Microdeletions and Microduplications at 15q11.2 in Autism Spectrum Disorder in a Costa Rican Cohort

Background: Recently, de novo copy number variants (CNVs) at 15q11.2 were reported to represent rare risk factors for schizophrenia in a large population-based study. An inherited deletion at the same locus was previously reported in a boy with neurological disorder, mental retardation and speech impairment. Furthermore, cases with Angelman syndrome due to the larger class I deletion, which includes the 15q11.2 region, are more likely to meet criteria for autism spectrum disorders (ASDs). The Central Valley of Costa Rica (CVCR) is a well-studied founder population that has been studied for neuropsychiatric disorders and ASDs.

Objectives: The broad goal of this study was to investigate the role of copy number variants (CNVs) in CVCR subjects with an ASD.

Methods: All ASD cases are being screened for CNVs using array-based comparative genomic hybridization (CGH) analysis and/or SNP arrays. QPCR and multiplex ligation-dependent probe amplification (MLPA) assays are being used to validate potential CNVs.

Results: We identified three subjects with ASDs in the CVCR that showed a CNV at 15q11.2. The CNV occurs between BP1 and BP2 (chr15:20089383-20630718, NCBI Build 35) in 15q, and includes four genes (TUBGCP5, CYFIP1, NIPA2 and NIPA1). A CNV loss arose do novo in one case, while two CNV gains were inherited. All three subjects are male with an autism diagnosis. One case with a CNV gain has had reported convulsive episodes, and has macrocephaly and moderate cortio-subcortical atrophy, but EEG tests are normal in all cases. In the case with the de novo CNV loss, the mother had language delay (spoke late and had articulation problems). Functional analyses in mice indicate that haploinsufficiency of CYFIP1 can result in abnormalities in synaptic plasticity.

Conclusions: CNVs at 15q11.2 were found in ASD in subjects from the CVCR. Functional studies support a role for one of these genes in synaptic plasticity. Our findings provide further support for a common pathogenic pathway in some cases of schizophrenia and of ASDs. We are examining whether this results from pleiotropic effects of the implicated genes or from an intermediate phenotype that might increase risk for both ASD and schizophrenia.