| Yield potential is very important trait, not only for rice growing under irrigated condition but also in area with limited water. Improving yield potential can also increase water use efficiency. In rice and other cereal crops, grain yield is determined by the source-sink relationship and translocation of photosynthesis products. Many studies have been carried out to investigate the QTL underlying varietal differences for morphological traits of the flag leaf in rice under well watered condition. Most previous study revealed the source-sink related QTL under well watered condition using low throughput DNA markers. However, the possible relationship between source and sink traits under water stress condition as well as the genetic mechanism underlied still remain unknown.The objective of this study to elucidate correlation of sink and source related traits in different water regimes, to detect QTL underlying sink and source related traits in different water conditions and to analyze genetic background effects on the related QTL detection. Three sets of populations consisted of226lines (BC2F8) in the MH63background (MH63-ILs),229lines (BC2F8) in the02428background (02428-ILs) and262(F2:7) recombinant inbreed lines (RIL) were evaluated under well watered condition at Beijing and Hainan in2011/2012and the reciprocal introgression lines (MH63-ILs and02428-ILs) were evaluated under stress irrigation at Hainan in2011/2012. A total of24traits including6source traits (FLL, FLW, FLA, SLL, SLW, SLA),3sink traits (PN, GNP, and TGW),9agronomic traits (HD, PH, PL, FGN, UFGN, SPF, BIO, HI, GWP, GY) and5grain size traits (GL, GW, GT, GLWR, and GV) were used to assess the source sink related traits of the above populations under well watered and drought condition at Beijing in2011(BJ11) and Hainan in2011/2012(HN11/HN12). Genotyping was done by using265evenly distributed SNP markers. Both the additive QTLs and digenic epistatisis were identified. The main results are summarized as following:1. Under well watered condition, source traits were positively correlated with sink traits especially with GNP. Strong consistent correlation was observed for FLW, SLW, FLA, SLA (P<0.0001). Under drought condition, the positive correlation was observed especially among source traits with GNP. Source traits were negatively correlated with PN (panicle number), especially the FLW and SLW under both well watered and drought stress conditions. Mostly source traits were not correlated with TGW under both conditions. 2. Under well-watered condition, all6source traits were also positively correlated with agronomic traits especially with GWP, FGN, PL and BIO. Whereas under drought stress condition, the positive correlations were found between FLL and FLA with GWP and FGN, FLL, FLA, SLL and SLA with PL, and FLW, SLW and SLA with BIO. Under well-watered condition, source traits were not correlated with HI, SPF and UFGN, whereas under drought stress condition, positive correlation was observed between FLL with HI, FLL and FLA with SPF, SLA with UFGN. Their correlation with PH was not consistent, ranged by negative, insignificant to positive correlations. Source traits were positively correlated with GY, especially for the FLA and SLA in HN12across population under optimal condition and the FLL and FLA under drought stress condition. The source traits were found to be positively correlated with GW, GV in02428-ILs and RILs, correlated with GT in MH63-ILs negativly under well watered condition but positively under severe drought stress condition.3. Genetic background greatly affected the expression of detected QTL. In MH63-ILs which averagely have94.45%of MH63genome, the QTL detected were found to be with higher LOD and effect, in comparing to those in02428-ILs which possessed79.62%of02428genome. The least QTLs were found in RILs. A total of502main QTL were identified in three backgrounds for24traits under well watered and drought stress conditions, and most of them (97.0%) have2-11supporting evidences except11QTLs detected under well watered conditions and4QTL under drought stress condition with only single supporting case. A total of114QTL for6source traits were identified in this study on12chromosomes under different water regimes (R2=1.55-49.73%). Among them,49QTL for source traits were detected under well watered condition,51QTLs (44.7%) under both conditions, and14QTL under drought stress condition. Whereas a total of42QTL for sink traits were identified in this study (R2=1.6-42.8%), among them,22QTL (52.3%) were detected both conditions,11QTL under well watered condition and9QTL under drought condition. A total of241QTL for agronomic traits were identified in this study (R2=1.6-55.8%), among them,139QTL (57.7%) were detected under both conditions,59QTL under well watered condition and21QTL under drought condition. A total of105QTL for grain size traits were identified in this study (R2=1.6-66.1%), among them,48QTL(45.7%) were detected under both conditions,50QTL under well watered condition and7QTL under drought condition. Out of140(27.9%) background-independent loci detected with at least4supporting evidences,69QTL (49.2%) were detected across all populations. Under mild drought stress condition, drought induced QTL for SLL was more frequently detected than other source traits QTL (6QTL for SLL,2QTLs for FLA, FLL, SLA, SLW and1QTL for FLW), while FLA and FLL showing higher contribution to the phenotypic variance supported the consistent and higher phenotypic correlation of FLA and FLL with sink size traits under drought stress condition.4. In the QTL mapping level, genetic overlaps were found that some main QTLs for source traits are located to the similar regions where QTL for either sink size, agronomic or grain size traits were detected which supported the relationship among traits in this study. Close linkage or pleiotropy may be responsible for this kind of genetic overlaps. Of the502QTLs identified for24traits,246,278, and216QTLs were detected in02428-ILs, RILs, and MH63-ILs, respectively, shared29,32, and27regions containing at least5QTLs. Region M110-M114(27,293,771-35,249,935bp) on chr.4harboring background-independent loci for FLL, FLW, FLA, SLL, SLW, SLA, GNP detected under both conditions associated with region where QTL for source sink related traits and drought tolerance QTL had been identified in the previous study in our lab as a major gene (SS1). We also identified the genetic overlap region M19-M30(24,694,141-36,369,760bp), region M115-M118(591,451-7,310,209bp) on chr.5and region M222-M226(17,551,097-22,500,927bp) on chr.10where QTL for the source sink related traits were identified clustered with agronomic traits as well as grain size traits under both conditions. These clustered QTL involved independence loci will contribute to our understanding of the genetic control of source and sink under drought conditions at the sensitive reproductive stage and might be considered as candidate loci for MAS which be useful for rainfed rice improvement. Digenic interactions between QTL main effects for sink size traits, agronomic traits and grain size was observed under both well watered and drought conditions, especially in02428-ILs and RILs. In a word, the QTL, especially the genetic overlap regions for the source and sink traits would offer useful information for MAS breeding in rice to improve yield potential under both irrigation and stress conditions. |
PDF Full Text Request