QTL Mapping And Association Analysis Of Stigma Exsertion In Rice

Rice is one of the most important food crops of the developing countries and is the staple food of more than half of the world?s population.Application of hybrid rice is an important and reliable way to fulfill the increasing world population.Although hybrid rice has been commercialized,the relatively lower outcrossing rate leads to a lower rate of hybrid seed production and a higher cost for hybrid seed production and as a result,the adoption of hybrid rice is limited in parts of the world.Stigma exsertion is an important factor that influences the outcrossing ability of rice.Since the seed setting rate of the spikelet with exserted stigma is more than three times higher than those without exserted stigma,improving the stigma exsertion of rice sterile lines can increase their outcrossing rate.For the past few years,researchers have detected many QTLs that influence stigma exsertion but most of them are with minor effect and till now,no one has successfully used stigma exsertion QTLs to improve the stigma exsertion of rice sterile lines.In this study,we identified two major stigma exsertion QTLs,q SE2 and q SE6 through linkage analysis.Through association analysis,we detected three major loci and which were just co-located with three grain shape genes and we confirmed the function of these three loci.The major results of the research are as follows:1.Using a BC1F2 population developed from HL type CMS line,Chaotai A and a PTGMS line,Guangzhan 63 S,a linkage map containing 146 SSR markers was constructed and used for mapping QTL conferring stigma exsertion.A major QTL,q SE2,was located on the short arm of the second chromosome,explaining 46% of the variation for stigma exsertion.NIL(Chaotai A)had a fifty percent higher stigma exsertion than NIL(Guangzhan 63S)and NIL(Guangzhan 63S)displayed a pollen abortive type.Thus,we concluded that the QTL q ES2 was the gene tms5.2.Molecular markers that were closely linked to tms5 were developed to detect the fertility of the F2 population derived from Pokhreli and HD9802 S and 192 homozygous fertile plants were finally used to construct the linkage map.Three QTLs conferring stigma exsertion,q SE4,q SE6 and q SE9,were detected.The major QTL was q SE6 which explained 29% of the variation for stigma exsertion in the F2 population.However,the allele from HD9802 S which had a lower stigma exsertion could increase the stigma exsertion of the population and NIL(Pokhreli)displayed a 30% lower stigma exsertion rate than NIL(HD9802S).3.Genome Wide Association Study was performed to identify loci associated with stigma exsertion and several floral traits.More than 50 significant loci associated with stigma exsertion were identified.Stigma exsertion has high heritability in pure line and is a typic quantitative trait controlled by many minor genes.5 loci highly associated with stigma exsertion explained up to 39% of the phenotype variation in mixed linear model,indicating that the additive effect was also very significant.4.The predicted stigma exsertion genes for at three loci of stigma exsertion were co-located with grain size genes,namely GS3,GW5 and GW2.Transgenic plants with these genes were used to confirm the results of association analysis.Through further investigation on the stigma traits of these three transgenic plants,we found that GS3 could reduce the style length,GW5 could increase style length,stigma length and stigma width and GW2 could decrease the stigma size.The grain shape of these transgenic plants was the same as reported,which indicated that the effect of GS3,GW5 and GW2 on stigma exsertion was caused by their effect on the hull shape and the stigma size.5.Genome-wide association analysis of grain length and grain width suggested that GS3 and GW5 were the major genes controlling grain shape and they occur frequently in rice accessions.Functional markers were used to identify the genotypes of the 529 accessions.Through regression analysis,we found that GS3 and GW5 could independently explain more than 20% of the genetic variance and together could explain more than 30% of the variance,which indicated that GS3 and GW5 were two major genes affecting stigma exsertion.6.Four genotype combinations of GS3 and GW5 were divided to elucidate the feasibility of applying these two genes to improve the stigma exsertion of rice.By comparing the stigma exsertion and yield of these four genotype combinations,we found that the combination,GW5gs3,had the highest yield and the highest stigma exsertion.The genotype combination gw5GS3,which showed the highest frequency in cultivated rice shared the lowest stigma exsertion and yield.Thus,It is feasible to use the advantageous combination GW5gs3 to improve the stigma exsertion of rice.7.Combination GW5gs3 mainly existed in Ind II and Tr J and gw5GS3 mainly existed Ind I and Te J.These different combination distributions indicated that selection might have occurred in these subpopulations.We analyzed the linkage disequilibrium(LD)of these two genes and found that strong LD existed between them no matter taking the population structure into account or not.Through further calculating the nucleotide diversities of the four combinations,we noticed that GW5gs3 displayed the lowest diversity.So the selection of the two genes during the breeding process may lead to the different distribution of GW5gs3 and gw5GS3.