Combined hepatic deletion of A1cf and Rbm47 in double knockout (ARLKO) liver eliminates apoB RNA editing. (A) Western blot analysis of A1CF, RBM47, and APOBEC1 in livers of ARLKO mice, with or without Ad-Apobec1. Actin was used as a loading control. (B) Quantitative PCR analysis of hepatic A1cf and Rbm47 RNAs demonstrating greater A1cf mRNA abundance versus Rbm47 RNA. Data represents mean ± SE of four to five mice per genotype. (**) P-value < 0.01. (C) Quantitative PCR analysis of hepatic apoB RNA, showing no differences by genotype (mean ± SE of three to four mice per genotype). (D) Western blot of hepatic apoB protein in Rbm47LKO, A1cf−/−, and ARLKO mice. Actin was used as a loading control. (E) Western blot detection of FLAG-A1CF following adenovirus A1CF administration to ARLKO mice. The calculated percent C-to-U apoB RNA editing indicates apoB editing frequency at the canonical site C6666. Parentheses indicate number of clones containing the edited site vs. the total number of sequenced clones. (F) In vitro apoB RNA editing assay using liver S100 extracts (lane 1 negative control contains primer and bovine serum albumin only) prepared from either WT mice (lane 2), or from ARLKO mice (lanes 3–8), or using recombinant (rec) APOBEC1 (lanes 9,10), supplemented where indicated with recombinant A1CF (lanes 4,5) or RBM47 (lanes 6–8). ApoB RNA editing was determined by poison primer extension (Materials and Methods). The relative mobility of the unedited (C) and the edited (U) C6666 is indicated to the right. Representative of three independent assays. (G) Baseline hepatic editing efficiency of 3′UTR APOBEC1 RNA targets and hierarchy of RNA binding complementation factors (see Table 1 for data). Wild-type (black dots), Rbm47LKO (green dots), A1cf−/− (blue dots), ARLKO (orange dots), and Apobec1−/− (white dots). RNAs preferentially edited by RBM47 show decreased editing efficiency in Rbm47LKO and ARLKO and no change in A1cf−/−. In contrast, RNAs preferentially edited by A1CF show no change in the absence of A1CF but increased editing frequency in the absence of RBM47 and almost complete loss of editing in the absence of both A1CF and RBM47.