Viral RNase3 co-localizes and interacts with the antiviral defense protein SGS3 in plant cells

RESEARCH ARTICLE Viral RNase3 Co-Localizes and Interacts with the Antiviral Defense Protein SGS3 in Plant Cells Isabel Weinheimer 1¤a , Tuuli Haikonen 1¤b , Marjo Ala-Poikela 1¤c , Mirko Moser 2 , Janne Streng 1 , Minna-Liisa Rajamäki 1 *, Jari P. T. Valkonen 1 1Department of Agricultural Sciences, University of Helsinki, Helsinki, Finland, 2Foundation Edmund Mach, San Michele, Italy ¤a Current Address: Roche Diagnostics GmbH, Penzberg, Germany ¤b Current Address: Natural Resources Institute Finland (Luke), Horticultural production research, Piikkiö, Finland ¤c Current Address: Department of Plant, Soil and Entomological Sciences, University of Idaho, Moscow, Idaho, United States of America *[email protected] Abstract Sweet potato chlorotic stunt virus (SPCSV; family Closteroviridae) encodes a Class 1 RNase III endoribonuclease (RNase3) that suppresses post-transcriptional RNA interference (RNAi) and eliminates antiviral defense in sweetpotato plants (Ipomoea batatas). For RNAi suppression, RNase3 cleaves double-stranded small interfering RNAs (ds-siRNA) and long dsRNA to fragments that are too short to be utilized in RNAi. However, RNase3 can suppress only RNAi induced by sense RNA. Sense-mediated RNAi involves host suppressor of gene silencing 3 (SGS3) and RNA–dependent RNA polymerase 6 (RDR6). In this study, subcellular localization and host interactions of RNase3 were studied in plant cells. RNase3 was found to interact with SGS3 of sweetpotato and Arabidopsis thaliana when expressed in leaves, and it localized to SGS3/RDR6 bodies in the cytoplasm of leaf cells and protoplasts. RNase3 was also detected in the nucleus. Co-expression of RNase3 and SGS3 in leaf tissue enhanced the suppression of RNAi, as compared with expression of RNase3 alone. These results suggest additional mechanisms needed for efficient RNase3-mediated suppression of RNAi and provide new information about the subcellular context and phase of the RNAi pathway in which RNase3 realizes RNAi suppression. Introduction Sweet potato chlorotic stunt virus (SPCSV, genus Crinivirus; Closteroviridae) predisposes sweetpotato (Ipomoea batatas) to heavy yield losses by eliminating the basal antiviral defense based on post-transcriptional RNA interference (RNAi), which also renders plants susceptible to other unrelated viruses [1]. The dsRNA-specific Class 1 RNase III endoribonuclease (RNase3) encoded by SPCSV is sufficient for eliminating antiviral defense when expressed from a transgene in sweetpotato plants [2]. RNase3 can suppress sense RNA–mediated RNAi PLOS ONE | DOI:10.1371/journal.pone.0159080 July 8, 2016 1/16 a11111 OPEN ACCESS Citation: Weinheimer I, Haikonen T, Ala-Poikela M, Moser M, Streng J, Rajamäki M-L, et al. (2016) Viral RNase3 Co-Localizes and Interacts with the Antiviral Defense Protein SGS3 in Plant Cells. PLoS ONE 11 (7): e0159080. doi:10.1371/journal.pone.0159080 Editor: Rui Lu, Louisiana State University, UNITED STATES Received: March 8, 2016 Accepted: June 27, 2016 Published: July 8, 2016 Copyright: © 2016 Weinheimer et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Funding: This work was funded by the Academy of Finland (grants 1134335, 1253126 and 1276136 to JPTV), Viikki Doctoral Program in Molecular Biosciences (to IW), and the Finnish Doctoral Program in Plant Sciences (to TH). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. (gene co-suppression) but not RNAi induced by antisense RNA or dsRNA [2–4]. RNase3 can process long dsRNA and double-stranded short interfering RNAs (siRNA) in a lengthand sequence-independent manner [2,3], but certain structural anti-processing determinants (asymmetric bulges) typical for microRNA (miRNA) duplexes may interfere with processing [5]. Taken together, RNase3 can potentially target the dsRNA substrates of the Dicer-like dsRNA-specific endoribonucleases (DCL1-4), as well as the 21-, 22-, and 24-nucleotide (nt) siRNA duplexes produced by DCLs in plants and which guide the RNA-induced silencing complex (RISC) to target and cleave homologous RNAs [6]. RNA-dependent RNA polymerases (RDRs) are important in sense-mediated RNAi. They can copy ssRNA to dsRNA, which is subsequently processed to siRNA by DCLs. In this process, secondary siRNAs are produced also from regions outside the primary targeted area, which is called transitivity and enhances antiviral RNAi [7,8]. Previous studies found no influence of RNase3 on transitivity [3]. SGS3 is another plant protein involved in antiviral defense, but the mechanism is not well understood. For example, a lowered level of SGS3 mRNA enhances susceptibility of Arabidopsis thaliana (L.) Heynh. to Cucumber mosaic virus (genus Cucumovirus) but not Turnip vein clearing virus (genus Tobamovirus)orTurnip mosaic virus (genus Potyvirus)[9,10]. In potyvirus infection, SGS3-silencing reduces accumulation of viral RNA, as reported with Potato virus A and Soybean mosaic virus strain G7 [11,12]. The coordinated functions of SGS3 with RDR6 are pivotal in trans-acting siRNA (tasiRNA) pathways that regulate plant gene expression. The first step in the tasiRNA pathway is the miRNA-programmed cleavage of tasiRNA gene (TAS) transcripts, which are stabilized by SGS3 [13]. For example, TAS3 is conserved among plant species [14] and has two miRNA390 target sites, of which the 3’site is recognized and cleaved specifically by RISC containing the RNase H–like endoribonuclease Argonaute 7 (AGO7). Subsequently, RDR6 converts the 5’- cleavage fragment of TAS3 transcripts to dsRNA in SGS3/RDR6 bodies (also called siRNA bodies), and DCL4 processes the dsRNA to 21-nt siRNAs. Consistent with these functions, AGO7 co-localizes with the SGS3/RDR6 bodies in the cytoplasm [15,16]. Involvement of RDR6 and SGS3 in RNAi suggests that plant viruses may have evolved mechanisms to interfere with their functions. Indeed, the protein P6 of Rice yellow stunt virus (genus Nucleorhabdovirus) interacts with RDR6 of rice (Oryza sativa) and interferes with systemic silencing of RNAi, but cannot suppress RNAi locally in the leaves in which silencing is induced [17]. Therefore, it differs from RNase3 that suppresses RNAi locally [2]. On the other hand, the triple gene block protein (TGBp1) of Plantago asiatica mosaic virus (genus Potexvirus) interacts with both RDR6 and SGS3 to mediate their aggregation and inhibits SGS3/ RDR6-dependent dsRNA synthesis and tasiRNA accumulation [18]. Similarly, the P protein of Lettuce necrotic yellows virus (genus Cytorhabdovirus) interacts with both RDR6 and SGS3 in large protein aggregates, which inhibits RNAi amplification [19]. Furthermore, the RNAi-suppressing proteins V2 of Tomato yellow leaf curl virus (ssDNA genome, genus Begomovirus) [20], p2 of Rice stripe virus (negative-sense ssRNA genome; genus Tenuivirus)[21] and VPg of Potato virus A (ssRNA genome, genus Potyvirus)[11] interact with the SGS3 proteins of tomato (Solanum lycopersicum L.), rice (Oryza sativa L.), and potato (Solanum tuberosum L.), respectively. RNase3 is a unique suppressor interfering with RNAi in an endoribonuclease activitydependent manner. However, little is known about the subcellular localization and host interactions of RNase3 in plant cells. RNase3 interferes with sense-mediated RNAi but is unable to suppress RNAi induced by hairpin RNA [2], similar to the TGBp1, P, V2, p2 and VPg proteins mentioned above [11,18–21]. Therefore, the aim of this study was to examine possible interactions of RNase3 with SGS3 and RDR6 and interference with the RNAi pathway involving these host proteins. RNase3 Interacts with SGS3 PLOS ONE | DOI:10.1371/journal.pone.0159080 July 8, 2016 2/16 Results Subcellular localization of RNase3 in cytoplasm and nucleus Agroinfiltration was used to co-express RNase3 (RNase3-dsRED; red fluorescent protein fused to the C-terminus) in leaves of Nicotiana benthamiana Domin along with fibrillarin (Fib2-GFP, green fluorescent protein fused to the C-terminus), which is a major nucleolar protein also present in Cajal bodies [22,23]. Confocal microscopy of epidermal cells in the infiltrated areas at 2 days post-infiltration (dpi) revealed dsRED signals in punctate bodies in the cytoplasm (arrowheads, Fig 1A and 1C) and in the nucleus, and also in subnuclear bodies that differed from Cajal bodies (arrow in Fig 1A and 1C). The nucleolus and Cajal bodies did not show pronounced dsRED signals. Co-localization of RNase3 and RDR6 The punctate RNase3-containing bodies detected in N.benthamiana epidermal cells resembled the SGS3/RDR6 bodies detected in A.thaliana and Nicotiana tabacum L. [15,16]. Therefore, RNase3-dsRED was co-expressed with the RDR6 of A.thaliana or N.tabacum,bothwithGFP fused to the C-terminus (AtRDR6-GFP and NtRDR6-GFP, respectively), in leaves of N. benthamiana by agroinfiltration. Confocal microscopy showed that signals of AtRDR6-GFP (Fig 2A) and NtRDR6-GFP (Fig 2B) were observed in punctate cytoplasmic bodies that co-localized with the RNase3-dsRED-positive punctate bodies. Co-localization of AtRDR6 and NtRDR6 with RNase3 was also observed in protoplasts prepared from the agroinfiltrated leaf tissues (S1 Fig). Substitution of Asn37 and Glu44 with alanine in the catalytic site of RNase3 results in an RNase3-Ala mutant lacking dsRNA cleavage activity [3]. Localization studies carried out with RNase3-Ala fused with dsRED revealed signals in similar punctate bodies and similar co-localization with RDR6 (S2 Fig) as found with RNase3 (Fig 2). The RNase3-SGS3 complex co-localizes with SGS3/RDR6 bodies Co-localization of RNase3 and RDR6 prompted us to test the possible interactions of RNase3 with RDR6 or SGS3, which was done using bimolecular fluorescence complementation (BiFC) Fig 1. Subcellular localization of RNase3 in epidermal cells of Nicotiana benthamiana following expression by agroinfiltration. (a) RNase3-dsRED (red signals) was detected in cytoplasmic punctate bodies by confocal microscopy at 2 dpi. A few of the many bodies are pointed out with arrowheads. It was also present in the nucleus and subnuclear bodies (arrow). (b) Fibrillarin of A.thaliana (Fib2) was expressed as a fusion with GFP (green) and used as a marker for nucleolus (N) and Cajal bodies (C). (c) Merged image revealed localization of RNase3 in subnuclear bodies (arrow) other than nucleolus or Cajal bodies. Scale bars, 10 μm. doi:10.1371/journal.pone.0159080.g001 RNase3 Interacts with SGS3 PLOS ONE | DOI:10.1371/journal.pone.0159080 July 8, 2016 3/16 [24,25]. AtRDR6 and AtSGS3 of A.thaliana and RNase3 were introduced to BiFC vectors and expressed in translational fusion with the Nor C-terminal halves of yellow fluorescent protein (YFP; YN or YC, respectively). BiFC assay of these proteins with eIF(iso)4E served as a negative control (Fig 3, right panel). Co-expression of AtRDR6 with AtSGS3 served as a positive interaction control in the experiments and resulted in yellow fluorescence confined to cytoplasmic punctate bodies (Fig 3A), as expected [15]. RNase3 did not interact with AtRDR6 (Fig 3C). In contrast, interaction of RNase3 with AtSGS3 was readily observed in cytoplasmic punctate bodies (Fig 3E). Similarly, RNase3-Ala interacted with AtSGS3 (Fig 3G). The SGS3 coding sequence of sweetpotato (cv. Huachano) (designated IbSGS3) was cloned utilizing the transcriptome data of sweetpotato cv. Xushu18 available in NCBI Sequence Read Archive (accession number SRX090758; [26]) (see Methods and S3 Fig) and introduced into the BiFC vectors. IbSGS3 interacted with RNase3 in punctate bodies (Fig 3I) similar to those observed with the AtSGS3-RNase3 interaction (Fig 3E). Co-expression of YN-RNase3, AtSGS3-YC, and AtRDR6-mRFP revealed that cytoplasmic punctate bodies expressing BiFC signals (green) of the RNase3-SGS3 interaction co-localized with the signals (red) of RDR6-containing bodies (Fig 4). Taken together, the results indicated that RNase3 interacts with SGS3 and hence co-localizes with SGS3/RDR6 bodies. The catalytic site mutations in RNase3-Ala did not affect these interactions and subcellular localization. Co-expression of RNase3 and IbSGS3 enhances suppression of RNAi RNase3 can suppress sense-mediated RNAi, e.g., when silencing of the constitutively expressed gfp transgene is induced in N.benthamiana 16c following gfp overexpression by agroinfiltration [2]. Influence of IbSGS3, RNase3 and RNase3-Ala (a mutant of RNase3 debilitated for Fig 2. Co-localization of RNase3 and RDR6 in epidermal cells of N.benthamiana following co-expression by agroinfiltration. The red signals of RNase3-dsRED and green signals of (a) AtRDR6-GFP (A.thaliana) and (b) NtRDR6-GFP (N.tabacum) co-localized in cytoplasmic, punctate bodies detected by confocal microscopy at 2 dpi. Images in (a) and (b) illustrate optical planes in which many RDR6-containing bodies were observed. Scale bars, 10 μm. doi:10.1371/journal.pone.0159080.g002 RNase3 Interacts with SGS3 PLOS ONE | DOI:10.1371/journal.pone.0159080 July 8, 2016 4/16 Fig 3. Interactions of SGS3 with RDR6 and RNase3 in planta.Protein interactions were tested by BiFC and monitored in epidermal cells of N.benthamiana using epifluorescence microscopy. The name of the protein indicates whether the YFP half was fused to the Nor C-terminus of the test protein (e.g., YN-RNase3 and IbSGS3-YC, respectively). Co-expression of the test proteins with eIF(iso)4E of potato [27] was included as a negative control for interaction (panels b, d, f, h, j and k). (a) Interaction of AtRDR6 and AtSGS3 resulted in SGS3/RDR6 bodies (positive control); (b) negative control. (c) RNase3 did not interact with AtRDR6, but RNase3 Interacts with SGS3 PLOS ONE | DOI:10.1371/journal.pone.0159080 July 8, 2016 5/16 catalytic activity on dsRNA), as well as the effect of IbSGS3 co-expressed with RNase3 or RNase3-Ala, on sense-mediated gfp silencing was tested in N.benthamiana 16c. If RNase3 or RNase3-Ala was not used, the corresponding Agrobacterium strain was replaced with a strain expressing β-glucuronidase (GUS, negative control). Because the YN-tagged IbSGS3 (SGS3-YN) was used in these experiments, it was replaced, when not used, by an Agrobacterium strain expressing YN, so to maintain the sense-mediated gfp silencing pressure similar in all treatments. Following co-expression of gfp,YN and GUS, GFP fluorescence initially increased, but then decreased substantially by 6 dpi (Fig 5A and 5B), indicating no significant suppression of silencing. Similar results were obtained following co-expression of gfp and IbSGS3. These results were consistent with detectable accumulation of gfp-derived siRNA, as tested by northern analysis (Fig 5C). However, when leaf tissues were co-infiltrated with Agrobacterium strains for expression of GFP and RNase3, GFP fluorescence remained higher than the background fluorescence resulting from expression of the constitutively expressed gfp transgene of the 16c line (note the green fluorescent veins in Fig 5A and 5B) and the additional expression of the infiltrated gfp-expressing construct. GFP fluorescence was further enhanced in leaf tissue co-expressing GFP, IbSGS3, and RNase3 (Fig 5A and 5B). gfp mRNA levels were not much increased in leaf tissues co-infiltrated with gfp and RNase3, but they were clearly elevated in leaves co-infiltrated with gfp,IbSGS3 and RNase3, as compared with leaf tissues infiltrated with gfp and GUS,orgfp,GUS and IbSGS3 (Fig 5C). Accumulation of gfp-specific siRNAs was negatively correlated with gfp mRNA accumulation (Fig 5C) and was high in leaf tissues co-expressing gfp and RNase3-Ala (Fig 5C). In contrast, little gfp–derived siRNA was observed in leaf tissue co-expressing gfp,RNase3-Ala and IbSGS3 although suppression of silencing could not be observed visually in the leaves (Fig 5C). (e) interacted with AtSGS3 (A.thaliana) and (i) IbSGS3 (I.batatas). (g) Catalytically inactive RNase3-Ala was also able to interact with SGS3. Data were collected at 2 dpi. All images were acquired at the same magnification. Scale bar, 20 μm. doi:10.1371/journal.pone.0159080.g003 Fig 4. Interactions of RNase3 and AtSGS3, as assessed with BiFC and confocal microscopy, co-localize with cytoplasmic punctate bodies containing AtRDR6 in epidermal cells of N.benthamiana.The three proteins were co-expressed by agroinfiltration. Images were acquired at an optical plane in which many RDR6-containing bodies were observed at 2 dpi. The dashed circles point out some of the cytoplasmic punctate bodies that show signals of the RNase3-SGS3 interaction and co-localize with RDR6. Scale bar, 10 μm. doi:10.1371/journal.pone.0159080.g004 RNase3 Interacts with SGS3 PLOS ONE | DOI:10.1371/journal.pone.0159080 July 8, 2016 6/16 RNase3 Interacts with SGS3 PLOS ONE | DOI:10.1371/journal.pone.0159080 July 8, 2016 7/16 RNase3 and IbSGS3 were readily expressed in the tissues, as detected by western analysis (Fig 5D). These results were reproduced in five independent experiments, indicating enhanced suppression of gene silencing in the presence of both IbSGS3 and RNase3. tasiRNA production is not affected by RNase3 RNase3 has not been found to interfere with transitivity [3], but possible interference of RNase3 with the tasiRNA pathway has not been tested. Therefore, TAS3 transcript–derived siRNA were analyzed in transgenic sweetpotato plants (cv. Huachano) expressing RNase3 [2]. Computational analysis of the Sequence Read Archive database sequences of sweetpotato cv. Xushu18 (SRX090758) revealed that the A.thaliana TAS3 5’D7(+) transcript-specific tasiRNA probe [28] aligned well with a number of reads, which allowed partial reconstruction of the putative IbTAS3 transcript sharing high sequence identity with known TAS3 transcripts. The A.thaliana D7(+)-specific probe was used to detect tasiRNA in sweetpotato cv. Huachano and two RNase3-transgenic lines, but no differences in the tasiRNA amounts were detected (S4 Fig). Discussion Our results show that RNase3 co-localizes and is associated with SGS3/RDR6 bodies implicated in plant gene regulation and antiviral RNAi [9,29]. The association is likely mediated by the interaction of RNase3 with SGS3, as indicated by BiFC assay. Deprivation of SGS3 can reduce RNAi-based virus resistance and enhance accumulation of cucumoviruses in plants [9], whereas accumulation of potyviruses correlates positively with SGS3 accumulation [11,12]. In this respect it is noteworthy that co-expression of SGS3 with RNase3 slightly (but reproducibly) enhanced suppression of sense RNA–induced RNAi by RNase3. Yoshikawa et al.[13] have shown that in the micro-RNA (miRNA) directed process of trans-acting small interfering RNA (tasiRNAs) production from TAS2 gene transcript by RISC, the 3' cleavage fragment of the TAS2 transcript is protected from degradation. In this process, SGS3 binds to AGO1-RISC via the dsRNA formed by interaction of the miRNA with the target RNA. The authors also Fig 5. Co-expression of RNase3 and IbSGS3 enhances suppression of sense-mediated gene silencing (gene co-suppression). (a) and (b) Four sectors (1 to 4) of a leaf of N.benthamiana 16c constitutively expressing gfp were agroinfiltrated to co-express GFP and (1) GUS (negative control), (2) IbSGS3 and RNase3, (3) RNase3, or (4) IbSGS3. If RNase3 was not used, the corresponding Agrobacterium strain was replaced with a strain expressing β-glucuronidase (GUS, negative control). IbSGS3 was expressed with the N-proximal part of YFP fused to the C-terminus. If it was not used, it was replaced with an Agrobacterium strain expressing YN (N-proximal half of yfp) to maintain similar sense-mediated silencing pressure. The treatments are positioned differently in the two leaves in terms of the younger (basal) and older (tip) part of the leaf. Silencing of gfp was observed by the disappearance of GFP fluorescence (sectors 1 and 4), whereas GFP fluorescence above the background level indicated suppression of gfp silencing (sectors 2 and 3). The leaf was photographed under UV light at 6 dpi. Similar results were obtained in five independent experiments. (c) Northern analysis of gfp mRNA and gfp mRNA-derived siRNA in the agroinfiltrated leaf tissues. Co-expression of GUS or IbSGS3 with GFP by agroinfiltration in gfp-transgenic leaves resulted in gfp silencing, as shown by the readily detectable accumulation of gfp-derived siRNA (Fig 5C). In contrast, coexpression of GFP and RNase3 resulted only in low accumulation of gfp siRNA, and no gfp siRNA could be detected following co-expression of GFP, RNase3 and SGS3; however, accumulation of gfp mRNA was enhanced (Fig 5C). Co-expression of the RNase3-Ala mutant (disabled from catalytic activity on dsRNA) with GFP resulted in readily detectable accumulation of gfp siRNA, whereas co-expression of RNase3-Ala, SGS3 and GFP resulted in low accumulation of gfp siRNA (Fig 5C). 25S and 5S ribosomal RNA is shown as a loading control, respectively. (d) Western analysis of RNase3 and IbSGS3-YN in the agroinfiltrated leaf sectors illustrated in (a) by immunoblotting using anti-RNase3 and anti-GFP antibodies, respectively. Molecular masses of the detected proteins (kDa) were estimated by comparison with the protein marker run in the gel. doi:10.1371/journal.pone.0159080.g005 RNase3 Interacts with SGS3 PLOS ONE | DOI:10.1371/journal.pone.0159080 July 8, 2016 8/16 concluded that the SGS3-dependent stabilization of the 3' fragment of TAS2 RNA is crucial to tasiRNA production [13]. However, the results of our study may suggest that interaction of RNase3 with SGS3 is part of the protective functions of SGS3, or assists RNase3 in targeting and cleaving ds-siRNA or secondary viral dsRNA synthesized by RDR6. The catalytic activity on dsRNA is pivotal for suppression of RNAi by RNase3 and results in short fragments of ~14 base pairs that are too short to be incorporated in RISC and are inactive in RNAi [2,4]. It is noteworthy that RNase3 was found to interact with the SGS3 homologs of two unrelated plant species, A.thaliana and sweetpotato, of which the latter is the natural host of SPCSV. These findings imply that RNase3-SGS3 interactions could be a common and an important mechanism for SPCSV to control antiviral RNAi in its host plants. SGS3 and RDR6 are involved in tasiRNA pathways, but mutations in SGS3 or RDR6 do not always yield a discernable phenotypic change in A.thaliana [29,30]. Hence, also interference of RNase3 with the tasiRNA pathways would not necessarily cause any phenotype, whereas perturbation of miRNA homeostasis more often is associated with morphological changes [31]. The transgenic sweetpotato plants expressing RNase3 do not show any phenotype different from wild-type plants, except more pronounced purple pigmentation in leaves, similar to SPCSV-infected plants. Therefore, we tested the possible influence of RNase3 on TAS3 transcript–based tasiRNA production, which represents the most conserved tasiRNA gene in plants [14], but results were negative. Other tasiRNA pathways could be elucidated in sweetpotato and influence of RNase3 on them tested in future studies. The punctate bodies containing RNase3, SGS3 and RDR6 were cytoplasmic, which is consistent with previous studies reporting occurrence of the SGS3/RDR6 bodies in cytoplasm without any specific association with other cellular bodies or organelles [15]. In addition, RNase3 was detected in the nucleus. RDR6 is also found in the nucleus, in contrast to the SGS3/RDR6 bodies [32]. The size of RNase3-dsRED (ca. 51 kDa) could allow passive diffusion to the nucleus [25], but the nuclear RNase3-containing bodies, which were distinct from Cajal bodies, are intriguing and suggest that RNase3 might have a functional role in the nucleus. In general, recent studies question the view that post-transcriptional RNAi would occur only in the cytoplasm. dsRNA directed to an intron can silence a gene [32] and DCL4 responsible for production of the majority of the virus-derived siRNAs in A.thaliana [33] is detected only in the nucleus [15,16,32]. Furthermore, nuclear import of the P6 protein of Cauliflower mosaic virus is required for the suppression of antiviral RNAi [34]. Therefore, it is likely that antiviral RNAi occurs in both the cytoplasmic and nuclear compartments [32]. Finally, the studies of Garcia-Ruiz et al.[33] have shown that basal levels of antiviral RNAi and siRNA biogenesis remain in Arabidopsis mutants lacking RDR1, RDR2 and RDR6, which suggests that there is an additional, unknown pathway producing dsRNA. Hence, the possible interference of RNase3 with RNAi in the nucleus and the production of RNase3 via the postulated new pathway remains an interesting subject for further study. Taken together, there is limited but increasing precedence for interaction between viral RNAi suppressors with SGS3 and its impact on RNAi. V2 of Tomato yellow leaf curl virus (family Geminiviridae) [20,35], p2 of Rice stripe virus [21], TGBp1 of Plantago asiatica mosaic virus [18], protein P of Lettuce necrotic yellows virus [19], VPg of Potato virus A [11] and RNase3 of SPCSV represent examples reported from unrelated viruses. The unifying factor is that all six viral proteins are able to suppress only sense RNA–induced RNAi, indicating that they interfere with the steps of RNAi involving activities of RDR6 and SGS3. Hence, the results presented here advance understanding of the mechanisms needed for efficient RNase3-mediated suppression of RNAi and provide new information about the subcellular context and phase of the RNAi pathway in which RNase3 realizes RNAi suppression. RNase3 Interacts with SGS3 PLOS ONE | DOI:10.1371/journal.pone.0159080 July 8, 2016 9/16 28. Allen E, Xie Z, Gustafson AM, Carrington JC. MicroRNA-directed phasing during trans-acting siRNA biogenesis in plants. Cell. 2005; 121: 207–221. PMID: 15851028 29. Peragine A, Yoshikawa M, Wu G, Albrecht HL, Poethig RS. SGS3 and SGS2/SDE1/RDR6 are required for juvenile development and the production of trans-acting siRNAs in Arabidopsis. Genes Dev. 2004; 18: 2368–2379. PMID: 15466488 30. Adenot X, Elmayan T, Lauressergues D, Boutet S, Bouche N, Gasciolli V et al. DRB4-dependent TAS3 trans-acting siRNAs control leaf morphology through AGO7. Curr Biol. 2006; 16: 927–32. PMID: 16682354 31. Voinnet O. Induction and suppression of RNA silencing: insights from viral infections. Nat Rev Genet. 2005; 6: 206–220. PMID: 15703763 32. Hoffer P, Ivashuta S, Pontes O, Vitins A, Pikaard C, Mroczka A, et al. Posttrancriptional gene silencing in nuclei. Proc Natl Acad Sci U S A. 2011; 108: 409–414. doi: 10.1073/pnas.1009805108 PMID: 21173264 33. Garcia-Ruiz H, Takeda A, Chapman EJ, Sullivan CM, Fahlgren N, Brempelis KJ, et al. Arabidopsis RNA-dependent RNA polymerases and Dicer-like proteins in antiviral defense and small interfering RNA biogenesis during Turnip mosaic virus infection. Plant Cell. 2010; 22: 481–496. doi: 10.1105/tpc. 109.073056 PMID: 20190077 34. Haas G, Azevedo J, Moissiard G, Geldreich A, Himber C, Bureau M, et al. Nuclear import of CaMV P6 is required for infection and suppression of the RNA silencing factor DRB4. EMBO J. 2008; 27: 2102– 2112. doi: 10.1038/emboj.2008.129 PMID: 18615098 35. Fukunaga R, Doudna JA. dsRNA with 5´overhangs contributes to endogenous and antiviral RNA silencing pathways in plants. EMBO J. 2009; 28: 545–555. doi: 10.1038/emboj.2009.2 PMID: 19165150 36. Zhang D, Trudeau VL. The XS domain of a plant specific SGS3 protein adopts a unique RNA recognition motif (RRM) fold. Cell Cycle. 2008; 7: 2268–2270. PMID: 18635957 37. Haikonen T, Rajamäki ML, Valkonen JPT. Interaction of the microtubule associated host protein HIP2 with the viral helper component proteinase is important in infection with Potato virus A. Mol Plant Microbe Interact. 2013; 26: 734–744. doi: 10.1094/MPMI-01-13-0023-R PMID: 23489059 RNase3 Interacts with SGS3 PLOS ONE | DOI:10.1371/journal.pone.0159080 July 8, 2016 16 / 16