A rice diversity research set of germplasm (RDRS) was developed based on a genome-wide RFLP polymorphism survey of 332 accessions of cultivated rice (Oryza sativa L.). The accessions used in the initial survey were selected based on the passport data from the whole collection maintained at the Genebank of the National Institute of Agrobiological Sciences (NIAS). These accessions were analyzed using 179 nuclear RFLP markers. A total of 554 alleles were detected, and the number of alleles per locus ranged from 2 to 8 (mean 3.1). Principal coordinate analysis using RFLP data enabled to classify the accessions into three major groups, one Japonica and other two Indica. To develop a rice diversity research set of germplasm, the RFLP data on the 332 accessions were subjected to cluster analysis and 67 groups were recognized at a similarity index of 0.915. A single accession from each of the 67 groups was selected. These 67 accessions retained 91% of the alleles detected in the original 332 accessions, and covered the variation of the initial set of accessions in terms of several agro-morphological traits. The 69 accessions including varieties from 19 countries and the reference varieties, Nipponbare and Kasalath, were selected for the development of a rice diversity research set of germplasm. This collection which is presently well characterized at the molecular level will be used for the detailed genetic studies and rice improvement.

A mini core collection of Japanese rice landrace was developed based on a genome-wide SSR polymorphism survey of 236 accessions. The accessions used in the initial survey were selected based on the historical record and passport data from the collection maintained at the Genebank of the National Institute of Agrobiological Sciences (NIAS). These accessions were analyzed using 32 SSR markers. A total of 247 alleles were detected and the number of alleles per locus ranged from 3 to 15 (mean 7.7). Principal coordinate analysis using SSR data enabled classification of the accessions into three major groups, two Japonica groups and one Indica group. To develop a Japanese landrace mini core collection, the SSR data on the 236 accessions were subjected to cluster analysis and 50 groups were recognized at a similarity index of 0.88. A single accession from each of the 50 groups was selected. These 50 accessions retained 87.5% of the alleles detected in the original 236 accessions, and covered the variation of the initial set of accessions in terms of several agro-morphological traits.

Rice bran is abundant in bioactive compounds including tocotrienol (T3, unsaturated vitamin E). T3 has been reported about its potential functionalities (i.e., antiangiogenic effect), so much attention has been paid on usability of rice bran T3. Hence, we developed a rapid screening method for T3-rich rice bran by one-step equilibrium direct solvent extraction followed by normal phase high-performance liquid chromatography (HPLC). The method gave high-extraction rate of rice bran T3 and tocopherol (above 90%), and the determination of vitamin E by HPLC was completed within 15 min. Using the method, an average of total T3 content in 109 kinds of rice bran samples was 830 mug/g dry wt. Kouchi-Akamai, Joushuu, and Wataribune were found as the T3-rich rice bran varieties (1350-1430 microg T3/g dry wt). According to T3 ratio against total vitamin E (wt %), the average ratio was 61%. Hirayama, Moritawase, and Kaneko were found as the varieties having the highest T3 ratio (80-86%). Since T3 content in Koshihikari rice bran (the leading variety in Japan) was a little above the average, we cross-fertilized Koshihikari with T3-rich varieties and found that T3 content or ratio in F1 was improved compared with Koshihikari. The varieties found rich in T3 could be used for nutraceutical purpose.