DDX3 depletion represses translation of mRNAs with complex 5' UTRs

Abstract

DDX3 is an RNA chaperone of the DEAD-box family that regulates translation. Ded1, the yeast ortholog of DDX3, is a global regulator of translation, whereas DDX3 is thought to preferentially affect a subset of mRNAs. However, the set of mRNAs that are regulated by DDX3 are unknown, along with the relationship between DDX3 binding and activity. Here, we use ribosome profiling, RNA-seq, and PAR-CLIP to define the set of mRNAs that are regulated by DDX3 in human cells. We find that while DDX3 binds highly expressed mRNAs, depletion of DDX3 particularly affects the translation of a small subset of the transcriptome. We further find that DDX3 binds a site on helix 16 of the human ribosomal rRNA, placing it immediately adjacent to the mRNA entry channel. Translation changes caused by depleting DDX3 levels or expressing an inactive point mutation are different, consistent with different association of these genetic variant types with disease. Taken together, this work defines the subset of the transcriptome that is responsive to DDX3 inhibition, with relevance for basic biology and disease states where DDX3 is altered.

Figure 1.

Translational changes upon DDX3 depletion. (A) A workflow of the ribosome profiling experiments. (B) siRNA knockdown efficiency of DDX3 analyzed by western blot. Mock: untreated. NTC: nontargeting control siRNA. (C) The log₂ fold change (log₂FC) in ribosome profiling or RNA levels are plotted for all genes. Number genes in each category are indicated with selected genes labeled. Point size indicates the P-value of a significant change in translational efficiency. (D) Tracks showing RNA-seq (RNA) or ribosome profiling (RP) reads that map to the NT5DC2 mRNA. NT5DC2 is an mRNA with differential ribosome occupancy in (C). (E) Western blot analysis of nontargeting (siNTC) or siDDX3 translation lysate samples with antibodies indicated. (F) Renilla luciferase luminescence from in vitro transcribed reporter RNAs translated in vitro in siNTC or siDDX3 HEK 293T lysates. (G) Reporter RNAs as in (F) transfected into HCT 116 cells with or without IAA treatment to induce the DDX3 degron.

Figure 2.

DDX3-sensitive transcripts have complex 5′ leaders. (A) Strength of individual random forest model features that predict TE with R = 0.54 (Supplementary Figure S2A, Materials and Methods). Features in blue are plotted in (B). (B) The translation efficiency (TE) in wild-type cells, CERT motif score, GC-content of 5′ UTRs, and GC content of coding sequences in the indicated gene sets from Figure 1C were computed and are plotted as a density plot based on their importance in the random forest model. Vertical lines are medians. P-values for Wilcoxon rank sum and effect sizes from Cliff's delta (d) versus mixed_ns indicated. (C) The fold-change of the ratio in ribosome occupancy in the 5′ UTR versus the coding sequence upon DDX3 depletion as a density plot for transcripts in the indicated sets. A larger 5′ UTR skew value means that there are more ribosomes in the 5′ UTR compared to the coding sequence. P-values from Wilcoxon rank sum and effect sizes from Cliff's delta (d) versus mixed_ns indicated. (D) An example gene (HMBS) with increases in 5′ UTR ribosomes versus coding sequence ribosomes with tracks as in Figure 1D.

Figure 3.

DDX3 binding sites on rRNA identified by PAR-CLIP. (A) A workflow of the PAR-CLIP experiment. (B) PAR-CLIP T>C conversion locations on human rDNA (top) compared to iCLIP coverage from Oh et al. 2016 (bottom). Boxed regions refer to processed rRNA transcripts. (C) PAR-CLIP T>C conversion density on the 18S rRNA is visualized from gray to blue on the structure of the 48S ribosome (PDB 6FEC). The peak in (B) is contained in the helix in the upper left in (C), which is h16 of rRNA. Yellow: translation factors; blue–gray: rRNA; pink: ribosomal proteins.

Figure 4.

DDX3 binds to structured 5′ leaders of mRNA. (A) Sum of the DDX3 PAR-CLIP peaks mapping on different gene types and regions. (B) A metagene plot of DDX3 PAR-CLIP across all genes and PAR-CLIP data from other RNA binding proteins. eIF3b is a canonical initiation factor, FMR1 binds elongating ribosomes, and MOV10 is a 3′ UTR binding factor. TSS: transcription start site, TES: transcription end site. (C) T>C conversion specificity averaged across all PAR-CLIP peaks across indicated mRNA regions as in (B). (D) PAR-CLIP, ribosome profiling, and RNA-seq across part of the ODC1 gene. Blue peaks in the PAR-CLIP track indicate T>C conversion events. (E) RNA structure in a window of 100 nucleotides around PAR-CLIP peak summits in 5′ UTRs. (F) CERT motif scores averaged across PAR-CLIP peak summits in 5′ UTRs or for shuffled positions. (G) PAR-CLIP T>C conversion specificity in transcript regions in indicated gene sets from Figure 1.

Figure 5.

Translation changes caused by R326H mutant DDX3. (A) Western blots of degron DDX3 cells treated with IAA and transfected with empty vector or the indicated constructs. (B) Ribosome profiling and RNA-seq of DDX3 degron cells treated with IAA and transfected with either DDX3 wild-type or R326H mutant. (C) Western blots of cells treated with siDDX3 and transfected with the indicated constructs. (D) In vitro translation performed with indicated reporter RNAs as in Figure 1 in the lysates from panel (C). Reporter sequences are in Supplementary Table S4.

Figure 6.

A model of DDX3 in translational control. DDX3 binds to the small subunit via helix 16 (h16). Transcripts that are poorly translated in normal cells and that harbor increased intramolecular RNA structure (blue transcript; left) are sensitive to DDX3 depletion (right). Other mRNAs that are initially highly translated (black transcript) are unaffected.