【作者】 袁娟；

【导师】 武天龙；

【作者基本信息】 上海交通大学 ， 生物医学工程， 2009， 博士

【摘要】 扁豆(Lablab purpureus (L.)) Sweet是一种古老的作物,其品质优良,营养价值高,用途广泛。它可以用作粮食、蔬菜和草料。在我国南方地区,扁豆主要用作鲜荚蔬菜和草药,具有很大的栽培面积。目前,我国扁豆生产存在产量小、品质低的问题。如何提高扁豆的产量和品质一直困扰着育种工作者。本试验室经过多年的扁豆栽培观察,发现扁豆花序性状是决定扁豆产量的主要因素之一,果实性状是构成扁豆品质和商品性的重要因子,生育期性状影响扁豆产量和品质。因此本课题对扁豆的花序性状、果实性状和生育期性状等农艺性状进行QTL定位以及对花序的分化、发育进行了形态学研究。本试验使用花序等性状差异大的扁豆农家品种‘眉豆2012’ב南汇23’的F2群体构建了一张扁豆遗传连锁图谱。图谱包含131个标记位点,其中122个RAPD分子标记位点,9个形态标记。整个图谱由14个连锁群构成,涵盖1302.4 cM ,平均间距9.9 cM。在F2群体和F3株系中,应用复合区间作图法,重点对花轴长度、花序长度、花序至叶腋间的花梗长度、叶腋到第一朵花的花梗长度、花序两端花间的花梗长度、花序节数、花节间距、初花节位和最低花节9个花序性状进行QTL定位,共检测出49个QTL,分布在10个连锁群上。有些不同性状的QTL位于相同的连锁群区域内,表明控制花序诸性状的基因存在基因多效现象。这些QTL在分子标记辅助育种中将起到重要的作用。运用贝叶斯模型选择法,本试验对F2群体的荚长度、荚宽度、荚厚度、开花期、结荚期、成熟期和两年F3株系的果实和生育期8个农艺性状:荚长度、荚宽度、荚厚度、荚面积、荚体积、开花期、结荚期和成熟期进行了QTL定位,并在F3株系中检测了QTL与环境的互作效应(QEs)。结果表明,当阈值2lnBF设定为2.1时,在F2群体中共检测到21个主效QTL和22对上位性效应。主效QTL的表型变异量在2.8%至9.6%之间变化。在F3株系中,8个性状共检测到30个主效QTL,分布在9个连锁群上,其贡献率均较小。同时,还检测到24对上位性效应,其贡献率比主效QTL和QEs的大。在所有的主效QTL中,有13个主效QTL与环境间存在互作关系。环境对果实性状影响较小,但对生育期性状影响较大。此外,有些检测到的主效QTL和QEs存在一因多效现象。在F2群体中,有两个主效QTL表现出基因多效性。在F3株系中,存在4个多效性QTL和2个多效性QEs。在这些多效性位点中,有2个主效QTL在三个世代中均能检测到。分别是第4连锁群上同时影响开花期和成熟期性状的主效QTL;第6连锁群上同时影响开花期、结荚期和成熟期三个性状的主效QTL。这些多效性基因具有稳定性,可以应用到分子标记辅助育种工作中来提高扁豆的产量和商品品质。为了进一步的研究扁豆花序性状对产量的影响,试验以长花序品种‘南汇23’和短花序品种‘眉豆2012’为材料,采用石蜡切片法观察花序分化和发育进程。结果表明,两品种的花序分化过程基本一致。扁豆花序分化始于真叶展平期,止于4叶期。5叶期开始现蕾。花序分化具体细分为花序未分化期、花序分化始期、花序分化期、花原基分化期、萼筒分化期、花瓣分化期、雄蕊分化期、雌蕊分化期、花序伸长期。对扁豆花序轴发育研究表明,两扁豆品种的花序轴在花序分化过程中存在显著差异。‘南汇23’的花序轴宽度从花序分化开始一直大于‘眉豆2012’的花序轴宽度,其花序轴长度从花序分化期后期便小于‘眉豆2012’的花序轴长度。花序分化完成后的生长发育过程中,两扁豆品种花序轴各性状的发育规律不尽相同,并且‘南汇23’完成花序轴发育的时间比‘眉豆2012’的长两天。此外,环境显著影响花序轴各性状的发育。花序分化过程的划分和花序轴发育研究有利于在扁豆生产上制定相应的栽培技术措施。更多还原

【Abstract】 Lablab purpureus (L.) Sweet is an ancient legume species, commonly kown as lablab, who serves as vegetable, foodstuff and herb. Lablab has many outstanding qualities. In China, lablab serves as vegetable and Chinese medicinal herb. In recent years, the market demand for output and market quality in lablab gradually has exceeded the supply. In addition, lablab breeding obviously lags behind other legume plants in the world. It is needed to improve the output and market quality. Through years’observation of the inflorescence and fruit development in cultivated lablab varieties, we found that the florescence traits were significantly associated with output of lablab, fruit traits were the important factors of market quality in lablab and the growth phenological traits had influence on output and market quality of lablab to a certain degree.The objective of this study was to identify quantitative trait loci (QTLs) associated with the agronomic traits: floral axis rachis length, inflorescence length, peduncle from axis to axillae, peduncle from axillae to first flower, peduncle between extreme flowers, node number of inflorescence, rachis internode length, node of the first inflorescence, node of lowest florescence, pod length, pod diameter, pod flesh thickness, pod area, pod volume, flowering time, podding time and harvest maturity period in lablab, which were key factors of the output and market quality in lablab. And it was to observe the regulation of inflorescence differentiation, growth and development of florescence in lablab.A molecular linkage map was constructed by using 136 F2 progeny derived from a cross of‘Medou 2012’בNanhui 23’. The map included 131 loci (122 RAPD and nine morphological loci), and the fourteen linkage groups covered 1302.4 cM with the average marker interval of 9.9 cM. Using the F2 population and F3 derived lines with composite interval mapping method, a total of 49 QTLs were detected on ten linkage groups for nine agronomic traits (floral axis rachis length, inflorescence length, peduncle from axis to axillae, peduncle from axillae to first flower, peduncle between extreme flowers, node number of inflorescence, rachis internode length, node of the first inflorescence, node of lowest florescence). Some QTLs were detectable in the same linkage regions among different generation/season combinations, suggesting genes which control inflorescence development were pleiotropic or coincident involving more than one trait. Thus these QTLs can be tagged for marker assisted selection in breeding high yield vegetable cultivars of lablab.Using 136 F2 population derived from a cross of‘Medou 2012’בNanhui 23’, the six agronomic traits, namely pod length, pod diameter, pod flesh thickness, flowering time, podding time and harvest maturity period were measured. Bayesian model selection was used to locate interacting QTLs for these traits. A total of 21 main-effect QTLs and 22 pairs of epistatic interactions were detected under that the threshold of statistic 2lnBF was 2.1 for declaring significance. The phenotypic variance explained by these QTLs ranged from 2.8% to 9.6%. In addition, we repeatedly measured eight agronomic traits, including five fruit traits (pod length, pod diameter, pod flesh thickness, pod area and pod volume ) and three growth phenological traits (flowering time, podding time and harvest maturity period) in F3 lines from two planting years in order to map the QTLs for these traits in lablab. Bayesian model selection was used to analyze main-effect QTL, epistatic QTL and main-effect QTL by environment interaction as well. As a result, a total of 30 main-effect QTLs were identified on 9 linkage groups with relative small phenotypic variances, and of which, 13 main-effect QTLs had interacted with two planting years. Twenty four pairs of epistatic QTLs were also found which accounted for large proportions of the phenotypic variations. In both F2 population and F3 lines, Some QTL and QEs were found to be pleiotropy. Two QTLs in F2 population, four QTLs and two QEs in F3 lines were pleiotropic. Among these QTLs, two QTLs were detectable among three generations, one responsible for flowering time, podding time and harvest maturity period, and another for flowering time and harvest maturity period simultaneously. These QTL were useful in MAS to improve the output and market quality of lablab.In order to probe into the regulation of growth and development of florescence in lablab, the inflorescence differentiation in the way of paraffin-cut section was observed with‘Nanhui 23’and‘Meidou 2012’as materials. Results showed that the process of inflorescence differentiation in lablab was from first-leaf time to four-leaf time, and could be divided into nine stages: unifferentiation phase, inflorescence differentiation elementary period, iInflorescence differentiation phase, single flower differentiation phase, bract differentiation phase, petal differentiation phase, stamen differentiation phase, pistil differentiation phase, inflorescence protraction phase. In the course of the whole inflorescence differentiation, the diameter of floral axis rachis in‘Nanhui 23’was larger than that in‘Meidou 2012’, while the length of floral axis rachis in‘Nanhui23’was less than that in‘Meidou2012’. And in the development of floral axis rachis, there was the distinct difference in two accessions, and the development of floral axis rachis in‘Nanhui 23’was two day long than that in‘Meidou 2012’. In addition, the developments of all traits about floral axis rachis were evidently influenced by environment.更多还原