Search for a command to run...
Viral infections have a significant impact on plant growth and development and can lead to severe crop losses. In recent years, substantial progress has been made in understanding host–virus interactions, leading to the introduction of several highly effective countermeasures in agriculture. However, the evolution of viral virulence continues to exert pressure to better understand viral diseases. Plants possess a sophisticated, multitiered immune system that provides protection against invading pathogens, including viruses (Lopez-Gomollon & Baulcombe, 2022). The plant immune system, unlike that of animals, is fundamentally innate, consisting of two main layers: pattern-triggered immunity (PTI) and effector-triggered immunity (ETI). While PTI is activated by pathogen-associated molecular patterns (PAMPs), ETI is initiated by effectors that pathogens deploy to suppress PTI. Double-stranded RNA is a common PAMP generated during viral replication. In addition to PTI, it can also induce RNA silencing, which not only acts as a key gene regulatory mechanism but also plays an important antiviral role in plants. Unlike PTI and ETI, RNA silencing does not confer broad-spectrum resistance, as its antiviral effect is based on the sequence specificity provided by virus derived 21–24 nt RNAs (viral small interfering RNAs (vsiRNAs)). vsiRNAs are associated with Argonaute (AGO) proteins, which act as executors of antiviral RNA silencing. Two AGOs, AGO1 and AGO2, have been implicated most frequently in plant antiviral responses. AGO1, the prototype member of the AGO family, was isolated > 25 yr ago during a mutagenesis screen aimed at identifying novel genetic loci that control leaf development (Bohmert et al., 1998). Shortly after its discovery, AGO1 was recognized as a key regulator of plant gene expression, affecting almost all aspects of metabolism and development, and was shown to play a prominent role in various stress responses, including viral infections (Z.X. Zhao et al., 2023). Null mutants of AGO1 display a severe pleiotropic phenotype that is often lethal. Importantly, not only the absence of the protein but also its overproduction is detrimental to the plant. Disabling the evolutionarily conserved miR168-AGO1 autoregulatory feedback loop leads to overexpression of the protein, resulting in a severe phenotype that closely resembles null mutants of genes involved in miRNA biogenesis (Vaucheret et al., 2004). Thus, proper tuning of AGO1 expression appears to be of utmost importance for the plant. Initially, it was thought that AGO2 played a secondary role in antiviral protection, complementing the action of AGO1 (Harvey et al., 2011). However, increasing evidence suggests that AGO2 actually provides the primary defense of the plant against a number of viruses (Alvarado & Scholthof, 2012; Ludman et al., 2017, 2023; Brosseau et al., 2020). Unlike AGO1, mutation of AGO2 does not cause any phenotypic changes, suggesting that AGO2 is not involved in the regulation of plant growth and development (Harvey et al., 2011; Ludman et al., 2017; L. Zhao et al., 2023). In analogy to the miR168-dependent regulation of AGO1, AGO2 expression is thought to be modulated by miR403, providing a molecular basis for the hierarchical role of AGO1 and AGO2 in antiviral defense (Harvey et al., 2011). Despite its importance however, the details of this process have not yet been directly investigated. Therefore, we set out to explore the consequences of disabling this regulatory circuit in planta. CRISPR/Cas9 genome editing was used to mutate the miR403-binding site in the 3′UTR of the AGO2 gene in the virological model plant Nicotiana benthamiana. A biallelic T0 plant was obtained, carrying a T insertion and an 8-nt deletion in the miR403-binding site. Through self-pollination, this plant produced a homozygous line with an 8-nt deletion in the miR403-binding site (ΔmiR403bs line) (Fig. 1a). This mutation is expected to abolish the binding of miR403 to the 3′UTR of AGO2. Importantly, the mutant did not show any phenotypic alterations compared with the wild-type (WT) plant (Fig. 1b). Next, we monitored AGO2 expression in Potato virus X (PVX) infected and uninfected ΔmiR403bs plants using reverse transcription quantitative polymerase chain reaction (RT-qPCR) (Fig. 1c). WT plants were also used as controls in these experiments. AGO2 expression was very low in uninfected WT plants, but by 7 days post infection (dpi), the gene was robustly induced. At a later stage of infection however, AGO2 expression fell back to the uninfected level. Remarkably, ΔmiR403bs plants constitutively expressed the AGO2 gene at high levels, which was only moderately affected by virus infection. As a control, we also monitored the expression of AGO5, which has occasionally been found to exert complementary antiviral activity with AGO2 (Brosseau & Moffett, 2015; Ludman et al., 2023) (Fig. 1d). AGO5, unlike AGO2, showed identical expression dynamics in both plant genotypes: After a strong initial surge, gene expression declined to baseline in later stages of infection. Since AGO1 is directly involved in the regulation of AGO2, we also examined the expression of N. benthamiana AGO1 homeologs, but none showed substantial changes during PVX infection (Fig. 1e,f). AGO2 protein levels were also monitored by western blot analysis using a polyclonal rabbit antibody raised against the N terminus of N. benthamiana AGO2 (Fig. 1g). The specificity of the antibody was confirmed using ago2 protein lysate as a negative control. In WT plants, the AGO2 protein level was quite low and was only moderately increased by virus infection. Consistent with the strong constitutive expression of the gene, AGO2 protein level was high in ΔmiR403bs plants and was unaffected by virus infection. To exclude possible nonspecific expression changes associated with the inoculation method, the experiments were repeated using an alternative infection method (rubbing leaves with viral RNA containing extract instead of agroinfiltration). The two methods gave effectively the same results (Fig. S1). In conclusion, our findings clearly support that AGO2 expression is under tight control by a miR403-dependent posttranscriptional mechanism. Furthermore, it is worth noting that, unlike AGO1, high AGO2 levels are well-tolerated by the plant, apparently affecting neither its viability nor its phenotype. Since AGO2 plays a key role in plant defense against numerous viruses, we tested the susceptibility of ΔmiR403bs plants to viral infection compared with WT and ago2 plants. PVX accumulated at lower levels in inoculated and systemic leaves of ΔmiR403bs plants than in similar leaves of WT plants (Fig. 2a). Furthermore, by 28 dpi, ΔmiR403bs plants recovered more efficiently from PVX infection than WT controls, whereas ago2 plants necrotized, consistent with a previous report (Ludman et al., 2017). Turnip crinkle virus (TCV) reached apical leaves more slowly in ΔmiR403bs plants than in WT or ago2 controls, indicating that AGO2 plays a critical role in suppressing systemic spread of the virus. Furthermore, TCV-infected ΔmiR403bs plants showed milder symptoms than WT and especially ago2 plants (Fig. 2b). Next, we examined the ability of plants to recover from infection with Tomato bushy stunt virus (Fig. 2c). Since the p19 protein of Tombusviruses is a very strong silencing suppressor and can mask the antiviral activity of AGO2, we used the p19-deficient mutant virus (TBSVΔp19) instead of the WT for infection (Ludman et al., 2017). There was no difference in virus levels between WT and the ΔmiR403bs plants at the early stage of infection. As the disease progressed, generally less viral RNA was detected in the mutants than in the WT plants, but this difference did not appear to be statistically significant. Importantly however, ΔmiR403bs plants exhibited considerably milder symptoms than WT plants at 28 dpi. Consistent with the critical role of AGO2 in TBSV defense, ago2 mutants were necrotic by this time. Finally, we examined the dynamics of AGO2 expression in TCV- and TBSVΔp19-infected plants (Fig. S2). Both viruses were able to elicit strong transcriptional induction of AGO2, as evidenced by a robust increase in the gene's expression even in Δmir403bs plants, in which the miR403-dependent posttranscriptional regulation of AGO2 was disabled. In plants infected with TBSVΔp19, AGO2 expression was strongly reduced at the late stage of infection, similar to what was observed during PVX infection. Interestingly, in TCV-infected plants, AGO2 expression remained at high levels in both WT and Δmir403bs plants until the late stage of the disease. The observations described above raise the intriguing possibility that viral infections have a critical early phase when high AGO2 levels may positively influence the later outcome of the infection. The explanation for this phenomenon is unknown, but it can be assumed that early, robust silencing provides the plant with more time to more effectively activate other innate antiviral defense responses (Mandadi & Scholthof, 2013). Considering the previously described points, investigating the mutual dynamics of different antiviral mechanisms may be an important topic for future research. Notwithstanding, our results clearly demonstrate that specific inactivation of the miR403-dependent posttranscriptional regulation of the AGO2 gene results in robust, constitutive overexpression of the AGO2 protein, which confers enhanced resistance to viral infection. It is important to note that our results may also have significant practical implications. The miR403-dependent posttranscriptional regulation of AGO2 appears to be evolutionarily highly conserved in dicots (Jagtap & Shivaprasad, 2014). The miR403-binding site and its surrounding are particularly well-preserved in the 3′UTR of the AGO2 gene of solanaceous plants (Fig. S3). Given the emerging evidence that AGO2 is also an important component of antiviral defense of these plants, it is conceivable that overproduction of the AGO2 protein may enhance their resistance to viral infections. This can be achieved relatively easily with the approach presented here, which also has several additional advantages: (1) Since the AGO2 protein is produced by the plant's own gene, native cis-regulatory elements (including those yet to be identified) still regulate gene expression (with the exception of miR403 of course), ensuring a natural-like expression pattern; (2) as AGO2 is not overproduced from a transgene, the problem of silencing, which often thwarts similar approaches in plants, would not be an issue here; (3) the risk of random mutagenesis associated with the integration of the transgene into the chromosome can also be excluded; and (4) plant lines free of foreign DNA (the CAS9 transgene and the sgRNA expression cassette) can be easily obtained by backcrossing. Finally, it should be emphasized that AGO2 is involved not only in antiviral but also in antibacterial defense, as well as in thermal and genotoxic stress responses (Zhang et al., 2011; Wei et al., 2012; Zaheer et al., 2024). It will be an important topic for future studies to assess whether overexpression of AGO2 increases the resilience of plants to these biotic and abiotic stresses as well. Plasmids were constructed using standard techniques (Sambrook et al., 1989). For the generation of the ΔmiR403bs mutant N. benthamiana Domin line, the SaCas9-based editing system was employed as described previously (Ludman & Fátyol, 2021). The specific target regions were selected using the CCTop-CRISPR/Cas9 target online predictor tool (Stemmer et al., 2015). Structures of all constructs were verified by sequencing. Sequences of oligonucleotides used for vector construction and sequencing are listed in a Table S1. Genome-edited plant lines were generated using the traditional leaf-disk transformation protocol, as detailed elsewhere (Ludman et al., 2017, 2023; Ludman & Fátyol, 2021). Virus infections were performed as described before (Ludman et al., 2017, 2023; Ludman & Fátyol, 2021). Briefly, PVX and TCV infections were initiated by infiltrating the appropriate Agrobacterium strains (pGR106 or pCB301-TCV) (Jones et al., 1999; May et al., 2018) into N. benthamiana leaves. Alternatively, leaves were inoculated by rubbing with total RNA preparations extracted from infected plants. TBSVΔp19 inoculations were performed using in vitro transcribed full-length viral transcripts. All infections were repeated at least three times and representative results are presented. RNA samples were prepared from leaf tissues as described earlier (Ludman et al., 2017) and treated with Turbo DNase (ThermoFisher, Budapest, Hungary) according to the manufacturers' instructions. DNase-treated RNA was subsequently used as template for the production of cDNA employing the High-Capacity cDNA Reverse Transcription Kit (ThermoFisher). Using the generated cDNAs as templates, RT-qPCR analyses were performed with the FastStart Essential DNA Green Master kit (Roche) according to the manufacturers' instruction. Measurements were taken on a LightCycler 96 Instrument (Roche). AGO2, AGO5, AGO1A, AGO1B mRNA and virus RNA levels were determined using appropriate primers. Sequences of oligonucleotides used for the RT-qPCR analyses are listed in a Table S1. The measured AGO mRNA and virus RNA levels were normalized between samples by actin mRNA levels as internal control. Measurements were carried out in three biological replicates. Statistical significance of changes in RNA levels were determined using unpaired Student's t-test. Western analyses of proteins were carried out as described before (Fátyol et al., 2016). The PVX replicase-specific antibody was a kind gift of Kristiina Mäkinen (Merits et al., 1999). The N. benthamiana AGO2-specific antibody was generated by ImmunoGenes Ltd, Budakeszi, Hungary. Briefly, genetically modified rabbits overexpressing the neonatal Fc receptor (Bender et al., 2007) were injected subcutaneously four times with the Keyhole Limpet Hemocyanin conjugated C-Ahx-ADLVAYRGRMFQEVLMEMQSP-amide peptide and adjuvant. Serum from rabbits bled on day 70 was used in our experiments. This work was supported by the National Research Development and Innovation Office, Hungary (K142626 and STARTING152648). None declared. KF planned and designed the research. ML and KF performed experiments and analyzed data. KF and ML obtained grant support. KF wrote the manuscript. The authors declare that all data supporting the results of this study are available in the article and in Supporting Information (Figs S1–S3; Table S1). Fig. S1 Testing the susceptibility of plants to PVX infection. Fig. S2 Measurement of AGO2 mRNA levels in virus infected plants. Fig. S3 Multiple sequence alignment of the 3′UTR of the AGO2 genes of solanaceous plants. Table S1 Sequences of oligonucleotides used in this study. Please note: Wiley is not responsible for the content or functionality of any Supporting Information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. The New Phytologist Foundation remains neutral with regard to jurisdictional claims in maps and in any institutional affiliations.