The role of the zinc finger protein ZC3H32 in bloodstream-form Trypanosoma brucei

Fig 1. Myc-tagged ZC3H32 is in the cytoplasm. ZC3H32-myc was detected by anti-myc antibody and DNA (nuclei and kinetoplasts) were detected with diamino pimelic acid (DAPI). Differential interference contrast (DIC) images are on the left. The lowest panels show that lambdaN-ZC3H32-myc has a similar location.

Fig 2. No evidence for association of ZC3H32 with polysomes. (A) A typical optical density profile after sucrose density gradient centrifugation of extracts from trypanosomes expressing V5-tagged ZC3H32, and treated briefly with cycloheximide. Fraction numbers are below; the heaviest fraction is fraction 11. (B) Typical Western blot showing the distribution of V5-ZC3H32 for (A), trypanothione reductase (TR) and ribosomal protein S9. (C) Quantitation of (B), with results adjusted for loading. Note the interruption of the scale. (D) As in (B), but the trypanosomes were UV-irradiated prior to lysis in order to bind proteins to RNA. 10 fractions were obtained. (E) Quantitation of (D)—as in C. Results are also similar to (C). (F) As (B) but 100 μg/ml RNase was included in the polysome buffer. (G) Quantitation of (F).

Fig 3. ZC3H32 is essential in bloodstream-form trypanosomes. (A) Results with a trypanosome clone expressing the tet-repressor and with a V5 tag sequence integrated into the genome in frame with ZC3H32 (S2B Fig). RNAi was induced by addition of tetracycline, in 3 clones, and cells were diluted as necessary. Cumulative cell numbers after various days of incubation (means and standard deviation) are shown. (B) Expression of V5-ZC3H32 after induction of RNAi—control for (A). Expression is decreased on day 1, but has recovered on days 4–5; cell growth is also restored (panel A). Antibody to ribosomal protein S9 serves as a loading control. (C) Cells containing an inducible copy of ZC3H32-myc, and no endogenous copies (S2E Fig) were grown with or without tetracycline. ZC3H32-myc expression was lost after withdrawal of tetracycline. Cumulative cell numbers after various days of incubation are shown for three clones. Cells were diluted as necessary. (D) Expression of ZC3H32-myc after withdrawal of tetracycline, experiment (C). (E) Growth of two cloned procyclic-form cell lines with tetracycline-inducible expression of ZC3H32-myc. (F) Western blots showing expression of the myc-tagged protein…

Fig 4. Most ZC3H32 is not associated with MKT1. (A) Trypanosomes expressing V5-ZC3H32 from the endogenous locus, or with induced (RNA- polymerase-1-mediated) expression of ZC3H32-myc, with or without YFP-tagged MKT1. Extracts were immunoprecipitated with anti-GFP and proteins detected by Western blotting. The loading (in cell-equivalents relative to the input) is indicted below the blots. An artificially darkened image (increased exposure but no contrast adjustment) is also shown for V5-ZC3H32. (B) As (A), but with Immunoprecipitation of V5-ZC3H32 and detection of YFP-MKT1. (C) Anti-myc immunoprecipitation of extracts from cells expressing V5-4EIP, with or without ZC3H32-myc.

Fig 5. Tethering of ZC3H32 to a reporter mRNA represses expression. (A) Principle of the tethering assay. The reporter mRNA has a chloramphenicol acetyltransferase (CAT) open reading frame. The capped spliced leader is shown as a circle. Five boxB sequences (shown as a single lollipop) bind to the lambdaN peptide. In our case the lambdaN peptide is at the N-terminus of a ZC3H32 open reading frame with a C-terminal myc tag. The negative control reporter (lower diagram) has no boxB. (B) Effect on CAT enzyme expression from tethering lambdaN-ZC3H32-myc to the CAT reporter mRNA. CAT activity is shown as a proportion of that seen in cells with no tethered protein. Results are arithmetic mean ± standard deviation for three clones. (C) Effect on CAT mRNA of tethering lambdaN-ZC3H32-myc in the 3'-UTR. A typical Northern blot is shown. (D) Expression of lambdaN-ZC3H32-myc in the six different clones.

Fig 6. Both the N- and C-terminal portions of ZC3H32 repress expression in the tethering assay. A map of the complete tagged lambdaN-ZC3H32-myc protein is at the top, with the Cx7Cx5Cx3H zinc fingers in red, the Cx9Cx5Cx3H zinc finger in orange, and the putative MKT1 binding motif (NNPY from 541–545) in blue. Below, maps of expressed proteins are on the left and the effect on the reporter are on the right; cyan is the cell line without tetracycline and dark blue, the line with tetracycline. Western blots showing expression of the myc-tagged proteins are on the right. Only the anti-myc panels are shown; loading is shown in S4 Fig. Sometimes, there was easily detectable expression in the absence of tetracycline. (A) shows CAT activities relative to the amount with no tethered protein at all (top bars). Results are mean and standard deviations for 3 independent clones except for fragments 1–253 and 211–655, where only 1 and two clones, respectively, were measured. (B) shows relative RNA amounts, measured by quantitative Northern blotting using rRNA or 7SL RNA as standards. Note that the Western blot used for construct 211–655 is identical to that in panel (A), since this was the same…

Fig 7. Tethering of ZC3H32 represses translation initiation. (A) Polysome profile of cells expressing the CAT reporter from Fig 5A without box B, and with expression of lambdaN-ZC3H32-myc. The top panel is the optical density reading across the gradient. Fraction 1 has the lowest sucrose concentration and fraction 14 the highest. A synthetic ß-globin RNA was added to the fractions before RNA preparation, as a control for the RNA purification efficiency. Below are the results from a typical Northern blot for the gradient fractions, hybridised with probes for CAT, histone H4 (HISH4) and ß-globin. (B) As (A), but with CAT reporter from Fig 5A with box B. (C) Percent of CAT RNA in each fraction, relative to the total signal of the whole gradient. The signals on the Northerns were measured from the phosphorimager files. Numbers of experiments are indicated and results are arithmetic mean ± standard deviation. (D) As (C) but for the control HISH4 mRNA.

Fig 8. Transcriptome results. (A) ZC3H32 RNAi was induced for 24h and the transcriptome (single experiment) was compared with two wild-type control transcriptomes using DESeq2. The result for each coding region (RNAi/Wild-type, log2-transformed) is on the y-axis and log2-transformed mRNA lengths (as annotated in TritrypDB) on the x-axis. The regression line (plotted using Microsoft Excel) and correlation coefficient are also shown. (B) Tandem-affinity tagged ZC3H32 was purified on an IgG column. RNA was purified from the bound and unbound fractions and analysed by RNASeq. The bound/unbound ratio was calculated based on the normalized read counts (reads per million reads). The graph is as in (A), except that shows the log2-transformed average ratio for triplicate experiments is on the y-axis.