Fol-milR1, a pathogenicity factor of Fusarium oxysporum, confers tomato wilt disease resistance by impairing host immune responses

Abstract

Although it is well known that miRNAs play crucial roles in multiple biological processes, there is currently no evidence indicating that milRNAs from Fusarium oxysporum f. sp. lycopersici (Fol) interfere with tomato resistance during infection. Here, using sRNA-seq, we demonstrate that Fol-milR1, a trans-kingdom small RNA, is exported into tomato cells after infection. The knockout strain ∆Fol-milR1 displays attenuated pathogenicity to the susceptible tomato cultivar 'Moneymaker'. On the other hand, Fol-milR1 overexpression strains exhibit enhanced virulence against the resistant cultivar 'Motelle'. Several tomato mRNAs are predicted targets of Fol-milR1. Among these genes, Solyc06g007430 (encoding the CBL-interacting protein kinase, SlyFRG4) is regulated at the posttranscriptional level by Fol-milR1. Furthermore, SlyFRG4 loss-of-function alleles created using CRISPR/Cas9 in tomato ('Motelle') exhibit enhanced disease susceptibility to Fol, further supporting the idea that SlyFRG4 is essential for tomato wilt disease resistance. Notably, our results using immunoprecipitation with specific antiserum suggest that Fol-milR1 interferes with the host immunity machinery by binding to tomato ARGONAUTE 4a (SlyAGO4a). Furthermore, virus-induced gene silenced (VIGS) knock-down SlyAGO4a plants exhibit reduced susceptibility to Fol. Together, our findings support a model in which Fol-milR1 is an sRNA fungal effector that suppresses host immunity by silencing a disease resistance gene, thus providing a novel virulence strategy to achieve infection.

Keywords: Fusarium oxysporum f. sp. lycopersici; immunity response; plant-pathogen interactions; resistant gene; tomato wilt disease; trans-kingdom miRNA.

Fig. 1

Fol‐milR1 is exported into tomato host cells during infection. (a) Tomato wilt disease symptoms caused by infection with Fol for 2 wk in susceptible cultivar ‘Moneymaker’ (MM) and resistant cultivar ‘Motelle’ (Mot). (b) Detection of Fol‐milR1 in treated tomato roots using low molecular weight RNA gel blots. 40 μg of total RNA was separated by electrophoresis on 8% sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS‐PAGE) gels and transferred to a nylon N+ membrane. (γ‐32P)ATP‐labelled specific oligonucleotide probe sequences were used for hybridization. The snRNA gene U6 was used as a loading control. No hybridization could be detected for Fol‐milR3 or Fol‐milR4 (shown) or the other four Fol‐milRNAs (data not shown). (c) Fol‐milR1 expression was confirmed using quantitative real‐time polymerase chain reaction (qRT‐PCR) with specific primers. Asterisks indicate significant difference when compared to the corresponding control plants in the same treatment, according to the Chi‐square test (*, P < 0.05). Error bars represent the SD of three replicates. (d) Fol‐milR1 was detected in Fol, Fol protoplasts and tomato root protoplasts using qRT‐PCR with specific primers. Asterisks indicate significant difference when compared to the corresponding control plants in the same treatment, according to the Chi‐square test (*, P < 0.05). Error bars represent the SD of three replicates. (e) Relative levels of fungal biomass were presented by ribosomal intergenic spacer region (IGS) amplified from genomic DNA correlates withFol biomass in Fol, Fol protoplasts and tomato root protoplasts, using quantitative‐PCR with specific primers. Asterisks indicate significant difference when compared to the corresponding control plants in the same treatment, according to a chi‐squared test (*, P < 0.05). Error bars represent the SD of three replicates.

Fig. 2

Fol‐milR1 is essential for Fol pathogenicity. (a) The pre‐miRNA‐like stem‐loop structure of precursor Fol‐milR1. (b) Identification of Fol‐milR1‐KO (knockout) and Fol‐milR1‐OE (overexpression) strains using low molecular weight RNA gel blots. Three Fol‐milR1‐KO and three Fol‐milR1‐OE strains (highlighted in red) were recruited for subsequent experiments. (c) The Fol‐milR1‐KO and Fol‐milR1‐OE strains and control wild‐type Fol were used to inoculate tomato seedlings. Wilt disease symptoms were photographed 2 wk after inoculation. (d) Generation of Fol‐milR1 site‐mutated strains. Mutated sites were highlighted in yellow. The mutated sites were confirmed by sequencing. (e) The Fol‐milR1‐site‐mutated strains and control wild‐type Fol were used to inoculate tomato seedlings. Wilt disease symptoms were photographed 2 wk after inoculation. Cotton blue staining results reflect the abundance of Fol in the stem of tomato plants. More intense cotton blue staining correlates with higher levels of Fol. (f) Disease grades for all pathogen infection assays at 14 d post inoculation (dpi). The asterisks indicate significant differences in the wilt disease symptoms of Fol‐milR1‐KO strains vs wild‐type Fol in ‘Moneymaker’ (MM), and Fol‐milR1‐OE strains vs wild‐type Fol in ‘Motelle’ (Mot) according to the Chi‐square test (*, P < 0.05). Error bars represent the SD of three replicates. (g) Disease grades for Fol infection assays at 14 dpi. The asterisks indicate significant differences in the wilt disease symptoms of Fol‐milR1‐SM strains vs wild‐type Fol in ‘Moneymaker’ according to the Chi‐square test (*, P < 0.05). Error bars represent the SD of three replicates.

Fig. 3

Fol‐milR1 regulates SlyFRG4 expression at the posttranscriptional level. (a) SlyFRG4 mRNA levels are repressed after Fol infection in both ‘Moneymaker’ (MM) and ‘Motelle’ (Mot). Asterisks indicate significant difference when compared to the corresponding control plants in the same treatment, according to the Chi‐square test (*, P < 0.05). Error bars represent the SD of three replicates. (b) Level of SlyFRG4 target mRNA during co‐infiltration experiments in Nicotiana benthamiana. Quantitative real‐time polymerase chain reaction (qRT‐PCR) was used to determine the relative levels of SlyFRG4 in N. benthamiana leaves expressing SlyFRG4 only, SlyFRG4 + Fol‐milR1 or SlyFRG4 + control miRNA (Sly‐miR166). Values were normalized to N. benthamiana actin. Asterisks indicate significant difference when compared to the corresponding control plants in the same treatment, according to the Chi‐square test (*, P < 0.05). Error bars represent the SD of three replicates. (c) SlyFRG4‐GFP fusion protein was detected by Western blot using anti‐GFP antibody. Crude protein extracts prepared from N. benthamiana leaves in (b) were electrophoresed on sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS‐PAGE) gels and blotted onto nitrocellulose membranes (top panel). A duplicate gel was Ponceau S‐stained as a loading control (bottom panel). A minimum of 10 individual leaf samples were used for each experiment. (d) The cleavage site in the SlyFRG4 mRNA was determined using 5′RLM‐RACE. The arrow indicates the 5′ terminus of miRNA‐guided cleavage products and the frequency of clones (8/12) is shown. The cDNA of SlyFRG4 contains one single large exon.

Fig. 4

SlyFRG4 is required for Fol resistance. (a) Schematic diagram of the CRISPR/Cas9 cassette used for mutation of SlyFRG4. (b) clustalx nucleic acid sequence alignments of genomic sequences obtained for SlyFRG4‐LOF plants. The sequence of the gRNA is highlighted with red. (c) Loss of function (LOF) of SlyFRG4 attenuates the resistance to Fol in ‘Motelle’. Cotton blue staining results reflect the abundance of Fol in the stem of tomato plants. More intense cotton blue staining correlates with greater abundance of Fol. (d) Relative levels of fungal ribosomal intergenic spacer region (IGS) amplified from genomic DNA correlates with Fol biomass in tomato plants at 2 wk after inoculation with Fol. The asterisks indicate significant differences in the Fol biomass of SlyFRG4 loss‐of‐function alleles vs ‘Motelle’ after Fol infection according to the Chi‐square test (*, P < 0.05). Error bars represent the SD of three replicates. (e) Disease grades for Fol infection assays at 14 d post inoculation (dpi). The asterisks indicate significant differences in the wilt disease symptoms of SlyFRG4‐LOF alleles vs ‘Motelle’ after Fol infection according to the Chi‐square test (*, P < 0.05). Error bars represent the SD of three replicates. KO, knockout; MM, cv Moneymaker; Mot, cv Motelle; PAM, protospacer adjacent motif.

Fig. 5

Fol‐milR1 associates with SlyAGO4a to suppress host immunity. (a) Association of Fol‐milR1 with SlyAGO4a during infection. SlyAGO4a was immunoprecipitated (IP) from Fol‐infected roots harvested at 24 h after inoculation using a SlyAGO4a polyclonal antibody. Total RNA was extracted from the SlyAGO4a‐IP fraction and used for stem‐loop quantitative real‐time polymerase chain reaction (qRT‐PCR). (b) Novel‐m0003‐3p, homologous to Fol‐milR1, was detected in the SlyAGO4a‐IP sample using sRNA‐sequencing. (c) SlyAGO4a‐VIGS plants exhibit reduced disease susceptibility to Fol compared with the susceptible ‘Moneymaker’. In total, 30 SlyAGO4a‐VIGS plants were generated, and 25 exhibited reduced disease susceptibility to Fol. phytoene desaturase (PDS). TRV‐silenced plants (TRV‐PDS) and TRV‐vector plants were used as positive controls for silencing. (d) Disease grades for Fol infection assays at 14 dpi. The asterisks indicate significant differences of the wilt disease symptoms of SlyAGO4a‐VIGS plants vs ‘Moneymaker’ after Fol infection according to the Chi‐square test (*, P < 0.05). Error bars represent the SD of three replicates. MM, cv Moneymaker; Mot, cv Motelle; PDS, phytoene desaturase; VIGS, virus‐induced gene silencing.