Monday, February 23, 2009

Compound Mutants: A Better Model for Autism?

ResearchBlogging.orgDr. Mriganka Sur and colleagues at MIT's Picower Institute for Learning and Memory published a paper in Proceedings of the National Academy of Sciences earlier this month on a compound mutant mouse model of brain enlargement and abnormal social behavior. Both increased brain size and social deficits are hallmarks of autism spectrum disorders (ASDs), a very hot research topic these days. The paper shows that these mice, which have mutations in two genes called Pten and SLC6A4, exhibit some of the characteristic symptoms of ASDs. The key finding is that the combined mutations exacerbate the mices' symptoms, such that a double mutant exhibits more severe problems than a mouse with mutations in Pten alone or SCL6A4 alone. From this, the authors conclude that these two ASD candidate genes play off of each other during development, providing further evidence that while there is no single "autism gene," combined genetic risk factors seem to play a large role in the disease's etiology.

Pten and SLC6A4 were already known to play individual roles in ASDs and other neurological disorders. Pten is a regulator of a highly complex series of interlinked biochemical processes called the PI3-kinase pathway. This pathway regulates a huge number of cellular functions, including cell growth, differentiation, proliferation, and survival. Because of this, the pathway is closely linked to tumor formation -- indeed, Pten is found to be absent in many kinds of tumors. But aside from its role in cancer, Pten seems to help regulate normal brain development. One effect of Pten disregulation is macrocephaly -- fancy science language for "big heads." This molecule may be required to keep brain growth in check. Relatedly, a subset of individuals with autism and other forms of cognitive impairment are haploinsufficient for Pten, meaning that they carry one normal copy of the gene and one inactivated mutant copy. Macrocephaly is commonly associated with ASDs, and this may be one reason why. (TSC1 and TSC2, the genes that cause tuberous sclerosis, are found in the same pathway. Tuberous sclerosis, it's worth noting, is a tumor-forming disease that is also the most common comorbid disorder with autism.) People with Pten mutations display a broad range of symptoms, however -- some are far more impaired than others. For this reason, the gene has been suggested to sensitize individuals to the effects of other mutations, which in turn establish the severity of cognitive impairment in these patients.

Meanwhile, SLC6A4 is a gene that codes for a serotonin transporter. The protein made by this gene helps to regulate levels of serotonin in the extracellular space. One common biological marker for ASD is an increase in peripheral serotonin levels, so it seems that there is some association between autism and the regulation of this key neurotransmitter. In addition, some patients with ASD have been found to carry a mutant SLC6A4 allele which greatly reduces the transporter's expression. These patients, like individuals with mutant Pten, tend to have overgrown brains and some form of social impairment.

Previous research had made some associations between the PI3-kinase pathway and the serotonin pathway in vitro -- studies of protein-protein interactions and protein expression in cultured cell lines. Dr. Sur's study, however, makes the first direct link between PI3-kinase/serotonin pathway interaction and ASD symptoms in vivo. By breeding together two mutant mouse lines -- a Pten heterozygous knockout and a SLC6A4 loss-of-function line -- the researchers were able to study the combined effect of these two mutations in a living animal.

Both Pten and SLC6A4 single mutants had previously been shown to exhibit macrocephaly. When you combine the two mutations, though, the mice develop brains enlarged more severely than those of mice with individual mutations alone. The biochemical mechanisms allowing these genes to cause brain overgrowth have not been completely explained, but it seems that they are acting on different but related cues in brain development. I was unable to tell from the published figures whether the brain size increase in the compound mutants was greater than the sum of the increases of the two individual mutants, however. It's possible to imagine a situation in which the two genes were acting on completely unrelated pathways that could still cause an additive effect in the compound mutant. Just as a simplified example, we might someday learn that Pten regulates overall cell size, whereas SLC6A4 regulates the pruning of superfluous neurons during development. If this is the case, then you'd expect that combining the two mutations might lead to brain size greater than that seen in either of the single mutants (because then you'd have too many cells and the individual cells would be overgrown). A more convincing link is established if you prove that the change in brain size is somehow worse than would be predicted by the effects of the individual mutations. I wish this paper had provided more explicit measurements for the brain size experiments, to allow for this sort of comparison. Still, the study indicates that the compound mutant displays a more pronounced ASD-like phenotype than either individual mutant, which, combined with the combinatorial mutation theory of ASD, makes these mice an interesting model.

After measuring macrocephaly in these mutant mice, Dr. Sur and colleagues examined their behavior. ASDs are primarily diagnosed based on behavioral deficits, so it's important to establish whether the social behavior of an ASD mouse model is affected by the experimental manipulations being performed. The experiment used a fairly standard test of mouse sociability, in which the mice are given a choice between spending time in a chamber with an unknown "stranger" mouse (defined as the "social" chamber) or a novel inanimate object ("nonsocial" chamber). Most normal mice prefer to spend time getting to know a new mouse than playing with a new object. The Pten single mutant mice did show a deficit in this social behavior assay, spending roughly equal time in each chamber rather than showing a preference for social interaction. The SLC6A4 mutants also showed a decreased preference for the "social" chamber when compared to normal mice, although the effect was not statistically significant. Compound mutants, meanwhile, showed a social deficit more pronounced than that of either of the single mutants. This result seems to confirm the additive effect of the two mutations implied by the brain size measurements.

Other tests of social behavior, including social approach/avoidance and social recognition, also found deficits in socialization among the mutant mice. A test of prepulse inhibition, however, which measures the sensitivity of sensory and motor systems that can be disrupted in ASD patients, did not show an additive effect in the compound mutants. The researchers theorize that the two genes studied here are therefore involved in brain development relevant to social behavior, but not necessarily involved in other ASD symptoms such as sensory and motor defects. For a disease like autism, which involves symptoms in many different systems, this is not an unreasonable theory to espouse.

In the discussion section of their article, Dr. Sur and colleagues propose a few mechanisms by which the serotonin and PI3-kinase pathways could interact in ASD patients. Perhaps Pten is directly affecting the serotonin pathway by binding to serotonin receptors:
One possibility is that Pten and serotonin receptors may physically interact in a regulatory manner to influence brain development. In neurons, Pten binds the 5-HT2c [serotonin] receptor and, via its phosphatase activity, limits agonist-induced activation of this receptor and modulates the firing rate of dopaminergic neurons in the ventral tegmental area. It is interesting to note that 3 drugs that have been reported as alleviating symptoms of autism -— the atypical antipsychotics risperidone and olanzapine and the antidepressant fluoxetine -— all have antagonistic effects on the 5-HT2c [serotonin] receptor, in addition to well-known effects targeting other members of the serotonin and dopamine pathways.

Alternately, serotonin signaling might regulate the PI3-kinase pathway, as has been implied by several tissue culture experiments. Either theory could provide a mechanism by which the two pathways interact to regulate neuron growth, development, and function in a manner that might explain some ASD symptoms. They posit that the interaction between these two pathways could provide new therapeutic targets for treatment of ASD, allowing development of new drugs that ameliorate both abnormal cell growth and dysfunctional intercellular signaling.

Regardless of the exact mechanism, this study shows that a combination of genetic factors can be used to create a better animal model of ASDs. The scientists go on to discuss the fact that Pten dysfunction may increase susceptibility to spontaneous additional mutations and deleterious effects of environmental toxins in haploinsufficient patients. They propose using Pten mutations in combination with other genetic and environmental perturbations to further elucidate the role this gene may play in enhancing ASD symptoms in the presence of additional risk factors. If this line of research pans out, we may learn that while autism has no single genetic cause, there may be certain genes that greatly increase the effect of other risk factors that have yet to be fully characterized. If genes like Pten are responsible for increased sensitization to other mutations, they may provide a single target for treatment of a multifactorial and poorly-understood disease.

A press release from the Picower Institute has more information about this study.

D. T. Page, O. J. Kuti, C. Prestia, M. Sur (2009). Haploinsufficiency for Pten and Serotonin transporter cooperatively influences brain size and social behavior Proceedings of the National Academy of Sciences, 106 (6), 1989-1994 DOI: 10.1073/pnas.0804428106

1 comment:

  1. Hi Laura: I am the father of a 7 year old child who has Autism. We have recently learned via sequencing of three of the genes thought highest in likelihood for asd causation; that he has mutations in two of them. He has two mutations in the shank 3 gene and one in the cntnap2 gene. A frameshift in part of the sequence thought to transcribe but not code of the shank3 gene. A missense mutation later in the shank 3 gene(of unknown significance),and a missense mutation in the cntnap2 gene.

    As a non scientist(I am an engineer)it is difficult for me to know this meaning. However, i find it difficult to believe that the combined effects aren't causitive for ASD. Your thoughts? Also, are there researchers working on the combination of genetic risk factors? if so Who?

    ReplyDelete