CHEN Chang-long,YUAN FangLI Xiao-ying,MA Rong-cai,XIE Hua

1 Beijing Agro-Biotechnology Research Center,Beijing Academy of Agriculture and Forestry Sciences,Beijing 100097,P.R.China

2 Beijing Key Laboratory of Agricultural Genetic Resources and Biotechnology,Beijing 100097,P.R.China

Abstract Chinese cabbage (Brassica rapa subsp. pekinensis) suffers from soft rot disease caused by Pectobacterium carotovorum(Pc).To uncover the mechanisms underlying the defense response of Chinese cabbage to Pc,we constructed a suppression subtractive hybridization (SSH) library from Pc-infected cabbage and obtained 1 919 non-redundant expressed sequence tags (ESTs),which were used for cDNA microarray.We detected 800 differentially expressed genes (DEGs) in cabbage at different time points post-Pc inoculation,which were further confirmed by quantitative real-time PCR.One quarter of these DEGs were involved in the biotic stress pathways visualized by MapMan.Among them,8,8,1,3,and 2 DEGs were related to jasmonic acid (JA),ethylene (ET),JA+ET,auxin,and abscisic acid (ABA) signaling pathways,respectively,while no DEG was detected for salicylic acid (SA) signaling.Assessment of phytohormone production in the Pc-infected leaves showed that JA and ET production was increased,while SA production was decreased.Treatment with JA,methyl jasmonate (MeJA),the ET precursor 1-aminocyclopropane-1-carboxylate (ACC),or combinations thereof,reduced the disease severity,and the JA and JA+ACC treatments were superior and performed equally well.Our findings suggest that JA and ET may act synergistically against Pc infection in Chinese cabbage,and JA-mediated signaling might be the most significant.

Keywords:Brassica rapa,Pectobacterium carotovorum,gene expression,defense response,hormone signaling

1.Introduction

Chinese cabbage (Brassica rapasubsp.pekinensis;Brassicaceae) is an important vegetable crop in many countries,including China.However,one of the major hurdles limiting Chinese cabbage production is soft rot disease (SRD) caused byPectobacterium carotovorum(Pc);a ubiquitous necrotrophic bacterial pathogen (Rohet al.2010).Pectobacteriumcauses SRD and leads to substantial production losses during growth and post-harvest storage(Rohet al.2010).Pectobacteriumhas a broad host range,and plant resistance to this pathogen is complex and does not appear to involve single resistance genes (Davidssonet al.2013).Uncovering the defense responses of Chinese cabbage toPcis thus important for elucidating the underlying mechanisms of the immune response and might facilitate the development of strategies for enhancing plant disease resistance to this pathogen.

In the plant immune system,plant cells perceive pathogen-associated molecular patterns (PAMPs) through pattern recognition receptors,and this recognition initiates pattern-triggered immunity (PTI).PTI is considered to be the primary plant defense strategy againstPectobacterium(Davidssonet al.2013).Once PTI has been initiated,general defense responses are induced,including the generation of reactive oxygen species (ROS) (Camejoet al.2016),induction of cell wall defense mechanisms (Miedeset al.2014),mitogen-activated protein kinase (MAPK)pathway (Meng and Zhang 2013) and phytohormonemediated defense networking (Bigeardet al.2015),and the synthesis of phenolic compounds (Luzzattoet al.2007a).As central mediators of plant–pathogen interactions and plant defense,phytohormone signaling pathways have been well studied,and the crosstalk between phytohormone signaling pathways dominates the defense response in many plant species (Bruceet al.2007;Pieterseet al.2012) and determines the balance between the signaling pathways and the response to a particular stimulus (Aaltoet al.2012).The phytohormones mediating plant defense responses constitute an interactive set of endogenous signaling molecules,such as jasmonic acid (JA),ethylene (ET),salicylic acid (SA),abscisic acid (ABA),and indole-3-acetic acid (IAA) (Andersonet al.2004;Sharaf and Farrag 2004).Among them,SA,JA,and ET are the three major phytohormones,and the signaling pathways mediated by SA,JA,and ET interact extensively(Dodds and Rathjen 2010).

Traditionally,SA is thought to activate defense signaling against biotrophic and hemibiotrophic pathogens,while the JA and ET signaling pathways are considered to be associated with defense responses against necrotrophs (Kunkel and Brooks 2002;Glazebrook 2005).However,the signaling network triggered byPectobacteriumnecrotrophs appears to be more complex.It was reported that the induction of JA-mediated,ET-mediated,or SA-mediated defenses could enhance the resistance againstPectobacterium(Palvaet al.1994;Vidalet al.1998;Norman-Setterbladet al.2000;Kariolaet al.2003;Liet al.2004;Montesanoet al.2005;Luzzattoet al.2007a,b;Luzzatto-Knaanet al.2014),and the major phytohormone signaling pathway in the defense response toPcinfection tends to vary across studies of different plant–Pcpathosystems.

In the model plantArabidopsis thaliana,Norman-Setterbladet al.(2000) discovered thatP.carotovorumor its secreted cell wall-degrading enzymes trigger two distinct defense pathways.The first is mediated through the synergistic action of ET and JA,potentiated by SA,and leads to the expression of marker genes of JA and ET,while the second pathway is strictly JA mediated,inhibited by ET or SA,and controls the expression of a marker gene of JA.Another study showed thatP.carotovorumactivates both SA-and JA/ET-dependent plant defense signalingviathe induction of marker genes,which corroborates the report of Liet al.(2004).In other plants such as tobacco,the addition of SA to growing seedlings rendered them almost fully resistant to subsequentPcinfection,and this resistance was caused by the induction of plant defense responses (Palvaet al.1994).In contrast,Vidalet al.(1997,1998) demonstrated that the tobacco plant defense responses induced by cell-free culture filtrates (CFs) ofPcwere SA-independent.In potato,the production of ET and JA was up-regulated,while SA levels were down-regulated,in leaf tissues in response to CFs ofPc.Furthermore,the addition of ET and JA was demonstrated to induce transcript accumulation of plant defense-related genes,suggesting that these two signaling molecules might play a role in potato defense signaling againstP.carotovorum(Montesanoet al.2005).In calla lily (Zantedeschia aethiopica),MeJA was able to reduce disease symptoms dramatically relative to the other two chemicals Bion and β-aminobutyric acid (BABA)related to SA-pathway (Zimmerliet al.2000;Cohen 2001).Furthermore,the disease symptoms were repressed for a longer period,indicating that the JA/ET-signaling pathway is required for long-term resilience,while the SA-signaling pathway acts in the short term in the defense response(Luzzattoet al.2007b).MeJA treatment was also found to trigger a concerted priming of protein expression inZ.aethiopicafollowing infection withPc,suggesting that the JA/ET signaling pathway plays a role in the protein priming involved in the plant defense response toPc(Luzzatto-Knaanet al.2014).

Although substantial progress has been made in understanding the phytohormone signaling network in the defense againstPcinfection,few studies have directly explored the effect of hormone signaling in Chinese cabbage in response toPc(Liuet al.2019).Herein,we constructed a suppression subtractive hybridization (SSH) cDNA library of Chinese cabbage infected withPc,and obtained 6 282 expressed sequence tags (ESTs).Using cDNA microarray in combination with biochemical and physiological analyses,we examined the global gene expression profiles and signaling pathways inPc-infected Chinese cabbage.Our data demonstrated that JA-and ET-mediated signaling pathways play key roles in the defense againstPcinfection in Chinese cabbage.

2.Materials and methods

2.1.Plant materials and treatments

Seeds of Chinese cabbage cultivar Longbai 2 were grown in sterile soil (mixture of vermiculite:turf soil:clay at 1:1:1(v/v)) at 30% relative humidity (RH) with a 14 h light (8 000 to 10 000 lux)/10 h dark cycle for 4 weeks before inoculation.Pcstrain BC1 (PcBC1),isolated from Chinese cabbage,was cultured overnight at 28°C in Luria Broth (LB) medium.Four weeks later,at the six–eight true-leaf stage,three leaves (the third,fourth,and fifth leaves) from each plant were inoculated with bacterial suspension (8 µL,OD600=0.2),using a sterile micropipette to prick the midrib 2 cm from the end of the petiole.Inoculated plants were then incubated at 28°C and 95% RH under the above light/dark cycle.

2.2.RNA isolation and SSH cDNA library construction

Leaf samples from 10 plants either inoculated withPcBC1 suspension in LB medium or inoculated with LB medium only were collected for total RNA extraction at 6,12,24,36,48,and 72 h post-inoculation (hpi).Total RNA was isolated using an RNeasy Plant Kit (Qiagen,Hilden,Germany),and the concentration and quality of the RNA were measured using a NanoDrop 2000c (Thermo Fisher,Waltham,MA,USA).Equal amounts of total RNA from the different time points for thePc-infected or control plants were combined by pooling the RNA samples for enrichment of the genes.The mRNA was purified using an mRNA Purification Kit(Amersham Pharmacia Biotech,Piscataway,NJ,USA).SSH was performed to identify the genes up-regulated inPc-infected Chinese cabbage using a PCR-Select™cDNA Subtraction Kit (Clontech,Mountain View,CA,USA)according to the manufacturer’s instructions.Samples containing a total of 2 μg mRNA fromPc-infected and control leaves were used as tester and driver samples,respectively.Subtracted cDNAs were purified using a Concert™ Rapid PCR Purification System (GIBCOBRL®,Grand Island,NY,USA) and cloned into the pGEM-T Easy vector (Promega,Madison,WI,USA) to construct the cDNA library.About 3×105clones were obtained after electrotransformation intoEscherichia coliDH5α.

2.3.EST sequencing and sequence analysis

DNA sequencing was performed on a CEQ™ 8000 Genetic Analysis System (Beckman Coulter Inc.,Brea,CA,USA).The vector sequences and low-quality ambiguous sequences were trimmed using the ‘Sequencing’ module of the CEQTM8000 Genetic Analysis System Software.DNA sequences generated from each cDNA clone were edited to remove poly(A) sequences.Sequences that were shorter than 150 bp or had >4% ambiguity were not processed further.A total of 6 282 high-quality sequences were submitted to GenBank (accession numbers DN960311 to DN960318 and DN960326 to DN966599).A homology search was performed using BLASTN to find the best matches in theB.rapagenome database (ftp://ftp.ensemblgenomes.org/pub/plants/release-33/fasta/brassica_rapa/dna/) with a cutoff ofE-value<1e–05.The functions of the hit genes were annotated using the BLASTX program against the Nr database with anE-value<1e–05.

2.4.Preparation of the cDNA microarray

After removing highly similar ESTs,1 919 ESTs derived from the SSH library of Chinese cabbage were used for preparing the cDNA microarray.Thus,the 1 919 cDNA probes were printed on poly-lysine coated glass in triplicate using an Array Spotter Generation III (Genomic Instrumentation Services,Foster City,CA,USA).The array platform generated three technical replications per hybridization.Six housekeeping genes (tubulin,histone H4,histone H2B,rubisco activase,chlorophyll A/B-binding protein,and citrate synthase)were used as internal controls,and a response regulator(GenBank accession no.Q47279),extracellular protease(GenBank accession no.AAA24858),β-(1,4)-glucan glucanohydrolase (GenBank accession no.AAC02965),and HrpN (GenBank accession no.AAD01739) fromPc,and eight yeast intergenic sequences,were used as external negative controls in the cDNA microarray.

2.5.Hybridization of cDNA microarray and data analysis

Leaf tissues (about 0.5 cm) from around the lesions were collected at 0.5,2,6,12,and 24 hpi fromPcBC1-or water-inoculated plants for total RNA extraction.The cDNA microarray hybridization was performed at CapitalBio Co.Ltd.(Beijing,China).Hybridization images were obtained using a ScanArray Express scanner (Packard Bioscience,Meriden,CT,USA) and analyzed with GenePix Pro 4.0(Axon Instruments,USA).For each experiment,the microarray probe was represented in triplicate and the experiments were performed in duplicate by dye swap.Six data points were thus produced for each gene.Since data derived from low signal intensities are less reliable,a signal intensity cut-off value of 1 000 was used.Genes with |log2FC|>1 were identified as DEGs at each time point.

2.6.Bioinformatics analysis

The DEGs obtained by microarray were annotated against the GO and KEGG databases before the classification figure was illustrated using the tools on BMKCloud (https://www.biocloud.net/) under default parameters.These DEGs were also classified by Mercator (Lohseet al.2014) into MapMan functional plant categories.MapMan was used to visualize the biotic stress pathways the DEGs were possibly involved in (Thimmet al.2004).Venn diagrams,heatmaps,and cluster patterns of gene expression were constructed using the tools available on BMKCloud.

2.7.Quantitative real-time PCR (qPCR) and semiquantitative RT-PCR validation of the microarray data

Leaf samples of about 0.5 cm from around the inoculated site from nine leaves on three plants (three leaves/plant)per treatment were harvested at the indicated time points,flash frozen in liquid N2,and kept at −80°C.Total RNA was extracted using an RNAprep Pure Plant Kit (Tiangen,Beijing,China) and treated with DNase I.The cDNA was synthesized from total RNA using oligo(dT) primers and the reverse transcriptase from the PrimeScript™ RT Master Mix(TaKaRa,Dalian,China).TB Green™ PremixEx Taq™II (Tli RNaseH Plus) (TaKaRa,Dalian,China) was used for qPCR analysis with three technical replicates for each sample.The thermal cycling program was 95°C for 30 s,followed by 40 cycles of 95°C for 5 s and 60°C for 34 s.Melting curve analysis was used to ensure the specificity of the product.Normalization of gene expression across different samples was performed using glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as an internal control(Qiet al.2010) and the relative expressions were calculated using the 2−ΔΔCtmethod.Primer sequences are shown in Appendix A.We also conducted semi-quantitative RT-PCR using β-tubulin gene as a reference for validation.The gene-specific primer pairs used for the RT-PCR are listed in Appendix B.The RT-PCR products were analyzed by agarose gel electrophoresis.

2.8.Measurements of phytohormone levels

ET emissions of 0–6,6–12 and 12–24 hpi from infected and control Chinese cabbage were measured by placing nine excised seedlings in a 5-L plastic bag connected to a gas collection bag,with three replicates.The collected gas was then subjected to gas chromatograph (GC-14B,Shimadzu)with column temperature at 65°C,inlet temperature at 120°C,detector temperature at 150°C,stationary phase of GDX-102 and ethylene retention time of 1.27 min.SA was extracted and analyzed by HPLC using a modified method with three replicates (Dewdneyet al.2000).JA was quantified using the internal control (±)-9,10-dihydro-JA with three replicates as described previously (Baldwinet al.1997).

2.9.Effects of phytohormones on the resistance of Chinese cabbage to Pc

Each seedling was sprayed with either the ET precursor ACC(1 mmol L–1),JA (50 µmol L–1),or MeJA (50 µmol L–1) in single water solutions or in combinations,using H2O as a negative control.The seedlings were incubated in a chamber at 28°C under 95% RH with a 14 h light (162 µmol m−2s−1)/10 h dark photoperiod.To observe the induced resistance toPc,the plants were inoculated with 5 µL of the bacterial suspension(OD600=0.2) at 2 h after the phytohormone treatment and then incubated at 28°C and 95% RH in a climate chamber.The relative disease severity (RDS) of 27 leaves on nine seedlings (that is,three leaves for each seedling) for each treatment was calculated by comparing the lesion length with the largest disease length at 45 hpi.

To exclude antimicrobial ability of the three hormones,Pccultured overnight in LB only or with added ACC (1 mmol L–1),JA (50 µmol L–1) and MeJA (50 µmol L–1),respectively,was used for determining the virulence on Chinese cabbage petiolesin vitro(Liet al.2020).The necrotic area was recorded at 24 hpi.Three petri dishes with three petioles per dish for each treatment were tested (n=9).

2.10.Data statistical analysis

The significance analysis of data differences was performed by one-way ANOVA in SPSS 13.0.

3.Results

3.1.Gene expression profiling of the SSH cDNA library of Chinese cabbage in response to Pc infection

An SSH cDNA library was constructed using the leaves ofPc-infected and mock-inoculated plants.A total of 6 282 high-quality EST sequences were obtained.Of these,6 036 ESTs matched the sequences in theB.rapagenome database.After removing highly similar ESTs,1 919 ESTs derived from the SSH library of Chinese cabbage were used for the cDNA microarray.The microarray result showed that 1 627 ESTs covering 1 627 genes were expressed for all time points.Among these genes,800 genes were identified as differentially expressed genes (DEGs) with|log2(fold change (FC))|>1 at any time point (Appendix C).The number of DEGs (both up-regulated and downregulated) in response toPcinoculation increased with time(from 0.5 to 24 hpi) with 61,142,357,545,and 689 DEGs at 0.5,2,6,12,and 24 hpi,respectively (Fig.1-A and B).At each time point,there were more up-regulated DEGs than down-regulated DEGs (Fig.1-B).The Venn diagram in Fig.1-A indicates that there were some DEGs specific to each time point,with the number of specific DEGs being highest at 24 hpi (184).However,most of the DEGs were shared between different time points,with a large portion being shared by 2–4 time points and only four DEGs being shared by all five time points (Fig.1-A).These four genes wereBra005104(predicted:probable WRKY transcription factor 33 (B.rapa)),Bra006692(predicted:zinc finger protein ZAT12-like (B.rapa)),Bra018160(predicted:asparagine synthetase (glutamine-hydrolyzing) (B.rapa)),andBra039217(predicted:tetraspanin-8 (B.rapa)).They were all down-regulated at 0.5 hpi and up-regulated from 2 to 24 hpi,during which their relative expression levels increased (Appendix C).

The DEGs were subclassified into three groups according to their expression patterns (Appendix D).The fold change of most DEGs (480 DEGs in total) shown in subcluster 1 and subcluster 2 increased from 0.5 to 24 hpi,while the other 320 DEGs shown in subcluster 3 decreased in expression with infection time.The magnitude of the fold change of gene expression in subcluster 1 was greater than in subcluster 2.

To confirm the results of the microarray data,26 DEGs were selected for qPCR.The expression of 22 genes at different time points was consistent with that of the microarray data (Appendix E),confirming the reliability of the microarray data.Four genes (Bra012939,Bra029319,Bra028753,andBra037954) exhibited different expression regulation patterns on the whole in comparison to the microarray data at different time points.In addition,we also conducted semi-quantitative PCR of 12 genes to confirm the microarray data,which showed consistent results(Appendix F).

Fig.1 The expression patterns of differentially expressed genes (DEGs) in Chinese cabbage induced by Pectobacterium carotovorum infection analyzed by microarray.A,Venn diagram of the DEGs at five different time points post-inoculation.B,the numbers of up-and down-regulated genes at five different time points.

3.2.Functional annotation analysis of the DEGs

Nearly all of the DEGs (798 out of 800 DEGs) were annotated in the Non-redundant (Nr) protein sequence dataset,and 787 and 327 DEGs were annotated in the Gene Ontology (GO)and Kyoto Encyclopedia of Genes and Genomes (KEGG)Databases,respectively.In the biological process category of the GO annotation,more DEGs were classified into metabolic process,cellular process,and single-organism process and responses to stimulus,while in the molecular function category,more DEGs were involved in catalytic activity and binding.According to the cellular component category,more DEGs were localized in cell (part),organelle(part),and membrane (Fig.2-A).KEGG pathway analysis allocated 253 DEGs to 82 KEGG pathways,and 50 of the most enriched pathways are shown in Fig.2-B.The top three pathways with the highest number of DEGs included carbon fixation in photosynthetic organisms,photosynthesis,and plant–pathogen interaction.The plant hormone signal transduction pathway also harbored a certain number of DEGs (9 DEGs),indicating its involvement in the response toPcinfection.Based on the MapMan functional classification,with the exception of 18.19% of the DEGs that could not be assigned,the majority of the DEGs were classified into the category of protein (10.56%) (Fig.2-C).Interestingly,6.34 and 2.93% of the DEGs were classified in stress and hormone metabolism,respectively (Fig.2-C).

Fig.2 Classification of the differentially expressed genes (DEGs) in Chinese cabbage induced by Pectobacterium carotovorum infection according to Gene Ontology (GO) database,Kyoto Encyclopedia of Genes and Genomes (KEGG) database,and MapMan function analysis.A,GO categories of the DEGs.The number and percentage of DEGs in each subcategory for the three main categories of biological process,cellular component,and molecular function are indicated.B,KEGG classification of the DEGs.The x-axis shows the percentage of the annotated genes in each category,and the number of genes is indicated at the top of the bar.C,MapMan functional plant categories of the DEGs.The percentage of DEGs in each category is shown.

3.3.Visualization of the biotic stress pathways of the DEGs

Fig.3 Visualization of the biotic stress pathways associated with the differentially expressed genes (DEGs) in Chinese cabbage following Pectobacterium carotovorum infection at different time points.Red and blue dots represent the genes that were up-and down-regulated,respectively,in Chinese cabbage infected with P.carotovorum compared to the mock-inoculation control.The change in magnitude of gene expression is indicated by the intensity of the color.hpi,hours post-inoculation.

MapMan was used to visualize the putative biotic stress pathways that the DEGs participated in.The result showed that a fair number of DEGs (208 out of 800) in response toPcinfection were mapped for 219 data points putatively involved in biotic stress pathways,including redox state,signaling,transcription factors,cell wall,PR-proteins,and secondary metabolites (Fig.3).From 0.5 to 24 hpi,increasing numbers of genes were expressed,and the same pattern was observed for all 800 DEGs (Figs.1 and 3).Several DEGs were predicted to be involved in hormone signaling,namely,(i) eight JA signaling DEGs,including two lipoxygenase genes (Bra003526andBra008037),one allene oxide cyclase 3 gene (Bra017350),one allene oxide synthase gene (Bra035320),three TIFY genes(Bra022254,Bra025713,andBra031065),and one nitrilespecifier protein 5 gene (Bra037506);(ii) eight ET signaling DEGs,including three 1-aminocyclopropane-1-carboxylate oxidase genes (Bra006713,Bra015380,andBra027049),one 1-aminocyclopropane-1-carboxylate synthase 6 gene (Bra035236),three ET-responsive transcription factor genes (Bra010880,Bra012939,andBra029319),and a methylesterase 9-like gene (Bra011722);(iii) one JA+ET signaling DEG encoding a CBS domain-containing protein CBSX2 (Bra017627);(iv) three auxin signaling DEGs,including one IAA-amino acid hydrolase ILR1-like 4 precursor gene (Bra030419),one cytochrome b561 and DOMON domain-containing protein gene (Bra035384),and one myrosinase-like gene (Bra039825);and (v) two ABA signaling DEGs,including one zeaxanthin epoxidase gene(Bra012127) and one abscisic acid 8´-hydroxylase 3-like gene (Bra021965) (Table 1).These genes were classified as related to hormone synthesis,hormone signal transduction,or hormone responsive genes by MapMan.Of these genes,Bra003526,Bra008037,Bra017350,andBra035320are related to JA synthesis;Bra006713,Bra015380,Bra027049,andBra035236are related to ET synthesis;andBra017627is related to both JA and ET synthesis.In addition,the number of DEGs associated with JA and ET signaling pathways was greater than that of auxin and ABA,while no DEG was found to be involved in SA or brassinosteroid signaling (Fig.3).These results indicated that the JA,ET,auxin,and ABA signaling pathways are suggested to be involved in the defense response towardsPcinfection in Chinese cabbage.

3.4.Expression patterns of the DEGs involved in phytohormone signaling

An expression heatmap of the DEGs involved in phytohormone signaling of the biotic stress pathways suggested by MapMan (Table 1) was drawn based on the microarray data.It showed that there was only one gene related to ET signaling,namelyBra035236(predicted:1-aminocyclopropane-1-carboxylate synthase 6 (B.rapa))that was differentially expressed (down-regulated) at 0.5 hpi,after which increasing numbers of DEGs were induced with time afterPcinfection.At 24 hpi,these genes were all significantly regulated byPcinfection (Fig.4).As most of the DEGs in phytohormone signaling were related to the JA and ET pathways,we focused on the expression of these DEGs.The expression of most of the DEGs associated with the JA and ET pathways increased relative to the controlfrom 0.5 to 24 hpi (Fig.4).The number of DEGs related to these two pathways increased with time,and a large portion of the DEGs (six out of eight DEGs for JA and five out of nine DEGs for ET) started to become up-regulated at 6 hpi(Fig.4),which indicates that these two pathways became increasingly activated followingPcinfection.Among these DEGs,the JA synthesis-related genesBra017350(annotated as allene oxide cyclase 3),Bra008037(annotated as lipoxygenase 4),andBra017627(annotated as CBS domain-containing protein CBSX2) were up-regulated from 6 to 24 hpi.Another gene,Bra035320(annotated as allene oxide synthase),was also up-regulated at 24 hpi,while yet another gene,Bra003526(annotated as lipoxygenase),was down-regulated.From 6 to 24 hpi,four out of five ET synthesis-related genes,namelyBra035236(annotated as 1-aminocyclopropane-1-carboxylate synthase 6),

Bra006713(annotated as 1-aminocyclopropane-1-carboxylate oxidase homolog 12-like),Bra027049(annotated as 1-aminocyclopropane-1-carboxylate oxidase),andBra017627(CBS domain-containing protein CBSX2),were up-regulated,while the other ET synthesis-related geneBra015380(annotated as 1-aminocyclopropane-1-carboxylate oxidase 4) was not up-regulated until 24 hpi.

Table 1 Hormone signaling-related differentially expressed genes (DEGs) suggested by MapMan in Chinese cabbage in response to Pectobacterium carotovorum infection

Fig.4 The microarray expression heatmap of the differentially expressed genes (DEGs) related to hormone signaling responses in Chinese cabbage to Pectobacterium carotovorum infection at the indicated time points post-inoculation.* indicates up-or downregulation with |log2(fold change)|>1.ET,ethylene;JA,jasmonic acid;ABA,abscisic acid;hpi,hours post-inoculation.

3.5.Phytohormone production in response to Pc infection

We measured the levels of JA,ET,and SA at different time points afterPcinfection in order to elucidate how their production was affected byPcinfection.The results showed that free JA levels in thePc-infected leaves increased threefold at 12 hpi compared to the water-inoculated control leaves,while no difference was observed between the two treatments at 6 and 24 hpi (P>0.05) (Fig.5-A).Significantly higher ET levels were detected in the infected leaves in comparison to the control at each time interval of either 0–6,6–12 or 12–24 hpi (P<0.05) (Fig.5-B).In contrast,free SA levels in thePc-infected leaves were generally lower than those in the water-inoculated leaves (P<0.05) during the experiment.These results indicated that JA and ET production was induced,while SA production was inhibited,in thePc-infected leaves.

3.6.JA and ET treatments induced resistance in Chinese cabbage in response to Pc infection

Fig.5 Phytohormone production in water-inoculated and Pectobacterium carotovorum-inoculated Chinese cabbage leaves at the indicated time points post-inoculation.A,free jasmonic acid (JA) levels.B,total ethylene (ET) production levels.C,free salicylic acid (SA) levels.Each column shows the mean and standard deviation (n=3).Statistical analyses were conducted by one-way ANOVA among different columns.Different letters indicate significant difference at P<0.05.

The above analyses showed that the production and signaling pathway-related genes of JA and ET were highly induced followingPcinoculation.To further confirm the roles of these two phytohormones in the resistance againstPcinfection in Chinese cabbage,the leaves of Chinese cabbage were treated with single applications or combinations of 1-aminocyclopropane-1-carboxylate (ACC,an ET precursor),JA,and MeJA,and then inoculated withPc.The results showed that the leaves treated with ACC,JA,or MeJA alone or with mixtures of JA+ACC or MeJA+ACC all exhibited lower RDS than the H2O controls (Fig.6),indicating that these hormones contributed to the resistance againstPc.Of these treatments,the JA and JA+ACC treatments produced the lowest RDS (almost full resistance) in the Chinese cabbage afterPcinoculation,indicating a major role for JA in the resistance againstPc.MeJA+ACC produced a lower RDSthan the individual phytohormone treatments with MeJA or ACC (Fig.6),suggesting that the JA signaling pathway works in conjunction with the ET pathway to increase plant resistance againstPc.However,treatments with JA or JA+ACC were associated with similar RDS,possibly because the JA treatments masked the interactive effect of JA+ACC (Fig.6).In addition,the direct antimicrobial effects of either ACC,JA or MeJA on the virulence ofPcwere excluded by an experiment performed on Chinese cabbage petiolesin vitro.It showed the necrotic area caused byPccultured in LB added with ACC,JA or MeJA,respectively,had no difference withPccultured in LB alone (Appendix G).In summary,these results demonstrate that the plant JA and ET signaling pathways can work in synergy to defend againstPcinfection,and furthermore,that JA might play a significant role.

Fig.6 Relative disease severity of Chinese cabbage to Pectobacterium carotovorum infection following treatment with different types of hormones.JA,jasmonic acid;MeJA,methyl jasmonate;ACC,1-aminocyclopropane-1-carboxylate.Each column shows the mean and standard deviation (n=27).Statistical analyses were conducted by one-way ANOVA among different columns.Different letters indicate a significant difference at the P<0.05 level.

4.Discussion

This study aimed to elucidate the defense mechanisms underlying the response of Chinese cabbage toPcinfection.By sequencing the ESTs from SSH cDNA libraries,we were able to ensure that the probes on the cDNA microarray truly represented the genes induced in Chinese cabbage followingPcinfection.The roles of the JA and ET signaling pathways in the defense responses of other plants toPchave been previously assessed using various physiological and biochemical tools (Luzzattoet al.2007a,b).However,this is the first attempt at investigating the molecular basis of this response using a microarray approach in Chinese cabbage.

A central feature of the plant defense signaling network is the cross-talk among different but interrelated signaling pathways (Pieterseet al.2012).Induced defenses against microbial pathogens are regulated by a network of interconnecting signaling pathways in which plant signaling molecules/hormones (SA,JA,and ET) play dominant roles.Our results showed that genes involved in JA-and ETmediated signaling pathways participated in the defense response toPcinfection.In thePc-infected Chinese cabbage,several genes related to the JA and ET signaling pathways,including JA and ET synthesis-related genes,were up-regulated,being accompanied by increasing levels of JA and ET as well as an increased defense response in plants treated with JA,MeJA,ACC or JA/MeJA+ACC.This corroborates previous reports where JA-and ETmediated signaling pathways were shown to be important in plant resistance to the soft rot pathogenP.carotovora(Norman-Setterbladet al.2000;Liuet al.2019).However,the role of SA in plant defense againstP.carotovorahas not previously been established.SA appears to have a minor role or no role in activating defense gene expression in response to CF ofPc(Vidalet al.1998;Norman-Setterbladet al.2000),but inhibited the induction ofPRgenes elicited byP.carotovorain tobacco (Vidalet al.1997).However,some contrasting studies indicate that plant resistance toPccan be induced by SA-mediated defense (Liet al.2004;Kariolaet al.2005;Liuet al.2019).In Chinese cabbage,no DEGs related to SA signaling were identified (Fig.3),and free SA levels all decreased followingPcinfection (Fig.5).Thus,the SA signaling pathway might not participate in the defense response of Chinese cabbage toPcinfection.This is consistent with the decreased SA levels in potato in response toPcinfection (Montesanoet al.2005).As we constructed an SSH cDNA library using mixed samples of different time points,the identified DEGs all constituted induced genes overall,but they may have been up-or down-regulated at different time points.As the SA levels were decreased,the genes related to SA signaling may have been down-regulated overall duringPcinfection and were not identified from the SSH library.This discrepancy could also possibly be attributed to the different strains ofPcor different plants used in this study in comparison to other studies.

Our results showed that both the JA-and MeJAtreatments resulted in lower disease severity in Chinese cabbage in comparison to the water-treated control (Fig.6).Although the concentrations of JA and MeJA used in the present study were the same,MeJA induced a weaker resistance,which was likely caused by the higher volatility of MeJA,as some of the MeJA might have evaporated during the treatment period.Moreover,MeJA needs to be hydrolyzed to JA to be active,and not all of the MeJA could be hydrolyzed to JA.Thus,the effective concentration of MeJA could have been less than that of JA.As there was one DEG,namelyBra017627(predicted:CBS domain-containing protein CBSX2,chloroplastic-like (B.rapa)),that overlapped in JA and ET signaling related to biotic stress in the MapMan analysis(Fig.4;Appendix H),we wanted to determine whether JA and ET could work synergistically in the resistance againstPcinfection in Chinese cabbage.The effects of treatments of JA,MeJA,ACC,or their combinations on the RDS ofPcinfected Chinese cabbage truly suggested that JA/MeJA and ACC could work synergistically to defend againstPcinfection,though the influence of JA might be greater.

In addition to SA,JA,and ET,the auxin and ABA signaling pathways have been reported to be involved in plant defense against pathogens (Sharaf and Farrag 2004;Bruceet al.2007).We also discovered that the DEGs related to auxin and ABA signaling probably participated in the defense response of the infected cabbage (Fig.3).In the bryophytePhyscomitrella patens,the auxin pathway was activated,while ABA levels remained unchanged,in response toP.carotovorumsubsp.carotovoruminfection (Alvarezet al.2016),which contrasts with our results.This disparity might be due to the different subspecies or strains ofPcused in the two studies,as we usedP.carotovorumsubsp.brasiliense.Alternatively,it could bedue to the different host plants.

5.Conclusion

The present study detected 800 DEGs in cabbage at different time points post-Pcinoculation using the SSH library and cDNA microarray method.Among the DEGs,certain genes were related to JA and ET rather than SA signaling.JA and ET production was increased while SA production was decreased in thePc-infected leaves of Chinese cabbage.Treatments with JA,MeJA,the ET precursor ACC,or combinations thereof reduced the disease severity,and the JA and JA+ACC treatments were superior to the others and performed equally well.In summary,these results suggested that JA and ET may act synergistically againstPcinfection in Chinese cabbage,and JA-mediated signaling might be the most significant.

Acknowledgements

We thank Prof.Chen Huajun and Dr.Deng Wenhong of the Beijing Forestry University,China,for help in plant signalcompound measurements,and LetPub for its linguistic assistance during the preparation of this manuscript.This research was funded by the Beijing Leafy Vegetables Innovation Team of Modern Agro-industry Technology Research System,China (BAIC07),and the Beijing Natural Science Foundation,China (5051002).

Declaration of competing interest

The authors declare that they have no conflict of interest.

Appendicesassociated with this paper are available on http://www.ChinaAgriSci.com/V2/En/appendix.htm