The Pythium periplocum elicitin PpEli2 confers broad-spectrum disease resistance by triggering a novel receptor-dependent immune pathway in plants

Abstract Elicitins are microbe-associated molecular patterns produced by oomycetes to elicit plant defense. It is still unclear whether elicitins derived from non-pathogenic oomycetes can be used as bioactive molecules for disease control. Here, for the first time we identify and characterize an elicitin named PpEli2 from the soil-borne oomycete Pythium periplocum, which is a non-pathogenic mycoparasite colonizing the root ecosystem of diverse plant species. Perceived by a novel cell surface receptor-like protein, REli, that is conserved in various plants (e.g. tomato, pepper, soybean), PpEli2 can induce hypersensitive response cell death and an immunity response in Nicotiana benthamiana. Meanwhile, PpEli2 enhances the interaction between REli and its co-receptor BAK1. The receptor-dependent immune response triggered by PpEli2 is able to protect various plant species against Phytophthora and fungal infections. Collectively, our work reveals the potential agricultural application of non-pathogenic elicitins and their receptors in conferring broad-spectrum resistance for plant protection.


Introduction
Phytophthora is an oomycete genus causing some of the most notorious plant diseases in the world [1][2][3][4]. For example, Phytophthora capsici is a highly dynamic and destructive pathogen, which attacks important vegetables such as pepper, tomato, eggplant, and cucurbits [5]. Phytophthora parasitica causes the notable tobacco diseases of brown rot, foot rot, and black shank, which affect almost all tobacco-growing areas worldwide, with losses as high as 100% [6]. Phytophthora sojae infection leads to soybean root rot, causing an annual loss of $1-2 billion worldwide [7,8]. Chemical pesticides and plant disease resistance (R) genes can be used to defend Phytophthora pathogens. However, Phytophthora pathogens evolve rapidly to overcome fungicides and R genemediated resistance, and this leads to a continuous demand for sustainable approaches to confer broad-spectrum resistance (BSR) against Phytophthora and other pathogens.
Development of novel Phytophthora control methods relies on the dissection of the plant immune system [9]. Plants use pattern-recognition receptors (PRRs) at the cell surface to perceive evolutionarily conserved microbe-or pathogen-associated molecular patterns (MAMPs or PAMPs) [10,11], and thereby trigger immunity (MAMP-or PAMP-triggered immunity, MTI or PTI) [12, (BAK1) and SUPPRESSOR OF BAK1-INTERACTING RECEPTOR-LIKE KINASE 1 (SOBIR1) [16,35]. BAK1 is involved in both RLK and RLP signaling [36], while SOBIR1 mainly regulates RLPmediated signal transmission [35]. Some PRRs are characterized only in a narrow range of plant species. For example, EFR exists exclusively in Brassicaceae species. ELICITIN RESPONSE (ELR) is identified specifically from Solanum microdontum genotype mcd360-1 [29]. Ectopic expression of EFR in tomato and Nicotiana benthamiana confers BSR against bacteria despite plant species or genotype specificity [37]. Introduction of ELR into cultivated potato varieties enhances their resistance to P. infestans. These observations suggest that PRRs and the PAMPs/MAMPs they recognize are valuable resources for developing disease-resistant crops.
In recent decades, crop disease control has mainly depended on the use of chemical pesticides. However, several pathogens have evolved resistance to frequently used pesticides [38]. In addition, some pesticides are not readily broken down into simple and safer constituents, and eventually exist as toxic residues in the soil [39]. Increasing public concerns on environmental and health issues associated with synthetic chemicals are causing a shift towards more sustainable and eco-friendly disease management practices [38,40]. Given this situation, a large number of biological control agents have been developed and applied [41]. PRRs and MAMPs are potential biocontrol agents for conferring BSR against oomycete and fungal pathogens.
MAMPs from non-pathogenic microbes are of particular interest for biocontrol. The mycoparasite Pythium periplocum and its sister species Pythium oligandrum are non-pathogenic oomycetes that colonize the plant root ecosystem of a large number of plant species [42]. Distinguished from pathogenic oomycetes, P. oligandrum triggers plant immune responses through its MAMPs, and produces the auxin precursor tryptamine to promote plant growth. P. oligandrum has been successfully used as a biocontrol agent for controlling various plant diseases, including those caused by Phytophthora pathogens [42][43][44]. MAMPs identified in P. oligandrum include POD-1/2, NLPs, and oligandrins (Oli-D1 and Oli-D2) classified as elicitins [45][46][47]. Elicitins are structurally conserved extracellular proteins found in Phytophthora and Pythium species. Their encoding genes occur as complex multigene families and form diverse subclasses of elicitin and elicitin-like genes [48]. All elicitins share a highly conserved 98-amino-acid (aa) domain with six cysteine residues forming three disulfide bridges. Elicitin-like proteins also have six structurally conserved cysteine residues, but the lengths of their elicitin domains and cysteine-spacing patterns vary within each phylogenetic clade member [48,49]. Elicitins are able to trigger hypersensitive response (HR) cell death, ROS burst, and disease resistance in several plants [48]. LRR-RLP ELR from S. microdontum has been identified as a PRR to recognize the P. infestans elicitin INF1 [29]. However, elicitin PRRs in other plant species are largely unknown.
Here, we identified a novel elicitin-like PpEli2 from P. periplocum. PpEli2 induces HR and ROS burst, which enhance plant BSR to multiple Phytophthora and fungal pathogens. We further demonstrated that an N. benthamiana LRR-RLP mediates the perception of PpEli2, and named it REli (Receptor of PpEli2). REli is associated with NbBAK1 and NbSOBIR1. PpEli2 enhances the interaction between REli and BAK1 in planta. Ectopic expression of PpEli2 is able to enhance BSR against various Phytophthora and fungal pathogens in tomato, pepper, and soybean. Our work demonstrates that PpEli2 is a novel MAMP with promising biocontrol applications.

PpEli2 induces cell death and plant resistance in N. benthamiana
To identify putative elicitins in P. periplocum, we used a hidden Markov model of the conserved elicitin domain (Pfam PF00964) to search the proteome of P. periplocum strain CBS 532.74. Meanwhile, the SignalP v3.0 program was used to identify the N-terminal signal peptides. In total, 40 elicitin-like genes, but no elicitin gene, were found in the P. periplocum genome. The encoded proteins were named PpEli1-40, with none of their elicitin domains being the typical 98 aa in length (Supplementary Data Table S1). All 40 PpEli genes were cloned and then transiently expressed in leaves of N. benthamiana for cell death screening. Five PpElis were found to induce cell death in N. benthamiana (Supplementary Data Fig. S1). Among them, PpEli3 is a homolog of Oli-D1/D2 from P. oligandrum. PpEli10-12 encode identical aa sequences. PpEli2, PpEli3, and PpEli10-12 are evolutionarily separated (Fig. 1A).
PpEli2 was chosen for further investigation due to its strong cell death-and ROS-inducing activity ( Fig. 1B-D). In relevant assays, we used two elicitins, P. infestans INF1 [50] and P. oligandrum Oli-D2, as two positive controls, and GFP as a negative control. To evaluate plant immunity responses in N. benthamiana, we examined the expression levels of two cell death markers, HSR203J and HARPIN-INDUCED 1 (HIN1) [46,51,52], as well as two PTI markers, NbCYP71D20 and NbPTI5 [9,53]. Quantitative reverse transcription polymerase chain reaction (qRT-PCR) analysis showed that transient expression of PpEli2 significantly upregulated all four marker genes ( Fig. 1E and F), indicating its strong immune response-inducing activity. Furthermore, PpEli2 expression was induced during the interaction between P. periplocum and N. benthamiana (Fig. 1G). Ectopic expression of PpEli2 significantly reduced P. capsici colonization in N. benthamiana. We used Agrobacterium tumefaciens-mediated transient expression technology to express GFP and PpEli2 in N. benthamiana; then, the leaves were inoculated with P. capsici mycelia plugs at 24 h after infiltration; the lesion and P. capsici biomass were measured after 36 hours of wet culture at 25 • C in the dark (Fig. 1H). Similarly, PpEli2 also conferred resistance of N. benthamiana to P. parasitica (Fig. 1I). Together, the results demonstrate that PpEli2 is able to enhance resistance against various Phytophthora pathogens.

PpEli2 triggers MAMP-triggered immunity and broad-spectrum resistance in N. benthamiana
Purified recombinant PpEli2 protein produced from Escherichia coli strain BL21 (DE3) ( Fig. 2A, Supplementary Data Table S2) significantly induced cell death and H 2 O 2 accumulation in N. benthamiana leaves (Fig. 2B). Protein infiltration assays showed that the HR cell death-inducing ability of PpEli2 was positively correlated with protein concentration (Fig. 2C), negatively correlated with temperature, and decreased gradually over time that can be still induce HR saved after >1 month at 4 • C and 2 weeks at 25 • C (room temperature) ( Fig. 2D and E). Meanwhile, we found PpEli2 was able to induce cell death and H 2 O 2 accumulation under different pH values (Fig. 2F). In a luminol-based chemiluminescence assay, PpEli2 induced a rapid ROS burst in N. benthamiana leaves (Fig. 3A). Furthermore, immunoblotting using the phosphor-p44/42 MAPK antibody showed that, like the classic PAMP f lg22, PpEli2 induced MAPK activation in N. benthamiana within 30 minutes (Fig. 3B). Both assays demonstrated that PpEli2 is able to trigger PTI in plants. To further examine PpEli2-mediated protection against oomycete and fungal pathogens, leaves of N. benthamiana were Agroinfiltrated leaves were collected at the indicated time points to analyze the expression levels of NbHIN1, NbHSR203J, NbPTI5, and NbCYP71D20 by qRT-PCR. GFP was used as a negative control. Error bars represent mean ± standard deviation (n ≥ 8). (G) Time-series expression profile of PpEli2 after leaf inoculation with P. periplocum. Total RNAs were extracted from P. periplocum mycelia (0 hpi) or inoculated N. benthamiana leaves at 3-72 hpi. Transcript accumulation levels of PpEli2 during plant-P. periplocum interaction was determined by qRT-PCR. The P. periplocum Actin gene was used as the housekeeping gene. Error bars represent mean ± standard deviation (n = 8; n represents sample number). (H, I) PpEli2 enhanced plant resistance of N. benthamiana leaves to P. capsici (H) and P. parasitica (I) infections. Scale bar = 1 cm. Relative biomass was determined by qPCR using references P. capsici or P. parasitica Actin and N. benthamiana EF1α. A representative immunoblot shows the protein levels of transiently expressed GFP and PpEli2 in planta. Error bars represent mean ± standard deviation (n ≥ 10). Data were analyzed by the Shapiro-Wilk test to determine normality and log normality across various groups. Groups passing the normality test were then analyzed by the unpaired t test ( * * , P < .01). These experiments were performed with least three biological replicates, with similar results. pretreated with either 200 nM PpEli2 or GFP 24 h before the inoculation of P. capsici mycelium plugs (Fig. 3C). Pretreatment with PpEli2 significantly reduced lesions and pathogen biomass compared with GFP (Fig. 3D). Similarly, PpEli2 enhanced N. benthamiana resistance against P. parasitica and Sclerotinia sclerotiorum (Fig. 3D). Together, these results suggest that PpEli2 protein is able to induce plant resistance rapidly and can potentially be used as a bioactive molecule.

PpEli2-mediated immunity requires REli and co-receptors BAK1 and SOBIR1
To identify PpEli2 perception PRRs, we screened LRR-RLK/RLP genes in N. benthamiana using a high-throughput Tobacco rattle virus (TRV)-induced gene silencing (VIGS) assay. The results showed that PpEli2-induced cell death was significantly impaired in an LRR-RLP silencing line (Niben101Scf02826g01005.1), and the corresponding protein was defined as REli ( Fig. 4A-C,  Fig. S2). The PpEli2-induced ROS burst was also significantly weakened in REli-silenced plants using a luminol-based chemiluminescence assay (Fig. 4D). PpEli2induced cell death was suppressed in REli-silenced plants but could be recovered by co-overexpressing the full-length REli gene as complementation ( Fig. 4A-C, Supplementary Data Fig. S3). Furthermore, PpEli2-mediated resistance to P. capsici was impaired in REli-silenced N. benthamiana ( Fig. 4E and F). Collectively, these results suggest that REli mediates a PpEli2-triggered immune response, likely by serving as its recognition receptor.
BAK1 and SOBIR1 are two co-receptor PRRs for MAMP recognition [54]. To test their potential roles in PpEli2 perception, we silenced BAK1 and SOBIR1 individually via TRV-based VIGS in N. benthamiana, and then infiltrated GFP, PyolNLP7, or PpEli2 protein in these silenced plants after 3 weeks. Cell death and H 2 O 2 accumulation triggered by PpEli2, but not PyolNLP7, were abolished in plants with the silencing of either BAK1 or SOBIR1 (Fig. 4A-C), which demonstrates that they are both required for the PpEli2triggered immune response.

REli binds to PpEli2 and co-receptors BAK1 and SOBIR1
To test physical interaction between PpEli2 and REli, C-terminal Flag-tagged PpEli2 and HA-tagged REli were generated to perform Scale bar = 1 cm. Lesion diameters and relative biomass were measured at the indicated time points. Error bar represents mean ± standard deviation (n ≥ 9). Data were analyzed by the Shapiro-Wilk test to determine normality and log normality across groups. Groups passing the normality test were then analyzed by the unpaired t test ( * * , P < .01). These experiments were performed three times with similar results. a co-immunoprecipitation assay in N. benthamiana. The result showed that PpEli2 can bind REli in planta (Fig. 5A). We also showed that REli binds to its two co-receptors, BAK1 and SOBIR1 (Fig. 5B,C), with REli-BAK1 interaction being significantly enhanced by PpEli2 treatment (Fig. 5B). According to the results of protein-protein interaction, we speculate a model that REli recruits BAK1 and SOBIR1 to form a complex for PpEli2 perception.

PpEli2 confers broad-spectrum resistance against filamentous pathogens in various plant species
We identified REli orthologs in soybean (Glycine max), tomato (Solanum lycopersicum), and pepper (Capsicum annuum) via a BLASTP search (Fig. 6A). We propose that PpEli2 can improve BSR in various plant species by interacting with REli orthologs. To test this hypothesis, a luminol-based chemiluminescence assay was conducted firstly to show that PpEli2 induced a (F) P. capsici-caused disease lesions and relative biomass of P. capsici were measured at the indicated time points. Error bars represent mean ± standard deviation (n ≥ 11). Data were analyzed by the Shapiro-Wilk test to determine normality and log normality across groups. Groups passing the normality test were then analyzed by one-way ANOVA with the post hoc Dunnett's multiple comparisons test ( * * , P < .01; ns, no significant difference). All the experiments were performed with at least four biological replicates. significant ROS burst in tomato, pepper, and soybean leaves (Fig. 6B-D). In China, Hang pepper and cherry tomato are important vegetables suffering substantial losses from P. capsici infection, and thus we assessed controlling activity of PpEli2 against the leaves and fruit rot. Tomato and pepper leaves and fruits were sprayed with 6 μM PpEli2 protein, and then inoculated with P. capsici mycelium plugs at 2 hours post-treatment (Fig. 7A). Compared with the negative control, PpEli2-treated leaves had reduced disease lesions ( Fig. 7B and C), tomato fruits exhibited lower disease indexes ( Fig. 7D and E), and pepper fruits showed smaller disease lesions and attenuated water-soaking symptoms ( Fig. 7F and G).
Furthermore, we investigated the ability of PpEli2 to confer P. sojae resistance in soybean. Etiolated soybean seedlings were incubated in solution containing 6 μM GFP or PpEli2 for 6 hours, and subsequently challenged with P. sojae for 36 hours (Fig. 8A). Compared with the GFP control, pretreatment of PpEli2 significantly reduced disease lesions and P. sojae biomass on soybean hypocotyls (Fig. 8B-D). Together, our results demonstrate that PpEli2 can confer BSR against Phytophthora pathogens in various plant species. The ability of PpEli2 to confer resistance against the notorious fungal pathogen S. sclerotiorum, which has a broad host range that includes several important plants [55][56][57], was also assessed. Tomato, pepper, and soybean leaves were pretreated with either PpEli2 or GFP followed by S. sclerotiorum mycelium inoculation for 48 hours. PpEli2 significantly reduced leaf lesions caused by S. sclerotiorum in all three plants (Fig. 8E and F), suggesting that PpEli2 can also confer fungal resistance in various vegetable plants.

Discussion
Phytophthora disease management is an ongoing challenge despite significant improvements in chemical pesticides and plant resistance gene breeding. Phytophthora is notoriously adept at overcoming these control strategies [58,59]. Pesticide residues negatively affect food safety, human health, and the environment [60]. The long cycle time in conventional breeding limits its contribution to Phytophthora disease control. A bottleneck in the genetic engineering approach is the availability of promising transgenic candidates that can confer high-level BSR with minimal side effects. Given this situation, biological control, as an environmentfriendly disease management method, has received remarkable attention in recent years. P. periplocum is potentially a promising biocontrol agent due to its aggressive parasitism on numerous plant pathogenic fungi and oomycetes [61]. In this study, we for the first time identify the elicitin-like PpEli2 as a P. periplocum MAMP and demonstrate its suppression effect on Phytophthora and fungal pathogens in planta.
We dissected the PpEli2-triggered plant immune pathway in detail using recombinant PpEli2 protein produced by E. coli. Significantly upregulated during plant-P. periplocum interaction, PpEli2 is directly perceived by a novel LRR-RLP, REli, which is the first elicitin receptor identified in N. benthamiana and is conserved in various vegetable crops. It is likely that successful PpEli2-REli interaction requires proper protein folding, which may explain the observations that PpEli2 stability and function are temperatureand pH-sensitive. REli interacts with co-receptors BAK1 and SOBIR1 to transduce immune signals, which is common for LRR-RLPs lacking the cytoplasmic kinase domain [62][63][64]. For example, perception of Valsa mali PAMP VmE02 requires its RLP receptor Error bars represent mean ± standard deviation (n ≥ 10). (G) Disease severity of pepper fruit rot at 72 hpi. Data were analyzed by the Shapiro-Wilk test to determine normality and log normality across groups. Groups passing the normality test were then analyzed by the unpaired t test ( * * , P < .01). These experiments were performed with five biological replicates, with similar results.
RE02 to form a complex with the two RLKs SOBIR1 and BAK1 [65]. Interestingly, PpEli2 significantly enhances the interaction between REli and BAK1. Downstream events of PpEli2-REli, REli-BAK1, and REli-SOBIR1 interactions include HR cell death, ROS burst, and enhanced plant resistance to various filamentous pathogens.
In the plant apoplastic space between plant-microbe interaction, the natural concentration of MAMPs is usually low, at which the molecule may not cause HR. Here, to screen and identify the objective proteins efficiently, we used the overexpression system in N. benthamiana, and then confirmed the phenotype with prokaryotic expression in E. coli. We found that PpEli2 induces HR-cell death in a concentration-dependent manner; however, a lower concentration of PpEli2 protein that failed to induce cell death still induced plant immunity, suggesting that PpEli2induced cell death is not essential for the protective effect of PpEli2 in plants against Phytophthora. In addition, the higher concentrations of PpEli2 may cause side effects in the crop product. Here, we found the spray of PpEli2 protein did not cause any obvious phenotype in tomato, pepper, and soybean, and therefore the treatment of plants at a reasonable concentration will avoid adverse effects on plants.
For resistance breeding, BSR is a desirable trait as it confers resistance against more than one pathogen species or against the majority of races or strains of the same pathogen [66]. PpEli2 confers resistance to diseases caused by two Phytophthora species (P. capsici and P. sojae) and a fungus (S. sclerotiorum). The dicotyledonous crops tested include tomato, pepper, and soybean. It is likely that testing more vegetable crops and oomycetes/fungi could further expand the resistance spectrum of PpEli2 on both the plant and the pathogen side. Disease resistance in transgenic plants can be established by expressing either elicitors [67] or receptors [29]. Both the elicitin PpEli2 and the receptor REli characterized in this study are promising candidates for the development of resistant cultivars. Given the sensitivity of PpEli2 to high temperature (i.e. >37 • C), its protein sequence could be engineered to improve tertiary structure stability but still retain high binding affinity to REli. This enhanced version of PpEli2 would be an ideal biocontrol agent conferring BSR in real-world field conditions. GFP was used as a negative control. Error bars represent mean ± standard deviation (n ≥ 12). Data were analyzed by the Shapiro-Wilk test to determine normality and log normality across groups. Groups passing the normality test were then analyzed by the unpaired t test ( * * , P < .01). All the experiments were performed with five biological replicates.
In conclusion, our work shows that the P. periplocum MAMP PpEli2 significantly improves BSR to Phytophthora and fungal pathogens in N. benthamiana, tomato, pepper, and soybean. REli, a novel LRR-RLP receptor, recognizes PpEli2 and forms a complex with two co-receptors, BAK1 and SOBIR1, to trigger the downstream immune response. PpEli2 can potentially be used as a bioactive molecule to control plant diseases in an environmentfriendly manner. PpEli2 induces HR and immunity in a dosagedependent manner.

Plasmid construction
The binary vector pBINHA was used for Agrobacterium-mediated transient gene expression. Genes were inserted via the SmaI site. For VIGS assays, pYL156 BamHI and EcoRI sites were used for gene cloning. For prokaryotic expression, genes were inserted into pET32a via BamHI and XhoI sites. Gene sequences used in this study are listed in Supplementary Data Table S1. Primers used in this study are listed in Supplementary Data Table S3.
To evaluate disease resistance in agroinfiltrated N. benthamiana, leaves were inoculated with P. capsici mycelium plugs at 24 hours after infiltration. Disease lesions were measured at 36 hours post-infiltration (hpi) [70]. Agroinfiltrated leaves were collected at 48 hours after infiltration for western blot assays. Relative biomass of P. capsici in infected leaves was determined by qPCR as previously described [71]. Disease resistance was also assessed after recombinant protein infiltration. Each N. benthamiana leaf was infiltrated with 200 nM GFP control on one half and the same concentration of PpEli2 on the other half. Methods used for P. capsici inoculation assay and disease development evaluation were similar to those used for transient expression in N. benthamiana [53,63].
In spray inoculation tests of tomato and pepper, equal amounts of recombinant protein (6 μM) were sprayed evenly on leaves and fruits. P. capsici mycelium plugs were inoculated into leaves 2 hours after spraying. Disease lesion diameters or lengths were measured at designated time points after inoculation. The disease index and disease scores were calculated and classified as previously described [47,[72][73][74]. P. parasitica and S. sclerotiorum inoculation methods were similar to that for P. capsici. For P. sojae inoculations, 4-day-old etiolated soybean seedling hypocotyls were kept in purified protein solution for 6 hours without light and then inoculated with P. sojae mycelium as previously described [68]. Soybean leaves from 10-day-old plants were challenged with S. sclerotiorum for 48 hours.

Agrobacterium-mediated transient expression and VIGS in N. benthamiana
Leaves of 4-to 6-week-old N. benthamiana leaves were agroinfiltrated with Agrobacterium containing the constructs and the P19 silencing suppressor in a 1:1 ratio with optical density (OD) of 0.3 at 600 nm [75][76][77]. For VIGS assays, Agrobacterium cells with the pTRV1, pTRV2:BAK1, pTRV2:SOBIR1, or pTRV2:REli construct were harvested and resuspended in infiltration buffer to an OD of 0.8. A mixture of Agrobacterium cultures containing the pTRV1 and pTRV2 constructs was injected into two primary leaves of a four-leaf-stage plant [53,63].

Prokaryotic expression of recombinant proteins and stability evaluation
The coding sequence of PpEli2 was amplified and cloned into the pET32a vector. Then, pET32a-PpEli2 was transformed into E. coli strain BL21, Rosetta, or BL21(DE3). Protein expression was induced by 0.1 mM isopropylthio-β-galactoside (IPTG) and incubation at 16-37 • C for 12 hours [78]. ImageJ software was used to quantify protein concentrations in SDS-PAGE gel bands as described previously [79]. For the evaluation of protein stability, the recombinant protein was incubated at 4-75 • C for between 3 hours and 2 months before testing its activity.

Co-immunoprecipitation assay
Proteins were extracted using HEPES extraction with protease inhibitor cocktail (Roche) [80], and then incubated with α-HA beads (Abmart) for 8-10 h. The beads were collected and then washed six times with 1 × TBS at 4 • C. Proteins were released from the beads by incubating at 100 • C for 5 minutes with 1 × TBS. Immunoprecipitates were separated by SDS-PAGE gels and then immunoblotted with anti-Flag (Sungene) or anti-HA (Sungene). The ECL Western Blotting Detection Kit (GE) was used to detect the blots [22,63].

Electrolyte leakage assay and DAB staining
N. benthamiana leaf disks were soaked in 5 ml of distilled water at room temperature for 2 hours. The conductivity of the bathing solution was then measured using a conductivity meter (Con 700, Consort) and calculated as previously described [81]. For DAB staining, N. benthamiana leaves were stained with 1 g/l DAB solution for 8 hours in the dark, and decolored with ethanol [2].

Statistical analysis
GraphPad Prism 8.3.0 software was used for statistical analysis of all data. The Shapiro-Wilk test was used to determine normality and log normality across groups. Groups passing the normality test were analyzed by either one-way ANOVA with the post hoc Dunnett's multiple comparisons test or the unpaired t test. Results are expressed as the mean ± standard deviation of replicates.