The wild-type strain of V. mali 03-8 and Sclerotinia sclerotiorum were obtained from the Laboratory of Integrated Management of Plant Disease in the College of Plant Protection, Northwest A&F University, Shaanxi, Yangling, China (Wu et al., 2018). Valsa mali, S. sclerotiorum, and Phytophthora capsici were cultured on potato dextrose agar (PDA) at 25°C. Agrobacterium tumefaciens strains were routinely cultured on a yeast-extract and peptone (YEP) medium at 28°C and Escherichia coli strains DH5α were cultured on a Luria-Bertani medium (LB) at 37°C.

The tissue-cultured plantlets of apple (M. domestica “Gala 3”) were grown on a Murashige and Skoog (MS) agar medium containing 0.3 mg/L 6-benzyl aminopurine (6-BA) and 0.2 mg/L indoleacetic acid (IAA) (Sun et al., 2018). They were cultured under conditions of 25°C, 60 μmol/m²/s, and a 14 h photoperiod. Nicotiana benthamiana were grown in a greenhouse under a 16 h 25°C:8 h 22°C, high light intensity:darkness regime.

Valsa mali and S. sclerotiorum were cultured on PDA for 2 days. The agar disks containing mycelium of V. mali or S. sclerotiorum were taken from the edge of a growing colony. The stab inoculation method was used on the infiltrated leaves of apple and N. benthamiana (Wu et al., 2018). The inoculated samples were incubated at 25°C for 36 h. The size of lesions on apple and N. benthamiana were measured by Image J software and the crossing method, respectively. The apple leaves were then collected to measure the biomass of V. mali. All the treatments were performed on at least 30 apple leaves or 10 N. benthamiana leaves, and all the experiments were repeated three times.

Valsa mali biomass was measured using a quantitative polymerase chain reaction (qPCR) with V. mali-specific glyceraldehyde-3-phosphate dehydrogenase (G6PDH) primers (Yin et al., 2013) related to the apple elongation factor 1 alpha gene (EF-1a) (Yin et al., 2016) at 36 h post infiltration (hpi). Genomic DNA was isolated using a Super Plant Genomic DNA Kit (Polysaccharides and Polyphenolics-rich) (Tiangen, Beijing, China) from 0.5 g apple leaves that included all the infected tissue and healthy tissue without petioles. Quantification of V. mali biomass assay was performed three times, and each experiment with three replicates.

All the constructs were cloned by homologous recombination with 15–20 bp of vector sequences at the 5′ terminus of each primer using a ClonExpress II One Step Cloning Kit (Vazyme Biotech, Nanjing, China). The VmEP1 (lacking the signal peptide-encoding region) was cloned from the cDNA of V. mali using gene-specific primers and was inserted into PICH86988, resulting in PICH86988-HA-VmEP1 used to perform a co-immunoprecipitation (Co-IP) assay. MdPR10 and NbPR10, without terminator codons, were ligated into pCAMBIA1302 between the Nco1 and Spe1 sites, resulting in pCAMBIA1302-MdPR10-GFP and pCAMBIA1302-NbPR10-GFP for a Co-IP assay and a transient expression assay. The VmEP1 and MdPR10 amplified fragments were cloned into nYFP or cYFP, resulting in nYFP-MdPR10 and cYFP-VmEP1 used for carrying out a bimolecular fluorescence complementation (BiFC) assay. Individual colonies of each construct were tested by PCR and verified by sequencing. For yeast two-hybrid (Y2H) assays, VmEP1 fragment without signal peptide-encoding region was cloned into the pKGBKT7 vector using the EcoR1 and BamH1 sites to form the bait plasmid BD-VmEP1. MdPR10 was cloned into the pGADT7 vector using the EcoR1 and BamH1 sites to form the AD-MdPR10 plasmid. Loss-of-function MdPR10 plasmid was produced by cloning the reverse partially specific sequences of gene MdPR10 into pFGC5941, carrying the NcoI and BamHI digestion sites and resulting in pFGC5941-MdPR10 to produce siRNA (hpRNAi-MdPR10) (Zhang Q. et al., 2018). The constructs used for virus-induced gene silencing (VIGS) in N. benthamiana were generated in the tobacco rattle virus 2 (TRV2) vector (Liu et al., 2002) using the N. benthamiana cDNA library for gene fragment amplification. All the constructs were validated by sequencing in Sangon (Sangon Biotech, Shanghai, China). All the primers used above are listed in Supplementary Table 1.

The constructs were transformed into Agrobacterium strain GV3101 (pSoup-P19) by electric shock transformation. GV3101 (pSoup-P19) containing plasmids were grown in YEP medium supplemented with appropriated antibiotics at 28°C for 16–18 h until optical density (OD)₆₀₀ = 0.8. The bacterial cells were harvested and resuspended in the infiltration buffer (10 mM MES, 10 mM MgCl₂, and 200 μM acetosyringone), adjusted to the required OD₆₀₀, and incubated for 3 h before use. The plasmids were infiltrated into 4-week-old N. benthamiana leaves by injection; each plasmid was infiltrated into more than five N. benthamiana plants (∼20 leaves). For co-expression of multiple constructs, the Agrobacterium suspensions carrying the different constructs were thoroughly mixed before infiltration.

The transient expression on the apple leaves was performed as previously reported (Zhang Q. et al., 2018), with slight improvements. Agrobacterium suspensions used for transient expression in apple were used to prepare 50 ml, and the processing method was as described above. The pCAMBIA1302 constructs and pFGC5941 constructs were infiltrated into the leaves of 5-week-old tissue-cultured plantlets under a vacuum of 100 kPa for 10 min, and the infiltrated plantlets were returned to the MS medium and cultured in a growth cabinet. The infiltrated plantlets of overexpression and gene silencing were used for the experiments after 2 and 7 days of culture, respectively. Each plasmid was infiltrated into more than the eight apple plantlets (∼40 leaves).

A Y2H system was performed to screen for VmEP1-interacting proteins (Ito et al., 2001). The cDNA library was constructed into the prey vector pGADT7 using mRNA isolated from the junction of diseased (infected by V. mali wild type strain 03-8) and healthy twigs. The recombinant BD-VmEP1 and pGADT7-cDNA libraries were co-transformed into yeast strain AH109 using the lithium acetate method according to the Yeast Protocols Handbook (Clontech, United States). To verify the interaction between VmEP1 and MdPR10, BD-VmEP1 and AD-MdPR10 were co-transformed into the yeast strain AH109. The yeast transformants were cultured on synthetic dropout medium lacking leucine, tryptothan, and medium lacking leucine, tryptophan, and histidine (SD/-Leu-Trp-His), and the positive colonies were transferred to synthetic dropout medium lacking leucine, tryptophan, and histidine and containing X-α-Gal (SD/-Leu-Trp-His + X-α-Gal) for confirmation of the interaction. The yeast transformation and interaction tests were performed following the instructions in the Yeast Protocols Handbook (Clontech, United States).

Virus-induced gene silencing (VIGS) in plant was performed as previously described (Lu et al., 2003). A 300-bp sequence of NbPR10, predicted by sol genomics network (SGN) VIGS tool¹, was amplified and cloned into pBinary TRV2 vectors between the BamH1 and EcoR1 sites in the antisense orientation. A TRV2 construct expressing green fluorescent protein (GFP) served as a control. The primers used for PCR amplification are listed in Supplementary Table 1. The two largest leaves of four-leaf-stage N. benthamiana plants were pressure-infiltrated with GV3101 A. tumefaciens strains containing a mixture of TRV1 (OD₆₀₀ = 0.4) and the VIGS construct or GFP control of each target protein at OD₆₀₀ = 0.5. The plants were used for assays or check gene-silencing levels by real-time quantitative PCR (qRT-PCR) 3 weeks later. A. tumefaciens transient expressions in combination with P. capsici infection were carried out as described previously described.

The samples were collected and ground in liquid nitrogen. For each sample, the total RNA was extracted using an EasyPure^® Plant RNA Kit (cat. ER301, Transgen, Beijing, China) following the instructions of manufacturer, and genomic DNA was freed by treatment with DNAse 1. A NanoDrop spectrophotometer (ND-1000, Wilmington, DE, United States) was used to quantify RNA. cDNA was synthesized using MultiScribe^TM Reverse Transcriptase (Thermo Fisher Scientific, MA, United States). The qRT-PCR was performed using a Roche LightCycler^® 96 SW1.1 Real-Time PCR System (Roche, IN, United States) with 2 × RealStar Green Power Mixture (GeneStar, Beijing, China). The EF-1a from Malus was used as an endogenous reference gene (Yin et al., 2016). The relative transcript level of each gene was calculated by the 2^−ΔΔCT method (Livak and Schmittgen, 2001). All the treatments were performed in the three replicates, and all the experiments were repeated three times. The data from the three replicates were used to calculate the means and SDs. A statistical analysis was performed using Student’s t-test implemented by SPSS software (IBM Corp., Armonk, NY, United States) (P < 0.05). The PCR was performed under the following conditions: 95°C (10 min), 40 cycles at 95°C (15 s), 60°C (30 s), and 72°C (30 s), followed by a melting curve step, 95°C (15 s), 60°C (60 s), and 95°C (15 s). The gene-specific primers used for the qRT-PCR are listed in Supplementary Table 1.

The heterologous expression of VmEP1 was verified by semi-quantitative PCR (the primers are shown in the Supplementary Table 1). cDNA products measuring 2 μl served as templates and were amplified with 2xTaq Plus MasterMix (Dye) (CWBIO, China), and in the presence of the specific primers for VmEP1. The gene-specific primers used for RT-PCR are listed in Supplementary Table 1. The reactions were carried out using a Life Technologies ProFlex PCR System (United States). The PCR was performed under the following conditions: 95°C (3 min); 30 cycles at 95°C (30 s), 58°C (30 s), and 72°C (30 s); 72°C (2 min); 4°C holding. The PCR products were loaded onto 1% agarose gels with Gel Green (COFIT^® Bioscience, China). Images of the agarose gels were acquired with a G:BOX F3 Gel Documentation System (Syngene, MD, United States).

The protein fusions were overexpressed in N. benthamiana using Agrobacterium-mediated transient expression. The leaf samples were collected at 48 hpi. The proteins were extracted using native lysis buffer (50 mM Tris–HCl pH 7.5, 150 mM NaCl, 1 mM EDTA, 0.5% NP-40, 1 × protease inhibitor mixture) with protease inhibitor mixture and 1 mM PMSF. To immunoprecipitated GFP-tagged MdPR10 or NbPR10, the protein extracts were incubated with GFP-Trap^®_A magnetic beads (Chromotec) for 2 h on a rotator at 4°C. The resulting samples were analyzed by immunoblotting. The samples were loaded onto a 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) gel run with 1 × SDS running buffer for 30 min at 10 mA and 2 h at 15 mA. The gels were blotted onto a nitrocellulose membrane for 2 h at 65 V, then stained with Ponceau solution to show loading and transfer. The membranes were blocked in 5% milk in 1 × TBST [TBS (50 mM Tris–HCl, 150 mM NaCl, and pH 7.4) with Tween-20 0.1% (vol/vol)] for 1 h before the addition of primary antibodies overnight: either a monoclonal GFP antibody raised in mouse at 1:1,000 dilution (cat. AF0159; Beyotime) or a monoclonal hemagglutinin (HA) antibody raised in mouse at 1:1,000 (cat. no. 26183; Abcam). The membrane was washed with 1 × TBST (0.1% Tween 20) three times for 10 min each before the addition of the secondary antibody at 1:8,000 dilution with goat anti-mouse lgG-HRP antibody (cat. A0216; Beyotime) for 1 h. ECL Western Blotting Substrate (cat. PE0030; Beijing Solarbio Science and Technology, Beijing, China) was used according to the instructions of manufacturer.

Callose deposition of N. benthamiana and apple leaves were stained with aniline blue as described (Adam and Somerville, 1996) and observed using a fluorescence microscope. Callose accumulation was quantified using Image J software and data were analyzed by Student’s t-test or Tukey’s multiple comparisons test.

The patches of N. benthamiana leaves were imaged 48 h after agro-infiltration using an Olympus FV3000 laser scanning microscope (Olympus Corporation, Tokyo, Japan). The GFP and yellow fluorescent protein (YFP) were imaged at excitation/emission wavelengths of 488 and 561 nm, respectively.

The VmEP1 was previously demonstrated to contribute to V. mali full virulence (Li et al., 2015). As an approach to illustrate this, VmEP1 lacking the signal peptide was transiently expressed in the apple leaves followed by inoculation with V. mali. As a result, the average number of lesions in the leaves expressing VmEP1 increased by ∼49% when compared with those expressing GFP control (Figures 1A,B). The relative biomass of V. mali in the apple leaves expressing VmEP1 was about 1.3 times that of the control (Figure 1C). The expression of VmEP1 was determined by semiquantitative RT-PCR (Figure 1D). These results indicated that the transient expression of VmEP1 in the apple leaves enhanced the host susceptibility to V. mali.

We also tested whether VmEP1 could facilitate the infection of other phytopathogens. For this, the VmEP1 was transiently expressed in the model plant N. benthamiana prior to inoculation with P. capsici and S. sclerotiorum. As shown (Supplementary Figures 1A–D), the average lesion diameters caused by P. capsica and S. sclerotiorum in the leaves expressing VmEP1 were ∼19 and ∼14% higher than those expressing the GFP control, respectively. These results indicated that VmEP1 could enhance the plant susceptibility to both oomycetes (P. capsici) and fungi (V. mali and S. sclerotiorum).

To clarify the underlying virulence mechanism of VmEP1, we aimed to find those host components that it targets. For this, we adopted the Y2H assay approach to screen the candidate targets using VmEP1 as bait. MdPR10 was one of the candidate proteins screened by Y2H assay (Figure 2A). To validate the VmEP1-MdPR10 interaction, a reciprocal BiFC assay was first performed. It showed that VmEP1 interacted with MdPR10 in vivo (Figure 2). To further confirm their interaction, we performed a Co-IP assay in N. benthamiana. As shown in Figure 2C, a specific signal for VmEP1-HA was clearly observed in the MdPR10-GFP immunoprecipitated (Figure 2C), indicating that VmEP1 interacted with MdPR10 in the plants.

Since VmEP1 also facilitates pathogen infection in N. benthamiana, we analyzed the homologous sequence of MdPR10 in N. benthamiana (Supplementary Figure 2A) and tested whether it interacted with VmEP1 similarly. As expected, the BiFC and Co-IP assays showed that the yellow fluorescence and a specific signal for VmEP1 were observed (Supplementary Figures 2B,C), indicating that the VmEP1 also interacted with N. benthamiana PR10 (NbPR10). In addition, we tested the interaction between an empty vector cYFP and nYFP. The yellow fluorescence was not detected (Supplementary Figure 3), indicating the results observed were all positive results.

To investigate the role of MdPR10 during phytopathogen infection, we infiltrated A. tumefaciens cells carrying an MdPR10 fusion construct in the apple leaves, followed by inoculation with V. mali. The average lesion diameter of apple leaves expressing MdPR10 was ∼43% smaller than that of the control (Figures 3A,B). Moreover, the relative biomass of V. mali in the apple leaves expressing MdPR10 was ∼0.6 times that of the control (Figure 3C). The qRT-PCR results showed that the gene expression level of MdPR10 increased significantly (Figure 3D) and immunoblot analysis of MdPR10-GFP indicated MdPR10 was successfully expressed (Figure 3E). We also inoculated the apple leaves with Colletotrichum gloeosporioides. The results showed that the lesion diameter on the apple leaves expressing MdPR10 decreased (Supplementary Figure 4). To expound on these results, we further silenced MdPR10 in the apple leaves by Agrobacterium-mediated transient expression and inoculated the leaves with V. mali. The average lesion diameter of the apple leaves silencing MdPR10 was ∼25% higher than that of the control (Figures 3F,G). The relative biomass of V. mali in the apple leaves silencing MdPR10 was ∼1.27 times that of the control (Figure 3H). The qRT-PCR analysis showed that the transcript levels of MdPR10 were reduced by ∼50% (Figure 3L). Figure 3M illustrates a schematic diagram of the construction of the MdPR10 silencing vector. These results collectively suggested that MdPR10 positively contributed to the apple immunity and disease resistance.

To further confirm the above results, we transiently expressed MdPR10 in N. benthamiana and inoculated with P. capsici. The average lesion diameter of N. benthamiana leaves expressing MdPR10 was ∼32% smaller than that of the control (Supplementary Figures 5A,B). Furthermore, we silenced NbPR10 in N. benthamiana by VIGS and inoculated P. capsici. It showed that the average lesion diameter of N. benthamiana leaves silencing NbPR10 was ∼29% larger than that of the control (Supplementary Figures 5C,D). The silencing efficiency of NbPR10 was tested by qRT-PCR, and the result showed that the NbPR10 expression level was reduced by ∼80% (Supplementary Figure 4E). These results indicated that NbPR10 positively contributed to the plant immunity and disease resistance as well.

To address if the PR10 is a virulence target of VmEP1, VmEP1 deletion mutants (Li et al., 2015) were inoculated on the apple leaves silencing MdPR10. The average lesion diameter of the VmEP1 deletion mutants was ∼14% higher than that of the control (Figures 3I,J). The relative biomass of V. mali in the apple leaves silencing MdPR10 was significantly greater than that of the control (Figure 3L). These results suggested that PR10 was a functional target for VmEP1. The relative lesion diameter growth rate of V. mali (∼25%) was higher than that of the VmEP1 deletion mutants (∼15%), indicating PR10 was essential for VmEP1 virulence function. These results indicated that the PR10 was a functional target for VmEP1.

We detected the callose deposition by staining with aniline blue after transient expression of MdPR10. The results showed that the accumulation of callose was much greater in the apple leaves expressing MdPR10 than that of the empty vector (Figure 4). However, the callose deposition in the apple leaves co-expressing MdPR10 and VmEP1 was clearly attenuated, compared with the experimental group expressing MdPR10 alone (Figure 4). To validate these results, we detected the transcript level of MdCalS5, a key enzyme for callose synthesis (Xie et al., 2012), by qRT-PCR. It showed that the expression level of MdCalS5 in the apple leaves expressing MdPR10 was about two times higher than the control expressing the empty vector alone. Additionally, the co-expression of MdPR10 and VmEP1 attenuated the transcript level of MdCalS5 by ∼34% compared with the apple leaves expressing MdPR10 (Figure 4C). These results indicated MdPR10 triggered callose deposition, and VmEP1 inhibited the response mediated by MdPR10.

To verify whether the expression of VmEP1 affected the pathogen-associated molecular pattern (PAMP)-triggered callose accumulation, we examined the callose deposition in N. benthamiana leaves after treatment with INF1 elicitin (Kamoun et al., 1997), a PAMP of Phytophthora infestans (Kamoun et al., 1998). The result showed that VmEP1 significantly suppressed the callose deposition induced by INF1 (Supplementary Figure 6). Together, these results indicated that VmEP1 could effectively inhibit the accumulation of callose in the plants.

Since VmEP1 attenuated the accumulation of callose induced by MdPR10, we speculated that VmEP1 might affect the resistance of MdPR10 to phytopathogens. To test this hypothesis, MdPR10 and VmEP1 were transiently co-expressed in apple. A V. mali inoculation assay showed that the average lesion diameter of the apple leaves co-expressing VmEP1 and MdPR10 was similar with that of the control, and was about 43% higher than the average lesion diameter of the apple leaves expressing MdPR10 (Figures 5A,B). These results indicated that the function of MdPR10 to suppress the infection of V. mali was weakened by VmEP1. The changed trend of relative V. mali biomass was consistent with the average lesion diameter (Figure 5C). Immunoblot analysis showed that MdPR10 successfully expressed (Figure 5D). These results showed that VmEP1 promoted infection of phytopathogens by inhibiting the callose deposition induced by MdPR10.