Identifying Autism Genes in Large Multiplex Families

Background: Autism spectrum disorders (ASD) are a group of neurodevelopmental disorders characterized by impairments in language, social interaction and repetitive or stereotyped behaviours. ASD has a strong genetic basis with heritability estimated at approximately 90% and up to 20% of relatives of individuals with ASD displaying the “broader autism phenotype” (BAP).  Underlying genetic aetiology is heterogeneous, with the majority of cases remaining unsolved.  Novel methodologies are required to disentangle the complex genetic architecture of these disorders.

Objectives: To clinically phenotype large families with multiple affected individuals with ASD and the BAP, and to map these families to identify novel ASD genes of major dominant effect via linkage analysis.

Methods: We phenotyped participants from large multiplex families using multiple standardized measures and a semi-structured interview including specific tasks to evaluate the BAP.  Data was gathered regarding developmental, medical, psychiatric, vocational, and educational history, as well as hobbies and interpersonal relationships.  Participants were formally evaluated for cognition, language skills, and executive function.  A BAP rating system was developed to quantify specific BAP traits.  Hierarchical cluster analysis of these BAP scores was used to identify phenotypic subgroups.  Phenotypic patterns were scrutinized to determine affected status. Genotyping was performed and parametric linkage analysis conducted.

Results: We phenotyped 64 individuals from two large families and mapped ASD traits in both pedigrees. Five endophenotypic clusters of BAP traits were identified via hierarchical cluster analysis.  Phenotypic patterns varied between families, with Family A demonstrating more individuals with disordered pragmatic language as their strongest feature, while aberrant social function was the most common endophenotype in Family B.  In Family A we identified an 8Mb region suggestive of linkage at chr7q21.11-7q21.3 with a parametric LOD score of 2.76.  Haplotype analysis demonstrated segregation with 15/17 affected individuals.  In Family B we obtained a genome wide significant maximum LOD score of 3.3 for a 0.5Mb region at chr17p13.3. The critical haplotype was identified in 14/15 affected individuals.

Conclusions: We mapped ASD and BAP traits in two large multiplex families by employing novel methodology to identify the BAP. Phenotypes varied between the two families but phenotypic patterns showed intra-familial similarities, in keeping with the recognised phenotypic heterogeneity within ASD.  Linkage data indicates a high probability that a causative mutation lies within the linkage region for each family.  Next generation sequencing will be used to identify these variants. Gene identification will inform the neurobiology of ASD and could potentially lead to novel treatments.