Molecular Investigation of An Autism Risk Association Region on Chromosome 5p14.1

Background: Autism is a highly heritable neuropsychiatric disorder.  To date, however, no single causative gene has been identified through linkage or association studies, suggesting a complex genetic etiology.  We have recently identified and validated a significant association in a novel region on chromosome 5 p14.1 (see Ma et al., this meeting).  Though this region lacks any known genes or regulatory sequences, the clustering of association signals suggests that one or more nearby functional variants is responsible for the association. 

Objectives: To use molecular techniques to examine the candidate region and determine the genetic variant on chromosome 5p14.1 responsible for the association peak.

Methods: We used a three pronged molecular approach to examine the candidate region.  First, evolutionarily conserved regions in the peak area of association, as well as the exons of the nearest genes, CDH9 and CDH10, were sequenced in 100 autism cases and 100 controls.  Secondly, CGH arrays were used to detect copy number variations (CNVs) located within the peak region and CDH9 and CDH10.  Finally, a combination of RT and RACE PCR was used to evaluate the expression of unannotated ESTs in the peak region and search for previously unidentified exons at the 5’ end of CDH10 and 3’ end of CDH9 that would extend them nearer to the association peak.

Results: Sequence analysis of 100 cases and 100 controls has identified variation at 50 known SNPs and 37 novel variants within the conserved regions near the association peak, 9 known and 26 novel variants in CDH10, including 2 nonsynonymous amino acid changes, and 12 known and 9 novel variants in CDH9, including 5 nonsynonymous amino acid changes.  Novel CNVs were detected in a single autistic individual and arrays of higher density of coverage in this area are currently in process.  RT-PCR suggested the expression of predicted ESTs near the association region in brain tissue.  Finally, RACE PCR indicated a 150bp extension of the 5’UTR of CDH10, though not a novel exon.

Conclusions: Our preliminary results suggest that several molecular factors in the autism association peak region on chromosome 5p14.1 may contribute to the association signal.  Novel sequence variations and CNVs are currently being validated in a larger dataset to assess their frequency in cases vs. controls.  Moreover, analysis of the predicted ESTs and new 5’UTR extension of CDH10 may reveal variations which confer increased risk of autism.