Article

• The EMBO Journal (2008) 27, 1596 - 1608
• doi:10.1038/emboj.2008.87

Published online: 8 May 2008

• Subject Categories:

  • RNA | Microbiology and Pathogens

3' adenylation determines mRNA abundance and monitors completion of RNA editing in T. brucei mitochondria

Ronald D Etheridge¹, Inna Aphasizheva¹, Paul D Gershon² and Ruslan Aphasizhev¹

1. Department of Microbiology and Molecular Genetics, School of Medicine, University of California, Irvine, CA, USA
2. Department of Molecular Biology and Biochemistry, University of California, Irvine, CA, USA

Correspondence to:

Ruslan Aphasizhev, Department of Microbiology and Molecular Genetics, School of Medicine, University of California, Irvine, B240 Medical Sciences I, Irvine, CA 92697, USA. Tel.: +1 949 824 7845; Fax: +1 949 824 9394; E-mail: ruslan@uci.edu

Received 24 October 2007; Accepted 2 April 2008

• Keywords:

  • mitochondria,
  • mRNA editing,
  • mRNA stability,
  • polyadenylation,
  • Trypanosoma

Introduction

RNA processing in the mitochondria of trypanosomatid protozoa begins with a nucleolytic partitioning of multi-cistronic transcripts into ribosomal RNAs and pre-mRNAs, which are then subjected to 3' polyadenylation and U-insertion/deletion RNA editing. In Trypanosoma brucei, editing is required for expression of 12 of the 18 protein-encoding genes. Multi-protein complexes carry out editing reactions (20 S editosome), guide RNA (gRNA) 3' uridylylation (RNA editing TUTase 1 (RET1)), and mRNA–gRNA annealing (mitochondrial RNA binding proteins 1 and 2 (MRP1/2)) (reviewed by Simpson et al, 2004; Stuart et al, 2005). In the steady-state RNA population, pre-edited, highly heterogeneous partially edited and fully edited forms have been distinguished (Abraham et al, 1988; Decker and Sollner-Webb, 1990; Sturm and Simpson, 1990).

Polyadenylation ubiquitously regulates RNA stability, even though the exact role of the poly(A) tail may differ among cellular organelles or within the same compartment (reviewed by Martin and Keller, 2007). The virtually universal nuclear polyadenylation of eukaryotic mRNAs is accomplished by macromolecular complexes bearing poly(A) polymerase (PAP) as a catalytic subunit and is coupled to transcription, splicing, and nuclear export (Proudfoot, 2004). Conversely, adenylation of intergenic transcripts and various aberrant RNAs by the non-canonical Trf4/5 PAPs leads to RNA degradation in the nucleus of budding yeast (reviewed by Houseley et al, 2006). Cytoplasmic polyadenylation generally stimulates translation and is often targeted to specific mRNAs through RNA binding proteins (Huang and Richter, 2004; Stevenson and Norbury, 2006). Despite the mitochondrion's monophyletic origin, the roles of polyadenylation in this organelle are diverse. For example, yeast mitochondrial mRNAs are not polyadenylated, whereas in plant mitochondria (Gagliardi et al, 2004) and bacteria (Kushner, 2004), the poly(A) tail is a signal for degradation. In human mitochondria, polyadenylation stabilizes mRNAs whereas de-adenylation by polynucleotide phosphorylase has the opposite effect (Nagaike et al, 2005).

Existing data suggest that polyadenylation has a key role in regulating T. brucei mitochondrial genome expression by altering mRNA stability. In addition to the sequence heterogeneity created by mRNA editing, mitochondrial transcripts have either short (20–25 nucleotides (nt)) or long (120–250 nt) A-tails (Bhat et al, 1992). The length of the poly(A) tail appears to correlate with the editing status of the mRNA: pre-edited forms possess only short A-tails whereas never-edited and edited mRNAs have both short and long A-tails (Bhat et al, 1992; Read et al, 1994; Militello and Read, 1999). In vitro experiments have demonstrated that synthetic pre-edited mRNAs with short A-tails degrade rapidly, whereas partially or fully edited messages were relatively stable in mitochondrial extracts (Ryan et al, 2003; Kao and Read, 2005). However, PAP activity has never been detected in mitochondrial extracts and the abundance of mitochondrial transcripts has not been linked to their polyadenylation status in vivo.

We have previously pointed out that the RNA editing TUTases, RET1 (Aphasizhev et al, 2002) and RET2 (Aphasizhev et al, 2003; Ernst et al, 2003), are homologous to several proteins encoded in trypanosomal genomes (Aphasizhev, 2005). Indeed, the non-mitochondrial TUT3 (Aphasizhev et al, 2004) and TUT4 (Stagno et al, 2007b) have been characterized as RNA uridylyl transferases. Beyond Kinetoplastida, however, data mining of eukaryotic genomes has identified 'non-canonical' PAPs, such as animal cytoplasmic Gld-2-type and yeast nuclear 'quality control' Trf4/5 PAPs, as the proteins most closely related to TUTases (not shown). This apparent ambiguity was partially resolved by examining the X-ray structures of TUT4 in complexes with UTP or ATP, and the demonstration that TUTase's active site may be hospitable to ATP binding (Stagno et al, 2007a). Therefore, we hypothesized that some TUTase-like proteins in trypanosomes are non-canonical PAPs. To search for the mitochondrial PAP, we set out to analyse the NTP substrate specificity and cellular function of a candidate TUTase-like protein with a putative mitochondrial importation signal, previously designated TUT5 (Aphasizhev, 2005).

Here we show that TUT5, re-named kinetoplast poly(A) polymerase 1 (KPAP1), is a mitochondrial PAP essential for parasite viability and mitochondrial function. This enzyme is directly involved in the synthesis of short A-tails that are required and sufficient for the steady-state abundance of never-, partially, and fully edited mitochondrial mRNAs. The pre-edited transcripts were either unaffected or increased because of the loss of KPAP1 and the ensuing lack of short A-tails. Editing directed by a single gRNA is sufficient to impose a requirement for the short A-tail in partially and fully edited molecules. Further, A/U heteropolymers, previously thought to be long poly(A) tails, are appended to the fully edited mRNAs upon completion of the editing process. Thus, the A/U-tail may represent a hallmark of translationally competent mRNAs. Proteomic analysis of the affinity-purified KPAP1 complex revealed a novel RNA processing particle composed of RNA binding and pentatricopeptide repeat (PPR) proteins, and polypeptides with no similarities beyond Kinetoplastida.

KPAP1 is a mitochondrial PAP

Protein sequences of RET1, RET2, TUT3, and TUT4 were used to search Kinetoplastida genome databases (http://www.genedb.org), leading to the identification of the KPAP1 gene (Supplementary Figure S1). Trypanosomal nuclear PAP (Mair et al, 2000) was not detected in these searches. The recombinant KPAP1 (Figure 1A) showed a preference for ATP in a nucleotidyl transfer assay (Figure 1B). To verify the enzymatic identity, KPAP1 with a D97A mutation in the catalytic site was purified and found to be inactive (Figure 1B). Apparent molecular mass determined by gel filtration on Sepharose 12 column in 150 mM of KCl (not shown) closely matched the predicted value for the His-tagged protein (61 kDa), indicating that recombinant KPAP1 is a monomer. To assess the processivity of ATP polymerization by KPAP1, the reactions were carried out under competitor challenge conditions. Following pre-incubation of the RNA substrate with enzyme, extensions were initiated by addition of ATP plus increasing concentrations of unlabelled RNA. The product length was linearly reduced from 20–25 to 2–3 adenosines, illustrating a distributive nucleotidyl transferase activity (Figure 1C).

Figure 1.

KPAP1 is a mitochondrial PAP. (A) Purification of the recombinant KPAP1 from E. coli. M: mol. mass marker; MAC: metal affinity chromatography; CEX: cation exchange column. Proteins were separated on 8–16% gradient SDS–PAGE. (B) NTP specificity of KPAP1. The 5'-labelled 24-mer 6[A] RNA was incubated with KPAP1 at increasing concentrations of NTPs (1, 10, and 100 M). The mutant D97A protein was tested in the presence of ATP. C: control RNA. The products were resolved on a 15% polyacrylamide/8 M urea gel. (C) Primer challenge assay. The 5'-labelled 6[A] RNA and KPAP1 were pre-incubated at final concentrations of 25 and 50 nM, respectively, for 10 min at 27°C. Reactions were started by simultaneous addition of ATP to 100 M and unlabelled 6[A] RNA to 0, 25, 50, 100, 200, 400, and 800 nM. Incubation was continued for 30 min. (D) Subcellular fractionation of procyclic T. brucei. Approximately 10 g of protein from hypotonic cell extract, cytosolic fraction, and purified mitochondria was separated on SDS–PAGE. Western blotting was performed with antibodies against KPAP1 and mitochondrial (MP81), cytoskeletal (-tubulin), and cytosolic (HSP70.4) proteins. (E) Intracellular distribution of KPAP1. The C-terminal KPAP1–eYFP fusion was expressed in pLew79-based vector (Wirtz et al, 1999). Fluorescent images were captured in the presence of MitoTracker Red CMX ROS and DAPI stains.

View full figure (141 KB)

Western blotting of T. brucei subcellular fractions demonstrated that KPAP1 is localized in the mitochondrial matrix. As shown in Figure 1D, KPAP1 was enriched in the density gradient-purified mitochondrial fraction. A subunit of the 20 S editosome, MP81 (Drozdz et al, 2002), showed a similar distribution, indicating that KPAP1 is localized to the mitochondrial matrix. A truncated form of KPAP1 was also observed in the mitochondrial fraction. The cytoskeletal (-tubulin) and cytosolic (HSP70.4) proteins were not detected, verifying the correct segregation of subcellular compartments. KPAP1 localization in live T. brucei cells was confirmed by expressing its C-terminal fusion with enhanced yellow fluorescence protein (eYFP). The KPAP1–eYFP colocalized with the membrane potential-dependent MitoTracker Red CMX-Ros dye (Figure 1E). Interestingly, the apparent concentration of KPAP1 was observed at two punctate antipodal regions adjacent to the kDNA disk.

KPAP1 is essential for cell viability and mitochondrial function

As seen from RNA interference (RNAi) analysis, the KPAP1 gene is essential for the viability of insect (procyclic form (PF)) T. brucei. Upon RNAi induction with tetracycline, cumulative cell growth was monitored in comparison to mock-treated culture. The growth inhibition was followed by elimination of live cells from the culture (Figure 2A). The KPAP1 protein was undetectable after 50 h of RNAi induction (Figure 2B). To address the appearance of a second band in the mitochondrial extract (Figure 1D, KPAP1 panel), RNAi was induced for 50 h and subcellular fractionation was carried out. Both bands disappeared in the mitochondrial fraction depleted of KPAP1 by RNAi (Figure 2C).

Figure 2.

KPAP1 is an essential gene required for mitochondrial function. (A) Cumulative cell growth after KPAP1 RNAi induction in procyclic T. brucei. (B) Depletion of KPAP1 protein in PF T. brucei. Cell lysates from the parental cell line (29-13) and tet-induced cells were separated on SDS–PAGE and probed with anti-KPAP1 antibody. (C) Subcellular fractions obtained from KPAP1 RNAi cell line were probed with anti-KPAP1 antibody. The Coomassie blue staining was used as loading control.

View full figure (121 KB)

Maintenance of the mitochondrial membrane potential (m) is essential for PF and BF (bloodstream) trypanosome viability (Schnaufer et al, 2005; Brown et al, 2006). The depolarization of the mitochondrial membrane over the course of KPAP1 RNAi was monitored in the PF T. brucei (Supplementary Figure S2). RNAi was induced at 24-h intervals and individual cultures were maintained in the presence of tetracycline. Cell staining with membrane potential-sensitive dye MitoTracker Red CMX-Ros and FACS analysis for all time points were performed at the same time. Treatment with CCCP (carbonylcyanide m-chlorophenylhydrazone) protonophore, which discharges the pH gradient across the inner membrane, was included as a control for the loss of membrane potential. In this experiment, cell division arrest was observed between 72 and 96 h, which coincided with the appearance of a predominant population of cells lacking a membrane potential. Therefore, KPAP1 is essential for cell viability and mitochondrial function.

KPAP1 interacts with RNA editing complexes, but its depletion does not affect editing activity

Previous work has suggested that polyadenylation exerts different effects on the stability of edited and pre-edited mRNAs in mitochondrial extracts (Kao and Read, 2005). RET1 was also implicated in the UTP-dependent mRNA degradation in organello (Militello and Read, 2000; Ryan and Read, 2005). To investigate KPAP1 interactions with RNA editing complexes, immunoprecipitations with anti-KPAP1 antibody were performed in the mitochondrial extract. The core components of the 20 S editosome, RNA editing ligase 1 (REL1) and MP81, and RET1 were detected in the co-immunoprecipitated (co-IP) material (Figure 3A). These interactions do not appear to be RNA-mediated, as RNase A and high-salt treatment had no effect on co-IP. Insignificant depletion of editing proteins was observed when all detectable KPAP1 was immunoprecipitated from the mitochondrial extract (not shown), suggesting that only minor fractions of the 20 S editosome and RET1 are stably associated with KPAP1.

Figure 3.

Inhibition of KPAP1 expression does not affect RNA editing complexes. (A) Co-immunoprecipitation of KPAP1 with 20 S editosome and RET1. Mitochondrial extract (200 l, 5 mg protein/ml) from T. brucei was incubated for 1 h with 10 l of magnetic beads pre-coated with antigen-purified anti-KPAP1 antibodies. Additional 30 min washes with 1% Triton or RNase A (0.1 mg/ml) in PBS buffer, or with 0.5 M KCl, were performed to assess the stability of KPAP1 interactions. Immunoprecipitated material was adenylated on beads, separated on SDS–PAGE, and probed for RET1 and MP81 on immunoblotting. Co-IP with antibodies against glutamate dehydrogenase (GDH) served as a negative control. (B) Sedimentation of the 20 S editosome from KPAP1-depleted mitochondrial extracts. Mitochondrial extract was fractionated on a 10–30% glycerol gradient. Fractions were incubated with [-³²P]ATP to detect editing ligases REL1 and REL2 and separated on SDS–PAGE. Sedimentation standards (catalase (11 S), thyroglobulin (19 S), and E. coli 30 S ribosome subunit) migrated as indicated by arrows. (C) U-insertion editing activity in the peak gradient fraction. RNA substrates for the pre-cleaved editing assay were assembled as follows: (1) 5' fragment, no proteins added; (2) 5' fragment; (3) 5' fragment+'guide' RNA; (4) 5' fragment+3' fragment; (5) fully assembled substrate for +2 addition, 5' fragment+3' fragment+'guide' RNA. Positions of +2 guided U-insertions and ligation of edited product are shown by arrows. (D) gRNA labelling. Total RNA was isolated from the parental cell line (29-13), KPAP1, and RET1 RNAi cells. Guide RNAs were 5' labelled with [-³²P]GTP in the presence of vaccinia virus guanylyltransferase and separated on 10% polyacrylamide/urea gel. The unidentified cytosolic RNA labelled with [-³²P]GTP was used as a loading control.

View full figure (189 KB)

To assess the effects of KPAP1 RNAi on RNA editing complexes and editing activity in vivo, mitochondrial extracts were obtained from the T. brucei parental cell line (29-13), and cells were subjected to KPAP1 RNAi for 50 h. The extracts were fractionated on 10–30% glycerol gradients and each fraction was tested for the 20 S editosome (Figure 3B). This approach reportedly detects even minor alterations in the editosome structure (Aphasizhev et al, 2003; Kang et al, 2006), yet no apparent differences were observed. The peak fraction 9 from both gradients, adjusted for REL1 adenylation signal and total protein amount, showed virtually identical U insertion (Figure 3C) and U-deletion (data not shown) editing activities on a pre-cleaved mRNA substrate (Igo et al, 2000).

Next we analysed gRNA abundance and size distribution. The 5' labelling of primary transcripts, such as gRNAs, with guanylyltransferase in the presence of [-³²P]GTP (Blum et al, 1990) detected no appreciable difference caused by KPAP1 RNAi (Figure 3D). As a control for the loss of oligo[U] tail in gRNAs (Aphasizhev et al, 2003), RNA isolated from cells depleted of RET1 by RNAi was subjected to the same labelling procedure. These results show that KPAP1 interacts with RNA editing components, but its depletion does not affect the integrity of editing complexes, editing activity, or gRNA abundance.

Purification of mitochondrial polyadenylation complex

To analyse complexes associated with KPAP1, mitochondrial extract from T. brucei expressing a TAP-tagged (Puig et al, 2001) fusion protein was subjected to sedimentation in a 10–30% glycerol gradient. Fractions were pre-incubated with [-³²P]ATP to self-adenylate editing ligases, separated by gradient Tris–glycine PAGE in the presence or absence of SDS, and transferred to nitrocellulose membrane (Figure 4A). Western blotting with PAP reagent, which detects the protein A moiety in the TAP tag, and exposure to phosphor storage screen were used to detect polyadenylation and editing complexes, respectively. SDS–PAGE analysis showed that KPAP1 complexes sediment in the 15–30 S range, whereas native gel separation revealed KPAP1-containing particles with apparent molecular weights ranging from 0.4 to 1.2 MDa (Figure 4A). Unassociated KPAP1–TAP, which would be expected to migrate at 80 kDa in the native gel and sediment at the top of the gradient, was also detected in the mitochondrial extract.

Figure 4.

Purification of the KPAP1 complex. (A) Glycerol gradient fractionation. Mitochondrial extract from T. brucei cells expressing C-terminally TAP-tagged KPAP1 was fractionated on 10–30% glycerol gradients. Fractions were analysed for the presence of RELs (adenylation) and tagged protein (TAP detection). (B) Tandem affinity purification of the KPAP1 complex. The final fraction from the calmodulin column was separated on 8–16% SDS–PAGE and stained with Sypro Ruby. Carryover of TEV protease is indicated by an asterisk. The same fraction was probed along with the mitochondrial extract with antibodies against KPAP1, RET1, and MRP1. Before gel separation, samples were incubated with [-³²P]ATP to detect editing ligases REL1 and REL2. MRP1 has been shown to dimerize even under denaturing conditions (Zikova et al, 2008). (C) PAP and TUTase activities of the purified KPAP1 complex. Left panel: the 89-mer fragment of pre-edited RPS12 mRNA (100 nM) was incubated with KPAP1 complex (10 nM KPAP1, as estimated for the KPAP1-CBP band in Sypro Ruby-stained SDS gel) for 5, 15, and 45 min in the presence of 100 M ATP or UTP, or both. Radiolabelled [-³²P]ATP and [-³²P]UTP were added as indicated. C: RPS12 RNA labelled at the 5' end. Right panel: recombinant KPAP1 (20 nM) and RET1 (20 nM) were incubated with RPS12 RNA under the same conditions. Reaction products were separated on 8% polyacrylamide/urea gel. (D) Synthesis of a short A-tail by the recombinant KPAP1. The 5'-labelled RPS12 mRNA fragment (20 nM) was incubated with 100 M ATP and increasing concentrations of KPAP1 (5, 10, 20, 50, 100, and 200 nM) for 30 min.

View full figure (230 KB)

Tandem affinity purification of KPAP1 complex from mitochondrial extract produced a set of 15 polypeptides, which was further analysed by western blotting (Figure 4B) and mass spectrometry (Table I). Consistent with immunoprecipitation data, the RET1 and 20 S editosome components were detectable in the affinity-purified material whereas the abundant gRNA binding protein MRP1 (gBP21) (Koller et al, 1997) was not.

Table 1: Proteins identified in TAP tag affinity-purified KPAP1 complex

View full table

The nanoLC-MS/MS analysis of the major gel bands and the entire complex identified several proteins that had been previously implicated in mitochondrial RNA processing as well as predicted mitochondrial proteins lacking any discernable functional motifs (Table I). Mitochondrial oligo[U] binding protein, TBRGG1, previously reported as a part of unknown high-molecular-weight complex (Vanhamme et al, 1998), was among the most prominent polypeptides. Multiple peptides were matched with high confidence to five PPR proteins, including PPR1 as one of the major components. Remarkably, RNAi knockout of PPR1 inhibits the formation of the long A-tails in never-edited and edited mRNAs, and is detrimental for oxidative phosphorylation (Mingler et al, 2006; Pusnik et al, 2007). Among 25 predicted trypanosomal mitochondrial PPR proteins, six have been shown to function in mitochondrial ribosomes (Pusnik et al, 2007); none of these proteins (PPR2–7) was detected in the polyadenylation complex.

Five proteins with no apparent functional motifs (Tb927.4.4150, Tb927.7.2570, Tb927.2.3800, Tb927.1.3010, and Tb11.02.5390) have been also identified as subunits of a putative mitochondrial RNA binding complex, put-MRB (Panigrahi et al, 2007). KPAP1 was not reported in this study, which prompted us to find out whether KPAP1–put-MRB association is mediated by RNA. To distinguish such contacts, affinity purification was carried out in the presence of nuclease cocktail including RNases A, T1, T2, micrococcal nuclease, RNase H+oligo[dT]. NanoLC-MS/MS analysis of the same amount of purified material demonstrated that some of the put-MBP polypeptides were no longer detected while coverage decreased for others indicating RNA-based interactions (Table I). Finally, the predicted mitochondrial elongation factor Tu was abundant in all preparations.

Enzymatic activities of the polyadenylation complex

We next wished to see whether the purified KPAP1 complex would produce elongation patterns similar to those of the recombinant protein (Figure 1B and C) or long A-tails observed in vivo. To generate an RNA molecule that resembles an in vivo substrate for KPAP1, the 3' fragment of the pre-edited mRNA for small ribosomal protein subunit 12 (RPS12) was fused to the modified HDV ribozyme sequence (Supplementary data). The ribozyme self-cleavage produced a defined 3' end terminating with five encoded uracil residues. In the presence of ATP, activities of the recombinant enzyme and purified KPAP1 complex, adjusted for KPAP1 concentration, were virtually indistinguishable (Figure 4C, left panel). Addition of UTP into the reaction revealed the presence of a processive TUTase activity, which is in agreement with KPAP1–RET1 co-IP (Figure 3A) and immunoblotting of the purified KPAP1 complex (Figure 4B).

Reactions with KPAP1 complex did not produce the expected 20–25 nt A-tail of the pre-edited mRNAs, likely because of low enzyme concentration. Given the distributive nature of KPAP1 activity (Figure 1C), we tested whether increasing the concentration of the recombinant protein would lead to the formation of such a structure. Surprisingly, at concentrations above 50 nM, short A-tail was synthesized but the additions effectively stopped at 20–25 nt (Figure 4D). These findings indicate that KPAP1 does not require associated factors for synthesis of short A-tail but is unable to add longer moieties because of an intrinsic self-limiting mechanism.

Inhibition of KPAP1 expression reduces the abundance of never-edited and edited mRNAs

To assay the changes in mRNA abundance caused by KPAP1 RNAi, we carried out 'poisoned' primer extensions with oligonucleotides that hybridize downstream of editing domains (Aphasizhev et al, 2002). The 3' regions of cytochrome oxidase subunit 3 (CO3) mRNA (Feagin et al, 1988) and ATP synthase subunit 6 (A6) (Bhat et al, 1990) were chosen as representative pan-edited mRNAs. Cytochrome oxidase subunit 2 (CO2) mRNA was analysed because editing is limited to the insertion of only four uridines (Benne et al, 1986) and is guided by a cis-acting RNA element (Golden and Hajduk, 2005 and references therein). The never-edited cytochrome oxidase subunit 1 (CO1) mRNA was tested with -tubulin mRNA serving as a loading control (Figure 5A). The edited forms of all mRNAs, as well as the CO1 transcript, decreased in abundance. To further assess KPAP1 RNAi effects, the relative abundance of eight pre-edited, corresponding edited, and four never-edited mRNAs was determined by quantitative RT–PCR (qRT–PCR). In a uniform pattern, the rRNAs and pre-edited mRNAs either remained unaltered or increased, indicating lack of effects on transcription and nucleolytic processing of multi-cistronic transcripts. In contrast, all edited transcripts showed a 45–80% reduction (Figure 5B). Never-edited mRNAs also declined to various extents (Figure 5C).

Figure 5.

Polyadenylation determines the abundance of never-edited and edited mRNAs. (A) 'Poisoned' primer extension analysis. DNA oligonucleotides were hybridized with total RNA and reverse transcription was performed in the presence of ddGTP (Missel et al, 1997). Reaction products were separated on 10% polyacrylamide/urea gel. Single-extension products were observed with never-edited CO1 mRNA and -tubulin. In the A6 panel, primers for -tubulin, A6, and CO1 mRNA were used in a single reaction. Only extension products are shown, and 5'-labelled DNA primers are omitted. The relative decrease in abundance was calculated assuming the mitochondrial mRNA/-tubulin mRNA ratio in mock-induced cells as 100%. (B) Quantitative RT–PCR analysis of edited mRNAs. The RNA levels averaged from three replicates were normalized to -tubulin mRNA. The thick line at 1 indicates no change in relative abundance. (C) Relative abundance of never-edited mRNAs and 12 S ribosomal RNA. (D) Northern blotting of the never-edited (CO1), cis-edited (CO2), 5'-edited (Cyb), and pan-edited (A6) mRNAs. Gel bands corresponding to mRNAs with long-tail (LT) or short-tail (ST) adenylation patterns are shown by arrows. Total RNA (25 g) isolated from uninduced (-) and induced (+) KPAP1 RNAi cells was separated on 1.4% agarose-formaldehyde gel. (E) RPS12 mRNA decay in the course of KPAP1 RNAi. Total RNA was separated on 6% polyacrylamide/urea gel and sequentially probed for edited and pre-edited RPS12 transcripts, and 9 S mitochondrial ribosomal RNA. Long-tail (LT), short-tail (ST), and non-adenylated (NA) forms are shown by arrows. C: untreated RNA; (-), RNase H digestion minus DNA oligo.

View full figure (234 KB)

To determine whether depletion of KPAP1 led to a loss of poly(A) tails, we used northern blotting to examine changes in transcript sizes (Figure 5D). In agreement with primer extension and qRT–PCR data, the abundance of both short A-tail (ST) and long A-tail (LT) forms of CO1 mRNA declined, whereas no changes were observed for -tubulin mRNA and 12 S rRNA. The CO2 probe, which hybridizes to both pre-edited and edited mRNAs (+4Us), detected a loss of the slower-migrating band, which would correspond to mRNA with a long A-tail. Probes that specifically bind to pre-edited and edited forms, and to both species, were used for the cytochrome b (Cyb) transcript. As expected, the edited mRNA was reduced by 80%, but the pre-edited form increased by 20% and became shortened. A similar pattern was observed for the pan-edited A6 mRNA. The accumulation of non-adenylated never-edited or edited transcripts in the steady-state mRNA population was not detected, likely because of rapid degradation. Furthermore, the loss of edited mRNAs apparently does not depend on the number of editing events or the positioning of editing blocks within an mRNA. Collectively, these results demonstrate a uniform reduction of edited and never-edited transcripts in KPAP1 RNAi cells. The lack of apparent defects in transcription, nucleolytic processing of multi-cistronic transcripts (Figure 5), and RNA editing processes (Figure 3) implies that the increase in the decay rates of never-edited and edited transcripts is a most likely consequence of KPAP1 knockout.

Edited mRNAs with short and long A-tails are equally impacted by KPAP1 RNAi

The next step investigated whether the abundance of mRNAs distinguished by short and long A-tails changes synchronously with RNAi progression. Northern blotting of the never-edited CO1 revealed heterogeneous A-tails of more than 200 nt, whereas edited mRNAs possessed more homogenous long A-tails of 200 nt (Figure 5D). For a more precise length assertion of short and long A-tails, we analysed the mRNA encoding RPS12 (Maslov et al, 1992). Its smaller size made the transcript amendable to high-resolution separation on acrylamide/urea gels. As seen in Figure 5E (edited RPS12 panel), long tail (LT) and short tail (ST) forms of edited RPS12 mRNA showed a simultaneous decrease upon RNAi induction. The pre-edited form lost 25 nucleosides (NA, Figure 5E, pre-edited RPS12 panel) but its abundance remained unaffected.

Because the non-adenylated edited mRNAs were undetectable by northern blotting, we set out to determine the length of poly(A) tail in both forms of edited RPS12. Total RNA was incubated with 18-mer oligo[dT] and treated with RNase H (Figure 5E, RNase H panel). In the presence of oligo[dT], ST and LT edited forms collapsed into a single band corresponding to the predicted length (325 nt) of the fully edited RPS12 mRNA. The quantitation of the underexposed gels allowed us to estimate the length of the short and long A-tails as 20–25 and 240–250 nt, respectively. In summary, these results indicate that the presence of a short poly(A) tail is essential for the maintenance of edited mRNAs in mitochondria. Conversely, the steady-state level of pre-edited mRNAs does not depend on 3' adenylation.

KPAP1 adds short A-tails to pre-edited and never-edited mRNAs

The short A-tails apparently have distinct roles in maintaining the abundance of pre-edited and never-edited transcripts, although both types are thought to originate from the nucleolytic processing of the same precursor RNA. The non-adenylated pre-edited mRNAs are readily detectable by northern blotting whereas non-adenylated never-edited transcripts are not (Figure 5). Therefore, the question arose if KPAP1 is responsible for the short A-tail addition to never-edited mRNAs. To directly analyse poly(A) tails in pre-edited and never-edited mRNAs, total RNA was isolated from cells subjected to KPAP1 RNAi for various time periods, followed by circularization with T4 RNA ligase, transcript-specific cDNA synthesis, and PCR amplification in a procedure known as circular RT–PCR (cRT–PCR; Brogna, 1999). Primers were designed to allow sequencing analysis of poly(A) tails plus the 5' and 3' regions of respective mRNAs species (Supplementary Figure S3). The cRT–PCR produced products of expected sizes for adenylated and non-adenylated forms of pre-edited RPS12 mRNA (Figure 6A). Products obtained from the parental cell line and after 48 h of RNAi induction were cloned and sequenced. The statistical analysis of the short A-tail in pre-edited RPS12 mRNA showed an average length of 25 nt (Figure 6B). In agreement with northern blotting data (Figure 5D and E), the A-tail was lost in all 30 clones obtained from KPAP1 RNAi cells. Several sequences also had heterogeneous 3' deletions (Supplementary Table S1) extending into the mRNA for up to 90 nt.

Figure 6.

Short A-tails are synthesized by KPAP1. (A) cRT–PCR amplification of pre-edited RPS12 mRNA. PCR products were separated on 2% agarose gel. The pre-edited adenylated (ST) and non-adenylated (NA) forms are shown by arrows. (B) Distribution of A's per tail in pre-edited RPS12 mRNA. Box plot shows the distribution of A's per tail with possible outlying points in the 25–75 percentile range. The middle line identifies the median sample value. The upper and lower 'whiskers' highlight the range of non-outlying data points. The diamond illustrates the sample mean and 95% confidence interval. The line across each diamond represents the group mean. The vertical span of each diamond represents the 95% confidence interval for each group. (C) cRT–PCR amplification of the CO1 mRNA. (D) Distribution of A's per short A-tail in CO1 mRNA after KPAP1 RNAi.

View full figure (114 KB)

Applying cRT–PCR to never-edited CO1 mRNA generated products consistent with the presence of the short A-tails (Figure 6C). We were unable to detect PCR products corresponding to CO1 mRNA with a long A-tail. The analysis of A-tails in 47 CO1 clones from the parental cell line and 45 clones from KPAP1 RNAi demonstrated a statistically significant reduction in the number of adenosines per A-tail in KPAP1 RNAi (Figure 6D). Approximately 22% of all clones contained non-adenylated transcripts and RNAs missing 3–4 encoded nucleotides. In contrast to pre-edited RPS12 mRNA, the lack of more extensive 3' deletions in CO1 may indicate a processive degradation of non-adenylated never-edited mRNAs. Importantly, mRNA intermediates consistent with 5'–3' degradation pathway have not been detected among sequenced RPS12 or CO1 clones. Although the relative decay rates have not been measured, the structure of degradation intermediates suggests that the decrease in edited and never-edited mRNA abundance was caused by the 3'–5' degradation of non-adenylated transcripts.

In agreement with earlier reports (Decker and Sollner-Webb, 1990; Read et al, 1992), we have found that the short A-tails of both pre-edited RPS12 and never-edited CO1 mRNAs contain randomly incorporated U's, one or two at a time. Because of the low U content, the A/U ratio has not changed with any statistical significance in KPAP1 RNAi. To conclude, the short A-tails of pre-edited and never-edited mRNAs are synthesized by KPAP1 and appear to be slightly different in length but similar in nucleoside composition.

Editing of a single block converts the short A-tail into a stabilizing signal

As polyadenylation is likely to precede RNA editing (Koslowsky and Yahampath, 1997), we examined whether editing alters the function of the short poly(A) in the pre-edited transcripts from a neutral feature into a signal required for edited mRNA maintenance. Amplification of the pre-edited RPS12 mRNA did not discriminate transcripts that were unedited or edited in a region flanked by the primer binding site (A479) and the A-tail (positions 131–221; Supplementary Figure S3). Therefore, it was possible to observe a correlation between the editing and polyadenylation patterns within the same molecule. Analysis of 29 clones showed that in the parental cell line, six clones contained sequences that were fully edited in the seven editing sites adjacent to the mRNA's 3' end. In addition, 17 more clones were correctly edited in six sites. Further, extensive random editing upstream of this region was observed in 21 clones (Supplementary Figure S4 and Supplementary Table S1). Similar random editing patterns have been reported for Cyb and CO3 mRNAs (Decker and Sollner-Webb, 1990). Only five out of 29 clones carried unedited sequences. In contrast, among the 30 RPS12 clones obtained from the KPAP1 RNAi cells, only four sequences were edited in the first seven sites, and four more underwent random editing to a much lesser extent (Supplementary Figure S4).

Editing of these seven sites constitutes the first editing block, which is covered by a single gRNA, for example, E09 (KISS database; Ochsenreiter et al, 2007). The complete editing of this block most likely creates a binding site for the next gRNA, thus ensuring a 3'–5' polarity of the editing process (Maslov and Simpson, 1992). Because the 20 S editosome and gRNA processing machineries remain intact in KPAP1 RNAi cells (Figure 3), and the editing took place in the first editing block (Supplementary Figure S4), it is unlikely that the loss of edited sequences was due to a compromised editing of non-adenylated mRNAs. Rather, a switch to poly(A)-tail-dependent stabilization of the edited mRNAs occurs upon editing of the first 3' block.

Long poly(A/U) tails are the hallmarks of fully edited mRNAs

The editing-dependent switch of function for the short A-tail must be followed, at some point, by addition of 200 nt. This extended structure does not have an apparent effect on mRNA stability in vitro (Kao and Read, 2005) and its synthesis requires KPAP1 activity (Figure 5E). We next set out to analyse the structure of long tails and the temporal order of their appearance in edited mRNAs. The poly(A) tails are presumably difficult to amplify and clone because of the instability of homopolymeric DNA regions in Escherichia coli. Indeed, application of cRT–PCR to edited RPS12 mRNA produced two PCR products migrating with 100 bases difference (Figure 7A). We were able to clone LT products only from the parental cells; four out of 100 clones contained sequences extending beyond the short A-tail (Figure 7B).

Figure 7.

The A/U extension constitutes the 'long A-tail' of fully edited RPS12 mRNA. (A) cRT–PCR amplification of the long A-tail (LT) and short A-tail (ST) forms of the fully edited RPS12 mRNA. (B) Distribution of A's per tail in edited RPS12 mRNA following RNAi knockout of KPAP1. (C) Schematic representation of the A/U extensions in fully edited RPS12 mRNA derived from LT PCR products. (D) Northern blotting of pre-edited and partially edited RPS12 mRNAs in RNA isolated from RET1 and KPAP1 RNAi cells. The schematic positioning of DNA probes (solid line) within mRNA (dotted line) is shown under each panel; the sequences are provided in Supplementary Figure S3. [dT]: RNase H treated; C: control RNA; 9 S rRNA: mitochondrial small subunit rRNA. (E) Hybridization of the same membrane as in (D) with a probe for the fully edited RPS12 mRNA (Supplementary Figure S3).

View full figure (203 KB)

As expected, the region between the PCR primer binding site and the A-tail was fully edited in the ST and LT forms. In the ST form, a statistically significant reduction in the number of A's per short A-tail was observed as a consequence of KPAP1 RNAi (Figure 7B). We also noticed a 35% increase in the number of edited RPS12 transcripts missing the entire A-tail. However, heterogeneous deletions extending into the encoded region of the pre-edited form were not detected (Supplementary Figure S4). This indicates a processive decay of non-adenylated edited RPS12 mRNA similar to that observed for non-adenylated never-edited CO1 mRNA.

Remarkably, sequencing of four LT PCR products (Figure 7B) revealed that the homogenous 5' segment of 20 adenosine nucleosides, or the short A-tail, was extended by an 80-nt-long random addition of A (70%) and U (30%) residues (Figure 7C). The distinct U-rich structure of the long tail suggested that uridylyl transferases may be involved in the 3' processing of mitochondrial mRNAs, which is consistent with RET1 association with polyadenylation complex (Figure 4).

The long A/U-tails appear to be a characteristic of edited mRNAs (Figures 5 and 6), but the question remained whether these additions are present solely in the fully edited mRNAs, that is, in which editing of the 5' ends has been completed. Alternatively, the A/U extension may be added to the partially edited mRNA, in which editing has proceeded beyond the first 3' editing block. To resolve the temporal order of the long A/U-tail addition with respect to 3'–5' polarity of editing (Maslov and Simpson, 1992), we hybridized RNA isolated from RET1 and KPAP1 RNAi cells with a 38-mer DNA probe for the 5' end of pre-edited RPS12 mRNA (A605, positions 33–70; Supplementary Figure S3). This region is edited by the insertion of 47 and deletion of 3 U's, which would prevent hybridization if editing took place (Figure 7D, right panel). In addition, the PCR-generated DNA probes (Supplementary Figure S3) against pre-edited RPS12 mRNA were hybridized with the same membrane (Figure 7D, left panel). Sequential hybridization with a probe for the fully edited form is shown in Figure 7E. In the parental and mock-induced cells, partially edited molecules were three-fold more abundant than pre-edited mRNAs, but long A/U-tails were not detected (Figure 7D, right panel). In RNA isolated from RET1-depleted cells, the partially and fully edited transcripts showed an 90% decrease in abundance, as would be expected from impeded RNA editing due to loss of gRNA 3' uridylylation (Aphasizhev et al, 2003). The slower-migrating 550- and 700-nt-long RPS12 species also appeared but their identities have not been investigated further. In RET1 RNAi cells, the abundance of pre-edited RPS12 transcripts increased while their A-tails remained intact (Figure 7D). Apparently, inhibition of RNA editing did not affect the stability of the pre-edited mRNAs. Most importantly, partially edited mRNA (Figure 7D, right panel) did not have the long A/U extension characteristic of the fully edited form (Figure 7C and E). Consistent with the northern blotting results, analysis of the cRT–PCR clones confirmed that the 5' ends of RPS12 mRNAs with long A/U extensions contained only fully edited sequences (not shown). Hence, we have concluded that the A/U extension is added to the short A-tail, which is required and sufficient for the maintenance of mRNA undergoing U-insertion/deletion, upon completion of the editing process.

We report on the identification and functional characterization of the trypanosomal mitochondrial PAP termed KPAP1. Collectively, our results show that the short (20–25 nt) A-tails are essential for maintaining the steady-state level of never-edited and edited mRNAs. In the edited mRNAs, the long (100–200 nt) poly(A/U) extensions are added to the short A-tails upon completion of the editing process. It is possible that A/U structures are characteristic of translationally competent fully edited mRNAs. Although the nucleotide composition of 'long A-tails' in never-edited mRNA remains to be established, one may speculate that it should be similar to that of the fully edited transcripts. Our inference that the long 'A-tail' is a specific feature of never-edited and fully edited, but not partially edited, mRNAs disagrees with an earlier report (Militello and Read, 1999). We used northern hybridization with a selective probe for the extensively edited 5' end in RPS12 mRNA, and sequencing analysis of the 5' region to verify this conclusion.

The degradation of mRNAs lacking poly(A) tails appears to be a processive 3'–5' reaction that nevertheless does not affect non-adenylated pre-edited mRNAs. The rapid loss of non-adenylated edited mRNAs, and evidently functional RNA editing and gRNA processing machineries in KPAP1 RNAi cells prompted a search for the signal that converts the A-tail into a stabilizing feature of edited mRNA. We found that the U-insertion/deletion events beyond the first editing block impose a requirement for 3' adenylation to maintain the transcript. It is conceivable that upon binding of the next gRNA, the editosome dissociates or moves to the upstream block, exposing the edited mRNA to 3'–5' degradation. In cases of limited editing at the 5' end of mRNA (Cyb) or editing guided by a cis-gRNA (CO2), the edited mRNAs still depend on 3' adenylation. This may indicate cross-talk between the editing and polyadenylation complexes or deposition of editing factors on pre-existing short A-tails.

Purification of the polyadenylation complex supported the notion that RNA editing TUTase 1, an enzyme responsible for gRNA 3' uridylylation (Aphasizhev et al, 2003), is involved in mRNA 3'-end processing (Ryan and Read, 2005). Analyses of KPAP1 and RET1 RNAi cell lines suggested that synthesis of a short A-tail by KPAP1 is an initial processing step for pre-edited and never-edited mRNAs. The short tail's length is consistent with the extension patterns of recombinant KPAP1 (Figure 4D). It is appealing to speculate that after cleavage of the multi-cistronic precursor, the short A-tail is added to pre-edited and never-edited transcripts, which then become inefficient substrates for KPAP1. Quite possibly, the re-activation of such substrates for KPAP1 to synthesize longer extensions is achieved by uridylylation. Therefore, the A/U extensions may be synthesized by recycling of poly(A) and poly(U) polymerase activities, which are both present in the purified polyadenylation complex. The in organello phenomenon of UTP-stimulated, RET1-dependent mRNA degradation (Ryan and Read, 2005) may be rationalized assuming that an increase in the U content of A/U extensions leads to mRNA destabilization.

The critical question of signalling between completion of mRNA editing and 3'-end A/U-addition remains. In a possible scenario, the direct readout of RNA sequence generated de novo by RNA editing may be followed by the recruitment of an RET1-containing particle, in which KPAP1 and RET1 compete for the RNA substrate. Identification of five PPR proteins in the KPAP1 complex highlights these molecules as potential transcript-specific sensors. Indeed, PPR proteins ubiquitously function as sequence-recognition factors in post-transcriptional RNA processing in organelles (reviewed in Delannoy et al, 2007). Although the biological roles of all proteins reported in this study must be investigated further, it should be noted that PPR proteins may have general and transcript-specific roles. The results of PPR1 knockdown, for example, are consistent with its general role in the synthesis of 'long A-tails'. Upon induction of PPR1 RNAi, these structures were lost and mitochondrial function was compromised (Mingler et al, 2006; Pusnik et al, 2007). Predictably, the short A-tails and mRNA abundance were not affected, which leaves defects in translation of mRNAs missing A/U extensions as a plausible cause. Identification of the EF-Tu elongation factor as one of the major polypeptides in the KPAP1 complex also indicates a probable coupling of mitochondrial polyadenylation and translation processes.

Expression, mutagenesis, and purification of KPAP1

The KPAP1 gene was PCR-amplified, re-sequenced, and inserted into pET28c vector (Novagen) to generate N-terminal 6xHis fusion protein. Enzymatic assays and RNA substrates are described in Supplementary data.

Trypanosome culture and RNAi

The RNAi expression plasmids were generated by cloning a fragment of the KPAP1 gene (24–600) into p2T7-177 vector that allows for tetracycline-inducible, T7 RNA polymerase-driven expression (Wickstead et al, 2002). The construct was transfected into procyclic 29-13 T. brucei strains (Wirtz et al, 1999) followed by clonal selection of phleomycin-resistant cell lines. RNAi was performed as described by Djikeng et al (2004).

Mitochondrial extract preparation and fractionation

Mitochondria purification, lysate preparation, glycerol gradient fractionation, and adenylation reactions were carried out according to Pelletier et al (2007). Tandem affinity purification and sample preparation for mass spectrometry were performed as described (Aphasizhev and Aphasizheva, 2007).

Western blotting and immunoprecipitation

Rabbit polyclonal antibodies against the recombinant KPAP1 were purified on antigen-affinity column. For western blotting, 2.5 10⁶ cells were lysed in SDS loading buffer, separated on 8–16% gradient SDS–PAGE, and transferred to nitrocellulose membrane. For immunoprecipitation, purified mitochondria were resuspended at 0.3 g (wet weight)/ml in 50 mM HEPES (pH 8.0), 60 mM KCl, 10 mM MgCl₂, and 0.4% NP-40 and sonicated three times at 9 W for 5 s. The extract was spun at 18 000 g for 30 min to remove the debris. KPAP1 antibody (2.5 g) was pre-bound to 10 l of Dynabead-Protein A (Invitrogen) and incubated with 200 l of mitochondrial extract for 1 h at 4°C followed by three washes for 30 min with 0.3% of Triton X-100 in PBS.

Fluorescent microscopy

A KPAP1–eYFP fusion gene was cloned into the pLew79 vector, which allows for tetracycline-inducible expression, and the construct was transfected into procyclic 29-13 T. brucei (Wirtz et al, 1999). Parasites were incubated with 100 nM MitoTracker Red CMX ROS.

RNA analysis

Total RNA was isolated by a modified procedure (Chomczynski and Sacchi, 1987). Detailed methods for primer extension, qRT–PCR, northern blotting, RNase H digestion, gRNA labelling, mRNA 3' end cloning and statistical analysis, and oligonucleotides used are described in Supplementary data.

This paper is dedicated to the memory of our mentor Professor Lev L Kisselev. This work was supported by NIH grant AI064653 to RA. We thank Larry Simpson, Ken Stuart, Ulrich Göringer, and Jay Bangs for providing antibodies. Kind gifts of pET-HisD1/D12 and pLew79-MH-TAP plasmids from Stewart Shuman and Marilyn Parsons, respectively, are acknowledged. We appreciate the expert advice from Jason Carnes on qRT–PCR and Menna Hodge on flow cytometry. We thank Dmitry Maslov and members of the Aphasizhev laboratory for reading the manuscript and reviewers for helpful comments.

Abraham J, Feagin J, Stuart K (1988) Characterization of cytochrome c oxidase III transcripts that are edited only in the 3' region. Cell 55: 267–272 | Article | PubMed | ISI | ChemPort |

Aphasizhev R (2005) RNA uridylyltransferases. Cell Mol Life Sci 62: 2194–2203 | Article | PubMed | ChemPort |

Aphasizhev R, Aphasizheva I (2007) RNA editing uridylyltransferases of trypanosomatids. Methods Enzymol 424: 51–67

Aphasizhev R, Aphasizheva I, Simpson L (2003) A tale of two TUTases. Proc Natl Acad Sci USA 100: 10617–10622 | Article | PubMed | ChemPort |

Aphasizhev R, Aphasizheva I, Simpson L (2004) Multiple terminal uridylyltransferases of trypanosomes. FEBS Lett 572: 15–18 | Article | PubMed | ISI | ChemPort |

Aphasizhev R, Sbicego S, Peris M, Jang SH, Aphasizheva I, Simpson AM, Rivlin A, Simpson L (2002) Trypanosome mitochondrial 3' terminal uridylyl transferase (TUTase): the key enzyme in U-insertion/deletion RNA editing. Cell 108: 637–648 | Article | PubMed | ISI | ChemPort |

Benne R, Van den Burg J, Brakenhoff J, Sloof P, Van Boom J, Tromp M (1986) Major transcript of the frameshifted coxII gene from trypanosome mitochondria contains four nucleotides that are not encoded in the DNA. Cell 46: 819–826 | Article | PubMed | ISI | ChemPort |

Bhat GJ, Koslowsky DJ, Feagin JE, Smiley BL, Stuart K (1990) An extensively edited mitochondrial transcript in kinetoplastids encodes a protein homologous to ATPase subunit 6. Cell 61: 885–894 | Article | PubMed | ISI | ChemPort |

Bhat GJ, Souza AE, Feagin JE, Stuart K (1992) Transcript-specific developmental regulation of polyadenylation in Trypanosoma brucei mitochondria. Mol Biochem Parasitol 52: 231–240 | Article | PubMed | ISI | ChemPort |

Blum B, Bakalara N, Simpson L (1990) A model for RNA editing in kinetoplastid mitochondria: 'Guide' RNA molecules transcribed from maxicircle DNA provide the edited information. Cell 60: 189–198 | Article | PubMed | ISI | ChemPort |

Brogna S (1999) Nonsense mutations in the alcohol dehydrogenase gene of Drosophila melanogaster correlate with an abnormal 3' end processing of the corresponding pre-mRNA. RNA 5: 562–573 | Article | PubMed | ISI | ChemPort |

Brown SV, Hosking P, Li J, Williams N (2006) ATP synthase is responsible for maintaining mitochondrial membrane potential in bloodstream form Trypanosoma brucei. Eukaryot Cell 5: 45–53 | Article | PubMed | ChemPort |

Chomczynski P, Sacchi N (1987) Single-step method of RNA isolation by acid guanidium thiocyanate–phenol–chloroform extraction. Anal Biochem 162: 156–159 | Article | PubMed | ISI | ChemPort |

Decker CJ, Sollner-Webb B (1990) RNA editing involves indiscriminate U changes throughout precisely defined editing domains. Cell 61: 1001–1011 | Article | PubMed | ISI | ChemPort |

Delannoy E, Stanley WA, Bond CS, Small ID (2007) Pentatricopeptide repeat (PPR) proteins as sequence-specificity factors in post-transcriptional processes in organelles. Biochem Soc Trans 35: 1643–1647 | Article | PubMed | ChemPort |

Djikeng A, Shen S, Tschudi C, Ullu E (2004) Analysis of gene function in Trypanosoma brucei using RNA interference. Methods Mol Biol 265: 73–84 | PubMed | ChemPort |

Drozdz M, Palazzo SS, Salavati R, O'Rear J, Clayton C, Stuart K (2002) TbMP81 is required for RNA editing in Trypanosoma brucei. EMBO J 21: 1791–1799 | Article | PubMed | ISI | ChemPort |

Ernst NL, Panicucci B, Igo Jr RP, Panigrahi AK, Salavati R, Stuart K (2003) TbMP57 is a 3' terminal uridylyl transferase (TUTase) of the Trypanosoma brucei editosome. Mol Cell 11: 1525–1536 | Article | PubMed | ISI | ChemPort |

Feagin JE, Abraham J, Stuart K (1988) Extensive editing of the cytochrome c oxidase III transcript in Trypanosoma brucei. Cell 53: 413–422 | Article | PubMed | ISI | ChemPort |

Gagliardi D, Stepien PP, Temperley RJ, Lightowlers RN, Chrzanowska-Lightowlers ZM (2004) Messenger RNA stability in mitochondria: different means to an end. Trends Genet 20: 260–267 | Article | PubMed | ISI | ChemPort |

Golden DE, Hajduk SL (2005) The 3'-untranslated region of cytochrome oxidase II mRNA functions in RNA editing of African trypanosomes exclusively as a cis guide RNA. RNA 11: 29–37 | Article | PubMed | ChemPort |

Houseley J, LaCava J, Tollervey D (2006) RNA-quality control by the exosome. Nat Rev Mol Cell Biol 7: 529–539 | Article | PubMed | ISI | ChemPort |

Huang YS, Richter JD (2004) Regulation of local mRNA translation. Curr Opin Cell Biol 16: 308–313 | Article | PubMed | ISI | ChemPort |

Igo RP, Palazzo SS, Burgess ML, Panigrahi AK, Stuart K (2000) Uridylate addition and RNA ligation contribute to the specificity of kinetoplastid insertion RNA editing. Mol Cell Biol 20: 8447–8457 | Article | PubMed | ISI | ChemPort |

Kang X, Gao G, Rogers K, Falick AM, Zhou S, Simpson L (2006) Reconstitution of full-round uridine-deletion RNA editing with three recombinant proteins. Proc Natl Acad Sci USA 103: 13944–13949 | Article | PubMed | ChemPort |

Kao CY, Read LK (2005) Opposing effects of polyadenylation on the stability of edited and unedited mitochondrial RNAs in Trypanosoma brucei. Mol Cell Biol 25: 1634–1644 | Article | PubMed | ChemPort |

Koller J, Muller UF, Schmid B, Missel A, Kruft V, Stuart K, Goringer HU (1997) Trypanosoma brucei gBP21. An arginine-rich mitochondrial protein that binds to guide RNA with high affinity. J Biol Chem 272: 3749–3757 | Article | PubMed | ISI | ChemPort |

Koslowsky DJ, Yahampath G (1997) Mitochondrial mRNA 3' cleavage polyadenylation and RNA editing in Trypanosoma brucei are independent events. Mol Biochem Parasitol 90: 81–94 | Article | PubMed | ISI | ChemPort |

Kushner SR (2004) mRNA decay in prokaryotes and eukaryotes: different approaches to a similar problem. IUBMB Life 56: 585–594 | Article | PubMed | ISI | ChemPort |

Mair G, Shi HF, Li HJ, Djikeng A, Aviles HO, Bishop JR, Falcone FH, Gavrilescu C, Montgomery JL, Santori MI, Stern LS, Wang ZF, Ullu E, Tschudi C (2000) A new twist in trypanosome RNA metabolism: cis-splicing of pre-mRNA. RNA 6: 163–169 | Article | PubMed | ISI | ChemPort |

Martin G, Keller W (2007) RNA-specific ribonucleotidyl transferases. RNA 13: 1–16 | Article | PubMed | ChemPort |

Maslov DA, Simpson L (1992) The polarity of editing within a multiple gRNA-mediated domain is due to formation of anchors for upstream gRNAs by downstream editing. Cell 70: 459–467 | Article | PubMed | ISI | ChemPort |

Maslov DA, Sturm NR, Niner BM, Gruszynski ES, Peris M, Simpson L (1992) An intergenic G-rich region in Leishmania tarentolae kinetoplast maxicircle DNA is a pan-edited cryptogene encoding ribosomal protein S12. Mol Cell Biol 12: 56–67 | PubMed | ChemPort |

Militello KT, Read LK (1999) Coordination of kRNA editing and polyadenylation in Trypanosoma brucei mitochondria: complete editing is not required for long poly(A) tract addition. Nucleic Acids Res 27: 1377–1385 | Article | PubMed | ChemPort |

Militello KT, Read LK (2000) UTP-dependent and -independent pathways of mRNA turnover in Trypanosoma brucei mitochondria. Mol Cell Biol 20: 2308–2316 | Article | PubMed | ChemPort |

Mingler MK, Hingst AM, Clement SL, Yu LE, Reifur L, Koslowsky DJ (2006) Identification of pentatricopeptide repeat proteins in Trypanosoma brucei. Mol Biochem Parasitol 150: 37–45 | Article | PubMed | ChemPort |

Missel A, Souza AE, Norskau G, Goringer HU (1997) Disruption of a gene encoding a novel mitochondrial DEAD-box protein in Trypanosoma brucei affects edited mRNAs. Mol Cell Biol 17: 4895–4903 | PubMed | ISI | ChemPort |

Nagaike T, Suzuki T, Katoh T, Ueda T (2005) Human mitochondrial mRNAs are stabilized with polyadenylation regulated by mitochondria-specific poly(A) polymerase and polynucleotide phosphorylase. J Biol Chem 280: 19721–19727 | Article | PubMed | ChemPort |

Ochsenreiter T, Cipriano M, Hajduk SL (2007) KISS: the kinetoplastid RNA editing sequence search tool. RNA 13: 1–4 | Article | PubMed | ChemPort |

Panigrahi AK, Zikova A, Dalley RA, Acestor N, Ogata Y, Anupama A, Myler PJ, Stuart KD (2007) Mitochondrial complexes in Trypanosoma brucei: a novel complex and a unique oxidoreductase complex. Mol Cell Proteomics 7: 534–545.  | Article | PubMed | ChemPort |

Pelletier M, Read LK, Aphasizhev R (2007) Isolation of RNA binding proteins involved in insertion/deletion editing. Methods Enzymol 424: 69–96

Proudfoot N (2004) New perspectives on connecting messenger RNA 3' end formation to transcription. Curr Opin Cell Biol 16: 272–278 | Article | PubMed | ISI | ChemPort |

Puig O, Caspary F, Rigaut G, Rutz B, Bouveret E, Bragado-Nilsson E, Wilm M, Seraphin B (2001) The tandem affinity purification (TAP) method: a general procedure of protein complex purification. Methods 24: 218–229 | Article | PubMed | ISI | ChemPort |

Pusnik M, Small I, Read LK, Fabbro T, Schneider A (2007) Pentatricopeptide repeat proteins in Trypanosoma brucei function in mitochondrial ribosomes. Mol Cell Biol 27: 6876–6888 | Article | PubMed | ChemPort |

Read LK, Myler PJ, Stuart K (1992) Extensive editing of both processed and preprocessed maxicircle CR6 transcripts in Trypanosoma brucei. J Biol Chem 267: 1123–1128 | PubMed | ISI | ChemPort |

Read LK, Stankey KA, Fish WR, Muthiani AM, Stuart K (1994) Developmental regulation of RNA editing and polyadenylation in four life cycle stages of Trypanosoma congolense. Mol Biochem Parasitol 68: 297–306 | Article | PubMed | ISI | ChemPort |

Ryan CM, Militello KT, Read LK (2003) Polyadenylation regulates the stability of Trypanosoma brucei mitochondrial RNAs. J Biol Chem 278: 32753–32762 | Article | PubMed | ChemPort |

Ryan CM, Read LK (2005) UTP-dependent turnover of Trypanosoma brucei mitochondrial mRNA requires UTP polymerization and involves the RET1 TUTase. RNA 11: 763–773 | Article | PubMed | ISI | ChemPort |

Schnaufer A, Clark-Walker GD, Steinberg AG, Stuart K (2005) The F1-ATP synthase complex in bloodstream stage trypanosomes has an unusual and essential function. EMBO J 24: 4029–4040 | Article | PubMed | ChemPort |

Simpson L, Aphasizhev R, Gao G, Kang X (2004) Mitochondrial proteins and complexes in Leishmania and Trypanosoma involved in U-insertion/deletion RNA editing. RNA 10: 159–170 | Article | PubMed | ChemPort |

Stagno J, Aphasizheva I, Aphasizhev R, Luecke H (2007a) Dual role of the RNA substrate in selectivity and catalysis by terminal uridylyl transferases. Proc Natl Acad Sci USA 104: 14634–14639 | Article | ChemPort |

Stagno J, Aphasizheva I, Rosengarth A, Luecke H, Aphasizhev R (2007b) UTP-bound and Apo structures of a minimal RNA uridylyltransferase. J Mol Biol 366: 882–899 | Article | ChemPort |

Stevenson AL, Norbury CJ (2006) The Cid1 family of non-canonical poly(A) polymerases. Yeast 23: 991–1000 | Article | PubMed | ChemPort |

Stuart KD, Schnaufer A, Ernst NL, Panigrahi AK (2005) Complex management: RNA editing in trypanosomes. Trends Biochem Sci 30: 97–105 | Article | PubMed | ISI | ChemPort |

Sturm NR, Simpson L (1990) Partially edited mRNAs for cytochrome b and subunit III of cytochrome oxidase from Leishmania tarentolae mitochondria: RNA editing intermediates. Cell 61: 871–878 | Article | PubMed | ISI | ChemPort |

Vanhamme L, Perez-Morga D, Marchal C, Speijer D, Lambert L, Geuskens M, Alexandre S, Ismaïli N, Göringer U, Benne R, Pays E (1998) Trypanosoma brucei TBRGG1, a mitochondrial oligo(U)-binding protein that co-localizes with an in vitro RNA editing activity. J Biol Chem 273: 21825–21833 | Article | PubMed | ISI | ChemPort |

Wickstead B, Ersfeld K, Gull K (2002) Targeting of a tetracycline-inducible expression system to the transcriptionally silent minichromosomes of Trypanosoma brucei. Mol Biochem Parasitol 125: 211–216 | Article | PubMed | ISI | ChemPort |

Wirtz E, Leal S, Ochatt C, Cross GA (1999) A tightly regulated inducible expression system for conditional gene knock-outs and dominant-negative genetics in Trypanosoma brucei. Mol Biochem Parasitol 99: 89–101 | Article | PubMed | ISI | ChemPort |

Zikova A, Kopecna J, Schumacher MA, Stuart K, Trantirek L, Lukes J (2008) Structure and function of the native and recombinant mitochondrial MRP1/MRP2 complex from Trypanosoma brucei. Int J Parasitol (in press)

Main navigation

• Search EMBO Journal

  Search

• Advanced search

• The EMBO Journal

  • Home
  • Aims and scope
  • Current issue
  • Advance online publication
  • Web focuses
  • Archive:
  • Browse by issue
  • Browse by subject
  • Browse by category
  • Free online sample issue
  • Press releases

• Authors and Referees

  • Editorial process
  • Guide for authors
  • Submit an article
  • Guide for referees
  • Editors and Editorial Board
  • Contact editorial office

• Customer Service

  • Subscribe
  • Order sample copy
  • Purchase articles
  • Reprints and permissions
  • Contact NPG
  • Advertising