Dysfunctional ADAM22 implicated in progressive encephalopathy with cortical atrophy and epilepsy

Objective: To identify the molecular genetic basis of a syndrome characterized by rapidly progressing cerebral atrophy, intractable seizures, and intellectual disability. Methods: We performed exome sequencing in the proband and whole-genome single nucleotide polymorphism genotyping (copy number variant analysis) in the proband-parent trio. We used heterologous expression systems to study the functional consequences of identified mutations. Results: The search for potentially deleterious recessive or de novo variants yielded compound heterozygous missense (c.1202G>A, p.Cys401Tyr) and frameshift deletion (c.2396delG, p.Ser799IlefsTer96) mutations in ADAM22, which encodes a postsynaptic receptor for LGI1. The deleterious effect of the mutations was observed in cell surface binding and immunoprecipitation assays, which revealed that both mutant proteins failed to bind to LGI1. Furthermore, immunoprecipitation assays showed that the frameshift mutant ADAM22 also did not bind to the postsynaptic scaffolding protein PSD-95. Conclusions: The mutations identified abolish the LGI1-ADAM22 ligand-receptor complex and are thus a likely primary cause of the proband's epilepsy syndrome, which is characterized by unusually rapidly progressing cortical atrophy starting at 3–4 months of age. These findings are in line with the implicated role of the LGI1-ADAM22 complex as a key player in nervous system development, specifically in functional maturation of postnatal synapses. Because the frameshift mutation affects an alternatively spliced exon with highest expression in postnatal brain, the combined effect of the mutations is likely to be hypomorphic rather than complete loss of function. This is compatible with the longer survival of the patient compared to Lgi1−/− and Adam22−/− mice, which develop lethal seizures during the first postnatal weeks.


Appendix e-1 Exome variant analysis under recessive and de novo inheritance models
Variants with the following Variant Effect Predictor 1  Given that the healthy parents of the study subject were not exome sequenced, de novo mutations could not be assessed directly. Instead, in the exome variant filtering strategy aiming to identify pathogenic de novo variants, heterozygous variants absent from the above three variant databases were included, after which the candidates were subjected to segregation analysis (figure e-1). Additionally, dbSNP build 138 (http://www.ncbi.nlm.nih.gov/SNP/) variants were excluded except those with clinical association in NCBI ClinVar (http://www.ncbi.nlm.nih.gov/clinvar/).
After applying the filtering criteria, low quality variants were excluded based on IGV visualization of sequence reads. 10 The qualifying variants were subjected to segregation analysis by capillary sequencing.
Even though mtDNA is not included in the SureSelect exome capture kit, an average of 35.7× sequencing coverage in the mitochondrial genome was obtained, due to the abundance of mitochondrial DNA in the cells. We called mtDNA variants using samtools. 11 The MITOMAP database (www.mitomap.org) was used to identify known mtDNA mutations and polymorphisms.

Cell surface binding assay
COS7 cells were seeded onto poly-d-lysine 12-mm cover slips in a six-well cell culture plate (3 × 10 5 cells/well) and co-transfected with LGI1-FLAG and ADAM22.
At 24 h after transfection, cells were fixed with paraformaldehyde/120 mM sucrose/100 mM HEPES (pH 7.4) at room temperature for 10 min and blocked with PBS containing 10 mg/ml bovine serum albumin for 10 min on ice. The fixed cells were stained with anti-FLAG antibody, followed by Cy3-conjugated secondary antibody. Then, the cells were permeabilized with 0.1% Triton X-100 for 10 min, blocked with PBS containing 10 mg/ml BSA, and stained with anti-ADAM22 polyclonal antibody, followed by Alexa488-conjugated secondary antibody and staining with Hoechst dye (33342, Invitrogen). Fluorescent images were taken with a confocal laser microscopy system (Carl Zeiss LSM 510; Carl Zeiss). For double staining of Ser799IlefsTer96 mutant and an ER marker (anti-KDEL antibody), transfected COS7 cells were fixed, permeabilized and blocked as described above.
Then, cells were stained with anti-ADAM22 and anti-KDEL antibodies, followed by Alexa488-and Cy3-conjugated secondary antibodies, respectively.

Immunoprecipitation
The transfected HEK293T cells were washed with PBS and subsequently lysed with

Exome sequencing metrics
Exome sequencing of the proband produced 4.4 Gb of sequence with a mean coverage of 84.68× in the target regions, of which 96.2% and 92.8% were captured with at least 5× and 10×, respectively.

Lack of biallelic mutations in ADAM22 in other datasets
To attempt to identify more patients with mutations in ADAM22, we utilized exome data from our 29 other in-house exomes from patients with severe epilepsy syndromes (unpublished data, Laari A., Muona, M. et al.). In addition, we accessed 178 exomes or whole genomes of epileptic encephalopathy cases generated in EuroEPINOMICS Rare Epilepsy Syndromes consortium and exomes from >1000 children included in the Deciphering Developmental Disorders project, 13 where 24% of the cases present with seizures. However, we did not identify rare, biallelic mutations in ADAM22 in these data.       indicates the conservative threshold that was used by the authors to define an 'expressed' gene. 18 The graph was generated based on data released by the Human Brain Transcriptome project 18 using exon-level microarray data of the transcriptome in various brain regions (Gene Expression Omnibus accession ID GSE25219). The project used 1,340 tissue samples collected from 57 developing and adult postmortem brains of clinically unremarkable donors. Similar results for ADAM22 expression are obtained from BrainSpan database with exon array and RNA sequencing data from developing and adult human brain (BrainSpan: Atlas of the