Genetics of Neurodevelopmental Disorders AP Monaco
Genetics of Neurodevelopmental
Disorders
AP Monaco
Neurodevelopmental Disorders
• Autism
• Specific Language Impairment (SLI)
• Developmental Dyslexia
• Attention Deficit Hyperactivity Disorder
Summary of Talk
• Why we think autism has genetic origins.
• Strategies to identify genes involved in autism.
• What have we discovered thus far.
• The Autism Genome Project.
• The search for further autism susceptibility genes.
AUTISM: a severe neurodevelopmental disorder
‘…… an innate inability to form the usual biologically provided affective contact with people.’ Leo Kanner, 1943.Leo Kanner, 1943.
Characterised by impairments in three principal areas:• Verbal and non-verbal communication• Reciprocal social interaction• Repetitive and stereotyped patterns of interests and
behaviours
Onset apparent before 3 years of age and persists throughout life
Population prevalence of autism is ~17-61 per 10,000.Male to female ratio of ~4:1
Evidence for a genetic origin
• Large number of chromosome abnormalities associated with autism.
• Familial clustering of autism is well above the normal population prevalence.
• Twin based studies.
Twin Studies
• Identical twins share 100% of their genes.
• Non-identical twins share on average 50% of their genes.
mother father
child 1 child 2
Identical twins
child 3 child 4
mother father
Non-Identical twins
Twin StudiesIdentical twins Non-Identical twins
? ?
60% concordance 0-3% concordanceaffected
unaffected
GENETIC FACTORS IN AUTISM
The Wellcome Trust Centre for Human Genetics.
•Evidence from twin studies suggests a monozygotic (MZ) to dizygotic (DZ) concordance rate of 60%:0% (Bailey et al., 1995)
•Heritability estimates of >90%
•The rate among siblings of an autistic proband is ~3%
•Autism is one of the most strongly genetic of the childhood-onset psychiatric disorders but no known mode of inheritance
•Statistical modelling suggests the epistatic interaction of 3 genes, but possibly up to 10 loci
•Family and twin studies indicate evidence for a broader autistic phenotype including combinations of milder but related social and/or communicative abnormalities in people of normal intelligence
•MZ:DZ twin concordance rate for the broader phenotype ~92%:10%
Finding autism susceptibility genes:
I. Chromosome abnormalities
II. Linkage Studies
III. Association Studies
The Human Genome
a human
a human cell 23 pairs of chromosomes
23 from mother
23 from father
23 pairs of chromosomes
• Principle: To identify an individual (or several members of a family) with a chromosomal translocation and autism spectrum disorder.
Finding autism susceptibility genes:I. Translocations
Potential site for physical break in gene or regulatory element
• Principle: To identify an individual (or several members of a family) with a chromosomal translocation and reading impairment.
Finding dyslexia susceptibility genes:I. Translocations
23 pairs of chromosomes
Finding autism susceptibility genes:
• Principle: To identify regions of the genome inherited more often than by chance in individuals with autism.
child 1 child 2
mother father
Family 1
mother
child 1 child 2
father
Family 2
child 1 child 2
mother father
Family 3
Linked to autism
.... Family 300
II. Linkage Studies
Finding autism susceptibility genes:
• Principle: To identify a genetic variant that is disproportionately more frequent in individuals with autism.
III. Association Studies
Variant Freq. (%) in controls
Freq. (%) in autistic cases
1 25 24
2 11 13
3 28 27
4 17 34
5 11 12
Examples of genetic variants: - microsatellites
- single nucleotide polymorphisms (SNPs)
- insertions and deletions (INDELS)
- combinations of the above (haplotypes)
Finding dyslexia susceptibility genes:
• Principle: To identify a genetic variant that is disproportionately more frequent in individuals with autism.
Associated to autism
III. Association Studies
Variant Freq. (%) in controls
Freq. (%) in autistic cases
1 25 24
2 11 13
3 28 27
4 17 34
5 11 12
Examples of genetic variants: - microsatellites
- single nucleotide polymorphisms (SNPs)
- insertions and deletions (INDELS)
- combinations of the above (haplotypes)
The Wellcome Trust Centre for Human Genetics.
THE INTERNATIONAL MOLECULAR GENETIC STUDY OF AUTISM CONSORTIUM (IMGSAC)
Initial aims:
Identify 200 multiplex families with two or more individuals with autism
Clear inclusion (ADI and ADOS) and exclusion criteria
Carry out a genome screen for autism susceptibility loci.
340 families now identified through clinics internationally U.K, Italy, Denmark, France, Greece, Germany, The Netherlands, U.S.A.
Autism Pathway
A
B
C
D
E
F
Patient1
A*
C*
D*
Patient2
B*
C*
E*
F*
Patient3
A*
C*
E*
Patient4
B*
C*
E*
Normal
B*
The Wellcome Trust Centre for Human Genetics.
GENOME SCREEN RESULTS
Chromosome
1 2 4 6 8 10 12 15 18 223 5 7 9 11 13 14 16 17 19 20 21 X
0
0.5
1
1.5
2
2.5
3
3.5
AS
PEX
MLS
152 ASP
219 ASP
MLS > 3.6MLS > 2.2MLS > 1
The Wellcome Trust Centre for Human Genetics.
AUTISM GENOME SCREEN RESULTS8 genome screens for autism susceptibility loci carried out to date
Genetic heterogeneity between studies?Some encouraging convergence of linkage findings eg chr 2, 7
The Wellcome Trust Centre for Human Genetics.
POSITIONAL CANDIDATE GENE SCREENING
•Common Disease, Common Variant model-High Resolution SNP/haplotype association mapping-Regional or Whole Genome Association studies
Strategy depends on the model for autism susceptibility
•Multiple rare variants model -Exhaustive Candidate gene screening-No clear aetiological mutations involved in the majority of families -Rare mutations identified: NLGN, NRXN, SHANK3
Chromosome
1 2 4 6 8 10 12 15 18 223 5 7 9 11 13 14 16 17 19 20 21 X
0
0.5
1
1.5
2
2.5
3
3.5
AS
PEX
MLS
Chromosome 2, 219 ASP
0
0.5
1
1.5
2
2.5
3
0 50 100 150 200 250 300
Position (cM)
AS
PE
X M
LS
The Wellcome Trust Centre for Human Genetics.
REGIONS FOR ILLUMINA SNP GENOTYPING
Chromosome 7, 219 ASP
0
0.5
1
1.5
2
2.5
3
0 50 100 150 200 250
Position (cM)A
SP
EX
MLS
International HapMap Project• Genotyped individuals from four populations:
– Yoruba (Nigerian)– Japanese– Han Chinese– Utah (US) = CEPH samples:
• Northern and western European ancestry.
• PHASE I to genotype 1,000,000 SNPs
• PHASE II to genotype 5,000,000 more SNPs
Summary Strategy Overview• Download CEPH HapMap genotypes.
• Process genotypes (filter & clean).
Summary Strategy Overview• Download CEPH HapMap genotypes.
• Process genotypes (filter & clean).
• Create haplotype blocks:– Regions of LOW recombinations.
Haploview
Daly Lab at the Whitehead Institute
LOW
recombination
HIGH
recombination
Summary Strategy Overview• Download CEPH HapMap genotypes.
• Process genotypes (filter & clean).
• Create haplotype blocks.
• Identify haplotype-tagging SNPs (htSNPs).
1 2 3 4 5 6 7 8 9 Freq
A T A A G C T A G 0.75C C A G A T C A G 0.10C C C A A T T G A 0.08C C A A A T T A G 0.06
1 2 3 4 5 6 7 8 9^ ^ ^A T A A G C T A GC C A G A T C A GC C C A A T T G AC C A A A T T A G
SNP number
1234
Haplotype number
Summary Strategy Overview• Download CEPH HapMap genotypes.
• Process genotypes (filter & clean).
• Create haplotype blocks.
• Identify haplotype-tagging SNPs (htSNPs).
• Align haplotype blocks with genes.
Important Points: Gene not covered by haplotype block, so can not be tested.
Haplotype block not covering any gene, therefore less need
to genotype htSNPs.
Segments of genes not covered by
haplotype block, so can not be tested.
Summary Strategy Overview• Download CEPH HapMap genotypes.
• Process genotypes (filter & clean).
• Create haplotype blocks.
• Identify haplotype-tagging SNPs (htSNPs).
• Align haplotype blocks with genes.
• Genotype 1536 htSNPs on the Illumina platform, utilising the GoldenGate assay, in the selected samples (cases and controls).
G/G A/C A/T G/C T/T A/CT/CG/CT/T
Summary Strategy Overview• Download CEPH HapMap genotypes.
• Process genotypes (filter & clean).
• Create haplotype blocks.
• Identify haplotype-tagging SNPs (htSNPs).
• Align haplotype blocks with genes.
• Genotype 1536 htSNPs on the Illumina platform, utilising the GoldenGate assay, in the selected samples (cases and controls).
• Reconstruct haplotypes based on the htSNPs’ genotypes.
G/G A/C A/T G/C T/T A/CT/CG/CT/T
TGC 0.80
TCG 0.15
AGC 0.05
AT 0.60
AA 0.25
CT 0.15
AT 0.60
AA 0.25
CT 0.15
AT 0.60
AA 0.25
CT 0.15
Summary Strategy Overview• Download CEPH HapMap genotypes.
• Process genotypes (filter & clean).
• Create haplotype blocks.
• Identify haplotype-tagging SNPs (htSNPs).
• Align haplotype blocks with genes.
• Genotype 1536 htSNPs on the Illumina platform, utilising the GoldenGate assay, in the selected samples (cases and controls).
• Reconstruct haplotypes based on the htSNPs’ genotypes.
• Test haplotypes for association to autism.
G/G A/C A/T G/C T/T A/CT/CG/CT/T
association p-values0.021 0.7480.451 0.001
TGC 0.80
TCG 0.15
AGC 0.05
AT 0.60
AA 0.25
CT 0.15
AT 0.60
AA 0.25
CT 0.15
AT 0.60
AA 0.25
CT 0.15
Summary Strategy Overview• Download CEPH HapMap genotypes.
• Process genotypes (filter & clean).
• Create haplotype blocks.
• Identify haplotype-tagging SNPs (htSNPs).
• Align haplotype blocks with genes.
• Genotype 1536 htSNPs on the Illumina platform, utilising the GoldenGate assay, in the selected samples (cases and controls).
• Reconstruct haplotypes based on the htSNPs’ genotypes.
• Test haplotypes for association to autism.
• Attempt to replicate any significant resultsin INDEPENDENT samples.
G/G A/C A/T G/C T/T A/CT/CG/CT/T
TGC 0.80
TCG 0.15
AGC 0.05
AT 0.60
AA 0.25
CT 0.15
AT 0.60
AA 0.25
CT 0.15
AT 0.60
AA 0.25
CT 0.15
association p-values0.021 0.7480.451 0.001
Autism Candidate Gene
0.847 0.002
IMMP2L
PTPRZ1/2 NM015328
Chromosome 7
Strong evidence for association (90%)
Positive evidence for association (75%)
Probands v/s unrelated controls (population-based analysis)Experiment-wise posterior probability of association
Family Based Analysis
Probands v/s internal controlsExperiment wise posterior probability of association
Strong evidence for association 90%Positive evidence for association 75%
CUTL1
MUC3A/B
LHFPL3 IMMP2L
WNT16
Summary of molecular genetics research
The Wellcome Trust Centre for Human Genetics.
• Encouraging convergent evidence for linkage in some regions.
• Many candidate gene screening and association studies have been carried out, but no etiological variants have been so far conclusively identified in the majority of families with autism.
• High density SNP genotyping for association to autism on chromosome 7q and 2q have identified several positive risk haplotypes that need to be replicated.
• Several groups have carried out genome screens for linkage to autistic disorder.
Autism Genome Project
• Genome-wide SNP genotypes (10k, 1M)
• Copy Number Variation (CNV)
• Linkage analyses
• Association analyses
• Trait subsets, endophenotypes, QTLs
Autism Genome Project• Consortium of consortia
• Autism Genetics Cooperative (AGC)• Autism Genetics Research Exchange (AGRE)• Collaborative Programs of Excellence in Autism (CPEA)• International Molecular Genetic Studies of Autism
Consortium (IMGSAC)
• Pool autism family samples, phenotype data and expertise
• Phase I• Genome-wide genotyping (10k)• Initial linkage analysis
• Phase II• Association and Copy Number Variants
Autism Genome Project: Phase I
• ~1400 multiplex families• Affymetrix 10k SNP genotypes• Basic linkage analysis
• “autism” = affected (categorical Dx)• Narrow and broad autism and “heterogeneous
ASDs”
• Copy Number Variation• Use signal intensity data from adjacent/contiguous
SNPs to infer copy number gains or losses
• Linkage analyses in data subsets• Sex: Male-only, Female-containing• Ancestry: Western European• CNV: Removing CNV by different algorithms
Autism Genome ProjectCopy Number Variation and
Genetic Linkage
Copy Number Variants
1 2
2p16 deletion:2 affected siblings
NRX1
1q21 dups/dels:three families
Previously implicated in MR
17
17p12 dups:three families SMS, CMT
22
22q11.2 deletions:two families
Interpretation complicated
17 de novo CNVs (10 found in both ASP) 18 CNVs overlap ASD-related
rearrangements Numerous overlapping/recurrent CNVs Families with transmission of maternal 15q
gains
Copy Number Variation and Autism Risk:De Novo Mutation vs. Inherited Risk
CNV in Simplex and Multiplex Autism Families
Sebat et al, Science 2007
Are CNVs more frequent in simplex families?
Sebat et al, Science 2007
Are autism associated CNVs more likely to be sporadic “mutations” or inherited risk factors?
AGP Phase IIDesign
Summary
• Autism has a complex genetic etiology• CNV is a major class of autism risk and causation• Need to clarify what is disease-related vs rare polymorphism• New information from genetic linkage studies• 11p and other regions targeted for candidate gene analysis• Whole genome association (WGA) studies will (hopefully)
identify common liability alleles– Common = common in the general population
• WGA does not identify genes with rare or infrequent risk or causative alleles– Sequence (autism genomes?)
• Higher resolution CNV analysis• Collaboration is important to make progress in such complex
disorders
AcknowledgmentsNuala SykesInes SousaAlistair PagnamentaRichard HoltKirsty Wing Gaby BarnbyPenny Farrar Elena BonoraTom Scerri Elena MaestriniAndrew MorrisJanine LambTony Monaco Tony Bailey
FundingMRC, Wellcome Trust, NLM Family Foundation, Simons Foundation, EU
The Wellcome Trust Centre for Human Genetics.
International Molecular Genetic Study of Autism Consortium (IMGSAC)
http://www.well.ox.ac.uk/~maestrin/iat.html
The Wellcome Trust Centre for Human Genetics.
• U.K.• Dept of Psychiatry, University of Oxford• Wellcome Trust Centre, Univ of Oxford• Institute of Psychiatry, London• University of Cambridge, Dept of
Psychiatry • Guy’s Hospital, London• University of Newcastle, School of
Clinical Medical Sciences• University of Manchester, School of
Epidemiology & Health Science• ECACC, Porton Down
• Denmark, Centre for Autisme, Bagsvaerd• France, Hôpital La Grave, Toulouse• Germany, Molecular Genome Analysis,
HeidelbergDept of Psychiatry, Frankfurt
• Greece, Agia Sophia Children’s Hospital, Athens
• Italy, University of Bologna• The Netherlands, Dept of Psychiatry,
Utrecht• U.S.A
University of Chicago, Dept of PsychiatryUCLA Center for Neurobehavioural Genetics, Psychiatric InstituteYale UniversityUniversity of Pittsburgh, Dept of Human Genetics and BiostatisticsUniversity of Michigan, Autism & Communicative Disorders Center
Genome-Wide Analysis of Linkage and Copy Number Variation
Autism Genome Project Consortium:Autism Genetics Collaborative (AGC)
Autism Genetics Resource Exchange (AGRE)Collaborative Programs of Excellence in Autism (CPEA)
International Molecular Genetics Study of Autism Consortium (IMGSAC)
Funding:UK: Medical Research Council
USA: Autism SpeaksIreland:HRB
Canada: Genome Canada
Autism Genome ProjectAutism Genome Project