Food Chemistry 170 (2015) 271–280

Contents lists available at ScienceDirect

Food Chemistry journal homepage: www.elsevier.com/locate/foodchem

Chemical investigation of commercial grape seed derived products to assess quality and detect adulteration Tom S. Villani a,b, William Reichert a, Mario G. Ferruzzi c, Giulio M. Pasinetti d,e, James E. Simon a,b,⇑, Qingli Wu a,b,⇑ a

New Use Agriculture and Natural Plant Products Program, Department of Plant Biology and Pathology, Rutgers University, New Brunswick, NJ 08901, United States Department of Medicinal Chemistry, Ernest Mario School of Pharmacy, Rutgers University, Piscataway, NJ 08854, United States Department of Food Science, Purdue University, West Lafayette, IN 47907, United States d Department of Neurology, Mount Sinai School of Medicine, New York, NY 10029, United States e Geriatric Research, Education and Clinical Center, James J. Peters Veterans Affairs Medical Center, Bronx, NY 10468, United States b c

a r t i c l e

i n f o

Article history: Received 5 March 2014 Received in revised form 28 July 2014 Accepted 5 August 2014 Available online 26 August 2014 Keywords: Grape seed extract LC–MS Polyphenol Proanthocyanidins Adulteration Quality control

a b s t r a c t Fundamental concerns in quality control arise due to increasing use of grape seed extract (GSE) and the complex chemical composition of GSE. Proanthocyanidin monomers and oligomers are the major bioactive compounds in GSE. Given no standardized criteria for quality, large variation exists in the composition of commercial GSE supplements. Using HPLC/UV/MS, 21 commercial GSE containing products were purchased and chemically profiled, major compounds quantitated, and compared against authenticated grape seed extract, peanut skin extract, and pine bark extract. The antioxidant capacity and total polyphenol content for each sample was also determined and compared using standard techniques. Nine products were adulterated, found to contain peanut skin extract. A wide degree of variability in chemical composition was detected in commercial products, demonstrating the need for development of quality control standards for GSE. A TLC method was developed to allow for rapid and inexpensive detection of adulteration in GSE by peanut skin. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Grapes and grape-derived products are rich in numerous bioactive dietary polyphenols which have been attributed to improving health and nutrition (Castaneda-Ovando, Pacheco-Hernández, Páez-Hernández, Rodríguez, & Galán-Vidal, 2009; Welch, Wu, & Simon, 2008). Polyphenols from grape-derived products have been associated with the prevention of numerous diseases including cardiovascular diseases, neurodegenerative diseases such as Alzheimer’s disease, as well as several forms of cancers (Aziz, Kumar, & Ahmad, 2003; Bertelli & Das, 2009; Ho et al., 2013; Pasinetti & Ho, 2010; Renaud & de Lorgeril, 1992). Epidemiological and experimental evidence supports the hypothesis that specific grape polyphenols may serve as disease preventative agents (Aziz et al., 2003; Bertelli & Das, 2009; Jang et al., 1997; Pasinetti

⇑ Corresponding authors at: New Use Agriculture and Natural Plant Products Program, Department of Plant Biology and Pathology, Rutgers University, New Brunswick, NJ 08901, United States. Tel.: +1 848 932 6239; fax: +1 732 932 9441 (J.E. Simon). Tel.: +1 848 932 6238; fax: +1 732 932 9441 (Q. Wu). E-mail addresses: [email protected] (J.E. Simon), [email protected]. edu (Q. Wu). http://dx.doi.org/10.1016/j.foodchem.2014.08.084 0308-8146/Ó 2014 Elsevier Ltd. All rights reserved.

& Eberstein, 2008; Renaud & de Lorgeril, 1992). These compounds include proanthocyanidin monomers and oligomers such as catechin, epicatechin, and proanthocyanidin dimers, which are the major constituents in extracts of grape seed (Bertelli & Das, 2009; Fuleki & da Silva, 1997; Ho, Yemul, Wang, & Pasinetti, 2009; Wang et al., 2008). Proanthocyanidins (PACs) are oligomeric conjugates of any combination of the four isomers (±)-catechin and (±)-epicatechin (Fuleki & da Silva, 1997). Two distinct classes of PACs can be defined based on chemical structure, known as A-type (Lou et al., 1999) and B-type PACs (Manach, Williamson, Morand, Scalbert, & Rémésy, 2005). A-type PACs (not found in GSE) contain a 4b ? 8 C–C bond and a 2b ? O ? 7 C–O bond between the two monomer units, whereas the B-type PACs contain only the 4b ? 8 bond, and occasionally the 4b ? 6 bond (Passos et al., 2007) (Fig. 1). Proanthocyanidin oligomers containing ()-epicatechin also occur with differing degrees of galloylation (Rasmussen, Frederiksen, Struntze Krogholm, & Poulsen, 2005). While proanthocyanidins come with a wide range of degree of polymerization (DP), interest is focused on the proanthocyanidin monomers and dimers (DP < 3) because researchers have found that only the monomers and dimers are absorbed into human intestinal tissue and into

272

T.S. Villani et al. / Food Chemistry 170 (2015) 271–280

A

EC

C GA

B2

P2 P2

P2G1

C

P3

P1G1

P3

B

EC

P2 A2 P3 B2 P2G1 P1G1

High weight oligomer “bulge”

12022905.D: UV Chromatogram, 280

C

A2

High weight oligomer “bulge”

5

10

15

20

25

30

Time [min]

Fig. 1. Structures of major compounds found in grape seed extract, GSE (top); representative UV chromatograms at 280 nm of (A) authentic GSE; (B) authentic pine bark extract; (C) authentic peanut skin extract. Abbreviations: GA, gallic acid; C, catechin; EC, epicatechin; P, proanthocyanidin unit; G, galloyl unit; B2, proanthocyanidin B2 [()epicatechin-(4b ? 8)-()-epicatechin]; A2, proanthocyanidin A-type dimer.

circulation in significant quantities (Ferruzzi et al., 2009; Manach et al., 2005; Passos et al., 2007; Wang et al., 2012). A wide range of studies have evaluated the bioactivity of GSE derived products. Protection against gastric mucosal damage induced by ethanolic HCl was shown in rats administered GSE (Saito, Hosoyama, Ariga, Kataoka & Yamaji, 1998). Antibacterial activity of GSE has been demonstrated in a number of studies, both in gram positive and gram negative bacterial strains (Jayaprakasha, Selvi & Sakariah, 2003; Furiga, Lonvaud-Funel & Badet, 2009; Baydar, Özkan, & Sag˘diç, 2004) including methicillin resistant Staphylococcus aureus (Al-Habib, Al-Saleh, Safer, & Afzal, 2010). GSE has also been shown to inhibit pancreatic lipase and lipoprotein lipase, as well as inhibition of lipolysis in 3T3-L1 adipocytes (Moreno, Ilic, Pouley, Brasaemle, Fried & Raskin, 2003). In grape seed extract, typically the B-type PACs have been reported, with highest concentrations of catechin and epicatechin (Wu, Wang, & Simon, 2003), as well as proanthocyanidin B2 and its isomers (Xu et al., 2011). A number of studies have evaluated the proanthocyanidin constituents in grape seed extract. While some variability exists due to source and process used in preparation, reviewed in detail by Liu and White (2012), grape seed extract is typically composed of primarily of PAC monomer and low weight oligomers, with a much smaller concentration of tetramers or greater polymers. Other natural sources of PACs have been reported. Peanut skin extract, for example, is composed principally of A-type PACs with negligible quantities of the monomers, catechin and epicatechin (Sarnoski, Johnson, Reed, Tanko, & O’Keefe, 2012). Pine bark extract has been reported to contain both A and B-type PACs, as well as the monomers (Karonen, Loponen, Ossipov, & Pihlaja, 2004), reported to contain 55 mg/g total PACs (Hellström & Mattila, 2008).

Given the increasing use and myriad commercial products containing grape seed extracts, our research team focused on examining the quality of such plant-based dietary supplements. The Office of Dietary Supplements (ODS) and Food and Drug Administration (FDA) have long raised serious questions as to the safety, efficacy and quality of dietary supplements, especially those derived from plants, citing concerns relating to adulteration, contamination, lack of bioactive compounds, and a lack of compositional standardization which is the principle source of batch to batch inconsistency (Betz, 2006; Ho, Simon, Shahidi, & Shao, 2006; Wang, Liang, Wu, Simon, & Ho, 2006). This study was conducted as part of a larger program which has been examining the uses of grape seed extracts and specific compounds found within grape seed in the treatment of Alzheimer’s disease and mitigation against cognitive memory loss (Ferruzzi et al., 2009; Wang et al., 2012). A wide degree of variability was found during routine vetting of GSE vendors—leading us to question the overall quality and product consistency available to consumers when purchasing this dietary supplement. The objective of this study was to evaluate commercially available GSE derived products by chemical profiling, identify whether adulteration of GSE by other common PAC containing material such as peanut-skin extract or pine-bark extract, and develop a TLC technique to identify adulteration.

2. Materials and methods 2.1. Materials and reagents Twenty-one commercially available products containing grape seed extract were obtained and stored at room temperature in

T.S. Villani et al. / Food Chemistry 170 (2015) 271–280

the dark until use. These 21 products were obtained from a variety of sources including vitamin supplement retailers, supermarkets, and online vendors in order to gather a representative sampling of the products available to consumers. All products purchased were also well within their lot expiration dates as provided by the manufacturer. To ensure no bias introduced based on product branding, sample codes were assigned to each product to obfuscate the identity of the commercial products; this code was assigned by a researcher not involved in subsequent analysis. The products were capsules containing GSE powder; some of the products contained excipients, binders, and/or flow agents which was accounted for in quantitative analysis by normalizing by the ratio of the labeled GSE to mass of powder in each capsule (shown in Table 1). Three of the products (01, 02 and 10) contained labeled resveratrol. Authenticated reference materials were provided by BannerBio Nutraceuticals Inc. (Shenzhen, China). HPLC grade acetonitrile (ACN), methanol (MeOH), ethanol, formic acid (FA), acetic acid, hydrochloric acid, toluene, acetone and HPLC grade water were obtained from Fisher Scientific Co. (Fair Lawn, NJ). Trolox, ABTS, potassium persulfate, and sodium carbonate were obtained from Sigma Aldrich (St. Louis, MO). Folin Ciocalteau Phenol Reagent was obtained from MP Biomedicals LLC (Solon, OH). Vanillin and gallic acid were obtained from Acros Organics (Geel, Belgium). Silica XG TLC plates, 200 lm thickness, with UV254 indicator, were obtained from Sorbent Technologies (Norcross, GA) and stored sealed and dry until use. 2.2. Sample preparation Samples were prepared and analyzed by adapting a method developed previously by our group for the analysis of grape derived products (Xu et al., 2011). All samples were capsules containing powdered material. 10–20 capsules were opened, powder combined, and the capsule shells discarded. Approximately 50 mg of the sample was weighed and placed into a 25.00 mL volumetric flask. 70% methanol (aqueous) with 0.1% acetic acid was added to the flask, and the flask was sonicated for 30 min, conditioned to room temperature, and brought back to volume. Grape seed extract, pine bark extract, and peanut skin extract reference materials were prepared in the same fashion. While authentic grape seed extract is soluble at this concentration, a number of commercial samples contained insoluble material. HPLC samples were prepared by filtering each extract through a 0.45 lm nylon syringe filter into an amber HPLC autosampler vial. Samples were stored at 20 °C until analysis. These solutions were also used for TLC analysis. 2.3. HPLC–UV–MS conditions HPLC separation was performed on a Polaris amide-C18 column, 250  4.6 mm, 5 lm (Varian Inc., Palo Alto, CA). For LC–MS analysis, a Hewlett Packard Agilent 1100 Series LC/MS (Agilent Technologies, Waldbronn, Germany) equipped with autosampler, quaternary pump system, DAD detector, degasser, MSD trap with an electrospray ion source (ESI) was applied, and software for data processing was HP ChemStation, Bruker Daltonics 4.2 and Data Analysis 4.2. HPLC separation was performed with the mobile phase containing solvent A (0.1% formic acid in water) and B (0.1% formic acid in acetonitrile) in gradient: 0–20 min, linear gradient from 10% to 20% B; 20–30 min, linear gradient from 20% to 30% B; 30–40 min, isocratic elution at 30% B; 40–50 min, linear gradient from 30% to 50%; 50–60 min, linear gradient from 50% to 60%. The flow rate was set at 1.0 mL/min with a 1:10 splitter after the output of the UV detector leading 100 lL/min to the ESI-MS. The UV detector was set at 254, 280, 370, 520 nm. The eluent was monitored by

273

electrospray ion mass spectrometer (ESI-MS) scanned from m/z 100 to 1200. ESI was conducted by using a needle voltage of ±3.5 kV (positive and negative mode). High purity nitrogen (99.999%) was used as dry gas at a flow rate of 8 L/min and capillary temperature was at 350 °C. Nitrogen was used as nebulizer at 45 psi, and helium as collision gas. 2.4. Identification and quantification of compounds Gallic acid, as well as proanthocyanidins monomers and oligomers were quantitated or estimated. The identities of catechin, epicatechin, gallic acid, and proanthocyanidin B2 were confirmed by comparison to standard compounds. PAC derivatives were detected at 280 nm and the tentative structures assigned by analyzing the MS and UV data. Quantitation of catechin, epicatechin, and gallic acid was achieved by construction of a calibration curve by serial dilution of standardized reference materials, obtained from ChromaDex (Irvine, Ca.). Quantitation of PAC dimers was achieved by comparison of response to authenticated proanthocyanidin-B2 [()-Epicatechin-(4b ? 8)-()-epicatechin] reference material, obtained from ChromaDex (Irvine, Ca.). Stock solutions of known concentration were prepared in 70% methanol with 0.1% acetic acid and stored in amber vials at 20 °C until use. To estimate the quantity of the proanthocyanidin dimers (P2) of unknown stereochemical configuration, and O-galloylated derivatives of monomer (P1G1) and dimer (P2G1) the calibration response for catechin (for P1G1) or proanthocyanidin B2 (for P2G1 and PAC trimer P3) was adjusted by the ratio of the molecular weights of the derivative to PAC B2. Final results are expressed as a combined measure of stereo/regioisomers. 2.5. Statistical analysis Statistical clustering techniques to reduce the dimensionality of large sets of collected data, which allows for detection of emergent patterns in underlying data sets (Jackson, 2004) is common in quality control analysis of natural products. Clustering analysis is a valuable technique used in the analysis of complex sample matrices, allowing for a grouping of samples based on compositional similarity (Lawaetz, Schmidt, Staerk, Jaroszewski, & Bro, 2009; Xie et al., 2006; Liang, Xie, & Chan, 2004). We hypothesized that this technique could be applied to allow for the differentiation between products containing authentic grape seed extract, and products adulterated with pine bark and/or peanut-skin based the differences in quantitative composition. Clustering analysis was accomplished using R mathematical programming language (R Development Core Team, 2011). Each sample was treated as a linearly independent vector whose dimensions are spanned by the vector-space described by the quantitative data gathered through HPLC analysis. The data was normalized for each individual analyte. Clustering analysis was performed using the Ward clustering method, using the Euclidean distance. 2.6. Assay methods The method for determination of total antioxidants was used as described (Zululeta, Esteve, & Frígola, 2009) with minor modifications, based on the capacity of a sample to inhibit the ABTS radical compared with to TroloxÒ(water-soluble vitamin E derivative). The ABTS radical was generated by single electron oxidation by potassium persulfate (K2S2O8), accomplished by preparing a solution containing 7.46 mM ABTS and 2.44 mM K2S2O8, in deionized H2O. This solution was allowed to stand in darkness at room temperature for 12–16 h (the time required for formation of the radical). The working solution was prepared by taking a volume of the previous solution and diluting it in ethanol until its absorbance at

274

Table 1 Quantitative composition, antioxidant capacity and total phenol content of commercial grape seed extract (GSE) samples and reference materials.a C (mg/g)

EC (mg/g)

P2 (mg/g)

P1G1 (mg/g)

P2G1 (mg/g)

P3 (mg/g)

Total PACs (mg/g)

TEAC (mmol/g)

GAE (mg/g)

Labeled content of GSE/capsule (mg/g)

102.0 ± 5.84 N.D. N.D. N.D. N.D. N.D. 22.1 ± 1.63 N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. 11.5 ± 0.69 10.9 ± 0.55 N.D. 29.7 ± 1.97 35.5 ± 0.77 0.4 ± 0.01 0.7 ± 0.02 N.D. N.D.

16.7 ± 1.31 39.9 ± 0.48 7.9 ± 0.17 22.2 ± 0.04 2.6 ± 0.05 16.7 ± 1.38 15.9 ± 0.33 2.7 ± 0.02 4.4 ± 0.01 79.7 ± 5.54 17.2 ± 0.23 3.0 ± 0.12 3.7 ± 0.14 4.0 ± 0.09 19.8 ± 0.32 N.D. 1.6 ± 0.15 34.9 ± 0.8 34.1 ± 0.78 19.2 ± 1.33 37.0 ± 0.09 58.6 ± 0.94 54.1 ± 0.83 61.1 ± 2.34 N.D. 19.7 ± 1.95

12.6 ± 0.48 20.4 ± 0.35 4.0 ± 0.02 18.3 ± 1.32 N.D. 10.3 ± 0.46 21.1 ± 0.01 3.3 ± 0.12 6.8 ± 0.51 46.2 ± 3.12 14.3 ± 0.75 N.D. N.D. N.D. 14.0 ± 0.49 N.D. N.D. 26.5 ± 1.18 31.6 ± 1.29 16.3 ± 0.72 36.1 ± 3.39 68.4 ± 3.74 66.0 ± 1.11 64.0 ± 2.79 N.D. 8.9 ± 0.68

43.6 ± 1.67 4.1 ± 0.09 7.1 ± 0.26 28.8 ± 1.57 1.5 ± 0.02 21.4 ± 2.01 125.9 ± 10.81 15.8 ± 0.6 51.9 ± 0.91 N.D. 24.3 ± 1.04 6.6 ± 0.23 5.5 ± 0.14 2.2 ± 0 11.2 ± 0.2 N.D. 5.4 ± 0.12 98.0 ± 4.4 154.5 ± 7.56 103.5 ± 7.46 68.6 ± 4.14 154.0 ± 4.95 166.5 ± 11.03 148.3 ± 3.58 N.D. 16.1 ± 1.12

N.D. N.D. 6.2 ± 0.51 12.0 ± 0.10 N.D. N.D. N.D. N.D. 4.1 ± 0.00 N.D. 3.3 ± 0.25 N.D. N.D. N.D. N.D. N.D. N.D. 25.0 ± 1.88 N.D. N.D. N.D. 17.5 ± 1.35 14.7 ± 0.68 21.6 ± 0.39 N.D. 15.5 ± 1.33

70.3 ± 0.03 N.D. N.D. 20.7 ± 0.8 N.D. N.D. N.D. 5.7 ± 0.03 N.D. N.D. 15.6 ± 0.06 N.D. N.D. N.D. N.D. N.D. N.D. 55.6 ± 4.78 22.9 ± 1.05 7 ± 0.3 137.5 ± 6.96 32.0 ± 0.05 29.1 ± 0.72 48.4 ± 4.25 N.D. N.D.

7.2 ± 0.11 N.D. 8.2 ± 0.82 10.3 ± 0.15 N.D. N.D. N.D. 3.2 ± 0.20 23.1 ± 0.87 N.D. 8.1 ± 0.07 N.D. N.D. N.D. N.D. N.D. N.D. 12.2 ± 0.89 23.2 ± 1.64 11.6 ± 0.69 22.1 ± 1.68 31.6 ± 1.10 28.4 ± 2.71 49.6 ± 2.85 N.D. 20.4 ± 1.69

252.5 64.4 33.2 112.3 4.2 48.4 185.0 30.6 90.2 125.9 82.9 9.6 9.2 6.2 44.9 0.0 7.1 263.7 277.1 157.6 331.0 397.6 359.1 393.7 252.5 64.4

3773.5 ± 70.91 8410.2 ± 9.42 4813.9 ± 153.85 3978.6 ± 68.92 5512.5 ± 43.64 6416.2 ± 166.98 11490.8 ± 184.03 4145.2 ± 61.84 12061.4 ± 110.88 4618.3 ± 11.93 6166.2 ± 15.48 10308.1 ± 93.19 5381 ± 51.35 6865.4 ± 51.35 5210.9 ± 131.75 5991.5 ± 159.2 4317.3 ± 42.33 10721.4 ± 52.66 11480.8 ± 52.17 3612.2 ± 62.75 8363.8 ± 73.52 14555.1 ± 21.42 12272.8 ± 20.55 13981.2 ± 27.37 11695.1 ± 61.07 10203.6 ± 36.24

167.6 ± 2.54 211.5 ± 12.58 102.7 ± 11.42 96.6 ± 6.96 138.2 ± 5.62 179.4 ± 8.36 1052.8 ± 35.1 779.3 ± 24.12 461.1 ± 4.24 120.3 ± 3.28 88.1 ± 16.11 543.6 ± 6.06 105.8 ± 13.55 204.8 ± 5.60 194.7 ± 3.00 134.8 ± 13.10 74.4 ± 9.75 713 ± 23.49 731.9 ± 19.55 91.2 ± 3.24 96.9 ± 8.80 955.4 ± 48.45 936.7 ± 45.45 961.1 ± 46.85 933.2 ± 42.81 791.8 ± 62.26

160 600 670 965 982 959 951 964 959 250 953 968 991 984 958 840 962 972 975 870 984 – – – –

a GA = gallic acid, C = catechin, EC = epicatechin, P = proanthocyanidin unit, G = galloyl unit, TEAC = Trolox equivalent anti-oxidant capacity, GAE = total phenolics as gallic acid equivalents, N.D. = not detected. Values reported are the average of at least 3 measurements.

T.S. Villani et al. / Food Chemistry 170 (2015) 271–280

01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 GSE1 GSE2 GSE3 Peanut Pine

GA (mg/g)

T.S. Villani et al. / Food Chemistry 170 (2015) 271–280

k = 734 nm was 0.70 ± 0.02. The measuring was done using an Agilent G1111AA UV–Vis spectrophotometer. For this purpose, 990 ll of the ABTS radical was added to the cuvette; the absorbance was measured, and 10 ll of sample or standard solution were added immediately. The final absorbance was measured. There is a quantitative relationship between the reduction of the absorbance and the concentration of antioxidants present in the sample. A calibration curve was prepared using Trolox at a range of 0.0359– 1.65 mM in ethanol. Trolox Equivalent Anti-oxidant Capacity (TEAC) was expressed as the content in mmol Trolox/g sample required to achieve the same response as measured in the sample. The concentration of total phenolics was measured by adapting the method reported previously (Kim, Jeong, & Lee, 2003), originally described by Singleton and Rossi. A serial dilution series was prepared using gallic acid at a concentration ranging from 31.2 lg/mL to 4.00 mg/mL. A reagent blank using ddH2O was prepared. Folin–Ciocalteau reagent was diluted to 10% in water. 40 lL each sample was added to an Eppendorf tube. 900 lL 10% Folin– Ciocalteau reagent was added. After 5 min, 400 lL of 15% Na2CO3 solution was added with mixing. After incubation for 90 min at room temperature, the absorbance versus prepared blank was read at 750 nm. Total phenolic contents were expressed as mg gallic acid equivalents per gram sample material. Sample solutions were prepared as described in Section 2.2, except without the addition of 0.1% acetic acid to the solvent. The concentrations were adjusted such that the sample results fell within the calibration range of each assay. All samples were analyzed in triplicate. 2.7. Thin layer chromatography Developing solvent consisting of a ratio of 3:1:3 acetone/acetic acid/toluene was selected (Sun, Leandro, Ricardo da Silva, & Spranger, 1998), and prepared fresh before development. 50 lL each sample solution (as prepared in Section 2.2) was spotted onto the origin of the plate. Plates were thoroughly dried before development. After development 8–10 cm, plates were removed from developing chamber and allowed to dry thoroughly at room temperature. Plates were visualized by spraying with (or submerging in) vanillin stain (10 g vanillin dissolved in 50 mL concentrated HCl), followed by gently heating with a heat-gun. 2.8. Price analysis Prices for each product were obtained from multiple retail outlets and internet retailers. Prices from at least three different retailers were used in the calculation of the average prices for each product. Since each bottle contained a varying number of capsules with a variable content of GSE, the total quantity of GSE per bottle was obtained by multiplying the number of capsules by the selfreported content of GSE in each capsule. This allowed for the calculation of a unit price representing the ratio of price to quantity GSE claimed used in further analyses. 3. Results and discussion Twenty-one commercial products containing grape seed extract were analyzed for their profile and content of PAC monomers and dimers. The samples were chemically profiled, and major PAC constituents were estimated as described above. Under the optimized LC–MS conditions and on the basis of analysis of the MS and UV data and in comparison with the authenticated standards and reported data, the major constituents were successfully separated, identified, and quantitated/estimated (Fig. 1). Results of quantitative analysis are shown in Table 1. Six different proanthocyanidin

275

B-type dimers were found, including Proanthocyanidin B2 ([]epicatechin dimer), which was the major PAC dimer detected by molecular ion [MH] at 577 m/z, [M+H]+ at 579 m/z, with major fragment at 291 m/z. The sum of all six isomers for each sample is reported in Table 1. Three galloylated proanthocyanidin monomers were detected by [MH] ion detected at 441 m/z in negative mode, [M+H]+ ion detected at 443 m/z in positive mode, with a strong fragment peak at 291 m/z; stereochemical or regiochemical configuration was not determined. Proanthocyanidin trimer (P3) was detected by molecular ion [MH] at 865 m/z. A-type dimer was detected in peanut skin extract and pine bark extract by molecular ion [MH] at 575 m/z, and [MH]+ at 577 m/z. Mass spectral data for selected compounds is shown in the Supplementary material (S1). The average total content of the major components found in GSE contained by the three authentic grape seed reference standards was found to be 383.5 ± 21.13 mg/g, with proanthocyanidin B dimers (P2) making up 41%, and catechin and epicatechin cumulatively representing 32% of the total content detected by HPLC (Table 1). Many of the components contained in GSE were also detected in authentic pine bark extract reference material. Pine bark extract contained only 80.6 ± 6.77 mg/g of the seven major components found in GSE, with only 20% accounted for by PAC B dimers (P2). This result is consistent with results found in the literature (Hellström & Mattila, 2008). Gallic acid and O-galloylated B dimers (P2G1) were not detected. Proanthocyanidin trimers (P3) were a major constituent of pine bark extract, embodying 25% by mass, compared to only 6% found in GSE. None of the major compounds detected in GSE were detected in peanut skin extract. A comparison of the 280 nm UV chromatograms for authentic GSE, pine bark extract, and peanut skin extract is shown in Fig. 1. These results show that GSE and pine bark extract contain a number of PAC B dimers and galloylated derivatives, whereas peanut skin extract contains mostly high-weight oligomers and PAC A-dimers. To understand more specific relationships about the variation of components across the population of commercial and authentic GSE, the coefficient of correlation (R value) was calculated for each pair of components in the GSE sample population. As expected, the two isomers catechin and epicatechin, have the strongest correlation (R = 0.8461). Epicatechin content is also correlated with PAC dimer (P2) content (R = 0.7770), which is reasonable to expect since the major PAC dimer is PAC B2, a dimer of epicatechin. The same logic follows for the correlation between epicatechin and PAC trimer (P3). Unsurprisingly, P2 content was correlated to P3 content (R = 0.7568), since PAC dimers are obviously precursors to PAC trimers. A comparison of the relative and absolute compositions of authentic GSE, commercial GSE, and pine bark is shown in Fig. 2a and b, respectively. The average composition of the commercial GSE samples reveals a large degree of inhomogeneity, the average of the total GSE constituents was 101.7 ± 103.9 mg/g—a relative standard deviation of 102%. The data also shows that the average relative composition of the commercial samples is very similar to that of authentic GSE reference material and the absolute quantity contained in the commercial samples is on average 73% less than the quantity in authentic GSE. The relative composition of pine bark extract is significantly different from that of GSE reference materials. Very few of the commercial GSE samples contained an overall content of PACs and catechins at a level comparable to authentic GSE, which raises a serious concern. As can be seen in Fig. 2c, there are 9 commercial samples containing less than 50 mg/g GSE PACs— less than 15% of the content contained in samples of authentic GSE. Five samples contain less than 10 mg/g of major compounds found in GSE. In sample 16, none of these compounds which should be present in GSE were even detected. Overall, six of the commercial

276

T.S. Villani et al. / Food Chemistry 170 (2015) 271–280

A

B

C

Fig. 2. (A) Relative and (B) absolute content of grape seed extract (GSE) reference material, the average of commercial GSE samples (n = 21) and pine bark extract; (C) quantitative results of Table 1, sorted by total content (mg/g).

samples could be considered seriously adulterated, perhaps counterfeit, while another five samples contained considerably less PACs and catechins than the remaining commercial samples. Scrutiny of the chromatograms from the six worst quality samples, with the smallest content of GSE compounds (16, 5, 14, 17, 13 and 12) in comparison to peanut skin extract, demonstrates that these samples are likely adulterated with peanut skin extract. A comparison of the 280 nm chromatograms of 16 to authentic GSE and peanut skin extract show remarkable similarity (data shown in the Supplementary material S2). Examination of 280 nm chromatograms and MS data for samples, 15, 6 and 3, show that they contain PAC A-type dimers, suggesting that these three samples are a mixture of GSE and peanut skin extract. In summation, nine samples out of 21 have evidence of adulteration with peanut skin extract. This result was surprising and represents a concern relative to food safety and human health. A number of samples contained PACs and catechins at a level more similar to pine bark extract than to GSE; samples 9, 11, and 3 show a high degree of similarity to authentic pine bark extract. A comparison of the 280 nm UV chromatogram of sample 3 to authentic GSE and pine bark extract is shown in the Supplementary material (S3). Two samples, 2 and 10, contain only catechin and epicatechin, likely a result of extraction of white grape variety seeds, which have been shown to contain principally PAC monomers (Freitas & Glories, 1999); however the reported content in white grape seed extract is 4–5 higher than detected for 2 and 10. A comparison of the 280 nm UV chromatograms for a number of representative samples is included in the Supplementary material (S4).

There were six samples which contained concentrations of GSE PACs at a high level. Samples 01, 18, 19 and 21 each contained greater than 200 mg/g GSE PACs. Samples 07, 20 contained greater than 150 mg/g. More importantly, each of these samples contained the diversity of components expected in GSE—greater than twice the concentration in pine bark extract. It is troubling to note the wide degree of variation in the total content among the tested products; variation can be accounted for by differing sources of grape seeds, grape variety, and extraction techniques. Such widespread variation in product composition creates difficulty for consumers who intend to use GSE for specific pharmacological effects. In order to group the commercial samples and reference materials according to their underlying compositional similarities, clustering analysis was applied to the normalized quantitative results obtained from HPLC analysis. Clustering analysis is amenable to quality control of natural products due to the complexity of natural products. This technique allows for establishing groups based on similarity, permitting the reduction of the overall complexity of the data for analysis. Results of clustering analyses are shown in Fig. 3a. The samples split into three major clusters: the first containing the authentic GSE (outlined in green) and samples high in PAC content, the second containing authentic pine bark extract (outlined in yellow) and commercial GSE samples with a lower content of PACs, and the last containing peanut skin extract (outlined in red). The grouping of the samples is congruent with the interpretations of the HPLC data offered in the preceding discussion. The samples which group with authentic GSE are the same six samples containing the highest content of GSE PACs. The samples which grouped in the cluster with peanut skin extract are the

277

400

7

20

18

19

24

23

21 22

1 8

3

6

15

2

26

4

9 11

5

14

25

13

16

12

17

B

10

0

200

Height

600

800

A

1000

T.S. Villani et al. / Food Chemistry 170 (2015) 271–280

Fig. 3. (A) Cluster dendrogram with corresponding quantitative data. Height of a branch indicates degree of similarity among the nodes on that branch; a large height means the branches are significantly different. Samples group into three major clusters. Only six samples group with the reference grape seed extract (GSE). (B) Representation of the quantitative composition of each sample from HPLC. Black outlines denote cost per gram of GSE reported. As can be seen, there is little correlation between price and content. Abbreviations: GA, gallic acid; C, catechin; EC, epicatechin; P, proanthocyanidin unit; G, galloyl unit.

same samples which contain less than 10 mg/g GSE PACs, reinforcing the argument that these samples are of unacceptably poor quality. The most difficult cluster to interpret is the central cluster (outlined in yellow). The samples that fall into this cluster can be divided into two categories: those which may contain pine bark extract, and those that contain low levels of GSE PACs with respect to authentic GSE. Pine bark extract is found in this cluster due to its similarity to these samples, suggesting that these samples may possibly contain pine bark extract. It should be noted, however, that GSE and pine bark extract contain a remarkably similar qualitative profile of PAC monomers and dimers, and as such it is difficult to distinguish between samples containing pine bark extract, and those with a low content of GSE compared to the reference standards selected for this analysis. The samples contained in this cluster differ significantly than the high-content GSE cluster which contains the authentic GSE selected for this study. The variation may be accounted for due to differences in grape variety, white vs. red grape seeds, or extraction technique. The samples in this cluster, while they may be authentic GSE, contain a far lower overall content of PACs than the samples contained in the green cluster. To better understand the relationship between the price of the product and its quality, vis-à-vis its value to the consumer, comprehensive pricing data for each product was compiled for analysis. This relationship was described by using the unit cost ($USD/g) GSE as paid for by the end consumer. To calculate the total mass in each bottle purchased for this study, the number of capsules was multiplied by the mass GSE reported for each capsule on the

product label. For the mass to cost ratio, an average of at least 3 prices offered at retail stores or through ecommerce were used. Detailed pricing data used for this study is contained in the Supplementary material (S5). The unit price ($/g) is shown in Fig. 3b. This data is shown with the bar-graph representing the quantities of each compound detected by HPLC and below the clustering diagram in Fig. 3a. The content of GSE detected in the product, a measure of the intrinsic quality, does not correlate with the price paid for each product. This is further demonstrated in Fig. 4a, a correlation scatter plot between the unit price ($/g) and the total content GSE determined by HPLC (mg/g), where the coefficient of determination (R2) is 0.0116, suggesting that there is no correlation between price and quality as determined by HPLC. These results indicate that relative to commercial GSE products and their associated label claims; consumers are paying an arbitrary price with regard to quality. The complexity of the chemical composition of grape seed extract and its primary adulterants demonstrate the difficulty involved in the elucidation of the quality of GSE in manufacturing. A major problem which may contribute to the high proportion of adulterated and low-quality GSE is the analytical techniques available to the herbal products industry. Typically, due to cost and the lack of expertise in commercial analytical labs, and the industry employs rapid, lower cost screens such as simple colorimetric and spectrophotometric assays. These assays are used to estimate total phenolic content or anti-oxidant activity, and therefore as a screen for content/activity. The two major methods used are the Folin–Ciocalteau total polyphenol (FCTP) assay (Nakamura, Tsuji,

278

T.S. Villani et al. / Food Chemistry 170 (2015) 271–280

$5.00

A

R² = 0.0116

mmol TEAC / g

$/g

$4.00

B 14,000

$3.00 $2.00 $1.00 $0.00 0

100

200

300

400

R² = 0.7703

12,000 10,000 8,000 6,000 4,000 2,000 0

Total content (mg/g) determined by HPLC

D R² = 0.3085

15000

GAE (mg/g)

TEAC mmol/g

C 20000 10000 5000 0 0

200 HPLC content (mg/g)

1000 mg GAE / g

1500

R² = 0.1244

0

400

E

1200 1000 800 600 400 200 0

500

100 200 300 HPLC Total Content (mg/g)

400

F

Fig. 4. Correlation diagrams for (A) HPLC content to amount per dollar (mg/$); (B) correlation between TEAC assay (mmol/g) and FCTP assay (mg/g) (C) TEAC (mmol/g) to HPLC content; (D) total phenolics in GAE (mg/g) to HPLC content; (E) TEAC mmol/g to amount per dollar (mg/$); (F) total phenolics in GAE mg/g to amount per dollar (mg/$).

& Tonogai, 2003; Zululeta et al., 2009; Škerget et al., 2005), and the Trolox Equivalent Antioxidant Capacity (TEAC) assay (Prior & Cao, 2000; Singleton & Rossi, 1965). The Folin–Ciocalteau assay provides approximated quantitative measure of phenols as a broad class, a major drawback being a lack of selectivity to other reducible components and compounds other than phenols that react with the reagent. The Trolox Equivalent Antioxidant Capacity assay measures the absorption of the ABTS (2,20 -azino-bis(3-ethylbenzthiazoline-6-sulphonic acid) radical cation by substrates relative to absorption by Trolox (a water soluble Vitamin E derivative). These assays are commonly employed in industry for quality control. Yet, as these assays cannot discriminate between different types of phenols and compounds, it would not be possible to detect adulteration of grape seed extract with peanut skin extract, or any phenol containing adulterant for that matter. A standing hypothesis was that the results of the Trolox Equivalent Antioxidant Capacity and Total Phenol content for the authentic samples of GSE would fall in the same order of magnitude as pine bark extract and peanut skin extract. This hypothesis was confirmed (Table 1): authentic GSE ranged from 12,272.8– 14,555.1 TEAC lmol/g, peanut skin at 11,695.1 TEAC lmol/g, and pine bark extract at 10,203.6 TEAC lmol/g. The results for the Folin–Ciocalteau assay were similar: 936.7–961.1 GAE mg/g for authentic GSE, 933.2 GAE mg/g for peanut skin, and 791.8 GAE mg/g. Given that these assays show no selectivity over the components of GSE, pine bark, and peanut skin, they cannot be employed as a method for screening for adulteration and should only be used as a preliminary indicator of quality. It is not therefore surprising

that the commercial samples showed a great degree of inhomogeneity with respect to TEAC and GAE content, averaging 6840.0 ± 2836.24 TEAC lmol/g and 299.5 ± 290.11 mg GAE/g. As expected, there was a strong correlation (R2 = 0.7703) between TEAC and GAE content for the all results of these assays (Fig. 4b), which can be accounted for by the well-known radical-scavenging activity of polyphenols (Kim et al., 2003). To examine whether the results of FCTP and TEAC assays of commercial GSE can be correlated to the content of GSE PAC monomers and dimers, the results from these assays for each sample were plotted against the corresponding HPLC data (Fig. 4c, 4d). As expected, the FCTP and TEAC assays were poorly correlated to HPLC results (R2 = 0.1244 and R2 = 0.1594, respectively. This can be accounted for due to the selectivity of HPLC allowing for the quantitation of specific components related to GSE, compared to the indirect assays like the FCTP and TEAC assays. It can be inferred that the antioxidant capacity and polyphenol content of the samples had little to do with the overall PAC content of compounds in commercial GSE, suggesting that there are other components than PACs in the commercial samples that are responsible for antioxidant capacity and/or polyphenol content. It was hypothesized that a relationship may exist between the unit cost ($USD/g) and the TEAC and FCTP results. To examine this, the results of each assay were correlated to the unit price of each commercial GSE sample. There is a slightly negative correlation (R2 = 0.2219) between anti-oxidant capacity and unit price (Fig. 4e). The total phenol content had almost no correlation (R2 = 0.0738) to unit price, as shown in Fig. 4f.

T.S. Villani et al. / Food Chemistry 170 (2015) 271–280

279

Fig. 5. TLC results. Lanes ordered corresponding to cluster groupings given by Fig. 3a. Grape seed extracts (GSE) exhibits a bright spot for catechin/epicatechin, as well as PAC B2 isomers. PAC A2 dimers were found only in peanut-skin containing samples. Pine bark extract contains more dilute content of catechin and PAC dimer. Identity of B2 was confirmed by cospot with reference standard (data not shown). The identity of A2 was confirmed by heavily spotting a fresh TLC plate, visualizing with UV, and scraping the spot of the plate. The scraped silica was extracted with 0.1% acetic acid in methanol, filtered through 0.45 lm syringe filter, and injected directly into ESI-MS in positive mode. Abbreviations: C, catechin; PAC, proanthocyanidin; A2, proanthocyanidin A-type dimer; B2, proanthocyanidin B-type dimer.

TLC was evaluated as a rapid, simple, and inexpensive technique to be used to distinguish authentic GSE from its adulterants. A TLC method was applied to separate the PACs and visualized concomitantly using a simple vanillin/HCl stain and allows for the rapid differentiation between high quality GSE and adulterated GSE (Fig. 5). The TLC lanes are organized in the same order as in the cluster dendrogram and data shown in Fig. 3. Examining the first cluster, outlined in red, depicts the poor-quality samples adulterated with peanut skin. Here, it can be readily seen that these samples contain PAC A2 dimers, whereas authentic grape seed samples (outlined in green) contain only PAC B dimers. As such, TLC has been demonstrated as a rapid and effective method to detect adulteration of GSE by peanut skin extract. However, it remains difficult to detect the difference between pine bark extract and GSE due to the similarity in composition. 4. Conclusion In conclusion, adulteration of grape seed extract in commercial products is a significant problem. Out of the 21 commercial products analyzed blind—without the researcher’s knowledge of the product source, six samples contained no detectable quantities of grape seed extract, and were composed primarily of peanut skin extract as determined by comparison to authentic peanut skin. PAC A-dimers were detected in 3 additional samples, suggesting adulteration with peanut skin. Adulteration with peanut skin extract represents a significant concern to safety with regards to allergens. Researchers have found by telephone survey that 1.3% of 4855 U.S. households self-report an allergy to either peanuts or tree-nuts (Sicherer, Muñoz-Furlong, Godbold, & Sampson, 2010). With the increasing development and sale of herbal remedies and natural products, adulteration with a common allergen represents a considerable risk to public safety. Adulteration can occur purposefully and/or inadvertently. The motivation behind purposeful adulteration in commercial products is for economic gain. Peanut skin extract, which is a high-volume byproduct of the peanut industry, is less expensive and typically available at a much greater volume than grape seed extract. Thus, a bulk distributor of grape seed extract or another manufacturer or actor along the value chain can take advantage of the chemical similarity between GSE and peanut skin extract since the simpler assays typically used in industry cannot discriminate between the two. Due to reliance on inferior proximate assays across the value-chain, adulteration can then go undetected by others

downstream in the commodity chain, such as those involved in distribution, packaging, wholesale, and retail sales. The inherent problem is that many of these manufacturers, relying on inferior quality control procedures, do not know their products may be adulterated or counterfeit, leading to the perpetuation of low quality products in the market-place. Consumers of the commercial product rely on label claims and other information provided directly from the supplier. This is one of many examples of the issues that dominate the international marketplace due to a lack of standardized quality control methodology, allowing for adulteration to perpetuate and go undetected (Betz, 2006; Foster, 2011; Ho et al., 2006; Wang et al., 2006). Economic adulteration may have significant impact to the consumers of grape seed extract, as well as the distributors of high quality, legitimate grape seed extract. Ultimately, consumers expect to the product which is purchased to be labeled accurately. The purpose of this study is to demonstrate the necessity of chromatographic techniques (TLC, HPLC, etc.) for the screening of grape seed extract for quality and adulteration. Only with chromatographic techniques can one differentiate between the components common to GSE, and peanut skin extract. Clearly, for the benefit of the consumers, as well as the distributors, more rigorous methods for quality control need to be applied to the GSE industry.

Acknowledgements This work was supported by the National Institute of Health grant PO1AT004511. We also thank the Rutgers New Use Agriculture and Natural Plant Products Program and the New Jersey Agricultural Experiment Station, Rutgers, The State University of New Jersey for their partial funding and support. We thank Xiaoyou Zhang in BannerBio Nutraceuticals Inc. (Shenzhen, China) for providing the authenticated reference materials. We thank Paul Coates, Joe Betz, Gordon Cragg and others at the NIH ODS for their interest in this work. We thank Ray Fatto, Wendy Wang, and Grace Lin for their invaluable help in this work.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.foodchem.2014. 08.084.

280

T.S. Villani et al. / Food Chemistry 170 (2015) 271–280

References Al-Habib, A., Al-Saleh, E., Safer, A. M., & Afzal, M. (2010). Bactericidal effect of grape seed extract on methicillin-resistant Staphylococcus aureus (MRSA). The Journal of toxicological sciences, 35(3), 357–364. Aziz, M. H., Kumar, R., & Ahmad, N. (2003). Cancer chemoprevention by resveratrol: In vitro and in vivo studies and the underlying mechanisms (review). International Journal of Oncology, 23(1), 17. Baydar, N. G., Özkan, G., & Sag˘diç, O. (2004). Total phenolic contents and antibacterial activities of grape (Vitis vinifera L.) extracts. Food Control, 15(5), 335–339. Bertelli, A. A., & Das, D. K. (2009). Grapes, wines, resveratrol, and heart health. Journal of Cardiovascular Pharmacology, 54(6), 468–476. Betz, J. M. (2006). Botanical quality initiatives at the Office of Dietary Supplements, National Institutes of Health. In Wang, M. S. Sang, L. S. Hawang, & C. T. Ho (Eds.), Herbs: Challenges in chemistry and biology. American chemical society symposium series 925 (pp. 2–13). Washington, D.C. USA: American Chemical Society. Castaneda-Ovando, A., Pacheco-Hernández, M. D. L., Páez-Hernández, M. E., Rodríguez, J. A., & Galán-Vidal, C. A. (2009). Chemical studies of anthocyanins: A review. Food Chemistry, 113(4), 859–871. Ferruzzi, M. G., Lobo, J. K., Janle, E. M., Cooper, B., Simon, J. E., Wu, Q. L., et al. (2009). Bioavailability of gallic acid and catechins from grape seed polyphenol extract is improved by repeated dosing in rats: Implications for treatment in Alzheimer’s disease. Journal of Alzheimer’s Disease, 18(1), 113–124. Foster, S. (2011). A brief history of adulteration of herbs, spices, and botanical drugs. HerbalGram, 92, 42–57. Freitas, V. A. P. D., & Glories, Y. (1999). Concentration and compositional changes of procyanidins in grape seeds and skin of white Vitis vinifera varieties. Journal of the Science of Food and Agriculture, 79(12), 1601–1606. Fuleki, T., & da Silva, J. M. R. (1997). Catechin and procyanidin composition of seeds from grape cultivars grown in Ontario. Journal of Agricultural and Food Chemistry, 45(4), 1156–1160. Furiga, A., Lonvaud-Funel, A., & Badet, C. (2009). In vitro study of antioxidant capacity and antibacterial activity on oral anaerobes of a grape seed extract. Food chemistry, 113(4), 1037–1040. Hellström, J. K., & Mattila, P. H. (2008). HPLC determination of extractable and unextractable proanthocyanidins in plant materials. Journal of Agricultural and Food Chemistry, 56(17), 7617–7624. Ho, L., Ferruzzi, M. G., Janle, E. M., Wang, J., Gong, B., Chen, T. Y., et al. (2013). Identification of brain-targeted bioactive dietary quercetin-3-O-glucuronide as a novel intervention for Alzheimer’s disease. The FASEB Journal, 27(2), 769–781. Ho, C. T., Simon, J. E., Shahidi, F., & Shao, Y. (2006). Dietary supplements: An overview. In C. T. Ho, Y. Shao, & J. E. Simon (Eds.), Dietary supplements. American chemical society symposium series 925 (pp. 2–8). Washington, D.C. USA: American Chemical Society. Ho, L., Yemul, S., Wang, J., & Pasinetti, G. M. (2009). Grape seed polyphenolic extract (GSE) as a novel therapeutic reagent in tau-mediated neurodegenerative disorders. Journal of Alzheimer’s Disease, 16, 433–439. Jackson, J. E. (2004). Frontmatter, in a user’s guide to principal components. Hoboken, NJ, USA: John Wiley & Sons Inc [doi: 10.1002/0471725331.fmatter]. Jang, M., Cai, L., Udeani, G. O., Slowing, K. V., Thomas, C. F., Beecher, C. W., et al. (1997). Cancer chemopreventive activity of resveratrol, a natural product derived from grapes. Science, 275(5297), 218–220. Jayaprakasha, G. K., Selvi, T., & Sakariah, K. K. (2003). Antibacterial and antioxidant activities of grape (Vitis vinifera) seed extracts. Food Research International, 36(2), 117–122. Karonen, M., Loponen, J., Ossipov, V., & Pihlaja, K. (2004). Analysis of procyanidins in pine bark with reversed-phase and normal-phase high-performance liquid chromatography–electrospray ionization mass spectrometry. Analytica Chimica Acta, 522(1), 105–112. Kim, D. O., Jeong, S. W., & Lee, C. Y. (2003). Antioxidant capacity of phenolic phytochemicals from various cultivars of plums. Food Chemistry, 81(3), 321–326. Lawaetz, A. J., Schmidt, B., Staerk, D., Jaroszewski, J. W., & Bro, R. (2009). Application of rotated PCA models to facilitate interpretation of metabolite profiles: Commercial preparations of St John’s wort. Planta Medica, 75(3), 271–279. Liang, Y. Z., Xie, P., & Chan, K. (2004). Quality control of herbal medicines. Journal of chromatography B, 812(1), 53–70. Liu, S., & White, E. (2012). Extraction and characterization of Proanthocyanidins from grape seeds. Open Food Science Journal, 6, 5–11. Lou, H., Yamazaki, Y., Sasaki, T., Uchida, M., Tanaka, H., & Oka, S. (1999). A-type proanthocyanidins from peanut skins. Phytochemistry, 51(2), 297–308.

Manach, C., Williamson, G., Morand, C., Scalbert, A., & Rémésy, C. (2005). Bioavailability and bioefficacy of polyphenols in humans. I. Review of 97 bioavailability studies. The American Journal of Clinical Nutrition, 81(1), 230S–242S. Moreno, D. A., Ilic, N., Poulev, A., Brasaemle, D. L., Fried, S. K., & Raskin, I. (2003). Inhibitory effects of grape seed extract on lipases. Nutrition, 19(10), 876–879. Nakamura, Y., Tsuji, S., & Tonogai, Y. (2003). Analysis of proanthocyanidins in grape seed extracts, health foods and grape seed oils. Journal of Health Science, 49(1), 45–54. Pasinetti, G. M., & Eberstein, J. A. (2008). Metabolic syndrome and the role of dietary lifestyles in Alzheimer’s disease. Journal of Neurochemistry, 106(4), 1503–1514. Pasinetti, G. M., & Ho, L. (2010). Role of grape seed polyphenols in Alzheimer’s disease neuropathology. Nutrition and Dietary Supplements, 2010(2), 97. Passos, C. P., Cardoso, S. M., Domingues, M. R. M., Domingues, P., Silva, C. M., & Coimbra, M. A. (2007). Evidence for galloylated type-A procyanidins in grape seeds. Food Chemistry, 105(4), 1457–1467. Prior, R. L., & Cao, G. (2000). Analysis of botanicals and dietary supplements for antioxidant capacity: A review. Journal of AOAC International, 83(4), 950–956. Rasmussen, S. E., Frederiksen, H., Struntze Krogholm, K., & Poulsen, L. (2005). Dietary proanthocyanidins: Occurrence, dietary intake, bioavailability, and protection against cardiovascular disease. Molecular Nutrition & Food Research, 49(2), 159–174. R Development Core Team. (2011). R: A language and environment for statistical computing. 3-900051-07-0. Vienna, Austria: R Foundation for Statistical Computing [URL: http://www.R-project.org/]. Renaud, S. D., & de Lorgeril, M. (1992). Wine, alcohol, platelets, and the French paradox for coronary heart disease. The Lancet, 339(8808), 1523–1526. Saito, M., Hosoyama, H., Ariga, T., Kataoka, S., & Yamaji, N. (1998). Antiulcer activity of grape seed extract and procyanidins. Journal of Agricultural and Food Chemistry, 46(4), 1460–1464. Sarnoski, P. J., Johnson, J. V., Reed, K. A., Tanko, J. M., & O’Keefe, S. F. (2012). Separation and characterisation of proanthocyanidins in Virginia type peanut skins by LC–MSn. Food Chemistry, 131(3), 927–939. Sicherer, S. H., Muñoz-Furlong, A., Godbold, J. H., & Sampson, H. A. (2010). US prevalence of self-reported peanut, tree nut, and sesame allergy: 11-year follow-up. Journal of Allergy and Clinical Immunology, 125(6), 1322–1326. Singleton, V. L., & Rossi, J. A. (1965). Colorimetry of total phenolics with phosphomolybdic-phosphotungstic acid reagents. American Journal of Enology and Viticulture, 16(3), 144–158. Škerget, M., Kotnik, P., Hadolin, M., Hraš, A. R., Simonicˇ, M., & Knez, Zˇ. (2005). Phenols, proanthocyanidins, flavones and flavonols in some plant materials and their antioxidant activities. Food Chemistry, 89(2), 191–198. Sun, B., Leandro, C., Ricardo da Silva, J. M., & Spranger, I. (1998). Separation of grape and wine proanthocyanidins according to their degree of polymerization. Journal of Agricultural and Food Chemistry, 46(4), 1390–1396. Wang, J., Ferruzzi, M. G., Ho, L., Blount, J., Janle, E. M., Gong, B., et al. (2012). Braintargeted proanthocyanidin metabolites for Alzheimer’s disease treatment. The Journal of Neuroscience, 32(15), 5144–5150. Wang, J., Ho, L., Zhao, W., Ono, K., Rosensweig, C., Chen, L., et al. (2008). Grapederived polyphenolics prevent Ab oligomerization and attenuate cognitive deterioration in a mouse model of Alzheimer’s disease. The Journal of Neuroscience, 28(25), 6388–6392. Wang, M., Liang, C. P., Wu, Q. L., Simon, J. E., & Ho, C. T. (2006). Instrumental analysis of popular botanical products in the U.S. market. In Wang, M. S. Sang, L. S. Hawang, & C. T. Ho (Eds.), Herbs: Challenges in chemistry and biology. American chemical society symposium series 925 (pp. 39–54). Washington, D.C. USA: American Chemical Society. Welch, C. R., Wu, Q., & Simon, J. E. (2008). Recent advances in anthocyanin analysis and characterization. Current Analytical Chemistry, 4(2), 75. Wu, Q. L., Wang, M., & Simon, J. E. (2003). Determination of proanthocyanidins in grape products by liquid chromatography/mass spectrometric detection under low collision energy. Analytical Chemistry, 75(10), 2440–2444. Xie, P., Chen, S., Liang, Y. Z., Wang, X., Tian, R., & Upton, R. (2006). Chromatographic fingerprint analysis—a rational approach for quality assessment of traditional Chinese herbal medicine. Journal of Chromatography A, 1112(1), 171–180. Xu, Y., Simon, J. E., Welch, C., Wightman, J. D., Ferruzzi, M. G., Ho, L., et al. (2011). Survey of polyphenol constituents in grapes and grape-derived products. Journal of Agricultural and Food Chemistry, 59(19), 10586–10593. Zululeta, A., Esteve, M. J., & Frígola, A. (2009). ORAC and TEAC assays comparison to measure the antioxidant capacity of food products. Food Chemistry, 114(1), 310–316.