved throughout evolution (figure 2B,C).No SNP in this location was previously reported in the 1000

genomes project or NCBI’s dbSNP.Analysis of the impact of the mutation on protein function

was predicted to be deleterious using Polyphen (‘ProbablyDamaging’ with a score of 1.000) and Panther (‘Deleterious’with 88.4% probability), while SIFT tool predicted the mutationto be ‘Tolerated’ with a score of 0.22. PredictProtein analysis ofthe normal and mutant proteins gave similar results with noindication of change.

Figure 3 (A) Expression of Vps37A in multiple tissue samples. Humanadult multiple tissue cDNA panel was amplified with the sequences atcoding regions of Vps37A and G3PDH as primers. Vps37A was expressedin all human adult tissues tested. It was expressed at a high level inpancreas, placenta and heart, and at a lower level in skeletal muscle andbrain (semiquantitative). (B) Vps37A expression in healthy and affectedlymphocytes and fibroblasts. cDNA from fibroblasts and lymphocytes wasextracted using RT-PCR, and amplified by multiplex PCR using Vps37Aand b-actin primers. Repeated experiments demonstrated similar Vps37Aexpression in both tissues. (C) Endogenous expression of Vps37A incontrol and complex hereditary spastic paraparesis (CHSP) cells. Westernblots of lysates from fibroblasts and lymphocytes. Western blot of proteinextracts from control (lanes 2 and 4) and CHSP cells (lanes 1 and 3)stained with anti-Vps37A and anti-actin (protein loading control). Bothcontrol and CHSP cells expressed comparable amounts of Vps37A protein.(D) Interaction of Vps37A with members of the ESCRT-I complex. Co-IPwith an anti-Vps37A antibody on protein extracts from control (lane 1) andCHSP (lane 2) fibroblasts. Western blot was probed for Vps37A, Vps28and Tsg101. Both Vps37A (wild type) and Vps37A (K382N) showedinteraction with Vps28 and Tsg101 (lanes 1 and 2). The negative control,beads only co-IP on protein lysates of wild type (lane 3) and mutant(lane 4) fibroblasts did not show binding with Vps37A, Vps28 or Tsg101.


New disease loci


http://jmg.bmj.com/


Lysine 382 is located at the C’ terminus of the protein, in anunstructured region. Analysis of the tertiary structure predictedthat no structural change would be caused by the substitution.Prediction of the impact of the mutation on the protein stability,using the MUpro tool, was inconclusive, with some of themethods used showing increased stabilisation while othersshowing it to decrease.

Previously published causative mutations for HSP in Pales-tinians, in exon 8 and 13 of the KIF1A gene,12 13 were examinedand ruled out.

Disease mutation cosegregationUsing direct sequencing and restriction analysis we showedcomplete segregation and thus full penetrance of the mutationVps37A c.1146A>T with the disease phenotype; all affectedindividuals were found homozygous to the change, obligatecarriers and some siblings of the affected were heterozygous, andno healthy members of kindreds 1 and 2 were homozygotes. Theabsence of the identified point substitution in 428 chromosomesof healthy ethnically matched control samples refuted thepossibility of a unique ethnic polymorphism. Interestingly,random analysis of DNA samples from 50 healthy individuals,residents of the same village, but not related to our patients,

revealed three carriers for the point substitution, none of themhomozygous, pointing indeed towards a founder mutation inthis genetic isolate.

Vps37A mRNA expressionVps37A mRNA transcript (Human Multiple Tissue cDNA Panel,BD Biosciences) was observed in the eight normal tissue typesexamined: heart, brain, placenta, lung, liver, skeletal muscle,kidney, and pancreas, with lower levels in the brain and skeletalmuscle (figure 3A). Expression levels of Vps37A mRNA derivedfrom lymphocytes and fibroblasts of affected individuals werefound equivalent to those of healthy individuals (figure 3B).

Western blotting and coimmunoprecipitation assayEqual amounts of protein lysates, as determined by BCA assayfrom both affected and control fibroblasts and lymphocytes,showed equivalent amounts of Vps37A protein on western blotprobed with a Vps37A-specific antibody (Santa Cruz Biotech-nology) (figure 3C), indicating that the novel missense mutationdoes not cause destabilisation of the Vps37A protein. We foundthat both Vps28 and Tsg101 coimmunoprecipitated withVps37A (wild type) and Vps37A (K382N) in equal amounts(figure 3D, lanes 1 and 2). The beads-only co-IP, which served as

Figure 4 Zebrafish model for Vps37Acomplex hereditary spastic paraparesis;zebrafish embryos were injected at theone-cell stage with either 1 ng ofstandard morpholino (stdMO) (A) or1 ng of splice site MO (vps37aMO) (B)and showed no obvious defects ingross morphology analysed at 96 hpostfertilisation (hpf). (C) Embryosinjected with control MO (n¼23),Vps37A MO (n¼23), vps37a MO plusVPS37A RNA (n¼26) or vps37a MOplus VPS37A RNA with mutation K382N(n¼25) were subjected to a touchresponse test to determine the degreeof larval mobility. At 96 hpf, the meandistance per embryo was significantlyreduced in Vps37A MO-injected andembryos. Graph showing the meandistance per embryo (mm) in the touchresponse test: controls (StdMO)(4866768045, n¼23), vps37aMO(488961453, n¼23), vps37aMO+VPS37A RNA (5427868094, n¼26),vps37aMO+VPS37A K382N RNA(783361424, n¼25). Values shown asmean 6 SEM. Data were analysedusing the ManneWhitney U test.*p<0.0001, **p<0.0001,***p<0.0001, ****p¼0.0374 (NS).


New disease loci


http://jmg.bmj.com/


a negative control, did not show a specific immunoprecipitationwith the Vps37A, Vps28 or Tsg101 protein (figure 3D, lanes 3and 4).

Zebrafish and morpholino injectionZebrafish embryos injected with 1 ng of vps37a MO showed noobvious dysmorphology at 96 hpf (figure 4A,B). The Vps37AMO-injected zebrafish at 96 hpf showed striking and significantloss of motility in comparison with standard MO-injectedcontrols (figure 4C), providing evidence for Vps37A involvementin CHSP. Rescue with wild type but not with mutated VPS37Aproved specificity of this morpholino approach (figure 4C).

DISCUSSIONA mutation in Vps37A causes a novel form of autosomalrecessive CHSPWe characterised a novel form of autosomal recessive CHSP intwo Arab Moslem kindreds. We showed a homozygous missensemutation in Vps37A, c.1146A>T (pK382N), to be the cause ofthis pathology. The causative relation between this mutationand autosomal recessive CHSP is substantiated by several find-ings: The disease and the region on chromosome 8p22 are linkedwith statistical significance and a common haplotype isinherited with full segregation. Mutation c.1146A>Tof Vps37A,which is in full segregation with the disease in both families, isnot a known SNP according to 1000 Genomes Project anddbSNP, and was not detected in any of 428 control chromosomesfrom the same ethnic origin. The haplotype fully correspondswith the mutation in affected individuals and is absent inhealthy family members. The two kindreds reside in the samesmall genetically isolated village and are highly consanguineous.The amino acid K382 is located in the mod(r) motif in Vps37Aand is highly if not completely conserved among species.Further, our demonstration of significantly reduced mobility inVps37A MO-injected zebrafish compared with standard MO-injected controls with specific rescue experiments supportsVps37A involvement in CHSP. A number of unique mutationshave been identified in similar isolated populations residing invillages in the same area.16e19

Vps37A and the ESCRT-I complexVps37A encodes a subunit of the ESCRT-I complex. The lysine inposition 382 is highly conserved throughout evolution,appearing in all examined organisms. The impact of the muta-tion on protein function was predicted to be deleterious usingPolyphen and Panther, while SIFT tool predicted the mutation tobe ‘Tolerated’. PredictProtein analysis of the normal and mutantproteins gave similar results with no indication of change.Analysis of the tertiary structure of VPS37A predicted that nostructural change would be caused by the substitution, and weshow that mutant Vps37A protein expression does not desta-bilise the ESCRT-I complex. The observation that Vps37A(p.K382N) is still able to interact with members of the ESCRT-Icomplex Vps28 and Tsg101 suggests that the conserved K382 isnot involved in ESCRT-I assembly, but rather affects normalmuscle tone and strength via a different biological pathway.Thus, this position may be related to interactions with otherproteins or other forms of regulation of the protein which arecurrently unknown.

Bache and coworkers demonstrated that in mammalian cellsthe ESCRT-I complex, namely Vps37, Tsg101, Vps28 andMVB12, is required for downregulation of the epidermal growthfactor receptor via degradation.20 While the tissue and cell-

specific role of the ESCRT machinery is not well characterised inhumans, ESCRT-I is known to play a role in HIV budding andinfection,21 as well as in melanosome biogenesis in epidermalmelanocytes.22 Bache et al20 identified hVps37A as the humancounterpart of Vps37 in yeast, and the equivalent of the proteinhepatocellular carcinoma related protein 1. Severe consequencesof downregulation of expression of other ESCRT-I members orfailure of formation of the ESCRT-I have been reported.Homozygous Tsg101 -/-mouse embryos have been shown to failto develop past day 6.5 of embryogenesis;23 interference ofTsg101 expression has been shown to lead to neoplastic trans-formation.24 Loss of TSG101 function in endo-lysosomal traf-ficking plays a significant role in the pathogenesis of spongiformneurodegeneration in Mgrn1 null mutant mice.25

Expression and function of Vps37AOur observation of Vps37A expression in cDNA of eight normalhuman tissues, including the brain, supports previous evidencethat Vps37A is expressed in most human tissues, with thehighest expression in the liver.26

While the functional role of K382 is still unknown, K382 inthe mod(r) domain is thought to be involved in regulation ofprotein expression through its binding to ubiquitin. This bindingsignals protein degradation, both by the proteasome and lyso-some.27 Interaction between Vps37A and ubiquitin remainsspeculative despite the similarity between the C-terminal ofVps37A and the ubiquitin E2 variant, located in the proteinTsg101.20 27 Another possibility is that K382N impairs lysineacetylation. Lysine acetylation is not limited to modification ofhistones but occurs in other proteins as well, and participates incellular regulatory processes such as silencing and proteinstability.28

Web resources

< Ensemble Genome Browser (http://www.Ensemble.org/index.html)

< GeneCard (http://www.genecards.org/index.shtml)< Marshfield genetic Map (http://research.marshfieldclinic.org/

genetics/home/index.asp)< NCBI-MapView (http://research.marshfieldclinic.org/genetics/

home/index.asp)< PedTool (http://bioinfo.cs.technion.ac.il/pedtool/)< UCSC Genome Browser (http://genome.ucsc.edu/cgi-bin/

hgGateway)< NCBI BLAST (http://www.ncbi.nlm.nih.gov/BLAST/)< UniProt (Universal Protein) Resource (http://www.expasy.org/

sprot)< ConSeq website (http://conseq.tau.ac.il)< Superlink online (http://bioinfo.cs.technion.ac.il/superlink-

online)< 1000 Genome project (http://www.1000genomes.org/)< dbSNP (http://www.ncbi.nlm.nih.gov/projects/SNP/)< RCSB PDB (http://www.rcsb.org/pdb/home/home.do)< UCSC Chimera (http://www.cgl.ucsf.edu/chimera)< Polyphen tool (http://genetics.bwh.harvard.edu/pph2/index.

shtml)< SIFT tool (http://sift.jcvi.org/)< PANTHER tool (http://www.pantherdb.org/)< PredictProtein package (www.predictprotein.org/)< MUpro tool (http://www.ics.uci.edu/wbaldig/mutation.html)


New disease loci


http://jmg.bmj.com/


It is possible that Vps37A-p.K382N causes CHSP by impairinga mechanism that is not related to endosomal protein sorting.ESCRT subunits are known to be involved in functions notrelated to membrane trafficking in a wide range of diseasepathologies.29 30

Defects in membrane trafficking and neurodegenerative diseaseThe current study shows for the first time the involvement ofthe ESCRT-I member, Vps37A, in CHSP. Spastin, atlastin,REEP1 and spartin are other proteins involved in ESCRTcomplexes that are encoded by HSP causative genes.31e39 Otherdefects in membrane trafficking have also been associated withHSP. Infantile-onset ascending hereditary spastic paralysis16 andALS219 are two recently documented examples. Furthermore, wenote that defects in ESCRT compounds have been implicatedwith neurodegenerative diseases other than HSP, such as fron-totemporal dementia and amyotrophic lateral sclerosis.29 40 Thecurrent findings enable accurate genetic counselling to patientsand their families, carrier testing and early prenatal diagnosis ina large consanguineous population in northern Israel, andpossibly in other CHSP families worldwide. The molecularmechanisms underlying the phenotype are not yet known,but are possibly related to vesicular trafficking or abnormalubiquitination.

Author affiliations1Institute of Human Genetics, Western Galilee Hospital-Nahariya, Nahariya, Israel2The Rappaport Faculty of Medicine and Research Institute, Technion - Israel Instituteof Technology, Haifa, Israel3Section on Human Biochemical Genetics, Medical Genetics Branch, National HumanGenome Research Institute, National Institutes of Health, Bethesda, Maryland, USA4Department of Child Development, Western Galilee Hospital-Nahariya, Nahariya,Israel5Department of Neurology, Western Galilee Hospital-Nahariya, Nahariya, Israel6The Galilee Faculty of Medicine, Bar Ilan University, Safed, Israel7Department of Pathology, Western Galilee Hospital-Nahariya, Nahariya, Israel8Sherutei Briut Clalit, Haifa and Western Galilee District, Israel9Department of Computer Sciences, Technion - Israel Institute of Technology, Haifa,Israel10Pediatric Neurology Unit, Dana Children’s Hospital, Tel Aviv Souraski Medical Center,Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel11Metabolic Unit, Edmond and Lily Safra Children’s Hospital, Sheba Medical Center,Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel12Nephro-Urology Unit, Great Ormond Street Hospital, London, UK13Division of Medicine, University College London, London, UK

Acknowledgements We thank the patients and their parents for participating inthis study, and the nurses from the prenatal and postnatal clinics of the village fortheir assistance with its realisation. We thank Julia Fekecs (NHGRI, NIH, Bethesda,Maryland, USA) for assisting with the preparation of the figures. The authors aremost thankful to Dr Sara Amit, former head of the neuro-developmental departmentin the Western Galilee Hospital, who first took care of the patients described in thisreport, initiated and supported this project, but unfortunately did not live to see itssuccessful completion.

Contributors YZE as part of her PhD thesis performed DNA linkage analyses andsequencing expression studies. WW performed the biochemical and cell biologyexperiments. NK participated in ascertaining the patients and families, and providedgenetic counselling to the families. DS, AB and RR performed the neurological andclinical investigation on the patients. YS performed sequencing studies andbioinformatics analyses. MK participated in ascertaining the patients and families,and in performing the molecular and linkage analyses. BG and WS performed theneurological, clinical and laboratory investigation on the patients and families. HC andVS performed immunohistochemistry studies and participated in the cellular biologystudies. TF ascertained the patients, performed clinical analyses and follow-up. DGparticipated in the statistical analyses for the linkage studies. AFV participated in theclinical evaluation of the patients and performed botulinum treatments. YAparticipated in the biochemical analyses and the design of experiments. AMWperformed the zebrafish studies. RK participated in the zebrafish studies, linkageanalyses and designing of the experiments. TCFZ is the principal investigator of thisproject. She designed the research, recruited and examined patients, performed and

supervised research, analysed clinical, molecular and biochemical data. All authorshelped in writing and reviewing the manuscript.

Funding This work was supported by the Rappaport Institute for Research Haifa toTCFZ, Israel; Microsoft Inc 2009331 to TCFZ; and the Intramural Research Program ofthe National Human Genome Research Institute, National Institute of Health,Z99HG999999 to WW.

Competing interests None.

Patient consent All study participants and parents of minors signed locally approvedand appropriate informed consent forms, delivered by board certifiedgeneticists/genetic counselors able to communicate in the patients’ local language,following local and international Helsinki committees’ directives. The anonymity ofpatients is kept.

Ethics approval The ethics approval was provided by the Ethics committee ofWestern Galilee Hospital, Nahariya, Israel and by the supreme Ethics committee of theIsraeli Ministry of Health.

Provenance and peer review Not commissioned; externally peer reviewed.

REFERENCES1. Strumpell A. Beitrage zur Pathologia des Ruckenmarks. Arch Psychiatr

Nervenkrankheiten 1880;10:676e717.2. Fink JK. Hereditary spastic paraplegia Overview. In: Pagon RA, Bird TD, Dolan CR,

et al, eds. Gene Reviews. University of Washington, Seattle (WA), 1993.3. Hedera P, Bandmann O. Complicated autosomal recessive hereditary spastic

paraplegia: a complex picture is emerging. Neurology 2008;70:1375e6.4. Reid ER, Elena I. Hereditary spastic paraplegias. In: Scriver C, et al, eds,

Metabolic and Molecular Bases of Inherited Disease. OMMBID. New York:McGraw-Hill, 2010.

5. Coutinho P, Barros J, Zemmouri R, et al. Clinical heterogeneity of autosomalrecessive spastic paraplegias: analysis of 106 patients in 46 families. Arch Neurol1999;56:943e9.

6. Hurley JH, Boura E, Carlson LA, et al. Membrane budding. Cell 2010;143:875e87.7. Reid E, Connell J, Edwards TL, et al. The hereditary spastic paraplegia protein

spastin interacts with the ESCRT-III complex-associated endosomal protein CHMP1B.Hum Mol Genet 2005;14:19e38.

8. Renvoise B, Parker RL, Yang D, et al. SPG20 protein spartin is recruited tomidbodies by ESCRT-III protein Ist1 and participates in cytokinesis. Mol biology cell2010;21(19):3293e303.

9. Lee JA, Gao FB. Neuronal functions of ESCRTs. Exp Neurobiol 2012;21:9e15.10. Genzer-Nir M, Khayat M, Kogan L, et al. Mammary-digital-nail (MDN) syndrome:

a novel phenotype maps to human chromosome 22q12.3-13.1. Eur J Hum Genet2010;18:662e7.

11. Silberstein M, Tzemach A, Dovgolevsky N, et al. Online system for faster multipointlinkage analysis via parallel execution on thousands of personal computers. Am JHum Genet 2006;78:922e35.

12. Erlich Y, Edvardson S, Hodges E, et al. Exome sequencing and disease-networkanalysis of a single family implicate a mutation in KIF1A in hereditary spasticparaparesis. Genome Res 2011;21:658e64.

13. Klebe S, Lossos A, Azzedine H, et al. KIF1A missense mutations in SPG30, anautosomal recessive spastic paraplegia: distinct phenotypes according to the natureof the mutations. Eur J Hum Genet 2012;20:645e9.

14. Westerfield M, McMurray JV, Eisen JS. Identified motoneurons and theirinnervation of axial muscles in the zebrafish. J Neurosci 1986;6:2267e77.

15. Fassier C, Hutt JA, Scholpp S, et al. Zebrafish atlastin controls motility and spinalmotor axon architecture via inhibition of the BMP pathway. Nat Neurosci2010;13:1380e7.

16. Eymard-Pierre E, Lesca G, Dollet S, et al. Infantile-onset ascending hereditaryspastic paralysis is associated with mutations in the alsin gene. Am J Hum Genet2002;71:518e27.

17. Falik-Zaccai TC, Kfir N, Frenkel P, et al. Population screening in a Druze community:the challenge and the reward. Genet Med 2008;10:903e9.

18. Falik-Zaccai TC, Khayat M, Luder A, et al. A broad spectrum of developmentaldelay in a large cohort of prolidase deficiency patients demonstrates markedinterfamilial and intrafamilial phenotypic variability. Am J Med Genet BNeuropsychiatr Genet 2010;153B:46e56.

19. Hadano S, Otomo A, Kunita R, et al. Loss of ALS2/Alsin exacerbates motordysfunction in a SOD1-expressing mouse ALS model by disturbing endolysosomaltrafficking. PloS One 2010;5:e9805.

20. Bache KG, Slagsvold T, Cabezas A, et al. The growth-regulatory protein HCRP1/hVps37A is a subunit of mammalian ESCRT-I and mediates receptor down-regulation.Mol Biol Cell 2004;15:4337e46.

21. Stuchell MD, Garrus JE, Muller B, et al. The human endosomal sorting complexrequired for transport (ESCRT-I) and its role in HIV-1 budding. J Biol Chem2004;279:36059e71.

22. Truschel ST, Simoes S, Setty SR, et al. ESCRT-I function is required for Tyrp1transport from early endosomes to the melanosome limiting membrane. Traffic2009;10:1318e36.


New disease loci


http://jmg.bmj.com/


23. Ruland J, Sirard C, Elia A, et al. p53 accumulation, defective cell proliferation, andearly embryonic lethality in mice lacking tsg101. Proc Natl Acad Sci U S A2001;98:1859e64.

24. Li L, Cohen SN. Tsg101: a novel tumor susceptibility gene isolated by controlledhomozygous functional knockout of allelic loci in mammalian cells. Cell1996;85:319e29.

25. Jiao J, Sun K, Walker WP, et al. Abnormal regulation of TSG101 inmice with spongiform neurodegeneration. Biochim Biophys Acta2009;1792:1027e35.

26. Xu Z, Liang L, Wang H, et al. HCRP1, a novel gene that is downregulated inhepatocellular carcinoma, encodes a growth-inhibitory protein. Biochem Biophys ResCommun 2003;311:1057e66.

27. Pickart CM, Fushman D. Polyubiquitin chains: polymeric protein signals. Curr OpinChem Biol 2004;8:610e16.

28. Sadoul K, Wang J, Diagouraga B, et al. The tale of protein lysine acetylation in thecytoplasm. J Biomed Biotechnol 2011;2011:970382.

29. Raiborg C, Stenmark H. The ESCRT machinery in endosomal sorting of ubiquitylatedmembrane proteins. Nature 2009;458:445e52.

30. Saksena S, Emr SD. ESCRTs and human disease. Biochem Soc Trans2009;37:167e72.

31. Bakowska JC, Jupille H, Fatheddin P, et al. Troyer syndrome protein spartin ismono-ubiquitinated and functions in EGF receptor trafficking. Mol Biol Cell2007;18:1683e92.

32. Ciccarelli FD, Proukakis C, Patel H, et al. The identification of a conserved domain inboth spartin and spastin, mutated in hereditary spastic paraplegia. Genomics2003;81:437e41.

33. Connell JW, Lindon C, Luzio JP, et al. Spastin couples microtubule severingto membrane traffic in completion of cytokinesis and secretion. Traffic2009;10:42e56.

34. Evans K, Keller C, Pavur K, et al. Interaction of two hereditary spastic paraplegiagene products, spastin and atlastin, suggests a common pathway for axonalmaintenance. Proc Natl Acad Sci U S A 2006;103:10666e71.

35. Fonknechten N, Mavel D, Byrne P, et al. Spectrum of SPG4 mutations in autosomaldominant spastic paraplegia. Hum Mol Genet 2000;9:637e44.

36. Park SH, Zhu PP, Parker RL, et al. Hereditary spastic paraplegia proteins REEP1,spastin, and atlastin-1 coordinate microtubule interactions with the tubular ERnetwork. J Clin Invest 2010;120:1097e110.

37. Patel H, Cross H, Proukakis C, et al. SPG20 is mutated in Troyer syndrome, anhereditary spastic paraplegia. Nat Genet 2002;31:347e8.

38. Sanderson CM, Connell JW, Edwards TL, et al. Spastin and atlastin, two proteinsmutated in autosomal-dominant hereditary spastic paraplegia, are binding partners.Hum Mol Genet 2006;15:307e18.

39. Yang D, Rismanchi N, Renvoise B, et al. Structural basis for midbody targeting ofspastin by the ESCRT-III protein CHMP1B. Nat Struct Mol Biol 2008;15:1278e86.

40. Rusten TE, Vaccari T, Lindmo K, et al. ESCRTs and Fab1 regulate distinct steps ofautophagy. Curr Biol 2007;17:1817e25.

Journal of Medical Genetics online

Visit JMG online for free editor’s choice articles, online archive, email alerts, blogs or to submityour paper. Keep informed and up to date by registering for electronic table of contents atjmg.bmj.com.


New disease loci


http://jmg.bmj.com/


doi: 10.1136/jmedgenet-2012-1007422012

2012 49: 462-472 originally published online June 20,J Med Genet Yifat Zivony-Elboum, Wendy Westbroek, Nehama Kfir, et al. spastic paraparesisautosomal recessive complex hereditary A founder mutation in Vps37A causes

http://jmg.bmj.com/content/49/7/462.full.htmlUpdated information and services can be found at:

These include:

Data Supplement http://jmg.bmj.com/content/suppl/2012/07/11/jmedgenet-2012-100742.DC1.html

"Supplementary Data"

References http://jmg.bmj.com/content/49/7/462.full.html#ref-list-1

This article cites 38 articles, 12 of which can be accessed free at:

serviceEmail alerting

the box at the top right corner of the online article.Receive free email alerts when new articles cite this article. Sign up in

CollectionsTopic

(224 articles)Neuromuscular disease � (17 articles)Motor neurone disease �

(752 articles)Genetic screening / counselling � (218 articles)Dermatology �

(278 articles)Calcium and bone � Articles on similar topics can be found in the following collections

Notes

http://group.bmj.com/group/rights-licensing/permissionsTo request permissions go to:

http://journals.bmj.com/cgi/reprintformTo order reprints go to:

http://group.bmj.com/subscribe/To subscribe to BMJ go to:


http://jmg.bmj.com/content/49/7/462.full.html

http://jmg.bmj.com/content/suppl/2012/07/11/jmedgenet-2012-100742.DC1.html

http://jmg.bmj.com/content/49/7/462.full.html#ref-list-1

http://jmg.bmj.com/cgi/collection/calcium_and_bone

http://jmg.bmj.com/cgi/collection/dermatology

http://jmg.bmj.com/cgi/collection/genetic_screening_counselling

http://jmg.bmj.com/cgi/collection/motor_neurone_disease

http://jmg.bmj.com/cgi/collection/neuromuscular_disease

http://group.bmj.com/group/rights-licensing/permissions

http://journals.bmj.com/cgi/reprintform

http://group.bmj.com/subscribe/

http://jmg.bmj.com/


new disease loci - biumedweb.md.biu.ac.il/research/tzipora-falik-zaccai/images/yifat_2012.pdf ·...

Documents