Posted by jlwile on August 17, 2011
Autism is a poorly-understood neurological disorder that affects many people throughout the world. Unfortunately, because it is poorly-understood, there is an tendency for people to blame autism on anything they don’t like. For example, there are those who try to claim that vaccines cause autism. When confronted with the overwhelming scientific evidence against such a claim, many of those people simply ignore the data.
For example, not long ago, I did an online debate on whether or not vaccines cause autism. The debate was heavily-publicized by the anti-vaccination group that hosted it, but after the debate, all mention of it was removed from the group’s website. Why? Because I simply presented the data that clearly show there is no way autism could be related to vaccination. The group decided to pull all mention of the debate rather than risk some of their readers learning what the data actually say about vaccines and autism!
Fortunately, most people are more interested in finding the real causes of autism. Thus, they have looked at the data and realize that vaccines simply aren’t a possibility. As a result, they have moved on and are looking at other possible causes. About a year ago, I blogged about a study that tried to pin down the genetic causes of autism. Since autism is a highly heritable disease1, it makes sense that the cause should be genetic. However, rather than implicating just a few genes, the study came to the conclusion that there are a lot of genes involved in autism. That made the results rather disheartening, because it is hard enough to treat a disease that is caused by only one or two genes. How can you possibly treat a disease that is caused by lots and lots of genes?
Well, researchers from UCLA might have found an answer to that question.
In the study, the researchers compared brain tissue samples that came from 36 dead people. 19 of them did not have autism, and 17 of them did. Rather than looking at the tissue structure, however, they looked at gene activity within the tissue.2 How do you measure gene activity in dead tissue? You look at the RNA found there.
In order for a cell to use the information contained in a gene, the gene must be copied by an RNA molecule, and the RNA molecule must then leave the cell’s nucleus. The RNA molecule must then go to a ribosome, where the information that has been copied by the RNA can be translated into a protein. So if you look at the RNA in a tissue sample, you are looking at what genes were being copied shortly before the tissue died. Since the cell only copies genes when it needs to use them, this is essentially a measure of what genes were active at the time of death.
When the researchers compared gene activity in the frontal lobe of each brain (see picture above) to gene activity in the temporal lobe of each brain, they found something quite amazing. In the brains from people without autism, gene activity was quite different between the lobes. That makes perfect sense, since those two lobes are involved in rather different neurological processes. The portions of the frontal lobe that were examined deal with high-level reasoning tasks such as planning and impulse control. The parts of the temporal lobe that were studied deal with language and how people relate to one another. Thus, you would expect the genes that are active in each region to be quite different.
However, when comparing gene activity in the tissue samples from the brains of autistic people, the researchers found very little difference between the two regions. To give you an idea of how striking the difference was, there were more than 500 genes that varied in activity between the two regions of the non-autistic brains. There were only 8 genes that varied in activity between the two regions of the autistic brains! In addition, while the pattern of gene activity was quite varied among the non-autistic brains, the pattern of gene activity among the autistic brains was remarkably similar.
This seems to indicate that even though the genetic causes of autism might be rather varied, the practical results of those causes are rather similar. As one of the authors in the study noted:3
It looks like there’s a common pathology in autism, which is a surprising thing…In spite of having many different causes, there’s some shared convergence.
Hopefully, this shared convergence will lead to an effective treatment. Of course, before that happens, the conclusions of this study will need to be verified and fleshed out with larger, more detailed studies.
1. Pickles A, et al., “Latent-class analysis of recurrence risks for complex phenotypes with selection and measurement error: a twin and family history study of autism,” American Journal of Human Genetics 57 (3):717-26, 1995.
Return to Text
2. Irina Voineagu, et al., “Transcriptomic analysis of autistic brain reveals convergent molecular pathology,” Nature 474:380–384, 2011.
Return to Text
3. Laura Sanders, “Genetic Analysis Reveals Clues to Autism’s Roots,” Science News June 18, 2001, p. 5.
Return to Text