SGK1 mutations lead to aberrant splicing and hyperstable protein isoforms. (A) Distribution of mutations across the SGK1 locus from 5 published sequencing studies. The variant nucleotides depicted were identified in 5 or more cases. (B) Distribution of high-frequency variant nucleotides (identified in 10 or more cases) at mutation hotspots flanking exon 2. The magnified region shows the position of the 2 most frequent mutations highlighted in red. (C) Agarose gel electrophoresis showing polymerase chain reaction amplicons from 5' RACE performed on CRISPR-edited U2932 control and splice-mutant clones. The transcript neoisoform identified from Sanger sequencing of each band is shown. (D) Representative Sashimi plots from analysis of RNA sequencing data showing splicing across SGK1 intron 1. Data from a representative case are shown for each genotype. The number of reads corresponding to each splice junction is indicated. (E) An intron 1 retention score was calculated for each patient classified one of the genotypes. Significance was calculated using a Kruskal-Wallis test and pairwise comparisons with the Wilcoxon rank sum test. Adjusted P values: ***P < .001; ****P < .0001. (F) Immunoblot showing expression of SGK1 protein isoforms in multiple CRISPR-edited Defauw control and intronic splice mutant clones. The lower band is nonspecific (ns). (G) Predicted amino acid sequence from open reading frames identified in the WT and 2 aberrantly spliced neoisoforms. Predicted translational start sites are highlighted in yellow, the SGK1 kinase domain in pink, and the ubiquitination domain in orange. (H) Half-life study of SGK1 protein isoforms in CRISPR-edited control or splice-mutant clones showing immunoblot at the indicated times after addition of cycloheximide. (I) Immunoblot showing SGK1 protein isoform expression in Defauw-Cas9 cells transduced with single gRNAs targeting the indicated SGK1 exons. NTC, nontargeting control single gRNA.