1USDA-ARS, Department of Botany and Plant Pathology, 1155 Lilly Hall, Purdue University, West Lafayette, IN 47907-1155, USA; 2IPO-DLO, PO Box 9060, 6700 GW Wageningen, Netherlands

Background and objectives
Septoria tritici blotch, caused by Mycosphaerella graminicola, is an important disease of wheat worldwide. Recently, populations of M. ;graminicola were analysed for genetic variation using RFLP markers [1]. These studies showed that most populations of M. ;graminicola world-wide contained high levels of genetic diversity, and that sexual reproduction probably occurs commonly. Although these studies provided an excellent initial picture of global genetic diversity, many important gaps remain. The major wheat-growing regions of the central USA have not been sampled extensively and it is not known whether results from other parts of the world are representative of the central United States. The different market classes of wheat in the central United States (hard red spring, hard red winter, soft red winter, durum wheat) might select for different pathogen populations, and differences in the growing seasons might affect the frequency of sexual recombination.

Addressing these questions will require more extensive samples and faster markers. The RFLP technology used in previous studies provides excellent resolution, but is slow and labour intensive compared with other markers. The goal of this research was to develop additional markers and perform a preliminary analysis of genetic diversity within and among populations of M. ;graminicola in the central USA.

Materials and methods
Isolates of M. ;graminicola were obtained from infected leaves of hard red spring and soft red winter wheat kindly provided by D. Long (USDA-ARS, Cereal Rust Laboratory). RAPD analysis was performed according to standard protocols. Genetic analyses of RAPD bands were done using progeny of a cross between two Dutch isolates of M. ;graminicola [2]. Genetic data were analysed using MAPMAKER and HAPMAP. Specific polymorphic RAPD bands were cloned using a TA Cloning Kit and sequenced by automatic sequencer. DNA sequences were aligned using DNASIS and primers designed using OLIGO.

Results and conclusions

Analyses of 59 isolates of M. ;graminicola from Minnesota, North Dakota, Indiana and Ohio with 20 RAPD primers revealed that almost every isolate had a unique genotype. Thus, sexual reproduction probably occurs commonly in these populations. The only exceptions were isolates from different lesions on the same leaf, which usually had identical genotypes. This pattern is consistent with epidemics initiated by ascospores, with subsequent spread on the same leaf by asexual pycnidiospores.

Genetic analyses of more than 100 putative RAPD loci on 99 progeny isolates revealed that most behaved as simple Mendelian markers. The loci associated into several loose linkage groups that will be integrated into a more complete genetic map of M. ;graminicola. Most of the loci segregated for the presence or absence of a band. Knowledge of the genetic basis for each RAPD phenotype simplified scoring and ensured that only reliable bands were retained.

Four loci segregated as if they had 'codominant' alleles, i.e., each isolate possessed exactly one band within a certain size range. Putative alternative alleles at each locus were cloned and sequenced. In each case tested, the different bands were identical except for an insertion or deletion of 20-60 ;bp. Thus, they were alternative alleles at single genetic loci. These loci were converted to sequence characterized amplified regions (SCARs) by designing specific primer pairs that amplified the variable regions. The longer primer lengths and increased specificity of SCARs eliminate the problems with RAPDs. Scoring is more reliable because the alleles differ in size rather than varying in plus or minus. These markers are now available for analysing genetic variability within populations of M. ;graminicola.

1. McDonald BA, Zhan J, Yarden O, Hogan K, Garton J, Pettway RE, 1998. Proceedings of the Long Ashton Septoria Conference, in press.

2. Kema GHJ, Verstappen ECP, Todorova M, Waalwijk C, 1996. Current Genetics 30, 251-258.