Mapping quantitative trait loci for resistance to downy mildew in Pearl Millet: Field and glasshouse screens detect the same QTL

• Published in: Crop science 2002-07, Vol.42 (4), p.1316-1323
• Main Authors: JONES, E. S, BREESE, W. A, LIU, C. J, SINGH, S. D, SHAW, D. S, WITCOMBE, J. R
• Format: Article
• Language: English
• Subjects: Agricultural production; Agricultural research; Agriculture; Agronomy. Soil science and plant productions; Airborne microorganisms; Analysis; Biological and medical sciences; Cellular biology; Crop losses; Crops; Diseases and pests; Fundamental and applied biological sciences. Psychology; Fungi, Phytopathogenic; Gene mapping; Genetics; Genetics and breeding of economic plants; Grasses; Millet; Pearl millet; Pest resistance; Plant immunology; Plant pathogens; Tropical environments; Varietal selection. Specialized plant breeding, plant breeding aims
• ISSN: 0011-183X; 1435-0653; 1435-0635
• DOI: 10.2135/cropsci2002.1316

Downy mildew, caused by the pathogen Sclerospora graminicola (Sacc.) J. Schrot, can cause devastating yield losses in pearl millet [Pennisetum glaucum (L.) R. Br.]. Breeding for resistance to downy mildew is facilitated by an artificial glasshouse screening method that can be used, worldwide. Quantitative trait loci (QTLs) mapping was used to determine whether resistance QTLs identified under field conditions in India were also detected in glasshouse screens carried out in India and the UK. Quantitative trait loci were mapped using 114 individual pearl millet progeny derived from a resistant x susceptible cross: molecular marker mapping was carried out in an [F.sub.2] population with restriction fragment length polymorphisms (RFLPs), and disease incidence was assessed on [F.sub.4] families. Composite interval mapping (CIM) was used to detect associations between [F.sub.4] family means and marker genotypes. Despite key environmental and methodological differences between the disease screens, the same two QTLs were detected in each screening environment. One QTL had a major effect and explained up to 60% of the phenotypic variation, while the other had a minor effect and explained up to 16% of the phenotypic variation. Two additional QTLs were also consistently detected across screens by examining pair-wise marker interactions. Multiple-trait interval mapping detected all of the QTLs that had been detected in individual screens, including the QTLs that had only been detected by examining pair-wise marker interactions, demonstrating its increased power over single trait mapping. Quantitative trait locus x environment interactions were significant at each QTL due to differences in the magnitude, rather than direction, of QTL effects. The differences in magnitude appeared to be a consequence of the degree of normality of the disease distribution, rather than any differences between screening methods.