The Trypanosoma brucei RNA-binding protein DRBD18 ensures correct mRNA trans splicing and polyadenylation patterns

Figures

• 

  FIGURE 1.. Effect of DRBD18 RNAi on RBP10 mRNA and expression levels. (A) Example of cumulative growth curves of bloodstream forms with and without induction of DRBD18 RNAi. (B) DRBD18 and RBP10 protein levels after RNAi induction, corresponding to A. Ribosomal protein S9 is the loading control. Anti-RBP10 antibodies (Wurst et al. 2012) were used for detection of RBP10 and a nonspecific band (x). (C) RNA from the samples used in A was used to examine effects on RBP10 mRNA. The methylene blue-stained membrane is shown on the left, with RBP10 hybridization in the center. The membrane was then stripped and hybridized with a β-tubulin probe as a further control; *RBP10 signal that remained after stripping. (D) Visualization of RNA-seq results after 12 h tetracycline-mediated induction of DRBD18 RNAi, using the Integrated Genomics Viewer (Robinson et al. 2011; Thorvaldsdóttir et al. 2013). Three replicates are shown. Note that these results are shown on a linear scale whereas those in Figure 1A are on a log scale. Proposed full length mRNAs are shown below the tracks; positions of (U)12 or (U)6C(U)6 sequences are indicated in cyan but there are numerous other polypyrimidine tracts present. The…

• 

  FIGURE 2.. Effect of MEX67 RNAi on RBP10 mRNA and expression levels. (A) Cell densities in two cultures 24 h after addition of tetracycline to induce MEX67 RNAi. (B) Northern blots using RNA from the two cultures, with detection of rRNA (methylene blue) and of RBP10 and β-tubulin mRNAs.

• 

  FIGURE 3.. Proteins associated with DRBD18. (A) Volcano plot showing proteins that were significantly enriched with DRBD18-TAP relative to TAP-GFP. The x-axis shows log2 enrichment, while the y-axis is −log10 of the P-value generated by the Perseus algorithm (Tyanova et al. 2016). Proteins outside the curved lines are calculated to be significantly enriched according to the method described in Keilhauer et al. (2015), using a double-sided two sample t-test with permutation-based FDR of 0.01, and with parameter S0 (for minimal fold change) set to 0.1. Spots for proteins that are mentioned in the text and showed specific enrichment with ZC3H18 are labeled; these were present in all three DRBD18 preparations, had P-values of less than 0.01, and were greater than fourfold enriched as calculated in Perseus. Spots that are not labeled include ribosomal proteins, abundant enzymes, and RNA processing factors. In addition, we have not labeled spots for proteins with anomalous enrichment values and probabilities due to the replacement, by the Perseus algorithm (Tyanova et al. 2016), of absent values (protein not detected) by random ones. For the DRBD18-GFP comparison, these include ALBA2, DRBD4,…

• 

  FIGURE 4.. Long RBP10 mRNAs remain in the nucleus after DRBD18 depletion. Individual RBP10 mRNAs were detected by single-molecule fluorescent in situ hybridization with or without DRBD18 depletion for 12 or 24 h. In each case, 50 cells were analyzed. (A) Map showing the positions of the probes on the RBP10 mRNA, approximately to scale. The 3′-UTR divisions on the map correspond to those in Figure 2. The bars indicate the extent of mRNAs detected by the probe combinations. The dotted lines indicate that RNAs hybridizing to the different probe sets could have various lengths. The distances next to the bars indicate the position of the probe end on the mRNA. “Exp.”: Experiment. (B) Example image of RBP10 mRNAs in a wild-type cell, detected by single-molecule fluorescent in situ hybridization using probes 1, 2, and 4 (experiment I in panel A). DNA was stained with DAPI. The white arrow indicates an mRNA that hybridizes only to the coding region (open reading frame, ORF) and 3′ probe 2; the yellow arrow points to one of the four mRNAs that hybridize to probes 1, 2, and 4. (C) Mean numbers of long, medium, and short mRNAs per cell under different conditions. The data were from 50 cells. “None”:…

• 

  FIGURE 5.. DRBD18 RNA binding and effects on the transcriptome. (A) Tagged DRBD18 was pulled down from bloodstream-form extracts, then associated (bound) RNAs were released using TEV protease. RNA that did not bind to the affinity beads served as the negative control (“unbound”). The average bound/unbound ratio for three biological replicates is on the y-axis and the annotated 3′-UTR length on the x-axis, both on a log scale. Each spot represents one open reading frame, with just one representative of repeated genes. The orange spots are genes with at least 10 reads in all samples and with at least twofold enrichment in all three comparisons. (B) Annotated mRNA lengths and 3′-UTR lengths for bound mRNAs (bloodstream forms, at least twofold enrichment in all three comparisons) and for mRNAs with, on average, less than onefold enrichment are plotted. The box plot shows the median and the box includes the interquartile range. Whiskers are up to 1.5-fold the inter-quartile ranges, and spots are outliers. (C) The 3′-UTRs of the 163 bound mRNAs (bloodstream forms) were compared with those of a length-matched set of mRNAs with less than onefold enrichment using MEME, using the discriminative mode…

• 

  FIGURE 6.. Effects of DRBD18 RNAi on the RNAs from Tb927.8.1580 and Tb927.8.1590. (A) Results from the Integrated Genome Viewer in the region containing Tb927.8.1580 and Tb927.8.1590. The direction of transcription and a scale bar are at the top. Representative views for all six types of samples are shown. The scales on the y-axis are different (see numbers on the right); the thin lines are drawn at intervals that show read depths of approximately 100. A map of the detected mRNAs is below, as for Figure 1. (B) Effects of RNAi on coding region reads (log2 fold change from DeSeq2). (C) Method to estimate the relative abundances of wild-type and alternatively processed reads. For more detail, see Supplemental Figure S5. Reads that spanned the sites were designated as wild-type, and those with polyadenylation or spliced leader sequences were classified as alternatively processed. “All”—whole-cell RNA; Cyt: cytoplasmic RNA. (D) The upper two bar graphs show numbers of reads (normalized to total data set size) that were either wild-type or alternatively processed at the different poly(A) sites; all of these events are directed by trans splicing at a single downstream site, results for which are…

• 

  FIGURE 7.. Effects of DRBD18 RNAi on additional mRNAs. For each gene, the upper two bar graphs show numbers of reads (normalized to total data set size) that were either wild-type or alternatively processed at the different poly(A) sites, and the lower panel shows mRNAs with or without the 3′-UTR-internal splice site. Details are as in Figure 6.

• 

  FIGURE 8.. Effects of DRBD18 depletion show chromosomal clustering. The effects of DRBD18 depletion (log2 fold change) for unique genes (without paralogs) were plotted as orange bars according to the order of the genes on the chromosomes (positions not to scale). Start sites according to histone modifications (Siegel et al. 2009) and transcript orientation are upward blue bars, and stop sites are downward bars. Neighboring smaller bars indicate the orientation. The scale of the y-axis has been truncated so that all effects greater than 2.8-fold are shown as log2(1.5). Black horizontal arrows indicate the directions of selected affected transcription units.

• 

  FIGURE 9.. Model for the interaction between DRBD18 and RBP10 mRNA. A key is on the Figure, which is explained in the text. RBP: RNA-binding protein.

• 

  Tania Bishola Tshitenge.