• The genetic basis of nonhost resistance of barley to nonadapted formae speciales of Blumeria graminis is not known, as there is no barley line that is susceptible to these nonadapted formae speciales, such as the wheat powdery mildew pathogen, Blumeria graminis f.sp. tritici (Bgt).
• Barley accessions with rudimentary susceptibility to an isolate of the nonadapted Bgt were identified. Those accessions were intercrossed in two cycles and two lines, called SusBgtSC and SusBgtDC, with substantial susceptibility to Bgt at the seedling stage were selected.
• The quantitative variation among barley accessions and in the progenies after convergent crossing suggests a polygenic basis for this nonhost resistance. Both lines allowed an unusually high level of haustorium formation and colony development by Bgt. The SusBgt lines and their ancestor lines also allowed haustorium formation and conidiation by four out of seven isolates of other nonadapted B. graminis forms. Analysis of the infection process suggested that nonhost resistance factors are specific to the form and developmental stage of B. graminis. Resistances to establishment (first haustorium), colonization (subsequent haustoria) and conidiation are not associated.
• The lines developed will be of use in elucidating the genetic basis of nonhost resistance to Bgt in barley, and in gene expression and complementation studies on nonhost resistance.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
The majority of plant species are immune to the majority of potentially phytopathogenic microbial invaders. This is the most common and broadest spectrum resistance, termed nonhost resistance. It is defined as resistance shown by all members of a plant species against all members of a given pathogen species (Heath, 2000; Thordal-Christensen, 2003). Nonhost resistance relies on a variety of preformed and inducible mechanisms of defense (Thordal-Christensen, 2003; Mysore & Ryu, 2004). Pathogens that are able to overcome the preformed defense mechanisms encounter extracellular surface receptors that recognize general elicitors, the so-called pathogen-associated molecular patterns (PAMPs) (Nürnberger et al., 2004; Chisholm et al., 2006). The pathogen has to suppress PAMP-triggered immunity (PTI) in order to establish a compatible interaction and successfully colonize the plant (Chisholm et al., 2006). Those pathogen species that are not able to suppress PTI are often called nonadapted, inappropriate or heterologous pathogens. It is presumed that would-be pathogens deliver effectors to the attacked plant cells that may suppress PTI if there is sufficient recognition between effectors and their respective operative targets (Block et al., 2008; van der Hoorn & Kamoun, 2008). The degree of correspondence between effectors and targets may determine the host status of a plant with respect to a would-be pathogen (Niks & Marcel, 2009). Two interesting questions are whether different stages of development of a pathogen in a plant require different sets of effectors and whether different strains of a pathogen species have the same operative targets in a particular plant species.
Powdery mildew in cereals is caused by the biotrophic Ascomycete fungus Blumeria graminis, formerly known as Erysiphe graminis. Blumeria graminis has a high level of biological specialization for its cereal and grass hosts. Barley (Hordeum vulgare) is considered a strict nonhost plant species for the wheat (Triticum aestivum)forma specialis Blumeria graminis f.sp. tritici (Bgt), and also for the forms of B. graminis that are pathogenic to oat (Avena sativa) and rye (Secale cereale) (Hardison, 1944; Eshed & Wahl, 1970;Menzies & MacNeill, 1989). At the macroscopic level, no barley accession has been identified that allowed Bgt colonies to develop (Menzies & MacNeill, 1989; Olesen et al., 2003; Atienza et al., 2004). At the microscopic level, however, some barley accessions have been reported to permit some development of Bgt (Tosa & Shishiyama, 1984; Trujillo et al., 2004). Penetration of the plant epidermal cell walls and subsequent haustorium formation, hereafter called establishment, is a critical stage in the infection process. In compatible interactions of barley with the adapted powdery mildew pathogen, Blumeria graminis f.sp. hordei (Bgh), the fungus can successfully penetrate through epidermal cell walls and form haustoria to obtain nutrients from the host cells. In nonhost interactions of barley with nonadapted formae speciales such as Bgt, establishment typically fails. At sites of attempted cell wall penetration cell wall appositions (papillae) are formed, and in some cases attacked plant cells may mount a hypersensitivity reaction (HR). In the interaction of barley with the adapted powdery mildew, Bgh, a proportion of attempted germlings are also stopped by such defense mechanisms (Tosa & Shishiyama, 1984; Hückelhoven et al., 2001; Trujillo et al., 2004).
It is unknown which genetic factors determine that barley is a host to Bgh but a nonhost to nonadapted formae speciales of B. graminis. As there is no barley line that is susceptible to nonadapted formae speciales such as Bgt, inheritance studies are not possible. Some candidate genes have been implicated as playing a role in resistance of barley to the wheat forma specialis (Bgt) through transient overexpression or transient-induced gene silencing (TIGS) in single epidermal cells (Eichmann et al., 2004; Douchkov et al., 2005; Miklis et al., 2007; Schweizer, 2007), but it is not known whether allelic variation between these genes explains the host status difference between wheat and barley with respect to Bgt. Barley is indeed a model plant in which to study the molecular basis of basal and nonhost resistance to powdery mildews (Hückelhoven, 2007; Schweizer, 2007; Collinge et al., 2008). We set out to identify the genes that are responsible for natural variation in host status and nonhost resistance level.
In the present paper, barley accessions with rudimentary macroscopic susceptibility to the nonadapted wheat powdery mildew (Bgt) were identified and those accessions were intercrossed in two cycles to develop experimental barley lines with substantial susceptibility to Bgt. The components of the development of infection with Bgh, Bgt and isolates of several other nonadapted powdery mildew formae speciales and species on these experimental lines were quantified to determine the specificity of resistance factors.
Materials and Methods
Plant and pathogen material
A collection of 439 accessions of barley was evaluated at the seedling stage for resistance to the nonadapted wheat powdery mildew fungus, Bgt. The collection consisted of 136 accessions of Hordeum spontaneum (C. Koch) and 303 accessions of Hordeum vulgare L.: 227 accessions (landraces and wild H. spontaneum barley from the International Center for Agricultural Research in the Dry Areas (ICARDA) barley germplasm collection) were provided by R. K. Varshney, 54 accessions were obtained from the Centre for Genetic Resources (CGN), Wageningen, the Netherlands and 22 parental lines of mapping populations of barley were available at the Laboratory of Plant Breeding, Wageningen University. The seeds of barley H. vulgare cv. Turkey 290 were kindly provided by H. E. Bockelman (National Small Grains Collection, US Department of Agriculture, Aberdeen, ID, USA). An isolate of Blumeria graminis (D.C.) Speer f.sp. tritici Em. Marchal (Bgt; Swiss field isolate FAL92315), kindly provided by P. Schweizer (IPK, Gatersleben, Germany), was propagated on the susceptible wheat (Triticum aestivum) cv. Vivant and used for inoculation. The lines with some degree of susceptibility to Bgt were intercrossed in two cycles, resulting in two lines with substantial susceptibility to Bgt (see Results section). These lines were named SusBgtSC and SusBgtDC. SusBgt lines and the parental lines were tested with isolates of several nonadapted powdery mildew formae speciales (ff.spp.) and species. The isolates will be referred to according to their source host. The following nonadapted ff.spp. of B. graminis were tested: Blumeria graminis f.sp. avenae Em. Marchal (Bga), the powdery mildew of oat (Avena sativa L.); Blumeria graminis f.sp. secalis Em. Marchal (Bgs), the powdery mildew of rye (Secale cereale L.); and four isolates collected from wild grasses: Blumeria graminis f.sp. hordei-murini (Bghm) from Hordeum murinum L.; Blumeria graminis f.sp. agropyri (Bgar) from Agropyron repens L.; Blumeria graminis f.sp. bromi (Bgb) from Bromus mollis L., and Blumeria graminis f.sp. dactylidis (Bgd) from Dactylis glomerata L. All of these isolates were collected in the Wageningen region and maintained in isolation on their respective host plants. Isolates of two nonadapted species, namely Oidium neolycopersici L. Kiss, the powdery mildew of tomato (Solanum lycopersicum L.), and Podosphaera fusca (Fr.) U. Braun & N. Shishkoff, the powdery mildew of cucumber (Cucumis sativus L.), were also tested.
Three seedlings of each barley accession were grown in boxes 37 × 39 cm in size filled with compost soil. They were watered as required and no fertilizer or plant protection chemical was applied. The susceptible wheat cultivar Vivant was sown in each box to monitor the effectiveness of inoculation. Barley accessions Turkey 290, SusPmur and SusPtrit were included as references. They have been reported to allow relatively high haustorium formation by Bgt (Tosa & Shishiyama, 1984; Trujillo et al., 2004). Inoculation took place during daylight. The fully expanded primary leaves of 12-d-old seedlings were fixed to the soil horizontally, adaxial side up. Each plant box was placed in a carton (58 × 39 × 37 cm) that served as a settling chamber. Freshly produced conidia of a Bgt isolate growing on the susceptible wheat cultivar Vivant were blown into the settling chamber using compressed air. The density of inoculum was monitored with a haemocytometer. Applied inoculum densities were c. 50 conidia per mm2. The inoculated plants were incubated in a growth chamber at 18–20°C with 70% relative humidity and a photoperiod of 16 h (200 μmol m−2 s−1).
Macroscopic and microscopic evaluation of the barley–Bgt interaction
First, macroscopic evaluations of all tested accessions were carried out using a magnifying glass (10×) 8 d after inoculation (dai). Those accessions that macroscopically showed some degree of susceptibility were selected. These accessions were sown and inoculated again for quantification of haustorium formation and conidiation under the microscope. At two time-points, 72 h after inoculation (hai) and 8 dai, a leaf segment c. 3 cm long was collected from the middle part of the infected primary leaves. Leaf segments were transferred to a solution of acetic acid–ethanol (3 : 1 v/v) and cleared for at least 3 h. The leaf segments were then stained in Coomassie brilliant blue according to the method of Wolf & Frič (1981). The experiment was carried out in two consecutive replications. Per barley accession, 200 observed germlings were scored for the result of the first penetration attempt to establish a haustorium. Different types of host epidermal cells show different degrees of resistance to penetration (Lin & Edwards, 1974; Koga et al., 1990) and penetration attempts by different germlings on the same cell may induce resistance or susceptibility (Carver et al., 1999; Olesen et al., 2003). Therefore, only cells that were attacked by a single conidium and only epidermal cells adjacent to stomata and interstomatal epidermal cells (epidermal cells of types A and B following the terminology of Koga et al., 1990) were considered. Histological data were subjected to logistic transformation. Comparison of means was based on Fisher’s Least Significant Difference (LSD) at α = 0.05.
Evaluation of susceptibility of SusBgt lines
SusBgt lines and the parental lines were tested with nonadapted powdery mildew isolates in order to quantify the degrees of haustorium formation and hypersensitive and nonhypersensitive mechanisms of defense and conidiation. At 72 hai, leaf samples were cut and placed in a solution of 1 mg mL−1 3,3′-diaminobenzidine (DAB)-HCL, pH 3.8, which allows whole-cell accumulation of H2O2, for 8 h and subsequently transferred to a solution of acetic acid–ethanol (3 : 1 v/v) and cleared overnight. The leaf segments were then stained with 0.1 mg ml−l trypan blue in alcoholic lactophenol as described by Peterhänsel et al. (1997). H2O2 accumulation appears as a reddish-brown coloration of epidermal cells and indicates the degree of HR (Thordal-Christensen et al., 1997).
Genotypic variation in the interaction of barley with the wheat powdery mildew pathogen
The majority of the 439 barley accessions were immune, showing no macroscopic symptoms or signs of infection whatsoever with the nonadapted pathogen (Bgt) even 10 dai. However, a few barley accessions were identified that showed a low degree of susceptibility to this Bgt isolate. On these accessions very small lesions, or tiny colonies, were visible using a 10× magnifying glass or the naked eye, indicating some haustorium formation by the nonadapted pathogen. Six such accessions were examined under the microscope to quantify the amount of haustorium formation in comparison with lines Turkey 290, SusPtrit and SusPmur and with the barley cultivar Vada which is immune to Bgt. The cellular reactions of these selected lines to Bgt are shown in Fig. 1. On barley cv. Vada, all germlings were stopped at the penetration stage and they were associated with the deposition of cell wall appositions (papillae). The fungus did not develop further to establish any haustoria (Fig. 6a,b). On the other accessions, the large majority of germlings were also stopped at the penetration stage, but 12–25% of the germlings succeeded in penetrating through the epidermal cell wall and established a haustorium and elongated secondary hyphae (ESH). Some accessions with similar rates of establishment of Bgt, such as Hsp17 and SusPtrit, differed greatly in the rate of conidium formation (Fig. 1). The results confirm earlier reports (Tosa & Shishiyama, 1984; Trujillo et al., 2004) that there are differences among barley accessions in the degree to which they allow the nonadapted wheat powdery mildew pathogen to establish haustoria. Barley accessions with some degree of susceptibility to this nonadapted mildew are rare; such accessions were used as parental lines to accumulate genes for susceptibility to Bgt.
Development of SusBgt lines, susceptible to the nonadapted wheat powdery mildew
Based on macroscopic and microscopic observations, including the degree of conidiation, four accessions of barley, namely Hiproly (CGN00588; Egypt), Trisuli Bazar 9 (CGN00931; Nepal), Chame 2 (CGN01042; Nepal) and SusPtrit (Atienza et al., 2004) were selected as the most promising lines for crossing to accumulate genes for susceptibility to Bgt. Pairwise crosses were made in these four lines: Trisuli Bazar 9 × Hiproly and SusPtrit × Chame 2 (Fig. 2). The second generation offspring (F2) derived from each of the two crossing combinations were first evaluated macroscopically for susceptibility to the Bgt isolate, and then a few F2 plants from each crossing combination that seemed to be more susceptible than the parents were tested microscopically for penetration efficiency and conidiation. A double cross (DC) was made between a selected individual F4 plant derived from the cross Trisuli Bazar 9 × Hiproly and a selected individual F3 plant derived from the cross SusPtrit × Chame 2. Four DC-S1 plants originating from the same single DC plant were used to make doubled haploid (DH) lines by microspore culture. We had expected additional transgression in the DC (because it had four susceptible ancestors) compared with the single cross (SC) with only two ancestors. However, during the procedure, the progeny from the cross SusPtrit × Chame 2 seemed so susceptible that we also produced a DH from that lineage. Two selected F4 individuals derived from one selected F3 plant from the single cross (SC) SusPtrit × Chame 2 were used to make DH lines. Progeny from selfed DH plants were first evaluated macroscopically, and those lines that showed clear susceptibility were analyzed microscopically. As the selected DC and SC lines seemed similarly susceptible, we decided to test them both extensively. The most Bgt-susceptible lines derived from the SC and DC were named SusBgtSC and SusBgtDC.
Susceptibility of SusBgt lines to the adapted barley (Bgh) and nonadapted wheat (Bgt) powdery mildew pathogens
The SusBgt lines and all the parental lines showed a compatible interaction (infection type 4 according to the scale of Mains & Dietz (1930)) with the barley powdery mildew. Interestingly, SusPtrit and both SusBgt lines stopped at least 50% of the infection units of the adapted pathogen, Bgh, from developing haustoria. This indicates a moderate degree of basal resistance to Bgh (Fig. 3).Therefore, the substantial susceptibility of these lines to Bgt did not bring about extreme susceptibility to Bgh.
SusBgt lines allowed the isolate of the nonadapted Bgt to form many colonies that were clearly visible to the naked eye at 8 dai and even earlier (Fig. 4a). In the barley accessions, a proportion (depending on genotype, 12–27%) of challenged epidermal cells without detectable haustoria responded with HR (Fig. 5a).
There were substantial differences among the accessions in the frequency of cells in which a haustorium was formed. Vada was fully resistant to penetration by Bgt. Both SusBgtSC and SusBgtDC lines showed a higher level of haustorium formation by Bgt than the parental lines (Fig. 5a). In SusBgtSC and SusBgtDC, 51% and 59%, respectively, of germlings succeeded in penetration and haustorium formation, while in Chame 2, the most susceptible parental line, only 35% of germlings successfully formed a first haustorium. In SusBgtDC, which showed the highest frequency of successfully penetrated cells, a larger proportion of penetrated cells with a haustorium (59%) were killed by single-cell HR (Fig. 6c) than in SusBgtSC (42%) and the parental lines. The highest frequency of successfully penetrated epidermal cells without HR (Fig. 6d) was detected in SusBgtSC (29%). The majority of those germlings that had formed a primary haustorium were stopped at the secondary penetration attempts. In both SusBgtSC and SusBgtDC, a considerable proportion of successfully penetrated germlings succeeded in establishment of a secondary haustorium (44 and 42%, respectively, Fig. 5d). Interestingly, the secondary and subsequent haustoria were almost always detected in the same cell in which the primary haustorium had been established, typically resulting in two to five haustoria in one epidermal cell (Fig. 6e). Subsequent penetration attempts in epidermal cells located away from the first infected cell usually failed (Fig. 6f). At 8 dai, lines differed in the proportion of established germlings that had proceeded to the formation of at least some conidiophores (Fig. 6g,h). This conidiation was most conspicuous (34% of those germlings that had formed a haustorium) in SusBgtSC, and much less so in SusBgtDC (6%) (Fig. 5g). In the parental line Trisuli Bazar 9, a similar proportion of established germlings reached conidiation as in SusBgtSC (Fig. 5g), but the rate of establishment was lower than in SusBgtSC (Fig. 5a). In summary, the results showed that SusBgt lines show substantial susceptibility to the nonadapted Bgt, and that susceptibility to first haustorium formation may (SusBgtSC) or may not (SusBgtDC) be associated with successful conidiation. Both lines are of value in the study of the inheritance of nonhost resistance in barley to powdery mildews.
The degree of susceptibility of SusBgt lines to other nonadapted powdery mildew pathogens
To determine whether the SusBgt lines are susceptible to isolates of other nonadapted powdery mildew pathogens, the four parental lines, the SusBgt lines and the reference barley cultivar Vada were inoculated with one isolate each of other nonadapted formae speciales and species of powdery mildew pathogens.
SusBgt lines and their parental lines showed relatively high haustorium formation and conidiation by the powdery mildew isolate of Hordeum murinum, Bghm (Figs 5b,e and 7b). At the macroscopic level, SusBgtSC showed a very strong HR reaction at 8 dai (Fig. 4b). All tested accessions allowed at least some haustorium formation and conidiation by the rye powdery mildew isolate (Bgs) (Fig. 5c,f), but the SusBgt lines were not more susceptible to this mildew form than the other lines. SusBgtSC and SusPtrit, which allowed relatively high rates of establishment of Bgt (> 50%), allowed considerably less establishment by Bgs (< 26%). Surprisingly, Vada, which did not allow any haustorium formation by Bgt, allowed relatively high haustorium formation by the isolate of Bgs (31%). Relatively few germlings of Bgs that had formed a first haustorium proceeded to form a secondary haustorium (< 22%) and conidia (< 14%) compared with Bgt (< 45 and < 35%, respectively) (compare Fig. 5d with Fig. 5f, and note the difference in the scale of the ordinates). The tested barley lines were all nearly immune to the mildew isolates of oat, B. mollis and D. glomerata. At most, 10% of the germlings formed a haustorium, and none of the germlings formed conidia. Both SusBgt lines allowed substantial haustorium formation (35–51%) by the Agropyron powdery mildew isolate (Bgar), but the rate of established germlings reaching conidiation was considerably higher in SusBgtSC than in SusBgtDC (80% and 21%, respectively) (data not shown). As with Bgt, during infections by the other nonadapted mildew strains, except Bghm (Fig. 7a), second and subsequent haustoria were almost always formed in the same cell as the first haustorium, and failed penetration attempts along the spreading hyphae were commonly observed. The isolates of the two additional nonadapted mildew species, O. neolycopersici and P. fusca, were not observed to form any haustoria on any of the barley accessions.
Several interesting conclusions regarding the specificity and genetic basis of the basal resistance of barley to nonadapted B. graminis mildews have emerged.
Firstly, the resistance is quantitative, as suggested by the quantitative differences among accessions. The transgressive segregation found when the parental lines were combined suggests at least two genes to be responsible. Crossing and selection among some barley accessions with some degree of susceptibility resulted in the accumulation of susceptibility factors (or depletion of resistance factors) to the nonadapted wheat powdery mildew Bgt. These lines showed a substantial level of haustorium formation and colony development by this nonadapted target mildew.
Secondly, genes for basal resistance have pathogen developmental stage-specific effects. This is especially well illustrated by observations that suggest that susceptibility to first haustorium formation may (SusBgtSC) or may not (SusBgtDC) be associated with successful conidiation (Fig. 5). If quantitative trait loci (QTLs) indeed represent operative targets for pathogen-delivered effectors to suppress aspects of the defense (Jafary et al., 2006; Niks & Marcel, 2009), suppression of the defense at the first penetration stage may require different effector–target interactions than conidiation. Also, it seems that making the first haustorium (suppression of the defense of the first plant cell) does not imply that the defenses of the surrounding cells are also suppressed easily. The observation that several haustoria occurred in the same cell as the first haustorium (Fig. 6e) suggests that, after the first haustorium was initiated, the defense was suppressed in that cell. Such induced accessibility has been well documented (e.g. Lyngkjær & Carver, 1999; Olesen et al., 2003). Apparently this induction did not spread to surrounding cells to allow haustorium formation there by Bgt and Bga, and the fungal effectors did not seem to be efficient at allowing access to these surrounding cells (Fig. 6f). By contrast, the isolate of Bghm was able to induce accessibility of surrounding cells, as it formed haustoria in them (Fig. 7a). Arabidopsis penetration (pen) mutants allow higher frequencies of penetration by the nonadapted powdery mildew pathogen (Bgh), but other genes (Senescence-associated gene Sag101 and Phytoalexin deficient gene Pad4) compromise subsequent stages in the infection process (Lipka et al., 2005). Consonni et al. (2006) reported that the Arabidopsis thaliana mlo2 Atmlo2pen1 double mutant allows more cell entry by Golovinomyces cichoracearum, but no significant increase in conidiophore production. For later stages of development, it seems that enhanced disease susceptibility gene Eds1, Sag101 and Pad4need to be silenced.
Thirdly, genes for basal resistance are mildew forma specialis-specific. SusBgt lines and SusPtrit are remarkably accessible to Bgt and allow relatively high frequencies of fungal establishment. Yet, they show a moderate degree of basal resistance to the adapted powdery mildew pathogen, Bgh. Vada had a high degree of resistance to Bgt, but was not equally resistant to the isolate of rye mildew Bgs. The factors in SusBgtDC that hampered conidiation by Bgt compared with SusBgtSC (Fig. 5g) seem also to be effective against Bgar (Fig. 7c,d), but not against Bghm (Fig. 5h,b) or Bgs (Fig. 5i). Selection for high susceptibility to Bgt has not affected the degree of resistance to isolates of Bga and other nonadapted mildews.
The results of the present study show many parallels with the barley–rust interaction (Atienza et al., 2004; Jafary et al., 2006, 2008; Marcel et al., 2007). In both pathosystems, accumulation of genes for susceptibility to nonadapted pathogen forms and species appears feasible. Nonhost resistance to rust species and to nonadapted formae speciales of powdery mildew seems similar to quantitative, basal host resistance, and the resistance mechanism in both pathosystems is mainly based on nonhypersensitive mechanisms.
Powdery mildew pathogens must suppress PTI to establish basic compatibility. This suppression of PTI requires compatibility between the effectors and the operative targets in the plant. Effectors (O’Connell & Panstruga, 2006) and their operative targets in the plant (van der Hoorn & Kamoun, 2008) both seem to be under strong diversifying selection. For each plant species, a different set of effectors may be required to suppress PTI (Almeida et al., 2009; Niks & Marcel, 2009), implying that each mildew form probably has a different set of effectors, each tuned to the operative targets in their host plant species. Depending on the fungal development stage, different plant genes involved in defense should be reprogrammed, again requiring different effectors. By common evolutionary ancestry, however, some compatible variants of operative target motifs may occur in some nonhost plant accessions. Nonadapted mildews cannot be expected to suppress PTI of barley effectively, but, in the SusBgt lines, variants of the targets that can be suppressed relatively easily by Bgt effectors may have been accumulated. An alternative hypothesis is that specialization of powdery mildew to different host species depends on the timing of the delivery of effectors in plant cells. Bgt has been reported to develop more slowly than Bgh (Boyd et al., 1996). This difference in timing is reflected in different timing of the expression of barley genes involved in defense against Bgh and Bgt (Boyd et al., 1996). The SusBgt lines could have been selected to have a timing of responses that makes them more susceptible to Bgt.
New evidence has emerged on the involvement of some candidate genes in the basal host and nonhost resistance of barley against mildew pathogens (Douchkov et al., 2005; Miklis et al., 2007). An example of such a candidate gene is H. vulgare Synaptosomal-associated protein HvSNAP34, which plays a role in basal host and nonhost resistance of barley to B. graminis. Transient induced gene silencing of this candidate gene strongly increased the rate of successful haustorium formation of both adapted (Bgh) and nonadapted (Bgt) powdery mildew fungi (Douchkov et al., 2005). HvSNAP34 could be one of the operative targets of effectors to enhance successful establishment of haustoria by invading mildew fungi. The lines developed here will allow inheritance studies to be carried out on the nonhost resistance of barley to Bgt and other nonadapted powdery mildew, similar to studies performed on rusts by Jafary et al. (2006). Finally, the existence of relatively high degrees of susceptibility of SusBgt lines to more than one forma specialis of B. graminis may facilitate the cross-hybridization between different ff.spp. of B. graminis, as was carried out in the shared host Triticum urartu for crossing f.sp. tritici with f.sp. agropyri by Tosa (1989). Such pathogen and plant inheritance studies can contribute greatly to our understanding of the evolutionary and genetic basis of the host–pathogen interactions in Gramineous powdery mildew pathosystems.
RA was supported by the Agricultural Research and Education Organization (AREO) and the Ministry of Science Research and Technology of Islamic Republic of Iran. Anton Vels is gratefully acknowledged for his excellent technical assistance in the glasshouse experiments. We thank Dr Patrick Schweizer (Leibniz-Institute of Plant Genetics and Crop Plant Research (IPK), Germany) for providing the wheat isolate of mildew. Bert Essenstam from the Unifarm of Wageningen University is acknowledged for providing the cucumber powdery mildew and production of wheat powdery mildew inoculum. We thank Dr Harold E. Bockelman (National Small Grains Collection, USDA, USA) for providing the seeds of barley cv. Turkey 290, the Centre for Genetic Resources, the Netherlands (CGN) for proving the seeds of 54 accessions of barley, and Dr Rajeev K. Varshney from IPK, Germany, for providing the seeds of 227 accessions of barley (ICARDA collection) used in this study. ICARDA (Syria), which prepared the seeds, is acknowledged for its indirect contribution to the study. We thank Dr Viktor Korzun and Clemens Springmann (PLANTA Angewandte Pflanzengenetik und Biotechnologie, Germany) for making the DHs. Dr Chris Maliepaard is acknowledged for his help with the data analysis. We thank Prof. Richard G. F. Visser for critically reading the manuscript and for his continuous support. We thank the anonymous reviers for their excellent suggestions.