Characterization of NIST food-matrix standard reference materials for their vitamin C content Jeanice B. Thomas, James H. Yen & Katherine E. Sharpless
Analytical and Bioanalytical Chemistry ISSN 1618-2642 Volume 405 Number 13 Anal Bioanal Chem (2013) 405:4539-4548 DOI 10.1007/s00216-013-6891-4
1 23
Your article is protected by copyright and all rights are held exclusively by Springer-Verlag Berlin Heidelberg (outside the USA). This eoffprint is for personal use only and shall not be self-archived in electronic repositories. If you wish to self-archive your article, please use the accepted manuscript version for posting on your own website. You may further deposit the accepted manuscript version in any repository, provided it is only made publicly available 12 months after official publication or later and provided acknowledgement is given to the original source of publication and a link is inserted to the published article on Springer's website. The link must be accompanied by the following text: "The final publication is available at link.springer.com”.
1 23
Author's personal copy Anal Bioanal Chem (2013) 405:4539–4548 DOI 10.1007/s00216-013-6891-4
RESEARCH PAPER
Characterization of NIST food-matrix standard reference materials for their vitamin C content Jeanice B. Thomas & James H. Yen & Katherine E. Sharpless
Received: 17 January 2013 / Revised: 26 February 2013 / Accepted: 27 February 2013 / Published online: 26 March 2013 # Springer-Verlag Berlin Heidelberg (outside the USA) 2013
Abstract The vitamin C concentrations in three foodmatrix Standard Reference Materials (SRMs) from the National Institute of Standards and Technology (NIST) have been determined by liquid chromatography (LC) with absorbance detection. These materials (SRM 1549a Whole Milk Powder, SRM 1849a Infant/Adult Nutritional Formula, and SRM 3233 Fortified Breakfast Cereal) have been characterized to support analytical measurements made by food processors that are required to provide information about their products’ vitamin C content on the labels of products distributed in the United States. The SRMs are primarily intended for use in validating analytical methods for the determination of selected vitamins, elements, fatty acids, and other nutrients in these materials and in similar matrixes. They can also be used for quality assurance in the characterization of test samples or in-house control materials, and for establishing measurement traceability. Withinday precision of the LC method used to measure vitamin C in the food-matrix SRMs characterized in this study ranged from 2.7 % to 6.5 %. Published in the topical collection Functional Foods and Dietary Supplements with guest editors Melissa M. Phillips and Catherine A. Rimmer. J. B. Thomas (*) : K. E. Sharpless Chemical Sciences Division, Material Measurement Laboratory, National Institute of Standards and Technology, 100 Bureau Drive Stop 8392, Gaithersburg, MD 20899-8392, USA e-mail:
[email protected] K. E. Sharpless e-mail:
[email protected] J. H. Yen Statistical Engineering Division, Information Technology Laboratory, National Institute of Standards and Technology, 100 Bureau Drive Stop 8980, Gaithersburg, MD 20899-8980, USA e-mail:
[email protected]
Keywords Ascorbic acid . Food . Reference material . Standard reference material . Liquid chromatography . Absorbance detection
Introduction Well-characterized reference materials (RMs) and reliable analytical methods are needed in the food testing and nutrition communities to help facilitate compliance with nutritional labeling laws and to provide accurate information on product labels so that consumers can make sound dietary choices. Information about vitamin C content is required on nutrition labels, thereby making certified reference materials with assigned values for vitamin C particularly important for laboratories trying to establish the accuracy of their methods’ results. NIST food-matrix reference materials can be used for method validation and for quality assurance in the analysis of routine or in-house (secondary) control samples, as well as to establish traceability to the International System of Units (SI). Vitamin C functions in a number of major roles in humans: in protein metabolism and the biosynthesis of collagen, L-carnitine, and neurotransmitters; as an aid in the absorption of dietary iron; and as an antioxidant [1]. In its role as an antioxidant, it has been studied for the prevention and treatment of various cancers and cardiovascular disease. Studies have linked high serum levels of vitamin C to a reduced risk of cancers [2]. This reduced risk may be due to vitamin C or may be due to consumption of healthful foods that coincidentally contain vitamin C. Because the human body cannot manufacture vitamin C, it must be obtained through consumption foods or in the form of supplements, thus foods containing high levels might be considered “functional foods” or “nutraceuticals.” Vitamin C is found at high concentrations in fruits and vegetables, particularly citrus, peppers, broccoli, and
Author's personal copy 4540
strawberries; the vitamin C content of a variety of foods can be found in the national nutrient database maintained by the United States Department of Agriculture [3]. A Dietary Reference Intake for vitamin C is established by the National Academies, Institute of Medicine [4]. Because of the human body’s need for dietary sources of vitamin C, it is one of the few compounds found in food for which the U.S. Food and Drug Administration allows a label claim: “high in antioxidants” [5]. Two Standard Reference Materials (SRMs) based on fruits and/or vegetables have been available from NIST for a number of years: SRM 2383 Baby Food Composite and SRM 2385 Slurried Spinach [6–8]. Unfortunately we were unable to assign values for vitamin C in these materials because vitamin C degrades at the high temperatures at which these materials were processed. To fill this void, vitamin C (as ascorbic acid) concentrations have recently been determined in three NIST food-matrix SRMs using a liquid chromatographic (LC)-absorbance method developed at NIST. These SRMs include SRM 1549a Whole Milk Powder, SRM 1849a Infant/Adult Nutritional Formula, and SRM 3233 Fortified Breakfast Cereal. SRM 1549a [9], was prepared to replace SRM 1549 NonFat Milk Powder and RM 8435 Whole Milk Powder. SRM 1849a, a nonfat milk-based hybrid infant/adult nutritional powder, was prepared to replace SRM 1849 Infant/Adult Nutritional Formula [10–12] and SRM 3233 Fortified Breakfast Cereal [13] was prepared as a new addition to the suite of food-matrix SRMs distributed across the full range of fat, protein, and carbohydrate contents. NIST has used a ninesector triangle, developed by AOAC International, in which foods are positioned on the basis of their fat, protein, and carbohydrate content (Fig. 1): one or two foods within a sector are expected to be representative of other foods within that sector. If an analytical method yields accurate results for those one or two foods, it is expected to yield accurate results for other foods within the same sector [14, 15]. NIST extended this thinking to reference materials, and has produced a number of food-matrix SRMs distributed across the triangle. The positions of SRMs 1549a, 1849a, and 3233 in relation to the other food-matrix SRMs are provided in Fig. 1. In a study performed at NIST in 1997, vitamin C in SRM 1846 Infant Formula (which was superseded by SRM 1849) was determined by using ion-exchange chromatography employing a polyamine column with electrochemical detection [16, 17]. This method was initially developed for the measurement of ascorbic acid in human serum and was applied to the analysis of two food matrices, SRM 1846 and SRM 2383 Baby Food Composite, which were being characterized for their vitamin C content at that time. (As mentioned above, SRM 2383 was found to be a poor candidate for certification of vitamin C.) Since the development of the ion-exchange method, a reversed-phased LC method with absorbance detection has been employed at NIST for
J.B. Thomas et al.
the determination of ascorbic acid in food-related SRMs. This LC method allows reliable quantitative measurements in the various matrices with within-day precision ranging from 2.7 % to 6.5 % and between-day precision ranging from 3.4 % to 4.5 %. In this paper, the procedure used to extract vitamin C from the different food matrices mentioned above and the LC-absorbance method used to characterize these SRMs are described. Results are compared to those provided by the other data used to value assign vitamin C in each material.
Materials and methods1 Calibration standards Stock solutions of ascorbic acid (vitamin C; CAS-50-81-7; Fluka, BioChemika, Buchs, Switzerland) were prepared by dissolving each compound in 0.1 mol/L HCl. Three calibration standards (ranging from 1.0 mg/g to 4.0 mg/g of ascorbic acid) were independently prepared from these solutions. The concentrations of vitamin C in the calibration standards were determined spectrophotometrically using molar absorptivity 550 dL/g•cm±30 dL/g•cm at 243 nm. Because the maintenance of pure and stable primary reference compounds for this analytes is difficult, detector responses were calibrated against solutions whose concentrations were determined by spectrophotometry with corrections made for purity (>99 %) as determined by LC with absorbance measured at 243 nm. The uncertainty in the purity measurement represents the standard deviation of a single measurement and is less than 1 %. Karl Fischer analyses performed by Galbraith Laboratories, Inc. (Knoxville, TN, USA) also confirmed the moisture content (< 0.020 %) of the vitamin C reference compound. 4-Pyridoxic acid (4-PA; Sigma-Aldrich, St. Louis, MO, USA) was used as the internal standard. HPLC-grade solvents were used without further purification. Description of NIST food-related SRMs characterized in this study SRM 1549a Whole Milk Powder SRM 1549a is powdered whole milk containing about 26 % fat. Seven 50-pound bags of whole milk powder were purchased from a commercial manufacturer. The material was packaged as received in heat-sealed nitrogen-flushed aluminized plastic bags by High-Purity Standards (Charleston, SC). Each packet
1
Certain commercial products are identified to specify adequately the experimental procedure. Such identification does not imply endorsement or recommendation by the National Institute of Standards and Technology, nor does it imply that the materials identified are necessarily the best available for the purpose.
Author's personal copy Characterization of NIST food-matrix standard reference materials
4541
Fig. 1 AOAC fat-proteincarbohydrate triangle with the location of SRM 1549a Whole Milk Powder, SRM 1849a Infant/Adult Nutritional Formula, SRM 3233 Fortified Breakfast Cereal and other NIST food-matrix SRMs (SRM 1544 Fatty Acids and Cholesterol in Frozen Diet Composite, SRM 1546 Meat Homogenate, SRM 1548a Typical Diet, SRM 1566b Oyster Tissue, SRM 1845a Egg Powder, SRM 1946 Lake Superior Fish Tissue, SRM 1947 Lake Michigan Fish Tissue, SRM 2383a Baby Food Composite, SRM 2384 Baking Chocolate, SRM 2385 Slurried Spinach, SRM 2387 Peanut Butter, SRM 3234 Soy Flour, SRM 3274 Botanical Oils containing Omega-3 and Omega-6 Fatty Acids, SRM 3287 Blueberry Fruit)
contains 11 g of whole milk powder. The material is stored under refrigeration (4 °C). SRM 1849a Infant/Adult Nutritional Formula SRM 1849a is a milk-based hybrid infant/adult nutritional powder prepared by a manufacturer of infant formula and adult nutritional products. (This is a non-commercial batch and is not an infant formula; SRM 1849a did not meet the content requirements of the Infant Formula Act of 1980.) A base liquid containing all constituents was conventionally heat processed, homogenized, and spray-dried. The product was packaged into single-use nitrogen-flushed foil pouches, each containing 10 g of powder. The material was stored below 0 °C following packaging and is stored at NIST at −80 °C to enhance longterm stability. Along with many of the ingredients currently found in a milk-based infant formula (e.g., vitamins, longchain polyunsaturated fatty acids, nucleotides, and elements), this material contains some nutrients at levels not permitted in infant formula. SRM 3233 Fortified Breakfast Cereal SRM 3233 is a wheat-based fortified breakfast cereal. Four hundred kilograms of fortified breakfast cereal were purchased, and the cereal was ground and sieved to 180 μm (80 mesh), blended, and bottled by High-Purity Standards (Charleston, SC, USA). The cereal powder was placed in 4 oz amber bottles (each containing 60 g of cereal) that had been flushed with nitrogen. The bottles are stored at controlled room temperature.
Sample preparation Ten packets/bottles of each of the three SRMs described above were selected for analysis using a stratified random sampling scheme. Duplicate samples of 2 g from each of 10 packets/bottles of SRM 1849a, SRM 1549a, and SRM 3233 were dissolved in 30 g to 35 g of HPLC-grade water. Two grams of the 4-PA solution (348 mg 4-PA in 673 g 0.1 mol/L HCl) were added gravimetrically to each mixture. Two grams of 40 % (mass fraction) solution of metaphosphoric acid (MPA; 40 g MPA in 100 g water) was added to stabilize the vitamin C in the mixture. 0.5 g to 1 g of dithiothreitol (DTT) solution (100 mg in 10 mL of 0.5 mol/L potassium phosphate dibasic) was added to the mixture to convert dihydroascorbic acid to total ascorbic acid. The mixture was placed in an ultrasonicating bath for 30 min and then centrifuged (1,000×G) at room temperature for 15 min. A 1 mL aliquot of the extracted mixture was removed and filtered using a 0.45 μm nylon filter prior to LC analysis using the conditions described below. SRM 1849 Infant/Adult Nutritional Formula was prepared and analyzed for quality control. LC apparatus and conditions The LC system used for this work consisted of a ternarypump LC solvent delivery system, a variable-wavelength UV-visible absorbance detector equipped with a deuterium lamp, an autosampler, and a controller/integrator. A 5 μm analytical column (YMC C18 Pro; 250×4.6 mm; YMC, Inc.,
Author's personal copy 4542
J.B. Thomas et al.
Wilmington, NC) held at 26 °C with UV-visible absorbance detection at 243 nm was used for ascorbic acid; 260 nm was used for the internal standard 4-PA. A linear gradient at 0.7 mL/min using as solvent A: initial concentration of 0.02 mol/L potassium phosphate dibasic buffer (pH adjusted to 3.1±0.3 with 7.5 mL phosphoric acid) and solvent B: acetonitrile was employed. Note that the initial concentration of solvent A is 0.02 mol/L and that the pH adjustment will alter the final concentration. All mobile phase compositions were based on volume fractions. The gradient was initially held at 100 % solvent A for 13 min followed by a linear gradient to 75 % solvent B in 20 min. The final composition of 75 % solvent B was held for 5 min before the system was returned to initial conditions of 100 % solvent A over 5 min and re-equilibrated for 10 min. The sample injection volume was 5 μL.
Results and discussion Extraction of ascorbic acid from the SRM matrices Vitamin C occurs naturally in two forms, L-ascorbic acid and Ldehydroascorbic acid (shown in Fig. 2). Ascorbic acid is stable at low pH (
