Identification and characterization of quantitative trait loci affecting spikelet number per panicle in rice (Oryza sativa L.)

Improvement of rice (Oryza sativa L.) yields through breeding requires a good understanding of genetic factors that control component traits, such as spikelet number per panicle (SN/PnN). The objective of this study was to identify and characterize quantitative trait loci (QTLs) for SN/PnN with emphasis on their interactions with growth environments using recombinant inbred (RI) and heterogeneous inbred family (HIF) lines. One-hundred-and-ninety-one RI lines (F-7) derived from a cross between two genetically divergent high-yielding varieties, Milyang 23 (M23, Indica type with a large SN/PnN) and Akihikari (AK, Japonica type), were grown for three crop seasons in both Joetsu, Japan (temperate climate) and Los Banos, Philippines (tropical climate) to detect QTLs for SN/PnN at the LOD threshold of 2.0 using simple interval mapping. QTLs were detected repeatedly in eight genomic regions on chromosomes (chr.) 1 (two regions), 2 (two), 4 (two) and 6 (two). Two closely linked QTLs located on the short arm of chr. 1 (the nearest RFLP marker, XNpb359) increased SN/PnN with the M23 allele more significantly in Joetsu (LOD = 14.0-18.2) than in Los Banos (LOD = 5.7-9.6). The QTL effect detected on chr. 4 (C513) was likewise enhanced in Joetsu, and those on chr. 2 (XNpb298) and 6 (XNpb12) for the Los Banos dry season. Thus, the interaction between QTL and environment was greatly involved in the determination of SN/PnN. For detailed characterization of the large-effect QTLs detected on chr. 1 (XNpb359), 20 HIF lines that segregated for the genomic region with the M23 and AK homozygotes as being isogenic for other regions were evaluated in Los Banos for SN/PnN and other agronomic traits. For the heading stage, the M23 homozygote resulted in a 29% decrease in tiller number relative to the AK homozygote, and for the harvesting stage, the M23 homozygote resulted in >25% increase and decrease in SN/PnN and panicle number (PnN), respectively. These results demonstrate that the QTLs on chr. 1 primarily affect tillering, thereafter SN/PnN and PnN in opposite directions depending on the alleles present, while interacting with the environment. SN/PnN QTLs are further discussed with respect to their positional relations to PnN QTLs previously reported, thereby showing key QTLs for improving either trait without suppressing the other trait. The information on the QTL locations, effects and interactions with the environments would be helpful for genetic improvement of SN/PnN and ultimately for increasing rice yields over different locations. (C) 2004 Elsevier B.V. All rights reserved.