【作者】 吴迅；

【导师】 黄玉碧； 黎裕；

【作者基本信息】 四川农业大学 ， 作物遗传育种， 2013， 博士

【摘要】 对玉米种质资源进行精细遗传评价不仅能为揭示重要自交系形成的遗传基础提供理论支撑,还是等位基因发掘、复杂性状关联分析的重要基础。玉米株型性状作为重要的产量影响因子,在抗倒性、光能利用效率以及增加籽粒产量等方面表现出丰富的遗传多样性。探讨具有理想株型性状的玉米种质中所蕴藏的优异等位基因以及株型性状较差的玉米种质中隐蔽的等位基因是发掘控制理想株型性状QTL位点以及基因克隆的必要研究工作,不仅能够揭示理想株型形成的遗传基础,还能为育种实践中的株型性状改良提供分子遗传学依据。本研究在前期构建的多样性种质群体基础上,筛选出一套包含367份重要自交系的关联作图群体。以该群体为试验材料,在表型水平上,对不同玉米生态区、不同年份的株型相关性状进行鉴定：同时在分子水平上利用包含56110个SNP标记的高密度玉米芯片对其遗传多样性、群体结构、亲缘关系和连锁不平衡(LD)进行分析。此外,本研究还对群体中所包含的43份黄早四衍生系进行遗传组分比较。最后,基于全基因组扫描的关联分析策略定位到158个与株型相关性状显著关联的SNP位点,其中97个位于基因内,其余SNP位点则位于基因间隔区域。主要研究结果如下：1.基于56110个SNP标记的遗传多样性分析共检测到83638个等位变异,平均基因多样性为0.364,平均PIC为0.291。群体结构分析表明,367份重要自交系群体首先被划分成2个亚群,分别对应于中国玉米育种中的外引种质和本地种质。深入分析发现,以遗传相似性比例≥50%为标准,该群体被进一步细分为5个亚群,分别对应于中国玉米育种中的5大杂种优势类群：瑞德、兰卡斯特、唐四平头、温热Ⅰ和P群,各亚群在所有材料中所占的比例分别为8.45%、8.99%、9.54%、42.23%和5.18%。另外,还有94份自交系与任何亚群的遗传相似性比例均<50%,因此将这些材料划分为一个亚群,称为混合亚群,占总材料的25.61%。亚群遗传多样性分析显示,温热Ⅰ亚群的遗传多样性最高,其余依次是兰卡斯特、瑞德、唐四平头,而P群的遗传多样性水平最低。亲缘关系分析结果显示,94.97%的配对亲缘关系系数分布于0.05-0.28之间；0.17%的配对亲缘关系系数等于0；其余配对亲缘关系系数则分布于0.30-0.50之间,说明该群体内个体之间存在着中等程度的亲缘关系。黄改系遗传组分比较显示,黄早四的遗传组分在其衍生系785中所占比例最低为1.35%,在72-125中所占比例最高为93.79%,变异幅度较大。共发现15个黄早四的特征区段,分别位于染色体1,2,3,4,5,6,8和10上,这些区段在超过60%的黄早四衍生系中表现出一致性,揭示了骨干自交系黄早四极其衍生系的遗传相似性,为其重要表型性状利用和改良提供了很好的分子遗传学依据。LD评价结果表明,367份重要自交系群体的平均LD衰减距离为74.08kb,不同连锁群的LD存在显著差异,其中Chr1的LD衰减距离最短为48.31kb, Chr10的LD衰减距离最长为183.04kb。另外,不同连锁群的LD受群体大小和群体组成的影响程度也各不相同。当群体样本量减小时,LD衰减距离会增大；群体遗传组成复杂会导致遗传多样性增加从而降低LD衰减距离；不同连锁群上或同一连锁群上不同遗传区域间也表现出LD水平的显著差异。2.方差分析显示,株型相关性状在不同自交系间存在极显著差异,其中株高的变异系数为43.83%,雄穗一级分支数和穗位高的变异系数分别为34.59%和22.83%,雄穗主轴长的变异系数最小为11.87%,其次是“穗位高/株高”,变异系数为14.69%。另外,5个株型相关性状的Shannon-Weaver多样性指数均在2以上,揭示了本研究群体丰富的表型多样性。3.全基因组关联分析结果显示,在P<0.0001水平下,共检测到28个与株高性状显著相关的SNP位点,其中6个在2个环境条件下均被检测到；3个位点同时在3个环境中被检测到。在50个与穗位高显著关联的SNP位点中,10个在两个环境条件下被检测到；1个位点在3个环境条件下被检测到；3个位点在4个环境条件下均被检测到,分别为PZE-105098995、SYN31958和PZE-105099028,均位于第5染色体上；另外一个穗位高显著相关的SNP位点(PZE-104109619)在5个环境条件下均被检测到,位于第4染色体上；其余穗位高相关的SNP位点则仅在单个环境条件下被检测到。在34个与“穗位高/株高”显著关联的SNP位点中,4个在2个环境条件下被检测出来；8个位点在3个环境条件下被检测到；2个位点在4个环境条件下均被检测出来,分别为PZE-105090603和PZE-105090633,均位于第5染色体上；说明这些区域可能与控制“穗位高/株高”的基因位于相近区域；其余“穗位高/株高”相关的SNP位点则仅在单个环境条件下被检测到。14个与雄穗长显著关联的SNP位点以及32个与雄穗分支数显著关联的SNP位点均仅在单环境条件下被检测到。BlastN比对显示,在株型性状显著相关的SNP中,部分位点位于已知基因所在区域,如Rht基因所在区域共发现3个株型显著关联的SNP位点,其中PZE-101080319在两个环境下均与株高和穗位高显著关联；PZE-101137671和SYN2469则分别与雄穗主轴长显著关联。sdl基因所在区域发现PZE-102120220与穗位高显著关联。除此之外,大量与株型性状显著关联的SNP位点位于一些预测基因内部或基因间隔区域,说明了这些基因或基因间隔区域可能在玉米株型相关性状的进化、改良等过程中起一定作用,关于这些位点的连锁验证仍在继续中。更多还原

【Abstract】 Characterization of maize germplasm is an important work for founding allelomorphic gene, association analysis of complex traits, and breeding practice. Plant architecture trait is one of the main factors in maize yield formation. Plant architecture trait showes higher variantation and the ideal plant architecture will improve maize lodging resistance, photosynthesis, grain yield and so on. It is a primary work before QTL analysis and gene clone to found excellent allelomorphic gene included in maize germplasm with good plant architecture trait, which can not only for uncovering the genetic basis of plant architecture but also supplling molecular genetics proofs for plant architecture improving.Based on the larger and diverser collection of maize germplasm collected in our previous studies, a subset of367main inbred lines was constructed. Which was planted under different envirments among two years, five plant architecture related traits were investigated. On the molecular level, genetic diversity, population structure, pairvise kinship, and linkage disequilibrium (LD) were evaluated using high throughput’MaizeSNP50’including56110SNPs. Gentic comparing was done among43inbred lines derived from Huangzaosi. And158SNPs associated with plant architecture related traits were found. Main results are listed as follow:1. A total of41,819informative SNPs with minor allele number (MAF) of more than0.05were used to estimate the genetic diversity and relatedness, and16,827SNPs with MAF>0.1wore selected to estimate the LD decay. Totally1015SNPs which were evenly distributed in the genome were selected randomly to evaluate the population structure of these accessions. The results showed that two groups could be classified in the accessions, i.e. the introduced germplasm and the local germplasm. Further, five subgroups corresponding to different heterotic groups, that is, Reid, Lancaster, P, TSPT, and Tem-tropic I, were clustered. By using the re-sampling method, the genetic diversity of each subgroups was estimated, with the highest in the Tem-Tropic I and the lowest in the P. Most lines in this panel showed weak or modest relatedness with each other. Comparisons of gene diversity (GD) showed that there existed some conserved genetic regions for specific subgroup across10chromosomes of maize, i.e. seven in Lancaster, seven in Reid, six in TSPT, five in P, and two in Tem-Tropical I. Additionally, the results also revealed that there existed fifteen conservative regions transmitted from Huangzaosi, an important foundation parent, to its descendants, which are important for further studies since the outcomes may provide clues to understand why Huangzaosi could become a foundation parent in Chinese maize breeding. For the entire set, average LD distance was74.08kb and varied among different chromosomes as well as in different genomic regions of one chromosome. This analysis uncovered a rich natural genetic diversity of the elite maize inbred set, suggesting that the panel can be used in association study, esp. for temperate regions.2. Among this subset of367inbred lines, ANOVA analysis showed that there was significant deviation among5plant architecture related traits between any two inbred lines. The coefficient variation ranged from22.83%to11.87%. Plant height showed the highest coefficient variation of43.83%. The second two were Tassel primary branch number and Ear height with coefficient variation of34.59%and22.83%, respectively. The smallest coefficient variation was found in phenotype of Tassel length with11.87%, coefficient variation of "Ear height/Plant height" was14.69%. Otherwise, all the Shannon-Weaver indexes of5plant architecture related traits were beyond2. These results suggested that this subset used here were diverser in phehotypes.3.158SNPs were significant association with plant architecture related traits with P<0.0001.28SNPs were significant association with plant height, of which, six SNPs were found under two envirments at the same time, three SNPs were found under three envirments at the same time.50SNPs were significant association with ear height, of which, ten SNPs were found under two envirments at the same time, one SNPs were found under three envirments at the same time, three SNPs named PZE-105098995, SYN31958and PZE-105099028were found under four envirments at the same time, another important ear related SNP named PZE-104109619is on Chrom4, which was found under five envirments at the same time, the other ear height SNPs were found under one envirment only.34SNPs were significant association with "ear height/plant height", of which,4SNPs were found under two envirments at the same time, Eight SNPs were found under three envirments at the same time, two important SNPs named PZE-105090603and PZE-105090633, which were found under four envirments at the same time.14tassel length related SNPs and32tassel primary branch number related SNPs were found under one envirment only. These results revealed that genetic regions including these markers were close to genetic regions containing plant architecture related genes.BlastN search showed, among the158plant architecture related SNPs, some were in the regions with known plant architecture related genes. For example,3SNPs named PZE-101080319, PZE-101137671and SYN2469were close to the region with Rht gene which was related with plant height. PZE-102120220was close to the region with sdl gene which was related with plant architecture related traits. Except that, most significant association with plant architecture were found in the predication genes or intergenic regions, these variation may play important role in maize domestication and plant architecture formation.更多还原