lZENECA Agrochemicals, Jealotts Hill Research Station, Bracknell, Berks RG42 6EY, UK; 2IACR-Long Ashton Research Station, University of Bristol, Bristol BS18 9AF, UK

Background and objectives
As discovery of new agricultural fungicides becomes an increasingly rational process it will become more likely that antifungal target proteins and particularly the genes that determine them will be well characterised prior to commercial launch. This provides us with the opportunity to apply recombinant DNA technologies to aid in the evaluation of the risk of resistance development and in the future monitoring and management of the initial anti-resistance strategies. Molecular diagnostic technologies have progressed rapidly in the medical field driven by such problems as HIV and cystic fibrosis and fuelled by the human genome project. Sequencing of the yeast genome provides a similar impetus to fungal studies.

Results and conclusions
Sequences of the gene involved in MBC resistance have been determined for several plant pathogenic fungi. More recently a mutation in the 14-demethylase gene of Uncinula necatorthat correlates with resistance to a sterol biosynthesis inhibitor has been shown to be due to a single substitution in the target protein [1]. The cytochrome b gene which codes for the target protein of the recently introduced strobilurin fungicides has been sequenced in many species. We have recently sequenced several plant pathogens. Key codons were highly conserved across species.
Techniques such as multiplex PCR, quantitative multiple competitive PCR(QMC-PCR) and the amplification refractory mutation system(ARMSTM) [2] provide tools to analyse polymorphic regions of individual atieles in complex DNA's. ARMS provides a way to selectively multiply mutant alieles in a background of wild type alieles, thus allowing detection of rare mutations in the DNA sample. We can probe for single base pair substitutions. Quantification techniques are improving and QMC-PCR has been used to determine DNA copy number of HIV-1 DNA in a single reaction tube. We are approaching the area where we may be able to accurately determine gene frequencies not only within populations but also where mitochondrial genes are involved (e.g. cytochrome b) within individual clones. Visualising gene frequencies and not just phenotypic frequencies should improve quantitative selection modeling and help us be more predictive about resistance risk. Improvements in molecular diagnostic assay processing in the form of high throughput screens and automated data capture through fluorescence chemistry should enable us to sample populations more thoroughly as part of ongoing monitoring efforts.
It is here that some of the initial challenges lie. We have to find better ways of predicting which mutant alieles will be important in the field and which will only flourish within the laboratory in order to ensure that recombinant diagnostics are comprehensive and relevant. The diagnostic assay can only be as good as the DNA sample it is processing. In this respect we must think carefully about the questions we wish to address and the most efficient means of sampling the population to tackle them.

1. Delye C, Laigret F, Corio-Costet M-F, 1997. Applied and Environmental Microbiology 63, 2966-2970.
2. Newton CR, Graham A, Heptinstall LE, Powell SJ, Summers C, Kaisheker N, Smith JC, Markham AF, 1989. Analysis of any point mutation in DNA. The amplification refractory mutation system (ARMS). Nucleic Acid Research, 17, 2503-2515.