SCAMP5, NBEA and AMISYN: three candidate genes for autism

26 Pages·2010·0.86 MB·English

Human Molecular Genetics, 2010, Vol. 19, No. 7 1368–1378
doi:10.1093/hmg/ddq013
Advance Access published on January 12, 2010
SCAMP5, NBEA and AMISYN: three candidate
genes for autism involved in secretion of large
dense-core vesicles
Dries Castermans1, Karolien Volders1, An Crepel2, Liesbeth Backx2, Rita De Vos3,
Kathleen Freson4, Sandra Meulemans1, Joris R. Vermeesch2, Connie T.R.M. Schrander-
Stumpel8, Peter De Rijk9,10, Jurgen Del-Favero9,10, Chris Van Geet4,5, Wim J.M. Van De Ven6, D
o
Jean G. Steyaert2,7,9, Koen Devriendt2,{ and John W.M. Creemers1,(cid:2),{ wn
lo
a
d
1Department of Human Genetics, Laboratory for Biochemical Neuroendocrinology, 2Center for Human Genetics, ed
Division of Clinical Genetics, 3Department of Medical Diagnostic Sciences, Morphology and Molecular Pathology from
Section,4DepartmentofMolecularandCellularMedicine,CenterforMolecularandVascularBiology,5Departmentof h
ttp
Pediatrics,6DepartmentofHumanGenetics,LaboratoryforMolecularOncologyand7DepartmentofChildPsychiatry, s
Catholic University of Leuven, 3000 Leuven, Belgium, 8Centre for Clinical Genetics, University of Maastricht, ://ac
a
6201 BL Maastricht, the Netherlands, 9Applied Molecular Genomics Group, Department of Molecular Genetics, de
m
VIB, Belgium and 10University of Antwerp (UA), 2610 Antwerp, Belgium ic
.o
u
p
ReceivedNovember10,2009;RevisedandAcceptedJanuary11,2010 .co
m
/h
m
g
Autism is a neurodevelopmental disorder characterized by impaired social reciprocity, impaired communi- /a
cation and stereotypical behaviors. Despite strong evidence for a genetic basis, few susceptibility genes rtic
le
have been identified. Here, we describe the positional cloning of SCAMP5, CLIC4 and PPCDC as candidate -a
b
genes for autism, starting from a person with idiopathic, sporadic autism carrying a de novo chromosomal stra
translocation. One of these genes, SCAMP5 is silenced on the derivative chromosome, and encodes a c
t/1
brain-enriched protein involved in membrane trafficking, similar to the previously identified candidate 9
/7
genes NBEA and AMISYN. Gene silencing of Nbea, Amisyn and Scamp5 in mouse b-TC3 cells resulted in /1
3
a 2-fold increase in stimulated secretion of large dense-core vesicles (LDCVs), while overexpression sup- 6
8
pressed secretion. Moreover, ultrastructural analysis of blood platelets from the patients with haploinsuffi- /29
0
cieny of one of the three candidate genes, showed morphological abnormalities of dense-core granules, 1
1
4
which closely resemble LDCVs. Taken together, this study shows that in three independent patients 6
b
with autism three different negative regulators of LDCV secretion are affected, respectively, suggesting y
g
that in at least a subgroup of patients the regulation of neuronal vesicle trafficking may be involved in the u
e
s
pathogenesis of autism. t o
n
1
5
A
p
ril 2
0
INTRODUCTION 1
There is compelling evidence of a genetic cause for autism 9
Autism and autism spectrum disorder is a frequent and severe in the majority of cases (2). In the last few years, the
developmental disorder of the central nervous system, with an generalideahasgrownthatgenesinvolvedinthepathogenesis
estimated prevalence of about 6/1000 (1). Given the lack of of autism play a role in the overall processes of neuronal
reliablediagnosticneurobiologicalorgeneticmarkers,thediag- signaltransmission,includingmoleculessuchassynapticscaf-
nosis of autism is still achieved on purely clinical grounds. foldingproteins,celladhesionmolecules,proteinsinvolvedin
(cid:2)To whom correspondence should be addressed at: Department of Human Genetics, Herestraat 49, B-3000 Leuven, Belgium. Tel: þ32 16346080;
Fax:þ3216346073;Email:[email protected]
†Theauthorswishittobeknownthat,intheiropinion,thelasttwoauthorsshouldberegardedasjointLastAuthors.
# The Author 2010. Published by Oxford University Press. All rights reserved.
For Permissions, please email: [email protected]
Human Molecular Genetics, 2010, Vol. 19, No. 7 1369
second-messenger systems, secreted proteins, receptors and Molecular analysis
transporters (3–7). Moreover, a number of cell adhesion
proteins coordinating synaptic connectivity between dendrites Genome-wide array Comparative Genome Hybridisation
and axons were shown to be involved in autism, e.g. the post- (aCGH, BAC/PAC array with resolution of 1Mb) revealed
synaptic neuroligins (NLGN3 and 4) (8–12), a presynaptic that no additional deletions or duplications larger than 1Mb
neurexin (NRXN1) (13,14), binding partner of the neuroligins were present in the genome of the reported subject. By
SH3 and multiple ankyrin repeat domains 3 (SHANK3) means of Fluorescent in situ Hybridization (FISH), the break-
(15,16)andneurexinsuperfamilymembercontactin-associated point on chromosome 1 was found to be flanked by BAC
protein-like2(CNTNAP2)(17–19).Recently,Ca2þ-dependent RP11-373M8 (NCBI AL445648) and cosmid c418b12 at the
activator protein for secretion 2 (CADPS2), involved in the telomeric side, and BAC RP3-398I9 (NCBI AL023096) and
regulation of large dense-core vesicle (LDCV) release and cosmid 208b4 at the centromeric side (Supplementary
synaptic vesicle fusion (20), has been added to this list of Material, Fig. S1A). BAC RP11-108J9 (NCBI AL662924)
D
genes implicated in autism (21,22). spanned the breakpoint (Supplementary Material, Fig. S1B). o
w
The identification and characterization of additional genes Thebreakpointmustthereforebelocatedinthechlorideintra- n
lo
involved in autism is expected to increase our understanding cellular channel 4 gene (CLIC4), most likely between exon 1 a
d
of the pathogenesis of this disorder. Chromosomal anomalies and exon 4 (Supplementary Material, Fig. S1A). Since there ed
are found in (cid:2)5% of individuals with autism and their study are no alternative CLIC4 transcripts known lacking the 50 fro
m
hascontributedmuchtowardstheidentificationofautismcan- end,thetranslocationispredictedtodisruptCLIC4expression h
didate genes. The characterization of balanced translocations fromthisallele.Thiswasconfirmedonnorthernblot,showing ttp
is particularly useful compared with chromosomal deletions adecreasedexpressionoftheCLIC4geneinanEpstein–Barr s://a
or duplications, since the identification of the breakpoint virus (EBV) transformed leukocytes from the patient (Sup- c
a
regions allows the selection of a small number of candidate plementaryMaterial,Fig.S1C).Usingamultipletissuenorth- de
m
genes at or near the breakpoints (23,24). ern blot, expression was detected in all tissues investigated, ic
Here we report on the positional cloning of novel candi- including in brain (Supplementary Material, Fig. S1D). .ou
date genes starting from a patient with autism carrying a Onchromosome 15,BACRP11-151H2(NCBI AC015720) p.c
de novo balanced chromosomal translocation. Together andcosmids32h10and76d1spannedthebreakpoint(Fig.1A om
with two previously reported candidate genes for autism, and B). Southern blot analysis with a probe from these /hm
NBEA (25) and AMISYN (26), we present SCAMP5 as a cosmids showed rearranged fragments in the DNA of the g/a
third novel candidate gene predicted to play a role in patientusingtwodifferentrestrictionenzymes(Fig.1C),loca- rtic
neuronal vesicle trafficking. Functional studies show that lizing the breakpoint to a 3.18kb XbaI restriction fragment. le
thethreeproteinsfunctionasnegativeregulatorsofsecretion Thisindicatedthatthebreakpointislocatedbetweentheputa- -ab
s
of LDCVs. tive exons 1 and 3 of the recently characterized PPCDC gene tra
(Fig. 1A). A CpG island (CpG34 - covering exon 1 of ct/1
PPCDC) and a putative promoter were identified by the Pro- 9
/7
RESULTS moter Inspector program (Genomatix). The gene encodes a /1
3
transcript of 2.3kb, with an ATG start in exon 2. However, 6
8
Case report there is evidence for the presence of an alternative splice /2
9
The subject is a 40-year-old male, living in a group home for variant lacking the first 2 exons and containing a truncated 01
adultswithautisticdisorder.Heisthesecondchildofhealthy, last exon (PPCDC(cid:2), Fig. 1A) (28), and we cannot exclude 146
unrelated parents. Family history is negative with regard to that the transcription of this shorter transcript is unaffected by
autism or mental retardation. Pregnancy and delivery were by the translocation. The PPCDC gene is ubiquitously gu
uneventful, and he was born at term with birth weight expressed and codes for an enzyme involved in the biosyn- es
4.5kg. Early psychomotor developmental milestones were thesis of coenzyme A. Since this cofactor is essential in the t on
mildlydelayed.Languageandsocialdevelopmentweremark- synthesis and oxidation of fatty acids, and the oxidation of 15
edlydelayed. At5years,he wasreferred forchildpsychiatric pyruvate in the citric acid cycle, we do not further consider A
p
assessment. A diagnosis of ‘early infantile autism’ was made PPCDC a candidate gene for autism. ril 2
based on a lack of responsiveness to other people, a very dis- The SCAMP5 gene is located 10kb centromeric to the 0
1
turbed language development and stereotyped movements. breakpoint on chromosome 15q (Fig. 1A). This gene is 9
Testing with the Merrill-Palmer scale showed an IQ of 63 almost specifically expressed in brain (Fig. 1D). Expression
withanunevenprofile.Visuomotorskillsweregood,language studies were carried out by means of quantitative real-time
skillswere impaired. Psychiatricexamination attheage of 15 PCR(qRT-PCR)oncDNAfromEBVtransformedleukocytes
years confirmed a diagnosis of autistic disorder according to ofthepatientversusfivenormalcontrols.SCAMP5expression
the DSM-III-R criteria (27). Ritualistic behavior and stereo- was robustly detectable (Ct(cid:3)30) but slightly variable
typedbodymovementswereverypronounced.Otherpsychia- between five control individuals (1.0+0.23, n¼5). Relative
tric problems at this age included mood changes and anxiety, quantification of SCAMP5 expression in two different RNA
episodic aggression and automutilation. Clinical examination extracts revealed a reduction in expression to ,50%
at the age of 18 years revealed no dysmorphic features. (0.39+0.08, n¼2) in the patient, consitent with haploinsuf-
Height was 188cm, head circumference 54cm. He carried a ficiency (Fig. 1E). By single nucleotide polymorphism (SNP)
de novo apparently balanced translocation with karyotype analysis, we investigated the expression of this gene in the
46,XY t (1;15)(p36.11;q24.2). patient. At the genomic level, the patient was found to be
1370 Human Molecular Genetics, 2010, Vol. 19, No. 7
D
o
w
n
lo
a
d
e
d
fro
m
h
ttp
s
://a
c
a
d
e
m
ic
.o
u
p
.c
o
m
/h
m
g
/a
rtic
le
-a
b
s
tra
c
t/1
9
/7
/1
3
6
8
/2
9
0
1
1
4
6
b
y
g
u
e
s
t o
n
1
5
A
p
ril 2
0
1
9
Figure1.ThePPCDCgeneisdisruptedandSCAMP5expressionisaffectedbythetranslocationon15q24.2.(A)Physicalmapof15q24.2(UCSCGenomeBrowser,
July2003version—centromereontheleft).ThepositionoftheBACends(AC125435andAC113208)andcosmidsusedforFISHanalysisareshown.Arrowsindi-
catethepositionofthetranslocationbreakpointwithregardtothegenomicclone(rightarrow,distal;leftarrow,proximal).Genesinthegenomiclocusaredepicted
(SCAMP5,PPCDCandalternativetranscriptPPCDC(cid:2)),andexonsofallthesegenesarenumbered,indicatingdirectionoftranscription.(B)FISHanalysisonmeta-
phasespreadofthepatientshowsthatBACAC015720(left)andcosmid32h10spanthebreakpointonchromosome15.Der,derivativechromosome.(C)Southern
blotanalysiswithaprobefromC32h10revealsrearrangedfragments(indicatedbyarrow)inthepatient(P)comparedwithacontrol(C).(D)Multipletissuenorthern
blot(Heart,Brain,Placenta,Lung,Liver,Skeletalmuscle,Kidney,Pancreas)forhumanSCAMP5.(E)QuantitativeRT–PCRexpressionstudyofSCAMP5inleu-
kocytecDNAshowshaplo-insufficiencyforthepatientwiththetranslocation.Relativequantificationresultsandrespectivestandarddeviationsareshownfortwo
differentpatientcDNAsamples,normalizedtothemeanexpressionvaluein5controls.(F)ExpressionanalysisofSCAMP5inthepatient.Sequenceanalysison
gDNAofthepatientshowedheterozygosityforrs8033925(A/G),anSNPatthe30-UTRofSCAMP5.cDNAanalysisrevealedexpressionoftheAallelewassig-
nificantlyreduced,comparedwithequalexpressionfrombothallelesobservedinfourcontrols(oneshown).Sequencinganalysisofrs8033925inthedissectedampli-
fiedDNAfromder(15)ofthepatientshowedtheAallele,consistentwithaffectedexpressionfromtheaberrantchromosome15.
Human Molecular Genetics, 2010, Vol. 19, No. 7 1371
D
o
w
n
lo
a
d
e
d
fro
m
h
ttp
s
://a
c
a
d
e
m
Figure2.NBEA,SCAMP5andAMISYNareinvolvedintheregulatedsecretionofLDCVs.(A)Analysisoftheendogenousexpressionofthecandidategenes ic
inendocrineandneuronalmousecelllines(a-TC1-6,b-TC3,AtT-20andNeuro2A)byRT–PCR(A)andwesternblotanalysis(B).Sincenoantibodydirected .o
u
againstScamp5wasavailable,endogenousScamp5couldnotbedetected.(C)WesternblotanalysisshowingthesilencingcapacityoftheRNAiconstructs(R) p
.c
bymeansoftransientco-transfectionofb-TC3cellswithacorrespondingexpressionconstruct(E).(M)indicatesthemock-transfectedlanes.Foreachgene,the o
m
resultsofthemU6proRNAiconstructwiththebestsilencingcapacityisshown.(D)Effectofsilencingandoverexpressionofthegenesofinterestonregulated /h
secretioninb-TC3cells.Stimulatedsecretion(additionofIBMXandforskolin)ofFlag-AGRPwasassayed3daysaftertransientco-transfectionwithashRNA m
(Nbea,Scamp5,Amisyn,Clic4,SytIXandMALT1)orfull-lengthexpressionconstruct(Scamp5,AmisynandClic4).Numberofexperimentsaresixforeach g/a
construct.Meandataarepresented,standarddeviationsaredepictedandasterisksindicateP-values ,0.01. rtic
le
-a
b
s
heterozygous for rs8033925, an SNP (A/G) in the 30-UTR of expressed in all cell lines investigated, whereas Amisyn was tra
c
SCAMP5 (Fig. 1F). Subsequent analysis of the corresponding only expressed in the insulinoma cell line b-TC3. RT–PCR t/1
mRNA fragment was performed in duplo on two different confirmed that Synaptotagmin IX (SytIX), used as a control, 9/7
RNA extracts from the patient. This study revealed altered is alsoexpressed in thiscell line(data not shown). Constructs /1
3
expression of SCAMP5, since the expression from the A expressing silencing short hairpin RNAs (shRNA) were 68
allele was consistently found to be strongly reduced designed and cloned for the candidate genes Nbea, Scamp5, /29
0
(Fig. 1F), unlike four heterozygous controls (one shown in Amisyn and Clic4 as well as for the controls SytIX and 1
1
Fig. 1F). Microdissection of the derivative chromosome 15 MALT1. Knockdown efficiency of these shRNA constructs 4
6
containing the SCAMP5 gene, and subsequent amplification was tested in b-TC3 cells by transfection with the silencing b
y
and sequencing showed the presence of the A allele on this construct and an epitope tagged recombinant construct repre- gu
e
aberrant chromosome (Fig. 1F), consistent with affected senting the gene of interest (Fig. 2C). s
expression of the SCAMP5 gene from the translocated b-TC3 cells have previously been shown to display low t on
chromosome. tonic release of recombinant AGRP while high secretion was 15
observedafterstimulationwithsecretagogues,suchasforsko- A
p
Novel candidate genes involved in regulated secretion lin and IBMX (29). In a pilot experiment it was found by ril 2
means of immunofluorescence analysis that endogenous 0
of LDCVs 19
Nbea protein was almost undetectable 72h after transfection
To investigated the role of SCAMP5, CLIC4 and the pre- with the appropriate silencing construct (data not shown).
viously identified candidate genes NBEA and AMISYN in the This timepoint was chosen for further experiments.
secretion of LDCVs, RNA interference (RNAi)-mediated Results of the regulated secretion assay are shown in
gene silencing and overexpression experiments were per- Figure2D.KnockdownoftheSNAREregulatorSytIXresulted
formed. In this assay, we used recombinant Agouti-related in a 2-fold reduction of stimulated secretion of LDCVs, con-
protein (AGRP) as a marker for LDCV cargo. AGRP is a sistent with reported data (30). As expected, we observed no
soluble protein that is efficiently sorted to the regulated effect with the shRNA construct for human MALT1 (31), as
secretory pathway (29). it does not have a target in mouse. Knockdown of Nbea,
By means of RT–PCR and western blot analysis, we Scamp5 and Amisyn resulted in a significant 1.8–2.2-fold
studied the endogenous expression of the four candidate increaseofregulatedsecretion.Consistentwiththesefindings,
genes in both neuronal and endocrine mouse cell lines overexpressionofScamp5andAmisynresultedinasignificant
(Fig. 2A and B). Nbea, Scamp5 and Clic4 were found to be 4-foldand2-folddecreaseofregulatedsecretion,respectively.
1372 Human Molecular Genetics, 2010, Vol. 19, No. 7
D
o
w
n
lo
a
d
e
d
fro
m
h
ttp
s
://a
c
a
d
e
m
ic
.o
u
p
.c
o
m
/h
m
g
/a
rtic
Figure3.Electronmicrographsofbloodplateletsinthereportedpatients.Themorphologyoftheplateletsofacontrol(A)andthethreereportedpatients(B–D) le
were examined. In the patients with chromosomal rearrangements, (cid:2)50% of the observed dense granules showed an altered morphology. A representative -ab
ecxoanmtropllepiastieshnotwrenp.rIensseenttsedshaowreaguhlaigrhreorumndagdneinfisceaticoonreofsutrhreouDnCdeGdobfythaesdmeaplilctheadlopl(aAteleints(elto)w. eInr)paantdienotnsewoifthantohtehecrhprolamteolseotm(uaplpreera)r.raDnegnesmeegnrtasnuaflefescotifnga strac
genesNBEA(Binset),SCAMP5(Cinset)andAMISYN(Dinset)thedensecoreofthegranulesaresmaller,moreirregular,frequentlyperipherallylocalized t/1
andpoorlydelineated. 9/7
/1
3
6
8
In b-TC3 cells, Nbea could not be overexpressed detectably, peripherally localized, less electron-dense, poorly delineated /2
9
probably related to its large size ((cid:2)9kb cDNA). For Clic4, and surrounded by an unclear halo (Fig. 3B–D insets). 0
1
no effect on the regulated secretion of LDCVs was observed, 1
4
neither by overexpression nor by knockdown. 6 b
Mutation analysis y
g
u
In order to establish a genotype–phenotype correlation e
s
Ultrastructural analysis of dense-core granules in blood between candidate genes described above and autism, t o
n
platelets of the patients mutation analyses were performed. Resequencing of all 7 1
5
Inordertoprovideevidenceforadefectinregulatedsecretion coding exons of SCAMP5, including the intron–exon bound- A
p
irneptohretepdatpieantites,nutsltrwaastsrupcetrufroarlmaenda.lyPsilsatoeflebtslowoderpelactheloestesnofasthea amriuetsa,tioinnsa. Gcoivheonrt iotfs 1la9r2gepesriszoen,smwuittahtioauntisamnalryesvisealoefdtnhoe ril 20
1
NBEA gene was limited to the detection of small deletions 9
model because they are accessible and they contain dense
or duplications. In a cohort of 248 persons with autism, no
granules. Many components involved in the secretion of
copy-number alterations were detected in eight regions span-
dense granules are conserved with LDCVs in neuronal and
ning the NBEA gene.
endocrinecells(32).RT–PCRconfirmedthatallofthecandi-
date genes are expressed in platelets (data not shown). The
morphology of the platelets of controls (Fig. 3A) and all
DISCUSSION
three patients with chromosomal rearrangements (Fig. 3B–
D) was examined by electron microscopy. In the patients Thisstudydescribestheidentificationofnovelautismcandidate
with chromosomal rearrangements, (cid:2)50% of the observed genesinapersonwithautism,carryingadenovotranslocation
dense-core granules (DCGs) presented with an altered involvingchromosomes1and15:46,XYt(1;15)(p36.11;q24.2).
pattern (represented in Fig. 3B–D) while the remaining In the subject, two genes were found to be directly disrupted,
DCGs presented with a normal morphology (data not CLIC4(ChLorideIntracellularChannel4gene)atchromosome
shown).Thedensecoreofthealteredgranuleswasfrequently 1,andPPCDC(PhosphoPantothenoylCysteineDeCarboxylase
Human Molecular Genetics, 2010, Vol. 19, No. 7 1373
gene) at chromosome 15. Since recently several autism genes alsoSCAMP4,mightplayaroleasnegativeregulatorofexo-
havebeenproposedwitharoleinsynapticprocesses(3,4,33), cytosis, possibly by influencing the formation of fusogenic
CLIC4,codingforaproteinpredictedtoplayaroleinneuronal SNARE complexes.
vesicletrafficking(34),wasconsideredagoodfunctionalcandi- AMISYNhaspreviouslybeendescribedasamemberofthe
dategeneforautism.Incontrast,thePPCDCgenecodesforan inhibitory class of SNAREs or i-SNAREs (52,53). The pres-
enzyme involved in the biosynthesis of coenzyme A (35,36). ence of a VAMP2-like C-terminal coiled-coil domain, inter-
Since this metabolic process is unlikely to be involved in acting with t-SNARE complex syntaxin-1/SNAP25, together
autism, this gene was not included in any further studies. with the unique N-terminus lacking the hydrophobic stretch
However, SCAMP5 (Secretory Carrier Membrane Protein 5), that may serve as a transmembrane anchor to the vesicle,
located 10kb proximal to this breakpoint on chromosome AMISYN cannot function as a classical v-SNARE (45). The
15q, was considered an excellent additional functional candi- formation of so-called nonfusogenic complexes prevents
dategene.SCAMP5ismostabundantlyexpressedinbrainlate proper v/t-SNARE assembly and subsequent fusion of
D
during development, coinciding with the elaboration of vesicle with the plasma membrane (46). Not only do our o
w
maturesynapses,andhasaprobablefunctioninmembranetraf- data support this role for AMISYN as negative regulator of n
lo
ficking(37,38).Chromosomalbreakpointsmayhaveeffectson secretion, we also confirmed the dominant inhibitory effect a
d
e
theexpressionofgeneslocatedatadistance(26,39–41),andin on exocytosis observed by overexpressing the C-terminal d
thepatientwithatranslocationbreakpoint10kbdownstreamof coiled-coil domain (Supplementary material, Fig. S2) as fro
m
SCAMP5,thegene’sexpressionwasfoundreducedto (cid:2)40%. reported earlier in rat PC12 cells (45). h
Analysisofrs8033925,anSNPinthe30-UTRofSCAMP5,sup- DisruptionoftheinteractionofproteinkinaseA(PKA)with ttp
s
ports this haploinsufficiency for SCAMP5 in the reported AKAPsinXenopuslaevismelanotrophiccellshasbeenshown ://a
patient. Moreover, our SNP data on microdissected derivative to result in induced secretion (54), consistent with negative c
a
d
chromosome 15 is consistent with reduced expression of the regulation of secretion. Besides compartmentalization of e
m
translocatedallele. PKA, the interaction withAKAPs, has been suggested to par- ic
Besides SCAMP5 and CLIC4, two other autism candidate ticipate in regulating PKA activity (54). Tight regulation of .ou
p
genes we recently identified, NBEA (25) and AMISYN (26), kinase activity is essential for many processes, e.g. the phos- .c
o
also play a possible role in neuronal vesicle trafficking and phorylationstatusofSNAREs,botht-SNAREsandregulating m
synaptogenesis (42–46). Therefore, the genes considered as i-SNAREs, alters the ability to assemble and form fusogenic /hm
candidate genes in all further studies are SCAMP5, CLIC4, SNARE complexes (46,55–57). g/a
AMISYNandNBEA.Allthesegenesareexpressedinneuronal These observations support our finding that all three pro- rtic
and/orendocrinecells,specializedcelltypesinwhichregulated teins identified function as negative regulators of LDCV le
-a
secretory pathways are present. Neuronal vesicle transport secretion, possibly viatheregulationofSNAREcomplexfor- b
s
includes LDCVs and small synaptic vesicles (SSVs), whereas mation. Therefore, haploinsufficiency of SCAMP5, NBEA tra
c
(neuro)endocrinecellscontainonlyLDCVs(47).Sinceallcan- and/orAMISYNinthepatientsmostlikelyresultsinincreased t/1
didate genes are endogenously expressed in the endocrine secretion of LDCVs, and possibly of SSVs. In line with this 9/7
b-TC3 cell line, we were able to specifically investigate their hypothesis, elevated expression of t-SNARE syntaxin 1A in /1
3
role in regulated secretion of LDCVs. However, it should be lymphocytes was observed to be associated with high func- 68
statedthatourdatadonotexcludeanadditionalroleinsecretion tioning autism (58), likely resulting in elevated formation of /29
0
of SSVs. Our results indicate that SCAMP5, NBEA and fusogenic vesicles and subsequent secretion. Interestingly, a 1
1
AMISYN function as negative regulators of secretion, knock-in mouse model of the gain-of-function mutation 4
6
whereasforCLIC4,noeffectcouldbedemonstrated. R451C in NLGN3, previously reported to be involved in b
y
SCAMPs (secretory carrier membrane proteins) are pro- autism (8), showed an increase in inhibitory synaptic trans- gu
e
posed to play a role as positive regulators of exocytosis mission with no apparent effect on excitatory synapses (59). s
(38), most likely by setting up SNARE interactions between Although allthese reportssupportthe ideathatelevated neur- t on
secretory vesicle and plasma membranes or facilitate fusion onal vesicle trafficking may contribute to the pathogenesis of 15
pore formation (48). However, we show here that Scamp5 (a subgroup of) persons with autism, the latter is not in line A
p
functions as a negative regulator of exocytosis. Most likely with the theoretical model postulating that some forms of ril 2
this is due to the absence of an N-terminal NPF repeat, autismarecausedbyanincreasein‘noiseinformation’result- 0
1
9
found only in SCAMP1-3, combined with the presence of ing from an increased ratio of excitation/inhibition in neural
the so-called E-peptide, located between transmembrane circuits that mediate language and social behaviors (60).
regions 2 and 3 and conserved in all SCAMPs. This However, several functional studies on the physiological con-
11-residue peptide inhibits the fusion of secretory vesicles sequences of certain mutations in candidate genes have
with plasma membrane in mast cells (49,50) and PC12 cells reported ‘contradictory and incompatible results’, suggesting
(48), whereas fusion of the N-terminal NPF repeat of that affecting processes such as exocytosis or synaptogenesis
Scamp2 with full-length Scamp5 resulted in a 2.5-fold in either direction might lead to autism. For instance, the
increase in secretion in our assay (Fig. S2). In addition, it NLGN4 gene on chromosome X, is associated with autism
has been recently reported that expression of SCAMP5 is and mental retardation (8). Further functional analysis of the
markedly increased in the striatum of Huntington disease reported D396X frame shift mutation showed intracellular
patients, an increase positively regulates the aggregation of retention of the NLGN4 mutant protein and loss of synaptic
mutant huntingtin protein via the endocytotic pathway (51). function (61). In addition, an R87W missense mutation
Taken together, this suggests that SCAMP5, and possibly was shown to impair neuroligin-4 folding and endoplasmic
1374 Human Molecular Genetics, 2010, Vol. 19, No. 7
reticulum export (11), and a novel NLGN4 isoform lacking syndrome that includes SCAMP5, autism has not been
exon 4 resulting in monoallelic expression was identified in recorded(73,74),whereasthethreecaseswithamicrodeletion
an autistic female (62). Although all these reports support including NBEA have autism (75–77). In addition to that,
the idea that a point mutation in NLGN4 can cause autism NBEA is located in a linkage region for autism (78,79). In
by a loss-of-function mechanism, very recently, a de novo order to determine to what extent these candidate genes con-
mutation in the regulatory sequence of NLGN4 associated tribute to the cause of autism in the whole population,
with autism was shown to result in increased expression of mutation and association studies need to be performed. Dena-
the gene (12), suggesting that also a gain in functional turing high performance liquid chromatography (DHPLC)
NLGN4 protein can cause autism. analysis and resequencing of the coding exons and all
Regulatedsecretionisnotrestrictedtoneuronsand(neuro)- intron–exon boundaries of SCAMP5 did not reveal any func-
endocrine cells, but also occurs in a conserved manner in tionalmutationsinacohortof192personswithautism.Simi-
specialized cell types like blood platelets (32). It has been larly, the previously reported mutation analysis of AMISYN
D
stated that platelet research can be used to unravel mechan- revealed no functional mutations in a cohort of 227 patients o
w
isms involved in a neurological disorder, such as depression (26). Since NBEA appears to be located at the common n
lo
and autism (63). Moreover, platelet hyperserotonemia fragile site FRA13A (80,81), the region might be predisposed a
d
e
(described in one-third of patients with autism), the only to breakage and rearrangements. Nevertheless, Multiplex d
biochemical anomaly repeatedly observed in autism (64,65), Amplicon Quantification (MAQ) analysis of the gene in a fro
m
supports the idea of blood platelets representing an endophe- cohort of 248 persons with autism did not reveal any copy- h
notypic model for autism. Since platelets contain DCGs that numbervariations(CNVs)inthisgenomicregion.Thesefind- ttp
s
resemble LDCVs both biochemically and morphologically, ings are in agreement with previous reports showing a very ://a
and, most importantly, platelets are easily accessible from low frequency of mutations and CNVs in established candi- c
a
d
patients,ourinvitroresultspromptedustoperformultrastruc- date genes for autism (for instance NLGN3 and NLGN4) e
m
turalanalysisinvivoonbloodplateletsofallthreepatientsand (82–85), underscoring the high genetic heterogeneity of this ic
a number of controls. We detected similar structural abnorm- condition (86). .ou
p
alities of the DCGs in the platelets of all three patients. .c
o
Moreover, analysis of platelet ultrastructure in an autistic m
patient with a deletion of the 22q13 region, including the /hm
well-established autism candidate gene SHANK3, revealed MATERIALS AND METHODS g/a
tmheatesraimale, Fsitgru.cSt3u)r.al aberration in the DCGs (Supplementary Electronic-databaseandprimerinformationisprovidedonline rticle
as Supplementary material. -a
Althoughwedonotexcludeanadditionalroleforthesepro- b
teinsinsecretionofSSVsorendocyticprocesses,weshoweda Positional cloning of the translocation breakpoints (includ- stra
ing strategy, genomic resources, FISH, aCGH, Southern and c
ccLlaDenaCdriVdfaus.tnecItnigoennnaeeluraonpndrso,dmutohcrtepssheoilnLogDitCchaeVlsinrecvgoounlvltaaetimendennteseuocrforetartliolopnthhirneosef, n(2o5rt,8h7er)n.Prbimloetrtisnugs)edwfoasrdepseirgfnoirnmgetdheSasoutdheesrcnriabneddnobrtehfoerrne t/19/7/1
like brain-derived neurotrophic factor (BDNF) and nerve probes are listed as Supplementary material. 368
growth factor. These factors regulate neuronal structure and /29
0
synaptic plasticity, and their uncleaved precursors mediate 1
1
apoptosis of neurons (66). In line with the hypothesized 4
6
increased regulated secretion of LDCVs in our patients, two qRT-PCR and calculation of relative expression b
y
recent studies have reported increased concentrations of Primersweredesignedusingtheweb-basedRocheProbefinder gu
e
BDNF in a group of persons with autism (67,68). It has software and ordered from Eurogentec (Seraing, Belgium). s
been suggested that these impaired levels reflect an abnormal Sequences of these primers are available as Supplementary t on
status of prenatal or early postnatal neuronal development, material. Primer pairs were validated with a standard dilution 15
which may lead to autism (67,69). However, since the series. The 3 primer pairs for housekeeping genes ACTB, A
p
autism susceptibility gene CADPS2 is a positive regulator of GUSB, CLK2 were validated as stable housekeeping gene ril 2
neurotrophin secretion (21,22), it is also possible that both primer pairs in EBV cell-lines with the online GENorm soft- 0
1
9
increased and decreased LDCV release may contribute to ware(88).qRT-PCRwascarriedoutontheRocheLightcycler
the development of autism. Since LDCV transport is impli- 480 instrument using Roche Sybrgreen Mastermix in 15mL
cated in a wide range of neuronal processes, such as prolifer- total volumes. cDNA samples of the patient (two different
ationofneuronalprecursors,apoptosis,outgrowthofneurites, RNA extracts) and five independent controls were diluted
synapseformationandsynaptictransmission,itisnotyetclear 1:15, all reactions were done at least in duplicate and a non-
howalteredregulatedsecretionaffectsthebiologyofneuronal template control was included in each series. If necessary,
cells. serieswere replicated toobtainaDCt ,0.3forallduplicates.
Recently, high-resolution techniques, such as aCGH, have Allnon-amplificationandnon-templatecontrolswerenegative
detected de novo submicroscopic chromosomal aberrations withCt .35.Relativegene-expressionvalueswerecalculated
in a significant number of individuals with idiopathic autism with qBase software v.1.3.5 by use of a DCt relative quantifi-
(70–72). For both SCAMP5 and NBEA several independent cation model withPCR efficiency correction (89). The results
patients carrying microdeletions harboring the gene are werenormalizedtothemeanexpressionlevelsofSCAMP5in
reported.Inthefivecaseswiththenovel15q24microdeletion the five control persons.
Human Molecular Genetics, 2010, Vol. 19, No. 7 1375
Study of gene expression using SNPs interestmakinguse ofAmbions’siRNATargetFinder (http://
www.ambion.com/techlib/misc/siRNA_finder.html). Several
Exonic SNPswere used to study the effect of a breakpoint on
target sequences were selected at different positions along
the expression of the SCAMP5 gene nearby the breakpoint at
the length of the gene. 60-mer oligonucleotides encoding
chromosome 15q24. First, by sequencing the genomic DNA
the corresponding silencing short hairpin RNA (shRNA)
(gDNA)corresponding tocodingexonsand50 and30 untrans-
were designed as described (29) and cloned in the mU6pro
latedregions(UTR)ofthesegenes,weshowedthatthesubject
vector (kindly provided by Dr Turner, MI, USA). The
washeterozygousforreportedSNPrs8033925.Next,thepres-
efficiency was confirmed in HEK293T and b-TC3 cells
enceorabsenceofthisheterozygousSNPswasanalyzedinthe
by co-transfection of the silencing shRNA construct with
patientsmRNA(performedinduplooneachofthetwodiffer-
expression vectors encoding the target mRNA in a 4/1 ratio.
ent RNA extracts) by sequencing the corresponding cDNA
In general, the expression construct encoded the complete
fragments, and biallelic expression was tested in four hetero-
coding sequence. However, for Nbea the construct only
zygous controls. PCR amplification and subsequent sequen- D
encoded a C-terminal part (AA1863–2566) including its o
cing was performed in three independent experiments. w
target sequence. n
Isolation of gDNA and mRNA from the patient’s EBV lo
In order to obtain full-length expression constructs for a
transformed leukocytes, reverse transcription of mRNA to de
Clic4, Scamp5 and SytIX, PCR was performed on cDNA d
cDNA and gDNA/cDNA SNP analysis by subsequent PCR obtained by RT–PCR of total RNA from mouse embryonic fro
amplification and sequencing was performed as described m
(41). Specific primers used for amplification and sequencing brain. PCR fragments were purified (QIAquick gel extraction h
kit, Qiagen), followed by a digestion and another purification ttp
oftheregionsofinterestarelistedasSupplementarymaterial. s
step (QIAquick PCR purification kit, Qiagen) before the ://a
inserts were cloned in the expression vectors. Inserts were c
a
d
Microdissection of derivative chromosome 15 cloned into the pcDNA3.1/HisA vector (Invitrogen), except e
m
for Scamp5 that was cloned into the pcDNA3 vector (Invitro- ic
Microdissection was performed as described (90), with minor gen) with addition of an N-terminal HA-tag. Although a .ou
modifications. To identify the der (15) a paint probe for 1pter p
full-length mouse Nbeaexpressionconstruct wasmade (over- .c
wasdesignedbyrandomprimelabelingof10subsequentDOP o
expression was observed in HEK293T cells), we were unable m
amplified BAC clones from the 1-Mb clone set at 1pter. This to obtain detectable overexpression in the b-TC3 cell line. /hm
rcdehesrruo(lm1te5od)s.oiRnmeetvhe1ersahenypdbaroiindntiiz1napgtitoeunrsitnsrgiagnntshaleolscDaoOtnePd1atpomtetprhleiofindedits,htaedlisnesonerdcmteoadfl ctaornDysetmrtauaictlteeirdoinailn.offoramllatthioensecpolnacsmerindisngisthperovpirdimederassuSsuepdpfleomr ethne- g/article
DNA labeled with Cy3-dCTPs by random prime labeling -a
b
showed hybridization signals to the proximal part of chromo- stra
some 15q but not to the distal end of chromosome 15q and to Cell lines c
trhanetdicstharloemnodsoofmcehro1m5oswoamsed1ips.seTchtiesdcoannfidrmneodtthtahtethneoarbmear-l Ctheellsmforoumsethpeanhcuremaatincecmebllryloinneics kbid-TnCey3caenlldlinae-THCE1K-62,93thTe, t/19/7/1
chromosome 15. PCR amplification and subsequent sequen- mouse pituitary tumor cell line AtT-20 and the mouse neur- 368
cing for the SNP rs8033925 using the dissected amplified onal cell line Neuro2A were grown in Dulbecco’s modified /29
DNA as template, including gDNA as a positive control and 0
Eagles Medium (DMEM) supplemented with 10% fetal 1
alistbeldanaksaSsupnpelgeamtievnetacryonmtroalt,erwiaals(Spteurdfoyrmofedgeunseinegxpprreismsieorns fboorvmineedseursuinmgatF3u7g8eCne,56%TCrOan2s.fTercatinosnfecRtieoangeonftce(lRlsocwhaes)pfeor-r 146 b
y
using SNPs). HEK293T cells, and Lipofectamine 2000 (Invitrogen) with gu
e
Plus reagent (Invitrogen) for b-TC3 cells. s
t o
Mutation analysis n
1
5
MutationanalysiswasdonebymeansofDHPLC(Supplemen- Western blot analysis A
p
tcaordyinmgaetxeroinasl)aannddexsuobns–eiqnutreonnt bseoquunednacriinesg oafnathlyesiSsCoAnMPal5l 1C2e%llsgwlyecreerloyls,e4d%inSSDaSm)palendBpurfofeterin(6s0inmMmeTdriias–wHeCrelpprHec6ip.8i-, ril 20
1
gene. The NBEA gene was screened for CNVs using MAQ 9
tated in 4 volumes of methanol (2208C) in the presence of
(Supplementary material).
albumin and dissolved in Sample Buffer containing 4% 2-
mercaptoethanol. Proteins were separated by SDS–PAGE on
10% Tris–Glycine gels, with the exception of endogenous
Analysis of endogenous gene expression in cell lines
Nbea (5% SDS–PAGE) and Agouti-related protein (AGRP)
Endogenous expression of the candidate genes was tested by samples of the ‘regulated secretion’ assay (NuPage 10%
RT–PCR on mRNA from mouse endocrine and neuronal Bis–Tris Gel (Invitrogen). Gels were electroblotted and pro-
cell lines (a-TC1-6, b-TC3, AtT-20 and Neuro2A). cessed as described (29). The polyclonal rabbit antibody
used to detect endogenous neurobeachin was raised against
the synthetic peptide TKVSDDILGNSDRPGS-Cys-KLH-
Gene silencing and overexpression constructs
MBS (Eurogentec). Primary antibodies used for detection of
For RNA interference, 19-mer target regions were designed recombinant tagged constructs are anti-Xpress (Invitrogen),
inside the coding sequences of the different mouse genes of anti-HA(Roche)andanti-FlagM2(Sigma-Aldrich).Anti-Clic4
1376 Human Molecular Genetics, 2010, Vol. 19, No. 7
andanti-AmisynantibodieswerekindlyprovidedbyDrYuspa ACKNOWLEDGEMENTS
(Bethesda, MD, USA) and Dr Scheller (Stanford, CA, USA),
Wethank thepatients andtheir families fortheir cooperation.
respectively. Secondary antibodies used are horseradish per-
Reinhilde Thoelen and Rob Deckers are gratefully acknowl-
oxidase-linked (DAKO). Chemiluminescent signal (Renais-
edged for their excellent technical assistance and Bernard
sance, Perkin Elmer Life Sciences) was either detected by
Thienpont for performing the aCGH experiments.
X-ray film (Fujifilm) or directly quantified using Kodak
Digital Science (Kodak Imager with 1D Image Analysis
Conflict of Interest statement. None declared.
Software, version 3.0).
Regulated secretion assay
FUNDING
The assay was performed essentially as described previously
(29,91).Briefly,b-TC3cellsweretransientlytransfected(Lipo- This work was supported by grants from Cure Autism Now, D
o
fectamine2000)withanshRNAmU6proconstructoroverex- the Interuniversity Attraction Poles Program-Belgian Science wn
pression construct, together with the recombinant AGRP Policy (IUAP 5/25), ‘Geconcerteerde Onderzoeksacties’ loa
construct(C-terminallyFlag-tagged).ForScamp5knockdown GOA/2006/12andGOA/2008/16,‘FondsvoorWetenschappe- de
d
experiments, the AGRP construct was co-transfected with lijk Onderzoek Vlaanderen (FWO)’ G0264.04 and G.0124.02 fro
20mM of synthetic siRNA (Dharmacon) instead of an shRNA and ‘interdisciplinaire onderzoeksprogramma (IDO)’. K.V. is m
cthoenstFrluacgt-.taAgsgaedcoAntGroRl,Pb-cToCns3trcuecltlsawnderethceo-termanpstfyectmedU6wpirtho svuapnpIonrnteodvabtyieadgoroarntwfertoemnscthheapInesntittueucthnvooloorgdieeiananVmlaoaenddigeirneng https
vectororthe‘non-targeting’siRNArespectively.After2days (IWT). J.S. is supported by a grant of the Clinical Research ://a
c
the cells were washed and incubated for 18h in serum-free Fund U.Z. Leuven. K.D. is a Senior Clinical Investigator of ad
e
mediumtoreduce thetonicreleaseofstoredmaterial induced FWO, K.F. is postdoctoral fellow of FWO, A.C. is PhD m
by serum factors. Subsequently, the cells were incubated for fellow of FWO. ic.o
u
3h in fresh serum-free medium (tonic release fraction), fol- p
.c
lowed by 3h incubation in serum-free medium containing o
m
secretagogues [10mM forskolin (Sigma) and 0.1mM IBMX REFERENCES /h
m
(1-Methyl-3-isobutylxanthine,Sigma)](inducedsecretionfrac- g
tpihoons)p.Fhoodrsikesotleinraisseaninahdiebniytolar,tebcoytchlassetiamctuilvaattionrgansdecIrBeMtioXnisoaf 12.. JcFohrheilindtsarogenn,,CwC.Mi.tPh..a(a2un0tdi0sm7M)ysTephresec,tgrSue.Mmne.tdi(ci2sso0or0df7e)arusId.tiePsnteitcdifiidacitasrtoiicrodsn,er1as2n0dan,ed1v1ait8lsu3ac–tli1ion2ni1c5aol.f /article
LDCVsthroughactivationofthePKApathway. relevance:areviewoftheliterature.Mol.Psychiatry,12,2–22. -a
b
Both medium fractions and lysates were collected and the 34.. SStteepyhaearnt,,DJ.G.A..a(n2d00D8e)UlanMraavreclhineg,Wau.ti(s2m0.08A)mW.Jh.aHt’sumne.wGeinneat.u,t8is2m,?7–9. stra
amount of AGRP was quantified. Regulated secretion was c
dweifithnefdorsaksoltihne/IBamMoXunmtionfusAtGonRicPreselecarseetedandaftneorrmstaimlizueldatifoonr 5. EBsyuentraa.pnJtc.iucPr,ceeCdli.la,-taSrd.a,hke1us6ria7oi,n,1Tp0.a9ta1hnw–d1aB1yu0sx1ib.natuhme,pJa.Dth.o(g2e0n0e9si)sTohfeaeumtiesmrgisnpgercotrluemof t/19/7/1
the total amount of AGRP in medium samples plus lysates. disorders.TrendsNeurosci.,32,402–412. 36
Regulatedsecretionintheabsenceofanysilencingoroverex- 6. Bourgeron,T.(2009)Asynaptictrektoautism.Curr.Opin.Neurobiol., 8/2
19,231–234. 9
pression constructs was arbitrarily set at 1, and the other 0
7. Kumar,R.A.andChristian,S.L.(2009)Geneticsofautismspectrum 1
samples were adjusted using the same correction factor. disorders.Curr.Neurol.Neurosci.Rep.,9,188–197. 14
6
8. Jamain,S.,Quach,H.,Betancur,C.,Rastam,M.,Colineaux,C.,Gillberg, b
I.C.,Soderstrom,H.,Giros,B.,Leboyer,M.,Gillberg,C.etal.(2003) y g
Transmission electron microscopy MutationsoftheX-linkedgenesencodingneuroliginsN8LGN3and u
e
Platelet-rich fractions from the patients and controls were pre- 9. LNaLuGmNo4nnaireer,aFs.s,oBcoiantneedt-wBirtihlhaauutlits,mF..,NGaotm.Gote,nMet..,,B3l4a,n2c,7R–.2,9D.avid,A., st on
pared by centrifugation (15min at 150g) of whole blood anti- Moizard,M.P.,Raynaud,M.,Ronce,N.,Lemonnier,E.,Calvas,P.etal. 1
5
coagulatedwith3.8%(w/v)trisodiumcitrate(9:1)asdescribed (2004)X-linkedmentalretardationandautismareassociatedwithamutation A
intheNLGN4gene,amemberoftheneuroliginfamily.Am.J.Hum.Genet., p
i(m92m).eBdioatthelpyefillxetesdoifnb2-.T5C%3gcleultlasraanlddehpyladteeleint-r0ic.1hMplapshmoaspwheartee 10. 7P4a,m5p5a2n–o5s,5A7..,Volaki,K.,Kanavakis,E.,Papandreou,O.,Yourouko8s,S., ril 201
buffer, pH 7.2 at 48C overnight. After 1h post-fixation in 1% Thomaidis,L.,Karkelis,S.,Tzetis,M.andKitsiou-Tzeli,S.(2009)A 9
osmium tetroxide and 0.3% potassium ferricyanide in 0.1M SubstitutionInvolvingtheNLGN4GeneAssociatedwithAutistic
phosphate buffer at 48C, the samples were dehydrated in BehaviorintheGreekPopulation.Genet.Test.Mol.Biomarkers,13,
611–615.
gradedseriesofalcoholandembeddedinepoxyresin.Ultra-thin
11. Zhang,C.,Milunsky,J.M.,Newton,S.,Ko,J.,Zhao,G.,Maher,T.A.,
sectionsof50–60nmwerecut,stainedwithuranylacetateand Tager-Flusberg,H.,Bolliger,M.F.,Carter,A.S.,Boucard,A.A.etal.
lead citrate and examined at 50kV using a Zeiss EM 900 (2009)Aneuroligin-4missensemutationassociatedwithautismimpairs
electron microscope (Oberkochen, Germany). Images were neuroligin-4foldingandendoplasmicreticulumexport.J.Neurosci.,29,
10843–10854.
recorded digitally using a Jenoptik Progress C14 camera
12. Daoud,H.,Bonnet-Brilhault,F.,Vedrine,S.,Demattei,M.V.,Vourc’h,P.,
systemoperatedusingImage-Proexpresssoftware.
Bayou,N.,Andres,C.R.,Barthelemy,C.,Laumonnier,F.andBriault,S.
(2009)AutismandNonsyndromicmentalretardationassociatedwithade
novomutationintheNLGN4XGenepromotercausinganincreased
SUPPLEMENTARY MATERIAL expressionlevel.Biol.Psychiatry,66,906–910.
13. Feng,J.,Schroer,R.,Yan,J.,Song,W.,Yang,C.,Bockholt,A.,Cook,
Supplementary Material is available at HMG online. E.H.Jr,Skinner,C.,Schwartz,C.E.andSommer,S.S.(2006)High
Human Molecular Genetics, 2010, Vol. 19, No. 7 1377
frequencyofneurexin1betasignalpeptidestructuralvariantsinpatients 32. Reed,G.L.,Fitzgerald,M.L.andPolgar,J.(2000)Molecularmechanisms
withautism.Neurosci.Lett.,409,10–13. ofplateletexocytosis:insightsintothe‘secrete’lifeofthrombocytes.
14. Kim,H.G.,Kishikawa,S.,Higgins,A.W.,Seong,I.S.,Donovan,D.J., Blood,96,3334–3342.
Shen,Y.,Lally,E.,Weiss,L.A.,Najm,J.,Kutsche,K.etal.(2008) 33. Persico,A.M.andBourgeron,T.(2006)Searchingforwaysoutofthe
Disruptionofneurexin1associatedwithautismspectrumdisorder. autismmaze:genetic,epigeneticandenvironmentalclues.Trends
Am.J.Hum.Genet.,82,199–207. Neurosci.,29,349–358.
15. Durand,C.M.,Betancur,C.,Boeckers,T.M.,Bockmann,J.,Chaste,P., 34. Chuang,J.Z.,Milner,T.A.,Zhu,M.andSung,C.H.(1999)A29kDa
Fauchereau,F.,Nygren,G.,Rastam,M.,Gillberg,I.C.,Anckarsater,H. intracellularchloridechannelp64H1isassociatedwithlargedense-core
etal.(2007)Mutationsinthegeneencodingthesynapticscaffolding vesiclesinrathippocampalneurons.J.Neurosci.,19,2919–2928.
proteinSHANK3areassociatedwithautismspectrumdisorders.Nat. 35. Strauss,E.andBegley,T.P.(2003)Stereochemicalstudieson
Genet.,39,25–27. phosphopantothenoylcysteinedecarboxylasefromEscherichiacoli.
16. Gauthier,J.,Spiegelman,D.,Piton,A.,Lafreniere,R.G.,Laurent,S., Bioorg.Med.Chem.Lett.,13,339–342.
St-Onge,J.,Lapointe,L.,Hamdan,F.F.,Cossette,P.,Mottron,L.etal. 36. Strauss,E.,Zhai,H.,Brand,L.A.,McLafferty,F.W.andBegley,T.P.
(2009)NoveldenovoSHANK3mutationinautisticpatients.Am.J.Med. (2004)Mechanisticstudiesonphosphopantothenoylcysteine
Genet.B.Neuropsychiatr.Genet.,150B,421–424. decarboxylase:trappingofanenethiolateintermediatewitha D
o
17. Arking,D.E.,Cutler,D.J.,Brune,C.W.,Teslovich,T.M.,West,K.,Ikeda, mechanism-basedinactivatingagent.Biochemistry,43,15520–15533. w
M.,Rea,A.,Guy,M.,Lin,S.,Cook,E.H.etal.(2008)Acommongenetic 37. Fernandez-Chacon,R.andSudhof,T.C.(2000)NovelSCAMPslacking nlo
variantintheneurexinsuperfamilymemberCNTNAP2increasesfamilial NPFrepeats:ubiquitousandsynapticvesicle-specificformsimplicate a
d
riskofautism.Am.J.Hum.Genet.,82,160–164. SCAMPsinmultiplemembrane-traffickingfunctions.J.Neurosci.,20, e
d
18. Alarcon,M.,Abrahams,B.S.,Stone,J.L.,Duvall,J.A.,Perederiy,J.V., 7941–7950. fro
Bomar,J.M.,Sebat,J.,Wigler,M.,Martin,C.L.,Ledbetter,D.H.etal. 38. Liao,H.,Zhang,J.,Shestopal,S.,Szabo,G.,Castle,A.andCastle,D. m
(2008)Linkage,association,andgene-expressionanalysesidentify (2008)Nonredundantfunctionofsecretorycarriermembraneprotein h
CNTNAP2asanautism-susceptibilitygene.Am.J.Hum.Genet.,82, isoformsindensecorevesicleexocytosis.Am.J.Physiol.CellPhysiol., ttp
s
19. 1P5o0ot–,1M59.,.Beyer,V.,Schwaab,I.,Damatova,N.,Van’tSlot,R.,Prothero, 39. 2P9fe4i,feCr,79D7.,–KCi8s0t,9R. .,Dewar,K.,Devon,K.,Lander,E.S.,Birren,B., ://ac
J.,Holder,S.E.andHaaf,T.(2010)DisruptionofCNTNAP2and Korniszewski,L.,Back,E.andScherer,G.(1999)Campomelicdysplasia ad
e
additionalstructuralgenomechangesinaboywithspeechdelayand translocationbreakpointsarescatteredover1MbproximaltoSOX9: m
autismspectrumdisorder.Neurogenetics,11,81–89. evidenceforanextendedcontrolregion.Am.J.Hum.Genet.,65,111– ic
20. Renden,R.,Berwin,B.,Davis,W.,Ann,K.,Chin,C.T.,Kreber,R., 124. .ou
Ganetzky,B.,Martin,T.F.andBroadie,K.(2001)DrosophilaCAPSisan 40. DeBaere,E.,Dixon,M.J.,Small,K.W.,Jabs,E.W.,Leroy,B.P., p.c
essentialgenethatregulatesdense-corevesiclereleaseandsynaptic Devriendt,K.,Gillerot,Y.,Mortier,G.,Meire,F.,VanMaldergem,L. o
m
vesiclefusion.Neuron,31,421–437. etal.(2001)SpectrumofFOXL2genemutationsin /h
21. Sadakata,T.,Washida,M.,Iwayama,Y.,Shoji,S.,Sato,Y.,Ohkura,T., blepharophimosis-ptosis-epicanthusinversus(BPES)families m
g
Katoh-Semba,R.,Nakajima,M.,Sekine,Y.,Tanaka,M.etal.(2007) demonstratesagenotype–phenotypecorrelation.Hum.Mol.Genet.,10, /a
Autistic-likephenotypesinCadps2-knockoutmiceandaberrantCADPS2 1591–1600. rtic
splicinginautisticpatients.J.Clin.Invest.,117,931–943. 41. Castermans,D.,Vermeesch,J.R.,Fryns,J.P.,Steyaert,J.G.,VandeVen, le
22. Sadakata,T.andFuruichi,T.(2009)Developmentallyregulated W.J.,Creemers,J.W.andDevriendt,K.(2007)Identificationand -a
b
Ca2þ-dependentactivatorproteinforsecretion2(CAPS2)isinvolvedin characterizationoftheTRIP8andREEP3genesonchromosome10q21.3 s
BDNFsecretionandisassociatedwithautismsusceptibility.Cerebellum, asnovelcandidategenesforautism.Eur.J.Hum.Genet.,15,422–431. tra
c
8,312–322. 42. Wang,X.,Herberg,F.W.,Laue,M.M.,Wullner,C.,Hu,B., t/1
23. Vorstman,J.A.,Staal,W.G.,vanDaalen,E.,vanEngeland,H., Petrasch-Parwez,E.andKilimann,M.W.(2000)Neurobeachin:Aprotein 9
/7
Hochstenbach,P.F.andFranke,L.(2006)Identificationofnovelautism kinaseA-anchoring,beige/Chediak-higashiproteinhomologimplicatedin /1
candidateregionsthroughanalysisofreportedcytogeneticabnormalities neuronalmembranetraffic.J.Neurosci.,20,8551–8565. 3
6
associatedwithautism.Mol.Psychiatry,11,1,18–28. 43. Su,Y.,Balice-Gordon,R.J.,Hess,D.M.,Landsman,D.S.,Minarcik,J., 8
/2
24. Castermans,D.,Wilquet,V.,Steyaert,J.,VandeVen,W.,Fryns,J.P.and Golden,J.,Hurwitz,I.,Liebhaber,S.A.andCooke,N.E.(2004) 9
0
Devriendt,K.(2004)Chromosomalanomaliesinindividualswithautism: Neurobeachinisessentialforneuromuscularsynaptictransmission. 1
1
astrategytowardstheidentificationofgenesinvolvedinautism.Autism, J.Neurosci.,24,3627–3636. 4
6
8,141–161. 44. Medrihan,L.,Rohlmann,A.,Fairless,R.,Andrae,J.,Doring,M.,Missler, b
25. Castermans,D.,Wilquet,V.,Parthoens,E.,Huysmans,C.,Steyaert,J., M.,Zhang,W.andKilimann,M.W.(2009)Neurobeachin,aprotein y g
Swinnen,L.,Fryns,J.P.,VandeVen,W.andDevriendt,K.(2003)The implicatedinmembraneproteintrafficandautism,isrequiredforthe u
e
neurobeachingeneisdisruptedbyatranslocationinapatientwith formationandfunctioningofcentralsynapses.J.Physiol.,587,5095– s
idiopathicautism.J.Med.Genet.,40,352–356. 5106. t o
n
26. Castermans,D.,Thienpont,B.,Volders,K.,Crepel,A.,Vermeesch,J.R., 45. Scales,S.J.,Hesser,B.A.,Masuda,E.S.andScheller,R.H.(2002) 1
5
Schrander-Stumpel,C.T.,VandeVen,W.J.,Steyaert,J.G.,Creemers, Amisyn,anovelsyntaxin-bindingproteinthatmayregulateSNARE A
J.W.andDevriendt,K.(2008)Positioneffectleadingto complexassembly.J.Biol.Chem.,277,28271–28279. p
haploinsufficiencyinamosaicringchromosome14inaboywithautism. 46. Gerst,J.E.(2003)SNAREregulators:matchmakersandmatchbreakers. ril 2
Eur.J.Hum.Genet.,16,1187–1192. Biochim.Biophys.Acta.,1641,99–110. 0
1
27. APA(1987)DiagnosticandStatisticalManualofMentalDisorders,ed.3 47. Liu,Y.,Krantz,D.E.,Waites,C.andEdwards,R.H.(1999)Membrane 9
Revised(DSM-III-R),AmericanPsychiatricAssociation,Washington. traffickingofneurotransmittertransportersintheregulationofsynaptic
28. Kent,W.J.,Sugnet,C.W.,Furey,T.S.,Roskin,K.M.,Pringle,T.H., transmission.TrendsCellBiol.,9,356–363.
Zahler,A.M.andHaussler,D.(2002)Thehumangenomebrowserat 48. Liu,L.,Liao,H.,Castle,A.,Zhang,J.,Casanova,J.,Szabo,G.andCastle,
UCSC.GenomeRes.,12,996–1006. D.(2005)SCAMP2interactswithArf6andphospholipaseD1andlinks
29. Creemers,J.W.,Pritchard,L.E.,Gyte,A.,LeRouzic,P.,Meulemans,S., theirfunctiontoexocytoticfusionporeformationinPC12cells.Mol.
Wardlaw,S.L.,Zhu,X.,Steiner,D.F.,Davies,N.,Armstrong,D.etal. Biol.Cell,16,4463–4472.
(2006)Agouti-relatedproteinisposttranslationallycleavedbyproprotein 49. Guo,Z.,Liu,L.,Cafiso,D.andCastle,D.(2002)Perturbationofavery
convertase1togenerateagouti-relatedprotein(AGRP)83-132:interaction latestepofregulatedexocytosisbyasecretorycarriermembraneprotein
betweenAGRP83-132andmelanocortinreceptorscannotbeinfluenced (SCAMP2)-derivedpeptide.J.Biol.Chem.,277,35357–35363.
bysyndecan-3.Endocrinology,147,1621–1631. 50. Liu,L.,Guo,Z.,Tieu,Q.,Castle,A.andCastle,D.(2002)Roleof
30. Fukuda,M.(2004)RNAinterference-mediatedsilencingof secretorycarriermembraneproteinSCAMP2ingranuleexocytosis.Mol.
synaptotagminIX,butnotsynaptotagminI,inhibitsdense-corevesicle Biol.Cell,13,4266–4278.
exocytosisinPC12cells.Biochem.J.,380,875–879. 51. Noh,J.Y.,Lee,H.,Song,S.,Kim,N.S.,Im,W.,Kim,M.,Seo,H.,Chung,
31. Thome,M.(2004)CARMA1,BCL-10andMALT1inlymphocyte C.W.,Chang,J.W.,Ferrante,R.J.etal.(2009)SCAMP5links
developmentandactivation.Nat.Rev.Immunol.,4,348–359. endoplasmicreticulumstresstotheaccumulationofexpanded