2.2.7S
bTHE FUTURE IS IN THE PAST: THE IMPACT OF EVOLUTIONARY HISTORY ON CURRENT POPULATION DYNAMICS IN THE USTILAGO MAYDIS-ZEA MAYS PATHOSYSTEM

G MAY1,3, J GARTON1,3 and JV GROTH2,3

1Department of Plant Biology, 2Department of Plant Pathology, and 3Center for Community Genetics, University of Minnesota, St Paul, MN 55108, USA

Background and objectives
In this study, we ask what is the impact of modern agricultural practices on the evolutionary interactions of a pathogen, Ustilago maydis, with its host plant, Zea mays? Our specific objectives are: (1) to provide estimates of gene flow in smut populations which allow distinction of historical and current gene flow; (2) to determine the history of selection in U. maydis populations using the b mating type locus. Many of our inferences regarding population dynamics of plant pathogens make the implicit assumption that current, ongoing processes, such as gene flow, selection and drift, are responsible for the observed patterns of genetic differentiation. In this study, we are testing the hypothesis that smut populations followed maize through domestication from teosinte and that, as a consequence, most of the important incidences of gene flow and selection in this pathogen occurred historically.

Materials and methods
Polymorphic RFLP markers were chosen from partial genomic libraries and blots with total genomic DNA probed to obtain allele frequency and geographic distribution data. Genealogies of the b mating type locus were generated from sequence information obtained from the hypervariable region of the b mating type locus. Collections and genetic analyses of U. maydis from geographic locations were made by obtaining galls, isolating haploid isolates and obtaining DNA as described [2].

Results and conclusions
The results of our genealogical analyses of the b mating type locus combined with allele frequencies at RFLP markers demonstrate that gene flow is not high between widely separated geographic distances and suggest that much gene flow occurred as smut followed maize through domestication and host range expansion. Selection on new b types occurred largely in the past when population sizes of host and pathogen were smaller. From our results to date, we suggest that analyses of genetic differentiation in current pathogen populations using allele frequency data is largely measuring the impact or "echo" of historical processes. Further, our information to date suggests that genetic variation in this pathogen is largely maintained by neutral processes and that metapopulation models [1] better fit observations of population dynamics than do classical migration/selection models. Further analyses are planned to include sampling different host populations and estimates of inbreeding.

References
1. Thrall PH, Burden JJ, 1997. Journal of Ecology 85, 743-753.
2. Zambino P, Groth JV, Lukens L, Garton JR, May G, 1997. Phytopathology 87, 1233-1239.