Using a modified molecular inversion probe (MIP) approach, a postdoc and a technician spent six weeks pinning down six new genes involved in autism. In addition, they confirmed a new autism pathway. And this is just the beginning, said study author Evan Eichler.
While previous studies uncovered hundreds of de novo mutations in children with autism spectrum disorders (ASD), each of these mutations was typically found in just a single individual. In order to prove that such a mutation significantly increases the risk of ASD, it would have to be found in several autism sufferers. And that would require the screening of thousands more people, which, in turn, requires a significant investment of time and money.
Enter Eichler and Jay Shendure, an associate professor at the University of Washington. Together, their labs, especially shared postdoc Brian O'Rk, optimized an inexpensive solution to this challenge by modifying an MIP strategy that has been used to detect SNPs and exons.
MIPs are single-stranded DNA probes that capture an exon of interest. The MIP is homologous at either of its ends to two regions of a gene, and when the two ends anneal to the gene of interest, the MIP forms a partial loop, similar to a gymnast doing a backbend. The space in between the hands and the feet of the backbend is then filled in with nucleotides, and a ligase enzyme binds the completed loop together. After the loop is released and cleaved, PCR primers surround the now-captured exon to amplify the gene in preparation for sequencing.
Because these MIPs were quite variable in the past, the team needed to increase the efficiency for this study by designing more specific MIPs. In addition, the group automated the workflow and used small amounts of DNA input. Most importantly, they multiplexed; thousands of MIP probes were placed in one test tube with genomic DNA, capturing multiple exons accurately at the same time, thanks in part to improved PCR barcoding. As a result, each sample cost less than $15 to analyze, according to Eichler.
Using this modified technique, the team screened 44 candidate genes in 2446 people with ASD, uncovering 27 de novo mutations in 16 genes. Most interestingly, they found six genes mutated in multiple ASD sufferers. Five of these are involved in a single pathway, called the beta-catenin/chromatin remodeling network, which is involved in neuronal proliferation and cell growth, confirming the importance of this pathway.
In the future, Eichler hopes to match particular mutations to a patient’s specific ASD phenotype. “We will start to break down the umbrella of autism into different subtypes,” he said. The team has already found that people who possess a mutation in the CHD8 gene have a larger head circumference, while those with a mutation in the DYRK1A gene have smaller ones.
Given these advances, physicians may one day screen babies in utero for ASD gene mutations, and perhaps even distinguish the type of autism by which gene is mutated. “There are behavioral therapies that can make a huge difference, but they’re time dependent. You need to get to the kids early,” said Eichler.
- O’Roak BJ et al. 2012. Multiplex targeted sequencing identifies recurrently mutated genes in autism spectrum disorders. ScienceExpress, November 15.