RBP38, a novel RNA-binding protein from trypanosomatid mitochondria, modulates RNA stability

Abstract

We describe here the isolation and characterization of a novel RNA-binding protein, RBP38, from Leishmania tarentolae mitochondria. This protein does not contain any known RNA-binding motifs and is highly conserved among the trypanosomatids, but no homologues were found in other organisms. Recombinant LtRBP38 binds single and double-stranded (ds) RNA substrates with dissociation constants in the 100 nM range, as determined by fluorescence polarization analysis. Downregulation of expression of the homologous gene, TbRBP38, in procyclic Trypanosoma brucei by using conditional dsRNA interference resulted in 80% reduction of steady-state levels of RNAs transcribed from both maxicircle and minicircle DNA. In organello pulse-chase labeling experiments were used to determine the stability of RNAs in mitochondria that were depleted of TbRBP38. The half-life of metabolically labeled RNA decreased from approximately 160 to approximately 60 min after depletion. In contrast, there was no change in transcriptional activity. These observations suggest a role of RBP38 in stabilizing mitochondrial RNA.

FIG. 1.

Isolation of LtRBP38 from mitochondrial extracts of L. tarentolae by using two different purification protocols. (A and B) Method 1 is a gel shift assay with a partially paired duplex RNA (substrate 3, panel C). Mitochondrial extracts were chromatographically enriched for RNA-binding activity by using Blue-Sepharose, heparin, and MonoQ columns. The final MonoQ column was developed with a 0 to 1 M KCl gradient. The collected fractions were used in the gel shift assay shown. (B) The 500 mM KCl peak fraction (indicated in panel A by an arrow) was analyzed on a SDS-polyacrylamide gel electrophoresis, and the gel was stained with Coomassie blue. The fraction contained a major 38-kDa band (arrow) that was isolated and subjected to microsequencing. (C) Method 2 is a gel shift assay for RNA binding to single-stranded RNA using uniformly labeled T7-transcribed gRPS12-III gRNA as the probe. The mitochondrial extract was chromatographically enriched with Q-Sepharose, S-Sepharose, and MonoQ columns. The final MonoQ column was developed by using a gradient of 50 to 500 mM KCl, and the peak activity was subjected to UV cross-linking, RNase digestion, and SDS-gel electrophoresis. The major radioactively labeled cross-link migrated at 38 kDa as a single band (Fig. 1C). This band was eluted from the gel (pooled from lanes 3 to 5) and subjected to N-terminal sequencing. The sequence was identical to the predicted amino acid sequence of the protein isolated by method 1: 27 amino acids from the N terminus.

FIG. 2.

Nucleotide and amino acid sequence of LtRBP38. Nucleotide sequence and the predicted amino acid sequence are shown. The putative mitochondrial targeting signal consisting of the first 26 amino acids is shown in boldface italics. The six tryptic peptides used for cloning are highlighted in boldface. The 93-amino-acid fragment expressed for antiserum production is indicated in brackets. For the expression of the full-length gene, the boxed 37 codons, representing rare E. coli codons, were mutated as indicated by lowercase letters.

FIG. 3.

Sequence alignment of RBP38 from T. brucei (Tb, accession no. AY187285), L. tarentolae (Lt, accession no. AY187286), and L. major (Lm, accession no. AL138558) showing the high conservation within the trypanosomatids. Identical amino acids are boxed in black, conservative changes are boxed in gray.

FIG. 4.

Binding of recombinant LtRBP38 to various RNA substrates. (A) The RNAs were 5′ labeled with fluorescein, and the binding was monitored in a Beacon 2000 fluorescence polarization system. The binding data were normalized, and the curves for an average of three measurements are shown. (B) The K_d values were calculated from the fitted curves. (C) Substrate RNAs: 1, ND7.4x (5); 2, gND7x[+3] (5); 3, RNA1+RNA2 annealed; 4, ND7-AU; 5, ND7-CU; 6, Cyb-CU (14). The following RNAs did not show binding: 7, TAR (13); 8, NR; 9, NR+comp1 annealed; 10, NR+comp2 annealed; 11, NF (25).

FIG. 5.

Downregulation of TbRBP38 expression in procyclic T. brucei by inducible RNAi. RNAi was induced by the addition of 1 mg of tetracycline/liter. (A) Cumulative growth curve with or without tetracycline. Cell growth is expressed as the number of cell divisions. (B) Analysis of TbRBP38 downregulation in induced and mock-treated cells. A total of 20 μg of total protein was loaded per lane and then analyzed with anti-LtRBP38 antisera. p70 was used as a loading control. Next, 35 μg of RNA per time point was analyzed by primer extension as described in Materials and Methods. Calmodulin was used as a loading control.

FIG. 6.

Downregulation of TbRBP38 affects the steady-state level of total mitochondrial RNA in vivo. Total RNA was isolated at the indicated time points after induction of RNAi. Relative abundance of specific RNAs was determined by poisoned primer extension with 35 μg of RNA per reaction. The cytoplasmic calmodulin mRNA was used as the normalization signal. (A) Representative autoradiogram detecting the maxicircle-encoded Co1 transcript. (B) Minicircle-encoded gRNAs (total population) were visualized by labeling 1 μg of total RNA with guanylyltransferase and [γ-³²P]GTP. The cytoplasm-specific 5S rRNA was also labeled and was used as a loading control. In the lower panel is shown the abundance of a specific gRNA (gCyb-II), monitored by primer extension as described above. (C) Quantitation of relative steady-state levels of nine mitochondrial transcripts during induction of RNAi.

FIG. 7.

Downregulation of TbRBP38 affects RNA stability in organello. Mitochondria were isolated ∼72 h after induction of RNAi (plus tet) or mock treatment (minus tet). Mitochondrial RNA was metabolically labeled with [α-³²P]GTP, harvested at the indicated time points, and quantitated by using a filter-binding assay. (A) Degradation kinetics of mitochondrial RNA in organello. The kinetic data collected from six independent mitochondrial isolations were linearized, and the decay rates were calculated from the pooled data. (B) Kinetics of [α-³²P]GTP incorporation into mitochondrial RNA as a marker for transcription.

FIG. 8.

Downregulation of TbRBP38 does not affect RNA editing in vivo, as shown by primer extension analysis of mitochondrial mRNAs (35 μg of RNA per reaction) during RNAi. (A) Time course analysis of Cyb (left panel) and ND7 (right panel) mRNAs detecting edited (ed) and preedited (pre) RNA. The cytoplasmic calmodulin mRNA (calm) was analyzed in the same reaction as an internal control for loading and assay performance. (B) Quantitative analysis of five edited mRNAs. The percentage of edited mRNA was calculated.