Classifying autism etiology by expression networks in neural progenitors and differentiating neurons
Massachusetts General Hospital
Basic & Clinical
In many subjects with autism, disruption of a gene is thought to have contributed strongly to the development of the disorder. However, genetic research has identified that the same gene is not involved in every case of autism. Rather, many distinct genes have been identified which contribute to the development of autism when they are inactivated. The actual mechanism by which any given inactivated gene leads to autism is not known, but the genes that have been implicated appear on the surface to be quite different from each other, as some are regulators of other genes, some are involved in adhesion of nerve cells to each other and to other cells, some are directly involved in the structures that permit electrical transmission between nerve cells, and still others have no known function. The differences between these autism genes might mean that there are many "autisms" with each one having a mechanism that has a different starting point (a different gene), a different process (the biochemical and developmental consequences of the gene) and a common end-point (autism) and therefore each one would require a different mechanism-based treatment. However, it is more likely that the different genetic triggers lead ultimately to consequences that converge on the same shared processes that produce autism. If true, this would mean that autism subjects could be grouped based upon which shared process is involved in their case and target treatments to the limited number of shared processes that are identified across all autism subjects. This hypothesis will be tested by examining the consequences of inactivating individual autism genes in cultured human immature nerve cell precursors and allowing them to differentiate into mature nerve cells. As they do, their gene expression profiles will be monitored, measuring the level of expression of all of their genes using a powerful sequence-based technology that determines whether inactivation of different autism genes leads to overlapping sets of shared changes in gene expression. This approach promises to allow the clustering of autism genes based upon which shared process that they trigger and thereby permit the development and testing of treatments applicable to a broader array of autism subjects than would be possible if each gene necessitated its own individual treatment.