玉米代表性品种籽粒代谢物的数量遗传学解析

Maize varieties in China have been renewed several times, and it is of vital importance for variety improvement to explore the genetic mechanism generated by representative varieties formed in different periods. Regarded as typical quantitative traits, metabolites are bridges connecting genotypes and phenotypes. Although metabolites have been applied to functional gene detection and phenotype prediction, the genetic model of metabolites has not been available. In this project, we plan to use nine parental inbred lines of five representative maize varieties including Zhongdan2, Danyu13, Yedan13, Zhengdan958 and Xianyu335, and 72 hybrids derived from the complete diallel cross among nine inbred lines as materials. The untargeted LC-MS metabolic profiling technology will be applied to detect metabolites. Quantitative genetic approach will be used to analyze the metabolites with yield and quality traits. The objective is to study the relationship between the various hybrids and their parents in metabolites; to detect the cytoplasm effect, the heterosis and the difference of combining ability; to identify the genetic model of metabolites; to detect the potential metabolite biomarker related with yield and quality traits using our proposed LASSO method that has the ability to perform statistical tests. Results will not only help find out the inheritance of formation of elite varieties at the metabolic level but also provide a novel idea for genetic improvement of maize.

我国玉米品种已经历了多次更新换代，了解不同时期代表性品种的遗传机制对于玉米品种改良具有重要意义。代谢物既是衔接基因型和表型的桥梁，又具备数量性状的特征，已被广泛用于功能基因挖掘和表型预测，然而，至今对代谢物自身的遗传模型尚少有研究。本项目拟以玉米代表性品种中单2号、丹玉13、掖单13、郑单958和先玉335的9个亲本自交系为材料，采用完全双列杂交设计配制正反杂交种，基于LC-MS的非靶向技术测定亲本和正反杂交种籽粒中的代谢物含量，结合它们的产量性状和品质性状，应用数量遗传学分析方法，研究杂交种和其亲本在代谢物含量上的关系，检测代谢物的细胞质效应、杂种优势以及配合力差异，鉴定各代谢物含量的遗传模型，并利用前期发展的一种可进行自变量显著性检验的LASSO新方法挖掘与玉米产量及品质相关的代谢标志物。研究结果既有助于从代谢水平解析代表性品种形成的遗传规律，又可为玉米优良品种的遗传改良提供新的思路。

代谢物既是衔接基因型和表型的桥梁，又具备数量性状的特征，已被广泛用于功能基因挖掘和表型预测，但针对代谢物自身的遗传信息尚少有研究。本项目对中单.2号、丹玉13、掖单13、郑单958和先玉335和苏玉16号这6个代表性玉米品种的11份亲本自交系，及其组配的正反交杂交组合的籽粒代谢物进行了遗传分析。利用非靶向代谢物谱技术鉴定得到了玉米籽粒中的208种代谢物。聚类分析和主成分分析均表明，籽粒代谢谱可以有效区分杂种及其亲本。进一步方差分析表明144个代谢物在亲本间差异显著，109个代谢物在杂交种间差异显著，163种代谢物在亲本和杂交种间差异显著，40种代谢物在正反交间差异显著。代谢物的遗传效应和杂种优势分析结果表明，玉米杂交种与其亲本的籽粒代谢物呈现多种关系，在所有杂交种中，约2/3的代谢物表现为超显性效应，其中36.8%表现为正向超显性，31%表现为负向超显性，在所有组合中，有27.5%的组合表现出显著的中亲优势，20.4%的组合表现出显著的超亲优势。利用LASSO方法鉴定了与11个农艺性状相关的25个显著代谢标志物，在这些代谢标志物中，有7个代谢标志物同时与多个性状显著相关。研究结果为进一步在代谢水平进行玉米遗传改良提供了重要基础。在该项目的资助下，发表学术论文9篇，授权软件著作权1个，培养硕士研究生2人，后续研究工作获国家自然科学基金面上项目资助。