Previously it has been shown that Familial Encephalopathy with neuroserpin inclusion bodies (FENIB) patients develop abnormalities that partially overlap with Autism Spectrum disorders (ASD), confounded with additional features that could be explained by inclusion body formation not expected in subjects with inclusion body forming SERPINI1 mutations. There is no described human neuroserpin deficiency phenotype. The neuroserpin knockout mouse phenotype also suggests a possible overlap with autism. Neuroserpin could contribute directly at the synapse or through altered neuron migration during early development leading to the "underconnectivity" that underlies autism by potentially contributing to the excess of short connections and not enough long ones seen in autistic brains, possibly due to an imbalance in pruning of neurons and synapses early in life. It is thus proposed to sequence the neuroserpin gene in initially 20, and subsequently up to 100 idiopathic autistic patients selected as having the language impairment and perseveration endophenotypes.
It is proposed to perform a pilot study to ask whether genetic variation at SERPINI1 is contributing to idiopathic autism. To maximize the chance of finding such variants, the perseveration and language delay endophenotypes (simpler intermediate traits) will be used to select twenty autistic multiplex (and if necessary singleplex) families with idiopathic autistic patients who have been evaluated with the Autism Diagnostic Interview-revised (ADI-R) and their nuclear families ( siblings and their parents) will be identified by Dr. Liptak and Dr. Lebel with the assistance of Ms Klausner and recruited. Patients with syndromic autism due to known causes such as fragile X syndrome, Tuberous sclerosis, Down syndrome, and Neurofibromatosis type I will be excluded. The entire SERPINI1 coding region (nine exons), one kilobase of promoter region (Chen, 2007) and at least 200 bp of intron flanking each exon will be sequenced in an index case and the segregation of mutations identified investigated within the families. All mutations that appear to be good functional candidates will be compared in frequency amongst autistic and control populations in order to determine if a case can be made that SERPINI1 could be contributing to autism. In addition, known neuroserpin SNPs identified through the Hapmap project (www.hapmap.org) will be investigated and tested for evidence of transmission distortion/disequilibrium by linkage/association analysis. Depending on the results from this initial experiment it is proposed to subsequently sequence up to a further 80 local autism patients (maximum of 100 patients). Additional studies would be performed as follow up studies to similarly sequence the genes of other serpins expressed in the brain; Plasminogen activator 1, PA1, SERPINE1 located at the MLS linkage peak at 7q22.1 seen in IMGSAC study (Lamb, 2002) while Proteinase nexin 1, PN1 SERPINE2 at 2q36.1,close to previously identified linkage peaks (Autism Genome Project consortium, 2007) and potentially their putative serine protease targets (tPA at 8p11.21 and uPA at 10q22.2) and the upstream neuroserpin activators (ALK6, AMH and BMP2) (Lebeurrier, 2008). tPA expression is increased by events that require synaptic plasticity (Yepes, 2002) and enhances NMDA receptor signaling by cleavage of its NR1 subunit (Pawlack, 2002). All interesting leads will be pursued by obtaining additional samples from the Autism Genetic Resource Exchange (AGRE)( www.agre.org), a publically available recourse of phenotypic data and bio-materials which provides diagnostic information and DNA from 100's of multiplex ASD families.