节点文献

橡胶树分子遗传图谱的构建及相关性状的初步定位

Molecular Linkage Map Construction and Preliminary QTL Analysis of Hevea Brasilienses

【作者】 和丽岗

【导师】 郑学勤; 王文泉;

【作者基本信息】 华南热带农业大学 , 作物遗传育种, 2007, 博士

【摘要】 天然橡胶产业在我国属于典型的资源约束型产业,受气候条件的严重制约,产品具有明显的地域性、不可替代性,在国民经济建设和国家安全建设中具有重要的地位。近年来,随着经济的迅速增长,我国对天然橡胶的需求量大幅增加,自给率已由上世纪90年代的约50%下降到目前的不足1/3,且供需缺口在逐年加大。橡胶树优良品种的选育和推广是我国天然橡胶产业持续发展的基础。但橡胶树育种周期较长,橡胶树遗传基础研究薄弱,优良品种选育十分艰难。橡胶树的选育种主要是从不同的有性群体中筛选出高产单株并进行检验的过程。为了缩短育种年限,提高选育种效率,各国橡胶育种工作者发明了试割法、刺检法、形态预测法、叶脉胶法、小叶柄胶法等进行产量预测。但是这些方法都是从产量相关基因的表型进行的分析,受外界因素的影响较大,具有一定的局限性。从DNA水平研究橡胶树干胶产量、木材积蓄量、胶乳代谢生理指标等性状相关基因的信息,在橡胶树选育种中具有重要的指导意义。本研究用云南省热带作物科学研究所执行“九五”国家攻关项目期间选配的一个橡胶树人工授粉杂交组合GT1×IAN873的195个F1代群体,构建了AFLP连锁遗传图。提取了该群体中的174个单株的基因组DNA,用EcoRI和MseI酶切系统,筛选了13对选择性扩增引物,用ABI PRISM377 DNA测序仪电泳后,得到675个多态性位点,其中487个符合孟德尔分离规律,用Joinmap 3.0进行连锁分析,可以分为18个连锁群,连锁图含261个标记。18个连锁群中,第一群的图距最长,为125.64cM,含有43个标记;第16群的图距最短,图距为47.55cM,含有7个标记;第9、11连锁群标记最少,均为5个。连锁图谱的总图距为1455.57cM,各标记间平均图距为5.58cM。对该群体的性状鉴定表明,离地1.00米处的茎围在26.50-36.00cm之间,变异系数为0.209。2003年试割前三刀平均干胶产量A3和后五刀平均干胶产量A5分别在0.97-2.28和1.43-3.45克/株·次之间,变异系数分别为0.588和0.589。2006年试割前三刀和后五刀的平均干胶产量(B3和B5)分别在1.55-4.85和3.12-7.80克/株·次之间,变异系数分别为0.684和0.564。胶乳中的干胶含量在30.53%-38.32%之间,变异系数为0.159。群体胶乳蔗糖含量在10.48-19.88mM之间,变异系数为0.427。胶乳无机磷含量在17.73-25.53mM之间,变异系数为0.298。胶乳硫醇含量在0.26-0.46mM之间,变异系数为0.432。这些性状都有较大的变异幅度,出现明显分离,而且属于正态分布。说明橡胶树胶乳中的干胶含量、蔗糖含量、无机磷含量和硫醇含量等生理性状也可以进行QTL定位分析。用复合区间作图法对各性状进行了QTL初步定位。用WinQTLCart 2.5在LOD值大于或等于2.0的条件下,检测出与茎围相关的13个QTL,其中3个有加性效应,10个有负效应,贡献率在10.90%-20.74%之间。在LOD值大于或等于2.0的条件下,仅检测到两个与干胶含量相关的、具有加性效应的QTL,贡献率均为5.39%。在LOD值大于或等于6.0的条件下,CIM扫描得到10个与胶乳蔗糖含量相关的QTL,其中6个有加性效应,4个有负效应,贡献率在19.67%-31.00%之间。在LOD值大于或等于2.5的条件下,检测到与无机磷含量相关的QTL 10个,其中4个有加性效应,6个有负效应,贡献率在9.35%-14.15%之间。在LOD值大于或等于9.5的条件下,CIM扫描得到10个与硫醇含量相关的QTL,其中4个有加性效应,6个有负效应,贡献率在28.41%-34.25%之间。2003年试割的前三刀产量A3在LOD≥2.5时检测到13个QTL,其中4个有加性效应,贡献率在14.97%—19.70%之间,9个有负效应,贡献率在13.48%—21.25%之间。后五刀产量A5在LOD≥2.2时检测到14个QTL,8个有加性效应,贡献率在13.21%—22.86%之间,6个有负效应,贡献率在10.08%—21.76%之间。A3和A5检测到9个位置的效应均相同的QTL,贡献率在10.08%—22.50%之间。2006年试割的前三刀产量B3在LOD≥2.1时检测到16个QTL,其中6个有加性效应,贡献率在13.50%—29.66%之间,10个有负效应,贡献率在12.28%—25.09%之间。后五刀产量B5在LOD≥2.5时检测到9个QTL,5个有加性效应,贡献率在17.11%—31.97%之间,4个有负效应,贡献率在31.31%—35.37%之间。B3和B5检测到6个位置的效应均相同的QTL,贡献率在21.44%—35.37%之间。四次产量测定(即A3、A5和B3、B5)中,没有找到同时影响产量性状的QTL。但在其中三次测产有4个QTL同时影响产量。其中位于第五连锁群的两个QTL具有加性效应,贡献率在17.83%-24.42%之间;位于第18连锁群的两个QTL有负效应,贡献率在19.72%-21.93%之间。由于目前仍然缺乏一个可以用胶乳生理指标来综合判断各种橡胶树育种材料产胶潜力的量化模型,所以本研究也未用这些生理参数校正各个体的产量和生长量。橡胶树很难像短期农作物那样利用近交系和高世代群体构建遗传图谱,但橡胶树的作图群体可长时间保存,图谱完成后可用于长期的多方面的研究,这是其他作物图谱所不能比拟的,因此橡胶树分子遗传图谱的构建和相关性状的QTL定位,对研究橡胶树产量、生长量、抗性等农艺性状有重要意义。

【Abstract】 Natural rubber plays an important role in national economy and security since its specific characteristics, although the industry is strictly confined by the ecological conditions in China. The demand of natural rubber is increasing rapidly along with Chinese economic development. The recent self-support is declined to 1/3 from 1/2 in 1990s.New planting material innovation becomes a formidable work since rubber tree improving requires a long evaluation period, relatively poor in genetic fundamental studies, as well as the narrow genetic bases.Rubber tree breeding and selection is a procedure of create new seedling populations and evaluate superior individuals. In order to shorten breeding cycle and improve selection efficiency, a variety of yield prediction methods are invented, such as young tree test tapping, puncture test, morphological prediction, leaf vein latex test, petiole latex test, and so on. However, these phenotypic tests may be greatly affected by environmental factors. Studies on genes or DNA segments related to rubber tree growth, yield and latex physiological characters could be informative.An AFLP linkage map is constructed using an artificial cross between two commercial Hevea brasiliensis clones, GT1 and IAN873. The F1 population is composed of 195 seedlings. 174 DNA samples were extracted. 13 EcoRI and MseI primers were selected. AFLP electrophoresis was conducted by ABI PRISM377 DNA Sequencer. There are 487 loci accord with Medal segregation among 675 poly-morphological loci. The map is consist of 261 markers and can be divided to 18 linkage groups. Group 1 is the longest with a distance of 125.64cM, contains 43 markers. Group 16 is the shortest with a distance of 47.55cM, contains 7 markers. Group 9 and 11 contain the least makers, both contain 5 markers. The total distance of the map is 1455.57cM. The average interval is 5.58cM.The seedling trial was established in Yunnan Institute of Tropical Crops in 1999. The range of girths (100cm above ground) of the population was from 26.50 to 36.00 cm; the variation is 0.209 in 2006. The ranges of average yield of first 3 tapping (A3) and next five tapping (A5) were 0.97 to 2.28 g/t/t and 1.43 to 3.45g/t/t; the variations were 0.588 and 0.589 in 2003. The trees were tapped in 2006 again, the range of B3 and B5 were 1.55 to 4.85g/t/t and 3.12 to 7.80g/t/t, and the variations were 0.684 and 0.564 respectively. The dry rubber contents of latex (DCR %) were ranged from 30.53% to 38.32%, the variation was 0.159. The contents of sucrose (Sue) were ranged from 10.48 to 19.88mM, the variation was 0.427. The contents of inorganic phosphorus in latex (Pi) were ranged from 17.73 to 25.53mM. the variation was 0.298.Thiol contents in latex (R-SH) were ranged from 0.26 to 0.46mM, the variation was 0.432. The trait segregations were normally distributed which indicated the physiological parameters could be analyzed by QTL scanning.Composite interval mapping method was employed in this study. There were 13 putative QTL were related to tree girth. Three of them showed positive additive effect while 10 showed negative additive effect. Phenotypic variation explained (PVE) by the putative QTL varied from 10.90% to 20.74%. Only two QTL were related to DCR%. Both showed positive additive effect, and PVE were 5.39. There were 10 QTL related to Sucrose content. 6 of them showed positive additive effect while 4 loci showed negative effect. PVE varied from 19.67% to 31.00%. There were 10 QTL related to R-SH content in latex. 4 of them were showed positive while 6 showed negative additive effect. The PVE ranged from 28.41 % to 34.25%.There were 13 QTL detected related to A3. 4 of them showed positive additive effect while 9 showed negative. The PVE ranged from 13.48% to 21.25%. There were 14 QTL related to A5, 8 of them showed positive while 6 loci showed negative additive effect. The PVE ranged from 10.08% to 22.86%. There were 9 loci showed the same position and additive effect in A3 and A5. The PVE varied from 10.08% to 22.50%.There were 16 QTL related to B3. 6 of them showed positive additive effect while 10 showed negative. The PVE ranged from 12.28% to 29.66%. There were 9 QTL related to B5, 5 of them showed positive while 4 loci showed negative additive effect. The PVE varied from 17.11% to 35.37%. There were 6 loci showed the same position and additive effect in B3 and B5. The PVE varied from 21.44% to 35.37%.No common QTL was detected among the 4 yield trait test, i e A3, A5, B3 and B5. However, 4 loci were common in 3 test tapping. Two loci showed positive and the other two showed negative additive effect. The PVE ranged from 17.83% to 24.42%.A quantitative model is lacked for evaluation yield and growth potential of the rubber tree according to latex physiological parameters. The yield and growth of the individuals could not be adjusted by the physiological parameters in this study.It is very hard to establish multi-backcross inbred lines or near-isogenic lines for rubber tree like annual crops. Nevertheless, the mapping population of rubber tree can be conserved easily. Once the map is well constructed, the agronomic traits can be evaluated by many ways. So the strategy will be useful for improving yield, growth, disease or cold resistance of the tree.

节点文献中: 

本文链接的文献网络图示:

本文的引文网络