The tobacco N gene confers resistance to tobacco mosaic virus (TMV) and encodes a Toll-interleukin-1 receptor/nucleotide binding site/leucine-rich repeat (TIR-NBS-LRR) class protein. We have developed and used a tobacco rattle virus (TRV) based virus induced gene silencing (VIGS) system to investigate the role of tobacco candidate genes in the N-mediated signalling pathway. To accomplish this we generated transgenic Nicotiana benthamiana containing the tobacco N gene. The transgenic lines exhibit hypersensitive response (HR) to TMV and restrict virus spread to the inoculated site. This demonstrates that the tobacco N gene can confer resistance to TMV in heterologous N. benthamiana. We have used this line to study the role of tobacco Rar1-, EDS1-, and NPR1/NIM1- like genes in N-mediated resistance to TMV using a TRV based VIGS approach. Our VIGS analysis suggests that these genes are required for N function. EDS1-like gene requirement for the N function suggests that EDS1 could be a common component of bacterial, fungal and viral resistance signalling mediated by the TIR-NBS-LRR class of resistance proteins. Requirement of Rar1- like gene for N-mediated resistance to TMV and some powdery mildew resistance genes in barley provide the first example of converging points in the disease resistance signalling pathways mediated by TIR-NBS-LRR and CC-NBS-LRR proteins. The TRV based VIGS approach as described here to study N-mediated resistance signalling will be useful for the analysis of not only disease resistance signalling pathways but also of other signalling pathways in genetically intractable plant systems.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Plant resistance (R) genes that confer resistance to various phytopathogens have been cloned from different plant species (Dangl and Jones, 2001). With the exception of Pto and PBS1 genes, all the R genes isolated and characterized to date encode proteins with leucine-rich repeats (LRRs). Members of the major class of R genes, the NBS-LRRs, contain a nucleotide-binding site (NBS) and C-terminal LRR of various lengths. The NBS-LRR class of R genes can be further grouped into TIR-NBS-LRR and CC-NBS-LRR classes based on their amino-terminal sequence. Members of the TIR-NBS-LRR class of proteins contain an N-terminal domain that is similar to the cytoplasmic domains of the Toll and the interleukin-1 receptor (called TIR domain). Members of the CC-NBS-LRR class of proteins contain a coiled-coil domain (CC) at the amino terminus.
The tobacco N gene confers resistance to tobacco mosaic virus (TMV) and encodes a protein belonging to TIR-NBS-LRR class (Whitham et al., 1994). The N protein, either by itself or in a protein complex, is hypothesized to specifically recognize the 50 kDa helicase domain of the TMV replicase protein (Abbink et al., 1998; Erickson et al., 1999; Padgett et al., 1997) and trigger a signal transduction cascade leading to induction of hypersensitive response (HR; localized cell death) and restriction of virus spread. This initial response makes N plants resistant to subsequent infection by TMV or other pathogens in a process called systemic acquired resistance (SAR) (Ross, 1961). During SAR, salicylic acid (SA) levels increase throughout the plant and defense genes such as pathogenesis related (PR) genes are expressed (Ryals et al., 1996).
A comprehensive structure-function analysis of the N gene suggests that TIR, NBS and LRR domains are necessary for N function (Dinesh-Kumar et al., 2000). The N gene encodes two transcripts, NS and NL, via alternative splicing in vivo, that are required to confer resistance to TMV (Dinesh-Kumar and Baker, 2000). The molecular and physiological processes immediately following TMV infection and preceding the HR in the N–mediated signal transduction pathway remain obscure. It has been widely observed that HR is preceded by a rapid outburst of reactive oxygen intermediates (ROI). In NN tobacco, the burst of ROI occurs as early as 10 min following TMV infection and is NADPH- and Ca2+-dependent (Doke and Ohashi, 1988). ROI may play a role in restricting the pathogen to the infection site by inducing cell death in a manner similar to mammalian neutrophils (Dangl et al., 1996; Levine et al., 1994). In barley, a cytoplasmic Zn2+ binding protein, Rar1, functions upstream of H2O2 accumulation and HR induction (Shirasu et al., 1999). Rar1 is required for the function of many Mla R genes in barley (Schulze-Lefert and Vogel, 2000). However, its requirement in resistance to other pathogens, including TMV, has not been tested.
SA plays an important role in the N-mediated resistance to TMV (Delaney et al., 1994). During the N-mediated HR to TMV, SA accumulates around the HR lesions and throughout the plant (Malamy et al., 1990). Furthermore, nahG transgenic tobacco NN plants in which SA gets converted into catechol, fail to accumulate SA, and fail to develop SAR (Delaney et al., 1994). In Arabidopsis, NPR1/NIM1 encodes a putative transcription factor regulator that controls defense gene expression and operates downstream of SA (Cao et al., 1997; Ryals et al., 1997). HRT-mediated resistance to turnip crinkle virus (TCV) in Arabidopsis requires SA but is independent of NPR1 (Kachroo et al., 2000). The role of NPR1 in the N-TMV pathway is not known.
In this paper, we have used a tobacco rattle virus (TRV) based VIGS system to test the role of the tobacco homologues Rar1, EDS1, and NPR1/NIM1 genes in N-mediated resistance to TMV. VIGS is initiated when a recombinant virus carrying a sequence from a host gene infects the plant. The endogenous gene transcripts homologous to the insert in the VIGS vector are degraded by a post-transcriptional gene silencing mechanism (PTGS) (Baulcombe, 1999).
We show that suppression of Rar1-, EDS1-, and NPR1/NIM1- like genes compromise N function suggesting that these proteins play an essential role in the N-mediated resistance pathway.
Development of a TRV based VIGS system
TRV is a bipartite positive sense RNA virus (MacFarlane, 1999). RNA1 encodes 134 and 194 kDa replicase proteins from the genomic RNA, a 29-kDa movement protein and a 16-kDa cysteine-rich protein from subgenomic RNAs (Figure 1a). RNA2 encodes the coat protein from the genomic RNA and two non-structural proteins from the subgenomic RNAs (Figure 1a). TRV RNA1 can replicate and move systemically without RNA2. To develop TRV as a VIGS vector, we constructed T-DNA vectors containing cDNA clones of RNA1 and RNA2 of Ppk20 strain (Figure 1b). The cDNAs corresponding to RNA1 and RNA2 were inserted immediately following the duplicated CaMV 35S promoter transcriptional start site so that no non-TRV nucleotides are present at the 5′ end after transcription. In addition, a self-cleaving ribozyme sequence (Turpen et al., 1993) was included at the 3′ end so that only three non-viral nucleotides are predicted to be present after transcription. In the TRV RNA2 cDNA construct, the non-structural genes were replaced with a multiple cloning site (MCS) useful for cloning the target gene sequences for VIGS (Figure 1b).
The biological activity of the TRV clones was confirmed by mixing Agrobacterium cultures containing RNA1 and RNA2 T-DNA constructs and infiltrating the mixture into the leaves of N. benthamiana. An initial test with two independent RNA1 and RNA2 cDNA clones proved to be less infectious than the wild type virus (data not shown); possibly due to sequence changes introduced during RT–PCR. Therefore, sequences that were consistent among each of three independent RNA1 and RNA2 clones, though different from the GenBank sequence, were considered as correct sequences. Based on this information, we reconstructed the RNA1 and RNA2 cDNA clones. The infectivity of this new TRV clone was comparable to that of wild type Ppk20 virus (data not shown). We will refer to the T-DNA constructs corresponding to RNA1 and RNA2 as pTRV1 and pTRV2, respectively, throughout this paper.
The TRV based VIGS vector we have developed differs in sequence and design from the previously reported TRV-VIGS vector (Ratcliff et al., 2001). In terms of design: (i) it includes a double 35S promoter; (ii) RNA1 cDNA is stable at room temperature in the pBIN19 vector in E. coli and therefore disruption of the polymerase ORF by inclusion of an intron is not required; (iii) transcription initiation is predicted in vivo at the first nucleotide of TRV cDNA; and (iv) inclusion of a ribozyme at the 3′ end ensures generation of a precise 3′ end of the RNA; inclusion of ribozyme sequence in TMV-based expression vectors has been shown to increase the infectivity 3-fold (Turpen et al., 1993). In terms of sequence: (i) in RNA1, there are 9 base changes (G1266A, T2291C, C3094T, C3130T, A3634G, G4123A, G4254T, G4642A, G5559A) and three amino acid changes (S355N and S697P in polymerase ORF; V78I in 29K movement protein); and (ii) in RNA2, there are many base insertions and changes (base insertion at nt 287, 380, 3490, 3662 and 3756; base changes T338C, A339T, C340A, G342C, C343G, A344C, T654C, C3509T, A3660G). The comparisons of sequences between our TRV and that of Ratcliff et al. (2001) can be found at our web site http://www.yale.edu/plantfunctionalgenomics.
We assessed the gene silencing efficiency of our TRV-VIGS clones by suppressing the expression of the phytoene desaturase (PDS) gene in N. benthamiana. A mixture of Agrobacterium culture containing the pTRV2-PDS and pTRV1 was infiltrated on to the 4-leaf stage N. benthamiana plants (for details see Experimental procedures). The silencing effect was monitored in the upper leaves of the plant 5–7 days post Agro-inoculation. Silencing of PDS leads to the inhibition of carotenoid synthesis, causing the plants to exhibit a photo-bleached phenotype (Figure 1c; Kumagai et al., 1995). The PDS suppression phenotype was visible 5 days post Agro infiltration in the upper leaves of the plant and persisted indefinitely. This result indicated that our TRV-VIGS system could be successfully used to induce silencing of other desirable endogenous plant genes. Moreover, PDS suppression effect was visible uniformly throughout the entire leaf (Figure 1c). Uniform suppression of target gene is essential to study disease resistance using VIGS because often these experiments involve secondary infection with another pathogen.
Generation of N-containing transgenic N. benthamiana to study TMV resistance
The N gene response to TMV infection is routinely studied using N. tabacum cv. Xanthi nc or N. tabacum cv. Samsun NN or transgenic N. tabacum cv. SR1 containing the N transgene (Holmes, 1934; Whitham et al., 1994). In tobacco, recombinant TMV-GFP virus fails to move systemically (Y.L., M.S., and S.P.D-K, unpublished). However, TMV-GFP virus moves systemically in N. benthamiana. Therefore, we generated transgenic N. benthamiana lines containing the N gene construct pSPDK694. This construct contains the N promoter + cDNA-NS+ Intron III containing the alternative exon + N genomic-3′end. This is the minimum N sequence required to confer resistance to TMV (Dinesh-Kumar and Baker, 2000). We transformed N. benthamiana plants with pSPDK694 and generated 10 independent primary transgenic lines. Six independent transgenic lines showed HR upon infection with TMV-U1 strain with no symptoms elsewhere in the plant. We isolated homozygous line, MS4-5, containing a single copy of T-DNA and was used for the experiments described in this paper. The transgenic MS4-5 line exhibits HR lesions on the inoculated leaf 48–72 h after infection with TMV (Figure 2a). The HR lesion phenotype observed in MS4-5 plants is similar to that observed in the wild type N-containing tobacco plants (Whitham et al., 1994). In contrast, the non-transformed N. benthamiana plants fail to induce HR in response to TMV (Figure 2b).
In our VIGS assay described below, we use TMV-GFP virus to monitor TMV movement under UV illumination. Therefore, we tested the effect of TMV-GFP virus infection on MS4-5 and N. benthamiana plants. In MS4-5 plants, GFP fluorescence was observed 24 h after infection on inoculated leaves (Figure 2c); but the upper un-inoculated leaves of the infected plant did not show systemic spread of TMV-GFP even 10 days after infection (Figure 2d). In N. benthamiana control plants, GFP fluorescence was visible in the inoculated leaf after 24 h (Figure 2e). The infection foci as indicated by the area of GFP fluorescence spot were much bigger as compared to that of MS4-5 plants (compare Figure 2e versus Figure 2c). After 10 days, TMV-GFP was observed in the upper leaves of the control plants (Figure 2f). These results suggest that transgenic N-containing MS4-5 plants exhibit HR upon TMV infection and that the virus is restricted to the infection site.
To confirm our phenotypic data, RNA gel blots were prepared using RNA extracted from mock and TMV-infected plants and hybridized with a probe derived from the movement protein (MP) gene of TMV. In mock inoculated N. benthamiana (Figure 2g; lane 1) and MS4-5 plants (Figure 2g; lane 4), no hybridization signal was observed. As expected, in susceptible wild-type N. benthamiana plants, TMV-RNA accumulated to a high level in inoculated leaves (Figure 2g; lane 2) as well as in the systemic upper leaves (Figure 2g; lane 3). In MS4-5 transgenic plants, a low level of TMV-RNA was detected in the inoculated leaf (Figure 2g; lane 5) and no viral RNA was detected in the upper un-inoculated leaves (Figure 2g; lane 6). These results confirm that the MS4-5 line restricts TMV to the inoculated leaf.
In NN tobacco plants, HR and SAR to TMV are associated with the induction of PR proteins (Hooft van Huijsduijnen et al., 1986; Ward et al., 1991). We investigated the induction of PR1a message in TMV inoculated leaves of wild type N. benthamiana and transgenic MS4-5 lines using RNA gel blots. No PR1a message was detected in either mock-infected MS4-5 plants (Figure 2h; lane 2) or in TMV infected N. benthamiana plants (Figure 2h; lane 1). On the other hand, PR1a expression was induced in TMV infected N-containing MS4-5 transgenic plants (Figure 2h; lane 3). These results suggest that transgenic MS4-5 lines induce PR1a message just as wild-type N-containing tobacco plants do.
Taken together, our phenotypic and molecular analyses of the MS4-5 line demonstrate that the tobacco N gene can confer resistance to TMV in heterologous N. benthamiana plants and can be used to study N-mediated signalling in TMV resistance.
Suppression of N function using TRV based VIGS
We tested the ability of the TRV-VIGS system to induce silencing of the N gene to determine whether the system could be used for TMV resistance pathway gene function studies. We hypothesized that the suppression of N using TRV-VIGS should result in loss-of-resistance to TMV. First, we needed to observe whether TRV infection alone has an effect on N-mediated resistance to TMV. A mixture of Agrobacterium containing pTRV1 and pTRV2 was infiltrated into MS4-5 and wild type N. benthamiana plants (Figure 3a). Eight days post infiltration, the upper leaves of these plants were infiltrated with Agrobacterium containing the TMV-GFP construct to monitor resistant and susceptible phenotypes. In MS4-5 plants, an HR phenotype was observed within 72 h and the GFP fluorescence signal was restricted to the infiltrated area (Figure 3b), and after 6 days, the signal was reduced significantly (data not shown). Systemic leaves of these plants showed no sign of TMV-GFP fluorescence even after 10 days (Figure 3c). Wild type N. benthamiana plants showed no HR at the site of infiltration (Figure 3d) and TMV-GFP was able to spread throughout the plant in 10 days (Figure 3e) with no reduction in the TMV-GFP signal in the infiltrated leaf.
We confirmed that the TMV-GFP is restricted to the Agro-inoculation site in MS4-5 plants by Northern blot analysis (Figure 3f) using the MP gene of TMV as a probe. In TRV infected wild-type N. benthamiana plants, TMV-GFP RNA was present in the upper un-infiltrated leaves (Figure 3f; lane 1). There was no TMV-GFP RNA in the upper un-infiltrated leaves of TRV infected MS4-5 plants (Figure 3f; lane 2). These results, together with the above-described phenotypic data, indicate that TRV infection alone has no effect on TMV resistance or susceptibility and could be used to study genes involved in N-mediated resistance to TMV using the TRV-based VIGS system.
To suppress N function using TRV-VIGS, we cloned a 1129-bp fragment of the N cDNA corresponding to nt 86–1215 (Whitham et al., 1994) into pTRV2. A mixture of Agrobacterium cultures containing pTRV1 and pTRV2-N was infiltrated onto MS4-5 plants (Figure 4a). Eight days post-infiltration, the upper leaves were infiltrated with Agrobacterium containing a TMV-GFP plasmid. These plants were then observed for at least 20 days for movement of TMV-GFP from the infiltrated site. The TRV-N infected MS4-5 plants failed to show characteristic HR phenotype in the infiltrated leaf (Figure 4b) and the TMV-GFP was able to spread to the upper leaves (Figure 4c). Movement of TMV-GFP into the upper parts of the MS4-5 plants was detectable 6 days after TMV-GFP infiltration. Presence of TMV-GFP RNA in systemic leaves was confirmed by RNA blots hybridized to a probe derived from the MP gene of TMV (Figure 3f; lane 3). These results indicate that the N transgene function is suppressed in MS4-5 plants by TRV-VIGS.
We performed semi-quantitative RT–PCR, using total RNA extracted from MS4-5 plants infected with TRV-N or TRV alone, to confirm the VIGS of the N gene at the molecular level. In TRV-N infected plants, the N message was reduced by more than 88% compared to the TRV infected control (Figure 5a versus Figure 5b). In both tissue RNA samples, EF1α RNA levels were similar (Figure 5c versus Figure 5d) and served as an internal control for RNA quality and RT–PCR amplification. These results show that the TRV based VIGS system efficiently suppresses targeted host genes and can be used as a rapid means for assaying the role of candidate genes in N-mediated resistance to TMV. Therefore, we set out to suppress tobacco Rar1, EDS1 and NPR1/NIM1 homologues in MS4-5 plants to investigate their role in N-mediated signalling.
Cloning of tobacco homologues of Rar1, EDS1, and NPR1/NIM1
In order to study the role of Rar1, EDS1 and NPR1/NIM1 in N-mediated resistance to TMV, we cloned tobacco homologues of these genes. We used the amino acid sequence of barley Rar1 to search the Institute of Genomic Research (TIGR) tomato database using TBLASTN. Tomato EST clone TC96555 showed significant homology to barley Rar1. Based on this information, we designed primers described in the Experimental procedures section and performed RT–PCR to clone full-length Rar1 from N. tabacum. The nucleotide sequence of tobacco Rar1 has been deposited in GenBank (AF480487). Amino acid sequence analysis of tobacco Rar1 using a BLAST search shows 63% identity and 73% similarity to barley Rar1 (Figure 6). Similar to barley Rar1, tobacco Rar1 contains two CHORD domains and one plant-specific CCCH domain (Shirasu et al., 1999). In addition, the tobacco Rar1 contains conserved strings of invariant cysteine and histidine residues within the CHORD domains (Figure 6). The amino acid sequences outside the CHORD and CCCH domains are significantly different between tobacco Rar1 and barley Rar1 (Figure 6).
To clone the tobacco EDS1 homologue, EST clones TC91460 and TC95587, which showed significant homology to Arabidopsis EDS1, were identified. We cloned full-length tobacco EDS1 using RT–PCR as described in the Experimental procedures section. The tobacco EDS1sequence is deposited in GenBank (AF480489). Amino acid sequence analysis of tobacco EDS1 shows 43% identity and 58% similarity to Arabidopsis EDS1 (Figure 7). Like Arabidopsis EDS1, tobacco EDS1 contains three lipase catalytic residues, S125, D186, H322 (Falk et al., 1999). In addition, the consensus sequence around the predicted catalytic S125 is highly conserved between tobacco EDS1 and Arabidopsis EDS1. Sequence of the eds1–1 allele in Arabidopsis indicated a change in the E466K (Falk et al., 1999), an amino acid that is conserved in tobacco EDS1.
A TIGR tomato database search using ArabidopsisNPR1/NIM1 identified EST clones TC95582, TC91366 and AW399343, which showed significant homology to the ArabidopsisNPR1. A full-length tobacco homologue of NPR1 was cloned using RT–PCR as described in the Experimental procedures section. The nucleotide sequence of full-length tobacco NPR1 has been deposited in GenBank (AF480488). Amino acid sequence analysis of tobacco NPR1 shows 52% identity and 72% similarity to Arabidopsis NPR1 (Figure 8). Amino acid changes that lead to loss-of-function of NPR1 in Arabidopsis (Cao et al., 1997) are conserved in the tobacco NPR1 sequence (Figure 8). In addition 4 out of 5 amino acids required for nuclear localization of Arabidopsis NPR1 (Kinkema et al., 2000) are also conserved in tobacco NPR1 (Figure 8).
Role of tobacco Rar1, EDS1 and NPR1/NIM1 in the N-mediated resistance to TMV
In order to determine if Rar1, EDS1 and NPR1/NIM1 play important roles in N-mediated resistance, we cloned fragments of these genes into pTRV2 as described in the materials and methods section. A mixture of Agrobacterium cultures containing pTRV1 with pTRV2; pTRV2-Rar1; pTRV2-EDS1; or pTRV2-NPR1/NIM1 were each infiltrated on to 4-leaf stage MS4-5 plants. Eight days after infiltration, Agrobacterium containing the TMV-GFP construct was infiltrated on to upper leaf of each of the plants. These plants were visualized under UV light 10 days later for movement of TMV-GFP. In the Rar1, EDS1 and NPR1/NIM1 suppressed plants, TMV-GFP spread into the upper leaves (Figure 9a-c) as visualized by the presence of GFP in the systemic leaves. More GFP fluorescence was consistently observed in the EDS1 and Rar1 than the NPR1/NIM1 silenced plants (compare Figure 9a-b versus Figure 9c). Taken together, the phenotypic data suggest that suppression of Rar1-, EDS1- and NPR1/NIM1-like genes compromise N function and lead to loss-of-resistance to TMV.
To see whether the TMV-GFP is present in the upper un-infiltrated leaves of EDS1, Rar1 and NPR1/NIM1 silenced plants, we analysed them for the presence of TMV-GFP RNA. RNA gel blots were prepared from the upper un-infiltrated leaves and probed with the TMV MP gene. As expected, more TMV-GFP RNA was observed in EDS1 (Figure 9d; lane 4) and Rar1 (Figure 9d; lane 5) silenced plants than the NPR1 (Figure 9d; lane 6) silenced plants. The amount of TMV-GFP RNA present in EDS1 and Rar1 silenced plants was less than the amount present in susceptible N. benthamiana plants (compare Figure 9d; lane 1 versus 4 and 5) and more than TRV-N suppressed plants (compare Figure 9d; lane 3 versus 4 and 5). These results corroborate our phenotypic data that the suppression of Rar1-, EDS1- and NPR1/NIM1-like genes lead to loss-of-resistance to TMV.
To confirm the VIGS of Rar1, EDS1 and NPR1/NIM1 genes at the molecular level, we performed semiquantitative RT–PCR using total RNA extracted from MS4-5 plants infected with TRV, TRV-Rar1, TRV-EDS1 and TRV-NPR1/NIM1. In TRV-Rar1 infected plants, the Rar1 message was reduced by more than 95% compared to the TRV-infected control (Figure 5e versus Figure 5f). In TRV-EDS1 and TRV-NPR1/NIM1 infected plants, the EDS1 (Figure 5g versus H) and NPR1/NIM1 (Figure 5i versus Figure 5j) messages were reduced by more than 88% and 78%, respectively, compared to the TRV-infected control (Figure 5a versus Figure 5b). In all these cases, EF1α RNA levels were similar (data not shown) and served as an internal control for RNA quality and RT–PCR amplification.
In this report, we have shown that the TRV based VIGS system can be used as an efficient reverse genetics tool to identify components of N-mediated resistance signalling. This technique offers a rapid means of gaining insight into gene function in plants.
Transgenic MS4-5 plants exhibit resistance to TMV by induction of HR lesions and containment of TMV to the infection site similar to wild type N-containing tobacco plants (Whitham et al., 1994) and transgenic N-containing tomato plants (Whitham et al., 1996). Tobacco and N. benthamiana belong to the same genus and therefore may contain conserved signal transduction components required for N-mediated resistance to TMV in both species. This is consistent with the observation reported by (Tai et al., 1999) that R genes may function in restricted heterologous species belonging to the same genus or family of plants. In fact, many R genes like pepper BS2, potato Rx1, and tomato Pto and Cf9 function in the heterologous Nicotiana host (Bendahmane et al., 1999; Hammond-Kosack et al., 1998; Rommens et al., 1995; Thilmony et al., 1995). Unlike tobacco, N. benthamiana plants show less auto-fluorescence under UV illumination and support movement of the TMV-GFP recombinant virus. Therefore, the MS4-5 N-containing N. benthamiana transgenic line will be useful in studies aimed towards understanding cellular and physiological events associated with N-mediated HR using TMV-GFP recombinant virus in the future.
Requirement of tobacco Rar1-, EDS1-, and NPR1/NIM1-like genes for the function of the N gene described in this report provide the first evidence for the role of these genes in a viral resistance pathway. From this work and that of others, EDS1 constitutes the converging point of signalling pathways mediated by the functional TIR-NBS-LRR class of R genes. In addition to EDS1, NDR1 is involved in converging race-specific resistance pathways in Arabidopsis (Aarts et al., 1998). However, this convergence involves distinct subclasses of NBS-LRR R genes. NDR1 is required for the function of the CC-NBS-LRR class of R genes. Requirement of Rar1 for virus resistance (this report) and some powdery mildew resistance genes in barley (Schulze-Lefert and Vogel, 2000) provide another example of converging points in the disease resistance signalling pathways. Moreover, Rar1 represents the first example of a signalling component shared by CC-NBS-LRR (Mla12) and TIR-NBS-LRR (N) type of race-specific R genes.
Our observations suggest that the tobacco NPR1/NIM1-like gene is required for TMV resistance. However, NPR1/NIM1 is not required for function of the HRT-mediated resistance to the virus TCV in Arabidopsis (Kachroo et al., 2000). The Arabidopsis NPR1/NIM1 has been shown to interact with members of the basic leucine zipper (bZIP) family of transcription factors like TGA2, TGA3 and TGA6 (NIF1) (Despres et al., 2000; Zhang et al., 1999; Zhou et al., 2000). To date, the biological significance of these NPR1/NIM1–TGA interactions is not known. Our VIGS phenotypic data suggests that suppression of individual TGA factors has no effect on N-mediated resistance to TMV (Y.L., M.S., S.P.D-K, unpublished observations). However, suppression of multiple TGA factors simultaneously by mixed infections results in loss-of-resistance to TMV. Further molecular and biochemical analyses of this phenotypic data are necessary to understand the exact role of NPR1-TGA factors in N-mediated resistance to TMV. We have tested the role of 18 additional candidate genes like SIPK, SIPKK, WIPK, NtMEK1 and NtMEK2, COI1, WRKY factors, etc. Our phenotypic analysis suggests that 10 of these genes may play a role in N-mediated resistance to TMV (Y.L., M.S., and S.P.D-K, unpublished observations). Further functional analysis of these genes at the molecular and biochemical level should help to better understand N-mediated signal transduction pathway leading to TMV resistance.
VIGS is known to suppress sequences of highly homologous genes. Therefore, one should be cautious in interpreting VIGS results. In Arabidopsis, EDS1, NPR1 and Rar1 are single copy genes. Our N. benthamiana genomic DNA blot analysis suggests that there are two copies of EDS1, NPR1 and Rar1 (Figure 10). This is consistent with the aneuploidy nature of N. benthamiana (1n number = 19 chromosome) compared to the diploid members of Nicotiana (1n number = 12 chromosome) (Smith, 1979). Therefore, data presented in this report indicates that the Rar1-, EDS1- and NPR1/NIM1-like genes of tobacco are required for function of the N gene.
In RNAi, the RNA degradation is triggered by double stranded RNA (dsRNA) and occurs in a 2-step process. In the first step, double stranded RNA (dsRNA) is processed into shorter, 21–25 nucleotide long sense and antisense units. These small RNAs, called short interfering RNA or siRNA, in the second step act as guide sequences to identify homologous transcripts and target them for destruction (Nishikura, 2001). Recent evidence suggests that there are secondary siRNAs appear to derive from the action of RNA-dependent RNA polymerase (RdRp) (Sijen et al., 2001). The RdRp plays a role in cyclic amplification of initial siRNAs into secondary siRNAs. These secondary siRNAs exhibited a distinct polarity, 5′ to 3′ on the antisense strand. Therefore, in the future to overcome the silencing of other highly homologous genes, one could target the 5′ untranslated region of the gene for silencing.
The availability of the Arabidopsis genome sequence (AGI, 2001) and a large number of plant-expressed sequence tags (ESTs) (The Institute of Genomic Research) provide a wealth of information about the plant genome. Therefore, the VIGS assay described here will offer a means to test the function of homologous gene sequences in N. benthamiana. Moreover, our analysis suggests that the TRV based VIGS system described here induces efficient gene silencing in Arabidopsis and tomato plants (Y.L., M.S., and S.P.D-K. manuscript in preparation). Therefore, TRV-VIGS will offer an efficient reverse genetics tool to test gene function in different plant systems of choice. Even though the VIGS approach is a rapid method to ascertain function by gene inactivation, one of the main disadvantages is that the phenotype observed is not transmittable to the next generation because of this, it is not possible to perform genetic crosses, suppressor or enhancer screens and other long-term genetic manipulations. However, transgenic expression of a replicating PVX (termed PVX amplicon) containing the plant exon sequence consistently induces the silencing of the corresponding endogenous gene in subsequent generations (Angell and Baulcombe, 1997). Therefore, the generation of a TRV-based amplicon containing transgenic lines of Rar1, EDS1 and NPR1 will provide us with an invaluable resource for further genetic analyses. So far, Dangl (1999) reports that classical forward genetics screens to identify components of a given resistance response have yielded only a few genes because of redundancy or lethality (Dangl, 1999). The VIGS approach described here may help to overcome this problem because the VIGS ‘phenotype’ is conditional, loss of mutations due to organismal lethality should not occur. In the future, large-scale screens using a normalized cDNA library in the TRV-based VIGS system, in conjunction with microarray analysis and two-hybrid experiments should facilitate identification of additional components of the defense pathways in plants.
pTRV1 (RNA1). The first strand cDNA of TRV-RNA1 was derived from total RNA extracted from TRV-Ppk20 infected N. benthamiana leaves using primer OYL64 (5′-CGGCCCGGGCCCGTTTCGTCCTTTAGGGACTCGTCAGTGTACTGATATAAGTACAGACGGGCGTAATAACGCTTACGTAGGCGAGGGGTTTTACC− 3′) and superscript reverse transcriptase (Gibco/BRL). The primer OYL64 contains an XmaI restriction site (bold), a ribozyme sequence (underlined), and a sequence complementary to TRV-RNA1 bases 6755–6791 (italicized). This first strand cDNA was used as template with upstream primer OYL61 (5′-ATAAAACATTTCAATCCTTTGAACGCGGTAGAACG-3′) corresponding to TRV- RNA1 bases 1–35 and the downstream primer OYL64 to PCR amplify the full-length cDNA of TRV-RNA1. The PCR amplified product was digested by XmaI and cloned into StuI-XmaI–cut pYL44, which is a derivative of pBIN19 binary T-DNA vector (Frisch et al., 1995) carrying the duplicated cauliflower mosaic virus (CaMV) 35S promoter from pCASS2 (Shi et al., 1997) and nopaline synthase (NOS) terminator. The entire sequence of cDNA corresponding to TRV Ppk20 strain RNA1 is deposited in GenBank (AF406990).
pTRV2 (RNA2). Two cDNA fragments corresponding to bases 1–1646 and 3470–3855 of TRV Ppk20 RNA2 (GenBank Z36974) were amplified by RT–PCR and cloned into StuI-SacI restricted pCASS2 (Shi et al., 1997) to obtain pYL36. During RT–PCR, multiple restriction enzyme sites were included between two fragments at base 1646 for cloning foreign DNA sequences, and a self-cleaving ribozyme sequence was also engineered at the 3′-end of viral RNA2 cDNA. Plasmid pTRV2 was generated by subcloning the HindIII-EcoICR1 restricted fragment of pYL36 into HindIII-HpaI restricted pCAMBIA0390 T-DNA vector. The complete sequence of pTRV2 is deposited in GenBank (AF406991).
pTRV2-derivatives. The cDNA fragments corresponding to PDS, NPR1/NIM1, Rar1, EDS1 and N were PCR amplified and cloned into pTRV2. A 369-bp fragment of tobacco PDS that corresponds to nt 878–1246 of tomato PDS (GenBank #M88683) was amplified using a forward primer (5′-CTG ACG AGC TTT CGA TGC AGT GCA T-3′) and a reverse primer (5′-ATA TAT GGA CAT TTA TCA CAG GAA C-3′). A 1129-bp fragment corresponding to nt 86–1215 of N cDNA (Whitham et al., 1994) was amplified using a forward primer (5′-ATG GAG CTA TGA TGT TTT CTT AAG TTT TAG-3′) and a reverse primer (5′-GAA GGC CTT TAG CAT AAT TTA CTA CCT C-3′). A 468-bp fragment corresponding to nt 186–654 of tobacco Rar1 was amplified using primers: 5′-AGG AAA GCA CAC AAC AGA AAA ACC-3′ and 5′-GTG CCA TCC TTT GGT GCA TGG AGG-3′). A 548-bp fragment corresponding to nt 1290–1837 of tobacco EDS1 was amplified using a forward primer (5′-GAG TAT CAG ACC AAG TGT GAT ATC CG-3′) and a reverse primer (5′-GCT GAG GTG GGA GTG TTT TCC ACC-3′). A 753-bp fragment corresponding to nt 1014–1767 of tobacco NPR1 was amplified using primers: 5′-GAA AGA GCC TAA AAT TGT AGT GTC-3′ and 5′-CTA TTT CCT AAA AGG GAG CTT ATT-3′. The identity of these constructs was confirmed by DNA sequencing.
pSPDK661 (TMV-GFP): The TMV cDNA fragment from 30B TMV-GFP (Shivprasad et al., 1998) was cloned into pBIN19 derivative pYL44. A ribozyme sequence (Turpen et al., 1993) based on the satellite virusoid of subterranean clover mottle virus was engineered at the 3 ′ end of the TMV cDNA.
Cloning Rar1-, EDS1- and NPR1/NIM1-like genes from Tobacco
The 5′ and 3′ ends of tobacco Rar1, EDS1 and NPR1/NIM1 genes were cloned using SMRT RACE cDNA amplification kit (CLONTECH, CA, USA). Rar1 5′ RACE product was generated using a nested universal primer (NUP) from the kit as a forward primer and a Rar1 gene specific primer (5′-CCT TTC ATC CGG TCA TGG AAG ATA GCG-3′) as a reverse primer. Rar1 3′ RACE product was generated using a gene specific forward primer (5′-AGG AAA GCA CAC AAC AGA AAA ACC-3′) and a universal reverse primer (UPM) from the kit. EDS1 5′-and 3′ RACE products were generated using NUP and an EDS1 gene specific primer (5′-GTT TCT TAG TTC CTC CAC TTC TGC-3′) and EDS1 gene specific primer (5′- GAG TAT CAG ACC AAG TGT GAT ATC CG-3′) and UPM, respectively. NPR1/NIM1 5′ RACE product was generated by using NUP and NPR1/NIM1 specific primer (5′-CAA CGT GGA AAG AAG CGT TTT CCA AG-3′). NPR1/NIM1 3′ RACE product was generated using NPR1/NIM1 specific primer (5′-TCT TGC TAT GGC AGG CGA TGA TTT G-3′) and UPM. In all cases, the RACE products were cloned into TOPO cloning vector (INVITROGEN, Carlsbad, CA, USA). At least 5 independent 5′ and 3′ RACE clones were sequenced. RACE 5′ and 3′ overlapping sequences were assembled using the DNA STAR SEQMAN program to obtain full-length sequences of Rar1, EDS1 and NPR1/NIM1.
N. benthamiana plants were transformed with pSPDK694 using Agrobacterium-mediated leaf disc transformation (Horsch et al., 1985) and kanamycin (150 mg l−1) selection. Presence of the transgene was confirmed by PCR using N specific primers.
VIGS assay and GFP imaging
N. benthamiana plants were grown in pots at 25°C in a growth chamber under 16 h light/8 h dark cycle. For VIGS assay, pTRV1 or pTRV2 and its derivatives were introduced into Agrobacterium strain GV2260 by electroporation (BIO-RAD, CA, USA). Agrobacterium cultures at O.D.600 = 0.8 containing TRV or TRV-derivative plasmids were mixed in 1 : 1 ratio and infiltrated onto the lower leaf of 4-leaf stage plants using a 1-ml needleless syringe. Experiments in which the suppression effect of N, EDS1, Rar1 and NPR1/NIM1 on TMV resistance is investigated, these plants received a secondary infiltration with Agrobacterium cultures at O.D.600 = 0.5 containing TMV-GFP construct 8 days after TRV infiltration. At this time, TRV-PDS infected plants exhibit silenced phenotype for PDS. Each silencing experiment was repeated at least 5 times and each experiment included at least four independent plants. In experiments where TMV-GFP virus was used, the inoculum was prepared from systemic infected leaves of N. benthamiana plants infiltrated with Agrobacterium containing TMV-GFP plasmid pSPDK661. GFP imaging was done using UV illumination and photographs were taken using OLYMPUS CAMEDIA E10 digital camera.
RNA isolation, Northern blot and RT–PCR analysis
Total RNA was extracted from silenced and non-silenced N. benthamiana plants using RNAwiz solution (Ambion, TX, USA) and treated with RNase-free DNase (Gene Hunter, TX, USA). First strand cDNA was synthesized using 10 µg of total RNA, oligo d(T)primer and superscript reverse transcriptase (Gibco/BRL, MD, USA). Semi-quantitative RT–PCR was performed as described in (Burton et al., 2000). For RT–PCR, primers that anneal outside the region targeted for silencing were used to ensure that the endogenous gene is tested. The intensities of PCR generated fragments were analysed and quantified using Gel Doc 2000 and Quantity One Version 4.2.1 (BIO-RAD, CA, USA).
RNA blots were prepared using 5 or 10 µg of total RNA following the method described in (Ausubel et al., 1998). To determine TMV or TMV-GFP transcript levels, RNA blots were hybridized with a probe derived from the MP gene of TMV. To determine PR1a message level, a fragment of PR1a derived from tobacco PR1a cDNA (Payne et al., 1988) was used as a probe.
DNA gel blot analysis
The DNA gel blot analysis was performed as described in Dellaporta and Moreno (1994). DNA from N. benthamiana was purified using Qiagen plant DNeasy extraction kit. Ten micrograms of genomic DNA was digested with restriction enzymes, fractionated on 0.8% agarose gel, and blotted onto a Zetaprobe membrane (BIO-RAD). The [α-32P]dCTP-labelled probe corresponding to the fragments of EDS1, NPR1 and Rar1 genes used in silencing were made by the random priming method (Pharmacia Corporation, Peapack, NJ, USA).
We thank Janet Stewart for editing the manuscript. We thank an anonymous reviewer and members of S.P.D-K lab for thoughtful comments and critical reading of the manuscript. We thank David Robinson for TRV Ppk20 virus; W. O. Dawson and S. Shivprasad for 30B TMV-GFP; R. Jefferson and K. S. Ravi for pCAMBIA vectors; S-W. Ding for the pCASS2 vector. The National Science Foundation Plant Genome Grant DBI-0077510 to S.P.D-K supported this work.