Figure 5: eIF1A augments miR-451 biogenesis.
From: eIF1A augments Ago2-mediated Dicer-independent miRNA biogenesis and RNA interference
(a) In vitro activity of eIF1A in Ago2-mediated cleavage of 5′-32P end-labelled synthetic pre-miR-451. RNAs were analysed by northern blotting and autoradiographs analyses. (b) Statistical analyses of the data in a. Data of three independent experiments were shown (mean±s.d., n=3, Student’s t-test). (c) Northern blot assays show the effects of eIF1A mutants on miR-451 biogenesis in HEK293 cells. Total RNAs were extracted from stable HEK293 cells expressing Flag-eIF1A and Flag-eIF1A mutants. Signals show mature miR-451 and miR-144 levels with U6 as a loading control. (d) Northern blot assays show the effects of eIF1A(K56A) mutant on miR-451 generation in the developing zebrafish. Total RNAs were extracted from zebrafish with/without treatments of control MO, eif1axb MO (62.5 μM, 1 nl), eif1axb MO (62.5 μM, 1 nl) plus heIF1A (mRNA, 1 μg μl−1, 1 nl), eif1axb MO (62.5 μM, 1 nl) plus heIF1A (K56A) (mRNA, 1 μg μl−1, 1 nl). (e) Haemoglobin (brown) was visualized by the oxidation of o-dianisidine (o-das) in zebrafish at 48 h.p.f. Haemoglobinized cells accumulate in wild-type (group I) and are reduced in eif1axb MO-injected embryos (group II). Human eIF1A, but not K56A mutant (group III, with less haemoglobinized cells), rescues haemoglobinized cells in eif1axb MO-injected zebrafish, while miR-451 spares K56A-caused rescue defects. Scale bar, 100 μm. (f) Percentage of embryos with haemoglobinized cells in zebrafish. Control group (n=40). eif1axb MO injection (n=42). Human eIF1A (n=35), but not K56A mutant (n=35), increased the percentage of embryos with normal haemoglobinized cells, while miR-451 (n=31) suppresses the decreased percentage by K56A mutant (P<0.001). (g) Erythrocytes stained with May-Grünwald/Giemsa are representative of the mean of each group at 56 h.p.f. Scale bar, 5 μm. (h) Scatterplot of the nuclear cytoplasmic ratio (N:C) for each group as a readout of erythrocyte maturation (Wilcoxon rank-sum test after Bonferroni correction, P<10−15). Human eIF1A (n=120 erythrocytes), but not K56A mutant (n=145), rescued erythrocyte maturation impaired by eif1axb MO injection, while miR-451(n=118) spares distribution difference in the heIF1A(K56A) mutant group. ***P<0.01.
