Fig. 3: EZH2-92aa translation is regulated by DDX3.

a MES28 GSCs were subjected to RNA immunoprecipitation using an anti-DDX3 antibody or IgG (negative control). Immunoprecipitates were analysed by RT-qPCR with specific primers for the circEZH2 IRES sequence (P = 3.35e−05), GAPDH (negative control, P = 0.0636) and BCL2 (positive control, P = 7.09e−04). b DDX3 and circEZH2 RNA levels in MES28 and GSC23 cells transfected with DDX3 siRNAs. MES28, si-DDX3-1, P = 8.66e−10, si-DDX3-2, P = 9.31e−08; GSC23, si-DDX3-1, P = 5.77e−05, si-DDX3-2, P = 4.06e−05. c Protein levels of DDX3 and EZH2-92aa in MES28 and GSC23 cells transfected with DDX3 siRNAs. d Expression of DDX3 in GBM datasets from TCGA (GBM and LGG dataset, plotted by GEPIA2, [http://gepia.cancer-pku.cn/]) and CGGA (dataset mRNAseq_325, [http://www.cgga.org.cn/index.jsp]). Data are presented as boxes containing the median (centre line), the first and third quartiles (box limits). The whiskers indicate the maxima and minima. One-way ANOVA test. For TCGA dataset, *P < 0.01. For CGGA dataset, P = 0.00078. e Protein levels of DDX3 and EZH2-92aa in NHAs and several patient-derived GSC lines. The data in (a), (b), (c) and (e) are pooled from three independent experiments. The data are presented as the mean ± SD. Unpaired two-tailed Student’s t test was used to determine the significance of differences between the indicated groups. *P < 0.05; **P < 0.01; ***P < 0.001. Source data are provided as a Source data file.