UCLA

Language is a complex communicative behavior unique to humans, though its genetic basis is still poorly understood. Genes associated with human speech and language disorders have provided a basis for study, originating with the FOXP2 transcription factor, a mutation in which is the source of an inherited form of developmental verbal dyspraxia. Subsequently, targets of FOXP2 regulation have been investigated for their associations with language-related disorders. One such target, contactin associated protein-like 2 (CNTNAP2), is associated with autism and specific language impairment. Due to the exclusivity of language to humans, no single animal model is sufficient to study the complete behavioral effects of these genes. However, some animals do possess components of language. One such component is vocal learning, which though rare in the animal kingdom, is shared with songbirds. Here, I use the zebra finch songbird as an animal model to investigate the role of Cntnap2 in birdsong. Cntnap2 is enriched in several song production nuclei in the zebra finch brain, including the striatopallidal nucleus area X, and the cortical lateral magnocellular nucleus of the anterior nidopallium and the robust nucleus of the arcopallium (RA). In adult RA, the distribution of Cntnap2 protein corresponds to the sexually dimorphic singing behavior of this species: males sing, and have enrichment of Cntnap2-expressing neurons, whereas females display neither the behavior nor the enrichment. In juveniles, however, there is comparable enrichment in RA in both sexes until the onset of sensorimotor learning in males, at which time the percentage of Cntnap2-expressing neurons in female RA declines. The neurons in RA that express Cntnap2 are projection neurons that directly innervate the motor neurons that control the vocal organ, analogous to human layer 5 pyramidal neurons of the primary motor cortex that innervate the motor neurons of the larynx. To test the function of Cntnap2 in zebra finch song, I designed and tested RNA interference constructs, which can be used to knock down Cntnap2 in RA. The songbird model can be used to understand the impact of Cntnap2 and other vocal learning genes as they relate to human speech and language.