- Email: [email protected]
Isolation and analysis of candidate ascochyta blight defence genes in chickpea. Part II. Microarray expression analysis of putative defence-related ESTs
Physiological and Molecular Plant Pathology 66 (2005) 201–210 www.elsevier.com/locate/pmpp
Isolation and analysis of candidate ascochyta blight defence genes in chickpea. Part II. Microarray expression analysis of putative defence-related ESTs Tristan E. Coram*, Edwin C.K. Pang School of Applied Sciences, Biotechnology and Environmental Biology, RMIT University, Building 223, Level 1, Plenty Road, Bundoora, Vic. 3083, Australia Received 28 January 2005; revised 18 July 2005; accepted 2 August 2005
Abstract Using microarray technology and previously identified defence-related ESTs from chickpea (Cicer arietinum L.), the ascochyta blight (Ascochyta rabiei (Pass.) Labrousse) resistance response was studied in a highly resistant chickpea accession (ICC3996) and a susceptible cultivar (Lasseter). The time-series expression patterns of 20 defence-related ESTs were studied after inoculation with A. rabiei. Ten of the defence-related ESTs displayed up- or down-regulation in ICC3996 and/or Lasseter compared to uninoculated control samples. Hierarchical clustering grouped the ESTs into clusters of similar observations, revealing that three defence related ESTs showed differential up-regulation in ICC3996 when compared to Lasseter—a leucine zipper protein, SNAKIN2 antimicrobial peptide precursor, and elicitor-induced receptor protein. The potential involvement of these ESTs in effective chickpea defence against ascochyta blight is discussed. The microarray expression analysis of defence-related ESTs in chickpea represents the first study of this type, and future work will focus on large-scale expression studies aimed at further elucidating the function of genes involved in ascochyta blight resistance and the pathway of their action. q 2005 Elsevier Ltd. All rights reserved. Keywords: Chickpea; Ascochyta blight; Genomics; EST; Microarray
1. Introduction Plants deal with a vast range of pathogens, and effective disease resistance stems from the ability of the host to recognise pathogens and then initiate defence mechanisms. Common defence responses include the hypersensitive response (HR), the oxidative burst to produce reactive oxygen intermediates, and the delayed accumulation of salicylic acid necessary for systemic acquired resistance (SAR) [1]. Transcription factors play an integral role in the signalling and control of these pathways, which are also mediated by plant hormones (jasmonic acid and ethylene), elevation of cytosolic calcium, and activation of protein kinases [2]. Additionally, host defences also involve structural alterations such as the strengthening of cell
* Corresponding author. Tel.: C61 3 9925 7140; fax: C61 3 9925 7110. E-mail address: [email protected] (T.E. Coram).
0885-5765/$ - see front matter q 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.pmpp.2005.08.002
walls by phenolic compounds, and/or the production of defensive compounds including pathogenesis-related (PR) proteins [3], phenylpropanoids [4], and protease inhibitors [5] to name a few. At the genomic level, plant defence responses are complex and diverse, and every gene involved in the defence response, from recognition to signalling to direct involvement, forms part of a coordinated response network. Chickpea (Cicer arietinum L.) is the third most important pulse crop in the world [6], but a major factor limiting world production is a severe and destructive fungal disease known as ascochyta blight caused by Ascochyta rabiei (Pass.) Lab. [7]. Currently, the level of resistance in cultivated chickpeas is not sufficient to withstand disease pressure under conditions that are favourable to A. rabiei. However, the world collection of chickpea germplasm contains resistant accessions, such as ICC3996, that remain uncultivated due to poor agronomic characteristics, but do possess a strong capacity for A. rabiei resistance [8]. Understanding of the chickpea defence response to ascochyta blight at the genomic level is a prerequisite for
202
T.E. Coram, E.C.K. Pang / Physiological and Molecular Plant Pathology 66 (2005) 201–210
developing resistant cultivars. Until now the methods for studying the chickpea defence response have been limited to mapping of quantitative trait loci (QTL) associated with A. rabiei resistance [9–11], differential screening of cDNA libraries [12], and mapping of resistance gene analogs to existing linkage maps [13]. Several biochemical studies have also been performed that implicate certain compounds in the chickpea defence response, such as pathogenesisrelated proteins beta-1,3-glucanase [14], chitinase [15], as well as polyphenyloxidase and catalase [16]. The disadvantage of previous studies is that the overall coordinated defence response remains largely uncharacterised, but quantitative methods for global and simultaneous analysis of expression profiles, through microarray analysis, can improve the overall understanding of the coordinated defence response at a molecular level. In fact, microarray analysis has been successful in studying the defence responses of plants such as maize [17], rice [18], soybean [19] and Arabidopsis thaliana [20]. The availability of a set of chickpea ESTs [21] enables the development of efficient and accurate methods of gene expression profiling, including the identification of genes whose expression is changed by a biological phenomenon, such as disease pressure, to suggest functional involvement. The expression pattern of several genes may also be used as an indicator representing a certain state of a cell or tissue, such as resistance or susceptibility to a disease. DNA microarrays are powerful tools for comprehensive characterisation of different plant processes, such as pathogen defence, at the transcription level. In this study, chickpea ESTs functionally classified as having potential involvement in plant defence [21] were employed in microarray experiments. The aim was to generate EST expression profiles over a time-series, after inoculation with A. rabiei spores, in the ascochyta blight resistant ICC3996 chickpea accession and a susceptible chickpea cultivar known as Lasseter.
2. Materials and methods 2.1. Plant material & fungal isolate Seeds of C. arietinum (ICC3996) and C. arietinum (Lasseter) were obtained from the Australian Temperate Field Crops Collection (ATFCC) in Horsham, Victoria, Australia. Seven isolates of A. rabiei collected by Coram and Pang [21] were prepared and maintained on V8 agar according to the method of Collard et al. [8]. Spore suspensions were prepared according to the method of Coram and Pang [21]. 2.2. Cultivation, inoculation & RNA extraction Thirty plants each of ICC3996 and Lasseter were cultivated and inoculated in a glasshouse (20G4 8C)
according to the method of Coram and Pang [21] with three plants potK1 and 10 replicates, of which two replicates served as an uninoculated controls. About 500 mg of stem/ leaf tissue was extracted from all inoculated plants at each of 12, 24, 48, and 96 h post-inoculation. Additionally, stem and leaf tissue samples were also taken from uninoculated plants of ICC3996 and Lasseter at each time point. To confirm that A. rabiei infection had been effective, plants were checked for expected disease symptoms at 14 days post-inoculation based on known infection reactions. For each biological replicate, total RNA was extracted from pooled stem and leaf samples for ICC3996 and Lasseter at each time point (including control samples) using the RNeasyw Plant Mini Kit (QIAGENe). The quantity and quality of the total RNA samples were assessed by OD260/ OD280 ratios and gel electrophoresis, respectively. Total RNA quality was also confirmed by ion-pair reversed-phase High Performance Liquid Chromatography (HPLC) [22], where degraded samples were detected by the absence of sharp elution peaks. 2.3. Microarray construction From a previously synthesised and characterised EST collection constructed from cDNA of the ascochyta blight resistant ICC3996 chickpea accession [21], 20 nonredundant ESTs classified as defence-related were used for the construction of microarrays according to minimum information about a microarray experiment guidelines (MIAME) [23]. In total, the cDNA of 25 ESTs was used after the selection of five housekeeping ESTs as internal normalisation controls (Table 1). The cDNA inserts of the 25 ESTs were amplified to 2000 ng, purified with multiscreen-PCR plates (Millipore), resuspended in 8 mL 50% (v/v) dimethylsulphoxide:water at 250 ng/mL, and arrayed onto amino-silanized slides (Corning) using a Virtek Chipwiter (Virtek Vision International, Inc.) with one pin. Blank buffer spots were also incorporated as negative controls. Each element was deposited in duplicate with a volume of approximately 6 nL and diameter of 200 mm. Slides were pre-hybridised by blocking in 5! SSC, 0.1% SDS, 25% Formamide, 1% BSA for 45 min at 42 8C, rinsed in distilled water and dried. 2.4. Microarray probe preparation & hybridisation Total RNA of 50 mg from each sample was reverse transcribed using Superscript II reverse transcriptase (Invitrogen) and oligo(dT) 23mer (Invitrogen). Aminoallyl dUTP (Sigma) was incorporated during the reverse transcription process. Briefly, the RNA and 5 mg of oligo(dT) primer were denatured at 70 8C for 10 min and cooled on ice before adding first strand buffer to a final concentration of 1!, aa-dUTP/dNTPs mix (final concentrations of 0.5 mM dATP, 0.5 mM dGTP, 0.5 mM dCTP, 0.2 mM dTTP, 0.3 mM aa-dUTP), DTT to a final
T.E. Coram, E.C.K. Pang / Physiological and Molecular Plant Pathology 66 (2005) 201–210
203
Table 1 Identity of the 25 ESTs used for microarray construction, where CON01 to CON05 are normalisation controls and DEF01 to DEF20 represent 20 unique defence-related ESTs EST
Putative identity after BLASTX and BLASTN searches
Category
GenBank dbEST Accession
CON01 CON02 CON03 CON04 CON05 DEF01 DEF02 DEF03 DEF04 DEF05 DEF06 DEF07 DEF08 DEF09 DEF10 DEF11 DEF12 DEF13 DEF14 DEF15 DEF16 DEF17 DEF18 DEF19 DEF20
5.8S/18S/26S ribosomal RNA Ribulose 1,5-bisphosphate carboxylase small subunit Chloroplast 4.5S/5S/16S/23S messenger RNA Chlorophyll a/b binding protein ATP Synthase C chain (EC 3.6.1.34) Extensin-like disease resistance protein Gamma-thionin defensin/protease inhibitor Avr9/Cf-9 rapidly elicited protein 65 Pathogen-induced translation initiation factor nps45 S1-3 pathogen-induced protein Putative disease resistance protein Transcription factor EREBP-1 Caffeoyl-CoA-methyltransferase (EC 2.1.1.104) Pathogenesis-related protein 4A Beta-1-3-glucanase Protein with leucine zipper Pathogen-induced transcription factor Leucine-zipper containing protein Cinnamyl alcohol dehydrogenase (CAD1) Nematode resistance protein Hs1pro-1 Multi-resistance transporter protein Putative flavonol glucosyl transferase SNAKIN2 antimicrobial peptide precursor Elicitor-induced receptor protein Pathogenesis-related protein class 10
Control Control Control Control Control Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence
N/A N/A N/A N/A N/A CV793587 CV793588 CV793589 CV793590 CV793591 CV793593 CV793594 CV793595 CV793597 CV793598 CV793599 CV793600 CV793601 CV793602 CV793603 CV793605 CV793607 CV793608 CV793609 CV793610
concentration of 10 mM, and 150 units Superscript II in a total reaction volume of 30 mL. Reverse transcription was carried out at 42 8C for 2 h before hydrolysis of RNA template with NaOH for 15 min at 65 8C and neutralisation with HEPES (pH 7.0). The cDNA was purified and postlabelled using a PCR purification kit (QIAGENe) and Cy3/Cy5 mono-NHS esters (Amersham) resuspended in 0.1 M Na2CO3 (pH 9.0). cDNA samples were applied to QIAquicke columns and washed/dried according to manufacturers instructions, before adding the appropriate resuspended CyDye to the column membrane and incubating for 1 h at room temperature in the dark. Following incubation, labelled samples were eluted, appropriate Cy3 and Cy5 probes were combined, and purification was repeated. Purified combined probes were resuspended in 2! hybridisation buffer (5! SSC, 0.2% SDS, 50% formamide, 25 mg Cot1 DNA (Invitrogen), 0.4 mg polyA (Sigma), 0.5 mg salmon sperm DNA (Sigma), made up to 60.0 mL with sterile water) and applied to the array after denaturation at 100 8C for 2 min. The slide was covered by a 60!25 mm Lifter slip (Grale Scientific) and incubated in a 42 8C water bath for 16–20 h in a waterproof and humidified hybridisation chamber (Corning) in the dark. Each hybridisation was performed in triplicate and two biological replicates were performed for each hybridisation, incorporating dyeswapping (i.e. reciprocal labelling of Cy3 and Cy5) to eliminate any dye bias.
2.5. Scanning & data analysis After hybridisation, slides were washed for 5 min in each of 1! SSC/0.2% SDS and 0.1! SSC/0.2% SDS, and twice for 2 min in 0.1! SSC. Slides were scanned at 532 nm (Cy3 green laser) and 660 nm (Cy5 red laser) at 10 mm resolution using an Affymetrixw 428e array scanner, and captured with the Affymetrixw Jaguare software (v. 2.0). Image analysis was performed using Imagenee 5 (Biodiscovery) image analysis software. Grids were predefined and manually positioned over the image to ensure optimal spot recognition. Spots with high intensity due to dust particles or other artefacts were manually flagged. Spots were individually quantified using the fixed circle method; sample values were measured as the mean of pixels within the spot circle and the local background in a five-pixel diameter ring that began five pixels outside the spot circle. Automatic flagging eliminated empty spots, negative spots (signal mean!background mean), and poor spots (contaminated background, ignored pixelsO25%, open perimeterO25%, offset from expected positionO60%). Spots with mean signal intensity less than two times the local background were also manually flagged, as well as duplicate spots on the same subgrid with significantly different signal means. Quantified spot data was then compiled and transformed using GeneSighte 3 (Biodiscovery). Data transformations consisted of a background correction (median intensity of specified blank spots subtracted from
204
T.E. Coram, E.C.K. Pang / Physiological and Molecular Plant Pathology 66 (2005) 201–210
signal intensity of all other spots), omitting flagged spots, creating a Cy5/Cy3 mean signal ratio, normalisation by dividing the ratio values of all spots by the mean ratio of the internal normalisation controls whose ratio is expected to equal one, and combination of duplicated spot data. Following data transformation, expression ratios were used for hierarchical cluster analysis, performed with SPSSw version 13.0 (SPSS, Inc., Chicago, IL) using average distance linkage between groups and Euclidean metrics. To determine the significant cut-offs for up- or downregulation, a separate replicated hybridisation was performed using identical total RNA for both Cy3 and Cy5 labelling. Subsequently, ESTs showing up- or down-regulation at one or more time points in ICC3996 or Lasseter were subjected to time-series analysis.
3. Results 3.1. Microarray construction & hybridisation Chickpea cDNA microarrays containing 25 ESTs, of which 20 consisted of previously characterised defencerelated ESTs and the remaining five comprised internal normalisation controls (Table 1), were constructed. Each EST was printed in duplicate spots, and used to study the expression patterns of defence-related ESTs in an A. rabiei resistant (ICC3996) and susceptible (Lasseter) chickpea over a time-series after inoculation with A. rabiei. Each hybridisation was made in reference to an uninoculated control, was made in triplicate, and biologically replicated twice for each time point. The transcript level for each cDNA was calculated firstly as the average intensity of the duplicated spots, then the average intensity of the triplicate hybridisations, and finally the average intensity of the two replicates. The result of the separate replicated hybridisation to determine the significant cut-offs for up- or down-regulation yielded a scatter plot with all spots lying within a two-fold difference range (Fig. 1), therefore, cDNAs were regarded as differentially expressed where they showed a greater than two-fold increase or decrease compared to control samples. These cut-offs translate into up-regulated cDNAs having a ratio S2.0, and down-regulated cDNAs%0.5. Additionally, the distribution of the ratio data for the five normalisation controls in every hybridisation for both ICC3996 and Lasseter also resulted in a scatter plot where all spots lay within a two-fold difference range (Fig. 2).
Fig. 1. Scatter plot of mean signal intensity (Cy3 v. Cy5) from the self:self hybridisation. Broken lines show the two-fold difference range from equal ratio (solid line).
points. Fig. 3, representing the number of up- or downregulated ESTs in both ICC3996 and Lasseter at each time point, shows that differential expression peaked 24 h postinoculation, and the majority of ESTs had returned to normal regulation by 96 h post-inoculation. Additionally, Fig. 3 shows that the main response to A. rabiei inoculation amongst the differentially expressed ESTs was up-regulation (19 instances) in favour of down-regulation (9 instances). The 10 differentially expressed cDNAs were subjected to time-series analysis to independently compare expression levels in ICC3996 and Lasseter (Fig. 4). Of the 10 ESTs, seven were involved in up-regulation (DEF09, DEF10, DEF11, DEF16, DEF18, DEF19, and DEF20),
3.2. Time-series analysis Of the 20 specific defence-related ESTs included in the microarray, 10 exhibited differential expression in at least one time point of either ICC3996 or Lasseter compared to uninoculated control samples. The remaining 10 ESTs failed to show differential expression over the sampled time
Fig. 2. Scatter plot of mean signal intensity (Cy3 vs. Cy5) for the five normalisation controls over every hybridisation for ICC3996 and Lasseter. Broken lines show the two-fold difference range from equal ratio (solid line).
T.E. Coram, E.C.K. Pang / Physiological and Molecular Plant Pathology 66 (2005) 201–210
205
contained two ESTs that displayed significant up-regulation in Lasseter for the all or most of the time-series, and either brief up-regulation in ICC3996 (DEF16), or unchanged regulation in ICC3996 (DEF09).
4. Discussion
Fig. 3. Distribution of the number of differentially expressed ESTs in both ICC3996 and Lasseter over the time-series, where up-regulated ESTs are shaded in grey and down-regulated in black.
whilst three were down-regulated (DEF08, DEF13, and DEF17). Importantly, the three down-regulated ESTs showed the same pattern of reduced expression in both ICC3996 and Lasseter. However, only one of the seven upregulated ESTs showed similar differential expression in both ICC3996 and Lasseter (DEF10). DEF09 displayed upregulation in Lasseter and no change in ICC3996, DEF16 and DEF20 showed a higher level of up-regulation in Lasseter than the up-regulation observed in ICC3996, whilst DEF11, DEF18, and DEF19 showed up-regulation in ICC3996 and no significant change in Lasseter. Fig. 4 also shows that every differentially expressed EST achieved upor down-regulation by 24 h post-inoculation at the latest, and all but DEF16 had returned to normal regulation by 96 h. 3.3. Hierarchical clustering To statistically analyse the gene expression dataset and divide it into groups of similar observations, agglomerative hierarchical clustering was performed. To calculate dissimilarities between observations, the average linkage between groups and Euclidean metrics method was used, resulting in five different clusters (Fig. 5). Cluster I contains the 10 ESTs whose expression did not show any significant up- or downregulation over the time series. Cluster II includes three ESTs (DEF11, DEF18, and DEF 19) whose expression was temporarily up-regulated in ICC3996 only, before returning to baseline expression at 96 h post-inoculation. Clusters III and IV were represented by ESTs that showed downregulation (DEF08, DEF13, and DEF17), and up-regulation (DEF10 and DEF20), respectively, in both ICC3996 and Lasseter, before both clusters returned to normal regulation at the completion of the time-series. Finally, cluster V
The aim of this study was to examine the changes that occur in the transcript level of 20 previously identified A. rabiei defence-related ESTs. The ascochyta blight resistant chickpea, ICC3996, and susceptible cultivar, Lasseter, were inoculated with A. rabiei spores before extracting total RNA over a time-series. Microarray technology was used to assess the expression of the 20 defence-related ESTs in each RNA sample when compared to uninoculated control RNA samples. The use of a timeseries also enabled the putative detection of gene induction over the sampled period. The five normalisation controls included in the present study (5.8S/18S/26S rRNA, Rubisco, chloroplast 4.5S/5S/16S/23S mRNA, chlorophyll a/b, and ATP synthase) are all involved in general biochemical pathways and housekeeping activities, thus their expression levels after A. rabiei inoculation were not expected to alter. Importantly, the expression ratios of these ESTs over the times-series did not exceed a two-fold change in either direction (Fig. 2), which provided validation of the sampling and hybridisation methods used. This observation also supports the up- and down-regulation cut-off values defined by the self:self hybridisation (Fig. 1), giving increased significance to values that did show a greater than two-fold expression change. The blank buffer spots incorporated as negative controls were all automatically flagged by the scanning analysis software (Imagenee 5, Biodiscovery), a result that confirmed the desired level of hybridisation stringency. Considering that the ESTs employed in this study were functionally annotated as defence-related, it was assumed that the main observed response to A. rabiei challenge would be significant differential expression. However, 10 of the ESTs failed to display any sign of either increased or decreased expression, which may be explained by the limitation of the time-series used or by the absence of a specific role for those ESTs in the A. rabiei infection response. Additionally, it is important to note that significantly different levels of basal expression between ICC3996 and Lasseter could not be determined by this study, as hybridisations were not performed between control samples of ICC3996 and Lasseter. Thus, some of the 10 non-differentially expressed ESTs may possibly possess high constitutive expression that allows them to be effective in the A. rabiei response. Four of the non-differentially expressed ESTs; Avr9/Cf-9 rapidly elicited protein 65 (DEF03), Pathogen-induced translation initiation factor nps45 (DEF04), Transcription factor EREBP-1 (DEF07),
206
T.E. Coram, E.C.K. Pang / Physiological and Molecular Plant Pathology 66 (2005) 201–210
Fig. 4. Time-series plots of the 10 defence-related ESTs showing a differential expression ratio at one time point or more in ICC3996 (solid line) or Lasseter (broken line). Standard error bars are included for each measurement. Dashed lines at 2.00 and 0.50 expression ratio represent up- and down-regulation, respectively, in relation to control samples.
and Pathogen-induced transcription factor (DEF12) are putatively involved in signal transduction/defence-activating pathways. Considering that signal transduction follows pathogen recognition as a very early stage in a plantpathogen interaction, a possible explanation for the lack of observed expression changes for these ESTs may be that the time-series was unable to capture these changes. The earliest sampling in this study was 12 h post-inoculation, but in Arabidopsis it has been shown that the earliest
detectable changes in gene expression for an incompatible pathogen reaction are as early as 6 h post-inoculation [24]. Therefore, the use of earlier sampling points may have revealed differential expression for these ESTs. The Extensin-like disease resistance protein (DEF01) and Cinnamyl alcohol dehydrogenase (CAD1) (DEF14) were two other non-differentially expressed ESTs that are involved in cell-wall resistance. Extensins are cell-wall proteins implicated in pathogen perception [25] whilst
T.E. Coram, E.C.K. Pang / Physiological and Molecular Plant Pathology 66 (2005) 201–210
207
Fig. 5. Dissimilarity dendrogram for the expression dataset of the defence-related ESTs in ICC3996 and Lasseter, showing hierarchical clustering into five groups of significantly similar observations. The steps of the dendrogram show the combined clusters and the values of the distance coefficients at each step, where the values have been rescaled to numbers between 0 and 25, preserving the ratio of the distances between steps.
CAD1 is an enzyme of the phenylpropanoid pathway responsible for lignin biosynthesis [4]. Again, considering extensins are involved in pathogen recognition, earlier sampling may have captured differential expression. However, the unchanging CAD1 expression may be explained by another upstream enzyme of the lignin biosynthesis pathway, Caffeoyl-CoA-methyltransferase (DEF08). DEF08 was found to be down-regulated in both ICC3996 and Lasseter (Fig. 4), indicating that lignin biosynthesis was not occurring, and hence there existed no requirement for CAD1 up-regulation. The Nematode resistance protein HsPro-1 (DEF15) and Gamma-thionin/defensin protease inhibitor (DEF02) also possessed unchanged expression over the time-series. This observation was not unexpected considering that DEF15 has previously only been implicated in chickpea resistance to nematodes, and protease inhibitors are generally only involved in insect defence [5]. DEF15 and DEF02 were included in this study only on the possibility that they may represent broad-spectrum disease resistance proteins. The final two ESTs with unchanging expression were the S1-3 pathogen-induced protein (DEF05) and Putative disease resistance protein (DEF06). This result was not surprising, considering that the public database hits that these ESTs matched to were only tentatively characterised as defence-related. The previous classification of the 20 ESTs as defencerelated [21] was supported by the observation that 10 (50%) showed up- or down-regulation in ICC3996 or Lasseter for at least one time point. Furthermore, the distribution of
the number of differentially expressed ESTs (Fig. 3) showed that there existed a tendency toward up-regulation rather than down-regulation. This demonstrates that the A. rabiei inoculation provoked a significant response that could be witnessed over a wide-range of ESTs involved in various defensive pathways. Differentially expressed ESTs included Caffeoyl-CoA-methyltransferase (DEF08) that, as described earlier, forms part of the phenylpropanoid pathway involved in the biosynthesis of lignin. Pathogenesis-related protein 4A (DEF09), beta-1,3-glucanase (DEF10), and pathogenesis-related protein 10 (DEF20) are all classified as pathogenesis-related (PR) proteins. DEF09 belongs to class four chitinase-like PR proteins, DEF10 is a member of class two that act to dissolve fungal cell walls, and DEF20 is a class 10 PR protein with putative acidic and RNAse activity [26]. The putative flavonol glucosyl transferase (DEF17) also belongs to the phenylpropanoid pathway where it is involved in the production of flavonol glycosides and anthocyanins, which have been implicated in phytoalexin production [27]. The two leucine-zipper containing proteins (DEF11 and DEF13) may represent putative basic region/leucine-zipper motif (bZIP) transcription factors that are involved in the regulation of various plant processes including pathogen defence [28]. The SNAKIN2 antimicrobial precursor (DEF19) represents a putative elicitorinduced antimicrobial protein, similar to a PR protein, active against fungal pathogens [29]. Finally, both the multiresistance transporter protein (DEF16) and elicitor-induced receptor protein (DEF19) are putative defence-related proteins whose mode of action has not yet been fully
208
T.E. Coram, E.C.K. Pang / Physiological and Molecular Plant Pathology 66 (2005) 201–210
characterised, although the identification of DEF19 as a receptor protein indicates that it may form part of a signal transduction pathway. Hierarchical clustering confirmed that 10 of the 20 ESTs were not differentially expressed (Cluster I, Fig. 5), whilst five were co-regulated with either increased (Cluster IV) or decreased (Cluster III) expression in both the A. rabiei resistant (ICC3996) and susceptible (Lasseter) chickpea. Such co-regulation may imply that ESTs in these clusters are not effective in A. rabiei defence, as they cannot be used to explain the phenotypic difference between ICC3996 and Lasseter. The co-regulated ESTs displaying up-regulation were PR proteins beta-1,3-glucanase (DEF10) and Pathogenesis-related protein 10 (DEF20). Although beta-1,3-glucanase has previously been found to accumulate in chickpea after A. rabiei infection [14], that study did not compare resistant and susceptible varieties, and the present results suggest that DEF10 and DEF20 possess little effectiveness in A. rabiei resistance. The down-regulated ESTs in ICC3996 and Lasseter were Caffeoyl-CoA-methyltransferase (DEF08), Putative flavonol glucosyl transferase (DEF17), and Leucine-zipper containing protein (DEF13). Both DEF08 and DEF17 are part of the phenylpropanoid pathway that, together with the earlier described CAD1 (DEF14), does not show any sign of becoming active in response to A. rabiei infection. The deposition of lignin and formation of phytoalexins, both products of this pathway, usually occur rapidly after infection in legumes [4]. However, this pathway appears to be depressed by A. rabiei infection in chickpea, suggesting that it may not be involved in effective resistance. The leucinezipper protein (DEF13) is a bZIP transcription factor, and considering these proteins can have various regulatory roles ranging from pathogen defence to flower development, it appears DEF13 may not be involved in defence. In fact, DEF13 may have been down-regulated to allow for more efficient energy utilisation in the defence pathways, as is often observed for non-defensive proteins. Cluster V (Fig. 5) is of interest as it contains two ESTs that showed significantly higher expression in the susceptible Lasseter compared to ICC3996; Pathogenesis-related protein 4A (DEF09) and Multi-resistance transporter protein (DEF16). DEF09 is a PR protein and DEF16 is a putative uncharacterised transporter protein with potential involvement in broad resistance mechanisms. The increased expression of these proteins in Lasseter implies a lack of effectiveness in A. rabiei resistance, considering that Lasseter is highly susceptible to ascochyta blight. Overall, of the four ESTs up-regulated in Lasseter, three were PR proteins (DEF09, DEF10, DEF20) and one was a putative uncharacterised protein (DEF16). The most important group of ESTs are members of Cluster II, as these ESTs were up-regulated in the resistant ICC3996 and showed no change in Lasseter. Subsequently, these ESTs may possess an effective role in A. rabiei resistance. The first member of this cluster is the Protein with leucine-zipper (DEF11), another bZIP transcription factor. These proteins, made up of a basic region that binds
DNA and a leucine-zipper dimerization motif, have been studied in Arabidopsis where they regulate a variety of plant processes. There exists a group of the bZIP transcription factors that participate in pathogen defence, specifically by regulating the production of salicylic acid (SA) to induce the expression of PR proteins [28]. Studies of one Arabidopsis protein, NPR1, which is essential for regulating PR protein gene expression, have shown that NPR1 interacts strongly with a bZIP transcription factor [30]. Thus, the upregulation of DEF11 in ICC3996 may indicate that it is involved in activating effective defence mechanisms against A. rabiei, such as PR proteins, and future work will focus on the identification of potential ESTs that do interact with DEF11. The second member of Cluster II is a SNAKIN2 antimicrobial peptide precursor (DEF18) whose activity has been studied extensively in potato [29]. SNAKIN2 peptides are basic globular antimicrobial peptides rich in Cys residues that form stabilising disulphide bridges, but the exact mechanism of their action remains unknown. However, it is known that SNAKIN2 is induced by fungal infection of potato tubers, and that a SNAKIN2 EST has been associated with potato leaves infected with Phytophthora infestans [29]. Further, in potato, SNAKIN2 synergistically accumulates with another member of its family known as SNAKIN1, although they share only 38% sequence similarity. Subsequently, the up-regulation of SNAKIN2 in ICC3996 may also be associated with an up-regulation of a SNAKIN1-like protein. The evidence to date suggests that SNAKIN2 is involved in pathogen defence, so the up-regulation of DEF18 in ICC3996 compared to Lasseter may indicate that DEF18 has some effect in A. rabiei resistance. Future studies of DEF18 may involve the use chickpea transformation to study knockout mutants, and will focus on the potential isolation of a SNAKIN1-like EST from the previously synthesised chickpea EST library [21]. The final member of Cluster II is the Elicitor-induced receptor protein (DEF19). This protein was first isolated from A. thaliana, where it was putatively identified according to sequence structure alone [31]. DEF19 represents the only subsequent isolation of a protein matching the A. thaliana protein, thus information on the biochemical activity of DEF19 is scarce. However, the identification of DEF19 as an elicitor-induced receptor protein indicates that it may be membrane-bound, involved in signal transduction, and up-regulated by pathogensecreted elicitor molecules. The up-regulation of DEF19 in ICC3996 does support a potential role for DEF19 in effective A. rabiei resistance, but further study on this protein is required to identify the mode of its action and confirm any potential involvement in the chickpea resistance mechanism to ascochyta blight. A major observation for the differentially expressed ESTs was that they all achieved either up- or down-regulation by 12 or 24 h post-inoculation (Fig. 4). This reflects a rapid growth and proliferation of A. rabiei within the host tissues,
T.E. Coram, E.C.K. Pang / Physiological and Molecular Plant Pathology 66 (2005) 201–210
resulting in a significant host response at these time points. Other post-inoculation gene expression studies have also reported this rapid change in gene expression, such as the soybean and Phytophthora sojae interaction, where expression changes peaked at 24 h post-inoculation [19]. In the present study, the expression changes also peaked at either 24 or 48 h, and the majority of the differentially expressed ESTs returned to baseline expression by 96 h postinoculation. This shows that, for the ESTs included in this study, the potential defence mechanism against A. rabiei is likely to occur within 48 h of inoculation. In summary, this study was the first example of the use of cDNA microarrays to study the chickpea resistance response to ascochyta blight. Expression profiles were generated for 20 defence-related ESTs, leading to the identification of potentially effective, and ineffective, proteins conferring A. rabiei resistance to chickpea. The results indicate that significant differences exist between the infection response of the A. rabiei resistant (ICC3996) and susceptible (Lasseter) chickpea. In particular, ICC3996 expressed three defencerelated ESTs not observed in Lasseter, which may form part of an effective ascochyta blight resistance, and will be studied further. Additionally, these ESTs may represent novel chickpea defence-related proteins considering that they have not previously been implicated in A. rabiei resistance. Considering that the plant response to pathogen challenge is associated with massive changes in gene expression, future work will focus on the generation of large-scale cDNA microarrays that incorporate previously synthesised ESTs from a wide range of functional categories. The sampling of such a wide range of ESTs will aid in the identification of complete pathways of defence activation, and is required to illuminate the overall mechanism of resistance.
Acknowledgements The authors would like to acknowledge the Grains Research and Development Corporation (GRDC) for their generous granting of a research scholarship. The authors thank Dr Chris Pittock, Dr Trevor Bretag, and Kristy Hobson (Department of Primary Industry, Horsham) for the supply of seeds and fungal cultures, as well as the Australian Genome Research Facility (Melbourne) for the use of microarray equipment.
References [1] Devoto A, Muskett P, Shirasu K. Role of ubiquitination in the regulation of plant defence against pathogens. Curr Opin Plant Biol 2003;6:307–11. [2] Grant M, Mansfield J. Early events in host–pathogen interactions. Curr Opin Plant Biol 1999;2:312–9. [3] Van Loon L, Van Strien E. The families of pathogenesis-related proteins, their activities, and comparative analysis of PR-1 type proteins. Physiol Mol Plant Pathol 1999;55:85–97.
209
[4] Dixon R, Achnine L, Kota P, Lui C, Reddy M, Wang L. The phenylpropanoid pathway and plant defence—a genomics perspective. Mol Plant Pathol 2002;3:371–90. [5] Koiwa H, Bressan R, Hasegawa P. Regulation of protease inhibitors and plant defense. Trends Plant Sci 1997;2:379–84. [6] Food and agriculture organization (Accessed: 30 August 2004) FAO statistical databases—agricultural production. Updated: 24 May 2004 http://apps.fao.org/page/collections. [7] Nene Y. A review of ascochyta blight of chickpea. Trop Pest Manage 1982;28:61–70. [8] Collard B, Ades P, Pang E, Brouwer J, Taylor P. Prospecting for sources of resistance to ascochyta blight in wild cicer species. Australas Plant Pathol 2001;30:271–6. [9] Collard B, Pang E, Ades P, Taylor P. Preliminary investigation of QTLs associated with seedling resistance to ascochyta blight from Cicer echinospermum, a wild relative of chickpea. Theor Appl Genet 2003;107:719–29. [10] Flandez-Galvez H, Ford R, Pang E, Taylor P. An intraspecific linkage map of the chickpea (Cicer arietinum L.) genome based on the sequence tagged microsattelite site and resistance gene analog markers. Theor Appl Genet 2003;106:1447–56. [11] Cho S, Chen W, Muehlbauer F. Pathotype-specific genetic factors in chickpea (Cicer arietinum L.) for quantitative resistance to ascochyta blight. Theor Appl Genet 2004;109:733–9. [12] Ichinose Y, Tiemann K, Schwenger-Erger C, Toyoda K, Hein F, Hanselle T, et al. Genes expressed in Ascochyta rabieiinoculated chickpea plants and elicited cell cultures as detected by differential cDNA-hybridization. Z Natuforsch 2000;55c: 44–54. [13] Rajesh P, Tekeoglu M, Gupta V, Ranjekar P, Muehlbauer F. Molecular mapping and characterisation of an RGA locus RGAPtokin1-2 171 in chickpea. Euphytica 2002;128:427–33. [14] Hanselle T, Barz W. Purification and characterisation of the extracellular PR-2b (beta)-1,3-glucanase accumulating in different Ascochyta rabiei-infected chickpea (Cicer arietinum L.) cultivars. Plant Sci 2001;161:773–81. [15] Vogelsang R, Barz W. Cloning of a class III acidic chitinase from chickpea. Plant Physiol 1993;103:297–8. [16] Sarwar N, Sarwar M, Jamil F. Role of polyphenoloxidase and catalase in ascochyta blight resistance in chickpea. Pak J Bot 2003; 35:111–5. [17] Baldwin D, Crane V, Rice D. A comparison of gel-based, nylon filter and microarray techniques to detect differential RNA expression. Curr Opin Plant Biol 1999;2:96–103. [18] Rao Z, Dong H, Zhuang J, Chai R, Fan Y, Li D, et al. Analysis of gene expression profiles during host-Magnaporthe grisea interactions in a pair of near isogenic lines of rice. Acta Genet Sinica 2002;29:887–93. [19] Moy P, Qutob D, Chapman B, Atkinson I, Gijzen M. Patterns of gene expression upon infection of soybean plants by Phytophthora sojae. Mol Plant Microbe Int 2004;17:1051–62. [20] Huitema E, Vleeshouwers V, Francis D, Komoun S. Active defence responses associated with non-host resistance of Arabidopsis thaliana to the oomycete pathogen Phytophthora infestans. Mol Plant Pathol 2003;4:487–500. [21] Coram T, Pang E. Isolation and analysis of candidate ascochyta blight resistance genes in chickpea. I. Generation and analysis of an expressed sequence tag (EST) library. Physiol Mol Plant Pathol (doi: 10.1016/j.pmpp.2005.08.003) [22] Azarani A, Haefele R, Hecker KH. RNA analysis by ion-pair reversed-phase HPLC. Nat Biotechnol Short Rep 2000;11:20. [23] Brazma A, Hingamp P, Quackenbush J, Sherlock G, Spellman P, Stoeckert C, et al. Minimum information about a microarray experiment (MIAME)-towards standards for microarray data. Nat Genet 2001;29:365–71. [24] Tao Y, Xie Z, Chen W, Glazebrook J, Chang H, Han B, et al. Quantitative nature of Arabidopsis responses during compatible and
210
[25]
[26]
[27] [28]
T.E. Coram, E.C.K. Pang / Physiological and Molecular Plant Pathology 66 (2005) 201–210 incompatible interactions with the bacterial pathogen Pseudomonas syringae. Plant Cell 2003;15:317–30. Silva NF, Goring DR. The proline-rich, extensin-like receptor kinase1 (PERK1) gene is rapidly induced by wounding. Plant Mol Biol 2002;50:667–85. Elbez M, Kevers C, Hamdi S, Rideau M, Petit-Paly G. The plant pathogenesis-related PR-10 proteins. Acta Bot Gallica 2002;149: 415–44. Winkel-Shirley B. Biosynthesis of flavonoids and effects of stress. Curr Opin Plant Biol 2002;5:218–23. Jakoby M, Weisshaar B, Droge-Laser W, Vicente-Carbajosa J, Tiedemann J, Kroj T, et al. bZIP transcription factors in Arabidopsis. Trends Plant Sci 2002;7:106–11.
[29] Berrocal-Lobo M, Segura A, Moreno M, Lopez G, Garcia-Olmedo F, Molina A. Snakin-2, an antimicrobial peptide from potato whose gene is locally induced by wounding and responds to pathogen infection. Plant Physiol 2002;128:951–61. [30] Despres C, Chubak C, Rochon A, Clark R, Buthune T, Desveaux D, et al. The Arabidopsis NPR1 disease resistance protein is a novel cofactor that confers redox regulation of DNA binding activity to the basic domain/leucine zipper transcription factor TGA1. Plant Cell 2003;15:2181–91. [31] Sato S, Nakamura Y, Kaneko T, Katoh T, Asamizu E, Kotani H, et al. Structural analysis of Arabidopsis thaliana chromosome 5.X. Sequence features of the regions of 3,076,755 bp covered by sixty P1 and TAC clones. DNA Res 2000;7:31–63.
Isolation and analysis of candidate ascochyta blight defence genes in chickpea. Part II. Microarray expression analysis of putative defence-related ESTs Tristan E. Coram*, Edwin C.K. Pang School of Applied Sciences, Biotechnology and Environmental Biology, RMIT University, Building 223, Level 1, Plenty Road, Bundoora, Vic. 3083, Australia Received 28 January 2005; revised 18 July 2005; accepted 2 August 2005
Abstract Using microarray technology and previously identified defence-related ESTs from chickpea (Cicer arietinum L.), the ascochyta blight (Ascochyta rabiei (Pass.) Labrousse) resistance response was studied in a highly resistant chickpea accession (ICC3996) and a susceptible cultivar (Lasseter). The time-series expression patterns of 20 defence-related ESTs were studied after inoculation with A. rabiei. Ten of the defence-related ESTs displayed up- or down-regulation in ICC3996 and/or Lasseter compared to uninoculated control samples. Hierarchical clustering grouped the ESTs into clusters of similar observations, revealing that three defence related ESTs showed differential up-regulation in ICC3996 when compared to Lasseter—a leucine zipper protein, SNAKIN2 antimicrobial peptide precursor, and elicitor-induced receptor protein. The potential involvement of these ESTs in effective chickpea defence against ascochyta blight is discussed. The microarray expression analysis of defence-related ESTs in chickpea represents the first study of this type, and future work will focus on large-scale expression studies aimed at further elucidating the function of genes involved in ascochyta blight resistance and the pathway of their action. q 2005 Elsevier Ltd. All rights reserved. Keywords: Chickpea; Ascochyta blight; Genomics; EST; Microarray
1. Introduction Plants deal with a vast range of pathogens, and effective disease resistance stems from the ability of the host to recognise pathogens and then initiate defence mechanisms. Common defence responses include the hypersensitive response (HR), the oxidative burst to produce reactive oxygen intermediates, and the delayed accumulation of salicylic acid necessary for systemic acquired resistance (SAR) [1]. Transcription factors play an integral role in the signalling and control of these pathways, which are also mediated by plant hormones (jasmonic acid and ethylene), elevation of cytosolic calcium, and activation of protein kinases [2]. Additionally, host defences also involve structural alterations such as the strengthening of cell
* Corresponding author. Tel.: C61 3 9925 7140; fax: C61 3 9925 7110. E-mail address: [email protected] (T.E. Coram).
0885-5765/$ - see front matter q 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.pmpp.2005.08.002
walls by phenolic compounds, and/or the production of defensive compounds including pathogenesis-related (PR) proteins [3], phenylpropanoids [4], and protease inhibitors [5] to name a few. At the genomic level, plant defence responses are complex and diverse, and every gene involved in the defence response, from recognition to signalling to direct involvement, forms part of a coordinated response network. Chickpea (Cicer arietinum L.) is the third most important pulse crop in the world [6], but a major factor limiting world production is a severe and destructive fungal disease known as ascochyta blight caused by Ascochyta rabiei (Pass.) Lab. [7]. Currently, the level of resistance in cultivated chickpeas is not sufficient to withstand disease pressure under conditions that are favourable to A. rabiei. However, the world collection of chickpea germplasm contains resistant accessions, such as ICC3996, that remain uncultivated due to poor agronomic characteristics, but do possess a strong capacity for A. rabiei resistance [8]. Understanding of the chickpea defence response to ascochyta blight at the genomic level is a prerequisite for
202
T.E. Coram, E.C.K. Pang / Physiological and Molecular Plant Pathology 66 (2005) 201–210
developing resistant cultivars. Until now the methods for studying the chickpea defence response have been limited to mapping of quantitative trait loci (QTL) associated with A. rabiei resistance [9–11], differential screening of cDNA libraries [12], and mapping of resistance gene analogs to existing linkage maps [13]. Several biochemical studies have also been performed that implicate certain compounds in the chickpea defence response, such as pathogenesisrelated proteins beta-1,3-glucanase [14], chitinase [15], as well as polyphenyloxidase and catalase [16]. The disadvantage of previous studies is that the overall coordinated defence response remains largely uncharacterised, but quantitative methods for global and simultaneous analysis of expression profiles, through microarray analysis, can improve the overall understanding of the coordinated defence response at a molecular level. In fact, microarray analysis has been successful in studying the defence responses of plants such as maize [17], rice [18], soybean [19] and Arabidopsis thaliana [20]. The availability of a set of chickpea ESTs [21] enables the development of efficient and accurate methods of gene expression profiling, including the identification of genes whose expression is changed by a biological phenomenon, such as disease pressure, to suggest functional involvement. The expression pattern of several genes may also be used as an indicator representing a certain state of a cell or tissue, such as resistance or susceptibility to a disease. DNA microarrays are powerful tools for comprehensive characterisation of different plant processes, such as pathogen defence, at the transcription level. In this study, chickpea ESTs functionally classified as having potential involvement in plant defence [21] were employed in microarray experiments. The aim was to generate EST expression profiles over a time-series, after inoculation with A. rabiei spores, in the ascochyta blight resistant ICC3996 chickpea accession and a susceptible chickpea cultivar known as Lasseter.
2. Materials and methods 2.1. Plant material & fungal isolate Seeds of C. arietinum (ICC3996) and C. arietinum (Lasseter) were obtained from the Australian Temperate Field Crops Collection (ATFCC) in Horsham, Victoria, Australia. Seven isolates of A. rabiei collected by Coram and Pang [21] were prepared and maintained on V8 agar according to the method of Collard et al. [8]. Spore suspensions were prepared according to the method of Coram and Pang [21]. 2.2. Cultivation, inoculation & RNA extraction Thirty plants each of ICC3996 and Lasseter were cultivated and inoculated in a glasshouse (20G4 8C)
according to the method of Coram and Pang [21] with three plants potK1 and 10 replicates, of which two replicates served as an uninoculated controls. About 500 mg of stem/ leaf tissue was extracted from all inoculated plants at each of 12, 24, 48, and 96 h post-inoculation. Additionally, stem and leaf tissue samples were also taken from uninoculated plants of ICC3996 and Lasseter at each time point. To confirm that A. rabiei infection had been effective, plants were checked for expected disease symptoms at 14 days post-inoculation based on known infection reactions. For each biological replicate, total RNA was extracted from pooled stem and leaf samples for ICC3996 and Lasseter at each time point (including control samples) using the RNeasyw Plant Mini Kit (QIAGENe). The quantity and quality of the total RNA samples were assessed by OD260/ OD280 ratios and gel electrophoresis, respectively. Total RNA quality was also confirmed by ion-pair reversed-phase High Performance Liquid Chromatography (HPLC) [22], where degraded samples were detected by the absence of sharp elution peaks. 2.3. Microarray construction From a previously synthesised and characterised EST collection constructed from cDNA of the ascochyta blight resistant ICC3996 chickpea accession [21], 20 nonredundant ESTs classified as defence-related were used for the construction of microarrays according to minimum information about a microarray experiment guidelines (MIAME) [23]. In total, the cDNA of 25 ESTs was used after the selection of five housekeeping ESTs as internal normalisation controls (Table 1). The cDNA inserts of the 25 ESTs were amplified to 2000 ng, purified with multiscreen-PCR plates (Millipore), resuspended in 8 mL 50% (v/v) dimethylsulphoxide:water at 250 ng/mL, and arrayed onto amino-silanized slides (Corning) using a Virtek Chipwiter (Virtek Vision International, Inc.) with one pin. Blank buffer spots were also incorporated as negative controls. Each element was deposited in duplicate with a volume of approximately 6 nL and diameter of 200 mm. Slides were pre-hybridised by blocking in 5! SSC, 0.1% SDS, 25% Formamide, 1% BSA for 45 min at 42 8C, rinsed in distilled water and dried. 2.4. Microarray probe preparation & hybridisation Total RNA of 50 mg from each sample was reverse transcribed using Superscript II reverse transcriptase (Invitrogen) and oligo(dT) 23mer (Invitrogen). Aminoallyl dUTP (Sigma) was incorporated during the reverse transcription process. Briefly, the RNA and 5 mg of oligo(dT) primer were denatured at 70 8C for 10 min and cooled on ice before adding first strand buffer to a final concentration of 1!, aa-dUTP/dNTPs mix (final concentrations of 0.5 mM dATP, 0.5 mM dGTP, 0.5 mM dCTP, 0.2 mM dTTP, 0.3 mM aa-dUTP), DTT to a final
T.E. Coram, E.C.K. Pang / Physiological and Molecular Plant Pathology 66 (2005) 201–210
203
Table 1 Identity of the 25 ESTs used for microarray construction, where CON01 to CON05 are normalisation controls and DEF01 to DEF20 represent 20 unique defence-related ESTs EST
Putative identity after BLASTX and BLASTN searches
Category
GenBank dbEST Accession
CON01 CON02 CON03 CON04 CON05 DEF01 DEF02 DEF03 DEF04 DEF05 DEF06 DEF07 DEF08 DEF09 DEF10 DEF11 DEF12 DEF13 DEF14 DEF15 DEF16 DEF17 DEF18 DEF19 DEF20
5.8S/18S/26S ribosomal RNA Ribulose 1,5-bisphosphate carboxylase small subunit Chloroplast 4.5S/5S/16S/23S messenger RNA Chlorophyll a/b binding protein ATP Synthase C chain (EC 3.6.1.34) Extensin-like disease resistance protein Gamma-thionin defensin/protease inhibitor Avr9/Cf-9 rapidly elicited protein 65 Pathogen-induced translation initiation factor nps45 S1-3 pathogen-induced protein Putative disease resistance protein Transcription factor EREBP-1 Caffeoyl-CoA-methyltransferase (EC 2.1.1.104) Pathogenesis-related protein 4A Beta-1-3-glucanase Protein with leucine zipper Pathogen-induced transcription factor Leucine-zipper containing protein Cinnamyl alcohol dehydrogenase (CAD1) Nematode resistance protein Hs1pro-1 Multi-resistance transporter protein Putative flavonol glucosyl transferase SNAKIN2 antimicrobial peptide precursor Elicitor-induced receptor protein Pathogenesis-related protein class 10
Control Control Control Control Control Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence Defence
N/A N/A N/A N/A N/A CV793587 CV793588 CV793589 CV793590 CV793591 CV793593 CV793594 CV793595 CV793597 CV793598 CV793599 CV793600 CV793601 CV793602 CV793603 CV793605 CV793607 CV793608 CV793609 CV793610
concentration of 10 mM, and 150 units Superscript II in a total reaction volume of 30 mL. Reverse transcription was carried out at 42 8C for 2 h before hydrolysis of RNA template with NaOH for 15 min at 65 8C and neutralisation with HEPES (pH 7.0). The cDNA was purified and postlabelled using a PCR purification kit (QIAGENe) and Cy3/Cy5 mono-NHS esters (Amersham) resuspended in 0.1 M Na2CO3 (pH 9.0). cDNA samples were applied to QIAquicke columns and washed/dried according to manufacturers instructions, before adding the appropriate resuspended CyDye to the column membrane and incubating for 1 h at room temperature in the dark. Following incubation, labelled samples were eluted, appropriate Cy3 and Cy5 probes were combined, and purification was repeated. Purified combined probes were resuspended in 2! hybridisation buffer (5! SSC, 0.2% SDS, 50% formamide, 25 mg Cot1 DNA (Invitrogen), 0.4 mg polyA (Sigma), 0.5 mg salmon sperm DNA (Sigma), made up to 60.0 mL with sterile water) and applied to the array after denaturation at 100 8C for 2 min. The slide was covered by a 60!25 mm Lifter slip (Grale Scientific) and incubated in a 42 8C water bath for 16–20 h in a waterproof and humidified hybridisation chamber (Corning) in the dark. Each hybridisation was performed in triplicate and two biological replicates were performed for each hybridisation, incorporating dyeswapping (i.e. reciprocal labelling of Cy3 and Cy5) to eliminate any dye bias.
2.5. Scanning & data analysis After hybridisation, slides were washed for 5 min in each of 1! SSC/0.2% SDS and 0.1! SSC/0.2% SDS, and twice for 2 min in 0.1! SSC. Slides were scanned at 532 nm (Cy3 green laser) and 660 nm (Cy5 red laser) at 10 mm resolution using an Affymetrixw 428e array scanner, and captured with the Affymetrixw Jaguare software (v. 2.0). Image analysis was performed using Imagenee 5 (Biodiscovery) image analysis software. Grids were predefined and manually positioned over the image to ensure optimal spot recognition. Spots with high intensity due to dust particles or other artefacts were manually flagged. Spots were individually quantified using the fixed circle method; sample values were measured as the mean of pixels within the spot circle and the local background in a five-pixel diameter ring that began five pixels outside the spot circle. Automatic flagging eliminated empty spots, negative spots (signal mean!background mean), and poor spots (contaminated background, ignored pixelsO25%, open perimeterO25%, offset from expected positionO60%). Spots with mean signal intensity less than two times the local background were also manually flagged, as well as duplicate spots on the same subgrid with significantly different signal means. Quantified spot data was then compiled and transformed using GeneSighte 3 (Biodiscovery). Data transformations consisted of a background correction (median intensity of specified blank spots subtracted from
204
T.E. Coram, E.C.K. Pang / Physiological and Molecular Plant Pathology 66 (2005) 201–210
signal intensity of all other spots), omitting flagged spots, creating a Cy5/Cy3 mean signal ratio, normalisation by dividing the ratio values of all spots by the mean ratio of the internal normalisation controls whose ratio is expected to equal one, and combination of duplicated spot data. Following data transformation, expression ratios were used for hierarchical cluster analysis, performed with SPSSw version 13.0 (SPSS, Inc., Chicago, IL) using average distance linkage between groups and Euclidean metrics. To determine the significant cut-offs for up- or downregulation, a separate replicated hybridisation was performed using identical total RNA for both Cy3 and Cy5 labelling. Subsequently, ESTs showing up- or down-regulation at one or more time points in ICC3996 or Lasseter were subjected to time-series analysis.
3. Results 3.1. Microarray construction & hybridisation Chickpea cDNA microarrays containing 25 ESTs, of which 20 consisted of previously characterised defencerelated ESTs and the remaining five comprised internal normalisation controls (Table 1), were constructed. Each EST was printed in duplicate spots, and used to study the expression patterns of defence-related ESTs in an A. rabiei resistant (ICC3996) and susceptible (Lasseter) chickpea over a time-series after inoculation with A. rabiei. Each hybridisation was made in reference to an uninoculated control, was made in triplicate, and biologically replicated twice for each time point. The transcript level for each cDNA was calculated firstly as the average intensity of the duplicated spots, then the average intensity of the triplicate hybridisations, and finally the average intensity of the two replicates. The result of the separate replicated hybridisation to determine the significant cut-offs for up- or down-regulation yielded a scatter plot with all spots lying within a two-fold difference range (Fig. 1), therefore, cDNAs were regarded as differentially expressed where they showed a greater than two-fold increase or decrease compared to control samples. These cut-offs translate into up-regulated cDNAs having a ratio S2.0, and down-regulated cDNAs%0.5. Additionally, the distribution of the ratio data for the five normalisation controls in every hybridisation for both ICC3996 and Lasseter also resulted in a scatter plot where all spots lay within a two-fold difference range (Fig. 2).
Fig. 1. Scatter plot of mean signal intensity (Cy3 v. Cy5) from the self:self hybridisation. Broken lines show the two-fold difference range from equal ratio (solid line).
points. Fig. 3, representing the number of up- or downregulated ESTs in both ICC3996 and Lasseter at each time point, shows that differential expression peaked 24 h postinoculation, and the majority of ESTs had returned to normal regulation by 96 h post-inoculation. Additionally, Fig. 3 shows that the main response to A. rabiei inoculation amongst the differentially expressed ESTs was up-regulation (19 instances) in favour of down-regulation (9 instances). The 10 differentially expressed cDNAs were subjected to time-series analysis to independently compare expression levels in ICC3996 and Lasseter (Fig. 4). Of the 10 ESTs, seven were involved in up-regulation (DEF09, DEF10, DEF11, DEF16, DEF18, DEF19, and DEF20),
3.2. Time-series analysis Of the 20 specific defence-related ESTs included in the microarray, 10 exhibited differential expression in at least one time point of either ICC3996 or Lasseter compared to uninoculated control samples. The remaining 10 ESTs failed to show differential expression over the sampled time
Fig. 2. Scatter plot of mean signal intensity (Cy3 vs. Cy5) for the five normalisation controls over every hybridisation for ICC3996 and Lasseter. Broken lines show the two-fold difference range from equal ratio (solid line).
T.E. Coram, E.C.K. Pang / Physiological and Molecular Plant Pathology 66 (2005) 201–210
205
contained two ESTs that displayed significant up-regulation in Lasseter for the all or most of the time-series, and either brief up-regulation in ICC3996 (DEF16), or unchanged regulation in ICC3996 (DEF09).
4. Discussion
Fig. 3. Distribution of the number of differentially expressed ESTs in both ICC3996 and Lasseter over the time-series, where up-regulated ESTs are shaded in grey and down-regulated in black.
whilst three were down-regulated (DEF08, DEF13, and DEF17). Importantly, the three down-regulated ESTs showed the same pattern of reduced expression in both ICC3996 and Lasseter. However, only one of the seven upregulated ESTs showed similar differential expression in both ICC3996 and Lasseter (DEF10). DEF09 displayed upregulation in Lasseter and no change in ICC3996, DEF16 and DEF20 showed a higher level of up-regulation in Lasseter than the up-regulation observed in ICC3996, whilst DEF11, DEF18, and DEF19 showed up-regulation in ICC3996 and no significant change in Lasseter. Fig. 4 also shows that every differentially expressed EST achieved upor down-regulation by 24 h post-inoculation at the latest, and all but DEF16 had returned to normal regulation by 96 h. 3.3. Hierarchical clustering To statistically analyse the gene expression dataset and divide it into groups of similar observations, agglomerative hierarchical clustering was performed. To calculate dissimilarities between observations, the average linkage between groups and Euclidean metrics method was used, resulting in five different clusters (Fig. 5). Cluster I contains the 10 ESTs whose expression did not show any significant up- or downregulation over the time series. Cluster II includes three ESTs (DEF11, DEF18, and DEF 19) whose expression was temporarily up-regulated in ICC3996 only, before returning to baseline expression at 96 h post-inoculation. Clusters III and IV were represented by ESTs that showed downregulation (DEF08, DEF13, and DEF17), and up-regulation (DEF10 and DEF20), respectively, in both ICC3996 and Lasseter, before both clusters returned to normal regulation at the completion of the time-series. Finally, cluster V
The aim of this study was to examine the changes that occur in the transcript level of 20 previously identified A. rabiei defence-related ESTs. The ascochyta blight resistant chickpea, ICC3996, and susceptible cultivar, Lasseter, were inoculated with A. rabiei spores before extracting total RNA over a time-series. Microarray technology was used to assess the expression of the 20 defence-related ESTs in each RNA sample when compared to uninoculated control RNA samples. The use of a timeseries also enabled the putative detection of gene induction over the sampled period. The five normalisation controls included in the present study (5.8S/18S/26S rRNA, Rubisco, chloroplast 4.5S/5S/16S/23S mRNA, chlorophyll a/b, and ATP synthase) are all involved in general biochemical pathways and housekeeping activities, thus their expression levels after A. rabiei inoculation were not expected to alter. Importantly, the expression ratios of these ESTs over the times-series did not exceed a two-fold change in either direction (Fig. 2), which provided validation of the sampling and hybridisation methods used. This observation also supports the up- and down-regulation cut-off values defined by the self:self hybridisation (Fig. 1), giving increased significance to values that did show a greater than two-fold expression change. The blank buffer spots incorporated as negative controls were all automatically flagged by the scanning analysis software (Imagenee 5, Biodiscovery), a result that confirmed the desired level of hybridisation stringency. Considering that the ESTs employed in this study were functionally annotated as defence-related, it was assumed that the main observed response to A. rabiei challenge would be significant differential expression. However, 10 of the ESTs failed to display any sign of either increased or decreased expression, which may be explained by the limitation of the time-series used or by the absence of a specific role for those ESTs in the A. rabiei infection response. Additionally, it is important to note that significantly different levels of basal expression between ICC3996 and Lasseter could not be determined by this study, as hybridisations were not performed between control samples of ICC3996 and Lasseter. Thus, some of the 10 non-differentially expressed ESTs may possibly possess high constitutive expression that allows them to be effective in the A. rabiei response. Four of the non-differentially expressed ESTs; Avr9/Cf-9 rapidly elicited protein 65 (DEF03), Pathogen-induced translation initiation factor nps45 (DEF04), Transcription factor EREBP-1 (DEF07),
206
T.E. Coram, E.C.K. Pang / Physiological and Molecular Plant Pathology 66 (2005) 201–210
Fig. 4. Time-series plots of the 10 defence-related ESTs showing a differential expression ratio at one time point or more in ICC3996 (solid line) or Lasseter (broken line). Standard error bars are included for each measurement. Dashed lines at 2.00 and 0.50 expression ratio represent up- and down-regulation, respectively, in relation to control samples.
and Pathogen-induced transcription factor (DEF12) are putatively involved in signal transduction/defence-activating pathways. Considering that signal transduction follows pathogen recognition as a very early stage in a plantpathogen interaction, a possible explanation for the lack of observed expression changes for these ESTs may be that the time-series was unable to capture these changes. The earliest sampling in this study was 12 h post-inoculation, but in Arabidopsis it has been shown that the earliest
detectable changes in gene expression for an incompatible pathogen reaction are as early as 6 h post-inoculation [24]. Therefore, the use of earlier sampling points may have revealed differential expression for these ESTs. The Extensin-like disease resistance protein (DEF01) and Cinnamyl alcohol dehydrogenase (CAD1) (DEF14) were two other non-differentially expressed ESTs that are involved in cell-wall resistance. Extensins are cell-wall proteins implicated in pathogen perception [25] whilst
T.E. Coram, E.C.K. Pang / Physiological and Molecular Plant Pathology 66 (2005) 201–210
207
Fig. 5. Dissimilarity dendrogram for the expression dataset of the defence-related ESTs in ICC3996 and Lasseter, showing hierarchical clustering into five groups of significantly similar observations. The steps of the dendrogram show the combined clusters and the values of the distance coefficients at each step, where the values have been rescaled to numbers between 0 and 25, preserving the ratio of the distances between steps.
CAD1 is an enzyme of the phenylpropanoid pathway responsible for lignin biosynthesis [4]. Again, considering extensins are involved in pathogen recognition, earlier sampling may have captured differential expression. However, the unchanging CAD1 expression may be explained by another upstream enzyme of the lignin biosynthesis pathway, Caffeoyl-CoA-methyltransferase (DEF08). DEF08 was found to be down-regulated in both ICC3996 and Lasseter (Fig. 4), indicating that lignin biosynthesis was not occurring, and hence there existed no requirement for CAD1 up-regulation. The Nematode resistance protein HsPro-1 (DEF15) and Gamma-thionin/defensin protease inhibitor (DEF02) also possessed unchanged expression over the time-series. This observation was not unexpected considering that DEF15 has previously only been implicated in chickpea resistance to nematodes, and protease inhibitors are generally only involved in insect defence [5]. DEF15 and DEF02 were included in this study only on the possibility that they may represent broad-spectrum disease resistance proteins. The final two ESTs with unchanging expression were the S1-3 pathogen-induced protein (DEF05) and Putative disease resistance protein (DEF06). This result was not surprising, considering that the public database hits that these ESTs matched to were only tentatively characterised as defence-related. The previous classification of the 20 ESTs as defencerelated [21] was supported by the observation that 10 (50%) showed up- or down-regulation in ICC3996 or Lasseter for at least one time point. Furthermore, the distribution of
the number of differentially expressed ESTs (Fig. 3) showed that there existed a tendency toward up-regulation rather than down-regulation. This demonstrates that the A. rabiei inoculation provoked a significant response that could be witnessed over a wide-range of ESTs involved in various defensive pathways. Differentially expressed ESTs included Caffeoyl-CoA-methyltransferase (DEF08) that, as described earlier, forms part of the phenylpropanoid pathway involved in the biosynthesis of lignin. Pathogenesis-related protein 4A (DEF09), beta-1,3-glucanase (DEF10), and pathogenesis-related protein 10 (DEF20) are all classified as pathogenesis-related (PR) proteins. DEF09 belongs to class four chitinase-like PR proteins, DEF10 is a member of class two that act to dissolve fungal cell walls, and DEF20 is a class 10 PR protein with putative acidic and RNAse activity [26]. The putative flavonol glucosyl transferase (DEF17) also belongs to the phenylpropanoid pathway where it is involved in the production of flavonol glycosides and anthocyanins, which have been implicated in phytoalexin production [27]. The two leucine-zipper containing proteins (DEF11 and DEF13) may represent putative basic region/leucine-zipper motif (bZIP) transcription factors that are involved in the regulation of various plant processes including pathogen defence [28]. The SNAKIN2 antimicrobial precursor (DEF19) represents a putative elicitorinduced antimicrobial protein, similar to a PR protein, active against fungal pathogens [29]. Finally, both the multiresistance transporter protein (DEF16) and elicitor-induced receptor protein (DEF19) are putative defence-related proteins whose mode of action has not yet been fully
208
T.E. Coram, E.C.K. Pang / Physiological and Molecular Plant Pathology 66 (2005) 201–210
characterised, although the identification of DEF19 as a receptor protein indicates that it may form part of a signal transduction pathway. Hierarchical clustering confirmed that 10 of the 20 ESTs were not differentially expressed (Cluster I, Fig. 5), whilst five were co-regulated with either increased (Cluster IV) or decreased (Cluster III) expression in both the A. rabiei resistant (ICC3996) and susceptible (Lasseter) chickpea. Such co-regulation may imply that ESTs in these clusters are not effective in A. rabiei defence, as they cannot be used to explain the phenotypic difference between ICC3996 and Lasseter. The co-regulated ESTs displaying up-regulation were PR proteins beta-1,3-glucanase (DEF10) and Pathogenesis-related protein 10 (DEF20). Although beta-1,3-glucanase has previously been found to accumulate in chickpea after A. rabiei infection [14], that study did not compare resistant and susceptible varieties, and the present results suggest that DEF10 and DEF20 possess little effectiveness in A. rabiei resistance. The down-regulated ESTs in ICC3996 and Lasseter were Caffeoyl-CoA-methyltransferase (DEF08), Putative flavonol glucosyl transferase (DEF17), and Leucine-zipper containing protein (DEF13). Both DEF08 and DEF17 are part of the phenylpropanoid pathway that, together with the earlier described CAD1 (DEF14), does not show any sign of becoming active in response to A. rabiei infection. The deposition of lignin and formation of phytoalexins, both products of this pathway, usually occur rapidly after infection in legumes [4]. However, this pathway appears to be depressed by A. rabiei infection in chickpea, suggesting that it may not be involved in effective resistance. The leucinezipper protein (DEF13) is a bZIP transcription factor, and considering these proteins can have various regulatory roles ranging from pathogen defence to flower development, it appears DEF13 may not be involved in defence. In fact, DEF13 may have been down-regulated to allow for more efficient energy utilisation in the defence pathways, as is often observed for non-defensive proteins. Cluster V (Fig. 5) is of interest as it contains two ESTs that showed significantly higher expression in the susceptible Lasseter compared to ICC3996; Pathogenesis-related protein 4A (DEF09) and Multi-resistance transporter protein (DEF16). DEF09 is a PR protein and DEF16 is a putative uncharacterised transporter protein with potential involvement in broad resistance mechanisms. The increased expression of these proteins in Lasseter implies a lack of effectiveness in A. rabiei resistance, considering that Lasseter is highly susceptible to ascochyta blight. Overall, of the four ESTs up-regulated in Lasseter, three were PR proteins (DEF09, DEF10, DEF20) and one was a putative uncharacterised protein (DEF16). The most important group of ESTs are members of Cluster II, as these ESTs were up-regulated in the resistant ICC3996 and showed no change in Lasseter. Subsequently, these ESTs may possess an effective role in A. rabiei resistance. The first member of this cluster is the Protein with leucine-zipper (DEF11), another bZIP transcription factor. These proteins, made up of a basic region that binds
DNA and a leucine-zipper dimerization motif, have been studied in Arabidopsis where they regulate a variety of plant processes. There exists a group of the bZIP transcription factors that participate in pathogen defence, specifically by regulating the production of salicylic acid (SA) to induce the expression of PR proteins [28]. Studies of one Arabidopsis protein, NPR1, which is essential for regulating PR protein gene expression, have shown that NPR1 interacts strongly with a bZIP transcription factor [30]. Thus, the upregulation of DEF11 in ICC3996 may indicate that it is involved in activating effective defence mechanisms against A. rabiei, such as PR proteins, and future work will focus on the identification of potential ESTs that do interact with DEF11. The second member of Cluster II is a SNAKIN2 antimicrobial peptide precursor (DEF18) whose activity has been studied extensively in potato [29]. SNAKIN2 peptides are basic globular antimicrobial peptides rich in Cys residues that form stabilising disulphide bridges, but the exact mechanism of their action remains unknown. However, it is known that SNAKIN2 is induced by fungal infection of potato tubers, and that a SNAKIN2 EST has been associated with potato leaves infected with Phytophthora infestans [29]. Further, in potato, SNAKIN2 synergistically accumulates with another member of its family known as SNAKIN1, although they share only 38% sequence similarity. Subsequently, the up-regulation of SNAKIN2 in ICC3996 may also be associated with an up-regulation of a SNAKIN1-like protein. The evidence to date suggests that SNAKIN2 is involved in pathogen defence, so the up-regulation of DEF18 in ICC3996 compared to Lasseter may indicate that DEF18 has some effect in A. rabiei resistance. Future studies of DEF18 may involve the use chickpea transformation to study knockout mutants, and will focus on the potential isolation of a SNAKIN1-like EST from the previously synthesised chickpea EST library [21]. The final member of Cluster II is the Elicitor-induced receptor protein (DEF19). This protein was first isolated from A. thaliana, where it was putatively identified according to sequence structure alone [31]. DEF19 represents the only subsequent isolation of a protein matching the A. thaliana protein, thus information on the biochemical activity of DEF19 is scarce. However, the identification of DEF19 as an elicitor-induced receptor protein indicates that it may be membrane-bound, involved in signal transduction, and up-regulated by pathogensecreted elicitor molecules. The up-regulation of DEF19 in ICC3996 does support a potential role for DEF19 in effective A. rabiei resistance, but further study on this protein is required to identify the mode of its action and confirm any potential involvement in the chickpea resistance mechanism to ascochyta blight. A major observation for the differentially expressed ESTs was that they all achieved either up- or down-regulation by 12 or 24 h post-inoculation (Fig. 4). This reflects a rapid growth and proliferation of A. rabiei within the host tissues,
T.E. Coram, E.C.K. Pang / Physiological and Molecular Plant Pathology 66 (2005) 201–210
resulting in a significant host response at these time points. Other post-inoculation gene expression studies have also reported this rapid change in gene expression, such as the soybean and Phytophthora sojae interaction, where expression changes peaked at 24 h post-inoculation [19]. In the present study, the expression changes also peaked at either 24 or 48 h, and the majority of the differentially expressed ESTs returned to baseline expression by 96 h postinoculation. This shows that, for the ESTs included in this study, the potential defence mechanism against A. rabiei is likely to occur within 48 h of inoculation. In summary, this study was the first example of the use of cDNA microarrays to study the chickpea resistance response to ascochyta blight. Expression profiles were generated for 20 defence-related ESTs, leading to the identification of potentially effective, and ineffective, proteins conferring A. rabiei resistance to chickpea. The results indicate that significant differences exist between the infection response of the A. rabiei resistant (ICC3996) and susceptible (Lasseter) chickpea. In particular, ICC3996 expressed three defencerelated ESTs not observed in Lasseter, which may form part of an effective ascochyta blight resistance, and will be studied further. Additionally, these ESTs may represent novel chickpea defence-related proteins considering that they have not previously been implicated in A. rabiei resistance. Considering that the plant response to pathogen challenge is associated with massive changes in gene expression, future work will focus on the generation of large-scale cDNA microarrays that incorporate previously synthesised ESTs from a wide range of functional categories. The sampling of such a wide range of ESTs will aid in the identification of complete pathways of defence activation, and is required to illuminate the overall mechanism of resistance.
Acknowledgements The authors would like to acknowledge the Grains Research and Development Corporation (GRDC) for their generous granting of a research scholarship. The authors thank Dr Chris Pittock, Dr Trevor Bretag, and Kristy Hobson (Department of Primary Industry, Horsham) for the supply of seeds and fungal cultures, as well as the Australian Genome Research Facility (Melbourne) for the use of microarray equipment.
References [1] Devoto A, Muskett P, Shirasu K. Role of ubiquitination in the regulation of plant defence against pathogens. Curr Opin Plant Biol 2003;6:307–11. [2] Grant M, Mansfield J. Early events in host–pathogen interactions. Curr Opin Plant Biol 1999;2:312–9. [3] Van Loon L, Van Strien E. The families of pathogenesis-related proteins, their activities, and comparative analysis of PR-1 type proteins. Physiol Mol Plant Pathol 1999;55:85–97.
209
[4] Dixon R, Achnine L, Kota P, Lui C, Reddy M, Wang L. The phenylpropanoid pathway and plant defence—a genomics perspective. Mol Plant Pathol 2002;3:371–90. [5] Koiwa H, Bressan R, Hasegawa P. Regulation of protease inhibitors and plant defense. Trends Plant Sci 1997;2:379–84. [6] Food and agriculture organization (Accessed: 30 August 2004) FAO statistical databases—agricultural production. Updated: 24 May 2004 http://apps.fao.org/page/collections. [7] Nene Y. A review of ascochyta blight of chickpea. Trop Pest Manage 1982;28:61–70. [8] Collard B, Ades P, Pang E, Brouwer J, Taylor P. Prospecting for sources of resistance to ascochyta blight in wild cicer species. Australas Plant Pathol 2001;30:271–6. [9] Collard B, Pang E, Ades P, Taylor P. Preliminary investigation of QTLs associated with seedling resistance to ascochyta blight from Cicer echinospermum, a wild relative of chickpea. Theor Appl Genet 2003;107:719–29. [10] Flandez-Galvez H, Ford R, Pang E, Taylor P. An intraspecific linkage map of the chickpea (Cicer arietinum L.) genome based on the sequence tagged microsattelite site and resistance gene analog markers. Theor Appl Genet 2003;106:1447–56. [11] Cho S, Chen W, Muehlbauer F. Pathotype-specific genetic factors in chickpea (Cicer arietinum L.) for quantitative resistance to ascochyta blight. Theor Appl Genet 2004;109:733–9. [12] Ichinose Y, Tiemann K, Schwenger-Erger C, Toyoda K, Hein F, Hanselle T, et al. Genes expressed in Ascochyta rabieiinoculated chickpea plants and elicited cell cultures as detected by differential cDNA-hybridization. Z Natuforsch 2000;55c: 44–54. [13] Rajesh P, Tekeoglu M, Gupta V, Ranjekar P, Muehlbauer F. Molecular mapping and characterisation of an RGA locus RGAPtokin1-2 171 in chickpea. Euphytica 2002;128:427–33. [14] Hanselle T, Barz W. Purification and characterisation of the extracellular PR-2b (beta)-1,3-glucanase accumulating in different Ascochyta rabiei-infected chickpea (Cicer arietinum L.) cultivars. Plant Sci 2001;161:773–81. [15] Vogelsang R, Barz W. Cloning of a class III acidic chitinase from chickpea. Plant Physiol 1993;103:297–8. [16] Sarwar N, Sarwar M, Jamil F. Role of polyphenoloxidase and catalase in ascochyta blight resistance in chickpea. Pak J Bot 2003; 35:111–5. [17] Baldwin D, Crane V, Rice D. A comparison of gel-based, nylon filter and microarray techniques to detect differential RNA expression. Curr Opin Plant Biol 1999;2:96–103. [18] Rao Z, Dong H, Zhuang J, Chai R, Fan Y, Li D, et al. Analysis of gene expression profiles during host-Magnaporthe grisea interactions in a pair of near isogenic lines of rice. Acta Genet Sinica 2002;29:887–93. [19] Moy P, Qutob D, Chapman B, Atkinson I, Gijzen M. Patterns of gene expression upon infection of soybean plants by Phytophthora sojae. Mol Plant Microbe Int 2004;17:1051–62. [20] Huitema E, Vleeshouwers V, Francis D, Komoun S. Active defence responses associated with non-host resistance of Arabidopsis thaliana to the oomycete pathogen Phytophthora infestans. Mol Plant Pathol 2003;4:487–500. [21] Coram T, Pang E. Isolation and analysis of candidate ascochyta blight resistance genes in chickpea. I. Generation and analysis of an expressed sequence tag (EST) library. Physiol Mol Plant Pathol (doi: 10.1016/j.pmpp.2005.08.003) [22] Azarani A, Haefele R, Hecker KH. RNA analysis by ion-pair reversed-phase HPLC. Nat Biotechnol Short Rep 2000;11:20. [23] Brazma A, Hingamp P, Quackenbush J, Sherlock G, Spellman P, Stoeckert C, et al. Minimum information about a microarray experiment (MIAME)-towards standards for microarray data. Nat Genet 2001;29:365–71. [24] Tao Y, Xie Z, Chen W, Glazebrook J, Chang H, Han B, et al. Quantitative nature of Arabidopsis responses during compatible and
210
[25]
[26]
[27] [28]
T.E. Coram, E.C.K. Pang / Physiological and Molecular Plant Pathology 66 (2005) 201–210 incompatible interactions with the bacterial pathogen Pseudomonas syringae. Plant Cell 2003;15:317–30. Silva NF, Goring DR. The proline-rich, extensin-like receptor kinase1 (PERK1) gene is rapidly induced by wounding. Plant Mol Biol 2002;50:667–85. Elbez M, Kevers C, Hamdi S, Rideau M, Petit-Paly G. The plant pathogenesis-related PR-10 proteins. Acta Bot Gallica 2002;149: 415–44. Winkel-Shirley B. Biosynthesis of flavonoids and effects of stress. Curr Opin Plant Biol 2002;5:218–23. Jakoby M, Weisshaar B, Droge-Laser W, Vicente-Carbajosa J, Tiedemann J, Kroj T, et al. bZIP transcription factors in Arabidopsis. Trends Plant Sci 2002;7:106–11.
[29] Berrocal-Lobo M, Segura A, Moreno M, Lopez G, Garcia-Olmedo F, Molina A. Snakin-2, an antimicrobial peptide from potato whose gene is locally induced by wounding and responds to pathogen infection. Plant Physiol 2002;128:951–61. [30] Despres C, Chubak C, Rochon A, Clark R, Buthune T, Desveaux D, et al. The Arabidopsis NPR1 disease resistance protein is a novel cofactor that confers redox regulation of DNA binding activity to the basic domain/leucine zipper transcription factor TGA1. Plant Cell 2003;15:2181–91. [31] Sato S, Nakamura Y, Kaneko T, Katoh T, Asamizu E, Kotani H, et al. Structural analysis of Arabidopsis thaliana chromosome 5.X. Sequence features of the regions of 3,076,755 bp covered by sixty P1 and TAC clones. DNA Res 2000;7:31–63.