| Cotton is an important industrial crop worldwide. With the development of society and the demand of life improvement for people, breeding for varieties with high yield and high quality become more and more important. Conventional breeding is to pyramid more excellent traits by hybridization between cultivars. But the narrow genetic base makes cotton breeding come into a plateau and it difficult to make breakthrough. Wild and semi-wild species have many favorable genes and are important genetic resources for cotton breeding. But they are very hard to transfer to cultivated species in cotton breeding because of linkage drag. To break this linkage drag, one of the important ways is to develop chromosome segment introgression lines (CSILs) which have only one chromosome segment from donor in genetic background of cultivated species. After that, these CSILs will be used to carry out the genetic research on QTLs and genes related to agronomic traits. These CSILs also become basic materials for cotton breeding.G. hirsutum race Latifolium is one of the seven semi-wild upland cotton species. It has the closest relationship with the cultivated cotton varieties. The present upland cotton cultivars were domesticated from the existing race, Latifolium, which were subjected from long-term natural selection in different geographical conditions. It remains favorable genes about insect resistance, disease resistance, stress tolerance and so on. Because it has the same number of chromosomes (2n=4x=52) with the cultivated species, the hybrid between them and their progenies are fertile. In this study we use G. hirsutum race Latifolium acc.TX-48as donor parent and upland cotton genetic standard line TM-1as recipient parent to build chromosome segment introgression lines.1. The screening of parental polymorphism primerIn this study, total6208SSR primers were employed to screen the polymorphisms of two parents, upland genetic standard line TM-1and G. hirsutum race Latifolium acc.TX-48.230polymorphic SSR primers were obtained. The ratio of polymorphism is3.7%.230pairs of polymorphic SSR primers were used to amplify the BC1population with125individuals. There were240polymorphic loci produced. Chi-square test was conducted for the240polymorphic loci. There were21distorted segregation, occupying8.75%. Among of them,11biased towards TM-1and10biased towards donor parent.2. The construction of genetic linkage mapWe used the software Joinmap3.0to construct linkage groups at LOD≥3. It produced51linkage groups in total, containing215SSR loci. The full-length was1989.17cM with the average distance of9.23cM between loci.Using the high-density genetic maps of G.barbadense and G.hirrsutum as a reference, there were40linkage groups (including the189loci) assigned to21chromosomes. The11small groups were not located on any chromosome and named LG01-LG11. The remaining15SSR loci were not assigned to any linkage group.There were13groups with77loci in A-subgenomes.The full-length was716.93cM with the average distance of9.311cM between loci. The13groups were assigned to8chromosomes of A-subgenomes.No linkage groups were assigned to A2, A4, A6, A8and A12. There were24linkage groups with112loci in D-subgenomes. The full-length was1036.197cM with the average distance of9.25cM between loci. The11small linkage groups contained26loci, and the full-length were236.04cM and the average distance between markers was9.08cM.3. The development of chromosome segment introgression lines from G hirsutum race Latifolium acc.TX-48in the background of upland genetic standard line TM-1The development of CSILs:The F1progeny were backcrossed with TM-1to BC3and then self-pollinated to produce BC3F2. The identification was conducted to identify the donor chromosome segments in individuals of BC3F2. There were3-35donor chromosome segments in the individuals of BC3F2with the average of18.91. No single fragment introgression lines were found in BC3F2. The percentage of genetic background was81.25%-98.33%with the average of89.16%. Duo to more donor fragments in a single individual backcross must be performed to isolate the single donor fragments in their progenies. The statistical analysis on traits within-family in BC3F2showed that the genetic diversity was large, while the mean of the traits were close to recurrent parent. Due to each individual with different donor segments within a single family, their traits shows difference each other but the values within a single family were subjected to the genetic background.4. QTL mapping for yield traits was conducted using chromosome fragment introgression linesQTL mapping for four yield traits was conducted using chromosome fragment instrogression lines. Thirteen QTLs were found for four yield traits based on single marker analysis, one for boll number QTL, two for fruiting branches number QTL, nine for single boll weight QTL, one for lint percentage QTL. Seven additive QTLs were found by likelihood ratio test based on step wise regression for additive QTL (RSTEP-LRT-ADD), one for boll number additive QTL, two for fruiting branches number additive QTL,nine for single boll weight additive QTL, one for lint percentage additive QTL.We found a multi-effect additive QTL for boll number and fruiting branches number. And we found that the boll weight additive QTL with a negative value.It showed that the enhance gene could come from both donor parent and the background parent. Forty pairs of epistasis QTL were found based on likelihood ratio test based on stepwise regression for digenic epistasis (RSTEP-LRT-EPI). |
PDF Full Text Request