American Journal of Medical Genetics (Neuropsychiatric Genetics) 114:125± 128 (2002)

Analysis of Ten Candidate Genes in Autism by

Association and Linkage
Anne Philippe,1,2,5 Michel Guilloud-Bataille,4 Maria Martinez,3 Christopher Gillberg,6
Maria RaÊstam,6 Eili Sponheim,7 Mary Coleman,8 Michele Zappella,9 Harald Aschauer,10
Christiane Penet,2 Josue Feingold,4 Alexis Brice,2 and Marion Leboyer,1,5* and the Paris Autism
Research International Sibpair Study
1
INSERM U 513, HoÃpital Henri Mondor, CreÂteil, France
2
INSERM U 289, HoÃpital de la PitieÂ-SalpeÂtrieÁre, Paris, France
3
INSERM U 358, HoÃpital Saint-Louis, Paris, France
4
Laboratoire d'Anthropologie Biologique, Unite de Recherche de GeÂneÂtique EpideÂmiologique, Paris, France
5
Service de Psychiatrie, HoÃpitaux Albert Chenevier et Henri Mondor, AP-HP, CreÂteil, France
6
Department of Child and Adolescent Psychiatry, GoÈteborg University, GoÈteborg, Sweden
7
Centre for Child and Adolescent Psychiatry, University of Oslo, Oslo, Norway
8
Department of Pediatrics, Georgetown University School of Medicine, Washington D.C.
9
Divisione di Neuropsichiatria Infantile, Azienda Ospedaliera Senese, Siena, Italy
10
Department of General Psychiatry, University Hospital, Vienna, Austria

We studied the possible involvement of ten subtilisin/kexin type 2 (opioid metabolism);

candidate genes in autism: proenkephalin,
prodynorphin, and proprotein convertase
Paris Autism Research International Sibpair Study: Sweden.
Department of Child and Adolescent Psychiatry, GoÈteborg
University, GoÈteborg: Christopher Gillberg, Maria RaÊstam,
Carina Gillberg, Agneta NydeÂn. France. Consultation speÂcialiseÂe
pour l'autisme infantile, HoÃpital Robert DebreÂ, Paris: Marion
Leboyer, Manuel Bouvard, Anne Philippe, Nadia Chabane and
Marie-Christine Mouren-Simeoni; Service universitaire d'Explo-
rations Fonctionnelles et Neurophysiologie en PeÂdopsychiatrie,
HoÃpital Bretonneau, Tours: Catherine BartheÂlemy. Norway.
National Centre for Child and Adolescent Psychiatry, University
of Oslo, Oslo: Eili Sponheim, Ingrid Spurkland; Department of
Pediatrics, Rikshospitalet, University of Oslo, Oslo: Ola H.
Skjeldal. USA. Department of Pediatrics, Georgetown University
School of Medicine, Washington D.C.: Mary Coleman; Children's
National Medical Center, George Washington University School
of Medicine, Washington, D.C.: Philip L. Pearl; New York State
Institute for Basic Research in Developmental Disabilities, Staten
Island, New York: Ira L. Cohen, John Tsiouris. Italy. Divisione di
Neuropsichiatria Infantile, Azienda Ospedaliera Senese, Siena:
Michele Zappella, Grazia Menchetti, Alfonso Pompella. Austria.
Department of General Psychiatry, University Hospital, Vienna:
Harald Aschauer.
Grant sponsor: Assistance Publique-Hopitaux-de-Paris; Grant
number: PHRC AOM95076 and CRC Number 932413;
Grant sponsor: Association FrancËaise Contre les Myopathies;
Grant sponsor: France Telecom; Grant sponsor: Fondation Lilly;
Grant sponsor: Swedish Medical Research Council; Grant num-
ber: K97-21X-11251-03CK.
*Correspondence to: Marion Leboyer, INSERM U513, HoÃpital
Henri Mondor, 8, rue du GeÂneÂral Sarrail, 94010 CreÂteil Cedex,
France. E-mail: Marion.Leboyer@im3.inserm.fr
Received 14 June 2000; Accepted 2 August 2001
tyrosine hydroxylase, dopamine receptors
D2 and D5, monoamine oxidases A and B
(monoaminergic system); brain-derived neu-
rotrophic factor, and neural cell adhesion
molecule (involved in neurodevelopment).
Thirty-eight families with two affected sib-
lings and one family with two affected half-
siblings, recruited by the Paris Autism
Research International Sibpair Study
(PARIS), were tested using the transmission
disequilibrium test and two-point affected
sib-pair linkage analysis. We found no evi-
dence for association or linkage with intra-
genic or linked markers. Our family sample
has good power for detecting a linkage
disequilibrium of 0.80. Thus, these genes
are unlikely to play a major role in the
families studied, but further studies in a
much larger sample would be needed to
highlight weaker genetic effects.
ß 2002 Wiley-Liss, Inc.
KEY WORDS: autism; linkage disequili-
brium; linkage; sib-pair
DATABASES: PENK (GDB no. 156549);
PDYN (GDB no. 376650);
PCSK2 (GDB no. 215822);
TH (GDB no. 225009); DRD2
(GDB no. 270166); BDNF
(GDB no. 162545); NCAM
(GDB no. 377827); MAOA
(GDB no. 156455); MADB
(GDB no. 155726)

ß 2002 Wiley-Liss, Inc.

DOI 10.1002/ajmg.10041
126 Philippe et al.
INTRODUCTION
Sibling recurrence risk and twin studies have
demonstrated the existence of a genetic component in
the etiology of autism (MIM No. 209850), the mode of
inheritance of which is unknown [Smalley, 1997]. Many
hypotheses have been proposed to account for the
symptoms of autism. One is that autism results from
excessive brain opioid activity during the neonatal
period [Panksepp, 1979]. We have demonstrated that
there are large differences between plasma C- and N-
terminally directed b-endorphin immunoreactivities in
autistic children [Leboyer et al., 1994].
We explored this further by studying three candidate
genes involved in opioid metabolism: proenkephalin
(PENK), prodynorphin (PDYN), and proprotein con-
vertase subtilisin/kexin type 2 (PCSK2). Individuals
with autism display biochemical abnormalities in
monoamines. This led us to test genes involved in
dopaminergic metabolism as well [Martineau et al.,
1992]. A positive association between the Taq I RLFP
polymorphism of the dopamine D2 receptor and autism
has been reported [Comings et al., 1991]. The tetra-
nucleotide repeat polymorphism in intron 1 and the Bgl
II polymorphism of the tyrosine hydroxylase gene have
not been shown to be associated with autism [Comings
et al., 1991; HeÂrault et al., 1993], but these studies used
a case-control design. We assessed the following
candidate genes involved in monoaminergic transmis-
sion: dopamine receptors D2 (DRD2) and D5 (DRD5),
tyrosine hydroxylase (TH), and monoamine oxidases A
(MAOA) and B (MAOB).
Finally, autistic patients display morphological
changes in the brain and low serum levels of neural
cell adhesion molecule (NCAM) [Plioplys et al., 1990].
We therefore studied two genes involved in nervous
system development, those encoding brain-derived
neurotrophic factor (BDNF) and NCAM. We investi-
gated possible association and/or linkage between these
candidate genes and autism using the transmission
disequilibrium test and two-point affected sib-pair
analysis.
MATERIALS AND METHODS
Families
Families with at least two siblings or half-siblings
ful®lling the DSM IV criteria for autistic disorder
[American Psychiatric Association, 1994] and the
Autism Diagnostic Interview algorithm for ICD-10
childhood autism [Lord et al., 1994] were recruited for
a collaborative autism sib-pair project (PARIS). The
families included in this study were recruited before
1996. As part of this project, a genome-wide scan of
these families (plus additional families recruited in
1997) was carried out and has been reported [Philippe
et al., 1999]. Subjects were included after physical and
neurological examination, neuropsychological testing,
standard karyotyping and fragile-X testing, brain
imaging, and blood and urine analysis. Subjects with
known medical conditions associated with autism were
excluded. The study was approved by the ethical
committees of the participating organizations. Inform-
ed consent forms were completed by the parents.
Genotyping
Blood samples were collected from children and their
parents and DNA was extracted from lymphocytes by
standard procedures. All polymorphisms and primer
sequences are available on-line from the Genome
Database (GDB: http://gdbwww.gdb.org/gdb/). The
intragenic microsatellites for PENK (GDB accession
no: 156549), PDYN (GDB accession no: 376650), PCSK2
(GDB accession no: 215822), TH (GDB accession no:
225009), DRD2 (GDB accession no: 156179), DRD5
(GDB accession no: 270166), BDNF (GDB accession no:
162545), NCAM (GDB accession no: 377827), MAOA
(GDB accession no: 156455), and MAOB (GDB acces-
sion no: 155726) were ampli®ed by PCR using standard
procedures. Reaction conditions were varied according
to the primers used. PCR products were separated by
electrophoresis in denaturing 6% polyacrylamide gels,
capillary-blotted onto charged nylon membranes,
probed with a primer 50 -labeled with (g32P)-dATP and
detected by autoradiography. For each candidate gene,
we genotyped neighboring microsatellite markers:
D4S412, D4S403 for DRD5; D8S532, D8S285,
D8S166, D8S1763 for PENK; D11S992 for TH;
D11S904, D11S935 for BDNF; D11S898, D11S4142
for DRD2 and NCAM; D20S113 for PDYN; D20S189,
D20S101 for PCKS2; DXS1202, DXS1068, DXS1003 for
MAOA, and MAOB. They were ampli®ed using ¯uor-
escent end-labeled primers and analyzed as previously
described [Philippe et al., 1999].
Statistical Methods
Data were analyzed by linkage disequilibrium and
linkage tests, for a total of 38 families with two affected
siblings for autosomal markers and in 39 families for X
chromosome markers.
To test for linkage disequilibrium, we used the
likelihood-based approach of the multiallelic transmis-
sion/disequilibrium test (MTDT), developed by Terwil-
liger [Terwilliger, 1995], as implemented in the
TDTlike program of the Analyze package [Spielman
and Ewens, 1996]. At each locus, if the sum of
transmitted and non-transmitted alleles was less than
®ve, these alleles were eliminated. MTDT signi®cance
is based on a global test of all alleles. In multiplex
families, the MTDT is valid as a test of association only
if transmissions to a single affected individual per
family are counted [Spielman and Ewens, 1996]
because of the lack of independence between siblings.
Therefore, association analysis with intragenic poly-
morphisms for the candidate genes tested was con-
ducted by selecting, at random, one affected child per
family. Only families in which both parents had been
genotyped were included in the association study for
autosomal markers.
Two-point affected sib-pair linkage analysis was
performed for polymorphisms intragenic and adjacent
to candidate genes using the SIBPAIR program
[Terwilliger, 1996], which provides a likelihood-based
 Analysis of Ten Candidate Genes in Autism 127

test statistic for linkage. The likelihood over all families TABLE II. Maximum Lod Score (MLS) Values for Two-Point
is maximized as a function of the frequency of marker Affected Sib-Pair Analysis
alleles identical by descent (y) among affected sibs. The Two-point analysis
likelihood ratio test statistic, T, is calculated, and used Recombination
to test the null hypothesis of y ˆ 0.5 (T ˆ 2Ln[L(y)/ Locus fractiona MLS P value
L(y ˆ 0.5)]). This statistic follows a chi-square distribu-
tion with one degree of freedom and can thus be D4S412 39.7% 0 0.500
DRD5 7.4% 0 0.500
expressed as a lod score, MLS ˆ T/2ln(10). D4S403 0 0.500
D8S532 20.4% 0 0.500
RESULTS D8S285 10.8% 0 0.500
Thirty-eight families with two affected siblings and PENK 3.2% 0.120 0.224
one family with two affected half-siblings were ana- D8S166 2.5% 0 0.500
D8S1763 0.010 0.460
lyzed. Full genotype information was available for all
parents except three fathers. All families were Cauca- TH 4.3% 0.210 0.164
sian; 13 came from France, 12 from Sweden, ®ve from D11S922 0.190 0.170
Norway, four from the USA, three from Italy, and two D11S904 8.0%
from Austria. The mean age of the probands was 16.5 BDNF 9.0% 0.090 0.420
years (range: ®ve to 46) and the sex ratio was 3.1 (59 D11S935 0 0.500
males and 19 females). None of the candidate gene D11S898 19.1% 0.020 0.370
markers was signi®cantly associated with autism, DRD2 3.3% 0 0.500
according to MTDT (Table I). Two-point affected sib- NCAM 6.7% 0.009 0.420
pair analyses provided no evidence for linkage between D11S4142 0 0.450
autism and markers located within or close to the tested PDYN 0.01% 0.040 0.330
genes (Table II). For MAOA and MAOB, there was no D20S113 0.220 0.160
difference in the distribution of alleles identical by D20S189 8.6% 0 0.500
descent between male-male, female-male, and female- PCKS2 11.1% 0 0.500
female sib-pairs (data not shown). D20S101 0 0.500
DX1202 17.6% 0.410 0.084
DISCUSSION MAO B 5.5% 0.500 0.064
MAO A 0% 0 0.500
We investigated the possible involvement of ten DXS1068 14.9% 0.090 0.260
candidate genes in autism by multiallelic transmis- DXS1003 0.001 0.470
sion/disequilibrium tests and two-point affected sib- a
Recombination fractions were estimated with respect to the next marker.
pair analysis in 39 families including 78 affected Allele frequencies were estimated from the data using the ILINK program
individuals. None of the markers of candidate genes from the Linkage software package [Lathrop and Lalouel, 1984].
we examined showed statistical signi®cance for asso-
ciation or linkage in this sample.
Our results con®rm the absence of an association the study of Comings et al. [1995] is a classical case-
between the tetranucleotide repeat polymorphism in control study with a high risk of false-positive associa-
intron 1 of the TH gene and autism, as previously tion due to population strati®cation; we controlled this
reported in a case-control study [HeÂrault et al., 1993]. risk by using internals controls (TDT). We also found no
Unlike Comings et al. [1995], we found no evidence of evidence of linkage with intragenic markers of DRD2
association between DRD2 and autism. However, our but this is not inconsistent with the association
results are not strictly comparable with those of reported by Comings et al. [1995].
Comings et al. [1995] because the initial polymorph- Our sample has good statistical power for detecting a
isms tested differ between the two studies. In addition, linkage disequilibrium of 0.8, suggesting that the
polymorphisms of the candidate genes tested are not
TABLE I. Multiallelic Transmitted Disequilibrium Test for the consistent with a strong etiologic role in autism.
Markers Located Within the Candidate Genes Nevertheless, our power to detect associations of
weaker effect was limited by the available sample size
Total Number
number of alleles and additional large-scale studies are required to rule
Locus Location of alleles tested P value out the involvement of these genes in autism.
DRD5 4p15-p15.3 10 7 0.487
PENK 8q11.23-q12 4 2 0.393 ACKNOWLEDGMENTS
TH 11p15.5-p15 5 5 0.405
BDNF 11p11.2-p13 4 4 0.152 We thank the subjects and their families for their
DRD2 11q23.1 4 4 0.500 cooperation. We thank Armelle Faure and Nathalie
NCAM 11q23.1 10 7 0.245 CheÂron for technical help in the GeÂneÂthon lab and
PDYN 20pter.p12 5 3 0.168 Jacky Bou, Yolaine Pothin, and AgneÁs Camuzat for
PCKS2 20p11.1-11.2 6 5 0.500 technical assistance in the INSERM U289 lab. The
MAOA Xp11.23 7 6 0.464
work pro®ted by a grant from the Fondation pour la
MAOB Xp11.23 6 5 0.459
Recherche MeÂdicale (A.P.).
128 Philippe et al.
REFERENCES
American Psychiatric Association. 1994. Diagnostic and statistical manual
of mental disorders, 4th ed. Washington, DC.
Comings DE, Comings BG, Muhleman D, Dietz G, Shahbahrami B, Tast D,
Knell E, Baumgarten R, Kovacs B, Levy DL, Smith M, Borison R, Evans
DD, Klein D, MacMurray J, Tosk JM, Sverd J, Gysin R, Flanagan D.
1991. The dopamine D2 receptor locus as a modifying gene in
neuropsychiatric disorders. JAMA 266:1793±1800.
Comings DE, Gade R, Muhlemen D, Sverd J. 1995. No association of a
tyrosine hydroxylase gene tetranucleotide repeat polymorphism in
autism, Tourette syndrome, or ADHD. Biol Psychiatry 37:484±486.
HeÂrault J, Perrot A, BartheÂleÂmy C, BuÈchler M, Cherpi C, Leboyer M,
Sauvage D, Lelord M, Mallet J, MuÈh JP. 1993. Possible association of C-
harvey-ras-1 (HRAS-1) marker with autism. Psychiatry Res 46:261±
267.
Lathrop GM, Lalouel JM. 1984. Easy calculations of LOD scores and
genetic risks on small computers. Am J Hum Genet 36:460±465.
Leboyer M, Bouvard MP, Recasens C, Philippe A, Guilloud-Bataille M,
Bondoux D, Tabuteau F, Dugas M, Panksepp J, Launay JM. 1994.
Difference between plasma N- and C-terminally directed b-endorphin
immunoreactivity in infantile autism. Am J Psychiatry 151:1797±1801.
Lord C, Rutter M, Le Couteur A. 1994. Autism diagnostic interview-
revised: a revised version of a diagnostic interview for caregivers of
individuals with possible pervasive developmental disorders. J Autism
Dev Disord 24:659±685.
Martineau J, BartheÂleÂmy C, Jouve J, Muh JP, Lelord G. 1992. Monoamines
(serotonin and catecholamines) and their derivatives in infantile
autism: age-related changes and drug effects. Dev Med Child Neurol
34:593±603.
Panksepp J. 1979. A neurochemical theory of autism. Trends Neurosci
2:174±177.
Philippe A, Martinez M, Guilloud-Bataille M, Gillberg C, RaÊstam M,
Sponheim E, Coleman M, Zappella M, Aschauer H, van Maldergerm L,
Penet C, Feingold J, Brice A, Leboyer M, Paris Autism Research
International Sibpair Study. 1999. Genome-wide scan for autism
susceptibility genes. Hum Mol Genet 8:805±812.
Plioplys AV, Hemmens SE, Regan CM. 1990. Expression of a neural cell
adhesion molecule serum fragment is depressed in autism. J Neurop-
sychiatry Clin Neurosci 2:413±417.
Smalley SL. 1997. Behavioral genetics '97. Genetic in¯uences in childhood-
onset psychiatric disorders: autism and attention-de®cit/hyperactivity
disorder. Am J Hum Genet 60:1276±1282.
Spielman RS, Ewens WJ. 1996. The TDT and other family-based tests
for linkage disequilibrium and association. Am J Hum Genet 59:983±
989.
Terwilliger J. 1995. A powerful likelihood method for the analysis of
linkage disequilibrium between trait loci and one or more polymorphic
marker loci. Am J Hum Genet 56:777±787.
Terwilliger J. 1996. Program SIBPAIR: sibpair analysis on nuclear
families. ftp://linkage.cpmc.columbia.edu.