Fig 1.
Schematic representation of reporter constructs.
All constructs include the Rluc (Renilla luciferase) reporter gene and a T7 promoter site upstream of the 5’ UTR for in vitro transcription. The ToRSV-Rasp1 RNA2 is shown at the top along with the AUG1 and AUG3 putative start codons and the UAA stop codon. The VRV reporter transcript consists of the Rluc ORF (black box) flanked by the 5’ region (first 443 nts) and complete 3’ UTR of ToRSV-Rasp1 RNA2 (black lines). The control ARA transcript, includes the 5’ and 3’ UTRs from the Arabidopsis thaliana actin gene (blue lines). A 50 nts stretch of polyA sequence followed by a HindIII restriction site was added at the 3’ end of all constructs. Additional restriction sites that facilitate cloning of derivative constructs are shown by coloured triangles.
Fig 2.
Regions of sequence identity in the 5’ and 3’ regions of the genomic RNAs of selected ToRSV isolates.
(A) Schematic diagram of the ToRSV RNA2 showing regions (orange bars) with high sequence identities amongst ToRSV isolates. The 5’ region of the two genomic RNAs of many ToRSV isolates are nearly identical in the first 900 nts. The first 150 nts and last 650 nts of the 3’ UTRs of the two genomic RNAs are also highly conserved amongst ToRSV isolates. (B) Sequences alignments of the first 500 nts of the two genomic RNAs (RNA1: R1, RNA2: R2) from selected ToRSV isolates showing putative start codons. The three putative start codons are shown in red. The predicted complementary stem sequences of a putative stem-loop (5’ SL) structure located after the first AUG are shown with arrows. NCBI accession number are as follows: ToRSV-13C280 R1 (KM083890), ToRSV-13C280 R2 (KM083891), ToRSV-GYV R1 (KM083892), ToRSV-GYV R2 (KM083893), ToRSV-Rasp1 R1 (KM083894), ToRSV-Rasp1 R2 (KM083895). Please see S1 and S2 Figs for additional sequence alignments of the 5’ region and 3’ UTR of ToRSV RNAs.
Fig 3.
In vitro translation of the VRV transcript initiates at the first AUG of the viral 5’ UTR.
(A) Schematic representation of the wild-type reporter transcripts (VRV) with the three putative AUG-start codons (AUG1, AUG2, AUG3). Fusion Rluc proteins produced from each AUG are shown with rectangles along with their predicted molecular mass. The solid black rectangles represent the Rluc protein and the grey rectangles represent the portion of viral polyprotein sequences fused in-frame with Rluc after translation initiation at AUG1 or AUG2. (B) In vitro translation of WT and AUG mutant transcripts. Detection of Rluc by radiolabelling or after probing with Rluc antibody shows the Rluc proteins produced from the wild-type VRV transcript and from the AUG to AGG mutant derivatives. Pictures of EtBr-stained RNA show the integrity of the corresponding transcripts used for the in vitro translation. Migration of molecular mass size markers is shown on the left.
Fig 4.
The ToRSV 5’ region and 3’ UTR are both required for efficient translation from reporter transcripts.
(A) Schematic representation of VRV and derivative transcripts. In the derivative transcripts, the viral 5’ region and/or 3’ UTR (black lines) are replaced by the actin 5’ and/or 3’ UTR (blue lines). Grey rectangles represent the fusion of viral sequences to Rluc after translation initiation at the first AUG codon. The black pentagon structure represents the 7-methylguanylate cap on the ARA transcript (cARA, for capped ARA) that is enzymatically linked during the in vitro transcription reaction. (B) Luciferase activity of in vitro translated VRV and derivative transcripts. Rluc activity of the capped ARA (cARA) control transcript is comparable to that of the uncapped VRV transcript (no significant difference between VRV and cARA, P>0.05). Rluc activities of uncapped derivative transcripts missing the 5’ region and/or 3’ UTR from ToRSV-Rasp1 RNA2 (VRA, ARV and ARA) are significantly lower (**, P<0.001) compared to VRV (Student’s t-test, two-tailed, n = 4). Luminescence values obtained after translation of each derivative reporter transcripts were normalized to that obtained for the VRV transcript. Results are shown as the average of four independent experiments, each with three technical repeats. Error bars represent the standard deviation. (C) Protein blot showing the Rluc expression from transcripts shown in B. Proteins were visualized by immunoblotting with an anti-Rluc antibody. EtBr staining of RNA transcripts run on an agarose gel shows the integrity of the transcripts. (D) Schematic representation of VRV and V76RV derivative transcripts. V76RV includes a deletion of the viral polyprotein coding region (deletion from AUG1 up to the nucleotide immediately upstream of AUG3). (E) Luciferase activity of in vitro translated VRV and V76RV reporter transcripts. Rluc activity of in vitro translated V76RV is significantly lower than that of the VRV transcripts (**, P<0.001) (Student’s t-test, two-tailed, n = 5). Luminescence (Rluc) values are normalized to that obtained for the VRV transcripts. Results are shown as the average of five independent experiments, each with three technical repeats. Error bars represent the standard deviation. (F) Protein immunoblot showing the Rluc expression from transcripts shown in E. Pictures of EtBr-stained RNA show the integrity of the transcripts.
Fig 5.
Cap analog inhibits translation of uncapped VRV transcripts.
(A) Cap analog was added at increasing concentrations ranging from 0 to 150 picomoles per 50 μl translation reaction volume that also contained one picomole of reporter transcripts (uncapped VRV or capped ARA) and the luminescence activity of the translation reactions was measured. Experiments were repeated four times using varying concentrations of cap and a representative result is shown. Error bars represent the standard deviation of three technical repeats. (B) Translation inhibition of uncapped VRV or capped ARA (cARA) transcripts in the presence of a 125-molar excess of cap analog. Reactions were conducted as above. Translation of 1 picomole of either transcript (VRV or cARA) was reduced significantly in the presence of 125 picomoles of cap analog (**, P<0.001, Student’s t-test, two-tailed, n = 4). Results are presented as an average of four independent experiments, each with three technical repeats. Error bars represent the standard deviation.
Fig 6.
Translation of capped ARA control transcripts is inhibited by the viral 3’ UTR supplied in trans.
(A) Luciferase assay showing translation inhibition of capped reporter transcripts (cARA) with increasing concentration of viral 3’ UTR (V3’) supplied in trans. Reporter transcript (cARA) was used at a concentration of one picomole per 50 μl of WGE translation reactions. Either a control transcript (C, derived from vector sequences) or the V3’ transcript were supplied at various concentrations (0, 2, 5, 10 and 20 picomoles per 50 μl of translation reactions) and the luminescence was measured after in vitro translation. The luminescence values of reporter transcripts with added inhibitory transcripts (C or V3’) were normalized to those obtained for the cARA in the absence of inhibitory transcripts. Results are shown as the average values obtained from two independent experiments, each with three technical repeats. Error bars represent the standard deviation. (B) Inhibition of translation of cARA transcripts in the presence of a 20-fold excess of control or viral 3’ UTR provided in trans. Translation reactions were programmed as in A with 1 picomole of cARA transcripts and 20 picomoles of either V3’ or C transcripts. Translation of cARA was reduced significantly (**, P <0.001) in the presence of 20 picomoles of V3’ transcripts (Student’s t-test, two-tailed, n = 3). Results are shown as an average of three independent experiments, each with three technical repeats. Error bars indicate the standard deviation.
Fig 7.
A 386 nucleotides region of the viral 3’ UTR is necessary and sufficient for translation of reporter transcripts in conjunction with the viral 5’ region.
(A) Graphical representation of VRV derivative transcripts containing non-overlapping deletions of ~400 nts regions in the 3’ UTR. Numbers in parentheses indicate the region of the ToRSV-Rasp1 RNA2 3’ UTR included in the construct (WT and Δ124) or the regions that were deleted from constructs Δ1 to Δ4 (numbering from the 5’ end of RNA2). (B) Luminescence activity from the in vitro translated VRV and deletion derivative transcripts. Luciferase activities obtained from the deletion derivative transcripts VRVΔ1 and VRVΔ124 are comparable (P>0.05) to that of the WT VRV transcript. Those of VRVΔ2 and VRVΔ4 are only slightly reduced (25% and 33% reduction, respectively, *, P<0.05) compared to WT VRV. In contrast, the luminescence activity obtained from the VRVΔ3 transcript is significantly reduced (>70% reduction, **, P<0.001) compared to WT VRV (Student’s t-test, two-tailed, n = 3). The luminescence values from the derivative transcripts are normalized to the value obtained from VRV transcripts and the averages of three independent experiments, each with three technical repeats, are shown. Error bars represent the standard deviation.
Fig 8.
Delineation of sequences required for translation of uncapped VRV transcripts in region 3 of the 3’ UTR.
(A) Graphical representation of VRV derivative transcripts containing non-overlapping 100 nts deletions in region 3 of the 3’ UTR. Numbers in parentheses indicate the region of the ToRSV-Rasp1 RNA2 3’ UTR included in the construct (WT and Δ124) or the regions that were deleted from constructs Δ3 and Δ3a to Δ3d (numbering from the 5’ end of RNA2). (B) Luminescence values from the in vitro translated VRV and deletion derivative transcripts. As shown in Fig 7, deletion of the entire region 3 (Δ3) resulted in a significant reduction of translation activity (**, <0.001) (Student’s t-test, two-tailed, n = 2). Translation of the VRV reporter transcript was also significantly decreased upon deletion of the first two 100 nts stretches (Δ3a and Δ3b) in region 3 of the 3’ UTR (*, P<0.05) (Student’s t-test, two-tailed, n = 2). The luminescence values from the derivative transcripts are normalized to the value obtained from VRV transcripts and the average of two independent experiments, each with three technical repeats, are shown. Error bars represent the standard deviation.
Fig 9.
Predicted secondary structures of the 5’ region of ToRSV-Rasp1 RNA2 and of region 3a-b of the 3’ UTR.
(A) Predicted secondary structure of the 5’ region (first 443 nts up to the third AUG codon) of ToRSV-Rasp1 RNA2. The three AUG codons are shown in green. The predicted 5’ SL located 17 nts downstream of the first AUG is shown along with the predicted free energy. A putative base pairing (sequence complementarity) with a sequence from region 3a-b of the 3’ UTR shown in panel C is also indicated with the light blue highlight (B) Close-up of the 5’ SL structure. Sequence variation occurring in ToRSV isolates are shown with the following color code: ToRSV-Rasp1 RNA2 (black), ToRSV-Rasp1 RNA1 (green), ToRSV-GYV RNA1 and RNA2 (red), ToRSV-13C280 RNA2 (purple), ToRSV-13C280 RNA1 (blue). Please note that complementarity of sequences in the stem is conserved for all isolates. Please see sequence alignment of the corresponding region in Fig 2. (C) Predicted secondary structure of region 3a-b (nts 6789–6988) of the 3’ UTR of ToRSV-Rasp1 RNA2. A putative base pairing (sequence complementarity) with a sequence from the 5’ region shown in panel A is indicated in light blue. For all panels, secondary structures were predicted using the mFold webserver [58]. The obtained dot-bracket format files were then loaded into RNA2Drawer (web application) [59] for drawing and visualization. Polypyrimidine stretches in exposed loops or bulges in the predicted structures are highlighted in yellow. Please see S4 Fig for secondary structure predictions of the same RNA regions for other selected ToRSV isolates.
Fig 10.
Comparison of identified translation-enhancing elements and predicted secondary structures in BRV and ToRSV-Rasp1 RNA2s.
(A) Schematic representation of the location of translation-enhancing elements. The RNA2s of ToRSV and BRV are shown with the coding region of the polyprotein represented by thin horizontal black lines and the UTRs by thicker horizontal black lines. Regions in the 3’ UTRs identified as playing a critical role in translation are shown in red. A region of the BRV 3’ UTR identified as playing a modest role in translation is shown in orange. (B) Predicted secondary structures of the ToRSV-Rasp1 RNA2 5’ region and 3’ UTR region 3a-b. Structures discussed in the text are shown. For a more detailed representation of ToRSV RNAs features please refer to Figs 9 and S4. (C) Predicted secondary structures of the BRV RNA2 5’ region and 3’ UTR region A2. Structures discussed in the text are shown. For a more detailed representation of BRV RNA2 features (including a predicted stem-loop in region C2), please refer to S6 Fig. For all panels, identified regions of sequence complementarities between the 5’ and 3’ UTRs are shown in light blue. Polypyrimidine stretches are shown in yellow. The first AUG and the stop codon are highlighted in green and red, respectively.