1Sainsbury Laboratory, John lnnes Centre, Colney Lane, Norwich NR4 7UH, UK; 2Monsanto European Crop Research Centre, Parc Scientific Fleming, Rue Laid Burniat 5, B-1348 Louvain-la-Neuve (Sud), Belgium

Background and objectives
The objective of our research is to understand the signalling pathways which activate the plant defence response. We are studying the interaction between Lycopersicon esculentum and Cladosporium fulvum, causal agent of leaf mould disease. A number of semi-dominant Cf genes confer resistance to C. fulvum races expressing the corresponding avirulence (Avr) genes. The Cf genes encode proteins with large leucine-rich repeat regions, thought to interact with the products of the Avr genes, small cysteine-rich peptides which can be retrieved in intercellular washing fluid (IF) from infected leaves. A rapid response in the incompatible interaction is the production of reactive oxygen species (ROS) such as superoxide radicals and hydrogen peroxide [1]. The Cf-9 gene has been shown to retain its activity and specificity in tobacco plants and cell cultures. An Avr9-dependent oxidative burst has been characterized in cell cultures which can be inhibited by diphenyleniodonium (DPI), an inhibitor of the mammalian phagocyte NADPH oxidase. We are using these cell cultures as a system in which to study gene expression during the defence response and its dependence upon the oxidative burst.

Materials and methods
Tobacco cell cultures were subjected to three treatments: IF-Avr9, IF+ vr9, and IF+Avr9 pre-treated with DPI, and harvested after 30 min. In order to identify genes whose expression patterns change, the cDNA-AFLP technique was used [2]. This involves the selective PCR amplification of restriction fragments from a total digest of cDNA and is based on the AFLP fingerprinting technique. Once differentially expressed fragments were identified they were sequenced and BLAST database searches were carried out in an attempt to identify the function of these genes.

Results and conclusions
AFLP display of cDNA fragments revealed 292 which are differentially expressed. Less than 10% of these changes in gene expression were inhibited by the addition of DPI, suggesting that most changes are not dependent on the presence of ROS. Some induced fragments showed homology to genes in the sequence databases, including components of signalling pathways. A subset of induced fragments was used to screen a cDNA library, constructed from cell cultures harvested 30 min after Avr9 treatment, in order to gain full-length clones for use in further studies. We conclude that there are many changes in gene expression after elicitation with Avr9. Although the oxidative burst may play a role in mediating some changes in gene expression, the majority of changes in our system do not require the presence of ROS. The alternative model is that an independent signal cascade from Cf-9 controls most gene expression. Components of this cascade may include ion channels, protein kinases and transcription factors.

This PhD studentship is funded by the Gatsby Charitable Foundation.

1. May MJ, Hammond-Kosack KE, Jones JDG, 1996. Plant Physiology 110, 1367-1379.
2. Bachem CWB, van der Hoeven RS, de Bruijn SM et al., 1996. Plant Journal 9, 745-753.