Seeds of the Tartary buckwheat variety “Manten-Kirari” (trace rutinosidase variety) were milled using a test mill (Quadrumat® Junior, Brabender® GmbH & Co., Duisburg, Germany) at the flour milling percentage of 63%. The flour was stored at −20˚C until used for experiments.

Tartary buckwheat flour (100 mg) and 1.0 mL of DMSO were suspended and incubated at 37˚C for 3 hours. Next, to extract rutin and quercetin from the DMSO-suspended flour, 7.2 mL of methanol and 1.8 mL 0.1% phosphoric acid were added to the mixture and stored at 37˚C for 16 hours. After extraction, the sample was centrifuged at 5000 g, 10 min at 20˚C, the resultant supernatant was analyzed using HPLC [17] , rutin and quercetin concentrations were determined. The extracts were filtered through a 0.45 mm filter and applied to HPLC. HPLC was performed on an CAPCELL PAK C18 column (SHISEIDO, Japan) at a flow rate of 1.5 ml/min. Elution gradient program was 0 - 20 min linear gradient from solvent A [methanol-water-phosphoric acid (30:69.7:0.3)] to solvent B [methanol]. A chromatograph was monitored at 360 nm.

Salmonella typhimurium TA100 [22] , TA1535 [22] , TA98 [23] , TA1537 [24] , and Escherichia coli WP2 uvrA were used for the Ames test. Phenotype confirmation, genotype and mutation type detected are shown in Table 1. A preliminary experiment was conducted to optimize the test solution. Tartary buckwheat flour did not completely dissolve in any of the following solutions tested: water, acetone, N,N-dimethylformamide, 1,4-diox- isane and 1,4-epoxybutane. Of these, DMSO resulted in the best dispersion condition; therefore, we employed DMSO as the test solution. Tartary buckwheat flour and DMSO were mixed (12 - 50,000 μg flour/mL_DMSO) and subjected to sonication for approximately 2 minutes.

The S9 microsomal fraction was used as a metabolic activation system. As a positive control, B[a]P and 2AA were used in the presence of S9, while AF-2, NaN₃ and ICR-191 were used in the absence of S9. The DMSO was used as a negative control. A standard preincubation assay [25] [26] was performed. In the plate incorporation method, two replicates were conducted per dose group.

In 2 mL of the overlay agar without S9, 0.1 mL of bacterial culture (2.3 - 5.7 × 10⁹ bacteria), 0.1 mL of sample solution and 0.5 mL of 100 mM sodium phosphate buffer (pH 7.4) were added and mixed. Bacterial concentration was calculated using optical density. In the S9 added overlay agar, 0.5 mL of S9 mix was substituted for 0.5 mL of the sodium phosphate buffer, and the remainder of the protocol was as for the culture medium without S9 mix. Next, the molten overlay agar was added to the minimum salts agar containing 0.6% (w/v) agar and 0.5% (w/v) NaCl. After allowing the medium to harden, the plates were incubated at 37˚C for 48 h and the number of revertants was recorded. These experiments were performed at BML, Inc. (BML General Laboratory, Kawagoe, Saitama, Japan) under contract from the New Drug Research Center, Inc. (Eniwa, Hokkaido, Japan).

Although the rutinosidase activity in “Manten-Kirari” is two or three orders magnitude less than that of the common variety, some rutinosidase activity remains. Therefore, we investigated the hydrolysis of rutin by trace amounts of rutinosidase in “Manten-Kirari” flour in DMSO. Rutin concentration of DMSO-suspended flour were almost same compared with intact flour (Figure 2). In addition, aglycone of rutin quercetin, which is the product of rutinosidase activity, was not increased in DMSO-suspended flour (Figure 2). This indicates that the rutin in “Manten-Kirari” was not hydrolyzed in DMSO. Therefore, the extract contained rutin as the major polyphenolic flavonoid.

To date, there have been few reports dealing with the mutagenic activity of rutin or Tartary buckwheat. Therefore, prior to start detailed examination, we first investigated a range of Tartary buckwheat concentrations, 12 to 50,000 μg/mL, as shown in Table 2. The positive controls showed an increase in the number of revertants both with and without S9 compared to the negative control. Also, the S9 mix or sample solution was confirmed as sterile. In the tested samples, colony numbers in “Manten-Kirari” flour did not differ from the negative control at all concentrations with or without metabolic activation for all bacteria tested. Notably, we observed precipitation with Tartary buckwheat concentrations >3130 μg/mL. In response, we used the concentration range 200 - 3130 μg/mL for a detailed examination, and the results are shown in Table 3. The Ames test showed no

Data are means of two independent experiments.

Data are means of two independent experiments.

increase in the number of revertants for each bacterial strain tested with or without S9. In this paper, all experiments were performed in duplicate; therefore, statistical analysis could not be applied. However, the results of “Dose optimization of Tartary buckwheat flour for bacterial mutagenic assessment with or without S9” (Table 2) and “Effect of various doses of Tartary buckwheat flour on bacterial mutagenicity with or without S9” (Table 3) were almost identical in the number of revertants for each sample. Therefore, although statistical analysis could not be conducted, the results show high reproducibility. From these results, it is suggested that the flour of the Tartary buckwheat variety “Manten-Kirari” does not exhibit genotoxicity. In addition, dough at a dose of 5000 mg flour/ kg is non effect level at acute and subacute test using experimental animals [27] . From these results, “Manten-Kirari” flour should be safe. The concentration of rutin in “Manten-Kirari” flour is about 15 mg/g flour. Among several crops, Tartary buckwheat contains a notably high polyphenol concentration. Some papers have described the effect of polyphenols on mutagenic activity, assessed by the Ames test [28] . Therefore, we hypothesized that rutin may affect the number of revertants in the Ames test. However, in this experiment, “Manten-Kirari” flour did not affect the number of revertants in the Ames test, suggesting that rutin is not mutagenic.

“Manten-Kirari” is a promising Tartary buckwheat variety for use in rutin-rich food products; therefore, the results of our mutagenesis analysis provide important information for optimizing its use in the food industry.