Objectives: In this presentation two sets of data will be presented that converge on a system’s approach to understanding potential roles of the cerebellum in autism.
Methods: We have created a comprehensive time-series analysis of the genome-wide RNAs expressed by the cerebellum in the mouse from embryonic day 11 and each 24-hour period until birth and then 72-hour intervals until postnatal day 9 (http://www.cbgrits.org). Using bioinformatics tools to query these data we can construct or test hypotheses that emerge from event-based or gene-based approaches. Additionally, to explore the role of the Purkinje cell in cognitive function we have used a modification of the Lurcher mutant mouse that loses all of its Purkinje cells in a time-frame that is equivalent to the last trimester in humans. By combining Lurcher mutant and littermate wild-type 4-cell embryos to make experimental mouse chimeras, we produce a series of mice that have varying numbers of Purkinje cell loss. These mice have then been tested for stereotypy, attention, spatial memory, and behavioral inflexibility.
Results: The transcriptome work permits a molecular analysis of gene expression that might underlie ASD. From a gene-centric perspective, members of the neuroligin family have been implicated in autism. By querying the CbGRiTS database we find that Neuroligins 1 and 3 are dynamically expressed in cerebellar development and one can identify other molecules that are expressed in a similar pattern or in a manner that proceeds or precedes Nlgn gene expression. In this manner we can build a gene regulation network where Nlgn is a key molecule. We can also query from an event related perspective relative to key events in cerebellar development, such as genes related to the birth of Purkinje cells or the major period of parallel fiber synaptogenesis. The connections between the cerebellum and other brain regions are presumably altered in Purkinje cell-deficient mice. Accordingly, we find that cognitive tasks are impacted by decreased numbers of Purkinje cells. Our behavioral studies have demonstrated a correlation between higher cognitive functions and the number of cerebellar Purkinje cells. We find significantly correlations with deficits in executive function, working memory and repetitive behavior and Purkinje cell number. For example, in two different paradigms we find an increased number of perseverative errors in mice with decreased numbers of Purkinje cells.
Conclusions: This work provides insights into the molecular and functional underpinnings of autism-related phenotypes. We have begun to build gene networks that emerge from genes found in humans that are implicated in ASD and to explore how to explain our functional findings as a disconnection between the cerebellum and forebrain structures with a focus on the prefrontal cortex.