Functional analysis of disease-associated CASK missense mutations

Abstract

The calcium/calmodulin-dependent protein serine kinase CASK is a member of the MAGUK (membrane-associated guanylate kinase) family of proteins. CASK fulfills several functions in neurons by interacting with several interaction partners. When binding to Liprin and Neurexin, CASK acts as a presynaptic scaffolding protein. When binding to Sap97, CASK can regulate the trafficking of postsynaptic glutamatergic receptors. When binding to Tbr1 and CINAP, CASK can regulate the transcription of genes with T-element promoters, such as GRIN2B, which encodes the NR2B subunit of the NMDA receptor. The CASK gene is located on the X-chromosome, meaning that males can only be hemizygous for CASK mutations. Heterozygous female and hemizygous male patients with CASK loss-of-function mutations exhibit a recognizable set of phenotypes called microcephaly with pontine and cerebellar hypoplasia with intellectual deficiencies (MICPCH + ID). These symptoms are too severe to correlate to any specific molecular function of CASK. However, CASK missense mutations have also been reported in male patients who often exhibit either a subset or a milder form of the symptoms observed in patients with CASK loss-of-function mutations. One possible explanation for the less severe symptoms observed in patients with CASK missense mutations is that CASK missense mutations may not disrupt all of CASK’s molecular functions, but a subset of them. The single amino acid replacements may disrupt the ability of CASK to bind to specific interaction partners, and in turn, disrupt the function associated with each lost interaction partner. In principle, it should be possible to correlate the symptoms of patients with CASK missense mutations with the molecular phenotypes of each missense mutation. In this project, I studied 12 CASK missense mutations identified in male hemizygous patients. My goals were to determine, first, whether each mutations was pathogenic; second, the pathogenic mechanism underlying each missense mutation; third, whether certain pathogenic mechanisms are shared across multiple missense mutations; and fourth, if certain pathogenic mechanisms correlated with patient symptoms. In the first part of the project, I performed coimmunoprecipitation experiments to determine which partners were affected by which mutation. With the exception of Liprin-α2, which bound to all 12 CASK mutants, all other partners were affected by at least one mutation. CASK:Sap97 binding was disrupted by the L354P mutation, suggesting that this mutation affects CASK’s ability to regulate postsynaptic glutamatergic receptor trafficking. The CASK:Tbr1 and CASK:CINAP interactions were disrupted by the Y723C and W914R mutations, suggesting these mutations disrupt CASK’s ability to regulate gene transcription. However, the CASK:Neurexin interaction stood out as particularly sensitive, being mildly affected by the R489W and M507I mutations, and completely disrupted by the G521V, Y723C, and W914R mutations. This suggests that the loss of CASK:Neurexin interaction could be a major pathogenic mechanism underlying CASK missense mutations. I then performed functional assays to determine whether the CASK functions associated with each partner were disrupted by the various mutations. The L354P mutation slightly decreased the trafficking of NMDA receptors compared to AMPA receptors, in accordance with previous literature, but not significantly. I did not observe any effect of CASK on the transcriptional activity of Tbr1, which made it impossible to study the effect of the Y723C and W914R mutations on this aspect of CASK function. Studies of other MAGUKs suggest that CASK should be able to oligomerize when bound to Neurexin, a presynaptic adhesion protein. Oligomerization upon binding to Neurexin would both greatly aid in CASK’s presynaptic scaffolding function and localize it to the correct subcellular compartment. Using split-YFP experiments, I confirmed that CASK oligomerizes in the presence of Neurexin. Furthermore, the G521V, Y723C and W914R mutations, which completely disrupted CASK:Neurexin interaction, also disrupted CASK’s ability to oligomerize even in the presence of Neurexin, further supporting the hypothesis that CASK oligomerizes in the presynapse to scaffold the assembly of protein complexes. Until now, the main proposed pathogenic mechanism underlying CASK loss-of-function mutations has been the disruption of its ability to upregulate the transcription of genes because some of the genes it regulates, such as Reelin, are necessary for brain laminar development. My results instead suggest that it is the loss of CASK:Neurexin interaction, which mediates CASK’s presynaptic scaffolding role, that may be the main pathogenic mechanism underlying CASK missense mutations. This notion is further supported by a recent study that also suggests that loss of the CASK:Neurexin interaction could also cause microcephaly in patients.