Graft Transmission of RNA Silencing to Non-Transgenic Scions for Conferring Virus Resistance in Tobacco

RNA silencing is a mechanism of gene regulation by sequence specific RNA degradation and is involved in controlling endogenous gene expression and defense against invasive nucleic acids such as viruses. RNA silencing has been proven to be transmitted between scions and rootstocks through grafting, mostly using transgenic plants. It has been reported that RNA silencing of tobacco endogenous genes, NtTOM1 and NtTOM3, that are required for tobamovirus multiplication, resulted in high resistance against several tobamoviruses. In the present study, we examined the graft transmission of RNA silencing for conferring virus resistance to non-transgenic scions of the same and different Nicotiana species grafted onto rootstocks in which both NtTOM1 and NtTOM3 were silenced. Non-transgenic Nicotiana tabacum (cvs. Samsun and Xanthi nc) and N. benthamiana were used as scions for grafting onto the rootstocks silenced with both genes. Short interfering RNA (siRNA) of NtTOM1 and NtTOM3 was detected in both the scions and the rootstocks eight weeks after grafting. The leaves were detached from the scions and inoculated with several tobamoviruses. The virus accumulation was tested by ELISA and northern blot analysis. The viruses were detected in grafted scions at extremely low levels, showing that virus resistance was conferred. These results suggest that RNA silencing was induced in and virus resistance was conferred to the non-transgenic scions by grafting onto silenced rootstocks. The effect of low temperature on siRNA accumulation and virus resistance was not significantly observed in the scions.


Introduction
RNA silencing is sequence-specific gene regulation, which is widely conserved among eukaryotes including fungi, plants and animals. One of the characteristics is a mobile signal that can spread to neighboring cells through plasmodesmata or systemically to the entire plant through the vascular system [1][2][3][4]. Long distance or systemic silencing involves an amplification of the same silencing signal utilizing cellular RNA-dependent RNA polymerase (RDR) [4], [5]. Double-stranded (ds) RNA is the key trigger for RNA silencing. There are different ways for generating dsRNA: as partial ds structure of viral RNA, by which they can induce RNA silencing and become targets for RNA degradation [6], transcription of inverted repeats (IR) [7], [8] or RDR-mediated synthesis that uses aberrant RNA (aRNA) as a template [9]. Dicer, a ribonuclease (RNase) III family, is involved in the cleavage of the dsRNA into 21-25 nucleotide (nt) duplexes, referred to as short interfering RNA (siRNA) [10], [11]. The next step in the RNA silencing process is the incorporation of siRNA into Argonaute containing (Ago-containing) multicomponent RNA-induced silencing complex (RISC). After an ATP-dependent incorporation of (2) siRNA into RISC, the Ago acts as a sequence specific ribonuclease (Slicer) that cleaves single stranded RNA at discrete positions [11][12][13].
Arabidopsis thaliana genes TOM1 and TOM3 are involved in tobamovirus multiplication [14][15][16]. In the double null mutants of these genes, tobamovirus cannot multiply [15]. These genes are conserved in other plants including tomato, tobacco and melon [17]. It is reported that N. tabacum carries two endogenous homologous genes NtTOM1 and NtTOM3 which function in parallel to support tobamovirus multiplication and that silencing of both genes resulted in high resistance against several tobamoviruses [17].
It is widely accepted that the siRNA signal in plants can spread systemically. The siRNA signal moves through sieve tubes and serves an antiviral function [18]. Recently, siRNA was shown to be able to pass through the graft junction and that siRNA triggers RNA silencing in recipient cells [19], [20]. Most experiments on the graft transmissions of RNA silencing have been shown with transgenic plants [1], [21][22][23][24][25]. The transmission of post-transcriptional gene silencing (PTGS) to transgenic scions occurred from transgenic tobacco plants silenced with both the transgene and endogenous gene of the nitrate reductase gene (Nia) but that PTGS was never transmitted to non-transgenic scions [1]. More recently it is reported that silencing of endogenous genes in nontransgenic scions was induced from transgenic rootstocks in which the companion cell specific promoter driven hairpin construct (hp) of the glutamate-1-semialdehyde aminotransferase (GSA) gene was introduced by Agrobacterium but not from silenced rootstocks with cauliflower mosaic virus (CaMV) 35S promoter-driven hpGSA [26]. Another report on grafted apple showed that graft transmission of RNA silencing to GUS expressing scions occurred in an in vitro system from GUS silenced transgenic apple but that it did not to non-transgenic scions from rootstocks silenced with an endogenous Malus domestica anthocyanindin synthase (Mdans) gene [27]. In this study, we show that RNA silencing and the resultant virus resistance was induced in non-transgenic scions through grafting onto transgenic silenced rootstocks, where CaMV 35S promoter-driven hp construct of the endogenous gene of tobacco NtTOM1 and NtTOM3 were silenced. Our findings may shed light on a practical approach of RNA silencing to confer desired traits to non-transgenic plants.

Plant Materials
Transgenic Sd1 (Silencing double mutant 1) line, in which both NtTOM1 and NtTOM3 genes were silenced, was produced by crossing transgenic S-tom1 and S-tom3 plants [17]. The Sd1 line was selected after selection by kanamycin. Non-transgenic Nicotiana tabacum (cvs. Samsun and Xanthi nc) and N. benthamiana were used as scions for grafting onto the Sd1 rootstocks. Seeds were sown on K MS plates and one week later the seedlings were transferred to plastic plant boxes containing K MS with kanamycin (50 mg/ml). Approximately 2 weeks later the plants were transferred to pots containing autoclaved soil (Fujimi Co. Ltd., Shizuoka, Japan) in growth chambers. The plants were grown at 24uC under a 16light/8-h dark cycle with cool fluorescent light.

Grafting Procedure
Approximately 8-week-old plants were used for the cleft grafting, as described previously [21]. Rootstocks were prepared by removing the shoots above at least two basal leaves and then creating a vertical cut of 1-2 cm at the center of the stem. Scions (3-5 cm stature) were prepared by removing leaves and trimming the base of the scion to a wedge. The scion/rootstock junction was wrapped with Parafilm and a clip. Plants were covered with plastic bags to avoid dehydration for 1 week or until the graft was completed.

DNA Extraction and PCR Analysis
DNA was extracted from leaves using a cetyltrimethylammonium bromide-based (CTAB-based) extraction procedure [28]. Quantity and purity of DNA was measured by using a spectrophotometer (Gene Spec I, Hitachi Co. Ltd., Tokyo, Japan). Genomic DNA (100 ng) was used as a template for PCR. The forward primer GUS-linker-F and the reverse primers NtTOM1-R or NtTOM3-R were used to amplify the NtTOM1 or NtTOM3 transgene, respectively. The PCR reaction was 2 min of 94uC preheating, followed by a 25-cycle amplification program; 1 min at 94uC for denaturation, 1 min at 55uC for annealing and 1 min at 72uC for extension, and a final extension at 72uC for 5 min. PCR products were analyzed by electrophoresis on a 1% agarose gel. Forward and reverse primer pairs are listed in Table S1.

RNA Isolation and Reverse Transcription PCR
Total RNA was extracted from the young leaves of the transgenic Sd1 line and non-transgenic plants using a TRI reagents kit (Molecular Research Centre, Cincinnati, OH, USA) according to the manufacturer's instructions, and residual genomic DNA was removed by DNase I treatment for 30 min at 37uC (Takara, Kyoto, Japan). Total RNA (1 mg) was reverse-transcribed using RevertAid TM reverse transcriptase (Fermentas, Hanover, CA, USA) in a 20 ml reaction containing 5X RT buffer [250 mM Tris-HCl (pH 8.3 at 25uC), 250 mM KCl, 20 mM MgCl 2 , 50 mM DTT], 10 mM dNTP, 20U RNase Inhibitor and 20 mM of GUS-linker-R, NtTOM1-R or NtTOM3-R primer. The mixture containing RNA and primers was heated at 70uC (10 min) and chilled (10 min) before addition of 5X buffer and RNase inhibitor, and then incubated at 42uC (60 min). The reaction was inactivated at 70uC (10 min). The cDNA was amplified by PCR under the following conditions: 95uC for 1 min, followed by 30 cycles of denaturing at 95uC for 30 sec, annealing at 60uC for 30 sec, and extension at 68uC for 1 min, and a final extension at 68uC for 10 min. The forward primer NtTOM1-F or NtTOM3-F and the reverse primer NtTOM1-R, NtTOM3-R or GUS-linker-R were used to amplify the endogenous/transgene NtTOM1 or NtTOM3, respectively. Their sequences are listed in Table S1. PCR products were analyzed by electrophoresis on a 1% agarose gel.

Probe DNA Preparation
The cDNA clones, pBI-NtTOM1 and pBI-NtTOM3, were digested with the combination of XbaI/SacI and BamHI/SacI, respectively. The gel-purified DNA fragments (876 and 910 bp) of pBI-NtTOM1 and pBI-NtTOM3 were used as NtTOM1 and NtTOM3 probes, respectively. To detect virus accumulation, the 600 bp EcoRI/BamHI fragment of pTMVOM-L (Kawamura and Ishikawa, unpublished), 1964 bp KpnI/EcoRI fragment of pL11A-2-35 [29], and 900 bp EcoRI DNA fragment of a WMoV cDNA clone, pTW62-2 [30], were used as probes to detect the accumulation of TMV, ToMV and WMoV, respectively. The GUS fragment from pBI221 was prepared after digestion with BamHI/SacI. The [a-32 P]dCTP-labeled probe DNA was prepared using the Megaprime DNA Labeling System (Amersham Biosciences, Piscataway, NJ, USA) according to the manufacturer's instructions.

Southern Blot Analysis
Southern blot analysis was performed according to the method described previously [31]. In brief, genomic DNA (20 mg) was subjected to gel-electrophoresis and transferred to a nylon membrane (Hybond N + , General Electric, New York, USA) with a few modifications. The membrane was hybridized with the [a-32 P]dCTP-labeled cDNA probe of the GUS (1.8 kbp) fragment. Then the membranes were analyzed using a Bio Image Analyzer (BAS-2000, Fuji Photofilm, Tokyo, Japan) as described previously [32].

Northern Blot Analysis
Total RNA was extracted from the young leaves of plants using the TRI reagents kit according to the manufacturer's instructions. The total RNA gel blot analysis for transcript was performed as described previously [32] and the membrane was hybridized with the [a-32 P]dCTP-labeled cDNA probes prepared from the pBI221, pTMVOM-L, pL11A-2-35 or pTW62-2 as mentioned above.

siRNA Detection
Small RNA was prepared as described previously [32]. The enriched small RNA (30 mg) was separated by electrophoresis and electrotransferred to a nylon membrane. The probe was prepared as described above. Hybridization and siRNA detection was carried out with the labeled cDNA probes (GUS, NtTOM1 or NtTOM3) as described previously [32].

Evaluation of Virus Resistance
Purified ToMV-L [33], WMoV (TMV-W [34]) or TMV-OM [35] was used as inoculum. Detached 5 or 6th true leaves were cut out from the top of the scion and the base of the rootstock, respectively, and inoculated with a suspension of virus (10 mg/ml in 10 mM sodium phosphate buffer, pH 7.0). Detection of virus infection was performed 15 days after inoculation using ELISA and northern blot analysis. ELISA was performed using antiserum specific for each tobamovirus (Japan Plant Protection Association, Tokyo, Japan) and anti-rabbit IgG conjugated with alkaline phosphatase (Sigma-Aldrich, St. Louis, MO, USA).

Results and Discussion
Characterization of Sd1 Line Silenced with Both NtTOM1 and NtTOM3 The seeds of the crossed line between S-tom1 and S-tom3 were grown in the presence of kanamycin. All plants appeared normal in morphology and development. From the resulting collection of the progeny, all the plants of one Sd1 line showed two DNA bands on Southern blots probed by the GUS fragment ( Figure 1B). On the other hand, another line showed only a single DNA band ( Figure S2A). PCR analysis showed that the plants of that line with only the upper DNA band on the Southern blot carry the NtTOM3-IR transgene ( Figure S2B). The total RNA from the plants of Sd1 was used for northern blot analyses. As a result, transgene expression was observed ( Figure 1C) coinciding with the result reported previously [17].

Graft Transmission of RNA Silencing from Sd1 line to Non-transgenic Tobacco Plants
Non-transgenic scions of the same or different species of tobacco, N. tabacum L. cvs. Samsun and Xanthi nc or N. benthamiana, were grafted onto transgenic silenced Sd1 rootstock. Eight weeks after grafting, total RNA was extracted from leaves. Endogenous mRNA expression of both genes were highly detected in the non-transgenic scions before grafting but reduced extremely after grafting onto the Sd1 rootstocks, where it was under the detection level, (Figure 2A B). Furthermore, NtTOM3 transgene transcript was only detected in rootstocks but not in scions ( Figure 2C). The same result was also obtained with NtTOM1 transgene (data not shown). The transgene derived siRNA was detected in both rootstocks and grafted scions (Figure 3). The RNA blots were re-hybridized with other probes specific to NtTOM1 or NtTOM3. Using either probe, the corresponding siRNA was detected (Figure 3) even though the level was lower in scions than in rootstocks. Approximately 50-60% non-transgenic scions of tobacco showed the induction of RNA silencing (Table 1). From these results, it was confirmed that RNA silencing was induced in the non-transgenic grafted scions of not only the same species but also different species of tobacco.

Virus Resistance Assay of Scions after Grafting
It has been reported that the Sd1 lines show strong resistance to tobamoviruses [17]. Thus we examined whether any virus resistance was induced in the grafted scions. The leaves were detached from the scions 8 weeks after grafting and inoculated with several tobamoviruses including TMV-OM, ToMV-L and WMoV. Then virus accumulation was tested 16 days after inoculation by ELISA and northern blot analysis. As a result, an extremely low amount of virus was detected in grafted scions (Figure 4 and 5) showing that virus resistance was conferred. The non-transgenic scions grafted onto the non-transgenic rootstocks showed a high amount of virus accumulation similar to that in the non-transgenic scions before grafting onto the Sd1 rootstocks ( Figure S3). Cucumber mosaic virus could multiply efficiently in both the non-transgenic scions and transgenic silenced rootstocks suggesting no resistance to other virus group than tobamovirus (unpublished data). This observation confirmed that non-transgenic scions were post-transcriptionally silenced after grafting onto the transgenically silenced rootstocks.

Effect of Temperature on siRNA Accumulation and Virus Resistance
RNA silencing is negatively affected by low temperature [36]. Thus we examined the induction of RNA silencing in the scions of grafted plants at 24 and 15uC. After the grafted plants were kept at either 24 or 15uC for two months, siRNA accumulation was tested in both scions and rootstocks. There was no significant decrease in siRNA accumulation at low temperature (15uC) compared to that at 24uC (data not shown). The ToMV-L inoculation test showed that all scions and rootstocks were resistant to the virus at different temperatures ( Figure S3). Moreover, almost similar levels of virus resistance were found at different temperatures in N. tabacum cv. Samsun and Xanthi nc although little difference, if any, in virus titer was found in N. benthamiana. These findings indicate that the resistance level is not remarkably different and rather similar between 24 and 15uC.
In this study, we showed the graft transmission of RNA silencing to non-transgenic scions from the silenced transgenic rootstocks where endogenous tobacco genes, NtTOM1 and NtTOM3, were silenced by the hp of these genes driven by the 35S promoter. In micrografting experiments using N. benthamiana, it was reported Figure 2. RT-PCR analysis of NtTOM1 and NtTOM3 in grafted plants. One microgram of total RNA was reverse-transcribed with NtTOM1-R, NtTOM3-R or GUS-linker-R primer. The cDNA was amplified by PCR under the conditions mentioned in Materials and Methods. The forward primer NtTOM1-F or NtTOM3-F and the reverse primer NtTOM1-R or NtTOM3-R were used to amplify the NtTOM1 or NtTOM3 endogenous gene, respectively. NtTOM3 transgene transcript was amplified with NtTOM3-F and GUS-linker-R. PCR products were analyzed by electrophoresis on a 1% agarose gel. SB, scion (before grafting); RS, rootstock; GS, grafted scion (after grafting). Horizontal bracket represents a grafted plant. doi:10.1371/journal.pone.0063257.g002 that RNA silencing of the endogenous GSA gene is grafttransmissible to non-transgenic scions after grafting onto the transgenic rootstocks with hp of GSA gene driven by a companion cell-specific promoter [26]. However, it is not the case with transgenic rootstocks with the same hp GSA driven by the 35S promoter. This finding contradicts our results, which may be because their grafting duration was only for 2 weeks while ours was 8 weeks. Thus the prolonged duration may cause the successful graft transmission of RNA silencing in non-transgenic scions through grafting. In apple graft transmission of RNA silencing to a transgenic GUS expressing scions occurred in an in vitro system from GUS silenced transgenic rootstocks while the transmission to non-transgenic scions of high anthocyan natural variety did not from the Mdans silenced transgenic rootstocks [27]. It is not known why RNA silencing of the endogenous gene was not induced in the non-transgenic apple scions. It may be because the levels of siRNA in rootstocks and of Mdans transcript might have been insufficient to induce silencing in the non-transgenic scions. From our results, it is clear that RNA silencing can be induced in non-transgenic scions after grafting onto the transgenic silenced rootstocks.
Induction of RNA silencing in scions of different species, N. benthamiana, was also observed ( Figure 3 and Table 1). Considering 97.2% identity of the nucleotide sequence between N. tabacum l. cv. Samsun (NtTOM1, AB193039 [17]) and N. benthamiana (TOM1 homologue, AM261863 [37]), it is reasonable to suggest that RNA  silencing was induced in the scion of N. benthamiana. It was shown that a transgenic silenced line of N. benthamiana showed resistance to TMV [37]. This approach may be applicable to other species of tobacco or crops of interest if the nucleotide sequence identity is high enough to induce RNA silencing. Based on siRNA analysis and virus resistance assay against tobamoviruses, it was confirmed that the engineered transgene sequence in the Sd1 line as well as non-transgenic grafted scions was post-transcriptionally silenced. We analyzed the Sd1 rootstocks by Southern and northern blot analyses. It revealed that Sd1 plants carried one copy of each transgene and the expression of the transgenes was detected in the rootstocks ( Figure 1B and C). The non-silenced non-transgenic scions grafted onto the silenced rootstocks showed accumulation of NtTOM1 or NtTOM3 siRNA and resistance to the tobamoviruses (Figure 3, 4, 5 and Table 1). These results suggest that the silencing signal which triggers silencing is transmitted from the silenced rootstocks to the nontransgenic scions and induced silencing.
Low temperature inhibited the accumulation of siRNA in insect, plant and mammalian cells [36], [38][39][40] and leads to enhancement of virus susceptibility and to a dramatic reduction in the level of siRNA [36]. However, another report showed that in N. tabacum no obvious decrease in siRNA accumulation was found at 15uC compared with that at 24uC [40]. In this study, we found no significant decrease in the accumulation of siRNA at low temperature in N. tabacum compared to that at 24uC. There was little or no reduction in the level of siRNA in N. benthamiana (15uC) (data not shown) in contrast to the remarkable reduction in siRNA accumulation [36]. However, in N. tabacum, there is no significant difference between 24 and 15uC [40]. Virus resistance assays showed that all scions were resistant to ToMV-L( Figure S3). Only a slight difference in virus titer by ELISA was found in the scions of N. benthamiana between the two temperatures. Thus effect of temperature on RNA silencing may depend upon the plant species, which might be due to the properties of silencing components in each plant species. These findings indicate that the resistance level might be similar at both 24uC and 15uC in this system.
In conclusion, RNA silencing can be induced in non-transgenic scions by grafting onto transgenic silenced rootstocks. This approach of grafting with non-transgenic scions onto transgenic rootstocks could be very useful for any crop, where grafting has been frequently used to confer desired traits. Grafting is conventionally used for vegetable and horticultural crops including tomato, cucurbitaceous plants and fruit trees. Their rootstocks are very highly resistant to soil-born disease pathogens and nematodes.
Our findings may open the way to apply RNA silencing to crops for conferring other desired traits such as higher nutritional value, improved flavor, and sugar contents. The [a-32 P]dCTP-labeled cDNA probe was prepared as mentioned in Materials and Methods. SB, scion (before grafting); RS, rootstock of Sd1; GS, grafted scion (after grafting), CGS, grafted scion on control rootstock. Horizontal bracket represents a grafted plant. (TIF) Figure S4 Effect of temperature on virus resistance in the grafted scions. Purified ToMV-L [33] was used as inoculum. Detached leaves were inoculated with a suspension of virus (10 mg/ml). Detection of viruses was performed 15 days after inoculation using ELISA. The mean absorbance values are shown with SD. Each absorbance value was analyzed in three independent ELISA assays. Asterisks indicate significant difference from 24uC (*p,0.01). (TIF)