Ashwin Prakash M.D., Predoctoral Fellow, Department of Medicine, University of Toledo, email: ashwin.prakash@utoledo.edu
Could non-coding RNAs be involved in development and propagation of Autistic Spectrum Disorder?
Ashwin Prakash*, Theodor Rais, Sadik Khuder, Alexei Fedorov
University of Toledo-HSC, Cardiovascular sciences program
The Autistic Spectrum Disorders (ASD), are known for their considerable heterogenic clinical presentation. Linkage studies have pointed toward very diverse areas of the genome related to the brain and associated with ASD. We hypothesized that small non-coding RNA (ncRNA), like miRNA, siRNA or orphan snoRNAs, could be involved in the pathogenesis of ASD. The complex regulatory roles of miRNA.s have been implicated in diseases like Tourette.s syndrome and Fragile X syndrome. Since ncRNAs have been found abundantly in introns, we set out to find evolutionarily conserved areas in the introns of ASD associated genes, which may play a causal role in some clinical phenotypic forms of ASD.
A group of 73 genes associated with ASD were selected, proven through linkage studies, presence of biological markers (evidence of SNP.s or haplotypes) or candidate genes association studies. A control group of 75 genes, which were predominantly expressed in non-brain tissues was selected. We procured in this fashion an ASD set, which contained 540 introns and a non-brain set, which consisted of 522 introns. Using the protein sequences of these two sets of genes, their orthologs in dog, mouse and rat were obtained. We then developed a PERL program to investigate the degree of conservation among the different species. orthologous intronic sequences aligned by ClustalW. A threshold of 50% was set to qualify a segment of sequence as conserved, and a minimum of 50 nucleotides was used as the window size so as to exclude highly conserved but short regions, such as transcription factor binding sites from our results. Several filters were created within our program so as to exclude the possibility of the conserved sequence being a part of alternatively spliced exons.
For each intron in the two sets of genes we quantified the total number and length of conserved segments, and used it to compare the two groups. The results showed that the average number of conserved segments per intron in autism specific genes was 3.9, while it was 1.4 in the case of the genes minimally expressed in the brain. A non-parametric statistical test (Mann-Witney) showed a significant difference in degree of conservation between the two groups. Application of RNA Vienna package for prediction of RNA secondary structures demonstrated that many conserved segment sequences have stable RNA structures and their cross-species evolutionary conservation. Our findings could help to shed light on possible etiologies of development and genetic susceptibility of ASD.
Support: National Science Foundation Grant No. 0643542