RNA silencing is activated by N gene-mediated hypersensitive response and plays a key role in local and systemic virus resistance

Abstract

In plants, recognition between resistance gene (R) and virus induces a local hypersensitive response (HR), which is accompanied by systemic acquired resistance (SAR). The dominant resistance gene N in tobacco confers resistance to tobacco mosaic virus (TMV) at both locally inoculated tissues and systemically infected tissues. However, the mechanisms underlying HR- and SAR-mediated viral inhibition are not fully revealed. In this study, we find that Nicotiana glutinosa RNA-binding protein (NgRBP) is an RNA silencing suppressor which enhances TMV-triggered HR. Stronger HR could result in stronger local and systemic RNA silencing as well as SAR. Enhanced RNA silencing in the systemically infected leaves induced by the NgRBP gene is compromised by transient expression of NahG. These results indicate that RNA silencing is activated by HR and plays a crucial role in local and systemic virus resistance. Our results reveal a crosstalk between N gene-mediated virus resistance and RNA silencing which would deepen our understanding of the established HR and SAR models.

Background

The hypersensitive response (HR) and systemic acquired resistance (SAR) are essential components of plant defence mechanisms against pathogens. HR against virus infection is always initiated upon recognition of virus avirulence factors by plant resistance (R) proteins. Most dominant R genes recognizing viruses condition broad-scale changes in plant biochemistry and physiology that are activated and regulated by various signal transduction pathways. These induced changes include localized cell death and the upregulation of resistance against many types of pathogens throughout the plant. This is conveyed via a signal transduction cascade involving MAP kinases, reactive oxygen species (ROS) and nitric oxide (NO) production, leading to programmed cell death (Jones & Dangl 2006, Mur et al. 2006, Fu & Dong 2013).

After HR, SAR is initiated by a signal transmitted from the local infection site to distal tissues, providing systemic protection against secondary infections (Mur et al. 2006). This concerns the generation of mobile signals like salicylic acid (SA), methyl salicylate (MeSA), and lipid-transfer proteins (Vlot et al. 2009, Klessig et al. 2018). SAR leads to the transcriptional reprogramming of a wide array of defence-related genes. This involves not only pathogenesis-related protein (PR) genes but also genes involved in fortifying cell walls, detoxifying ROS, and other stress-related responses (Fu & Dong 2013). Studies indicate that SAR also involves epigenetic modifications, including histone modifications and DNA methylation, which play a part in the regulation of defence gene expression (Luna & Ton 2012). SA accumulates in the tissues where SAR is activated and binds to nonexpressor of pathogenesis-related proteins 1 (NPR1) to activate the expression of PR genes throughout the plant (Durrant & Dong 2004, Choudhary et al. 2007). Despite considerable research on the mechanisms of HR and SAR, it is still poorly understood how virus infection is inhibited and restricted during the HR in the initially infected tissues and in the systemically infected tissues exhibiting SAR (Carr et al. 2010).

The N gene confers resistance to TMV and encodes a TIR-NBS-LRR protein (Whitham et al. 1994). The N protein is proved to recognize the 50 kD helicase domain of TMV replicase mediated by the protein NRIP1 (Burch-Smith et al. 2007). This recognition triggers a signal transduction cascade that induces HR and limits virus transmission (Abbink et al. 1998, Zhang et al. 2019). The local HR often leads to SAR characterized by nonspecific general defence response throughout the plant (Ryals et al. 1996). N gene-mediated recognition activates downstream signal transduction pathways involving various plant hormones, such as SA and jasmonic acid (JA), and triggers programmed cell death, thereby limiting the spread of the pathogen in local tissues. This is a strategy of sacrificing local cells to protect the whole plant (Jones & Dangl 2006). Our previous study showed that NgRBP, an RNA-binding protein encoded by Nicotiana glutinosa (NN), could enhance TMV resistance (Huang et al. 2019). In this study, HR, SAR, and RNA silencing in TMV-infected N. glutinosa were investigated using NgRBP. The results showed that the TMV-induced HR-activated RNA silencing which could spread to the systemically infected leaves of N. glutinosa. This indicates that RNA silencing is involved in virus-induced HR and SAR. As a transmissible and broad-spectrum antiviral mechanism, RNA silencing might play a key role in virus restriction and inhibition in SAR.

Results

NgRBP promotes HR and SAR induced by TMV

Local TMV infection induces SAR in plants harbouring the N gene, which confers resistance to TMV via gene-for-gene interactions. We observed that transient expression of NgRBP, an endogenous RNA silencing suppressor encoded by N. glutinasa (Huang et al. 2019), enhanced HR induced by TMV infection. To investigate whether the local resistance enhanced by NgRBP affects the SAR, NgRBP and an empty vector were transiently infiltrated into N. glutinasa leaves separately. One day later, the infiltrated leaves of the plants were inoculated with TMV, and their upper leaves were re-inoculated with TMV at 3 d post-inoculation (dpi) (Fig. 1a). The results showed that the necrotic spots in the systemically infected leaves of NgRBP-infiltrated plants were smaller than those in the (empty vector) control plants (Fig. 1c). Using the virus-induced gene silencing (VIGS) technology, the effect of NgRBP on virus resistance was further investigated. N. glutinosa leaves were first infiltrated with tobacco rattle virus (TRV)-NgRBP, and TMV was inoculated twice in the upper leaves at 7 and 11 dpi; TRV vector infiltration was used as a control (Fig. 1b). The necrotic spots on the upper leaves of NgRBP VIGS plants were larger and more intense than those in the controls (Fig. 1c). RT-qPCR showed that TMV accumulation levels in the systemically infected leaves were highest in NgRBP VIGS plants, followed by the empty vector control, and lowest in NgRBP infiltrated plants (Fig. 1d). Gene expression analysis showed that the expression levels of PR genes in the three treatments were inversely related to the TMV accumulation levels (Fig. 1e). These results indicated that NgRBP enhanced the SAR induced by TMV infection.

Fig. 1

NgRBP enhances systemic acquired resistance (SAR) triggered by tobacco mosaic virus (TMV) infection in Nicotiana glutinosa. a Schematic representation of two times TMV inoculation in NgRBP transient-infiltrated plants. pBI121-NgRBP and empty vector were transiently infiltrated in N. glutinasa leaves separately; thereafter, TMV was inoculated in the infiltrated leaves at 1 d post-infiltration (dpi). At 3 dpi, the infiltrated plants were reinoculated with TMV in their upper leaves (the third leaf from the top), and the symptoms were observed 3 days later. b Schematic representation of two TMV inoculations in NgRBP VIGS plants. Firstly, TRV-NgRBP and empty TRV vectors were infiltrated in N. glutinosa leaves, respectively. TMV was inoculated in the upper leaves at 7 dpi, and TMV was reinoculated in the new upper leaves after 4 days. c Symptom observation of the second TMV challenge. d Virus accumulation analysis using TMV capsid protein (CP) of qRT-PCR. e Levels detection of PR1, PR2, and PR5 using qRT-PCR. Gene expression values of the plants that were directly inoculated with TMV were set at 1.0. The mRNA levels of the analyzed genes were normalized by Actin1, which served as an internal standard. The data were analyzed using Student's t-test, and asterisks denote significant differences compared with control plants (* P ≤ 0.05, *** P ≤ 0.001)

To investigate whether the function of HR enhancement is specific to NgRBP, two other viral RNA silencing suppressors (RSSs), P19 and HVT063, whose RSS activity has been reported previously (Silhavy et al. 2002, Jing et al. 2011), were applied to detect their effects on TMV-induced HR. P19, HVT063, and NgRBP were transiently expressed in N. glutinosa leaves, with empty vector infiltration as a control. One day later, TMV was mechanically inoculated into the infiltrated regions. Necrotic symptoms were observed at 3 dpi of TMV, and the leaves were sampled for virus accumulation detection using RT-qPCR for the TMV capsid protein (CP) gene. Similar to NgRBP, P19 and HVT063 expression resulted in fewer and smaller necrotic lesions compared with the empty vector controls (Fig. 2a). RT-qPCR showed less TMV accumulation in P19 and HVT063 expressing leaves than in the controls (Fig. 2b). All three detected RSSs enhanced HR, suggesting that RSS activity may be responsible for increased resistance, as inhibiting RNA silencing in the primary stages of TMV infection ensures that there is enough virus level to trigger stronger HR. This finding aligns with the results obtained by Pruss et al. (2004), who showed that TMV infection in HC-Pro-expressing plants carrying the N gene leads to a reduced number and size of lesions compared with the control plants.

Fig. 2

P19 and HVT063 enhance systemic acquired resistance triggered by tobacco mosaic virus (TMV) infection in Nicotiana glutinosa. a Observation of necrotic symptoms. b Virus accumulation analysis. First, pBI121-NgRBP, pBI121-P19, and pBI121-HVT063 (right) or empty vectors (left) were infiltrated into N. glutinosa leaves, and one day later, TMV was mechanically inoculated in the above-infiltrated leaves. At 3 d post-infiltration of TMV inoculation, necrosis symptoms in the infiltrated area were observed, and the tissues of the infiltrated area were sampled for virus accumulation analysis using TMV capsid protein (CP) of RT-qPCR. The mRNA levels of CP were normalized to those of Actin1, which served as an internal standard. The data were analyzed using Student's t-test, and asterisks denote significant differences compared with control plants (** P≤ 0.01, ***P ≤0.005)

Gene expression profiling of enhanced SAR mediated by NgRBP

To determine the effect of NgRBP on SAR, gene expression profiling was performed using RNA-seq. N. glutinosa leaves were infiltrated with NgRBP and inoculated with TMV one day later. Empty vector-infiltrated and TMV-inoculated plants were used as controls. RNA sequencing profiles were obtained from the upper leaves of these plants at 3 dpi of TMV. Using a false discovery rate threshold of 0.05 and an absolute Log₂ fold change ≥1 (P ≤ 0.05) as the significance cutoff, 1002 differentially expressed genes (DEGs) were found to be significantly regulated by NgRBP compared with controls. Overall, 833 genes were upregulated and 169 downregulated (Additional file 1: Figure S1).

To match the functional enrichment of commonly up- or down-regulated genes during SAR enhancement, gene ontology (GO) enrichment analysis was performed. Common overrepresented GO terms are shown in Additional file 1: Figure S2, and 19 significant GO terms were identified. A total of 1002 DEGs were divided into cellular component (CC), molecular function (MF), and biological process (BP) categories according to their GO functional annotations. The proportion of DEGs in each subclass of the above categories was counted. Among these, the BP category was significantly enriched in cellular protein metabolic processes (GO:0044267) (156) and macromolecular modifications (GO:0043412) (126). Among the MF GO terms, phosphotransferase activity, alcohol group as an acceptor (GO:0016773) (89), kinase activity (GO:0016301) (88), and protein kinase activity (GO:0004672) (83) were considerably enriched. No subclass was enriched in CC categories (Additional file 1: Figure S2). Similar to the GO functional annotation, 1002 DEGs from NgRBP and control groups were annotated using the Kyoto Encyclopedia of Genes and Genomes database. Sixty-nine enriched pathways were identified, of which five were significantly increased (P < 0.05). The most enriched was the ribosomal pathway (KO00940), which included 20 DEGs. The second step involved protein processing in the endoplasmic reticulum containing 11 DEGs. Furthermore, ten DEGs were involved in plant and pathogen interactions, eight in phenylpropanoid biosynthesis, and six in ribosome biogenesis in eukaryotes (Additional file 2: Table S1).

Eight resistance-related genes that were upregulated in NgRBP-expressing plants, based on RNA-seq data, were selected for RT-qPCR analysis (Additional file 2: Table S2). These genes showed higher levels in the systemically infected leaves of NgRBP-expressing plants than in the controls. Among these, PRR1, RGA3, and PRRP2 were significantly upregulated (Fig. 3a). To further study the effect of NgRBP on SAR, nine SAR-related genes were selected. The proteins encoded by these genes were divided into six categories: leucine-rich repeat protein (LRR), cell wall-related proteins including glycine-rich protein precursor, reactive oxygen species-related proteins including glutathione S-transferase, pathogen-related proteins including PR-1a, SAR induced protein 8.2e (SAR8.2) family which responds to agents that induce SAR, and SAR marker genes including PR1, PR2, PR5, and NPR1. Previous studies indicated that LRR1 as a receptor-like protein kinase might play a role in signal transduction in the resistance response of tobacco to TMV (Jacques et al. 2006). The results showed that the expression of these genes was upregulated in NgRBP-expressing plants compared with that in the controls. Among them, PR1, PR2, PR5, LRR1, and SAR8.2e levels were significantly increased in the NgRBP group (Fig. 3b). These results indicated that NgRBP promotes the expression of resistance-related genes in systemically infected leaves where SAR occurs and plays a key role in SAR enhancement.

Fig. 3

The expression levels of disease resistance-related genes in systemic Nicotiana glutinosa leaves. a Expression detection of the up-regulated disease resistance-related genes in RNA-seq using RT-qPCR. b Expression detection of systemic acquired resistance (SAR)-related genes. pBI121-NgRBP and pBI121 were infiltrated in N. glutinosa leaves, respectively, and tobacco mosaic virus was infiltrated one day later. The upper uninfected leaves were sampled to extract total RNA at 3 d post-infiltration. The expression levels of disease resistance-related genes selected from differentially expressed genes and SAR-related genes were detected by RT-qPCR. The standard deviation was based on the results of three biological repetitions. The data were analyzed using Student's t-test, and asterisks denote significant differences between the two groups (* P ≤ 0.05, ** P ≤ 0.01, *** P ≤ 0.005)

TMV-induced HR enhances both local and systemic RNA silencing

RNA silencing is a natural plant defence mechanism against viruses. It is triggered by double-stranded RNAs (dsRNAs), formed during viral replication, and inhibits viral amplification by specifically degrading viral RNAs. To determine the effect of TMV infection on RNA silencing, N. glutinosa leaves were transiently infiltrated with Agrobacterium carrying the GFP gene, and TMV was inoculated into the infiltrated leaves the next day. Simultaneously, the upper leaves were infiltrated with Agrobacterium carrying GFP gene. GFP fluorescence of the lower and upper leaves was observed under UV light at 6, 12, 24, 36, 48, and 60 h post-TMV inoculation (hpi), respectively. Compared with the control (the leaves mock-inoculated with PBS buffer), the GFP fluorescence of TMV-inoculated leaves was weaker at all observation times, especially at 36 and 48 hpi. At 36 hpi, fluorescence induced by cell necrosis appeared in TMV-inoculated leaves. At 60 hpi, TMV infection resulted in large cell necrosis areas (Fig. 4a). The GFP fluorescence of the upper leaves of TMV-inoculated plants was weaker at all observation times compared with that of the mock-inoculated plants (Fig. 4b). These results suggest that TMV-induced HR enhances both local and systemic RNA silencing in N. glutinosa.

Fig. 4

The effect of tobacco mosaic virus (TMV) infection on RNA silencing in GFP transient-expressing leaves of Nicotiana glutinosa. a GFP fluorescence of GFP transiently infiltrated and TMV inoculated leaves. First, the leaves were infiltrated with Agrobacterium carrying GFP, and one day later, TMV was inoculated into the GFP-infiltrated leaves. GFP fluorescence was observed under a long-wave UV lamp at 6, 12, 24, 36, 48, and 60 h post-TMV inoculation (hpi), respectively. b GFP fluorescence in the systemic leaves of N. glutinosa plants whose lower leaves were inoculated with TMV or PBS. The lower leaves of the plants were inoculated with TMV, and simultaneously, the upper leaves were infiltrated with Agrobacterium carrying GFP. GFP fluorescence was observed under a long-wave UV lamp at 6, 12, 24, 36, 48, and 60 hpi, respectively

Enhanced HR is associated with stronger local RNA silencing

Our previous study showed that NgRBP is an endogenous silencing suppressor (Huang et al. 2019), and the transient expression of NgRBP in N. glutinosa leaves could promote HR induced by TMV infection. To explore the role of NgRBP in TMV defence, GFP and NgRBP or empty vector were co-infiltrated into N. glutinosa leaves, and TMV was inoculated into the co-infiltrated leaves. Changes in the GFP fluorescence over time were observed under UV light. At 6 hpi, stronger fluorescence was observed in the co-infection area of GFP and NgRBP compared with GFP and empty vector co-infiltration, although both fluorescence intensities were relatively weak. This difference may be due to the silencing-suppressive activity of NgRBP. From 12 to 48 hpi, GFP fluorescence in the area co-infiltrated with GFP and NgRBP became weaker than that of the empty vector control. Necrosis was observed at 48 hpi in TMV-inoculated area, and GFP fluorescence in both treatments disappeared at 60 hpi (Fig. 5a).

Fig. 5

NgRBP enhances local RNA silencing triggered by tobacco mosaic virus (TMV) infection. a GFP fluorescence observation under UV illumination. b Detection of GFP mRNA and siRNA accumulation using Northern blot. First, pBI121-GFP together with pBI121-NgRBP or an empty vector were co-infiltrated into Nicotiana glutinosa leaves, and 1 d later, TMV was inoculated in the above co-infiltrated leaves. The change of GFP fluorescence with time was observed under UV illumination at 6, 12, 24, 36, 48, and 60 h post-TMV inoculation. Total RNA and small RNA from the co-infiltrated patches were extracted for Northern blot analysis. Ethidium bromide-stained rRNAs and tRNAs were used as loading controls for mRNA and siRNA analyses, respectively

Co-infiltrated tissues were sampled at different time points for Northern blot analysis to verify this observation. The results showed that GFP mRNA levels were consistent with the fluorescence observations. GFP siRNA analysis showed little accumulation in GFP and empty vector co-infiltrated tissues at any time. In contrast, small amounts of siRNA were detected in GFP and NgRBP co-infiltrated leaves at 24, 36, and 48 hpi, and GFP siRNA levels increased with time (Fig. 5b). At 60 hpi, no accumulation of GFP siRNA was detected in any of the co-infiltrations, which may have been caused by cell death induced by the HR response. These results suggest that the stronger the HR, the stronger the local RNA silencing.

HR mediated by N gene prompts systemic RNA silencing

To investigate the effect of HR on systemic RNA silencing, pBI121-NgRBP and an empty vector were transiently infiltrated into N. glutinosa leaves. One day later, infiltrated leaves were inoculated with TMV and infiltrated with Agrobacterium carrying pBI121-GFP (Fig. 6a). In addition, N. glutinosa was infiltrated with Agrobacterium carrying TRV-NgRBP and empty vectors. After 7 d, the systemic leaves were inoculated with TMV, and the upper leaves were infiltrated with pBI121-GFP simultaneously (Fig. 6b). GFP expression in each treatment was monitored under UV light 3 d after pBI121-GFP infiltration. It was observed that the GFP fluorescence of pBI121-NgRBP infiltration was weaker than that of the empty vector control, and the GFP fluorescence of TRV-NgRBP infiltration was more potent than that of TRV alone (Fig. 6c). Northern blot analysis at 3 dpi showed that the mock-inoculation and TMV inoculation treatments accumulated a large amount of GFP mRNA in the infiltrated zone; whereas the other treatments showed little GFP mRNA accumulation in the infiltrated zones. GFP mRNA accumulation in pBI121-NgRBP was slightly lower than that of empty pBI121, and GFP mRNA of TRV-NgRBP was slightly higher than that of TRV. A large amount of GFP siRNA was detected in the systemically infected leaves of the pBI121-NgRBP and empty vector TRV-infiltrated plants. A small amount of GFP siRNA accumulated in the systemically infected leaves of pBI121 and TRV-NgRBP infiltrated groups. GFP siRNAs in TMV-infected and mock-inoculated plants without Agrobacterium infiltration were almost undetectable (Fig. 6d). These results suggest that NgRBP-enhanced HR also prompts systemic RNA silencing in N. glutinosa.

Fig. 6

NgRBP enhances systemic RNA silencing triggered by tobacco mosaic virus (TMV) infection. a Schematic representation of TMV inoculation together with transiently up-regulated NgRBP. b Schematic representation of TMV inoculation together with VIGS down-regulated NgRBP. c GFP fluorescence observation under UV illumination of NgRBP up-regulated and down-regulated leaves. d Detection of GFP mRNA and siRNA accumulation using Northern blot of NgRBP-up-regulated and down-regulated N. glutinosa leaves. pBI121-NgRBP and the empty vector were transiently infiltrated to N. glutinosa leaves, respectively. One day later, the above-infiltrated leaves were inoculated with TMV, and simultaneously, the upper leaves were infiltrated with Agrobacterium carrying pBI121-GFP. First, Agrobacterium carrying TRV-NgRBP and empty vectors were infiltrated into N. glutinosa leaves. The upper leaves were inoculated with TMV at 7 d post-infiltration (dpi), and the upper leaves were infiltrated with pBI121-GFP simultaneously. At 3 dpi, GFP fluorescence was observed under UV illumination, and total RNA and small RNA of the co-infiltrated patches were extracted for Northern blot analysis. Ethidium bromide-stained rRNA and tRNA were used as loading controls for mRNA and siRNA analyses, respectively

To verify the effect on RNA silencing, the systemic leaves of N. glutinosa plants with NgRBP expression and TMV infection were sampled for RT-qPCR. The results showed that the detected genes in the RNA silencing pathway were upregulated compared to those in the empty vector control. Among them, DCL4 and AGO7 were significantly upregulated (Fig. 7a). Previous studies have reported that SA plays a crucial role as an SAR signal to activate resistance to various pathogens, including viruses (Thomma et al. 1998, Métraux et al. 1990). To determine whether SA affects RNA silencing mediated by HR, NgRBP and TMV were infiltrated or inoculated into N. glutinosa leaves, and thereafter NahG, a bacterial gene encoding salicylate hydroxylase that can significantly decrease SA accumulation by converting it to catechol (Yamamoto et al. 1965) together with GFP, was transiently expressed in systemically infected leaves. The results showed that GFP fluorescence was weaker in the upper leaves of the plants where NgRBP was locally expressed before TMV infection than in the empty vector control. However, when NahG was transiently expressed in systemically infected leaves, the weaker GFP fluorescence was restored (Fig. 7b). These results indicate that SA may play a key role in systemic RNA silencing activated by local NgRBP transient expression together with TMV inoculation.

Fig. 7

RNA silencing was enhanced by NgRBP and hindered by NahG expression. a Expression analysis of RNA silencing pathway genes in the systemic leaves of plants with NgRBP transient expression and tobacco mosaic virus (TMV) infection. pBI121-NgRBP and the empty vector pBI121 were infiltrated into the leaves of N. glutinosa, respectively. One day later, TMV was inoculated in the above-infiltrated area, and the systemic leaves were sampled for gene expression analysis via RT-qPCR. The data were analyzed using Student's t-test, and asterisks denote significant differences between the two groups (* P ≤ 0.05, ** P ≤ 0.01, *** P ≤ 0.005). b Transient expression of NahG blocked RNA silencing enhancement mediated by NgRBP. First, pBI121-NgRBP and the empty vector pBI121 were infiltrated into the leaves of N. glutinosa, respectively. One day later, TMV was inoculated in the above-infiltrated area, and simultaneously, the systemic leaves were infiltrated with 35S-GFP and pBI121 or 35S-GFP and pBI121-NahG, respectively. GFP fluorescence was observed under UV light 3 d post inoculation of TMV

Discussion

R gene-virus interactions always result in local acquired resistance (LAR) and SAR. Although there have been many studies on cytopathy and signal transduction in HR and SAR, the mechanisms of LAR and SAR remain unclear. The localized infection mediated by HR is a highly effective mechanism that enables plants to resist infection by viruses. The zone surrounding a TMV lesion exhibits resistance to not only TMV but also other viruses (Kang et al. 2005). Numerous cytopathic changes lead to the collapse and necrosis observed during TMV-induced HR. This process begins with a reactive oxygen burst (Allan et al. 2001), followed by swelling and distortion of the chlorophyll lamella (Weintraub & Ragetli 1964), increased electrolyte leakage, and a notable increase in the superoxidic radical monohydroascorbate (Fodor et al. 2001). HR involves various physiological and biochemical processes, including programmed cell death (PCD), reversible-protein phosphorylation, and protein degradation (Dunigan & Madlener 1995, Lacomme & Santa Cruz, 1999, del Pozo & Lam 2003). Furthermore, during HR, there is a sharp increase in the levels of several hormones such as SA, methyl jasmonate (MeJA), and ethylene (Mur et al. 1997, Malamy et al. 1990, Darby et al. 2000). None of these changes occur in systemically infected susceptible plants.

Some hypotheses have been raised to explain virus localization in HR, which believe that the death of cells in a necrotic lesion may localize or inactivate the virus. However, this explanation is not satisfactory, as virus particles are found in apparently viable cells outside the necrotic areas (Murphy et al. 2001, Wright et al. 2000). More research support the viewpoint that HR-induced cell death and inhibition of virus replication are two different processes (Loebenstein et al. 1980, Loebenstein & Gera 1981, Cole et al. 2001). Barrier substances such as callose were also suggested to be involved in virus localization (Wu & Dimitman 1970). However, callose depositions were also observed in systemic necrosis, where the virus does not remain localized (Shimomura & Dijkstra 1975). Canto and Palukaitis (2002) found that cucumber mosaic virus (CMV) 1a gene could prompt TMV spread in NN tobacco. While CMV 1a protein maintains the silencing suppressor activity of the 2b protein and might affect virus restriction in HR through regulating RNA silencing (Watt et al. 2020). Based on our findings, we suggest that the TMV-induced HR activates RNA silencing, which plays a crucial role in LAR. RNA silencing has been established as a robust mechanism that can directly degrade the genomic RNA of plant viruses. Therefore, our discovery also explains why LAR mediated by TMV is effective not only against TMV but also against other viruses.

SAR can be activated by R gene-virus interactions, and it is effective against a wide range of pathogens, including viruses, bacteria, fungi, and oomycetes (Ryals et al. 1996, Sticher et al. 1997, Durrant & Dong 2004, Liu et al. 2010). The signal of SAR has gained particular attention due to its ability to transmit over long distances. Initially, both SA/MeSA and JA were thought to be SAR signals, but several studies have provided evidence to the contrary (Clarke et al. 2000, Truman et al. 2007, Attaran et al. 2009). Ethylene perception is also believed to be necessary for the generation of systemic signal molecules and SAR development (Verberne et al. 2003). SAR is characterized by increased expression of PR genes, although their roles in establishing SAR are unclear. SAR generally results from the concerted effects of many PR proteins rather than from a specific PR protein (van Loon & van Kammen 1970, Durrant & Dong 2004, Fu & Dong 2013). The activation of six groups of genes, including 1-aminocyclopropane-1-carboxylic acid oxidase gene (ACCO), 3-hydroxy-3-methylglutaryl-CoA reductase gene (HMGR), ascorbate peroxidase gene (APX), pathogenesis-related protein 3 gene (PR3), SAR8.2e, and proteinase-activated peceptor 1 gene (PAR-1), is associated with the development of SAR induced by TMV. Exogenous application of SA transiently induces these genes (Guo et al. 2000). However, the antiviral role of these genes remains unclear.

Although much progress has been made in SAR signalling, the exact mechanisms of SAR, particularly viral inhibition, are not yet completely understood. Our findings indicate that RNA silencing is activated by TMV-mediated HR and spreads to systemic tobacco leaves, playing a key role in viral resistance. In plants, RNA silencing is a crucial mechanism against viruses, and its systemic spread was initially demonstrated in grafting experiments (Yoo et al. 2004, Hewezi et al. 2005, Shaharuddin et al. 2006). Systemic silencing serves an antiviral function by initiating a resistant response in tissues where the virus has not yet infiltrated (Schwach et al. 2005). Obviously, RNA silencing exhibits systemic transmission and broad-spectrum antiviral properties, aligning with the characteristics of SAR. Studies have demonstrated that fewer viral siRNAs are produced when a virus infects transgenic plants with reduced SA accumulation, indicating that SA affects RNA silencing (Alamillo et al. 2006). A recent study showed that Bacillus subtilis 26D triggers SAR against Rhopalosiphum padi by regulating RNAi and microRNA pathways in wheat. This indicates that the connection between SAR and RNA silencing may be universal (Rumyantsev et al. 2023). To summarize, TMV-induced HR promoted local and systemic RNA silencing, and RNA silencing contributes to both LAR and SAR which are triggered against viruses. These results would enrich the established HR and SAR models.

Conclusions

In this study, by using the NgRBP gene which confers stronger HR in N. glutinosa (NN), we found that a stronger HR led to intensified local and systemic RNA silencing, resulting in reduced virus accumulation in both local and systemic tissues. This observation suggests that RNA silencing plays a crucial role in inhibiting viruses during HR and SAR. Our findings uncover an intriguing interaction between N gene-mediated resistance and RNA silencing.

Methods

Plant materials and plasmid constructs

N. glutinosa plants were raised in a greenhouse at 24℃ under a 16 h light/8 h dark photoperiod. The full-length open reading frame (ORF) of 471-nt segment of NgRBP (Accession no. AF005359) was amplified from N. glutinosa total RNA via RT-PCR using high-fidelity DNA polymerase (TransGen, Beijing, China) (Huang et al. 2019). The PCR product was ligated to the pMD-19T vector and cloned into the binary vector pBI121 between the 35S promoter and the Nos terminator. The full-length sequence of NgRBP, except for the start codon, was inserted into TRV infectious clone to generate a VIGS vector. All fragments generated using PCR were confirmed via DNA sequencing. The recombinant plasmids were transformed using the freeze-thaw method into Agrobacterium tumefaciens strain GV3101, which contained the helper plasmid pJIC SA Rep (Höfgen & Willmitzer 1988).

Viral inoculation and detection

The N. tabacum leaves displaying TMV symptoms were homogenized in 0.1 M phosphate buffer (pH 7.2) using a sterile pestle and mortar, and the cell fragments were removed via centrifugation. Standardized volumes of infectious sap (or phosphate buffer as a control) were mechanically inoculated onto the upper surface of fully expanded N. glutinosa leaves sprayed with carborundum. The accumulation of virus in infected tissues was determined using reverse transcription (RT)-quantitative polymerase chain reaction (qPCR) for the TMV CP gene. Inoculated plants were grown in an insect-free greenhouse at 24℃, and the viral symptoms were monitored. Each experiment was replicated three times, including five independent plants.

Co-infiltration and GFP imaging

N. glutinosa plants at the five- or six-leaf stage were locally infiltrated with Agrobacterium GV3101, carrying the aforementioned constructs, using a previously described method (Brigneti et al. 1998). Each Agrobacterium culture (OD₆₀₀ = 1.0) was incubated for 3 h and then mixed with other culture (s) at a 1:1 (v/v) ratio before infiltration. Local and systemic RNA silencing was determined by observing GFP fluorescence in both infiltrated and newly emerging leaves under long-wavelength (365 nm) UV light (Spectroline Model SB-100P/A; Spectronics Corporation, Lexington, KY, USA) and photographed with a Fujifilm FinePix S8000fd digital camera (Fujifilm Holdings Corporation).

Virus‑induced gene silencing

For the NgRBP VIGS, TRV-based vectors were used as previously described (Chung et al. 2004). To ensure specific silencing of NgRBP, the full-length CDS with an ATG deletion was inserted into TRV2 to generate pTRV2:NgRBP. No homologous genes of NgRBP were found in N. glutinosa using National Center for Biotechnology Information (NCBI) BLAST. The pTRV1 and pTRV2:00 (empty vector) or pTRV2:NgRBP constructs were transformed into Agrobacterium GV3101. Agrobacterium cultures were resuspended in an infiltration buffer containing 10 mM MgCl₂, 10 mM MES, and 200 mM acetosyringone (pH 5.6; OD₆₀₀ = 0.8). Agrobacterium cultures harbouring pTRV1 were mixed with pTRV2:00 or pTRV2:NgRBP in a 1:1 ratio and infiltrated into the lower leaves of 3-week-old N. glutinosa plants. The infiltrated plants were placed in the greenhouse at 24℃ and 70% RH under a long-day (16 h light/8 h dark) photoperiod. At 7 dpi, empty vector control, and NgRBP-silenced plants were used for NgRBP expression analyses and TMV inoculation assays.

RT‑qPCR analysis

Total RNAs were isolated from leaves using TRIzol reagent (TaKaRa, Shiga, Japan) according to the manufacturer's instructions and treated with DNase I at 37°C for 30 min prior to reverse transcription. cDNA was synthesized from 1 μg of total RNA using the TIAN Script RT Kit (Tiangen, Beijing, China). RT-qPCR was performed using the Talent SYBR Green Kit (Tiangen, Beijing, China). The Actin1 gene was used as an internal reference. Each reaction was conducted in triplicate and repeated three times. The results were analyzed using the Bio-Rad CFX Manager software (Bio-Rad, California, USA).

Total RNA and siRNA Northern blot analysis

As described previously, total and low-molecular-weight RNAs were extracted from the leaves (Jing et al. 2011). From each sample, 20 µg whole RNA aliquots were separated on 1% formaldehyde agarose gels and transferred to Hybond-N⁺ membranes (GE Healthcare, Marlborough, MA, USA) by upward capillary transfer in 20× SSC buffer (3 M sodium chloride, 0.3 M sodium citrate, pH 7.0). The membranes were hybridized with digoxigenin (DIG)-labelled probes corresponding to the full-length ORFs of GFP and NgRBP, respectively. For siRNA detection, 15 µg low molecular weight RNAs were separated on a 5% polyacrylamide-7 M urea gel and transferred to a Hybond-N⁺ membrane in 0.5× TBE (44.5 mM Tris, 44.5 mM boric acid, 1 mM EDTA, pH 8.0) at 0.8 mA/cm² for 1 h. After being UV-crosslinked and incubated at 80°C for 2 h, the membrane was hybridized with a DIG-labelled probe of GFP mRNA. Chemiluminescent detection was performed using a DIG Northern Starter Kit (Roche, Basel, Switzerland), according to the manufacturer's instructions.