Genetic Analysis For Traits Relating To Outcrossing And Their Sensitivity To Exogenous GA₃ In Rice (Oryza Sativa L.)

Traits relating to outcrossing influencing yield of hybrid seed production directly inrice. They even became limited factors for large-scale popularization of hybrid rice.Identifying favorable alleles of traits influencing outcrossing is a foundation for improvingthe characters relating to outcrossing of male sterile lines through polymerizing meritalleles. Exogenous gibberellin (GA₃) application has been a key technology in hybrid riceseed production. Improving outcrossing rate and sensitivity to GA₃ of male sterile lineshave important role in increasing yield and lowering cost of hybrid rice seed production.In this dissertation, three studies have been conducted using a set of 98 backcrossinbred lines (BILs), which derived from a backcross of Nipponbare (japonica)/Kasalath(indica)//Nipponbare by the single seed descent methods. First, we analyzed the geneticsegregation pattern of 5 traits influencing outcrossing (UIL, PNL, PEL, FDS, AGO) anddetected their QTLs under 3 growing environments (E1 environment: Nanjing, 2005; E2environment: Nanjing, 2006; E3 environment: Sihong, 2006). Second, we studiedsensitivity of 3 seedling traits to GA₃ and the QTLs controlling the sensitivity. Third, westudied sensitivity of 4 traits influencing outcrossing to GA₃ at heading stage and the QTLsgoverning the sensitivity under 2 growing environments (E2 and E3). The results were asfollows:1. Genetic segregation analysis for 5 traits relating to outcrossing were conducted byusing major gene-polygene mixed inheritance models under 3 different growingenvironments. We found:(1) The uppermost internode length was controlled by 3 pairs of major genes pluspolygene. Both the major genes and the polygene had additive-epistatic (additive×additive) effects. Major gene heritability was 55.53%, 92.12% and 77.08%, polygeneheritability was 41.36%, 3.97% and 19.38% under 3 environments, respectively. Additiveeffect of the first pair of major gene was biggest among the 3 major genes.(2) Panicle neck length was controlled by 3 pairs of major genes plus polygene. Both the major genes and the polygene had additive-epistatic (additive×additive) effects.Major gene heritability was 52.60%, 78.43% and 78.32%, polygene heritability was38.30%, 16.30% and 16.44% under 3 environments, respectively. Additive effect of thefirst pair of major gene was biggest among the 3 major genes.(3) Percentage of exerted stigma was controlled by 2 pairs of major genes plus polygene.The major genes expressed cumulated effects. The polygene expressed additive-epistatic(additive×additive) effects. Major gene heritability was 77.12%, 59.79% and 52.41%,polygene heritability was 17.69%, 38.48% and 46.24% under 3 environments,respectively. Cumulated effects of the 2 pairs of major genes were 9.15%, 10.83% and7.97%, respectively, under 3 environments.(4) Flowering duration of spikelet was controlled by 2 pairs of major genes pluspolygene. Both the major genes and the polygene had additive-epistatic (adciitive×additive) effects. Major gene heritability was 49.32%, 29.49% and 39.39%, polygeneheritability was 49.04%, 64.20% and 60.23% under 3 environments, respectively.Additive effect of the first pair of major gene was equal to additive effect of the second pairof major gene, were 9.68min, 9.11min and 10.80min under 3 environments, respectively.(5) For angle of glumes opening, it was controlled by 2 pairs of major genes pluspolygene under 3 environments. Under E1 and E3 environments, both the major genesand the polygene had additive-epistatic (additive×additive) effects. However, in E2environment, the major genes expressed additive effects, and polygene expressedadditive-epistatic (additive×additive) effects. Major gene heritability was 58.34%,55.26% and 57.45%, polygene heritability was 28.10%, 33.00% and 27.93% under 3environments, respectively. Additive effect of the first pair of major gene was biggestamong the 3 major genes.2. We detected QTLs of the 5 traits at genome-wide typeâ… error 5% level by usingcomposite interval mapping method of Win Cartgrapher 2.5 software. The results showedthat:(1) For the uppermost intemode length, 4 QTLs were detected in total, and theseQTLs located on chromosome 1, 3 (2), 6, explained 5.64%-15.21% of observed phenotypicvariance. The qUIL-6 was detected in all 3 environments, explained 5.64%-12.80% ofobserving phenotypic variance, Nipponbare allele increased trait value 1.83-2.87cm cm respectively. The other 3 QTLs were detected in 2 of the environments, explained7.74%-15.21% of observing phenotypic variance, Kasalath carried positive allele, increased2.31-2.81 cm of trait value.(2) 4 QTLs for panicle neck length were detected, which were located onchromosome 1,3,5,10,explained 6.8%-17.76% of observed phenotypic variance. TheqPNL-10 was detected in all 3 environments, explained 11.57-17.76% of observedphenotypic variances, Nipponbare allele increase 1.39-2.04 cm of trait value; qPNL-5 wasdetected in 2 of the environments, explained 6.8-12.88% of observed phenotypic variance,Nipponbare allele increase 1.55cm and 1.67cm of trait value. The other 2 QTLs weredetected only in single environment, explained 10.31% and 13.46% of observed phenotypicvariance, on qPNL-1, Kasalath allele increase 1.72cm of trait value, on qPNL-3,Nipponbare allele increase 1.72cm of trait value.(3) For percentage of exerted stigma, 3 QTLs were detected totally, located onchromosome 3, 4, 5 respectively, and these QTLs explained 4.07-18.39% of observedphenotypic variance. The qPES-4 was detected in 3 environments, explained4.07%-10.27% of observed phenotypic variance, Kasalath allele increased 2.24-4.57% ofPES. qPES-3,qPES-5 were detected in 2 of the environments, qPES-3 explained9.77%-18.39% of observed phenotypic variance, positive alleles was from Nipponbare,increased 3.60% and 4.03% of PES in E1 and E3 environments, qPES-5 explained5.08-16.25% observed phenotypic variance, Kasalath allele increased 2.45% and 5.94% ofPES in E2 and E3 environments.(4) For angle of glumes opening, 5 QTLs were detected totally, located onchromosome 7, 8(2), 12(2) respectively, and these QTLs explained 8.21-11.77% ofobserved phenotypic variance, additive effect range of 0.91-1.13 degree. Only 1 QTL,qAGO-7 was detected in 2 environments, explained 8.82-10.87% of observed phenotypicvariance, its positive allele was from Kasalath, increased 0.95 and 0.91 degree of AGO,others were detected in single environments, qAGO-8a and qAGO-8b positive alleles werefrom Nipponbare, which increased 0.96 and 1.02 degree of AGO, explained 9.45% and11.77% of observed phenotypic variance. Positive allele of qAGO-12a and qAGO-12bwere from Kasalath, increased 1.05 and 1.13 degree of AGO, explained 8.21% and 10.3%observed phenotypic variance.(5) For flowering duration of spikelet, 1 QTL was detected on chromosome 5 in 3environments, explained 8.94%-18.77% of observed phenotypic variance, positive allele were all from Kasalath, increased 6.81-9.43 minutes of trait value, respectively.(6) QTLs of qUIL-3 and qPNL-3, qPES-5 and qFDS-5 were simultaneously detectedin the same region on chromosome 3 and 5.3. A set of 98 backcross inbred lines (BC₁F₁₂), was employed studying differentdisposal of GA₃ in inducing seedling growing and leaf sheath elongating in this research,and mapping quantitative trait loci(QTL) of responsive index of GA₃ by composite intervalmapping method. The result showed that the first leaf sheath length increasedsignificantly than control, but seedling height and the second leaf sheath increased slightlyin GA₃ soaking experiment. In GA₃ spraying experiment, seedling height and the secondleaf sheath length increased significantly, but the first leaf sheath length was closed tocontrol. For responsive index of the first leaf sheath length, 1 QTL were detected onchromosome 1, which explained 14.0% of the phenotypic variance. For responsive indexof the second leaf sheath length, 2 QTLs were detected on chromosome 3 and 12,respectively, which totally explained 25.2% of the phenotypic variance, qIFL-1 andqFSL-la located the same marker interval (C742-R2414), qISL-12 located betweenqSSL-12a and qSSL-12b (C732-G193), indicates that the QTLs of sensitivity to GA₃ wereQTLs of the traits.4 At heading stage, there times of exogenous GA₃ were sprayed on every line of theBIL populations, and QTLs of sensitivity to GA₃ (using responsive index as indicator) weredetected. The result showed that:(1) Five QTLs for responsive index of the uppermost internode length were detected.They located on chromosome 1,2,8(2),12, explained 14.45%-24.80% of phenotypicvariation, respectively. Positive alleles of the two QTLs located on chromosome 8 camefrom Nipponbare, others from Kasalath. The chromosome regions of the QTLs for IUILwere different from that of QTLs for UIL, and also different from that of QTLs forsensitivity to GA₃ of leaf sheath length in seedling stage.(2) One QTL for responsive Index of flowering duration of spikelet was detectedunder E3 environment, explained 21.26% of phenotypic variation. Nipponbare alleleincreased responsive index of sensitivity to GA₃. The chromosome region of the QTLsfor IFDS was different from that of QTLs for FDS.(3) For responsive index of percentage of exerted stigma, 4 QTLs were detected onchromosome 7, 8(2), 9, explained 10.94%-23.99% of phenotypic variation, respectively.Nipponbare allele increased IPES in E2 environment, while Kasalath allele increased IPES in E3 environment. The chromosome regions of the QTLs for IPES were different fromthat of QTLs for PES.(4) One QTL for responsive index of angle of glumes opening was detected in E2environment, explained 21.2% of phenotypic variation, Kasalath allele increase 0.012 ofresponsive index. The chromosome region of the QTLs for IAGO was different from thatof QTLs for AGO.