Autism as a quantitative phenomenon: beyond single genes and dichotomous outcomes

Autism spectrum disorders (ASD) and autistic-like traits often run in families. Over the past decade, scientists have identified hundreds of risk genes for ASD, as a first step towards better treatments. However, the same genetic studies now also point out that ASD is almost never caused by a single gene, but most likely due to multiple genetic variants together.

Studies in animals offer the unique opportunity to examine how genetic variation contributes to the development of individual differences in brain and behavior. So far, however, animal models for ASD have several limitations. First, animal models are mostly based on single genetic mutations, while there is increasing evidence that ASD is most likely caused by multiple genetic variants. Secondly, most animal studies have focused on independent traits measured at adult age, while ASD is developmental disorder that emerges early in life.

Chapter 1 describes the development of a longitudinal test battery for the characterization of mouse behavioral and cognitive domains across developmental stages, from preadolescence into adulthood.Our findings reveal different developmental trajectories of repetitive behavior across four mouse inbred strains, which may be relevant to those observed in people with ASD. The results also show that this approach provides sufficient sensitivity to detect differences at different developmental stages.

In chapter 2, these longitudinal trajectories are characterized in mice in a genetic model for ASD and developmental retardation, in which the Cntn4 gene is deleted. Against our expectations, we find that deletion of Cntn4 in the C57BL/6J background does not affect specific autism-related phenotypes in developing or adult mice, but causes subtle non-disorder specific changes in sensory behavioral responses and cognitive performance.A possible explanation for this limited impact of Cntn4-deficiency is the role of genetic background variation.

In chapter 3, we investigated whether exposure to environmental change can be used to enhance the expression of behavioral phenotypes in animal models for ASD. Results from systematic literature review and additional experiments showed that behavioral tests of short duration and with a high degree of novelty can help to detect rigid and inflexible behaviors. These findings may facilitate the discovery of genotype-phenotype relationships in the context of ASD.

Chapter 4 shows that some autistic-like behavioral traits in mice are heritable, and that behavioral differences are influenced by many genes. Mice with genetically diverse genetic backgrounds were screened to see whether they showed a lot or only a few repetitive patterns of behavior when exposed to a new environment. Mice with the same genetic background were found to be very similar, while mice with different genetic backgrounds showed very large differences in behavior. Subsequently, these differences could be mapped to multiple genetic loci, some of which contained genes associated with ASD in humans.

In summary, these findings show that mice can be used to study the developmental trajectories of ASD-like behaviors, and that populations of mice can be used to better investigate the nature of autistic behavioral traits. These findings will likely contribute to the development of new therapies for ASD in humans.