Defining Chromosome Regions Related with Stripe Rust Resistance QTLs as the Foundation for Characterizing Resistance Gene Pyramids

A. Castro (1), A. Corey (1), T. Filichkin (1), P. Hayes (1), J. Korte (1), J.S. Sandoval-Islas (2), T. Toojinda (3), H. Vivar (4)

(1)Dept. of Crop and Soil Science
Oregon State University
Corvallis, OR 97331, USA
(2)Instituto de Fitosanidad
Colegio de Postgraduados
Chapingo, Mexico
(3) National Center for Genetic Engineering and Biotechnology
Kasetsart University
Nakorn Pathom 73140, Thailand
(4) ICARDA/CIMMYT
Apdo. 6-641
06600 Mexico, DF, Mexico.

Barley stripe rust (caused by Puccinia hordei fsp. hordei) is an important barley disease in North and South America. Qualitatively and quantitatively inherited sources of resistance have been identified and mapped in several populations. The latter are potentially more durable than the former but are more difficult to use in cultivar development. The resistance QTLs we have mapped are distributed across the genome, which will allow combining different QTL alleles from different sources in a single line. Marker assisted selection (MAS) can help to achieve this goal, but it requires adequate definition of the target regions in all the parents involved in the cross, as well as availability of markers that discriminate between the parents. In order to pyramid different QTL alleles, 115 DH lines were derived from a complex cross that combined resistance QTL alleles on chromosomes 4 and 7 from the variety Orca (mapped in the Cali-sib/Bowman population) with a resistance allele on chromosome 5 from the experimental line DI-72 (mapped in the Shyri/Galena population). The complex cross also included the malting variety Harrington. The original mapping populations were analyzed and compared for the three target regions and SSRs were screened for their ability to discriminate among parents. The genotyping of the 115 lines with the SSRs will allow precisely definition of the effects of each QTL and all their joint effects. Preliminary results show that a simple additive model is not adequate to explain the effects of the QTLs in different genetic backgrounds.