1a. Objectives (from AD-416)
Objective 1: Develop and evaluate procedures for quantitative extraction and/or fractionation of food materials by polarity. Sub-objective 1.A.: Develop an extraction procedure for sequential fractionation of the major groups of components from plant materials. Sub-objective 1.B.: Develop optimized extraction procedures for accurate quantification of individual phytochemicals in plant materials. Objective 2: Develop and evaluate spectral fingerprinting and chromatographic profiling methods to characterize components in lipid soluble, water soluble, and intermediate fractions of food materials. Sub-objective 2.A.: Develop spectral fingerprinting methods for identification of plant materials and individual components using direct analysis (no prior chromatographic separation) and pattern recognition algorithms. Sub-objective 2.B.: Develop chromatographic profiling methods for identification and quantification of individual components in plant materials. Objective 3: Develop methods to determine variability of biologically active components in food materials through profiles and/or fingerprints.
1b. Approach (from AD-416)
Objective 1: Methods will be developed for the quantitative extraction of macro and micro components from plant materials using commercial, high pressure/temperature extraction instrumentation. Soybeans will be tested initially as they contain both lipid soluble and highly polar molecules of health interest. Sequential and parallel extraction will be investigated. Extracted materials will be characterized using liquid chromatography with diode array and electrospray ionization/mass spectrometric detection (LC-DAD-ESI/MS). The new method will be applied to foods and botanical materials. In addition, optimized methods will be developed for specific families of compounds such as water-soluble vitamins, lipid-soluble vitamins, phenolic acids, and flavonoids. Objective 2: Spectral fingerprinting methods will be developed based on ultraviolet and visible molecular absorption (UV/Vis), infrared (IR), near-infrared (NIR), and mass spectrometric (MS) detection. The overlapping complex spectra will be interpreted using pattern recognition programs. The patterns will be used to determine the sensitivity of the different detection systems for discriminating between plant materials based on genera, species, variety, growing year, growing site, and processing conditions. These methods will be developed using 3 food materials and 3 botanical supplement materials. Repeat samples will be examined over a period of years to determine the stability of the spectra and the ability to compare spectra of new materials to archived spectra. The phenolic and vitamin content of the plant materials will also be determined using chromatographic profiling using LC-DAD-ESI/MS. This will make it possible to determine which compounds are contributing most to differences arising from the various growing factors. Objective 3: The spectral fingerprints can be used with nested analysis of variance to determine the relative variance contributed by each growing factor: species, variety, site, year, plant-to-plant variation, and analytical uncertainty. Samples will be obtained from collaborators across the country and representing a variety of foods and botanical supplements. UV/Vis, IR, and NIR spectra will provide variance data for the integrated chemical composition of the plant materials and MS will provide variance data for specific masses and, with the assistance of chromatographic profiling, specific compounds of health interest.
3. Progress Report
At the end of FY2010, with the support of Thermo Scientific and the Office of Dietary Supplements at NIH, FCMDL was able to purchase a high resolution mass spectrometer (HRMS) that is essential to support future cutting-edge research. The HRMS permits high accuracy measurement of mass which in turn can be translated into a chemical formula and then a compound. The HRMS permits acquisition of metabolomic data of foods, supplements, and botanical materials. We have currently identified more than 300 anthocyanins (the dark pigment flavonoid found in blueberries, plums, etc.) in foods. This data will go into a database that will be available to MS users. During this fiscal year, FCMDL has worked on a quantification scheme for flavonoids. In the first phase, we have identified an approach for quantifying flavonol and flavone glycosides and hydroxycinnamic acid derivatives using a single standard. This scheme is critical since there are more than 4600 of these compounds and only 100 standards (fewer than 3%). FCMDL worked with AOAC INTERNATIONAL to develop guidelines for botanical identification methods. This project is particularly critical to supplement and botanical producers who now must establish appropriate quality assurance methods for all ingredients under FDA’s new GMPs (Good Manufacturing Practices) for dietary supplement ingredients. The guidelines and explanatory appendices are based largely on research at FCMDL. During this fiscal year, FCMDL established initial contact with the University of Maryland Chemistry Department and FDA for the use of FDA's high-field nuclear magnetic resonance (NMR) spectrometer for metabolomic research. The extreme stability and the universal quantification capability of proton NMR make it the ideal tool for characterizing food materials for small molecules and providing reference spectra for different genus and species. FCMDL developed methods for the extraction of phenolic compounds from wheat and determined the phenolic concentration in each ground wheat fraction. We compared the phenolic acid composition and the antioxidant capacity of five spring wheat cultivars from eastern Washington state: two soft white cultivars (Alpowa and Louise), two hard wheat cultivars (Blanca Grande and Macon), and a hard red cultivar (Westbred 936). To date we have compared three commonly used extraction solvent mixtures for extracting soluble phenolics from the grain samples.
1. Identification of Panax species. Identification of plant species and varieties is important for economic, nutritional, and health reasons. This is particularly true for botanical supplements. With the support of an interagency agreement from NIH, FCMDL developed a spectral fingerprinting method that, when used with pattern recognition programs, can differentiate between American ginseng (Panax quinquefolius), Chinese ginseng (P. quinquefolius), and P. notoginseng. Direct analysis of solids or extracts by ultraviolet (UV), near-infrared (NIR), or mass spectrometry (MS) readily identified and differentiated the 3 species and the red and white preparations of the Chinese ginseng. MS and NIR could also identify the growing location (China, U.S., or Canada) for American ginseng. This work demonstrated that differences in the chemical composition of these materials can be readily detected using relatively simple methods. The methods can be used to correlate composition with health benefits and determine levels of adulteration.
Luthria, D.L., Mukhopadhyay, S., Lin, L., Harnly, J.M. 2011. A statistical evaluation of spectral fingerprinting methods using analysis of variance and principal component analysis. Food Chemistry. 65:80-85.