Considerable clinical heterogeneity has been well documented amongst individuals with autism spectrum disorders (ASD). However, little is known about the biological mechanisms underlying phenotypic diversity. Genetic studies have established a strong causal relationship between ASD and molecular defects in the SHANK3 gene. Individuals with various defects of SHANK3 display considerable clinical heterogeneity. Different lines of Shank3 mutant mice with deletions of different portions of coding exons have been reported recently. Variable synaptic and behavioral phenotypes have been reported in these mice, which makes the interpretations for these data complicated without the full knowledge of the complexity of the Shank3 transcript structure.

We systematically examined alternative splicing and isoform-specific expression of Shank3 across different brain regions and developmental stages by regular RT-PCR, quantitative real time RT-PCR (q-PCR), and western blot. With these techniques, we also investigated the effects of neuronal activity and epigenetic modulation on alternative splicing and isoform-specific expression of Shank3. We explored the localization and influence on dendritic spine development of different Shank3 isoforms in cultured hippocampal neurons by cellular imaging.

The Shank3 gene displayed an extensive array of mRNA and protein isoforms resulting from the combination of multiple intragenic promoters and extensive alternative splicing of coding exons in the mouse brain. The isoform-specific expression and alternative splicing of Shank3 were brain-region/cell-type specific, developmentally regulated, activity-dependent, and involved epigenetic regulation. Different subcellular distribution and differential effects on dendritic spine morphology were observed for different Shank3 isoforms.

Our results indicate a complex transcriptional regulation of Shank3 in mouse brains. Our analysis of select Shank3 isoforms in cultured neurons suggests that different Shank3 isoforms have distinct functions. Therefore, the different types of SHANK3 mutations found in patients with ASD and different exonic deletions of Shank3 in mutant mice are predicted to disrupt selective isoforms and result in distinct dysfunctions at the synapse with possible differential effects on behavior. Our comprehensive data on Shank3 transcriptional regulation thus provides an essential molecular framework to understand the phenotypic diversity in SHANK3 causing ASD and Shank3 mutant mice.