Autism and schizophrenia: two sides of the same coin? (Crespi et al. 2009, PNAS)
Autism and schizophrenia are two disorders that I wouldn’t think to compare to each other. Autism is usually evident by the time a child turns three years old, and is normally characterized by a lack of social and communication skills. Schizophrenia, on the other hand, is typically later-onset, in the teens and twenties, and stereotyped by imaginary friends and thus talking to oneself.
In 1943, Leo Kanner theorized a connection between the two, and placed autism as a type of very early-onset schizophrenia. (Imagine the Venn diagram with the little autism circle inside of the big schizophrenia circle.) He later renounced this theory, putting them in separate, unrelated circles.
A new study from Simon Fraser University in British Columbia by Bernard Crespi, Philip Stead, and Michael Elliot has brought these two disorders back together again in their PNAS (Proceedings from the National Academy of Sciences) paper, “Comparative genomics of autism and schizophrenia,” published early online on November 30, 2009 (doi: 10.1073/pnas.0906080106). The authors, however, argue for a new model: that they are actually diametric, or opposite, conditions.
As with many genetic disorders, there are often many different mutations that can occur at the DNA level which show the same symptoms. (For a review of DNA and mutations, check out my DNA Basics Primer.) The authors of this study did a genome-wide scan, collected the most common mutations associated with these disorders, and used data from a public database to compare rates of these specific mutations between autism- and schizophrenia- diagnoses to determine whether these disorders are related, and how.
Frankly, their results are astounding. First of all, they found 20 gene mutations which are common to both disorders, “results inconsistent with a separate and independent relationship of autism and schizophrenia.” They also found several genes that were common to autism but not schizophrenia, rejecting Kanner’s original model of autism being a subset of schizophrenia. They thus conclude that these disorders are related in some way.
Here’s the amazing part — at 4 different gene loci, they found opposite results for autism and schizophrenia. That is, where a duplication of a gene section was associated with autism, a deletion of the same gene was associated with schizophrenia. Their statistical work places these negative associations far from the realm of chance.
This means that autism overproduces 2 proteins which are non-functional in schizophrenics, and schizophrenics overproduce 2 proteins that are absent in autistic individuals. Previous studies have found that both disorders are associated with improper regulation of genes involved in cell growth-signaling pathways (PI3K, Akt, and mTOR), activating or deactivating the growth, proliferation, differentiation, and death of cells (apoptosis). Autism has excess signaling while schizophrenia has reduced signaling, especially in the brain. Further supporting this is the collected evidence from multiple studies that autistic individuals often have increased brain size in childhood, while schizophrenic individuals have reduced brain size.
If these results hold with future tests, which I’m sure will come, it could mean a lot for both diagnosis and treatment of these disorders. For example, a treatment for autism could lead to immediate research into a related treatment for schizophrenia, if the relationship between these disorders becomes clear.
Let’s not get our panties in a twist just yet. While these 4 genes have very strong diametric associations and the authors’ work should be applauded, there are hundreds of genes associated with each of these disorders and the causes of autism and schizophrenia are each unclear. It’s easy to get excited over clear patterns in scientific study — I’m guilty of it on a regular basis. But one study doesn’t prove a theory; much more work needs to be done to fully clarify this relationship.
Crespi, B., Stead, P., & Elliot, M. (2009). Comparative genomics of autism and schizophrenia Proceedings of the National Academy of Sciences DOI: 10.1073/pnas.0906080106