MPMI PhytoFrontiers Phytobiomes all journals
Skip main navigation
Close Drawer MenuOpen Drawer Menu
Previous
Next
Analytical and Theoretical Plant PathologyFree Access icon

Simultaneous Transcriptome Analysis of Host and Pathogen Highlights the Interaction Between Brassica oleracea and Sclerotinia sclerotiorum

    Affiliations
    Authors and Affiliations
    • Yijuan Ding1 2
    • Jiaqin Mei1 2
    • Yaru Chai1 2
    • Yang Yu3
    • Chaoguo Shao1
    • Qinan Wu1
    • Joseph Onwusemu Disi4
    • Yuhua Li1
    • Huafang Wan1 2
    • Wei Qian1 2 †
    1. 1College of Agronomy and Biotechnology, Southwest University, Chongqing 400715, China;
    2. 2Academy of Agricultural Sciences, Southwest University, Chongqing 400715, China;
    3. 3College of Plant Protection, Southwest University, Chongqing 400716, China; and
    4. 4Department of Entomology, University of Georgia, Athens, GA 30602, U.S.A.
    Published Online:12 Mar 2019https://doi.org/10.1094/PHYTO-06-18-0204-R

    Abstract

    White mold disease caused by Sclerotinia sclerotiorum is a devastating disease of Brassica crops. Here, we simultaneously assessed the transcriptome changes from lesions produced by S. sclerotiorum on disease-resistant (R) and -susceptible (S) B. oleracea pools bulked from a resistance-segregating F2 population. Virulence genes of S. sclerotiorum, including polygalacturonans, chitin synthase, secretory proteins, and oxalic acid biosynthesis, were significantly repressed in lesions of R B. oleracea at 12 h postinoculation (hpi) but exhibited similar expression patterns in R and S B. oleracea at 24 hpi. Resistant B. oleracea induced expression of receptors potentially to perceive Sclerotinia signals during 0 to 12 hpi and deployed complex strategies to suppress the pathogen establishment, including the quick accumulation of reactive oxygen species via activating Ca2+ signaling and suppressing pathogen oxalic acid generation in S. sclerotiorum. In addition, cell wall degradation was inhibited in the resistant B. oleracea potentially to prevent the expansion of Sclerotinia hyphae. The transcriptome changes in S. sclerotiorum and host revealed that resistant B. oleracea produces strong responses against S. sclerotiorum during early infection.

    Sclerotinia sclerotiorum Lib. de Bary is a cosmopolitan fungal pathogen that infects >400 hosts, including rapeseed (Brassica napus), the third largest oilseed crop in the world (Bolton et al. 2006). The mycelium or ascospores of S. sclerotiorum infect the organs of rapeseed, such as flower petals, leaves, and stems, resulting in necrotic lesions, premature wilting, stem breakage, and plant lodging (Amselem et al. 2011; Garg et al. 2010). Yield losses in oilseed Brassica species vary between 5 and 100% each year (Saharan and Mehta 2008).

    S. sclerotiorum secretes cell wall-degrading enzymes (e.g., pectinases, cellulases, and hemicellulases) to facilitate plant cell wall degradation, hyphae colonization, and lesion expansion (Amselem et al. 2011; Bashi et al. 2012; Li et al. 2004a, b, c; Seifbarghi et al. 2017). These processes are accompanied by the production of the nonhost-selective toxin oxalic acid (OA) to manipulate the host redox environment, which is a benefit to colonization of the host plant, suppression of host autophagy, and inhibition of plant defense responses (Cessna et al. 2000; Kabbage et al. 2013; Kim et al. 2008; Williams et al. 2011). In 2014 and 2015, a large number of secretory proteins serving as potential effectors were predicted in the genus Sclerotinia with the use of bioinformatics tools (Guyon et al. 2014; Heard et al. 2015), and several genes encoding effector have been identified (Lyu et al. 2016; Yang et al. 2018; Yu et al. 2017; Zhu et al. 2013).

    In rapeseed, a number of quantitative trait loci (QTLs) and expressed sequence tags in association with resistance were identified in the partially resistant accessions, such as cultivars Zhongshuang 9 and Zhongyou 821 (Li et al. 2004a, b, c; Wu et al. 2016; Zhao et al. 2006). Several pathways and biological processes were revealed to associate with the defense against the genus Sclerotinia, such as oxidative burst, cell wall enforcement or modification, secondary metabolism and calcium binding, signaling, antioxidation, glutathione metabolism, glucosinolate metabolism, and biosynthesis of lignin (Joshi et al. 2016; Wei et al. 2016; Wu et al. 2016; Yang et al. 2007; Zhao et al. 2007, 2009). However, these studies only emphasized the response from the pathogen or the host during infection, producing an incomplete view of the interaction. For a comprehensive understanding of plant-pathogen interactions, it is valuable to analyze gene expression alterations in both pathogen and host (Kawahara et al. 2012; Westermann et al. 2012; Yazawa et al. 2013; Zhuang et al. 2012). The lack of an available host genotype with high resistance has limited the depth of understanding of the mechanistic interaction between S. sclerotiorum and its host. In 2011 and 2013, a genotype cultivar C01 of B. incana (a wild B. oleracea species) related to rapeseed, with nearly 20-fold higher stem resistance than partially resistant B. napus cultivar Zhongyou 821, was identified, and resistance QTLs were mapped in a resistance segregation population derived from a cross between C01 and a susceptible B. oleracea (Mei et al. 2011, 2013). In this study, the stems of extreme disease-resistant (R) and -susceptible (S) lines in this segregation population were inoculated with the genus Sclerotinia, and the transcriptome changes in both host and pathogen in lesions were analyzed. Our data suggested that the R B. oleracea quickly perceives S. sclerotiorum secretory proteins or/and pathogen-associated molecular pattern (PAMPs), limits the expression of virulence genes, and quickly accumulates reactive oxygen series (ROS) to inhibit the S. sclerotiorum establishment by activating Ca2+ signaling and suppressing pathogen OA generation. Meanwhile, the gene expressions of pectin methylesterase inhibitors (PMEIs) and pectinesterase inhibitors (PEIs) in R B. oleracea were promoted to suppress the cell wall degradation and prevent hyphae expansion. This study expands our understanding of the interaction between the genus Brassica and S. sclerotiorum.

    MATERIALS AND METHODS

    Plant materials and inoculation of the genus Sclerotinia.

    In a previous study (Mei et al. 2013), the stem resistance against S. sclerotiorum was investigated in an F2 population composed of 149 vegetative cloned lines derived from the cross between a resistant genotype C01 (B. incana) and a susceptible genotype cultivar C41 (B. oleracea var. alboglabra). Based on the resistance evaluation, four extreme resistant F2 cloned lines were chosen together with C01 as the R group, whereas four highly susceptible F2 cloned lines and the susceptible parent C41 formed the S group. Plants were grown in the growth chamber at 22°C in the light (fluorescent lamp, 100 lux) and 16°C in the dark. At the end of flowering, for each time point, three plants in each line were inoculated with 6-mm mycelia plugs obtained from actively growing colony edges of S. sclerotiorum ‘1980’ grown on the potato dextrose agar medium (20% potato, 2% dextrose, and 1.5% agar). Three sites on the main stem of each plant were inoculated and secured with Parafilm. The infection temperature was maintained at 22°C under 85% humidity. The epidermal stem tissues composed of the symptomatic lesion and the 10-mm region extending beyond the lesion margin were excised at 0, 12, and 24 h postinoculation (hpi) for total RNA extraction with the RNAprep pure Plant Kit (DP 432; Tiangen Biotech [Beijing] Co., Ltd.). RNA was bulked separately at each time point within the R and S groups (yielding six samples encoded with R0, R12, R24, S0, S12, and S24) and sequenced with three biological replicates.

    RNA sequencing and data analysis.

    The sequencing library was generated using the NEBNext Ultra RNA Library Prep Kit for Illumina (NEB, USA) following the manufacturer’s recommendation and sequenced on an Illumina Hiseq 2000 platform, which yields 100-bp paired end reads. The raw reads were filtered to obtain high-quality clean reads by removing adaptor sequences, duplicated sequences, reads containing >5% “N” (i.e., ambiguous bases in reads), and reads in which >50% of the bases showed a Q value (i.e., Bonferroni-adjusted P value) ≤5. Clean reads were aligned to the reference genome of B. oleracea (http://brassicadb.org/brad/downloadOverview.php) and S. sclerotiorum (http://fungidb.org/common/downloads/Current_Release/Ssclerotiorum1980UF-70/) by using the TopHat program (http://ccb.jhu.edu/software/tophat/index.shtml) with default parameters, except that the Q value was set to 100. Gene expression was quantified using htseq-count 0.6.1p2 (http://www-huber.embl.de/users/anders/HTSeq/doc/count.html). The raw counts were normalized by TMM (the weighted trimmed mean of M-values) normalization using the edgeR package (Robinson et al. 2010), and the differential expression analysis was carried out using the DEGseq package (Wang et al. 2010). The threshold determining the significance of differentially expressed genes (DEGs) among multiple tests was set at a false discovery rate (FDR) ≤ 0.001 and |log2 ratio| ≥ 1 (Mao et al. 2018). GO and KEGG enrichment analyses were performed with an FDR ≤ 0.05 as the threshold using BiNGO (Maere et al. 2005) and KOBAS 3.0 (http://kobas.cbi.pku.edu.cn/), respectively.

    Validation of the RNA-seq.

    Quantitative reverse transcription polymerase chain reaction (qRT-PCR) analysis was performed to test the gene expression in three replications with iTaq Universal SYBR Green Supermix (Bio-Rad) in the CFX96 Real-Time PCR Detection System. The polymerase chain reaction cycling conditions included 1 cycle of 95°C for 30 s and then 39 cycles of 95°C for 5 s and 55°C to 70°C for 1 min followed by a melting curve ramping from 65°C to 95°C, with temperature increasing by 0.5°C every 5 s (1 cycle). The genes BoActin3 and SsTubulin were used as the internal controls for the expression analyses of 20 B. oleracea and 15 S. sclerotiorum DEGs, respectively. The primers for qRT-PCR are listed in Supplementary Table S1. Pearson’s simple correlation coefficients were calculated among the results of qRT-PCR and RNA-seq.

    Identification of pathogen secretory proteins.

    To identify the differentially expressed pathogen secretory proteins, the sequences of DEGs in S. sclerotiorum were first screened by SignalP 4.1 (http://www.cbs.dtu.dk/services/SignalP/) in default settings for signal peptide cleavage sites (Nielsen 2017), and then, we screened the transmembrane domains with TMHMM (http://www.cbs.dtu.dk/services/TMHMM/) (Möller et al. 2001). Protein-encoding signal peptides but no transmembrane domains were considered candidate secretory proteins (Lum and Min 2011).

    Hydrogen peroxide staining.

    To detect hydrogen peroxide (H2O2) accumulation in situ, inoculated stems of extreme R and S B. incana × B. oleracea var. alboglabra F2 lines were stained with 3′3′-diaminobenzidine (DAB; Sigma-Aldrich; 1 mg/ml DAB-HCl, pH 7.4) under a gentle vacuum for 5 h and then washed three times in distilled water.

    Resistance assay of homologous Arabidopsis mutant.

    The homologous Arabidopsis T-DNA (Transfer DNA) mutants, AT3G47380.1: SALK_015169C (homologous to Bol023155 [PMEI]) and AT3G58790.1: SALK_137818C (homologous to Bol044481 [GAUT15]), were acquired from the Arabidopsis Biological Resource Center at The Ohio State University. The pathogenicity assays with three replicates were conducted as described by Wang et al. (2015) with minor modifications. Briefly, five leaves at 3 weeks old of every line were inoculated with a 2-mm mycelia plug obtained from actively growing colony edges of S. sclerotiorum ‘1980’. The lengths and widths of lesions were measured with a caliper at 24 hpi, and the lesion size (S = π × length × width/4) was calculated.

    RESULTS

    Phenotypic variances and general transcriptome changes in pathogen and host.

    Based on 2 years of evaluation of stem resistance (Mei et al. 2013), cloned lines with extreme resistance were chosen in the F2 population and reevaluated in the other 2 years. A total of 4 years of resistance data revealed a significant difference for the lesion size between the R and S B. oleracea groups (Fig. 1). The lesion lengths of the R and S B. oleracea groups averaged 2.45 cm (1.75 to 2.90 cm) and 5.50 cm (4.95 to 6.11 cm) at 72 hpi, respectively.

    Fig. 1.

    Fig. 1. Resistance against Sclerotinia sclerotiorum in disease-resistant (R) and -susceptible (S) Brassica oleracea groups. A, Lesion size of R and S of B. oleracea 72 h postinoculation (hpi) across 4 years. B, Symptom of infection by S. sclerotiorum among R and S plants in B. oleracea at 72 hpi. The control, cultivar Zhongshuang 9, is a partially resistant rapeseed variety against S. sclerotiorum. * represents significant difference was detected between the R and S pool (*P < 0.05, Student’s t test).

    Download as PowerPoint

    The cDNA (complementary DNA) libraries of six samples (R0, R12, R24, S0, S12, and S24) with three biological replicates were sequenced on the Illumina Hiseq 2000 platform and produced an average of 48.2 million clean reads for each sample (Table 1). Among these, 2.5 million mean clean reads per sample mapped to the reference genome of S. sclerotiorum, covering 73.8% (10,706 of 14,509) of the S. sclerotiorum genes except for R0 and S0, where <0.01% clean reads mapped to the S. sclerotiorum reference genome; 37 million mean clean reads per sample were mapped to the reference genome of B. oleracea, covering 77.1% (35,262 of 45,759) of the B. oleracea genes on average (Table 1). To validate the data obtained by RNA-seq, we performed qRT-PCR analyses by choosing 15 S. sclerotiorum genes and 20 B. oleracea genes of interest and confirmed that the expression obtained from qRT-PCR and RNA-seq was highly consistent (R = 0.899, P < 0.01) (Supplementary Fig. S1).

    TABLE 1. Summary of sequence read alignments to the reference genomes of Brassica oleracea and Sclerotinia sclerotiorum averages by three biological replicates

    We detected 326 DEGs of S. sclerotiorum from lesions between R and S B. oleracea at 12 hpi (Ss-DEGs-12) and 582 Ss-DEGs-24 (Fig. 2A). The GO biological processes “carbohydrate metabolic process,” “oxalate metabolic process,” and “pectin catabolic process” and the KEGG pathways “ribosome” and “pentose and glucuronate interconversions” were specifically enriched in Ss-DEGs-12, whereas Ss-DEGs-24 were significantly enriched in “response to oxygen-containing compound,” “translation,” and “ribosome” (Fig. 2B).

    Fig. 2.

    Fig. 2. Analysis of differentially expressed genes (DEGs) of Sclerotinia sclerotiorum. A, Venn diagram of DEGs of S. sclerotiorum in lesions between disease-resistant (R) and -susceptible (S) groups and B, the GO biological processes (black) and KEGG terms (gray), which were significantly enriched within these DEGs.

    Download as PowerPoint

    A total of 1,768 and 2,882 DEGs of B. oleracea (Bol-DEGs) in the R B. oleracea and 4,130 and 2,335 Bol-DEGs in the S B. oleracea were found during 0 to 12 and 12 to 24 hpi, respectively. Common responses between the R and S B. oleracea involved several GO biological processes, such as “response to hormone,” “salicylic acid/jasmonic acid mediated signaling pathway,” “defense response,” “immune response,” “response to fungus,” “response to chitin,” “Mitogen-activated protein kinase (MAPK) cascade,” and “systemic acquired resistance,” and several KEGG pathways, such as “photosynthesis,” “biosynthesis of secondary metabolites,” “circadian rhythm,” “plant-pathogen interaction,” “photosynthesis-antenna proteins,” “flavonoid biosynthesis,” and “plant hormone signal transduction” (Supplementary Fig. S2). Nevertheless, specific DEGs analysis revealed different responses between the R and S B. oleracea involved in the biological process “cell wall organization or biogenesis” and the pathways “plant-pathogen interaction,” “pentose and glucuronate interconversions,” “starch and sucrose metabolism,” and “regulation of autophagy” (Fig. 3).

    Fig. 3.

    Fig. 3. Analysis of specific differentially expressed genes (DEGs) between disease-resistant (R) and -susceptible (S) Brassica oleracea. GO biological processes in black boxes and KEGG terms in red boxes were significantly enriched with specific DEGs between R and S B. oleracea groups during the A, early (0 to 12 h postinoculation [hpi]) and B, late (12 to 24 hpi) infection against Sclerotinia sclerotiorum. FC = fold change.

    Download as PowerPoint

    Interaction between pathogen and host.

    The KEGG pathway “Plant-pathogen interaction” (ko04626) was significantly enriched in both R-specific up-regulated DEGs and S-specific down-regulated DEGs of B. oleracea during 0 to 12 hpi (Fig. 3). Specific categories included pathogen perception, Ca2+ signaling, MAPK-WRKY signaling, ROS activation, and disease resistance proteins. Host perception of pathogen-secreted proteins and PAMPs may determine infection success after infection (Dangl et al. 2013; Wirthmueller et al. 2013). A total of 93 Ss-DEGs encoding the protein with signal peptides, but no transmembrane domains were regarded as the candidate genes of secretory proteins (Supplementary Table S2). Of these, 83 genes exhibited higher expression in lesion of S B. oleracea than R B. oleracea, whereas 10 genes had the same expression level in R and S B. oleracea at 12 hpi. Additionally, all 93 genes exhibited the same expression levels in R and S B. oleracea at 24 hpi. It seems to indicate that these 83 candidate secretory proteins may be involved in the virulence of S. sclerotiorum, especially in the early infection.

    Chitin is an important component of fungal cell wall, and it is regarded as a PAMP (Yamaguchi et al. 2017). Chitin synthase is associated with virulence of S. sclerotiorum (Andrade et al. 2016). We detected 11 chitin synthase genes with higher expression level in lesion of S B. oleracea than R B. oleracea at 12 hpi and without expression difference between R and S B. oleracea at 24 hpi. It suggests that the expression of chitin synthase genes is limited in R B. oleracea during 0 to 12 hpi.

    Proteins with the domain of leucine-rich repeat (LRR), LRR kinase, lectin kinase, serine/threonine kinase, or LysM domain-containing GPI-anchored protein 2 precursor (LysM2) are regarded as candidate receptors to perceive pathogen effectors and PAMPs during early infection (Dangl et al. 2013). A total of 51 Bol-DEGs in B. oleracea were identified as candidate receptor genes, including 22 DEGs specifically induced in R B. oleracea during 0 to 12 hpi and 29 DEGs with the same expression pattern in both R and S B. oleracea (Supplementary Table S3). Specifically, two receptor genes (Bol036654 and Bol021625) showed higher up-regulated expression of around threefold in R B. oleracea as revealed by qRT-PCR but no changes in S B. oleracea during 0 to 12 hpi. We further investigated the expression of these two receptors in the parental B. oleracea lines. Both Bol036654 and Bol021625 were up-regulated in the R parental line, but there were no changes in the S parental line at 9 hpi (Supplementary Fig. S3). These findings potentially indicate an effective perception of S. sclerotiorum by the R B. oleracea. It was in accordance with the observation of a delay of infection establishment in R B. oleracea. Among 11 compound appressorium-associated Sclerotinia DEGs detected, 8 exhibited lower expression at 12 hpi in lesion of R B. oleracea than S B. oleracea (Fig. 4A; Supplementary Table S4). Furthermore, the observation of scanning electron microscopy revealed that the infection cushions were successfully established in S B. oleracea at 9 hpi but not in R B. oleracea, and there were more and larger infection cushions in S B. oleracea than in R B. oleracea at 12 hpi (Fig. 4B).

    Fig. 4.

    Fig. 4. Infection establishment of Sclerotinia sclerotiorum on stems of disease-resistant (R) and -susceptible (S) Brassica oleracea. A, Expression of compound appressorium-associated differentially expressed genes (DEGs) in S. sclerotiorum during the infection and B, the infection cushions observation in R and S B. oleracea at 6, 9, and 12 h postinoculation (hpi) revealed by electron microscopy.

    Download as PowerPoint

    Transient elevation of cytosolic Ca2+ occurs after pathogen perception to activate downstream signaling cascades (Takahashi et al. 2011; Tena et al. 2011). A total of 45 B. oleracea genes involved in the Ca2+ signaling showed differential expression patterns between R and S B. oleracea during 0 to 12 hpi. Among them, 28 DEGs were up-regulated in the R B. oleracea but stably expressed (24 genes) or down-regulated (4 genes) in the S B. oleracea, whereas the other 17 DEGs were stably expressed in the R B. oleracea but down-regulated (15 genes) or up-regulated (2 genes) in the S B. oleracea, indicating a stronger Ca2+ signaling in the R B. oleracea in early infection. This was in accordance with the qRT-PCR analysis in R and S parental lines. For example, calcium-dependent protein kinase 6 (Bol042113) showed up-regulated expression at 6 hpi (4.5-fold) and 9 hpi (6.6-fold) in the R parental line but was not significantly induced in the S parental line.

    Plant Ca2+ signaling regulates the generation of ROS by modulating the activity of respiratory burst oxidase homologs (RBOHs) (Arthikala et al. 2014; Keller et al. 1998; Li et al. 2015). Four differentially expressed RBOHF genes (Bol022639, Bol000420, Bol033122, and Bol043072) detected during 0 to 12 hpi exhibited a higher levels of up-regulated expressions in the R B. oleracea than in the S B. oleracea. We chose an RBOH (Bol022639) for qRT-PCR analysis in R and S parental lines and found that it showed up-regulated expression at 9 hpi of 2.88-fold in the R but that it was not significantly induced in the S parental line from 6 to 9 hpi. These findings seem to indicate increased production of ROS in R B. oleracea than in S B. oleracea during early infection. This speculation was confirmed by DAB staining, which showed an earlier accumulation of H2O2 in the R than in the L lines (Fig. 5).

    Fig. 5.

    Fig. 5. Hydrogen peroxide accumulation in extreme disease-resistant (R) and -susceptible (S) Brassica incana × B. oleracea var. alboglabra F2 lines after inoculation of Sclerotinia sclerotiorum. hpi = hours postinoculation.

    Download as PowerPoint

    Cell wall changes in the host.

    The host cell wall is a physical defense against pathogens. A total of 66 Ss-DEGs were enriched into GO and KEGG terms related to cell wall degradation, such as “carbohydrate metabolic,” “pectin catabolic,” and “pentose and glucuronate interconversions” (Fig. 2), including 30 carbohydrate-active enzymes and auxiliary proteins (CAZymes) involved in the degradation of plant cellulose (5 genes), hemicellulose (10 genes), and pectin (15 genes) (Supplementary Table S5). Of these, 27 CAZymes showed lower expression in the lesion of R B. oleracea than in S B. oleracea at 12 hpi, whereas 3 CAZymes exhibited a similar expression pattern in R and S B. oleracea. It seems to indicate that the expressions of pathogen cell wall degradation enzymes were suppressed when infecting the R B. oleracea.

    Correspondingly, a total of 314 Bol-DEGs involved in the cell well metabolism were enriched in the GO term “cell wall organization or biogenesis” (66 genes) and the KEGG terms “pentose and glucuronate interconversions” (125 genes) and “starch and sucrose metabolism” (232 genes) (Fig. 3), with 49 overlapping DEGs among the three terms (Supplementary Table S6). Those 49 DEGs were specifically induced in R B. oleracea but not in S B. oleracea during 0 to 12 hpi, including eight PMEIs and seven PEIs, which inhibit the catalyzation of plant cell wall polygalacturonans (Lionetti et al. 2012). The Bol023155 (PMEI) showed the highest up-regulated expression at 210-fold in R B. oleracea during 0 to 12 hpi. In addition, the expressions of three galacturonosyltransferases (GAUT14, GAUT15, and GATL8), which are key enzymes in pectin biosynthesis, and one cellulose synthase-like protein (CSLD2) were detected specifically up-regulated in R B. oleracea. The Arabidopsis T-DNA mutants homologous to Bol023155 (PMEI) and Bol044481 (GAUT15) were evaluated for the genus Sclerotinia resistance. Both mutant lines showed enhanced susceptibility to S. sclerotiorum in comparison with the wild-type Col-0 (Columbia-0) (Fig. 6). These findings indicate that R B. oleracea genotypes inhibit lesion expansion by possibly inhibiting cell wall decomposition and promoting the cell wall reorganization or rearrangement.

    Fig. 6.

    Fig. 6. Resistance assay of Arabidopsis mutants homologous to BolPMEI and BolGAUT15. Agar plugs containing actively growing cultures of the wild-type (WT) strain cultivar 1980 of Sclerotinia sclerotiorum were inoculated onto leaves of Arabidopsis WT (Columbia-0) and select Arabidopsis PMEI (Atpmei: homologous to Bol023155) and GAUT15 (Atgaut15: homologous to Bol044481) mutant plants. A, The lesion phenotype at 24 h postinoculation (hpi), and B, lesion size was monitored from three biological replicates. *Significant difference at P < 0.05 (two-tailed t test).

    Download as PowerPoint

    OA.

    S. sclerotiorum pathogenesis accumulates high levels of OA during the successful establishment of colonization in the host and modulates the host redox environment to block the host oxidative burst (Kabbage et al. 2013; Williams et al. 2011). The activation of oxaloacetate acetylhydrolase (SsOAH; SS1G_08218) is required in the final step of OA biosynthesis, which is regulated by the zinc finger transcription factor SsPAC1 (SS1G_07355) (Liang et al. 2015a, 2015b; Rollins 2003). We found that both SsOAH (Fig. 7; Supplementary Table S7) and SsPAC1 (Fig. 7) exhibited lower expressions at 12 hpi in lesions of R B. oleracea than in S B. oleracea but similar expressions at 24 hpi in the two groups. The oxalate decarboxylase (SsODC2; SS1G_10796) (Liang et al. 2015a, 2015b) exhibited similar expression in lesions of R and S B. oleracea at 12 hpi but higher expression at 24 hpi in R B. oleracea than in S B. oleracea.

    Fig. 7.

    Fig. 7. Oxalic acid (OA) secreted by Sclerotinia sclerotiorum induced different responses on plant autophagy in disease-resistant (R) and -susceptible (S) Brassica oleracea. TCA = tricarboxylic acid cycle, AKIN = Arabidopsis SNF1 kinase homolog, and ATG = Autophagy.

    Download as PowerPoint

    Pathogen OA is reported to suppress host autophagy (Kabbage et al. 2013). We found that the pathway “regulation of autophagy” was significantly enriched in S B. oleracea but not in R B. oleracea during 0 to 12 hpi with 19 specific down-regulated DEGs, including 8 ATG8 and 1 ATG3 (Fig. 7; Supplementary Table S8), which are the key genes to modulate plant autophagy (Bassham et al. 2006; Kabbage et al. 2013). It suggests that the autophagy pathway is significantly suppressed in S B. oleracea in the early infection.

    DISCUSSION

    In this study, the interaction between the genus Brassica and the genus Sclerotinia was revealed by simultaneously investigating the transcriptome changes in S. sclerotiorum and R and S B. oleracea. In comparison with the S B. oleracea, the R B. oleracea seems to have a stronger ability to defend Sclerotinia inoculation by hindering the expression of Sclerotinia virulence genes, such as genes encoding secretory proteins PAMPs and CAZymes, blocking appressorium establishment, and causing an earlier oxidative burst at the inoculate site. Meanwhile, the degradation of the plant cell wall was suppressed and autophagy was maintained in R B. oleracea.

    The initiation of plant defense against a pathogen starts with the perception of pathogen effectors or PAMPs (Wirthmueller et al. 2013). Several secretory proteins have been shown to be involved in the pathogenesis of S. sclerotiorum, such as an integrin-α N-terminal domain superfamily SsITL, an Rhs repeat-containing protein SsRhs1, a putative Ca2+ binding EF (elongation factors)-hand motif Sscaf1, a γ-glutamyl transpeptidase SsGgt1, and a cysteine-rich protein SsSSVP1 (Guyon et al. 2014; Li et al. 2012; Lyu et al. 2016; Xiao et al. 2014; Yu et al. 2017; Zhu et al. 2013). Here, we identified 93 genes encoding secretory proteins in S. sclerotiorum and found that most of them exhibited lower expression levels in lesions of R B. oleracea than in lesions of S B. oleracea. Except for 14 genes known to be effectors (Guyon et al. 2014; Li et al. 2012; Seifbarghi et al. 2017; Yu et al. 2017), the other 79 genes were newly detected in this study. Correspondingly, 22 receptors were specifically induced in R B. oleracea in the early infection. Except for three kinds of receptors, LRR family proteins, wall-associated kinase-like proteins, and L-type lectin receptors, which were reported up-regulated expressed after inoculation of S. sclerotiorum in B. napus (Wei et al. 2016; Wu et al. 2016; Zhao et al. 2009), we detected several other receptors that were induced in B. oleracea, including LysM domain-containing proteins perceived in fungal chitin fragments (Petutschnig et al. 2010; Shinya et al. 2012; Wan et al. 2012), cysteine-rich RLKs, serine/threonine kinase, BRI1-associated receptor kinase, and somatic embryogenesis receptor kinase. However, interactions between receptors and these secretory proteins need to be investigated in more detail in the future. Our findings indicate that the R B. oleracea may effectively perceive and limit the expression of the S. sclerotiorum secretory proteins and/or PAMPs by triggering various receptors.

    WRKY transcription factors are downstream of pathogen perception and regulate resistance genes in the host during infection (Li et al. 2004a, 2004b, 2004c; Tena et al. 2011). In this study, we detected 11 WRKYs and 5 resistance gene likes with specifically up-regulation expression in R B. oleracea during the early infection. Included in these, three toll-interleukin receptor (TIR) domain-encoding genes (Bol024282, Bol018676, and Bol029861) were detected. In the genus Arabidopsis, RLM3 containing a TIR domain was reported to be involved in broad-range immunity to several necrotrophic pathogens (Staal et al. 2008). Wu et al. (2016) also detected up-regulated expression of TIR domain-encoding genes after inoculation of the genus Sclerotinia in resistant B. napus. Among the 11 WRKYs, 2 WRKY50 (Bol012741 and Bol022304), 1 WRKY51 (Bol015902), and 2 WRKY70 (Bol044396 and Bol044275), which are positive modulators of the downstream salicylic acid-mediated signaling (Gao et al. 2010; Li et al. 2004a), were specifically up-regulated in R B. oleracea. However, these genes show an opposite expression pattern in the inoculated leaves of B. oleracea and B. napus (Mei et al. 2016; Wu et al. 2016). It seems to indicate different responses against S. sclerotiorum between stems and leaves.

    The fungus expresses genes encoding cell wall-degrading enzymes to break down the cell wall matrix of its host, such as genes that encode the CAZymes to decompose plant pectin, cellulose, and hemicellulose (Amselem et al. 2011; Kars et al. 2005). Conversely, the host expresses genes to biochemically modify the cell wall components and inhibit cell wall degradation (Luna et al. 2011; Vorwerk et al. 2004). PMEIs, PEIs, and GAUTs are key enzymes of pectin metabolism in plants (Caffall et al. 2009; Lionetti et al. 2017). In this study, the processes “carbohydrate metabolic,” “pectin catabolic,” and pathway “pentose and glucuronate interconversions” were significantly enriched by the Ss-DEGs-12, which is coincident with the study in B. napus (Seifbarghi et al. 2017), suggesting that these processes possible play roles in the pathogenicity in the early infection. We detected 30 Sclerotinia CAZymes involving in these processes and found that most of them showed lower expression levels in lesions of R B. oleracea, whereas the corresponding genes, such as PMEIs, PEIs, and GAUTs, were specifically up-regulated in R B. oleracea in the early infection. PMEIs and PEIs are regarded as mediators of cell wall integrity maintenance in the plant against fungal, bacterial, and viral pathogens (An et al. 2008; Lionetti et al. 2007, 2014, 2017; Volpi et al. 2011). In the B. napus-S. sclerotiorum pathosystems, PGIPs, but not PMEIs, were the major enzyme inhibitors against Sclerotinia CAZymes (Wu et al. 2016; Zhao et al. 2009). These specifically induced PMEIs and PEIs in this study suggest that R B. oleracea has a possibly differential Sclerotinia resistance mechanism from B. napus to inhibit cell wall degradation. Meanwhile, a series of glycosyl hydrolase, glycoside hydrolase, and glycosyl transferring genes, which are involved in disassembly of the fungal cell wall of S. sclerotiorum (Zhuang et al. 2012), were specifically up-regulated in R B. oleracea. These findings seem to indicate that R B. oleracea may disassemble the fungal cell wall during infection.

    S. sclerotiorum has long been considered to be a prototypical necrotrophic pathogen (Hegedus and Rimmer 2005). However, recent evidence showed that this fungus may exhibit a brief biotrophic phase in the early infection (Kabbage et al. 2013, 2015; Seifbarghi et al. 2017). Initially, the genus Sclerotinia generates a reducing environment in host cells that suppresses host oxidative burst and calluses deposition, akin to compatible biotrophic pathogens (Kabbage et al. 2013; Williams et al. 2011). After the pathogen is successfully established, S. sclerotiorum switches to the necrotic stage to induce the host ROS and programmed cell death to promote hyphal expansion (Hegedus and Rimmer 2005; Kabbage et al. 2013, 2015; Mei et al. 2016). Ca2+ signaling induces the production of ROS via activating the expression of RBOHs in host (Davies et al. 2006; Harding et al. 1997; Keller et al. 1998; Takahashi et al. 2011). We found that Ca2+ signaling was first induced in R B. oleracea during 0 to 12 hpi but suppressed during 12 to 24 hpi. The earlier accumulation of H2O2 was subsequently observed in R B. oleracea at 6 hpi (Fig. 5). These results indicate that the R B. oleracea produces an oxidative environment to inhibit the establishment of the pathogen in the biotrophic stage and then inhibits the ROS accumulation via suppressing the Ca2+ signaling during necrotic stage. It is consistent with the low expression of Sclerotinia compound appressorium-associated genes and weak establishment of infection cushions in R B. oleracea at 12 hpi (Fig. 4). Taken together, our findings indicate that the resistance B. oleracea stems trigger complex strategies to effectively perceive and defend against S. sclerotiorum.

    LITERATURE CITED

    The author(s) declare no conflict of interest.

    Y. Ding and J. Mei contributed equally to this work.

    Funding: This study was financially supported by Key Projects in National Science and Technology grant 2014BAD01B07, National Nature Science Foundation of China grants 31671726 and 31801395, 973 Program grant 2015CB150201, Science and Technology Innovation Program for the Social Undertakings and the People’s Livelihood in Chongqing grants cstc2016shmsx0674 and cstc2017shms-xdny80050, and Fundamental Research Funds for the Central Universities grants XDJK2018AA004 and XDJK2018B022.