Powdery mildew virulence survey data collated from 10-11 European countries in 1996 to 1998 showed that many commonly deployed specific resistance genes (notably Pm1, Pm2, Pm3c, Pm3f, Pm4b, pm5, Pm6, Pm7, Pm8, Pm2,6) in wheat cultivars were matched by high corresponding virulence frequencies in the mildew populations in most regions. However, some resistance genes (e.g. Pm3a, Pm3b, Pm4a, Pm17) had low corresponding virulence frequencies in some countries in some years. Following a multi-location test, a Core Differential Set of wheat cultivars/lines was established for use in collaborative powdery mildew survey work.
Introduction
Powdery mildew (Blumeria graminis (DC.) E. O. Speer f sp tritici Em. Marchal) is a major disease problem in wheat crops in Europe, particularly in northern and western areas. Many countries conduct surveys to ascertain the virulence composition of the mildew population but there has been little standardisation in methodology and use of differential cultivars.
Under the auspices of COST Action 817 - Population studies of airborne pathogens of cereals as a means of improving strategies for disease control - attempts were made in the period 1995-1999 to:
1) collate mildew virulence survey data from different countries to provide a pan-European picture;
2) standardise virulence survey methodology and define a core differential set of test cultivars.
This paper summarises the main results from the work of the Wheat Mildew Sub-group of Working Group 1 - Surveys on virulence and aggressiveness - of COST 817.
Materials and Methods
Virulence surveys
Methods employed by different laboratories varied considerably and little standardisation was possible. Mildew isolates were variously obtained by collection of infected leaves from crops and trials or from exposed trap plants in mobile or static nurseries. The single colony isolates obtained were subsequently tested on detached leaves or seedlings of sets of differential wheat cultivars/lines with known specific resistance genes, again varying according to laboratory, to ascertain the corresponding virulence genes present. Infection type was assessed using the method of Moseman et al, 1965. Numbers of isolates tested by each laboratory each year varied between 17 and 618; the virulence frequencies corresponding to specific resistance genes and gene combinations were calculated. In Norway, bulk isolates were tested and the mean percentage disease severity was recorded instead of virulence frequency.
Data were collated from as many countries as possible in the years 1995 to 1998 and the data were studied for possible trends across Europe. Table 1, 2 and 3 show virulence frequencies corresponding to the most commonly used resistance genes throughout the countries in the years 1996-1998; 1995 data were incomplete and are not shown here. Full data, including virulence frequencies corresponding to many less common resistance genes deployed in individual countries, have been published in COST 817 Annual Reports (Anon. 1997, 2000).
Core differential set
Six laboratories participated in a multi-location test of differential cultivars in 1997. Seed of each laboratory's own differential cultivars for the most common mildew resistance genes was exchanged with all other laboratories and used to produce young wheat seedlings. Detached leaves of the latter were then inoculated with 10-20 representative mildew isolates at each laboratory and infection assessed on the standard 0-4 scale (Moseman et al, 1965). Results were collated at NIAB Cambridge, UK: the most consistent cultivars, in terms of infection response, for each resistance gene were selected for a Core Differential Set
Results
Virulence surveys
The collated results from 11 countries in 1996 are shown in Table 1, from 10 countries in 1997 (Table 2) and 11 countries in 1998 (Table 3). Results for 1995 were less extensive and were published previously (Anon. 1997).
Examination of all virulence survey data from 1996-1998 has clearly shown certain trends. A number of common specific resistance genes - Pm1, Pm2, Pm3c, Pm3f, Pm4b, pm5, Pm6, Pm7, Pm8, Pm2,6 - had high corresponding virulence frequencies in the mildew populations in most countries where they were deployed. Therefore, these resistance genes were no longer providing adequate disease control, despite being initially effective. Other common virulence genes were at high frequencies in most regions, but were at sufficiently low frequencies in some areas in some years that the corresponding resistance genes may have offered mildew control, e.g. Pm3a in France (1998), Denmark (1996) and Switzerland (1997, 1998); Pm3b in France (1998), Switzerland (1997, 1998) and the UK (all years); Pm3d in Denmark (1996), France (1998) and Germany (all years) and Pm17 in France (1998), Switzerland (1996, 1997) and the UK (1998). In eastern European countries, some differences from western areas were found: e.g. Pm4a was at low frequencies in Hungary and Slovakia and Pm1,2,9 at low frequencies in Slovakia in all three survey years.
Further analysis of data showed that mildew isolates are composed of complex pathotypes: e.g. the UK population is dominated by one pathotype carrying 6 virulence genes (Clarkson & Slater, 1999). Ideally, virulence data should be accompanied by information on areas sown with corresponding mildew resistance genes, as this largely determines the local mildew population, but these data were not generally available.
Core differential set
Detailed analysis of the data was carried out and the consistency of mildew reaction on all the differential cultivars from each laboratory was compared. This culminated in the compilation of a proposed Core Differential Set of wheat mildew cultivars (Table 4), representing the most common resistance genes deployed. It was suggested that this set should be used for collaborative work on wheat mildew virulence surveying and gradually introduced at as many laboratories as possible, although there would still be a need for individual laboratory's differential sets for some studies. The differential set will be multiplied at Nickersons Seeds in the UK (the author can be contacted for information).
Discussion
Collated virulence survey data for 1995-1998 clearly showed that many specific resistance genes had high corresponding virulence frequencies in the mildew populations in most European countries surveyed, particularly for genes Pm1, Pm2, Pm3c, Pm3f, Pm4b, pm5, Pm6, Pm7, Pm8 and Pm2,6. Other resistance genes and gene combinations had high corresponding virulence in some regions but low in others, where they may have provided effective disease control in certain years. This would be largely dependent on the extent and length of time of deployment of the resistance genes in wheat cultivars within each individual country. This also illustrates the need for virulence survey data to be accompanied by information on areas of wheat cultivars grown. Therefore, it is apparent that specific resistance genes may only provide relatively short-term control of wheat mildew, necessitating heavy reliance on routine fungicide use, particularly in northern and western Europe (JDS Clarkson, unpublished data). However, other work in COST action 817 has shown that partial resistance is present in many wheat varieties and is providing effective mildew control (K Flath, M Jalli, personal communications).
Standardisation of virulence survey methodology is essential for meaningful interpretation and comparison of data. Accordingly, a Core Differential Set of wheat cultivars/lines was compiled for current common resistance genes/gene combinations. This Set is being multiplied and maintained in the UK and should be used for collaborative survey work and also introduced by new workers in the field. The Set will need to be reviewed regularly and new differential cultivars added as new resistance genes are introduced into wheat cultivars.
Control of powdery mildew in European winter and spring wheat varieties thus demands an integrated approach. As specific resistance may be unreliable in the medium and long term, use of partial resistance and cultural control methods need to be optimised to reduce the great reliance on routine fungicide usage.
Acknowledgements
The author would like to acknowledge the leadership and guidance of Dr Hanne Ostergaard, Riso National Laboratory, Roskilde, Denmark, throughout the duration of COST Action 817. I would also like to thank colleagues in the following countries, who have contributed data from virulence surveys and others who have participated in the work of the Wheat Mildew Sub-group of COST 817:-
Bulgaria: I. Iliev.
Czech Republic: P. Bartos; R. Hanusova; K. Klem.
Denmark: B. Boesen; M. Hovmoller.
Finland: M. Jalli.
France: G. Doussinault; F. Godet.
Germany: A. Andersch; F. Felsenstein; K. Flath; U. Sperling.
Hungary: L. Szunics; G. Vida.
Netherlands: A. Engels.
Norway: H. Skinnes; Y. Tarkegne.
Poland: A. Strzembicka.
Slovakia: J. Huszar.
Switzerland: G. Schachermayr; M. Winzeler.
UK: R. Bayles; P. Fenwick; P. Oldridge; S. Slater.
Yugoslavia: R. Jevtic.
References
Anon., 1997. EUR 18015 - COST 817 - Population studies of airborne pathogens of cereals as a means of improving strategies for disease control - Annual Report 1996, pp 51-56.
Anon., 2000. EUR 19222 - COST 817 - Population studies of airborne pathogens of cereals as a means of improving strategies for disease control - Annual Report 1998, pp 37-43.
Clarkson JDS, Slater SE, 2000. Mildew of wheat. UK Cereal Pathogen Virulence Survey 1999 Annual Report, pp 22-28.
Moseman JG, Macer RCF, Greely LW, 1965. Genetic studies with cultures of Erysiphe graminis f sp hordei virulent on Hordeum spontaneum. Transactions of the British Mycological Society 48, 479-489.