• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of plntphysLink to Publisher's site
Plant Physiol. Apr 2010; 152(4): 2053–2066.
Published online Feb 19, 2010. doi:  10.1104/pp.109.151829
PMCID: PMC2850024

Nonhost Resistance of Barley to Different Fungal Pathogens Is Associated with Largely Distinct, Quantitative Transcriptional Responses1,[W][OA]

Abstract

Nonhost resistance protects plants against attack by the vast majority of potential pathogens, including phytopathogenic fungi. Despite its high biological importance, the molecular architecture of nonhost resistance has remained largely unexplored. Here, we describe the transcriptional responses of one particular genotype of barley (Hordeum vulgare subsp. vulgare ‘Ingrid’) to three different pairs of adapted (host) and nonadapted (nonhost) isolates of fungal pathogens, which belong to the genera Blumeria (powdery mildew), Puccinia (rust), and Magnaporthe (blast). Nonhost resistance against each of these pathogens was associated with changes in transcript abundance of distinct sets of nonhost-specific genes, although general (not nonhost-associated) transcriptional responses to the different pathogens overlapped considerably. The powdery mildew- and blast-induced differences in transcript abundance between host and nonhost interactions were significantly correlated with differences between a near-isogenic pair of barley lines that carry either the Mlo wild-type allele or the mutated mlo5 allele, which mediates basal resistance to powdery mildew. Moreover, during the interactions of barley with the different host or nonhost pathogens, similar patterns of overrepresented and underrepresented functional categories of genes were found. The results suggest that nonhost resistance and basal host defense of barley are functionally related and that nonhost resistance to different fungal pathogens is associated with more robust regulation of complex but largely nonoverlapping sets of pathogen-responsive genes involved in similar metabolic or signaling pathways.

Nonhost resistance usually is defined as durable resistance of all known genotypes of a plant species to all known races or isolates of a pathogen species. Although this definition does not cover all known cases of nonhost resistance that can also operate at the subspecies level, such as formae speciales (f. sp.) of Blumeria graminis, it reflects a low level of genotype dependence on the efficiency of resistance. Despite its extreme importance for natural plant populations as well as its promises for agriculture, nonhost resistance is only beginning to be better understood and remains widely unexploited in agricultural practice to date (Ellis, 2006; Schweizer, 2007). The analysis of mutants of Arabidopsis (Arabidopsis thaliana) has led to the identification of several genes that contribute to nonhost resistance against the barley powdery mildew fungus B. graminis f. sp. hordei (Bgh; Collins et al., 2003; Lipka et al., 2005; Stein et al., 2006). These results led to the hypothesis of multilayered nonhost resistance in plants, with the plant cell wall being the first and rapid cell death the second line of defense. In wild-type plants, inappropriate pathogens to which Arabidopsis is a nonhost are usually stopped at the preinvasive stage of penetration. This penetration resistance is associated with the formation of large cell wall appositions (papillae) enriched in callose, lignin-like material, and hydrogen peroxide. Upon breaching of this first defense layer, pathogen growth is stopped by a hypersensitive reaction of attacked cells, which is associated with autofluorescence and a hydrogen peroxide burst and which leads to cell death (Schweizer, 2007). Recently, it has been shown that genes of Arabidopsis identified to play an important role in nonhost resistance against powdery mildews also contribute to resistance against nonhost rust fungi such as Phakopsora pachyrhizi (Loehrer et al., 2008). Despite progress made in the model plant Arabidopsis, one of the main obstacles to a better understanding of the genes and pathways underlying nonhost resistance is the lack of genetically tractable systems segregating for this type of resistance, which often operates at the species level.

Barley (Hordeum vulgare subsp. vulgare) has been reported to be a nonhost to the wheat powdery mildew (PM) fungus B. graminis f. sp. tritici (Bgt), the wheat leaf rust (Rust) fungus Puccinia triticina (Ptrit), and isolate CD180 (CD) from the genus Magnaporthe (Mag) that is associated with the host Pennisetum species (Tosa and Shishiyama, 1984; Hoogkamp et al., 1998; Zellerhoff et al., 2006). In all three nonhost systems, the incompatibility appears to be based mostly on the first layer of defense (i.e. penetration resistance; Hoogkamp et al., 1998; Trujillo et al., 2004; Zellerhoff et al., 2006). This highly effective nonhost resistance contrasts sharply with the susceptibility of many barley genotypes (such as cv Ingrid selected for this study) to the corresponding appropriate host pathogens of the same genera, such as Bgh, Puccinia hordei (Phor), and Magnaporthe oryzae (TH). In the nonhost interactions of barley with inappropriate rust fungi, a better understanding of the genetic basis of nonhost resistance was achieved recently by accumulating susceptibility alleles in a series of consecutive crosses, which resulted in two barley lines with essentially full susceptibility to nonhost rusts (Atienza et al., 2004). Segregation analysis of a progeny resulting from crosses between “normal” nonhost-resistant parents and one of the new nonhost-susceptible lines led to the identification of a number of quantitative trait loci for nonhost resistance (Jafary et al., 2006). However, the map position of these quantitative trait loci varied greatly depending on (1) the nonhost rust species and (2) the genotype of the nonhost-resistant parent. It was concluded, therefore, that nonhost resistance, at least to rust fungi, might depend on a complex and functionally redundant set of genes in barley.

Here, we describe the transcriptional responses of barley during host/nonhost interactions with three different fungal pathogens that all share a lifestyle with at least an initial biotrophic phase but that differ in the mode of penetration and the leaf tissue they feed from. The powdery mildew pathogens Bgh and Bgt exclusively attack epidermal tissue and, in the case of host susceptibility, grow on the leaf surface, where they sporulate abundantly approximately 5 d after initial host contact. The rust fungi Ptrit and Phor exclusively attack mesophyll tissue, and compatible isolates grow in the mesophyll until they breach the epidermis in order to release urediniospores. The blast fungi CD and TH first attack the leaf epidermis and, in susceptible interactions, invade the underlying mesophyll approximately 48 h after initial host contact. While powdery mildews and rust are biotropic fungi, blast is considered as a hemibiotroph. Thus, blast grows on susceptible hosts in the first penetrated epidermal cell without any sign of cell death, which in turn is caused by the release of toxins during mesophyll colonization. The results provide a comparative view of transcriptional events that are associated with nonhost resistance in barley.

RESULTS

Experimental Setup for Transcript Profiling

Three pairs of host-nonhost interactions were used for this study (Table I). The time range for the analysis was selected in order to study the initial phases of the interactions until the completion of conditioning toward accessibility or resistance. All experiments were planned in a nonrandomized split-plot design, as shown in Supplemental Figure S1. For the interactions of barley with Mag and Rust, peeled abaxial epidermis and the entire leaf were used for RNA isolation, respectively. For the interaction with PM, peeled epidermis and the remaining leaf tissue containing approximately 75% of nonpeeled epidermis were collected separately.

Table I.
Pairs of host/nonhost interactions of fungal pathogens with barley analyzed

Transcript Profiling

For the study presented here, the in-house-produced barleyPGRC1 macroarray carrying 10,450 spotted cDNA features was used (Schweizer, 2008). This array was enriched in cDNAs from powdery mildew-attacked epidermis and, therefore, well suited for the type of experiments presented here. It also contains approximately 2,000 spotted unigenes (nonredundant EST singletons or contigs) from attacked epidermis that are not represented by a community resource, the Barley1 genome chip (Close et al., 2004) from Affymetrix. However, the design of the array did not cause any bias for signal detection in leaf epidermis only and was therefore as well suited for inoculation experiments with PM that grows in the epidermis as for experiments with Rust that grows exclusively in inner leaf tissue (Supplemental Fig. S2). Technical replication experiments, as shown in Supplemental Figure S3A, revealed a high degree of data reproducibility. We also hybridized probes from biological replicates onto different barleyPGRC1 macroarray membranes or onto the Barley1 chip (Supplemental Fig. S3, B and C). The correlation of the identified gene regulation events induced by Bgh attack in barley leaf epidermis was similar in both the intraplatform and the interplatform comparisons. This demonstrates the suitability of the barleyPGRC1 array for the detection of gene regulation events upon pathogen attack.

Barley plants for inoculation with PM, Mag, or Rust were grown under different conditions, which were adapted for each interaction (for details, see “Materials and Methods”). In order to estimate the influence of the different growth chamber or greenhouse environments on the overall transcriptome, a principal component analysis (PCA) of samples from noninoculated control plants was performed (Supplemental Fig. S4). This showed that tissue type had the largest effect (PC1 explaining 56% of variation). The epidermal samples for Mag and PM inoculations clustered together with respect to PC1 (tissue type), whereas they were separated by PC2. This separation might indicate the effect of the mock treatment for Mag inoculation (spray with detergent-containing solution). Entire-leaf samples were more similar according to PC2 but showed some shift along the PC1 axis, although with considerable overlap between PM and Rust experiments. Because of the shift of entire-leaf control samples along the PC1 axis, we also performed PCA of the regulation factors of transcript abundance of all spotted unigenes on the array, not only the ones exhibiting statistically significant regulation by pathogen inoculations. This approach was expected to give a robust estimation of the overall comparability of pathogen-induced transcriptional changes because, per pathosystem, approximately 40% of the spotted unigenes showed a pathogen-induced change in transcript abundance by a factor of at least 1.5 (data not shown). PCA of the regulation factors showed that PM- and Rust-inoculated leaf samples clustered together at specific time points, indicating comparability of overall transcriptional responses in the two pathosystems (Supplemental Fig. S5). This conclusion is supported by the analysis of overlap of significantly regulated genes between PM and Rust (see below).

Over all analyzed pathosystems, a total of 1,686 spotted features (PCR products) of the barleyPGRC1 cDNA array produced differential signals by a factor of at least 2.0 between control and inoculated samples, with a q value of q < 0.05 corresponding to a false-discovery rate of less than 5%. These spotted features correspond to 1,667 unigenes, because some were spotted as duplicates from independent PCRs at different positions on the array for control purposes. A total of 93 spotted features of the EST library “HO,” which were found to produce signals exclusively in PM-inoculated samples, were eliminated from the analysis. These were likely to represent fungal transcripts, because the HO library was prepared from PM-inoculated leaf epidermis. The fungal identities of most of the 93 removed feature sequences were confirmed using BLAST (data not shown). The remaining eliminated unigenes did not result in significant BLAST results to sequences from any organism and were removed because fungal origin could not be excluded. A complete list of unigenes corresponding to differentially accumulating transcripts upon pathogen attack can be found in Supplemental Table S1.

PCA of the 1,667 unigenes that correspond to differentially accumulating transcripts in response to pathogen attack revealed a major influence of tissue type (PC1 explaining 59% of variation) followed by treatment effects (14% of data variation; Fig. 1). In both epidermis and entire leaf samples, the corresponding host/nonhost pairs of interactions induced similar changes in transcript profiles that resulted in similar shifts of sample mean values along the axes of the two components. In the epidermis, attack by TH or CD induced fewer transcriptional changes at early time points during the (pre)penetration phase of the fungus, whereas in entire leaf samples, attack by Rust fungi induced fewer changes at the latest time point analyzed (48 h after inoculation), demonstrating the transient nature of the transcriptional response even in the susceptible interaction. This suggests that Mag produced fewer pathogen-associated molecular patterns (PAMPs) that were perceived by barley during the early interaction compared with PM, whereas Phor might have suppressed defense responses efficiently by effector molecules as soon as first haustoria were established. Similar to our observations, a remarkably low number of pathogen-regulated genes were observed early during the interaction of rice (Oryza sativa) with Mag (99 out of 21,500 analyzed genes; Vergne et al., 2007). Very little information is currently available for barley-Rust interactions without clear evidence for strong suppression of defense gene expression during compatible interactions (Neu et al., 2003).

Figure 1.
PCA of transcript profiles of different plant inoculations and tissues. Mean values from three to four biological replicates of the 1,667 unigenes corresponding to differentially accumulating transcripts upon pathogen attack were used for the analysis. ...

Transcriptional changes induced by two pairs of adapted and nonadapted pathogens were analyzed in the epidermis and entire leaf, respectively, and were compared within each pair and between pairs. For this purpose, the overlap of transcripts fulfilling the statistical criterion for differential accumulation (greater than 2-fold change in transcript abundance between control and corresponding treatment with q < 0.05) was compared, as shown in Figure 2. The host and nonhost blast fungi (Mag) induced transcriptional changes in barley epidermis that overlapped considerably with the responses to PM: approximately 15% of PM-regulated and approximately 40% of Mag-regulated genes were responding to attack by both pathogens (Fig. 2B). An even higher degree of overlap (approximately 42% of Rust-regulated and 70% of PM-regulated genes) was found in entire leaf samples responding to either PM or Rust, despite the fact that the inner leaf tissues, which contributed to approximately 95% of extracted RNA (Zierold et al., 2005), were in direct contact with the Rust fungi but responded only indirectly to PM, whose development is restricted to the epidermis. Thus, the two obligate biotrophic pathogens PM and Rust affected the barley transcriptome in a more similar way compared with PM and Mag. In contrast, the overlap of the nonhost-specific transcriptional responses was clearly less between the different pathogens (Fig. 2, gray areas and columns).

Figure 2.
Nonhost resistance of barley to fungal pathogens is associated with largely nonoverlapping sets of pathogen-responsive transcripts. A, Venn diagram of transcripts differentially regulated during the interaction of barley with PM and Mag (left panel) or ...

Effect of the Nonhost Status on Transcriptional Responses

In order to identify those transcripts that might accumulate differentially during host and corresponding nonhost interactions, inoculated samples from resistant nonhost interactions and the matching inoculated samples from corresponding susceptible host interactions were compared in a pairwise manner using the set of 1,667 genes with significantly pathogen-regulated transcript abundance (see above). This resulted in the identification of 70 and 195 differentially accumulating transcripts in the epidermis during host and nonhost Mag and PM interactions, respectively, that met the chosen significance thresholds (P < 0.05 and q < 0.1; Supplemental Table S2). During Rust interactions, no significant transcript differences between host and nonhost interactions were found, despite the fact that the initial Venn analysis suggested the existence of interaction-specific differences also upon attack by these pathogens (data not shown). Thus, tissue complexity of the whole-leaf samples and the corresponding complex overlap of tissue-specific transcriptional profiles may have prevented the identification of significant interaction-specific differences that may be quantitative rather than qualitative. This interpretation was confirmed by the analysis of interaction-specific differences between Bgh- and Bgt-attacked whole-leaf samples. In contrast to the above-mentioned 195 identified unigenes in epidermis, only four differentially accumulating transcripts were identified in whole-leaf samples (data not shown). This demonstrates the importance of using peeled epidermis as a method to reduce biological complexity for the identification of rather subtle effects on the barley transcriptome. Figure 3 shows the results of a hierarchical clustering of interaction-specific transcripts during the interactions of barley epidermis with Mag or PM fungi. From this type of analysis, it became clear that both nonhost interactions were characterized by a stronger up- or down-regulation of transcript levels compared with corresponding host interactions.

Figure 3.
Differentially regulated transcripts between host and nonhost interactions. A hierarchical clustering of pathogen-regulated transcripts was performed that showed a significant (P < 0.05, q < 0.1) quantitative difference of expression in ...

The observed, more pronounced up- or down-regulation of transcript levels during nonhost interactions in the epidermis was in agreement with the higher number of statistically significant gene regulation events during nonhost interactions with Bgt and CD, as compared with host interactions with Bgh and TH (Fig. 2A). Such quantitative changes can also be displayed by calculating the differential index (DI) of transcript abundance. DI of host versus nonhost interactions is defined as

An external file that holds a picture, illustration, etc.
Object name is pp151829E01_LW.jpg

(where t = time post inoculation and Int = normalized signal intensity) and provides an integrative, robust, and normalized measure of differences in the abundance of individual transcripts. DI varies between −1 and 1, whereas the lowest and highest values are obtained if a transcript is only detected during the host and nonhost interaction, respectively. Using DI, a major quantitative effect of the mlo resistance allele on the host transcriptome of PM-attacked barley epidermis was previously described (Zierold et al., 2005). Figure 4 shows that DI of most up-regulated, interaction-specific transcripts (Fig. 3; Supplemental Table S2) was positive, which means that they are triggered to a higher abundance during the nonhost interaction. On the other hand, DI was mostly negative in the case of genes with down-regulated transcript abundance, which means that the expression of these was repressed more strongly during the nonhost interaction. Therefore, attack by the inappropriate Bgt or CD isolates induced higher amplitudes of transcriptional changes compared with the corresponding host interactions. A selection of unigenes associated with absolute DI values larger than 0.2 and 0.3 for Mag and PM interactions, respectively, is shown in Table II.

Figure 4.
Stronger regulation of transcript abundance in nonhost-resistant interactions. Distribution of the DI for the interaction-specific marker transcripts identified in barley epidermis (according to Supplemental Table S2) is shown. The transcripts were grouped ...
Table II.
Selection of transcripts showing most differential accumulation between host and nonhost interactions

The analysis of overlapping gene sets suggests strong differences in the transcriptional response to different nonhost pathogens (Fig. 2). This, however, does not exclude the possibility that among the genes with significantly regulated transcript levels in response to both pathogen genera of a specific comparison (inside the thick border lines in Fig. 2A), similar quantitative differences between host and nonhost interactions exist, which would be reflected in a similar distribution of DI values. Therefore, we analyzed the correlation of DI values of PM, Mag, and Rust interactions. As shown in Table III, there was a highly significant positive correlation of DI between PM interactions in the epidermis and entire leaf samples, suggesting a similar, (non)host-specific response of the different tissues to PM attack. By contrast, no correlation of DI values was found by comparing PM-to-Mag and PM-to-Rust interactions in the epidermis and entire leaf samples, respectively. In agreement with this observation, only four of 195 and 70 nonhost-marker transcripts of the PM and Mag interaction, respectively, were overlapping, and these four transcripts showed opposite regulation trends (Supplemental Table S2). In conclusion, the nonhost status of barley for the different fungal pathogens was reflected by clearly distinct signatures of the transcriptional response.

Table III.
Dissimilarity of transcript profiles of barley during different host/nonhost pairs of interactions

Regulation of Functional Groups of Genes

Are levels of transcripts belonging to specific functional groups of genes preferentially up- or down-regulated in response to attack by host or nonhost pathogens? This question was addressed using the recently introduced binning system of MapMan Barley (Sreenivasulu et al., 2008). Here, only the uppermost level of functional categories (superbins) was considered, because otherwise, unigene numbers per bin were too low for χ2 analysis. Figure 5 shows that several functional categories were significantly overrepresented or underrepresented among the groups of genes with pathogen-regulated transcript levels. Overrepresented among genes with up-regulated transcript abundance were “Amino acid metabolism,” “Redox regulation, ascorbate, glutathione,” “Stress,” “Transport,” and “Miscellaneous.” Braking down superbin Miscellaneous into specific bins (Supplemental Fig. S6) revealed that more than 75% of the contained unigenes belong to pathogenesis-related or stress-related multigene families. It seems, therefore, that superbin Miscellaneous represents the basal response of barley to different pathogens. Superbin “Photosynthesis” was overrepresented in a highly significant manner among the genes with down-regulated transcript levels, suggesting a shift of the leaves away from photosynthetic carbon assimilation to defense. One striking difference between the different pathogens was the highly significant overrepresentation of superbin “Lipid metabolism” in Mag-induced transcripts, whereas an overrepresentation of superbin Amino acid metabolism was seen during PM and Rust interactions. Comparing the transcripts specifically changed in abundance during nonhost interactions and the ones responding to both host and nonhost pathogens, we found no obvious differences in the overrepresentation or underrepresentation of functional categories. However, because numbers of nonhost-specific genes were generally lower, the statistical power of analysis was also reduced, thereby providing indicative rather than statistically significant results. Despite this limitation, two functional categories of genes with down-regulated transcript levels were identified that were significantly overrepresented either during host or nonhost Rust interactions. First, “Major CHO metabolism,” including starch and major carbohydrate metabolic enzymes, was more prominently down-regulated during the nonhost interaction with Ptrit. The significant overrepresentation of Major CHO metabolism during the nonhost-resistant Rust interaction is paralleled by the observation that, in Bgt-attacked epidermis, several genes of this functional category had more strongly down-regulated transcript levels (Table II). Second, during the interaction with Phor, genes belonging to “Minor CHO metabolism,” including biosynthetic enzymes of secondary cell wall building blocks such as hemicelluloses, were found to be significantly overrepresented among genes with down-regulated transcript abundance. Therefore, this fungus might cause specific repression of the expression of genes encoding biosynthetic enzymes of secondary cell wall or other minor carbohydrates, which might facilitate the penetration of mesophyll cells.

Figure 5.
Different pathogens induce similar changes in signaling or metabolic pathways. Statistical significance of the overrepresentation or underrepresentation of regulated transcripts belonging to a specific functional transcript category was calculated relative ...

In summary, during the interactions of barley with different pairs of host or nonhost pathogens, common metabolic or signaling pathways appeared to be activated or inactivated, without a clear difference between generally pathogen-responsive and nonhost-specific sets of genes.

DISCUSSION

This is a comparative study to describe the transcriptional responses of a particular barley genotype to three pairs of host/nonhost pathogens causing the important diseases of powdery mildew, rust, and blast on cereals. Although the barleyPGRC1 cDNA array used for this purpose represents not more than approximately 25% of the estimated complexity of the barley transcriptome, its complexity was sufficient for a comparative and statistical analysis of global trends of gene expression between the different pathosystems. Moreover, the barleyPGRC1 array carries approximately 2,000 unique sequences not represented by the Barley1 genome chip (Affymetrix) and was found to be specifically well suited for studying plant-pathogen interactions in epidermal tissue (Gjetting et al., 2007).

When comparing gene expression within each host/nonhost pair of pathogens by PCA and by pairwise analysis, it became clear that barley did not activate fundamentally different responses depending on the host or nonhost status of the interactions (Figs. 1 and and3).3). Rather, a complex quantitative difference of the transcriptional response was revealed using DI as a sensitive measure for differences of transcript abundance between inoculated samples from host and nonhost interactions (Fig. 4). However, during PM and Mag interactions, 195 and 70 unigenes possessed different transcript abundance when comparing samples from Bgh- versus Bgt-inoculated and TH- versus CD-inoculated epidermis, respectively (Table II; Supplemental Table S2). Unigenes of these subsets associated with absolute DI values larger than 0.2 and 0.3 for Mag and PM interactions, respectively, are briefly discussed here. Four of the most prominently accumulating marker transcripts for the nonhost interaction with CD encode lipid transfer proteins, further supporting an important, defense-related role of this protein family in barley (Molina and Garcia-Olmedo, 1997) that is also impacting malt quality and plays a role as a major food allergen (Broekaert et al., 1997; Breiteneder and Mills, 2005; Stanislava, 2007). Among the genes with more strongly down-regulated transcript levels during CD interaction are four chlorophyll a/b-binding proteins. Because barley epidermis is known to be mostly devoid of chlorophyll (except for stomatal cells), these signals might represent minor contamination of the epidermal samples by adhering mesophyll cells, detected due to the very high abundance of the corresponding transcripts. This assumption is supported by the finding that transcript levels of three of the four genes encoding chlorophyll a/b-binding proteins were also down-regulated upon PM attack in entire leaf samples (data not shown). Therefore, photosynthesis in leaf mesophyll might be more strongly suppressed during the nonhost interaction with CD than during the compatible interaction with TH, which points at a role of a senescence-related defense program (Buchanan-Wollaston et al., 2003). During the nonhost interaction with Bgt, many transcripts encoding unknown or hypothetical proteins accumulated more strongly. In addition, genes with up-regulated transcript abundance encoding polyubiquitin, an actin-related protein, as well as anthranilate phosphoribosyltransferase were markers of nonhost resistance, in agreement with suggested functions of protein ubiquitination, actin reorganization, and aromatic amino acid synthesis in basal defense of barley to PM (Dong et al., 2006; Miklis et al., 2007; Hu et al., 2009). Among the marker genes of nonhost resistance to Bgt with down-regulated transcript abundance were three genes of major carbohydrate metabolism encoding Suc synthase, 6-phosphogluconolactonase, and a mitochondrial hexokinase. This might indicate selective reduction of cellular homeostasis, which shifts cells to a predisposition to execute a programmed cell death upon further stimuli. Indeed, in the case of mitochondrial hexokinase, silencing was recently shown to induce programmed cell death in tobacco (Nicotiana tabacum; Kim et al., 2006). The relevance of the individual barley genes belonging to the host/nonhost-specific regulons remains to be further examined by direct functional assays or by genetic approaches. In the barley-PM system, a single-cell transient assay for gene silencing is available to test corresponding hypotheses in high throughput (Douchkov et al., 2005). In the Mag pathosystem, different approaches, such as virus-induced gene silencing, TILLING, and stable RNA interference, would be required in order to directly address gene function (Holzberg et al., 2002; Himmelbach et al., 2007; Weil, 2009).

When comparing gene sets with significantly altered transcript levels between the different pathosystems, we found a much higher overlap of genes that were generally pathogen responsive compared with those whose transcript levels only changed during nonhost-resistant interactions (Fig. 2). The lower overlap of significantly regulated transcript levels during nonhost interactions was in agreement with the low or absent correlation between DI values of the genes that were responding to the different host/nonhost pairs of pathogens (Table III). This phenomenon becomes very clear when comparing the overlap of genes between PM and Rust in entire leaf samples: more than 70% of genes responding to PM were also responding to Rust fungi, but only 14% of Bgt-responsive genes were also responding to Ptrit inoculation. Thus, the nonhost status of one barley genotype to three different pathogens was reflected by largely nonoverlapping, quantitative transcriptional responses.

We also addressed the question of which transcripts significantly changed in abundance during all three compatible host interactions. This revealed six and 24 transcripts with down- and up-regulated levels, respectively (Supplemental Table S3). While most of the genes with down-regulated transcript abundance were not assigned to a known function, the up-regulated group included members involved in sugar, amino acid, and phosphate mobilization and transport processes, which might indicate cooption of the corresponding genes by the successful host pathogens for nutrient uptake. In addition, transcripts of jasmonate and ethylene biosynthesis genes and of pathogenesis-related genes such as peroxidases, glutathione S-transferase, and proteinase inhibitors were induced, reflecting a general stress response of the colonized tissue.

Using the MapMan tool of barley (Sreenivasulu et al., 2008), we analyzed meta-trends of the transcriptional response in the different host/nonhost pairs of interactions and found a number of significantly overrepresented or underrepresented functional transcript categories. This revealed an astonishingly similar pattern of activated or repressed pathways or cellular responses in barley responding to the different pathogens, irrespective of their genus or host/nonhost status. One of the most pronounced, overrepresented categories of genes with up-regulated transcript levels encoded for proteins of amino acid metabolism. Breaking down of the superbin revealed that most of the encoded enzymes were involved in amino acid synthesis and that about 50% of those were synthesizing aromatic or sulfur-containing amino acids (data not shown). This probably reflects the fact that shikimate-derived amino acids such as Trp and Phe are important precursors of defensive compounds such as indole alkaloids, monolignols, and lignin-like materials and confirms previous reports on transcript profiling and gene silencing in Bgh-attacked barley (Matsuo et al., 2001; Caldo et al., 2004; Hu et al., 2009). The enhanced number of genes with up-regulated transcript abundance encoding Met and Cys may reflect an enhanced requirement of attacked tissues for activated methyl groups as cofactors for biosynthetic pathways and/or reduced sulfhydryl groups in order to control cellular redox status. Interestingly, amino acid metabolism appears not to be stimulated in Mag-attacked barley. This might suggest that papilla formation and cell wall lignification are not the most abundant defense mechanisms in Mag-attacked barley, although fluorescent papillae occurred frequently at penetration sites (Zellerhoff et al., 2006). Instead, a strong overrepresentation of accumulating transcripts involved in lipid metabolism was characteristic for this interaction, thereby separating the barley-Mag from the two other interactions and providing a first clue to the higher importance of lipid(-like) molecules in stress signaling or defense of barley epidermis attacked by blast fungi compared with PM. Several of the corresponding transcripts encode lipid transfer proteins (Supplemental Table S1), which might indicate that lipid metabolism is a key step in basal defense of plants against this pathogen. Moreover, signaling by cutin monomers as cuticle breakdown products might be important, because Cutinase2 has been shown to be involved in virulence of the fungus and because cutin monomers were found to induce genes in rice encoding lipid transfer proteins that enhanced resistance in transgenic rice to Mag upon overexpression (Patkar and Chattoo, 2006; Skamnioti and Gurr, 2007; Kim et al., 2008c). Perception of cutin monomers by several plant species including barley has been reported previously (Schweizer et al., 1996a, 1996b; Park et al., 2008). Noteworthy, analysis with a reporter gene construct fused to the promoter of a LTP1 gene in rice has shown that its expression is restricted to the lesion, suggestion a role of LTPs in the establishment of a physical barrier (Guiderdoni et al., 2002). Genes with up-regulated transcript abundance of superbin “RNA” were underrepresented, and this effect was statistically significant in entire leaf samples. Most transcripts within this superbin encode transcription factors, suggesting that most relevant transcription factors are regulated posttranslationally in pathogen-attacked barley.

Upon pathogen attack, host plants perceive PAMPs, leading to the initiation of a PAMP-triggered immunity response (Jones and Dangl, 2006). Pathogen effectors have been found to subsequently quench this response, thereby reestablishing host susceptibility. In several plants including barley, Mlo genes have been identified as negative regulators of PAMP-triggered immunity, and nonfunctional alleles of Mlo cause enhanced PAMP-triggered immunity, resulting in strong and durable resistance (Buschges et al., 1997; Consonni et al., 2006). Several lines of evidence point to a functional link of strong PAMP-triggered host immunity against Bgh mediated by recessive mlo alleles and nonhost resistance to inappropriate PM isolates (Trujillo et al., 2004; Humphry et al., 2006; Schweizer, 2007). These observations support the hypothesis that, in many cases, nonhost resistance is actually a manifestation of PAMP-triggered immunity, which is inefficiently suppressed by nonhost pathogens because these secrete nonadapted effectors that were specifically shaped during coevolution to fit targets of their corresponding host. Accordingly, nonhost resistance based on this mechanism is expected to be robust and durable. We tested the hypothesis that nonhost resistance shares similarity with PAMP-triggered immunity by performing transcript profiling experiments in a near-isogenic pair of barley differing in the allelic status of the Mlo gene (Ingrid [Mlo] versus Ingrid BC mlo5) and by comparing the distribution of DI values between pathogen-regulated transcript sets in barley exhibiting either nonhost resistance or strong PAMP-triggered immunity mediated by the mlo5 allele (for 684 transcripts differentially regulated by Bgh in the epidermis of Ingrid BC mlo5, see Supplemental Table S1). As shown in Table IV, there was indeed a highly significant correlation between DI values of the PM-responsive transcriptome of nonhost resistant plants and near-isogenic host plants differing in the absence/presence of the mlo5 resistance gene. This supports the view that nonhost resistance to PM in barley is mechanistically related to mlo-mediated host resistance. Interestingly, a highly significant correlation of DI values was also found between mlo-mediated resistance to PM and nonhost resistance to the blast fungus CD. Therefore, nonhost resistance against at least two fungal pathogens may be generally regarded as a manifestation of PAMP-triggered immunity in barley. Since there was no correlation of DI values between mlo-mediated PM resistance and nonhost resistance to the rust Ptrit, it seems that the impact of the mlo allele on effective defense is restricted to epidermal tissue.

Table IV.
Significant correlation of transcriptional changes of barley between different host/nonhost pairs of pathogens and mlo-mediated resistance against Bgh

The obvious relation of the nonhost response of barley to mlo-mediated resistance led to the model of selective suppression of host responses shown in Figure 6. In this model, compatible isolates of host pathogens selectively suppress the part of the PAMP-triggered immunity regulon via different effector molecules, which during coevolution turned out to be critical for the success or failure of the parasitic interaction. In consequence, those genes that are under host pathogen-specific suppression define the sets of transcripts with more strongly regulated levels during nonhost interactions. In accordance with our observations, these transcript sets are predicted to be largely nonoverlapping. Several effectors, such as AvrBS3 and XopD, may directly bind to target promoters affecting expression of the corresponding genes (Kay et al., 2007; Kim et al., 2008a). The complexity and mostly quantitative nature of the nonhost-specific transcriptional response, therefore, might be due to rather pleiotropic, indirect effector action. Good candidates for genes mediating such indirect effects are several WRKY transcription factors that were found to repress basal defense in barley and Arabidopsis (Eckey et al., 2004; Shen et al., 2007; Kim et al., 2008b). Despite the fact that the nonhost-specific regulons were largely nonoverlapping between the different host/nonhost interactions of barley, we observed a similar distribution of overrepresented or underrepresented functional categories between nonhost-specific regulons and generally pathogen-responsive transcripts across two or even three pathosystems. Therefore, the compatible pathogens did address similar, probably important pathways of defense by effector molecules in order to establish host susceptibility, although they suppressed the accumulation of different sets of transcripts involved in such pathways.

Figure 6.
Model of selective suppression of host defense genes by different pathogens. The observed small overlap of the nonhost-specific part of the transcriptional response of barley may be due to pathogen-specific suppression of basal defense, which is also ...

MATERIALS AND METHODS

Plant and Fungal Material

For all inoculation experiments, 7- to 10-d-old barley (Hordeum vulgare subsp. vulgare ‘Ingrid’ Mlo) seedlings were used. In addition, for PM inoculation, 7-d-old seedlings of experimental line ‘Ingrid BC mlo5’ (Buschges et al., 1997) were used.

For PM inoculation, plants were routinely grown in compost soil from IPK nursery without fertilization in a growth chamber at 20°C with 60% to 70% relative humidity and a 16-h photoperiod (216 μmol m−2 s−1). Blumeria graminis f. sp. hordei, strain CH4.8 carrying AvrMla9, was cultivated by weekly inoculation of 7-d-old seedlings of barley cv Golden Promise.

For Mag inoculation, plants were routinely grown in commercial soil ED73 (Balster Einheitserdewerk) without fertilization in a growth chamber at 18°C with 50% to 60% relative humidity and a 16-h photoperiod (207 μmol m−2 s−1). Mag isolate TH6772 of the host plant rice (Oryza sativa) was received from the Institute of Biochemistry, Facility of Agriculture, Tamagawa University. Isolate CD180 of Pennisetum was kindly provided by D. Tharreau (Centre de Coopération Internationale en Recherche Agronomique pour le Dévelopment). Both Magnaporthe isolates were alternately grown on rice leaf agar (water extract of 50 g L−1 rice leaves, 10 g L−1 water-soluble starch, 2 g L−1 yeast extract [Gerbu], and 15 g L−1 agar-agar), HSA agar (oat flake starch agar, water extract of 10 g L−1 water-soluble starch, 2 g L−1 Faex medicinales [yeast extract; Gerbu]), and potato dextrose agar (Becton Dickinson). Isolates were incubated at 24°C under black light (310–360 nm) for 2 weeks under a 16-h-day/8-h-night regime.

For Rust inoculation, plants were grown in commercial soil (Floradur B fein; Floragard) with additions of approximately 25% sand and perlite in a greenhouse cabinet at 26°C with average 70% humidity in the light and at 19°C with average 70% humidity in the dark with a photoperiod of 16 h. Natural daylight was supplemented with artificial light as soon as the external light intensity dropped below 6,000 lux. Approximately 25 seeds each of cv Ingrid barley were grown in pots (11 cm diameter) until 10 d old and fertilized using a 1:1,000 dilution of Kamasol grün 10+4+7 (Compo). The Puccinia hordei isolate I80 that is fully virulent on cv Ingrid was obtained from the Julius Kühn Institute and propagated on barley cv Astrid. Puccinia triticina (field isolate collected by BASF near Limburgerhof, Germany) was propagated on wheat (Triticum aestivum) cv Monopol.

Experimental Design

All inoculation experiments were set up as split-split-plot experiments (treatment × fungal isolate or treatment × host genotype [Mlo/mlo5]). All plots of one experiment were inoculated at the same time and harvested at the times indicated.

Plant Inoculations

Plants of cv Ingrid or near-isogenic line Ingrid BC mlo5 (Buschges et al., 1997) were inoculated with PM conidia by shaking inoculated plants over test plants in a settling tower of approximately 60 × 60 × 60 cm. Inoculations were done 1 to 2 h after the onset of light at an average density of 10 conidia mm−2 leaf area. Control plants were left nontreated. Control and inoculated plants were incubated at a constant temperature of 20°C and natural daylight (no direct sunlight) until RNA extraction.

For inoculations with Mag, conidia of both isolates were harvested from 2-week-old agar plates by rinsing with distilled water and filtering through three layers of gauze. The resulting spore suspension was diluted 1:1 (v/v) with the spraying solution (0.1% [w/v] gelatin, 0.05% [v/v] Tween 20) to a final concentration of 200,000 conidia mL−1. Primary leaves of 7-d-old plants were spray inoculated and incubated in a dark moist chamber at 26°C and 100% relative humidity. Mock-treated plants were sprayed in parallel with the same solution containing no spores.

Plants were inoculated with Rust uredospores that had previously been dehydrated for at least 4 d. Plants were dry inoculated with spores (1 g m−2) using an automated inoculation device. After application of spores, the plants were dusted with water and moved for 24 h to a dark room with 23°C/20°C day/night and 95% humidity. Mock-treated plants were only dusted with water. Afterward, plants were maintained as described for the generation of plant material.

Tissue Sampling and RNA Extraction

For transcript profiling experiments with PM and Mag, abaxial epidermis of primary leaves from control and inoculated plants was stripped 6, 12, and 24 h post inoculation and immediately frozen in liquid N2. Epidermis of all four samples to be harvested per time point (two treatments times two isolates or genotypes) was stripped in parallel in order to exclude genes that are under circadian regulation. For transcript profiling experiments with PM, the remaining leaves with approximately 50% of the adaxial epidermis and the entire abaxial epidermis still attached were also frozen in liquid nitrogen, in order to compare transcript regulation events of the epidermis and the entire leaf. The removal of approximately 25% of epidermis from leaves was considered to be negligible; therefore, no discrimination between nonpeeled leaf samples of the Rust experiments and partially peeled samples of the PM experiments was made. For transcript profiling experiments with Rust, entire leaves were sampled at 12, 24, and 48 h after inoculation and immediately frozen in liquid nitrogen.

Total RNA of the inoculations with Mag (first three experiments), PM (all experiments), and Rust (all experiments) was extracted using the acid guanidinium thiocyanate/phenol/chloroform method (Chomczynski and Sacchi, 1987). The fourth experiment of the Mag inoculation was extracted using the hot phenol method (Dudler and Hertig, 1992).

Transcript Profiling

The design and production of the barleyPGRC1 10K cDNA arrays have been described elsewhere (Schweizer, 2008). The array contains a total of 10,450 spotted PCR fragments corresponding to 10,297 unigenes. The labeling and hybridization of barley cDNA probes by [33P]dCTP was performed as described (Zierold et al., 2005). Radioactive signals on nylon membranes were detected using a Fuji BAS3000 phosphor imager.

Transcriptome Data Analysis

Spots were detected and signals were quantified using the AIDA Image Analyzer 4.08 and ArrayVision 8.0 software packages, respectively. For background subtraction, dynamic definition of background spots was used by selecting four spots producing the lowest signals per subarray and by subtracting the mean intensity of those four spots from all spots of the corresponding subarray. The rationale behind this background subtraction was the presence of four empty spots per subarray that, however, did not always produce the lowest signals, due to sporadic overshining from neighboring spots with extremely strong hybridization signals. Spot intensity values were normalized by median centering of the signal distribution per array hybridization (Sreenivasulu et al., 2002). Spotted unigenes producing signals above 2.0× local background in less than three to four hybridizations per analyzed plant-pathogen combination (corresponding to three to four biological replicates) were excluded from the analysis. Pathogen-regulated transcripts per plant-pathogen combination were identified by paired, static-match analysis and correction for multiple testing using the EDGE software (Storey and Tibshirani, 2003; http://faculty.washington.edu/jstorey/edge/). The significance thresholds for true positives and false discoveries were set to P < 0.05 and q < 0.05, respectively. Pathogen-regulated transcripts were subjected to PCA and hierarchical clustering using the MeV version 4.0 software (The Institute for Genomic Research). Settings for PCA were as follows: log2 transformation of signal intensities or regulation factors, mean centering of samples, 10 neighbors for KNN imputation. Settings for hierarchical clustering were as follows: log2 transformation of signal intensities, median centering of transcript abundance signals, Pearson's correlation, complete linkage. For the analysis of transcript patterns between host and corresponding nonhost interactions (Bgh versus Bgt, TH versus CD, Phor versus Ptrit), the set of 1,667 unigenes corresponding to differentially accumulating transcripts was subjected to a second pairwise, static-match analysis as specified above with significance thresholds of P < 0.05 and q < 0.1.

The entire transcript profiling data from this article have been deposited in the ArrayExpress database and are available under experiment identifiers E-IPKG-4 to E-IPKG-9.

Supplemental Data

The following materials are available in the online version of this article.

Supplemental Figure S1. Schematic summary of the experimental design.

Supplemental Figure S2. Distribution of the ratio of signal intensities obtained from hybridization of epidermal versus whole-leaf mRNA samples.

Supplemental Figure S3. Reproducibility of macroarray experiments.

Supplemental Figure S4. PCA of all transcript-derived signals on the barleyPGRC1 array hybridized with noninoculated control samples.

Supplemental Figure S5. PCA of regulation factors (inoculated with PM or Rust/control) of all spotted unigenes in leaf samples.

Supplemental Figure S6. Breakdown of the functional transcript category “miscellaneous” from the MapMan binning file.

Supplemental Table S1. Summary data of all regulated genes during host or nonhost interactions or during interaction of Bgh with Ingrid BC mlo5.

Supplemental Table S2. Unigenes with significantly different transcript abundances between matching host-nonhost pairs of interactions in barley epidermis.

Supplemental Table S3. Summary of genes regulated robustly during all analyzed host interactions.

Supplementary Material

[Supplemental Data]

Acknowledgments

The technical assistance of Ines Walde and Tanja Kempf is acknowledged. We thank Dr. Matthias Lange for MAGE-ML export of array primary data to ArrayExpress.

References

  • Atienza SG, Jafary H, Niks RE. (2004) Accumulation of genes for susceptibility to Rust fungi for which barley is nearly a nonhost results in two barley lines with extreme multiple susceptibility. Planta 220: 71–79 [PubMed]
  • Breiteneder H, Mills C. (2005) Nonspecific lipid-transfer proteins in plant foods and pollens: an important allergen class. Curr Opin Allergy Clin Immunol 5: 275–279 [PubMed]
  • Broekaert WF, Cammue BPA, DeBolle MFC, Thevissen K, DeSamblanx GW, Osborn RW. (1997) Antimicrobial peptides from plants. Crit Rev Plant Sci 16: 297–323
  • Buchanan-Wollaston V, Earl S, Harrison E, Mathas E, Navabpour S, Page T, Pink D. (2003) The molecular analysis of leaf senescence: a genomics approach. Plant Biotechnol J 1: 3–22 [PubMed]
  • Buschges R, Hollricher K, Panstruga R, Simons G, Wolter M, Frijters A, van Daelen R, van der Lee T, Diergarde P, Groenendijk J, et al. (1997) The barley Mlo gene: a novel control element of plant pathogen resistance. Cell 88: 695–705 [PubMed]
  • Caldo RA, Nettleton D, Wise RP. (2004) Interaction-dependent gene expression in Mla-specified response to barley powdery mildew. Plant Cell 16: 2514–2528 [PMC free article] [PubMed]
  • Chomczynski P, Sacchi N. (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162: 156–159 [PubMed]
  • Close TJ, Wanamaker SI, Caldo RA, Turner SM, Ashlock DA, Dickerson JA, Wing RA, Muehlbauer GJ, Kleinhofs A, Wise RP. (2004) A new resource for cereal genomics: 22K barley GeneChip comes of age. Plant Physiol 134: 960–968 [PMC free article] [PubMed]
  • Collins NC, Thordal-Christensen H, Lipka V, Bau S, Kombrink E, Qiu JL, Huckelhoven R, Stein M, Freialdenhoven A, Somerville SC, et al. (2003) SNARE-protein-mediated disease resistance at the plant cell wall. Nature 425: 973–977 [PubMed]
  • Consonni C, Humphry ME, Hartmann HA, Livaja M, Durner J, Westphal L, Vogel J, Lipka V, Kemmerling B, Schulze-Lefert P, et al. (2006) Conserved requirement for a plant host cell protein in powdery mildew pathogenesis. Nat Genet 38: 716–720 [PubMed]
  • Dong WB, Nowara D, Schweizer P. (2006) Protein polyubiquitination plays a role in basal host resistance of barley. Plant Cell 18: 3321–3331 [PMC free article] [PubMed]
  • Douchkov D, Nowara D, Zierold U, Schweizer P. (2005) A high-throughput gene-silencing system for the functional assessment of defense-related genes in barley epidermal cells. Mol Plant Microbe Interact 18: 755–761 [PubMed]
  • Dudler R, Hertig C. (1992) Structure of an mdr-like gene from Arabidopsis thaliana: evolutionary implications. J Biol Chem 267: 5882–5888 [PubMed]
  • Eckey C, Korell M, Leib K, Biedenkopf D, Jansen C, Langen G, Kogel KH. (2004) Identification of powdery mildew-induced barley genes by cDNA-AFLP: functional assessment of an early expressed MAP kinase. Plant Mol Biol 55: 1–15 [PubMed]
  • Ellis J. (2006) Insights into nonhost disease resistance: can they assist disease control in agriculture? Plant Cell 18: 523–528 [PMC free article] [PubMed]
  • Gjetting T, Hagedorn PH, Schweizer P, Thordal-Christensen H, Carver TLW, Lyngkjaer MF. (2007) Single-cell transcript profiling of barley attacked by the powdery mildew fungus. Mol Plant Microbe Interact 20: 235–246 [PubMed]
  • Guiderdoni E, Cordero MJ, Vignols F, Garcia-Garrido JM, Lescot M, Tharreau D, Meynard D, Ferriere N, Notteghem JL, Delseny M. (2002) Inducibility by pathogen attack and developmental regulation of the rice Ltp1 gene. Plant Mol Biol 49: 683–699 [PubMed]
  • Himmelbach A, Zierold U, Hensel G, Riechen J, Douchkov D, Schweizer P, Kumlehn J. (2007) A set of modular binary vectors for transformation of cereals. Plant Physiol 145: 1192–1200 [PMC free article] [PubMed]
  • Holzberg S, Brosio P, Gross C, Pogue GP. (2002) Barley stripe mosaic virus-induced gene silencing in a monocot plant. Plant J 30: 315–327 [PubMed]
  • Hoogkamp TJH, Chen WQ, Niks RE. (1998) Specificity of prehaustorial resistance to Puccinia hordei and to two inappropriate Rust fungi in barley. Phytopathology 88: 856–861 [PubMed]
  • Hu PS, Meng Y, Wise RP. (2009) Functional contribution of chorismate synthase, anthranilate synthase, and chorismate mutase to penetration resistance in barley-powdery mildew interactions. Mol Plant Microbe Interact 22: 311–320 [PubMed]
  • Humphry M, Consonni C, Panstruga R. (2006) mlo-based powdery mildew immunity: silver bullet or simply non-host resistance? Mol Plant Pathol 7: 605–610 [PubMed]
  • Jafary H, Szabo LJ, Niks RE. (2006) Innate nonhost immunity in barley to different heterologous Rust fungi is controlled by sets of resistance genes with different and overlapping specificities. Mol Plant Microbe Interact 19: 1270–1279 [PubMed]
  • Jones JDG, Dangl JL. (2006) The plant immune system. Nature 444: 323–329 [PubMed]
  • Kay S, Hahn S, Marois E, Hause G, Bonas U. (2007) A bacterial effector acts as a plant transcription factor and induces a cell size regulator. Science 318: 648–651 [PubMed]
  • Kim JG, Taylor KW, Hotson A, Keegan M, Schmelz EA, Mudgett MB. (2008a) XopD SUMO protease affects host transcription, promotes pathogen growth, and delays symptom development in Xanthomonas-infected tomato leaves. Plant Cell 20: 1915–1929 [PMC free article] [PubMed]
  • Kim KC, Lai ZB, Fan BF, Chen ZX. (2008b) Arabidopsis WRKY38 and WRKY62 transcription factors interact with histone deacetylase 19 in basal defense. Plant Cell 20: 2357–2371 [PMC free article] [PubMed]
  • Kim M, Lim JH, Ahn CS, Park K, Kim GT, Kim WT, Pai HS. (2006) Mitochondria-associated hexokinases play a role in the control of programmed cell death in Nicotiana benthamiana. Plant Cell 18: 2341–2355 [PMC free article] [PubMed]
  • Kim TH, Park JH, Kim MC, Cho SH. (2008c) Cutin monomer induces expression of the rice OsLTP5 lipid transfer protein gene. J Plant Physiol 165: 345–349 [PubMed]
  • Lipka V, Dittgen J, Bednarek P, Bhat R, Wiermer M, Stein M, Landtag J, Brandt W, Rosahl S, Scheel D, et al. (2005) Pre- and postinvasion defenses both contribute to nonhost resistance in Arabidopsis. Science 310: 1180–1183 [PubMed]
  • Loehrer M, Langenbach C, Goellner K, Conrath U, Schaffrath U. (2008) Characterization of nonhost resistance of Arabidopsis to the Asian soybean Rust. Mol Plant Microbe Interact 21: 1421–1430 [PubMed]
  • Matsuo H, Taniguchi K, Hiramoto T, Yamada T, Ichinose Y, Toyoda K, Takeda K, Shiraishi T. (2001) Gramine increase associated with rapid and transient systemic resistance in barley seedlings induced by mechanical and biological stresses. Plant Cell Physiol 42: 1103–1111 [PubMed]
  • Miklis M, Consonni C, Bhat RA, Lipka V, Schulze-Lefert P, Panstruga R. (2007) Barley MLO modulates actin-dependent and actin-independent antifungal defense pathways at the cell periphery. Plant Physiol 144: 1132–1143 [PMC free article] [PubMed]
  • Molina A, Garcia-Olmedo F. (1997) Enhanced tolerance to bacterial pathogens caused by the transgenic expression of barley lipid transfer protein LTP2. Plant J 12: 669–675 [PubMed]
  • Neu C, Keller B, Feuillet C. (2003) Cytological and molecular analysis of the Hordeum vulgare-Puccinia triticina nonhost interaction. Mol Plant Microbe Interact 16: 626–633 [PubMed]
  • Park JH, Suh MC, Kim TH, Kim MC, Cho SH. (2008) Expression of glycine-rich protein genes, AtGRP5 and AtGRP23, induced by the cutin monomer 16-hydroxypalmitic acid in Arabidopsis thaliana. Plant Physiol Biochem 46: 1015–1018 [PubMed]
  • Patkar RN, Chattoo BB. (2006) Transgenic indica rice expressing ns-LTP-Like protein shows enhanced resistance to both fungal and bacterial pathogens. Mol Breed 17: 159–171
  • Schweizer P. (2007) Nonhost resistance of plants to powdery mildew: new opportunities to unravel the mystery. Physiol Mol Plant Pathol 70: 3–7
  • Schweizer P. (2008) Tissue-specific expression of a defence-related peroxidase in transgenic wheat potentiates cell death in pathogen-attacked leaf epidermis. Mol Plant Pathol 9: 45–57 [PubMed]
  • Schweizer P, Felix G, Buchala A, Mueller C, Metraux JP. (1996a) Perception of free cutin monomers by plant cells. Plant J 10: 331–341
  • Schweizer P, Jeanguenat A, Whitacre D, Metraux JP, Mosinger E. (1996b) Induction of resistance in barley against Erysiphe graminis f.sp. hordei by free cutin monomers. Physiol Mol Plant Pathol 49: 103–120
  • Shen QH, Saijo Y, Mauch S, Biskup C, Bieri S, Keller B, Seki H, Ulker B, Somssich IE, Schulze-Lefert P. (2007) Nuclear activity of MLA immune receptors links isolate-specific and basal disease-resistance responses. Science 315: 1098–1103 [PubMed]
  • Skamnioti P, Gurr SJ. (2007) Magnaporthe grisea cutinase2 mediates appressorium differentiation and host penetration and is required for full virulence. Plant Cell 19: 2674–2689 [PMC free article] [PubMed]
  • Sreenivasulu N, Altschmied L, Panitz R, Hahnel U, Michalek W, Weschke W, Wobus U. (2002) Identification of genes specifically expressed in maternal and filial tissues of barley caryopses: a cDNA array analysis. Mol Genet Genomics 266: 758–767 [PubMed]
  • Sreenivasulu N, Usadel B, Winter A, Radchuk V, Scholz U, Stein N, Weschke W, Strickert M, Close TJ, Stitt M, et al. (2008) Barley grain maturation and germination: metabolic pathway and regulatory network commonalities and differences highlighted by new MapMan/PageMan profiling tools. Plant Physiol 146: 1738–1758 [PMC free article] [PubMed]
  • Stanislava G. (2007) Barley grain non-specific lipid-transfer proteins (ns-LTPs) in beer production and quality. J Inst Brew 113: 310–324
  • Stein M, Dittgen J, Sanchez-Rodriguez C, Hou BH, Molina A, Schulze-Lefert P, Lipka V, Somerville S. (2006) Arabidopsis PEN3/PDR8, an ATP binding cassette transporter, contributes to nonhost resistance to inappropriate pathogens that enter by direct penetration. Plant Cell 18: 731–746 [PMC free article] [PubMed]
  • Storey JD, Tibshirani R. (2003) Statistical significance for genomewide studies. Proc Natl Acad Sci USA 100: 9440–9445 [PMC free article] [PubMed]
  • Tosa Y, Shishiyama J. (1984) Defense reactions of barley cultivars to an inappropriate forma-specialis of the powdery mildew fungus of gramineous plants. Can J Bot 62: 2114–2117
  • Trujillo M, Troeger M, Niks RE, Kogel KH, Huckelhoven R. (2004) Mechanistic and genetic overlap of barley host and non-host resistance to Blumeria graminis. Mol Plant Pathol 5: 389–396 [PubMed]
  • Vergne E, Ballini E, Marques S, Mammar BS, Droc G, Gaillard S, Bourot S, DeRose R, Tharreau D, Notteghem JL, et al. (2007) Early and specific gene expression triggered by rice resistance gene Pi33 in response to infection by ACE1 avirulent blast fungus. New Phytol 174: 159–171 [PubMed]
  • Weil CF. (2009) TILLING in grass species. Plant Physiol 149: 158–164 [PMC free article] [PubMed]
  • Zellerhoff N, Jarosch B, Groenewald JZ, Crous PW, Schaffrath U. (2006) Nonhost resistance of barley is successfully manifested against Magnaporthe grisea and a closely related Pennisetum-infecting lineage but is overcome by Magnaporthe oryzae. Mol Plant Microbe Interact 19: 1014–1022 [PubMed]
  • Zierold U, Scholz U, Schweizer P. (2005) Transcriptome analysis of mlo-mediated resistance in the epidermis of barley. Mol Plant Pathol 6: 139–151 [PubMed]

Articles from Plant Physiology are provided here courtesy of American Society of Plant Biologists
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...