CASKIN1 was identified as an ASD candidate gene based on having a p-value < 0.001 following DeNovoWEST analysis of de novo variants in 16,877 ASD trios from the Simons Simplex Collection, the Autism Sequencing Consortium, the MSSNG cohort, and the SPARK cohort in Zhou et al., 2022; among the de novo variants observed in ASD cases in this analysis were a de novo loss-of-function variant and three damaging de novo missense variants (defined as having a REVEL score > 0.5). Subsequent gene-based meta-analysis involving de novo variant enrichment, transmission disequilibrium testing (TDT) of rare, inherited LoFs from unaffected parents to affected offspring, and comparisons of loss-of-function variants in cases vs population controls in this report found that CASKIN1 exhibited a nominal enrichment of loss-of-function variants in cases vs. controls (p = 0.02). CASKIN1 was also identifeid as an ASD candidate gene based on having a false discovery rate (FDR) < 0.001 following joint analysis of protein-truncating variants, missense variants, and copy number variants in a cohort of 63,237 individuals in Fu et al., 2022. Wilfert et al., 2021 identified two ultra-rare inherited frameshift variants in CASKIN1 that were exclusively transmitted to ASD probands from two independent families. Caskin1-knockout mice were found to exhibit differences in gait, enhanced nociception, anxiety-like behavior, strong freezing responses in contextual and cued-fear conditioning tests, and low memory retention in the Barnes Maze test, suggesting that Caskin1 contributes to a wide spectrum of behavioral phenotypes (Katano et al., 2018).
Molecular Function
Enables identical protein binding activity. Predicted to be involved in signal transduction. Predicted to be active in cytoplasm and membrane. Immunoprecipitation studies in Tabuchi et al., 2002 demonstrated that Caskin1 stably binds to CASK in the brain, while affinity chromatography experiments determined that Caskin1 coassembles with CASK on the immobilized cytoplasmic tail of neurexin-1. Bencsik et al., 2019 found that, In cultured Caskin knockout hippocampal neurons, overexpressed Caskin1 was enriched in dendritic spine heads and increased the amount of mushroom-shaped dendritic spines; furthermore, the authors found that Shank2, a master scaffold of the postsynaptic density, and Caskin1 co-localized within the same complex.
Caskin1 knockout mice exhibited abnormal gait, enhanced nociception, and abnormal anxiety-like behavior relative to their wild-type littermates. In addition, the knockouts exhibited an increased freezing response to fear conditioning, with or without a cue tone, in contextual and cued-tests as well as decreased long-term spatial memory retention in the Barnes Maze test.
References
Type
Title
Author, Year
Additional
Distribution of Caskin1 protein and phenotypic characterization of its knockout mice using a comprehensive behavioral test battery
Model Type:
Genetic
Model Genotype:
Heterozygous
Mutation:
Caskin1-floxed mice were produced by using the embryonic stem (ES) cell line RENKA. To yield heterozygous knock-out (Caskin1+/-) mice, Caskin1^flox/+ mice were crossed with TLCN-Cre mice, by which recombination is induced throughout the whole body.
Allele Type: Knockout
Strain of Origin: C57BL/6N
Genetic Background: C57BL/6N
ES Cell Line: RENKA
Mutant ES Cell Line: Model Source: NIH
Description: The difference in activity level in the social interaction test was only significant during the day (p=0.0167), but not at night (p=0.0569).
Description: Brake phase in hind paws was greater in Caskin1 knockout compared to wildtype mice, and stance width of hind paws was smaller in Caskin1 knockout compared to wildtype mice. In stride duration, brake phase was smaller and propulsion phase was greater in hind paw than in front paw.
Description: In Caskin1 knockout mice, withdrawal latency from thermal stimulation in the hot-plate test was significantly reduced compared to wildtype mice.
Description: In the social interaction test, particle number and activity level were significantly lower in Caskin1 knockout mice than in wild-type mice. Specifically, particle number did not significantly differ in the light period, but particle number during the dark period and activity level in the light period significantly lower in the mutant animals.
Description: Distance traveled and time spent in the light chamber were significantly reduced in Caskin1 knockout mice. However, the first latency to enter the light chamber was significantly shorter in Caskin1 knockout mice.
Description: Distance moved in 6 s following the first and second foot shock (for conditioning of fear memory formation) was significantly higher in the Caskin1 knockout mice than in wildtype mice.
Description: Time spent around the correct hole did not differ significantly between genotypes at 1 day after training, but was significantly shorter in Caskin1 knockout mice 1 month later.
Cued or contextual fear conditioning: Memory of context1
Increased
Description: In the contextual test at 1 day after conditioning, Caskin1 knockout mice exhibited a significant increase in freezing response and a significant decrease in distance traveled relative to the wildtype mice.
Cued or contextual fear conditioning: Memory of cue1
Increased
Description: In the altered context chamber in the cued test (i.e., with the tone), the freezing response was significantly stronger in Caskin1 knockout than in wildtype mice 1 day after conditioning. These differences in the freezing response were also observed without the tone cue in the altered context chamber.
Description: Using the anti-Caskin1/2C antibody, Caskin1 was detected as a 180-kDa protein band in the spinal dorsal horn and hippocampus of wildtype mice, but not in Caskin1 knockout mice. There was no reactivity of Caskin1 in homogenates from cerebrum, heart, pancreas, lung, liver, skeletal muscle, kidney, and testis in Caskin1 knockout mice. When analyzing fractions from the central (cerebrum, cortex, hippocampus, midbrain, cerebellum, olfactory bulb, medulla and spinal ventral and dorsal horn) and peripheral (dorsal root ganglia, and sciatic nerve) nervous systems, neither antibody detected any Caskin1 in in Caskin1 knockout mice.
Exp Paradigm: anti-Caskin1/2C antibody
Description: The signals from both anti-Caskin1 antibodies were significantly reduced in Caskin1 knockout mouse brains and spinal cord. Caskin1 co-localized with pre- and post-synaptic marker proteins synapsin I and PSD-95 in CA1 and the spinal dorsal horn.
