|Holland, Jim - Jim|
Submitted to: American Oat Workers Conference Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 8/1/2002
Publication Date: 8/1/2002
Citation: ECKSTEIN, P.E., SCOLES, G.J., HOLLAND, J.B., MCELORY, A.R. STRATEGIES FOR THE DEVELOPMENT AND IMPLEMENTATION OF MOLECULAR MARKERS IN PLANT BREEDING PROGRAMS.. AMERICAN OAT WORKERS CONFERENCE PROCEEDINGS. 2002. Interpretive Summary: Genetically modified maize containing Bt genes are engineered to be resistant to lepidopteran insects, including corn borers. Farmers in the U.S. rapidly adopted Bt corn hybrids, with 19% of the U.S. corn acreage planted to Bt hybrids in the year 2000. Corn breeders develop new inbred lines designed to be used to produce hybrid cultivars. Breeders need to know if the value of their inbred lines as hybrid parents depends on whether they are crossed with conventional or Bt lines. This question was investigated by crossing 97 new tropical maize breeding lines to both conventional and Bt hybrids. The Bt and conventional hybrids used in this study differed only by the presence or absence of the Bt gene. The crosses were evaluated in seven environments, and average yields did not differ if lines were crossed to the Bt or conventional hybrid. The Bt hybrid did not have higher yield than the conventional hybrid, but the tropical corn crosses had significantly greater yield than the commercial hybrids, suggesting that tropical corn is a more important source of yield improvement genes than are insect resistance transgenes.
Technical Abstract: Molecular markers have been slow to meet expectations, but markers are currently providing real benefits in a number of breeding programs. Improved technologies provide opportunities to increase the usefulness of markers to breeding programs by reducing their costs and labor requirements and by increasing their informativeness and speed. Implementation of marker-assisted selection (MAS) requires first the development of useful markers. The choice of marker technologies depends primarily on: (1) their utility in terms of cost, ease of use, screening potential (ability to screen the entire genome quickly and evenly), polymorphism potential, and conversion potential; (2) developing the polymorphism into an economical and accurate diagnostic tool, such as microsatellites, SCARs (dominant or codominant, and potentially allele-specific), flanking markers, and can depend on the ability to multiplex, and to tag fluorescently; and (3) scaling up the process, which may require quick DNA template preparation methods, fluorescent tags, and liquid handling/robotics stations. Effective application of MAS also requires delineation of situations in which MAS will be more effective than phenotypic selection. The most promising areas for implementation of MAS include: (1) backcrossing, (2) selection for traits that are under simple genetic control but costly or difficult to phenotype; (3) selection in off-season nurseries or greenhouses, where the desired traits cannot be evaluated phenotypically, and (4) selection for favorable alleles in exotic or wild species germplasm. Development of MAS for these situations may help to economize the use of markers in selection for more complex or less valuable traits. Finally, plant breeders must consider several practical issues when integrating MAS into a breeding program. The first, most basic aspect is that of resources: how can relatively small breeding programs acquire the infrastructure and expertise needed to evaluate breeding lines? The development of contract services and `marker labs' that serve several programs within an institution will alleviate this obstacle. The availability of `user-friendly' markers such as SCARS is essential for MAS, but markers must also be `robust' (give consistent, clear results under slightly different lab conditions) to be useful. The structure of the breeding protocol may require some modifications as MAS is introduced. For example, it may be more cost effective to fix some desirable alleles in parental genotypes rather than screen large numbers of progenies in later generations for the desired phenotype. As well, the choice of parents will be affected by polymorphism patterns. The breeder must choose parents based on polymorphism patterns as well as trait combinations, but limited polymorphism among desirable parents could limit the application of some markers. Use of multiple markers per trait and the use of highly polymorphic classes of markers address this issue as well.