【作者】 刘丹；

【导师】 陈温福；

【作者基本信息】 沈阳农业大学 ， 作物学， 2014， 博士

【摘要】 近年来,随着分子标记技术的快速发展,传统育种技术与分子标记技术的有效结合己成为当前作物育种的重要方向。品种资源是育种家选育亲本的基础,其遗传多样性水平可为种质资源利用和杂交亲本选配提供理论依据,而穗部性状是育种家选育新品种的重要指标。良好的穗部形态直接决定着水稻的产量和品质。本研究从品种的选育地区和推广年代两个方面入手,系统分析了东北粳稻近40年来育成品种的遗传多样性,并基于籼粳杂交衍生系群体对水稻穗部性状进行遗传分析和基因定位研究,旨在为开展东北粳稻分子标记辅助选择育种(MAS)奠定基础。试验主要结果如下：1.东北粳稻遗传多样性水平较低。54个SSR多态性引物共检测到195个等位基因,平均每个位点的等位基因数为3.61。不同地区的粳稻品种的遗传多样性存在差异,其中吉林省粳稻品种多样性水平较高,而黑龙江省水稻品种较低；不同年代粳稻品种的遗传多样性分析表明,近年东北粳稻品种多样性水平略有提高。分子方差分析(AMOVA)结果证明,等位基因变异主要来自于群体内部,其中不同省份地区之间的种内变异为88.18%,而不同推广年代间的种内变异为96.86%。通过Neighbor-joining(NJ)聚类分析将东北粳稻品种划分为三个类群,分别对应于辽宁、吉林和黑龙江三个省份。群体结构分析表明,东北粳稻遗传结构存在明显的分化,且黑龙江省品种与辽宁省品种存在较大差异。2.部分产量相关功能基因在东北粳稻育种中得到利用。对8个产量相关的功能基因检测发现,东北粳稻中除了IPA1和GW2两个基因位点未存在等位基因变异外,其它6个产量相关的功能位点均存在等位基因变异,表明GS3,GS5,qSW5,Gn1a,qGW8以及DEP1这6个功能基因位点均部分的被固定到了现代粳稻品种中。3.东北地区近现代以来育成品种的籼型基因频率逐渐增加。对近年来育成的粳型超级稻品种遗传结构进行分析可知,参试15份粳型超级稻品种均含有一定的籼型血缘,且不同地区超级稻品种所含籼型基因频率(Fi)有所不同。对不同年代育成的参试材料的籼型基因频率(Fi)进行比较分析发现,以2005-2011年育成的超级稻品种籼型基因型频率最高,Fi为0.068；12份1977-1999年育成的栽培品种次之,Fi为0.033；9份1963-2000年育成的日本品种几乎不含有籼型血缘,Fi为0.011。4.检测到14个控制穗部性状的QTLs,包括2个穗长QTLs,1个穗重QTL,2个一次枝梗数QTLs,2个二次枝梗数QTLs,2个颖花数QTLs,3个结实率QTLs,1个千粒重QTL和1个着粒密度QTL,它们分布于第1,3,4,5,6,7,11和12号染色体上进一步分析发现,控制穗部性状的QTL多分布在第1号染色体上,数量占总数的40%以上。控制穗部性状的单个QTL的贡献率在8.06%-58.62%之间,其中有6个增效等位基因来源于七山占,8个来源于秋光。5.检测到27个控制粒形性状的QTLs,包括3个粒长QTLs,11个粒宽QTLs和13个粒厚QTLs,它们分布于第1,2,3,4,5,11和12号染色体上,可分别解释14.45-38.48%、28.98-52.36%和38.77-44.23%的表型变异。在2011年检测到7个QTLs,包括3个粒宽QTLs和4个粒厚QTLs;2012年检测到11个QTLs,包括2个粒长QTLs,5个粒宽QTLs和4个粒厚QTLs;2013年检测到9个QTLs,包括1个粒长QTL,3个粒宽QTLs和5个粒厚QTLs。检测到两个重演性较好的QTLs,分别为控制粒宽的qGW5a和控制粒厚的qGT12c,它们在三个年份中均稳定表达。控制粒形性状的单个QTL的贡献率在5.58-26.90%之间,其中有6个增效等位基因来源于七山占,14个来源于秋光。6.sp突变体是一个散穗型突变体材料,田间表现为穗部一次枝梗向外延伸,与穗轴夹角增大,穗向周围散开。与亲本相比,株高极显著增加,穗重极显著减少。遗传分析表明,该性状受一对显性核基因控制。利用sp与02428构建的F2群体进行定位研究,初步将sp基因定位在第4号染色体长臂端,位于E3和RM17578之间的62.9Kb区域内,跨越OSJNBb0022F16和OSJNBa0071I132个BAC克隆群。7.lax(t)俐突变体是一个散穗型突变体材料,与野生型相比,其一次枝梗正常,二次枝梗小穗发育受阻,小穗退化,只长有末端籽粒。利用极端个体分组与隐性基因组分析法(Bulked extremes and recessive class analysis, BERCA),将lax(t)基因初步定位在第4号染色体长臂RM16883和MM1466之间159.6kb区域。进一步通过扩大F2定位群体和筛选合成新的多态性引物,将lax(t)基因精细定位在MM1406和RM16890之间47.8kb区域。利用RAP-DB进行基因预测分析,发现该区域只有一个候选基因Os04g0396500,其编码一个参与调节水稻AM形成的蛋白因子。测序分析表明,lax(t)与日本晴存在三个位点上的碱基差异,推测lax(t)可能是lax2的一个新的等位基因。更多还原

【Abstract】 With the development of molecular marker technology, the traditional breeding in combination with marker-assistant selection has become the mainstream of the crop breeding in recent years. Germplasm resource is the base of parents selection. Genetic diversity of the germplasm resources provides us with the theoretical reference to the utilization of the germplasm,The panicle traits decides the yield and quality of rice directly which is regarded as an important factor in new variety selection for breeders. In this study, the genetic diversity of japonica rice cultivars released from1970’s to now and the fine mapped gene controlling the panicle traits in the northeast of China were analyzed which will benefit the breeding community in japonica rice.The major results are as follows:1.There is a lower genetic diversity level of’the japonica rice cultivars in northern China and the cultivars genetic diversity are different among release years and geographic locations. A total of195alleles (Na) are detected with an average of3.61alleles per locus. Further analysis showed that the genetic diversity of the cultivars from Jilin province is the highest among the three geographic distribution zones while the Heilongjiang is lower. According to the genetic diversity among different release years, it is showed that the cultivars genetic diversity increased slightly these years. The Analysis of molecular variance (AMOVA) reveals that genetic differentiations are more diverse within the population than that among the populations and the intraspecific variation in different location and release year are88.18%and96.86%, respectively. The Neighbor-joining (NJ) tree indicates that cultivar clusters based on geographic distribution represent three independent groups, with on behalf of Heilongjiang, Jilin and Liaoning province. The population structure of Heilongjiang cultivars is significantly different to the cultivars from Liaoning, a significant differentiation of population structure in japonica rice collection.2. Many of functional gene related yield were used in japonica rice breeding in northern China. Eight of functional gene tags were detected in japonica rice cultivars, which showed that most of functional gene tags had allelic variation in japonica rice cultivars except for IPA1and GW2. It implied that these6functional genes such as GS3, GS5, qSW5, Gnla, qGW8and DEP1are fixed into the modem japonica rice varieties.3. There is a gradually increasing of indica-allele frequencies in northern China. To analyze the genetic components of super japonica rice, we found that indica linage have already introgressed into the genomes of 15super-rice varieties at different level which represented by indica-type frequency (F,). The F, variation of super-rice varieties reflected registration periods difference, varieties bred in2005-2011are highest (F,=0.068), followed by1977-1999(F1=0.03) and rarely in1963-2000(0.011).4. A total of14QTLs which control panicle traits were detected, including2QTLs for PL,1QTL for PW,2QTLs for PBN,2QTLs for SBN,2QTLs for SNP,3QTLs for SSR,1QTL for GD and1QTL for TGW, which were identified on chromosomes1,3,4,5,6,7,11and12. Further analysis showed that the number of QTLs which detected on chromosome1accounted for more than40%of the totals. It is implied that the expression of gene on chromosome1plays an important role in rice panicle traits. Furthermore, the contribution of single QTL in this study was between8.06%-58.62%, and there were8efficiency alleles from the typical japonica rice’Akihilari’and6efficient alleles from indica rice’Qishanzhan’.5. From2011to2013, a total of27QTLs of grain shape traits were detected on chromosomes1,2,3,4,5,11and12, including3QTLs for grain length,11QTLs for grain width and13QTLs for grain thickness, which explained the phenotypic variation14.45-38.48%,28.98-52.36%and38.77-44.23%, respectively. Seven QTLs were detected in2011including3QTLs of grain width and4QTLs of grain thickness; Eleven QTLs were detected in2012including2QTLs of grain length,5QTLs of grain width and4QTLs of grain thickness; Nine QTLs were detected in2013including1QTL of grain length,3QTLs of grain width and5QTLs of grain thickness. The qGW5a and qGT12c were detected in three years, which implied that they were stable expression and stronger repeatability. Further analysis showed that the QTL controlling grain traits are mainly located on chromosome3and12, and accounted for more than44.44%of QTLs. The contribution of single QTL in this study was between5.58-26.90%. There were6efficient alleles from ’Qishanzhan’and14efficient alleles from ’Akihikari’.6. sp mutant is a spreading panicle mutant material with the panicle branch extending outward, the angle between primary branch and rachis increasing and the panicle growing around. Comparing with wild type parents, the plant height of sp significantly increased and the panicle weight has significantly decreased. The genetic analysis showed that the phenotype of sp was controlled by a single dominant nuclear gene. Primary mapping based on the F2derived line between sp and02428showed that the sp gene was located on the long arm of chromosome4, narrowed down to a62.9kb region between marker E3and RM17578and included two BAC such as OSJNBb0022F16and OSJNBa0071I13.7. A lax panicle natural mutant lax(i) was found from the recombinant inbred lines (RILs) which derived from a cross between ’Akihikari’(japonica) and ’Qishanzhan’(indica). From phenotype identification in the field, we find that the lax(i) mutant showed the second branch disappearance and lateral spikelet degradation. Genetic analysis showed that lax(t) phenotype was controlled by a single recessive nuclear gene. By map-based cloning based on BERCA, the target gene was located on a159.6kb region between marker RM16883and MM1466on the chromosome4. Fine mapping by expanding the mapping population and designing new markers, we finally narrowed down to a47.8kb region between markers MM1406and RM16890. Gene prediction shows that there was only one candidate gene Iax2in this region and it guide the AM formation. Sequence analysis revealed that there were three mutant locates between lax(t) and Nippobare, it is indicated that lax(t) is anew allelic genes of lax2possibly.更多还原