Note: Most Internet Explorer 8 users encounter issues playing the presentation videos. Please update your browser or use a different one if available.

Relevance of Common CNVs for Autism Etiology

Friday, 3 May 2013: 14:00-18:00
Banquet Hall (Kursaal Centre)
I. C. Conceicao1,2,3, C. Correia1,2,3, B. Oliveira1,2,3, J. Coelho1, C. Café4, J. Almeida4, S. Mouga4,5, F. Duque4,6, G. Oliveira4,5,6,7 and A. M. Vicente1,2,3, (1)Instituto Nacional de Saúde Doutor Ricardo Jorge, Lisbon, Portugal, (2)Instituto Gulbenkian de Ciência, Oeiras, Portugal, (3)Center for Biodiversity, Functional & Integrative Genomics, Lisbon, Portugal, (4)Unidade de Neurodesenvolvimento e Autismo – Centro de Desenvolvimento Luís Borges (CDLB), Hospital Pediátrico Carmona da Mota (HP) – Centro Hospitalar e Universitário de Coimbra (CHUC), Coimbra, Portugal, (5)Instituto Biomédico de Investigação em Luz e Imagem, Faculdade de Medicina da Universidade de Coimbra, Coimbra, Portugal, (6)Faculdade de Medicina da Universidade de Coimbra, Coimbra, Portugal, (7)Centro de Formação e Investigação e Formação Clínica (CIFC), Hospital Pediátrico Carmona da Mota (HP) – Centro Hospitalar e Universitário de Coimbra (CHUC), Coimbra, Portugal
Background: Recent reports by the Autism Genome Project (AGP) consortium and other groups show that Copy Number Variants (CNVs), while individually rare, collectively may explain a large fraction of the etiology of Autism Spectrum Disorders (ASD). 

Objectives: The goal of this study was to establish the relevance for ASD etiology of potentially pathogenic CNVs identified in a Portuguese population sample by the AGP whole genome CNV analysis.

Methods: A total of 14218 CNVs were identified in 342 Portuguese probands (35 females, 307 males), genotyped by the AGP using the Illumina Infinium1M SNP microarray. We selected 1062 CNVs (present in 300 individuals) not overlapping by more than 50% with CNVs in 8000 controls from available databases and explored recurrence rates, genic content, regulatory elements, inheritance patterns and clinical correlations.

Results: The identified CNVs ranged from about 5 Kb to 3.7 Mb, with 54% being deletions. Larger CNVs (>500 Kb) were more frequently duplications than deletions. There were 65% genic CNVs, ranging from one gene (67% of all genic CNVs) to 22 genes in a single CNV. CNVs were inspected for recurrence rates and inclusion of ASD-implicated or candidate genes. Interesting CNVs were validated by qPCR, including SHANK3 and NRG1. Although a large percentage of CNVs were present in only one individual (~94%), 13 or ~1.4% were common CNVs, defined as CNVs with a frequency of 1% or greater in the sample population. These CNVs were present in between 3 and 19 individuals, summing a total of 115 individuals (~38%) with common CNVs. Four CNVs not spanning any gene were identified in 3, 4, 5 and 17 individuals. The other nine CNVs include one gene or parts of it, namely ATRX (N=30 individuals), DPYD (N=22), NRP1 (N=13), and VAULTRC2 (N=5), with the exception of one CNV which affects four genes (ATP7A, PGAM4, COX7B, MAGT1 – N=4) that are in tandem with ATRX. We further compared data for autistic traits in the parents (using the BAPQ and SRS questionnaires) with the type of inheritance (inherited vsde novo). We observed a significant excess of autistic traits in the fathers that transmitted a CNV, mainly in the “rigid” personality, which is defined as little interest in change or difficulty adjusting to change.

Conclusions: Here we highlight the importance of studying common CNVs for understanding the etiology of autism. Also, we re-affirm the previously suggested presence of subthreshold autistic traits in the parents of children with ASD, in particular in patients with inherited CNVs. This exhaustive analysis of CNVs for clinical interpretation is necessary for the efficient translation of this knowledge into clinical practice, aiming at the development of molecular tools that may assist behavioral screening procedures for early diagnosis in ASD as well as genetic counseling.

| More