Zn²⁺ deficiencies and excess all harms the growth of rice plant and Zn²⁺ excess toxicity easily occurs in acid soil (Obata, 1995). It was reported that no significant effects on rice growth were observed until 600ppm of Zn²⁺ concentration (Takenaga, 1995). In this study, the 71 recombinant inbred (RI) lines from a cross of Asominori with IR24, the molecular data and RFLP map with 375 markers (Tsunematsu et al. 1996), kindly provided by Prof. A. Yoshimura (Kyushu University, Japan) was used to detect QTLs for zinc toxicity tolerance (ZNT) in rice. In this experiment, ten germinated seeds for each RI line including 'Asominori' and 'IR24' were sown at each hill [2.5(L) x 2.5(W) x 4.4(H) cm] in a nursery seeding bed (Made in Takii Seed Co. Japan) filled with commercial heated soil (Made in Miyazaki Yamamune Commercial and placed at outdoor on May, 17, 2004, and soaked in 1000ppm Zn²⁺ solution (pH = 5.4) from the 20th day after sowing. The solution was replaced every two days and the toxicity in rice seedlings was examined 20d after treatment and classified into 11 (from 0 to 10) degrees according to the leaf bronzing severity, where one plant without almost leaf bronzing was determined as "0" and one with all leaf bronzing as "10". All treatments were replicated two times and the average values were used for QTL analysis. The QTL analysis was performed with QTL Cartographer version. 2.0 (Wang et al., 2003) using composite interval mapping (CIM) method. The CIM analysis was calculated using forward regression, the walk speed of 2 cM, and the window size of 10 cM. The locus with LOD score greater than 2.0 was considered as indicative of the presence of a QTL.

Frequency distribution of leaf bronzing degrees in RI population is shown in Fig. 1. There

was a clear difference for ZNT between Asominori (high tolerant) and IR24 (susceptible). Three QTLs, tentatively designated as qZNT-1, qZNT-3, and qZNT-10, were detected for ZNT with LOD value of 6.0 (chromosome 1), 3.2 (chromosome 3) and 2.2 (chromosome 10) (Table 1 and Figure 2) and explained 21.9%, 8.9% and 7.6% of total phenotypic variation respectively. The Asominori allele in qZNT-1 contributed to the increase of ZNT, whereas qZNT-3, and qZNT-10 alleles decreased the trait, which confirmed the continuous variation and transgressive segregation for leaf bronzing degrees in RI population (Fig. 1). The results and the tightly linked molecular markers that flank the QTLs might be useful in breeding for high ZNT varieties in rice.

Acknowledgements

We are greatly indebted to Professor A. Yoshimura (plant breeding laboratory, Agricultural faculty of Kyushu University, Japan) for kindly providing materials, molecular data and valuable advices. We want to thank Japan Society for the Promotion of Science (JSPS) providing the first author to postdoctoral fellowships.

References

Tsunematsu H., A. Yoshimura, Y. Harushima, Y. Nagamura, N. Kurata, M. Yano and N. Iwata, 1996. RFLP framework map using recombinant inbred lines in rice. Breeding Science 46: 279-284.

Wang S., C.J. Basten and Z.-B. Zeng, 2003. Windows QTL Cartographer 2.0. Department of Statistics, North Carolina State University, Raleigh, NC. UAS (http://statgen.ncsu.edu/qtlcart/WQTLCart.htm).

Obata H., 1995. Micro essential elements. In Volume Two (Physiology) of Science of Rice Plant (Matsuo et al., eds) Food and Agriculture Policy Research Center, Tokyo, Japan pp: 402-419.

Takenaga H., 1995. Nutrient absorption of rice plant. In Volume Two (Physiology) of Science of Rice Plant (Matsuo et al., eds) Food and Agriculture Policy Research Center, Tokyo, Japan pp: 247-294.