Two members of the velvet family, VmVeA and VmVelB, affect conidiation, virulence and pectinase expression in Valsa mali

Abstract

Velvet protein family members are important fungal-specific regulators which are involved in conidial development, secondary metabolism and virulence. To gain a broader insight into the physiological functions of the velvet protein family of Valsa mali, which causes a highly destructive canker disease on apple, we conducted a functional analysis of two velvet protein family members (VmVeA and VmVelB) via a gene replacement strategy. Deletion mutants of VmVeA and VmVelB showed increased melanin production, conidiation and sensitivity to abiotic stresses, but exhibited reduced virulence on detached apple leaves and twigs. Further studies demonstrated that the regulation of conidiation by VmVeA and VmVelB was positively correlated with the melanin synthesis transcription factor VmCmr1. More importantly, transcript levels of pectinase genes were shown to be decreased in deletion mutants compared with those of the wild-type during infection. However, the expression of other cell wall-degrading enzyme genes, including cellulase, hemi-cellulase and ligninase genes, was not affected in the deletion mutants. Furthermore, the determination of pectinase activity and immunogold labelling of pectin demonstrated that the capacity for pectin degradation was attenuated as a result of deletions of VmVeA and VmVelB. Finally, the interaction of VmVeA with VmVelB was identified through co-immunoprecipitation assays. VmVeA and VmVelB play critical roles in conidiation and virulence, probably via the regulation of the melanin synthesis transcription factor VmCmr1 and their effect on pectinase gene expression in V. mali, respectively.

Keywords: Apple Valsa Canker; conidiation; immunogold; melanin accumulation.

Figure 1

Structure and sequence analyses of velvet proteins in Valsa mali. (A) The open reading frame (ORF) of VmVeA consists of 1799 bp which is interrupted by a single 77‐bp intron and encodes a protein with 573 amino acids. VmVelB consists of a 1836‐bp ORF interrupted by five introns and encodes a protein with 459 amino acids. No introns were found in VmVelC and VmVosA. All four velvet genes contain the common velvet domain. nt, nucleotide. (B) A phylogenetic tree for the V. mali velvet genes was constructed using neighbour‐joining analysis with 1000 bootstrap replicates. The numbers on the branches represent the percentage of replicates supporting each branch. The labels on the right indicate the accession numbers in GenBank and the fungal species. The bar represents 20% sequence divergence.

Figure 2

Colour changes in VmVeA and VmVelB deletion mutants. (A) Mycelial growth on potato dextrose agar (PDA) for 2 days. (B) Up‐regulation of the melanin biosynthesis‐related genes (Table S1, see Supporting Information). RNA samples were isolated from 4‐day‐old cultures of the wild‐type (WT), deletion mutants and complemented mutants of VmVeA and VmVelB, and the transcript level of each gene was determined as described. The experiment was carried out in triplicate and the data were analysed using the protected Fisher's least significant difference (LSD) test.

Figure 3

Effects of inactivation of VmVeA, VmVelB (A) and/or VmCmr1 (B) on conidiation (asexual reproduction). Quantification of pycnidia produced on potato dextrose agar (PDA) was carried out as described. Arrows indicate pycnidia on PDA medium. WT, wild‐type.

Figure 4

Effects of inactivation of VmVeA and VmVelB on the response of Valsa mali to abiotic stresses. (A) Cultures of the wild‐type (WT), VmVeA and VmVelB deletion mutants, and complemented mutants, grown on potato dextrose agar (PDA) supplemented with 0.5 m KCl, 3 mm H₂O₂, 200 mg/L Congo red (CR) or 0.01% sodium dodecyl sulfate (SDS). Images were taken after 7 days on KCl and 3 days on the other inhibitors. (B) Inhibition of growth rate on PDA with inhibitor compared with PDA without stress (given in Table 1). Different letters represent a statistically significant difference (P < 0.05). Bars indicate standard deviations from the mean of three replicates.

Figure 5

Phenotypes of leaves and twigs inoculated with VmVeA and VmVelB deletion mutants. (A) Apple leaves were inoculated with mycelium agar plugs from the wild‐type (WT), deletion mutants and complemented mutants. Lesion sizes were determined at 3 days post‐inoculation (dpi). PDA, potato dextrose agar. (B) Apple twigs were inoculated with mycelium agar plugs from the wild‐type, deletion mutants and complemented mutants. Lesion sizes were determined at 9 dpi. (C) Quantification of lesion sizes on apple leaves at 3 dpi (twigs at 9 dpi). Different letters represent a statistically significant difference in respective lesion size (P < 0.05). Bars indicate standard deviations of the mean of three individual host plants. The experiment was repeated three times.

Figure 6

Transcript levels of pectinase, hemi‐cellulase, cellulase and ligninase genes determined by quantitative real‐time polymerase chain reaction (qRT‐PCR) in the wild‐type (WT), VmVeA and VmVelB deletion mutants. RNA samples were isolated from the border of Valsa mali‐colonized apple tree bark at 3 days post‐inoculation (dpi), and transcript levels of pectinase, hemi‐cellulase, cellulase and ligninase genes in the wild‐type, VmVeA and VmVelB deletion mutants were quantified by qRT‐PCR. The transcript level of the glyceraldehyde‐6‐phosphate dehydrogenase gene (G6PDH) was used to normalize different samples. Transcript levels of the wild‐type were set to unity. The mean and standard deviation were calculated with data from three independent biological replicates. Data from three replicates were analysed with Student's t‐test. Asterisks represent a significant difference in transcript levels (P < 0.05).

Figure 7

Inactivation of VmVeA and VmVelB differentially affects pectinase formation in Valsa mali. (A) Pectinase activities of supernatants of wild‐type (WT), VmVeA and VmVelB deletion mutants on synthetic medium (SM) supplemented with pectin as sole carbon source were assayed at 6, 12, 24, 48 and 72 h after pre‐growth in potato dextrose agar (PDA) for 48 h. Enzymatic activities are given in arbitrary units and related to the respective mycelium dry weight of the strain at the respective time point. Bars indicate standard deviations of the mean of three replicates. (B) Pectinase activities of samples from lesion and healthy bark border (3 mm in lesion and 2 mm in healthy bark) were calculated with similar amounts of sample. Experiments are the means of six biological replicates and the standard deviations are given by vertical bars. (C) Immunogold labelling of pectin on infected and uninfected bark tissue. Pectin labelling of host cell wall (HCW). Gold particles (arrows) were dense in intact bark, but reduced labelling was found in proximity to the hyphae (H). Bars, 0.5 μm. (D) Labelling density of pectin on healthy and infected tissues [at 3 days post‐inoculation (dpi)] using different strains. The density of labelling is expressed as the number of gold particles/μm². Bars indicate standard deviations of the mean from 15 micrographs. Mean densities of different strains were analysed using the protected Fisher's least significant difference (LSD) test. Different letters represent a statistically significant difference (P < 0.05).

Figure 8

Co‐immunoprecipitation assays for the interactions between VmVeA and VmVelB. Total protein (Total) isolated from VmVeA‐His, VmVelB‐Flag and VmVeA‐His/VmVelB‐Flag transformants. Proteins eluted from anti‐FLAG immuno‐affinity columns were visualized with anti‐FLAG and anti‐His antibodies.