【作者】 杨加银；

【导师】 盖钧镒；

【作者基本信息】 南京农业大学 ， 作物遗传育种， 2010， 博士

【摘要】 借鉴玉米、高粱和水稻等作物上的成功经验,利用杂种优势是实现大豆产量突破的重要途径。随着大豆“三系”的配套和杂交种的选育成功,大豆杂种优势的研究愈来愈受到关注。但是要使杂交种真正应用于生产,仍然有许多问题需要研究和探索,大豆杂种优势相关的遗传基础研究必须引起高度重视。本研究选用来源于我国黄淮地区和美国的熟期组Ⅱ～Ⅳ的8个大豆亲本品种,2002-2004年每年按Griffing方法Ⅳ配制28个双列杂交组合,2003-2005年连续3年进行田间农艺性状鉴定和室内品质检测。选用300个SSR标记,对8个大豆亲本品种进行全基因组扫描。本研究分析了该组大豆亲本间杂种主要农艺与品质性状的表现和杂种优势、产量性状的亲本配合力、遗传距离与杂种优势的相关性；应用基于数量性状主基因+多基因遗传模型的主-微位点组分析方法,解析了杂种产量的主、微位点组遗传构成及其效应；并应用基于回归的单标记分析方法,检测与大豆产量相关的SSR标记位点,剖析杂交组合的等位变异构成。为深入理解杂种优势的遗传构成和大豆杂种产量的聚合育种提供依据。主要研究结果如下：1黄淮大豆亲本间主要农艺与品质性状的杂种优势黄淮地区8个大豆亲本间普遍存在中亲优势,产量、单株荚数和单株粒数相对较大,百粒重无优势。生育期及品质性状(蛋白质和脂肪含量)不明显。大豆亲本间存在产量超亲优势,超亲优势率平均20.39%,组合间差异甚大,变幅-5.34%～76.88%,优选出4个高产高优势组合,即豫豆22×晋豆27、淮豆4号×晋豆27、诱变30×蒙90-24和菏豆12×晋豆27,超亲优势率分别为76.88%、29.90%、34.42%和43.16%,超标率均在19%以上。2黄淮大豆主要产量性状的亲本配合力分析8个大豆亲本间产量的一般配合力和特殊配合力存在显著差异,且与年份存在显著互作,一般配合力与年份互作大于特殊配合力与年份互作,由加性和非加性基因共同决定。3个亲本即晋豆27、诱变30和淮豆4号表现出较好的一般配合力效应。2个组合即豫豆22×晋豆27和诱变30×蒙90-24表现出较好的特殊配合力效应。大豆亲本间产量杂种优势既与双亲一般配合力之和及特殊配合力有关,又不完全相关。高优势高产组合的亲本产量配合力特点为亲本之一具有较高的一般配合力,或双亲具有较高的一般配合力之和,兼有较高的特殊配合力。单株荚数和单株粒数的情况与产量一致。3黄淮大豆亲本间遗传距离与杂种优势的相关性分析8个大豆亲本品种间的分子标记遗传距离为0.4367～0.8635,平均0.6877；系谱遗传距离为0.3124～1.0000,平均0.7679。按亲本系数聚类和按SSR标记遗传相似系数聚类揭示的8个大豆亲本间的遗传关系相对一致,均分为两组,一组包含6个黄淮中、南部品种,另一组包含1个山西和1个美国品种。大豆亲本间遗传距离与产量中亲优势之间的相关程度中等,尤其是分子标记遗传距离与中亲优势之间的相关系数为0.5209,达极显著水平。要获得高优势高产大豆杂交组合,亲本间必须具有一定的遗传距离,但遗传距离大的并不一定都高产高优势,还有其他因素决定杂种优势,其中生态适应性十分关键。4黄淮大豆杂种产量的主-微位点组遗传解析8个大豆亲本间产量由6个主位点组加微位点组构成,主位点组、微位点组分别解释表型变异的75.98%和10.81%,广义遗传率为86.79%。主位点组加性效应(aJ)分别为140.10、259.65、1.95、151.35、-32.70和45.00(kg hm-2),显性效应(dJ)分别为177.15、314.25、105.75、75.90、242.85和171.00(kg hm-2).8个大豆亲本间杂种产量的遗传构成包括主位点组杂合显性效应、主位点组纯合加性效应、微位点组杂合显性效应和微位点组纯合加性效应4部分,相对重要性依次递减,以显性效应为主,加性效应为辅。大豆产量主、微位点组及其遗传效应的解析阐释了各杂交组合的遗传特点,还提供了进一步挖掘大豆遗传潜力进行产量优势改良的基础。5黄淮大豆杂种产量相关的SSR位点及等位变异剖析在300个SSR标记中,检测有38个与杂种产量显著相关的标记基因位点,分布于17个连锁群上,其中D1a和M等连锁群上较多,有8个位于连锁定位的QTL区段内(±5cM)。单个位点分别解释杂种产量表型变异的11.95%～30.20%。8个大豆亲本间杂种产量的位点构成包括有增效显性杂合位点、增效加性纯合位点、减效加性纯合位点和减效显性杂合位点4部分,其相对重要性依次递减。以杂合位点为主,纯合位点为辅。从38个显著相关的SSR标记位点中,遴选出Satt449.Satt233和Satt631等9个优异标记基因位点,Satt449-A311、Satt233-A217和Satt631-A152等9个优异等位变异,以及Satt449-A291/311、Satt233-A202/207和Satt631-A152/180等9个优异杂合基因型位点。这些结果为理解杂种优势的遗传构成和大豆杂种产量的聚合育种提供了依据。6黄淮大豆优良杂交组合产量的数量遗传学解析4个大豆优良杂交组合产量的位点效应构成中,杂合位点占主导地位,且显性效应相对较大；纯合位点占次要地位,且加性效应相对较小。不同的优良组合具有不同优异杂合位点。一个优良的杂种应该具备杂合和纯合位点两方面的优势,聚合更多的杂合位点,实现优良位点基因型值的最大化。更多还原

【Abstract】 The exploitation of heterosis was one of few milestones in the revolution of agricultural sciences and technologies leading to big jump in crop improvement in 20th century. Utilization of heterosis is one of the most effective ways to increase yield and improve quality in several major crops, including maize, sorghum and rice. In order to apply heterosis more efficiently, scientists have made every effort for a long time to probe into the genetic basis and the method of prediction for heterosis. The development of molecular markers and its application in biological researches had provided technological base for dissecting heterosis in crops.Soybean [Glycine max (L.) Merr.] is the leading oilseed crop produced in the world, and one of the most important sources of vegetable protein and edible oil worldwide, however the yield is relatively low compared to other important field crops such as maize, rice etc. The use of hybrid could not only increase the yield but also accelerate the genetic gain per year. Soybean breeders keep trying to find ways to use heterosis. Even through the first hybrid "HybSoyl" was released in 2002, its commercialization has not been achieved in large scale yet. In recent years, breeding for hybrid cultivars of soybean for utilization of heterosis has been paid great attention, but there are few reports published on the fundamental aspects regarding the heterosis in soybean. In fact, for a real utilization of hybrid soybean, the important prerequisite is high heterosis. Therefore, a fundamental effort in hybrid breeding is the choice of parents and identification of superior hybridized combinations.In this study,8 summer soybean cultivars (lines) from different origins from Huang-Huai Valley of China and US in maturity groupⅡ-Ⅳ, were used to develop 28 crosses according to a GriffingⅣmating design in 3 years (2002-2004). These crosses, together with their parents, were used to test for yield and quality traits in 3 years (2003-2005) in Huaian, Jiangsu, China. The genotyping data of 300 SSR polymorphic markers on 8 accessions were obtained. The present study was aimed at evaluating the heterosis performed in F1 generation of 14 yield and quality traits as to provide guidelines of parental selection in breeding for hybrid cultivars. In the paper, combining ability of yield-related traits was analyzed, and relationships between F1 yield heterosis and its pedigree-based and SSR-based genetic distances were investigated. Furthermore, the analysis of major-minor locus groups of yield based on major gene plus polygene mixed inheritance model was used to explore the genetic structure of hybrids among a group of soybean materials, additive and dominance effects of major-minor locus groups were estimated. Finally, the molecular data of 300 SSR markers on 8 parental materials were analyzed for association between SSR markers and hybrid yield using the single marker regression analysis. The hybrids were dissected into their allele constitution and the effects of alleles and genotypic value of each locus was estimated. The main objectives of this study were to understand the genetic basis of heterosis and lay a foundation for hybrid soybean breeding by design. The main results were as follows:1 Study on heterosis of agronomic and quality traits in hybrid soybean in Huang-Huai ValleyThere appeared mid-parent heterosis among 8 soybean parents. Heterosis of yield, pods and seeds per plant was relatively larger, while no obvious heterosis for 100-seed-weight, and no obvious heterosis in days to flowering and maturity and quality (protein and oil contents) traits. There were heterobeltiosis in yield among 8 soybean parents with average of 20.39%, and a big difference among hybridized combinations with a range from-5.34% to 76.88%. Among the combinations, Yudou 22 X Jindou 27, Huaidou 4 X Jindou 27, Youbian 30 X Meng 90-24 and Hedou 12 X Jindou 27 had the heterobeltiosis in yield of 76.88%,29.90%,34.42% and 43.16%, respectively.2 Analysis on combining ability of yield-related traits among key parental materials in soybean in Huang-Huai ValleyThere were significant differences among the parents for general combining ability (GCA) and crosses for specific combining ability (SCA) for yield-related traits studied. The GCA variance was significant, and larger than SCA variance, which indicated that yield traits were controlled by additive and non-additive gene effects. The interaction of year by GCA and SCA was significant, and the interaction of year by GCA was larger than the interaction of year by SCA. Among the parents, Jindou 27, Youbian 30 and Huaidou 4 were the best general combiners for yield. The best specific crosses for yield were Yudou 22 X Jindou 27 and Youbian 30×Meng 90-24. Yield heterosis among parents was related to GCA and SCA. One of soybean parents has high GCA or both have high GCA and high SCA in high-yield crosses. Combining ability of pods and seeds per plant was relatively in accord with combining ability for yield.3 Relationship between parental genetic distance measured by SSR markers and pedigree with heterosis in soybean in Huang-Huai ValleyGenetic distance among 8 soybean cultivars measured by SSR markers varied from 0.4376 to 0.8635 with the mean of 0.6877 and from 0.3124 to 1.0000 with the mean of 0.7679 based on the pedigree. SSR-based and pedigree-based cluster analysis revealed that genetic relationships for 8 soybean parents were basically consistent, and 8 soybean parents were grouped into 2 groups, one including 6 cultivars from middle and south of Huang-Huai Valley, the other consisting of one from Shanxi and one from America.The correlation between the genetic distance and mid-parent heterosis was moderate, especially, the correlative coefficient was 0.5209 which was significant at 0.01 level between mid-parent heterosis and the genetic distance measured by SSR between parents. Therefore, certain genetic distance is required for a cross with high heterosis and high yield, but genetic distance is not an only determinant factor for high heterosis and yield.4 Genetic analysis in terms of major-minor locus group constitutions of yield of hybrid soybean in Huang-Huai ValleyThere were 6 major locus groups plus minor locus groups detected in the genetic system of the 8 soybean parents and their hybrids. Genetic variation of the major locus groups and the minor locus groups explained 75.98% and 10.81% of the phenotypic yield variation, respectively, which indicated that major locus groups were the major source of genetic variation, with their additive effects (aJ) of 140.10,259.65,1.95,151.35,-32.70 and 45.00 (kg hm-2) and dominance effects (dJ) of 177.15,314.25,105.75,75.90,242.85 and 171.00 (kg hm-2), respectively, while the minor locus groups were a supplement source among the soybean hybrid. The genetic constitutions of the soybean hybrids were composed of heterozygous dominance effects of major locus groups, homozygous additive effects of major locus groups, heterozygous dominance effects of minor locus groups and homozygous additive effects of minor locus groups, with their relative importance in a descending order. The dissection of the relative importance of the genetic effects of major-minor locus groups helps to explain the genetic characteristics of the hybrid among the parents and provides the genetic basis for further mining the genetic potential of the soybean parental materials in the improvement of hybrid.5 Analysis of SSR loci and alleles associated with hybrid yield in soybean in Huang-Huai Valley38 SSR loci located on 17 linkage groups were identified to associate with hybrid yield in the diallel crosses with more loci on linkage groups D1a, M, etc, and 8 of the 38 loci were located within a region of±5 cM apart from known QTL identified from family-based linkage (FBL) mapping in the literature. Each of the loci explained 11.95%～30.20% of the phenotypic variance of hybrid yield. The allele pairs of the hybrids were composed of 4 parts, i.e. positive dominant heterozygous loci, positive additive homozygous loci, negative additive homozygous loci and dominant heterozygous loci, with their relative importance in a descending order. Among the 38 loci associated with hybrid yield, nine elite loci such as Satt449, Satt233 and Satt631 and nine elite alleles such as Satt449-A311, Satt233-A217 and Satt631-A152 were identified. Meanwhile, nine heterozygous allele pairs such as Satt449-A291/311, Satt233-A202/207 and Satt631-A152/180 were detected. These results will provide with some relevant information for understanding the genetic basis of heterosis and lay a foundation for hybrid soybean breeding by design.6 Quantitative genetic dissection of yield of elite crosses in soybean in Huang-Huai ValleyAmong genetic composition of hybrid yield in the 4 elite soybean crosses, heterozygous loci played a leading role, and the dominance effects were relatively larger; homozygous loci located in a secondary position, and the additive effects were relatively smaller. Different elite crosses had different elite heterozygous loci. A superior hybrid should take advantages of both heterozygous and homozygous loci, pyramiding more heterozygous loci to maximize the genotypic value of elite loci.更多还原