【作者】 李骏智；

【导师】 蒋建雄； 袁有禄；

【作者基本信息】 湖南农业大学 ， 遗传学， 2009， 硕士

【摘要】 棉花生产在我国国民经济中起着重要的作用。棉属包括46个二倍体种和5个四倍体种,其中四倍体陆地棉（Gosspium hirsutum L.）和海岛棉（Gossypium barbadense L.）分别占世界棉花产量的90%和8%左右,二倍体亚洲棉（Gossypium arboreum L.）和草棉（Gossypium herbaceum L.）共占2%左右。陆地棉产量高,适应广,而海岛棉品质优,产量低。结合陆地棉和海岛棉的优良性状,是当前棉花品种改良的方法之一。本研究利用优异的陆地棉栽培品种中棉所36和海岛棉品系海1杂交,中棉所36为轮回亲本,构建含BC₁F₁、BC₂F₁和BC₁S₁三个世代的回交群体,利用BC₁F₁分离群体以SSR标记构建连锁图谱,采用复合区间作图方法（CIM）对产量性状和纤维品质性状进行QTL定位研究,以将海岛棉优异基因导入陆地棉优异栽培品种,来拓宽陆地棉狭窄的遗传基础,为品种改良提供更丰富的资源材料。1.用亲本和F₁对新开发的2102对SSR引物进行多态性筛选,共筛选到317对含有海1显性带的引物,占筛选引物总数的15.08%；最终利用其中的275对引物对BC₁F₁群体进行扩增,获得306个SSR标记差异位点。连锁分析表明（LOD=6.5）,有254个标记位点连锁,分布在42个连锁群中,覆盖2252.36 cM,约占棉花基因组的50.05%；平均每个连锁群有6.08个标记,覆盖53.63 cM；标记间平均间距为8.87 cM。2.利用BC₁F₁、BC₂F₁和BC₁S₁三个不同世代分离群体产量性状数据,共定位16个产量性状QTL,解释表型变异5.77%～19.86%。其中,衣分6个,铃重6个,籽指4个。有9个增效基因来自陆地棉亲本中棉所36,7个增效基因来自海岛棉亲本海1,说明了表型性状较差的品种同样可能含有可用于性状改良的增效基因。控制衣分的3个QTL可在不同的世代稳定检测到,效应稳定,增效基因均来自高值亲本陆地棉,为进一步分子标记辅助选择奠定了基础。3.依据BC₁F₁、BC₂F₁和BC₁S₁三个不同世代的纤维品质数据,共定位纤维品质5个性状的31个QTL。其中,长度5个QTL,强度4个QTL,麦克隆值8个QTL,整齐度8个QTL,伸长率6个QTL,每个QTL分别解释6.43%-18.84%的表型变异。有3个纤维长度和1个与纤维强度的QTL可在两个不同世代中稳定检测到。更多还原

【Abstract】 Cotton has been playing an important role in the economy of China. Cotton genus comprises of about 46 diploid and 5 tetraploid species, two tetraploid species, G. hirsutum L. and G. barbadense L., accounted for 90 and 8%, two diploid species, G. herbaceum L. and G. arboreum L., totally accounted for 2%, respectively, of the world’s cotton production.The yield of G hirsutum L. is higher than that of G barbadense L., but the fiber quality of G barbadense L. is better than the G. hirsutum L. It is one of the important ways to Combine the fine traits of G hirsutum L. and G. barbadense L for improving cotton varieties. In order to introgress elite QTL alleles of G barbadense L. for fiber yield and quality into G. hirsutum L. and enlarge the narrow genetic base of G. hirsutum L. and provide more new germplasm resources for the variety development, the commercial G. hirsutum L. variety’CCRI36’and G. barbadense L.’Hai 1’were respectively used as recurrent and donor parent to produce backcross populations, including BC1F1, BC1S1 and BC2F1. QTL analysis of yield-related trait and fiber quality on three phenotypic data sets were proceeded by the Composite Interval Mapping method.1.A total of 2102 pairs of SSR primer were used to screen polymorphism among the parents CCRI36, Hai 1 and their F1, which resulted in 317 pairs of polymorphic primer with the special band for Hai 1, and the polymorphism primer accounted for 15.08% of the total primer, and 275 of them were used to screen the BC1F1 population. Linkage test (LOD=6.5) indicated 254 of 306 polymorphic locus derived from 275 pairs of primer could be mapped into 42 linkage groups and covered a total genetic distance of 2252.36cM, approximately 50.05% of cotton genome. The number of the marker locus per linkage group was 6.08, which spanned 53.63cM of the cotton genome. And the average distance of the neighboring marker locus were 8.87cM.2.16 QTLs for yield-related traits were detected from the three generations of BC1F1, BC2F1, and BC1S1 using CIM. QTL numbers for lint percentage, boll weight and seed index were 6,6 and 4, respectively, which accounted for 5.77%～19.86% of phenotypic variance. Among the 16 QTLs,9 with positive additive effect derived from CCRI36 and 7 QTLs with negative additive effect from Hail, which indicated the materials with poor phenotypic characteristics may contain the useful gene for improving the same traits.3 QTLs for lint percentage could be stably detected in the different generations, and from high-value parent of CCRI36, which should be useful for the marker-assisted selection.3.31 QTLs for fiber quality traits were mapped among the three generations. QTL numbers for fiber length, strength, micronaire, uniformity and elongation ratio were 5, 4,8,8,6, respectively, which could explain 6.43%～18.84% of phenotypic variance. Three QTLs for fiber length and one QTL for fiber strength could be detected in the two generations.更多还原