Copy number variants (CNVs) are segments of DNA that vary in copy number between different individuals. CNVs confer significant risk of neuropsychiatric disorders, including autism and schizophrenia. Notably, there appears to be a reciprocal relationship between copy number and brain size for certain genetic loci. For example, deletions of the genomic region 16p11.2 tend to be associated with autism and increased head circumference, whereas duplications of the same segment tend to be associated with schizophrenia and smaller head circumference. The contrasting clinical phenotypes that are associated with reciprocal changes in gene dosage could represent opposite extremes of the same neurodevelopmental process.

Jonathan Sebat and his team have sought to understand the developmental processes that are affected by CNVs through the genetic analysis of participants and by in-depth clinical and molecular analysis of individuals with clinically relevant CNVs. Microarray analysis of 165 families recruited for this project from Rady Children’s Hospital-San Diego revealed positive genetic findings in 18 individuals. Participants carrying CNVs at two loci of interest (1q21.1 and 16p11.2) were re-contacted for clinical evaluation, including 3D craniofacial imaging. Craniofacial analysis and gene expression profiling of CNVs at the 16p11.2 locus were performed on additional participants from the Simons Variation in Individuals Project (Simons VIP) and NIMH Human Genetics Initiative, respectively.

In collaboration with Curtis Deutsch, analysis of 3D photographic images was performed on individuals that carry CNVs at 1q21.1 and 16p11.2. Features that were significantly influenced by 16p11.2 dosage included the width of the cranial base and length of the nasal bridge, and frontal bossing was observed in a substantial proportion of deletion carriers. Individuals with duplication of 1q21.1 exhibited a wide cranial base and frontal bossing as well, but also exhibited other anomalies. These findings suggest that the dosage of certain genes influences craniofacial variation.

Gene expression analysis of participants’ lymphoblastoid cell lines showed that 16p11.2 dosage had a significant effect on gene expression within the CNV region, while a limited number genes were dysregulated in trans. Subsequent studies, in collaboration with Lilia Iakoucheva, concentrated on elucidating pathways under the regulation of 16p11.2 genes in the human brain. Network analysis of spatiotemporal expression data from human brain tissue showed that co-expression of 16p11.2 genes was highly enriched in the late mid-fetal period of cortical development¹. In addition, co-expression and protein interaction data implicate RhoA signaling in the regulation of brain size and connectivity during development and suggest that its dysregulation by de novo mutations may be a potential determinant of 16p11.2 deletion and duplication phenotypes.