AGRE Researcher Identifies Genetic Association Between Autism and ENGRAILED2 (EN2)

October 14, 2007

AGRE Researcher James Millonig and colleagues have published a genetic association study focusing on a gene called ENGRAILED2 (EN2), which is involved in the formation of the brain during early stages of post natal development.

Specifically, EN2 was shown in several studies, to be essential for the proper formation of specific types of cells called Purkinje cells in the cerebellum. Eliminating this gene in mice reduced the number of these types of cells in the cerebellum.

Remarkably, individuals with autism have reduced levels of these Purkinje cells in the cerebellum, a brain structure involved in activities such as learning and language acquisition. These activities are often a source of difficulty for individuals with autism.
Previous genetic evidence had shown that a gene in the vicinity of EN2, if not EN2 itself, had a statistical likelihood of being involved in autism. In 2001, scientists at Columbia University and UCLA, among others, carried out a genome scan on 110 families who participated in the AGRE program. In a genome scan, scientists look for excessive sharing of genetic variants among affected individuals. These variants serve as markers of the genetic terrain. If a particular set of markers shows excessive sharing among individuals affected with autism, this implies that autism is linked to that part of the genetic map. EN2 has been genetically mapped to an area on the short arm of chromosome 7 (7q), which showed this type of linkage.

Dr. Millonig and his colleagues realized that all lines of evidence implicated the EN2 gene and chose to test its potential genetic association using the Transmission Disequilibrium Test (TDT), a fairly common and sensible method in genetic research. Their group used a public database of genetic variants or Single Nucleotide Polymorphisms (SNPs, pronounce as SNiPs) which are single chemical changes in the DNA code that vary among individuals at a particular location on the genetic map. Four of these SNPs were chosen and tested on 138 triads (2 parents and 1 affected child) in the AGRE collection.

The TDT test simply asks the question whether each parent was more or less likely to transmit form A versus form B of the variant to the autistic child. If the parents were statistically more likely to give A than B, then A is either the cause of the susceptibility to autism or, as is often the case, that A is a good marker for that susceptibility.

The research team chose two SNPs in areas that actually serve as the blueprint for the EN2 protein. Curiously, those two variants did not show any association with autism. However, the two variants that did show an association with autism are located in a region previously considered "junk" DNA. It is now known that these "junk" regions often contain sequences that serve as a signal for when genes get turned on, how often, and in response to which triggers.

Future work will probably include looking at a greater number of families in the AGRE collection in order to replicate these findings. In addition, the discovery and testing of additional SNPs will help to better define the area that seems to be most associated with autism susceptibility.

This work adds to the growing body of evidence that genes involved in brain development are important in autism susceptibility and highlights the potential role of the cerebellum in autism. Finally, this study provides one of the most promising candidate genes in autism susceptibility.

Funding for this study was provided in part by NAAR and the March of Dimes.

Download Complete Text of Publication (PDF Format)
Association of the homeobox transcription factor, ENGRAILED 2, 3, with autism spectrum disorder

See also: Clara Lajonchere, Ph.D., AGRE Program Director, issues an alert to members last week saying the New Jersey study provides one of the "most promising candidate genes in autism susceptibility."