Fig. 6: E3 ligase loss or mutation increases APOBEC signature mutations.
A Schematic of mutREAD sequencing to detect APOBEC signature mutations. UNG2-deficient RKO cells expressing DOX-inducible Cas9 and with or without mCherry-P2A-3xHA-A3H-I overexpression were transduced with sgRNAs targeting UBR4, UBR5, or HUWE1. Following sorting for sgRNA-positive cells, gene editing was induced with DOX for up to 10 days, after which genomic DNA was isolated for mutREAD sequencing. B Fraction of APOBEC signature related mutations over all identified mutations in mutREAD sequencing control samples (n = 3 biological replicates, two-sided Fisher’s exact test, no adjustments for multiple comparisons, test statistic = Odds Ratio = 0.5588, p-value = 4.41 × 10−72, CIlow = 0.5239, CIhigh = 0.5960). C Best-subset signature refitting of samples expressing exogenous A3H-I, averaged per genotype, using APOBEC-associated and colon carcinoma signatures. Each bar represents the mean of three technical replicates scaled to one. D Fraction of APOBEC-related mutations in samples with exogenous A3H-I expression (n = 3 biological replicates, two-sided Fisher’s exact test with Bonferroni correction; odds ratio, 95% confidence intervals, and exact p-values shown). E Pentanucleotide context preference of APOBEC mutations in control samples expressing A3H-I. F Schematic of mutational signature analysis using PCAWG data. G TCGA and ICGC cancer samples were grouped by wild-type or mutated status of UBR4, UBR5, and HUWE1. Mutational signatures were normalized to the total number of mutations in each sample. The level of SBS13 APOBEC signature was compared between the groups. “E3scomb” are all samples in which at least one of the E3s is mutated. Wilcoxon rank-sum test, two-sided, ns: p > 0.05, n = 2703 samples). Box plots show median, interquartile range, and whiskers to 1.5× IQR; y-axis is log10-transformed. H Model: lack of RNA-binding determines A3 nuclear localization and simultaneously targeting by th eE3 ligases, ensuring low nuclear A3 levels.
