Fig. 2

miR-21 upregulates YAP expression via inhibiting RUNX1. a heat map of 50 genes of dataset GSE74706; b downstream genes of miR-21 and significant differentially expressed genes of dataset GSE74706 predicted by mirDIP and starBase and the Venn map of human transcription factor obtained from Cistrome, with BCL11A, E2F3, ERG, ETS1, KLF5, MEF2A, RUNX1 and TTF2 as the intersection genes; c PPI map of RUNX1 and its related genes obtained from GeneMANIA; d Venn map of genes related to RUNX1 obtained from GeneMANIA and target genes of RUNX1 predicted by Cistrome in lung adenocarcinoma, with YAP1 (YAP) as the only intersection gene; e correlation of RUNX1 and YAP1 in lung adenocarcinoma and squamous cell carcinoma of TCGA database analyzed by GEPIA (P = 2.3e−14); f binding site sequences of miR-21 and RUNX1; g relationship between miR-21 and RUNX1 measured by dual luciferase reporter gene assay, * p < 0.05 vs. miR-21; h binding of miR-21 and RUNX1 analyzed by RIP, * p < 0.05 vs. anti-IgG1; i the expression of RUNX1 and YAP in tumor tissues measured by immunohistochemistry, * p < 0.05 vs. normal mice; j the expression of RUNX1 after transfection in tissues; k the expression of YAP when RUNX1 was silenced; l the binding of RUNX1 and YAP detected by ChIP assay, * p < 0.05 vs. IgG; m the binding of RUNX1 and YAP verified by dual luciferase reporter gene assay; n the expression of miR-21 in cells measured by RT-qPCR; o the expression of RUNX1 and YAP in cells measured by western blot analysis; p the correlation of miR-21 and RUNX1, miR-21 and YAP, RUNX1 and YAP analyzed by Pearson. Measurement data were expressed as mean ± standard deviation. When data were homogeneous and of normal distribution, data between two unpaired groups were compared by unpaired t-test. Measurement data among multiple groups were checked by ANOVA with Tukey’s post hoc test. Pearson’s correlation analysis was used to analyze the relationships between miR-21, RUNX1 and YAP