Cap-Independent Translational Enhancement by the 3′ Untranslated Region of Red Clover Necrotic Mosaic Virus RNA1

Plasmid clones.Full-length cDNA clones of RCNMV Australian strain RNA1 (pRC1|G) and RNA2 (pRC2|G) were kindly provided by S. A. Lommel (50). Plasmids carrying the Renilla Luc gene (pSP64-RLUC), from which poly(A)₃₀-tailed mRNA transcripts could be generated by in vitro transcription with SP6 RNA polymerase (M. Ishikawa, Y. Chiba, and S. Naito, unpublished results), were kindly provided by M. Ishikawa. Plasmid p3′-8, which contains a 60-nt poly(A) tract, was a generous gift from K. A. White.

To facilitate the construction of deletion clones, an NcoI site was created at the p27 open reading frame (ORF) initiation codon in pRC1|G by using PCR-based in vitro mutagenesis (18). In the description of PCR primers, underlining indicates an introduced restriction enzyme site and boldface indicates a mutated nucleotide. A cDNA fragment of RCNMV RNA1 from nt 1 to 131 was amplified by PCR from pRC1|G using primers 5′RS1T7 (5′-GCGAGCTCTAATACGACTCACTATAGACAAACGTTTTACCGGTTTG-3′; the T7 promoter sequence is in italics) and AC5′UTR/Nco− (5′-AAAACCCATGGACTGGTACGAAAAGTAG-3′). A cDNA fragment of RCNMV RNA1 from nt 113 to 3890 was amplified by PCR from pRC1|G using primers AC5′UTR/Nco+ (5′-CGTACCAGTCCATGGGTTTTATAAATC-3′) and 3′R/C1 (5′-TACCCGGGGTACCTAGCCGTTATAC-3′). Both DNA fragments were denatured and annealed, and the resulting mixture was used as a template for PCR. The newly synthesized DNA duplex was amplified by PCR using primers 5′RS1T7 and 3′R/C1. The amplified 3.9-kb DNA fragment was digested with SacI and SmaI and cloned into the corresponding enzyme sites of pUC118 (Takara Bio Inc., Otsu, Japan) to create pURC1/Nco. Plasmids pR1-5′-NS, pR1-5′-XhS, pR1-5′-MS, and pR1-5′-SS were constructed as follows. pURC1/Nco was digested with NcoI/SphI (to construct pR1-5′-NS), XhoI/SphI (pR1-5′-XhS), MluI/SphI (pR1-5′-MS), and SacII/SphI (pR1-5′-SS) and blunt ended with T4 DNA polymerase. Small DNA fragments from each reaction were purified by 1% low-melting-point agarose gel (FMC Bioproducts, Rockland, Maine) electrophoresis and cloned into pSP-luc+ (Promega, Madison, Wis.), which had been digested with XbaI and HpaI, followed by treatment with T4 DNA polymerase. Small SacI/NcoI DNA fragments of pURC1/Nco were cloned into the NdeI/NcoI sites of each plasmid. All DNA fragments with incompatible ends used here were ligated after blunting with T4 DNA polymerase. Plasmid pR1-XbS was constructed as follows. A cDNA fragment of RCNMV RNA1 from nt 3446 to 3890 was amplified by PCR from pURC1/Nco using primers Aus1-3′/Xba (5′-GCTCTAGATTGGTTCTTTTAAGTGTAGCC-3′) and M4 (5′-GTTTTCCCAGTCACGAC-3′), and a 0.5-kb XbaI/SphI DNA fragment was cloned into the XbaI/HpaI sites of pSP-luc+. Plasmids pR1-5′-XbS and pR1-5′ were constructed by cloning the SacI/NcoI DNA fragment of pURC1/Nco into the NdeI/NcoI sites of pR1-XbS and pSP-luc+, respectively. To construct pR1-5′-XbSdSB, pR1-5′-XbS was digested with SacII and BstPI, blunt ended with T4 DNA polymerase, and religated. Plasmids pID1, pID2, p3′D1, p3′D2, p3′D3, and p3′D4 were constructed by ligation of SacII/SmaI-digested pR1-5′-XbS with a similarly digested PCR fragment that had been generated from pR1-5′-XbS with primers ID1 (5′-TCCCCGCGGAACCGGCATCGGACCCTGGG-3′) plus 3′R/C1 (pID1), ID2 (5′-TCCCCGCGGCTAGCGTATTAATAGGTCGC-3′) plus 3′R/C1 (pID2), Aus1-3′/Xba plus 3′D1 (5′-CCCCCGGGAATATTCTCCTGTTTTATTAAC-3′) (p3′D1), Aus1-3′/Xba plus 3′D2 (5′-TCCCCCGGGTTTTCTACTCTCAGGGGCC-3′) (p3′D2), Aus1-3′/Xba plus 3′D3 (5′-TCCCCCGGGAAAAAGGAAATAACTACAAC-3′) (p3′D3), and Aus1-3′/Xba plus 3′D4 (5′-TCCCCCGGGTCTTCCGACAACGACGTGCGC-3′) (p3′D4), respectively. Plasmid p3′TE-DR1/Lm1 was constructed as follows. Two DNA fragments were amplified by PCR from pR1-5′-XbS using two primer sets, Luc+3′FW (5′-TGTGGACGAAGTACCGAAAGGTCTTACCGG-3′) plus Aus1/3′TE-Lm1− (5′-GGTACCTGGGAAACAGGGTCCGATGCCGG-3′) and Aus1/3′TE-Lm1+ (5′-GTTTCCCAGGTACCTAGCGTATTAATAGG-3′) plus Luc+3′RV (5′-GCAGCCGAACGACCGAGCGCAGCGAGTCAG-3′). Both DNA fragments were denatured and annealed, and the resulting mixture was used as a template for PCR. The newly synthesized DNA duplex was amplified by PCR using primers Luc+3′FW and Luc+3′RV. The amplified 0.7-kb DNA fragment was digested with XbaI and SmaI and cloned into the corresponding enzyme sites of pR1-5′-XbS. Plasmid pRC1mL1 was constructed by cloning the SacII/SmaI DNA fragment of p3′TE-DR1/Lm1 into the corresponding sites of pRC1|G. Plasmid pLUCpA30 was constructed by exchanging the Renilla Luc (R-Luc) gene of pSP64-RLUC with a firefly Luc gene. Plasmid pLUCpA60 was generated by ligating the HindIII/SacI DNA fragment of pLUCpA30 into HindIII/AccI-digested p3′-8, which contains a 60-nt poly(A) tract (11). All constructs were verified by sequencing with an ABI 310 automated sequencer.

RNA preparation.RNA transcripts were synthesized in vitro from the linearized vector with T7 or SP6 RNA polymerase in the absence or presence of the cap structure analog (m⁷GpppG; New England Biolabs, Beverly, Mass.) and purified with a Sephadex G-50 fine column (Amersham Pharmacia Biotech, Buckinghamshire, United Kingdom). The RNA concentration was determined spectrophotometrically, and its integrity was verified by 1% agarose gel electrophoresis. RCNMV RNA1 and RNA2 were transcribed from SmaI-linearized pRC1|G and pRC2|G, respectively, with T7 RNA polymerase. RC1mL1 was transcribed from SmaI-linearized pRC1mL1 with T7 RNA polymerase. The genomic RNA of a tobamovirus, tobacco mild green mosaic virus (TMGMV), was transcribed from SalI-linearized pTGKJW (35) with T7 RNA polymerase. Control mRNAs, R-Luc mRNA, and firefly Luc mRNA were transcribed from EcoRI-linearized pSP64-RLUC and pLUCpA30, respectively, with SP6 RNA polymerase. Luc mRNA which contains a 60-nt poly(A) tract was transcribed from EcoRI-linearized pLUCpA60 with T7 RNA polymerase. Several Luc mRNAs which contained RCNMV-derived sequences were transcribed as follows. R1-5′-NS, R1-5′-XhS, R1-5′-MS, R1-5′-XbS, R1-5′-SS, R1-5′-XbSdSB, ID1, ID2, 3′D1, 3′D2, 3′D3, 3′D4, and 3′TE-DR1/Lm1 (named for their parent plasmids minus the “p” prefix) were transcribed from SmaI-linearized plasmids with T7 RNA polymerase. R1-5′ was transcribed from EcoRI-linearized pR1-5′ with T7 RNA polymerase, R1-XbS was transcribed from SmaI-linearized pR1-XbS with SP6 RNA polymerase, and R1-5′-XbB was transcribed from BstPI-linearized pR1-5′-XbS with T7 RNA polymerase.

Protoplast transfection.Cowpea plants (Vigna unguiculata cv. California Blackeye) were grown as previously described (6). Cowpea protoplasts were prepared from 12- to 16-day-old plants essentially as previously described (17), except that 0.5 M mannitol was used as an osmotic stabilizer in all solutions to which the protoplasts were exposed. Transfection and incubation of protoplasts were carried out as previously described (25).

Dual-Luc assay.Protoplasts (3.0 ×10⁵ cells) transfected with 2.0 pmol of Luc mRNA and 0.7 pmol of R-Luc mRNA were incubated at 17°C for 6 h. Cells were lysed in 500 μl of passive lysis buffer (Promega) and subjected to one or two freeze-thaw cycles to accomplish complete lysis of the cells. Aliquots (20 μl) of cell lysate were assayed with the Dual-Luciferase reporter assay system (Promega). The luminescence of firefly Luc was normalized with luminescence of R-Luc. Each experiment was repeated at least three times with different batches of protoplasts.

Northern blot analysis.Protoplasts (3.0 × 10⁵ cells) inoculated with 1.5 μg of transcripts were incubated at 17°C. Total RNAs extracted from protoplasts were subjected to Northern blot analysis as previously described (4). The RNA probe specific to the RCNMV 3′ UTR was also previously described (34). The digoxigenin (DIG)-labeled RNA probe for plus strand-specific detection of TMGMV (35) was transcribed with T7 RNA polymerase according to the manufacturer's protocols (Roche Molecular Biochemicals, Indianapolis, Ind.). The DIG-labeled RNA probe specific to the Luc gene was transcribed from NcoI-linearized pSP-Luc+ (Promega) with T7 RNA polymerase. The RNA signals were detected with a luminescence image analyzer (LAS 1000 Plus; Fuji Photo Film).

Detection of 5′ capped RNA.Virions of RCNMV and TMGMV were purified as previously described (37). Virions of Broad bean mottle virus (BBMV) were isolated from N. benthamiana plants 10 days postinoculation by the method of Lane (26) with a minor change. Infected-leaf homogenate clarified with chloroform was subjected directly to two cycles of differential centrifugation (low-speed centrifugation of 5,000 × g for 20 min and high-speed centrifugation at 100,000 × g for 90 min). The final virus pellet was suspended in a storage buffer (50 mM sodium acetate, 10 mM acetic acid, 1 mM Na₂EDTA, 1 mM MgCl₂). Virion RNAs of RCNMV, TMGMV, and BBMV were extracted from purified virus (25). Virion RNAs were fractionated on 1.5% agarose gel, transferred to a Hybond N+ membrane (Amersham Pharmacia Biotech), and then cross-linked on the blots by incubation at 80°C for 2 h. After blotting, membranes were immersed in a blocking solution (Roche Molecular Biochemicals) for 60 min and incubated with a 1/5,000-diluted monoclonal antibody against a 2,2,7-trimethyl guanosine-containing cap structure (Synaptic Systems, Göttingen, Germany) for 60 min. After being washed, membranes were reacted with an alkaline phosphatase-conjugated anti-immunoglobulin secondary antibody, followed by visualization with CDP-Star (Roche Molecular Biochemicals). Chemiluminescence was detected with a luminescence image analyzer (LAS 1000 Plus). Total RNAs blotted onto a membrane were detected with methylene blue staining (14). 5′ rapid amplification of cDNA ends (RACE) analysis was performed as previously described (34).

Virion RNAs of RCNMV lack a 5′ cap structure.It was previously reported that genomic RNAs of RCNMV, both RNA1 and RNA2, had an m⁷GpppA cap structure at the 5′ ends (29, 49). However, the fact that uncapped in vitro transcripts of RCNMV RNA1 and RNA2 still showed infectivity comparable to that of capped transcripts in N. benthamiana (50) and in cowpea protoplasts (Fig. 1) indicated that the 5′ cap structure was not needed for viral RNA replication in RCNMV. In contrast, the 5′ cap structure is necessary for the replication of TMGMV, a member of the genus Tobamovirus, which has a capped genome (Fig. 1). These results led us to reexamine whether virion RNAs of RCNMV were capped.

• Open in new tab
• Download powerpoint

FIG. 1.

Amplification of capped and uncapped transcripts of RCNMV RNA1 and TMGMV genomic RNA in cowpea protoplasts, which were incubated at 17°C. Total RNA extracted from protoplasts at 3 and 24 h postinoculation was separated by gel electrophoresis and blotted onto a membrane, which was probed with DIG-labeled RNA specific to RCNMV RNA1 or TMGMV genomic RNA.

To address this question directly, we used monoclonal antibodies that specifically reacted with the m⁷G cap (3). RCNMV virion RNAs were fractionated on a denaturing gel, transferred to a membrane, and reacted with the cap structure-specific antibody (Fig. 2A) and stained with methylene blue to detect total RNAs (Fig. 2B). As controls, virion RNAs of TMGMV and BBMV were also fractionated on the same gel. Genomic RNA of TMGMV reacted positively to the cap structure-specific antibody (Fig. 2A). BBMV RNA1, -2, -3, and -4, which have the m⁷GpppA cap structure at each 5′ end (5, 41), also reacted positively to the cap structure-specific antibody (Fig. 2A). However, the antibody did not react with virion RNA1 and -2 of RCNMV (Fig. 2A). This indicated that a detectable 5′ m⁷GpppA cap structure was not present in RCNMV virion RNAs. The lack of 5′ cap structure in RCNMV RNAs was further confirmed by sequencing the products obtained by 5′ RACE with RCNMV virion RNAs in which cytidine residues complementary to the m⁷G in the cap structure, if present, were not incorporated into the viral RNA 5′ ends of the RACE products. For Brome mosaic virus, Striped jack nervous necrosis virus, and TMGMV, which have capped 5′ ends, a cytidine residue complementary to the m⁷G in the cap structure was incorporated into the 5′ RACE products (20; H. Hamada and T. Okuno, unpublished results). From these results, we conclude that RNA1 and RNA2 of RCNMV packaged into virus particles do not have detectable 5′ cap structures.

• Open in new tab
• Download powerpoint

FIG. 2.

Immunodetection of capped RNA with a cap-specific antibody. (A) Virion RNAs of TMGMV, BBMV, and RCNMV were fractionated on a denatured agarose gel, transferred onto a Hybond N+ membrane, and reacted with a cap-specific antibody. The concentration of each virion RNA was adjusted to load the same molar amount (1.5 pmol) of TMGMV genomic RNA, BBMV RNA3, and RCNMV RNA1. Identities of the RNAs are indicated on the left. (B) Total virion RNAs blotted onto the membrane used for immunoassay were detected with methylene blue staining.

The 3′ UTR of RCNMV RNA1 enhances cap-independent translation.It was expected that genomic RNAs of RCNMV would contain an element(s) responsible for cap-independent translation. To identify the sequence element(s) of RCNMV RNA1 involved in cap-independent translation, translational activity was quantitatively evaluated with the Dual-Luciferase reporter assay system (Promega). Plasmids with the firefly Luc gene placed between the complete 5′ UTR of RNA1 and a series of other regions of RNA1, including the 3′ UTR, were constructed (Fig. 3). Equimolar capped and uncapped in vitro transcripts from these plasmids were transfected into cowpea protoplasts together with capped and polyadenylated R-Luc mRNA. The luciferase activity was measured after 6 h of incubation at 17°C, an optimum temperature for RCNMV replication (34). The luminescence of R-Luc was used as an internal control.

• Open in new tab
• Download powerpoint

FIG. 3.

Mapping of the RCNMV RNA1 sequences involved in cap-independent translation in vivo. Luc mRNAs containing various parts of RCNMV RNA1 diagrammed to the left were cotransfected with R-Luc mRNA (internal control) into cowpea protoplasts. The genome organization and restriction enzyme sites of RCNMV RNA1 are shown at the top. Asterisks, artificially introduced restriction enzyme sites; shaded boxes, ORFs for p27, p88, and CP; boldface black line, UTR. Maps below the genome depict transcripts coding for firefly Luc (open box) containing the viral sequence (boldface lines), vector sequence (thin lines), and poly(A) tails (A₃₀) in their UTRs. Bent lines, deleted regions. Firefly Luc activity from capped transcripts (white bars) and uncapped transcripts (black bars) for each construct are presented as firefly Luc/R-Luc ratios. Error bars, standard errors. Assays were performed at least three times.

Translational activity of an uncapped control Luc mRNA containing only vector-derived sequences and a poly(A) tail of 30 nt was less than 0.6% of that of a corresponding capped control Luc mRNA (Fig. 3, pA). Extending a poly(A) tail to 60 nt in the control Luc mRNA resulted in 1.3-fold increase in cap-dependent translational activity compared to that of capped Luc mRNA with a poly(A) of 30 nt but did not increase translational activity of uncapped Luc mRNA (data not shown). Capped R1-5′-NS, in which the Luc gene was inserted between the 5′ UTR and p27 ORF of RCNMV RNA1, produced a fourfold increase in activity compared to capped control Luc mRNA. In contrast, uncapped R1-5′-NS gave an over-200-fold increase in activity compared to uncapped control Luc mRNA (Fig. 3). These results indicate that RCNMV RNA1 has nucleotide sequences that enable uncapped mRNAs to be translated.

To define the sequence(s) of RNA1 required for cap-independent translation, Luc mRNAs having various regions of RNA1 sequences in their 3′ UTRs were tested for translational activity. All Luc mRNAs having the 3′ UTR of RNA1 were efficiently translated in the absence of a cap structure (Fig. 3). Interestingly, R1-5′-MS and R1-5′-XbS with large deletions from NcoI₁₂₃ to MluI₃₃₉₁ or to XbaI₃₄₄₆ yielded enhanced translational activity in both cap-independent and cap-dependent manners compared to R1-5′-NS or R1-5′-XhS (Fig. 3). Spacing between the initiation codon of the Luc gene and the 3′ UTR of RCNMV RNA1 might affect cap-independent translational activity. Alternatively, nucleotide sequences of RNA1 between the XhoI₁₉₃₃ and MluI₃₃₉₁ sites might negatively affect translational efficiency. It is also possible that deletions of the viral sequences increase RNA stability.

The 5′ UTR of RNA1 alone did not confer translational activity to Luc mRNA with no poly(A) tail either in the presence or absence of the 5′ cap structure (Fig. 3, R1-5′). In contrast, the 3′ UTR of RNA1 alone significantly enhanced cap-independent translational activity (Fig. 3, R1-XbS), although the activity was approximately 60% that of R1-5′-XbS, which has both precise 3′ and 5′ UTRs of RNA1. These results indicate that the 3′ UTR of RCNMV RNA1 is a primary determinant for cap-independent translation.

Nucleotide sequences in the 3′ UTR of RCNMV RNA1 required for cap-independent translation.To delimit regions in the 3′ UTR of RNA1 required for cap-independent translation, we evaluated the translational activity of Luc mRNAs which contain the precise viral 5′ UTR and the Luc gene followed by viral 3′ UTRs with different regions deleted. Deletion from the XbaI site to the SacII site (nt 3446 to 3568) in the viral 3′ UTR (R1-5′-SS) reduced translational activity of uncapped mRNA to 10% of that of uncapped R1-5′-XbS (Fig. 4A). Although deletion between nt 3571 and nt 3595 (ID1) also greatly reduced translational activity of uncapped mRNA to 2.9% of that of uncapped R1-5′-XbS, the activity was still 50-fold higher than that of uncapped control Luc mRNA (Fig. 4A). An additional deletion, extending to nt 3625 (ID2) or to nt 3672 (R1-5′-XbSdSB) abolished cap-independent translational activity, whereas capped constructs of these mRNAs showed 40 to 70% of the translational activity of capped R1-5′-XbS (Fig. 4A).

• Open in new tab
• Download powerpoint

FIG. 4.

(A) Deletion mapping of the 3′ UTR sequence of RCNMV RNA1 involved in cap-independent translation in vivo. Transcripts containing the precise viral 5′ UTR and the Luc gene followed by various parts of the 3′ UTR are diagrammed on the left. Restriction enzyme sites used for deletion are shown at the top. Boldface lines and thin lines, virus-derived and vector-derived sequences, respectively; bent lines, deleted regions. The 3′TE-DR1 (nt 3596 to 3732) is shaded. Firefly Luc activity from capped transcripts (white bars) and uncapped transcripts (black bars) for each construct are expressed as percentages of uncapped R1-5′-XbS activity; error bars, standard errors. Assays were performed at least three times. (B) Stability of RNAs in cowpea protoplasts transfected with uncapped R1-5′-XbS, ID1, or 3′D4. Total RNA extracted from protoplasts at the indicated times after transfection was separated by gel electrophoresis, blotted onto a membrane, and probed with DIG-labeled RNA specific to Luc mRNA.

Deletion of the 3′-proximal stem-loop structure (nt 3862 to 3887), which is almost identical between RNA1 and RNA2 (29, 49), did not significantly reduce cap-independent translational activity (Fig. 4A, 3′D1). Further deletions extending to nt 3835 (3′D2) and 3761 (3′D3) reduced cap-independent translational activity, but the activities relative to those of corresponding capped mRNA remained high (Fig. 4A). Although deletion to nt 3733 (3′D4) greatly reduced translational activity of uncapped mRNA to 1.9% of that of uncapped R1-5′-XbS, the activity was still 30-fold higher than that of uncapped control Luc mRNA (Fig. 4A). Deletion to the BstPI site from the 3′ end (nt 3673 to 3890) abolished cap-independent activity (Fig. 4A, R1-5′-XbB). It was possible that the deletions introduced into 3′ UTRs altered mRNA stability and affected cap-independent translational activity. To test this, uncapped RNA transcripts of R1-5′-XbS, ID1, and 3′D4 were transfected into cowpea protoplasts under the same conditions as those for the Luc assay. RNA degradation was monitored by Northern blot hybridization using an RNA probe that specifically hybridized with Luc mRNA. Results showed that accumulation patterns of the mRNAs with various deletions did not significantly differ from each other (Fig. 4B). Similar results were obtained with uncapped ID2 and R1-5′-XbB (data not shown). These results suggest that differences in translational activity of uncapped mRNA were not caused by differences in stability of mRNAs. Taken together, results obtained here suggest that sequences necessary for highly efficient cap-independent translation of RNA1 in vivo lie between nt 3446 and 3760 and that sequences between nt 3596 and 3732 are indispensable for cap-independent translation. Hereafter, we refer to this region (nt 3596 to 3732; highlighted in Fig. 4A) as a 3′ translation element of Dianthovirus RNA1 (3′TE-DR1).

A stem-loop structure conserved among several plant RNA virus genomes is responsible for cap-independent translation of RCNMV RNA1.To identify functional domains in 3′TE-DR1 required for cap-independent translation, we predicted its secondary structure by using the program Dynalign (30). The 3′TE-DR1 is predicted to have five stem-loop structures (SL1, SL2, SL3, SL4, and SL5) and base pairing (stem) between the ends of the region (Fig. 5A). SL1 is conserved in all dianthoviruses, RCNMV (34, 49), Sweet clover necrotic mosaic virus (10), and Carnation ringspot virus (42). Interestingly, as shown in Fig. 5B, the SL1 of RCNMV RNA1 is almost identical to one of three stem-loop structures predicted for the 3′ UTR of BYDV genomic RNA (11). The BYDV region, including the three stem-loop structures (TE105), facilitates cap-independent translation in vitro (11, 47). The loop sequence of SL-I, GGAAA, fits the consensus of a GNR(N)A pentaloop, a structure that forms a GNRA tetraloop fold (16). In BYDV, changing nucleotide sequences from GGAAA to UUUCC in the loop corresponding to RCNMV SL1 eliminates cap-independent translational activity in vitro (11). These previous results led us to test the effects of the same mutation in the SL1 of 3′TE-DR1 on translational activity. As shown in Fig. 5C, 3′TE-DR1/Lm1 possessing this mutation in R1-5′-XbS was not translated in the absence of the 5′ cap structure, although it was efficiently translated in the presence of the 5′ cap structure.

• Open in new tab
• Download powerpoint

FIG. 5.

(A) Secondary structure predicted by the computer algorithm Dynalign (30) for the 3′TE-DR1 of RCNMV (Australian strain) determined in concert with the 3′ UTR of the other RCNMV strain (RCNMV-Can) RNA1 (34). The calculation was performed with a maximum distance of 25, a gap penalty of 0.4 kcal mol⁻¹, and no single-base-pair inserts allowed. Numbering indicates nucleotide positions in RCNMV RNA1. (B) Secondary structure of SL1 of RCNMV RNA1 (left) and SL-I of BYDV (right). Boldface, nucleotide differences between RCNMV and BYDV. (C) Translational activity of the 3′TE-DR1 mutant (3′TE-DR1/Lm1) in vivo. For others, refer to Fig. 4A. (D) Stability of RNAs in cowpea protoplasts transfected with uncapped R1-5′-XbS or the SL1 mutant (3′TE-DR1/Lm1). For others, refer to Fig. 4B.

It is possible that 3′TE-DR1, including SL1, stimulates gene expression by increasing RNA stability. The RNA degradation pattern of uncapped RNA transcripts of R1-5′-XbS and 3′TE-DR1/Lm1 was monitored by Northern blot hybridization. The RNA transcripts with either the wild-type or the mutant SL1 did not differ significantly in degradation rate (Fig. 5D). These results and the results shown in Fig. 4 suggest that 3′TE-DR1 and sequences around 3′TE-DR1 are responsible for cap-independent translation.

The 3′TE-DR1 is required for replication of uncapped full-length RCNMV RNA1.To investigate whether the cap-independent translational nature observed in Luc mRNAs having 3′ UTR sequences of RCNMV RNA1 is related to the replication of RCNMV RNA1, we introduced the SL1 mutation into wild-type full-length RCNMV RNA1. Cowpea protoplasts were inoculated with the mutant RNA1 (RC1mL1), and wild-type RNA1 transcripts were synthesized with or without cap analogs. Total RNA was extracted from protoplasts at 3 and 24 h postinoculation and analyzed by the Northern blotting method using an RCNMV RNA1-specific probe. Uncapped transcripts of wild-type RNA1 were amplified to high levels, whereas RC1mL1 was not amplified in the inoculation of uncapped RC1mL1 transcripts (Fig. 6). Interestingly, inoculation with capped RC1mL1 transcripts did not cause viral RNA amplification, even though the SL1 mutation did not strongly affect translation efficiency in capped R1-5′-XbS (Fig. 5D). It is possible that the progeny RNAs from RC1mL1 were not translated in the absence of the 5′ cap and that the amount of viral replicases translated from the capped inoculum of RC1mL1 might be insufficient to produce progeny virus RNAs to the level detectable by Northern blot hybridization. Another possibility is that the mutation in SL1 knocked out a cis-acting sequence required for RNA replication as well as for cap-independent translation.

• Open in new tab
• Download powerpoint

FIG. 6.

Replication of RCNMV RNA1 and the SL1 mutant (RC1mL1) in cowpea protoplasts, which were incubated at 17°C. Total RNA extracted from protoplasts at 3 and 24 h postinoculation was separated by gel electrophoresis, blotted onto a membrane, and probed with DIG-labeled RNA specific to RCNMV RNA1.