Journal of Plant Physiology & Pathology ISSN: 2329-955X

All submissions of the EM system will be redirected to Online Manuscript Submission System. Authors are requested to submit articles directly to Online Manuscript Submission System of respective journal.

Research Article, J Plant Physiol Pathol Vol: 2 Issue: 4

Spatial Regulation of Defense-Related Genes Revealed by Expression Analysis using Dissected Tissues of Rice Leaves Inoculated with Magnaporthe oryzae

Shigeru Tanabe1, Naoki Yokotani2, Toshifumi Nagata2, Yukiko Fujisawa2, Chang-Jie Jiang2, Kiyomi Abe3, Hiroaki Ichikawa3,Nobutaka Mitsuda4, Masaru Ohme-Takagi4, Yoko Nishizawa2 and Eiichi Minami2*
1Quality Control Department, Sakata Seed Corporation, 1660 Hazawa-cho, Kanagawa-ku, Yokohama, 221-0863, Japan
2Disease Resistant Crops Research Unit, Genetically-Modified Organism Research Center, National Institute of Agrobiological Sciences, 2-1-2 Kannondai, Tsukuba, 305-8602, Japan
3Division of Plant Sciences, National Institute of Agrobiological Sciences, 2-1-2 Kannondai, Tsukuba, Ibaraki 305-8602, Japan
4Bioproduction Research Institute, National Institute of Advanced Industrial Science and Technology, Higashi 1-1-1, Tsukuba 305-8566, Japan
Corresponding author : Eiichi Minami
Disease Resistant Crops Research Unit, National Insitute of Agrobiological Sciences, 2-1-2 Kannondai, Tsukuba,Japan,
Tel: 81-29-838-7912
E-mail: [email protected]
Received July 07, 2014 Accepted September 06, 2014 Published October 06, 2014
Citation: Tanabe S, Yokotani N, Nagata T, Fujisawa Y, Jiang CJ, et al. (2014) Spatial Regulation of Defense-Related Genes Revealed by Expression Analysis using Dissected Tissues of Rice Leaves Inoculated with Magnaporthe oryzae. J Plant Physiol Pathol 2:4. doi:10.4172/2329-955X.1000135


Spatial Regulation of Defense- Related Genes Revealed by Expression Analysis using Dissected Tissues of Rice Leaves Inoculated with Magnaporthe oryzae

Gene expression profiles in response to inoculation with Magnaporthe oryzae at infected and adjacent cells obtained by microarray coupled with laser microdissection (LMD) were compared with results of microarray expression profiling using RNAs from pathogen infected whole leaves (WL). Genes whose expression was up-regulated following inoculation with the fungus were classified into two categories: group A indicates those whose increased expression was detected both in LMD- and WLmicroarrays, while group B indicates genes of which the expression was detected at significantly higher levels in WL-microarray than LMD-microarray. Interestingly, genes encoding enzymes for the biosynthesis of diterpenoid phytoalexins were exclusively found in group A, while pathogenesis-related (PR) genes were found in both groups A and B.

Keywords: Defense-related genes; Spatial regulation; Rice; Magnaporthe oryzae; Microarray


Defense-related genes; Spatial regulation; Rice; Magnaporthe oryzae; Microarray


qRT-PCR: Quantitative reverse-transcription coupled with PCR


Higher plants evoke a set of defense responses upon microbial attack, accompanied by a drastic change in gene expression. These inducible defenses are triggered by the recognition of pathogens by host receptors, followed by signal transduction events, ultimately leading to the activation of a variety of highly specific defense responses, among which include, for example, the production of pathogenesis-related (PR) proteins including anti-microbial proteins and phytoalexins [1]. The expression pattern of defense-related genes is considered to be tightly linked to the activation of broader defense responses, which are spatially, as well as temporally, regulated. In dicotyledonous plants such as tobacco and Arabidopsis, systemic acquired resistance (SAR) has been extensively studied as a model of spatial regulation of defense [2]. In SAR, mobile signal substances are generated in the primary infection site, triggering accumulation of a defense hormone, salicylic acid (SA), that plays essential roles in the establishment of resistance to pathogens in the non-infected region of the inoculated leaves or non-infected leaves of the same plant. In Arabidopsis, NPR1 was shown to activate PR genes as a pivotal regulator of SA-signaling [3]. In rice, an orthologous gene of Arabidopsis NPR1, OsNPR1, is highly conserved, and has been shown to be specifically up-regulated following treatment of plants with SA [4], as well as OsWRKY45 [5], an activator of SAR in dicotyledonous plants [6]. Taken together, these observations indicate that SA plays a key role in the systemic defense responses in monocotyledonous plants. However, the role of SA in monocotyledonous plants is still undefined.
Rice and Magnaporthe oryzae is one of the best-characterized pathosystems for the study of the interaction between plant and pathogenic fungi. The resistance/susceptibility signaling cascade follows the classical gene-for-gene relationship; indeed, both host resistance (R)- and pathogen avirulence (Avr)-genes have been isolated and characterized [7,8]. As a hemi-biotrophic pathogen, during compatible interactions, M. oryzae invades its host during the early stage of host association, followed by a tightly regulated necrotrophic phase. During incompatible interactions, the epidermal cells of rice exhibit the rapid induction of a hypersensitive response, ultimately leading to cell death [9]. To date, several studies using the rice-M. oryzae interaction have described the induction and expression of host genes in response to pathogen infection using both microarray analysis [10-12], and massively parallel sequencing [13]. In each of these studies, however, sampling was performed using infected whole leaves; thus, our knowledge remains limited on the regulation of expression of defenserelated genes at cells adjacent to, or distant from, the site of infection. To address the breadth and regulation of the spatial regulation after inoculation with M. oryzae, we compared microarray results that were obtained using the RNA extracted from laser-microdissected tissues with those using whole leaf RNA. In addition, we compared the expression of a select group of defense-related genes in inoculated versus healthy tissue. Our data suggest that the spatial regulation of some of these genes is dependent on SA.

Materials and Methods

Plant materials and fungal inoculation
Two isogenic lines of rice (Oryza sativa cv. Nipponbare), NP/++ and NP/Pia, compatible and incompatible with M. oryzae (isolate P91-15B, race 001.0), respectively, were used. NP/++ is a nullsegregant from NP/Pia and is considered to be genetically identical except a blast-resistance gene Pia. Transgenic rice lines expressing nahG from Pseudomonas putida under the control of a double 35S promoter (nahG-rice) were generated using the plant expression construct previously described in Yang et al. [14].
Genetic transformation of rice (O. sativa cv Nipponbare, Pia) was done by using Agrobacterium tumefaciens strain EHA105 [15], and the transgenic lines were inoculated with P91-15B and Ina86-137 (race 007.0) as an avirulent and virulent isolate of M. oryzae, respectively. Rice seedlings were grown by hydroponics as described in Tanabe et al. [16] in a nutrient solution containing 0.5 mM NH4H2PO4, 1 mM KNO3, 0.5 mM MgSO4, 12.5 μM Fe-EDTA, 0.5 mM CaCl2 and other micronutrients [17] in a chamber under 14 h light (28°C) and 10 h dark (24°C) for 2 weeks.
M. oryzae was cultured on oatmeal agar at 30°C. Conidia suspension (1 × 106/mL) was prepared as described previously [18], and sprayed on the fourth leaf of rice plants, followed by incubation in the dark at 25°C for 24 hrs. Plants were further incubated in a chamber under 14 h light (28°C) and 10 h dark (24°C) for up to 5 days. As a control, water was sprayed on rice leaves, then, processed in the same way as above.
Isolation of total RNA from rice leaves and microarray analysis
Total RNA was extracted from the inoculated fourth leaves of rice using the RNeasy Plant Mini kit (Qiagen, cat # 74904). Cyanine3- (Cy3-) CTP labeling was carried out using a Quick Amp Labeling Kit (Agilent Technologies) and the labeled cRNAs hybridized to a rice 4 × 44k oligoarray RAP-DB (Agilent Technologies, cat # G2519F). Probe hybridization and washing was performed according to the manufacturer’s instructions. Microarray experiments were carried out three times using RNAs extracted from independently inoculated rice leaves.
Preparation of micro-tissues by laser microdissection (LMD), isolation of RNA and microarray analysis
Excised segments of inoculated leaves were embedded in paraffin using a microwave processor (Energy Beam Sciences, http://www. as described by Suwabe et al. [19] except that leaves were fixed in 99.5% cold ethanol. Paraffin-embedded longitudinal section (10 μm thick) was prepared using a microtome (Microm HM335E, Karl Zeiss Co. Ltd) and subjected to LMD using the Veritas Laser Microdissection System LCC1704 (Molecular Devices, http:// www.molecular The total area of micro-tissues was > 2 mm2 (30 to 40 segments per sample).
Total RNA was extracted and amplified using Pico-Pure RNA Isolation Kit (Molecular Devices Co. Ltd) and WT-ovation Pico RNA Amplification System (NuGEN Co Ltd., cat # 3300-12), respectively. After purification using the QIAquick PCR Purification Kit (Qiagen, cat #28104), amplified cDNAs (2 μg) were labeled with Cy3- and Cy5- CTP using Genomic DNA Enzymatic Labeling Kit (Agilent Co. Ltd.) and hybridized using rice oligoarray as described above. Microarray experiments were carried out three times using total RNA extracted from independently inoculated rice leaves.
Statistical analysis
Statistical analysis was performed using Subio platform software (Subio Inc. Japan, and R/Bioconductor (http:// Three independent plants were used for each sample as biological replicates. For comparison of whole leaf microarray results with LMD microarray results, the LMD data were converted into Cy3-probe results using Subio platform software. Signal data were log transformed and z-score normalized to zero-mean and unit-variance. The normalized expression values were further analyzed by two-way analysis of variance (ANOVA) to determine whether either factor of sample (whole leaf or micro-tissue) or inoculation (mock, resistant or susceptible to M. oryzae) had a significant effect on the expression level of each gene. Interaction P-values of multiple tests were corrected using the false discovery rate (FDR) [20]. Genes differentially expressed were selected by applying the filter of the average max/min difference being > 10 and FDR threshold = 0.01. Data were z-score transformed across each gene and presented as a heat map represented by two-dimensional hierarchical clustering. Rice genome database ( was used to annotate the genes of group A and B.
Spot-inoculation and RNA isolation
Excised fifth leaves were spot-inoculated with 5 μl of M. oryzae conidia suspension (1 × 105/mL) including 0.02% (v/v) of India ink and 0.02% (v/v) of Tween 20 at 5 mm intervals. After incubation in the dark for 24 hrs at 25°C, leaves were further incubated at 25°C under continuous light, and spot-inoculated region and noninoculated (healthy) region were separately harvested using a razor blade. Total RNA was extracted using the Sepasol-RNA isolation kit (Nakarai Kagaku Co. Ltd., cat#30486-56).
Quantification of transcripts by PCR
Total RNA was reverse-transcribed using the PrimeScript RT reagent kit (TaKaRa, cat #RR037A) and the products (ca. 8.3 ng RNA equivalent) were subjected to quantitative PCR using SYBR Premix Ex Taq™ II (TaKaRa, cat # RR081A) in Mx3000P (STRATAGENE/ Agilent). For amplified cDNAs from tissues prepared by LMD, 10 ng portion was analyzed by quantitative PCR. In these experiments, the gene for OsEF1alpha (elongation factor 1alpha, Os03g0177500) was used as an internal control. The sequences of primers are listed in Table S1.
Assessment of resistance to M. oryzae
Excised fourth leaves were sprayed with the conidia suspension of M. oryzae at 1 × 105 /mL, incubated for 5 days and the area of lesions were measured using Image J ( The extent of resistance was represented as the ratio of area of lesions to the leaf. In some experiment, the fungal biomass in rice tissues was quantified by qRT-PCR of fungal 28S rDNA, using rice EF1alpha as an internal control.
Transactivation assay in yeast
To express the fusion protein with the N-terminal GAL4 DNAbinding domain (DB), the sequences encoding the full-length amino acid sequence of OsWRKY19 was cloned into pGBKT7 (Clonetech). Saccharomyces cerevisiae strain AH109 harboring the HIS3 and LacZ reporter genes were transformed with the recombinant plasmid. HIS3 activity was assessed by conducting a viability test on a histidine-lacking medium. LacZ activity was tested by performing the β-galactosidase filter assay according to the manufacturer’s instructions.
Luciferase (LUC) reporter assay
The GUS gene in pBI221 was replaced with the coding sequence of OsWRKY19. The plasmid expressing the GAL4 DNA-binding domain (DB) was used as a control effector. As a reporter plasmid, 35S-5W-TATA-LUC-NOS (35S-5W-LUC) carrying the 5 times multimerized W-box sequence was used and assayed in suspensioncultured rice cells as described previously [21].


Tissue sampling by laser microdissection
In this study, we used two lines of rice, NP/++ and NP/Pia, which are isogenic to each other except a blast-resistance gene, Pia. These lines are indistinguishable in morphology, fertility and sensitivity to M. oryzae strains that are deficient in AvrPia. Rice blast disease symptoms typically appear on the leaf blade. Microscopic observation revealed that hyphal invasion occurs on the epidermal cells of leaves including mortar cells. However, it is technically challenging to collect a sufficient numbers of epidermal cells; microscopic observation revealed that invaded epidermal cells in the transverse section occur at a very low frequency (unpublished observation). On the other hand, mortar cells in the longitudinal section showed significant levels of fungal invasion, and thus, we excised mortar cells and flanking mesophyll cells for analysis using LMD (laser microdissection; tissue isolated by LMD will hereafter be referred to as micro-tissue). A representative photograph of the excision of microtissue at 48 hpi is shown in Figure S1, where the invaded mortar cells were visible as brown-colored area. Total RNA isolated from microtissue was estimated to be sub-nanogram quantities and were able to be amplified as cDNAs at microgram order by using Ovation Pico RNA Amplification Kit, which was reported to suitable for the linear amplification of small amount of RNA [22]. The amplified cDNAs were used for microarray.
Statistical analysis of microarray results
To compare the microarray analysis of RNAs from whole leaf and micro-tissue (hereafter, WL- and LMD-microarray, respectively), raw data was processed as described in Materials and Methods. Using two-way ANOVA analysis, we observed that 1674 genes from rice were differentially expressed (as compared to uninoculated controls) following pathogen treatment in both WL and LMD (“inoculation (responsiveness to inoculation)” but not by “sample (WL or LMD)”; Figure S2a, Table S2). Of these, 1106 were identified as up-regulated (group A), while 568 were classified as down-regulated. Similarly, the expression pattern of 1028 genes was changed after inoculation differentially between WL and LMD (“sample” × “inoculation”). Hierarchical clustering (Figure S2b, Table S3) identified 235 genes that are up-regulated in WL at higher level than in LMD-microarray (group B). Namely, the group a genes are considered to be upregulated both in the outside and inside of the micro-tissue, while group B genes are up-regulated in the outside of the micro-tissue. The original data of microarray analysis will be freely available in a database, “RiceXpro” (for LMD-coupled microarray, see NCBI-GEO accession #GSE62422) [23].
Features of group A and B genes
Figure 1 indicates the classification of genes of groups A and B according to their putative functions, and no drastic difference was observed. However, detailed analysis revealed featured distribution of genes in groups A and B. In rice plants, the major phytoalexins are diterpenoids, phytocassanes, and momilactones, and genes encoding the enzymes in the biosynthetic pathway of these phytoalexins have been characterized [24]. In the current study, all of these genes were associated with group A. Group A and B included 23 and 6 genes for WRKY transcriptional factors (Table S4) respectively. Four WRKY genes that had been reported to confer resistance to rice, OsWRKY45 [5], OsWRKY47 [12], OsWRKY53 [25] and OsWRKY71 [26], were found in group A. Among 6 WRKY genes of group B, OsWRKY76 was reported as a negative regulator of defense in rice [21], and other WRKY genes have not been analyzed with the function in rice.
Figure 1: Histograms indicating the ratio of classified genes in group A and B. Numbers in parenthesis indicate the number of each group. An asterisk in group A indicates genes for phytoalexin synthesis.
Analysis of the healthy tissue distant from the infected site
We hypothesize that both group A and B include genes that are expressed in the not-infected tissue distant from the infected site. To test the induction of gene expression in the uninfected tissue that is distant from the infected cells, we inoculated rice leaves with drops of conidia suspension of M. oryzae and extracted total RNAs from the leaf segments of healthy tissue and spot-inoculated region (Figure S3) for qRT-PCR analysis. In this experiment, the spot-inoculated regions were comprised of infected cells equivalent to the micro-tissue and the adjacent cells, in addition to healthy regions. Conidia-suspension was colored with India ink to visualize the inoculated region, which did not give adverse effects on the appearance of lesions (data not shown).
Expression of diterpenoid phytoalexin-related genes in spot-inoculated leaves
Major phytoalexins of rice, phytocassanes and momilactones, are synthesized from geranylgeranyl diphosphate (GGDP), and OsCPS2 and OsCPS4 catalyze the first reaction from GGDP toward phytocassanes and momilactones, respectively [24]. In spotinoculated leaves, very low levels of transcripts from these two genes were detected in healthy regions compared to spot-inoculated regions (Figures 2a and 2b). The common precursor, GGDP, has been reported to be generated via the methylerythritol phosphate (MEP) pathway in suspension-cultured cells of rice treated with the elicitor N-acetylchitooligosaccharide [27]. Of the seven genes required for signaling of the MEP pathway, OsDXS3 and OsCMK belong to group A, and these two genes showed similar expression patterns to those of OsCPS2 and OsCPS4 (Figures 2c and 2d). These results suggest that the activation site of MEP pathway and biosynthesis site of diterpenoid phytoalexins in response to infection with M. oryzae are primarily in the infected site.
Figure 2: Expression of group A genes related to the biosynthesis of phytoalexins and JA in spot-inoculated leaves. Excised fifth leaves of resistant (R) and susceptible (S) lines of non-trasngenic rice were spotinoculated with M. oryzae and processed as described in Materials and Methods. Total RNAs extracted from healthy (H) and spot-inoculated (H+I) regions at 24 (open bars), 36 (gray bars) and 48 (solid bars) were subjected to qRT-PCR using primers specific to the genes indicated on top of each panel. Vertical axis indicates the relative amount of transcripts to that from OsEF1alpha.
Jasmonic acid (JA) is a phytohormone involved in defense signaling, and in part, regulates the biosynthesis of momilactones [28]. A gene for allen-oxide synthase (OsAOS) was included in group A and the mode of its regulation was also similar to those of OsCPS2 and OsCPS4 (Figure 2e), implying that biosynthesis of JA occurs locally in this experimental system.
PR genes not expressing in the healthy regions in spotinoculated leaves
We examined the expression patterns of nine PR genes (2 and 7 genes from group A and B, respectively). Of these, the levels of transcripts from four genes of group B were very low in healthy regions compared to those in spot-inoculated regions (Figure 3). Therefore, the site of expressions of these genes is considered to be restricted to the infected and the adjacent cells. On the other hand, transcripts from two and three PR genes of group A and B, respectively, were detected in the healthy region, i.e., distantly up-regulated in spotinoculated leaves (Figure 4).
Figure 3: Expression of PR genes of group B from spot-inoculated leaves. Excised fifth leaves of resistant (R) and susceptible (S) lines of nontransgenic rice were spot-inoculated with M. oryzae and processed as described in Materials and Methods. Total RNAs extracted from healthy (H) and spot-inoculated (H+I) regions at 24 (open bars), 36 (gray bars) and 48 (solid bars) were subjected to qRT-PCR using primers specific to the genes indicated on top of each panel. Vertical axis indicates the relative amount of transcripts normalized to the expression of OsEF1alpha.
In dicotyledonous plants, SAR is known where PR gene expressions are observed in the healthy leaves of pathogen-inoculated plants, and treatment with SA activates PR genes [2]. To elucidate the involvement of SA in the distant expression of defense-related genes in rice, we examined the expression of these genes in the spotinoculated rice leaves transformed with nahG, which encodes SAinactivating enzyme [14]. As shown in Figure 4, the accumulation of the transcripts of these genes was reduced, yet still detectable in the healthy region of inoculated leaves. This indicates that SA plays a limited role in the expression of the five PR genes in the healthy and spot-inoculated regions.
Figure 4: Expression of PR genes of group A (a, b) and B (c, d, e) in spotinoculated leaves. Excised fifth leaves of resistant (R) and susceptible (S) lines of non-transgenic rice (NT, left boxes) or nahG-rice (nahG, right boxes) were spot-inoculated with M. oryzae as described in Materials and Methods. Total RNAs extracted from healthy (H) and spot-inoculated (H+I) regions at 24 (open bars), 36 (gray bars) and 48 (solid bars) were subjected to qRT-PCR using primers specific to the genes indicated on top of each panel. Vertical axis indicates the relative amount of transcripts to that from OsEF1alpha.
Expression of OsWRKY19, OsWRKY76 and OsWRKY45
Among the WRKY genes associated with group A and B, OsWRKY45 (group A), OsWRKY19 and OsWRKY76 (group B) are known to be responsive to acibenzolar-S-methyl (BTH) application [5]. By analogy to dicotyledonous plants where BTH induces SAR through its action on SA-signaling pathway [6], it was speculated that these WRKY genes could be induced in the distant tissue from the infection site. Therefore, we tested these three WRKY genes for distant regulation in the spot-inoculated leaves. As shown in Figure 5, transcripts from these three WRKY genes were detected in the healthy tissue of spot-inoculated leaves of non-transgenic rice, and the levels of transcripts from these three genes remarkably reduced in the healthy and spot-inoculated regions in nahG-rice, unlike the five PR genes (Figure 4). Thus, it was indicated that SA is required for the induction of these three WRKY genes in the healthy region and spot-inoculated regions.
Figure 5: Expression of OsWRKY45, OsWRKY76 and OsWRKY19 in spot-inoculated leaves. Excised fifth leaves of non-transgenic rice (NT, left boxes) or nahG-rice (nahG, right boxes) were spot-inoculated with M. oryzae of incompatible (R) or compatible (S) combinations as described in Materials and Methods. Total RNAs were extracted from healthy (H) and spot-inoculated (H+I) regions at 24 (open bars), 36 (gray bars) and 48 (solid bars) and subjected to qRT-PCR using primers specific to OsWRKY45 (Os05g0322900; a), OsWRKY76 (Os09g0417600; b) and OsWRKY19 (Os05g0571200; c). Vertical axis indicates the relative amount of transcripts to that from OsEF1alpha.
Function of OsWRKY19 in defense responses
OsRac1 is known to be a key regulator of defense in rice [29], and a transcriptional activator, RAI, was recently reported to work as a regulator of OsWRKY19 in the downstream of OsRac1 [30]. Over expression of RAI and OsRac1 conferred blast-resistance in rice, implying that OsWRKY19 also plays positive roles in the resistance of rice [30]; however, over expression of OsWRKY19 did not alter the susceptibility to M. oryzae [5].
Thus, to analyze the function of OsWRKY19 in the defense response, we generated transgenic rice plants that express a chimeric gene for the repressor of OsWRKY19 (OsWRKY19-SRDX) [31]. These transgenic plants exhibited indistinguishable growth including fertility from non-transgenic rice (data not shown). The level of expression of the chimeric gene was examined by qRT-PCR and the selected three transgenic lines expressing the chimeric gene (T2 generation, Figure S4a) were tested for resistance to M. oryzae. The fungal biomass at 5 days post-inoculation (dpi) was significantly high in the T1 lines compared to those in a non-transgenic line or a vector-control line (Figure 6a). To examine the localization of OsWRKY19 in rice cells, a fusion gene of OsWRKY19 with GFP was constructed and introduced into the epidermal cells of leaf sheaths by particle bombardment. The fluorescence was localized in nucleus, indicating that OsWRKY19 is a nuclear protein (Figure S4b). The activity as a transcriptional factor of OsWRKY19 was demonstrated by GAL4 assay and LacZ assay in Saccharomyces cerevisiae, which survived on a medium without histidine and exhibited LacZ activity (Figure 6b). Furthermore, the effector/reporter assay indicated that OsWRKY19 activates the reporter gene via recognition of W-box in rice cells (Figure 6c). These results strongly indicated that OsWRKY19 positively regulates defense response as a transcriptional activator.
Figure 6: Functional analysis of OsWRKY19. (a) Effect of functional suppression of OsWRKY19 on the blast disease resistance. Fungal growth of M. oryzae in OsWRKY19-SRDX lines at 5 dpi was quantified by the measurement of fungal genomic DNA. Values are represented as mean values ± SE for four leaf blades. Stars indicate statistical significance in Student’s T-test at P < 0.05. (b) Transactivation activity of OsWRKY19 in yeast cells shown by growth test on His-deprived medium and LacZ assay. (c) Reporter assay using LUC driven by with 5 × W-box. A horizontal axis indicates the relative LUC activity.


To date, little is known about the long-distance regulation of gene expression during defense signaling in monocotyledonous plants relative to our understanding of this process in dicotyledonous plants. In the present study, we focused on the expression pattern of defense-related genes in cells distal to the primary infection site using a combination of LMD and microarray analysis, comparing the differential expression of genes in whole leaves versus microtissue. The profiles of gene expression derived from the whole leaves are considered to be the averages of the expression levels of genes in infected and non-infected cells, while those from microtissues prepared by LMD are representative of gene expression at the infection site. Thus, we posit that by comparing the expression profiles of these two materials during pathogen infection, we can identify candidate genes that function in long-distance signaling and regulation of defense.
By comparing the expression patterns of genes classified as group A and B genes, we could not find any bias in the category of induced genes, with the exception of genes involved in the biosynthesis of diterpenoid phytoalexins, which were included in group A (Figure 1). Previous studies using spot-inoculated leaves showed that the expression of OsCPS2 and OsCPS4 are locally restricted to the infection site and its adjacent region (Figure 2). In Arabidopsis, a leaf region inoculated with a virulent strain of P. syringae DC3000 contained higher level of camalexin, a phytoalexin of Arabidopsis, than healthy tissue region of the same leaf [32]. Sesquiterpenes are major phytoalexins of tobacco, and a sesquiterpene cyclase, EAS4, is a key enzyme of the biosynthesis of sesquiterpenes in tobacco. By histochemical assay, EAS4 was shown to be active in the edge of necrotic region induced by cryptogein [33]. The level of phytocassanes was higher in the edge of than distant region from blast lesions [34]. These results, taken together with the present study, indicate that the site of biosynthesis and accumulation of phytoalexins is the peripheral region close to the infection site.
It is remarkable that the two genes of MEP pathway, OsDXS3 and OsCMK, were found in group A and their expressions were regulated in a very similar manner to that of OsCPS2 and OsCPS4 (Figure 2). OsDXS3 encodes 1-deoxy-D-xylulose-5-phosphate synthase that catalyzes a rate-limiting step of MEP pathway in higher plants [35]. Therefore, the present results indicate the elevation of MEP pathway activity through up-regulation of OsDXS3. The final product of MEP pathway, GGDP, is a key precursor in isoprenoid pathway in plants to phytohormones and photosynthetic pigments such as GA, ABA, carotenoid and chlorophyll [36], in addition to phytoalexins. These results imply that sophisticated regulations of metabolic pathway of GGDP occur in the micro-tissue and its adjacent cells.
In suspension-cultured cells of rice, the genes for diterpenoid phytoalexin biosynthesis and MEP pathway are synchronously upregulated in response to N-acetylchitooligosaccharide elicitor [27,37]. Although the molecular mechanisms of regulation of MEP pathway in response to elicitor treatment are at present unknown, it has also been reported that a master transcriptional factor, OsTGAP1, regulates the expression of phytoalexin biosynthetic genes in suspension-cultured rice cells [37]. It is tempting to speculate that synchronous regulation systems of the gene expressions in the two metabolic pathways work locally in the peripheral region of infected site of rice upon infection of M. oryzae.
The mode of expression of OsAOS (group B) encoding allen oxide synthase in the biosynthetic pathway of JA was similar to those of genes for phytoalexin biosynthesis (Figure 2). This result indicates that the site of JA biosynthesis overlaps with that of PA biosynthesis. It has previously been observed that the induction of biosynthesis of momilactones is strongly dependent on JA in suspension-cultured rice cells [38] and partially in rice leaves [28]. However, the levels of phytocassanes in JA-deficient mutants were as high as that in their parental line, thus, JA regulates only momilactones but not phytocassanes in planta. Therefore, further studies are required to elucidate the regulatory signal that triggers biosynthesis of diterpenoid phytoalexins. Some of PR genes are known to be induced by JA, including PR8 (Os11g0700900) [39] which showed a similar pattern of expression to that of OsAOS in spot-inoculated leaves (Figure 2e and 3c). These results could lead us to speculate that the site of accumulation of JA is restricted to the infected site and its adjacent cells. In fact, it has been indicated that JA accumulates in the edge of excised leaf in response to wound stress [40].
The present study showed that PR genes in rice can be classified according to the pattern of expression in spot-inoculated leaves (i.e., those expressing or non-expressing in healthy regions of spotinoculated leaves; Figure 3 and 4). In tobacco and Arabidopsis, PR1a is a representative SA-responsive gene [6]. By using a promoter/GUS construct, PR1a was observed to be expressed locally in the edge of necrotic lesions, as well as systemically in the healthy tissues between TMV-induced lesions [41]. Consistent with these observations, the expression of a PR1 gene of rice (group A, Os07g0129200) was observed as up-regulated in the healthy regions in spot-inoculated leaves (Figure 4d). Similar up-regulation of transcript levels in the healthy region were observed for OsWRKY45 (group A), OsWRKY19 and OsWRKY76 (group B). Conversely, the levels of expression of OsWRKY45 and OsWRKY76 in the healthy region, as well as in the spot-inoculated region, were severely reduced in nahG-rice (Figure 5a and 5b), while that of PR1 gene in nahG-rice was not completely suppressed, as observed for four other PR genes (Figure 4) and OsWRKY19 (Figure 5c). Therefore, the expression of these PR genes and OsWRKY19 in the distant cell from the infection site likely requires unknown factor(s) in addition to SA in rice.
It has been argued about the role of SA in rice because the endogenous level of SA in rice is constitutively high compared to that in dicotyledonous plants and remains unchanged upon pathogen attack [42]. In rice treated with inducers of resistance, probenazole and 2,6-dichloroisonicotinic acid, OsSGT1 (Os09g0518200) encoding an enzyme that catalyzes the conversion of SA to glucoside form is up-regulated, and suppression of OsSGT1 compromises the induced resistance by these chemicals, indicating that the SA-glucoside is an active form of SA [43]. Participation of the glucoside form of SA in the distant regulation of OsWRKY45 and OsWRKY76 is not clear at present, because OsSGT1 was not responsive to the inoculation with M. oryzae (data not shown).
Transgenic rice expressing OsWRKY19-SRDX exhibited compromised resistance to M. oryzae infection (Figure 6a). The SRDX form of a transcriptional factor is known to act dominantly on other factors that are functionally redundant, including itself [31], and we don’t know whether such redundant factors occur or not for OsWRKY19, thus it is strongly indicated that OsWRKY19 definitely contributes to blast-resistance. OsWRKY19 expressed in the distant region from the site of infection by M. oryzae (Figure 5c, Table S3). Because such healthy region of inoculated leaves is likely to contain low levels of diterpenoid phytoalexins as already discussed, it is speculated that the defense response regulated by OsWRKY19 might commence at late stage of infection and be uncoupled with the production of diterpenoid phytoalexins.


We are grateful to Drs. Y. Nagamura, H. Takehisa, and Y. Sato of the Rice Genome Resource Unit at National Institute of Agrobiological Sciences (NIAS) for the use of the rice microarray analysis system, technical supports and useful discussions. We give special thanks to Ms. H. Kurano-Mochizuki and Dr. N. Ishii- Minami of NIAS for technical assistance in qRT-PCR and data analysis, and Mr. T. Ishizuka and Ms. Y. Takiguchi of AIST and Mr. K. Sato of GreenSogna Inc. and Drs. T. Tsuchida-Mayama and M. Shikata, and Ms. K. Iida-Okada and Mr. A. Horikawa (NIAS) for generation of OsWRKY19-transgenic rice plants. We also thank Dr. Y. Yang (University of Arkansas, USA) and Dr. R. B. Day (Michigan State University, USA) for providing the nahG gene plasmid and critical reading of the manuscript, respectively. This work was supported by grants from the Ministry of Agriculture, Forestry, and Fisheries of Japan (Genomics for Agricultural Innovation, RTR-0003, AMR-0001 and AMR-0004). The data analysis of this work was supported in part by a project of the Ministry of Agriculture, Forestry and Fisheries of Japan (Development of Genome Information Database System for Innovation of Crop and Livestock Production).


Track Your Manuscript