Rice Genetics IV - IRRI books - International Rice Research Institute

even at the juvenile stage. Several traits, although being controlled by a single gene,are quite laborious to measure. Most traits that fall under this category are consideredQTL that affect agronomic traits. Some types of disease resistance are quantitativeor partial in nature and require replicated testing to measure accurately. Examplesinclude resistance to blast (Wang et al 1994) and stem rot (Ni et al 2001). Althoughresistance conferred by one or two loci is not complete, it can be highly effectiveunder field conditions in reducing damage from the disease. For some abiotic resistancetraits such as salinity tolerance and submergence tolerance, selection in traditionalbreeding programs can only be done when the specific stress is present. Inmany cases, the trait must be measured in special screening nurseries grown fromprogeny seed. MAS has made it possible for breeders to perform indirect selectionwith the molecular markers closely linked with the genes. Although disease resistanceis often easy to score in screening nurseries, scoring is not possible for recessiveresistance genes during backcrossing unless an additional selfing generation is performed.Traits that can only be measured after the reproductive stage would be good candidatesfor marker-assisted selection. For example, amylose content is currently measuredafter harvest using chemical methods or sophisticated equipment. A microsatelliterepeat that is part of the Wx gene (Ayres et al 1997) can be effective in selecting forthis trait. PCR-based markers proved 85% accurate for identifying the thermosensitivemale sterility gene tms3 in the juvenile stage (Lang et al 1999). The ultimate exampleof this concept is genes that control traits that can only be observed in the progeny oftest crosses of individual plants. Two examples are the wide compatibility allele S5n(Liu et al 1997) and restorer genes for cytoplasmic male sterility (Akagi et al 1996,Ichikawa et al 1997, Yao et al 1997). The ability to identify such genes at the seedlingstage during a backcrossing program would offer a remarkable savings in time andeffort to transfer these genes into a specific genetic background.Pyramiding multiple genes. Gene pyramiding is considered a viable approach toattaining durable resistance to rice diseases or insect pests. Different resistance genesoften confer resistance to different isolates, races, or biotypes. Combining these resistancesbroadens the number of races or biotypes that a variety can resist, and there isevidence that multiple resistance genes make it more difficult for virulent races toevolve. Furthermore, combining major-gene and minor-gene resistance may lead toincreased durability (Wang et al 1994). When partial resistance is present in breedinglines, scoring for major-gene resistance can also be more difficult (Kelly and Miklas1998). In some cases, multiple pathogen races or insect biotypes can be used to detectplants with more than one resistance gene, but in practice this may be difficult orimpossible. Molecular markers can be used to select for these multiple resistancessimultaneously. When hybrid crops are the goal, additional options for pyramidingdifferent resistance-gene combinations into different parents exist (Witcombe and Hash2000).Hittalmani et al (2000) used marker-assisted selection to combine three blast resistancegenes, Pi1 on chromosome 11, Piz-5 on chromosome 6, and Pita on chromosome12, in a single genotype. For Piz-5, a single marker was used, whereas flanking142 Mackill and Junjian Ni