The Synaptic Vesicle Gene RIMS3 Is a Novel Candidate for Autism

Background: Chromosomal studies and copy number variant (CNV) analysis in patients with autism spectrum disorders (ASD) have successfully led to the identification of causative and susceptibility genes. A large number of these genes are important components of the pre- and postsynaptic machineries, including NLGN3, NLGN4, NRXN1, SHANK3, and CNTNAP2. These observations bolster the hypothesis that autism is a ‘disorder of the synapse’. Objectives: A child with autism and microcephaly was found to have a de novo 3.3-Mb deletion on chromosome 1p34. Here, we test the hypothesis that this microdeletion will contain a gene(s) that underlies the autism phenotype in this child and in other children with ASD.  Methods: To search for submicroscopic chromosomal rearrangements in the child, we performed CNV analysis using 1) a 19K whole-genome human BAC microarray; and 2) the Illumina 610-Quad BeadChip genotyping microarray. We used Ingenuity Pathway Analysis (IPA) to construct biological networks to identify candidate genes for autism. To identify functional variants in candidate genes, we performed mutation screening using PCR-based Sanger sequencing in 512 unrelated autism patients and 465 control subjects. Results: We identified a 3.3-Mb deletion containing 43 genes in chromosome 1p34.2 that we confirmed using fluorescence in situ hybridization (FISH). Microsatellite analysis determined that the deletion was de novo. IPA analysis and our review of the literature identified RIMS3 as the most promising autism candidate gene because 1) it forms biological networks with known autism genes, albeit indirectly; 2) it belongs to the RIMS protein family that constitute an important component of synaptic vesicle fusion and neurotransmitter release; 3) it is expressed in human fetal and adult brains; 4) its overexpression has been demonstrated to greatly facilitate Ca²⁺-triggered exocytosis; and 5) mostly notably, its expression in lymphoblasts has recently been shown to be dysregulated in patients with autism. Mutation screening of RIMS3 in 512 unrelated autism patients identified five inherited coding variants that were absent in 465 controls. One of these variants included E177A that is predicted to be deleterious and segregated with the autism phenotype in a sibship. Conclusions: Identification of a de novo microdeletion containing RIMS3 and discovery of autism-specific RIMS3 missense variants that segregate with autism and related phenotypes support a role for this gene in the etiology of ASD. Elucidating the functional impact of putative RIMS3 mutations such as the E177A substitution represents an important area of investigation . Towards this end, we have obtained RIMS3- and hGH-containing vector constructs to perform Ca²⁺-dependent exocytosis co-transfection experiments in PC12 cells to test the involvement of mutant RIMS3 variants in synaptic function using site-directed mutagenesis.