Viral Class 1 RNase III Involved in Suppression of
http://www.100md.com
病菌学杂志 2005年第11期
Department of Plant Biology and Forest Genetics, SLU, Box 7080, SE-750 07 Uppsala, Sweden
Department of Applied Biology, PO Box 27, FIN-00014 University of Helsinki, Finland
ABSTRACT
Double-stranded RNA (dsRNA)-specific endonucleases belonging to RNase III classes 3 and 2 process dsRNA precursors to small interfering RNA (siRNA) or microRNA, respectively, thereby initiating and amplifying RNA silencing-based antiviral defense and gene regulation in eukaryotic cells. However, we now provide evidence that a class 1 RNase III is involved in suppression of RNA silencing. The single-stranded RNA genome of sweet potato chlorotic stunt virus (SPCSV) encodes an RNase III (RNase3) homologous to putative class 1 RNase IIIs of unknown function in rice and Arabidopsis. We show that RNase3 has dsRNA-specific endonuclease activity that enhances the RNA-silencing suppression activity of another protein (p22) encoded by SPCSV. RNase3 and p22 coexpression reduced siRNA accumulation more efficiently than p22 alone in Nicotiana benthamiana leaves expressing a strong silencing inducer (i.e., dsRNA). RNase3 did not cause intracellular silencing suppression or reduce accumulation of siRNA in the absence of p22 or enhance silencing suppression activity of a protein encoded by a heterologous virus. No other known RNA virus encodes an RNase III or uses two independent proteins cooperatively for RNA silencing suppression.
INTRODUCTION
RNA silencing is a eukaryotic cellular surveillance mechanism (12) that defends against viruses (1, 20, 41), controls transposable elements (55), and participates in the formation of silent chromatin (62). RNA silencing is also involved in posttranscriptional regulation of gene expression during developmental processes (26, 37). Double-stranded RNA (dsRNA) is the most potent trigger for RNA silencing (17, 24). The inducing dsRNA is processed into two classes of small interfering RNAs (siRNAs) of ca. 21 to 24 nucleotides (nt) in plants (21, 22) by the RNase III (RNase III) class 3-like enzyme Dicer (3). The siRNA incorporates into a nuclease-containing RNA-induced silencing complex (RISC), and the antisense strand guides it to homologous RNAs that are subsequently degraded (23, 57). The amount of siRNA can be amplified because of transivity, i.e., the production of additional siRNAs from sequences located outside the original target region. Transivity occurs in a 3'-to-5' direction in Caenorhabditis elegans but is bidirectional in plants (25, 51). RNA silencing in plants involves an as-yet-unknown signal that moves out from the cells undergoing RISC-mediated RNA degradation, spreading the silencing effect over short distances (25), as well as to other parts of the plant (43, 60).
The host cell RNA-silencing mechanism targets replicating viruses. Viruses, in turn, must suppress RNA silencing to ensure successful invasion. The first RNA silencing-suppressing (RSS) proteins were found in plant viruses (2, 7, 31) and have since been identified in many unrelated plant (40) and animal (35) viruses. RSS proteins seem to be versatile in their function, since they can interfere with RNA silencing in plant, animal, and human cells (16, 35), consistent with the conserved nature of the RNA silencing mechanism in eukaryotes in general (12, 24). Different RSS proteins can interfere with different steps in the RNA-silencing pathway. For example, RSS proteins may prevent the initiation of silencing (7), the spread of the silencing signal (61), or the maintenance of the silenced stage (39).
Sweet potato chlorotic stunt virus (SPCSV; genus Crinivirus, family Closteroviridae) has a large bipartite (RNA, 9,407 nucleotides; RNA2, 8,223 nt), single-stranded, positive-sense RNA genome (32). It is an agriculturally important pathogen of sweet potato (Ipomoea batatas L.), especially because it breaks down resistance to unrelated viruses during a dual infection, significantly increasing disease severity (18). The synergistic effects of SPCSV on other viruses may be due to interference with RNA silencing, because they are associated with substantially increased accumulation of coinfecting viruses (30). SPCSV encodes proteins that are not found in other RNA viruses. For example, RNA1 contains an open reading frame (ORF) for a putative RNase III (RNase3) and also a 22-kDa protein (p22) that show no significant similarity to known proteins from any organism. The subgenomic RNAs (sgRNAs) encoding these proteins are expressed early in infection (32).
The detection of novel genes in the SPCSV genome prompted us to study whether any of the genes exhibit RSS functions, which might help to elucidate the mechanism by which SPCSV synergizes with other viruses in sweet potato plants. Our data reveal that p22 efficiently suppresses RNA silencing, which is further enhanced by the dsRNA-specific endonuclease activity of RNase3. These data provide the first evidence that RNase III enzymes are involved not only in RNA silencing but also in its suppression.
MATERIALS AND METHODS
Molecular cloning. Sequences of the primers used in this study will be provided upon request. Plasmids pAHCPro (50) and p35S-GUS-INT (58) have been described previously. They are based on the binary plant transformation vector pKOH200 and contain the helper component proteinase (HC-Pro) encoding sequence of Potato virus A (PVA; genus Potyvirus) and the the ?-galacturonidase (GUS) gene, respectively, flanked by the Cauliflower mosaic virus (CaMV; genus Caulimovirus) 35S promoter and the 3' terminator region of the nopaline synthase gene (3'nos). The GUS gene contains a plant intron to prevent GUS expression in Agrobacterium (58). The "cycle 3" gfp gene was amplified from pGFPuv (Clontech) by PCR, and the ORFs for RNase3, p22 (RNA1), and p28 (RNA2) were amplified by reverse transcription-PCR (RT-PCR) from purified viral SPCSV RNA. To obtain the constructs for agroinfiltration, the HC-Pro fragment in pAHCPRo was replaced by the aforementioned PCR products, resulting in constructs pAGFP, pARNase3, pAp22, and pAp28. The NotI-BamHI fragment of pAGFP, encompassing the N-terminal two-thirds (526 bp) of the gfp gene, was cloned into pKOH122 in the sense orientation between the 35S promoter and a BclI/XbaI-digested fragment of the IV2 intron (58). The resulting plasmid was subsequently digested with BamHI and SbfI, and the NotI/BamHI-digested fragment of gfp was cloned into this digested plasmid in the antisense orientation, together with the NotI/SbfI-digested 3' nos terminator fragment, to yield construct hpGFP.
PVX-p22, PVX-RNase3, PVX-p28, and PVX-HCpro vectors were constructed by cloning the corresponding genes into potato virus X (PVX), as described previously (65). pPVX201 contains the complete infectious cDNA of Potato virus X (genus Potexvirus) under the control of the CaMV 35S promoter and a duplicated copy of the coat protein gene sgRNA promoter from the plasmid pTXS.P3C2 (6). The SPCSV genes encoding p22, p28, and RNase3 proteins or the PVA HC-Pro-encoding region were amplified by PCR and inserted between the NheI and SalI restriction sites of pPVX201. Mutants of the p22 gene were generated based on pAp22 or PVX-p22 with the QuickChange XL Site-Directed Mutagenesis kit (Stratagene) according to the manufacturer's instructions.
For RNase3 protein expression in Escherichia coli, the ORF for RNase3 was amplified by RT-PCR from a viral RNA template using oligonucleotide primers that introduced the codons for a C-terminal hexahistidine (six-His) tag and the NcoI restriction site required for cloning. The resulting PCR product was digested with NcoI and cloned into the corresponding site of the dephosphorylated pET11d (Stratagene) vector for protein expression.
The predicted endonuclease domain (endoND; amino acids [aa] 1 to 152) and dsRNA-binding domain (dsRBD; aa 153 to 228) of RNase3 were cloned into the binary plant transformation vector pKOH200 as previously described (50). An intermediate construct, pKOH122-RNase3, was obtained by cloning the RNase3 gene into plasmid pKOH122 using the NotI and FseI restriction sites, flanked by the 35S promoter and the 3'nos terminator. The RNase3 mutant (RNase3-Ala37,44), in which Asp37 and Asp44 were substituted for Ala, was obtained by replacing the sequence in the pKOH122-RNase3 construct between NotI and FseI restriction sites with two equivalent PCR fragments into which mutations and a unique NarI restriction site had been introduced by PCR and which were digested with NotI/NarI and with NarI/FseI prior to ligation into the NotI/FseI-digested pKOH122-RNase3. The four point mutations within SPCSV RNase3 ORF resulted in the amino acid substitutions D37A and D44A and introduced a unique NarI restriction site. The whole cassette (35S-Rnase3-Ala37,44-3'nos) was then excised from the pKOH122 plasmid and inserted between the Sse8387I and PacI restriction sites of pKOH200. The resulting plasmid, pKOH-RNase3-Ala37,44, was subsequently used as a template for two-step PCR with nested primers. The resulting PCR product was digested with NcoI and cloned into the corresponding site of dephosphorylated pET11d. Upon expression, the resulting plasmid yielded a recombinant mutant RNase3 protein with a six-His tag at the C terminus and two amino acid substitutions, D37A and D44A, in the RNase III signature motif.
The fidelity of all constructs was confirmed by sequencing.
Protein expression and analysis. The six-His fusion proteins (RNase3 and RNase-Ala37,44) were expressed in E. coli BL21-RIL cells (Stratagene) according to the manufacturer's protocol. We purified recombinant RNase3 under native conditions by affinity chromatography using Ni-nitrilotriacetic acid agarose (QIAGEN) according to the manufacturer's protocol. Fractions containing the highest concentrations of recombinant RNase3 were pooled, dialyzed against enzyme storage buffer (30 mM Tris-HCl [pH 8.0], 500 mM KCl, 0.1 mM EDTA, 0.1 mM dithiothreitol, 50% glycerol [vol/vol]), and stored at –20°C.
For the production of polyclonal antibodies, the six-His RNase3 protein was dialyzed against buffered saline solution (0.9% NaCl, 10 mM Tris-HCl, pH 6.0). The protein (25 μg, in Freund's adjuvant) was injected into a rabbit; the immunization was carried out six times at 4-wk intervals at the National Veterinary Institute, Uppsala, Sweden.
RNase3 was detected by Western blot analysis with RNase3-specific polyclonal antibodies and anti-rabbit mouse monoclonal antibodies conjugated with horseradish peroxidase (Sigma) by standard procedures (48). Signals were detected by chemiluminescence by the ECL method (Amersham Pharmacia Biotech). In cross-protection assays, vector viruses were detected by double-antibody sandwich enzyme-linked immunosorbent assay (DAS-ELISA) as previously described (50) with anti-coat protein antibodies purchased from Boehringer and Adgen.
Nucleic acid substrates for enzymatic analysis. To study RNA cleavage in vitro, all RNAs were generated by transcription in vitro. The gfp gene was PCR amplified, digested with NcoI, and cloned into the corresponding dephosphorylated site in pET11d. Plasmids carrying gfp in the sense or antisense orientation were selected. Each plasmid was linearized by BamHI and transcribed by T7 RNA polymerase (Promega) according to the manufacturer's protocol. RNA transcripts were phenol-chloroform extracted, precipitated with ethanol, and dissolved in RNase-free water. Equimolar amounts of the in vitro-transcribed single-stranded RNAs (ssRNAs) were annealed to yield dsRNA substrates by being heated for 5 min at 95°C, followed by gradual cooling to room temperature in a water bath. Double-stranded DNA (dsDNA) substrate was prepared by linearizing pET11d-GFP with BamHI. The RNA-DNA hybrid substrate was generated by RT of the gfp RNA transcript with a green fluorescent protein (GFP)-specific primer. To obtain single-stranded DNA (ssDNA) substrate, the RNA-DNA hybrid substrate was treated with RNaseH (Promega).
The low-molecular-weight (LMW) RNA fraction containing siRNA was isolated from wild-type Nicotiana benthamiana plants coagroinfiltrated with GFP and hairpin GFP (hpGFP). Plant siRNAs were purified from a native 8% polyacrylamide gel as previously described (33). Synthetic 21-nt and 25-nt siRNAs (see below) visualized with ethidium bromide (EtBr) were used as size markers.
Synthetic gfp RNA oligonucleotides (11 nt, 21 nt, and 25 nt; sense and antisense) were purchased from Dharmacon. dsRNA oligonucleotides were obtained by annealing sense and antisense strands in equimolar amounts. The 21-nt and 25-nt dsRNAs were analogous to siRNA, as they contained 2-nt 3' overhangs and 3' hydroxyl groups.
The synthetic 21-nt and 25-nt RNAs (100 pmol) were phosphorylated in the presence of [-32P]ATP in a 50-μl reaction mixture using 30 U of T4 polynucleotide kinase (3'-Phosphatase Minus; Fermentas). Unincorporated nucleotides were removed with Micro Bio-Spin P-30 Tris chromatography columns (Bio-Rad). For electrophoretic mobility shift studies, various concentrations of recombinant six-His-tagged RNase3 or RNase3-Ala37,44 were mixed with radiolabeled RNA in the cleavage buffer and incubated at room temperature for 30 min. Complexes were resolved on a native 6% polyacrylamide gel that was subsequently exposed to Molecular Imager screens (Bio-Rad).
Cleavage assays for RNase III activity were performed with a cleavage buffer (30 mM Tris-HCl [pH 7.5], 10 mM MgCl2, 130 mM KCl, 5% glycerol [vol/vol]).
Agroinfiltration, virus inoculation, and detection of GFP. The transgenic N. benthamiana line 16c, which expresses GFP, has been previously described (47). Agroinfiltration was carried out as described by Johansen and Carrington (28) using Agrobacterium tumefaciens strain C58C1 (with Ti plasmid pGV3850 or pGV2260). For coinfiltration, the different Agrobacterium cultures were diluted with infiltration medium so that equal, final optical densities were achieved before combining the cultures in a 1:1 ratio for infiltration. If some coinfiltration treatments included fewer constructs than others, the missing volume was replaced by the Agrobacterium strain expressing GUS. Inoculation with PVX vectors and Tobacco mosaic virus (TMV)-GFP has been previously described (65).
GFP fluorescence was monitored by epi-illumination with a hand-held UV source (Blackray UVP 100 AP) as previously described (47). Photographs were taken with a digital camera, and images were prepared with Photoshop software (Adobe).
RNA analysis. Total RNA was extracted with TRIzol LS Reagent (Invitrogen, Ltd.) according to the manufacturer's recommendations. LMW and high-molecular-weight (HMW) RNA fractions were separated by precipitation of a HMW fraction with an equal volume of 4 M LiCl4 (final concentration, 2 M) at 4°C overnight. After centrifugation, the supernatant was transferred to a new tube to which an equal volume of isopropanol was added to precipitate LMW RNA. The quality of the extracted RNA was evaluated under UV light after electrophoresis in a standard formaldehyde gel and staining with ethidium bromide (48).
For Northern blot analysis of mRNA, 10 μg HMW RNA was separated by formaldehyde gel electrophoresis, transferred to a nylon membrane (Hybond-N; Amersham Biosciences AB), cross-linked with UV light, prehybridized, hybridized at 55°C, and washed in hybridization tubes (48).
To analyze siRNA, 30 μg LMW RNA was mixed with an equal volume of Tris-borate-EDTA--urea sample buffer (Bio-Rad), heated at 100°C for 5 min, and loaded onto a 15% polyacrylamide Tris-borate-EDTA-urea gel. The RNA was then transferred to a nylon membrane by capillary blotting (48), prehybridized, hybridized, and then washed as described previously (22).
Antisense [-32P]UTP-labeled gfp-specific RNA probes were synthesized with T3 RNA polymerase (Promega) from XbaI-digested pGFPuv according to the manufacturer's protocol. For hybridization to siRNAs, the probe was cleaved by alkaline hydrolysis to an average length of 50 bp, as described previously (22). After s final washing, the blots were wrapped in polyethylene plastic and exposed to an exposure cassette (Molecular Dynamics) for 1 to 48 h. The cassette was then scanned with a Molecular Imager FX, and the signal strength was quantified with Quantity One software (Bio-Rad).
RESULTS
RNase3 has enzymatic activities characteristic of RNase III. The deduced amino acid sequence of SPCSV RNase3 shows similarity to RNase III (32). RNase III belongs to a family of endoribonucleases that specifically recognize and cleave dsRNAs and are found in prokaryotes and eukaryotes (reviewed in references 13 and 49). Three classes of RNase III enzymes have been identified, based on the set of domains they possess. SPCSV RNase3 is similar to the class 1 enzymes (the simplest RNase III enzymes), which contain an endoND and a dsRBD. The sequence of endoND has a characteristic, highly conserved, 9-aa RNase III signature motif, in which three amino acid residues, corresponding to E40, G43, and D44 in SPCSV RNase3, are strictly conserved (5).
We overexpressed and purified the six-His-tagged SPCSV RNase3 and a mutant thereof in E. coli. The mutant, RNase3-Ala37,44, contained alanine instead of aspartic acid at each of two sites (D37 and D44) in the RNase III signature motif. Both proteins were stable and had electrophoretic mobilities corresponding to the expected molecular mass of 27 kDa (Fig. 1A and B). Both recombinant proteins were tested for cleavage of nucleic acid substrates derived from plasmids carrying the gfp gene under control of the T7 RNA polymerase promoter in the sense or antisense orientation. These substrates included ssRNA, dsRNA (Fig. 1C), ssDNA, dsDNA, and an RNA-DNA hybrid. The enzymes did not cleave ssRNA (Fig. 1D and E), ssDNA, dsDNA, or RNA-DNA hybrids (data not shown). However, RNase3 efficiently cleaved dsRNA at physiological concentrations of KCl (50 to 130 mM) (Fig. 1D). RNase3 activity on dsRNA required divalent cations such as Mg2+ (Fig. 1E) or Mn2+ (Fig. 1F), but Ca2+ did not support activity (Fig. 1G). Cleavage of a dsRNA substrate containing 69-nt and 31-nt ssRNA overhangs (Fig. 1C) was analyzed over time, revealing the accumulation of the two expected longer ssRNAs (cleavage products including the overhangs) and also shorter products (15 to 18 nt) (Fig. 1H) that probably represented unwound dsRNA cleavage products, consistent with earlier reports on exhaustive cleavage of long dsRNA by E. coli RNase III (42). Furthermore, as with E. coli RNase III (8), low concentrations of EtBr (125 μM) inhibited RNase3 cleavage of the dsRNA substrate (Fig. 1I). In all tests, RNase3 cleaved the hairpin RNA (hpRNA) substrate and the dsRNA substrate with similar efficiency (Fig. 1J). In contrast, none of the tested nucleic acid substrates were cleaved using RNase3-Ala37,44, although this mutant bound dsRNA at physiological salt concentrations (50 to 130 mM KCl) (Fig. 1K). Collectively, these data show that RNase3 exhibits the characteristic endonuclease activities of RNase III (13, 49).
p22 causes necrosis and interferes with cross-protection when expressed from a heterologous virus in Nicotiana benthamiana. The genes for p22, RNase3, and p28 of SPCSV were expressed from the 35S promoter-driven infectious cDNA of PVX (6, 65) by mechanical inoculation of the plasmids (PVX-p22, PVX-RNase3, and PVX-p28, respectively) into N. benthamiana plants. Following inoculation with PVX-p28 or PVX-RNase3, the plants developed similar symptoms and high PVX titers (measured by DAS-ELISA; data not shown) in upper noninoculated leaves, as observed with PVX-GF carrying a truncated gfp gene (half of the total gfp length) (65). In contrast, PVX-p22 and the PVX vector carrying the cistron for HC-Pro of PVA (genus Potyvirus) caused necrotic symptoms in systemically infected leaves at 7 days postinoculation (dpi), followed by death of the plant top (PVX-HCpro) or the whole plant (PVX-p22) (Fig. 2). Additionally, the leaves inoculated with PVX-p22 often developed a few necrotic lesions that expanded along the veins (Fig. 2A). Although reasons remain unknown, expression of RSS proteins from heterologous viral genomes such as PVX often results in necrotic symptoms in N. benthamiana (65).
Mutants of the p22 gene in PVX-p22 were produced by introducing a stop codon corresponding to amino acid positions 89, 150, 164, or 181 in p22 (191 aa), and the nucleotide following each new stop codon was deleted to produce a frameshift and ensure that the downstream part of the p22 sequence was not expressed. All mutated PVX-p22 constructs generated a systemic infection, PVX titers (determined by DAS-ELISA), and symptoms similar to those of PVX-GF in N. benthamiana, indicating the loss of necrotizing ability. We used reverse transcription PCR (RT-PCR) cloning, analysis of the amplification products by agarose gel electrophoresis, and sequencing of the p22 insert from the viruses in systemically infected leaves to confirm that lack of necrosis was not associated with loss or partial deletion of the inserts during virus replication (data not shown).
Initial testing for RSS by the SPCSV proteins p22, RNase3, and p28 was carried out using a previously described cross-protection assay in N. benthamiana (45). Leaves are first inoculated with PVX-GF and 3 days later with tobacco mosaic virus engineered with the gfp gene (65), which confines GFP fluorescence to coinoculated leaves (Fig. 2D), because replication of two viruses with partially homologous viral genomes in the same tissue triggers RNA silencing and prevents long-distance transport of the latter inoculated virus, TMV-GFP (45). Detection of a few GFP-expressing lesions on inoculated leaves is expected, since some initially infected cells may be infected only with TMV-GFP. Subsequently, we constructed derivatives of PVX-GF to express p22, RNase3, or p28 (PVX-p22-GF, PVX-RNase3-GF, and PVX-p28-GF, respectively) and coinoculated these constructs with TMV-GFP into N. benthamiana plants, as above, along with the p22 mutants described earlier. PVX-GF and a PVX chimera (PVX-HCpro-GF) expressing the HC-Pro of PVA were used as controls. In plants inoculated with PVX-RNase3-GF, PVX-p28-GF, any of the PVX-GF/p22-mutant constructs, or PVX-GF and subsequently (3 days later) with TMV-GFP, green fluorescence was confined to the inoculated leaves. In contrast, following inoculation with PVX-p22-GF (Fig. 2E and F) or PVX-HCpro-GF and subsequently TMV-GFP, green fluorescence spread to the stem and the upper noninoculated leaves at 6 to 8 dpi. The plants soon developed apical necrosis, however, as expected.
Three replicate experiments consistently showed that SPCSV p22 and PVA HC-Pro induce necrosis in N. benthamiana when expressed from a heterologous viral genome. These proteins interfered with cross-protection, which suggested that they suppress RNA silencing (7). These effects were not evident with SPCSV RNase3 and p28.
p22 suppresses silencing induced by dsRNA in N. benthamiana. An Agrobacterium tumefaciens infiltration assay (28) was used to test whether the SPCSV-encoded proteins p22, p28, and RNase3 could suppress intra- or intercellular RNA silencing triggered by a strong silencing inducer (i.e., dsRNA). In this assay, coexpression of gfp and an inverted-repeat construct of gfp (hpGFP) in agroinfiltrated leaves resulted in degradation of the GFP-specific RNA and suppression of GFP fluorescence, which in turn could be prevented by coexpression of RSS proteins to achieve strong green fluorescence (Fig. 3).
Coinfiltration for expression of GFP, hpGFP, and ?-glucuronidase (GUS) did not result in green fluorescence in the infiltrated areas (Fig. 3), and green fluorescence was not observed following coinfiltration with GFP, hpGFP, and RNase3 (Fig. 3) or p28 (data not shown). These data indicate that RNase3 and p28 do not exhibit RSS activity in this assay. In contrast, infiltration with p22 or HC-Pro resulted in strong fluorescence in the infiltrated leaf areas, commencing at 2 dpi and increasing until 5 dpi. The fluorescence achieved with HC-Pro was initially (5 dpi) (results not shown) stronger than with p22, but later the situation was reversed (10 dpi) (Fig. 3A, compare spot 2 and 4). The spots infiltrated for expression of p22 became chlorotic (10 dpi) (Fig. 3B, spot 2) and later necrotic (15 dpi) (Fig. 3D, spots 4 and 6). Other constructs did not induce chlorosis,; neither did the p22 mutants described earlier, which also did not suppress RNA silencing, as evidenced by the lack of green fluorescence in the coinfiltrated leaves (data not shown). Taken together, these data indicate that full-length p22 suppresses intracellular RNA silencing triggered by a strong RNA silencing inducer.
Cell-to-cell movement of the RNA silencing signal from the cells in which RNA silencing-based RNA degradation takes place can be detected in transgenic GFP-expressing N. benthamiana plants (e.g., line 16c) (47), following silencing of the gfp transgene by ectopic expression of gfp (25). As a result, the silencing signal exiting from the agroinfiltrated area shuts down GFP expression, which is observed as an appearance of a narrow, red border at the edge of the infiltrated spot. Our experiments with the N. benthamiana line 16c revealed short-distance movement of the silencing signal, following infiltration with GFP, hpGFP, and GUS (Fig. 4A) or RNase3 (Fig. 4D) but not HC-Pro (Fig. 4B) or p22 (Fig. 4C), indicating that HC-Pro and p22 interfere with short-distance intercellular signaling for silencing. Furthermore, strong green fluorescence was observed in the tissue infiltrated for expression of HC-Pro or p22, consistent with their ability to suppress RNA silencing (Fig. 4B and 4C).
In two additional experiments, the first fully developed leaves in line 16c plants were coinfiltrated with hpGFP and either p22, HC-Pro, p28, GUS, or RNase3 (five plants per experiment) to observe whether any of these proteins interfered with long-distance signaling for silencing of the gfp gene. Only plants infiltrated with p28, GUS, or RNase3 developed silencing in the upper noninfiltrated leaves (Fig. 4E) within the 2-month period the plants were monitored (data not shown), indicating that expression of either p22 or HC-Pro at the site of RNA silencing induction interfered with systemic silencing.
RNase3 enhances p22-mediated suppression of RNA silencing. Because the sgRNAs for RNase3 and p22 are expressed in a coordinated fashion early in SPCSV infection (32) and because RNase III-type proteins are involved in RNA silencing (3), we tested whether RNase3 influences specifically p22-mediated RSS or if it also affects the RSS functions of an RSS protein encoded by a heterologous virus (PVA HC-Pro). Expression of RNase3 was assayed by Western blotting with anti-RNase3 antibodies (Fig. 5B). Leaves were coinfiltrated with p22, RNase3, HC-Pro, or GUS along with GFP and hpGFP. Whole leaves rather than spots on leaves were coinfiltrated to ensure sufficient RNA yields for analysis of mRNA and siRNA accumulation. At 5 dpi, Northern blot analyses readily detected GFP mRNA in leaves coinfiltrated with GFP, hpGFP, and p22 or HC-Pro but not RNase3, GUS (Fig. 5A), or p28 (data not shown). Figure 5A illustrates that GFP mRNA was already expressed at 3 dpi in leaves coinfiltrated with p22 and RNase3 or with HC-Pro but was detected at a later time in leaves infiltrated with p22 only. These results correspond with the slower appearance of green fluorescence following coinfiltration with p22 than with HC-Pro in the experiments described earlier. RSS mediated by p22 and HC-Pro was also indicated in the increased amounts of RNase3 protein produced in the coinfiltrated leaves, compared to leaves infiltrated for expression of RNase3 alone (Fig. 5B). Importantly, in all experiments, leaves coinfiltrated with p22 and RNase3 showed at least twofold-higher gfp mRNA levels than the leaves coinfiltrated with p22 and GUS. Furthermore, the levels of 21-nt gfp siRNA decreased by approximately fivefold in leaves coinfiltrated with p22 and RNase3, compared with leaves coinfiltrated with p22 and GUS (Fig. 5A).
Experiments were carried out independently at least four times in each of two collaborating laboratories (SLU, Sweden, and University of Helsinki, Finland). Similar results were obtained, showing that even though RNase3 itself exhibits no detectable RSS activity in our assays, it enhances the RSS activity of p22 remarkably, as revealed by the increased accumulation of gfp mRNA and reduced accumulation of the corresponding siRNA.
Enhancement of p22-mediated RSS requires RNase3 endonuclease activity. RNase III can affect gene expression based on its properties as a dsRNA-processing and dsRNA-binding protein (13, 49). The specificity for and recognition of dsRNA by RNase III are associated with the dsRBD. The endoND mediates dsRNA cleavage and includes a highly conserved signature motif, within which a glutamic acid residue corresponding to D44 in SPCSV RNase3 is crucial for cleavage. Indeed, as described earlier, mutant RNase3-Ala37,44 cannot cleave dsRNA. However, the endonuclease activity of RNase III is not essential for substrate binding; on the other hand, dsRNA is cleaved in vitro by a truncated E. coli RNase III lacking the dsRBD (54). The independent activities of the endoND and dsRBD prompted us to investigate the requirement of each for RNase3-mediated enhancement of RSS.
The RNase3 gene fragments (Fig. 6A) encoding endoND (corresponding to aa 1 to 152 of RNase3) or dsRBD (aa 153 to 228) and RNase3-Ala37,44 were cloned into a binary vector, and each vector was used for Agrobacterium coinfiltration with the constructs expressing p22, GFP, and hpGFP. Accumulation of gfp mRNA and siRNA were tested in the high- and low-molecular-weight fractions of total RNA, respectively, at 5 dpi. Northern blot analyses indicated a consistent increase in gfp mRNA levels in the leaves coinfiltrated with p22 and RNase3 (Fig. 6C, top panel, lane 6), HC-Pro (lane 3), or HC-Pro and RNase3 (lane 10), compared with leaves infiltrated with p22 (lane 5) or coinfiltrated with p22 and dsRBD, endoND, or RNase3-Ala37,44 (lanes 7, 8, and 9, respectively). These data suggest that the RNase3 mutants do not enhance p22-mediated RSS. This conclusion was further supported by an analysis of siRNA accumulation. The two size classes of siRNA were electrophoretically separated on a denaturing polyacrylamide gel, and synthetic RNA oligonucleotides (21 nt and 25 nt) were run in the gel as size markers. In the leaves infiltrated for coexpression of p22 along with either endoND, dsRBD, or RNase3-Ala37,44, siRNA accumulated to the same level as in leaves infiltrated with p22 (Fig. 6C, three lower panels; compare lanes 7, 8, or 9 with lane 5), whereas in leaves coinfiltrated with p22 and RNase3 (lane 6), siRNA accumulation was significantly reduced (by up to 10 fold compared with GFP plus hpGFP), as before (Fig. 6C). The enhancing effect of RNase3 on RSS was p22 specific, since the presence or absence of RNase3 had no detectable effect on siRNA accumulation upon infiltration with HC-Pro (Fig. 6C, compare lanes 3 and 10), despite the high amounts of RNase3 and RNase-Ala37,44 proteins expressed in all infiltrated leaves (Fig. 5B). Taken together, these results indicate that the integrity of SPCSV RNase3 and its endonuclease activity are required to enhance RSS activity. Moreover, RNase3 seems to act as an enhancer of RSS activity only with the cognate protein (p22).
p22 shows little effect on transivity of RNA silencing. Transitivity in RNA silencing refers to the generation of secondary siRNAs from sequences located outside the inducer sequence. In plants, transivity can proceed in both 3' and 5' directions (25, 51). To investigate whether p22, RNase3, or their coexpression affects transivity in silencing of gfp mRNA, the membranes previously tested with the entire GFP probe were stripped and hybridized with two nonoverlapping probes (Fig. 6B). Probe 5'-GFP encompasses two-thirds of the gfp sequence and corresponds to the hpGFP (Fig. 1B) used as the gfp silencing inducer, whereas probe 3'-GFP encompasses the 3'-proximal third of the gfp sequence (Fig. 6B). The 3'-GFP probe was expected to reveal the secondary siRNA resulting from transivity. Northern blot hybridization and subsequent quantification showed similar relative siRNA accumulation in different samples, regardless of the probe used for siRNA detection (Fig. 6C, bottom two panels). Thus, the expression of p22, RNase3, or their coexpression had no detectable influence on transivity.
RNase3 cleaves 25-nt synthetic dsRNA but not the 21-nt and 23-nt siRNAs generated during gfp mRNA silencing. Because the endonuclease activity of RNase3 was needed for the enhancement of p22-mediated RSS activity, we hypothesized that RNase3 might cleave the two classes of siRNA. The minimum dsRNA substrate length for E. coli RNase III is 20 nt, which is approximately equivalent to two turns of A-form dsRNA (15) and close to the two size classes of siRNA generated from gfp mRNA in N. benthamiana (21 and 23 nt) (Fig. 6C). Therefore, the activity of purified recombinant RNase3 was tested on synthetic dsRNAs (21 nt and 25 nt) analogous to siRNA with 2-nt 3' overhangs and a 3' hydroxyl group, as well as on gfp siRNA extracted from leaves. Controls included siRNA denatured prior to the assay, the mutant RNase3-Ala37,44, and E. coli RNase III (Ambion). Neither RNase3 nor RNase-Ala37,44 cleaved the 21-nt or 23-nt siRNAs extracted from plants (Fig. 7A, lanes 6 to 9), in contrast to E. coli RNase III (Fig. 7A, lanes 10 and 11). However, RNase3 cleaved the synthetic 25-nt dsRNA but not the shorter synthetic 21-nt dsRNA (Fig. 7C), even though each carries 2-nt 3' overhangs as in siRNAs.
We also tested whether the 21-nt or 25-nt synthetic ssRNA or dsRNA could compete with or inhibit RNase3 cleavage of long dsRNA substrates (717 nt), but no such evidence was found in a time course analysis using a 600-fold molar excess of the short RNAs relative to long dsRNA (Fig. 7B). The finding that RNase3 did not cleave siRNA did not, however, exclude the possibility that it binds siRNA. To test this possibility, we performed gel-shift experiments with recombinant RNase3 and RNase3-Ala37,44. Both enzymes bound radiolabeled synthetic 21-nt and 25-nt RNAs, as indicated by band shifting (Fig. 7D; also Fig. 7C, lane 13.5 μM RNase3).
DISCUSSION
RNase III enzymes play various roles in RNA metabolism and gene expression and are divided into three classes according to their structure (9). SPCSV RNase III (RNase3) studied here and E. coli RNase III belong to the simplest class (class 1), containing a single endoND and dsRBD. Our data show that RNase3 enhances the suppression of RNA silencing mediated by a cognate RSS protein, p22. The RNase IIIs of classes 2 and 3 contain duplicate endoND domains and are distinguished by their N-terminal extensions. Drosha, a class 2 RNase III, is involved in maturation of microRNAs (miRNAs) that regulate animal development (34), whereas Dicer belongs to class 3 and is pivotal in the initiation and amplification of RNA silencing via its participation in RISC, cleaving silencing-inducing dsRNA molecules and homologous RNAs to 21- to 24-nt siRNAs (3, 23, 57). Taken together, the role of RNase III in the initiation and amplification of RNA silencing is well established (57), but our data now show that some RNase III enzymes interfere with this process. It therefore seems that different classes of RNase III endoribonucleases play different roles in RNA silencing.
The viral RSS proteins known to date, including p22 identified here, show no significant similarity to cellular proteins (40), in contrast to SPCSV RNase3 that is similar to the putative RNase III class 1 proteins of Arabidopsis thaliana (32) and rice Oryza sativa (BAD36550, for which functions have not been reported. Plants encode multiple, homologous Dicer-like RNase III enzymes. Four homologs are found in rice (a monocot) and A. thaliana (a dicot). Three of the A. thaliana homologs are implicated in different RNA silencing pathways: DCL1 in miRNA, DCL2 in antiviral siRNA, and DCL3 in endogenous siRNA biogenesis (64). Intriguingly, the genomes of SPCSV (32) and other plant viruses of the family Closteroviridae may have incorporated host genes, as evidenced by the genes in viral genomes that encode homologs of plant heat shock proteins and are required for virulence (29). The SPCSV RNase3 gene may likewise be of host origin, exemplifying a novel strategy by which viruses employ host proteins to enhance virulence or suppression of various types of host defenses (4), in this case via interference with RNA silencing. The finding that p22 suppresses RNA silencing without RNase3 is consistent with this theory and may also suggest that putative host factors can fulfill the function of RNase3 (although less efficiently). This scenario for the origin of RNase3, together with the fundamental role of RNA silencing in gene regulation in eukaryotic organisms (10), implies that some plant RNase IIIs may be involved in RSS. This would provide another mechanism for the negative regulation of RNA silencing in addition to that described for A. thaliana DCL1, the RNase III that positively regulates miRNA biogenesis but is itself controlled by miRNA-mediated RNA silencing (63).
To date, SPCSV is the first virus for which two independent proteins have been found to cooperatively control RSS functions. Very recently, another study has shown that Citrus tristeza virus belonging to the same family but a different genus (Closterovirus) encodes three RSS proteins, but their cooperative functions in RSS have not been studied (38). Results from several experimental approaches employed in this study consistently indicated that although p22 alone can efficiently suppress RNA silencing induced by dsRNA of various types, the endonuclease activity of RNase3 significantly enhances p22-mediated RSS, resulting in a remarkable reduction of siRNA accumulation compared with that observed when p22 is expressed alone.
The substitutions E37A and D44A made to RNAse3 in our study to abolish the endonuclease activity of the protein were based on the crystal structure and functional model of class 1 RNase III of Aquifex aeolicus and E. coli (5). Two hydrolysis events catalyzed by RNase III cleave dsRNA at sites two bases apart. The first RNA-cleaving site of A. aeolicus RNase III consists of residues E37 (corresponding to E37 in RNase3 of SPCSV) and E64, each from a different RNase III molecule of the dimer. The second RNA-cleaving site is composed of residues D44 (as in RNase3) and E110 (5). Site-directed amino acid substitutions E38V and D45A introduced to the E. coli RNase III (which correspond to E37 and D44 of RNase3) abolished its endonuclease activity in vivo but did not affect the RNA-binding ability (5). Similarly, RNase3 mutated at positions 37 (E37A) and 44 (D44A) bound long and short (21 to 25nt) dsRNA substrates efficiently but was devoid of endonuclease activity. These data suggest that no other known functions of RNase III (dimerization, for instance) were affected by the mutations introduced to RNase3. In support of our data, a mutant of E. coli RNase III that binds dsRNA but lacks the RNA cleavage activity exhibits only weak RSS functions in an agroinfiltration assay (36).
The mechanism of the coordinated functions of p22 and RNase3 in RSS, however, cannot yet be fully elucidated. Initial cleavage of dsRNA was probably not affected because Northern blots revealed no signals for the inducing dsRNA (hpGFP) in samples undergoing RSS. Transitivity was obviously not affected by p22 and/or RNase3, implying that RNA-dependent RNA polymerase-mediated amplification of the silencing signal was unlikely affected (14, 25, 41). The p21 RSS protein of Beet yellows virus (BYV; genus Closterovirus) (46) was recently shown to bind siRNA and miRNA, indicating that p21 may sequester siRNA such that these RNAs are unavailable to RISC (11). This hypothesis awaits testing with SPCSV p22, which is recalcitrant to purification (as initially was the case for BYV p21) (46). However, SPCSV p22 and BYV p21 show no significant sequence similarity and their functions differ, as p21 expression does not reduce siRNA accumulation or induce chlorosis in N. benthamiana leaves (46), in contrast to p22 studied here. Therefore, RSS mediated by p22 and enhanced by RNase3 may proceed via a novel mechanism. We favor a hypothesis whereby the specificity of RNase3 is modulated in the cellular environment via interaction with RISC and/or p22, such that RNase3 may cleave siRNA bound by RISC and/or p22; the efficiency by which RISC targets new homologous RNAs for cleavage is thereby reduced. This hypothesis is supported by studies with Saccharomyces cerevisiae RNase IIIs, the specificity of which is altered or endonuclease activity enhanced by interaction with specific proteins or protein complexes (19, 52). Additional RSS may result from interference of the viral RNase III with the loading of siRNAs into RISC by cellular RNase IIIs (44) or from interference of p22 and/or RNase3 with the incorporation of proteins or protein complexes into RISC, a process with which the P1/HC-Pro polyprotein of potyviruses has been proposed to interfere (11).
SPCSV is the first RNA virus shown to encode an RNase III enzyme. A few large, dsDNA-containing viruses infecting fish, insects, or algae encode putative RNase III proteins (27, 53, 56, 59), and the enzymatic activity of RNase III encoded by Paramecium bursaria Chlorella virus 1 has been characterized (66). Still, the functions of these enzymes in the viral infection cycle are unknown. Taken together, our data suggest an entirely novel mechanism in viral pathogenesis and virulence and reveal a role for RNase III class 1 enzymes in RNA silencing, an area in which they were previously not known to be involved.
ACKNOWLEDGMENTS
We are grateful to Anna Germundsson and Marjo Ala-Poikela for collaboration, David Baulcombe for providing line 16c, and Andrey Zamyatnin for help with preparing figures.
We gratefully acknowledge financial support from the European Union Programme for International Cooperation with Developing Countries (FP5 INCO-DEV, project ICA4-CT-2000-3007) and the Academy of Finland (grant 1102134).
J.F.K. and E.I.S. contributed equally to the study.
REFERENCES
Ahlquist, P. 2002. RNA-dependent RNA polymerases, viruses, and RNA silencing. Science 296:1270-1273.
Anandalakshmi, R., G. J. Pruss, X. Ge, R. Marathe, A. C. Mallory, T. H. Smith, and V. B. Vance. 1998. A viral suppressor of gene silencing in plants. Proc. Natl. Acad. Sci. USA 95:1379-1384.
Bernstein, E., A. A. Caudy, S. M. Hammond, and G. J. Hannon. 2001. Role for a bidentate ribonuclease in the initiation step of RNA interference. Nature 409:363-366.
Bilgin, D. D., Y. Liu, M. Schiff, and S. P. Dinesh-Kumar. 2003. P58IPK, a plant ortholog of double-stranded RNA-dependent protein kinase PKR inhibitor, functions in viral pathogenesis. Dev. Cell 4:651-661.
Blaszczyk, J., J. E. Tropea, M. Bubunenko, K. M. Routzahn, D. S. Waugh, D. L. Court, and X. Ji. 2001. Crystallographic and modeling studies of RNase III suggest a mechanism for double-stranded RNA cleavage. Structure 9:1225-1236.
Boevink, P., S. Santa Cruz, C. Hawes, N. Harris, and K. J. Oparka. 1996. Virus-mediated delivery of the green fluorescent protein to the endoplasmic reticulum of plant cells. Plant J. 10:935-941.
Brigneti, G., O. Voinnet, W. X. Li, L. H. Ji, S. W. Ding, and D. C. Baulcombe. 1998. Viral pathogenicity determinants are suppressors of transgene silencing in Nicotiana benthamiana. EMBO J. 17:6739-6746.
Calin-Jageman, I., A. K. Amarasinghe, and A. W. Nicholson. 2001. Ethidium-dependent uncoupling of substrate binding and cleavage by Escherichia coli ribonuclease III. Nucleic Acids Res. 29:1915-1925.
Carmell, M. A., and G. J. Hannon. 2004. RNase III enzymes and their role in initiation of gene silencing. Nat. Struct. Mol. Biol. 11:214-218.
Carrington, J. C., and V. Ambros. 2003. Role of microRNAs in plant and animal development. Science 301:336-338.
Chapman, E. J., A. I. Prokhnevsky, K. Gopinath, V. V. Dolja, and J. C. Carrington. 2004. Viral RNA silencing suppressors inhibit the microRNA pathway at an intermediate step. Genes Dev. 18:1179-1186.
Cogoni, C., and G. Macino. 2000. Post-transcriptional gene silencing across the kingdoms. Curr. Opin. Genet. Dev. 10:638-643.
Conrad, C., and R. Rauhut. 2002. Ribonuclease III: new sense from nuisance. Int. J. Biochem. Cell Biol. 34:116-129.
Dalmay, T., A. Hamilton, S. Rudd, S. Angell, and D. C. Baulcombe. 2000. An RNA-dependent RNA polymerase gene in Arabidopsis is required for posttranscriptional gene silencing mediated by a transgene but not a virus. Cell 101:543-553.
Dunn, J. J. 1982. Ribonuclease III, p. 485-499. In B. D. Boyer (ed.), The enzymes, vol. 15. Academic Press, New York, N.Y.
Dunoyer, P., C. H. Lecellier, E. A. Parizotto, C. Himber, and O. Voinnet. 2004. Probing the microRNA and small interfering RNA pathways with virus-encoded suppressors of RNA silencing. Plant Cell 16:1235-1250.
Fire, A., S. Xu, M. K. Montgomery, S. A. Kostas, S. E. Driver, and C. C. Mello. 1998. Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 391:806-811.
Gibson, R. W., I. Mpembe, T. Alicai, E. E. Carey, R. O. M. Mwanga, S. E. Seal, and H. J. Vetten. 1998. Symptoms, aetiology and serological analysis of sweet potato virus disease in Uganda. Plant Pathol. 47:95-102.
Giorgi, C., A. Fatica, R. Nagel, and I. Bozzoni. 2001. Release of U18 snoRNA from its host intron requires interaction of Nop1p with the Rnt1p endonuclease. EMBO J. 20:6856-6865.
Gitlin, L., S. Karelsky, and R. Andino. 2002. Short interfering RNA confers intracellular antiviral immunity in human cells. Nature 418:430-434.
Hamilton, A., O. Voinnet, L. Chappell, and D. Baulcombe. 2002. Two classes of short interfering RNAs in RNA silencing. EMBO J. 21:4671-4679.
Hamilton, A. J., and D. C. Baulcombe. 1999. A species of small antisense RNA in posttranscriptional gene silencing in plants. Science 286:950-952.
Hammond, S. M., E. Bernstein, D. Beach, and G. Hannon. 2000. An RNA-directed nuclease mediates post-transcriptional gene silencing in Drosophila cell extracts. Nature 404:293-296.
Hannon, G. J. 2002. RNA interference. Nature 418:244-251.
Himber, C., P. Dunoyer, G. Moissiard, C. Rizenthaler, and O. Voinnet. 2003. Transivity-dependent and -independent cell-to-cell movement of RNA silencing. EMBO J. 22:4523-4533.
Hutvágner, G., J. McLachlan, A. E. Pasquinelli, é. Bálint, T. Tuschl, and P. D. Zamore. 2001. A cellular function for the RNA-interference enzyme Dicer in the maturation of the let-7 small temporal RNA. Science 293:834-838.
Jakob, N. J., K. Muller, U. Bahr, G. Darai. 2001. Analysis of the first complete DNA sequence of an invertebrate iridovirus: coding strategy of the genome of Chilo iridescent virus. Virology 286:182-196.
Johansen, L. K., and J. C. Carrington. 2001. Silencing on the spot. Induction and suppression of RNA silencing in the Agrobacterium-mediated transient expression system. Plant Physiol. 126:930-938.
Karasev, A. V. 2000. Genetic diversity and evolution of closteroviruses. Annu. Rev. Phytopathol. 38:293-324.
Karyeija, R. F., J. F. Kreuze, R. W. Gibson, and J. P. T. Valkonen. 2000. Synergistic interactions of a potyvirus and a phloem-limited crinivirus in sweet potato plants. Virology 269:26-36.
Kasschau, K. D., and J. C. Carrington. 1998. A counterdefensive strategy of plant viruses: suppression of posttranscriptional gene silencing. Cell 95:461-470.
Kreuze, J. F., E. I. Savenkov, and J. P. T. Valkonen. 2002. Complete genome sequence and analyses of the subgenomic RNAs of Sweet potato chlorotic stunt virus reveal several new features for the genus Crinivirus. J. Virol. 76:9260-9270.
Lakatos, L., G. Szittya, D. Silhavy, and J. Burgyán. 2004. Molecular mechanism of RNA silencing suppression mediated by p19 protein of tombusviruses. EMBO J. 23:876-884.
Lee, Y., C. Ahn, J. Han, H. Choi, J. Kim, J. Yim, J. Lee, P. Provost, O. R?dmark, S. Kim, and N. Kim. 2003. The nuclear RNase III Drosha initiates microRNA processing. Nature 425:415-419.
Li, H., W. X. Li, and S. W. Ding. 2002. Induction and suppression of RNA silencing by an animal virus. Science 296:1319-1321.
Lichner, Z., D. Silhavy, and J. Burgyán. 2003. Double-stranded RNA-binding proteins could suppress RNA interference-mediated antiviral defences. J. Gen. Virol. 84:975-980.
Llave, C., Z. Xie, K. D. Kasschau, and J. C. Carrington. 2002. Cleavage of Scarecrow-like mRNA targets directed by a class of Arabidopsis miRNA. Science 297:2053-2060.
Lu, R., A. Folimonov, M. Shintaku, W. X. Li, B. W. Falk, W. O. Dawson, and S. W. Ding. 2004. Three distinct suppressors of RNA silencing encoded by a 20-kb viral RNA genome. Proc. Natl. Acad. Sci. USA 101:15742-15747.
Mallory, A. C., L. Ely, T. H. Smith, R. Marathe, R. Anandalakshmi, M. Fagard, H. Vaucheret, G. Pruss, L. Bowman, and V. B. Vance. 2001. HC-Pro suppression of gene silencing eliminates the small RNAs but not the transgene methylation or the mobile signal. Plant Cell 13:571-583.
Moissiard, G., and O. Voinnet. 2004. Viral suppression of RNA silencing in plants. Mol. Plant Pathol. 5:71-82.
Mourrain, P., C. Beclin, T. Elmayan, F. Feuerbach, C. Godon, J. B. Morel, D. Jouette, A. M. Lacombe, S. Nikic, N. Picault, K. Remoue, M. Sanial, T. A. Vo, and H. Vaucheret. 2000. Arabidopsis SGS2 and SGS3 genes are required for posttranscriptional gene silencing and natural virus resistance. Cell 101:533-542.
Nicholson, A. W. 1996. Structure, reactivity, and biology of double-stranded RNA. Prog. Nucleic Acid Res. Mol. Biol. 52:1-65.
Palaqui, J. C., T. Elmayan, J. M. Pollien, and H. Vaucheret. 1997. Systemic acquired silencing: transgene-specific post-transcriptional silencing is transmitted by grafting from silenced stocks to non-silenced scions. EMBO J. 16:4738-4745.
Pham, J. W., J. L. Pellino, Y. S. Lee, R. W. Carthew, and E. J. Sontheimer. 2004. A Dicer-2-dependent 80S complex cleaves targeted mRNAs during RNAi in Drosophila. Cell 117:83-94.
Ratcliff, F. G., S. A. MacFarlane, and D. C. Baulcombe. 1999. Gene silencing without DNA. RNA-mediated cross-protection between viruses. Plant Cell 11:1207-1216.
Reed, J. C., K. D. Kasschau, A. I. Prokhnevsky, K. Gopinath, G. P. Pogue, J. C. Carrington, and V. V. Dolja. 2003. Suppressor of RNA silencing encoded by Beet yellows virus. Virology 306:203-209.
Ruiz, M. T., O. Voinnet, and D. C. Baulcombe. 1998. Initiation and maintenance of virus-induced gene silencing. Plant Cell 10:937-946.
Sambrook, J., and D. W. Russel. 2001. Molecular cloning: a laboratory manual, 3rd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Saunders, L. R., and G. N. Barber. 2003. The dsRNA binding protein family: critical roles, diverse cellular functions. FASEB J. 17:961-983.
Savenkov, E. I., and J. P. T. Valkonen. 2001. Potyviral helper-component proteinase expressed in transgenic plants enhances titres of Potato leaf-roll virus but does not alleviate its phloem limitation. Virology 283:285-293.
Sijen, T., J. Fleenor, F. Simmer, K. L. Thijssen, S. Parrish, L. Timmons, R. H. A. Plasterk, and A. Fire. 2001. On the role of RNA amplification in dsRNA-triggered RNA silencing. Cell 107:465-476.
Spasov, K., L. I. Perdomo, E. Evakine, and R. N. Nazar. 2002. RAC protein directs the complete removal of the 3' external transcribed spacer by the Pac nuclease. Mol. Cell 9:433-437.
Stasiak, K., M. V. Demattei, B. A. Federici, and Y. Bigot. 2000. Phylogenetic position of the Diadromus pulchellus ascovirus DNA polymerase among viruses with large double-stranded DNA genomes. J. Gen. Virol. 81:3059-3072.
Sun, W., E. J. Jun, and A. W. Nicholson. 2001. Intrinsic double-stranded-RNA processing activity of Escherichia coli ribonuclease III lacking the dsRNA-binding domain. Biochemistry 40:14976-14984.
Tabara, H., M. Sarkissian, W. G. Kelly, J. Fleenor, A. Grishok, L. Timmons, A. Fire, and C. C. Mello. 1999. The rde-1 gene, RNA interference, and transposon silencing in C. elegans. Cell 99:123-132.
Tidona, C. A., and G. Darai. 1997. The complete DNA sequence of lymphocystis disease virus. Virology 230:207-216.
Tijsterman, M., and R. H. A. Plasterk. 2004. Dicers at RISC: the mechanism of RNAi. Cell 117:1-3.
Vancanneyt, G., R. Schmidt, A. O'Connor-Sanchez, L. Willmitzer, and M. Rocha-Sosa. 1990. Construction of an intron-containing marker gene: splicing of the intron in transgenic plants and its use in monitoring early events in Agrobacterium-mediated plant transformation. Mol. Gen. Genet. 220:245-250.
Van Etten, J. L., and R. H. Meints. 1999. Giant viruses infecting algae. Annu. Rev. Microbiol. 53:447-494.
Voinnet, O., and D. C. Baulcombe. 1997. Systemic signaling in gene silencing. Nature 389:553.
Voinnet, O., C. Lederer, and D. C. Baulcombe. 2000. A viral movement protein prevents spread of the gene silencing signal in Nicotiana benthamiana. Cell 103:157-167.
Volpe, T. A., C. Kidner, I. M. Hall, G. Teng, S. I. S. Grewal, and R. A. Martienssen. 2002. Regulation of heterochromatic silencing and histone H3 lysine-9 methylation by RNAi. Science 297:1833-1837.
Xie, Z., K. D. Kasschau, and J. C. Carrington. 2003. Negative feedback regulation of Dicer-like1 in Arabidopsis by microRNA-guided mRNA degradation. Curr. Biol. 13:784-789.
Xie, Z., L. K. Johansen, A. M. Gustafson, K. D. Kasschau, A. D. Lellis, D. Zilberman, S. E. Jacobsen, and J. C. Carrington. 2004. Genetic and functional diversification of small RNA pathways in plants. PLoS Biol. 2:642-652.
Yelina, N. E., E. I. Savenkov, A. G. Solovyev, S. Y. Morozov, and J. P. T. Valkonen. 2002. Long-distance movement, virulence, and RNA silencing suppression controlled by a single protein in hordei- and potyviruses: complementary functions between virus families. J. Virol. 76:12981-12991.
Zhang, Y., I. Calin-Jageman, J. R. Gurnon, T. J. Choi, B. Adams, A. W. Nicholson, and J. L. Van Etten. 2003. Characterization of a chlorella virus PBCV-1 encoded ribonuclease III. Virology 317:73-83.(Jan F. Kreuze , Eugene I.)
Department of Applied Biology, PO Box 27, FIN-00014 University of Helsinki, Finland
ABSTRACT
Double-stranded RNA (dsRNA)-specific endonucleases belonging to RNase III classes 3 and 2 process dsRNA precursors to small interfering RNA (siRNA) or microRNA, respectively, thereby initiating and amplifying RNA silencing-based antiviral defense and gene regulation in eukaryotic cells. However, we now provide evidence that a class 1 RNase III is involved in suppression of RNA silencing. The single-stranded RNA genome of sweet potato chlorotic stunt virus (SPCSV) encodes an RNase III (RNase3) homologous to putative class 1 RNase IIIs of unknown function in rice and Arabidopsis. We show that RNase3 has dsRNA-specific endonuclease activity that enhances the RNA-silencing suppression activity of another protein (p22) encoded by SPCSV. RNase3 and p22 coexpression reduced siRNA accumulation more efficiently than p22 alone in Nicotiana benthamiana leaves expressing a strong silencing inducer (i.e., dsRNA). RNase3 did not cause intracellular silencing suppression or reduce accumulation of siRNA in the absence of p22 or enhance silencing suppression activity of a protein encoded by a heterologous virus. No other known RNA virus encodes an RNase III or uses two independent proteins cooperatively for RNA silencing suppression.
INTRODUCTION
RNA silencing is a eukaryotic cellular surveillance mechanism (12) that defends against viruses (1, 20, 41), controls transposable elements (55), and participates in the formation of silent chromatin (62). RNA silencing is also involved in posttranscriptional regulation of gene expression during developmental processes (26, 37). Double-stranded RNA (dsRNA) is the most potent trigger for RNA silencing (17, 24). The inducing dsRNA is processed into two classes of small interfering RNAs (siRNAs) of ca. 21 to 24 nucleotides (nt) in plants (21, 22) by the RNase III (RNase III) class 3-like enzyme Dicer (3). The siRNA incorporates into a nuclease-containing RNA-induced silencing complex (RISC), and the antisense strand guides it to homologous RNAs that are subsequently degraded (23, 57). The amount of siRNA can be amplified because of transivity, i.e., the production of additional siRNAs from sequences located outside the original target region. Transivity occurs in a 3'-to-5' direction in Caenorhabditis elegans but is bidirectional in plants (25, 51). RNA silencing in plants involves an as-yet-unknown signal that moves out from the cells undergoing RISC-mediated RNA degradation, spreading the silencing effect over short distances (25), as well as to other parts of the plant (43, 60).
The host cell RNA-silencing mechanism targets replicating viruses. Viruses, in turn, must suppress RNA silencing to ensure successful invasion. The first RNA silencing-suppressing (RSS) proteins were found in plant viruses (2, 7, 31) and have since been identified in many unrelated plant (40) and animal (35) viruses. RSS proteins seem to be versatile in their function, since they can interfere with RNA silencing in plant, animal, and human cells (16, 35), consistent with the conserved nature of the RNA silencing mechanism in eukaryotes in general (12, 24). Different RSS proteins can interfere with different steps in the RNA-silencing pathway. For example, RSS proteins may prevent the initiation of silencing (7), the spread of the silencing signal (61), or the maintenance of the silenced stage (39).
Sweet potato chlorotic stunt virus (SPCSV; genus Crinivirus, family Closteroviridae) has a large bipartite (RNA, 9,407 nucleotides; RNA2, 8,223 nt), single-stranded, positive-sense RNA genome (32). It is an agriculturally important pathogen of sweet potato (Ipomoea batatas L.), especially because it breaks down resistance to unrelated viruses during a dual infection, significantly increasing disease severity (18). The synergistic effects of SPCSV on other viruses may be due to interference with RNA silencing, because they are associated with substantially increased accumulation of coinfecting viruses (30). SPCSV encodes proteins that are not found in other RNA viruses. For example, RNA1 contains an open reading frame (ORF) for a putative RNase III (RNase3) and also a 22-kDa protein (p22) that show no significant similarity to known proteins from any organism. The subgenomic RNAs (sgRNAs) encoding these proteins are expressed early in infection (32).
The detection of novel genes in the SPCSV genome prompted us to study whether any of the genes exhibit RSS functions, which might help to elucidate the mechanism by which SPCSV synergizes with other viruses in sweet potato plants. Our data reveal that p22 efficiently suppresses RNA silencing, which is further enhanced by the dsRNA-specific endonuclease activity of RNase3. These data provide the first evidence that RNase III enzymes are involved not only in RNA silencing but also in its suppression.
MATERIALS AND METHODS
Molecular cloning. Sequences of the primers used in this study will be provided upon request. Plasmids pAHCPro (50) and p35S-GUS-INT (58) have been described previously. They are based on the binary plant transformation vector pKOH200 and contain the helper component proteinase (HC-Pro) encoding sequence of Potato virus A (PVA; genus Potyvirus) and the the ?-galacturonidase (GUS) gene, respectively, flanked by the Cauliflower mosaic virus (CaMV; genus Caulimovirus) 35S promoter and the 3' terminator region of the nopaline synthase gene (3'nos). The GUS gene contains a plant intron to prevent GUS expression in Agrobacterium (58). The "cycle 3" gfp gene was amplified from pGFPuv (Clontech) by PCR, and the ORFs for RNase3, p22 (RNA1), and p28 (RNA2) were amplified by reverse transcription-PCR (RT-PCR) from purified viral SPCSV RNA. To obtain the constructs for agroinfiltration, the HC-Pro fragment in pAHCPRo was replaced by the aforementioned PCR products, resulting in constructs pAGFP, pARNase3, pAp22, and pAp28. The NotI-BamHI fragment of pAGFP, encompassing the N-terminal two-thirds (526 bp) of the gfp gene, was cloned into pKOH122 in the sense orientation between the 35S promoter and a BclI/XbaI-digested fragment of the IV2 intron (58). The resulting plasmid was subsequently digested with BamHI and SbfI, and the NotI/BamHI-digested fragment of gfp was cloned into this digested plasmid in the antisense orientation, together with the NotI/SbfI-digested 3' nos terminator fragment, to yield construct hpGFP.
PVX-p22, PVX-RNase3, PVX-p28, and PVX-HCpro vectors were constructed by cloning the corresponding genes into potato virus X (PVX), as described previously (65). pPVX201 contains the complete infectious cDNA of Potato virus X (genus Potexvirus) under the control of the CaMV 35S promoter and a duplicated copy of the coat protein gene sgRNA promoter from the plasmid pTXS.P3C2 (6). The SPCSV genes encoding p22, p28, and RNase3 proteins or the PVA HC-Pro-encoding region were amplified by PCR and inserted between the NheI and SalI restriction sites of pPVX201. Mutants of the p22 gene were generated based on pAp22 or PVX-p22 with the QuickChange XL Site-Directed Mutagenesis kit (Stratagene) according to the manufacturer's instructions.
For RNase3 protein expression in Escherichia coli, the ORF for RNase3 was amplified by RT-PCR from a viral RNA template using oligonucleotide primers that introduced the codons for a C-terminal hexahistidine (six-His) tag and the NcoI restriction site required for cloning. The resulting PCR product was digested with NcoI and cloned into the corresponding site of the dephosphorylated pET11d (Stratagene) vector for protein expression.
The predicted endonuclease domain (endoND; amino acids [aa] 1 to 152) and dsRNA-binding domain (dsRBD; aa 153 to 228) of RNase3 were cloned into the binary plant transformation vector pKOH200 as previously described (50). An intermediate construct, pKOH122-RNase3, was obtained by cloning the RNase3 gene into plasmid pKOH122 using the NotI and FseI restriction sites, flanked by the 35S promoter and the 3'nos terminator. The RNase3 mutant (RNase3-Ala37,44), in which Asp37 and Asp44 were substituted for Ala, was obtained by replacing the sequence in the pKOH122-RNase3 construct between NotI and FseI restriction sites with two equivalent PCR fragments into which mutations and a unique NarI restriction site had been introduced by PCR and which were digested with NotI/NarI and with NarI/FseI prior to ligation into the NotI/FseI-digested pKOH122-RNase3. The four point mutations within SPCSV RNase3 ORF resulted in the amino acid substitutions D37A and D44A and introduced a unique NarI restriction site. The whole cassette (35S-Rnase3-Ala37,44-3'nos) was then excised from the pKOH122 plasmid and inserted between the Sse8387I and PacI restriction sites of pKOH200. The resulting plasmid, pKOH-RNase3-Ala37,44, was subsequently used as a template for two-step PCR with nested primers. The resulting PCR product was digested with NcoI and cloned into the corresponding site of dephosphorylated pET11d. Upon expression, the resulting plasmid yielded a recombinant mutant RNase3 protein with a six-His tag at the C terminus and two amino acid substitutions, D37A and D44A, in the RNase III signature motif.
The fidelity of all constructs was confirmed by sequencing.
Protein expression and analysis. The six-His fusion proteins (RNase3 and RNase-Ala37,44) were expressed in E. coli BL21-RIL cells (Stratagene) according to the manufacturer's protocol. We purified recombinant RNase3 under native conditions by affinity chromatography using Ni-nitrilotriacetic acid agarose (QIAGEN) according to the manufacturer's protocol. Fractions containing the highest concentrations of recombinant RNase3 were pooled, dialyzed against enzyme storage buffer (30 mM Tris-HCl [pH 8.0], 500 mM KCl, 0.1 mM EDTA, 0.1 mM dithiothreitol, 50% glycerol [vol/vol]), and stored at –20°C.
For the production of polyclonal antibodies, the six-His RNase3 protein was dialyzed against buffered saline solution (0.9% NaCl, 10 mM Tris-HCl, pH 6.0). The protein (25 μg, in Freund's adjuvant) was injected into a rabbit; the immunization was carried out six times at 4-wk intervals at the National Veterinary Institute, Uppsala, Sweden.
RNase3 was detected by Western blot analysis with RNase3-specific polyclonal antibodies and anti-rabbit mouse monoclonal antibodies conjugated with horseradish peroxidase (Sigma) by standard procedures (48). Signals were detected by chemiluminescence by the ECL method (Amersham Pharmacia Biotech). In cross-protection assays, vector viruses were detected by double-antibody sandwich enzyme-linked immunosorbent assay (DAS-ELISA) as previously described (50) with anti-coat protein antibodies purchased from Boehringer and Adgen.
Nucleic acid substrates for enzymatic analysis. To study RNA cleavage in vitro, all RNAs were generated by transcription in vitro. The gfp gene was PCR amplified, digested with NcoI, and cloned into the corresponding dephosphorylated site in pET11d. Plasmids carrying gfp in the sense or antisense orientation were selected. Each plasmid was linearized by BamHI and transcribed by T7 RNA polymerase (Promega) according to the manufacturer's protocol. RNA transcripts were phenol-chloroform extracted, precipitated with ethanol, and dissolved in RNase-free water. Equimolar amounts of the in vitro-transcribed single-stranded RNAs (ssRNAs) were annealed to yield dsRNA substrates by being heated for 5 min at 95°C, followed by gradual cooling to room temperature in a water bath. Double-stranded DNA (dsDNA) substrate was prepared by linearizing pET11d-GFP with BamHI. The RNA-DNA hybrid substrate was generated by RT of the gfp RNA transcript with a green fluorescent protein (GFP)-specific primer. To obtain single-stranded DNA (ssDNA) substrate, the RNA-DNA hybrid substrate was treated with RNaseH (Promega).
The low-molecular-weight (LMW) RNA fraction containing siRNA was isolated from wild-type Nicotiana benthamiana plants coagroinfiltrated with GFP and hairpin GFP (hpGFP). Plant siRNAs were purified from a native 8% polyacrylamide gel as previously described (33). Synthetic 21-nt and 25-nt siRNAs (see below) visualized with ethidium bromide (EtBr) were used as size markers.
Synthetic gfp RNA oligonucleotides (11 nt, 21 nt, and 25 nt; sense and antisense) were purchased from Dharmacon. dsRNA oligonucleotides were obtained by annealing sense and antisense strands in equimolar amounts. The 21-nt and 25-nt dsRNAs were analogous to siRNA, as they contained 2-nt 3' overhangs and 3' hydroxyl groups.
The synthetic 21-nt and 25-nt RNAs (100 pmol) were phosphorylated in the presence of [-32P]ATP in a 50-μl reaction mixture using 30 U of T4 polynucleotide kinase (3'-Phosphatase Minus; Fermentas). Unincorporated nucleotides were removed with Micro Bio-Spin P-30 Tris chromatography columns (Bio-Rad). For electrophoretic mobility shift studies, various concentrations of recombinant six-His-tagged RNase3 or RNase3-Ala37,44 were mixed with radiolabeled RNA in the cleavage buffer and incubated at room temperature for 30 min. Complexes were resolved on a native 6% polyacrylamide gel that was subsequently exposed to Molecular Imager screens (Bio-Rad).
Cleavage assays for RNase III activity were performed with a cleavage buffer (30 mM Tris-HCl [pH 7.5], 10 mM MgCl2, 130 mM KCl, 5% glycerol [vol/vol]).
Agroinfiltration, virus inoculation, and detection of GFP. The transgenic N. benthamiana line 16c, which expresses GFP, has been previously described (47). Agroinfiltration was carried out as described by Johansen and Carrington (28) using Agrobacterium tumefaciens strain C58C1 (with Ti plasmid pGV3850 or pGV2260). For coinfiltration, the different Agrobacterium cultures were diluted with infiltration medium so that equal, final optical densities were achieved before combining the cultures in a 1:1 ratio for infiltration. If some coinfiltration treatments included fewer constructs than others, the missing volume was replaced by the Agrobacterium strain expressing GUS. Inoculation with PVX vectors and Tobacco mosaic virus (TMV)-GFP has been previously described (65).
GFP fluorescence was monitored by epi-illumination with a hand-held UV source (Blackray UVP 100 AP) as previously described (47). Photographs were taken with a digital camera, and images were prepared with Photoshop software (Adobe).
RNA analysis. Total RNA was extracted with TRIzol LS Reagent (Invitrogen, Ltd.) according to the manufacturer's recommendations. LMW and high-molecular-weight (HMW) RNA fractions were separated by precipitation of a HMW fraction with an equal volume of 4 M LiCl4 (final concentration, 2 M) at 4°C overnight. After centrifugation, the supernatant was transferred to a new tube to which an equal volume of isopropanol was added to precipitate LMW RNA. The quality of the extracted RNA was evaluated under UV light after electrophoresis in a standard formaldehyde gel and staining with ethidium bromide (48).
For Northern blot analysis of mRNA, 10 μg HMW RNA was separated by formaldehyde gel electrophoresis, transferred to a nylon membrane (Hybond-N; Amersham Biosciences AB), cross-linked with UV light, prehybridized, hybridized at 55°C, and washed in hybridization tubes (48).
To analyze siRNA, 30 μg LMW RNA was mixed with an equal volume of Tris-borate-EDTA--urea sample buffer (Bio-Rad), heated at 100°C for 5 min, and loaded onto a 15% polyacrylamide Tris-borate-EDTA-urea gel. The RNA was then transferred to a nylon membrane by capillary blotting (48), prehybridized, hybridized, and then washed as described previously (22).
Antisense [-32P]UTP-labeled gfp-specific RNA probes were synthesized with T3 RNA polymerase (Promega) from XbaI-digested pGFPuv according to the manufacturer's protocol. For hybridization to siRNAs, the probe was cleaved by alkaline hydrolysis to an average length of 50 bp, as described previously (22). After s final washing, the blots were wrapped in polyethylene plastic and exposed to an exposure cassette (Molecular Dynamics) for 1 to 48 h. The cassette was then scanned with a Molecular Imager FX, and the signal strength was quantified with Quantity One software (Bio-Rad).
RESULTS
RNase3 has enzymatic activities characteristic of RNase III. The deduced amino acid sequence of SPCSV RNase3 shows similarity to RNase III (32). RNase III belongs to a family of endoribonucleases that specifically recognize and cleave dsRNAs and are found in prokaryotes and eukaryotes (reviewed in references 13 and 49). Three classes of RNase III enzymes have been identified, based on the set of domains they possess. SPCSV RNase3 is similar to the class 1 enzymes (the simplest RNase III enzymes), which contain an endoND and a dsRBD. The sequence of endoND has a characteristic, highly conserved, 9-aa RNase III signature motif, in which three amino acid residues, corresponding to E40, G43, and D44 in SPCSV RNase3, are strictly conserved (5).
We overexpressed and purified the six-His-tagged SPCSV RNase3 and a mutant thereof in E. coli. The mutant, RNase3-Ala37,44, contained alanine instead of aspartic acid at each of two sites (D37 and D44) in the RNase III signature motif. Both proteins were stable and had electrophoretic mobilities corresponding to the expected molecular mass of 27 kDa (Fig. 1A and B). Both recombinant proteins were tested for cleavage of nucleic acid substrates derived from plasmids carrying the gfp gene under control of the T7 RNA polymerase promoter in the sense or antisense orientation. These substrates included ssRNA, dsRNA (Fig. 1C), ssDNA, dsDNA, and an RNA-DNA hybrid. The enzymes did not cleave ssRNA (Fig. 1D and E), ssDNA, dsDNA, or RNA-DNA hybrids (data not shown). However, RNase3 efficiently cleaved dsRNA at physiological concentrations of KCl (50 to 130 mM) (Fig. 1D). RNase3 activity on dsRNA required divalent cations such as Mg2+ (Fig. 1E) or Mn2+ (Fig. 1F), but Ca2+ did not support activity (Fig. 1G). Cleavage of a dsRNA substrate containing 69-nt and 31-nt ssRNA overhangs (Fig. 1C) was analyzed over time, revealing the accumulation of the two expected longer ssRNAs (cleavage products including the overhangs) and also shorter products (15 to 18 nt) (Fig. 1H) that probably represented unwound dsRNA cleavage products, consistent with earlier reports on exhaustive cleavage of long dsRNA by E. coli RNase III (42). Furthermore, as with E. coli RNase III (8), low concentrations of EtBr (125 μM) inhibited RNase3 cleavage of the dsRNA substrate (Fig. 1I). In all tests, RNase3 cleaved the hairpin RNA (hpRNA) substrate and the dsRNA substrate with similar efficiency (Fig. 1J). In contrast, none of the tested nucleic acid substrates were cleaved using RNase3-Ala37,44, although this mutant bound dsRNA at physiological salt concentrations (50 to 130 mM KCl) (Fig. 1K). Collectively, these data show that RNase3 exhibits the characteristic endonuclease activities of RNase III (13, 49).
p22 causes necrosis and interferes with cross-protection when expressed from a heterologous virus in Nicotiana benthamiana. The genes for p22, RNase3, and p28 of SPCSV were expressed from the 35S promoter-driven infectious cDNA of PVX (6, 65) by mechanical inoculation of the plasmids (PVX-p22, PVX-RNase3, and PVX-p28, respectively) into N. benthamiana plants. Following inoculation with PVX-p28 or PVX-RNase3, the plants developed similar symptoms and high PVX titers (measured by DAS-ELISA; data not shown) in upper noninoculated leaves, as observed with PVX-GF carrying a truncated gfp gene (half of the total gfp length) (65). In contrast, PVX-p22 and the PVX vector carrying the cistron for HC-Pro of PVA (genus Potyvirus) caused necrotic symptoms in systemically infected leaves at 7 days postinoculation (dpi), followed by death of the plant top (PVX-HCpro) or the whole plant (PVX-p22) (Fig. 2). Additionally, the leaves inoculated with PVX-p22 often developed a few necrotic lesions that expanded along the veins (Fig. 2A). Although reasons remain unknown, expression of RSS proteins from heterologous viral genomes such as PVX often results in necrotic symptoms in N. benthamiana (65).
Mutants of the p22 gene in PVX-p22 were produced by introducing a stop codon corresponding to amino acid positions 89, 150, 164, or 181 in p22 (191 aa), and the nucleotide following each new stop codon was deleted to produce a frameshift and ensure that the downstream part of the p22 sequence was not expressed. All mutated PVX-p22 constructs generated a systemic infection, PVX titers (determined by DAS-ELISA), and symptoms similar to those of PVX-GF in N. benthamiana, indicating the loss of necrotizing ability. We used reverse transcription PCR (RT-PCR) cloning, analysis of the amplification products by agarose gel electrophoresis, and sequencing of the p22 insert from the viruses in systemically infected leaves to confirm that lack of necrosis was not associated with loss or partial deletion of the inserts during virus replication (data not shown).
Initial testing for RSS by the SPCSV proteins p22, RNase3, and p28 was carried out using a previously described cross-protection assay in N. benthamiana (45). Leaves are first inoculated with PVX-GF and 3 days later with tobacco mosaic virus engineered with the gfp gene (65), which confines GFP fluorescence to coinoculated leaves (Fig. 2D), because replication of two viruses with partially homologous viral genomes in the same tissue triggers RNA silencing and prevents long-distance transport of the latter inoculated virus, TMV-GFP (45). Detection of a few GFP-expressing lesions on inoculated leaves is expected, since some initially infected cells may be infected only with TMV-GFP. Subsequently, we constructed derivatives of PVX-GF to express p22, RNase3, or p28 (PVX-p22-GF, PVX-RNase3-GF, and PVX-p28-GF, respectively) and coinoculated these constructs with TMV-GFP into N. benthamiana plants, as above, along with the p22 mutants described earlier. PVX-GF and a PVX chimera (PVX-HCpro-GF) expressing the HC-Pro of PVA were used as controls. In plants inoculated with PVX-RNase3-GF, PVX-p28-GF, any of the PVX-GF/p22-mutant constructs, or PVX-GF and subsequently (3 days later) with TMV-GFP, green fluorescence was confined to the inoculated leaves. In contrast, following inoculation with PVX-p22-GF (Fig. 2E and F) or PVX-HCpro-GF and subsequently TMV-GFP, green fluorescence spread to the stem and the upper noninoculated leaves at 6 to 8 dpi. The plants soon developed apical necrosis, however, as expected.
Three replicate experiments consistently showed that SPCSV p22 and PVA HC-Pro induce necrosis in N. benthamiana when expressed from a heterologous viral genome. These proteins interfered with cross-protection, which suggested that they suppress RNA silencing (7). These effects were not evident with SPCSV RNase3 and p28.
p22 suppresses silencing induced by dsRNA in N. benthamiana. An Agrobacterium tumefaciens infiltration assay (28) was used to test whether the SPCSV-encoded proteins p22, p28, and RNase3 could suppress intra- or intercellular RNA silencing triggered by a strong silencing inducer (i.e., dsRNA). In this assay, coexpression of gfp and an inverted-repeat construct of gfp (hpGFP) in agroinfiltrated leaves resulted in degradation of the GFP-specific RNA and suppression of GFP fluorescence, which in turn could be prevented by coexpression of RSS proteins to achieve strong green fluorescence (Fig. 3).
Coinfiltration for expression of GFP, hpGFP, and ?-glucuronidase (GUS) did not result in green fluorescence in the infiltrated areas (Fig. 3), and green fluorescence was not observed following coinfiltration with GFP, hpGFP, and RNase3 (Fig. 3) or p28 (data not shown). These data indicate that RNase3 and p28 do not exhibit RSS activity in this assay. In contrast, infiltration with p22 or HC-Pro resulted in strong fluorescence in the infiltrated leaf areas, commencing at 2 dpi and increasing until 5 dpi. The fluorescence achieved with HC-Pro was initially (5 dpi) (results not shown) stronger than with p22, but later the situation was reversed (10 dpi) (Fig. 3A, compare spot 2 and 4). The spots infiltrated for expression of p22 became chlorotic (10 dpi) (Fig. 3B, spot 2) and later necrotic (15 dpi) (Fig. 3D, spots 4 and 6). Other constructs did not induce chlorosis,; neither did the p22 mutants described earlier, which also did not suppress RNA silencing, as evidenced by the lack of green fluorescence in the coinfiltrated leaves (data not shown). Taken together, these data indicate that full-length p22 suppresses intracellular RNA silencing triggered by a strong RNA silencing inducer.
Cell-to-cell movement of the RNA silencing signal from the cells in which RNA silencing-based RNA degradation takes place can be detected in transgenic GFP-expressing N. benthamiana plants (e.g., line 16c) (47), following silencing of the gfp transgene by ectopic expression of gfp (25). As a result, the silencing signal exiting from the agroinfiltrated area shuts down GFP expression, which is observed as an appearance of a narrow, red border at the edge of the infiltrated spot. Our experiments with the N. benthamiana line 16c revealed short-distance movement of the silencing signal, following infiltration with GFP, hpGFP, and GUS (Fig. 4A) or RNase3 (Fig. 4D) but not HC-Pro (Fig. 4B) or p22 (Fig. 4C), indicating that HC-Pro and p22 interfere with short-distance intercellular signaling for silencing. Furthermore, strong green fluorescence was observed in the tissue infiltrated for expression of HC-Pro or p22, consistent with their ability to suppress RNA silencing (Fig. 4B and 4C).
In two additional experiments, the first fully developed leaves in line 16c plants were coinfiltrated with hpGFP and either p22, HC-Pro, p28, GUS, or RNase3 (five plants per experiment) to observe whether any of these proteins interfered with long-distance signaling for silencing of the gfp gene. Only plants infiltrated with p28, GUS, or RNase3 developed silencing in the upper noninfiltrated leaves (Fig. 4E) within the 2-month period the plants were monitored (data not shown), indicating that expression of either p22 or HC-Pro at the site of RNA silencing induction interfered with systemic silencing.
RNase3 enhances p22-mediated suppression of RNA silencing. Because the sgRNAs for RNase3 and p22 are expressed in a coordinated fashion early in SPCSV infection (32) and because RNase III-type proteins are involved in RNA silencing (3), we tested whether RNase3 influences specifically p22-mediated RSS or if it also affects the RSS functions of an RSS protein encoded by a heterologous virus (PVA HC-Pro). Expression of RNase3 was assayed by Western blotting with anti-RNase3 antibodies (Fig. 5B). Leaves were coinfiltrated with p22, RNase3, HC-Pro, or GUS along with GFP and hpGFP. Whole leaves rather than spots on leaves were coinfiltrated to ensure sufficient RNA yields for analysis of mRNA and siRNA accumulation. At 5 dpi, Northern blot analyses readily detected GFP mRNA in leaves coinfiltrated with GFP, hpGFP, and p22 or HC-Pro but not RNase3, GUS (Fig. 5A), or p28 (data not shown). Figure 5A illustrates that GFP mRNA was already expressed at 3 dpi in leaves coinfiltrated with p22 and RNase3 or with HC-Pro but was detected at a later time in leaves infiltrated with p22 only. These results correspond with the slower appearance of green fluorescence following coinfiltration with p22 than with HC-Pro in the experiments described earlier. RSS mediated by p22 and HC-Pro was also indicated in the increased amounts of RNase3 protein produced in the coinfiltrated leaves, compared to leaves infiltrated for expression of RNase3 alone (Fig. 5B). Importantly, in all experiments, leaves coinfiltrated with p22 and RNase3 showed at least twofold-higher gfp mRNA levels than the leaves coinfiltrated with p22 and GUS. Furthermore, the levels of 21-nt gfp siRNA decreased by approximately fivefold in leaves coinfiltrated with p22 and RNase3, compared with leaves coinfiltrated with p22 and GUS (Fig. 5A).
Experiments were carried out independently at least four times in each of two collaborating laboratories (SLU, Sweden, and University of Helsinki, Finland). Similar results were obtained, showing that even though RNase3 itself exhibits no detectable RSS activity in our assays, it enhances the RSS activity of p22 remarkably, as revealed by the increased accumulation of gfp mRNA and reduced accumulation of the corresponding siRNA.
Enhancement of p22-mediated RSS requires RNase3 endonuclease activity. RNase III can affect gene expression based on its properties as a dsRNA-processing and dsRNA-binding protein (13, 49). The specificity for and recognition of dsRNA by RNase III are associated with the dsRBD. The endoND mediates dsRNA cleavage and includes a highly conserved signature motif, within which a glutamic acid residue corresponding to D44 in SPCSV RNase3 is crucial for cleavage. Indeed, as described earlier, mutant RNase3-Ala37,44 cannot cleave dsRNA. However, the endonuclease activity of RNase III is not essential for substrate binding; on the other hand, dsRNA is cleaved in vitro by a truncated E. coli RNase III lacking the dsRBD (54). The independent activities of the endoND and dsRBD prompted us to investigate the requirement of each for RNase3-mediated enhancement of RSS.
The RNase3 gene fragments (Fig. 6A) encoding endoND (corresponding to aa 1 to 152 of RNase3) or dsRBD (aa 153 to 228) and RNase3-Ala37,44 were cloned into a binary vector, and each vector was used for Agrobacterium coinfiltration with the constructs expressing p22, GFP, and hpGFP. Accumulation of gfp mRNA and siRNA were tested in the high- and low-molecular-weight fractions of total RNA, respectively, at 5 dpi. Northern blot analyses indicated a consistent increase in gfp mRNA levels in the leaves coinfiltrated with p22 and RNase3 (Fig. 6C, top panel, lane 6), HC-Pro (lane 3), or HC-Pro and RNase3 (lane 10), compared with leaves infiltrated with p22 (lane 5) or coinfiltrated with p22 and dsRBD, endoND, or RNase3-Ala37,44 (lanes 7, 8, and 9, respectively). These data suggest that the RNase3 mutants do not enhance p22-mediated RSS. This conclusion was further supported by an analysis of siRNA accumulation. The two size classes of siRNA were electrophoretically separated on a denaturing polyacrylamide gel, and synthetic RNA oligonucleotides (21 nt and 25 nt) were run in the gel as size markers. In the leaves infiltrated for coexpression of p22 along with either endoND, dsRBD, or RNase3-Ala37,44, siRNA accumulated to the same level as in leaves infiltrated with p22 (Fig. 6C, three lower panels; compare lanes 7, 8, or 9 with lane 5), whereas in leaves coinfiltrated with p22 and RNase3 (lane 6), siRNA accumulation was significantly reduced (by up to 10 fold compared with GFP plus hpGFP), as before (Fig. 6C). The enhancing effect of RNase3 on RSS was p22 specific, since the presence or absence of RNase3 had no detectable effect on siRNA accumulation upon infiltration with HC-Pro (Fig. 6C, compare lanes 3 and 10), despite the high amounts of RNase3 and RNase-Ala37,44 proteins expressed in all infiltrated leaves (Fig. 5B). Taken together, these results indicate that the integrity of SPCSV RNase3 and its endonuclease activity are required to enhance RSS activity. Moreover, RNase3 seems to act as an enhancer of RSS activity only with the cognate protein (p22).
p22 shows little effect on transivity of RNA silencing. Transitivity in RNA silencing refers to the generation of secondary siRNAs from sequences located outside the inducer sequence. In plants, transivity can proceed in both 3' and 5' directions (25, 51). To investigate whether p22, RNase3, or their coexpression affects transivity in silencing of gfp mRNA, the membranes previously tested with the entire GFP probe were stripped and hybridized with two nonoverlapping probes (Fig. 6B). Probe 5'-GFP encompasses two-thirds of the gfp sequence and corresponds to the hpGFP (Fig. 1B) used as the gfp silencing inducer, whereas probe 3'-GFP encompasses the 3'-proximal third of the gfp sequence (Fig. 6B). The 3'-GFP probe was expected to reveal the secondary siRNA resulting from transivity. Northern blot hybridization and subsequent quantification showed similar relative siRNA accumulation in different samples, regardless of the probe used for siRNA detection (Fig. 6C, bottom two panels). Thus, the expression of p22, RNase3, or their coexpression had no detectable influence on transivity.
RNase3 cleaves 25-nt synthetic dsRNA but not the 21-nt and 23-nt siRNAs generated during gfp mRNA silencing. Because the endonuclease activity of RNase3 was needed for the enhancement of p22-mediated RSS activity, we hypothesized that RNase3 might cleave the two classes of siRNA. The minimum dsRNA substrate length for E. coli RNase III is 20 nt, which is approximately equivalent to two turns of A-form dsRNA (15) and close to the two size classes of siRNA generated from gfp mRNA in N. benthamiana (21 and 23 nt) (Fig. 6C). Therefore, the activity of purified recombinant RNase3 was tested on synthetic dsRNAs (21 nt and 25 nt) analogous to siRNA with 2-nt 3' overhangs and a 3' hydroxyl group, as well as on gfp siRNA extracted from leaves. Controls included siRNA denatured prior to the assay, the mutant RNase3-Ala37,44, and E. coli RNase III (Ambion). Neither RNase3 nor RNase-Ala37,44 cleaved the 21-nt or 23-nt siRNAs extracted from plants (Fig. 7A, lanes 6 to 9), in contrast to E. coli RNase III (Fig. 7A, lanes 10 and 11). However, RNase3 cleaved the synthetic 25-nt dsRNA but not the shorter synthetic 21-nt dsRNA (Fig. 7C), even though each carries 2-nt 3' overhangs as in siRNAs.
We also tested whether the 21-nt or 25-nt synthetic ssRNA or dsRNA could compete with or inhibit RNase3 cleavage of long dsRNA substrates (717 nt), but no such evidence was found in a time course analysis using a 600-fold molar excess of the short RNAs relative to long dsRNA (Fig. 7B). The finding that RNase3 did not cleave siRNA did not, however, exclude the possibility that it binds siRNA. To test this possibility, we performed gel-shift experiments with recombinant RNase3 and RNase3-Ala37,44. Both enzymes bound radiolabeled synthetic 21-nt and 25-nt RNAs, as indicated by band shifting (Fig. 7D; also Fig. 7C, lane 13.5 μM RNase3).
DISCUSSION
RNase III enzymes play various roles in RNA metabolism and gene expression and are divided into three classes according to their structure (9). SPCSV RNase III (RNase3) studied here and E. coli RNase III belong to the simplest class (class 1), containing a single endoND and dsRBD. Our data show that RNase3 enhances the suppression of RNA silencing mediated by a cognate RSS protein, p22. The RNase IIIs of classes 2 and 3 contain duplicate endoND domains and are distinguished by their N-terminal extensions. Drosha, a class 2 RNase III, is involved in maturation of microRNAs (miRNAs) that regulate animal development (34), whereas Dicer belongs to class 3 and is pivotal in the initiation and amplification of RNA silencing via its participation in RISC, cleaving silencing-inducing dsRNA molecules and homologous RNAs to 21- to 24-nt siRNAs (3, 23, 57). Taken together, the role of RNase III in the initiation and amplification of RNA silencing is well established (57), but our data now show that some RNase III enzymes interfere with this process. It therefore seems that different classes of RNase III endoribonucleases play different roles in RNA silencing.
The viral RSS proteins known to date, including p22 identified here, show no significant similarity to cellular proteins (40), in contrast to SPCSV RNase3 that is similar to the putative RNase III class 1 proteins of Arabidopsis thaliana (32) and rice Oryza sativa (BAD36550, for which functions have not been reported. Plants encode multiple, homologous Dicer-like RNase III enzymes. Four homologs are found in rice (a monocot) and A. thaliana (a dicot). Three of the A. thaliana homologs are implicated in different RNA silencing pathways: DCL1 in miRNA, DCL2 in antiviral siRNA, and DCL3 in endogenous siRNA biogenesis (64). Intriguingly, the genomes of SPCSV (32) and other plant viruses of the family Closteroviridae may have incorporated host genes, as evidenced by the genes in viral genomes that encode homologs of plant heat shock proteins and are required for virulence (29). The SPCSV RNase3 gene may likewise be of host origin, exemplifying a novel strategy by which viruses employ host proteins to enhance virulence or suppression of various types of host defenses (4), in this case via interference with RNA silencing. The finding that p22 suppresses RNA silencing without RNase3 is consistent with this theory and may also suggest that putative host factors can fulfill the function of RNase3 (although less efficiently). This scenario for the origin of RNase3, together with the fundamental role of RNA silencing in gene regulation in eukaryotic organisms (10), implies that some plant RNase IIIs may be involved in RSS. This would provide another mechanism for the negative regulation of RNA silencing in addition to that described for A. thaliana DCL1, the RNase III that positively regulates miRNA biogenesis but is itself controlled by miRNA-mediated RNA silencing (63).
To date, SPCSV is the first virus for which two independent proteins have been found to cooperatively control RSS functions. Very recently, another study has shown that Citrus tristeza virus belonging to the same family but a different genus (Closterovirus) encodes three RSS proteins, but their cooperative functions in RSS have not been studied (38). Results from several experimental approaches employed in this study consistently indicated that although p22 alone can efficiently suppress RNA silencing induced by dsRNA of various types, the endonuclease activity of RNase3 significantly enhances p22-mediated RSS, resulting in a remarkable reduction of siRNA accumulation compared with that observed when p22 is expressed alone.
The substitutions E37A and D44A made to RNAse3 in our study to abolish the endonuclease activity of the protein were based on the crystal structure and functional model of class 1 RNase III of Aquifex aeolicus and E. coli (5). Two hydrolysis events catalyzed by RNase III cleave dsRNA at sites two bases apart. The first RNA-cleaving site of A. aeolicus RNase III consists of residues E37 (corresponding to E37 in RNase3 of SPCSV) and E64, each from a different RNase III molecule of the dimer. The second RNA-cleaving site is composed of residues D44 (as in RNase3) and E110 (5). Site-directed amino acid substitutions E38V and D45A introduced to the E. coli RNase III (which correspond to E37 and D44 of RNase3) abolished its endonuclease activity in vivo but did not affect the RNA-binding ability (5). Similarly, RNase3 mutated at positions 37 (E37A) and 44 (D44A) bound long and short (21 to 25nt) dsRNA substrates efficiently but was devoid of endonuclease activity. These data suggest that no other known functions of RNase III (dimerization, for instance) were affected by the mutations introduced to RNase3. In support of our data, a mutant of E. coli RNase III that binds dsRNA but lacks the RNA cleavage activity exhibits only weak RSS functions in an agroinfiltration assay (36).
The mechanism of the coordinated functions of p22 and RNase3 in RSS, however, cannot yet be fully elucidated. Initial cleavage of dsRNA was probably not affected because Northern blots revealed no signals for the inducing dsRNA (hpGFP) in samples undergoing RSS. Transitivity was obviously not affected by p22 and/or RNase3, implying that RNA-dependent RNA polymerase-mediated amplification of the silencing signal was unlikely affected (14, 25, 41). The p21 RSS protein of Beet yellows virus (BYV; genus Closterovirus) (46) was recently shown to bind siRNA and miRNA, indicating that p21 may sequester siRNA such that these RNAs are unavailable to RISC (11). This hypothesis awaits testing with SPCSV p22, which is recalcitrant to purification (as initially was the case for BYV p21) (46). However, SPCSV p22 and BYV p21 show no significant sequence similarity and their functions differ, as p21 expression does not reduce siRNA accumulation or induce chlorosis in N. benthamiana leaves (46), in contrast to p22 studied here. Therefore, RSS mediated by p22 and enhanced by RNase3 may proceed via a novel mechanism. We favor a hypothesis whereby the specificity of RNase3 is modulated in the cellular environment via interaction with RISC and/or p22, such that RNase3 may cleave siRNA bound by RISC and/or p22; the efficiency by which RISC targets new homologous RNAs for cleavage is thereby reduced. This hypothesis is supported by studies with Saccharomyces cerevisiae RNase IIIs, the specificity of which is altered or endonuclease activity enhanced by interaction with specific proteins or protein complexes (19, 52). Additional RSS may result from interference of the viral RNase III with the loading of siRNAs into RISC by cellular RNase IIIs (44) or from interference of p22 and/or RNase3 with the incorporation of proteins or protein complexes into RISC, a process with which the P1/HC-Pro polyprotein of potyviruses has been proposed to interfere (11).
SPCSV is the first RNA virus shown to encode an RNase III enzyme. A few large, dsDNA-containing viruses infecting fish, insects, or algae encode putative RNase III proteins (27, 53, 56, 59), and the enzymatic activity of RNase III encoded by Paramecium bursaria Chlorella virus 1 has been characterized (66). Still, the functions of these enzymes in the viral infection cycle are unknown. Taken together, our data suggest an entirely novel mechanism in viral pathogenesis and virulence and reveal a role for RNase III class 1 enzymes in RNA silencing, an area in which they were previously not known to be involved.
ACKNOWLEDGMENTS
We are grateful to Anna Germundsson and Marjo Ala-Poikela for collaboration, David Baulcombe for providing line 16c, and Andrey Zamyatnin for help with preparing figures.
We gratefully acknowledge financial support from the European Union Programme for International Cooperation with Developing Countries (FP5 INCO-DEV, project ICA4-CT-2000-3007) and the Academy of Finland (grant 1102134).
J.F.K. and E.I.S. contributed equally to the study.
REFERENCES
Ahlquist, P. 2002. RNA-dependent RNA polymerases, viruses, and RNA silencing. Science 296:1270-1273.
Anandalakshmi, R., G. J. Pruss, X. Ge, R. Marathe, A. C. Mallory, T. H. Smith, and V. B. Vance. 1998. A viral suppressor of gene silencing in plants. Proc. Natl. Acad. Sci. USA 95:1379-1384.
Bernstein, E., A. A. Caudy, S. M. Hammond, and G. J. Hannon. 2001. Role for a bidentate ribonuclease in the initiation step of RNA interference. Nature 409:363-366.
Bilgin, D. D., Y. Liu, M. Schiff, and S. P. Dinesh-Kumar. 2003. P58IPK, a plant ortholog of double-stranded RNA-dependent protein kinase PKR inhibitor, functions in viral pathogenesis. Dev. Cell 4:651-661.
Blaszczyk, J., J. E. Tropea, M. Bubunenko, K. M. Routzahn, D. S. Waugh, D. L. Court, and X. Ji. 2001. Crystallographic and modeling studies of RNase III suggest a mechanism for double-stranded RNA cleavage. Structure 9:1225-1236.
Boevink, P., S. Santa Cruz, C. Hawes, N. Harris, and K. J. Oparka. 1996. Virus-mediated delivery of the green fluorescent protein to the endoplasmic reticulum of plant cells. Plant J. 10:935-941.
Brigneti, G., O. Voinnet, W. X. Li, L. H. Ji, S. W. Ding, and D. C. Baulcombe. 1998. Viral pathogenicity determinants are suppressors of transgene silencing in Nicotiana benthamiana. EMBO J. 17:6739-6746.
Calin-Jageman, I., A. K. Amarasinghe, and A. W. Nicholson. 2001. Ethidium-dependent uncoupling of substrate binding and cleavage by Escherichia coli ribonuclease III. Nucleic Acids Res. 29:1915-1925.
Carmell, M. A., and G. J. Hannon. 2004. RNase III enzymes and their role in initiation of gene silencing. Nat. Struct. Mol. Biol. 11:214-218.
Carrington, J. C., and V. Ambros. 2003. Role of microRNAs in plant and animal development. Science 301:336-338.
Chapman, E. J., A. I. Prokhnevsky, K. Gopinath, V. V. Dolja, and J. C. Carrington. 2004. Viral RNA silencing suppressors inhibit the microRNA pathway at an intermediate step. Genes Dev. 18:1179-1186.
Cogoni, C., and G. Macino. 2000. Post-transcriptional gene silencing across the kingdoms. Curr. Opin. Genet. Dev. 10:638-643.
Conrad, C., and R. Rauhut. 2002. Ribonuclease III: new sense from nuisance. Int. J. Biochem. Cell Biol. 34:116-129.
Dalmay, T., A. Hamilton, S. Rudd, S. Angell, and D. C. Baulcombe. 2000. An RNA-dependent RNA polymerase gene in Arabidopsis is required for posttranscriptional gene silencing mediated by a transgene but not a virus. Cell 101:543-553.
Dunn, J. J. 1982. Ribonuclease III, p. 485-499. In B. D. Boyer (ed.), The enzymes, vol. 15. Academic Press, New York, N.Y.
Dunoyer, P., C. H. Lecellier, E. A. Parizotto, C. Himber, and O. Voinnet. 2004. Probing the microRNA and small interfering RNA pathways with virus-encoded suppressors of RNA silencing. Plant Cell 16:1235-1250.
Fire, A., S. Xu, M. K. Montgomery, S. A. Kostas, S. E. Driver, and C. C. Mello. 1998. Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 391:806-811.
Gibson, R. W., I. Mpembe, T. Alicai, E. E. Carey, R. O. M. Mwanga, S. E. Seal, and H. J. Vetten. 1998. Symptoms, aetiology and serological analysis of sweet potato virus disease in Uganda. Plant Pathol. 47:95-102.
Giorgi, C., A. Fatica, R. Nagel, and I. Bozzoni. 2001. Release of U18 snoRNA from its host intron requires interaction of Nop1p with the Rnt1p endonuclease. EMBO J. 20:6856-6865.
Gitlin, L., S. Karelsky, and R. Andino. 2002. Short interfering RNA confers intracellular antiviral immunity in human cells. Nature 418:430-434.
Hamilton, A., O. Voinnet, L. Chappell, and D. Baulcombe. 2002. Two classes of short interfering RNAs in RNA silencing. EMBO J. 21:4671-4679.
Hamilton, A. J., and D. C. Baulcombe. 1999. A species of small antisense RNA in posttranscriptional gene silencing in plants. Science 286:950-952.
Hammond, S. M., E. Bernstein, D. Beach, and G. Hannon. 2000. An RNA-directed nuclease mediates post-transcriptional gene silencing in Drosophila cell extracts. Nature 404:293-296.
Hannon, G. J. 2002. RNA interference. Nature 418:244-251.
Himber, C., P. Dunoyer, G. Moissiard, C. Rizenthaler, and O. Voinnet. 2003. Transivity-dependent and -independent cell-to-cell movement of RNA silencing. EMBO J. 22:4523-4533.
Hutvágner, G., J. McLachlan, A. E. Pasquinelli, é. Bálint, T. Tuschl, and P. D. Zamore. 2001. A cellular function for the RNA-interference enzyme Dicer in the maturation of the let-7 small temporal RNA. Science 293:834-838.
Jakob, N. J., K. Muller, U. Bahr, G. Darai. 2001. Analysis of the first complete DNA sequence of an invertebrate iridovirus: coding strategy of the genome of Chilo iridescent virus. Virology 286:182-196.
Johansen, L. K., and J. C. Carrington. 2001. Silencing on the spot. Induction and suppression of RNA silencing in the Agrobacterium-mediated transient expression system. Plant Physiol. 126:930-938.
Karasev, A. V. 2000. Genetic diversity and evolution of closteroviruses. Annu. Rev. Phytopathol. 38:293-324.
Karyeija, R. F., J. F. Kreuze, R. W. Gibson, and J. P. T. Valkonen. 2000. Synergistic interactions of a potyvirus and a phloem-limited crinivirus in sweet potato plants. Virology 269:26-36.
Kasschau, K. D., and J. C. Carrington. 1998. A counterdefensive strategy of plant viruses: suppression of posttranscriptional gene silencing. Cell 95:461-470.
Kreuze, J. F., E. I. Savenkov, and J. P. T. Valkonen. 2002. Complete genome sequence and analyses of the subgenomic RNAs of Sweet potato chlorotic stunt virus reveal several new features for the genus Crinivirus. J. Virol. 76:9260-9270.
Lakatos, L., G. Szittya, D. Silhavy, and J. Burgyán. 2004. Molecular mechanism of RNA silencing suppression mediated by p19 protein of tombusviruses. EMBO J. 23:876-884.
Lee, Y., C. Ahn, J. Han, H. Choi, J. Kim, J. Yim, J. Lee, P. Provost, O. R?dmark, S. Kim, and N. Kim. 2003. The nuclear RNase III Drosha initiates microRNA processing. Nature 425:415-419.
Li, H., W. X. Li, and S. W. Ding. 2002. Induction and suppression of RNA silencing by an animal virus. Science 296:1319-1321.
Lichner, Z., D. Silhavy, and J. Burgyán. 2003. Double-stranded RNA-binding proteins could suppress RNA interference-mediated antiviral defences. J. Gen. Virol. 84:975-980.
Llave, C., Z. Xie, K. D. Kasschau, and J. C. Carrington. 2002. Cleavage of Scarecrow-like mRNA targets directed by a class of Arabidopsis miRNA. Science 297:2053-2060.
Lu, R., A. Folimonov, M. Shintaku, W. X. Li, B. W. Falk, W. O. Dawson, and S. W. Ding. 2004. Three distinct suppressors of RNA silencing encoded by a 20-kb viral RNA genome. Proc. Natl. Acad. Sci. USA 101:15742-15747.
Mallory, A. C., L. Ely, T. H. Smith, R. Marathe, R. Anandalakshmi, M. Fagard, H. Vaucheret, G. Pruss, L. Bowman, and V. B. Vance. 2001. HC-Pro suppression of gene silencing eliminates the small RNAs but not the transgene methylation or the mobile signal. Plant Cell 13:571-583.
Moissiard, G., and O. Voinnet. 2004. Viral suppression of RNA silencing in plants. Mol. Plant Pathol. 5:71-82.
Mourrain, P., C. Beclin, T. Elmayan, F. Feuerbach, C. Godon, J. B. Morel, D. Jouette, A. M. Lacombe, S. Nikic, N. Picault, K. Remoue, M. Sanial, T. A. Vo, and H. Vaucheret. 2000. Arabidopsis SGS2 and SGS3 genes are required for posttranscriptional gene silencing and natural virus resistance. Cell 101:533-542.
Nicholson, A. W. 1996. Structure, reactivity, and biology of double-stranded RNA. Prog. Nucleic Acid Res. Mol. Biol. 52:1-65.
Palaqui, J. C., T. Elmayan, J. M. Pollien, and H. Vaucheret. 1997. Systemic acquired silencing: transgene-specific post-transcriptional silencing is transmitted by grafting from silenced stocks to non-silenced scions. EMBO J. 16:4738-4745.
Pham, J. W., J. L. Pellino, Y. S. Lee, R. W. Carthew, and E. J. Sontheimer. 2004. A Dicer-2-dependent 80S complex cleaves targeted mRNAs during RNAi in Drosophila. Cell 117:83-94.
Ratcliff, F. G., S. A. MacFarlane, and D. C. Baulcombe. 1999. Gene silencing without DNA. RNA-mediated cross-protection between viruses. Plant Cell 11:1207-1216.
Reed, J. C., K. D. Kasschau, A. I. Prokhnevsky, K. Gopinath, G. P. Pogue, J. C. Carrington, and V. V. Dolja. 2003. Suppressor of RNA silencing encoded by Beet yellows virus. Virology 306:203-209.
Ruiz, M. T., O. Voinnet, and D. C. Baulcombe. 1998. Initiation and maintenance of virus-induced gene silencing. Plant Cell 10:937-946.
Sambrook, J., and D. W. Russel. 2001. Molecular cloning: a laboratory manual, 3rd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Saunders, L. R., and G. N. Barber. 2003. The dsRNA binding protein family: critical roles, diverse cellular functions. FASEB J. 17:961-983.
Savenkov, E. I., and J. P. T. Valkonen. 2001. Potyviral helper-component proteinase expressed in transgenic plants enhances titres of Potato leaf-roll virus but does not alleviate its phloem limitation. Virology 283:285-293.
Sijen, T., J. Fleenor, F. Simmer, K. L. Thijssen, S. Parrish, L. Timmons, R. H. A. Plasterk, and A. Fire. 2001. On the role of RNA amplification in dsRNA-triggered RNA silencing. Cell 107:465-476.
Spasov, K., L. I. Perdomo, E. Evakine, and R. N. Nazar. 2002. RAC protein directs the complete removal of the 3' external transcribed spacer by the Pac nuclease. Mol. Cell 9:433-437.
Stasiak, K., M. V. Demattei, B. A. Federici, and Y. Bigot. 2000. Phylogenetic position of the Diadromus pulchellus ascovirus DNA polymerase among viruses with large double-stranded DNA genomes. J. Gen. Virol. 81:3059-3072.
Sun, W., E. J. Jun, and A. W. Nicholson. 2001. Intrinsic double-stranded-RNA processing activity of Escherichia coli ribonuclease III lacking the dsRNA-binding domain. Biochemistry 40:14976-14984.
Tabara, H., M. Sarkissian, W. G. Kelly, J. Fleenor, A. Grishok, L. Timmons, A. Fire, and C. C. Mello. 1999. The rde-1 gene, RNA interference, and transposon silencing in C. elegans. Cell 99:123-132.
Tidona, C. A., and G. Darai. 1997. The complete DNA sequence of lymphocystis disease virus. Virology 230:207-216.
Tijsterman, M., and R. H. A. Plasterk. 2004. Dicers at RISC: the mechanism of RNAi. Cell 117:1-3.
Vancanneyt, G., R. Schmidt, A. O'Connor-Sanchez, L. Willmitzer, and M. Rocha-Sosa. 1990. Construction of an intron-containing marker gene: splicing of the intron in transgenic plants and its use in monitoring early events in Agrobacterium-mediated plant transformation. Mol. Gen. Genet. 220:245-250.
Van Etten, J. L., and R. H. Meints. 1999. Giant viruses infecting algae. Annu. Rev. Microbiol. 53:447-494.
Voinnet, O., and D. C. Baulcombe. 1997. Systemic signaling in gene silencing. Nature 389:553.
Voinnet, O., C. Lederer, and D. C. Baulcombe. 2000. A viral movement protein prevents spread of the gene silencing signal in Nicotiana benthamiana. Cell 103:157-167.
Volpe, T. A., C. Kidner, I. M. Hall, G. Teng, S. I. S. Grewal, and R. A. Martienssen. 2002. Regulation of heterochromatic silencing and histone H3 lysine-9 methylation by RNAi. Science 297:1833-1837.
Xie, Z., K. D. Kasschau, and J. C. Carrington. 2003. Negative feedback regulation of Dicer-like1 in Arabidopsis by microRNA-guided mRNA degradation. Curr. Biol. 13:784-789.
Xie, Z., L. K. Johansen, A. M. Gustafson, K. D. Kasschau, A. D. Lellis, D. Zilberman, S. E. Jacobsen, and J. C. Carrington. 2004. Genetic and functional diversification of small RNA pathways in plants. PLoS Biol. 2:642-652.
Yelina, N. E., E. I. Savenkov, A. G. Solovyev, S. Y. Morozov, and J. P. T. Valkonen. 2002. Long-distance movement, virulence, and RNA silencing suppression controlled by a single protein in hordei- and potyviruses: complementary functions between virus families. J. Virol. 76:12981-12991.
Zhang, Y., I. Calin-Jageman, J. R. Gurnon, T. J. Choi, B. Adams, A. W. Nicholson, and J. L. Van Etten. 2003. Characterization of a chlorella virus PBCV-1 encoded ribonuclease III. Virology 317:73-83.(Jan F. Kreuze , Eugene I.)