Fig. 1

Master microRNA-222 regulates cardiac microRNA maturation and triggers Tetralogy of Fallot. a Schematic illustration of the screening process for differentially expressed microRNAs (fold change >2, P < 0.05) using two independent datasets (GSE35490 and GSE40128). b RT-qPCR analysis of miR-222 levels in the RVs of aborted fetuses with TOF (n = 5) and normal controls (n = 5). c Abundance of upregulated microRNAs (fold change >2, P < 0.05) in TOF samples vs. normal using the GSE40128 dataset. d Abundance of miR-222, miR-1, miR-133a, and miR-195 in TOF samples vs. normal using the GSE40128 dataset. Abundance was calculated by the probe signal levels of indicated miRNA divided by total miRNA. e Temporal analysis of miR-222 expression during normal human heart development by microarray analyses (upper, left) and mouse heart development using the GSE105834, GSE82960, GSE105910, GSE82604, GSE82822, GSE82942, and GSE101175 datasets (lower, left) and differentiation of hESCs into CMs by RNA-seq (right). f (I–X) Stereoscopic images of whole hearts (I, II) and H&E-stained heart sections (III–X) from homozygous mir-222 KI and control mice, displaying human TOF-like phenotypes, such as (I) the control heart shows a normal caliber ratio of the pulmonary artery (PA) and aorta (Ao); (II) A mir-222 KI littermate of the animal in (I) shows narrowing of the pulmonary artery, RV hypertrophy (IV), overriding aorta (arrow, VI), VSDs (arrow VIII), and small aortic root (star, X) at P0.5; The scale bars in (I–IV) and (V–X) are 500 µm and 200 µm, respectively. g Schematic illustration of the screening approach for target genes of miR-222 using TargetScan prediction, RNA-seq, MGI database, and RT-qPCR verification. h Luciferase assays of H9C2 or HEK293T cells cotransfected with miR-222 or scramble control and luciferase reporter plasmids containing WT or mutant DICER1 3′UTR (upper) and AGO2 (lower) 3′UTR. i Biotin-labeled miR-222 and scramble control were transfected into hESC-CMs and the human DICER1 (upper, n = 3) or AGO2 (lower, n = 4) 3′-UTR pulled down by miR-222 or scramble control was quantified by RT-qPCR. j, k RT-qPCR and western blotting analysis of mRNA (j) and protein (k) levels, respectively, of DICER1 (upper) and AGO2 (lower) in hESC-CMs transfected with miR-222 or scramble control. l RT-qPCR analysis of the mRNA levels of DICER1 (left) and AGO2 (right) in the RVs of aborted fetuses with TOF and control fetuses (n = 5). m RT-qPCR analysis of the mRNA levels of Dicer1 and Ago2 in hearts of P0.5 neonatal mice of the indicated genotypes (n = 3). n RT-qPCR analysis of the mature miRNA levels of miR-1, miR-133a, miR-195, miR-451a, and Let-7 in the RVs of aborted fetuses with TOF and control fetuses (n = 5) (left); RT-qPCR analysis of mature mmu-mir-1 and mmu-mir-451a levels in hearts of P0.5 neonatal mice of the indicated genotypes (n = 3) (right). o RT-qPCR analysis of the levels of the mature forms of DICER1-dependent miRNAs (miR-1, miR-133a, miR-195) and AGO2-dependent miRNA (miR-451a) in hESC-CMs transfected with miR-222 or scramble control only (left) or cotransfected with miRNA and the indicated expression plasmids (middle and right) for 72 h (n = 4). p Immunofluorescence staining of WGA in heart sections from P0.5 neonatal mice and quantification of the cross-section area of CMs (upper). Immunofluorescence staining of the hESC-CMs markers α-actinin (red) and cTnT (green) in miR-222- or scramble control-transfected hESC-CMs and quantification of the surface area of hESC-CMs (lower). q Immunofluorescence staining (left) of α-actinin (red) in hESC-CMs expressing exogenous pri-miR-222 and transfected with the indicated genes and quantification (right) of the surface area of CMs. r Immunofluorescence staining of the hESC-CMs markers α-actinin (red) and cTnT (green) in miR-222- only, miR-222- and miR-133a- or scramble control-transfected hESC-CMs and quantification of the surface area of hESC-CMs; at least 100 cells were quantified in each group. The scale bars in (p, upper), (q, left), and (p, lower, r) represent 20 µm, 10 µm, and 50 µm, respectively. s FACS analysis and quantification of lipid ROS levels (C11-BODIPY) in hESC-CMs transfected with miR-222 or scramble control by Lipofectamine RNAimax with or without Fer-1 treatment. t Representative immunofluorescence staining of the lipid ROS marker 4-HNE in heart sections from P0.5 mice and normal controls with the indicated genotypes. u Representative immunofluorescence staining of NKX2-5 and the lipid ROS marker 4-HNE in heart sections from RV tissues from aborted fetuses with TOF (n = 3) and normal controls (n = 3). v H&E-stained heart sections from control mice and homozygous mir-222 KI treated with saline, enoxacin, or Fer-1 as indicated, which displayed human TOF-like phenotypes, such as pulmonary stenosis (VI), overriding aorta (arrow, X), and VSDs (arrow in II) at E13.5. The scale bars in (I–IV) and (V–XII) represent 500 µm and 100 µm, respectively. w Schematic diagram of the role of the miR-222-DICER1-AGO2 axis in the pathogenesis of TOF. miR-222: has-miR-222-3p (MI0000299); miR-187: has-miR-187-3p (MI0000274); miR-93*: has-miR-93-3p (MI0000095); miR-155: has-miR-155-5p (MI0000681); miR-499: has-miR-499-5p (MI0003183); miR-1: has-miR-1-3p; miR-133a: has-miR-133a-3p; miR-195: has-miR-195-5p; miR-451a: has-miR-451; mir-222: mmu-mir-222-3p; CMs: cardiomyocytes; +/+: WT; + /KI: heterozygous H11-Myh7-mir-222 knock-in, KI/KI: homozygous H11-Myh7-mir-222 knock-in; Ao: aorta; PA: pulmonary artery; RV: right ventricle; LV: left ventricle; IVS: interventricular septum; Rel.: relative; Luc.: luciferase; Ctrl: control. For all immunofluorescence staining experiments, DAPI was used for nuclear staining (blue). U6 or GAPDH was used as an internal control. Data are shown as means ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001