Supplementary Data from DDX3 Activates CBC-eIF3–Mediated Translation of uORF-Containing Oncogenic mRNAs to Promote Metastasis in HNSCC
Figure S1. DDX3 expression is correlated with cell invasive activity and ATF4 expression. Figure S2. Microarray-identified genes are enriched for pro-metastatic genes and ATF4 targets, and DDX3 is required for migration and ATF4 expression in all of the HNSCC cell lines examined. Figure S3. DDX3 and ATF4 are required for EMT gene expression. Figure S4. DDX3 does not regulate ATF4 protein stability. Figure S5. DDX3 regulate ATF4 translation independent of ER Stress. Figure S6. The DDX3/CBC/eIF3 complex directly activates the translation of uORF-containing mRNAs in vitro and in vivo. Figure S7. RT-qPCR analysis of luciferase mRNAs in the in vivo translation assay. Figure S8. The CBC and eIF3 complexes are involved in the translation of uORF-containing mRNAs. Table S1. List of sense shRNA or siRNA sequences Table S2. Univariate and multivariate cox regression analyses in HNSCC patient samples. Table S3. List of genes that showed more than two-fold changes upon DDX3 knockdown. Table S4. GO terms enriched in DDX3 knockdown-upregulated or -downregulated genes analyzed by DAVID software. Table S5. List of upstream transcription factors potentially involved in DDX3-regulated genes predicted by IPA software. Table S6. Percentages of siDDX3-downregulated, DDX3-upregulated or non-target genes overlapping with uTIS or non-uTIS genes from TIS-db or with uORF-containing genes from uORFdb. Table S7. Percentages of CBP20 CLIPs or eIF3 CLIPs overlapping with DDX3 CLIPs or eIF4A3 CLIPs. Table S8. Percentages of DDX3 CLIPs, CBP20 CLIPs, eIF3 CLIPs or eIF4A3 CLIPs overlapping with uTIS or non-uTIS genes from TIS-db.