Botrytis cinerea Transcription Factor BcXyr1 Regulates (Hemi-)Cellulase Production and Fungal Virulence

Abstract

ABSTRACT

Botrytis cinerea is an agriculturally notorious plant-pathogenic fungus with a broad host range. During plant colonization, B. cinerea secretes a wide range of plant-cell-wall-degrading enzymes (PCWDEs) that help in macerating the plant tissue, but their role in pathogenicity has been unclear. Here, we report on the identification of a transcription factor, BcXyr1, that regulates the production of (hemi-)cellulases and is necessary for fungal virulence. Deletion of the bcxyr1 gene led to impaired spore germination and reduced fungal virulence and reactive oxygen species (ROS) production in planta. Secreted proteins collected from the bcxyr1 deletion strain displayed a weaker cell-death-inducing effect than the wild-type secretome when infiltrated to Nicotiana benthamiana leaves. Transcriptome sequencing (RNA-seq) analysis revealed 41 genes with reduced expression in the Δbcxyr1 mutant compared with those in the wild-type strain, of which half encode secreted proteins that are particularly enriched in carbohydrate-active enzyme (CAZyme)-encoding genes. Among them, we identified a novel putative expansin-like protein that was necessary for fungal virulence, supporting the involvement of BcXyr1 in the regulation of extracellular virulence factors.

IMPORTANCE PCWDEs are considered important components of the virulence arsenal of necrotrophic plant pathogens. However, despite intensive research, the role of PCWDEs in the pathogenicity of necrotrophic phytopathogenic fungi remains ambiguous. Here, we demonstrate that the transcription factor BcXyr1 regulates the expression of a specific set of secreted PCWDE-encoding genes and that it is essential for fungal virulence. Furthermore, we identified a BcXyr1-regulated expansin-like gene that is required for fungal virulence. Our findings provide strong evidence for the importance of PCWDEs in the pathogenicity of B. cinerea and highlight specific PCWDEs that might be more important than others.

INTRODUCTION

Botrytis cinerea is a well-known plant-pathogenic fungus that is capable of infecting more than 1,000 plant species and is responsible for huge economic losses worldwide (1). During the infection process, B. cinerea secrets a wide range and large amounts of plant-cell-wall-degrading enzymes (PCWDEs) that macerate the plant tissue. Due to their high abundance and role in tissue degradation, PCWDEs have been considered virulence factors and have been studied for close to a century (2–4). However, despite intensive studies, there is no direct evidence for a role in pathogenicity for the vast majority of the B. cinerea PCWDEs (5–7).

The plant cell wall, which is composed mainly of cellulose, hemicellulose, and pectin, is among the first structures that phytopathogenic fungi encounter when colonizing plant tissues (8). The plant cell wall is a potential source of nutrients for pathogens, but intact walls are also an important barrier for effective defense against phytopathogenic fungi (7). The secreted PCWDEs contain a mixture of enzymes capable of degrading the different types of plant cell wall sugar polymers. The carbohydrate-active enzymes (CAZymes) of classes carbohydrate esterases (CEs), glycoside hydrolases (GHs), and polysaccharide lyases (PLs) are often known as plant-cell-wall-degrading enzymes due to their important roles in plant biomass decomposition (9). Necrotrophic plant pathogens, which derive nutrients from dead tissue, are characterized by the highest number and diversity of CAZymes in their genomes (5, 9). In B. cinerea, there are 275 putatively secreted CAZymes, and it is intuitive to assume that at least some of them contribute to the ability of the fungus to infect a wide range of host plants (8). A large number of B. cinerea PCWDEs have been mutated and analyzed; however, a clear effect on fungal virulence was demonstrated only in a small number of cases. The lack of a clear phenotype was attributed to functional redundancy, which interferes with efforts to identify PCWDEs as virulence determinants (10). Previously reported PCWDEs that contribute to fungal virulence include Bcpg1, Bcpg2, Xyn11A, BcAra1, BcXyl1, BcCBH, and BcEG (11–16). However, a recent study showed inconsistent results that no significant virulence defects were observed in mutants lacking Xyn11A or Xyl1, and Bcpg1 and Bcpg2 are the exclusive PCWDEs that were verified to be involved in the fungal virulence (17). Therefore, the general role of Botrytis PCWDEs in pathogenicity remains unclear.

In fungi, expression of PCWDE-encoding genes is usually governed by specific transcription factors. In Trichoderma reesei, the Xyr1 transcription factor functions as the main regulator of PCWDE-encoding genes (18–20), and its homologues are assumed to have a similar function in different fungi (21). Compared with saprophytic fungi, phytopathogenic fungi received less attention concerning the regulatory landscape associated with PCWDE production. Studies of the Xyr1/XlnR homologues in plant-pathogenic fungi showed functional differences between species. For example, Xyr1 regulated both xylanase and cellulase production in Fusarium graminearum, while XlnR in Fusarium oxysporum affected only xylanase gene expression (22, 23). In contrast, in Magnaporthe oryzae, XlnR regulated the pentose catabolic pathway but not genes encoding (hemi-)cellulolytic enzymes (24).

In order to elucidate the function of the Xyr1/XlnR homologue in B. cinerea and in particular its role in pathogenicity, we identified and analyzed bcxyr1, a B. cinerea homologue of T. reesei xyr1. Deletion of bcxyr1 resulted in reduced fungal virulence, and transcriptome sequencing (RNA-seq) analysis revealed that BcXyr1 positively regulates a total of 22 genes encoding predicted secreted proteins, which were enriched in CAZyme-encoding genes. Among them, a putative expansin-like protein was shown to be required for fungal virulence. These results unveil the previously unknown function of the PCWDE regulator BcXyr1 and its downstream target gene which contributes to virulence.

RESULTS

BcXyr1 is required for (hemi-)cellulase production.

A BLAST search of the B. cinerea genome ranked BcXyr1 (Bcin12g02060) as the closest homologue of T. reesei Xyr1 (GenBank AAO33577.1). The bcxyr1 gene encodes a predicted protein of 966 amino acids with a typical Zn₂Cys₆ fungal-type DNA-binding domain. To investigate the function of BcXyr1, we generated and characterized bcxyr1 deletion (Δbcxyr1) and overexpression (ox-bcxyr1) strains. Real-time PCR analysis showed a lack of transcript in the Δbcxyr1 strain and a 5.4-fold increase of bcxyr1 expression in the ox-bcxyr1 strain (see Fig. S1d in the supplemental material). Since the transcription factor Xyr1/XlnR is a major regulator in cellulose and xylan degradation in various fungi (21), we tested if BcXyr1 has a similar function. To this end, we performed the carboxymethyl cellulose (CMC)-containing Gamborg B5 plate assay and beechwood xylan-containing Gamborg B5 plate assay to compare the cellulase- and xylanase-producing ability among different strains. After incubation for 4 days, the Δbcxyr1 strain formed a significantly smaller halo diameter (1.39 cm) than the wild-type strain (2.04 cm), while the ox-bcxyr1 strain formed a larger halo (2.73 cm), confirming that BcXyr1 functions as an activator of cellulase production (Fig. 1a and b). When their xylanase-producing abilities were compared, however, there was difficulty in measuring the halo zones since all the three strains (wild type, Δbcxyr1, and ox-bcxyr1) formed irregularly shaped colonies when growing in the beechwood xylan-containing Gamborg B5 plates. For better assessment of (hemi-)cellulase production, fungi were grown in Gamborg liquid medium supplemented with 2% rice straw powder as a (hemi-)cellulase-inducing carbon source. The protein patterns of the culture supernatant displayed a distinguishable difference between the wild type and the mutant strains. After incubation for 24 h, the protein bands in the Δbcxyr1 sample were not only fewer but also lighter than those of the wild-type protein sample, while the sample of ox-bcxyr1 showed an opposite trend with stronger bands (Fig. 1c). Moreover, compared with the wild type, the carboxymethyl cellulase (CMCase) activity and xylanase activity of Δbcxyr1 were decreased by 69% and 84%, respectively. In contrast, the ox-bcxyr1 strain showed an 86% increase of CMCase and 32% increase of xylanase activity (Fig. 1d). These results confirmed that BcXyr1 regulates the expression of (hemi-)cellulase genes, similar to its homologues in some other fungi, such as Xyr1 in T. reesei (18, 19) and Xyr1 in Fusarium graminearum (22).

FIG 1

bcxyr1 is required for (hemi)cellulase production in B. cinerea. (a) Colonies were initiated by placing a 5-μL droplet containing 500 spores of wild-type (wt), Δbcxyr1, or ox-bcxyr1 strains onto a CMC plate, and the plates were incubated for 4 days and then stained by Congo red. (b) Average halo diameter of each strain after incubation for 4 days. (c) Three strains were grown in liquid Gamborg medium supplemented with 2% rice straw powder as a (hemi)cellulase inducer, and after 1 day of growth, 20 μL of culture supernatant of each strain was subjected to SDS-PAGE and then silver staining. (d) Xylanase activity and CMCase activity of culture supernatant samples of c were measured. Values that are statistically significantly different (P < 0.01) by two-tailed Student's t test from the wild-type values are indicated by an asterisk.

bcxyr1 is required for spore germination but dispensable for sporulation and mycelial growth.

Deletion of bcxyr1 had no effect on sporulation and mycelial growth (see Fig. S2 in the supplemental material). However, the bcxyr1 deletion strain had 50% reduced germination rates compared with the wild-type strain after incubation for 6 h on glass coverslips, whereas the overexpression strain had an opposite effect (Fig. 2a and b). Similarly, more than 70% of the wild-type spores germinated on Arabidopsis leaves at 8 h postinoculation (hpi), in comparison to almost no germination in the Δbcxyr1 strain and close to 100% spore germination in the ox-bcxyr1 strain (Fig. 2c).

FIG 2

Deletion of bcxyr1 affects spore germination. (a) Spore germination in CD + 1% sucrose medium. Spores were diluted to 10⁵/mL in liquid CD + 1% sucrose medium. A total of 15 μL of spore suspension of each strain was pipetted onto glass coverslips and incubated for 6 h. (b) Spore germination percentage of each sample after incubation for 6 h was counted under an inverted microscope. Four replicates were used for each sample. Values that are statistically significantly different (P < 0.01) by two-tailed Student’s t test from the wild-type values are indicated by an asterisk. (c) A total of 7.5 μL of spore suspension (10⁵/mL) in liquid CD + 1% sucrose medium was inoculated onto Arabidopsis leaves, images were taken at 8 hpi, and the scale bar indicates 20 μm.

BcXyr1 is essential for virulence.

Certain B. cinerea xylan-degrading enzymes have a cell-death-inducing activity (13, 15). Since (hemi-)cellulase production is significantly impaired in the Δbcxyr1 strain, it is possible that deletion of bcxyr1 also affects the secretion of cell-death-inducing proteins (CDIPs). To this end, the culture supernatants from strains which grew 24 h in Gamborg medium plus 2% rice straw powder (Fig. 1) were collected and filtered. The protein concentration in the filtrates was determined and diluted to 20 μg/mL in phosphate-buffered saline (PBS) and then injected into N. benthamiana leaves. After 5 days, treatment of N. benthamiana leaves with culture filtrate from the ox-bcxyr1 strain led to the strongest cell death, while treatment with culture filtrate from the Δbcxyr1 strain resulted in the weakest phenotype (Fig. 3a). The difference of cell death-inducing effect between strains suggests that there are secreted cell-death-inducing factors, which are regulated by BcXyr1.

FIG 3

bcxyr1 is required for fungal virulence. (a) The culture supernatants (20 μg/mL protein in PBS) from strains which grew 24 h in Gamborg medium supplemented with 2% rice straw powder were infiltrated into N. benthamiana leaves, and necrosis was observed 5 days after infiltration. PBS was used as a mock control. The black circles indicate the infiltrated area. (b) Arabidopsis leaves were inoculated with 7.5-μL droplets of spore suspension (10⁵ spores/mL in liquid CD + 1% sucrose medium), and typical lesions were developed at 80 hpi. (c) Average lesion size of sample b, and data are the means plus SD for eight lesions from three plants. (d) Arabidopsis leaves were inoculated with mycelial plugs (3 mm in diameter) from a CD + 1% sucrose agar plate, and typical lesions were developed at 34 hpi. (e) Average lesion size of sample d, and data are the means plus SD for eight lesions from three plants. Values that are statistically significantly different (P < 0.01) by two-tailed Student’s t test from the wild-type values are indicated by an asterisk.

A pathogenicity assay on Arabidopsis thaliana leaves showed reduced and increased lesion sizes for the Δbcxyr1 strain and the ox-bcxyr1 strain, respectively; at 80 hpi, the Δbcxyr1 mutant produced an average lesion of 0.19 cm compared with 0.35 cm by the wild-type and 0.63 cm by the ox-bcxyr1 strains (Fig. 3b and c). Infection with mycelial plugs reproduced these phenotypes (Fig. 3d and e), which ruled out a possible effect of spore germination rates. We obtained similar results by infection of tomato and cucumber leaves (see Fig. S3 in the supplemental material), confirming that BcXyr1 is necessary for the full virulence of this fungus.

bcxyr1 affects hyphal orientation and ROS induction in Arabidopsis.

To gain insight into the infection process, we stained infected Arabidopsis leaf tissue with lactophenol trypan blue, which stains dead cells (25). Although mostly germlings were observed for all the three strains in the inoculation spot at 1 day postinoculation (dpi) (Fig. 4), visually less fungal biomass was observed in leaves that were inoculated with the Δbcxyr1 strain. This difference is probably owing to the retarded spore germination of the Δbcxyr1 strain on Arabidopsis leaves since ungerminated spores can be easily washed off the plant surface during staining while germinated spores attach strongly through secretion of an extracellular matrix during germination (26). The ox-bcxyr1 strain developed oriented hyphae toward the periphery of the lesion at 2 dpi, and although the wild type formed oriented hyphae at 3 dpi, the ox-bcxyr1 showed a clearly far more ordered and aggressive hyphal pattern. Since the hyphal orientation is associated with the transition from local necrosis to a spreading lesion (27), a delay in hyphal orientation suggests weakened fungal virulence, which accords with the results of the pathogenicity test (Fig. 3b and c).

FIG 4

Arabidopsis leaves were inoculated with 7.5-μL droplets of spore suspension (10⁵ spores/mL in liquid CD + 1% sucrose medium). Leaves were cut at the designated time points, and samples were stained with lactophenol trypan blue. The large images show the entire lesion; squares show hyphal orientation and structure at the colony edge. Pictures were taken with the same magnification, and scale bar indicates 500 μm.

As a response to pathogen infestation, the plant releases large amounts of ROS to counteract the pathogen, known as oxidative burst (28, 29). It was shown that B. cinerea exploits this plant defense reaction and even contributes to oxidative burst by forming its own ROS (29). To check if H₂O₂ production correlates with BcXyr1-regulated fungal infection, 3,3'-diaminobenzidine (DAB) staining assay was performed to trace H₂O₂ production during the B. cinerea-Arabidopsis interaction. Plants that were inoculated with the ox-bcxyr1 strain accumulated more H₂O₂ than the wild-type-inoculated leaves at 24 hpi, while there was hardly any H₂O₂ detected in the leaf inoculated with the Δbcxyr1 strain at the same time point. Similar results were observed at 48 hpi that showed the highest accumulation of H₂O₂ in the ox-bcxyr1-inoculated leaves and the lowest accumulation in the Δbcxyr1-inoculated leaves (see Fig. S4 in the supplemental material). Therefore, the H₂O₂ level well reflected the BcXyr1-associated fungal virulence.

CAZymes are regulated by the BcXyr1.

In preparation for RNA-seq, B. cinerea wild-type strain B05.10 and Δbcxyr1 strains were precultured in liquid malt medium for 20 h, and then the mycelium was collected, washed, and transferred into liquid CD + 1% sucrose medium for another 7 h. Then samples were prepared for RNA isolation and further sequencing. After sequence quality control and mapping, the number of clean reads per sample ranged from 20.73 to 21.93 million, and resulted in clean bases per sample between 6.2 Gbp and 6.56 Gbp. We found 95 genes that were differentially expressed in the Δbcxyr1 strain compared with the wild-type strain, of which 41 were downregulated in Δbcxyr1 (see Table S1 in the supplemental material). These results were further validated by quantification of the expression of nine selected genes of the wild type and the Δbcxyr1 strain with real-time quantitative PCR (qRT-PCR) (see Table S2 in the supplemental material). We also analyzed gene expression in the ox-bcxyr1 strain and found that the ox-bcxyr1 strain and the Δbcxyr1 strain showed opposite expression trends in seven out of the nine selected genes. The two other genes (Bcin02g07640 and Bcin14g05500) were both regulated in the same trend in the deletion and the overexpression strains. A similar observation was reported in T. reesei, in which the Xyr1 overexpression strain did not show an opposite expression trend in all the target genes compared with a Xyr1-deficient strain (30); however, the reason for this phenomenon is currently unclear.

Further, the evolutionary genealogy of genes: nonsupervised orthologous groups (eggNOG) database was used to analyze the function classification of the BcXyr1-regulated genes. As shown in Fig. 5a, the main categories represented among the 41 downregulated genes in the Δbcxyr1 strain were those participating in carbohydrate transport and metabolism and amino acid transport and metabolism. In contrast, the 54 upregulated genes were enriched with those involved in secondary metabolite biosynthesis, transport, and catabolism. The Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis also indicated that deletion of bcxyr1 repressed the expression of genes involved in carbohydrate metabolic pathways, including starch and sucrose metabolism, other glycan degradation, galactose metabolism, and N-glycan biosynthesis, and stimulated the expression of genes involved in carbon metabolism and the biosynthesis of antibiotics (Fig. 5b).

FIG 5

Analysis of differentially expressed genes in the Δbcxyr1 mutant compared the wild-type strain. (a) eggNOG function classification of downregulated genes (left) and upregulated genes (right) in the Δbcxyr1 strain. (b) KEGG pathways enrichment analysis of downregulated genes (left) and upregulated genes (right) in the Δbcxyr1 strain.

Out of the 41 genes that were downregulated in the Δbcxyr1 strain, 22 genes harbor a predicted signal peptide, and half of them were CAZymes, including those participating in plant cell wall degradation (see Table S3 in the supplemental material). For instance, the previously reported virulence-associated gene bcpg2 (31) was downregulated in the Δbcxyr1 strain, which suggests that BcXyr1 positively regulates the expression of PCWDE genes. In addition to carbohydrate metabolism, several genes related to nitrogen transport and metabolism were downregulated, including bccrnA (Bcin01g06290) that encodes a putative nitrate/nitrite transporter, bcniaD (Bcin07g01270), which encodes nitrate reductase that is involved in the first step of nitrate assimilation (32), and a gene encoding putative amino acid permease (Bcin05g02810). Furthermore, bcxyr1 was required for the expression of the glucose-oxidase-encoding gene bcgod1 (Bcin14g05500) (33) and the galactose-oxidase-encoding gene bcgox1 (Bcin13g05710), which generate H₂O₂ as a by-product during the oxidation of glucose or galactose, respectively. Deletion of bcxyr1 also affected the expression of the CDIP BcNep2, which causes necrosis in dicotyledonous plant species (34). On the other side, genes upregulated in the Δbcxyr1 strain were enriched in transporters, particularly MFS transporters (Bcin06g06880, Bcin10g04810, Bcin09g05240, Bcin12g01880, and Bcin09g00730), and four genes (Bcin05g08400, Bcin11g02640, Bcin03g06680, and Bcin02g07640) involved in secondary metabolism were also negatively regulated by BcXyr1. Interestingly, a probable succinyl-coenzyme A (CoA)-ligase-encoding gene (Bcin11g03450) was upregulated in the Δbcxyr1 strain. Recently, succinate was shown to be connected to stress responses and cell death (35), suggesting this gene might play a role in the host-pathogen interaction.

To investigate the function of the downregulated CAZyme genes in the bcxyr1 deletion strain, we selected genes that encode secreted proteins, which have been detected in the secretome. Among these genes, the protein product of a putative expansin-like gene (Bcin01g02460) was found previously in the B. cinerea secretome (36–38), and therefore, we generated a deletion strain of this gene that we named Δbcexl1 and tested its pathogenicity. The Δbcexl1 strain was hypovirulent and caused smaller lesions than the wild-type strain (Fig. 6). Transformation of the Δbcexl1 strain with the native bcexl1 gene restored full virulence, confirming that the reduced pathogenicity of Δbcexl1 resulted from deletion of the bcexl1 gene. To the best of our knowledge, expansin genes have not been reported to be involved in fungal pathogenesis, and therefore, how this gene functions during the plant-pathogen interaction remains to be elucidated.

FIG 6

Deletion of bcexl1 affects fungal virulence. Spores were diluted to 10⁵/mL in liquid CD + 1% sucrose medium, and 7.5 μL of spore suspension was inoculated onto Arabidopsis leaves. (a) Typical lesions developed at 75 hpi. (b) Average lesion size at 75 hpi. Data are the means ± SD of eight leaves for each strain. Columns marked by different letters represented statistical differences (P < 0.01).

Virulence is further reduced but not abolished in the Δbcexl1Δbcxyr1 double deletion strain.

To estimate if the effect of losing both genes is greater than that of losing a single one, we generated a Δbcexl1Δbcxyr1 double-deletion strain by deleting the bcxyr1 gene on a background of the Δbcexl1 strain. Compared with either Δbcexl1 or Δbcxyr1, the double-deletion strain showed a weaker virulence at 84 hpi, although pathogenicity was not completely abolished (Fig. 7). These results suggest that (i) along with bcexl1, other genes regulated by bcxyr1 also contribute to fungal virulence; (ii) bcexl1 is not fully controlled by bcxyr1, and the remaining expression of bcexl1 in the Δbcxyr1 strain still plays a role in virulence, reflecting the importance of bcexl1 in pathogenicity; and (iii) repressing a portion of virulence-associated genes will probably be insufficient to achieve complete elimination of the fungal virulence. Therefore, bcexl1 and bcxyr1 play crucial but distinct roles in fungal virulence.

FIG 7

Double deletion of bcexl1 and bcxyr1 results in further reduction of fungal virulence. Spores were diluted to 10⁵/mL in liquid CD + 1% sucrose medium, and 7.5 μL of spore suspension was inoculated onto Arabidopsis leaves. (a) Typical lesions developed at 84 hpi. (b) Average lesion size at 84 hpi. Data are the means ± SD of eight leaves for each strain. Columns marked by different letters represented statistical differences (P < 0.01).

DISCUSSION

PCWDEs are considered essential components of B. cinerea virulence arsenal; however, conclusive evidence for their role in pathogenicity is limited, possibly due to functional redundancy. In this study, we showed that the B. cinerea transcription factor BcXyr1 regulates (hemi-)cellulase production as well as the expression of additional PCWDEs. Deletion of bcxyr1 resulted in reduced virulence, demonstrating the contribution of PCWDEs to fungal virulence. We further identified a BcXyr1-regulated expansin-like gene (bcexl1), which was required for full virulence.

The transcription factor Xyr1 (Xlr1/XlnR) is a major regulator of genes encoding xylanolytic and cellulolytic enzymes in different fungi (21). In this study, deletion of bcxyr1 reduced the secretion of xylanases and cellulases, confirming the regulation of (hemi-)cellulase genes by BcXyr1. This result was further confirmed by RNA-seq analysis, which revealed 41 genes that were positively regulated by BcXyr1, of which 13 genes were categorized as CAZymes. Deletion of bcxyr1 significantly affected fungal virulence (Fig. 3), in contrast to studies in other fungi, such as Magnaporthe oryzae (24), Fusarium oxysporum (23), and Fusarium graminearum (22), in which bcxyr1 homologues were dispensable for virulence. There could be several reasons underlying the difference. First, the function of Xyr1/XlnR varies in different fungi, for example, in M. oryzae, XlnR was found to be involved in the transcriptional control of the pentose catabolic pathway but not genes encoding (hemi-)cellulolytic enzymes (24), while in F. oxysporum, XlnR controlled xylanase gene expression but did not affect cellulase activity (23). Second, although disruption of Xyr1 regulated both cellulase and xylanase production in F. graminearum, the residual PCWDEs could be sufficient to achieve infection (22). In addition, M. oryzae and most Fusarium species are hemibiotrophs and the role of XlnR/Xyr1 might differ in these classes compared with necrotrophic pathogens, such as B. cinerea.

A working model proposed that during the early infection phase, B. cinerea secretes necrosis-inducing factors (CDIPs and toxins) that produce small spots of dead tissue, in which the fungus can establish itself and use as foci for spreading in the following stages (25). The N. benthamiana infiltration test showed that the secretome of the Δbcxyr1 strain displayed weaker cell-death-inducing activity than that of the wild type (Fig. 3a), suggesting that BcXyr1 may regulate the expression of certain PCWDEs that are also CDIPs (17, 39).

Among the 11 BcXyr1-regulated secreted CAZyme-encoding genes (Table S3), we investigated bcexl1 that encodes a putative expansin-like protein, and it was found essential for B. cinerea virulence. Unlike most other types of PCWDEs, BcExl1 is the only expansin-like protein found in the B. cinerea secretome, possibly explaining the clear phenotype of the Δbcexl1 strain. Therefore, the impaired virulence of the Δbcxyr1 strain is at least partially due to the reduced expression of bcexl1 and possibly other PCWDEs, which also correlates with the observation that the secretome of the Δbcxyr1 strain displayed a weaker cell-death-inducing ability. Expansins are proteins that are present primarily in the plant kingdom and are known to have cell-wall-loosening activity and to be involved in cell expansion and other developmental events (40). So far, expansin-like proteins have hardly been investigated in plant-pathogenic fungi, and it would be interesting and important to gain insight into the role of bcexl1 in the pathogen-plant interaction.

The development of CRISPR-Cas genome editing technology in B. cinerea enables the generation of multiple knockout mutants (17) and is expected to facilitate the evaluation of multiple genes at one time. Application of CRISPR-Cas will make it much easier to identify those virulence-associated ones from numerous downstream genes.

The CAZymes that are produced by plant-pathogenic fungi play a pivotal role in breaching the frontline of plant defense (5). Biotrophs, hemibiotrophs, and necrotrophs vary not only in the number of CAZyme genes but also in the expression pattern of CAZyme genes. For example, most CAZyme genes of the necrotroph B. cinerea were expressed during infections of lettuce leaves, ripe tomato fruit, and grape berries (8), while many CAZymes are not expressed in the biotroph Cladosporium fulvum during plant infection (41). Since Xyr1 (Xlr1/XlnR) is considered a major regulator of fungal cellulase and hemicellulose genes, we anticipate that a comparative analysis of the regulon of BcXyr1 and its homologues in pathogens of different lifestyles would contribute to better understanding the necrotrophic infection mechanisms of B. cinerea. Taken together, in this study, we identified the crucial PCWDE-regulating transcription factor BcXyr1 that regulates both (hemi-)cellulase production and fungal virulence in B. cinerea, providing evidence that supports the role of PCWDEs in the pathogenicity of B. cinerea.

MATERIALS AND METHODS

Fungi and plants.

Botrytis cinerea strain B05.10 was used as the wild-type strain. Potato dextrose agar (PDA; Acumedia, Lansing, MI) was used as maintenance and sporulation medium. Routine culture or growth assays were carried out in the following media: Gamborg B5 including vitamin mixture (GB5; Duchefa Biochemie) supplemented with the indicated carbon sources and Czapek Dox medium (CD; 0.3% NaNO₃, 0.05% KCl, 0.05% MgSO₄·7H₂O, 0.001% FeSO₄·7H₂O, 0.1% K₂HPO₄·3H₂O [pH 7.3]) supplemented with the indicated carbon sources. Solid media were prepared by adding 1.5% agar.

Arabidopsis thaliana ecotype Columbia (Col-0) plants were grown in a 14-/10-h light/dark cycle at 20°C. Tomato plants and cucumber plants were grown in a 16-/8-h light/dark cycle at 25°C. Mature leaves plants were used for infection assays. The collection of spores and infection procedures were performed as described previously (42).

Generation of B. cinerea mutant strains.

The B. cinerea B05.10 genome sequence was used to design primers (http://fungi.ensembl.org/Botrytis_cinerea/Info/Index). Name, purpose, and sequences of all the primers are listed in Table S4 in the supplemental material. Single-deletion mutants (Δbcxyr1) were generated by replacing the entire open reading frame (ORF) of bcxyr1 with a hygromycin resistance cassette (hph) (Fig. S1a). The hph cassette consists of the Aspergillus nidulans oliC promoter (PoliC), the hygromycin phosphotransferase gene (hph), and acr (B. cinerea enoyl- [acyl carrier protein] reductase) terminator (Tacr) (27). A similar tactic was applied to generate the bcexl1 deletion strain (see Fig. S5 in the supplemental material).

Constructs for the deletion of the bcxyr1 gene were produced by adding 500 bp of the 5′ flanking region of the desired gene on one side of the hph cassette and 500 bp of the 3′ flanking region on the other side by overlap PCR. A bcxyr1 complementation construct was prepared by assembling four PCR fragments using a Gibson assembly master mix kit (New England BioLabs [NEB]). The complementation cassette contained the upstream region of bcxyr1 and the bcxyr1 ORF, the Tcel5a termination signal (B. cinerea cel5a, GenBank accession number AY618929.1), a nr cassette that conferring nourseothricin resistance, and the downstream region of bcxyr1 (Fig. S1b).

A bcxyr1 overexpression cassette was prepared by placing the bcxyr1 ORF under the B. cinerea strong H2B promoter (27) and was flanked by a hph resistance gene (Fig. S1c). A fragment containing the 3′ part of the bcgapdh ORF (Bcin15g02120) and its termination sequence and a fragment of 600 bp adjacent to the bcgapdh termination sequence were then placed upstream and downstream of the PH2B-bcxyr1-hph cassette, respectively. The cassette was then inserted to the bcgapdh locus by homologous recombination.

To generate a bcexl1 complementation strain, an intact bcexl1 fragment (2,070 bp) containing 500 bp of the 5′ flanking sequence and 500 bp of the 3′ flanking sequence was amplified using genomic DNA as the template. The H1-green fluorescent protein (GFP) cassette (see Fig. S6 in the supplemental material) that was designed to tag the nuclei with GFP (25) was constructed by fusing the 3′ part of the histone H1 gene (Bcin02g04870), GFP, the Tcel5a termination signal, the nr gene, and the 3′ flanking region of the histone H1 gene together. The bcexl1 fragment and the H1-GFP cassette were cotransformed into the Δbcexl1 strain to generate the complementation strain. The H1-GFP cassette was transformed into wild-type strain to generate the H1-GFP strain.

A double-deletion mutant (Δbcexl1Δbcxyr1) was generated by replacing the bcxyr1 ORF with nr gene on the background of the Δbcexl1 strain. The bcxyr1 deletion cassette was constructed by adding 500 bp of the 5′ flanking region of bcxyr1 on one side of the nr gene and 500 bp of the 3′ flanking region on the other side by overlap PCR.

Genetic transformation of B. cinerea with the different DNA constructs was performed as described previously (27). Colonies that developed on the selection media were transferred to separate plates. At least 10 independent transformants for each mutant were obtained for each mutant, and genomic DNA was extracted from each colony and analyzed by PCR to verify the integration of the construct at the desired locus. Homokaryotic strains were obtained by single-spore isolation, and derived colonies were analyzed by PCR to verify that the strain is homokaryotic at the desired locus and additional rounds of single-spore isolation were performed in cases of impurity. At least four separate single-spore isolates were obtained for each strain, and similar experimental results were observed from independent transformants. In the initial analyses, we used bcxyr1 deletion strain Δbcxyr1-1a1 (abbreviated Δbcxyr1), bcxyr1 complementation strain Δbcxyr1-C1b (abbreviated Δbcxyr1-C), bcxyr1 overexpression strain ox-bcxyr1-6b (abbreviated ox-bcxyr1), bcexl1 deletion strain Δbcexl1-9a1 (abbreviated Δbcexl1), bcexl1 complementation strain Δbcexl1-C2i (abbreviated Δbcexl1-C), bcexl1 and bcxyr1 double-deletion strain Δbcexl1Δbcxyr1-3c (abbreviated Δbcexl1Δbcxyr1), and H1-GFP homologous transformant H1-GFP-7a (abbreviated H1-GFP).

(Hemi-)cellulase assay.

Spores were collected from 1-week-old PDA cultures as described previously (43). To compare (hemi-)cellulase production, 5 μL of spore suspensions (10⁵/mL) was inoculated into 1/2 Gamborg B5 solid medium supplemented with 0.5% carboxymethyl cellulose (241297; J&K Chemical Ltd., China) or 0.2% beechwood xylan (X4252; Sigma, Germany). After 4 days of fungal growth, a 0.1% Congo red solution and a 1 M NaCl solution was used to flood the plate sequentially, and the halo diameters of different strains were measured.

To further measure the (hemi-)cellulase activity of different strains, spore suspensions were added into the liquid malt medium in Erlenmeyer flasks to a final concentration of 10⁶/mL. Samples were incubated on an orbital shaker with agitation at 150 rpm for 20 h at 21°C. Then samples were centrifuged at 4,000 × g for 10 min, and the supernatant was removed. Mycelium pellets were washed with sterile deionized distilled water (DDW) and then centrifuged to decant the supernatant, and this washing cycle was repeated three times to get rid of the residual malt medium. Then 0.3 g of wet mycelial sample was inoculated into a 50-mL Erlenmeyer flask containing 10 mL of Gamborg B5 liquid medium supplemented with 2% rice straw powder as a (hemi-)cellulase inducer, and samples were incubated with agitation at 150 rpm for 24 h. Then samples were collected and centrifuged at 12,000 × g for 20 min, and the supernatants were filtered through 0.22-μm filters and subjected to (hemi-)cellulase activity measurement. Cellulase activity was determined as follows: 25-μL supernatant was incubated with a 25-μL CMC solution (2% in 0.05 M acetate buffer, pH 5.0) at 30°C for 30 min. Then 100 μL of dinitrosalicylic acid (DNS) solution was added and the mixtures were incubated at 95°C for 5 min. The absorbance at 540 nm was measured with a Multiskan GO microplate spectrophotometer (Thermo Scientific). Xylanase activity was determined as follows: 25-μL supernatant was incubated with 25-μL beechwood xylan solution (1% in 0.05 M acetate buffer, pH 5.0) at 30°C for 30 min. Then 100 μL of DNS solution was added and the mixtures were incubated at 95°C for 5 min. One international unit (IU) of the activity of cellulase or xylanase was defined as the amount of enzyme to liberate one micromole (μM) reducing sugars per minute from CMC or beechwood xylan.

Pathogenicity test.

Pathogenicity assays were performed on Arabidopsis, tomato, and cucumber plants as described previously with modifications (27). Briefly, leaves were inoculated with 7.5-μL droplets of spore suspension (10⁵/mL in liquid CD + 1% sucrose medium) or with mycelial plugs (3 mm in diameter) from a CD + 1% sucrose agar plate.

To test the induction of plant cell death by secreted proteins of different strains, the culture supernatants from strains which grew 24 h in Gamborg medium plus 2% rice straw powder were collected, filtered, and concentrated using 3-kDa MWCO Amicon Ultra-15 centrifugal filter units (Millipore) at 4°C. Then the concentrated proteins were diluted in PBS to a final concentration of 20 μg/mL and infiltrated into N. benthamiana leaves using syringes. Plants were kept in a growth chamber at 25°C for 5 days.

RNA isolation, sequencing, and data analysis.

Spore suspensions of B. cinerea wild-type strain 05.10 and Δbcxyr1 strain were added into liquid malt medium in Erlenmeyer flasks to a final concentration of 10⁶/mL, and three biological replicates were used for each sample. Samples were incubated on an orbital shaker with agitation at 150 rpm for 20 h at 21°C. Then samples were centrifuged at 4,000 × g for 10 min, and the supernatants were removed. Mycelial pellets were washed with sterile DDW and then centrifuged to decant the supernatant. This washing cycle was repeated five times to thoroughly get rid of the residual malt medium. For each sample, 0.5 g of wet mycelia was inoculated into a 50-mL Erlenmeyer flask containing 10 mL liquid CD + 1% sucrose medium and incubated with agitation at 150 rpm for 7 h. Then mycelia of each sample were harvested and ground under liquid nitrogen to extract total RNA using the RNAprep pure plant kit (TianGen, China).

A total amount of 1 μg RNA per sample was used as input material for the RNA sample preparations. Sequencing libraries were generated using NEBNext Ultra RNA library prep kit for Illumina (NEB, USA) following manufacturer’s recommendations, and index codes were added to attribute sequences to each sample. The clustering of the index-coded samples was performed on a cBot cluster generation system using TruSeq PE cluster kit v4-cBot-HS (Illumina) according to the manufacturer’s instructions. After cluster generation, the library preparations were sequenced on an Illumina platform and paired-end reads were generated.

Raw data (raw reads) of FASTQ format were first processed through in-house Perl scripts. In this step, the adaptor sequences and low-quality sequence reads were removed from the data sets. At the same time, Q20, Q30, GC content, and sequence duplication level of the clean data were calculated. These clean reads were then mapped to the B. cinerea 05.10 genome using the HISAT2 program (44). Only reads with a perfect match or one mismatch were further analyzed and annotated based on the reference genome. Gene expression levels were estimated by the number of fragments per kilobase per million fragments mapped (FPKM). Differential expression values were determined using DEseq (45), and the false discovery rate (FDR) of <0.05 and fold change of ≥2 were set as the thresholds for significantly differential expression.

Real-time quantitative PCR (qRT-PCR).

The RNA samples were reverse transcribed into cDNA using PrimeScript RT reagent kit with genomic DNA (gDNA) Eraser (TaKaRa, Dalian, China) according to the manufacturer’s protocol. For the reaction, SYBR Premix Ex Taq II (TaKaRa, Dalian, China) was used for 20-μL assays. Primers used were given in Table S4. Three replicates were performed per experiment. The bcgpdh gene (43) was used as a control gene to normalize data. qRT-PCR was performed using the CFX Connect real-time system (Bio-Rad). The experiments were repeated three times. Nine genes were chosen to validate the RNA-seq results using the qRT-PCR method.

Spore germination assay.

Spore suspension from each sample was prepared in liquid CD + 1% sucrose medium to a concentration of 10⁵/mL. A droplet of 15 μL was pipetted onto a microscope cover glass (0101050; Marienfeld, Germany), and kept in a humid chamber. Besides, 7.5 μL of spore suspension was pipetted onto Arabidopsis leaves and kept in a humid box. Spore germination was then examined using microscopes.

Lactophenol trypan blue staining.

Infected leaves were stained with lactophenol trypan blue as described previously (27). The Arabidopsis leaf tissue around the inoculated spot was cut and transferred into tubes containing 2 mL of fresh trypan blue staining solution (10 mL of lactic acid, 10 mL of glycerol, 10 mL of phenol [saturated with Tris-buffer, pH 8], and 10 mg of Trypan blue, mixed well with 10 mL of sterile DDW). The tubes were boiled for 2 min, and samples were then distained for 1 h in chloral hydrate solution (2.5 g chloral hydrate in 1 mL DDW). Stained samples were mounted on microscope slides and examined under an SMZ25 stereomicroscope (Nikon, Japan).

Hydrogen peroxide detection.

Arabidopsis leaves were inoculated with 7.5-μL droplets of spore suspension (10⁵/mL in liquid CD + 1% sucrose medium). After 24 h and 48 h, inoculated leaves were stained with 0.1% 3, 3′-diaminobenzidine (DAB) staining solution to detect H₂O₂ accumulation as described (46). Pictures were taken using a Nikon SMZ25 stereomicroscope. The brown intensity of DAB staining was quantified using ImageJ software (https://imagej.nih.gov/ij/).

Statistical analyses.

The statistical significance tests were performed by Student’s t test (*, P < 0.01; two-tailed test). In all graphs, results represent the mean value of at least three independent experiments, each with at least three replications per treatment.

Data availability.

Raw RNA-seq data can be accessed at the SRA database (BioProject identifier [ID] PRJNA762449).