【作者】 任光俊；

【导师】 周开达；

【作者基本信息】 四川农业大学 ， 作物遗传育种， 2001， 博士

【摘要】 过去的50年，全球主要粮食作物单产增长了两倍以上。这主要是通过增加单位土地面积上的光合作用以及增加作物生物产量向收获器官分配所获得的。由于多种作物的叶面积指数已经很高，收获指数已近极限，因此改良作物的光合效率、增加同化物的生产将成为进一步提高作物单产的重要技术途径。 作物杂种优势虽已广泛应用于农业生产，但关于其遗传机理，目前尚无一致的看法。经典数量遗传学把控制数量性状的基因作为一个系统进行研究，无法区分单个数量基因的效应以及在染色体上的精确位置。近年来，分子标记的发展、QTLs的研究，为分析作物杂种优势和数量性状的遗传基础提供了新的方法。 本研究以美国的高光效光身稻品种Lemont为母本，中国的优良籼稻保持系冈46作父本，构建了F_2:3群体，将160个F_2:3家系按随机区组2次重复进行试验。测定了叶绿素含量、单个叶片的净光合速率、单株生物干重等性状，采用RFLP和SSR标记，构建分子图谱，分析了各性状的QTL效应和上位性效应，主要结果如下： 1、在定位的173个共显性标记位点中，Lemont等位基因频率为22.9％，冈46等位基因频率为25.7％，杂合基因型频率为51.3％。单位点卡方检验表明，36个标记位点的等位基因显著地偏离预期的孟德尔比例（1：2：1）而表现出偏分离。这些位点分布在10条染色体上，尤其是第3、4、6、7、11染色体上的偏分离位点较多，共27个标记位点偏向籼稻冈46，9个标记位点偏向粳稻Lemont。 2、用180个多态性标记构建的分子图谱覆盖了水稻整个基因组的12个连锁群，图谱总长度为1758.7cM，标记间的平均间距为9.77cM。该图谱较好地整合了文献报道的水稻标记连锁图。 3、采用区间作图方法定位了14个性状的36个QTLs。各性状的QTL数目有较大的差异。叶绿素a含量的QTL数最多，有7个；净光合速率和叶面积均只检测到1个QTL。单个QTL解释的表型变异幅度为6.4％-25.0％；就单个性状而言，QTLs共同解释的表型变异幅度为7.1％-43.9％。 4、对于大多数性状，都检测到了增效和减效QTL位点，在双亲基因组中呈离散分布。同一性状，在低值亲本中，有增效QTL位点；在高值亲本中，有减效QTL位点。在所有QTL位点中，约为44．4％的QTLs与亲本表型值预期的基因作用方向相反。 5、对全基因组的两位点进行双向方差分析表明，上位性效应大量存在，且显著影响叶绿素含量、净光合速率等性状。在P＜0．001水平，影响叶绿素a含量的双基因互作对数最多，为64对；影响叶面积的互作位点对最少，为32对。 6、上位性类型以加性X加性为主，加性X显性或显性X加性次之，显性X显性很少。两位点间互作的形式以非QTL位点间互作最多，QTL与非QTL位点间互作次之，QTL位点间互作最少。单个上位性位点对的贡献率为6．3％－12．1％。 7、检测到QTL和上位性都存在一因多效，相同的QTL位点或同一互作位点对影响多个性状的表现。 8、上位性分介于水稻的整个基囚组中，在生理特性、农艺性状和抗病性中均检测到大量的上位性存在。这说明上位性效应对水稻的许多重要性状及其杂种优势均有显著影响。更多还原

【Abstract】 The average global grain yield per unit area of the major staple crops, wheat, rice and maize, more than dubled in the past fifty years and this trend continues. Increased yields have been achieved by increased or extended photosynthesis per unit land area and increased partitioning of crop bimass to the harvested product. Since values of leaf area index are generally high enough at present, and the upper limit for harvest index is being approached in most bred crops, it is appropriate to further increase yield and to focus on ways of improving crop photosynthetic efficiency and increasing assimilate production genetically. Plant heterosis has been widely used in agricultural production, but the views in the genetic mechanisms of heterosis were inconsistant. The genes controlling quantitative trails were analyzed systematically in the classic quantitative genetics, but it is difficult for this method to identify single gene with mini effect and to define its precise location on the chromosome. As the molecular markers developing and studing on QTLs in recent years, these provide a new method to detect the genetic basis of plant heterosis and quantitative trait loci. The marker-based genetic analysis of heterosis and quantitative traits were studied in F2.3 generation between Lemont and Gang 46. Lemont is the female parent with high photosynthetic rate originated from United States of America, Gang 46 is the maintainer line of indica rice of chinese origin. The 160 F23 lines were replicated twice in the field following a randomized complete block design. Fourteen traits including chlorophyll content, net photosynthetic rate of single leaf, etc. were measured in Chengdu in 1999. The molecular map were constructed by using RFLP and SSR markers. The main results are as follows. 1. Among the 173 co-dominant markers anchored, the frequencies of allelic genotypes for lemont, Gang 46 and heterozygote were 22.9%~ 25.7%~ 51.3%, respectively Chisquare test indicated that distorted segregation of 36 markers was found by significant diviation from the expected Mendelian segregation ratio (1:2:1) at P 5% level. These loci distributed on 10 91 chromosomes, especially on Chromosome 3, 4, 6, 7 and 11. 27 out of 36 ditorted seqregation loci were for Gang 46, and others for Lemont. 2. A molecular map was constructed for the rice genome comprised of 180 markers (176 RFLPs and 4 SSRs) covering 12 chromosomes. The map contains 1758.7 cM with an average interval size of 7.77 eM, which well integrated the markers from the other maps. 3. A total of 36 QTLs for fourteen traits were identified by interval mapping. The number of QTLs for each trait is quite different. Senven QTLs with highest number for chlorophyll a content were identified, and only one QTL both for net photosynthetic rate and leaf area was detected espectively. Single QTL explained 6.4%-25.0% of the genotypic variation. All coincided with QTLs for each trait account for 7.1%-43.9% of the phenotypic variation. 4. For the most traits, both positive and negative additive effects were dispersed in both parents?genome. For the same trait, the parent with the lower value of the phenotype contains positive effect QTL, one with the high. value of the phenotype including negative effect QTL. About 44.4% of all QTLs had allelic effects opposite to those predicted by the parental phenotypes. 5. The digenic interactions were detected in the entire genome. All更多还原