| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Plant Genome Mapping Laboratory, Univ. of Georgia, 111 Riverbend Road, Athens GA 30602
b Dep. of Crop and Soil Sciences, Univ. of Georgia, Coastal Plain Exp. Station, Tifton, GA 31793
c Dep. of Soil and Crop Sciences, Texas A&M Research and Extension Center, Lubbock TX 79403
d Center for Applied Genetic Technologies, Univ. of Georgia, 111 Riverbend Road, Athens GA 30602
e Dep. of Soil and Crop Sciences, Texas A&M University, College Station TX 77843
* Corresponding author (paterson{at}plantbio.uga.edu)
THE U.S. COTTON (Gossypium spp.) production system exemplifies the challenges that must be met to reduce the genetic vulnerability of a major crop. As an industrial crop that sustains one of the largest U.S. industries (textiles), more than 400000 domestic jobs are related to cotton production and processing, with an aggregate impact of over $40000 million on the U.S. gross domestic product.
The U.S. cotton gene pool is genetically impoverished. All cottons cultivated in the USA are tetraploid, thought to have arisen in the New World about 1 to 2 million years ago, as a result of an unusual hybridization event between an invasive A-genome diploid genotype and an indigenous D-genome diploid. Polyploid cottons are thought to be monophyletic, with contributions from only two of the eight extant diploid "genome types" in the genus, representing the first genetic bottleneck. A second bottleneck was associated with the domestication of cotton from a small subset of the wild genotypes. A third bottleneck was imposed by human sampling of tetraploid cotton genotypes from their center of diversity in Mexico and Central America, and spread northward into the USA, and also to China, India, Egypt, Australia, and other countries.
Growing concern about genetic vulnerability of the cotton gene pool to a wide range of biotic and abiotic hazards is exemplified by recent investigation of trends in yield improvement. Stagnation in yield improvement of G. hirsutum L. recently led the National Cotton Council to form a Blue Ribbon Committee of public and private scientists to determine if there truly was a yield plateau, and if so, how it should be addressed (Helms, 2000). The committee based its findings on current public data, largely from National Cotton Variety Tests (Rayburn et al., 1999) and the National Agricultural Statistics Service, USDA (NASS, 1998). On the basis of a linear model, over the past 39 yr, cotton yields have increased about 6.7 kg ha–1 yr–1 (1.3% annual rate). A polynomial model clearly indicates that this increase has not been linear. From 1970 through 1985, the rate of change in cotton yields rose at an increasing level. However, from 1985 through 1998, the rate of change in cotton yields declined and from 1992 through 1998, the actual yields declined. Stated another way, the rate of change in cotton yields has steadily declined since 1985. By 1998, absolute cotton yields (not just the rate of change) reached a disturbing rate of decline of about 16.8 kg–1 ha–1 yr–1 (3.3% annual rate). Accompanying this yield decline, year-to-year variations in yield were almost four times greater in the period from 1980 to 1998 than in 1960 to 1979. This increased volatility in yield translates directly into higher risk for the grower.
The yield plateau in cotton appears to be closely associated with increasing genetic vulnerability. In addition to the genetic bottlenecks imposed by polyploid formation, domestication, and migration, the U.S. cotton gene pool has been further eroded by the over-exploitation of a few genetic backgrounds during the past 15 yr. Commercial breeding programs repeatedly employ a few closely related genotypes to generate new cultivars (May et al., 1995). Growers are planting large areas to these closely related cultivars, resulting in a high level of field genetic uniformity (Van Esbroeck et al., 1998). This has been exacerbated by the widespread (about 60% of 1999 U.S. hectares) planting of transgenic cultivars that are the result of backcross breeding with an even smaller subset of closely related genotypes.
A solution may lie in the exploration of exotic genotypes. Although there exist five tetraploid Gossypium species each including a huge array of feral and wild forms, at present, no exotic cottons appear in the pedigrees of any modern U.S. cultivars or enhanced germplasm (Bowman et al., 1996; Bowman et al., 1997; Calhoun et al., 1997). Our early explorations in a small sampling of these genotypes show many transgressive QTL alleles (P. Chee, X. Draye, A.H. Paterson, in preparation), including some cases in which different taxa appear to have evolved complementary alleles at alternative homeologous loci (Saranga et al., 2001).
However, introgressive breeding involves new challenges that require fundamentally different approaches than mainstream plant breeding, and an equally long-term effort. Particularly prominent among the challenges of introgressive breeding appear to be a high level of epistasis (nonlinear interactions between unlinked loci)—in many cases individual QTL loci in combination with different unlinked introgressed alleles affect a phenotype to very different degrees. In a few cases, the same QTL allele has even shown statistically significant opposite effects in different genetic backgrounds (P. Chee, X. Draye, A.H. Paterson, in preparation). This complication, superimposed on the effects of linkage drag, hybrid breakdown, and other well-known characteristics of wide crosses all highlight the fact that successful introgressive breeding of quantitative traits will require a dedicated effort that is designed to address such challenges, rather than being a ‘side project’ in a mainstream plant breeding program. Further, this will clearly be a high-risk, high-reward activity, in the sense that many promising early leads will not translate into commercially viable products ... but those that do may confer larger incremental gains than are available in the present cultivated gene pool.
Implementing this solution requires integration of research, education, and extension activities. Genetic vulnerability is a complex problem that results from a crop's evolutionary history, trends in breeding and biotechnology practices, and grower decisions based on inadequate information being available, all responding to the inevitable pressures of processor and consumer preferences. Gaining the partnership of stakeholders is key. Toward this end, while the introgressive breeding process proceed, we are working with both researchers and extension personnel to create a Web-based resource to provide objective information about relatedness of genotypes, as a management tool for producers to better deploy the remaining variation in the gene pool to minimize genetic vulnerability of the crop, and a research tool for scientists to glean new information useful for crop improvement.
 |
ACKNOWLEDGMENTS |
|---|
Received for publication July 23, 2003.
|
REFERENCES |
|---|
 TOP REFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Plant Registrations | Soil Science Society of America Journal | ||||
| Journal of Natural Resources and Life Sciences Education |
Journal of Environmental Quality |
||||
