Gene Expression Analysis of Brassica napus in Response to Leptosphaeria biglobosa Infection Progress

HAO Lifen,YAN Mengjiao,FANG Yongyu,SHI Zhidan,LI Ziqin,HAN Bing

(1 College of Life Sciences,Inner Mongolia Agricultural University,Hohhot 010011,China; 2 Inner Mongolia Academy of Agricultural & Animal Husbandry Sciences,Hohhot 010031,China)

Abstract: Blackleg caused by Leptosphaeria biglobosa,is one of the damaging diseases of oilseed rape (Brassica napus). Based on histological observation and RNA-Seq,we analyzed the phenotypes and gene expression changes of oilseed rape after inoculation with L. biglobosa for 4,12,24,36,48 and 96 h to discuss the defense response and disease resistance mechanism of oilseed rape in response to the infection of L. biglobosa,provided a theoretical basis for comprehensively revealing the molecular mechanism of interaction between oilseed rape and L. biglobosa,and accumulated genetic information for breeding the disease resistance varieties of oilseed rape. The results showed that: (1) after inoculation for 4-96 h,the leaf spot gradually expanded,and the L. biglobosa formed mycelium within 48-96 h. (2) By RNA-Seq,a number of 3 384,2 270,3 802,5 811,6 155,7 153 differentially expressed genes (DEGs) were obtained at 4,12,24,36,48 and 96 h of L. biglobosa infection on oilseed rape. (3) qRT-PCR results showed that the levels of the 15 differentially expressed genes in oilseed rape were basically consistent with the results of RNA-Seq. (4) The differentially expressed genes were analyzed by short time series clustering and KEGG enrichment analysis. The results showed that the genes in the pathways of plant and L. biglobosa interaction,protein kinase,jasmonate/ethylene/salicylic acid signaling,and glucosinolate biosynthesis were strongly activated,and their gene expression showed dynamic change trend.

Key words: differentially expressed genes (DEGs); Leptosphaeria biglobosa; plant-pathogen interaction; MAPK; jasmonate/ethylene/salicylic acid; glucosinolates biosynthesis

Phoma stem canker(blackleg) is the most common disease of oilseed rape (Brassicanapus)[1]and is caused byLeptosphaeriamaculansand theL.biglobosacomplex,evolved from a common ancestor[2]. These two species have many similar biological characteristics,andBrassicais their main host[3]. Compared toL.maculans,L.biglobosais less virulent and generally does not cause significant yield loss[2]. However,it can result in a certain extent of damage in China and somewhere else,with a trend of gradually spreading[4-6].L.maculansis a hemibiotrophic fungal pathogen,whereasL.biglobosais necrotrophic[7]. In previous study,pretreatment of oilseed rape leaves withL.biglobosainfection can induce the oilseed rape produce local and systemic resistance toL.maculans[8-9]. However,the defense molecular mechanism of oilseed rape afterL.biglobosainduced was not clear,it is necessary to reveal the globally transcriptional responses directly after inoculated withL.biglobosa,and it is helpful to prevent and control the blackleg to oilseed rape.

Plant immune responses to necrotrophs and biotrophs may be different,which depends on the pathogen species and the primary determinant of virulence. Pathogen-associated molecular patterns (PAMPs)-triggered immunity (PTI) is the first barrier of plant defense against pathogens[10-11]. PAMPs contain macromolecules such as chitin,endopolygalacturonases and the proteins that trigger the plant immune response[12]. The second plant defense barrier is effector-triggered immunity (ETI),which recognizes pathogens through the plant R protein,detects effectors and initiates a defense response[13]. The host R gene mainly responds to the infection caused by biotrophic and hemi-biotrophic fungi and contributes to the breakthrough in plant disease resistance breeding. For example,theL.maculans-B.napusinteraction is a typical “gene-for-gene” model,with many varieties being bred with multiple sources of resistance genes[14]. However,the plant immune response to the necrotrophic fungi is very complicated. Therefore,to prevent necrotrophic diseases,it is important to understand the pathogenic mechanism of fungal pathogen and the defense mechanism of plants. Once the pathogen recognized PAMPs,MAPK and hormone signaling play major roles in signal transduction and defense reaction[15]. MAPK activates ROS production,signal transduction,transcription factors,production of plant antitoxin synthesis,and expression of defense response genes,as well as regulates cell death and hypersensitivity reaction[16]. It is well known that necrotrophic fungi mainly activate jasmonic acid and ethylene signaling pathways inArabidopsisthaliana[17],and biotrophic fungi mainly activate salicylic acid (SA) signaling and induce systemic acquired resistance (SAR)[18].

The molecular mechanism of oilseed rape responding toL.maculanshas been confirmed with elucidating receptor recognition and signal transduction and identification of resistance genes related to oilseed rape defense through transcriptome sequencing[19]. For example,the key transcription factors and glucosinolate biosyntheais were highly upregulated early in the response in resistant oilseed rape. It is consistent with the results in oilseed rape leaves infected with the necrotrophSclerotiniasclerotiorum[20]. Genomes and transcriptomes have been employed to look for important events in plant-fungal interactions between twoLeptosphaeriaspecies andBrassicanapus. It showed that during the first seven days of infectionL.biglobosaexpressed more cell wall degrading genes. However,L.maculansexpressed more genes in the carbohydrate binding module class of CAZy[7]. Thus,thorough analyses of RNA-seq profiles at the plant host response to theL.biglobosais very helpful to understand the defense molecular mechanism.

In this study,we applied RNA-sequencing technology and bioinformatics analysis to determine differential expression of the genes in oilseed rape infected withL.biglobosaat different time points to reveal the molecular mechanism of oilseed rape responding to this pathogen. Dynamic changes in the expression of responsive genes to pathogen inoculation were revealed by time-series expression cluster. The key pathways and genes were analyzed in main clustering profiles. This information provides insight into the molecular mechanism of necrotrophic fungus-plant interaction and changes in gene expression of oilseed rape in response toL.biglobosainfection.

1 Materials and methods

1.1 Materials

L.biglobosastrain nm-1 was originally isolated and identified from a diseased plant in oilseed rape field in Inner Mongolia,China[21]. The culture ofL.biglobosawas grown on potato dextrose agar (PDA) medium for 10 days. The conidia were collected and prepared for spore suspension with a concentration of 1×107spores/mL. Li[22]found the resistance agaistL.biglobosain 74 ChineseBrassicanapus cultivars even though the resistance is not very strong. A better resistant variety ‘Qingza 5’ was selected as the experimental material. The seeds were sown in sterilized garden soil,and the plants were grown in an illuminated incubator at 25/18 ℃ with a 16/8 h photoperiod for 30 days. At the five-leaf stage of plants,the second and third leaves were artificially inoculated with 10 μL of suspension (107spores /mL) ofL.biglobosaat the wound points punched by a bamboo needle.

1.2 Methods

1.2.1 Phenotypic characteristics ofB.napusafterL.biglobosainoculation and progress of infectionTo understand the interaction of oilseed rape andL.biglobosaat early stage of infection,we determined six time points at 4,12,24,36,48 and 96 h post inoculation to analyze phenotypic change ofB.napusand the pathogen growth status in the leaf using green fluorescent protein (GFP) tagging procedure. The recombinant plasmid PCH-sGFP was transformed intoL.biglobosathrough Agrobacterium-mediated genetic transformation,and T-DNA inserted intoL.biglobosamutant library was constructed,and the positive transformants were selected with the similar phenotypic and pathogenicity characteristics as wild typeL.biglobosa[23]. The positive transformants were inoculated into oilseed rape leaves according to the above method. The leaves tissues with inoculated sites were harvested at 4,12,24,36,48 and 96 h after inoculation,and cut into thin slices by hand and placed under a Laser Confocal Microscopy (Nikon A1RMP,Japan) to examine the process of infection of theL.biglobosain the leaves,each treatment had 3 replicates.

1.2.2 DEG library construction and sequencingThe leaf samples from 10 plants were collected at 4,12,24,36,48 and 96 h post-inoculation. Negative control leaves inoculated with sterile water were also collected at same time points. Each treatment had 3 replicates. All samples were immediately frozen in liquid nitrogen and stored at -80 ℃ prior to RNA extraction. Total RNA was extracted using a TaKaRa MiniBEST Plant RNA Extraction Kit (TaKaRa,Japan) following the manufacturer’s protocol. The concentration of RNA was determined using a Nanodrop 2000 spectrophotometer (Thermo Fisher Scientific,Wilmington,DE,USA). RNA quality was assessed using an Agilent Bioanalyzer Model 2100 (Agilent Technologies,Palo Alto,CA). DNase I was used to degrade any possible DNA contamination,and then with the oligo(dT) magnetic beads enriched mRNA. After mRNA was fragmented into short fragments,the first strand of cDNA was synthesized by using random N6 primer; then the second strand was synthesized. The synthetic double-stranded DNA reparation end,and sequencing adaptors were ligated to the fragments. Single-strand DNA reoccupied by thermotropy,then cyclized to single-stranded circular DNA by the bridge type primers to construct cDNA libraries. Quality control was performed on the sample libraries,and an Agilent 2100 Bioanalyzer and quantitative PCR (qRT-PCR) were employed to ensure the quality of the library. The 12 constructed cDNA libraries were sequenced at Beijing Genomics Institute (Shenzhen,China) using the BGISEQ-500 platform.

1.2.3 Analysis ofBrassicanapusDEGs in response toL.biglobosaSequencing data with a read length of 50 bp were defined as raw reads. Clean reads were obtained by removing adapter and primer sequences. All clean reads from each sample were mapped to theBrassicanapusreference genome (Darmor-bzh) (https://www.genoscope.cns) using HISAT[24]. After alignment of each sample,the relative abundance of transcripts was evaluated to profile gene expression levels using FPKM (fragments Per Kilobase of exon model per Million mapped reads) from RSEM software[25].

1.2.4 Cluster analysis of different time points and KEGG enrichment analysisAfter obtaining DEGs at the six time points,the matrix data of these genes were further screened so as to cluster analysis using STEM (Short Time-series Expression Miner) (version1.3.11)[26](http://www.cs.cmu.edu/stem/). The number of genes forming a cluster was set not less than 20,with a correlation indexR≥0.7 andP<0.05 of expression. Annotations of DEGs in each cluster were subjected to enrichment analysis using Kyoto Encyclopedia of Genes and Genomes (KEGG) databases.

1.2.5 Confirmation of gene expression level by quantitative real-time PCRRNA was extracted from inoculated leaves withL.biglobosafrom check at 4,12,24,36,48,and 96 h using the TaKaRa MiniBEST Plant RNA Extraction Kit (TaKaRa,Japan) and reverse transcription was conducted using TransScript One-Step gDNA Remover and cDNA Synthesis Kit (TransGen,China). The SYBR Green I kit (TaKaRa,Japan) was utilized to analyze relative expression abundance with the Roche Light Cycler 480 (Roche,USA). Gene-specific primers for 15 candidate genes (Table 1) were designed using the Primer Premier 5.0 software program (Premier Biosoft,Palo Alto,CA,USA).Actin1 (DQ665832) was used as the endogenous control. Each qRT-PCR reaction contained 1.0 ng of cDNA template from the reverse-transcription reaction,10.0 nmol/L gene-specific primers,and 10.0 μL of SYBR Green supermix with a final volume of 20.0 μL. Amplification was performed as follows: 95 ℃ for 30 s,followed by 40 cycles of 95 ℃ for 5 s,and 60 ℃ for 34 s. Three biological replicates were included for every sample. Expression data were transformed and analyzed using the 2-ΔΔCtmethod[27]. Expression levels were normalized usingActin1,and ratios are expressed as fold changes compared to the levels detected in the samples. All qRT-PCR data were collected from three biological replicates and three technical replicates for each sample.

Table 1 The primer sequences were used for qRT-PCR verification

2 Results and analysis

2.1 Phenotypic characteristics of B. napus hosts inoculated with L. biglobosa

Previous studies indicated that the infection of leaves by the conidia ofL.biglobosahas three stages: germination of the spores (swelled spores),growth of the spores (development of spore tubes),hyphae elongation during 4 days after inoculation[28]. It is important to understand the phenotypic changes ofB.napusafter inoculation with the pathogen in those early stages. The phenotypic characteristics ofB.napushosts inoculated withL.biglobosashowed no disease symptoms on leaves during the first 36 h post inoculation (Fig.1，A-D). Small dot lesions at the inoculated site of the leaves at 48 h were observed (Fig.1，E),and the obvious brown spot lesions on inoculated site occurred at 96 h (Fig.1，F). The observation of theL.biglobosain the leaves was conducted under laser confocal microscopy during this infection period. After 4 h,the conidia attached to the leaf surface but did not germinate (Fig.1，G); after 12 h the conidia swelled and some conidia reached to stoma (Fig.1，H); after 24 h,all conidia germinated and produced tubes at one side of spore (Fig.1，I); after 36 h the elongation of the germ tubes was continuing (Fig.1，J); After 48 h,the mycelia were formed and started spreading through the gap between the cells in the epidermis of the leaves (Fig.1，K),and after 96 h,the mycelia developed branches and began to spread further to the surrounding areas (Fig.1，L). The whole infection process can be divided into three segments including conidia swelled stage (4-12 h),germ tube produced stage (24-36 h) and mycelia formed stage (48-96 h).

The green color was L. biglobosa in leaves under laser confocal microscopy

2.2 Analysis and verification of DEGs of transcriptional reprogramming in response to L. biglobosa infection

To understand the transcriptional reprogramming in leaves during the early stage of infection at 4,12,24,36,48,and 96 h,we sampled the leaves inoculated withL.biglobosaand mock-inoculated with sterile water at each time point for RNA sequencing. 22 690 260 clean reads were obtained per sample on average,and the regular rate of clean reads mapping to reference genes was 74.38%,with a more than 90% clean reads mapping to the reference genome (http://www.genoscope.cns.fr/brassicanapus/) (Table 2). The gene expression libraries of theL.biglobosainoculated and mock inoculated at the 6 time points were compared,DEGs were defined with log2fold-change ≥1 andFDR<0.01. In total,3 384,2 270,3 802,5 811,6 155,and 7 153 DEGs were obtained at 4,12,24,36,48,and 96 h,which consisted of 1 561,1 393,2 368,1 699,3 373,4 523 upregulated genes and 1 823,877,1 434,4 112,2 782,2 630 downregulated,respectively (Fig. 2，A). It was noticed that the number of upregulated genes increased significantly at 48 and 96 h. It indicated that the deeper transcriptional reprogramming occurred after hyphae extension ofL.biglobosa. All upregulated and downregulated genes were converted to their log2fold change. The log2fold change for upregulated genes was 2-3,3-4,4-5,and > 6 greater than that of downregulated genes in the corresponding sectors. This result indicates that upregulated genes were significantly triggered,and they were related to theL.biglobosainfection (Fig. 2，B).

Table 2 Summary of reads mapped to the reference genome

A. The number of DEGs up-or down regulated at six different time points in inoculated and mock-inoculated oilseed rape; B. The rate of Log2 fold change in DEGs detected in oilseed rape

To examine the quality of 12 RNA-Seq libraries at the 6 time points,we calculated the fold change of 15 differentially expressed genes at each time point by qRT-PCR. The RNA-seq results are basically consistent with those of qRT-PCR (Fig. 3),and the fold change showed a high correlation (R2=0.812 7),indicating that the sequencing results are reliable.

The qRT-PCR fold change in the expression ratio (inoculated/mock-inoculated) (y-axis) was plotted against the values from RNA-Seq (x-axis)

2.3 Dynamic changes of DEGs with L. biglobosa infection by STEM and KEGG analyses

To understand the dynamic changes rules in the transcriptome of oilseed rape in response toL.biglobosainfection and its correlation with the measured time points,we clustered the total of 19 015 DEGs at the 6 time points in order to understand whether the DEGs were sequentially induced with pathogen infection. The results of clustering (P<0.05) showed significant differences in 12 profiles obtained from co-clustering and KEGG enrichment analysis (Fig.4). The profile 6 contained 1 310 DEGs that were downregulated sharply in the first 12 hours,and then continuously upregulated,with a slight fluctuation in the thereafter 3 time points. They were involved with plant-pathogen interaction,protein kinases,the MAPK signaling,the phosphatidylinositol signaling system and glycerophospholipid metabolism,which all related to the recognition of the pathogen and signal transduction. Profile 27 showed opposite change trend to profile 6 and the DEGSin profile 27 were enriched in mRNA splicing,ubiquitin system,plant hormone signal transduction and other related regulation pathways. All of those illustrated the transcriptional reprogramming event occurred at pathogen conidia prepared for germinating stage (4-12 h),the earliest response toL.biglobosainfection. The DEGs in profile 10 contained the KEGG pathways included glucosinolate biosynthesis,nitrogen metabolism,cytochrome P450,alpha-linolenic acid metabolism,polyketide sugar unit biosynthesis,which indicating that pathways mainly related in glucosinolate biosynthesis,The DEGs were downregulated sharply in the first 24 hours,and then continuously upregulated to the peak at 96 h. Profile 8 consisted of DEGs being firstly downregulated and then upregulated,which was mainly involved in photosynthesis,nitrogen metabolism,photosynthesis proteins,carbon fixation in photosynthetic organism,nitrogen metabolism,glycine,serine and threonine metabolism,all these pathways related photosynthesis. It suggested that photosynthesis is suppressed at early 24 h and then restored. DEGs in other pathways showed up and down regulation in the infected oilseed rape for primary metabolism may be the readjustment of plant in response to the pathogen infection. To survive the invasion of pathogens,plants require an effective response to suppress the further propagation of the pathogen. Such a quick response heavily depends on the gene network involving plant-pathogen interaction,MAPK signaling cascades,plant hormone signaling pathways and secondary metabolite biosynthesis. In this study,the DEGs of plant-pathogen interaction,MAPK signaling cascades,glucosinolate biosynthesis showed down regulation in the first time,then continuous up regulation to the peak at 96 h in profile 6 and profile 10,but the plant hormone shown opposite trend in profile 27,indicating that the genes of plant immunity was strongly activated in 96 h,which was the stage of mycelia formed network.

2.4 Plant-pathogen interaction involved in plant defense to L. biglobosa

Our study identified pattern recognition receptors (PRRs),which include bacterial flagellin receptorFLS2,elongation factor Tu (EF-Tu) receptorEFR. They were mainly upregulated at 12,24 and 96 h. The research also found that the expression of chitin elicitor receptor kinase1/Lysin motif receptor-like kinase1 (CERK1/LysMRLK1),CERK1(BnaC04g43970D) was upregulated at 24 h and 96 h. Twobak1 genes were upregulated at 24,48,96 h. TwoCDPKswere identified,which were mainly upregulated at 12,24 and 96 h and downregulated at 4 h. FourTIR-NBS-LRRdisease resistance genes (TIR-NBS-LRR class) were mainly upregulated at 24 and 96 h. It was also found that pathogenesis-related protein (BnaCnng17400D) was downregulated at 4 h and 24 h,but upregulated at 96 h (Fig.5).

Genes in red are upregulated; genes in green are downregulated. The same as below

2.5 Activation of the MAPK signaling in oilseed rape response to L. biglobosa infection

MAPK signaling cascades responding to the infection ofL.biglobosawere clustered in profile 6. NineteenMKKKs,twoMKKs,and eightMAPKsgenes were activated in expression (Fig.6). EightMKKK18swere induced at 96 h. TwoMKKK17swere induced at 4,24,and 96 h. TwoMKK9s(BnaC06g23550DandBnaA07g22640D) were stimulated notably at 96 h. ThreeMAPK17swere repressed at 12 h,activated at 96 h and were up at 96 h. TwoMAPK16swere went up at 48 h,and oneMAPK5(BnaC09g24030D) was increased at 12 and 96 h.

Fig.6 Heat maps of MKKK,MKK and MAPK genes induced after L. biglobosa infection

2.6 Plant hormone signal transduction in oilseed rape response to L. biglobosa infection

The DEGs related to plant hormone signal transduction were clustered in profile 27,showed firstly upregulated then downregulated. The DEGs related to abscisic acid (ABA),brassinosteroid (BR),cytokinin (CTK),ethylene (ET),gibberellin (GA),auxin (IAA),jasmonic acid (JA) and salicylic acid (SA),showed significant changes during the infection process. Downregulated genes related to abscisic acid (ABA),brassinosteroid (BR),cytokinin (CTK),gibberellins (GA),and auxin (IAA),had more DEGs than upregulated genes (Fig.7,A),indicating that these phytohormone signals were inhibited after the pathogen infection. However,more upregulated DEGs were found to be related to jasmonic acid (JA),ethylene (ET) and salicylic acid (SA),suggesting that DEGs in plant hormone were activated and maybe related to plant defense.

Sixty-eight percent JA-related DEGs were upregulated. The key genes in JA pathway wereCOI1,JAZandMYC.COI1 was upregulated at 4 and 48 h. Among 37MYCgenes,24 of them were upregulated. JAZ is a repressor of the JA response. It was upregulated at 4,12,24,and 96 h. Thirty-fiveJAZswere induced at 96 h. In respect to the dynamic change of whole gene expression,they were activated at 4 and 12 h,inhibited at 36 h and 48 h,and then significantly induced at 96 h. In considering the quantity of upregulated DEGs,JA is the signal pathway that plays a leading role for resistance in oilseed rape in response toL.biglobosainfection (Fig.7,B).

In total,46 upregulated and 37 downregulated DEGs belong to ET.EIN3 was downregulated at 4 h and 48 h,but upregulated at 96 h.ERF1/2 was downregulated at 48 h and upregulated at 12,24,and 96 h.ETRwas upregulated at 12,24,48 and 96 h,but downregulated at 36 h (Fig.7,B). DEGs were mainly downregulated at 36-48 h,but upregulated from 12,24 and 96 h. This dynamic change of expression was synergistically similar to JA signals pathway.

Salicylic acid (SA)-related genes were not largely altered at the all time points. DEGs of SA signal were mainly downregulated at 4-24 h,upregulated from 36 to 96 h,with the all 12 DEGs downregulated at 12 h and upregulated at 48 h.NPR1 was up regulated at 96 h andTGAupregulated at 24,48,and 96 h.PR-1 was highly induced at 4,24,36 h and 96 h (Fig.7,B),suggesting that the dynamic change was not synergistically similar to JA/ET signal pathway.

A. Changes in the number of DEGs related to phytohormone signal transduction； B. Time-series changes in the number of DEGs related to the jasmonate/ethylene/salicylic acid pathways； The number in the circle represents the number of genes

2.7 Glycosylates biosynthetic to increase oilseed rape resistance

Glucosinolates biosynthetic and cytochrome P450 were clustered in profile 27,two Cytochrome450 families,CYP79 and CYP83 involved in glucosinolates biosynthetic. CYP79 family catalyzes hydroxylation of amino acids to produce corresponding acetaldoxime,while CYP83 family catalyzes the acetaldoxime to glucosinolates. In our study,twenty-seven DEGs associated with Glucosinolate (GSLs) were enhanced. Interestingly,all DEGs were only upregulated at 4 h and 96 h,no differentially expressed at other timepoints,which covered eight genes expressed at both time points (Table 3). All DEGs in Cytochrome P450 family,including sixCYP79B,twoCYP83A,oneCYP83B,andCYP79Bwere upregulated at 96 h,sulfotransferase (SOT16,SOT17 andSOT18),were also upregulated at 96 h.MAM1 was upregulated at 4 h and 96 h. fourSUR1 were only upregulated at 96 h. Four genes encoding UDP-glucosyl transferase were also upregulated,suggesting that biosynthesis of glucosinolates is regulated byL.biglobosainfection,with an important role in plant disease resistance and a unique function in the oilseed rape-pathogen interaction.

Table 3 Genes related to glucosinolates biosynthesis at 4 and 96 h in response to L. biglobosa

3 Discussion

The fungal genusLeptosphaeriahave a range of Cruciferae hosts,especially,the most important disease inBrassicanapus(oilseed rape) worldwide[2]. In the present study we investigated the interaction between theBrassicanapusandL.biglobosainfection during a short time-series expression course. The results showed that 1310 DEGs initially downregulated sharply in the first 12 hours,and then continuously upregulated with a slight fluctuation in the thereafter 3 time points in profile 6,indicating that the genes associated with recognition pathogen are inhibited when the spores swelled. It might be due to the pathogen secrete effector proteins to suppress host defenses at early infection stage[29-31]. It is activated at the spores germinating and hyphae forming stage,and it might be the comprehensive defense system was formed to defend againstL.biglobosainfection. In addition,the spliceosome,ubiquitin,plant hormone signal transduction and transcription machinery were rapidly activated toL.biglobosainfection. The results indicated that the transcriptional reprogramming occurred at 4-12 h,then these pathways were turn down. Overall,the study showed that the process of infection is dynamic and the plant-pathogen interaction can result in the changes of relevant gene expression during the course of infection. In other plant-pathogen interaction researches,for example,BrassicanapusversesL.maculansandS.sclerotioruminfection,they also found dynamic changes in genes expression[10,19]. This indicates that the resistance-related genes are not continuously expressed,and their expression level maybe affected by the trade-off between plant and pathogen.

PTI is initiated in plants when PAMPs are recognized by pattern recognition receptors (PRRs). In this study,the gene of chitin receptor kinase (CERK1) was up-regulated,indicating that chitin ofL.biglobosawas one of the PAMPs,which was common case for majority of necrotrophic fungi. Chitin perception and signaling have been well characterized inArabidopsisresponse toBotrytiscinereainfection[32-33]. It implies that oilseed rape andArabidopsisthalianamay have a similar pattern in identifying the necrotrophic fungi[34]. ETI leads to a HR and localized host cell death[10]. Therefore,the infection of biotrophic pathogens could be inhibited due to the death of tissues. Interestingly,necrotrophic fungi activated R protein-mediated ETI to cause hypersensitivity (HR)-like cell death and created conditions for further infection. In general,plant immune systems are very complex and reflect the multiplicity of necrotroph virulence mechanisms. In our study,identifiedTIR-NBS-LRRgenes were mainly upregulated at 24 and 96 h,even though small lesions were observed at 48 h,suggesting that the HR occurred at 24-48 h. The phenotypic characteristics in infested leaves and network of mycelia formed were noticed at 96 h,which shown that defense line has been broken. However,many defense-related genes were induced at 96 h,suggesting that the defense response was lagging. This may be related to the resistance of someB.napusvarieties.

MAPK cascades minimally consist of a MKKK-MKK-MAPK module[16]. Sunetal. identified 66MKKKgenes inB.napusand found thatBnaMKKK18 andBnaMKKK19 could elicit HR-like cell death in the leaves ofNicotianabenthamiana[35]. A previous transcriptomic study proved thatBnaMKKK17 andBnaMKKK18 are responsive to several pathogens and defense hormones[36],suggesting thatBnaMKKK17/18/19 are crucial defense factors in the response of oilseed rape to pathogens. In our study,MKKK17/18/19 were strongly activated at 24 and 96 h,two copies ofMKK9 were generated at 96 h. They may play an important role in the initial stages of infection,A recent study found the interaction of MKKK19 with MKK9,which was confirmed in a yeast two-hybrid system and bimolecular fluorescence complementation analysis inB.napus[20]. It was uncovered that BnaMKKK19 probably mediated cell death through BnaMKK9. A research also revealed that Arabidopsis MKK9 is a regulator of cell death and can similarly accelerate cell death inN.benthamiana[37]. Liangetal. identified BnaMKK9-BnaMAPK1/2-BnaWRKY53 andBnaMKK9-BnaMPK5/9/19/20 in canola (Brassicanapus) and also found thatMKK9-MPK19-WRKY20 was responsive toS.sclerotioruminB.napus. MKK9 is reported to directly interact with a number of MAPKs,including BnaMPK1,-2,-5,-9,-19,and -20 inB.napusand AtMAPK5,-10,-17,and -20 inArabidopsis[38]. In our research,BnaMKK9andBnaMAPK5/12/17 were prompted in response toL.biglobosaat 96 h,and the obvious disease spots were noticed at 96 h,suggesting that the appearance of disease spots was maybe related to the MKKK19-MKK9 cascades.

SA positively adjusts plant defense against biotrophic pathogens,whereas JA/ET pathways commonly regulate resistance to necrotrophic pathogens[18]. Our study showed that the DEGs related to JA and ET pathways were up-regulated,and SA was even more up regulated,indicating that JA and SA synergistically played the roles in the defense againstL.biglobosainfection,and it is consistent with the results on oilseed rape leaves withL.biglobosainfection at 7 dpi[7]. The increase level of JA in response to pathogen infection clearly feature its activity in plant defense response[39]. In the absence of JA-Ile and JA,JAZ interacts with MYC2 and inhibits the JA response. Exogenous JA binds to its receptor F-box protein COI1,resulting in the degradation of JAZ by the 26S proteasome,allowingMYC2 to be upregulated and MYC2 to be phosphorylated. This leads to degradation of the JAZ protein. The MYC2 transcription factor removes inhibition,and the JA pathway is ultimately activated[40]. JAZ is an inhibitor of the JA pathway and contributes an important function in the balance of “growth-resistance”[41]. In our study,it shown that a number of resistance-related gene expression were largely increased at 96 h,but thejazgene was also upregulated at 96 h. Therefore,it is inferred that the growth regulation intensity was greater than the resistance regulation intensity after 96 h. ET plays diverse roles in plant defense response[18]. EIN3 is an important node in ET signal pathway as it integrates most plant hormones and stress signals to form a complicated transcriptional regulation network[42]. For example,EIN3/EIL1 can directly regulate expression of the immune receptor FLS2 and correlate positively the disease resistance inA.thaliana[43]. In this study,allEIN3 genes were upregulated at 96 h,more than 1 000 genes have been found to be directly regulated by EIN3[42],it can be inferred that a number of genes were regulated byEIN3 and expressed after 96 h. It is well known that ERFs confer resistance to several necrotrophic fungi[44]and activate expression of several defense-related genes[45]. Almost allETRswere upregulated in our study,suggesting that ET is an important signal in oilseed rape in response toL.biglobosa. Plant salicylic acid signal transduction is identified to play a key role in preventing the infection caused by biotrophic[18]. Our work found that a number of resistance genes related to salicylic acid signaling were upregulated too,indicating that the salicylic acid signal pathway plays a great role in the oilseed rape inoculated withL.biglobosa.

Glucosinolates are a diverse class of S- and N-containing secondary metabolites,which are important in plant defense. Indole glucosinolates were found to be involved in Arabidopsis plants defense againstB.cinerea. Mutant Arabidopsis plant lines with lack of camalexin or indolic or aliphatic glucosinolate biosynthesis were hypersusceptible toSclerotiniasclerotium[46]. In this study,it was confirmed that many genes in the glucosinolates biosynthesis pathway were upregulated,suggesting that glucosinolate is produced upon pathogen infection occurs,which plays an important role in plant disease resistance. Interestingly,it was only induced at 4 and 96 h,no difference in the expression during 12-48 h,suggesting that the glucosinolate can take early response toL.biglobosainfection. Glucosinolate inB.napuswas found toxic toL.maculans[19]andSclerotiniasclerotium[20],but the defense effects of glucosinolates on necrotrophic fungal pathogens for their specialized or broad-host-spectrum need be further studied.

Acknowledgements: We thank LIU Sheng-yi and CHEN Yu for their kind advice and help with writing this paper.