An RFLP and QTL Linkage Map in Gossypium hirsutum L.
Dec 21, 2018
Ninety-six F2.F3 bulk sampled families of Upland Cotton, Gossypium hirsutum L. from a cross of MARCABUCAG8US1-88 X HS 46 were analysed with 129 probe/enzyme combinations resulting in 138 RFLP loci. There were 84 co-dominant markers of which 76 fit a normal 1:2:1 ratio. There were 54 dominant markers of which 50 fit a normal 3:1 ratio. Using MAPMAKER\EXP program these were arranged into 31 linkage groups with 120 linked loci and 18 unlinked loci. These covered 865 cM or 18.6% of the estimated cotton genome. We used the mixed model analysis of Zhu and Weir (1998) to analyse QTLs associated with 19 agronomic and fiber traits scored mostly in F2.F5 families. The model also provided estimates of the additive and dominance genetic effects. We mapped 100 QTLs to 60 maximum likelihood locations in 24 linkage groups. Several QTLs influenced more than one trait. For example, in linkage group 14 we found a QTL with significant genetic effects for micronaire, and three Arealometer measurements. In linkage group 19 four closely linked QTLs had significant effects on strength, fineness, and maturity
Meredith (1992) reported the first RFLP evaluations in cotton when he studied heterosis and varietal origins of several cultivars using RFLP probes. The only RFLP linkage maps in a cross of two upland cotton (G. hirsutum) lines have been reported by Shappley (1994, 1996), Shappley et al. (1996), Shappley et al. (1998). Molecular markers provide increased potential for gathering useful information for cotton improvement. The identification of QTLs controlling traits of interest to breeders of upland cotton and their association with RFLP markers was the focus of our research.
Material and Methods
The two cultivars chosen for our study are very diverse in pedigree and in RFLP patterns. Agronomic and fiber properties are also very diverse between these two parents. We crossed MARCABUCAG8US-1-88 X HS 46 and screened 9 F1 plants for RFLP variability. Some variability was found so we chose one F1 plant and self pollinated it for the beginning of segregating generations for this study. The RFLP analysis was conducted by the commercial laboratory, Biogenetic Services, Brookings, SD. This laboratory developed most of the probes from a cDNA library using leaf material collected from six diverse cultivars. We grew F2.F3 families and harvested leaf samples. Shappley et al. (1998) contains details of the RFLP analysis All RFLP probes were scored visually on x-ray film after exposure to the proper conditions. Parameters for detection of linkage groups of RFLP loci were a LOD score of 3.0 or greater and a genetic distance of 50cM. The fiber analysis was by Starlab Inc., a commercial fiber laboratory, in Knoxville, TN. Lint for fiber analysis was from hand harvested bolls on plants in F2.F5 families, ginned on a 10-saw gin.
Results and Discussion
Most RFLP loci segregated normally for co-dominant or dominant loci. Twelve of 138 loci showed distorted segregation. Parallel research with cytogenetic deficiency stocks showed that three of the 12 markers also segregated abnormally in the cytogenetic studies. The cause of abnormal segregation is not known but abnormal segregation is not unusual with molecular markers. Schon et al. (1993) reported an excess of heterozygotes in his study of corn. Saha (1989) reported an excess of heterozygotes in his analysis of isozyme alleles in cotton. The detailed genetic linkage map is reported in Shappley et al. (1998).
We found 31 linkage groups with 2 to 10 markers each in our cross of two G. hirsutum lines. Reinisch et al. (1994) developed a detailed RFLP map of cotton; however, they used an interspecifc cross of G. hirsutum L. race “palmeri” by G. barbadense L. accession K101. Jiang et al. (1998) published a QTL linkage map showing several QTLs for fiber traits in an interspecific cross. Breeders have known for a long time that interspecific crosses show abnormal segregation. Thus, these linkage map developed from an intraspecific G. hirsutum cross may be more useful to breeders.
Segregation of phenotypic traits in our research, except number of nodes, was normal based upon the skewness and kurtosis values suggesting that these data were suitable for QTL analysis. The phenotypic mean and range for each trait are shown in Table 1.
Proceedings of the World Cotton Research Conference-2. Athens, Greece, September 6-12, 1998. pp.217-220.
J.N. Jenkins et al.