A Genome Wide Association Study in Families with MORE THAN ONE CHILD with ASDs

Background: Several family studies have indicated a high heritability for autism, with evidence for a strong genetic basis. The risk of autism in siblings of autistic individuals is reported as approximately 20%, which is higher than the general population prevalence (Baird et al., 2006). However, despite such evidences of a genetic susceptibility, a number of studies failed to identify an unique allele as the basis of an high risk of Autism Spectrum Disorders. Recent studies are focusing on locus heterogeneity by which the same phenotype is caused by risk alleles at multiple different loci (Amaral, Dawson and Geschwind, 2011). Population with high level of genetic transmitted diseases (ie: type 1 diabetes), like for example people living in geographically isolated places, might be particularly interesting to study for ASDs risk, due to their peculiar genetic homogeneity.

Objectives: This study aimed to identify risk alleles in families with more than one child with ASDs in a sample of Italian families, living in an island in the south of Italy (Sardinia) with high level of genetic homogeneity, by employing a genome-wide association methodology.

Methods: For the present study we selected 8 families, each one with two children affected by idiopatic autism. We realized a genome-wide association study (GWAS) and a copy number variant (CNV) analysis by the means of whole-genome genotyping arrays HumanOmni1-Quad. Specifically, we purified genomic DNA by salting out extraction from blood samples of patients and their parents. Samples concentration was evaluated by absorbance (Nanodrop 1000) and fluorometric reading (Pico-green/Qubit). Each sample was genotyped using commercial chip, that interrogate more than a million points in the genome, including single nucleotide polymorphisms (SNPs) and copy number variants (CNVs).

Results: Our analysis revealed genomic variations at the level of recurrent CNVs. In particular, some do not seem to co-segregate with the trait of interest. On the contrary, a CNV mapping on chromosome 17q12, shows, in approximately 30% of patients analyzed, only one copy of genomic DNA. This is a region extending from 1.5 to 3.3 Mb, in the different patients. The region contains the following genes: TBC1D3B, CCL3L1, TBC1D3C, CCL3L3, CCL4L1, TBC1D3H, TBC1D3C, TBC1D3G. The TBC family of genes encodes for proteins involved in RAB GTPase signaling and vesicle trafficking; the CCL family of genes encodes for cytokines, secreted proteins involved in immunoregulatory and inflammatory.

Conclusions: These preliminary results are suggestive of interesting developments but need to be replicated in a larger sample, in which we would endeavor to confirm the results highlighted by the chip, by real-time PCR. Moreover, it might be of interest to evaluate the possible genotype correlations identified through the present study.