DNA markers development
Molecular marker technology has progressed rapidly in the 1990s with the construction of the first genetic map of cotton published in 1994. We are collaborating with the Plant Genome Mapping Laboratory (Dr. Andrew Paterson, UGA, Athens, GA) to expand the cotton map that now includes over 2500 loci (Rong et al. 2004). While it is recognized that molecular markers are becoming an important tool in plant breeding, the complexity of DNA-based assays is hindering its use in practical plant breeding. The real business of plant breeding; i.e. crossing superior parents and planting large segregating populations in the field for selecting those few genotypes with superior or novel attributes; requires a DNA marker system that is cost effective, reliable, amenable to screening large populations, and is quick to produce results. The standard DNA marker system (RFLPs) for genetic mapping and QTL analysis in cotton is ill-suited for breeding programs. However, our lab is developing a DNA marker system based on sequence-tagged site PCR technology that satisfies the practical elements required for use in breeding programs (Chee et al. 2004). With recent improvements in DNA detection methods, we have extended the utility of this PCR technology by demonstrating how it can be used to target specific regions of a gene such as the intron sequences (Kumar et al. 2006).
Genetic dissection of quantitative traits
The availability of a large number of genetically anchored DNA markers has allowed us to study genome organization and evolution in polyploid cotton and the inheritance of traits important to cotton production. Some of the current or recently completed projects are listed below.
A genetic map of the diploid A genome of Gossypium has been contructed (Aparna et al. 2006) which represents the progenitor of the At subgenome of tetraploid cotton. Using a comparative genetic mapping approach, various chromosome structural rearrangements between the genomes of A diploids and their At tetraploid counterparts were detected. The data also indicates that the At genome has shown a substantial increase in recombination after its merger into the tetraploid nucleus, which may have promoted chromosomal structural changes and enhanced the evolution of genes with new functions subsequent to polyploidization. In a companion study, the genes conditioning flower morphology in the diploid and tetraploid cottons were mapped. While many of the genes mapped to the A genome showed orthologs in At and Dt subgenomes, others revealed interesting differences in the diploids and tetraploids and suggest possible clustering of genes with similar functions in the Gossypium genome.
Because cotton improvement relies mainly upon repeatedly crossing a few closely-related genotypes and inbreeding approaches that tend to preserve homozygosity and homogeneity, the lack of genetic progress in improving yield and fiber properties in recent years suggests that many favorable genes have reached fixation in the elite gene pool. The most immediate and promising opportunities to enhance the genetic diversity of Upland cotton derive from crossing elite cultivars with the four divergent allotetraploid relatives in the secondary gene pool, which include the cultivated species G. barbadense and three non-domesticated species G. mustelinum, G. darwinii and G. tomentosum. In collaboration with the Plant Genome Mapping Laboratory (Dr. Andrew Paterson, UGA, Athens, GA) and the Texas A&M Cotton Breeding Programs (Dr. Wayne Smith at College Station and Dr. John Gannaway at Lubbock), we have created advanced-backcross populations using allotetraploid species with the goals of detecting quantitative trait loci that contribute high-value traits to Upland cotton cultivars, toward developing new interspecific gene combinations. Advanced-backcross populations of G. hirsutum x G. barbadense have been characterized in detail (Jiang et al. 2000), and advanced backcross stocks of G. hirsutum x G. tomentosum and G. hirsutum x G. mustelinum are nearing completion.
Most traits such as lint yield and fiber properties are governed by many genes, each with a small effect. Managing these quantitative traits is difficult in classical breeding simply because of the number of genes involved. Frequently, the expression of genes controlling quantitatively inherited traits is also greatly influenced by the environment, reducing response to phenotypic selection. We have identified DNA markers associated with QTLs for improved fiber quality such as length, strength and uniformity from Pima cotton (Chee et al. 2005). The DNA markers that showed linkage to fiber length QTLs are currently being used in the lab to screen for Pima introgressant genes that confer enhanced fiber length properties in upland cotton as well as to develop a set of near-isogenic introgression lines.
We are currently using the same technology to identify DNA markers linked to root-knot nematode resistance in cotton. In collaboration with the USDA-ARS Crop Protection & Management Research Unit (Dr. Richard Davis, cotton nematologist, Tifton, GA), we have identified two chromosome regions significantly associated with the resistant phenotype. Current studies are being done to confirm the genetic linkage as well as to identify additional DNA markers more closely associated with the resistance genes and to determine if these markers can be used to select resistant progenies in breeding programs.
