One hundred twenty six doubled-haploid (DH) rice lines were evaluated in nine diverse Asian environments to reveal the genetic basis of genotype x environment interactions (GEI) for plant height (PH) and heading date (HD). A subset of lines was also evaluated in four water-limited environments, where the environmental basis of G x E could be more precisely defined. Responses to the environments were resolved into individual QTL x environment interactions using replicated phenotyping and the mixed linear-model approach. A total of 37 main-effect QTLs and 29 epistatic QTLs were identified. On average, these QTLs were detectable in 56% of the environments. When detected in multiple environments, the main effects of most QTLs were consistent in direction but varied considerably in magnitude across environments. Some QTLs had opposite effects in different environments, particularly in water-limited environments, indicating that they responded to the environments differently. Inconsistent QTL detection across environments was due primarily to non- or weak-expression of the QTL, and in part to significant QTL x environment interaction effects in the opposite direction to QTL main effects, and to pronounced epistasis. QTL x environment interactions were trait- and gene-specific. The greater GEI for HD than for PH in rice were reflected by more environment-specific QTLs, greater frequency and magnitude of QTL x environment interaction effects, and more pronounced epistasis for HD than for PH. Our results demonstrated that QTL x environment interaction is an important property of many QTLs, even for highly heritable traits such as height and maturity. Information about QTL x environment interaction is essential if marker-assisted selection is to be applied to the manipulation of quantitative traits.

Grain yield (GY) of rice is a complex trait consisting of several yield components. It is of great importance to reveal the genetic relationships between GY and its yield components at the QTL (quantitative trait loci) level for multi-trait improvement in rice. In the present study, GY per plant in rice and its 3 yield component traits, panicle number per plant (PN), grain number per panicle (GN), and 1000-grain weight (GW), were investigated using a doubled-haploid population derived from a cross of an indica variety IR64 and a japonica variety Azucena. The phenotypic values collected from 2 cropping seasons were analysed by QTLNetwork 2.0 for mapping QTLs with additive (a) and/or additive × environment interaction (ae) effects. Furthermore, conditional QTL analysis was conducted to detect QTLs for GY independent of yield components. The results showed that the general genetic variation in GY was largely influenced by GN with the contribution ratio of 29.2%, and PN and GN contributed 10.5% and 74.6% of the genotype × environment interaction variation in GY, respectively. Four QTLs were detected with additive and/or additive × environment interaction effects for GY by the unconditional mapping method. However, for GY conditioned on PN, GN, and GW, 6 additional loci were identified by the conditional mapping method. All of the detected QTLs affecting GY were associated with at least one of the 3 yield components. The results revealed that QTL expressions of GY were contributed differently by 3 yield component traits, and provide valuable information for effectively improving GY in rice.

Several biologically significant parameters that are related to rice tillering are closely associated with rice grain yield. Although identification of the genes that control rice tillering and therefore influence crop yield would be valuable for rice production management and genetic improvement, these genes remain largely unidentified. In this study, we carried out functional mapping of quantitative trait loci (QTLs) for rice tillering in 129 doubled haploid lines, which were derived from a cross between IR64 and Azucena. We measured the average number of tillers in each plot at seven developmental stages and fit the growth trajectory of rice tillering with the Wang-Lan-Ding mathematical model. Four biologically meaningful parameters in this model--the potential maximum for tiller number (K), the optimum tiller time (t(0)), and the increased rate (r), or the reduced rate (c) at the time of deviation from t(0)--were our defined variables for multi-marker joint analysis under the framework of penalized maximum likelihood, as well as composite interval mapping. We detected a total of 27 QTLs that accounted for 2.49-8.54% of the total phenotypic variance. Nine common QTLs across multi-marker joint analysis and composite interval mapping showed high stability, while one QTL was environment-specific and three were epistatic. We also identified several genomic segments that are associated with multiple traits. Our results describe the genetic basis of rice tiller development, enable further marker-assisted selection in rice cultivar development, and provide useful information for rice production management.