HETEROGENOUS ORIGIN OF A NEW PATHOTYPE OF MELAMPSORA EPITEA ON SALIX x MOLLISSIMA 'Q83', REVEALED BY AMPLIFIED FRAGMENT LENGTH POLYMORPHISM (AFLP) MH PEI, Z W YUAN, C RUIZ, T HUNTER and DJ ROYLE IACR-Long Ashton Research Station, University of Bristol, Long Ashton, Bristol BSI 8 9AF, UK Background and ObjecUves The emergence of new virulent types of plant pathogens, especially in highly specialised biotrophs such as rust fungi, can drastically break down resistance in crop plants and seriously affect their productivity. A pathogen population new to an area may have evolved locally or become established as a result of introduction by means of gene flow. To date, limited information exists on the origins of new, virulent types of plant pathogens. Willow plantings at Long Ashton, SW England, Markington, N Yorkshire, and Loughgall, N Ireland, were among the first biomass plantations to be grown for renewable energy production in the UK. A productive willow clone,Salix x mollissima (triandra x viminalis) 'Q83', which has been grown since the early 1 980s, was shown to be highly resistant to rust (Melampsora epitea)until 1992. Disease surveys, undertaken during 1987-91, confirmed that '083' was one of the most resistant clones at all sites investigated in W Europe and Canada. Severe infection on 'Q83' was detected, for the first time, simultaneously in the UK at Long Ashton, Markington and Loughgall in the autumn of 1992. Since then, infections on '083' have occurred at all three sites each year. The aim of this study was to use the AFLP technique to determine whether the rust first encountered on 'Q83' at the UK sites had originated from a common source or evolved separately. Materials and Methods UredinIospore samples from 'Q83' were collected at Long Ashton, Markington and Loughgall in Sept~ 1992. Seven willow clones, S. burjatica 'Korso' and 'Germany', S. disperma, S. x mollissima 'Q83', S. x stipularis, S. viminalis 'Mullatin' and 'Bowles Hybrid' were used as differentials in pathogenicity tests [1]. Young, fully expanded willow leaves were placed on tap-water-soaked filter paper in Petri dishes and inoculated with the rust collections. Rust isolates, 22 from Long Ashton and 22 from Markington were made from single pustules which developed on 'Q83' in pathogen icity tests. For each isolate, approximately 10 mg spores were used for DNA extraction. Genomic DNA was digested using restriction endonucleases Eco RI and Mse I. The two-step PCR procedure of Vos et al. [2] was adopted for selective amplification. EcoRI and Msel primers having 1-bp, 3-extension were used in the first step (preamplification) and the primers having 3-bp, 3'-extensions in the second. Both amplications were penformed in a Perkin Elmer Cetus Thermocycler 480. Results and conclusions All the rust collections were of the same pathotype since they showed the same pathogenicity patterns in inoculation tests using willow differentials. When isolates from Long Ashton and Markington were analysed using AFLP fingerprinting, those collected from within the same site showed less than 7% variation. However, the DNA profiles were distinct between the two different sites, with differences of more than 30%. The results conclusively indicate that the new pathotype on '083' was not spread from a common source, but evolved independently. The results suggest that local evolution of willow rust may play a more important role than previously thought in producing new, virulent forms to overcome host resistance. The AFLP has shown to be a powerful method to detect DNA polymorphisms and thus to analyse field populations of rust pathogens. References 1. Pei MH, Royle DJ, HunterT, 1996. Plant Pathology 45, 679-690. 2. Vos P, Hogers R, Bleeker M, Reijans M, van de Lee T, Homes M, Frijters A, Pot J, Peleman J, Kuiper M, Zabeau M, 1995. Nucleic Acid Research 23, 44074414.