Integrative genetic analysis of autism brain tissue

Dan Arking, Ph.D. Joel Bader, Ph.D.
Johns Hopkins University School of Medicine

Andrew West, Ph.D.
University of Alabama School of Medicine

Susceptibility to autism has a strong genetic component, but the search for autism risk genes has met with limited success. Despite having identified several autism-linked genes through genome-wide association studies, potentially many more exist that cannot be identified this way. Dan Arking and his colleagues at the Johns Hopkins University School of Medicine are taking a new approach to this search, combining measurements of altered gene expression — obtained directly from brain tissue using high-throughput, next-generation sequencing — with other genetic data.

The scientists recently found that some cases of autism can be linked to the gene CNTNAP2, which codes for a protein that regulates signaling between neurons, and SEMA5A, which encodes a protein involved in neuron development. Neither of these genes show sequence variations that modify the proteins directly, suggesting that the risk for autism associated with these genes results from changes in the amount, timing or location of protein expression. Arking and his colleagues also found that the amounts of messenger RNA expressed by these genes in brains obtained from people with autism were altered relative to those of age- and sex-matched controls.

To find more autism-associated genes, the researchers plan to screen the brains of individuals with autism and controls for differences in genetic sequences, DNA methylation and gene expression. The analyses may directly identify autism genes, or they may be combined with existing data from genome-wide association studies to uncover new genes and validate previous findings.

Their data may also identify important downstream targets — genes that are directly or indirectly regulated by the altered genes. The discovery of autism-linked genes and their downstream targets will point to compromised pathways and thus potential targets for new therapies. Arking’s team plans to make all of their data and software publicly available, allowing other researchers to benefit from this work.