【作者】 马金亮；

【导师】 吴建宇；

【作者基本信息】 河南农业大学 ， 作物遗传育种， 2010， 硕士

【摘要】 玉米重要性状QTL定位是分子标记辅助选择的前提条件,对于提高育种效率有重要意义。本研究以当前大面积推广的一个优良玉米杂交种郑单958的两个亲本（郑58×昌7-2）构建含有225个家系的F_2:3群体为基础材料,构建了SSR分子标记遗传连锁图谱,并对产量和相关性状进行了QTL作图。在5个环境条件下（2007年在郑州、济源、西昌,2008年在郑州、济源）进行田间试验,利用复合区间作图法对穗长、穗粗、穗行数、行粒数、轴粗、单穗重、穗粒重、百粒重9个产量相关性状和抽雄期、吐丝期、散粉期3个生育期性状进行QTL定位和效应分析;利用基于混合线性模型的复合区间作图法,分析产量相关性状和生育期性状的上位性及与环境互作效应。主要结论如下:1.以225个F₂单株为作图群体,利用180个多态性SSR标记,构建了覆盖玉米全基因组的分子标记遗传连锁图谱,覆盖玉米10条染色体,图谱总长度1 978.7 cM,平均间距为11.0 cM。同IBM群体连锁图谱相比,共同SSR标记在连锁图谱位置和顺序基本一致。在P<0.05和P<0.01水平上观察到22个（12.22%）分子标记发生了偏分离。个体基因型组成来源于母本郑58的纯合染色体片段在8.98%-33.78%之间,平均23.47%;基因型组成中来源于父本昌7-2的纯合染色体片段在10.67%-34.67%之间,平均23.53%。双亲的杂合片段在20.00%-57.78%,平均为47.71%。整个群体中,双亲染色体同源片段的基本分离符合1:2:1的理论比例,该F_2:3群体是一个随机群体,为下一步QTL定位奠定基础。2.在5个环境条件下,对亲本和225个F_2:3家系群体的9个产量相关性状和3个生育期性状进行调查。对亲本自交系及F_2:3群体的考察性状表型分析,亲本自交系在穗粗、穗行数、轴粗和出籽率性状的均值昌7-2明显大于郑58;而穗长上郑58要明显大于昌7-2;在行粒数上昌7-2和郑58差异不大;F1表现明显的超亲优势;F_2:3群体各个性状在所有环境下均表现不同程度的双向超亲分离,呈连续性正态分布。相关性分析表明:产量相关性状间和生育期性状间均存在显著或极显著相关性;同时进行方差分析:所有性状在家系间、环境间、家系和环境互作均呈现显著或极显著的差异。3. 5个环境条件下及联合分析共检测到111个与9个产量相关性状的QTL,单个QTL的贡献率为3.39%-31.15%,14个QTL的贡献率大于10%,其中穗长QTL qEL3-1和qEL4、穗粗QTL qED3、穗行数QTL qRN3、行粒数QTL qKR5、轴粗QTL qCD5-1和百粒重QTL qHKW1-1在5种环境条件下及合并分析均被检测到,表型贡献率较大且环境稳定性好,可以作为进一步精细定位和分子标记辅助选择的主要目标QTL。此外,还发现在第1染色体umc1703-umc1590、第3染色体bnlg1325-umc2369和bnlg420-umc1209、第5染色体的umc1800-umc2304和第6染色体的umc1257-phi031标记区间检测到多个产量及构成因子的QTL,表现为QTL的关键区段分布。部分显性和加性是产量相关性状的主要遗传基础。4.对9个产量及其构成因子分别进行上位性及与环境互作分析,没有检测到影响轴粗的上位性QTL,共检测到20对上位性QTL,存在加性×加性、加性×显性、显性×显性效应的互作,贡献率为0.04%-2.41%。说明上位性效应对产量及相关性状的形成也有一定作用。其中穗粒重的2对上位性QTL和出籽率的2对上位性QTL均与环境互作效应显著,贡献率0.33%-0.91%。大多数上位性QTL在单个位点主效应不显著,本身对性状表达没有效应,但通过位点间的互作影响性状的表现。5. 5个环境条件下及合并分析共检测到52个与3个生育期性状相关的QTL,单个QTL的贡献率为4.28%-19.59%,其中17个QTL的贡献率大于10%,其中抽雄期QTL qTE5-2、散粉期QTL qAN1-2和吐丝期QTL qSE3-1在4种环境条件及合并分析下均被检测到,表型贡献率较大且环境稳定性好,这些主效QTL可以直接用于分子标记辅助选择。在第1染色体1.05-1.07区域（qTE1-1、qTE1-2、qAN1-2、qAN1-3、qAN1-4、qSE1-2、qSE1-3）、第3染色体3.05-3.06区域（qTE3-1、qAN3-1、qSE3-1）、第5染色体5.08区域（qTE5-1、qAN5-6、qSE5-2）,发现影响生育期性状的多效性QTL区域,抽雄期、散粉期、吐丝期每一个性状的变异都会对生育期产生影响。6.对3个生育期性状进行上位性及与环境互作分析,没有检测到影响散粉期的上位性QTL,其它2个性状共检测到4对上位性QTL,存在加性×加性、加性×显性、显性×显性效应的互作,贡献率为0.02%-0.13%。更多还原

【Abstract】 QTL mapping for important agronomic traits in maize is basic work to marker assistant selection （MAS）. In this study, the F_2:3 population derived from an elite hybrid Zhengdan958 （Zheng58×Chang7-2） was used to detect the inheritance of yields and their related traits in maize. The partants, F1 and the F_2:3 population were evaluated in five environment. QTL mapping for nine yield traits and three flowering related traits, including Ear Length, Ear diameter, Row Number, Kernels per Row, Cod Diameter, Ear Weight, Kernel Weight per Ear, Rate of Kernel Production, 500 Kernel Weight, and days to tasseling, days to silking, days to anthesis were firstly analyzed by Composite Interval Mapping （CIM） method. Also, using Network 2.0 software with Mixed Linear Model （MCIM）, main genetic effect including epistasis and epistatic×environment interactions were analyze. Compared with the single locus QTL and the interaction effects of QTL-by-environment （QE）, These results will do great help in fine mapping QTL associated with yield characters and in marker-assisted selection in maize breeding. The main results were summarized as following:1.A genetic linkage map containing 180 SSR polymorphic markers was constructed using 225 F₂ population derieved from the F1 （Zheng58×Chang7-2） , which spanned a total of 1 978.7 cM with an average space between two makers of 11.0 cM. Compared with the IBM map from MaizeGDB website, the order of the markers on ten chromosomes were similar. Among these SSR markers, 22 （12.22%） markers showed the genetic segregation distortion （P<0.05 and P<0.01）. The molecular genotypes deriving from parent Zheng58 were 8.98%-33.78%, the average homozygous genotypes of Zheng58 was 23.47%; the molecular genotypes of Chang 7-2 were 10.67%-34.67%, and the average genotypes of Chang 7-2 was 23.53%; the molecular genotypes of F1 were 20.00%-57.78%, and the average genotypes of F1 was 47.71%. The genotypes of the two parents at the marker loci followed 1:2:1 theoretical ratio, so the F_2:3 population was a random one and was fit for QTL analysis.2.Under five environment, Nine yield traits and three flowering traits were evaluated among family lines as well as the two parents. Transgressive segregation was observed for all traits in F_2:3 population, Normal distribution was observed for all traits; The yield relate trait and flowering related traits were significant correlation; the results of analysis of variance （ANOVA） showed significant difference with the lines, the environments, the lines and environment interaction in most of the traits. 3.111 QTLs were detected for 9 yield related traits under five environments. Contribution of single QTL to phenotypic variation varied from 3.39% to 31.15%, there were 14 QTLs showing more than 10% of phenotypic variation; 7 major QTLs for EL （qEL3-1 and qEL4）, ED （qED3）, RN （qRN3）, KR （qKR5）, CD （qCD5-1） and HKW （qHKW1-1） were common detected under five different environments and in the combined analysis, with higher contributions and stability, and could be used as stable QTL for further fine mapping and MAS. Moreover, Many QTLs were detected with the umc1703-umc1590, bnlg1325-umc2369 and bnlg420-umc1209, umc1800-umc2304, umc1257-phi031 marker confidence intervals on chromosome 1, 3, 5 and 6, respectively. Partially dominance and additive effect played an important part in the inheritance of yield-related traits.4.Digenic interactions were also detected for yield and the related traits. Twenty pairs of epistasis QTLs were identified in most of the traits except Cod Diameter, all epistasis QTLs have additive×additive, additive×dominance and dditive×dominance effect, the contribution of epistasis varied from 0.04% to 2.41%. Among the Digenic interaction effects, two pairs of interaction for kernel weight and kernel product percent, respectively, showed significant interaction effects with environments, which account for 0.33%-0.91% phenotype variation. Single locus among most of epistatic QTL was not significant, which had little effects on trait variation. It is important to consider different genotypes and the environment in the breeding.5.52 QTLs were detected for 3 flowering related traits under five environments. Contribution of single QTL to phenotypic variation varied from 4.28% to 19.59%, 17 QTLs explained more than 10% of phenotype variation; There major QTLs for TE （qTE5-2）, AN （qAN1-2） and SE （qSE3-1） were stable under four different environments and in combined analysis. Moreover, Many pleitropic QTLs were also detected at 1.05-1.07 bin（qTE1-1、qTE1-2、qAN1-2、qAN1-3、qAN1-4、qSE1-2、qSE1-3）, 3.05-3.06 bin（qTE3-1、qAN3-1、qSE3-1）and 5.08 bin（qTE5-1、qAN5-6、qSE5-2）.6.Digenic interactions were detected for flowering related traits. Totally four pairs of epistasis QTLs were detected for days to tasseling and days to silking, and not detected epistatic QTL for days to anthesis, the type of epistasis QTL included additive×additive, additive×dominance and dominance×dominance effects, the contribution of epistasis varied from 0.02% to 0.13%.更多还原