A complete Protocadherin-19 ectodomain model for evaluating epilepsy-causing mutations and potential protein interaction sites

Cadherin superfamily members play a critical role in differential adhesion during neurodevelopment, and their disruption has been linked to several neurodevelopmental disorders. Mutations in protocadherin-19 (PCDH19), a member of the δ-protocadherin subfamily of cadherins, cause a unique form of epilepsy called PCDH19 clustering epilepsy. While PCDH19 and other non-clustered δ-protocadherins form multimers with other members of the cadherin superfamily to alter adhesiveness, the specific protein surfaces responsible for these interactions are unknown. Only portions of the PCDH19 extracellular domain structure had been solved previously. Here, we present a structure of the missing segment from zebrafish Protocadherin-19 (Pcdh19) and create a complete ectodomain model. This model shows the structural environment for 97% of disease-causing missense mutations and reveals two potential surfaces for intermolecular interactions that could modify Pcdh19's adhesive strength and specificity.

Identification of four novel PCDH19 Mutations and prediction of their functional impact

The PCDH19 gene encodes protocadherin-19, a transmembrane protein with six cadherin (EC) domains, containing adhesive interfaces likely to be involved in neuronal connection. Over a hundred mostly private mutations have been identified in girls with epilepsy, with or without intellectual disability (ID). Furthermore, transmitting hemizygous males are devoid of seizures or ID, making it difficult to establish the pathogenic nature of newly identified variants. Here, we describe an integrated approach to evaluate the pathogenicity of four novel PCDH19 mutations. Segregation analysis has been complemented with an in silico analysis of mutation effects at the protein level. Using sequence information, we compared different computational prediction methods. We used homology modeling to build structural models of two PCDH19 EC-domains, and compared wild-type and mutant models to identify differences in residue interactions or biochemical properties of the model surfaces. Our analysis suggests different molecular effects of the novel mutations in exerting their pathogenic role. Two of them interfere with or alter functional residues predicted to mediate ligand or protein binding, one alters the EC-domain folding stability; the frame-shift mutation produces a truncated protein lacking the intracellular domain. Interestingly, the girl carrying the putative loss of function mutation presents the most severe phenotype.