5. AOACSPDSMethods-2018AwardsV3

17

1474 O fitserova & N erkar : J ournal of AOAC I nternational V ol . 99, N o . 6, 2016 Table 2016.10C . The effect of any specific factor was evaluated by comparing the difference between the averages of two subsets of four experiments with 2 ×SD, where SD represents the SD between the replicates performed under the same conditions. Dry green tea extract capsules were used for the ruggedness trial and each experiment was performed in duplicate.

salts, as well as common inactive ingredients. The gelcaps contained glycerin and the softgels contained fish oil, caffeine, lecithin, glycerin, and several plant extracts. None of the dietary supplements had label claims indicating theanine content. NIST SRMs included SRM 3254 C. sinensis (green tea) leaves, SRM 3255 C. sinensis (green tea) extract, and SRM 3256 green tea-containing solid oral dosage form. Only reference (noncertified) mass fraction values for l -theanine were available from NIST. l -Theanine reference values represented data from a single laboratory using an LC-MS method. Performing a postcolumn reaction with ninhydrin reagent is specific for primary amino groups and allows for the selective detection of amino acids in complex matrixes. Lithium cation- exchange columns and lithium citrate buffers represent a chromatographic system designed to separate free amino acids. Only free amino acids and a very limited number of organic amines are retained in a lithium cation-exchange column under the analytical conditions used for this analysis, and therefore could be detected after reaction with ninhydrin postcolumn reagent. l -Theanine peak identity was confirmed by comparing the HPLC elution profiles of l -theanine standard solution with the profiles of the samples by using two types of cation-exchange columns and different sets of buffers as mobile phases. The peaks of l -theanine and l -norleucine (IS) were fully resolved from the other peaks present in the chromatograms with a resolution R S ≥1.5. The chromatograms of dietary supplements analyzed during the course of this study are shown in Figures 1–3. Table 1. SMPR for the determination of l -theanine in tea dietary ingredients and supplements (8) Analytical range, ppm 10–100000 LOQ, ppm ≤ 5 Method performance parameters by range 10–50 ppm 51–500 ppm >500 ppm Recovery, % 80–110 90–107 95–105 RSD r , % ≤7 ≤5 ≤5 RSD R , % ≤10 ≤8 ≤8 Selectivity

M. Calibration Curves

Sample analysis throughout the validation study was performed using a seven-point working calibration curve covering a range from 1 to 40 μg/mL (Table 2016.10A ). Additionally, the linearity of the method was evaluated by building extended calibration curves consisting of 13 mixed calibration standards that ranged from 0.5 to 100 μg/mL. Extended calibration curves were obtained on 6 separate days by two different analysts using different lots of reagents. Calibration standards for each curve were freshly prepared on the day of analysis.

Results and Discussion

Validation Study

The single-laboratory validation (SLV) study was conducted to compare performance characteristics of this method with the characteristics of AOAC SMPR 2015.014, “ Standard Method Performance Requirements for Determination of Catechins, Methyl Xanthines, Theaflavins, and Theanine in Tea ( Camellia sinensis ) Dietary Ingredients and Supplements” (8), which are listed in Table 1. Eight matrixes were used in the validation study: five green tea-containing dietary supplements and three NIST SRMs. The dietary supplements included tablets, dry capsules, liquid formulation, softgels, and gelcaps. According to label claims, all dietary supplements contained green tea extract. The liquid formulation contained up to 45% alcohol, whereas the tablets and dry capsules contained calcium and magnesium Matrixes

Figure 1. Chromatogram of a green tea softgel sample. The concentration of theanine in the sample was 0.1432 mg/g.