ND GUNN, JL BEYNON and EB HOLUB

Horticulture Research International, Wellesbourne, Warwick CV35 9EF, UK

Background and objectives
The interaction between the biotrophic oomycete, Peronospora parasitica and Arabidopsis thaliana is characterized by marked isolate/accession specificity. More than 20 RPP loci (recognition of Peronospora parasitica) have been identified in A. thaliana. The research reported here was undertaken to discriminate the number of specificities at a particular host locus and confirm correspondence of a single avirulence gene with each RPP gene.

Procedures for genetic analysis of avirulence in homothallic isolates of P. parasitica in A. thaliana have to be established. Outcrossing has been demonstrated for facultative oomycete pathogens using molecular markers to identify polymorphic loci that allow discrimination of outcrossed progeny from selfs of the two parental isolates [1]. Such markers were a prerequisite for genetic analyses of homothallic P. parasitica isolates. A PCR-based marker for identifying outcrossed progeny in controlled matings between P. parasitica isolates and the segregation of avirulence loci (ATR; Arabidopsis thaliana recognized) was studied in an F₂ population following selfing of a confirmed F₁ individual.

Results and conclusions
A co-dominant cleaved amplified polymorphic sequence (CAPS) marker [2] was produced which differentiated between two P. parasitica isolates: Maks9 and Emoy2. These isolates had been used previously to map several RPP genes in A. thaliana: RPP1 (chr3), RPP4 (chr4) and RPP8 (chr5) using Emoy2; and RPP13 (chr3) and RPP21 (chr5) using Maks9 [3]. The resulting CAPS products from these isolates were dimorphic, and readily enabled identification of hybrid (F₁) progeny. A population of selfed and outcrossed oospores was produced following inoculation of a susceptible A. thaliana accession with a mixture of asexual inoculum from both isolates. Ten oospore-derived cultures were established from this mixed population using the same susceptible accession as the baiting host. The CAPS marker and pathotypic response on a differential set of 20 A. thaliana genotypes provided the basis for distinguishing the progeny isolates; six progeny isolates expressed a hybrid pattern and the remaining four appeared to be selfs of Emoy2.

Asexual inoculum from one of the F₁ progeny was used to inoculate the same baiting host and produce an F₂ oospore population. Among 51 F₂ progeny derived from this population, non-parental pathotypes were observed when each isolate was tested on a differential set of A. thaliana genotypes carrying different combinations of the known RPP genes. Segregation of corresponding ATR loci was determined, and simple inheritance was observed in each case. ATR1 and ATR8 provided clear examples of avirulence controlled by dominant alleles at single loci; ATR13 and ATR21 appear to be dominant alleles at two closely linked loci; and ATR4 appeared to be controlled by partially dominant or recessive alleles at a single locus. Segregation of alleles at a single avirulence locus corresponding with RPP5 was also observed. This was unexpected, as RPP5 had not previously been separated from RPP4 by genetic recombination in the host. Additional F₂ progeny and recombinant inbred lines are currently being produced to provide a genetic resource for associating ATR loci with AFLP markers as a prelude to isolating avirulence genes from P. parasitica.

References
1. Francis DM, St Clair DA, 1993. Current Genetics 24, 100-106.
2. Konieczny A, Ausubel FM, 1993. Plant Journal 4, 403-410.
3. Holub EB, Beynon JL, 1996. Advances in Botanical Research 24, 228-273.