General enquiries on this form should be made to:Defra, Science Directorate, Management Support and Finance Team,Telephone No. 020 7238 1612E-mail: research.competitions@defra.gsi.gov.uk

SID 5 Research Project Final Report

SID 5 (Rev. 3/06) Page 1 of 25

NoteIn line with the Freedom of Information Act 2000, Defra aims to place the results of its completed research projects in the public domain wherever possible. The SID 5 (Research Project Final Report) is designed to capture the information on the results and outputs of Defra-funded research in a format that is easily publishable through the Defra website. A SID 5 must be completed for all projects.

This form is in Word format and the boxes may be expanded or reduced, as appropriate.

ACCESS TO INFORMATIONThe information collected on this form will be stored electronically and may be sent to any part of Defra, or to individual researchers or organisations outside Defra for the purposes of reviewing the project. Defra may also disclose the information to any outside organisation acting as an agent authorised by Defra to process final research reports on its behalf. Defra intends to publish this form on its website, unless there are strong reasons not to, which fully comply with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.Defra may be required to release information, including personal data and commercial information, on request under the Environmental Information Regulations or the Freedom of Information Act 2000. However, Defra will not permit any unwarranted breach of confidentiality or act in contravention of its obligations under the Data Protection Act 1998. Defra or its appointed agents may use the name, address or other details on your form to contact you in connection with occasional customer research aimed at improving the processes through which Defra works with its contractors.

Project identification

1. Defra Project code AR0712

2. Project title

Biology and genetics of durable resistance to biotrophic pathogens of cereals

3. Contractororganisation(s)

IGERPlas GogerddanAberystwythCeredigionSY23 3PD

                    

54. Total Defra project costs £ 914,053(agreed fixed price)

5. Project: start date................ 01 April 2003

end date................. 31 March 2007

SID 5 (Rev. 3/06) Page 2 of 25

6. It is Defra’s intention to publish this form. Please confirm your agreement to do so...................................................................................YES NO (a) When preparing SID 5s contractors should bear in mind that Defra intends that they be made public. They

should be written in a clear and concise manner and represent a full account of the research project which someone not closely associated with the project can follow.Defra recognises that in a small minority of cases there may be information, such as intellectual property or commercially confidential data, used in or generated by the research project, which should not be disclosed. In these cases, such information should be detailed in a separate annex (not to be published) so that the SID 5 can be placed in the public domain. Where it is impossible to complete the Final Report without including references to any sensitive or confidential data, the information should be included and section (b) completed. NB: only in exceptional circumstances will Defra expect contractors to give a "No" answer.In all cases, reasons for withholding information must be fully in line with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.

(b) If you have answered NO, please explain why the Final report should not be released into public domain

Executive Summary7. The executive summary must not exceed 2 sides in total of A4 and should be understandable to the

intelligent non-scientist. It should cover the main objectives, methods and findings of the research, together with any other significant events and options for new work. INTRODUCTION AND OBJECTIVESThe Defra Fellowship was awarded to Carver who led studies on powdery mildew resistance and stomatal dysfunction at IGER and advised on the wheat yellow rust studies led by Boyd at JIC. The work underpinned development of strategies for sustainable disease management using genetic resistance to reduce dependence on fungicides. The rapid evolution of virulence has nullified many disease resistance genes that have been deployed in UK cereal cultivars. Durable resistance is available but it is acknowledged generally to be partial and under complex physiologic control. Poor understanding of its cell biological basis, and therefore of the fundamental traits contributing to the resistance, has so far precluded identification of the controlling genetic factors. This work therefore aimed to increase understanding of the basis of durable resistance to the obligate biotrophic fungal pathogens that cause cereal powdery mildews and wheat yellow rust, in order to facilitate its exploitation through plant breeding. Increased understanding genetic control of durable resistance has been hampered by lack of appropriate plant material. Two key objectives of the current work were to generate such material for studies of powdery mildew (Objective 1) and to reassess the potential value of 21 wheat cvs previously identified by Dr Roy Johnson as having durable resistance (Objective 5).Evolution of virulence by powdery mildew populations has overcome virtually all single gene resistances deployed in UK cereal cvs. Plant breeders have, therefore, selected empirically for partial resistance. Objective 2 examined whether this form of resistance can be overcome by pathogen adaptation. A form of highly effective, broad spectrum penetration resistance is induced in cereal epidermal cells subjected to sequential powdery mildew attacks. This induced ‘inaccessibility’ is even more effective than that conditioned by mlo alleles in barley, and its constitutive expression in cereal epidermis would provide a novel and extremely efficient form of durable resistance. To achieve this requires detailed understanding of inaccessibility and Objective 3 aimed to discover its physiological basis and whether it relies on static or active processes. Chance observations made during this work led to establishment of Objective 7 which examined the effects of resistance expression on stomatal function in attacked leaves. Earlier MAFF/Defra funded studies at IGER, and very recent collaboration with Würzburg University (Prof. M. Riederer et al.) indicated that epicuticular leaf wax constitution can affect the ability of powdery mildews to form infection structures, and this imparts virtual immunity to the abaxial surface of ryegrass leaves. Objective 4 used introgression lines created at IGER (for a different purpose) with a view to identifying the basis of this resistance and possibilities of its exploitation.All of the above studies involving powdery mildew used microscopy to characterise host responses and study individual resistance mechanisms. To date, however, microscopy of responses to yellow rust attack have been severely constrained by the fact that under laboratory conditions yellow rust infection frequencies are extremely low. This impedes data collection and can therefore preclude identification of individual resistance mechanisms of quantitative effect. Thus, Objective 6 sought to manipulate

SID 5 (Rev. 3/06) Page 3 of 25

environmental conditions to improve infection frequency as a prelude to comparative work aimed at identifying resistance characteristics.

APPROACHES AND RESULTSObjective 1. F1 hybrids were made from crosses between oat cv. Maldwyn (durably resistant) with cv. Emotion, cv. Millennium and N237/6. F2 plants (200 per cross) and selected F3 descendants were assessed for powdery mildew resistance in a glasshouse. Juvenile and adult leaves of F4 selections were assessed microscopically for primary penetration resistance, colony growth and plant epidermal cell death. This showed considerable variation between lines. Leaf material from each generation was cryo- stored for future DNA analysis.Objective 2. Powdery mildew populations collected in 2002 from crops of barley cvs County and Optic were ‘trained’ by continuous, isolated culture on their source cv. and cv. Golden Promise (susceptible) for 2 years. Each year isolates were tested on County, Optic and Golden Promise and their penetration efficiency and colony growth was compared microscopically. Mildew from County was slightly more pathogenic than that from Optic, but both cvs maintained good adult plant resistance. Data indicated no evidence that training had led to fungal adaptation. Objective 3. Phosphate scavengers and phenolic biosynthesis inhibitors suppressed induced inaccessibility in oat, providing evidence that it is energy dependent and relates to lignin biosynthesis. Tests using cordycepin to inhibit gene transcription by inhibition were inconclusive; cordycepin was fungitoxic at recommended concentration and ineffective at lower concentration. Cytochalasin E, an inhibitor of cytoskeletal microfilament polymerisation, increased basal susceptibility but had no effect on induced inaccessibility suggesting that cytoskeleton reorganisation may not be a key factor. Objective 4. IGER hybrids between short lived ryegrass species and meadow fescue carrying a ‘stay green’ gene showed abaxial mildew susceptibility but examination of a derived mapping population suggested that the susceptibility and stay-greens factor are independent. This work could not be concluded as the short-lived hybrids died.Objective 5. Field trials assessed the 21 cvs inoculated with yellow rust isolates representing the race-profile of the UK yellow rust population. In 2005, 16 were apparently immune, three showed high levels of partial resistance, but 2 were now fully susceptible. In 2006 the cvs were tested at two sites, Norwich and Market Rasen. In Market Rasen all varieties supported more yellow rust infection and all previously immune varieties showed varying degrees of symptoms though most maintained good resistance. In detailed glasshouse assessments of five selected cvs, adult plant resistance did not affect disease latent period but fewer pustules formed on the three more resistant cvs, Bersee, Carstens V and Little Joss, which also produced fewer spores per pustule. CarstensV and Cappelle Desprez have been crossed to the yellow rust susceptible wheat cv. Lemhi to give F2 seed. Crosses have also been made between wheat cvs carrying known sources of durable yellow rust resistance and the susceptible wheat cvs Brigadier, Robigus and Glasgow and F1 seed is available.Objective 6. Environmental conditions pre- and post-inoculation were manipulated and, in a susceptible cv., high light input immediately before inoculation increased both spore germination rates and the frequency of rust germ tubes entering stomata. Transpiration measurement showed no post-inoculation differences between leaves due to light pre-treatment, indicating no effect of treatment on stomatal aperture post-inoculation. This suggests that high light pre-inoculation may stimulate production of volatile factor(s) acting as germination promotors and germ tube attractants. Comparisons between cvs. suggested variation in degree of germination/germ tube orientation stimulated by pre-inoculation high light, indicating a potential novel source of yellow rust resistance. Ojective 7. Measurement of leaf water conductance and light and cryo-scanning electron microscopy showed that epidermal cell death associated with three different single resistance genes caused stomata of powdery mildew-attacked barley to lock open and become unable to close in response to darkness, drought or application of abscisic acid. Conversely, following brown rust attack, single gene resistance of wheat and barley caused stomata to become locked shut. By contrast, stomata quickly recovered functionality where resistance was papilla-based and epidermal cells survived powdery mildew attack. DISCUSSIONThe generation of oat populations, stored genetic material and histological data provides future opportunity to identify genes for durable powdery mildew resistance: some work will be based on core budget while other funding options are being considered. Similarly, identification of wheats with durable yellow rust resistance indicates lines for study and as potential sources of resistance. A constraint to studies of yellow rust resistance has been its extremely low infection frequency under laboratory conditions. The discovery that high light input pre-inoculation substantially increases infection opens the way to identify of individual resistance mechanisms operating throughout the interaction. The hypothesis that the pre-inoculation effect is due release of volatile(s) that promote germination and germ tube orientation can now be explored and the finding that conditions affect wheat genotypes differentially may be indicate a novel form of resistance. Past MAFF/Defra work defined histological methods for analysing powdery mildew resistance, and here the quantitative resistance expressed in two modern barley cvs proved stable over a period of isolate ‘training’ equating to c. 140 asexual cycles or seven field seasons. Thus, empirical selection by plant breeders has delivered durable resistance in the absence of sexual recombination. Confirming previous

SID 5 (Rev. 3/06) Page 4 of 25

studies, induced inaccessibility was shown to convey very effective resistance but the work showed its requirement for energy and phenolic biosynthesis. That it is expressed strongly in all cereals, irrespective of their inherent susceptibility, indicates that it is regulation of gene function rather than the presence of particular genes that controls its expression. Thus, single cell transcript analysis, (collaboration with Lyngkjær et al.) holds the key to understanding its control. The suggestion that a gene linked to the stay-green factor in meadow fescue influences abaxial leaf surface susceptibility suggests a route to identifying gene(s) of importance by construction of a long-lived Fescue/Lolium mapping population. An important physiologic basis for cost of major gene resistance was identified as stomatal dysfunction consequent on epidermal death. This will have obvious consequences for photosynthesis, respiration, water use efficiency and respond to abiotic stresses. Papilla-based resistance is therefore more desirable. The Fellowship facilitated national/international collaborations that added value through progressing understanding of single cell genetic responses, induced resistance, early host-pathogen interactions and pea powdery mildew resistance (principally via: Gay, IGER; Gurr, Oxford; Lyngkjær, Denmark; Kunoh, Japan; Mur, UWA; Riederer, Germany; Prats, Spain; Zeyen, USA). Additionally, Defra’s interest has fostered relationships between Carver and Pavely (Defra Fellowship AR0511) that led to a proposal for work to assess costs of disease resistance in field situations, particularly in relation to crop water use efficiency.

Project Report to Defra8. As a guide this report should be no longer than 20 sides of A4. This report is to provide Defra with

details of the outputs of the research project for internal purposes; to meet the terms of the contract; and to allow Defra to publish details of the outputs to meet Environmental Information Regulation or Freedom of Information obligations. This short report to Defra does not preclude contractors from also seeking to publish a full, formal scientific report/paper in an appropriate scientific or other journal/publication. Indeed, Defra actively encourages such publications as part of the contract terms. The report to Defra should include: the scientific objectives as set out in the contract; the extent to which the objectives set out in the contract have been met; details of methods used and the results obtained, including statistical analysis (if appropriate); a discussion of the results and their reliability; the main implications of the findings; possible future work; and any action resulting from the research (e.g. IP, Knowledge Transfer).

SID 5 (Rev. 3/06) Page 5 of 25

8.1. Create and assess a range of recombinant inbred lines for oat powdery mildew resistance based on crosses between the cv. Maldwyn that has proven, durable mildew resistance and modern, susceptible cvs. (Proposal Objective 1). All milestones were completed.

INTRODUCTIONThe quantitative, adult plant resistance of oat cv. Maldwyn provides a rare example of powdery mildew resistance that has proven durable in the field since 1947. Early studies [1] showed the resistance is controlled by at least six minor genes with additive effect. To breed for resistance under such complex genetic control is extremely difficult. Previous MAFF/Defra projects (CE0120 and CE0154) showed that the resistance depends on several mechanisms: papilla-based resistance and low frequency cell death limit colony establishment, and where colonies form, reduced efficiency of haustoria limits growth extending disease latent period and reducing sporulation. This work and studies on the physiological bases of different resistance mechanisms has been collated [2]. It is totally impractical, however, to screen breeding populations histologically for combinations of different mechanisms but this might be achieved using marker-assisted selection. Towards this, the current work aimed to produce plant material in which relevant genetic markers of different resistance genes may be identified. This involved hybridising Maldwyn with selected, susceptible modern oat cvs with a view to producing progeny in which various resistance mechanisms limiting colony establishment and development could be identified, with a view towards providing a resource for future molecular genetic analyses.

MATERIALS AND METHODSJM Leggett (cyto-geneticist, IGER) hybridised cv. Maldwyn (male) with nine different oat lines including modern cvs and a high oil experimental line (N327-6). Six hybrids were produced and selfed to produce F2 seed (ripened in July 04). Three crosses were selected for further study: x cv. Emotion (spring oat, moderately mildew susceptible); x cv. Millennium (winter oat tolerant of spring sowing, moderately mildew susceptible) and x line N327-6 (very mildew susceptible). In August 2004, 200 F2 plants of each of the three F1 hybrids, together with 12 plants of the parental genotypes for reference, were grown in randomized blocks in a glasshouse. When their fourth-formed leaf was emerging, segments of their fourth leaf were collected, frozen in liquid nitrogen, and stored at -80ºC for potential future use for molecular genetic study. Plants were then inoculated with race 5 B. graminis f.sp. avenae, and an epidemic developed. Each plant was scored at the juvenile stage (5th leaf unrolling) and again as adults (flag leaf expanded) for percentage leaf area affected [3]; for juveniles the sum of scores from leaves 3, 2, 1, and for adults the sum of scores from the flag leaf, flag-1 and flag-2, were summed as an overall indicator of disease. In addition infection type (IT) was recorded on the uppermost leaf at each stage using a 0-4 scale where: 0 = no visible symptoms; 1 = colonies just visible, no apparent sporulation; 2 = small colonies, sporulation questionable; 3 = moderate, obviously sporulating colonies; 4 = large, freely sporulating colonies. Note was taken if colonies were associated with necrosis. Unfortunately, although the parent cv. Millennium is tolerant of spring sowing, with the August planting it failed to head. After self pollination, all available F3 seed was collected and stored.In April 2005, from each of the three hybrid populations, 10 F3 plants were grown from each of five F2 parents that had shown high resistance (low area affected, low IT), from five that had shown moderate high leaf area infected but with low IT, from five that had shown low area infected but with high IT, and from five that had shown high susceptibility (high area affected, high IT). Seed of Millennium and its derivatives were vernalised (2ºC, 2 wk) and all were planted at the end of April. Plants were assessed as for F2 populations, fourth leaf material was cryo-preserved and F4 seed was collected. In May 2006, from each of the three F3 populations, seven plants were selected. Three of these had shown high resistance as adults (low area affected, low IT), two had shown moderate resistance (one with low IT but high area infected; the other with high IT but low area infected), and two had shown low resistance (high area affected, high IT). Five F4 plants of each of the seven lines were grown in a spore-proof glasshouse together with four of the parental genotypes (Millennium and derivatives were vernalised). Leaf material (from tillers) was taken from all plants for cryo-storage. On three successive days (one hybrid population per day), the expanded fourth-formed leaf (juvenile) was excised from all plants and three 3-cm segments were cut from each. These were arranged floating on 45 mg L-1 aqueous benzimidazole solution [4], inoculated by settling tower (ca 10 conidia mm -2) and incubated at 20ºC, 12 h light (150 umol m-2 sec-1) until one segment from each leaf was fixed at 42 and another at 72 h for microscopy, the last being inspected daily for symptom development. Plants were allowed to grow on until the flag leaf had emerged when the flag-1 leaf (adult) was taken, cut into segments used for microscopy and symptom assessment. Eventually, seed was later collected.

RESULTS Symptoms appeared on many juvenile F2 plants within 5 d of inoculation and developed fast although there was great variation between individuals (data not shown). As expected, Maldwyn showed good adult plant resistance as a relatively small area of its upper three leaves was affected, while adult N327-6 remained extremely susceptible; the distribution of F2 adult plants derived from this cross was skewed towards susceptibility. Cultivar Emotion was only slightly more susceptible than cv. Maldwyn (no significant difference) and here the F2

SID 5 (Rev. 3/06) Page 6 of 25

population data were skewed towards resistance with most behaving similarly to their parents. A similar skew was evident in the Millennium F2 population), but no data could be obtained from Millennium itself because plants failed to head under the late summer conditions. (However, data from 2005 indicated that cvs Maldwyn and Millennium were similarly resistant). In all F2 populations, transgressive segregants were evident as some plants showing greater resistance than either parent while others were far more susceptible. There was also variation in IT between plants within F2 populations, although, again, the population derived from N327-6 tended to produce larger colonies overall (data not shown). Thus it was possible to select plants showing different resistance phenotypes to provide tester F3 families. Assessment of the F3 selections again segregated for resistance allowing the selection of resistant, susceptible and intermediate phenotypes for histological tests in the F4 generation.With advice from Dr R. Sanderson (Statistician, IGER), data for several characteristics describing success of attempted infection and others reflecting post-infection development, assessed in segments fixed 42 and 72 h after inoculation, respectively, were examined by various approaches and all data and analyses have been archived and copies given to interested IGER staff (contact Dr C. Howarth). For brevity, only data from line N327-6 is presented (Figs 8.1.1, 8.1.2). Figure 8.1.1 shows data that indicate there was a strong tendency for adult plant resistance to increase frequencies of effective papilla resistance, to decrease successful penetration but to have relatively less effect on cell death frequency. To assess post-infection development, data from live colonies were examined in detail if at least seven of the 10 colonies observed per plant line were associated only with living epidermal cells. Thus, these data reflected the living plant’s ability to restrict pathogenesis. In lines derived from N327-6 and Millennium, a large number of lines (>15) fulfilled this criterion. Figure 8.1.2 illustrates data for per colony numbers of hyphal tips, reflecting growth made prior to fixation, and of secondary appressoria and haustoria that indicate future growth potential. In cv. Emotion and its derivatives, cell death associated with established colonies was very common and 26 of 35 showed very frequent post-infection cell death which was, therefore, a key component of their resistance. Within all populations, individual plants could be identified showing high, low and intermediate values for each character.

Fig 8.1.1. Kernel densities indicating underlying distributions in papillae formation, cell death and susceptibility to penetration in juvenile (grey line) and adult (black line) leaves within the F4 population of oat line N327-6 fixed 42 h after inoculation with powdery mildew.

Den

sity

Papillae (%) Cell death (%) Penetration (%)

Fig 8.1.2. Kernel densities indicating underlying distributions in per fungal colony numbers of secondary appressoria, haustoria and hyphal tips in juvenile (grey line) and adult (black line) leaves within the F4 population of oat line N327-6 fixed 72 h after inoculation with powdery mildew.

Den

sity

Number of secondary

appressoria

Number of secondary haustoria

Number of hyphal tips

DISCUSSIONThe data show variation in susceptibility between plants within and between F2 populations, within and between families and populations in F3 and F4 generations, and histological analyses of the F4 plants identified individuals that express different resistance mechanisms with different degrees of efficacy. These include mechanisms acting to prevent colony establishment and then the growth of colonies that do establish. The former reduce effective inoculum potential and the latter reduces the rate of colony growth and extends the asexual generation time, so slowing the rate of epidemic development. Both papilla formation and cell death prevent colony establishment. Papillae produced by living cells attacked by powdery mildew act as a physical and/or chemical barrier that prevents penetration of the host and establishment of haustoria (feeding structures), so arresting colony establishment. In Maldwyn oat and in mlo barley cvs, papilla-based resistance has proved durable. Separate work reported here (Objective 7) shows that papilla formation has little, and only transient, effect on the function of

SID 5 (Rev. 3/06) Page 7 of 25

stomata close to attack sites, providing a form of resistance that will be particularly useful in sustainable cropping systems under a changing environment where plants are likely more often to suffer drought and other abiotic stresses. By contrast, cell death has the undesirable consequence of causing dysfunction of nearby stomata, so that this form of resistance may carry a hitherto unsuspected cost to the plant. Where colonies do establish, restriction of their growth rate would clearly be desirable by limiting direct effects of parasitism as well as indirect effects in limiting stomatal opening. Again, however, post-infection cell death is likely to affect stomatal function deleteriously.The genetic material of all the plants used in this study has been preserved. Given sufficient molecular genetic markers, it should prove possible in future to relate genetic information to the degree of resistance/susceptibility expressed by individuals, and from the F4 data to relate genetic information to the individual resistance mechanisms expressed by particular genotypes. The strength of the relationship between molecular marker and resistance mechanism can be assessed not only from comparison between individuals within and between generations but also from comparison between populations arising from the different parents. All data, seed and cryo-preserved leaf material is now stored at IGER and ongoing discussions will determine how the material will be used and how the work is to be funded.

REFERENCES 1. Jones IT. 1986. Ann. Appl. Biol. 109, 187-192.2. Carver TLW. 2000. Powdery mildew disease: studies of fungal biology and host resistance. D.Sc. Thesis, University of Wales.3. Large EC, Doling DA. 1962. Plant Pathol. 11, 47-57.4. Jones IT, Hayes JD. 1971. Ann. Appl. Biol. 68, 31-39.

-------------------------------------------------------

8.2. Test the durability of powdery mildew resistance mechanisms in modern barley cvs that express high levels of partial field resistance. (Proposal Objective 2). All milestones were met for 2003-04 when the work was terminated.

INTRODUCTIONThis study was an extension of work started under Defra project CE0154 to identify individual mechanisms contributing to the partial resistance of two selected cvs and to test their continued efficacy as isolates of B. graminis f.sp. hordei (Bgh) were grown continuously on them (i.e. 'trained') and so provided the opportunity to adapt to the resistances. Current cereal powdery mildew populations in the UK have evolved complex virulence capable of overcoming virtually all major gene resistances deployed [[1] and previous reports of the UKCPVS]. As a direct consequence of this pathogen adaptation, breeders have selected empirically for partial forms of resistance and many cvs on the UK Recommended Lists for cereals show relatively high resistance ratings for powdery mildew (scoring 6-9 in trials). Although these resistances are quantitative, and are expressed more strongly in adult than seedling or juvenile plants, it is unclear whether they are durable [2]. To test this possibility two spring barley cultivars, Optic and County, that appeared on the 2001 Recommended List, were selected for study. Both showed good partial resistance to powdery mildew (Recommended List resistance ratings 7 and 8, respectively). The approach was to inoculate juvenile plants of each cultivar with mildew samples collected from them in the field, and attempt to erode their resistance by continuously culturing the mildew on them in isolation, thus applying pressure for selection of powdery mildew genotypes better able to grow on their host cv. To provide a control where no differential selection pressure was applied, isolates were sub-cultured on cv. Golden Promise which is highly susceptible to all known isolates of Bgh.

MATERIALS AND METHODSPathogen material. Samples of Bgh from natural infections on cvs Optic and County were collected by S. Slater (NIAB) from Cambridge field plots in spring/summer 2002. They were multiplied separately in isolated spore-proof compartments (min. temp. 20ºC) on their donor hosts. When established, both populations were subdivided. One part, was maintained on its 'own' cv. and the other was transferred to a separate compartment in which it was inoculated to cv. Golden promise so that four populations were created. Each was then maintained continuously on juvenile-adult plants of its donor host (cv. Optic or County) or on cv. Golden Promise by introducing new, healthy plants into the compartments each week. Sporulating colonies developed on these introductions after ca 5 days. Older plants were destroyed after 2 wks in the culture chambers, except in September of each year when they were kept until fully mature in an attempt to encourage cleistothecia.Assessments of fungal development. Histological and macroscopic observations were made in summer 2003 and 2004 to determine whether isolated mildew populations were adapting to their hosts. Healthy adult (5 th-leaf expanded) and juvenile plants (3rd-leaf expanded) were produced for inoculation at the same time in a completely randomised design with four replicates containing 32 adult and juvenile plants each of cvs Optic, County and Golden Promise. Leaves of four sets of intact plants were inoculated on four successive days with each of the four mildew populations. In all cases, the adaxial surface of third leaves of four juvenile plants and of fifth leaves of four adults of all the cvs was simultaneously inoculated. One set was incubated for 30 h to provide material for

SID 5 (Rev. 3/06) Page 8 of 25

assessment of Bgh penetration success; two other sets were incubated for 72 and 96 h for assessments of colony development, while a final set was used for macroscopic assessment of latent period (days to symptom appearance), final disease severity (percentage leaf area affected; [3]) and disease phenotype a 0-4 scale [4]. Inoculation was by settling tower with ca 10 conidia mm-2. Inoculated plants were incubated at 20 C under 12 h light (06:00-18:00) at 200 μmol m-2 sec-1, 70% RH. For microscopy leaves were prepared as described previously [5].

RESULTSAttempts to generate Bgh sexual recombinants were unsuccessful as no cleistothecia were produced, possibly because only a single mating type was present or because the environmental conditions within isolation compartments were unsuitable. The results from assessments of disease development from the clonal populations cultured in isolation showed no good evidence that trained isolates from Optic or County had adapted to their host cv. The data obtained from the 2004 tests are presented here in detail. Penetration success and host responses: Penetration (haustorium formation) into leaves incubated for 30 h was assessed by examining 100 germlings on each leaf. Percentage data were calculated and transformed to angles before analysis of variance (ANOVA) using Genstat (inspection of residuals showed data conformed to normality). Table 8.2.1 shows the mean percentage (transformed) penetration by germlings from each mildew isolate on cvs Optic, County and Golden Promise. There were significant (P<0.001) main effects of leaf position and test cv. and, as expected, third leaves were more susceptible than the adult fifth leaves (means = 18.42 and 11.69) while Golden Promise was far more susceptible than Optic or County. Overall, the Bgh populations arising from County penetrated significantly (P<0.001) more frequently than those from Optic, but their culture on the highly susceptible Golden promise had no effect on their performance and there was no significant isolate x cv. interaction indicating that fungal penetration potential was not affected by ‘training’.

Table 8.2.1. Percentage (transformed) penetration by Bgh germlings from populations sourced from cvs Optic and County and then cultured continuously for 2 yrs on either their source cv. or on the highly susceptible cv. Golden Promise.

IsolateSource

Culture cultivar

Test cultivar

Optic County Golden Promise

Mean

Optic Optic 5.84 6.63 21.19 11.12Golden Promise 5.58 8.31 27.4 13.76

County County 13.02 13.37 27.46 17.95Golden Promise 5.93 15.17 31.03 17.39

Mean 7.59 10.8 26.78

The frequency with which appressoria of all isolates penetrated was significantly (P<0.001) higher in third than fifth leaves that expressed adult plant resistance. In all cases the great majority of failures was associated with defensive papillae that were autofluorescent indicating that they contained phenolic compounds [6]. However, some attacked epidermal cells showed whole-cell autofluorescence indicating that their death as a result of attack [7]. Cell death can be an important component of durable resistance [8] but here its frequency was significantly (P< 0.001) higher in third than fifth leaves (means = 8.45 and 4.67, respectively) presumably because papilla defence was more effective in the later-formed leaf. Nevertheless, in all cases cell death frequency was relatively rare (ranging from <1%-5%) and even though there were significant differences between isolates (P< 0.05) and test cv (P<0.01) the biological significance of these small differences is questionable.Colony development: Counts of secondary haustoria formed by a colony accurately reflect its past growth and future potential [9]. Ten colonies were examined on every leaf and the number of secondary haustoria counted. Data subjected to ANOVA (Genstat) were normally distributed and since trends from leaves fixed after 72 and 96 h incubation were similar, Table 8.2.2 presents data only from the older colonies. As expected, colonies formed significantly (P<0.001) more haustoria in third than fifth leaves (means = 13.46 and 6.80, respectively), and although most were formed in Golden Promise, the number formed by all isolates was greater in Optic than County. Overall, the populations sourced from County produced more haustoria than those from Optic, but their culture on the highly susceptible Golden promise had no effect on their performance. Furthermore, although there was a significant isolate x cv. interaction (P=0.001) shows that this was largely explained by the very large colonies formed on Golden Promise by the population sourced from, and cultured on cv. County. Thus, colony growth potential was not improved by ‘training’.

Table 8.2.2. Numbers of secondary haustoria formed by Bgh colonies arising from populations sourced from cvs Optic and County and then cultured continuously for 2 yrs on either their source cv. or on the highly susceptible cv. Golden Promise.

IsolateSource

Culture cultivar

Test cultivar

Optic County Golden Promise

Mean

Optic Optic 8.86 4.85 13.91 9.21Golden Promise 8.34 6.21 13.21 9.25

SID 5 (Rev. 3/06) Page 9 of 25

County County 9.75 4.96 20.43 11.71Golden Promise 11.32 4.36 15.32 10.34

Mean 9.57 5.10 15.72

Macroscopic symptoms: As expected, latent period was overall shorter (P< 0.001) on third than fifth leaves (means = 4.2 and 5.4 d, respectively), and on Golden Promise than on Optic and County and percentage leaf area affected and IT were greater for third leaves. The data for percent cover required subjective judgement and must be treated cautiously, because by 7 d after inoculation, some leaves were senescing. Overall, cv Optic showed the lowest area affected (Table 8.2.3), and although County gave a value similar to Golden Promise, colonies on Optic fifth leaves did not appear to be sporulating by 7 d, they were almost invariably associated with necrosis (indicated by ‘n’), and underlying tissues were chlorotic. Infection type on Optic was superficially similar, but colonies were sporulating although smaller and more frequently associated with necrosis than those on Golden Promise.

Table 8.2.3. Latent period (LP), leaf area affected at 7 d (% LA), and infection type [IT] resulting from inoculation with Bgh populations sourced from cvs Optic and County and then cultured continuously for 2 years on either their source cv. or on the highly susceptible cv. Golden Promise.

IsolateSource

Culture cultivar

Test cultivar

Optic LP %LA IT

CountyLP %LA IT*

Golden Promise LP %LA IT

Mean LP %LA

Optic Optic (5.1) 2.5 [3n,1n] (5.5) 9.4 [3n,2n] (4.0) 11.3 [4,3n] (4.8) 7.7

Golden Promise (4.3) 8.3 [3n,2n] (5.9) 12.5 [3n,2n] (4.0) 10.6 [4,3n] (4.7) 10.5

County County (4.9) 3.5 [3n,2n] (5.8) 13.8 [3n,2n] (4.0) 15.0 [4,3n] (4.9) 10.8

Golden Promise (4.5) 6.4 [2n,3n] (6.0) 17.0 [3n,2n] (3.6) 20.1 [4,3n] (4.7) 14.5

Mean (4.7) 5.2 (5.8) 13.2 (3.9) 14.3*NB. Colonies on cv. County fifth leaves were not sporulating by 7 d.

DISCUSSIONAfter continuous culture on their source cv. since late spring 2002, there was no good evidence to suggest that either of the original powdery mildew populations has adapted to give increased pathogenicity on either its source host or the other two cultivars employed in the tests. Throughout, both County and Optic maintained a good level of adult plant resistance that reduced both penetration success and post-penetration colony development. In this sense they remained effectively durable.At the time of the 2004 test, each mildew population is estimated to have been through at least 140 asexual generations growing at the near optimal conditions provided in the glasshouse. In the field, where environmental conditions are often far from optimal, it is likely that the fungus goes through no more than an average of four generations per month over a 5 month growing season for spring barley. Thus, the test applied here equates to approximately 7 years of continuous culture on the same cultivar under favourable conditions and in the absence of control measures. In view of the absence of sexual recombination it was impossible to determine whether this would have allowed adaptation of the pathogen.

REFERENCES1. Ann. Rep. 2002. Ed. JDS Clarkson. The United Kingdom Cereal Pathogen Virulence Survey Committee, Cambridge, UK. 2. Brown JKM. 2002. In: The Powdery Mildews: a Comprehensive Treatise. Eds R.R. Belanger, A.J. Dik, W.R. Bushnell and T.L.W. Carver. American Phytopathological Society, St. Paul, USA. pp. 107-125.3. Large EC, Doling DA. 1962. Plant Pathol. 11, 47-55.4. Moseman JG et al. 1965. Trans. Br. Mycol. Soc. 48, 479-489.5. Carver TLW et al. 1991. Physiol. Mol. Plant Pathol. 39, 269-287.6. Carver TLW et al. 1994. Physiol. Mol. Plant Pathol. 44, 261-272.7. Zeyen RJ et al. 1995. Physiol. Mol. Plant Pathol. 47, 119-140.8. Carver T L W et al. 1999. Physiol. Mol. Plant Pathol. 55, 183-196. 9. Carver TLW, Carr AJH. 1978. Ann. of Appl. Biol. 88, 171-178.

-------------------------------------------------------

8.3. Investigate the physiologic basis of locally induced resistance (inaccessibility) and susceptibility (accessibility) to cereal powdery mildew. (Proposal Objective 3). All milestones were completed.

INTRODUCTION

SID 5 (Rev. 3/06) Page 10 of 25

In the field, leaves are repeatedly attacked by pathogens as spores are deposited from aerial pool. A series of papers arising from a previous Defra project (CE0154, listed in Prats et al. (1) below) showed that cereal cells' history of attack by, and response to, powdery mildew attack dramatically affects their subsequent ability to resist attack, and this has dramatic consequences for field resistance. Where initial attack fails due to papilla defence, the cell and its near neighbours are 'induced' to show extremely high inaccessibility to subsequent attack. By contrast, where initial attack succeeds, the infected cell and its near neighbours are induced to show extremely high accessibility. This is true irrespective of the inherent resistance of the host. The implications of these observations are profound in terms of exploiting quantitative resistance; there was little understanding of the basis of these induced changes and this was explored here.

PUBLISHED OUTCOMES: These papers give all details of the relevance of the studies, their context, the methods used and results and, since they are publicly available, only the summaries are presented. [NB. Only Carver, Roberts and Thomas were funded by Defra so inputs by others gave value added from other funding sources.]

[1] This paper shows the effects of energy deprivation and inhibition of phenolic compound synthesis on expression of induced inaccessibility in oat. [1] Prats E, Carver TLW, Lyngkjær MF, Roberts PC, Zeyen RJ. (2006). Induced inaccessibility and accessibility in the oat powdery mildew system: insights gained from use of metabolic inhibitors and silicon nutrition. Mol. Plant Pathol. 7(1), 47-59.Fungal-induced inaccessibility in oat to Blumeria graminis requires active cell processes. These are reiterative de novo cell processes involved in inherent penetration resistance. Therefore, induced inaccessibility may well involve cellular memory of the initial attack. Phenylpropanoid biosynthesis inhibitors (AOPP and OH-PAS) and phosphate scavengers (DDG and D-mannose) strongly suppressed induced inaccessibility, but silicon nutrition had no effect. Induced accessibility was modulated by the presence of fungal haustoria inside cells. Haustoria actively suppress or reprogram infected plant cells toward a constant state of penetration susceptibility. Neither inhibitor treatments nor silicon nutrition affected fungal induced accessibility.

[2, 3] These papers show that accessibility to non-pathogens is induced by prior infection with a virulent form but, conversely that inaccessibility can be rapidly induced by extracellular materials released by non-pathogenic powdery mildew conidia.

[2] Olesen KL, Carver TLW, Lyngkjaer MF. 2003. Fungal suppression of resistance against inappropriate B. graminis ff.spp. in barley, oat and wheat. Physiol. Mol. Plant Pathol. 62, 37-50.When barley, wheat or oat leaf epidermal cells were attacked by their appropriate forma specialis (f.sp.) of Blumeria graminis DC. Speer (f.sp. hordei, tritici and avenae, respectively), many attempted penetrations succeeded, functional haustoria were formed and very few plant cells died. When attacked by either of the two possible inappropriate ff.spp., penetration attempts failed in association with papilla deposition by epidermal cells, attacked cells died, or, if visible haustoria were formed, the plant cell died very soon afterwards. Double inoculation experiments were performed where each cereal species was first attacked by its appropriate f.sp., as inducer, and later by the different ff.spp. as challenger. Infection by the appropriate inducer profoundly affected cellular responses to challenger attack. Suppression of defensive responses was dramatic within epidermal cells containing the inducer haustorium, evident to some extent in adjacent cells, but undetectable at two cells distance. Suppression of penetration resistance allowed most challenger attacks, even by inappropriate ff.spp., to form a haustorium. Furthermore, death of penetrated epidermal cells was also suppressed so that haustoria of the inappropriate ff.spp. functioned to support colony development. In oat, delayed epidermal cell death prevented full colony development by inappropriate ff.spp., but in barley and wheat, no cell death was apparent by four days after inoculation and colonies of the inappropriate ff.spp. produced extensive hyphae, secondary haustoria and conidial chains.

[3] Fujita A, Suzuki T, Kunoh H, Carver TLW, Thomas BJ, Gurr SJ , Shiraishi T. (2004). Induced inaccessibility in barley cells exposed to extracellular material released by non-pathogenic powdery mildew conidia. Physiol. Mol. Plant Pathol. 64, 169-178.

Before germinating, conidia of Blumeria graminis f. sp. hordei (Bgh) and tritici (Bgt) and Erysiphe pisi (Ep) rapidly released extracellular material (ECM) onto barley coleoptile cells. More was released if full rather than partial cell contact was made. Within 5-6.5 h of receiving ECM from Ep or Bgt (non-pathogens), cells showed induced inaccessibility to challenger Bgh. This effect was greater for Ep than Bgt where full contact was required. Abiotic particles and ECM from Bgh did not affect cells’ accessibility to subsequent challenger attack by Bgh. 　 This induced inaccessibility must be due to active component(s) within conidial ECM. [4, 5]. The possible involvement of nitric oxide (NO) in signalling within and between barley cells expressing papilla-based and hypersensitive resistance to powdery mildew was demonstrated for the first time. Thus, NO may be involved induction of accessibility not only in attacked cells but also in their near neighbours. The finding that NO was generated in stomatal complexes close to attacked cells led directly to the work reported under Objective 7.

SID 5 (Rev. 3/06) Page 11 of 25

[4] Prats E, Mur AJ, Sanderson R, Carver TLW. (2005). Nitric oxide contributes both to papilla-based resistance and the hypersensitive response in barley attacked by Blumeria graminis f. sp. hordei. Mol. Plant Pathol. 6, 65-78.

Blumeria graminis f.sp. hordei (Bgh) attack disrupted stomatal behaviour, and hence leaf water conductance (gl), in barley genotypes Pallas and Risø-S (susceptible), P01 (with Mla1 conditioning a hypersensitive response, HR), and P22 and Risø-R (with mlo5 conditioning papilla-based penetration resistance). Inoculation caused some stomatal closure well before the fungus attempted infection. Coinciding with epidermal cell penetration, stomatal opening in light was also impeded, although stomata of susceptible and mlo5 lines remained largely able to close in darkness. Following infection, in susceptible lines stomata closed in darkness, but opening in light was persistently impeded. In Risø-R, stomata recovered near-full function by around 30 h after inoculation, i.e. after penetration resistance was accomplished. In P01 stomata became locked open and unable to close in darkness shortly after epidermal cells died due to HR. In the P22 background, mlo5 penetration resistance was often followed by consequential death of attacked cells, and here too stomata became locked open, but not until ca 24 h after pathogen attack had ceased. The influence of epidermal cell death was localised, and only affected stomata within one or two cells distance. These stomata were unable to close not only in darkness but also after application of abscisic acid and in wilted leaves suffering drought. Thus, resistance to Bgh based on HR or associated with cell death may have previously unsuspected negative consequences for the physiological health of apparently ‘disease free’ plants. The results are discussed in relation to the control of stomatal aperture in barley by epidermal cells.

[5] Mur LAJ, Carver TLW, Prats E. (2006) NO way to live; the various roles of nitric oxide in plant-pathogen interactions. J. Exp. Bot. 57(3), 489-505.Consideration of NO in plant-pathogenic interactions usually focuses on its roles in eliciting or confining plant cell death in the hypersensitive response (HR). We examined NO action in the development of both HR resistance and susceptible responses in three plant-pathogen interactions; tobacco and Arabidopsis interacting with pathovars of the bacterium Pseudomonas syringae, and barley (Hordeum vulgare) challenged with the powdery mildew fungus, Blumeria graminis. We have developed a novel approach to measure in planta NO production in tobacco and Arabidopsis based on photoacoustic laser detection. In barley, resistance is conditioned not only by HR but also by prevention of penetration whereby living cells deposit localised wall appositions (papillae). Spatio-temporal NO production elicited by B. graminis was visualised using the NO-specific fluorescent dye, DAF-2DA.[6, 7] Since epidermal cells of even the most cereal susceptible genotype can be rendered virtually totally resistant to penetration by powdery mildew, it is clear that all genotypes have the potential to express this form of resistance and suggests the likelihood that it is gene regulation rather than structure that is key. These papers address the fact that transcript analysis of entire organs or tissues cannot reveal activity specifically associated with induced cells, or indeed the genetic basis of inherent resistance or susceptibility to penetration. Work demonstrating the isolation of transcripts from single epidermal cells of known response type, initiated under a former Defra project (CE0154), has now been published. This will make possible studies of genetic activity during expression of not only inherent resistance, but also induced inaccessibility.

[6] Gjetting T, Carver TLW, Skøt L, Lyngkjær MF. (2004).Gene expression profiling of individual barley epidermal cells attacked by powdery mildew. Mol. Plant-Micr. Inter. 17, 729-738.Resistance and susceptibility in barley to the powdery mildew fungus (Blumeria graminis f.sp. hordei) is determined at the single-cell level. Even in genetically compatible interactions, attacked plant epidermal cells defend themselves against attempted fungal penetration by localised responses leading to papilla deposition and reinforcement of their cell wall. This conveys a race non-specific form of resistance. However, this defence is not complete and a proportion of penetration attempts succeed in infection. The resultant mixture of infected and uninfected leaf cells makes it impossible to relate powdery mildew-induced gene expression in whole-leaves or even dissected epidermal tissues to resistance or susceptibility. A method for generating transcript profiles from individual barley epidermal cells was established and proven useful for analysing resistant and successfully infected cells separately. Contents of single epidermal cells (resistant, infected and unattacked controls) were collected and, after cDNA synthesis and PCR amplification, the resulting sample was hybridised to dot-blots spotted with genes including some previously reported to be induced upon pathogen attack. Transcripts of several genes, e.g. PR1a – encoding a pathogenesis related protein, and GLP4 – encoding a germin-like protein, accumulated specifically in resistant cells, while GRP94 – encoding a molecular chaperone, accumulated in infected cells. Thus, the single-cell method allows discrimination of transcript profiles from resistant and infected cells. The method will be useful for micro-array expression profiling for simultaneous analysis of many genes.

[7] Gjetting T, Hagedorn PH, SchweizerP, Thordal-Christensen H, Carver TLW, Lyngkjær MF. (in press). Single-cell transcript profiling of barley attacked by the powdery mildew fungus. Mol. Plant-Micr. Inter.In many plant-pathogen interactions there are several possible outcomes for simultaneous attacks on the same leaf. For instance, an attack by the powdery mildew fungus on one barley leaf epidermal cell may succeed in infection and formation of a functional haustorium whereas a neighboring cell attacked at the same time may resist fungal penetration. To date, the mixed cellular responses seen even in susceptible host leaves have made it difficult to relate induced changes in gene expression to resistance or susceptibility in bulk leaf samples. By micro-extraction of cell-specific mRNA and subsequent cDNA array analysis, we have successfully obtained separate gene expression profiles for specific mildew-resistant and -infected barley cells. Thus, for the first time it

SID 5 (Rev. 3/06) Page 12 of 25

is possible to identify genes that are specifically regulated in infected cells and presumably involved in fungal establishment. Further, although much is understood about the genetic basis of effective papilla resistance associated with mutant mlo barley, we provide here the first evidence for gene regulation associated with effective papilla-based non-specific resistance expressed in nominally ‘susceptible’, wild-type barley.

ADDITIONAL STUDIESCordycepin has been used to inhibit gene transcription in B. graminis-attacked barley [Schweizer P et al. 1996. Physiol. Mol. Plant Pathol. 49, 103-120]. A range of cordycepin concentrations spanning the published optimum was tested. It was applied by dipping inoculated, cut leaves into solution. Microscope analyses of fungal development indicated that cordycepin was fungistatic. Even at 0.2mM, (50% the recommended optimum), only 2% of appressoria penetrated compared to ca 39% in controls. A further experiment therefore used 0.1mM Cordycepin, and in this case there was no significant effect of treatment. Thus, with its apparent anti-fungal effects at higher concentration, and its lack of effect at lower concentration, it was concluded that cordycepin was unsuited for experimentation on induced (in)accessibility. Cytochalasin E (CytE) inhibits microfilament polymerisation and increases infection in the cowpea rust system [Skalamera D, Heath MC. 1998. The Plant Journal 16, 191-200]. Preliminary work tested effects on basal susceptibility: CytE was applied by bathing the mesophyll of leaves (from which the abaxial surface had been stripped) with various concentrations and 0.2 μg ml-1 gave the greatest increase in successful penetration compared to controls (treated with 0.02ul ml-1 DMSO). Concentrations of 0.5 and 1.0 μg ml-1 gave less penetration but significantly greater levels than their respective controls, while results from 0.1 μg ml-1 were variable. A double inoculation experiment therefore used 0.2 μg ml-1 CytE to test effects on induced (in)accessibility. The adaxial surface of eight first leaves of intact Pallas barley plants was inoculated with 10 conidia mm -2 as inducer and incubated for 24 h before removal of superficial fungal structures with latex. The abaxial epidermis was then removed and central segments of four leaves were floated on CytE and four on control solution, bathing the mesophyll. Non-induced controls were treated identically. Challenger inoculum was applied to the adaxial surface and incubated for 42 h. In non-induced controls, CytE significantly (P=0.005) increased basal susceptibility to penetration (ca 46% in CytE; 32% in DMSO). Effects on induced accessibility and inaccessibility were assessed in double inoculated leaves [1]. This showed that, as expected, control cells containing an inducer haustorium were highly accessible to the challenger (98% penetrated) while cells with an inducer papilla were totally resistant to challenger attack (0% penetrated). Interestingly, CytE treatment had no significant effect on either phenomenon (P>0.45 for both cases). Thus, it affected neither induction of inaccessibility nor inaccessibility or transmission of the effects to adjacent cells. This suggests that cytoskeleton reorganisation may not be key to either phenomenon. However, it also supports the view that induced inaccessibility is a stable and effective form of resistance, likely to be durable, and that breeding lines with high inherent penetration resistance should enhance their expression of induced inaccessibility. This would be of clear value in protecting crops produced under both low-input (including organic) and intensive systems.

-------------------------------------------------------

8.4. Explore the role of Graminaceous epicuticular leaf wax components in disrupting germling development by the powdery mildew fungus. (Proposal Objective 4)This study was proposed as a minor component of AR0712. All milestones were achieved apart from gaining sufficient data to add information regarding the location of a suspected gene controlling a disruptive leaf wax factor for inclusion in the genetic map of ryegrass. This was because of the death of plants contributing to the mapping population (see below).

INTRODUCTIONPrevious studies (MAFF and Defra projects CE0120 and CE0154) showed that early plant/pathogen interactions have a profound effect on the germination and germling development that lead to infection structures (appressoria) formation by powdery mildew fungi [1-4]. There is little evidence that cereals can greatly affect these processes, but there is clear evidence that failure of appressorium development on the abaxial surface of ryegrasses (Lolium spp) relates to leaf surface characteristics and probably to surface wax components [5] that are disruptive. However, in certain Lolium lines carrying a fragment of chromosome introgressed from a genotype of Festuca pratensis, the abaxial leaf surface supports normal appressorium formation. This characteristic appears linked to another which interferes with leaf senescence to convey a 'stay green' phenotype. This and other plant material generated by the IGER Molecular and Applied Genetics group (BBSRC-funded) provided a resource for the studies.

MATERIALS AND METHODS A number of genotypes of Lolium perenne (perennial ryegrass), L. multiflorum (Italian ryegrass) and L. temulentum bearing a segment of a chromosome carrying a stay-green gene from Festuca pratensis (Meadow fescue) were examined for powdery mildew resistance in glasshouse tests and under controlled conditions. Light microscopy [1] was used to determine genotypic effects on germling development.Leaf wax consist of a mix of long-chain (20-40 carbons), typically being mainly a mixture of primary alcohols, aldehydes, fatty acids, alkanes and esters. In order to identify components that promote or inhibit powdery mildew

SID 5 (Rev. 3/06) Page 13 of 25

germling development, it would be most useful to test fungal behaviour on waxes or their individual components applied to artificial substrata remote from the influences of other plant factors. To explore this possibility, epicuticular waxes were collected from oat leaves, by dipping in chloroform, and by collection with a chloroform-soaked cotton bud, from the adaxial and abaxial surface of L. multiflorum (Bb2364-94) where powdery mildew germling development is impaired on the abaxial leaf surface, and from an L. temulentum line carrying an F. pratensis segment conferring the ‘stay-green’ phenotype (cv. Ceres) and where the adaxial leaf surface supports a relatively high frequency of normal germling development. The chloroform solutions were dried, the different waxes were redissolved in di-ethyl ether and drops (16 μL, containing ca. 0.9 mg ml -1 of wax) were placed either on clean glass or on cellulose acetate film laid on 2% agar. The solvent was evaporated and surfaces were inoculated and incubated in darkness for 17 h at 20ºC, 100% RH before 100 conidia were examined on each surface.

RESULTS AND DISCUSSIONIn planta: Confirming preliminary data, where the F. pratensis segment was present in both Lolium chromosomes, powdery mildew conidia germinated and frequently formed normal functional appressoria and successful infections on the abaxial (lower) leaf surface. In contrast, the abaxial surface of wild-type Lolium lines (lacking the segment) remained mildew free because normal appressoria formed rarely. This supported the view that the F. pratensis segment carries genetic factor(s) that negate the inherent, durable resistance of Lolium’s abaxial surface. Until recently, the factor(s) contributing to this mildew susceptibility had been inseparable from the stay-green factor, raising the question of whether they were the same. However, as part of a separate IGER programme, progeny of an introgression arose that lacked the stay-green factor but were mildew susceptible on the abaxial leaf surface. This indicated that the two factors are independent. A mapping population was available to progress this study, offering the possibility of identifying the plant gene(s) that condition resistance of Lolium abaxial leaf surfaces. Nine Lolium multiflorum (cv. Meribel) lines carrying different length fragments of a chromosome segment carrying the ‘stay-green’ character from F. pratensis genotype Bf993/17 were examined. The L. multiflorum parent was mildew resistant on its abaxial surface, while the stay-green parent was susceptible. Of the nine lines, four were susceptible on both surfaces, two showed limited symptoms and three were completely resistant. However, the data were equivocal because resistance did not relate to the size of the introgressed segment. Unfortunately, these clones showed many characteristics of L. multiflorum, including a short lifespan. To maintain them necessitated repeated re-potting of young vegetative tillers and growth under short-day conditions; nevertheless, plants appeared stressed and ran quickly to head. The poor health of these plants may have affected their ability to produce leaf surface wax and therefore to influence mildew germling development. Attempts to propagate further clonal descendents to repeat the study failed because plants died and it was impossible to complete the planned study and, therefore, to add any reliable information regarding the genetic control of abaxial leaf surface-mildew resistance to the ryegrass genetic map. To progress this work further would require the generation of new introgression lines based on a parental line with a perennial habit.A different population derived from L. perenne x F. pratensis (not stay-green), likely to be perennial, and containing chromosome substitution lines for each of the seven F. pratensis chromosomes, became available. Unsurprisingly, both parents proved resistant to B. graminis on their abaxial surface because they do not carry the ‘stay green’ gene which has been associated with abaxial susceptibility. However, in a glasshouse test, the line carrying chromosome 3 was unique in that it alone developed some powdery mildew on its abaxial surface. Twelve back-cross (generation 2) lines derived from this substitution line were also tested, each being known to carry a segment of the F. pratensis chromosome 3 introgressed into the homeologous L. perenne chromosome. One of these (BX 462/920) also showed some mildew susceptibility on the abaxial leaf surface. Detailed histological analyses of the chromosome substitution lines were therefore undertaken using replicate leaves taken from four similar tillers inoculated and incubated under controlled conditions. The data indicated that the chromosome 3 substitution line did not differ significantly from any other substitution line in that a very low percentage of fungal germlings formed normal appressoria on the abaxial leaf surface of all lines (range = ca 8%-23% [angle-transformed = 16.6-29.1]; chromosome 3 line = 18%); this was significantly (P<0.001) lower than the percentage of normal appressoria formed on their adaxial surfaces (range = 92%-99% [angle-transformed = 73.9-84.4]; overall s.e.d. = 5.74). Furthermore, no appressoria penetrated successfully to form haustoria on the abaxial surface of any line, whereas many did so on the adaxial surface (range 33%-67%). This controlled experiment therefore indicates no effect of F. pratensis chromosome 3 on resistance of the abaxial leaf surface and suggests that the earlier observations were perhaps due to leaf waxes being damaged in the glasshouse. In vitro: When conidia were inoculated onto glass coated with leaf wax, almost all died immediately after forming a single short germ tube irrespective of the source of wax. A pool of liquid was often evident around the tip of the germ tube indicating that primary gem tubes had ruptured had ruptured and extruded the spore cytoplasm. Thus, use of wax on was unsuitable.

SID 5 (Rev. 3/06) Page 14 of 25

Table 8.4.1. Developmental stages reached by powdery mildew germlings (% angle transformed) incubated on cellulose acetate (controls) and on epicuticular waxes collected from oat, or from the ad- or abaxial surface of L. temulentum cv. Ceres (stay green) and L. multiflorum line Bb2364-94. Means based on 100 germinated conidia on each of four replicates per treatment.

Treatment Development of germlingsTwo germ tubes

formedLong germ tube

formedAppressorium differentiated

Control 70.4 55.5 42.5Oat wax 71.9 65.2 58.5cv. Ceres adaxial wax 75.2 64.4 59.1cv. Ceres abaxial wax 62.1 50.8 42.1L. multiflorum adaxial wax 70.2 66.4 62.3L. multiflorum abaxial wax 73.0 59.9 52.6L.S.D. [5%]

L.S.D. [5%] 9.2 6.2 5.4

When conidia germinated on wax deposited on cellulose acetate membrane overlying agar, Table 8.4.1 shows that a reasonably high proportion formed two germ tubes, and while many of these formed an elongated tube, less than 63% eventually differentiated an appressorium, even on wax from susceptible surfaces (oat or the adaxial surface of grasses). Furthermore, the lowest frequency of appressorium differentiation occurred on abaxial wax from cv. Ceres, the stay green line, whereas in planta observations showed a high proportion of apparently normal appressoria on its abaxial surface (ca 81%) compared to wild type Lolium (ca 26%). These contradictory findings suggest that the extraction and deposition of wax from solvents produces unreliable results, possibly because the chemical components become redistributed unnaturally. The ability to generate an axenic system for the bioassay of leaf waxes remains a valuable objective, but this apparently cannot be achieved simply.

ADDITIONAL STUDIES Published work attracted the attention of Prof. M Riederer (Department of Botany, Universität Würzburg, Germany) a world authority on leaf wax, and this led to a collaboration studying effects of pea leaf wax constitution on germling development by the pea powdery mildew fungus. The results were published and the reference and an abstract are given below. Gniwotta F, Vogg g, Gartmann V, Carver TLW, Markus Riederer M, Jetter R. 2005. What do microbes encounter at the plant surface? Chemical composition of Pisum sativum leaf cuticular waxes. Plant Physiol. 139, 519-530.In the cuticular wax mixtures from leaves of Pisum sativum cvs (cvs) Avanta, Lincoln and Maiperle more than 70 individual compounds were identified. They comprised unbranched C25–C28 alkanes, C24 –C32 primary alcohols, C24 –C32 aldehydes, C22–C30 fatty acids, and C40–C50 esters containing C14–C24 fatty acids condensed mainly with C26 primary alcohol. Additionally, small amounts of the secondary alcohols hentriacontan-16-ol, -15-ol and -14-ol as well as nonacosan-13-ol, -14-ol, and -15-ol were detected. The cuticular wax coverage on the adaxial leaf side of all three cvs was significantly lower than on the abaxial surface. The adaxial wax was characterized by very high amounts of primary alcohols (71%), while the abaxial wax consisted mainly of alkanes (73%). An aqueous glue of gum arabic was employed to selectively sample the epicuticular wax layer on pea leaves, and hence to analyze the composition of epicuticular crystals exposed at the outermost surface of leaves. The epicuticular layer was found to contain 74% and 83% of the total wax on adaxial and abaxial surfaces, respectively. The platelet-shaped crystals on the adaxial leaf surface consisted of a mixture dominated by hexacosanol, accompanied by substantial amounts of octacosanol and hentriacontane. In contrast, the ribbon-shaped wax crystals on the abaxial surface consisted mainly of hentriacontane (63%), with ca. 5% each of hexacosanol and octacosanol being present. Based on this detailed chemical analysis of the wax exposed at the leaf surface, their importance for early events in the interaction with host-specific pathogenic fungi can now be evaluated. On adaxial surfaces approximately 80% of Erysiphe pisi spores germinated, and 70% differentiated appressoria. In contrast, significantly lower germination efficiencies (57%) and appressoria formation rates (49%) were found for abaxial surfaces. In conclusion, the influence of the physical structure and the chemical composition of the host surface, and especially of epicuticular leaf waxes, on the pre-penetration processes of biotrophic fungi is discussed.

Much of the current and past work was reviewed in recent book chapters: Kunoh H, Carver TLW, Thomas BJ, Fujita K, Meguro A, Wright AJ. 2004. The extracellular matrix of conidia of powdery mildew fungi: Its functions and involvement in information exchange with host cells. In: Genomic and Genetic Analysis of Plant Parasitism and Defense. Eds: Tsuyumu, Leach, Shiraishi, Wolpert. APS press, MN, USA, pp. 150-163. Carver, TLW & Gurr SJ (2006). Filamentous fungi on plant surfaces. In: Biology of the Plant Cuticle. Annual Plant Reviews 23. Ed. Riederer. Blackwell, Oxford, pp. 368-397

REFERENCES1. Carver TLW, Ingerson SM. 1987. Physiol. Mol. Plant Pathol. 30, 359-372.2. Wright AJ et al. 2000. Physiol. Mol. Plant Pathol. 57, 281-301.

SID 5 (Rev. 3/06) Page 15 of 25

3. Wright AJ et al. 2002. Physiol. Mol. Plant Pathol. 61, 163-178.4. Wright AJ et al. 2002. Physiol. Mol. Plant Pathol. 61, 217-226. 5. Carver TLW et al. 1990. Physiol. Mol. Plant Pathol. 39, 573-583.

-------------------------------------------------------

8.5. Determine consequences of expression of disease resistance for plant stomatal function. (Proposal Objective 07) All milestones completed.

INTRODUCTIONWith Defra’s agreement, this objective was introduced in April 2005 as a result of the observation that nitric oxide (NO) accumulated in stomatal complexes adjacent to powdery-mildew attacked epidermal cells. Since NO is known to play a role in causing stomatal closure, this implied that the expression of resistance to powdery mildew might affect stomatal function with obvious and serious consequences for gas exchange and plant growth. Initial work [1] has been published and since the paper gives all details of its relevance, context, methods used and results, only the summary is presented. [NB. Only Carver and Thomas were funded by Defra so inputs by others gave value added from other funding sources.][1] Prats E, Gay AP, Mur LAJ, Thomas BJ, Carver TLW. (2006) Stomatal lock-open, a consequence of epidermal cell death, follows transient suppression of stomatal opening in barley attacked by Blumeria graminis. J. Exp. Bot. 57, 2211-2226.Blumeria graminis f.sp. hordei (Bgh) attack disrupted stomatal behaviour, and hence leaf water conductance (gl), in barley genotypes Pallas and Risø-S (susceptible), P01 (with Mla1 conditioning a hypersensitive response, HR), and P22 and Risø-R (with mlo5 conditioning papilla-based penetration resistance). Inoculation caused some stomatal closure well before the fungus attempted infection. Coinciding with epidermal cell penetration, stomatal opening in light was also impeded although stomata of susceptible and mlo5 lines remained largely able to close in darkness. Following infection, in susceptible lines stomata closed in darkness but opening in light was persistently impeded. In Risø-R, stomata recovered near-full function by around 30 h after inoculation, i.e. after penetration resistance was accomplished. In P01 stomata became locked open and unable to close in darkness shortly after epidermal cells died due to HR. In the P22 background, mlo5 penetration resistance was often followed by consequential death of attacked cells, and here too stomata became locked open, but not until ca 24 h after pathogen attack had ceased. The influence of epidermal cell death was localised, and only affected stomata within one or two cells distance. These stomata were unable to close not only in darkness but also after application of abscisic acid and in wilted leaves suffering drought. Thus, resistance to Bgh based on HR or associated with cell death may have previously unsuspected negative consequences for the physiological health of apparently ‘disease free’ plants. The results are discussed in relation to the control of stomatal aperture in barley by epidermal cells.NB. The Editor-in-Chief of the journal Plant Signaling & Behavior has invited us to write an addendum (commentary) on this work noting that such contributions are solicited for ‘the most significant recent and forthcoming papers, published elsewhere, to provide a short summary with additional insights, new interpretations or speculation on the relevant topic’. The finding that HR to B. graminis controlled by the Mla1 allele caused permanent stomatal lock-open raised the question of the generality of this effect in relation to HR conditioned by other alleles and loci. This was examined in a series of studies that are in prep. for publication. The approaches were similar to those used in [1] and the summary of the submitted work is given as reference [2].

[2] Prats E, Gay AP, Roberts PC, Thomas BJ, Sanderson R, Paveley ND, Lyngkjær MF, Carver TLW, Mur LAJ. (submitted). Temporal and spatial relationships between stomatal dysfunction and cell death due to different single gene resistances and non-host resistance in barley attacked by Blumeria graminis. J. Exp. Bot.Early, non-specific responses to Blumeria graminis f.sp. hordei decreased leaf water conductance (gl) in various barley isolines, and, later, established infection of a suscept reduced light gl although dark gl was unaffected until leaves senesced. Ultimately, isolines with Mla1, Mla3 or MlLa resistance all showed abnormally high dark gl. This followed execution of hypersensitive responses (HR) which caused nearby stomata to lock open. Although all resistant lines eventually gave a high frequency of HR, its spatiotemporal expression was unique to each gene. Rapid HR in Mla1, mostly limited to single epidermal cells, arrested fungal growth before colonies initiated secondary attacks. With Mla3, mesophyll HR preceded delayed HR of epidermal cells whose initial survival supported spreading colonies that produced secondary infections. With MlLa, mesophyll survived but attacked epidermal cells showed contrasting behaviour: some died immediately while others survived infection supporting colony growth and secondary infection until they eventually died. The timing and extent of increased dark gl

varied between barley lines according to the timing and frequency of HR and consequential stomatal lock-open. Lock-open also followed single epidermal cell HR caused by the non-pathogen B. graminis f.sp. avenae, although here barley genotype affected neither the expression of HR, nor the timing or extent of increased dark gl. Although chloroplast autofluorescence indicated that guard cells of locked open stomata were alive, they were unable close in response to either drought or abscisic acid. The data support the view that HR-based resistance carries a potential cost due to consequential stomatal dysfunction.

SID 5 (Rev. 3/06) Page 16 of 25

ADDITIONAL STUDIESEffects of brown rust resistance in barley and wheat The studies of powdery mildew interactions suggested that major (R) gene resistance associated with HR cell death has a common consequence of loss of stomatal function that has severe implications even if the attacked plants appear disease free. It was important to determine if R gene resistance in other cereal-pathogen systems had similar consequences. Barley and wheat brown rust interactions were considered in separate trials.Methods. Following experimental designs similar to [1], barley plants were grown to full expansion of their first leaf under standard conditions of 70% RH and under 12 h light (10:00-22:00 at 450 umol m-2 sec-1) at 20˚C and 12 h dark (22:00-10:00) at 15ºC. In wheat, second leaves were used because first leaves were too narrow for porometer measurements. Uredospores (100 mm-2) of the barley brown rust fungus, Puccinia hordei (race octal BRS 273, isolate 03-23), were applied to the first leaf adaxial surface of three barley cvs carrying different R genes. In prior tests (ERL Jones, pers. comm.) cv. Gold was fully susceptible while resistance was shown by cvs Estate (IT = 0c; R gene Rph 3) and Cebeda Capa (IT = 0n; R gene Rph 7). Similarly spores of wheat brown rust, P. triticina (syn. P recondita) isolate WBRS-04-02, were applied to second formed leaves of three wheat isolines with different R genes. In prior tests (ERL Jones, pers. comm.) cv. Thatcher was fully susceptible, while resistance was shown by Lr20 (IT = 0c,n) and Lr24 (IT = mainly 0c but with a few small pustules). After inoculation, all plants, including uninoculated controls, were held for 16 h in darkness under mist at 15˚C before being returned to standard conditions. Leaf water conductance (gl) was measured by porometer in the middle of light periods and 1 h before the end of dark periods for several days after inoculation.Results. Diurnal patterns of change in gl in healthy and inoculated barley and wheat lines are shown in Figures 8.5.1 and 8.5.2, respectively.

SID 5 (Rev. 3/06) Page 17 of 25

Gold

0

200

400

600

NSNS NS

NS

Estate

0

200

400

600

NS NS * *

24 41 48 65 72 89 96 113 120 137 144 1610

200

400

600 Cebeda Capa

NSNSNS NS

Hours after inoculation

Leaf water conductance

(mm

ol m-2 sec-1)

Fig. 8.5.1. Leaf water conductance in healthy (O) and P. hordei attacked (▲) leaves of barley cvs Gold (susceptible), Estate and Cebeda Capa in successive light (unshaded) and dark periods (grey shaded) after inoculation. Comparing within sample times: NS = no significant difference; * = P<0.05; all others P< 0.001.

Leaf water conductance

(mm

ol m-2 sec-1)

Fig.8.5.2. Leaf water conductance in healthy (O) and P. triticina attacked (▲) leaves of wheat lines Thatcher (susceptible), Lr20 and Lr24 in successive light (unshaded) and dark periods (grey shaded) after inoculation. Comparing within sample times: NS = no significant difference; * = P<0.05; ** = P<0.05; all others P< 0.001.

0

200

400

600 Thatcher

NS NS

24 41 48 65 72 89 96 113 120 137 144 161 168 185 192 209 2160

200

400

600 Lr24

NS NS * * *

*****

Lr20

0

200

400

600

**

Hours after inoculation

All healthy plants showed high gl in light periods when stomata were open, and low gl when stomata had closed in response to darkness. In the early stages, the gl of inoculated, susceptible plants (Gold barley, Thatcher wheat), was substantially and significantly (P<0.001) reduced in light, indicating that endophytic infection had impeded stomatal opening, and slightly, though often not significantly, reduced in darkness. However, when pustules started to erupt (at 113 h.a.i. in Gold, and 161 h.a.i. in Thatcher), dark gl increased indicating loss of water via eruption sites. Nevertheless, light gl of infected leaves never exceeded that of the healthy even when pustules were fully erupted (144 h.a.i. in Gold, 216 h.a.i. in Thatcher), indicating that stomata remained unable to open fully in response to light.During the early stages of interaction, both resistant barleys showed inoculation-induced suppression of gl and little effect on dark gl, as in cv. Gold. The same was true in Lr20 wheat, though here light gl was suppressed more than in Thatcher. However, in all these resistant lines, where no pustules formed light gl remained low, and there was no substantial late-stage increase in dark gl. Thus, despite lack of visible disease, expression of resistance led to a permanent impairment of stomatal opening. A different situation was evident in Lr24. Here, reduction in light gl was evident but less substantial than in Lr20, dark gl was little affected, and this situation did not change with time. This indicates that stomata largely retained functionality even though expression of resistance prevented all but very few, small pustules from forming.Discussion. As in susceptible barley attacked by powdery mildew [1, 2, 3] and bean attacked by rust [see 3], infection of susceptible barley and wheat by the respective brown rust fungi impaired stomatal opening, with obvious deleterious consequences for plant growth and productivity. However, the current studies provide the first evidence that expression of cereals’ resistance to biotrophic pathogens can also have serious consequences via effects on stomata. In barley expressing R gene or non-host resistance to powdery mildew, stomata were so seriously affected that they became permanently locked open and unable to respond to darkness, drought or application of abscisic acid. By contrast, in barley and wheat attacked by avirulent rusts, the stomata could become permanently locked shut. The reason for these opposite effects awaits explanation, but the clear evidence for stomatal dysfunction as a consequence of the expression of all these resistances points to a previously unsuspected factor contributing to the ‘cost of resistance’ which has been recognised but evaded good explanation [4, 5]. Thus, the work highlights serious implications of using such resistances in field crops where stomatal dysfunction would interfere with photosynthesis, respiration, ability to cope with abiotic stresses, and, importantly in the context of environmental change, water use efficiency. For sustainable production systems where powdery mildew is a threat, a good alternative is offered by broad spectrum, papilla-based resistance, which leads to only transient impairment of stomatal opening [1]. The current work showed, however, that some R gene resistances may cause little stomatal dysfunction. None were yet identified for powdery mildew resistance [2] but the wheat brown rust resistance of Lr24 is promising. The reason for this remains unknown but deserves investigation to facilitate breeding for improved resistance phenotype. In addition, more extensive physiological and field studies are required to explore the implications of HR for costs of resistance. Towards this requirement, and informed by established collaboration between Carver and ND Pavely (Defra Fellowships AR0712 and AR0511), a recent proposal to Defra, led by Pavely and involving Dr J. Foulkes (University of Nottingham), includes work aimed at identifying strategies to minimise the impact of disease on water use efficiency, by quantifying the trade off between water and dry matter loss due to disease, and the benefits and costs of water use of disease resistance.Additional References3. Ayres PG. 1981. Soc. Exp. Biol. Seminar Series 8 Stomatal Physiology. Eds Jarvis PG, Mansfield TA. Cambridge University Press, 205-222. 4. Purrington CB. 2000. Curr. Opin. Plant Biol. 3, 305-308.5. Brown JKM. 2002. Curr. Opin. Plant Biol. 5, 339-344.

-------------------------------------------------------

8.6. Characterisation of a collection of wheat varieties for durable resistance against current populations of the wheat yellow rust pathogen carrying complex virulences. (Proposal Objective 5). All milestones were completed.

INTRODUCTIONYellow rust of wheat, caused by Puccinia striiformis f.sp. tritici (Pst), is a major disease of UK wheat [1] controlled through a combination of resistance gene deployment and fungicide application. Many sources of yellow rust resistance have proved race-specific, and thus ephemeral. To support sustainable wheat production it is increasingly important to identify and genetically characterise durable sources of resistance [2]. Dr Roy Johnson identified a number of wheat cvs which appeared to retain good field resistance to yellow rust despite being grown over extensive acreages and years (Durable Set; Table 8.6.1) and therefore met the criteria he defined [3, 4] for ‘durability’ of resistance. However, the genetic and mechanistic basis of these durable resistances was never tested. The objective of this study was to retest the Durable Set against modern Pst isolates with complex virulence under different environmental conditions and to generate materials for a genetic study of durable yellow rust resistance.

SID 5 (Rev. 3/06) Page 18 of 25

MATERIALS AND METHODSField trials. The wheat Durable Set (Table 8.6.1) included 21 cvs that were field tested for yellow rust resistance in 2003/2004 at New Found Farm, Norwich, and again in 2004/2005 at Church Farm, Norwich and at Osgodby, Market Rasen, Lincolnshire. In each trial three replicates were sown in a completely randomised design, containing 10 plants of each cultivar per row. Yellow rust was introduced to sites on spreader plants of the cv. Lemhi. These were infected with one of four Pst isolates (Table 8.6.2) collected from wheat crops within the UK in recent years. Isolates were multiplied under containment, and their virulence genotypes confirmed using the wheat World and European differential sets for yellow rust virulence identification, and the AvocetS near-isogenic lines carrying known yellow rust, seedling expressed resistance genes (Yr R-genes; [5]). The tests used standard protocols [6, 7] and reactions were measured using a qualitative, Infection Type scale [8]. In the field, yellow rust infection was measured on two dates during May/June using the modified Cobb scale as a quantitative measure and a qualitative score of necrosis and chlorosis to assess plant response (Table 8.6.1; [7, 9, 10].Greenhouse tests of adult plant resistance to yellow rust. Five cvs from the Durable Set were selected for a greenhouse study of the adult plant yellow rust infection phenotypes (Table 8.6.1). Ten plants of each cv. were grown to first ear emergence under containment conditions and then simultaneously inoculated with a 1:1:1:1 mixture of the four Pst isolates (Table 8.6.2). To assess timing of yellow rust development, plants were examined daily from 10 days after inoculation (dai) and the time of appearance of fully sporulating pustules on flag leaves was recorded. To assess urediniospores production per pustule, segments of flag leaves (2cm long and 2 segments per plant) bearing pustules were cut at 20 dai. The approximate number of pustules on the segment was assessed and urediniospores were collected from each leaf segment and counted [6]. An estimate of the number of urediniospres produced per pustule was calculated for each leaf segment, and a mean taken to represent each cultivar. Analysis of variance and general linear modelling were applied using Genstat for Windows, v.8.0. Comparisons were considered significant at F- and t-value P <0.001.Crosses made between wheat cvs showing durable yellow rust resistance. Three plants of cvs CarstensV and Cappelle Desprez were crossed to cv. Lemhi as the pollen parent, to give F1 seed. Ten F1 plants from each cross were selfed to give F2 seed. The known, durable sources of yellow rust resistance, Yr18 (Avocet S*6/Yr18), Yr29 (Avocet S*6/Yr29) and Yr30 (Pavon) were crossed to the yellow rust susceptible UK wheat cvs Brigadier, Glasgow and Robigus to give F1 seed.

RESULTSField trials. Of the 21 wheat cvs comprising the Durable Set, five remained immune (Bersee; Elite Lepeuple; Flinor; Luke and Widgeon), 12 had partial resistance and four were susceptible (Anza; Bezostaya; Bouquet, Champlain) (Table 8.6.1). Similar yellow rust reactions were seen on all 21 cvs at the two sites near Norwich, in 2004 and 2005 (Table 8.6.1). At the field trial site near Market Rasen, Lincolnshire, all cvs supported more yellow rust infection than at the sites near Norwich, and previously resistant cvs Bouquet and Champlain were susceptible (Table 8.6.1).Yellow rust greenhouse, adult plant tests. The five cvs were selected for greenhouse, adult plant tests on the basis of yellow rust infection phenotypes seen in field trials (Table 8.6.1) and their susceptibility to the Pst isolates at the seedling growth stage. Yellow rust pustules developed on all five cvs, although fewer were seen on the three more resistant cvs, Bersee, Carstens V and Little Joss. There was no significant difference between cvs with regards to the timing of the first appearance of pustules (18-20 days in all cases). In general, cvs Bersee, Carstens V and Little Joss produced fewer urediniospores per pustule than Cappelle Desprez and Vilmorin 27.Crosses made using wheat cvs showing durable yellow rust resistance. CarstensV and Cappelle Desprez have been crossed to the yellow rust susceptible wheat cv. Lemhi to give F2 seed. Crosses have also been made between wheat cvs carrying known sources of durable yellow rust resistance, i.e. resistance genes Yr18, Yr29 and Yr30 and the yellow rust susceptible UK commercial wheat cvs Brigadier, Robigus and Glasgow. F1 seed is available for these crosses.

DISCUSSIONReassessment of the field resistance to yellow rust in the wheat cultivars of the Durable Set identified four cvs, Anza, Bezostaya, Bouquet and Champlain, in which yellow rust resistance had been lost. The remaining 17 cvs retained varying degrees of resistance towards yellow rust, indicating that the durability of the resistance in these cvs had been maintained. These cvs represent a potential source of useful, durable field resistance for yellow rust that can be incorporated into modern, commercial UK wheat cvs. The test of time is the only current means of identifying durable sources of disease resistance, and four cvs had failed this test. Genetic assessment of the remaining cvs would provide information on the nature and number of genes responsible for their durable yellow rust resistance, and thereby the ease with which this resistance could be transferred into new wheat cvs. However, it is clearly important to determine how durable resistance can be recognised, and studies of the component phenotypes of pathogen infection may identify micro-phenotypes that provide indicators of the potential durability of a source of resistance.

SID 5 (Rev. 3/06) Page 19 of 25

Table 8.6.1. Wheat cultivars constituting the Durable Set and their field responses 2003-2005 to a mixture of yellow rust isolates carrying complex virulence.

Wheat cultivar 2003/04NF Farm, Norwich

2004/05Church Farm, Norwich

2004/05Lincolnshire

Notes

Anza 60S 50S 60S susceptibleAtou 0R 0R 25MR/MS chlorotic flecksBerseea 0R 0R 1MR/R chlorotic flecksBezostaya 80S 70S 25MR susceptibleBouquet 0R 0R 60MS/MR chlorotic flecksCappelle Despreza,b 10MR 5MR 10MR partial resistanceCarstens Va 0R 0R 10MRChamplain 10R 10MR 80S susceptibleElite Lepeuple 0R 0R 2MR chlorotic flecksFlander 0R 10MR 3MRFlinor 5R 5R 1MRHoldfast 5R 5R 15MR extensive chlorosisHybrid 46 20R 10-20MR 30MSHuntsman 0R 10MR 3MRLittle Jossa,b 0R 0R 10MRLuke 5R 5R very necrotic chlorosisNugaines 50R 80S 20MS temp sensitive RStarke II 25-30MR 20MR 20MR partial resistanceVilmorin 27a,b 20MR 20MR 5MR partial resistanceWidgeon 0R 10R 5MRYeoman 5R 10R 15MRLemhi 80S 80S 90S susceptible controla cvs selected for greenhouse, adult plant yellow rust infection test.b cvs selected to test the effect of light quanta on yellow rust infection efficiency (see objective 6)R – Resistance; no visible pustule formationMR- Moderate resistance; pustules surrounded by necrotic tissueMS – Moderate susceptible; pustules surrounded by chlorotic tissueS - Susceptible; pustules surrounded by green tissue.

Table 8.6.2. Puccinia striiformis f.sp. tritici isolates used in yellow rust disease tests.

UK Pst isolate Year of first collection Virulent on Yr R-genes Virulent on wheat cvsWYR 96/502 1996 1,2,3,6,9,17 Madrigal; EquinoxWYR 95/12 1995 1,2,3,4,9,13,17 BrigadierWYR 93/24 1993 1,2,3,4,6,CV,13,14 HerewardWYR 90/505 1990 1,2,3,4,7,14 Brock

REFERENCES1. Boyd LA. 2005. J. Ag. Sci. 143, 233-2432. Boyd LA. 2006. J. Sci. Food Agric. 86, 2523-25263. Johnson R. 1981. Phytopathol. 71, 567-5684. Johnson R. 1988. Durable resistance to yellow (stripe) rust in wheat and its implications in plant breeding. In: Breeding Strategies for Resistance to the Rusts of Wheat. Eds. NW Simmonds & S Rajaram, Mexico:CIMMYT, pp. 63-75..5 Lewis CM. 2006. The genetic basis of Puccinia striiformis resistance in UK wheat germplasm. PhD thesis, Faculty of Biological Sciences, University of East Anglia.6. Boyd LA, Minchin PN. 2001. Euphytica 122, 361-368.7. Smith PH et al. 2004. Mol. Plant-Micr. Inter. 17, 1242-12498. Rodrigues P. 2004. Theor. Appl. Genet.109, 425-432.9. Boyd LA, et al. 2002. Genome 45, 1035-1040.10. Ramburan VP, et al. 2004. Theor. Appl. Genet. 108, 1426-1433.

-------------------------------------------------------

8.7. Define the effects of pre-inoculation lighting treatments on predisposition of wheat to infection by yellow rust, Puccinia striiformis f.sp. tritici (Pst). (Proposal Objective 6). All milestones were attained.

INTRODUCTIONA principal difficulty in studying Pst infection is the rarity with which germinated urediniospores locate and enter wheat leaf stomata to establish infection. Some evidence [1] suggested that light quantity received by seedlings prior to inoculation influenced this phenomenon. Here, therefore, microscopy was used to test rigorously this suggestion and define the effects of pre-inoculation light input. A subsequent study examined effects of wheat genetic variation on this light effect which may offer a new and potentially valuable form of resistance.

SID 5 (Rev. 3/06) Page 20 of 25

MATERIALS AND METHODSPlant material, inoculation and incubation. According to experiment, the susceptible wheat cv. Lemhi, and the partial, adult plant resistant cvs CappelleDesprez, Vilmorin 23 and Little Joss, were used as seedling and adult plants. Seedlings were grown for 10 d and adult plants until first ear emergence, in a light-proof compartment of a spore-free greenhouse at 18ºC/16 h under sodium lights (300 μmol m-2 s-1) and 15oC/8 h darkness. Plants were then subjected to various light treatments before inoculation (as published; [2]) with the virulent Pst isolate WYR75/20 (race 232E137) and incubation in a growth cabinet at 8oC and 95% humidity for 24 h, in darkness. They were then returned to the greenhouse under the same conditions in which they were grown and leaves were checked regularly for the emergence of sporulating pustules. Temperature and humidity were monitored throughout. Pre-inoculation light treatments. Three repeat experiments were performed following the light treatments reported by de Vallavieille-Pope et al. [1]. Accordingly, cv. Lemhi seedlings (GS12) were transferred to a 16/8 h cycle of low light (45 μmol m-2)/darkness, at 8oC for 3 d and then held in total darkness for 16 h at 8oC. Seedlings were then divided into five batches (ca. 40/batch) with one batch receiving the following pre-inoculation quantities of light for 6 h at 10oC: 1) 0 mol m-2 (total darkness); 2) ca. 6 mol m-2 (269 umol m-2 s-1); 3) ca. 10 mol m-2 (457 umol m-2 s-1); 4) ca. 12 mol m-2 (536 umol m-2 s-1); 5) ca. 17 mol m-2 (780 umol m-2 s-1); as an additional control, a sixth batch was maintained in the greenhouse under standard growth conditions compartment. After pre-treatments, all seedlings were inoculated with Pst and after 24 h under initial incubation conditions (above), all were removed to a glasshouse where temperature was maintained at between 20 and 22oC, with a relative humidity of between 40 and 45%. Half the seedlings from each treatment were used to measure leaf water conductance (gl) by cycling porometer; this reflected the extent to which the stomata were open. Remaining seedlings were sampled for light microscopy. Porometer readings (12 readings per time point) and samples for light microscopy (three samples per time point) were taken 2 h before inoculation, and 2, 23, 26, 48-50 and 72-74 h afterwards.The lighting regimes used above were complex and difficult to interpret, and involved an unrealistically lengthy dark period before inoculation. Therefore a more realistic, simplified procedure was adopted and a series of eight experiments were carried out, altering the length of time seedlings were kept in the dark and the amount of pre-inoculation quanta of light received, until the conditions for a robust, repeatable assay were attained. In this assay, seedlings of Lemhi were grown under standard greenhouse conditions for 10 d before being subjected to three light treatments for 18 h immediately before inoculation, viz: 1) total darkness, 2) low light (ca. 13 mol m -2) or, 3) high light (ca. 29 mol m-2). The three sets of seedlings were inoculated simultaneously with equal densities of spores and then all three sets of seedlings were subjected to the same light regime. For each set of seedlings the stomatal status was determined by measuring gl 1 hour before inoculation and at various times up to 74 h afterwards. In subsequent experiments using this assay procedure, effects of pre-inoculation light treatments were studied in seedling and adult plants of three wheat cvs from the Durable Set: Vilmorin 27, Little Joss and Cappelle Despez, (see Proposal Objective 5). Here again, plants were placed either in 1) total darkness, 2) low light (ca. 13 mol m -2) or, 3) high light (ca. 29 mol m-2) for up to 18 h before inoculation. From each light treatment 12 plants were used for porometry and 24 for microscopy. Porometer readings and samples for microscopy (4 samples/time point) were taken 1 h before inoculation and 2, 23, 26, 48-50 and 72-74 h afterwards. Light Microscopy. According to experiment, leaf segments (ca. 2 cm) cut from the first leaves (seedling) or flag and boot leaves (adult plants) were prepared for light microscopy as described by [3]. The following observations were made on each leaf segment: total numbers of: 1) Pst spores germinated (germlings); 2) germlings where the germ tube had grown over a stomatal opening; 3) germlings where the germ tube had entered a stomatal opening; 4) germlings that had entered a stomate and formed sub-stomatal vesicles (SSV). Data were subjected to analysis of variance and general linear modelling using Genstat for Windows, v.8.0. Comparisons were considered significant at F- and t-value P<0.001.

RESULTSEffects of pre-inoculation light treatment on seedlings of susceptible wheat cv. Lemhi. With the pre-inoculation light treatments used by de Vallavieille-Pope et al. [1], on Lemhi seedlings exposed pre-inoculation to prolonged darkness, no germlings had entered stomata by 72 h. When exposed to low quanta of light pre-inoculation, some germlings succeeded in entering stomata SSVs, but the numbers were significantly lower than where seedlings had been exposed to high quanta of light. Porometry indicated that, in seedlings exposed to prolonged pre-inoculation darkness, the normal diurnal opening of stomata had been compromised and might explain the failure of stomatal entry. This may have been because the pre-inoculation dark period used for this treatment was unrealistically long. Interestingly, using the simplified light regimes, porometry showed no differences in post-inoculation gl profiles between seedlings exposed to different pre-inoculation light treatments. Thus, in this more realistic system the normal diurnal movements of stomata had not been compromised by the different, pre-inoculation light treatments. Nevertheless, the pre-inoculation light treatments had a significant effect on Pst development and infection efficiency. This is illustrated in Figure 8.7.1 which shows the numbers of urediospores/germlings that had attained different developmental stages at each sample time. In general, the higher the quanta of pre-inoculation light received by a seedling, the larger the proportion of germinated spores observed (Fig.8.7.1; treatment F value=24.28; P<0.001). By 23 h (when seedlings were still in the incubation chamber at 8 oC in darkness) some

SID 5 (Rev. 3/06) Page 21 of 25

urediospores had formed germ tubes on seedlings that had received high, and to a lesser extent low, quanta of light prior to inoculation and even by this early stage some had entered stomata and differentiated a sub-stomatal vesicles (SSV). By contrast, on leaves kept in darkness pre-inoculation few spores were present (presumably indicating loss of ungerminated spores during preparation) and none had germinated. After seedlings had been returned to the greenhouse, and had been under the normal light regime for 2 or more h (26-74 h samples) a few urediospores on the dark treatment seedling had germinated, but very few ever entered stomata. By contrast, both pre-inoculation light treatments led to a number of germlings entered stomates relatively rapidly and formed SSVs.

Fig. 8.7.1. Temporal development of Puccinia striiformis f.sp tritici germlings on seedlings of wheat cv Lemhi inoculated after pre-inoculation treatments with darkness (D), low light quanta (L; ca. 13 mol m-2) or high light quanta (H; ca. 29 mol m-2).

Measured as a proportion of total germlings, however, most formed SSVs in seedlings which had received the highest quanta of light pre-inoculation (Fig. 8.7.1; treatment F value=13.61, P<0.001). This procedure showed, therefore, that since treatment had not affected stomatal movement some other effect of light must have affected the ability of Pst to locate and enter a stomatal cavity.Effects of pre-inoculation light treatment in seedling and adult plants of wheat cvs with field resistance. The same simplified light regimes were then used to compare seedlings of wheat cvs Vilmorin 27, Little Joss and Cappelle Desprez, against Lemhi as the susceptible control. Overall, the quanta of pre-inoculation light received by seedlings of all cvs affected both germination and infection efficiency. As seen previously, high quanta of light again increased spore germination on Lemhi and also Cappelle Desprez, but no such effect was evident in Vilmorin 27 or Little Joss. Nevertheless, the overall effect of high light quanta on enhancing stomatal entry and SSV formation was significant. It was most pronounced on cv Little Joss, followed by Cappelle Desprez and Lemhi. Importantly, however, no significant effect on stomatal entry or SSV formation was observed on Vilmorin 27. The results obtained when the pre-inoculation light regimes were applied adult plants provided surprising results that are hard to interpret. As in seedlings, spore germination was significantly higher on plants exposed to the highest quanta of light (ca. 29 mol m-2) compared to low light quanta (ca. 13 mol m -2), but the highest levels of germination were observed on plants receiving the dark treatment and the controls kept in the greenhouse under normal light conditions throughout. No significant effect of the light treatments was seen on germ tube entry of stomata and SSV formation. Further work, beyond the scope of the current project, will be required to determine why the effect of high light quanta seen in seedlings was not observed with adult plants.

DISCUSSIONA unique effect of light quanta on the receptivity of wheat seedlings to infection by the causal agent of yellow rust has been characterised. Both urediniospore germination, and the ability of the pathogen to establish a successful infection site are increased on seedlings having received high quanta of light prior to pathogen inoculation. The effect of light is not due to physiological dysfunction, but appears to bring about a physiological change that increases the levels of a plant factor used by the fungus to locate and enter stomata. In wheat seedlings the data indicate that genetic variation influences this effect of light and genotypes that do not show the effect may express a form of resistance that reduces Pst infection efficiency. Further, the work has demonstrated that by manipulating pre-inoculation light conditions, it is possible to increase infection efficiency thus increasing the numbers of germlings that attempt to establish biotrophic endophytic infections. This will greatly facilitate future studies aimed at identifying resistance mechanisms acting formation of SSVs to limit colony establishment and growth. Thus, the work undertaken here and in Proposal Objective 5 has set the foundations to identify new and valuable sources of durable resistance to yellow rust in wheat genotypes that have maintained efficacy over many years and that continue, therefore, to provide a resource for UK breeders.

SID 5 (Rev. 3/06) Page 22 of 25

REFERENCES1. De Vallavieille-Pope C, et al. 2002. Phytopathology 92, 1308-1314.2. Boyd LA, Minchin PN. 2001. Euphytica 122, 361-368.3.Melichar JPE. 2007. Genetic and physiological analysis of mutations in wheat showing enhanced adult plant resistance to yellow rust. PhD thesis, Faculty of Biological Sciences, University of East Anglia.

-------------------------------------------------------

8.8 OVERVIEW

The reported work aimed to facilitate development of strategies for sustainable disease management using durable resistance to key biotrophic cereal pathogens, thus allowing reduced dependence on fungicides. The oat (8.1) and wheat (8.6) germplasm generated and characterised for powdery mildew and yellow rust resistance provides a resource not only for immediate use by plant breeders but also for future studies aimed at identifying the mechanistic basis of resistances and their cell biological and molecular genetic basis. Such fundamental studies will, in turn feed back to further facilitate plant breeding for resistance that is likely to be under complex physiological and genetic control, and therefore to offer durability. Clearly, studies of models such as the oat or barley (8.2; 8.3; 8.5) and grass (8.4) powdery mildew systems and the wheat yellow rust (8.6; 8.7) and barley and wheat brown rust (8.5) systems have direct relevance to efforts aimed at controlling other biotrophic pathogens not only of the Gramineae but also of dicotyledonous plants (NB collaborations with Riederer et al. [8.4] and publications arising with Rubiales et al.). Further, the work has generated novel methodologies that are applicable generally to the study of cell-specific transcriptomics in plant biology (NB collaborations with Lyngkær et al. [8.4 and associated publications]). The breadth of the studies undertaken here was made possible only by the award of the Defra Fellowship which not only funded directly the association of expertise between Carver and Boyd but also led to close interaction with Pavely (ADAS) and important components of a research proposal from Pavely currently under consideration by Defra. Importantly, the Fellowship also allowed continued collaboration between Carver and many world leaders of plant biological research in the UK, throughout Europe and in the USA, Canada and Japan.

SID 5 (Rev. 3/06) Page 23 of 25

References to published material9. This section should be used to record links (hypertext links where possible) or references to other

published material generated by, or relating to this project.Olesen KL, Carver TLW, Lyngkjaer MF. (2003). Fungal suppression of resistance against inappropriate B. graminis formae speciales in barley, oat and wheat. Physiol. Mol. Plant Pathol. 62, 37-50. Fujita K, Wright AJ, Meguro A, Kunoh H, Carver TLW. (2004). Rapid pre-germination responses of Erysiphe pisi conidia to contact and light. J. Gen. Plant Pathol. 70, 75-84. Gjetting T, Carver TLW, Skøt L, Lyngkjær MF. (2004). Gene expression profiling of individual barley epidermal cells attacked by powdery mildew. Mol. Plant-Micr. Inter. 17, 729-738.Fujita A, Suzuki T, Kunoh H, Carver TLW, Thomas BJ, Gurr SJ , Shiraishi T. (2004). Induced inaccessibility in barley cells exposed to extracellular material released by non-pathogenic powdery mildew conidia. Physiol. Mol. Plant Pathol. 64, 169-178.Kunoh H, Carver TLW, Thomas BJ, Fujita F, Meguro A, Wright AJ. (2004). The Extracellular Matrix of Conidia of Powdery Mildew Fungi: Its Functions and Involvement in Information Exchange with Host Cells. In: Genomic and Genetic Analysis of Plant Parasitism and Defense. Eds: Shinji Tsuyumu, Jan E. Leach, Tomonori Shiraishi and Thomas Wolpert. APS press, MN, USA. pp. 150-163. Prats E, Mur LAJ, Sanderson R, Carver TLW. (2005). Nitric oxide contributes both to papilla-based resistance and the hypersensitive response in barley attacked by Blumeria graminis f. sp. hordei. Mol. Plant Pathol. 6,65-78.Gniwotta F, Vogg G, Carver TLW, Riederer M, Jetter R. (2005). What do microbes encounter at the plant surface? Chemical composition of Pisum sativum leaf cuticular waxes. Plant Physiol. 139, 519-530.Boyd LA. (2005) Centenary Review: Can Robigus defeat an old enemy? – Yellow rust of wheat. J. Agric. Sci. 143, 233-243.Zhang Z, Henderson C, Perfect E, Carver TLW, Thomas BJ, Skamnioti P, Gurr SJ. (2005). Of genes and genomes, needles and haystacks: Blumeria graminis and functionality. Mol. Plant Pathol. 6, 561-575.Fondevilla S, Carver TLW, Moreno MT, Rubiales D. (2006). Macroscopic and histological characterisation of genes er1 and er2 for powdery mildew resistance in pea. Eur. J. Plant Pathol. 115, 309-321.Prats E, Carver TLW, Lyngkjær MF, Roberts PC, Zeyen RJ. (2006). Induced inaccessibility and accessibility in the oat powdery mildew system: insights gained from use of metabolic inhibitors and silicon nutrition. Mol. Plant Pathol. 7, 47-59. Mur LAJ, Carver TLW, Prats E. (2006) NO way to live; the various roles of nitric oxide in plant-pathogen interactions. J. Exp. Bot. 57, 489-505Prats E, Gay AP, Mur LAJ, Thomas BJ, Carver TLW. (2006). Stomatal lock-open, a consequence of epidermal cell death, follows transient suppression of stomatal opening in barley attacked by Blumeria graminis. J. Exp. Bot. 57, 2211-2226.Carver TLW, Gurr SJ (2006). Filamentous fungi on plant surfaces. In: Biology of the Plant Cuticle. Ed. Markus Riederer. Annual Plant Reviews 23, Blackwell, Oxford. pp. 368-397.Fondevilla S, Carver TLW, Moreno MT, Rubiales D. (Published Online, Dec 21 2006). Identification and characterisation of sources of resistance to Erysiphe pisi Syd. in Pisum spp. Plant Breeding.Gjetting T, Hagedorn PH, Schweizer P, Thordal-Christensen H, Carver TLW, Lyngkjær MF. (In press). Single-Cell Transcript Profiling of Barley Attacked by the Powdery Mildew Fungus. Mol. Plant Micr. Inter.. Prats E, Carver TLW, Fondevilla S, Rubiales D. (2006). Cellular bases of resistance to different formae speciales of Blumeria graminis in Hordeum chilense, wheat, and the amphiploids tritordeum and agroticum Can. J. Plant Pathol. 28, 577-587.Boyd LA. (2006). Perspective: Can the durability of resistance be predicted? J. Sci. Food Agric. 86,2523-2526.Olesen KL, Carver TLW, Lyngkjær MF. (submitted). Brown rust attack affects accessibility of barley epidermal cells to powdery mildew. Mol. Plant Pathol.Prats E, Gay AP, Roberts PC, Thomas BJ, Sanderson R, Paveley ND, Lyngkjær MF, Carver TLW, Mur LAJ. (in prep.). Temporal and spatial relationships between stomatal dysfunction and cell death due to different single gene resistances and non-host resistance in barley attacked by Blumeria graminis. J. Exp. Bot.

SID 5 (Rev. 3/06) Page 24 of 25

SID 5 (Rev. 3/06) Page 25 of 25