2.2.3S
EXPERIMENTAL APPROACHES TO UNDERSTANDING THE CONTRIBUTIONS OF MATING SYSTEM, GENE FLOW AND SELECTION TO THE GENETIC STRUCTURE OF POPULATIONS OF PLANT PATHOGENIC FUNGI
BA MCDONALD1, CC MUNDT2 and J ZHAN2

1Department of Plant Pathology and Microbiology, Texas A&M University, College Station, Texas, 77843-2132 USA; 2Department of Botany and Plant Pathology, Oregon State University, Corvallis, OR 97331-2902, USA

Background and objectives
Hierarchical sampling combined with DNA RFLP markers has substantially increased our knowledge of the amount and distribution of genetic variation within and among populations of Mycosphaerella graminicola and Phaeosphaeria nodorum over the past 10 ;years [1]. While knowledge of genetic structure may offer significant insight into the evolutionary processes that affect the population genetics of pathogens, experimental approaches ultimately are needed to test hypotheses and validate evolutionary concepts that emerge from population surveys. Our objective was to utilize replicated field experiments to measure the relative impacts of mating system, immigration and selection on the genetic structure of a population of the fungus M. ;graminicola which causes septoria tritici leaf blotch on wheat.

Materials and methods
Wheat varieties Stephens and Madsen were grown alone and in 50:50 mixture in an RCBD with four replications. Three replications were inoculated artificially at the seedling stage with an equiproportional mixture of 10 isolates of M. ;graminicola. Five of the isolates (M1-M5) originated from Madsen and five isolates (S1-S5) originated from Stephens in a previous year. The fourth replication was inoculated naturally by local ascospores. The inoculated Stephens plots were sampled three times (early, mid, late) during the growing season, the Madsen and mixture plots were sampled in early and late season. A total of 100 infected leaves were collected from each plot for each sample and one fungal isolation was made from each leaf. Each isolate (1410 total) was assayed for nine individual RFLP loci and one DNA fingerprint. Measures of selection were based on changes in frequencies of the 10 inoculated genotypes over time. Proportions of novel genotypes in inoculated plots due to recombination and immigration were estimated by calculating exact probabilities for each haplotype based on allele frequencies in the control plots and inoculated plots. Putative recombinants identified via probability analysis were confirmed with DNA fingerprints.

Results and conclusions
Significant changes occurred in the frequencies of the 10 inoculated isolates. Selection coefficients ranged from 0.03 to 0.32 per isolate. Three isolates showed evidence of host specialization. The frequency of novel isolates increased from 0.03 to 0.34 over the course of the season. Among the novel isolates detected in the inoculated plots at the end of the season, we estimate that 43% were immigrants and 57% resulted from recombination among the 10 inoculated isolates. In the late season collection, 64% of the isolates were asexual progeny of the 10 inoculated isolates, 19% were sexual progeny of the 10 inoculated isolates, and 15% were immigrants. The proportion of the population due to recombinants increased over the growing season. These results suggest that sexual recombinants originating from within the crop canopy represent a significant source of inoculum during the growing season.

References
1. McDonald BA, Zhan J, Yarden 0, Hogan K, Garton J, Pettway RE, 1998. Proceedings 15th Long Ashton International Symposium: Understanding Pathosystems: A Focus on Septoria.