Theoretical and Applied Genetics

, Volume 117, Issue 2, pp 261–272

Exploiting regulatory variation to identify genes underlying quantitative resistance to the wheat stem rust pathogen Puccinia graminis f. sp. tritici in barley

Authors

  • Arnis Druka
    • Genetics ProgrammeScottish Crop Research Institute
  • Elena Potokina
    • School of BiosciencesUniversity of Birmingham
  • Zewei Luo
    • School of BiosciencesUniversity of Birmingham
  • Nicola Bonar
    • Genetics ProgrammeScottish Crop Research Institute
  • Ilze Druka
    • Genetics ProgrammeScottish Crop Research Institute
    • School of Computing and Creative TechnologiesUniversity of Abertay
  • Ling Zhang
    • Department of Crop and Soil SciencesWashington State University
  • David F. Marshall
    • Genetics ProgrammeScottish Crop Research Institute
  • Brian J. Steffenson
    • Department of Plant PathologyUniversity of Minnesota
  • Timothy J. Close
    • Department of Botany and Plant SciencesUniversity of California
  • Roger P. Wise
    • Corn Insects and Crop Genetics Research, USDA-ARS and Department of Plant PathologyIowa State University
  • Andris Kleinhofs
    • Department of Crop and Soil SciencesWashington State University
  • Robert W. Williams
    • Department of Anatomy and NeurobiologyUniversity of Tennessee
  • Michael J. Kearsey
    • School of BiosciencesUniversity of Birmingham
    • Genetics ProgrammeScottish Crop Research Institute
Original Paper

DOI: 10.1007/s00122-008-0771-x

Cite this article as:
Druka, A., Potokina, E., Luo, Z. et al. Theor Appl Genet (2008) 117: 261. doi:10.1007/s00122-008-0771-x

Abstract

We previously mapped mRNA transcript abundance traits (expression-QTL or eQTL) using the Barley1 Affymetrix array and ‘whole plant’ tissue from 139 progeny of the Steptoe × Morex (St/Mx) reference barley mapping population. Of the 22,840 probesets (genes) on the array, 15,987 reported transcript abundance signals that were suitable for eQTL analysis, and this revealed a genome-wide distribution of 23,738 significant eQTLs. Here we have explored the potential of using these mRNA abundance eQTL traits as surrogates for the identification of candidate genes underlying the interaction between barley and the wheat stem rust fungus Puccinia graminis f. sp. tritici. We re-analysed quantitative ‘resistance phenotype’ data collected on this population in 1990/1991 and identified six loci associated with barley’s reaction to stem rust. One of these coincided with the major stem rust resistance locus Rpg1, that we had previously positionally cloned using this population. Correlation analysis between phenotype values for rust infection and mRNA abundance values reported by the 22,840 GeneChip probe sets placed Rpg1, which is on the Barley1 GeneChip, in the top five candidate genes for the major QTL on chromosome 7H corresponding to the location of Rpg1. A second co-located with the rpg4/Rpg5 stem rust resistance locus that has been mapped in a different population and the remaining four were novel. Correlation analyses identified candidate genes for the rpg4/Rpg5 locus on chromosome 5H. By combining our data with additional published mRNA profiling data sets, we identify a putative sensory transduction histidine kinase as a strong candidate for a novel resistance locus on chromosome 2H and compile candidate gene lists for the other three loci.

Supplementary material

122_2008_771_MOESM1_ESM.pdf (289 kb)
Fig. S1 Interval mapping of the PC3 rust infection phenotype and transcript abundance of the two Hsp17 gene family members across all seven chromosomes. Vertical arrows show the positions of Hsp17 (one of the family members) and dRsmMx (resistance to BSMV) genes. (PDF 290 kb)
122_2008_771_MOESM2_ESM.pdf (237 kb)
Fig. S2 Association of the PC4 stem rust infection phenotype with eQTL of ADF6 and COR413-PM1. QTL scans across the markers mapped to chromosome 5H using the Interval Mapping function are shown. (PDF 237 kb)
122_2008_771_MOESM3_ESM.pdf (559 kb)
Fig. S3 Interval mapping of the mRNA abundance of the genes underlying eight probe sets identified by Zhang et al. (2006). (PDF 559 kb)
122_2008_771_MOESM4_ESM.pdf (401 kb)
Fig. S4 Interval mapping of IT2 phenotype. a Phenotypic scores from all 150 SM DHL were used. b Only lines that have the Steptoe allele at the dRpg1 locus (Rpg1) were used for mapping IT2. c Only lines that have Morex allele at the dRpg1 locus (rpg1) were used for mapping IT2. (PDF 402 kb)
122_2008_771_MOESM5_ESM.doc (283 kb)
Table S1 (DOC 283 kb)
122_2008_771_MOESM6_ESM.doc (168 kb)
Table S2 (DOC 169 kb)
122_2008_771_MOESM7_ESM.doc (492 kb)
Table S3 (DOC 493 kb)
122_2008_771_MOESM8_ESM.doc (66 kb)
Table S4 (DOC 67 kb)
122_2008_771_MOESM9_ESM.doc (54 kb)
Table S5 (DOC 54 kb)

Copyright information

© Springer-Verlag 2008