【作者】 杨占烈；

【导师】 程式华；

【作者基本信息】 中国农业科学院 ， 作物遗传育种， 2013， 博士

【摘要】 灌浆充实是水稻的一个重要特性，直接影响产量的高低。目前对灌浆速率等灌浆特性的研究主要集中在生理生化上，由于该性状测定繁琐，利用遗传群体在基因水平上的研究甚少。为探索水稻灌浆过程中的相关基因表达，以期获得在多环境下稳定表达的相关性状QTL，本研究利用超级稻协优9308重组自交系群体为研究材料，结合构建的遗传连锁图谱，多环境条件下对强势籽粒和弱势籽粒的灌浆速率、籽粒动态发育以及灌浆持续期、充实度、穗总粒数、穗实粒数、有效穗、结实率、Q酶活性、粒长、粒宽、粒厚、籽粒体积、籽粒长宽比、百粒重等相关性状进行了QTL定位分析，主要研究结果如下：1.采用winQTLCart2.5软件对强势籽粒在4个环境下（贵阳2010、贵阳2011、富阳2010、富阳2011）与弱势籽粒在两个环境下（富阳2010、富阳2011）的最大灌浆速率和平均灌浆速率进行QTL定位分析，共检测到13个主效应QTLs，分布于第1、2、3、4、5、6、8染色体上。其中，检测到强势籽粒最大灌浆速率5个QTLs，平均灌浆速率6个QTLs，单个QTL可解释其表型贡献率的5.12%-9.56%，这些QTLs能在两个环境中稳定检测到的有：qSGFRmax-4、qSGFRmax-8、qSGFRmean-4、qSGFRmean-8；检测到弱势籽粒最大灌浆速率4个QTLs，平均灌浆速率3个QTLs。单个QTL可解释其表型贡献率的7.46%-14.28%。2.对4个环境下强势籽粒灌浆的5个时期灌浆速率（花后第6d、12d、18d、24d、30d）和平均灌浆速率（1-6d、7-12d、13-18d、19-24d、25-30d）进行QTL定位分析，共检测到21个主效应QTLs，分布在第1、2、3、4、6、7、8、9、11染色体上，单个QTL可解释的各自性状表型贡献率介于5.05%-18.09%。其中，检测到5个时期灌浆速率14个主效应QTLs，平均灌浆速率13个主效应QTLs，二者共同QTLs有6个，上述QTLs中，qSGFR-3-1在3个环境下的多个时期稳定检测到，表型贡献率介于7.16%-18.09%，最大LOD值为5.64，增效等位基因来自高值亲本协青早B。2个环境条件下检测到影响弱势籽粒5个灌浆时期的灌浆速率和平均灌浆速率主效应QTLs分别为14个和11个，其中共同QTLs有7个，这些QTLs中，位于第6染色体上RM136-RM6302区间的qIGFR-6-1能在两组试验的同一时期检测到，同时在富阳2011的4个时期检测到，表型贡献率介于6.91%-14.69%。在第3染色体上RM282-RM6283区间和RM7370-RM16区间、第6染色体上RM136-RM6302区间和RM3724-RM3330区间、第7染色体上RM5436-RM3670区间均检测到控制强、弱势籽粒灌浆速率QTL。3.采用条件分析方法结合复合区间作图法，对籽粒发育动态进行QTL定位。4个环境下检测到强势籽粒5个灌浆时期籽粒重的非条件与条件QTLs共21个，分布在第1、2、3、4、6、7、8、9染色体上，单个QTL可解释各自性状的表型贡献率介于5.25%-18.86%，其中非条件QTLs为16个，条件QTLs11个，非条件与对应的条件QTLs有6个。2个环境下检测到弱势籽粒5个灌浆时期粒重的非条件与条件QTLs共16个，分布在第1、2、3、5、6、7、8和第10染色体上，单个QTL可解释各自性状的表型变异介于5.73%-18.05%，其中非条件QTLs为12个，条件QTLs为10个，非条件与对应的条件QTLs有6个。位于第6染色体上RM136-RM6302区间的qIWG-6-1在弱势粒灌浆4个时期都能检测到并存在对应的条件QTL，说明该QTL对弱势粒灌浆起着重要的作用。检测到位于第3染色体上RM282-RM6283区间和RM7370-RM16区间、第6染色体上RM136-RM6302区间和第7染色体上RM5436-RM3670区间存在控制强、弱势籽粒灌浆速率和阶段粒重QTL，这与灌浆速率同粒重高度正相关吻合，今后可将这4个SSR标记区间作为深入研究籽粒灌浆特性的重点区域。4.4个环境下检测到控制籽粒充实度A（GPA）、充实度B (GPB)、充实度C (GPC)、灌浆持续期(GPD)、穗实粒数(NFGP)、穗总粒数(GNPP)、结实率(SF)、有效穗(PP)和百粒重(100-GW)（3个环境）等9个性状主效应QTLs共59个，分布于第1、2、3、5、6、7、8、9、10、11染色体上，单个QTL对各自性状表型变异的贡献率为4.69%-23.74%。其中，充实度A检测到2个QTLs，充实度B检测到5个QTLs，充实度C检测到5个QTLs，检测到灌浆持续期、穗实粒数、穗总粒数、结实率、有效穗和百粒重的主效应QTLs分别为6、12、11、6、8和4个。上述QTLs中，能在2个以上环境检测到有：qGPC-1、qGFD-6-1、qGFD-7-1、qGFD-7-2、qNFGP-6-1、qGNPP-2、qGNPP-3-1、qGNPP-7-2、qPP-2、qPP-3-1、qWG-3-1、qWG-6-2。5.对富阳试验点2010年的2个灌浆时期（花后第10d、15d）和2011年的3个灌浆时期（花后第10d、15d、20d）的籽粒Q酶活性进行QTL定位，共检测到9个影响Q酶活性的主效应QTLs，分布在第2、3、5、6、7染色体上，单个QTL可解释其表型变异介于5.68%-11.59%。其中影响花后第10d籽粒Q酶活性的QTL为7个，花后第15d籽粒Q酶活性的QTL为2个，花后第20d籽粒Q酶活性的QTL为1个。影响花后第10d籽粒Q酶活性的qQ10-3与花后第15d籽粒Q酶活性的qQ15-3-2同处于第3染色体RM282-RM6283区间，该区间同时检测到控制灌浆速率的QTL，揭示了Q酶活性与灌浆速率呈正相关的内在联系。6.在3个环境下（富阳2010、富阳2011、贵阳2011）共检测到控制籽粒长、粒宽、粒厚、籽粒体积和籽粒长宽比5个性状的26个主效应QTLs。其中，粒长共检测到4个QTLs，分布于第3、7、10、11染色体上，单个QTL可解释的表型贡献率在3.98%-54.58%之间，位于第3染色体RM6283-RM7370区间的qGl3-1在3个环境下均被检测到；粒宽检测到4个QTLs，位于第3和第6染色体上，单个QTL可解释6.98%-9.77%的表型变异，位于第3染色体RM7370-RM16区间的qGw3-2在3个环境均能检测到；粒厚检测到8个QTLs，分布于第1、2、3、6、8、11、12染色体上，单个QTL可解释表型变异的4.91%-11.92%， qGt3-1在2011年贵阳和2010年富阳环境下被检测到；籽粒体积共检测到5个QTLs，位于第3、4、6、7、11染色体上，单个QTL可解释表型变异的5.07%-22.35%，位于第3染色体RM6283-RM7370标记区间的qGv3-1能在3个环境都被检测到，表达稳定；检测到籽粒长宽比QTL5个，分布于第3、10、11染色体上，单个QTL可解释表型变异的5.17%-47.81%，位于第3染色体SSR标记RM6283-RM7370区间qGlwr3-1在3个环境均被检测到，贡献率都超过30%。第3染色体上RM6283-RM7370区间均检测到控制粒长、粒厚、籽粒体积和籽粒长宽比的QTL，且表型贡献率较大。7.应用基于MCIM的分析方法，采用QTLNetwork2.0软件对多环境下籽粒灌浆速率等相关性状表型值进行联合检测，共检测到8个性状（强势粒最大灌浆率、强势粒阶段粒重、弱势粒阶段粒重、弱势粒阶段灌浆速率、灌浆持续期、粒宽、粒厚和粒长宽比）的24对加性×加性上位互作效应的QTL，贡献率介于0.26%-4.16%；检测到的主效应QTLs与环境互作效应远小于自身的加性效应。8.本研究发现多个重要的QTL聚集区，可同时影响籽粒灌浆速率、籽粒重、粒形、Q酶活性等性状，阐释了灌浆速率与产量相关性状的密切相关关系。例如在第3染色体上RM282-RM16区间检测到强弱势籽粒的最大灌浆速率、平均灌浆速率、阶段灌浆速率、阶段籽粒重以及粒形、Q酶活性等QTLs，第6染色体上RM136-RM6302区间检测到强弱势籽粒的阶段灌浆速率和阶段籽粒重、Q酶活性、籽粒充实度和穗实粒数等QTLs，第7染色体上RM5436-RM3670区间检测到强弱势籽粒的阶段灌浆速率和阶段籽粒重、Q酶活性、穗实粒数、穗总粒数和有效穗等QTLs。更多还原

【Abstract】 Grain filling, as an important feature in rice, directly affects the production. At present, the studiesof grain filling characters such as grain filling rate mainly focused on physiology and biochemistry.However, because the measurement of this trait is complicated, the research using genetic population atthe level of genes is very little. In order to explore the related gene expression during rice grain fillingand obtain stable QTLs with expression in multiple environments, recombinant inbred lines (RIL)population derived from Xieyou9308, which was combined with the molecular genetic linkage map,was used as materials to identify and analyze the QTLs for grain filling rate of superior and inferiorgrains, developmental behaviour of grains, grain filling duration, grain plumpness, grain number perpanicle, number of filled grain per panicle, productive panicles, spikelet fertility, Q enzyme activity,grain length, grain width, grain thickness, grain volume, grain length to width ratio and100-grainweight under different conditions. The main results were as follows:1. The QTLs for the maximum and average grain filling rate of superior grains across fourenvironments (Guiyang2010,2011; Fuyang2010,2011) and inferior grains across two environments(Fuyang2010,2011) were determined using winQTLCart2.5, and13major QTLs were detected onchromosomes1,2,3,4,5,6, and8. A total of5and6QTLs for the maximum and average grain fillingrate of superior grains were detected respectively, and the detected QTL individually accounted for5.12%-9.56%of the phenotypic variation. Among the QTLs, qSGFRmax-4, qSGFRmax-8,qSGFRmean-4and qSGFRmean-8could be detected under two environments. In addition, a total of4and3QTLs for the maximum and average grain filling rate of inferior grains were detected respectively,and the detected QTL individually accounted for7.46%-14.28%.2. Grain filling rate of superior grains at five stages (6d,12d,18d,24d, and30d after anthesis) andaverage grain filling rate of superior grains (1-6d,7-12d,13-18d,19-24d and25-30d) across fourenvironments were used for QTL analysis. A total of21major QTLs were detected on chromosomes1,2,3,4,5,6,7,8,9and11. The detected QTL individually accounted for5.05%-18.09%of thephenotypic variation. Among the QTLs,14and13QTLs for grain filling rate at five stages and averagegrain filling rate at five stages were detected respectively, and6common QTLs were found. qSGFR-3-1could be detected at multiple stages under three environments, and phenotypic variance explained byindividual QTL ranged from7.16%to18.09%. The maximum LOD value was5.64and positive allelecame from Xieqingzao B. A total of major14and11QTLs for grain filling rate of inferior grains at fivestages and average grain filling rate of inferior grains across two environments were detectedrespectively, and7common QTLs were found. Among these QTLs, qSGFR-3-1in the intervalRM136-RM6302on chromosome6could be detected at the same stages in the two trials. Furthermore,it could be detected at four stages in Fuyang at2011, and phenotypic variance explained by individualQTL ranged from6.91%to14.69%. QTLs for grain filling rate of superior and inferior grains weredetected in intervals including RM282-RM6283and RM7370-RM16on chromosome3, RM136-RM6302and RM3724-RM3330on chromosome6, and RM5436-RM3670on chromosome7.3. Developmental behaviour of grains at five stages across four environments was used for QTLanalysis by using conditional composite interval mapping method. A total of21unconditional andconditional QTLs for superior grain weight were detected on chromosomes1,2,3,4,6,7,8and9,explaining5.25%-18.86%of phenotypic variation. Among these QTLs, there were16unconditionalQTLs,11conditional QTLs, and6corresponding QTLs. A total of16unconditional and conditionalQTLs for inferior grain weight at five stages across two environments were detected on chromosomes1,2,3,5,6,7,8and10, explaining5.73%-18.05%of phenotypic variation. Among these QTLs, therewere12unconditional QTLs,10conditional QTLs, and6corresponding QTLs. qIWG-6-1in theinterval RM136-RM6302on chromosome6was detected at four inferior grain filling stages, and therewas corresponding conditional QTL, suggesting that this QTL may play an important role in inferiorgrain filling. QTLs for grain filling rate of superior and inferior grains and grain weight were detected inintervals including RM282-RM6283and RM7370-RM16on chromosome3, RM136-RM6302andRM5436-RM3670on chromosome7, which was consistent with the fact that grain filling rate shows asignificantly positive correlation with grain weight. The results may provide four important intervals forfurther study on grain filling.4. A total of59major QTLs for nine traits including GPA, GPB, GPC, GPD, NFGP, SF, PP, and100-GW across four environments were detected on chromosomes1,2,3,5,6,7,8,9,10and11. TheQTL individually accounted for4.69%-23.74%of the phenotypic variation. A total of2,5,5,6,12,11,6,8and4QTLs for GPA, GPB, GPC, GPD, NFGP, SF, PP, and100-GW were detected respectively.Among these QTLs, qGPC-1, qGFD-6-1, qGFD-7-1, qGFD-7-2, qNFGP-6-1, qGNPP-2, qGNPP-3-1,qGNPP-7-2, qPP-2, qPP-3-1, qWG-3-1, and qWG-6-2were detected across two or more environments.5. Q enzyme activities at two grain filling stages (10d and15d after anthesis) in2010and at threegrain filling stages (10d,15d, and20d after anthesis) in2011in Fuyang were used for QTL analysis. Atotal of9major QTLs were detected on chromosomes2,3,5,6, and7, explaining5.68%-11.59%ofphenotypic variation. There were7,2, and1QTLs for Q enzyme activity at10d,15d and20d afteranthesis, respectively. Both qQ10-3for Q enzyme activity at10d after anthesis and qQ15-3-2for Qenzyme activity at15d after anthesis were located in the interval RM282-RM6283on chromosome3.The QTL for grain filling rate was also detected in this interval, suggesting that Q enzyme activityshows positive correlation with grain filling rate.6. A total of26major QTLs for five traits including grain length, grain width, grain thickness, grainvolume and grain length to width ratio across three environments (Fuyang2010,2011; Guiyang2011)were detected. Among these QTLs,4QTLs for grain length were detected on chromosome3,7,10, and11, explaining3.98%-54.58%of phenotypic variation. qGl3-1in the interval RM6283-RM7370onchromosome3could be detected in the three trials.4QTLs for grain width were detected onchromosome3and6, explaining6.98%-9.77%of phenotypic variation. qGw3-2in the intervalRM7370-RM16on chromosome3could be detected in the three trials.8QTLs for grain thickness weredetected on chromosome1,2,3,6,8,11, and12, explaining4.91%-11.92%of phenotypic variation. qGt3-1could be detected in the trials in Guiyang at2011and in Fuyang at2010.5QTLs for grainvolume were detected on chromosome3,4,6,7, and11, explaining5.07%-22.35%of phenotypicvariation. qGv3-1in the interval RM6283-RM7370on chromosome3could be detected in the threetrials.5QTLs for grain length to width ratio were detected on chromosome3,10, and11, explaining5.17%-47.81%of phenotypic variation. qGlwr3-1in the interval RM6283-RM7370on chromosome3could be detected in the three trials. The maximum phenotypic variance explained by this QTL wasmore than30%. QTLs for grain length, grain thickness, grain volume and grain length to width ratiocould be detected in the interval RM6283-RM7370on chromosome3, and the phenotypic varianceexplained by individual QTL was large.7. QTLs for traits related to grain filling rate under different conditions were analyzed by MCIMusing QTL-Network2.0. A total of24QTL pairs with additive×additive epistatic effects for8traits(the maximum grain filling rate and grain weight of superior grains, grain weight and grain filling rateof inferior grains, grain filling duration, grain width, grain thickness and grain length to width ratio)were detected, explaining0.26%-4.16%of phenotypic variation. QTL×environment interaction effectsof the major QTLs were less than their additive effect.8. Many QTL aggregation intervals were determined in this study, which affected grain filling rate,grain weight, grain type and Q enzyme activity. The results illustrated the close relationship between thefilling rate and yield related traits. For example, QTLs for the maximum grain filling rate of superiorand inferior grains, average grain filling rate, grain filling rate and grain weight at each stage of superiorand inferior grains, grain type, and Q enzyme activity were detected in the interval RM282-RM16onchromosome3; QTLs for grain filling rate and grain weight at each stage of superior and inferior grains,Q enzyme activity, grain plumpness, and number of filled grain per panicle were detected in the intervalRM136-RM6302on chromosome6; QTLs for grain filling rate and grain weight at each stage ofsuperior and inferior grains, Q enzyme activity, number of filled grain per panicle, grain number perpanicle, productive panicles were detected in the interval RM5436-RM3670on chromosome7.更多还原