Sainsbury Laboratory, John Innes Centre, Norwich NR4 7UH, UK;
1Present address, Monsanto Services International SA, Rue Laid Burniat 5, European Crop Research Centre, B-1348, Louvain-la-Neuve, Belgium

Background and objectives
Disease resistance genes (R genes) have been used by plant breeders since the start of the 20th century to control plant diseases. Although of agricultural benefit, natural R genes have three intrinsic disadvantages: most exhibit race specificity thereby conferring resistance to only a few isolates of a single pathogenic species; resistance is frequently not durable because genetic shifts in the pathogen population cause the loss of the corresponding avirulence (Avr) gene and, finally, due to interspecies incompatibility only a finite source of resistant germplasm is available to breeders. Collectively this results in the highly undesirable, so-called 'Boom and Bust' cycle of disease control. In the past 5 years, several R genes conferring resistance to distinct fungal, viral, bacterial and nematode pathogens have been isolated [1, 2]. The predicted R proteins have several features in common, which suggests that plants have evolved similar mechanisms to combat different disease-causing pathogens. In addition it has been demonstrated that R genes provide resistance when transferred into related plant species. Therefore it should be possible, by manipulating R genes, to harness natural plant disease-resistance mechanisms more effectively and thereby achieve both broad-spectrum and durable disease control. In this introductory keynote paper, some of the possibilities and potential pitfalls of utilising cloned R genes in novel approaches of disease control will be illustrated. Demonstrated pitfalls arising from wild-type and engineered R genes expressed in heterologous species include gross alterations to plant development or necrotic lesion development in the absence of pathogen attack. One successful method of disease control developed at The Sainsbury Laboratory and termed genetically engineered acquired resistance (GEAR) will be described in detail.

Results and conclusions
Tomato plants that inherit a maize Dissociation, Ds transposon tagged Cf-9 fungal resistance gene, Cf-9 Ds, a 35S:Avr9 transgene and a stabilized Activator, Ac transposase gene sAc, show somatic excision of Ds from the Cf-9 Ds gene which somatically restores Cf-9 function. This permits recognition of the in planta-produced Avr9 peptide and gives rise to the localized activation of plant defence responses. Ultimately, these activated host cells die and give rise to small necrotic sectors. Phenotypically the plants show a variegation for a defence-related necrosis in the same manner that plants challenged with necrotizing pathogens develop somatic flecks of necrosis that are associated with the induction of systemic acquired resistance (SAR) [2]. The plants variegating for necrosis were found to be more resistant to three fungal pathogens tested than sibling non-variegated progeny. The three fungal species controlled were Cladosporium fulvum (leaf mould), Oidium lycopersici (powdery mildew) and Phytophthora infestans (late blight). Interestingly, the effectiveness of GEAR in suppressing plant disease appeared to be inversely related to sector size. This observation raised the possibility that by carefully manipulating the frequency of somatic restoration of Cf-9 function, even higher levels of plant protection could be developed. We are currently testing different single T-DNA GEAR constructs to try to control various diseases on the roots and leaves of potato plants.

The work at the Sainsbury Laboratory is supported by The Gatsby Charitable Foundation. MAT is funded by an EU Training Fellowship.

1. Hammond-Kosack KE, Jones JDG, 1997. Annual Review of Plant Physiology and Plant Molecular Biology 48, 575-607.
2. Anon. (multiple authors) 1996. Plant Cell, October Special Review Issue.