lSchool of Biological Sciences, The University of Birmingham, Birmingham B15 2TT, UK; 2AgrEvo UK Ltd, Chesterford Park, Saffron Walden CBI 0 1 XL, UK

Background and objectives
With increasing knowledge of the physiology and biochemistry of phytopathogenic fungi an expanding number of potential targets for biochemical fungicide design can be identified. To prevent wasted resources it is important to test at an early stage whether each putative target is in fact essential for infection and disease development, i.e. to validate the target. We are using Stagonospora (Septoria) nodorum to explore the use of gene disruption as a general method of fungicide target validation [1]. Gene disruption has not been demonstrated previously in this species.

Materials and methods
Nitrate reductase was chosen as a model target for these studies as the gene (NIA1) has been cloned from S.nodorum and disruptants should have a readily distinguishable phenotype being nitrate non-utilising and chlorate resistant [2]. Three approaches to targeted disruption were attempted: (a) cotransformation, (b) integrative gene disruption, (c) one step gene replacement. With each approach, hygromycin resistance was used as the selectable marker for transformation.

Results and conclusions
We succeeded in disrupting the NIAI gene by both cotransformation and gene replacement. Around 2% of hygromycin resistant transformants from the cotransformation experiments had the expected NIAI mutant phenotype and disruption was confirmed by Southern hybridisation. Half of the transformants with the gene replacement vector were unable to grow on medium containing nitrate as sole nitrogen source, suggesting a high level of targeted integration in this fungus. However, Southern analysis of these nitrate non-utilising products indicated that only around 50% possessed the expected gene replacement. Further tests indicated that these gene replacement transformants possessed the predicted associated phenotypic changes of loss of nitrate reductase activity and acquisition of chlorate resistance. The other 50% retained an intact NIAI gene but had integrated multiple copies of the vector elsewhere in their genome. The presence of a functional NIAI gene in the latter group was confirmed by enzyme assays and their chlorate sensitivity. How this phenotype arose is not clear; it might involve cosuppression or a complex homologous integration event as the vector did carry 1.1 kb of the nitrite reductase gene which is adjacent to NIAI. Mutants of both types retained full pathogenicity in detached leaf assays thereby invalidating both nitrate and nitrite reductases as fungicide targets.

1. Caten CE, Hollomon DW, 1995. Antifungal Agents, GK Dixon et al ed., pp31-47.
2. Newton AC, Caten CE, 1988. Trans.Br.mycol.Soc. 90, 199-207.