用(培矮64s/Nipponbare)F2群体对水稻产量构成性状的QTL定位分析

QTL Mapping for Yield Component Traits Using (Pei’ai 64s/Nipponbare) F_2 Population

用水稻测序品种培矮64s和Nipponbare为亲本构建的含137个SSRs标记的连锁遗传图谱和(培矮64s/Nipponbare)F2群体的180个单株,对水稻的单株有效穗数、穗粒数、穗实粒数、结实率、穗着粒密度、千粒重等6个产量构成性状进行了QTL定位分析。共检测到6个性状的22个QTLs,分布在第1、2、4、5、6、9、10、11、12等9条染色体的14个区域,表型贡献率5.0%～19.3%;相关性较强的性状之间具有较多共同或紧密连锁的QTLs;集中分布的QTLs之间既有同向连锁,也有反向连锁。对不同水稻群体定位的同源QTL进行了比较,对QTL在染色体上的集中分布,以及用QTL定位结果和生物信息学方法相结合预测基因的功能等进行了探讨。

Recently QTL （quantitative trait locus） mapping has mostly adopted such markers as RFLP marker, which is costly and difficult to operate. Besides, in such mapping the genome information of the mapping parents is not known, which makes it difficult to compare and share the outcomes achieved by different researchers. QTL mapping, using genetic linkage map constructed by sequenced rice cultivar, will facilitate the application of rice genome outcomes, and can provide a basis for the comparison of different rice QTL mapping and molecular marker assistant selection. It will prove to be a pathway to integrate QTL mapping and genome findings application. QTL for six yield-component traits including panicle number （PN）, spikelets per panicle （SPP）, filled grain number per panicle （ GPP）, seed setting （ SS）, density of panicle （DP） and kilo-grains weight （KW） were investigated using a F2 population consisted of 180 lines derived from the cross between a sequenced indica parent Nipponbare and a partial sequenced parent Pei＇ ai 64s. A genetic linkage map （Fig.2） including 137 SSRs markers was constituted by （Pei＇ ai 64s/ Nipponbare） F2 population , and interval mapping. The forecasts of gene function by QTL mapping and bio-informatics was discussed. The QTL results for the six yield-component traits were listed in Table 3 and Fig. 2. One putative QTL for PN was detected, on chromosome 5, explaining 10.0% of the phenotypic variation. Seven putative QTLs for SPP were detected on chromosome 1, 2, 4, 6, 9, 10 and 11, of which the two （ qSPPI-1 and qSPPIO） on chromosome 1 and 10 had much genetic effect, explained 14.0% and 10.7% of the phenotypic variation, respectively. Three QTLs for GPP were detected on chromosome 4, 5 and 6 the of which, one on chromosome 5 had more genetic effect, and explained 11.9% of the phenotypic variation. Three putative QTLs for SS were located on chromosomes 1, 5 and 6. Four putative QTI~ for DP were detected on chromosomes 1, 10 and 11, and two （ qDP1 and qDPIO-1 ） of them had more genetic effects, explaining 19.3% and 12.8% of the phenotypic variation, respectively. Four putative QTLs for KW on chromosomes 2, 5 and 12 were detected, and two （qKW2-1 and qKW12） of them had much genetic effects, explaining 11.0% and 12.2% of the phenotypic variation, respectively. Many of the traits shared the same QTL, which was consistent with its significant phenotypic correlations. For the QTLs located in the same regions, some of the linkages were beneficial while others were unfavorable. Eleven of the twenty-two QTLs investigated in this study shared similar or identical chromosome numbers and marker intervals with the QTLs obtained by other researchers in their recent studies （Table 4）. For example, the marker intervals of two QTLs for SPP on chromosome 1 and 9 were identical to those by Guo L-B et al., respectively. The marker interval of one QTL for SS on chromosome 6 was identical to that by Jiang G-H et al. These QTLs, repeatedly detected in different genetic backgrounds, are of much practical significance. Those main-effect QTLs are particularly important. This study predicted, by means of bio-informatics, that a protein gene （PO678F11.20） may increase the spikelets per panicle. The outcomes of QTL mapping by using genetic linkage maps constructed by sequenced cultivars, can be used for gene forecasts and may prove to be a new effective approach to expressing gene functions. The accuracy in gene forecasts relies on the density of the markers as well as the number of the mapping population. Thus high-density genetic linkage maps and adequate mapping populations will considerably increase the accuracy in gene forecasts.