Project Summaries

09-546  Project Manager: D. C. Jones


Peng W. Chee, University of Georgia Research Foundation, Inc.

Gossypium hirsutum L. has a narrow gene pool having experienced genetic bottlenecks during polyploid formation and divergence from its sister polyploid species. The domestication of a small subset of the wild genotypes and over-exploitation of only a few genetic backgrounds in modern Upland cotton breeding results in genetically uniform cotton germplasm. The slow genetic progress in improving fiber productivity and quality is indicative that many of the favorable genes may have reached fixation in elite gene pools. This lack of genetic variation has hampered the ability of breeders to provide low-cost intrinsic genetic solutions for cotton improvement.

Because of the over-exploitation of a few genetic backgrounds, new sources of genetic variation need to be introduced into the cotton gene pool to ensure future success in breeding new cotton cultivars. Our long-term breeding strategy is to broaden the genetic base of cotton by assessing new genetic variation from the four tetraploid Gossypium species, G. tomentosum, G. darwinii, G. barbadense and G. mustelinum. Wide-crosses from G. barbadense and G. tomentosum have been extensively studied by Chee's and Paterson's laboratories. Since G. darwinii is closely related to G. barbadense, G. mustelinum is a logical species for gene introgression exploration. The primary overall goal of this project is to develop a series of genetic populations with new gene combinations from G. mustelinum in an elite Upland cotton genetic background in order to increase the genetic diversity available for breeding improvements. Since G. mustelinum is considered to be the most distantly related cotton tetraploid to Upland cotton, we expect to find novel alleles and genetic combinations. We will use SSR markers to determine the size and chromosome locations of all G. mustelinum introgressions in each population and test the introgressed segments for genes useful in improving fiber productivity and quality.

G. mustelinum, a wild cotton species native to northern Brazil (Fryxell 1979), is the most distantly related tetraploid to Upland cotton and therefore, the progenies derived from interspecific crosses between G. mustelinum and Upland cotton should harbor the greatest number of novel alleles among the four tetraploids. In this proposed work, we will (1) develop a series of near-isogenic introgression lines (NIILs) from G. mustelinum, each carrying ca. 6.25% of the donor genome in an elite Upland genetic background; (2) identify the genomic location of all the introgressed chromosome segments; and (3) detect gene combination suitable for improvement of Upland cotton. We expect this project to not only address the long-term genetic vulnerability of Upland cotton through the release of new cotton germplasm but also deliver a short- to medium-term solution for the improvement of fiber quality through genomic tools suitable to expedite deployment of new gene combinations.

We have developed a BC3F3 population consisting of 35 families, each with about 150-200 individuals. The entire population was planted in a field near Tifton, GA in 2006 and DNA was collected for genotyping. The individuals and the genes controlling desirable fiber traits were identified using DNA marker technologies and will be pyramided into a common breeding line. In 2008, 12 BC3F3 families were planted near Tifton in two replications for the purpose of progeny testing to verify the QTLs identified in the BC3F2 populations. In 2009 the 12 families, now BC3F4, were again planted in two replications near Tifton to provide additional supportive data to verify the fiber quality QTLs identified in the BC3F2 populations. In 2010, the 12 families were planted to develop NIILs with homozygous introgressed G. mustelinum chromatin. There were 2,846 plots planted in 2008, 2,780 plots grown in 2009, and 1,378 plots grown in 2010. Each of the plots in 2008 and 2009 was sampled and ginned with the lint sent to the Cotton Incorporated Fiber Laboratory for HVI testing. The 2008 samples from 5 of the 12 families and the 2009 samples of all of the 12 families were tested by AFIS. In 2011, the single plant selections that represented the most common plant structure of each of the plots grown in 2010 were advanced a generation. In 2012 534 plots that were unsuccessful in 2011 were replanted from remnant seed. 345 plots were successfully recovered and along with the successful plots in 2011, seed for subsequent research needs is available. The lint has been sent for fiber quality analysis.

The crosses initiated using our breeding program's elite cultivars GA230 and GA2004303 with selected lines from these backcross families were completed in 2012. The selected lines are improved fiber quality lines that have introgressed G. mustelinum QTLs. The populations of these crosses will be used to test the utility of these fiber quality QTLs specifically as well as providing an estimate of overall value of G. mustelinum germplasm.

The development of populations from these QTL donor individuals was initiated with four background cultivars Acala SJ4 (Acala type), Paymaster HS 26 (picker type), Deltapine DP 50 (MidSouth type) and the University of Georgia's GA 2004089 (Southeastern type). The means for fiber length and strength in the BC3F2, BC3F3, and BC3F4 families were analyzed. As was expected for a majority of BC3F2, BC3F3, and BC3F4 families, the average phenotypes were poorer than that of the recurrent parent. This 'negative' transgression, generating a phenotype that is poorer than the recurrent parent, suggests that many new gene combinations formed by interspecific hybridization are undesirable. A similar pattern of transgressive segregation has been reported in advanced backcross populations involving another allotetraploid cotton G. barbadense for a host of fiber quality characters (Chee et al. 2005a, b). However, we have observed several BC3F2, BC3F3, and BC3F4 families with fiber length (Pop 10, 11, & 27) and strength (Pop 20, 27, 34, & 35) that were better than the recurrent parent. It is interesting to note that in the G. barbadense advanced backcross population, for the families in which the means for fiber fineness were superior to that of the recurrent parent, all contained at least one QTL with the favorable allele contributed by the G. barbadense parent (Chee et al. 2005a). Our preliminary association analysis suggests that QTLs are indeed present in the families in which the fiber qualities were better than the recurrent parent.

Although a majority of the QTLs detected in the G. barbadense population accounted for only small portions of the phenotypic variance, a few QTLs explaining more than 20% of the phenotypic variance were detected. Furthemorer, several of the QTLs were consistently expressed across different BC3 families. This supports the hypothesis that improvement of fiber quality for Upland cotton may be achieved by introgressing QTLs from other tetraploid species.

This year we will further evaluate the BC3 families for fiber quality data and continue to utilize association analysis between fiber quality data and the SSR markers to detect QTLs that may be segregating in these families. Furthermore, the backcross program to stack several exotic length and strength fiber quality QTLs from G. mustelinum into elite cotton germplasm will continue. The populations of these crosses will test the utility of these fiber quality QTLs specifically as well as providing an estimate of overall value of G. mustelinum germplasm.


Project Year: 2012

Search 2012 Projects:

▸ Alabama
▸ Arizona
▸ Arkansas
▸ California
▸ Florida
▸ Georgia
▸ Louisiana
▸ Mississippi
▸ Missouri
▸ New Mexico
▸ North Carolina
▸ Oklahoma
▸ South Carolina
▸ Tennessee
▸ Texas
▸ Virginia

▸ Cotton Incorporated Fellow
▸ Cottonseed
▸ Crop Improvement
▸ Farm-to-Mill
▸ Production Efficiency
▸ Sustainable Cotton
▸ Variety Improvement

Project Manager

▸ All Project Nos.

Share This: