Linkage Maps Construction In Cotton And QTL Mapping For Yield And Fiber-Related Traits

Cotton is one of the most important economic crops in the world, cotton fiber is the most important material for textile industry, and cotton has an important strategic status in national economy. Under the contiouous efforts of cotton breeders, cotton yield has been greatly improved, other traits such as fiber quality, anti-desease and anti-insects have also been improved. However, the development of cotton cultivers lagged behind the fast development of textile industry, especially for cotton quality. Cultivars with high yield and good quality are imperative, while cotton yield traits are negatively related with quality traits which make it difficult to improve yield and quality at the same time by traditional breeding ways. The development of molecular markers and molecular quanitative genetics make it easy to dissect quanitative traits into mendelian factors, and the genetic basis of quanitative traits could be better understood so that we could combine good traits into single individual intentively. In this study, linkage maps for tetraploid cotton were constructed and some QTLs for yield and fiber quality were mapped which would be useful for cotton breeding. The results were as follows:1. The application of SRAP in cotton molecular analysisSRAP (Sequence-related Amplified Polymorphism) is a new marker system based on PCR and preferely amplifies ORFs (open reading frames) by special primer design. SRAP combines simplicity, reliability, moderate through-put ratio and facile sequencing of selected bands and have been applied in map construction, comparative genetics, genetic diversity and gene mapping.In this study, SRAP was applied in cotton genome analysis. We developed an efficient PCR reaction system for detecting SRAP that showed reliable, repetitive and productive.SRAP marker was used to detect the polymorphisms between "Pima90" (G barbadense L.) and "Handan208" (G hirsutum L.), which showed comparatively high polymorphism, every primer combination producing about 50 clearly discriminable bands. Seventy six primer combinations with higher polymorphism were used to screen the population, and generated 285 polymorphic bands with an average of 3.75 polymorphic bands per primer pair and a maximum of 13 polymorphic bands. Linkage analysis showed that 237 loci mapped into 39 linkage groups with 48 orphan loci. This linkage map was 3030.7cM, covering 65.4% of the cotton genome. There were 2-13 loci in each linkage group; the longest one was 227.4cM and the shortest one covered 5cM. The distance between adjacent markers ranged from 0.2cM to 42.8cM with an average distance of12.79cM. All the markers distributed evenly among the linkage groups without clustering of loci. This was the first linkage map of cotton constructed by SRAP markers.The SRAP marker was also used to screen the intraspecific population of DH962 X Jimian5. Thirty seven primer combinations produced polymorphic between parents and generated 48 loci. Single primer combination could produce 1.3 loci in average and 3 loci at the maximum. Compared to RAPD and SSR, SRAP showed relatively higher polymorphism ratio in intraspecific population of upland cotton.All the results showed that SRAP marker could be widely used in cotton molecular analysis.2. Interspecific linkage map construction for tetraploid cotton and mapping QTLs for fiber-related traits using SRAP, SSR and RAPDTetraploid cotton is one of the most extensively cultivated species. Upland cotton (G. hirsutum L.) and Island cotton (G barbadense L.) are two impotant tetraploid cultivars, which account for 90% and 5% of the world cotton production respectively. Upland cotton is characterized as high yield and general fiber quality; while Island cotton is famous for good quality but low yield. By combining these two cultivars, it is possible to get their offsprings with high yield and good quality. To better understand the genetic basis of cotton fiber trait for the improvement of fiber quality, a genetic linkage map of tetraploid cotton using SRAPs, SSRs and RAPDs was constructed. A total of 121 SRAP primer combinations with better polymorphism from 238 primer combinations, 368 SSR primer pairs and 600 RAPD primers were used to screen polymorphisms between G hirsutum cv. Handan208 and G barbadense cv. Pima90 which revealed 749 polymorphic loci in total (205 SSRs, 107 RAPDs and 437 SRAPs). Sixty nine F2 progeny from the interspecific cross of "Handan208"x"Pima90" were genotyped with the 749 polymorphic markers.Five hundred sixty six loci were assembled into 41 linkage groups with at least 3 loci in each group. Twenty eight linkage groups were assigned to corresponding chromosomes by SSR markers with known chromosome locations. The map covered 5141.8cM with a mean interlocus space of 9.08cM.A X~2 test for the significance of deviations from the expected ratio (1:2:1 or 3:1) identified 135 loci (18.0%) with skewed segregation including 70 SSR loci (31.4%), 17 RAPD loci (15.9%) and 48 SRAP loci (11%). One hundred and four loci (77%) skewed towards the "Handan208" genotype, 17 loci (12.6%) skewed towards the "Pima90"genotype and 14 (10.4%) had the heterozygous genotype. One hundred and two loci mapped into the linkage map distributing in 34 linkage groups with the tendency of clustering at each bottom of the linkage groups.Totally 13 QTLs associated with fiber-related traits were detected by compositive interval mapping, among which 2 QTLs were for fiber strength, 4 for fiber length and 7 for micronaire value. These QTLs were in 9 linkage groups explaining from 16.18% to 28.92% of the trait variation. Six QTLs were located on the A subgenome, the genome donor of fiber trait; 6 QTLs on the D subgenome derived from an ancestor that does not produce spinnable fibers, and 1 QTL in an unassigned linkage group. Ten QTLs (76.9%) associated with SRAP markers indicated the power of SRAP markers to detect QTLs because of the preferentially targeting ORFs. There were 4 QTLs clustering in LG1, among which 3 QTLs for micronaire value. A QTL near m7ell-360 located in LG1 affected fiber length and micronaire value simultaneously.3. Intraspecific linkage map construction of tetraploid cotton using PCR markers and QTLs mapping for yield and fiber-related traitsIn plant breeding, intraspecific diversity is usually used; construction of linkage maps for targeting traits of breeding interest would greatly increase breeding efficiency. An intraspecific F2 segregation population of DH962 (an introgression line of G. thurberi with good fiber quality) and Jimian5 was constructed, and used for linkage map construction and QTL mapping for yield and fiber related traits.A total of 600 RAPD primers, 929 SSR primer pairs and 238 SRAP primer combinations were used to screen the parental polymorphisms. As a result, 15 RAPD primers (2.5%), 91 SSR primer pairs (6.25%) and 37 SRAP primer combinations (15.55%) showed polymorphism between parents. A total of 156 loci generated from these primers including 17 RAPD loci, 91 SSR loci and 48 SRAP loci. Single SRAP primer combination could produce 1.3 loci in average with 3 loci at the maximum.One hundred thirty nine loci mapped into 32 linkage groups including 79 SSR loci, 15 RAPD loci and 45 SRAP loci with 2-14 markers in each group. These groups ranged from 1.1 cM to 90.8cM with an average distance of 31.83cM, and the mean interlocus distance was 9.08cM. This map covered 1018.5cM, about 21.86%of the total cotton genome.Among the 156 loci, 10 loci (6.41%) showed distortion including 7 SSR markers, 3 SRAP markers, and no RAPD markers were distorted. Among the 10 loci, 7 were dominant markers and 3 were co-dominant ones; five markers skewed towards Jimian5, 3towards DH962 and 2 towards heterozygous genotype. Except BNL3932, the other 9 markers distributed in 7 groups; the two markers in LG30 were all distorted and the two distorted markers in LG7 were at the bottom.A total of 14 QTLs associated to yield-related traits were detected and explained 7.35%~43.78% of the trait variation. The 14 QTLs distributed in 10 groups, among which 4 on A sub-genome, 5 on D sub-genome and the other 5 in 3 unassigned linkage groups. Eight QTLs were detected for fiber-related traits distributing in 5 groups and explained 8.14%~30.05% of the trait variation. Two QTLs located on A sub-genome, 3 on D sub-genome and the other 3 in 2 unassigned linkage groups.As for the low coverage of the linkage map, one-way analysis of variance was conducted for the other mapped markers and unmapped markers for potential markers. One marker was detected for boll number/plant, 2 for seed cotton weight/boll, 2 for lint weight/boll, 2 for seed index, 1 for lint percentage and 2 for lint index. For fiber-related traits, 4 markers were detected for fiber length, 1 for fiber uniformity ratio, 3 for fiber strength, 1 for fiber elongation and 2 for macronaire value.Corelation analysis showed that most of the traits were corelated, so that corelated traits maybe controlled by the same QTLs. Some QTLs were detected on the same position on LG1, LG6, LG17 and LG18. The QTLs in LG1, LG6, LG17 and LG18 controlled seed cotton weight/boll and lint weight/boll simultaneously; the QTLs in LG6 controlled fiber strength and macronaire value simultaneously. One-way analysis of variance also showed that some markers were associated with multi-traits. MGHES-75 associated with cotton weight/boll, lint weight/boll, lint index and macronaire; BNL3875 associated with boll nimber/plant and fiber strength; Me5Em 1-900 associated with cotton weight/boll, seed index and fiber uniformity ratio; BNL3932 associated with fiber length and strength; Me8Eml3-380 associated with fiber elongation and macronaire value.