Phenolic Compounds Analysis in Foods and Dietary Supplements is not the Same Using Different Sample Preparation Procedures

Recent epidemiological studies suggest a positive correlation between diets rich in fruits and vegetables and a reduced incidence of chronic diseases. This beneficial effect is partially attributed to phenolic phytochemicals, a complex group of secondary metabolites that provide flavor and color to fruits, vegetables, and grains. More than 8,000 different phenolic phytochemicals belonging to different subgroups (polyphenols, phenolic acids, and a miscellaneous group) have been identified. The large number of phenolic compounds, their structural diversity, and their interaction with other cellular constituents present a considerable challenge in developing efficient procedures for optimum extraction and accurate analysis of different plant matrices. This paper illustrates the difficulties related to the extraction of phenolic compounds using examples from peer-reviewed literature. It discusses the importance of optimizing sample preparation procedures for accurate estimation of phenolic compounds from foods (eggplant, soybean and parsley) and dietary supplements (Black cohosh). A comparison of current and classical procedures for the extraction of phenolic phytochemicals is presented. The influence of different extraction parameters, such as solvent composition, particle size, pressure, temperature, solid-to-solvent ratio, and number of extraction cycles is also discussed. A systematic approach for optimum extraction of phenolic phytochemicals from different plant matrices has been included. Accurate quantification of bioactive phenolic phytochemicals will allow researchers to provide better guidelines on dietary intake levels necessary to achieve the optimum health. INTRODUCTION The importance of bioactive phytochemicals in human health and nutrition is well documented in peer-reviewed scientific literature and is also widely recognized by food, nutrition, and pharmaceutical industries, as well as by consumers worldwide (Hasler, 1998; Dufresne Farnworth, 2001; Wildman, 2001; Mark-Herbert, 2004; Mandel et al., 2005). Phenolic phytochemicals are secondary metabolites that are widely distributed throughout the plant kingdom (fruits, vegetables, and grains) and are known to provide protection against a wide range of diseases such as coronary heart disease, stroke, and certain types of cancers. Hence, accurate quantification of phenolic phytochemicals in foods and food products is of vital importance to precisely evaluate their role in health and nutrition. Over 8,000 phenolic phytochemicals with wide structural diversity and polarities have been isolated from plant sources, making accurate quantification of phenolic compounds a challenging task (Robbins, 2003; Luthria, 2006). This challenge is further exacerbated by the fact that phenolic compounds are not uniformly distributed in plants and can be covalently bound to other cellular components such as the cell membrane and other macromolecules (Antolovich et al., 2000; Naczk and Shahidi, 2004). Phenolic compounds can be grouped into three broad categories: phenolic acids, polyphenols, and a miscellaneous group. Phenolic acids can be further subdivided into two main subgroups, hydroxylcinnamic and hydroxylbenzoic acids. Similarly, polyphenols can be arranged Proc. IInd IS on Human Health Effects of F&V Ed.: B. Patil Acta Hort. 841, ISHS 2009 382 into two broad classes: tannins (gallic acid, catechin, or epicatechin polymers) or flavonoids (flavones, flavonols, flavanones, flavanols, anthocyanidins). The miscellaneous group is comprised of lignans, lignins, coumarins, stilbenes, and other phenolic compounds not included in the other two subgroups (Luthria, 2006). An analytical procedure generally consists of four steps: 1. sample preparation, 2. analytical separation, 3. detection, and 4. data collection and processing. Significant advancements have been made in the final three steps; however, there has been limited progress in the first, the sample preparation step. Sample preparation is often considered as a rate limiting step and is estimated to account for approximately 30% of the analytical error (Majors, 1999). Sample preparation encompasses multiple steps such as grinding, sieving, extraction, pre-concentration, filtration, and derivatization. This manuscript focuses on the significance of the extraction step, an important part of sample preparation, on the accurate quantification of phenolic phytochemicals from different plant matrices. MATERIALS AND METHODS Dried parsley (Petroselinum crispum) flakes and soybean (Glycine max [L.] Merr.) samples were purchased from Giant and MOM’s (My Organic Market) and local grocery stores in Beltsville, Maryland, USA. Fresh freeze-dried powder of black cohosh (Cimicifuga racemosa) from root and rhizome was obtained from Dr. David Lytle (Eclectic Institute, Oregon, USA). Black bell variety of eggplant (Solanum melongena L.) was obtained from a USDA farm in California, USA. The flesh was chopped into small pieces and lypholized. All samples were ground in a coffee grinder. All ground samples were passed through a standard 20 mesh sieve (particle size < 0.825 mm), mixed thoroughly, and subdivided into multiple aliquots in amber bottles. Each bottle was flushed with nitrogen, and stored in a freezer (<-60°C) until analyzed. Extraction was carried out with different techniques, such as stirring, Soxhlet, rotary shaker, ultrasonic irradiation, and pressurized liquid extractor (PLE). The influence of various solvent compositions on the extraction of phenolic compounds from different matrices was evaluated. The impact of a variety of parameters such as solvent, particle size, pressure, temperature, solid-to-solvent ratio, and number of extraction cycles, which are often ignored or generally considered trivial, were also studied. RESULTS AND DISCUSSION The variability in the phenolic phytochemicals content in foods is well documented in peer-reviewed literature. These variations may be attributed to various factors namely, genotypes, cultivar, growing, storage, processing, environmental, and/or analysis conditions (Escarpa and Gonzalez, 2001; USDA Database, 2003; Ninfali and Bacchiocca, 2003; Vallejo et al., 2003; Giuntini et al., 2005; Anttonen and Karjalainenb, 2005; Luthria et al., 2006; Li et al., 2007). Methodology of Extraction The effect of six commonly used extraction methods on the determination of isoflavones in soybeans by six commonly used extraction techniques (PLE, sonication, Soxhlet, shaker, vortex, and stirring) was described in a recent publication (Luthria et al., 2007). Isoflavones were extracted with a single solvent mixture (dimethyl sulphoxide:acetonitrile:water, 5:58:37, v/v/v) using the same solid-to solvent ratio. The results showed that the best yields (> 95%) of total isoflavones was achieved using the PLE and sonication procedures (Luthria et al., 2007). The efficiency of the other four classical procedures (stirring, Soxhlet, shaking and vortexing) was between 65.6 and 70.4%, as compared to PLE. Earlier, Rostagno, Araujo and Sandi (2002) had compared the extraction efficiencies of soybean isoflavones by sonication, Soxhlet, and supercritical fluid extraction. The authors reported that optimum extractions were obtained with sonication, as observed in our study. The extraction efficiency for Soxhlet was also similar to those reported in our study. The lowest yields (27.7%) of isoflavones were obtained with supercritical fluid extraction. However, in another study, Delmonte, Perry