FOOD COMPOSITION AND ADDITIVES Optimization of the Treatment of Wheat Samples for the Determination of Phytic Acid by HPLC with Refractive Index Detection ROSA AMARO,MIGUEL MURILLO,ZURIMA GONZÁLEZ,ANDRÉS ESCALONA, and LUÍS HERNÁNDEZ Universidad Central de Venezuela, Facultad de Ciencias, Escuela de Química, Centro de Química Analítica, Laboratorio de Métodos Cromatográficos, Apartado 47102, Caracas 1041-A, Venezuela The treatment of wheat samples was optimized before the determination of phytic acid by high-performance liquid chromatography with refractive index detection. Drying by lyophilization and oven drying were studied; drying by lyophilization gave better results, confirming that this step is critical in preventing significant loss of analyte. In the extraction step, washing of the residue and collection of this water before retention of the phytates in the NH2 Sep-Pak cartridge were important. The retention of phytates in the NH2 Sep-Pak cartridge and elimination of the HCl did not produce significant loss (P = 0.05) in the phytic acid content of the sample. Recoveries of phytic acid averaged 91%, which is a substantial improvement with respect to values reported by others using this methodology. P hytic acid (inositol hexakisphosphate, InsP 6 ) is widely distributed in commonly consumed foods. It is found in high concentrations in the seeds of grains, pulses, and oleaginous products, and in lesser amounts in tubers and garden produce. The amount of phytate in food has been of interest to nutritionists for many years. Originally perceived as an antinutrient for its ability to precipitate and decrease the bioavailability of minerals such as calcium, iron, and zinc, phytate is now also credited with positive attributes related to the prevention of oxidation, cancer, atherosclerosis, and kidney stones (1, 2). The determination of phytic acid has constituted an analytical problem for a long time. Until 1980, phytic acid was exclusively determined through nonspecific precipitation methods (3). Currently, procedures in which HPLC is included are more and more common (4). The problem with the classic methods of analysis is that they cannot differentiate between InsP 6 and its lower phosphate forms, which can overestimate InsP 6 content (5, 6). One of the chromatographic techniques used to determine InsP 6 in food is HPLC using ion-pairing with refractive index (RI) detection (7–16). Nevertheless, recent studies have reported low recoveries (<72%) of InsP 6 determined in cereals by this technique (16–18). In the work described in this paper, each of the wheat treatment steps for the extraction of phytic acid and its subsequent determination by HPLC with RI detection were studied, with the purpose of improving the recoveries achieved by this technique. Experimental Materials Sodium phytate, 79%, w/w (dodecasodium salt hydrate), determined by elemental inductively coupled plasma-optical emission spectrometry phosphorus analysis; phytic acid, dipotassium salt, 95%, w/w; tetraethylammonium hydroxide (TENOH; 40%, w/w, in aqueous solution); tetrabutyl- ammonium hydroxide (TBNOH; 1.0 M, in methanol solution); and trace select ultra HCl, 30%, w/w, were supplied by Aldrich Chemical Co. (Milwaukee, WI). NH2 Sep-Pak cartridges were supplied by Millipore Corp. (Milford, MA). Ion-exchange resin (AG 50W-X4 of H + , 50–100 mesh) was from Bio-Rad Laboratories (Richmond, CA); 99.8% high-purity methanol from Riedel-de Haën AG Co. (Seelze, Hannover, Germany); 90%, w/w, purity formic acid from BDH Laboratory Reagents Co. (Poole, UK), and 18 MW/cm deionized water. Instrumentation The HPLC system consisted of a Hewlett-Packard (Waldbronn, Germany) 1100 solvent delivery system, a Rheodyne 9125 6-port injection valve (Rheodyne, Cotati, CA) with a 20 mL PEEK (poly-ether-ether-ketone) injection loop, a Hamilton PRP-1 PEEK column (Hamilton, Reno, NV; 5 mm, 150 ´ 4.6 mm id), and an RI detector Model 1047 A (Hewlett-Packard). The analog output from the detector was connected to A/D converter Model SS420, and the data were processed with EZCHROM software, provided by Scientific Software, Inc. (Pleasanton, CA). Connections between injection valves, injection loop, and pump were made with 0.250 mm id, PEEK material pipe. Solvent was pumped at a flow rate of 0.6 mL/min. The mobile phase was 0.015 M formic acid solution containing 5%, w/w, methanol and 0.28%, w/w, TENOH. The column temperature was kept at AMARO ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 92, NO. 3, 2009 873 Received February 8, 2008. Accepted by SG October 17, 2008. Corresponding author’s e-mail: rosa.amaro@ciens.ucv.ve Downloaded from https://academic.oup.com/jaoac/article/92/3/873/5656010 by guest on 10 September 2021