Genotype-Phenotype Correlations of Phelan-Mcdermid Syndrome Based on Neurodevelopmental Gene Expression Patterns of the 22q13 Region

Background:  Phelan-McDermid Syndrome (PMS), also known as 22q13 deletion syndrome, is a rare genetic disorder that results from gene deletions on chromosome 22.  Hundreds of different deletions of varying size have been reported in this gene-rich region, which contains ~100 coding mRNAs as well as other, poorly characterized non-coding RNAs (ncRNAs).  The clinical phenotype of PMS is similarly heterogeneous, characterized by intellectual disability, autism spectrum disorder (ASD), hypotonia, severe delays or absence of speech, and dysmorphic physical features.  Investigations into the molecular mechanisms linking deletions on 22q13 to the clinical phenotype have focused on the synaptic scaffolding protein SHANK3, despite observations suggesting the phenotype of PMS is dependent on gene deletion size.  It is important to understand what role other RNAs on 22q13 may play in human neurodevelopment, and how their disruption may contribute to features of the PMS phenotype.  

Objectives:  This work attempts to (i) comprehensively assess in an unbiased manner the functional genomic landscape of the 22q13 region and then (ii) to interpret these discoveries in the context of genomic location on 22q13 and the phenotypic features of PMS.

Methods:  We used the Allen Brain Atlas database to assess the brain expression patterns of genes on the 22q13 region across typical neurodevelopment (16 brain regions; 2nd trimester through 40 years of age).  We identified a subset of genes in the 22q13 region with enhanced expression in early development and in autism-related brain regions, and then analyzed the interactions of these genes and their roles in molecular pathways using the Ingenuity Pathway Analysis (IPA) software.

Results:  Surprisingly, we found the majority (38/69; 55%) of genes on 22q13 are not expressed in the brain at all.  Furthermore, we found that 9/24 (38%) of SHANK3 exons do not have detectable expression during human neurodevelopment.  The spatio-temporal expression profiles of individual genes vary across development, and multiple trends emerge, such as a decrease in expression during early childhood (ages 2-4 years).  In particular, we found SHANK3 is not expressed in brain prenatally compared to postnatally.  In addition to SHANK3, we identified 4 genes on 22q13 as critical to typical neurodevelopment based on their expression: ATXN10, MAPK8IP2, MLC1, and SULT4A1.  Network analyses of these genes revealed significant interactions with SHANK3 and consistent enrichment on molecular pathways critical to synaptic development, cell-to-cell signaling, and plasma membrane transport.

Conclusions:  Our results provide insight into the genomic landscape of the 22q13 region and the role these genes play in neurodevelopment.  Most of the genes in this region are not expressed in the brain at all, and the expressed genes exhibit highly variable expression across development. Similar variation and unexpected lack of expression also exist for SHANK3 exons. We show SHANK3 is expressed significantly less prenatally compared to postnatally, which has potential implications for developing treatments of PMS.  We identified four additional genes on 22q13 that have very high neurodevelopmental expression, play critical roles in cell-to-cell signaling, membrane transport, and synaptic scaffolding pathways, and which may be additional strong candidates for genes underlying the PMS phenotype.