The Conserved Guanyl‐Specific Ribonuclease T1 Effector EC19 Is Required for Full Virulence of the Vascular Wilt Fungus Fusarium oxysporum on Tomato

ABSTRACT

Plant fungal pathogens secrete a plethora of effectors into host cells to facilitate their infection by interfering with the normal physiological processes of the host plant. However, the role of guanyl‐specific ribonuclease T1 effector protein in the virulence of fungal pathogens remains largely unexplored. Here, we show that an effector protein EC19, secreted by the soil‐borne fungus Fusarium oxysporum f. sp. lycopersici (Fol) that causes tomato wilt disease, contributes to the full virulence of Fol. In‐locus knockout of EC19 significantly reduced the virulence of Fol, while complementing EC19 in the knockout mutant restored the pathogenicity of Fol on tomato. EC19, which contains the ribonuclease T1‐like domain, is highly conserved across fungal species and exhibited the ability to cleave single‐stranded RNA isolated from tomato. EC19 was specifically induced during Fol infection. The AlphaFold3‐predicted structure of EC19 shows a high degree of structural similarity with ribonuclease T1 from Aspergillus oryzae, with limited similarity to RNase T2. Using Agrobacterium‐mediated transient assay in the leaves of Nicotiana benthamiana, we found that EC19 predominantly localised to the nucleus of plant cells. Stable transgenic susceptible tomato plants overexpressing EC19 exhibited increased susceptibility to Fol infection compared to wild‐type plants. RNA‐seq analysis of EC19‐overexpressing tomato lines revealed that downregulated differentially expressed genes were predominantly enriched in plant hormone signal transduction, phenylpropanoid biosynthesis, and plant–pathogen interaction pathways. Taken together, these findings indicate that EC19 with ribonuclease activity promotes Fol infection, expanding our understanding of the molecular mechanisms underlying Fol virulence.

EC19, a secreted ribonuclease T1 effector from Fusarium oxysporum f. sp. lycopersici (Fol), is required for tomato wilt virulence. EC19‐overexpressing tomato lines display increased susceptibility to Fol and altered transcriptional changes associated with plant hormone signal transduction, plant–pathogen interaction, and phenylpropanoid biosynthesis pathways.

1. Introduction

To counteract pathogen invasion, plants have evolved sophisticated defence mechanisms that integrate multiple physical barriers as the first line of defence with the activation of multiple‐layered immune responses (Cao et al. 2018). However, plant pathogens employ sophisticated strategies to break the first physical barriers and inhibit plant immune responses. These result in severe disease occurrence and yield losses, thereby posing a significant threat to global food security and ecosystem stability (Jones and Dangl 2006). Commonly, plant pathogens secrete specific enzymes and effector proteins to facilitate their infection. Bacteria, fungi, and oomycetes can secrete cell wall‐degrading enzymes (CWDEs) to break down plant cell walls (Cao et al. 2018). For example, the soft rot‐causing bacterial pathogen Dickeya zeae requires CWDEs for virulence (Hu et al. 2022). Glycoside hydrolase 12 from Phytophthora sojae and Verticillium dahliae has xyloglucanase activity to degrade xyloglucan, a major hemicellulosic polysaccharide in the primary cell walls, and is required for full virulence (Gui et al. 2017; Ma, Song, et al. 2015). In Botrytis cinerea, the deletion of the xylanase‐coding gene Xyn11A impairs its virulence by reducing the lesion size (Brito et al. 2006; Noda et al. 2010). Plant pathogens also secrete effectors to target host diverse physiological processes, such as RNA silencing and cell signal transduction, and effectively inhibit plant immune responses (Li, Newman, et al. 2024; Wang et al. 2022).

Ribonucleases (RNases) are a class of ribonucleolytic proteins with the ability to cleave and hydrolyse RNA into 3′‐mononucleotides, resulting in the inhibition of protein synthesis and cell death caused by apoptosis (Deshpande and Shankar 2002; Lacadena et al. 2007). Based on their molecular weight and substrate specificity, RNases can be classified into three families: RNase A, RNase T1, and RNase T2 (Luhtala and Parker 2010). RNase T1, secreted by fungi, specifically hydrolyses single‐stranded RNA at the 3′‐end of guanosine residues, participating in nutrient recycling and microbial interactions (Yoshida 2001; Lacadena et al. 2007). The RNase T2 family is present in plants, fungi, animals, and some bacteria and degrades internal phosphodiester bonds in single‐stranded RNA under acidic pH conditions (Deshpande and Shankar 2002; Lacadena et al. 2007). Several studies have reported that the RNase T2 secreted by fungal pathogens contributes to their growth and pathogenicity. For example, RNase T2 FocRnt2, secreted by Fusarium oxysporum f. sp. cubense (Foc) tropicalrace 4 causing banana Fusarium wilt, exhibits in vitro ribonuclease activity to degrade banana RNA; knockout of FocRnt2 inhibits Foc penetration and reduces Foc virulence on banana plants (He et al. 2024). In F. oxysporum f. sp. lycopersici (Fol), the secreted FoRnt2, belonging to the RNase T2 family, degrades host RNA in vitro, enhances host susceptibility, and contributes to fungal virulence without impacting fungal growth or sporulation (Qian et al. 2022). A secreted ribonuclease Zt6 from Zymoseptoria tritici, the causal agent of Septoria tritici blotch in wheat, exhibits dual functionality by possessing catalytic activity to degrade host rRNA as well as contributing to antimicrobial competition (Kettles et al. 2018). Two RNase T2‐type ribonucleases, Nuc1 and Nuc2, from the corn smut fungus Ustilago maydis degrade maize apoplastic RNA; the deletion mutants exhibit reduced virulence and delayed life cycle progression (Mukherjee et al. 2020). Interestingly, a secreted RNase F1 protein, SRE1, from Setosphaeria turcica contributes to fungal virulence and induces cell death in the leaves of maize and Nicotiana benthamiana through its enzymatic activity, while also enhancing N. benthamiana resistance to pathogenic fungi and oomycetes, suggesting a dual role in promoting infection and modulating host immunity (He et al. 2022). Fusarium graminearum virulence effector Fg12, a conserved fungal‐specific ribonuclease, manipulates the rhizosphere microbiome to promote Fusarium infection; guanosine monophosphate acts as an effective inhibitor by suppressing the antibacterial activity of Fg12 and mitigating disease in soybean and alfalfa (Yang et al. 2025). However, the virulence role of the secreted RNase T1 family protein from fungal pathogens remains largely unexplored.

Fol, the causal agent of tomato Fusarium wilt, is a globally significant soil‐borne pathogen that threatens tomato production and causes substantial yield losses (Adhikari et al. 2020; Li et al. 2026). Following root penetration, Fol preferentially colonises xylem vessels, blocking water and nutrient transport and ultimately causing leaf dehydration, wilting, stem lodging, and plant death (Srinivas et al. 2019; Yadeta and Thomma 2013). To date, the deployment of resistance genes in tomato represents an effective strategy to control Fusarium wilt. However, Fol secretes a variety of effectors into tomato cells to suppress plant immunity and modulate plant physiological processes, thereby promoting infection. For example, the Secreted‐in‐xylem 4 (SIX4) effector can suppress I‐2‐ and I‐3‐mediated resistance in tomato (Houterman et al. 2008). The effector protein SIX6 can inhibit I‐2‐mediated cell death in the leaves of N. benthamiana and is required for the virulence of Fol infecting tomato (Gawehns et al. 2014). Effector SIX8 interacts with tomato TPL1 and TPL2 to promote Fol infection, while loss‐of‐function TPL1 and TPL2 double mutants exhibit durable and non‐race‐specific resistance to Fol (Aalders et al. 2024). The effector FolSvp2 uses phase separation to sequester tomato iron–sulfur protein SlISP, thereby suppressing reactive oxygen species (ROS) production and promoting infection, while the tomato protein SlPR1 counters this process by blocking FolSvp2 entry into host cells (Li, Yang, et al. 2024). The effector FolSvp1, which is acetylated at K167 by the lysine acetyltransferase FolArd1 to prevent ubiquitination‐dependent degradation, translocates tomato SlPR1 from the apoplast to the nucleus, thereby blocking CAPE1 generation and promoting pathogen invasion (Li et al. 2022). The secreted aspartic protease effector FolAsp localises to the apoplast of plant cells, where it suppresses plant immunity, thereby enhancing fungal virulence without impacting growth or conidiation (Wang et al. 2023).

Our previous study demonstrated that the candidate effector protein EC19 from Fol possesses a functional signal peptide and that full‐length EC19 can induce cell death in the leaves of N. benthamiana (Sun et al. 2022). In this study, we further characterise the virulence function of EC19 following Fol infection and uncover its role in promoting virulence. We demonstrate that EC19 functions as a ribonuclease effector that likely regulates multiple biological pathways in tomato.

2. Results

2.1. In‐Locus Knockout of EC19 Compromises the Virulence of Fol

To examine the role of EC19 during Fol007 infectiion of tomato, three independent in‐locus EC19 knockout mutants (ΔEC19) were obtained using homologous recombination and protoplast‐mediated transformation (Figure S1). No growth differences in colony morphology were observed between the wild‐type (WT) and ΔEC19 mutant strains (Figure 1A). Pathogenicity assays were conducted on 10‐day‐old susceptible tomato seedlings using root‐dip inoculation. As shown in Figure 1B, Fol007‐inoculated WT plants exhibited typical Fusarium wilt symptoms, including wilting, stunting, and chlorosis at 4 weeks post‐inoculation, while tomato plants inoculated with ΔEC19 mutants displayed normal growth without any visible disease symptoms, similar to the water‐inoculated control. In addition, the disease index was recorded, and the fresh weight of above‐ground tissues was measured. Compared with the WT Fol007‐inoculated plants, ΔEC19‐inoculated tomato plants exhibited a significant reduction in disease index values and a comparable plant biomass with water‐inoculated plants (Figure 1C,D). Collectively, these results demonstrate that EC19 is required for the full virulence of Fol007 on tomato.

FIGURE 1.

Deletion of EC19 reduces the pathogenicity of Fusarium oxysporum f. sp. lycopersici (Fol) 007 on tomato. (A) Colony morphology of wild‐type (WT) Fol007 and three independent in‐locus EC19 knockout mutants (ΔEC19) on potato dextrose agar plates. The white scale bars represent 2 cm. (B) Disease symptoms of representative tomato plants at 4 weeks after inoculation. Ten‐day‐old seedlings of the susceptible tomato cultivar Moneymaker were inoculated with water (mock), Fol007 (WT), and three independent in‐locus ΔEC19 mutants. (C) Average above‐ground fresh plant weight of 15 plants at 4 weeks after inoculation with Fol strains or water (mock). Error bars indicate the standard deviation of three replicates. (D) Disease index classes of 15 plants at 4 weeks after inoculation. All experiments were repeated three times. Significant difference was calculated using one‐way ANOVA followed by the Tukey post hoc test (***p < 0.001). The white scale bars represent 5 cm.

2.2. Complementation of EC19 Restores the Virulence of ΔEC19 Mutants

To confirm the virulence function of EC19 during Fol infection of tomato, we constructed a complementation vector and reintroduced it into the previously generated EC19 mutant strain (ΔEC19‐2) via protoplast transformation. Three independent ΔEC19‐complemented strains (ΔEC19‐2+C‐1, ‐2 and ‐3) were obtained (Figure S2). The expression levels of EC19 in the three independent ΔEC19‐2+C strains were comparable to that in WT Fol007 (Figure S3). As shown in Figure 2A, the three ΔEC19‐2+C strains exhibited normal colony growth similar to that of the WT Fol007 strain on potato dextrose agar (PDA) plates. Tomato plants inoculated with ΔEC19‐2+C strains developed severe disease symptoms similar to those inoculated with WT Fol007 (Figure 2B). In addition, no significant differences in plant weight or disease index were observed between plants inoculated with ΔEC19‐2+C strains and those infected with Fol007 (Figure 2C,D). Collectively, our findings indicate that genetic complementation of EC19 fully restored the virulence of the ΔEC19 mutant, demonstrating that the reduced virulence of ΔEC19 is due to the loss of EC19 and EC19 is required for full virulence of Fol.

FIGURE 2.

Complementing the EC19 deletion mutant with EC19 restores its pathogenicity on tomato. (A) Colony morphology of wild‐type (WT) Fusarium oxysporum f. sp. lycopersici (Fol) 007 and three independent ΔEC19 complementation strains (ΔEC19+C) on potato dextrose agar plates. Scale bar = 2 cm. (B) Disease symptoms of representative tomato plants at 4 weeks after inoculation. Ten‐day‐old seedlings of the susceptible tomato cultivar Moneymaker were inoculated with water (mock), Fol007 (WT), and three independent ΔEC19+C strains. Scale bar = 8 cm. (C) Average above‐ground fresh plant weight of 15 plants at 4 weeks after inoculation with different Fol strains or water (mock). Error bars indicate the standard deviation of three replicates. (D) Disease index classes of 15 plants at 4 weeks after inoculation. All experiments were repeated three times. Significant difference was calculated using one‐way ANOVA followed by the Tukey's post hoc test (***p < 0.001).

2.3. EC19 Is Dispensable for Abiotic Stress Tolerance and Conidiation in Fol

To examine whether EC19 is involved in the Fol response to osmotic and cell wall integrity stresses, the WT strain Fol007, ΔEC19‐2 and ΔEC19‐2+C were cultured on PDA plates supplemented with NaCl, KCl, sorbitol, SDS, or Congo red (CR). As shown in Figure 3A,B, the colony morphology and growth rate of ΔEC19‐2 were similar to those of Fol007 and ΔEC19‐2+C, indicating that EC19 is not involved in the response of Fol to those abiotic stresses. In addition, the ability of ΔEC19‐2 and ΔEC19‐2+C strains to produce conidia was evaluated. Compared with the WT strain Fol007, the numbers of conidia produced by both ΔEC19‐2 and ΔEC19‐2+C were similar (Figure 3C). Collectively, these findings demonstrate that EC19 does not participate in the adaptation of Fol to the tested abiotic stresses and conidiation.

FIGURE 3.

EC19 is not involved in abiotic stress tolerance and conidiation in Fusarium oxysporum f. sp. lycopersici (Fol). (A) Colony morphology of wild‐type Fol007, EC19 knockout mutant (ΔEC19‐2) and ΔEC19 complementation strain (ΔEC19‐2+C) on potato dextrose agar (PDA) plates supplemented with 1 M NaCl, 1 M KCl, 1 M sorbitol, 0.02% SDS, or 0.02% Congo red (CR). Representative images were taken after incubation at 25°C for 7 days. (B) Colony diameters of the indicated strains grown on PDA supplemented with the different stress agents. (C) Quantification of the number of conidia produced by indicated strains grown on PDA or NO₃ medium. All experiments were repeated three times, and data are presented as the mean ± SD of three independent experiments. Significant difference was analysed using one‐way ANOVA, and different lowercase letters indicate significant differences among groups (p < 0.05).

2.4. EC19 Has RNase Activity and Is Conserved Across Fungal Species

A previous study showed that EC19 has homology to the guanyl‐specific ribonuclease T1 family and contains a functional secretion signal peptide (Sun et al. 2022). To test whether EC19 has RNase activity, EC19 fusion protein with an N‐terminal 10×His‐tag and a C‐terminal Myc‐tag was produced and purified from Escherichia coli Rosetta2 (DE3) (Figure S4). The RNase activity of the EC19 recombinant protein was examined by incubating the purified protein with total RNA isolated from tomato. Figure 4A showed that EC19 was capable of degrading the tomato RNA, similar to the positive control RNase A, while a buffer containing 50% (v/v) glycerol in nuclease‐free water was unable to degrade RNA. These results demonstrated that EC19 possesses RNase activity.

FIGURE 4.

EC19 has ribonuclease activity and is highly conserved across fungal species. (A) EC19 has ribonuclease activity. The ribonuclease activity was conducted using the recombinant protein EC19 with N‐terminal His‐tag and C‐terminal Myc‐tag expressed in Escherichia coli Rosetta2 (DE3). A buffer containing 50% (v/v) glycerol in nuclease‐free water served as a negative control, and RNase A was used as a positive control. (B) Phylogenetic dendrogram (neighbour joining) of the EC19 protein from Fusarium oxysporum and its related homologous sequences from Beauveria bassiana, Colletotrichum destructivum, Colletotrichum graminicola, Exserohilum turcica, Fusarium avenaceum, Fusarium chlamydosporum, F. oxysporum Fo47, Fusarium graminearum, Fusarium graminum, Fusarium solani, Fusarium tjaetaba, Fusarium torulosum, Ilyonectria destructans, Phyllosticta capitalensis, Plenodomus biglobosus, Xylaria flabelliformis, and Xylaria scruposa. (C) Protein sequence alignment between EC19 and its homologous sequences. Light‐green box indicates the locations of the Ribonuclease T1 domain. (D) AlphaFold3‐predicted structure of EC19 and structural alignments of EC19 with the known structures of ribonuclease T1 from Aspergillus oryzae (AoRnt1 with PDB ID: 2RNT) and ribonuclease T2 (FG12). Structural similarity was assessed using the DALI server, with a cut‐off for significant similarity defined as a DALI Z‐score greater than 2.

To explore the phylogenetic distribution of EC19 homologues, we performed a BLAST search using the EC19 protein sequence against the NCBI database. Phylogenetic tree analysis showed that EC19 is highly conserved among fungal species, including non‐pathogenic F. oxysporum strain Fo47 (Figure 4B). All identified homologues contained a conserved ribonuclease T1 domain, as predicted by the Pfam database (Figure 4C). In addition, the AlphaFold3‐predicted structural model of EC19 shares a highly conserved three‐dimensional fold with the known ribonuclease T1 structure from Aspergillus oryzae (AoRnt1), but not with ribonuclease T2, FG12 (Figure 4D). Collectively, these results indicate that EC19 is a functional RNase that is evolutionarily conserved across diverse fungal species.

2.5. EC19 Is Highly Induced During Infection, and Its Product Localises to the Nucleus of Plant Cells

To explore the expression profiles of EC19, we conducted reverse transcription‐quantitative PCR (RT‐qPCR) on RNA samples extracted from conidia, conidia incubated in liquid culture for 4, 8, 12 and 16 h, and roots of susceptible tomato inoculated with Fol at 2, 4 and 6 days post‐inoculation (dpi). Limited expression of EC19 was detected in conidia and during in vitro cultivation (Figure S5), whereas in roots EC19 expression increased at 2 dpi and peaked at 6 dpi (Figure 5A), indicating that EC19 is upregulated during Fol infection stages. To examine the subcellular localisation of EC19 in plant cells, EC19 lacking its signal peptide was N‐terminally fused to GFP and transiently expressed in the leaves of N. benthamiana using agroinfiltration. Fluorescence signals were visualised using confocal laser scanning microscopy at 48 h after infiltration. As shown in Figure 5B, the green fluorescence from the EC19‐GFP fusion protein was predominantly observed in the nuclei of plant cells, co‐localizing with the red fluorescence of the nuclear localisation marker H2B‐mCherry. These results indicate that EC19 is induced during infection and EC19 predominantly localises to the nucleus of plant cells.

FIGURE 5.

EC19 is highly induced during Fusarium oxysporum f. sp. lycopersici (Fol) infection and its product predominantly localises to the nucleus of plant cells. (A) Reverse transcription‐quantitative PCR analysis showing the expression of EC19 in conidiospores and Fol‐infected tomato roots at 2, 4 and 6 days post‐inoculation (dpi). EC19 expression level was normalised to Fol actin gene. Values represent means ± standard error (SE) from three independent experiments. Significant difference was calculated using one‐way ANOVA followed by the Tukey's post hoc test (*0.01 < p < 0.05, *** p < 0.001). (B) EC19‐GFP, lacking its signal peptide for secretion, and nuclear localisation marker H2B‐mCherry were transiently expressed in Nicotiana benthamiana leaves and visualised in epidermal cells using confocal microscopy 48 h after agroinfiltration. Free GFP was used as a negative control. The white scale bars represent 25 μm.

2.6. EC19 Increases the Susceptibility of Tomato to Fol Infection and Alters Host Transcriptional Responses

To assess the functional role of EC19 during Fol infection of tomato, transgenic tomato plants overexpressing EC19 lacking its signal peptide, driven by the CaMV 35S promoter, were generated (Figure S6). Three EC19‐overexpressing (EC19‐OE) lines with high expression levels of EC19 displayed no visible phenotype compared to wild‐type (WT) plants (Figures S7 and 6A). Tomato WT and EC19‐OE transgenic seedlings were inoculated with conidia of Fol007 using the root‐dip method and disease symptoms were observed 4 weeks after inoculation. Compared with WT plants, EC19‐OE transgenic lines exhibited more severe disease symptoms (Figure 6B), as evidenced by significantly reduced plant weights and increased disease index of the infected plants (Figure 6C,D), suggesting that EC19 functions as a virulence‐promoting factor.

FIGURE 6.

EC19‐overexpressing tomato lines exhibit increased susceptibility to Fusarium oxysporum f. sp. lycopersici (Fol). (A) Reverse transcription‐quantitative PCR analysis showing the expression levels of EC19 in three independent EC19‐overexpressing (EC19‐OE) tomato lines. (B) Disease symptoms of representative tomato plants at 4 weeks after inoculation. Ten‐day‐old seedlings of susceptible tomato cultivar Moneymaker and three EC19‐OE lines were inoculated Fol007 (wild‐type, WT). Mock controls were inoculated with water. (C) Average above‐ground fresh weight of 15 plants at 4 weeks after inoculation with Fol007. Mock controls were inoculated with water. Error bars indicate the standard deviation of three replicates. (D) Disease index classes of 15 plants at 4 weeks after inoculation. Significant difference was calculated using one‐way ANOVA followed by Tukey's post hoc test (**0.001 < p < 0.01; *** p < 0.001). The white scale bars represent 8 cm.

To dissect the molecular mechanism underlying EC19‐mediated susceptibility of tomato, RNA‐seq analysis on EC19‐OE and WT tomato lines without Fol inoculation was conducted. 225 differentially expressed genes (DEGs), including 77 upregulated DEGs and 148 downregulated DEGs, were identified (Figure 7A and Table S2). KEGG enrichment analysis showed that plant hormone signal transduction, plant–pathogen interaction, and phenylpropanoid biosynthesis were the most enriched pathways, suggesting that EC19 may affect hormone‐related signalling and plant defence responses (Figure 7B). To further validate the RNA‐seq data, we examined the expression levels of six randomly selected genes using RNA extracted from the roots of WT and EC19‐OE tomato plants. The expression patterns of these genes were consistent with those obtained from RNA‐seq, confirming the reliability of the transcriptomic data (Figure 7C). Taken together, these findings demonstrate that EC19 functions as a virulence factor that promotes Fol infection and is associated with transcriptional changes in pathways involved in plant defence responses and hormone signal transduction.

FIGURE 7.

RNA‐seq analysis of differentially expressed genes (DEGs) and reverse transcription‐quantitative PCR (RT‐qPCR) validation of selected genes in EC19‐OE tomato lines. (A) Volcano plot showing upregulated and downregulated DEGs in EC19‐OE tomato plants compared with wild‐type (WT) plants. Roots from 10‐day‐old plants were collected for RNA extraction and sequencing. (B) KEGG pathway enrichment analysis of DEGs between WT and EC19‐OE tomato plants. The top 20 enriched KEGG pathways are displayed. Point size indicates the number of DEGs assigned to each pathway, and point colour represents the enrichment p value, with black red to red indicating significant enrichment and blue indicating low significant enrichment. (C) RT‐qPCR validation of six randomly selected DEGs associated with the top 20 KEGG pathways. The expression levels of these selected genes were quantified and normalised to the tomato actin gene. Data are presented as means ± standard error (SE) from three independent experiments. Significant difference was calculated using a two‐tailed Student's t‐test (**0.001 < p < 0.01; ***p < 0.001).

3. Discussion

Fusarium oxysporum has been ranked fifth among the most important fungal pathogens due to its ability to infect a wide range of host plants, leading to vascular browning, leaf epinasty, stunting, progressive wilting, premature defoliation, and ultimately plant death (Dean et al. 2012). Fol has served as an excellent model system in which to study the molecular basis of resistance and susceptibility in tomato against the vascular pathogen (Takken and Rep 2010). Plant fungal pathogens secrete effectors, RNAs and other small molecules that counter or avoid plant immune responses, acquire nutrients, dampen plant health and modulate the microbiome to facilitate pathogen infection (Wilson and McDowell 2022; Guerreiro and Stukenbrock 2025; Li, Newman, et al. 2024). Previous studies reported that secreted fungal ribonuclease contributes to the virulence of pathogens (He et al. 2022; Yang et al. 2025; Kettles et al. 2018). This study reveals that the effector protein EC19, secreted by Fol, is essential for its full virulence in tomato to cause wilt disease. Knockout of EC19 significantly reduced the pathogenicity of Fol, while its complementation restored virulence. EC19, containing a ribonuclease T1‐like domain, is conserved across fungal species and can cleave tomato RNA. In N. benthamiana leaves, EC19 localised to the plant nucleus. Overexpressing EC19 in tomato plants increased susceptibility to Fol infection, and RNA‐seq analysis of EC19‐OE plants revealed that genes associated with plant hormone signal transduction, plant–pathogen interaction, and phenylpropanoid biosynthesis were significantly reduced, suggesting EC19 may interfere with host defence responses and hormone signal transduction. These results indicate that EC19 with RNase activity is required for the full virulence of Fol during tomato infection.

Our previous study demonstrated that EC19 contains a functional signal peptide for secretion, suggesting that the EC19 protein could be secreted outside of Fol or its hyphal structures, likely entering plant cells (Sun et al. 2022). In this study, we further demonstrated that in‐locus knockout of EC19 reduced the pathogenicity of Fol, while complementing EC19 in the mutant restored its virulence (Figures 1 and 2), strongly indicating that secreted EC19 contributes to Fol virulence. It has been documented that knockout of Fol effector genes, particularly SIX gene family, can reduce the virulence of Fol. Six out of the 14 SIX proteins, notably SIX1 (Rep et al. 2004), SIX2 (Gawehns et al. 2015), SIX3 (Houterman et al. 2009), SIX5 (Ma, Houterman, et al. 2015), SIX6 (Gawehns et al. 2014) and SIX8 (Aalders et al. 2024), have been demonstrated to be required for virulence because deletion mutants compromise the pathogenicity of Fol on susceptible tomato plants. Although SIX4 does not contribute to virulence on a susceptible tomato plant, a SIX4 knockout loses pathogenicity on tomato plants containing either the I‐2 or I‐3 gene, suggesting that SIX4 suppresses I‐2‐ and I‐3‐mediated resistance to promote fungal virulence on resistant plants (Houterman et al. 2008). However, SIX4 does not suppress I‐7‐mediated resistance against Fol race 1 (Gonzalez‐Cendales et al. 2016). Due to the fact that all SIX genes, except of SIX8, are located on chromosome 14 of Fol4287 and chromosome 13 of Fol007, this specific chromosome has been designated as the pathogenicity chromosome (Ma et al. 2010; Sun et al. 2025). EC19 is located on chromosome 12 of Fol4287, which is not a pathogenicity chromosome, but it is required for pathogenicity. This finding indicates that Fol virulence determinants are not restricted to pathogenicity chromosomes. Recently, more examples have shown that effector genes located on other chromosomes also contribute to Fol virulence. For example, the conserved crotonyltransferase FolRtt109, which is also conserved in the non‐pathogenic F. oxysporum Fo47, enhances the activities of FolChi, FolAra, and FolPl through lysine crotonylation, thereby promoting chitin degradation, host cell‐wall breakdown, and Fol virulence (Zhang et al. 2026; Li et al. 2026). Interestingly, EC19 is conserved in the non‐pathogenic strain Fo47 and is also present in a range of other fungal species (Figure 4), suggesting that EC19 does not function exclusively as a determinant of host‐specific pathogenicity or disease symptom development. EC19 might participate in more broadly conserved processes associated with infection, including fungal establishment and early host colonisation, rather than later disease development. However, functional characterisation of EC19 homologues in Fo47 or other fungal pathogens will be required to determine whether they contribute to non‐pathogenic root colonisation or early host colonisation. These findings highlight that Fol employs conserved effector genes located on both the pathogenicity and other chromosomes to promote infection. Further research is needed to identify additional effectors and potential targets for enhancing tomato resistance to Fol.

Our results found that the expression of EC19 was specifically induced during the Fol infection stage, with limited expression observed in the vegetative stage (Figures 5A and S5), strongly indicating that EC19 is required for Fol infection. We can speculate that the expression of EC19 is tightly regulated in Fol, which is likely induced by some signal molecules from tomato plants. Previous studies showed that the expression of Fol effector genes is tightly regulated by both complicated transcriptional and post‐translational mechanisms. Transcription factors such as Sge1, FTF1, and FTF2 in Fol are required for activating effector genes like the SIX family, as exemplified by RNA silencing or knockout of these factors reducing virulence and effector gene expression (van der Does et al. 2016). Additionally, autoacetylation of FolSas2 at lysine 269 inhibits proteasomal degradation, promoting histone modifications that activate effector genes during early infection, while deacetylation destabilises the protein and impairs pathogenicity (Song et al. 2025). These findings indicate the complex regulatory network controlling the expression of Fol effector genes, particularly those upregulated during infection. The molecular mechanism underlying the induction of EC19 expression during Fol infection demands further study.

Many plant pathogens employ secreted ribonuclease effectors to promote their infection. The reported secreted ribonucleases from plant pathogens can be classified into two families: RNase T1 and RNase T2 (Lv et al. 2022). EC19 belongs to the RNase T1 family and has RNase activity to degrade host plant RNA (Figure 3). Our findings are consistent with previous studies showing that Zt6 from Z. tritici (Kettles et al. 2018), SRE1 from S. turcica (He et al. 2022) and Fg12 from F. graminearum function as secreted RNases with the ability to degrade host plant RNA and contribute to the virulence of their respective pathogens (Yang et al. 2021). Similarly, secreted FoRnt2 from Fol and FocRnt2 from Foc, both belonging to the ribonuclease T2 family, possess RNase activity and are required for the virulence of their corresponding pathogens (Qian et al. 2022; He et al. 2024). However, the molecular mechanism underlying RNase effector‐mediated virulence remains largely unknown. Interestingly, the Zt6 effector from Z. tritici exhibited potent toxicity towards microorganisms, suggesting that it may play a role in antimicrobial competition and niche protection (Kettles et al. 2018). Although EC19 exhibits RNase activity, its site of action during host infection remains unknown. The functional signal peptide of EC19 supports its secretion from Fol (Sun et al. 2022), suggesting that EC19 may function in the apoplast or xylem. Alternatively, EC19 may enter host cells and target intracellular RNAs. However, direct evidence for the translocation and intracellular localisation of EC19 is currently lacking. Further localisation and substrate‐identification experiments are required to distinguish between these possible mechanisms.

Our previous studies showed that EC19 carrying its signal peptide induced cell death during transient expression in N. benthamiana (Sun et al. 2022), whereas stable expression of EC19 without the signal peptide caused no obvious growth defects in tomato (Figures 7 and S7). This observed difference may reflect distinct protein localisation, expression levels, or species‐specific immune responses. Recently, FG12 from F. graminearum was shown to exhibit RNase activity‐dependent antibacterial function, enabling FG12 to manipulate the plant‐associated rhizosphere microbiome and promote F. graminearum infection (Yang et al. 2025). Based on these findings, we cannot exclude the possibility that EC19 might function in the rhizosphere of the host as Fol is a soil‐borne pathogen. We speculate that it could be secreted into host cells to degrade host RNAs in the cytosol, using its RNase activity, to counteract host defence strategies such as the translocation of small RNAs from the host into pathogens to inhibit the expression of virulence genes (Qiao et al. 2021). Secondly, secreted EC19 might enter the host rhizosphere, where it could exert RNase‐dependent antimicrobial activity and regulate the tomato‐associated microbiome by inhibiting beneficial or competitive microbes, thereby facilitating Fol colonisation in the soil and infection of the host plant. Further research is needed to fully elucidate the precise mechanisms underlying EC19‐mediated RNA degradation and microbial regulation. Understanding these mechanisms could reveal potential targets for inhibiting Fol infection to enhance tomato resistance.

4. Experimental Procedures

4.1. Plants and Fungal Strains

Susceptible Solanum lycopersicum ‘Moneymaker’ without Fol‐related resistance genes and N. benthamiana were used in this study. Tomato plant seeds were obtained from a commercial company (Biorun) and N. benthamiana seeds were provided by Hebei Jingshu Agriculture Technology Company (Shijiazhuang, China). Plants were grown in a greenhouse maintained at 22°C under 16‐h light and 18°C under 8‐h dark. Fol race 2 strain Fol007 from the University of Amsterdam, carrying Avr2 and Avr3, was used in this study.

4.2. Generation of the Fol EC19 Knockout Mutants and Its Complemented Transformants

For the generation of in‐locus EC19 knockout mutants in Fol007, the 1000 bp upstream region of EC19 was amplified using primer pairs EC19‐Up‐PacI and EC19‐Up‐KpnI, while the 1000 bp downstream region was amplified using primer sets EC19‐DW‐XbaI and EC19‐DW‐BssHII from Fol007 genomic DNA. The obtained PCR products were cloned into the binary vector pRW2h (Ma, Houterman, et al. 2015). The resulting construct was used for protoplast‐mediated Fol transformation following the methods previously described (Ma et al. 2012). Hygromycin B (100 μg/mL) was used to select the transformants on PDA plates. DNA was extracted from putative transformants, and the deletion of EC19 was confirmed by PCR using primer set EC19‐F/EC19‐R (Table S1). To reintroduce EC19 into the ΔEC19 mutant, a DNA fragment containing the 882 bp upstream, the EC19 ORF and its 357 bp downstream sequences was amplified from genomic Fol007 DNA using primers EC19‐pRW1p‐F and EC19‐pRW1p‐R. The obtained product with XbaI and PstI restriction sites was subcloned into the pRW1p vector (Houterman et al. 2008). The resulting construct was used for transformation, and the presence of EC19 in the zeocin‐resistant transformants was confirmed by PCR using primers EC19‐F/EC19‐R (Table S1).

4.3. Disease Assay

Fol inoculation was conducted using the root‐dip method as previously described (Mes et al. 1999), and recording disease index and measuring plant weight were conducted following the method described previously (Houterman et al. 2009). Briefly, conidia of Fol strains were collected from NO₃ medium, and their concentration was adjusted to 1 × 10⁷ conidia/mL. Ten‐day‐old tomato seedlings were inoculated with the conidial suspension or water as a control for 10 min. The inoculated seedlings were planted in potting soil and grown under 25°C and 50% relative humidity with 16 h of light. Disease symptoms were checked at 4 weeks after inoculation. All experiments were repeated three times, and each treatment contained 15 plants.

4.4. Production and Purification of Recombinant Protein

The coding sequence of the EC19 gene lacking its signal peptide was cloned into a modified pET21a‐His/MYC expression vector containing an N‐terminal 10×His tag and C‐terminal Myc tag. The recombinant or empty vector was transformed into E. coli Rosetta2 (DE3). The expression and purification of recombinant protein were conducted following the protocols of the manufacturer (CUSABIO Technology LLC). The protein was dissolved and stored in 50% glycerol buffer. The purified EC19 protein was separated by 10% SDS‐PAGE gel and stained with Coomassie blue.

4.5. Bioinformatics Analysis

Homologous protein sequences of EC19 were identified by querying its amino acid sequence against the NCBI Non‐redundant (NR) protein database using the BLASTp program with default parameters. Sequences with significant similarity (E‐value < 1 × 10⁻⁵) were retained for subsequent analyses. Multiple sequence alignment of EC19 and homologous ribonuclease T1 proteins from different fungal pathogens was performed using ClustalW2. The resulting alignments were manually examined and trimmed to remove ambiguously aligned regions prior to phylogenetic analysis. A phylogenetic tree was generated using MEGA v. 5.0 with the neighbour‐joining (NJ) method and the Poisson substitution model, and node support was evaluated using 1000 bootstrap replicates. The final tree was visualised and annotated using iTOL v. 6. The three‐dimensional (3D) structure of EC19 protein was predicted using AlphaFold‐Multimer v. 3. Structural similarity between EC19 and known ribonuclease T1 proteins was examined with the DALI server, and protein pairs exhibiting a DALI Z‐score greater than 2.0 were considered structurally related.

4.6. Ribonuclease Activity Assay

The ribonuclease activity of EC19 was examined using an in vitro assay, as described previously (Yang et al. 2021). Briefly, approximately 1 μg of total tomato RNA was incubated with EC19 protein at 25°C for 30 min. An equal amount of RNase A was used as a positive control and 50% (v/v) glycerol in nuclease‐free water (protein buffer solution) was used as a negative control, with equal amounts of total RNA in each treatment. Following incubation, all samples were mixed with 1× loading buffer and subjected to electrophoresis on a 1% agarose gel.

4.7. Measurements of Fol Growth, Conidial Production and Abiotic Stress Tolerance

Wild‐type strain Fol007 (WT), EC19 knockout mutant (ΔEC19‐2), and the ΔEC19‐2 complemented strain (ΔEC19‐2+C) were cultured on PDA plates and incubated at 25°C for 7 days. After cultivation, colony morphology was photographed, and colony diameters were measured for statistical quantification. Conidial production by the three strains was quantified after incubation on PDA plates for 7 days and in NO₃ medium for 5 days. For abiotic stress assays, the osmotic stress agents were added to PDA plates at a final concentration of 1 M NaCl, 1 M KCl, or 1 M sorbitol, while cell wall stress agents were added at final concentrations of 0.02% SDS and 0.02% Congo Red (CR). All treatments were performed with three biological replicates. Significant differences were analysed using one‐way ANOVA, and different lowercase letters indicate significant differences among groups (p < 0.05).

4.8. Localisation Assay in the Leaves of N. benthamiana

For the localisation study, the coding sequence of EC19 lacking the signal peptide was amplified from Fol‐infected tomato cDNA library using primers pMDC43‐EC19‐F and pMDC43‐EC19‐R. The PCR products were cloned into the pMDC43 vector carrying an N‐terminal GFP tag. The obtained construct was transformed into Agrobacterium tumefaciens GV3101. Agrobacterium‐mediated transient expression in N. benthamiana leaves was conducted as previously described (Ma et al. 2012). Leaf discs 48 h after infiltration were imaged using LSM610 (Zeiss) confocal microscopy. Excitation of GFP was at 488 nm with an Ar‐ion laser, and emission was detected through a 505–530 nm pass filter. Excitation of mCherry was detected at 543 nm using HeNe laser. The 590–610 nm filter captured emission.

4.9. Generation of Stable Transgenic Tomato Plants

For stable tomato transformation, the EC19 ORF lacking its signal peptide was amplified using primers E19794_0S1‐F and E19794_0S1‐R (Table S1) from the Fol‐infected tomato cDNA library. The PCR product was cloned into the pBWA(V)HS vector. The resulting plasmid pBWA(V)HS::EC19 was used for Agrobacterium‐mediated transformation of susceptible tomato cultivar Moneymaker following the protocols of the manufacturer (Biorun).

4.10. RNA Extraction and Quantitative PCR

Plants of the T₁ generation of EC19‐overexpressing lines were grown for about 1 month, and leaves and roots were harvested and immediately frozen in liquid nitrogen and stored at −80°C prior to the extraction of total RNA. The tomato roots inoculated with EC19 mutant (ΔEC19) and ΔEC19 complemented strains were collected at 4 days post‐inoculation for RNA extraction. Purified conidia were incubated in PDA for 4, 8, 12, and 16 h, after which the germinating conidia were harvested for RNA extraction. RNA was extracted from the samples using Eastep Super Total RNA Extraction Kit (Promega), with on‐column DNase digestion performed using RNase‐free DNase (Qiagen) to remove genomic DNA following the protocols of the manufacturer. For RT‐qPCR analysis, cDNA was first synthesised with Eastep RT Master Mix Kit (Promega), followed by qPCR using the LightCycler 96 Instrument (Roche) with ChamQ Universal SYBR qPCR Master Mix (Vazyme). The primers used for RT‐qPCR are described in Table S1. C _t values were analysed according to the 2^−ΔΔCt method (Livak and Schmittgen 2001). The tomato and Fol actin served as reference genes. Statistical analysis was conducted using GraphPad Prism 6 (GraphPad Software). Significance comparison among multiple groups was examined using one‐way ANOVA followed by the Tukey's post hoc test, while pairwise comparisons were evaluated using a two‐tailed Student's t‐test. Values represent the mean ± SEM of three independent biological replicates from a representative experiment. Significant differences between treatments and controls are indicated by asterisks, with ***p < 0.001, **p < 0.01, and *p < 0.05.

4.11. RNA Sequencing and Data Analysis

EC19‐OE‐6 and wild‐type Moneymaker tomato plants were grown for about 10 days, and roots were harvested and immediately frozen in liquid nitrogen, and stored at −80°C. Total RNA was extracted using the TRIzol reagent (Invitrogen) according to the manufacturer's protocol. RNA purity and quantification were evaluated using the NanoDrop 2000 spectrophotometer (Thermo Scientific). RNA integrity was assessed using the 2100 Bioanalyzer (Agilent Technologies). Then the libraries were constructed using VAHTS Universal V10 RNA‐seq Library Prep Kit (Premixed Version) according to the manufacturer's instructions. The transcriptome sequencing and analysis were conducted by OE Biotech Co. Ltd. (Shanghai, China). The sequencing libraries were sequenced on an Illumina Novaseq 6000 platform to generate 150 bp paired‐end reads. After removing low‐quality reads using fastp (v. 0.20.1) (Chen et al. 2018), clean reads were obtained for subsequent data analysis. The clean reads were mapped to the reference genome of S. lycopersicum (accession: GCF_036512215.1) using HISAT2 (v 2.1.0) (Kim et al. 2015). Fragments per kilobase of transcript per million mapped reads (FPKM) (Roberts et al. 2011) of each gene was calculated and gene‐level read counts were obtained by HTSeq‐count (v. 0.11.2) (Anders et al. 2015). Differential expression analysis was performed using DESeq2 (v. 1.22.2) (Love et al. 2014). Genes with adjusted p value (p _adj) < 0.05 and a fold change > 2 or < 0.5 were set as the threshold for significant DEGs. Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis was performed using clusterProfiler (v. 3.10.0) based on the hypergeometric distribution to identify significantly enriched pathways (Kanehisa et al. 2008). KEGG enrichment plots and volcano plots were generated using R.

Author Contributions

Min Li: methodology, data curation, investigation. Yifei Wang: methodology, data curation, software, investigation, validation. Haoqian Li: methodology, data curation, investigation, formal analysis, validation. Xizhe Sun: methodology, data curation, validation, investigation. Jianbiao Wang: data curation, methodology, validation. Xiangling Fang: writing – review and editing, project administration, funding acquisition. Jianjun Zhao: conceptualization, funding acquisition, resources, writing – review and editing, supervision. Lisong Ma: conceptualization, supervision, writing – original draft, writing – review and editing, project administration, resources, visualization, funding acquisition.

Funding

This research was supported by the ‘Hundred Talents Program’ for the Introduction of high‐level overseas talents in Hebei Province (E2020100004), and the National Natural Science Foundation of China (31802126).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: PCR amplification to confirm the in‐locus deletion of EC19 with primer set EC19‐F/EC19‐R. ‘P’ indicates the positive control that is amplified using cDNA of Fol007 as template.

Figure S2: PCR amplification to confirm the reintroduction of EC19 into knockout mutant ΔEC19‐2. Three independent complemented transformants (ΔEC19‐2+C‐1, ΔEC19‐2+C‐2 and ΔEC19‐2+C‐3) were selected for further analysis.

Figure S3: Reverse transcription‐quantitative PCR analysis showing the relative expression of EC19 in tomato roots at 6 days post‐inoculation. Expression is shown for three independent knockout strains (ΔEC19‐1, ΔEC19‐2, and ΔEC19‐3), three independent complementation strains (ΔEC19‐2+C‐1, ΔEC19‐2+C‐2, and ΔEC19‐2+C‐3) and wild‐type Fol007.

Figure S4: Purified recombinant EC19 protein with N‐terminal 10×His tag and C‐terminal Myc tag was subjected to SDS‐PAGE analysis.

Figure S5: Reverse transcription‐quantitative PCR analysis showing the relative expression levels of EC19 in germinating conidia. Conidia were incubated on potato dextrose broth for 4, 8, 12, and 16 h, followed by RNA extraction and reverse transcription‐quantitative PCR analysis. RNA extracted from roots of tomato plants inoculated with Fusarium oxysporum f. sp. lycopersici at 2 days post‐inoculation (dpi) was included as well.

Figure S6: Reverse transcription‐quantitative PCR analysis showing the relative expression levels of EC19 in all EC19‐ overexpressing (EC19‐OE) tomato lines. Lines EC19‐OE‐2, EC19‐OE‐6 and EC19‐OE‐10 were selected for further analysis.

Figure S7: Phenotype of three independent EC19‐OE tomato lines and wild‐type tomato plants. Tomato lines EC19‐OE‐2, EC19‐OE‐6 and EC19‐OE‐10 were selected for observations at 30 days after planting.

Table S1: Primers used in this study.

Table S2: List of upregulated and downregulated differentially expressed genes between wild‐type (WT) and EC19‐OE tomato plants.

Data Availability Statement

The raw data of RNA‐seq have been deposited in the China National Center for Bioinformation under BioProject ID PRJCA070132. The other data supporting the conclusions of this study are available within the article and available from the corresponding authors upon request.