Roles of a Trypanosoma brucei 5'->3' exoribonuclease homolog in mRNA degradation

Abstract

The genome of the kinetoplastid parasite Trypanosoma brucei encodes four homologs of the Saccharomyces cerevisiae 5'-->3' exoribonucleases Xrn1p and Xrn2p/Rat1p, XRNA, XRNB, XRNC, and XRND. In S. cerevisiae, Xrn1p is a cytosolic enzyme involved in degradation of mRNA, whereas Xrn2p is involved in RNA processing in the nucleus. Trypanosome XRND was found in the nucleus, XRNB and XRNC were found in the cytoplasm, and XRNA appeared to be in both compartments. XRND and XRNA were essential for parasite growth. Depletion of XRNA increased the abundances of highly unstable developmentally regulated mRNAs, perhaps by delaying a deadenylation-independent decay pathway. Degradation of more stable or unregulated mRNAs was not affected by XRNA depletion although a slight decrease in average poly(A) tail length was observed. We conclude that in trypanosomes 5'-->3' exonuclease activity is important in degradation of highly unstable, regulated mRNAs, but that for other mRNAs another step is more important in determining the decay rate.

FIGURE 1.

Diagrammatic representation of T. brucei XRN exonuclease domain structures. The regions shown correspond to residues 1–671 of XRNA, 1–784 (XRNB), 1–557 (XRNC), and 1–600 (XRND). The lengths of the C-terminal domains not included in the figure are indicated on the right. The 5′-3′ exonuclease domain PF03159.7 was identified as follows: XRNA: 4e-130 (residues 1–229); XRNB: 1.1e-77 (residues 1–327); XRNC: 5e-38 (residues 1–245); XRND: 9.3e-104 (residues 1–251). XRNB has predicted zinc finger (SM00547 8.3e-09) from residues 976–1000. The conserved 5′→3′ exonuclease sequences are represented as black bars. Insertions that are unique to the individual proteins are shown with vertical stripes (XRNA), white (XRND), horizontal stripes (XRNB), or cross-hatches (XRNC). The XRNC sequence diverges considerably after residue 470, despite isolated 4–8-residue zones of conservation, so the C-terminal segment is shown in dark gray. The other Kinetoplastid sequences showed similar organization.

FIGURE 2.

(A) Effects of RNAi on XRNA and XRNB. Whole trypanosome lysates were analyzed by Western blotting using antisera to an XRNA peptide, with a cytosolic marker protein as control. Total trypanosome RNA was examined by Northern blotting with an XRNB probe, with the SRP RNA as control. The life-cycle stage is indicated as procyclic (PC) or bloodstream (BS). The RNA targeted by RNAi is indicated below the life-cycle stage. Cells were cultivated either without tetracycline (control, indicated as “−”) or with tetracycline (“+”) to induce RNAi (1 d for bloodstream forms, 2 d for procyclics). The signals on the Western blots were quantitated by scanning densitometry and normalized to the CSM control. One of the signals on each blot was then arbitrarily chosen to represent 100% and all other signals were compared to that one. The rather high signal for the XRNA RNAi on the left may be a consequence of the proximity of the nonspecific band or by a smear on the blot. The signal on the XRNB Northern blot was too low for accurate quantitation. (B) Effect of RNAi on expression of XRNC and XRND. The figure shows Western blots of total lysates from different cultures, detected with anti-peptide antibodies and labeled as in “A.” The values in parentheses were affected by the anomalously low CSM signal, which was not consistent with very similar Ponceau red staining of protein in all lanes.

FIGURE 3.

Effects of RNA interference targeting different XRN mRNAs on trypanosome growth. Transgenic trypanosomes containing tetracycline-inducible RNAi constructs were grown with or without tetracycline for the times indicated. Vertical lines indicate dilution of the cultures. (Black symbols and solid lines) cells without tetracycline, normal XRN levels; (open symbols and dotted lines) cells with tetracycline and induced RNAi. The nature of the targeted transcript and the life-cycle stage are indicated above each graph.

FIGURE 4.

Subcellular fractionation results. Trypanosomes were lysed with NP40 and fractions prepared as described by Zeiner et al. (2003). Aliquots representing equivalent numbers of cells were separated on SDS-polyacrylamide gels and the XRN proteins detected by Western blotting. The p34/p37 proteins were used as a nuclear marker (Nuc), and the cytoplasmic marker (Cyt) was CSM (see Materials and Methods). RNAi controls for the Westerns are also shown (labeled as in Fig. 2). (T) total lysate; (C) cytoplasmic fraction; (N) nuclear fraction. (A) Fractionation of procyclic trypanosomes, detection of XRNA and XRNC; nonspecific bands are indicated by *. (B) Fractionation of bloodstream trypanosomes expressing V5-tagged XRNB, detection of XRNA and the V5 tag. (C) Fractionation of procyclic trypanosomes, detection of XRND.

FIGURE 5.

Effect of XRNA depletion on actin (ACT) mRNA. (A) Trypanosomes with inducible RNAi targeting XRNA were treated with 1 μg/mL Sinefungin; mRNA was prepared and analyzed by Northern blotting. To induce XRNA RNAi (downward arrow), the cells were treated with 100 ng/mL tetracycline for 24 h before Sinefungin addition. The 0 time point was take before Sinefungin addition and the 5-min time point was taken by centrifuging the cells immediately after Sinefungin addition. A typical blot is shown, with signal recognition particle (SRP) RNA as a loading control. The levels of ACT mRNA were assessed using phosphorimaging. On the graph, each series of symbols represents the result of one experiment, and different symbol shapes are used for different experiments. Filled symbols show the amount of RNA in the absence of tetracycline, and open symbols the amount in cells with XRNA depletion. The lines were created by plotting the arithmetic means for each time point, then interpolating. (Solid line) without tetracycline (control); (dashed line) with tetracycline (XRNA depleted). The half-lives shown were estimated on the exponential part of the curve only, using Kaleidograph. (B) Cells were incubated for 5 min with Sinefungin, then (at time = 0) 10 μg/mL Actinomycin D was added. Results are means and standard deviations for four experiments, three of which were done with the XRNA RNAi cells and one in cells with the XRNA/B RNAi (which gave indistinguishable results). Open symbols are after RNAi. (C) Degradation after Sinefungin and Actinomycin D treatment, using procyclic trypanosomes; symbols as in A.

FIGURE 6.

Effect of XRNA depletion on CAT-GC-EP and G-CAT-EP mRNA degradation in bloodstream trypanosomes. Methodological details and symbols are all as in Figure 5. The probe was a PCR fragment encompassing the CAT-EP reporter (without a GC tract); it contains the 5′ UTR, the CAT gene, and the 3′ UTR. (A) Structures of the mRNAs detected in cells expressing CAT-EP mRNAs. The CAT open reading frame is shown in black and the untranslated regions are unfilled, except for the destabilizing U-rich tract, which is gray. The poly(A) tail is represented by “AA” and the G₃₀ or G₃₀C₃₀ are indicated as tangled loops. (B) Degradation of G-CAT-EP mRNA after addition of Sinefungin and Actinomycin D. (C) Degradation of CAT-GC-EP mRNA after addition of Sinefungin and Actinomycin D.

FIGURE 7.

Depletion of XRNA increases the steady-state abundances of other unstable mRNAs. RNAi against XRNA, XRNB, XRNC, or XRND was induced for 1 d (bloodstream) or 2 d (procyclics), and RNA was prepared and analyzed by Northern blotting using various probes. The PGK open reading frame probe hybridizes with the PGKA, PGKB, and PGKC mRNAs; AAT11 an amino acid transporter mRNA up-regulated in procyclics. SRP loading controls are also shown.

FIGURE 8.

Effect of XRNA results on poly(A) tail lengths. Transcription was inhibited using Actinomycin D (added at time 0). (A) The deadenylated transcript was made by incubating the RNA with oligo d(T) and RNase H (Irmer and Clayton 2001). RNA was examined by Northern blotting using probes for HISH4, PGK, and CAT. The trypanosome genome sequence shows 10 genes encoding histone H4 arranged as tandem repeats (Tb927.4.4170–4260) with identical 300-bp coding regions. The intergenic regions downstream of the first nine genes are also identical. Histone H4 mRNA cDNAs in the EMBL database have three different trans splicing acceptor sites spanning a region of 24 nt, and one polyadenylation site giving a 188-nt 3′ UTR, resulting in predicted mRNA lengths of ∼570 nt without poly(A); the band shown migrates ∼600 nt. The amounts of CAT-EP transcript were quantitated and expressed as a percentage of the amount present at time = 0; the amounts for the cells with and without tetracycline were assessed separately. (B) Northern blot showing 10 μg total RNA (T), and the poly(A)- and poly(A)+ fractions from 20 μg total RNA. RNA was prepared from the XRNA RNAi cell line expressing CAT-GC-EP mRNA, incubated with or without tetracycline for 1 d, before and 30 min after Actinomycin D treatment. The blot was hybridized with probes for the EP 3′-UTR, GPIPLC, CAT, and HISH4 with or without intermediate stripping as required. The ethidium bromide stain detecting rRNA is also shown. Probes are indicated in bold italics on the right, together with size indicators, and detected RNAs are shown on the left.