Previous PageTable Of ContentsNext Page

Mapping of QTLs for domestication-related and agronomic traits in a temperate japonica weedy rice

S.N. Ahn, C.S. Oh, S.J. Lee, D.B. Yoon

Dept of Agronomy, Coll. of Agri. & Life Sci., Chungnam Nat’l Univ., Daejeon 305-764, Korea (e-mail


This study was conducted to identify the genetic basis of the domestication-related traits in weedy rice. An RIL population consisting of 80 lines was developed from a cross between the japonica weedy rice, Hapcheonaengmi 3 and the Tongil-type cultivar Milyang 23. The population was genotyped with 133 DNA markers, and also evaluated for 20 traits related to domestication and agricultural performance. A total of 52 QTLs and two loci associated with qualitative variation for pericarp and base coloration were identified using single point and interval analysis. The number of QTLs per trait ranged from one to six. These 52 QTLs were located in 24 intervals distributed on 11 chromosomes. The results indicated that most domestication-related traits clustered in chromosomal blocks, and the positions of many of these clusters were consistent with those reported in previous studies. For 10 (40%) of the QTLs identified for agricultural performance in this study, the Hapcheonaengmi 3-derived allele contributed a desirable agronomic effect despite the overall undesirable characteristics of the weedy phenotype. Favorable wild alleles were detected for days to heading, panicle exertion and primary branch number. When compared with previous studies involving interspecific crosses, it can be concluded that weedy rice is useful as a source of valuable alleles for rice improvement.

Key words

Rice (Oryza sativa L.), quantitative trait locus (QTLs), domestication, weedy rice


Weedy rice usually shows a red pericarp, a high seed dispersal ability and a coloration in various organs including the hull, seed coat and pericarp (Suh & Ha 1994). Although the emergence of weedy forms remains unclear, the finding that weedy types of rice can occur where wild rice is not present, supports the possibility that weedy rice evolves through the degeneration of domesticated rice (Vaughan et al. 2003). Weedy rice has become important resources for breeding and for studying the domestication process of rice (Bres-Patry et al. 2001, Suh et al. 1997). Several studies of their genetic characteristics have been reported; weedy rice strains also appear to be differentiated into indica and japonica types. Several reports indicated that weedy rice possesses useful genes conferring tolerance to various biotic and abiotic stresses (Suh & Ha 1994, Suh et al. 1999). Quantitative trait locus (QTL) mapping using molecular markers has made it possible to understand the genetic architecture of quantitative traits (McCouch & Doerge 1995). In the past decade, a number of studies on Oryza species have identified numerous QTLs underlying various traits (for a summary, refer to The objectives of this study were to map genes and quantitative traits loci (QTLs) involved in the variation of weedy traits using an RIL population derived from a cross between the Koran japonica weedy rice and the Tongil-type rice

Materials and Methods

A population of 80 recombinant inbred lines (RILs) derived from a cross between “Milyang 23” and “Hapcheonaengmi 3”. F1 plants were backcrossed to Milyang 23 to produce BC1F1 plants. These plants were grown in the field and selfed for five generations via single seed descent. Eighty BC1F6 lines were grown in the field at the Chungnam National University, Daejeon, Korea during the summer of 2003. Twenty domestication-related and agronomic traits were evaluated in the BC1F6 families; days to heading (dth), culm length (cl), panicle length (pl), panicle number (pn), seedling height (sdh), panicle exertion (pe), spikelets per panicle (spp), percent seed set (pss), awn (awn), tiller angle (ta), primary branches per panicle (pb), secondary branches per panicle (sb), seed shattering (sh), seed dormancy (sd), grain length (gl), grain width (gw), grain thickness (gt) and length/width ratio of grain (lw).Two qualitative traits, pericarp coloration (pc) and base coloration (bc), were included and evaluated according to the absence or presence of the red color. DNA was extracted from young fresh leaves in bulk of BC1F6 plants as described in Causse et al. (1994). The polymorphic SSR markers between the parents were used for genotype analysis of the 80 BC1F6 lines. The chromosomal location of QTLs was determined by single-point analysis (SPA) and interval mapping (IM). QTL was declared if the phenotype was associated with a marker locus at P <0.001 or with two adjacent marker loci at P<0.05. The total phenotypic variation explained was estimated by fitting a model including all putative QTLs for the respective trait simultaneously.

Results and Discussion

A genetic linkage map was constructed with 2 morphological markers, Rc (pericarp color) and C (base color), 131 SSR markers and 1 telomeric marker, TEL-1a (Yang et al. 2003), for a total length of 1491.5cM with an average interval size 11.3cM (Fig. 1). Significant skewing of allele was observed for 17.4% of the marker used in this study at the significance level of P<0.01 (Fig. 1). Twenty-one (15.9%) of the markers were skewed toward the Milyang 23 allele, while 2 (1.5%) were skewed toward the Hapcheonaengmi 3.

Significant QTLs were identified for all domestication-related and agronomic traits (Fig. 1). A total of 52 QTLs and two genes controlling pericarp and base coloration were identified, and one to six QTLs were detected for the each trait. Most of the QTLs detected in this study were located in the same or adjacent regions as those reported in previous studies. They were distributed on all of the chromosomes except for chromosome 12, and showed clustering in several chromosomal regions (Fig. 1). The genetic basis of the domestication-related traits was relatively simple compared with most of the agronomic traits. For the majority of the domestication-related traits, a few genes with major effects explained most of the phenotypic variation in this population. For example, the seed shattering locus, sh1 explained 15.3% of the phenotypic variance. awn7 explained 20.3% of the phenotypic variance for awn, and pe1 explained 16.1% of the phenotypic variation associated with panicle exertion. These major QTLs were clustered in a few chromosomal regions along with QTLs for other domestication-related traits. A first cluster was located near RM323 on chromosome 1 with QTLs for spikelets per panicle and secondary branches per panicle. The finding that the Milyang 23 allele increased spikelets per panicle at spp1 and secondary branches per panicle at sb1, appear to indicate that a single gene has pleiotropic effects on these two traits. A second cluster was located near RM128 on chromosome 1 with QTLs for seed dormancy, culm length, and panicle

exertion. Of these, the seed dormancy QTL sd1 (R2 = 27.9%), the culm length QTL cl1 (R2 =31.9%) and the panicle exertion QTL pe1 (R2 = 16.1%). It is possible that a single gene has pleiotropic effects on two traits, because culm length and panicle exertion are associated with elongation of stems. Fine mapping of these clusters would be needed to differentiate between tight linkage and pleiotropy of the genes involved. A third cluster was located on chromosome 6, near RM539, with QTLs for shattering, awn, days to heading and grain shape traits, grain thickness and width. Another cluster was found on chromosome 11 with QTLs for awn, days to heading and seed shattering. The results of this study also support previous findings that a few genes with major effects explained a large proportion of the genetic variation associated with the domestication process.

Some of the major chromosomal regions associated with domestication-related and agronomic traits found in this study and the other study by Bres-Patry et al. (2001) were compared. Bres-Patry et al. (2001) evaluated 5 qualitative and 15 quantitative traits and identified 27 QTLs associated with 15 quantitative traits. Comparison indicated that a gene for pericarp coloration and three QTLs associated with three quantitative traits including days to heading were detected in the same or adjacent regions. For pericarp coloration, one gene corresponding to Rc on chromosome 7 was detected in this study whereas two genes corresponding to Rd and Rc were detected on chromosomes 1 and 7, respectively in the study by Bres-Patry et al. (2001). dth3 for heading date in this study was in the similar region as HD3 in the study by Bres-Patry et al. (2001). For seed shattering, the present study detected five QTLs, whereas one QTL was detected in the study by Bres-Patry et al. (2001). Among the five QTLs, sh1 explaining 15.3% of the variation in this study was in the similar region as SHT accounting for 30.2% of the phenotypic variation in the study by Bres-Patry et al. (2001). For awn, five and two QTLs were detected in the present study and in the study by Bres-Patry et al. (2001), respetively. Among them, one QTL, awn4 was in the similar region as AWN4 in the study by Bres-Patry et al. 2001). The QTL clustering phenomenon was documented in both studies; however, the genomic regions exhibiting QTL clustering were differently distributed throughout the genome. This is likely to be the result of different genotype x genotype (G x G) interactions.

The QTLs reported in this study were detected using single point and interval analysis, and a more precise estimate of their location and the magnitude of their effects is required. To address these objectives, a set of substitution lines are being developed, each containing only one or a few chromosomal segments from the Hapcheonaengmi 3 in the background of Milyang 23. Each of the introgression lines is nearly isogenic to Milyang23 and together, these lines provide complete coverage of the Hapcheonaengmi 3 genome (Ahn et al. 2002). The availability of a set of Hapcheonaengmi 3 introgression lines in Milyang 23 background will not only facilitate more rapid utilization of beneficial genes from the Hapcheonaengmi 3 parent, but also provide a powerful tool for the characterization and fine mapping of genes underlying QTLs associated with a wide variety of phenotypes.


This research was supported in part by grants from Crop Functional Genomics Center of the 21st Century Frontier Research Program (CG3121) by the Ministry of Science and Technology, Korea.

Literatures Cited

Ahn SN et al. (2002) Development of introgression lines of weedy rice in the background of Tongil-type rice. Rice Genet. Newsl. 19,14-15.

Bres-Patry et al. (2001) Heredity and genetic mapping of domestication-related traits in a temperate japonica weedy rice. Theor. Appl. Genet. 102, 118-126.

Causse MA et al. (1994) Saturated molecular map of the rice genome based on an interspecific backcross population. Genetics 138, 1251-1274.

McCouch SR, Doerge RW (1995) QTL mapping in rice. Trends Genet. 11,482-487.

McCouch SR et al. (2002) Development and mapping of 2240 new SSR markers for rice (Oryza sativa L.). DNA Research 9,199-207

Suh JP et al. (1999) QTL analysis of low temperature germinability in a weedy rice. Korean J. Breed. 31, 261-267.

Suh HS, Ha WG (1994) Character variations of Korean weedy rice. Rice Genet. Newsl. 11,69-72.

Suh HS, Sato YI, Morishima H (1997) Genetic characterization of weedy rice (Oryza sativa L.) based on morpho-physiology, isozymes and RAPD markers. Theor. Appl. Genet. 94,316-321.

Vaughan DA et al. (2003) Diversity in the Oryza genus. Curr. Opinion Plant Biology 6, 139-146

Previous PageTop Of PageNext Page