【作者】 钱能；

【导师】 张天真；

【作者基本信息】 南京农业大学 ， 遗传学， 2009， 博士

【摘要】 棉花的种质资源十分广泛,这些种质具有丰富的遗传多样性。对陆地棉基础种质及其主要衍生品种遗传多样性进行研究,可以从育种起源上揭示我国棉花的遗传基础,了解现有种质的遗传背景和遗传多样性的状况,为育种家高效发掘种质资源重要性状基因打下基础。遗传学研究的一个基本目标即是将生物的表现型与基因型相联系.关联分析以群体结构非固定的自然群体为研究对象,以长期重组后保留下来的基因间的连锁不平衡为基础,在获得群体表型数据与基因型数据后,采用统计方法检测遗传多样性与性状可遗传变异之间的关联,具有简单、高效和省时等优点。本文在分析所选81个陆地棉品种遗传多样性的基础上,利用TASSEL软件分析棉花基因组内的连锁不平衡水平,之后利用关联分析的方法结合多年多点田间调查数据,进行表型变异对标记变异的回归分析,检测数量性状相关位点。同时利用四个衍生系的原始亲本构建作图群体分析发掘与农艺及品质性状相关的数量性状基因位点。对我国自育品种和国外引进种质进行性状差异的综合比较,可以看出在农艺性状方面,国内品种的单株铃数增加、衣分提高和籽指降低,同时单株铃数、衣分和籽指的变异系数有较显著的提高；在纤维品质的差异上,国内品种的各项指标均较引进品种有所改善,同时强度、马克隆值和伸长率的变异系数均有了提高。这说明半个多世纪以来,我国棉花育种有了较快的发展。利用分子标记技术分析所选国内外棉花品种的差异时,国内品种中检测的等位变异数目和基因多样性的跨度较大,但从平均值上看仍然偏低,说明我国自育棉花品种在基因组水平上的差异比较丰富,但是应用到生产中的变异较少,拓宽棉花遗传基础的工作仍相当艰巨。对国内长江流域和黄河流域两大棉花主产区品种的表型性状差异进行分析,结果表明,单就性状表现来说,长江流域的品种要好于黄河流域,如单株铃数较多、单铃重大、籽指较小、衣指和衣分高,但是长江流域品种的变异系数普遍较黄河流域的低。利用分子标记研究两大产区的棉花品种,两流域品种的等位变异数、基因多样性和多态性信息含量均相当。对不同系谱衍生系的多样性分析,可以看到金字棉衍生系在果枝数、株高、单铃重等产量性状上较其他衍生系的均值高,纤维长度、纤维强度和整齐度等品质性状上均值最高,且其在株高、单株铃数、单铃重和马克隆值等性状上的变异系数最大。福字棉衍生系在衣指和衣分性状上均值最大,变异系数亦为最大。岱字棉和斯字棉曾被我国育种家广泛应用于育种工作,并培育了较多的品种,检测的品种也较多,因此这两个系统的衍生系所具有的等位变异数要高于其他系统来源的棉种,但等位变异的比例较低,表现为基因多样性和多态性信息含量较低.利用关联分析考察多个性状大多数QTL的关联位点及其等位变异,首先需要了解棉花基因组内的连锁不平衡(LD)情况。用TASSEL软件分析可知,棉花基因组内位点间的LD水平较低,常异花的授粉方式和品种培育过程中频繁的杂交选择是主要原因。通过关联分析,检测到大量标记位点与农艺品质性状的相关性,并发掘出这些标记位点的优异等位变异,以及含有该等位变异的棉花品种,为棉花育种工作的开展提供有益信息；同时本研究也为开展下一步更加深入的工作打下了基础。斯字棉2B、福字棉6号、岱字棉15和中棉所7号均是我国棉花品种培育的核心亲本,用这四个材料构建一个四交作图群体及其F2：3家系,利用SSR标记和JoinMapV3.0软件构建了一张陆地棉四交群体品种间的分子标记遗传图谱。该图谱图谱总长1129.1cM、含有102个多态位点,覆盖率为22.6%,两多态位点间的平均遗传距离为11.1cM。由31个连锁群组成,其中30个连锁群被定位到15条染色体。利用MATQTLV5.0定位了19个与农艺性状相关的QTLs和5个纤维品质QTLs.其中1个衣分QTL可在两世代中检测到；另有一个果枝数QTL和一个衣指QTL的定位结果与关联分析结果相吻合,均有较好的稳定性。利用海岛棉遗传标准系3-79和优良的细胞遗传学材料——置换系18,结合本实验室构建的饱和棉花遗传图谱,完成开放花蕾隐性基因ob2在四倍体棉第13染染色体色体(A13)上的精细定位,将ob2基因定位于该染色体上BNL2449b和TMP01两位点之间,与它们的遗传距离分别是1.1cM和1.7cM。此前本实验室已经利用TM-1和Sub18两个材料配置的作图群体,完成了开放花蕾ob1基因在四倍体棉第18染色体(D13)上的定位。结合两实验结果,验证了早期学者关于开放花蕾基因(ob)的遗传特性的研究,同时也为进一步研究并利用该基因打下了坚实的基础。两张连锁图之间有着共同的标记位点BNL2652、NAU817、NAU1141和BNL2571,这也从基因组学的角度再次验证了四倍体棉的染色体组具有部分同源结构,来自A亚组的第13染色体(A13)和来自D亚组的第18染色体(D13)是一对部分同源染色体,两染色体具有相近的功能区域。更多还原

【Abstract】 The germplasm resources of cotton have abundant genetic diversity. And genetic diversities are the basis of breeding excellent varieties. The research of genetic diversity about upland cotton and its offspring lines can reveal the genetic base of cotton in China on breeding origin, and help us to know the background and diversity of germplasm nowadays in use. And the research also can help breeders to mine important traits and use these good germplasm in high efficiency, and have great meaning of creating new germplasm. The goal of genetic research is to know the relationship between phonotype and genotype. Association mapping use natural population with unfixed population structure for research target, with the phonotype and genotype data obtained, we test the association between genetic diversity and heritable variation of traits use statistic method.In this research, on the base of analyzing the genetic diversity of 81 upland cottons, we used software TASSEL analyzed the LD level in cotton genome. And then combined with the traits and the markers data, we detected the quantity traits loci. We also constructed a four-way cross population to detect the QTLs concerned with yield and quality traits.In this paper, we compared the difference of traits between varieties domestic and introduced. We found that the domestic varieties have better representation in agronomic traits, such as the increase of NB (number of bolls per plant), LP (lint percentage), and the decrease of SI (seed index). And also, the variation coefficients increased in NB, LP, and SI. Domestic varieties also have better representation in many aspects of fiber quality, and the variation coefficients increased in FS (fiber strength), FM (fiber micronaire reading), and FE (fiber elongation). It shows that, for more than half a century, cotton breeding in China has developed quickly. When use molecular marker technique to analyze the difference between domestic and overseas varieties, we found that, domestic varieties have more alleles and big span of gene diversity, but the mean value is still low, indicate that though domestic varieties have many variation in genome level, but the variation is less applied to production, it’s still difficult to broaden the genetic basis of cotton.When analyzing the phenotypic difference of varieties between Yangtze River and Yellow River Basin, two main producing areas of cotton in our country. We found, varieties from Yangtze River is better if solely on the traits performance, such as more bolls per plant, big bolls, small seed index, high lint percentage and lint index. But the variation coefficients are lower. When analyze with molecular marker technique, the number of alleles, genetic diversity index and PIC value have no significant difference.When analysis the difference between varieties based on the pedigree, we see that the Kings have high mean value in plant branch, plant high, boll weight, fiber length, fiber strength and fiber uniformity ratio, and biggest variation coefficients in number of bolls per plant, boll weight and fiber micronaire reading. The Fosters have the biggest mean value and variation coefficients in lint index and lint percentage. The DPLs and Stonevilles had been used as parents in breeding wildly, and have bred many varieties. So the two pedigree sources have more alleles than others, but low in gene diversity and PIC value.In our research, in order to use association mapping to test association sites and alleles of several traits, we need to know the LD in cotton genome. With the help of software TASSEL, we see low level of LD in cotton genome, cross-pollination and frequent hybridization and selection in breeding is the main reason. We detected numerous associations between marker sites and agronomic or quality traits. After that, we tried to find out alleles with good variation of different markers, and the typical variety with these alleles. It will provide useful messages for cotton breeding, and for further research.Stoneville 2B, Foster 6, DPL 15 and CRI7 are core parents in cotton breeding in China. We constructed a four-way cross segregated population and its F2:3 inbreed lines with these varieties. A linkage map was developed for the four-way cross with SSR markers and JoinMap V3.0 software. The map covered 1129.1cM, and with 102 mapped loci, which was approximately 22.6% of the total recombination length of the cotton genome. The average distance between loci was 11.1cM. The map comprised 31 linkage groups, and these groups were assigned to 15 chromosomes. In the four-way cross population,19 QTLs concerned with agronomic traits and 5 QTLs with fiber quality were detected with MapQTL V5.0. And a QTL concerned with LP (lint percentage,%) was detected both in F2 and F2:3 generations, and is of value for MAS (Marker-assisted selection).In this paper, we used the G. barbadense genetic standard,3-79, and the good cytogenetics material, Sub.18. Combined with cotton genetic map constructed by our lab, we achieved to fine mapping the ob2 gene in chromosome 13 (A13) of tetraploid cotton, located this gene between two loci, BNL2449b and TMP01, the genetic distance is 1.1 cM and 1.7cM, separately. Former research in our lab had finished the mapping of the ob1 gene with the standard Upland stock, TM-1 and Sub.18. It locates in chromosome 18 (D13). Consider the result of two experiments together, we confirmed the result of former research about the hereditary feature of ob gene, and in the same time, laid a solid foundation of further researching and using this gene. BNL2571, BNL2652, NAU817, and NAU1141 also produced duplicate loci bridging of the homoeologous A13/D13 (At/Dt) chromosome pairs. Therefore, the expression of the ob gene in tetraploid cotton is a typical phenomenon of genome duplication.更多还原