Institute of Phytopathology, University of Kiel, Kiel, Germany

Background and objectives
Genetic structure refers to the amount and distribution of genetic variation within and between populations. Knowledge of the genetic structure of plant pathogen populations has direct applications to agricultural ecosystems. Unfortunately, many phytopathogenic fungi such asSeptoria tritici (teleomorph Mycosphaerella graminicola) do not possess clearly defined gene-for-gene interactions, that make them acceptable to genetic analysis using virulence markers. Several different types of markers have been developed. More recently, a new PCR-based technique has been developed [1] termed amplification fragment length polymorphism (AFLP).

Materials and methodsIn 1996 we collected 380 single spore isolates of S. tritici from five different locations and two different susceptible cultivars in Germany. The preamplification of the single spore isolate DNA was carried out with two primers having each a single selective nucleotide. The labelling reaction was performed with 33P-ATP. In the following amplification step we used two primers each having two selective nucleotides. After amplification each sample was loaded on a 4 % denaturing (sequencing) polyacrylamide gel.

The fragments from each technique were scored for presence or absence of bands showing the same mobility in the gel, regardless of their optical density. The fragment data were coded as a binary matrix where 1 designated presence and 0 absence of a particular band. Based on the binary data matrix, genetic similarities were calculated for all pair-wise isolate-comparisons. The genetic similarity between two isolates x and y was calculated according to the formula of DICE (1945), first applied to molecular data by Nei and Li (1979): Sxy=2nxy/(nx+ny), where nxy is the number of fragments in common between isolates x and y, and nx and ny are the total numbers of fragments in isolates x and y, respectively. Based on similarity values, associations among isolates were revealed by UPGMA cluster analysis specifying the DICE coefficient.

Results and conclusions
The present data showed a large genetic variation in the Septoria tritici populations. Isolates with the same genotype were found in the same pycnidium indicating that these could be clones. Furthermore there were often found more than one genotype within a single lesion, indicating that a lesion is often the result of coinfection of two or more genotypes. Different genotypes existed between the lesions on the same leaf and different locations in the field.

The populations showed similar levels of genetic diversity. The correlation between genetic distance and geographical distance depends largely on the dispersal ability of an organism. Organisms with the potential for widespread, long-distance dispersal will display greater genetic uniformity across local populations than organisms with very limited dispersal ability. The sexual state of S. tritici is now thought to be important for long-distance dispersal of the pathogen. The sexual cycle may also be an important source of genetic variation in S. tritici populations. Our data suggest that the 'geographical populations' we used are not evolving independently and therefore may be considered part of the same genetic population.

1. Vos P, R. Hogers, M Bleeker, M. Reijans, T Lee, M Hornes, A Frijters, J Pot J Peleman, M Kuiper and M Zabeau 1995. Nucleic Acids Research, 23 4407-4414.