Fig. 5
DHX9 binds downstream rG4 motifs in human cells. a Phosphorimage of SDS gel (left gel) resolving 32P-labeled RNAs crosslinked to DHX9. Immunoprecipitated samples prepared from HeLa cell lysates without UV light (254 nm) treatment are shown as a negative control. Immunoblot of the same membrane (right) probing DHX9 confirm the presence of DHX9 in the RNA-protein complex. b Gscore (red) and GC content (blue) of sequences around the center of iCLIP peaks. c Distribution of rG4 motifs, of the form G2 N3 (sequences of the form G2+N1-3G2+N1-3G2+N1-3G2+ where N is any base), around the center of iCLIP peaks. The gray area represents the median width of iCLIP peaks. d Consensus motif of rG4-forming sequences found upstream DHX9 iCLIP peaks. e rG4 and dsRNA predicted folding energies within 200 bases from the center of 5′-UTR DHX9 iCLIP peaks. Minimum folding energies were calculated using a sliding window of 35 nt. f Mapped DHX9 binding sites (red) and normalized ribosome density in control (black) and DHX9 depleted cells (green) within the DDX23 transcript. Depletion of DHX9 leads to an increase and a decrease of RPF within the 5′-UTRs, upstream of DHX9 binding sites, and associated CDSs respectively. Biophysical characterization of the DDX23 rG4 motif is reported in Figure S12 in Additional file 1. Fold changes in g TE and h RPFdist of transcripts directly bound by DHX9 in their 5′-UTR. Data are means ± s.e.m.; P values were assessed using two-sample Kolmogorov–Smirnov tests and compare the reported condition to background. *P < 0.05, **P < 0.01, ***P < 0.001