Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Patterns of differential gene expression in Brassica napus cultivars infected with Sclerotinia sclerotiorum

Patterns of differential gene expression in Brassica napus cultivars infected with Sclerotinia... INTRODUCTION Sclerotinia sclerotiorum is a necrotrophic plant pathogen that infects important agricultural crops, including canola ( Brassica napus ), soybean ( Glycine max ) and sunflower ( Helianthus annuus ) ( Boland and Hall, 1994 ). In the field, infection of B. napus by S. sclerotiorum occurs mainly through ascospore colonization of senescent flower petals that have lodged on leaves or petioles adjacent to the stem. Necrotic lesions develop on the stems, resulting in premature wilting, stem breakage and lodging of plants with subsequent yield loss. Partial resistance has been reported in semi‐winter‐type B. napus cultivars ( Zhao , 2004 ). Quantitative trait loci (QTLs) associated with partial resistance have been reported in two independent studies and indicate that resistant cultivars may possess different resistance genes ( Zhao and Meng, 2003 ; Zhao , 2006 ); however, the biochemical and genetic bases of resistance to S. sclerotiorum are not well understood. Plants have evolved complex mechanisms to recognize and respond to various pathogens. Plant–pathogen interactions have been investigated using microarray technology in model plant species, such as Arabidopsis thaliana , challenged with several pathogens and pest species ( AbuQamar , 2006 ; Maleck , 2000 ; Schenk , 2000 ; Thilmony , 2006 ). In crop plants, such studies have been confined to gene expression profiling based on expressed sequence tag (EST) analysis: for example, the response of rice to Magnaporthe grisea ( Jantasuriyarat , 2005 ), sorghum to aphid ( Zhu‐Salzman , 2004 ), cotton to Fusarium oxysporum ( Dowd , 2004 ), soybean to Pseudomonas syringae ( Zabala , 2006 ) and wheat to various biotic and abiotic stresses ( Chao , 2006 ). As a result of the high degree of synteny in gene sequences, A. thaliana microarrays have been used to examine gene expression profiles in cultivated Brassica species ( Girke , 2000 ; Lee , 2004 ), including transcriptional changes associated with the response of B. napus cultivars to S. sclerotiorum infection ( Yang , 2007 ; Zhao , 2007 ). These studies have revealed that a common set of genes is induced during the later stages [48–72 h post‐inoculation (hpi)] of the infection in partially resistant and susceptible cultivars. Genes expressed soon after inoculation, within 12 hpi, that might modulate the later defence response have not been examined. An extensive degree of genome duplication in Brassica species ( Parkin , 2005 ) gives rise to expanded gene families which are indistinguishable using A. thaliana microarrays. A clear example of this is the polygalacturonase inhibitor protein (PGIP) gene family in B. napus which contains at least 17 members, many of which respond differently to biotic and abiotic stresses and signalling hormones ( Hegedus , 2008 ). The A. thaliana genome contains only two PGIP genes. Recently, a gene‐specific oligonucleotide microarray representing 15 000 unique genes from B. napus has been developed ( Sharpe , 2006 ; Stasolla , 2008 ). This paper reports the use of this B. napus microarray to examine the patterns of gene expression in a resistant B. napus cultivar, ZhongYou 821 (ZY821), and a susceptible doubled haploid line of B. napus cv. Westar in response to S. sclerotiorum inoculation. Stem tissue, the site of natural infection, was used, and five sampling time points from 6 to 72 hpi were studied, as these were most likely to reveal clues about the mechanisms underlying the resistance. The expression of specific sets of regulatory and defence‐related genes was monitored following challenge with S. sclerotiorum , and differences in the response between resistant and susceptible cultivars were assessed. The expression of genes associated with key signalling and metabolic pathways is discussed. RESULTS AND DISCUSSION Disease development and expression of genes in response to S. sclerotiorum inoculation Natural infection of B. napus stems by S. sclerotiorum was simulated by the inoculation of unwounded plants with mycelial plugs at flowering. Visible lesions developed between 24 and 48 hpi. After 7 days, the lesions on stems of the resistant cultivar ZY821 were smaller than those on the susceptible cultivar Westar, and most cv. Westar plants died within 21 days of inoculation ( Fig. 1 ). 1 Phenotype of Brassica napus cv. ZY821 and Westar, 7 and 21 days after inoculation of stems with Sclerotinia sclerotiorum . To elucidate the mechanisms underlying the differences in the response of these two cultivars, the gene expression profiles of stem tissues were examined in the vicinity of the inoculation site over 3 days with the emphasis on the early stages of infection. Across all time points examined, the expression of 3659 and 4124 genes was altered significantly [analysis of variance ( anova ), false discovery rate (FDR) < 0.0001] in cv. ZY821 and Westar, respectively, by inoculation. In both cultivars, the number of differentially expressed genes increased steadily up to 48 hpi and then declined thereafter, with the greatest change occurring between 12 and 48 hpi ( Fig. 2 ). A complete list of the differentially expressed genes with at least a two‐fold difference in expression across the five time points is provided in Tables S1 and S2 (see Supporting Information). All lists include the ID number of the B. napus oligonucleotides on the microarray and the coordinate for the corresponding A. thaliana orthologue and annotation for the particular gene affected. The microarray data have been deposited in Gene Expression Omnibus (Accession no. GSE13262) and can be accessed at http://www.ncbi.nlm.nih.gov/geo/index.cgi . It should be noted that some differences between the two cultivars examined in this study could potentially be a result of variations in hybridization caused by allelic differences between the two cultivars and B. napus cv. DH12075, which was used to generate the sequences employed to design the oligonucleotides for the array ( Stasolla , 2008 ). In addition, 774 unique S. sclerotiorum cDNAs deposited in G en B ank , representing genes expressed on pectin medium or during infection of B. napus ( Li , 2004a ), were scanned against the 15 000 oligonucleotides on the array. Of these, only eight (CD645785, NADH:ubiquinone oxidoreductase subunit; CD645868, polyubiquitin; CD646053, histone H3; CD646110, ADP–ATP translocase; CD645779, severe depolymerization of actin; CD645744, G‐protein β‐subunit; CD645671, cytochrome C oxidase polypeptide Vib; CD646004, isocitrate dehydrogenase subunit I) exhibited a match better than 1 × 10 −5 and have the potential to cross‐hybridize. 2 Dynamics of differentially expressed genes in Brassica napus cv. ZY821 (squares) or cv. Westar (circles) after infection of stems with Sclerotinia sclerotiorum . The total (broken line) number of differentially expressed genes, as well as genes that were differentially expressed in only one cultivar (full line), are shown. The data presented are the number of differentially expressed genes detected by one‐way analysis of variance ( anova ) from three biological replicates of each cultivar at each time point. The number of genes that exhibited increased or decreased transcript abundance are represented as positive or negative values, respectively, on the y axis. The allotetraploid genome of B. napus , comprising contributions from B. rapa and B. oleracea , contains expanded gene families (paralogues) for which only a single orthologue is present in the A. thaliana genome. The B. napus microarray contains oligonucleotides that differentiate the expanded families corresponding to these orthologous A. thaliana genes. In cv. ZY821 alone, 1158 expanded gene families, containing two to eight members, were found to be differentially expressed 48 h after S. sclerotiorum inoculation (Table S3, see Supporting Information). In 767 gene families, transcripts from only one member were detected, indicating that the other members were either not differentially expressed or their transcript levels were undetectable using the microarray methodology employed. Trends in transcript abundance indicating either an increase or decrease in response to inoculation were observed with members of 369 families, whereas the members of 22 gene families responded differently (differential orthologues; Table S3). These included genes encoding proteins implicated in the defence response, such as chitinase and peroxidase. Verification of microarray data Sixteen genes were selected to examine the reliability of the use of microarray analysis to determine the relative changes in gene expression after S. sclerotiorum inoculation ( Fig. 3 ). In general, quantitative reverse transcriptase‐polymerase chain reaction (qRT‐PCR) analysis showed that the change in gene expression was greater than that indicated by microarray analysis, but the overall trend was consistent. qRT‐PCR analysis was also more sensitive and was able to detect elevated levels of transcripts 6–12 h earlier than microarray analysis, indicating that pathogen recognition and induction of the response occurred very soon after inoculation. The genes selected included five that encoded defence‐associated proteins, including that encoding a lectin whose transcripts increased 24 000‐fold within 48 hpi as determined by qRT‐PCR. Transcripts from genes encoding enzymes involved in jasmonic acid (JA) (BN15608, BnOPR2 ), ethylene (ET) (BN14781, BnSAM3 ; BN13547, BnACS6 ; BN18488, BnACO ) and auxin (BN24462, BnCYP79B2 ; BN15682, BnAA01 ) biosynthesis were detected within 6 hpi, as were those encoding enzymes required for the biosynthesis of glucosinolates (BN22452, BnSOT16 ) and phenylpropanoids (BN22021, BnOPCL ; BN13060, BnCCR ). As discussed below, there were also profound changes in the expression of genes associated with carbon metabolism, which was confirmed by qRT‐PCR analysis. 3 Verification of the expression of selected Brassica napus cv. ZY821 genes by quantitative reverse transcriptase‐polymerase chain reaction (qRT‐PCR) (black bars) and microarray (grey bars) analysis after inoculation of stems with Sclerotinia sclerotiorum . Errors bars for qRT‐PCR analysis show the standard deviation of three technical replicates for each of three biological replicates. Identification of genes induced in the resistant cultivar ZY821 Four genes were selected for qRT‐PCR analysis from those that showed higher transcript abundance in cv. ZY821 at 6–12 hpi compared with cv. Westar based on the microarray data ( Fig. 4 ). The genes encoded BnLOX1 (BN22526), a lipoxygenase (LOX), BnWRKY33 (BN17285), a WRKY‐like transcription factor, BnWRKY40 (BN23912), another WRKY‐like protein, and BnLectin (BN25790), a legume lectin family protein. qRT‐PCR analysis confirmed that the transcript level for each gene was significantly higher in cv. ZY821 than in cv. Westar at these early time points, suggesting that the resistant cultivar responds more quickly to the pathogen. 4 Quantitative reverse transcriptase‐polymerase chain reaction (qRT‐PCR) analysis of four genes that were differentially expressed in Brassica napus cv. ZY821 (white bars) and Westar (grey bars) stems at 6 and 12 h (indicated) after inoculation with Sclerotinia sclerotiorum . The increase in transcript levels was relative to a mock‐inoculated control after normalization to a B. napus reference gene ( Actin ). The error bars show the standard deviation, with means that were statistically different indicated by either **(Student t ‐test, P > 0.05) or *(Student t ‐test, P > 0.1) below each time point. A comparison of the lists of genes with increased transcript abundance in the two cultivars across all time points revealed 41 genes that were expressed only in cv. ZY821 ( Table 1 ). Transcripts from four genes encoding transcription factors were elevated, two of which were induced within 12 hpi. These may be partly responsible for regulating the expression of genes involved in the stress response or metabolism. Two genes encoding annexins (BN18917 and BN15878) were induced, one within 12 hpi and the other by 48 hpi. Annexins bind to membranes in a calcium‐dependent manner, but undergo cellular redistribution in response to stimuli, including environmental stress ( Clark , 2001 ). Some plant annexins have been shown to form calcium channels ( Gorecka , 2007 ) which may be important for the initiation of the signalling cascades needed to induce the expression of defence‐associated genes. Eleven other genes encoded proteins involved in some aspect of the stress response. Most noteworthy was a gene encoding a concanavalin‐like lectin (BN25790) that was induced five‐fold at 6 hpi and 109‐fold by 48 hpi based on microarray analysis, and several 1000‐fold based on qRT‐PCR analysis ( Fig. 3 ). Although such proteins have not been shown to exhibit anti‐mycotic activity, a gene encoding an endo‐β‐1,3‐glucanase was induced within 6 hpi with expression increasing steadily afterwards. A gene encoding BnPGIP1 (BN12243), a member of the expanded PGIP gene family in B. napus that is responsive to S. sclerotiorum infection ( Hegedus , 2008 ), was also induced. PGIPs inactivate polygalacturonases, which are major virulence determinants, and several are produced by S. sclerotiorum ( Li , 2004b ). Genes encoding a multidrug resistance protein (BN27037) and a UDP‐glucosyl transferase (BN22038) involved in secondary metabolite biosynthesis were induced within 12 and 24 hpi, respectively. The expression of these genes implies that a pharmacological battle is being waged between host and pathogen. 1 Classification of genes that were induced in only Brassica napus cv. ZY821 stems after Sclerotinia sclerotiorum infection. Oligo ID * Corresponding A. thaliana locus Description 6 † 12 24 48 72 Transcription BN11472 At3g57390 MADS‐box protein – 4.16 ‡ 6.28 12.8 4.95 BN26939 At2g47460 Myb transcription factor – 2.21 2.17 5.28 5.71 BN27665 At1g28360 ERF domain protein 12 – – 3.05 9.46 7.87 BN26396 At5g57520 Zinc finger – – 3.30 5.37 3.57 Stress BN25790 At3g16530 Lectin 5.18 15.8 50.5 109 108 BN22148 At3g57260 Endo β‐1,3‐glucanase 2.27 3.52 6.11 9.91 7.56 BN18917 At2g38750 Annexin – 3.65 2.66 25.1 33.5 BN27037 At4g22990 Multidrug resistance protein 2 – 2.20 2.57 8.02 9.01 BN22038 At4g34135 UDP‐glucosyl transferase – – 3.95 8.95 3.40 BN16777 At4g15160 Protease inhibitor – – 2.92 3.02 18.4 BN20902 At4g27670 Heat shock protein – – 2.65 7.35 3.71 BN11639 At1g69530 Expansin – – – 7.21 9.46 BN17923 At4g10265 Wound‐responsive protein – – – 6.03 7.79 BN18404 At5g27060 Leucine‐rich repeat – – – 5.58 5.00 BN15878 At5g65020 Annexin – – – 4.26 7.39 BN13492 At2g37640 Expansin – – – 3.90 7.79 BN12243 At5g06860 PGIP1 – – – 3.27 18.9 Metabolism BN22522 At4g27070 Tryptophan synthase 2.65 2.69 4.65 10.6 8.31 BN24404 At4g15417 Ribonuclease III 2.00 2.76 3.33 26.5 14.0 BN23990 At1g05680 UDP‐glucosyl transferase – 2.53 2.95 24.5 9.04 BN16989 At1g48920 Nucleolin – 2.08 4.04 6.67 4.41 BN18865 At1g07720 β‐Ketoacyl‐CoA synthase – – 3.09 28.8 15.7 BN15489 At1g17710 Phosphoric monoester hydrolase – – 3.01 73.0 54.4 BN23164 At4g18950 Ankyrin protein kinase – – 2.10 5.80 4.61 Others BN14638 At1g19770 Purine permease 2.13 2.10 4.59 10.9 10.1 BN20514 At1g76790 O ‐Methyltransferase 2.00 7.87 12.5 25.3 31.1 BN13257 At1g54020 Lipolytic enzyme – 3.15 14.2 – – BN14979 At2g30540 Glutaredoxin – 2.16 4.74 17.4 7.81 BN13468 At4g22620 Auxin‐induced protein – 2.05 2.16 – 8.11 BN26740 At2g31990 Exostosin – 2.04 2.17 10.4 7.10 BN23610 At5g43340 Inorganic phosphate transporter – – 4.66 27.5 14.8 BN20039 At1g15180 Membrane protein family – – 4.37 11.7 12.1 BN18437 At5g15440 Circadian clock coupling factor – – 3.75 13.4 11.5 BN24154 At5g01210 Transferase – – 3.63 5.48 7.55 BN17385 At5g10300 Hydrolase – – 2.14 3.91 7.93 BN26419 At5g17450 Copper homeostasis factor – – – 11.1 9.69 BN19106 At5g01750 Expressed protein – – – 9.41 6.90 BN20931 At5g60760 2‐Phosphoglycerate kinase – – – 7.38 4.21 BN24358 At3g54040 Photoassimilate‐responsive protein – – – 5.07 – BN17022 At3g13857 Expressed protein – – – 2.94 10.5 BN13468 At4g22620 Auxin‐induced protein – – – – 9.38 * Oligonucleotide designation from the B. napus gene‐specific array. † Hours post‐inoculation. ‡ Fold changes relative to mock‐inoculated control. Genes involved in hormone signalling Auxin and gibberellin Genes encoding products involved in aspects of hormonal signal transduction (Table S4, see Supporting Information) were identified in the response of cv. ZY821 to inoculation with S. sclerotiorum . Most plant defence genes are regulated by endogenous plant hormones, such as salicylic acid (SA), JA and ET. An indole‐3‐acetic acid amino synthetase has also been reported to be involved in the activation of basal immunity in rice ( Ding , 2008 ), indicating the role of auxin in plant defence. Two B. napus genes (BN13468 and BN22832), similar to At4g22620 and At5g35735, which are known to be responsive to auxin, were induced in cv. ZY821 at 12 hpi, but not in cv. Westar ( Table 2 ). Two other genes encoding enzymes involved in the biosynthesis of indole‐3‐acetic acid, cytochrome P‐450 CYP79B2 (BN24462) and aldehyde oxidase AA01 (BN15682), were induced at 6 and 12 hpi only in cv. ZY821 (Table S4). Gibberellins (GAs) also regulate plant growth and various developmental processes. Transcripts from a B. napus gene encoding gibberellin 2‐oxidase (BN19546), which is responsible for GA degradation ( Thomas , 1999 ), increased at 12 hpi in cv. ZY821 ( Table 2 ), suggesting that dynamic changes in plant growth and development are occurring. 2 Induction of regulatory genes in Brassica napus cv. ZY821 and Westar stems after Sclerotinia sclerotiorum inoculation. Oligo ID * Corresponding Signalling pathway Description ZY821 Westar A. thaliana locus 6 † 12 6 12 BN13468 At4g22620 Auxin IAA induced – 2.05 – – BN22832 At5g35735 Auxin IAA induced – 2.57 – – BN19546 At1g30040 Gibberellin GA degradation – 2.11 – – BN17492 At1g76650 Calcium Ca 2+ binding – – – 2.14 BN18528 At4g20780 Calcium Ca 2+ binding – 2.81 – 3.16 BN19238 At3g63380 Calcium Ca 2+ transport – 9.93 – 6.03 BN19735 At3g57330 Calcium Ca 2+ transport – 2.08 – 2.06 BN24299 At2g17290 Protein kinase – 2.80 – 2.61 BN10050 At3g45640 MAP kinase MPK3 2.00 – – – BN25533 At1g73500 MAP kinase MPK9 – 2.56 – – BN25534 At1g73500 MAP kinase MPK9 – – – 2.23 BN26459 At4g17500 Transcription ERF1 – 5.01 – 3.57 BN15727 At5g47220 Transcription ERF2 – – – 3.68 BN25670 At5g47230 Transcription ERF5 2.00 2.31 – 2.09 BN25738 At4g17490 Transcription EFR6 – – – 2.17 BN17976 At5g51190 Transcription AP2 2.97 3.11 – – BN11384 At2g46510 Transcription bHLH – – – 2.55 BN14283 At4g14365 Transcription RING finger 2.53 3.13 – 2.22 BN18692 At5g27420 Transcription RING finger – – – 3.96 BN26939 At2g47460 Transcription MYB12 – 2.21 – – BN14274 At3g23250 Transcription MYB15 – 2.71 – – BN22570 At3g46600 Transcription SCL11 – 3.05 – – BN17285 At2g38470 Transcription WRKY33 3.72 5.69 – 4.11 BN23912 At1g80840 Transcription WRKY40 2.98 8.05 – 2.73 BN10329 At1g27730 Transcription ZAT10 2.45 3.94 – – BN10330 At1g27730 Transcription ZAT10 4.41 8.12 2.08 6.40 BN14917 At5g59820 Transcription ZAT12 – 3.20 – – BN14919 At5g59820 Transcription ZAT12 – 5.39 – 2.70 BN22157 At3g21150 Transcription Zinc finger – – – – AP2, APETALA2; bHLH, basic helix–loop–helix; ERF, ethylene response factor; GA, gibberellin; IAA, indole acetic acid; MPK, mitogen activated protein kinase; MYB, myeloblastosis; RING, really interesting new gene; SCL, scarecrow‐like; ZAT, zinc‐finger protein from Arabidopsis thaliana . * Oligonucleotide designation from the B. napus gene‐specific array. † Fold change relative to mock‐inoculated control within each genotype at 6 and 12 hours post‐inoculation (hpi), respectively. Jasmonic acid and ethylene Of the classic defence hormones, JA and ET tend to broadly coordinate the response to mechanical attack from insects and necrotrophic pathogens, whereas SA is most often associated with the response to biotrophic pathogens ( Dong, 1998 ; McDowell and Dangl, 2000 ). This classification is not absolute as these signalling pathways may act individually, synergistically or antagonistically depending on the pathogen involved ( Berrocal‐Lobo , 2002 ; Glazebrook, 2005 ). For example, resistance to A. brassicicola in A. thaliana depends on JA signalling and camalexin production ( Glazebrook, 2005 ), but is also dependent on ET signalling ( Oh , 2005 ). Resistance to B. cinerea depends on genes co‐regulated by JA and ET, but SA, abscisic acid (ABA) and hydrogen peroxide‐dependent defences also contribute to the resistance ( Asselbergh , 2007 ; Glazebrook, 2005 ). JA is synthesized from α‐linolenic acid via four enzymatic steps ( Turner , 2002 ), the first of which yields 13‐hydroperoxylinolenic acid via the action of LOXs. In A. thaliana , LOX2 is primarily responsible for this reaction in response to wounding ( Bell , 1995 ), but five BnLOX genes were induced in cv. Westar within 24 hpi (Table S4). Interestingly, only two BnLOX genes were induced in cv. ZY821; however, one of these was induced more than five‐fold within 6 hpi. Hydroperoxide lyase (HPL) gene expression is also induced by wounding, and this activity can divert some of the 13‐hydroperoxylinolenic acid towards the synthesis of leaf alcohols and aldehydes involved in defence against insect attack ( Bate , 1998 ). In the current study, both cultivars exhibited a strong decrease in the abundance of transcripts for a gene encoding HPL (BN10615), which should serve to increase the pool of 13‐hydroperoxylinolenic acid available for JA synthesis. Allene oxide synthase (AOS) catalyses the formation of 12,13‐epoxylinolenate, which is then converted to 12‐oxophytodienoic acid by allene oxide cyclase (AOC). Similar to the BnLOX genes, three BnAOC genes were induced in cv. Westar within 24 hpi, whereas cv. ZY821 expressed only two, one of which was induced within 12 hpi. The final step in JA biosynthesis is catalysed by 12‐oxophytodienoate reductase (OPR). Both cultivars exhibited elevated levels of three BnOPR genes with similar expression patterns in this study. ET is synthesized by three enzymatic reactions from methionine ( Wang , 2002 ), with the first step being catalysed by S ‐adenosyl methionine synthase (SAM). A sharp decrease in BnSAM1 transcripts was detected in cv. Westar by 48 hpi. In contrast, BnSAM3 was induced in cv. ZY821 within 24 hpi. Two aminocyclopropane carboxylate synthase (ACS) genes were detected, BnACS6 and BnACS8 ; however, only BnACS6 was induced in cv. ZY821 and, once again, earlier than in cv. Westar. The final step in ET biosynthesis is catalysed by aminocyclopropane carboxylate oxidase (ACO). In cv. Westar, three BnACO genes were induced within 24 hpi, but only one in cv. ZY821 with the same expression profile as in cv. Westar. Salicylic acid Recently, Guo and Stotz (2007 ) reported that defence to S. sclerotiorum is dependent on JA, SA and ET signalling in A. thaliana . Likewise, an ABA‐deficient tomato line displayed an SA‐dependent defence response and was less susceptible to S. sclerotiorum and Erwinia chrysanthemi ( Asselbergh , 2007 ). In this study, genes encoding enzymes leading to JA and ET synthesis were rapidly induced on S. sclerotiorum inoculation, as were the genes responsive to them, such as chitinase, β‐1,3‐glucanase and plant defensin 1 (PDF1), whereas transcript abundance from several genes leading to SA biosynthesis decreased. Two orthologues of isochorismate synthase 1 (ICS1) required for the synthesis of SA in A. thaliana ( Wildermuth , 2001 ) were represented on the B. napus microarray (BN15208/BN15209 and BN18712). Although it is recognized that changes in gene expression do not necessarily reflect changes in metabolite levels, transcripts from both genes decreased similarly in cv. Westar and ZY821 (Tables S1 and S2). The coordination of the response to necrotrophic pathogens is obviously complex, and the role of SA and other signalling molecules cannot be discounted, as genes that exclusively respond to them, for example BnPR1 , were induced. Genes encoding transcription factors Rapid and directed induction of gene expression requires both a sensitive surveillance system and a highly responsive transcriptional apparatus. Several genes encoding zinc finger, WRKY, MYB‐like and APETALA2 (AP2)‐like transcription factors exhibited elevated levels of transcripts after inoculation with S. sclerotiorum ( Table 2 ). ET response factors (ERFs) are a class of AP2 domain‐containing transcription factor that recognize the ET responsive element in the promoters of ET‐responsive genes ( Riechmann and Meyerowitz, 1998 ). Transcript levels for BnERF1 (BN26459) were similar between the two cultivars; however, BnERF2 (BN15727) and BnERF6 (BN25738) were elevated in cv. Westar at 12 hpi and BnERF5 (BN25670) was elevated at 6 hpi in cv. ZY821 and in both cultivars at 12 hpi ( Table 2 ). The genes encoding additional AP2 domain, several zinc finger, RING finger, MYB and WRKY transcription factors were found to be induced exclusively, or at higher levels, in cv. ZY821 within 6–12 hpi. AbuQamar (2006 ) identified 30 genes encoding transcription factors that exhibited increased transcript abundance in A. thaliana leaves within 24 h after inoculation with B. cinerea . Functional redundancy precluded the establishment of a readily visible phenotype for most insertion inactivation mutants; however, a few were found to show reduced resistance. These included two zinc finger proteins with ankyrin repeats (C‐X 8 ‐C‐X 5 ‐C‐X 3 ‐H‐type), At3g55980 and At2g40140, which, when mutated, gave rise to plants with enhanced local susceptibility and sensitivity to ABA. In the current study, orthologues of these genes were not detected in either of the B. napus cultivars used, but genes encoding a similar zinc finger protein, BnZAT12 (BN14917 and BN14919), were induced soon after inoculation of cv. ZY821 ( Table 2 ). ZAT12 is required for the expression of ascorbate peroxidase, which provides some measure of resistance to hydrogen peroxide during oxidative stress ( Rizhsky , 2004 ). The genes encoding two WRKY factors similar to A. thaliana WRKY33 (BN17285) and WRKY40 (BN23912) were also rapidly induced in cv. ZY821 ( Table 2 ). Several pieces of evidence attest to the importance of WRKY factors in the mediation of the resistance response. A B. napus WRKY33 orthologue was induced in a partially resistant cultivar on infection by S. sclerotiorum and mapped to a QTL for resistance on N14 ( Zhao , 2007 ). Arabidopsis thaliana wrky33 mutants were more susceptible to B. cinerea and A. brassicicola , which corresponded to a reduction in the expression of the gene encoding the anti‐fungal protein PDF1.2 ( Zheng , 2006 ). Surprisingly, over‐expression of AtWRKY33 in A. thaliana enhanced resistance to necrotrophic fungal pathogens, but increased susceptibility to P. syringae , leading to the suggestion that WRKY33 might serve as a positive regulator of JA/ET‐responsive genes and a repressor of SA‐responsive genes ( Zheng , 2006 ). Genes encoding components of signal transduction pathways Genes coding for products that function upstream of transcription initiation ( Table 2 ) were identified. One gene (BN17492) encoding a calcium‐binding protein was induced in cv. Westar but not in cv. ZY821, whereas the expression of other genes (BN18528, BN19238 and BN19735) involved in calcium signalling was similar. Protein phosphorylation and dephosphorylation were implicated in the mediation of the response to the necrotrophic fungus Phoma macdonaldii in sunflower ( Alignan , 2006 ). Several genes encoding protein phosphatase 2C (Tables S1 and S2) were induced after exposure to S. sclerotiorum in both cultivars. These enzymes are negative regulators of mitogen‐activated protein kinase (MAPK) signalling pathways ( Rodriguez, 1998 ), including those leading to programmed cell death (PCD) ( Tamura , 2006 ), but are also produced after treatment with the necrosis‐inducing protein, Nep1, concomitant with a calmodulin‐like protein ( Keates , 2003 ). MAPK cascades are also required for the induction of defence responses to biotic and abiotic stresses. The inactivation of MPK6 and MPK4 by the protein phosphatase, AP2C1, in A. thaliana reduced innate immunity and increased susceptibility to B. cinerea ( Schweighofer , 2007 ). Two MAPK genes, BnMPK3 (BN10050) and BnMPK9 (BN25533/BN25334), were induced at 6 and 12 hpi, respectively, in cv. ZY821 ( Table 2 ); microarray studies in A. thaliana have shown that AtMPK3 is responsive to chitin oligomers. Defence‐related genes and pathways Of the genes that were induced within 6–12 hpi in both cultivars, those encoding pathogenesis‐related (PR) proteins and proteins involved in the response to biotic and abiotic stress were prominent ( Table 3 ). Several of these are known to possess anti‐fungal activity, such as chitinase and β‐1,3‐glucanase, but they may have only a limited role in providing resistance to S. sclerotiorum , as they were also induced in the susceptible cultivar. Plant lectins are members of a large group of carbohydrate‐binding proteins ( Peumans and Van Damme, 1998 ), some of which are induced by chitin ( Ramonell , 2005 ) and exhibit potent antimicrobial activity ( Koo , 2002 ). In this study, genes encoding concanavalin A‐like (BN25790) and curculin‐like mannose‐binding lectins (BN24426 and BN24464) were induced at 6–12 hpi, although these lectins have not been reported to show antimicrobial activity. As indicated previously, many genes were members of expanded families, with some members being commonly expressed in the two cultivars and a subset that was exclusive to each cultivar. This feature was most clearly evident in the expression pattern of a group of 11 glutathione‐ S ‐transferase (GST) genes, 10 of which were exclusive to either cv. ZY821 or cv. Westar within 6–12 hpi. 3 Induction of defence‐related genes in Brassica napus cv. ZY821 and Westar stems after Sclerotinia sclerotiorum inoculation. Oligo ID * A. thaliana Description ZY821 Westar locus 6 † 12 6 12 BN14925 At3g57240 β‐Glucanase 2.10 6.42 – 9.52 BN14926 At3g57240 β‐Glucanase – – 8.54 – BN15009 At2g43570 Chitinase 3.66 10.1 2.41 10.4 BN24266 At2g43590 Chitinase 7.21 10.9 10.1 20.1 BN24268 At2g43590 Chitinase – 3.93 – – BN11310 At3g12500 Chitinase – – – 2.35 BN10008 At2g29450 GST – 2.78 – – BN14800 At1g69930 GST – – – 3.00 BN15797 At4g02520 GST 2.11 6.60 2.21 – BN15802 At4g02520 GST – – – 3.82 BN16623 At2g02930 GST – – – 2.59 BN19123 At1g17170 GST – – – 6.20 BN19124 At1g17170 GST – 5.95 – – BN24382 At1g02920 GST – 2.96 – – BN25567 At1g27130 GST – 2.58 – – BN25568 At1g27130 GST – 3.47 – – BN26558 At2g29460 GST – 4.12 – – BN19079 At1g78820 Lectin – – 2.34 BN24426 At1g78860 Lectin – 6.31 – 4.84 BN24464 At1g78850 Lectin – 14.8 – 6.13 BN25790 At3g16530 Lectin 5.18 15.8 – – BN25791 At3g16530 Lectin – – 2.74 17.0 BN24866 At4g11650 Osmotin – 5.26 – – BN14490 At1g75830 PDF1.1 – – 2.25 6.18 BN14537 At2g26020 PDF1.2b – – – 3.15 BN11341 At2g38390 Peroxidase – – – 2.93 BN23917 At4g08780 Peroxidase 2.56 8.08 – – BN19400 At3g49110 Peroxidase 33 2.17 – – – BN24675 At3g49110 Peroxidase 33 – – – 2.65 BN15231 At4g37520 Peroxidase 50 – 2.88 – – BN12244 At5g06860 PGIP1 – 2.34 – 2.17 BN14604 At2g24810 Thaumatin 2.77 9.63 – 4.55 BN14605 At2g24810 Thaumatin – 3.60 – – BN23765 At4g33720 PR1 – 2.08 – – BN25356 At2g14610 PR1 – – – 7.14 BN25095 At3g11820 SYP121 – – – 2.13 BN24260 At3g52400 SYP122 – – – 2.44 GST, glutathione‐ S ‐transferase; PDF, plant defensin; PGIP, polygalacturonase inhibitor protein; PR1, pathogenesis‐related protein 1; SYP, syntaxin of plant. * Oligonucleotide designation from the B. napus gene‐specific array. † Fold change in inoculated stems vs. control 6 and 12 h post‐inoculation (hpi). Although PCD resulting from the release and accumulation of reactive oxygen species (ROS), such as superoxide and hydrogen peroxide, is well suited for containing biotrophic pathogens, the accumulation of ROS is directly proportional to the degree of infection for necrotrophs. Interestingly, transcript levels from two syntaxin genes, BnSYP121 and BnSYP122 (BN25095 and BN24260), which have been implicated in the suppression of PCD pathways ( Zhang , 2007 ), were elevated ( Table 3 ). Osmotin‐like PR proteins may exhibit antimicrobial activity ( Shatters , 2006 ), and one osmotin gene (BN24866) was induced early in the resistant cultivar and later in the susceptible cultivar. Osmotins accumulate normally in the cell apoplast during cold acclimation and provide general cryo‐ and osmotic protection. One report has indicated that osmotin, in conjunction with chitinase, forms a complex with actin, causing microtubule disruption and cytoplasmic aggregation ( Takemoto , 1997 ). Cytoskeletal reorganization is required for translocation of the nucleus to the site of pathogen interaction, a process that is thought to be important for rapid and targeted deployment of early defences. D'Angeli and Altamura (2007 ) reported that osmotin regulated the onset of PCD by blocking the Ca 2+ signalling necessary to initiate actin micro‐filament depolymerization. ROS are generated by normal mitochondrial electron transport mediated by the cytochrome c pathway and through the actions of oxidases ( Amirsadeghi , 2006 ). Several alternative oxidases are capable of serving as the terminal oxidase, accepting electrons from ubiquinone and diverting electron flow from the cytochrome c pathway ( Finnegan , 2004 ). In the current study, transcript abundance from a gene encoding an alternative oxidase (BN18515) increased early in the infection process in both cultivars (Tables S1 and S2), which should alter the rate of ROS generation. The relationship between alternative oxidase, ROS and PCD is not entirely clear as transgenic tobacco lines over‐expressing alternative oxidase exhibit smaller lesions ( Ordog , 2002 ), whereas wheat mutants with accentuated lesions show elevated alternative oxidase levels ( Sugie , 2007 ). Genes involved in secondary metabolism Certain secondary metabolites may play a role in plant defence. Altered patterns of expression were observed for genes associated with biochemical pathways leading to the biosynthesis of phenylpropanoids and glucosinolates in this study. The phenylpropanoid pathway is responsible for the synthesis of a wide variety of secondary metabolites, including lignins, flavonoids, phytoalexins and antioxidants, which play a role in plant defence ( Dixon and Paiva, 1995 ). Of the 11 differentially expressed genes associated with the phenylpropanoid biosynthetic pathway, most exhibited similar expression patterns in both cv. ZY821 and Westar (Table S4). Transcripts from a gene encoding a 4‐coumarate‐CoA ligase (BN22021), responsible for the formation of coumaroyl‐CoA, were elevated. Coumaroyl‐CoA is directed to several pathways, one of which leads to the formation of coniferyl aldehyde. The final step in this pathway is catalysed by cinnamoyl‐CoA reductase, and transcript abundance from three genes encoding this enzyme (BN23842, BN15787 and BN13060) was elevated in both cultivars. The fate of coniferyl aldehyde is less clear, as genes encoding alcohol dehydrogenases that would result in the formation of coniferyl alcohol were induced in cv. Westar (BN17669), but repressed in cv. ZY821 (BN22074). In contrast, transcript abundance from a gene that encodes a ferulate hydroxylase (BN18532), catalysing the first step in the pathway to ferulate and sinapate, which are precursors for lignin biosynthesis, decreased in cv. Westar. Glucosinolates are plant secondary metabolites found in the Brassicaceae family and can be grouped into three classes, aliphatic, indolyl and aromatic glucosinolates, according to the amino acid precursor used for their synthesis. Tryptophan biosynthesis could be accelerated, as transcript levels from genes encoding enzymes for chorismate biosynthesis and for four of the five biochemical reactions converting tryptophan to indolylmethyl‐glucosinolate, increased sharply in both cultivars. Indolylmethyl‐glucosinolate might be converted into different indolyl glucosinolate derivatives. Changes in the expression of genes associated with carbon metabolism The expression of genes encoding key enzymes involved in carbohydrate utilization and energy production [glycolysis, tricarboxylic acid (TCA) and Calvin cycles] was altered in both cultivars (Table S4), suggesting that changes in plant metabolism may occur following inoculation with S. sclerotiorum . A sharp decrease was detected in transcript abundance for five genes encoding glyceraldehyde‐3‐phosphate (G3P) dehydrogenase (BN11845, BN14336, BN24802, BN16865 and BN18780), which is responsible for the generation of G3P from 1,3‐diphosphoglycerate, the key intermediate required in carbon fixation via the Calvin cycle. Two genes further downstream in the pathway, encoding transketolase (BN25656) and ribose‐5‐phosphate isomerase (BN17967), were induced, and may serve to scavenge available G3P and other Calvin cycle intermediates by converting them to ribulose‐1,5‐biphosphate. This compound could be directed to the photorespiration pathway, where expression of a gene encoding glycolate oxidase (BN13732) was induced by as much as 50‐fold. Glycolate oxidase converts glycolate to glyoxylate and H 2 O 2 , the latter being an important component of the defence response. Transcript abundance from genes involved in the photorespiration pathways downstream of glyoxylate decreased, suggesting that glyoxylate is metabolized via a different mechanism. Glyoxylate may enter the TCA pathway by way of the glyoxylate shunt after conversion to malate. Transcript abundance from malate dehydrogenase (BN16831) and citrate synthase (BN24865 and BN15521) genes increased, although transcript abundance from two isocitrate lyase genes (BN26052 and BN16663) decreased. Many of the altered gene expression patterns appear to be directed towards increasing carbon flow to and accelerating flux through the TCA cycle. Multiple genes encoding each of the enzymes required to convert pyruvate to acetyl‐CoA (such as dihydrolipoyl dehydrogenase, dihydrolipoyl lysine‐residue acetyltransferase, pyruvate decarboxylase and pyruvate dehydrogenase) were induced, as were genes for several key steps in the TCA cycle. If carbon fixation via the Calvin cycle is impeded, the plant can access carbon reserves to drive the increased flux through the TCA cycle. A potential source of TCA cycle intermediates is the degradation of fatty acids by the β‐oxidation pathway. In this study, genes encoding enzymes that convert acyl‐CoA to acetyl‐CoA, such as acyl‐CoA oxidase (BN16426 and BN22545), enoyl‐CoA hydratase (BN13325) and 3‐ketoacyl‐CoA thiolase (BN11605), were induced; however, transcript abundance from several genes encoding long‐chain fatty acid‐CoA ligases (BN22268, BN13971 and BN20014), responsible for the initial stage of degradation, decreased (Table S4). Sucrose serves as a storage compound and can be transported throughout the plant. It can be converted into UDP‐glucose and fructose by sucrose synthase, or to glucose and fructose by invertase. Sucrose degradation via the latter pathway appears to be accelerated, as indicated by the induction of four genes encoding invertases (BN10428, BN18301, BN19126 and BN18327), whereas the expression of genes encoding enzymes involved in the alternative pathway, namely sucrose synthase (BN12649) and phosphoglucomutase (BN21073), was reduced in this study. In addition, the genes encoding hexokinase (BN25532) and glucose‐6‐phosphate isomerase (BN11549) were also induced, suggesting that a large pool of fructose‐6‐phosphate (F6P) is formed. F6P generated from sucrose degradation is converted to pyruvate by the glycolysis I pathway in the plant cytosol. A gene encoding a key enzyme in this pathway, phosphofructokinase (BN19519), which converts F6P into fructose‐1,6‐diphosphate, was induced in both cultivars. F6P can also be formed from starch degradation in plastids and is converted to pyruvate by the glycolysis II pathway; however, the means by which the carbon in F6P enters the TCA cycle does not appear to be so direct. Similar to that observed for the Calvin pathway, the generation of 3‐phosphoglycerate from G3P appears to be blocked by the repression genes encoding G3P dehydrogenases. It seems more likely that G3P is directed to the photorespiration pathway, resulting in the formation of glyoxylate and H 2 O 2 , with the glyoxylate entering the TCA cycle via the glyoxylate shunt. CONCLUSIONS In this study, a B. napus oligonucleotide microarray representing 15 000 unigenes was used to study the response to inoculation with S. sclerotiorum of stems in resistant and susceptible oilseed rape cultivars. A large number of genes were found to be differentially regulated after infection. The gene expression profiles during the latter stages of stem infection were similar to those of B. napus seedlings in response to S. sclerotiorum infection ( Yang , 2007 ; Zhao , 2007 ) and other necrotrophic pathosystems, namely Botrytis cinerea—A. thaliana ( AbuQamar , 2006 ) and Alternaria brassicicola—Brassica oleracea ( Cramer , 2006 ). Sampling was biased towards the very early stages of the infection, as it was hypothesized that events occurring soon after the initial interaction between pathogen and host were critical for resistance to aggressive necrotrophs, such as S. sclerotiorum . Infected stems were examined as these are the natural site for infection in this pathosystem. The resistant cultivar ZY821 effected the induction of defence‐associated genes earlier than the susceptible cultivar Westar. Differences were also observed in the patterns of a large group of regulatory genes, as well as genes involved in plant hormone synthesis and aspects of the host defence mechanisms against pathogens. Changes in the expression of genes involved in carbon metabolism suggest that carbon storage reserves (such as sucrose, starch and lipid) are accessed and shuttled through the photorespiration pathway. This pathway leads to the formation of glyoxylate, which can enter the TCA cycle, as well as hydrogen peroxide, which may form a part of the defence response ( Huckelhoven , 1999 ). The activity of the TCA cycle may be greatly increased, as evidenced by the induction of many genes encoding TCA cycle enzymes. This not only generates reducing power and energy, but also precursors for amino acid biosynthesis. A sharp increase in transcript abundance for genes encoding enzymes for tryptophan biosynthesis was observed, this amino acid being a precursor for both glucosinolates and phenylpropanoids. This study has provided new insights into aspects of the defence response to necrotrophic pathogens using an experimental design that closely resembles the actual circumstances occurring in the field. Several targets are being analysed to determine their contribution to the resistance in B. napus cv. ZY821 and other resistant cultivars. EXPERIMENTAL PROCEDURES Plant material, inoculation and tissue harvest The semi‐winter‐type B. napus cv. ZY821 and a spring‐type doubled haploid line (N‐o‐1) derived from the cultivar Westar (hereafter referred to simply as cv. Westar) were used. Brasicca napus cv. ZY821 is partially resistant to S. sclerotiorum ( Zhou , 1994 ), but cv. Westar is susceptible. Seeds of cv. ZY821 were sown in a Redi‐Earth soil mixture (Scotts Fertilizer Company) amended with coconut fibre to improve drainage and Osmocote 15‐7‐12 (Scotts Fertilizer Company, Marysville, OH, USA) for the controlled release of fertilizer. The plants were maintained in a growth chamber under 200 µmol/m 2 /s light with a 16‐h photoperiod provided by fluorescent and incandescent light. The temperature was maintained at 20 °C in the light and 16 °C in the dark. Two seeds were sown in each of 36 cells, and seedlings were thinned to one plant per cell 11 days after sowing. Four‐week‐old seedlings of the winter type were vernalized in a cold room at 5–8 °C for 4 weeks with an 8‐h photoperiod and half of the previous light intensity. After vernalization, each plant was transplanted to a 6‐in (15‐cm) pot and returned to the growth chamber. Seeds of cv. Westar were sown directly in 6‐in pots and plants were grown under the same conditions as for cv. ZY821, except for the vernalization treatment. Plants were selected for inoculation and sampling using a randomized design with three biological replicates for each cultivar. Each replicate consisted of 60 plants for five time points (6, 12, 24, 48 and 72 hpi) and two treatments (inoculated and mock‐inoculated controls). An S. sclerotiorum isolate (321; Kohli and Kohn, 1995 ), obtained originally from field‐sown canola plants, was used throughout this study. Plate cultures were maintained on minimal medium according to Li (2004a ). When the plants were at the full flowering stage, three sites on the main stem were inoculated at every other internode with 8‐mm plugs cut from the growing margin of a plate colony cultured on minimal salts‐glucose agar (20.0 g glucose, 2.0 g NH 4 NO 3 , 1.0 g KH 2 PO 4 , 1.0 g NaOH, 0.1 g MgSO 4 .7H 2 O, 0.1 g malic acid and 20.0 g agar per 1000 mL distilled water) modified from Cruickshank (1983 ). Mock‐inoculated plants were treated with an agar plug only. Plugs were secured to the stem with Parafilm™. Each line yielded 180 inoculation sites per biological replicate, or 540 sites in total. Epidermal stem tissues extending 10 mm beyond the inoculation site and 1 mm deep were excised at 6, 12, 24, 48 and 72 hpi with a razor blade. The tissues harvested from one biological replicate at each time point (six individual plants comprising 18 inoculation sites) were pooled as one sample. Harvested tissues were frozen immediately in liquid nitrogen and stored at −80 °C. Probe preparation and slide hybridization Total RNA was isolated from 1 g of ground tissue using the plant RNeasy kit (Qiagen, Mississauga, Ontario, Canada) and stored at −80 °C. RNase‐free DNase was applied during RNA isolation to remove any DNA contamination (Qiagen). The concentration of total RNA was determined using a Nanodrop spectrophotometer (Nanodrop Technologies, Montchanin, DE, USA) and verified by agarose gel electrophoresis. Probe preparation, hybridization and washing were performed using the methods of Stasolla (2008 ). Briefly, the Amino Allyl MessageAmp II aRNA amplification kit was used for cDNA synthesis and RNA amplification (Ambion, Austin, TX, USA) with 1 µg of total RNA, following the manufacturer's instructions. In vitro transcription to prepare amplified RNA (aRNA) was conducted at 37 °C for 14 h and then purified. aRNA was quantified using a Nanodrop spectrophotometer, and 5 µg were coupled with either cyanine‐3 (Cy3) or cyanine‐5 (Cy5) dye. The coupling reaction mixture was incubated at 20 °C for 30 min, quenched with 4 m hydroxylamine for 15 min and purified. Cy3/Cy5‐labelled aRNA was dried completely in a vacuum, dissolved in 9 µL double‐distilled H 2 O and fragmented by the addition of 1 µL of fragmentation reagent (Ambion, catalouge #8740), followed by incubation at 70 °C for 15 min. The reaction was stopped with 1 µL of stop solution and 47 µL of hybridization solution [25% formamide, 5 × standard saline citrate (SSC), 0.1% sodium dodecylsulphate (SDS), 0.1 mg/mL sheared salmon sperm DNA] was added. Hybridizations were performed at 37 °C for 16 h in the dark, after which the slides were washed in 2 × SSC, 0.1% SDS at 37 °C for 5 min, followed by two washes in 1 × SSC for 2 min each and two washes in 0.1 × SSC for 1 min each. The washed slides were dried by centrifugation at 225 g for 1 min at 20 °C. Microarray experimental design and data analysis Dye swap hybridizations were used to compare changes in gene expression between inoculated and mock‐inoculated samples at each time point (from 6 to 72 hpi) for each of the two cultivars. Each dye swap experiment comprised two hybridizations: the first consisted of two samples labelled with either Cy3‐dCTP or Cy5‐dCTP and hybridized onto a single slide, and the second with the same two samples labelled in reverse. Thus, the experimental design included five time points, two cultivars and three biological replicates, for a total of 60 slides. The slides were scanned using a Virtek Chip Reader (Bio‐Rad Laboratories, Hercules, CA, USA). For each scan, the signal intensity of the two channels was initially adjusted on the basis of the control spots so that an average red/green signal ratio of approximately unity was achieved and the spots were not saturated. Spot intensities from scanned slides were quantified using ArrayPro 4.0 (Bio‐Rad Laboratories) and the data were imported into the BioArray Software Environment (BASE version 1.0). Spots with intensities lower than the background were eliminated. Gene Spring 7.3 (Agilent Technologies, Inc., Mississauga, Ontario, Canada) was used to select spots with intensity values ≥50 in at least half of the samples (slides) for further analysis. Normalization was conducted using the per‐spot and per‐chip Lowess normalization procedure with a smooth factor of 20. A one‐way anova with a parametric test and variances not assumed to be equal was performed ( Craig , 2003 ). A multiple test correction ( Benjamini and Hochberg, 1995 ) was applied and an FDR of ≤0.0001 was used. A list of genes for which the expression varied by more than 2.0‐fold was generated. qRT‐PCR analysis qRT‐PCR analysis was conducted to verify some of the microarray data. Single‐stranded cDNA was synthesized from 4 µg of total RNA using the Superscript First‐Strand Synthesis System (Invitrogen, Burlington, ON, Canada); 6 ng of total cDNA was used in each RT‐PCR reaction. qRT‐PCR was performed in a DNA Engine Opticon 2 (MJ Research, Watham, MA, USA) system with the SYBR Green qPCR kit (Finnzymes Oy, Espoo, Finland) in a reaction volume of 20 µL containing 0.5 µ m of each primer. The PCR conditions were as follows: 95 °C for 15 min, followed by 45 cycles of 94 °C for 10 s, 60 °C for 30 s and 72 °C for 30 s. A melting curve of denatured double‐stranded cDNA was established to test the purity of the products. Data were obtained from each biological replicate and repeated in triplicate. The PCR primer amplification efficiency was tested using a six dilution series of the template, according to Pfaffl (2001 ). Expression ratios in the treatments relative to the mock‐inoculated controls were calculated using data corrected for PCR efficiency and normalized against the product from the B. napus Actin gene. Primers were designed using PRIMERTM ( http://www.workbench.sdsc.edu ) based on the homologues in B. napus DH12075, and the product size was limited to 100–300 nucleotides (Table S5, see Supporting Information). ACKNOWLEDGEMENTS This work was funded by Saskatchewan Agriculture and Food, the Saskatchewan Canola Development Commission and the Federal Matching Investments Initiative. We thank Branimir Gjetvaj for technical assistance. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Molecular Plant Pathology Wiley

Patterns of differential gene expression in Brassica napus cultivars infected with Sclerotinia sclerotiorum

Download PDF
15 pages

Loading next page...
 
/lp/wiley/patterns-of-differential-gene-expression-in-brassica-napus-cultivars-JGbXIC4SYg

References (73)

Publisher
Wiley
Copyright
Journal compilation © 2009 Blackwell Publishing Ltd. No Claim to Original US Government Works
ISSN
1464-6722
eISSN
1364-3703
DOI
10.1111/j.1364-3703.2009.00558.x
pmid
19694954
Publisher site
See Article on Publisher Site

Abstract

INTRODUCTION Sclerotinia sclerotiorum is a necrotrophic plant pathogen that infects important agricultural crops, including canola ( Brassica napus ), soybean ( Glycine max ) and sunflower ( Helianthus annuus ) ( Boland and Hall, 1994 ). In the field, infection of B. napus by S. sclerotiorum occurs mainly through ascospore colonization of senescent flower petals that have lodged on leaves or petioles adjacent to the stem. Necrotic lesions develop on the stems, resulting in premature wilting, stem breakage and lodging of plants with subsequent yield loss. Partial resistance has been reported in semi‐winter‐type B. napus cultivars ( Zhao , 2004 ). Quantitative trait loci (QTLs) associated with partial resistance have been reported in two independent studies and indicate that resistant cultivars may possess different resistance genes ( Zhao and Meng, 2003 ; Zhao , 2006 ); however, the biochemical and genetic bases of resistance to S. sclerotiorum are not well understood. Plants have evolved complex mechanisms to recognize and respond to various pathogens. Plant–pathogen interactions have been investigated using microarray technology in model plant species, such as Arabidopsis thaliana , challenged with several pathogens and pest species ( AbuQamar , 2006 ; Maleck , 2000 ; Schenk , 2000 ; Thilmony , 2006 ). In crop plants, such studies have been confined to gene expression profiling based on expressed sequence tag (EST) analysis: for example, the response of rice to Magnaporthe grisea ( Jantasuriyarat , 2005 ), sorghum to aphid ( Zhu‐Salzman , 2004 ), cotton to Fusarium oxysporum ( Dowd , 2004 ), soybean to Pseudomonas syringae ( Zabala , 2006 ) and wheat to various biotic and abiotic stresses ( Chao , 2006 ). As a result of the high degree of synteny in gene sequences, A. thaliana microarrays have been used to examine gene expression profiles in cultivated Brassica species ( Girke , 2000 ; Lee , 2004 ), including transcriptional changes associated with the response of B. napus cultivars to S. sclerotiorum infection ( Yang , 2007 ; Zhao , 2007 ). These studies have revealed that a common set of genes is induced during the later stages [48–72 h post‐inoculation (hpi)] of the infection in partially resistant and susceptible cultivars. Genes expressed soon after inoculation, within 12 hpi, that might modulate the later defence response have not been examined. An extensive degree of genome duplication in Brassica species ( Parkin , 2005 ) gives rise to expanded gene families which are indistinguishable using A. thaliana microarrays. A clear example of this is the polygalacturonase inhibitor protein (PGIP) gene family in B. napus which contains at least 17 members, many of which respond differently to biotic and abiotic stresses and signalling hormones ( Hegedus , 2008 ). The A. thaliana genome contains only two PGIP genes. Recently, a gene‐specific oligonucleotide microarray representing 15 000 unique genes from B. napus has been developed ( Sharpe , 2006 ; Stasolla , 2008 ). This paper reports the use of this B. napus microarray to examine the patterns of gene expression in a resistant B. napus cultivar, ZhongYou 821 (ZY821), and a susceptible doubled haploid line of B. napus cv. Westar in response to S. sclerotiorum inoculation. Stem tissue, the site of natural infection, was used, and five sampling time points from 6 to 72 hpi were studied, as these were most likely to reveal clues about the mechanisms underlying the resistance. The expression of specific sets of regulatory and defence‐related genes was monitored following challenge with S. sclerotiorum , and differences in the response between resistant and susceptible cultivars were assessed. The expression of genes associated with key signalling and metabolic pathways is discussed. RESULTS AND DISCUSSION Disease development and expression of genes in response to S. sclerotiorum inoculation Natural infection of B. napus stems by S. sclerotiorum was simulated by the inoculation of unwounded plants with mycelial plugs at flowering. Visible lesions developed between 24 and 48 hpi. After 7 days, the lesions on stems of the resistant cultivar ZY821 were smaller than those on the susceptible cultivar Westar, and most cv. Westar plants died within 21 days of inoculation ( Fig. 1 ). 1 Phenotype of Brassica napus cv. ZY821 and Westar, 7 and 21 days after inoculation of stems with Sclerotinia sclerotiorum . To elucidate the mechanisms underlying the differences in the response of these two cultivars, the gene expression profiles of stem tissues were examined in the vicinity of the inoculation site over 3 days with the emphasis on the early stages of infection. Across all time points examined, the expression of 3659 and 4124 genes was altered significantly [analysis of variance ( anova ), false discovery rate (FDR) < 0.0001] in cv. ZY821 and Westar, respectively, by inoculation. In both cultivars, the number of differentially expressed genes increased steadily up to 48 hpi and then declined thereafter, with the greatest change occurring between 12 and 48 hpi ( Fig. 2 ). A complete list of the differentially expressed genes with at least a two‐fold difference in expression across the five time points is provided in Tables S1 and S2 (see Supporting Information). All lists include the ID number of the B. napus oligonucleotides on the microarray and the coordinate for the corresponding A. thaliana orthologue and annotation for the particular gene affected. The microarray data have been deposited in Gene Expression Omnibus (Accession no. GSE13262) and can be accessed at http://www.ncbi.nlm.nih.gov/geo/index.cgi . It should be noted that some differences between the two cultivars examined in this study could potentially be a result of variations in hybridization caused by allelic differences between the two cultivars and B. napus cv. DH12075, which was used to generate the sequences employed to design the oligonucleotides for the array ( Stasolla , 2008 ). In addition, 774 unique S. sclerotiorum cDNAs deposited in G en B ank , representing genes expressed on pectin medium or during infection of B. napus ( Li , 2004a ), were scanned against the 15 000 oligonucleotides on the array. Of these, only eight (CD645785, NADH:ubiquinone oxidoreductase subunit; CD645868, polyubiquitin; CD646053, histone H3; CD646110, ADP–ATP translocase; CD645779, severe depolymerization of actin; CD645744, G‐protein β‐subunit; CD645671, cytochrome C oxidase polypeptide Vib; CD646004, isocitrate dehydrogenase subunit I) exhibited a match better than 1 × 10 −5 and have the potential to cross‐hybridize. 2 Dynamics of differentially expressed genes in Brassica napus cv. ZY821 (squares) or cv. Westar (circles) after infection of stems with Sclerotinia sclerotiorum . The total (broken line) number of differentially expressed genes, as well as genes that were differentially expressed in only one cultivar (full line), are shown. The data presented are the number of differentially expressed genes detected by one‐way analysis of variance ( anova ) from three biological replicates of each cultivar at each time point. The number of genes that exhibited increased or decreased transcript abundance are represented as positive or negative values, respectively, on the y axis. The allotetraploid genome of B. napus , comprising contributions from B. rapa and B. oleracea , contains expanded gene families (paralogues) for which only a single orthologue is present in the A. thaliana genome. The B. napus microarray contains oligonucleotides that differentiate the expanded families corresponding to these orthologous A. thaliana genes. In cv. ZY821 alone, 1158 expanded gene families, containing two to eight members, were found to be differentially expressed 48 h after S. sclerotiorum inoculation (Table S3, see Supporting Information). In 767 gene families, transcripts from only one member were detected, indicating that the other members were either not differentially expressed or their transcript levels were undetectable using the microarray methodology employed. Trends in transcript abundance indicating either an increase or decrease in response to inoculation were observed with members of 369 families, whereas the members of 22 gene families responded differently (differential orthologues; Table S3). These included genes encoding proteins implicated in the defence response, such as chitinase and peroxidase. Verification of microarray data Sixteen genes were selected to examine the reliability of the use of microarray analysis to determine the relative changes in gene expression after S. sclerotiorum inoculation ( Fig. 3 ). In general, quantitative reverse transcriptase‐polymerase chain reaction (qRT‐PCR) analysis showed that the change in gene expression was greater than that indicated by microarray analysis, but the overall trend was consistent. qRT‐PCR analysis was also more sensitive and was able to detect elevated levels of transcripts 6–12 h earlier than microarray analysis, indicating that pathogen recognition and induction of the response occurred very soon after inoculation. The genes selected included five that encoded defence‐associated proteins, including that encoding a lectin whose transcripts increased 24 000‐fold within 48 hpi as determined by qRT‐PCR. Transcripts from genes encoding enzymes involved in jasmonic acid (JA) (BN15608, BnOPR2 ), ethylene (ET) (BN14781, BnSAM3 ; BN13547, BnACS6 ; BN18488, BnACO ) and auxin (BN24462, BnCYP79B2 ; BN15682, BnAA01 ) biosynthesis were detected within 6 hpi, as were those encoding enzymes required for the biosynthesis of glucosinolates (BN22452, BnSOT16 ) and phenylpropanoids (BN22021, BnOPCL ; BN13060, BnCCR ). As discussed below, there were also profound changes in the expression of genes associated with carbon metabolism, which was confirmed by qRT‐PCR analysis. 3 Verification of the expression of selected Brassica napus cv. ZY821 genes by quantitative reverse transcriptase‐polymerase chain reaction (qRT‐PCR) (black bars) and microarray (grey bars) analysis after inoculation of stems with Sclerotinia sclerotiorum . Errors bars for qRT‐PCR analysis show the standard deviation of three technical replicates for each of three biological replicates. Identification of genes induced in the resistant cultivar ZY821 Four genes were selected for qRT‐PCR analysis from those that showed higher transcript abundance in cv. ZY821 at 6–12 hpi compared with cv. Westar based on the microarray data ( Fig. 4 ). The genes encoded BnLOX1 (BN22526), a lipoxygenase (LOX), BnWRKY33 (BN17285), a WRKY‐like transcription factor, BnWRKY40 (BN23912), another WRKY‐like protein, and BnLectin (BN25790), a legume lectin family protein. qRT‐PCR analysis confirmed that the transcript level for each gene was significantly higher in cv. ZY821 than in cv. Westar at these early time points, suggesting that the resistant cultivar responds more quickly to the pathogen. 4 Quantitative reverse transcriptase‐polymerase chain reaction (qRT‐PCR) analysis of four genes that were differentially expressed in Brassica napus cv. ZY821 (white bars) and Westar (grey bars) stems at 6 and 12 h (indicated) after inoculation with Sclerotinia sclerotiorum . The increase in transcript levels was relative to a mock‐inoculated control after normalization to a B. napus reference gene ( Actin ). The error bars show the standard deviation, with means that were statistically different indicated by either **(Student t ‐test, P > 0.05) or *(Student t ‐test, P > 0.1) below each time point. A comparison of the lists of genes with increased transcript abundance in the two cultivars across all time points revealed 41 genes that were expressed only in cv. ZY821 ( Table 1 ). Transcripts from four genes encoding transcription factors were elevated, two of which were induced within 12 hpi. These may be partly responsible for regulating the expression of genes involved in the stress response or metabolism. Two genes encoding annexins (BN18917 and BN15878) were induced, one within 12 hpi and the other by 48 hpi. Annexins bind to membranes in a calcium‐dependent manner, but undergo cellular redistribution in response to stimuli, including environmental stress ( Clark , 2001 ). Some plant annexins have been shown to form calcium channels ( Gorecka , 2007 ) which may be important for the initiation of the signalling cascades needed to induce the expression of defence‐associated genes. Eleven other genes encoded proteins involved in some aspect of the stress response. Most noteworthy was a gene encoding a concanavalin‐like lectin (BN25790) that was induced five‐fold at 6 hpi and 109‐fold by 48 hpi based on microarray analysis, and several 1000‐fold based on qRT‐PCR analysis ( Fig. 3 ). Although such proteins have not been shown to exhibit anti‐mycotic activity, a gene encoding an endo‐β‐1,3‐glucanase was induced within 6 hpi with expression increasing steadily afterwards. A gene encoding BnPGIP1 (BN12243), a member of the expanded PGIP gene family in B. napus that is responsive to S. sclerotiorum infection ( Hegedus , 2008 ), was also induced. PGIPs inactivate polygalacturonases, which are major virulence determinants, and several are produced by S. sclerotiorum ( Li , 2004b ). Genes encoding a multidrug resistance protein (BN27037) and a UDP‐glucosyl transferase (BN22038) involved in secondary metabolite biosynthesis were induced within 12 and 24 hpi, respectively. The expression of these genes implies that a pharmacological battle is being waged between host and pathogen. 1 Classification of genes that were induced in only Brassica napus cv. ZY821 stems after Sclerotinia sclerotiorum infection. Oligo ID * Corresponding A. thaliana locus Description 6 † 12 24 48 72 Transcription BN11472 At3g57390 MADS‐box protein – 4.16 ‡ 6.28 12.8 4.95 BN26939 At2g47460 Myb transcription factor – 2.21 2.17 5.28 5.71 BN27665 At1g28360 ERF domain protein 12 – – 3.05 9.46 7.87 BN26396 At5g57520 Zinc finger – – 3.30 5.37 3.57 Stress BN25790 At3g16530 Lectin 5.18 15.8 50.5 109 108 BN22148 At3g57260 Endo β‐1,3‐glucanase 2.27 3.52 6.11 9.91 7.56 BN18917 At2g38750 Annexin – 3.65 2.66 25.1 33.5 BN27037 At4g22990 Multidrug resistance protein 2 – 2.20 2.57 8.02 9.01 BN22038 At4g34135 UDP‐glucosyl transferase – – 3.95 8.95 3.40 BN16777 At4g15160 Protease inhibitor – – 2.92 3.02 18.4 BN20902 At4g27670 Heat shock protein – – 2.65 7.35 3.71 BN11639 At1g69530 Expansin – – – 7.21 9.46 BN17923 At4g10265 Wound‐responsive protein – – – 6.03 7.79 BN18404 At5g27060 Leucine‐rich repeat – – – 5.58 5.00 BN15878 At5g65020 Annexin – – – 4.26 7.39 BN13492 At2g37640 Expansin – – – 3.90 7.79 BN12243 At5g06860 PGIP1 – – – 3.27 18.9 Metabolism BN22522 At4g27070 Tryptophan synthase 2.65 2.69 4.65 10.6 8.31 BN24404 At4g15417 Ribonuclease III 2.00 2.76 3.33 26.5 14.0 BN23990 At1g05680 UDP‐glucosyl transferase – 2.53 2.95 24.5 9.04 BN16989 At1g48920 Nucleolin – 2.08 4.04 6.67 4.41 BN18865 At1g07720 β‐Ketoacyl‐CoA synthase – – 3.09 28.8 15.7 BN15489 At1g17710 Phosphoric monoester hydrolase – – 3.01 73.0 54.4 BN23164 At4g18950 Ankyrin protein kinase – – 2.10 5.80 4.61 Others BN14638 At1g19770 Purine permease 2.13 2.10 4.59 10.9 10.1 BN20514 At1g76790 O ‐Methyltransferase 2.00 7.87 12.5 25.3 31.1 BN13257 At1g54020 Lipolytic enzyme – 3.15 14.2 – – BN14979 At2g30540 Glutaredoxin – 2.16 4.74 17.4 7.81 BN13468 At4g22620 Auxin‐induced protein – 2.05 2.16 – 8.11 BN26740 At2g31990 Exostosin – 2.04 2.17 10.4 7.10 BN23610 At5g43340 Inorganic phosphate transporter – – 4.66 27.5 14.8 BN20039 At1g15180 Membrane protein family – – 4.37 11.7 12.1 BN18437 At5g15440 Circadian clock coupling factor – – 3.75 13.4 11.5 BN24154 At5g01210 Transferase – – 3.63 5.48 7.55 BN17385 At5g10300 Hydrolase – – 2.14 3.91 7.93 BN26419 At5g17450 Copper homeostasis factor – – – 11.1 9.69 BN19106 At5g01750 Expressed protein – – – 9.41 6.90 BN20931 At5g60760 2‐Phosphoglycerate kinase – – – 7.38 4.21 BN24358 At3g54040 Photoassimilate‐responsive protein – – – 5.07 – BN17022 At3g13857 Expressed protein – – – 2.94 10.5 BN13468 At4g22620 Auxin‐induced protein – – – – 9.38 * Oligonucleotide designation from the B. napus gene‐specific array. † Hours post‐inoculation. ‡ Fold changes relative to mock‐inoculated control. Genes involved in hormone signalling Auxin and gibberellin Genes encoding products involved in aspects of hormonal signal transduction (Table S4, see Supporting Information) were identified in the response of cv. ZY821 to inoculation with S. sclerotiorum . Most plant defence genes are regulated by endogenous plant hormones, such as salicylic acid (SA), JA and ET. An indole‐3‐acetic acid amino synthetase has also been reported to be involved in the activation of basal immunity in rice ( Ding , 2008 ), indicating the role of auxin in plant defence. Two B. napus genes (BN13468 and BN22832), similar to At4g22620 and At5g35735, which are known to be responsive to auxin, were induced in cv. ZY821 at 12 hpi, but not in cv. Westar ( Table 2 ). Two other genes encoding enzymes involved in the biosynthesis of indole‐3‐acetic acid, cytochrome P‐450 CYP79B2 (BN24462) and aldehyde oxidase AA01 (BN15682), were induced at 6 and 12 hpi only in cv. ZY821 (Table S4). Gibberellins (GAs) also regulate plant growth and various developmental processes. Transcripts from a B. napus gene encoding gibberellin 2‐oxidase (BN19546), which is responsible for GA degradation ( Thomas , 1999 ), increased at 12 hpi in cv. ZY821 ( Table 2 ), suggesting that dynamic changes in plant growth and development are occurring. 2 Induction of regulatory genes in Brassica napus cv. ZY821 and Westar stems after Sclerotinia sclerotiorum inoculation. Oligo ID * Corresponding Signalling pathway Description ZY821 Westar A. thaliana locus 6 † 12 6 12 BN13468 At4g22620 Auxin IAA induced – 2.05 – – BN22832 At5g35735 Auxin IAA induced – 2.57 – – BN19546 At1g30040 Gibberellin GA degradation – 2.11 – – BN17492 At1g76650 Calcium Ca 2+ binding – – – 2.14 BN18528 At4g20780 Calcium Ca 2+ binding – 2.81 – 3.16 BN19238 At3g63380 Calcium Ca 2+ transport – 9.93 – 6.03 BN19735 At3g57330 Calcium Ca 2+ transport – 2.08 – 2.06 BN24299 At2g17290 Protein kinase – 2.80 – 2.61 BN10050 At3g45640 MAP kinase MPK3 2.00 – – – BN25533 At1g73500 MAP kinase MPK9 – 2.56 – – BN25534 At1g73500 MAP kinase MPK9 – – – 2.23 BN26459 At4g17500 Transcription ERF1 – 5.01 – 3.57 BN15727 At5g47220 Transcription ERF2 – – – 3.68 BN25670 At5g47230 Transcription ERF5 2.00 2.31 – 2.09 BN25738 At4g17490 Transcription EFR6 – – – 2.17 BN17976 At5g51190 Transcription AP2 2.97 3.11 – – BN11384 At2g46510 Transcription bHLH – – – 2.55 BN14283 At4g14365 Transcription RING finger 2.53 3.13 – 2.22 BN18692 At5g27420 Transcription RING finger – – – 3.96 BN26939 At2g47460 Transcription MYB12 – 2.21 – – BN14274 At3g23250 Transcription MYB15 – 2.71 – – BN22570 At3g46600 Transcription SCL11 – 3.05 – – BN17285 At2g38470 Transcription WRKY33 3.72 5.69 – 4.11 BN23912 At1g80840 Transcription WRKY40 2.98 8.05 – 2.73 BN10329 At1g27730 Transcription ZAT10 2.45 3.94 – – BN10330 At1g27730 Transcription ZAT10 4.41 8.12 2.08 6.40 BN14917 At5g59820 Transcription ZAT12 – 3.20 – – BN14919 At5g59820 Transcription ZAT12 – 5.39 – 2.70 BN22157 At3g21150 Transcription Zinc finger – – – – AP2, APETALA2; bHLH, basic helix–loop–helix; ERF, ethylene response factor; GA, gibberellin; IAA, indole acetic acid; MPK, mitogen activated protein kinase; MYB, myeloblastosis; RING, really interesting new gene; SCL, scarecrow‐like; ZAT, zinc‐finger protein from Arabidopsis thaliana . * Oligonucleotide designation from the B. napus gene‐specific array. † Fold change relative to mock‐inoculated control within each genotype at 6 and 12 hours post‐inoculation (hpi), respectively. Jasmonic acid and ethylene Of the classic defence hormones, JA and ET tend to broadly coordinate the response to mechanical attack from insects and necrotrophic pathogens, whereas SA is most often associated with the response to biotrophic pathogens ( Dong, 1998 ; McDowell and Dangl, 2000 ). This classification is not absolute as these signalling pathways may act individually, synergistically or antagonistically depending on the pathogen involved ( Berrocal‐Lobo , 2002 ; Glazebrook, 2005 ). For example, resistance to A. brassicicola in A. thaliana depends on JA signalling and camalexin production ( Glazebrook, 2005 ), but is also dependent on ET signalling ( Oh , 2005 ). Resistance to B. cinerea depends on genes co‐regulated by JA and ET, but SA, abscisic acid (ABA) and hydrogen peroxide‐dependent defences also contribute to the resistance ( Asselbergh , 2007 ; Glazebrook, 2005 ). JA is synthesized from α‐linolenic acid via four enzymatic steps ( Turner , 2002 ), the first of which yields 13‐hydroperoxylinolenic acid via the action of LOXs. In A. thaliana , LOX2 is primarily responsible for this reaction in response to wounding ( Bell , 1995 ), but five BnLOX genes were induced in cv. Westar within 24 hpi (Table S4). Interestingly, only two BnLOX genes were induced in cv. ZY821; however, one of these was induced more than five‐fold within 6 hpi. Hydroperoxide lyase (HPL) gene expression is also induced by wounding, and this activity can divert some of the 13‐hydroperoxylinolenic acid towards the synthesis of leaf alcohols and aldehydes involved in defence against insect attack ( Bate , 1998 ). In the current study, both cultivars exhibited a strong decrease in the abundance of transcripts for a gene encoding HPL (BN10615), which should serve to increase the pool of 13‐hydroperoxylinolenic acid available for JA synthesis. Allene oxide synthase (AOS) catalyses the formation of 12,13‐epoxylinolenate, which is then converted to 12‐oxophytodienoic acid by allene oxide cyclase (AOC). Similar to the BnLOX genes, three BnAOC genes were induced in cv. Westar within 24 hpi, whereas cv. ZY821 expressed only two, one of which was induced within 12 hpi. The final step in JA biosynthesis is catalysed by 12‐oxophytodienoate reductase (OPR). Both cultivars exhibited elevated levels of three BnOPR genes with similar expression patterns in this study. ET is synthesized by three enzymatic reactions from methionine ( Wang , 2002 ), with the first step being catalysed by S ‐adenosyl methionine synthase (SAM). A sharp decrease in BnSAM1 transcripts was detected in cv. Westar by 48 hpi. In contrast, BnSAM3 was induced in cv. ZY821 within 24 hpi. Two aminocyclopropane carboxylate synthase (ACS) genes were detected, BnACS6 and BnACS8 ; however, only BnACS6 was induced in cv. ZY821 and, once again, earlier than in cv. Westar. The final step in ET biosynthesis is catalysed by aminocyclopropane carboxylate oxidase (ACO). In cv. Westar, three BnACO genes were induced within 24 hpi, but only one in cv. ZY821 with the same expression profile as in cv. Westar. Salicylic acid Recently, Guo and Stotz (2007 ) reported that defence to S. sclerotiorum is dependent on JA, SA and ET signalling in A. thaliana . Likewise, an ABA‐deficient tomato line displayed an SA‐dependent defence response and was less susceptible to S. sclerotiorum and Erwinia chrysanthemi ( Asselbergh , 2007 ). In this study, genes encoding enzymes leading to JA and ET synthesis were rapidly induced on S. sclerotiorum inoculation, as were the genes responsive to them, such as chitinase, β‐1,3‐glucanase and plant defensin 1 (PDF1), whereas transcript abundance from several genes leading to SA biosynthesis decreased. Two orthologues of isochorismate synthase 1 (ICS1) required for the synthesis of SA in A. thaliana ( Wildermuth , 2001 ) were represented on the B. napus microarray (BN15208/BN15209 and BN18712). Although it is recognized that changes in gene expression do not necessarily reflect changes in metabolite levels, transcripts from both genes decreased similarly in cv. Westar and ZY821 (Tables S1 and S2). The coordination of the response to necrotrophic pathogens is obviously complex, and the role of SA and other signalling molecules cannot be discounted, as genes that exclusively respond to them, for example BnPR1 , were induced. Genes encoding transcription factors Rapid and directed induction of gene expression requires both a sensitive surveillance system and a highly responsive transcriptional apparatus. Several genes encoding zinc finger, WRKY, MYB‐like and APETALA2 (AP2)‐like transcription factors exhibited elevated levels of transcripts after inoculation with S. sclerotiorum ( Table 2 ). ET response factors (ERFs) are a class of AP2 domain‐containing transcription factor that recognize the ET responsive element in the promoters of ET‐responsive genes ( Riechmann and Meyerowitz, 1998 ). Transcript levels for BnERF1 (BN26459) were similar between the two cultivars; however, BnERF2 (BN15727) and BnERF6 (BN25738) were elevated in cv. Westar at 12 hpi and BnERF5 (BN25670) was elevated at 6 hpi in cv. ZY821 and in both cultivars at 12 hpi ( Table 2 ). The genes encoding additional AP2 domain, several zinc finger, RING finger, MYB and WRKY transcription factors were found to be induced exclusively, or at higher levels, in cv. ZY821 within 6–12 hpi. AbuQamar (2006 ) identified 30 genes encoding transcription factors that exhibited increased transcript abundance in A. thaliana leaves within 24 h after inoculation with B. cinerea . Functional redundancy precluded the establishment of a readily visible phenotype for most insertion inactivation mutants; however, a few were found to show reduced resistance. These included two zinc finger proteins with ankyrin repeats (C‐X 8 ‐C‐X 5 ‐C‐X 3 ‐H‐type), At3g55980 and At2g40140, which, when mutated, gave rise to plants with enhanced local susceptibility and sensitivity to ABA. In the current study, orthologues of these genes were not detected in either of the B. napus cultivars used, but genes encoding a similar zinc finger protein, BnZAT12 (BN14917 and BN14919), were induced soon after inoculation of cv. ZY821 ( Table 2 ). ZAT12 is required for the expression of ascorbate peroxidase, which provides some measure of resistance to hydrogen peroxide during oxidative stress ( Rizhsky , 2004 ). The genes encoding two WRKY factors similar to A. thaliana WRKY33 (BN17285) and WRKY40 (BN23912) were also rapidly induced in cv. ZY821 ( Table 2 ). Several pieces of evidence attest to the importance of WRKY factors in the mediation of the resistance response. A B. napus WRKY33 orthologue was induced in a partially resistant cultivar on infection by S. sclerotiorum and mapped to a QTL for resistance on N14 ( Zhao , 2007 ). Arabidopsis thaliana wrky33 mutants were more susceptible to B. cinerea and A. brassicicola , which corresponded to a reduction in the expression of the gene encoding the anti‐fungal protein PDF1.2 ( Zheng , 2006 ). Surprisingly, over‐expression of AtWRKY33 in A. thaliana enhanced resistance to necrotrophic fungal pathogens, but increased susceptibility to P. syringae , leading to the suggestion that WRKY33 might serve as a positive regulator of JA/ET‐responsive genes and a repressor of SA‐responsive genes ( Zheng , 2006 ). Genes encoding components of signal transduction pathways Genes coding for products that function upstream of transcription initiation ( Table 2 ) were identified. One gene (BN17492) encoding a calcium‐binding protein was induced in cv. Westar but not in cv. ZY821, whereas the expression of other genes (BN18528, BN19238 and BN19735) involved in calcium signalling was similar. Protein phosphorylation and dephosphorylation were implicated in the mediation of the response to the necrotrophic fungus Phoma macdonaldii in sunflower ( Alignan , 2006 ). Several genes encoding protein phosphatase 2C (Tables S1 and S2) were induced after exposure to S. sclerotiorum in both cultivars. These enzymes are negative regulators of mitogen‐activated protein kinase (MAPK) signalling pathways ( Rodriguez, 1998 ), including those leading to programmed cell death (PCD) ( Tamura , 2006 ), but are also produced after treatment with the necrosis‐inducing protein, Nep1, concomitant with a calmodulin‐like protein ( Keates , 2003 ). MAPK cascades are also required for the induction of defence responses to biotic and abiotic stresses. The inactivation of MPK6 and MPK4 by the protein phosphatase, AP2C1, in A. thaliana reduced innate immunity and increased susceptibility to B. cinerea ( Schweighofer , 2007 ). Two MAPK genes, BnMPK3 (BN10050) and BnMPK9 (BN25533/BN25334), were induced at 6 and 12 hpi, respectively, in cv. ZY821 ( Table 2 ); microarray studies in A. thaliana have shown that AtMPK3 is responsive to chitin oligomers. Defence‐related genes and pathways Of the genes that were induced within 6–12 hpi in both cultivars, those encoding pathogenesis‐related (PR) proteins and proteins involved in the response to biotic and abiotic stress were prominent ( Table 3 ). Several of these are known to possess anti‐fungal activity, such as chitinase and β‐1,3‐glucanase, but they may have only a limited role in providing resistance to S. sclerotiorum , as they were also induced in the susceptible cultivar. Plant lectins are members of a large group of carbohydrate‐binding proteins ( Peumans and Van Damme, 1998 ), some of which are induced by chitin ( Ramonell , 2005 ) and exhibit potent antimicrobial activity ( Koo , 2002 ). In this study, genes encoding concanavalin A‐like (BN25790) and curculin‐like mannose‐binding lectins (BN24426 and BN24464) were induced at 6–12 hpi, although these lectins have not been reported to show antimicrobial activity. As indicated previously, many genes were members of expanded families, with some members being commonly expressed in the two cultivars and a subset that was exclusive to each cultivar. This feature was most clearly evident in the expression pattern of a group of 11 glutathione‐ S ‐transferase (GST) genes, 10 of which were exclusive to either cv. ZY821 or cv. Westar within 6–12 hpi. 3 Induction of defence‐related genes in Brassica napus cv. ZY821 and Westar stems after Sclerotinia sclerotiorum inoculation. Oligo ID * A. thaliana Description ZY821 Westar locus 6 † 12 6 12 BN14925 At3g57240 β‐Glucanase 2.10 6.42 – 9.52 BN14926 At3g57240 β‐Glucanase – – 8.54 – BN15009 At2g43570 Chitinase 3.66 10.1 2.41 10.4 BN24266 At2g43590 Chitinase 7.21 10.9 10.1 20.1 BN24268 At2g43590 Chitinase – 3.93 – – BN11310 At3g12500 Chitinase – – – 2.35 BN10008 At2g29450 GST – 2.78 – – BN14800 At1g69930 GST – – – 3.00 BN15797 At4g02520 GST 2.11 6.60 2.21 – BN15802 At4g02520 GST – – – 3.82 BN16623 At2g02930 GST – – – 2.59 BN19123 At1g17170 GST – – – 6.20 BN19124 At1g17170 GST – 5.95 – – BN24382 At1g02920 GST – 2.96 – – BN25567 At1g27130 GST – 2.58 – – BN25568 At1g27130 GST – 3.47 – – BN26558 At2g29460 GST – 4.12 – – BN19079 At1g78820 Lectin – – 2.34 BN24426 At1g78860 Lectin – 6.31 – 4.84 BN24464 At1g78850 Lectin – 14.8 – 6.13 BN25790 At3g16530 Lectin 5.18 15.8 – – BN25791 At3g16530 Lectin – – 2.74 17.0 BN24866 At4g11650 Osmotin – 5.26 – – BN14490 At1g75830 PDF1.1 – – 2.25 6.18 BN14537 At2g26020 PDF1.2b – – – 3.15 BN11341 At2g38390 Peroxidase – – – 2.93 BN23917 At4g08780 Peroxidase 2.56 8.08 – – BN19400 At3g49110 Peroxidase 33 2.17 – – – BN24675 At3g49110 Peroxidase 33 – – – 2.65 BN15231 At4g37520 Peroxidase 50 – 2.88 – – BN12244 At5g06860 PGIP1 – 2.34 – 2.17 BN14604 At2g24810 Thaumatin 2.77 9.63 – 4.55 BN14605 At2g24810 Thaumatin – 3.60 – – BN23765 At4g33720 PR1 – 2.08 – – BN25356 At2g14610 PR1 – – – 7.14 BN25095 At3g11820 SYP121 – – – 2.13 BN24260 At3g52400 SYP122 – – – 2.44 GST, glutathione‐ S ‐transferase; PDF, plant defensin; PGIP, polygalacturonase inhibitor protein; PR1, pathogenesis‐related protein 1; SYP, syntaxin of plant. * Oligonucleotide designation from the B. napus gene‐specific array. † Fold change in inoculated stems vs. control 6 and 12 h post‐inoculation (hpi). Although PCD resulting from the release and accumulation of reactive oxygen species (ROS), such as superoxide and hydrogen peroxide, is well suited for containing biotrophic pathogens, the accumulation of ROS is directly proportional to the degree of infection for necrotrophs. Interestingly, transcript levels from two syntaxin genes, BnSYP121 and BnSYP122 (BN25095 and BN24260), which have been implicated in the suppression of PCD pathways ( Zhang , 2007 ), were elevated ( Table 3 ). Osmotin‐like PR proteins may exhibit antimicrobial activity ( Shatters , 2006 ), and one osmotin gene (BN24866) was induced early in the resistant cultivar and later in the susceptible cultivar. Osmotins accumulate normally in the cell apoplast during cold acclimation and provide general cryo‐ and osmotic protection. One report has indicated that osmotin, in conjunction with chitinase, forms a complex with actin, causing microtubule disruption and cytoplasmic aggregation ( Takemoto , 1997 ). Cytoskeletal reorganization is required for translocation of the nucleus to the site of pathogen interaction, a process that is thought to be important for rapid and targeted deployment of early defences. D'Angeli and Altamura (2007 ) reported that osmotin regulated the onset of PCD by blocking the Ca 2+ signalling necessary to initiate actin micro‐filament depolymerization. ROS are generated by normal mitochondrial electron transport mediated by the cytochrome c pathway and through the actions of oxidases ( Amirsadeghi , 2006 ). Several alternative oxidases are capable of serving as the terminal oxidase, accepting electrons from ubiquinone and diverting electron flow from the cytochrome c pathway ( Finnegan , 2004 ). In the current study, transcript abundance from a gene encoding an alternative oxidase (BN18515) increased early in the infection process in both cultivars (Tables S1 and S2), which should alter the rate of ROS generation. The relationship between alternative oxidase, ROS and PCD is not entirely clear as transgenic tobacco lines over‐expressing alternative oxidase exhibit smaller lesions ( Ordog , 2002 ), whereas wheat mutants with accentuated lesions show elevated alternative oxidase levels ( Sugie , 2007 ). Genes involved in secondary metabolism Certain secondary metabolites may play a role in plant defence. Altered patterns of expression were observed for genes associated with biochemical pathways leading to the biosynthesis of phenylpropanoids and glucosinolates in this study. The phenylpropanoid pathway is responsible for the synthesis of a wide variety of secondary metabolites, including lignins, flavonoids, phytoalexins and antioxidants, which play a role in plant defence ( Dixon and Paiva, 1995 ). Of the 11 differentially expressed genes associated with the phenylpropanoid biosynthetic pathway, most exhibited similar expression patterns in both cv. ZY821 and Westar (Table S4). Transcripts from a gene encoding a 4‐coumarate‐CoA ligase (BN22021), responsible for the formation of coumaroyl‐CoA, were elevated. Coumaroyl‐CoA is directed to several pathways, one of which leads to the formation of coniferyl aldehyde. The final step in this pathway is catalysed by cinnamoyl‐CoA reductase, and transcript abundance from three genes encoding this enzyme (BN23842, BN15787 and BN13060) was elevated in both cultivars. The fate of coniferyl aldehyde is less clear, as genes encoding alcohol dehydrogenases that would result in the formation of coniferyl alcohol were induced in cv. Westar (BN17669), but repressed in cv. ZY821 (BN22074). In contrast, transcript abundance from a gene that encodes a ferulate hydroxylase (BN18532), catalysing the first step in the pathway to ferulate and sinapate, which are precursors for lignin biosynthesis, decreased in cv. Westar. Glucosinolates are plant secondary metabolites found in the Brassicaceae family and can be grouped into three classes, aliphatic, indolyl and aromatic glucosinolates, according to the amino acid precursor used for their synthesis. Tryptophan biosynthesis could be accelerated, as transcript levels from genes encoding enzymes for chorismate biosynthesis and for four of the five biochemical reactions converting tryptophan to indolylmethyl‐glucosinolate, increased sharply in both cultivars. Indolylmethyl‐glucosinolate might be converted into different indolyl glucosinolate derivatives. Changes in the expression of genes associated with carbon metabolism The expression of genes encoding key enzymes involved in carbohydrate utilization and energy production [glycolysis, tricarboxylic acid (TCA) and Calvin cycles] was altered in both cultivars (Table S4), suggesting that changes in plant metabolism may occur following inoculation with S. sclerotiorum . A sharp decrease was detected in transcript abundance for five genes encoding glyceraldehyde‐3‐phosphate (G3P) dehydrogenase (BN11845, BN14336, BN24802, BN16865 and BN18780), which is responsible for the generation of G3P from 1,3‐diphosphoglycerate, the key intermediate required in carbon fixation via the Calvin cycle. Two genes further downstream in the pathway, encoding transketolase (BN25656) and ribose‐5‐phosphate isomerase (BN17967), were induced, and may serve to scavenge available G3P and other Calvin cycle intermediates by converting them to ribulose‐1,5‐biphosphate. This compound could be directed to the photorespiration pathway, where expression of a gene encoding glycolate oxidase (BN13732) was induced by as much as 50‐fold. Glycolate oxidase converts glycolate to glyoxylate and H 2 O 2 , the latter being an important component of the defence response. Transcript abundance from genes involved in the photorespiration pathways downstream of glyoxylate decreased, suggesting that glyoxylate is metabolized via a different mechanism. Glyoxylate may enter the TCA pathway by way of the glyoxylate shunt after conversion to malate. Transcript abundance from malate dehydrogenase (BN16831) and citrate synthase (BN24865 and BN15521) genes increased, although transcript abundance from two isocitrate lyase genes (BN26052 and BN16663) decreased. Many of the altered gene expression patterns appear to be directed towards increasing carbon flow to and accelerating flux through the TCA cycle. Multiple genes encoding each of the enzymes required to convert pyruvate to acetyl‐CoA (such as dihydrolipoyl dehydrogenase, dihydrolipoyl lysine‐residue acetyltransferase, pyruvate decarboxylase and pyruvate dehydrogenase) were induced, as were genes for several key steps in the TCA cycle. If carbon fixation via the Calvin cycle is impeded, the plant can access carbon reserves to drive the increased flux through the TCA cycle. A potential source of TCA cycle intermediates is the degradation of fatty acids by the β‐oxidation pathway. In this study, genes encoding enzymes that convert acyl‐CoA to acetyl‐CoA, such as acyl‐CoA oxidase (BN16426 and BN22545), enoyl‐CoA hydratase (BN13325) and 3‐ketoacyl‐CoA thiolase (BN11605), were induced; however, transcript abundance from several genes encoding long‐chain fatty acid‐CoA ligases (BN22268, BN13971 and BN20014), responsible for the initial stage of degradation, decreased (Table S4). Sucrose serves as a storage compound and can be transported throughout the plant. It can be converted into UDP‐glucose and fructose by sucrose synthase, or to glucose and fructose by invertase. Sucrose degradation via the latter pathway appears to be accelerated, as indicated by the induction of four genes encoding invertases (BN10428, BN18301, BN19126 and BN18327), whereas the expression of genes encoding enzymes involved in the alternative pathway, namely sucrose synthase (BN12649) and phosphoglucomutase (BN21073), was reduced in this study. In addition, the genes encoding hexokinase (BN25532) and glucose‐6‐phosphate isomerase (BN11549) were also induced, suggesting that a large pool of fructose‐6‐phosphate (F6P) is formed. F6P generated from sucrose degradation is converted to pyruvate by the glycolysis I pathway in the plant cytosol. A gene encoding a key enzyme in this pathway, phosphofructokinase (BN19519), which converts F6P into fructose‐1,6‐diphosphate, was induced in both cultivars. F6P can also be formed from starch degradation in plastids and is converted to pyruvate by the glycolysis II pathway; however, the means by which the carbon in F6P enters the TCA cycle does not appear to be so direct. Similar to that observed for the Calvin pathway, the generation of 3‐phosphoglycerate from G3P appears to be blocked by the repression genes encoding G3P dehydrogenases. It seems more likely that G3P is directed to the photorespiration pathway, resulting in the formation of glyoxylate and H 2 O 2 , with the glyoxylate entering the TCA cycle via the glyoxylate shunt. CONCLUSIONS In this study, a B. napus oligonucleotide microarray representing 15 000 unigenes was used to study the response to inoculation with S. sclerotiorum of stems in resistant and susceptible oilseed rape cultivars. A large number of genes were found to be differentially regulated after infection. The gene expression profiles during the latter stages of stem infection were similar to those of B. napus seedlings in response to S. sclerotiorum infection ( Yang , 2007 ; Zhao , 2007 ) and other necrotrophic pathosystems, namely Botrytis cinerea—A. thaliana ( AbuQamar , 2006 ) and Alternaria brassicicola—Brassica oleracea ( Cramer , 2006 ). Sampling was biased towards the very early stages of the infection, as it was hypothesized that events occurring soon after the initial interaction between pathogen and host were critical for resistance to aggressive necrotrophs, such as S. sclerotiorum . Infected stems were examined as these are the natural site for infection in this pathosystem. The resistant cultivar ZY821 effected the induction of defence‐associated genes earlier than the susceptible cultivar Westar. Differences were also observed in the patterns of a large group of regulatory genes, as well as genes involved in plant hormone synthesis and aspects of the host defence mechanisms against pathogens. Changes in the expression of genes involved in carbon metabolism suggest that carbon storage reserves (such as sucrose, starch and lipid) are accessed and shuttled through the photorespiration pathway. This pathway leads to the formation of glyoxylate, which can enter the TCA cycle, as well as hydrogen peroxide, which may form a part of the defence response ( Huckelhoven , 1999 ). The activity of the TCA cycle may be greatly increased, as evidenced by the induction of many genes encoding TCA cycle enzymes. This not only generates reducing power and energy, but also precursors for amino acid biosynthesis. A sharp increase in transcript abundance for genes encoding enzymes for tryptophan biosynthesis was observed, this amino acid being a precursor for both glucosinolates and phenylpropanoids. This study has provided new insights into aspects of the defence response to necrotrophic pathogens using an experimental design that closely resembles the actual circumstances occurring in the field. Several targets are being analysed to determine their contribution to the resistance in B. napus cv. ZY821 and other resistant cultivars. EXPERIMENTAL PROCEDURES Plant material, inoculation and tissue harvest The semi‐winter‐type B. napus cv. ZY821 and a spring‐type doubled haploid line (N‐o‐1) derived from the cultivar Westar (hereafter referred to simply as cv. Westar) were used. Brasicca napus cv. ZY821 is partially resistant to S. sclerotiorum ( Zhou , 1994 ), but cv. Westar is susceptible. Seeds of cv. ZY821 were sown in a Redi‐Earth soil mixture (Scotts Fertilizer Company) amended with coconut fibre to improve drainage and Osmocote 15‐7‐12 (Scotts Fertilizer Company, Marysville, OH, USA) for the controlled release of fertilizer. The plants were maintained in a growth chamber under 200 µmol/m 2 /s light with a 16‐h photoperiod provided by fluorescent and incandescent light. The temperature was maintained at 20 °C in the light and 16 °C in the dark. Two seeds were sown in each of 36 cells, and seedlings were thinned to one plant per cell 11 days after sowing. Four‐week‐old seedlings of the winter type were vernalized in a cold room at 5–8 °C for 4 weeks with an 8‐h photoperiod and half of the previous light intensity. After vernalization, each plant was transplanted to a 6‐in (15‐cm) pot and returned to the growth chamber. Seeds of cv. Westar were sown directly in 6‐in pots and plants were grown under the same conditions as for cv. ZY821, except for the vernalization treatment. Plants were selected for inoculation and sampling using a randomized design with three biological replicates for each cultivar. Each replicate consisted of 60 plants for five time points (6, 12, 24, 48 and 72 hpi) and two treatments (inoculated and mock‐inoculated controls). An S. sclerotiorum isolate (321; Kohli and Kohn, 1995 ), obtained originally from field‐sown canola plants, was used throughout this study. Plate cultures were maintained on minimal medium according to Li (2004a ). When the plants were at the full flowering stage, three sites on the main stem were inoculated at every other internode with 8‐mm plugs cut from the growing margin of a plate colony cultured on minimal salts‐glucose agar (20.0 g glucose, 2.0 g NH 4 NO 3 , 1.0 g KH 2 PO 4 , 1.0 g NaOH, 0.1 g MgSO 4 .7H 2 O, 0.1 g malic acid and 20.0 g agar per 1000 mL distilled water) modified from Cruickshank (1983 ). Mock‐inoculated plants were treated with an agar plug only. Plugs were secured to the stem with Parafilm™. Each line yielded 180 inoculation sites per biological replicate, or 540 sites in total. Epidermal stem tissues extending 10 mm beyond the inoculation site and 1 mm deep were excised at 6, 12, 24, 48 and 72 hpi with a razor blade. The tissues harvested from one biological replicate at each time point (six individual plants comprising 18 inoculation sites) were pooled as one sample. Harvested tissues were frozen immediately in liquid nitrogen and stored at −80 °C. Probe preparation and slide hybridization Total RNA was isolated from 1 g of ground tissue using the plant RNeasy kit (Qiagen, Mississauga, Ontario, Canada) and stored at −80 °C. RNase‐free DNase was applied during RNA isolation to remove any DNA contamination (Qiagen). The concentration of total RNA was determined using a Nanodrop spectrophotometer (Nanodrop Technologies, Montchanin, DE, USA) and verified by agarose gel electrophoresis. Probe preparation, hybridization and washing were performed using the methods of Stasolla (2008 ). Briefly, the Amino Allyl MessageAmp II aRNA amplification kit was used for cDNA synthesis and RNA amplification (Ambion, Austin, TX, USA) with 1 µg of total RNA, following the manufacturer's instructions. In vitro transcription to prepare amplified RNA (aRNA) was conducted at 37 °C for 14 h and then purified. aRNA was quantified using a Nanodrop spectrophotometer, and 5 µg were coupled with either cyanine‐3 (Cy3) or cyanine‐5 (Cy5) dye. The coupling reaction mixture was incubated at 20 °C for 30 min, quenched with 4 m hydroxylamine for 15 min and purified. Cy3/Cy5‐labelled aRNA was dried completely in a vacuum, dissolved in 9 µL double‐distilled H 2 O and fragmented by the addition of 1 µL of fragmentation reagent (Ambion, catalouge #8740), followed by incubation at 70 °C for 15 min. The reaction was stopped with 1 µL of stop solution and 47 µL of hybridization solution [25% formamide, 5 × standard saline citrate (SSC), 0.1% sodium dodecylsulphate (SDS), 0.1 mg/mL sheared salmon sperm DNA] was added. Hybridizations were performed at 37 °C for 16 h in the dark, after which the slides were washed in 2 × SSC, 0.1% SDS at 37 °C for 5 min, followed by two washes in 1 × SSC for 2 min each and two washes in 0.1 × SSC for 1 min each. The washed slides were dried by centrifugation at 225 g for 1 min at 20 °C. Microarray experimental design and data analysis Dye swap hybridizations were used to compare changes in gene expression between inoculated and mock‐inoculated samples at each time point (from 6 to 72 hpi) for each of the two cultivars. Each dye swap experiment comprised two hybridizations: the first consisted of two samples labelled with either Cy3‐dCTP or Cy5‐dCTP and hybridized onto a single slide, and the second with the same two samples labelled in reverse. Thus, the experimental design included five time points, two cultivars and three biological replicates, for a total of 60 slides. The slides were scanned using a Virtek Chip Reader (Bio‐Rad Laboratories, Hercules, CA, USA). For each scan, the signal intensity of the two channels was initially adjusted on the basis of the control spots so that an average red/green signal ratio of approximately unity was achieved and the spots were not saturated. Spot intensities from scanned slides were quantified using ArrayPro 4.0 (Bio‐Rad Laboratories) and the data were imported into the BioArray Software Environment (BASE version 1.0). Spots with intensities lower than the background were eliminated. Gene Spring 7.3 (Agilent Technologies, Inc., Mississauga, Ontario, Canada) was used to select spots with intensity values ≥50 in at least half of the samples (slides) for further analysis. Normalization was conducted using the per‐spot and per‐chip Lowess normalization procedure with a smooth factor of 20. A one‐way anova with a parametric test and variances not assumed to be equal was performed ( Craig , 2003 ). A multiple test correction ( Benjamini and Hochberg, 1995 ) was applied and an FDR of ≤0.0001 was used. A list of genes for which the expression varied by more than 2.0‐fold was generated. qRT‐PCR analysis qRT‐PCR analysis was conducted to verify some of the microarray data. Single‐stranded cDNA was synthesized from 4 µg of total RNA using the Superscript First‐Strand Synthesis System (Invitrogen, Burlington, ON, Canada); 6 ng of total cDNA was used in each RT‐PCR reaction. qRT‐PCR was performed in a DNA Engine Opticon 2 (MJ Research, Watham, MA, USA) system with the SYBR Green qPCR kit (Finnzymes Oy, Espoo, Finland) in a reaction volume of 20 µL containing 0.5 µ m of each primer. The PCR conditions were as follows: 95 °C for 15 min, followed by 45 cycles of 94 °C for 10 s, 60 °C for 30 s and 72 °C for 30 s. A melting curve of denatured double‐stranded cDNA was established to test the purity of the products. Data were obtained from each biological replicate and repeated in triplicate. The PCR primer amplification efficiency was tested using a six dilution series of the template, according to Pfaffl (2001 ). Expression ratios in the treatments relative to the mock‐inoculated controls were calculated using data corrected for PCR efficiency and normalized against the product from the B. napus Actin gene. Primers were designed using PRIMERTM ( http://www.workbench.sdsc.edu ) based on the homologues in B. napus DH12075, and the product size was limited to 100–300 nucleotides (Table S5, see Supporting Information). ACKNOWLEDGEMENTS This work was funded by Saskatchewan Agriculture and Food, the Saskatchewan Canola Development Commission and the Federal Matching Investments Initiative. We thank Branimir Gjetvaj for technical assistance.

Journal

Molecular Plant Pathology – Wiley

Published: Sep 1, 2009

There are no references for this article.