ADAR1 regulates the time-dependent differences in the alternative splicing of the human ABCB1 gene.A, schematic diagram of the retention of intron 27 of the ABCB1 gene. Intron-retaining ABCB1 mRNA may be degraded through nonsense-mediated mRNA decay (NMD). B, ADAR1 is involved in the time-dependent differences in the production of retained-intron 27 ABCB1 transcripts. Cells were transfected with the ex27–ex28 minigene and treated with DEX. Minigene-derived normal splicing transcripts and retained-intron transcripts were assessed by semiquantitative RT-PCR at the indicated time points after DEX treatment. The percentage of retained intron was calculated by dividing the signal intensity of retained intron by the summed signal intensity of retained intron and normal splicing transcripts. Values are the mean with S.D. (n = 3). ∗p < 0.05; significant difference between the two groups (F3,8 = 11.373, p = 0.003; ANOVA with Tukey–Kramer’s post hoc test). C, ADAR1 binds to intron 27 of ABCB1 pre-mRNA. RPTECs were transfected with the ABCB1 ex27–ex28 minigene, and then RIP assay was conducted at 24 h posttransfection. Upper panel shows western blotting analysis of ADAR1-RIP immunoprecipitates. Lower panel shows quantitative real-time RT-PCR analysis of ADAR1-RIP immunoprecipitates using intron 27 specific primers listed in Table 1. Values are the mean with S.D. (n = 3). ∗∗p < 0.01; significant difference between the two groups (t4 = 12.964, unpaired t-test, two-sided). D, retained-intron 27 transcripts of the ABCB1 gene are degraded by NMD. Cells were transfected with the ex27–ex28 minigene and treated with 1 μM NMDI-14 for 6 h. Minigene-derived normal splicing transcripts and retained-intron transcripts were assessed by semiquantitative RT-PCR. Values are the mean with S.D. (n = 4). ∗∗p < 0.01; significant difference between the two groups (t6 = 8.022, unpaired t-test, two-sided). E, the stability of retained-intron 27 transcript of the ABCB1 gene. Cells were transfected with the ex27–ex28 minigene. Minigene-derived normal splicing transcripts and retained-intron transcripts were assessed by semiquantitative RT-PCR. The mRNA levels were normalized to that of 18S rRNA. Values are the mean with S.D. (n = 3). The value at 0 h (the time of the initiation of ActD) was set at 1.0. ∗p < 0.05; significant difference between the two groups at the corresponding time point (F5,12 = 13.641, p < 0.001; ANOVA with Tukey–Kramer’s post hoc test). F and G, NMD inhibition restores levels of ABCB1 mRNA and P-gp in ADAR1-KD RPTECs. Mock-transduced and ADAR1-KD cells were treated with 1 μM NMDI-14 for 24 h. The mRNA levels were assessed by quantitative real-time RT-PCR, and their levels were normalized to that of 18S rRNA. The levels of P-gp were normalized that of β-ACTIN. Values are the mean with S.D. (n = 4). The value of vehicle (0.01% DMSO)-treated mock cells was set at 1.0. ∗∗p < 0.01, ∗p < 0.05; significant difference between the indicated groups (F3,12 = 15.727, p < 0.001 for ABCB1 mRNA; F3,12 = 37.583, p < 0.001 for P-gp; ANOVA with Tukey–Kramer’s post hoc test). H, intracellular accumulation of digoxin in NMDI-14-treated ADAR1-KD RPTECs. Concentrations of digoxin were measured by LC-MS/MS analysis 1 h after incubation with the drug. Values are the mean with S.D. (n = 6). ∗∗p < 0.01; significant difference between the two groups (t10 = 4.868, unpaired t-test, two-sided).