tRNASec is transcribed by RNA polymerase II in Trypanosoma brucei but not in humans

Abstract

Nuclear-encoded tRNAs are universally transcribed by RNA polymerase III (Pol-III) and contain intragenic promoters. Transcription of vertebrate tRNA(Sec) however requires extragenic promoters similar to Pol-III transcribed U6 snRNA. Here, we present a comparative analysis of tRNA(Sec) transcription in humans and the parasitic protozoa Trypanosoma brucei, two evolutionary highly diverged eukaryotes. RNAi-mediated ablation of Pol-II and Pol-III as well as oligo-dT induced transcription termination show that the human tRNA(Sec) is a Pol-III transcript. In T. brucei protein-coding genes are polycistronically transcribed by Pol-II and processed by trans-splicing and polyadenylation. tRNA genes are generally clustered in between polycistrons. However, the trypanosomal tRNA(Sec) genes are embedded within a polycistron. Their transcription is sensitive to α-amanitin and RNAi-mediated ablation of Pol-II, but not of Pol-III. Ectopic expression of the tRNA(Sec) outside but not inside a polycistron requires an added external promoter. These experiments demonstrate that trypanosomal tRNA(Sec), in contrast to its human counterpart, is transcribed by Pol-II. Synteny analysis shows that in trypanosomatids the tRNA(Sec) gene can be found in two different polycistrons, suggesting that it has evolved twice independently. Moreover, intron-encoded tRNAs are present in a number of eukaryotic genomes indicating that Pol-II transcription of tRNAs may not be restricted to trypanosomatids.

Figure 1.

An intragenic T-stretch abolishes transcription of full-length human tRNA^Sec. (A) Predicted secondary structure of the human tRNA^Tyr and tRNA^Sec, respectively. Arrows indicate the positions where a synthetic Pol-III termination signal consisting of five adjacent thymidine (T₅) has been inserted into the genes of the indicated tRNAs. In the case of the tRNA^Tyr gene this position also corresponds to the position of the intron. (B) Left panel, total RNA samples isolated from control cells (neg), from transgenic cell lines overexpressing wild-type tRNA^Tyr (wt) or the tRNA^Tyr variant carrying the T-stretch were analyzed for the presence of tRNA^Tyr by specific oligonucleotide hybridization on northern blots. The positions of the intron-containing tRNA^Tyr (pre-tRNA^Tyr), the mature wild-type tRNA^Tyr (wt-tRNA^Tyr) and the abortive transcript corresponding to the 5′-half of the tRNA^Tyr (short-tRNA^Tyr) are indicated. Right panel, same as in the left panel but analysis was done for the human tRNA^Sec. The positions of the wild-type (wt-tRNA^Sec) and the abortive transcript corresponding to the 5′-half of the tRNA^Sec (short-tRNA^Sec) are indicated. For the experiments on the left panel the cell line expressing the wild-type-tRNA^Sec was used as a negative control (neg) whereas for the experiments on the right panel the control consisted of the cell expressing the wild-type tRNA^Tyr. The bottom panels serve as loading controls and show the ethidium bromide stained gel segment that contains the tRNAs.

Figure 2.

RNAi of Pol-III inhibits expression of human tRNA^Sec. (A) Real-time qPCR analysis of the mRNA levels of the core subunit of Pol-II (white bars) or Pol-III (grey bars) from HeLa cells transfected with either the empty plasmid (C) or with plasmids expressing shRNAs targeting the mRNAs of the core subunit of Pol-II or Pol-III, respectively. For both RNA polymerases, cells were transfected with two independent shRNA expressing plasmids that target distinct regions (A and B) of the corresponding mRNAs. All mRNAs levels have been normalized to the level of the 18S rRNA. The levels of mRNAs encoding the core subunits of Pol-II and Pol-III in the cell line transfected with the empty plasmid (C) was set to 1. (B) Northern analysis of 4 µg of total RNA isolated from control cells and from cells undergoing RNAi. Four days after transfection total RNA was isolated and analyzed for the presence of the indicated RNAs by specific oligonucleotide hybridization. The bottom panels serve as loading controls and show the ethidium bromide stained gel segment that contains the tRNAs. (C) Quantification of the results shown in (B). The signal in the control cells (C) that do not express shRNAs was set to 1.

Figure 3.

Schematic illustration of the two trypanosomal tRNA^Sec genes and their genomic context (drawn to scale). The two tRNA^Sec genes including the flanking sequences indicated in bold are identical. Tb09.160.1090 encodes a putative serine/threonine protein kinase, the two other ORFs are annotated as hypothetical proteins of unknown function. The position and sequence of the polyadenylation site (for Tb09.160.1090) and the splice acceptor site (for Tb09.160.1070) as determined by 3′ and 5′ RACE are indicated by A and S, respectively. The functional splice acceptor site detected by deep sequencing of a poly(A) enriched SL-containing cDNA library of procyclic and bloodstream T. brucei is indicated by S′.

Figure 4.

Ectopic expression of the trypanosomal tRNA^Sec gene requires an external promoter. (A) Predicted secondary structure of the tagged tRNA^Sec and tRNA^Met−i. The 2-nt changes introduced as tags are indicated. The tags allow the specific detection of the two tRNA variants by oligonucleotide hybridizations. (B) Cassette used for ectopic expression of the tRNA^Sec (the one encoded on the shorter intergenic region) and tRNA^Met−i, respectively. It contains the Pol-I procyclin promoter followed by two tetracycline operators and a splice acceptor site (SAS). The tagged tRNA^Sec was expressed in its own genomic context, whereas the tagged tRNA^Met−i was fused to the 5′-flanking region of a trypanosomal tRNA^Leu but retained its own 3′-flanking region (31). (C) Northern analyses of total RNA isolated from cell lines expressing the tetracycline repressor and transfected with the constructs shown in (A). tRNA^Sec, cell line expressing the tagged tRNA^Sec. tRNA^Met−i cell line expressing the tagged tRNA^Met−i. Top panel, hybridization with oligonucleotides that specifically recognize the tagged tRNA^Sec and the tagged tRNA^Met−i, respectively. Middle panel, same blot as above but reprobed with an oligonucleotide recognizing both the tagged and the endogenous tRNA^Sec. Bottom panel, ethidium bromide stained tRNA region of the gel used for the northern analyses. (D) To scale drawing of the wild-type Tb-VDAC locus (41) and the situation after homologous recombination leading to replacement of one allele by a tRNA^Sec/G418 resistance cassette. Relevant AvaII (A) restriction sites are indicated. The jagged line marks the probe used in the Southern analysis. Left panel: Southern analysis of genomic DNA isolated from the parental cell line (Tb-VDAC, +/+) and from a single knock out Tb-VDAC cell line (Tb-VDAC, +/−) containing the tRNA^Sec/G418 resistance cassette. Right panel, northern blot of total RNA isolated from the same cell lines that were analyzed by the Southern blot hybridized with a probe specific for the tagged tRNA^Sec (top panel). The same blot was reprobed with a probe recognizing both the tagged and the endogenous tRNA^Sec (middle panel). Bottom panel, ethidium bromide stained tRNA region of the gel used for the northern analyses.

Figure 5.

Transcription of the tRNA^Sec gene is α-amanitin-sensitive. (A) UTP-labeled RNA was synthesized in permeabilized trypanosomes in the absence and presence of 10 and 200 µg/ml of α-amanitin and hybridized to nitrocellulose filters containing immobilized cloned DNA of the indicated genes (pTZ18: vector control; SL: spliced leader). (B) PhosphorImager quantitation of the results as shown in panel (A). The samples without α-amanitin were set to 1. Values represent means of two experiments. The variation between the values of two experiments was <15%.

Figure 6.

Effect of RNAi-mediated ablation of Pol-II and Pol-III activities on steady state levels of different RNAs. (A) Growth of uninduced and induced procyclic T. brucei RNAi cell lines downregulated for RPB9 (30) or the largest subunit of Pol-III. Inset: downregulation of the mRNA encoding the largest subunit of Pol-III was verified by northern blot analysis 24 h after induction of RNAi. (B) Northern analyses of total RNA isolated from uninduced (0) and induced (24, 48 h) RPB9 and Pol-III RNAi cell lines. RNA was resolved in 8 M urea on a 10% polyacrylamide gel and hybridized with oligonucleotides specifically detecting the transcripts indicated on the right. The tRNA region of the corresponding ethidium bromide stained gel is shown at the bottom. (C) Quantitation of the northern blots shown in (B). Signals were normalized to the cytosolic M6 rRNAs not affected by ablation of either RPB9 or Pol-III. For tRNA^Sec and tRNA^Ile the means of three experiments are shown. The signal in the uninduced cells was set to 1. Standard errors and the relevant P-values (Student’s t-test, one-tailed, paired) are indicated.