As the number of genetic anomalies associated with autism continues to escalate, demand grows to understand the mechanisms through which these genes affect brain function and behavior. Mouse models engineered to express autism susceptibility genes are critical for achieving this goal, but their utility depends on behavioral assessment methods that have limited reliability, comprehensiveness, sensitivity and throughput.

Tecott and his colleagues at the University of California, San Francisco, developed a novel bioinformatics-based approach to behavioral assessment that addresses these limitations. The home-cage monitoring approach employs continuous automated high-resolution data collection and analysis that reveals an inherent hierarchical organization of behavioral elements. The ability to dissect behavioral patterns in this manner permits highly sensitive analysis of the impact of genes on diverse behaviors.

The researchers applied this approach to three autism mouse models. The first analysis focused on behavioral assessment of the BTBR inbred mouse strain, which has been extensively studied as a model of autism. The second looked at mice heterozygous for a SHANK3 gene mutation, and the third focused on mice bearing a duplication of chromosomal region 16p11.2.

Tecott and his team detected multiple previously unknown phenotypic perturbations of behavior in all three mouse models. The abnormalities were highly distinctive and different among the models. They included abnormal rhythms of multiple behavioral measures related to activity and feeding, and abnormal patterns of exploration. In addition, a novel assay of locomotor entropy revealed phenotypic changes indicative of enhanced behavioral stereotypy (repetitive movements). The researchers also observed abnormal responses to placement in a novel environment and to unexpected changes in the home cage environment. The findings indicate that this approach detects diverse behavioral consequences of genetic endowment with high sensitivity, substantially augmenting the value of rodent genetic models of autism.