sign in sign up
<<
Home , Leucoanthocyanidin, Polyphenol, Thearubigin

Lead Review Article November 1998: 31 7-333

Polyphenols: Chemistry, Dietary Sources, Metabolism, and Nutritional Significance Laura Bravo, Ph.D.

Polyphenols constitute one of the most numer- fore, have been studied for taxonomic purposes or to de- ous and ubiquitous groups of plant metabolites termine adulteration of food products. Polyphenols have and are an integral pat? of both human and animal several industrial applications, such as in the production diets. Ranging from simple phenolic molecules to of paints, paper, and cosmetics, as agents, and in highly polymerized compounds with molecular the food industry as additives (as natural colorants and weights of greater than 30,000 Da, the occurrence preservatives). In addition, some phenolic compounds, of this complex group of substances in plant foods the , have applications as and an- is extremely variable. Polyphenols traditionally have tidiarrheal, antiulcer, and anti-inflammatory agents, as well been considered by animal nutruon- as in the treatment of diseases such as hypertension, vas- ists, because of the adverse effect of , one cular fragility, allergies, hypercholesterolemia, and 0th- type of , on protein digestibility. How- ever, recent interest in food phenolics has in- Polyphenolic compounds are ubiquitous in all plant creased greatly, owing to their capac- ity (free radical scavenging and metal chelating organs and are, therefore, an integral part of the human activities) and their possible beneficial implications diet. Until recgntly, most of the nutritional interest in in human health, such as in the treatment and pre- polyphenolic compounds was in the deleterious effects vention of cancer, cardiovascular disease, and caused by the ability of certain polyphenols to bind and other pathologies. Much of the literature refers to precipitate , such as dietary protein, car- a single group of plant phenolics, the flavonoids. bohydrate, and digestive enzymes, thereby reducing food This review offers an overview of the nutritional digestibility. Recent interest, however, in food phenolics effects of the main groups of polyphenolic com- has increased greatly because of the antioxidant and free pounds, including their metabolism, effects on radical-scavenging abilities associated with some phe- bioavailabilityl and antioxidant activityl as nolics and their potential effects on human health. This well as a brief description of the chemistry of review offers an overview of the main nutritional effects polyphenols and their occurrence in plant foods. of polyphenolic compounds as well as a brief description of the chemistry of polyphenols and their occurrence in plant foods. Introduction In reviews of the abundant scientific literature on polyphenolic compounds, particularly those studies of For decades, plant polyphenols have interested scientists the physiologic effects of plant polyphenols, most experi- because they are essential to plant for their ments are devoted to the study of a specific group of contribution to plant morphology (ie., pigmentation),they phenolic compounds-the flavonoids. This review pro- are involved in growth and reproduction and provide vides a general overview of the nutritional significance of plants with resistance to pathogens and predators (by flavonoids as well as other types of food polyphenols, acting as or by increasing food astringency, including simple and tannins. thus making food unpalatable), they protect crops from plague and preharvest seed germination, and for other Chemistry of Phenolic Compounds reasons. The polyphenolic profiles of plants differ be- tween varieties of the same species. Polyphenols, there- Phenolic compounds or polyphenols constitute one of the most numerous and widely distributed groups of sub- Dr. Bravo is with the Departarnento de stances in the plant kingdom, with more than 8000 phe- Metabolism0 y Nutricion, lnstituto del Frio (CSIC), nolic structures currently Polyphenols are prod- Ciudad Universitaria s/n, Madrid 28040, Spain. ucts of the secondary metabolism of plants. They arise

Nutrition Reviews, Vol. 56, No. 11 317 biogenetically from two main synthetic pathways: the acids and many others. Associations with other com- shikimate pathway and the acetate path~ay.~This is an pounds, such as carboxylic and organic acids, amines, extremely wide and complex group of plant substances. and lipids, and linkages with other phenols are also com- Natural polyphenols can range from simple molecules, mon. such as phenolic acids, to highly polymerized compounds, According to Harb~rne,~polyphenols can be divided such as tannins. They occur primarily in conjugated form, into at least 10 different classes depending on their basic with one or more sugar residues linked to hydroxyl groups, . Table 1 illustrates the basic chemical although direct linkages of the sugar unit to an aromatic structure of the main polyphenolic compounds. Fla- carbon atom also exist. The associated sugars can be vonoids, which constitutethe most important single group, present as , disaccharides, or even as can be further subdivided into 13 classes, with more than oligosaccharides. Glucose is the most common sugar resi- 5000 compounds described by 19904 (Table 2).A brief de- due, although galactose, rhamnose, xylose, and arabinose scription of the main classes of food phenolics follows. are also found, as well as glucuronic and galacturonic For more comprehensive reviews, see references 4-1 0.

Table 1. Main Classes of Polyphenolic Compounds Class Basic Skeleton Basic Structure

Simple phenols '6

Benzoquinones '6 '

Phenolic acids

Acetophenones '6-'2

Phenylacetic acids

Hydroxycinnamic acids

Phenylpropenes

Coumarins,

Chromones

Naftoquinones '6-'4

Xanthones c,c,-c,

Stilbenes

Anthraquinones

Flavonoids c6-c3-c6

Lignans, neolignans (',5-'&

Lignins ('6-'3)"

318 Reviews, Vol. 56, No. 11 Table 2. Classification of Food Flavonoids as phenolic acids (e.g., gallic, vanillic, syringic, p- Basic Structure hydroxybenzoic) and aldehydes (e.g., , syringaldehyde,p-hydroxybenzaldehyde), also are fairly common in higher plants and ferns. Phenylacetic acids and acetophenones (C,-C,) are, however, less frequently described in the literature. All of these compounds can be Q)p found free, although their corresponding methyl and ethyl 0 and glycosides occur very commonly in free andor bound f~ms.~ derivatives (C,-C,) also are an im- portant group of low-molecular-weight phenolics. Chromones are less known than , with the latter occurring naturally as glycosides (e.g., umbilliferone,aes- culetin, scopoletin). The most important are the hydroxycinnamic acids (p-coumaric, caffeic, feru- lic, sinapic) and derivatives. Cinnamyl alcohols (coniferyl alcohol or guaiacyl, sinapyl alcohol or syringyl, and p- coumaryl alcohol orp-hydroxyphenyl) form the basic con- Dihydroflavonol stituent of , and thus represent one of the major groups of plant phenolics. Phenylpropanoids and more simple phenols (benzoic acid and benzaldehyde deriva- tives) are usually covalently linked to cell wall polysac- charides (predominantly -linked to arabinose units Flavanol of hemicellulose) or to the so-called core .11*'2 Flavonoids (Table 2) represent the most common and widely distributed group of plant phenolics. Their com- Flavandiol or mon structure is that of diphenylpropanes ((2,-C,-C,) and consists of two aromatic rings linked through three car- bons that usually form an oxygenated heterocycle. Figure 1 represents the basic structure and the system used for carbon numbering of the flavonoid nucleus. Biogeneti- cally, the A ring usually comes from a molecule of resorci- no1 or synthesized in the acetate pathway, whereas ring B is derived from the shikimate pathway.I3 Flavonoids occasionally occur in plants as aglycones, al- Biflavonoids though they are most commonly found as glycoside de- rivatives. Among the flavonoids, flavones (e.g., , luteolin, diosmetin), flavonols (e.g., , , ), and their glycosides are the most common

Proanthocyanidins or condensed tannins 3'

Simple Phenols and Flavonoids Among the most common and important low-molecular- weight phenolic compounds are simple phenolic deriva- tives and flavonoids. Simple phenols (C,), such as itself, cresol, , resorcinol, orcinol, etc., are wide- spread among different plant species, including hydro- 5 4 quinone and derivatives (e.g., arbutine, sesamol) and phlo- roglucinol. Phenolics with a C,-C, structure (Table l), such Figure 1. Basic structure and numbering system of flavonoids.

Nutrition Reviews, Vol. 56, No. 11 319 compounds. They are widespread in the plant kingdom, third group of tannins, the , are found only with the exception of algae and fungi. Flavonols occur as in marine brown algae and are not commonly consumed 0-glycosides, but flavone 0-glycosides and C-glycosides by humans.17 are very common,14 with the latter characterized for pos- Hydrolyzable tannins. Hydrolyzable tannins consist sessing a carbon-carbon linkage between the anomeric of and its dimeric condensation product, carbon of a sugar molecule and the C-6 or C-8 carbon of hexahydroxydiphenicacid, esterified to a polyol, which is the flavone nucleus. Unlike 0-glycosides, sugars in C- mainly glucose.18These metabolites can oxidatively con- glycosides are not cleaved by acid hydrolysis. Flavanones dense to other galloyl or hexahydroxydiphenic molecules (e.g., , ) also can occur as 0- or C- and form high-molecular-weightpolymers. As their name glycosides and are especially abundant in citrus foods indicates, these tannins are easily hydrolyzed with acid, and . The variability of this group of flavonoids is alkali, and hot water and by enzymatic action, which yield noteworthy, with about 380 flavonol glycosides and 200 polyhydric alcohol and phenylcarboxylic acid. According different quercetin and kaempherol glycosides described to the nature of the latter, hydrolyzable tannins can be to date? (e.g., , ), with ring further subdivided into , which are derived B of the flavone molecule attached to the carbon 3 of the from gallic acid, or , which are derived from heterocycle, especially occur in . hexahydroxydiphenic acid and which take their name from Flavonoids (e.g., , epicatechin, gallocatechin) the lactone . The best-known hydrolyzable tan- are the monomeric constituents of the condensed tannins, nin is (Figure 2), which is a con- although they are also very common as free monomers. sisting of a pentagalloyl glucose molecule that can further are the most important group of water- esterify with another five gallic acid units. soluble plant pigments and are responsible for the color Condensed tannins. Condensed tannins or proantho- of flowers and fruits of higher plants. The term azthocya- are high-molecular-weight polymers. The mo- nin refers to the glycosides of anthocyanidin (e.g., nomeric unit is a -3-01 (catechin, epicatechin, etc.), pelargonidin, malvidin, ). In addition to with a flavan-3,4-diol or leucoanthocyanidin molecule as glycosylation,common linkages with aromatic and aliphatic its precursor. Oxidative condensation occurs between car- acids, as well as methyl ester derivatives, also occur. An- bon C-4 of the heterocycle and carbons C-6 or C-8 of adja- thocyanins and polymeric pigments formed from antho- cent units. - cyanins by condensation with other flavonoids are re- Much of the literature on the condensed con- sponsible for the color of red wine.l* tent of different plants refers only to oligomeric proantho- Simple phenols and flavonoids represent the vast cyanidins (dimers, trimers, tetramers), because ofthe diffi- majority of plant phenolics. Most of these compounds are culty in analyzing highly polymerized molecules. of relatively low molecular weights and are soluble ac- , however, can occur as polymers with cording to their polarity and chemical structure (degree of degrees ofpolymerization of 50 and greater. The most com- hydroxylation, glycosylation, acylation, etc.). Some of monly described condensed tannins have molecular them, however, can be linked to cell wall components weights of approximately 5000 Da, although, as previously (, lignin). Because of the nature of the mentioned, polymers with molecular weights greater than ester linkages, these compounds can be solubilized in al- 30,000 Da have been discovered.16Autooxidativeor enzy- kaline conditions or are otherwise retained in the fiber matic of flavan-3-01 and flavan-3,4-diol matrix. units has been suggested as the process leading to the formation of condensed tannins.lV9Interflavanoid linkages Tannins

Unlike the previously described groups of plant pheno- OH lics, tannins are compounds of intermediate to high mo- lecular weight. Tannins with a molecular mass of up to 30,000 Da have been found in carob pods (Leguminosae).I6 Tannins are highly hydroxylated molecules and can form insoluble complexes with and protein. This function of plant tannins is responsible for the astrin- gency of tannin-rich foods, because of the precipitation of salivary proteins. The term “tannin” comes from the tanning capacity of these compounds in transforming ani- mal hides into leather by forming stable tannin-protein complexes with skin collagen. OH Plant tannins can be subdivided into two major groups: (1) hydrolyzable and (2) condensed tannins. A Figure 2. Structure of tannic acid.

320 Nutrition Reviews, Vol. 56, No. 11 are acid labile and yield during acid hy- polyphenolic compounds. As a result, information in the drolysis in alcoholic solutions. This reaction is used for literature on the content and composition of polyphenols determination of molecules. in plant foods is not only incomplete but sometimes also Phlobaphene-like substances also are formed when con- contradictory and difficult to compare. densed are heated in mineral acid solutions from Table 3'2' lists the polyphenolic content of different the further polymerization of these compounds.' foods and beverages. Most of the polyphenols listed are Oligomeric proanthocyanidins and low-molecular- phenolic acids and flavonoids (including anthocyanins, weight hydrolyzable tannins are soluble in different aque- , flavanones, flavanols, etc.); fewer are ous and organic solvents, such as acetone, methanol, and tannins. Nonetheless, as mentioned previously, tannins water. However, high-molecular-weight condensed and are often underestimated when polyphenols are analyzed hydrolyzable tannins are insoluble. In addition, when only in food extracts. tannins form complexes with protein or cell wall polysac- As illustrated in Table 3, the polyphenolic content of charides,they remain insoluble. This insolubility oftannins plant foods can vary by several orders of magnitude. In is responsible for significant errors in the quantification legumes and cereals, the main polyphenols are flavonoids, of the polyphenolic content of plants, because polyphe- phenolic acids, and tannins. Polyphenolic content in ce- nols usually are analyzed in extracts, often omitting the reals is usually less than 1% of dry matter, except for some quantification of insoluble or nonextractable tannins. sorghum (Sorghum bicolor) cultivars, which can have as much as 10%. Legumes with higher polyphenolic content Polyphenols in Foods are the dark varieties, such as red kidney beans, black Polyphenols are almost ubiquitous in plant foods (veg- beans (Phaseolus vulgaris), and black gram (Vigna etables, cereals, legumes, fruits, nuts, etc.) and beveqges ). Legumes also contain isoflavones, whereas veg- (wine, cider, beer, , cocoa, etc.). Their levels vary greatly etables are composed primarily of flavonoid glycosides. even between cultivars of the same species. For example, These are present mainly in the outer parts of the plant;14 the formation of flavone and flavonol glycosides greatly roots and tubers have very low concentrations of fla- depends on light; therefore, the highest concentrations vonoids, with the exception of certain plants, such as on- of these compounds are found generally in leaves and and Berries are characterized by their high outer parts of plants, with only trace amounts in the sub- content, whereas fruits such as and terranean parts of ~1ants.l~The presence of polyphenols citrus fruits are rich in phenolic acids and flavonoids, re- in plant foods is largely influenced by genetic factors and spectively. The predominant phenolic compound in fruits environmental conditions. Other factors, such as germi- is flavonol, and the highest concentrations occur in the nation, degree of ripeness, variety, processing, zpd stor- skin.L,2',22Nuts are rich in tannins; the polyphenols in oil age, also influence the content of plant phenoli~s.'~~~~~~~~~,~~seeds are mainly phenolic acids, and oil contains Polyphenols are partially responsible for the sensory both phenolic acids and hydrolyzable tanni11s.2~ and nutritional qualities of plant foods. The astringency The polyphenolic content of fruit is usually and bitterness of foods and beverages depends on the in the range of 2-500 mg/mL, although juices from cer- content of polyphenolic compounds. Oxidation of polyphe- tain varieties have much higher values (up to nols during processing or storage will result in either ben- 7000 mg/mL) owing to their extremely high eficial or undesirable characteristics in food products. For (hesperidin) ~ontents.*~,~~Fermentation of tea leads to example, oxidative changes such as the browning of co- important differences in the polyphenolic composition coa during processing or the oxidative polymerization of of tea leaves: is very rich in flavanols, whereas tea polyphenols during the manufacture of re- black tea contains large amounts of oxidized polyphe- sult in the development of distinctive and desirable orga- nols ( and ~).2~,~~ noleptic properties. Conversely, the enzymatic browning is the main phenolic constituent of beans. The reaction of phenolic compounds (catalyzed by polyphe- major polyphenol in cocoa beans is the flavanol no1 oxidase) and nonenzymatic browning reactions are epicatechin, and cocoa beans are also rich in anthocya- responsible for the formation of undesirable color and fla- nins and tannins. The polyphenols in wine include phe- vor in hits and vegetables.20J' nolic acids, anthocyanins, tannins, and other fla- There is a large body of literature on the polyphe- vonoids. There are significant variations between the nolic composition and content of plant foods and bever- polyphenolic content of white and red wines (200-300 ages. Because of the complexity of this wide group of versus 1000-4000 mg/mL, respectively) and between plant metabolites, however, many polyphenols remain young and aged wines, with important differences in unidentified.Moreover, it is difficult to compare data within the nature of the polyphenols present in aged wines the literature, owing to the lack of agreement on an appro- compared with those found in free-run juices and priate method to analyze the different types or families of young ~ines.'~~~~

Nutrition Reviews, Vol. 56, No. 11 32 1 Table 3. Polyphenolic Content of Different Plant Foods and Beverages1vz1 Food/Beverage" Total Polmhenols Food/Beverage" Total Polyphenoh Cereals (mg/lOO g dm) Fruits (mg/100 gfin) Barley 1200-1 500 Blackcmant 140-1200 Corn 30.9 135-280 Millet 590-1060 Cheny 60-90 Oats 8.7 Cowberry 128 Rice 8.6 77-247 Sorghum 170-1 0,260 Gooseberry 22-75 Wheat 22-40 Grape 50490 Grapesuit 50 Legumes (mg/lOO g dm) Orange 50-100 Black gram 540-1200 10-150 78-230 Pear 2-25 Cowpeas 175-590 Plum 4-225 Common beans 34-280 Raspberry 37-429 Green gram 440-800 Red currant 17-20 Pigeon peas 380-1710 Strawberry 38-218 Tomato 85-130 Nuts (% dm) Betel nuts 26-33 Fruit juices (ma) Cashew nuts 33.7 2-16 0.04 Orange juiceb 370-7100 nuts 8-14 660-1000 Beverages Vegetables (mg/100 g fin) Tea leaves (% dm) Brussels sprouts 6-1 5 Green 20-35 Cabbage 25 Black 22-33 Leek 20-40 Tea, cup (mg/200mL) 150-210 100-2025 Coffee beans (% dm) 0.2-10 Parsley 55-1 80 Coffee, cup (mg/150 mL) 200-550 Celery 94 Cacao beans(% dm) 12-18 wie(ma) Fruits (mg/lOO g fm) White 200-300 Apple 27-298 Red 1OO(r-4000 (6500) 30-43 Beer (mg/L) 60-100 Odm=dry matter; fm=fresh matter. bValuesfor different orange varieties.

Dietary Intake of Polyphenols than flavonoids. Thus, accurate estimation of total poly- phenolic intake is not available. Currently, there is no accurate information available on dietary intake of polyphenols; only a few estimations are Bioavailability of Polyphenolic Compounds available in the literature. Kiihnau' estimated the average daily intake of dietary flavonoids in the United States to It is important for a nutritionist to know not only a person's be between 1 and 1.1 g/day, depending on the season. daily intake of dietary polyphenols but also the Hertog et al?7 calculated the intake of two types of fla- bioavailability of those ingested polyphenols, since their vonoids-flavonols and flavones-in the Dutch diet, and nutritional significance and potential systemic effects will found it to be 23 mg/day. This figure is significantly smaller greatly depend on their behavior in the digestive tract. than Kiihnau's estimation of 115 mg/day for these two This is not a straightforward matter, however, and little is flavonoids, which is allegedly overestimated because of known about the absorption of polyphenols in the gas- the unreliability of the analytic methods employed during trointestinal tract, whether they are retained in the body the 1970s.2*More recently, Leth and JustesenZ9estimated after absorption,and what their biologic significancemight the intake of flavones, flavonols, and flavanones in Den- be. mark to be 28 mg/day, similar to that reported by Hertog et The absorption and metabolism of food phenolics are al. These studies, however, only contemplate the intake of determined primarily by their chemical structure, which some types of flavonoids and do not consider other phe- depends on factors such as the degree of glycosylationl nolic compounds. Moreover, it should be noted that the acylation, their basic structure (i.e., benzene or flavone actual content of polyphenols in foods is usually under- derivatives), conjugation with other phenolics, molecular estimated because of omission of the analysis of insoluble size, degree of polymerization, and . The enor- polyphenols, which may be quantitatively more important mous variability of this group of substances, as well as

322 Nutrition Reviews, Vol. 56, No. I1 their occurrence in plant materials as a complex mixture of bility illustrate their varying susceptibility to digestion, phenolic compounds, creates great difficulties in the study fermentation, and absorption within the gastrointestinal of their bioavailability and their physiologic and nutri- tract.3u* These findings prompted the authors to suggest tional effects. a classification of polyphenolic compounds for nutritional Efforts in this respect, however, have been made. Ex- purposes. Such classification distinguishes between ex- periments reported in the literature have used extracts of tractable and nonextractable polyphenols. Extractable different plant materials that contain a mixture of soluble polyphenols are low- and intermediate-molecular-mass phenolic compounds or pure standards used as supple- phenolics that can be extracted using different solvents ments in complex foods administered either to laboratory (water, methanol, aqueous acetone, etc.) and include some animals or to human volunteers. When plant extracts are hydrolyzable tannins and proanthocyanidins. Nonex- used, it is possible to gain information on the effect of tractable polyphenols are high-molecular-weight com- their constituent polyphenols as a group, but not on the pounds or phenols bound to or protein that digestive fate and specific effects of individual polyphe- remain insoluble in the usual solvent^.“^.^^ nols. Conversely, differences in the absorption, metabo- When different samples containing both extractable lism, and physiologic effects of food phenolics adminis- polyphenols and condensed tannins (nonextractable tered as supplements, compared with polyphenols that polyphenols) were treated in vitro with digestive enzymes are part of a complex food matrix, cannot be ruled out. (a-amylase, amyloglucosidase, and protease) and dialyzed Finally, the extrapolation of animal data to humans is not to simulate intestinal digestion and absorption, the pres- clear. Attention also should be given to the fact that in ence of both nonextractable and extractable polyphenols most cases, and mainly because of the difficulty in their was observed in the insoluble and soluble fiactions ob- analysis and characterization, the study of the dig%stive tained (Table 43b335).These results suggest the nonavail- fate and physiologic effects of insoluble polyphenols- ability of some polyphenolic compounds, mainly nonex- highly polymerized or bound tannins-is usually ne- tractable polyphenols. glected. All of these limitations represent difficulties to These results were confirmed by in vivo studies of overcome in studying the bioavailability of polyphenolic rats fed fruit products (apple pulp and grape pomace) that compounds and their nutritional significance; caution in contained boQ nonextractable and extractable polyphe- interpreting results is necessary. nols36,37or products with only nonextractable polyphe- Both in vivo and in vitro studies using polyphenolic nols (carob pod concentrate, which is rich in highly poly- compounds with different chemical structures and solu- merized condensed tannins) and extractable polyphenols

Table 4. Presence of Extractable (EPP) and Nonextractable Polyphenols (NEPP) in Different Plant Materials and in Soluble and Insoluble Fractions Obtained After Enzymatic Treatments (“h dry matter) Sample Insoluble Fraction Soluble Fraction % Drv Matter YoDrv Matter Yo Original Sample % Drv Matter % Original Sample Carob NEPP 17.9 17.1 95.5 ND - EPP 13 NIY - 0.72 55.48 Carob pod concentrate2l NEPP 27.6 12.7 46.0 ND - EPP 1.36 ND - 0.59 43.4 Spanish sainf~in~~ NEPP 7.0 5.6 80.0 ND - EPP 1.1 ND - 0.30 27.3 Apple pulp33 NEPP 1.6 1.oo 62.5 ND - EPP 0.34 ND - 0.2 1 61.7 White grape poma~e~~,~~ NEPP 14.5 ND - ND - EPP 4.02 1.70 42.3 1.14 28.4 Red grape p0mace~~3~ NEPP 36.4 ND - ND - EPP 329 1.44 43.8 0.68 20.7 WD=not determined.

Nutrition Reviews, Vol. 56, No. 11 323 (catechin and tannic Nonextractable polyphe- Table 5. Fecal Excretion of Extractable (EPP) and nols were extensively recovered in feces, confirming the Nonextractable Polyphenols (NEPP) in Rats Fed resistance of these compounds to intestinal digestion and/ Different Polyphenot-containingDiets or absorption (Table Conversely, extractable Intake Excretion polyphenols were excreted only in minor amounts, sug- (phveek) dweek YOof intake gesting that extensive digestion andor absorption of these Apple pulp36 polyphenolic compounds occurs in the gut. Similar re- NEPP 0.5 1 035 68.6 sults were obtained by Degen et al.,"3 who found that EPP 0.014 0.002 14.3 extractable polyphenols from Acacia saligna were virtu- ally absent in sheep and goat feces, whereas condensed Grape pomace37 NFPP 5.3 1 521 98.1 tannins were excreted in substantial amounts. EPP 0.19 0.06 31.6 Differences in the percentage of excretion between extractable and nonextractable polyphenols may result Carob pod ~oncentrate~~ because of chemical differences (i.e., molecular size, de- NEPP 6.1 1 5.97 97.7 gree of polymerization of soluble and insoluble polyphe- Tannic a~id~~,~~ nols, nature of the phenolic bound to other food compo- EPP 2.06 0.10 4.6 nents in nonextractable polyphenols, etc.). Distinct ab- Cate~hin~~ sorption of various extractable polyphenols, depending EET 2.11 0.07 3.1 on their extractability with different solvents, was reported by Jimenez-Ramsey et al." These authors used 14C-labeled phenolics from sorghum grains (Sorghum bicoloy) ex- tracted in water, ethanol, or aqueous acetone. When the ing absorption and metabolism of these polyphenols. A polymeric 14C-proanthocyanidin fraction (insoluble in fraction of 14C,made up of phenols resistant to microbial water and ethanol, but soluble in aqueous acetone) was degradation and not absorbed in the intestine, was ex- fed to chickens, no radioactivity was detected in plasma creted in feces. or tissues, with all the radioactivity recovered in the excre- Metabolism Phenolic Compounds ment and in the gastrointestinal tract and its contents. of These results illustrate the nonabsorbable nature of some It seems evident that some polyphenolic compounds, ei- soluble tannins with a low degree of polymerization (con- ther extractable polyphenols or solubilized phenolics, are densed tannins or proanthocyanidins). Conversely, the metabolized within the gastrointestinal tract. Aglycones 14C-nontanninphenolic fractions (monomeric and oligo- and free simple phenolic compounds, flavonoids (querce- meric polyphenols soluble in water and ethanol) were par- tin, genistein) and phenolic acids, can be directly absorbed tially absorbed from the intestinal tract and the label was through the small intestinal m~cosa.~~~Free phenolics extensively detected in all tissues and plasma. ( and derivatives such asp-coumaric, feru- Nonextractable polyphenols that are different from lic, caffeic, etc.) have been shown to be absorbed through highly polymerized tannins (i.e., polyphenols bound to the intestinal tract in both in vivo experiments of rats48 protein or dietary fiber) can be liberated under certain cir- and in vitro experiments of isolated rat jejun~m.4~Con- cumstances and thus made available for digestion. In an versely, glycosides must be hydrolyzed to their corre- interesting in vivo study using insoluble I4C-labeledphe- sponding aglycones before absorption. Because mammals nolic groups bound to cell wall polysaccharides from spin- lack the appropriate P-glycosidases, it was thought that ach (Spinacia oleracea), Buchanan et a1.4s showed that absorption in the small intestine did not occur.sss2 It has ferulic and p-coumaric acids were released from the cell been shown, however, that partial absorption of quercetin wall. Some phenolic acids were released in the upper in- glycosides takes place in the upper intestine, probably testine, partially because of the alkali-labile nature of their owing to the action of glycosidases from bacteria that ester bindings with cell wall polysaccharides and partially colonize the terminal Most glycosides, how- because of the action of fermentative bacteria in the small ever, pass into the large intestine, where they are hydro- intestine (especially in the terminal ileum). These free phe- lyzed by the cecal microflora, rendering free aglycones. nolics were promptly absorbed by the intestinal mucosa, The mediation of bacterial enzymes in the bioavailability with significant amounts of 14C-labelpresent in the body of phenolic glycosides was clearly proved by Griffiths tissues or excreted in urine. However, most 14C-labeled and who showed that flavonoid glycosides were phenolic groups reached the large intestine, where radio- excreted as such in the feces of germ-free rats. Bacterial activity was solubilized from cell wall and free phenolic fermentation of carbohydrates would also liberate pheno- groups (coumaric and ferulic) were released upon bacte- lics bound to dietary fiber, which would be metabolized rial fermentation of cell wall polysaccharides. Radioactiv- like the extractable polyphenols. ity could be detected in body tissues and urine, suggest- In the colon, aglycones are absorbed through the gut

324 Nutrition Reviews, Vol. 56, No. 11 epithelium and methylated and/or conjugated with glucu- echins corresponded to about 0.2-0.9% of the ingested ronic acid or sulfate in the liver. The main organ involved dose. in the metabolism of polyphenols is the liver, although the Only short-term data are available on the permanence implication of other organs such as the kidneys or the of absorbed phenolics in the body. This point is of great intestinal mucosa cannot be ruled out, since they contain importance, because some of the physiologic effects of the enzymes involved in polyphenol metaboli~m.~~ food polyphenols depend on their circulating levels (i.e., Conjugated and 3 '-0-methylated derivatives have antioxidant capacity). Manach et a1.46 observed that con- been detected in the plasma of rats administered flavanols centrations of quercetin metabolites in the plasma of rats (cate~hin~~?~~),flavonols (quercetin, rutin, and isorham- adapted or not adapted to a flavonoid-rich diet did not netin46,55,56), and isoflavones (geni~tein~').These metabo- drastically vary 16 hours after reaching maximum. These lites are secreted in the urine or in the bile. In this case, authors suggested that the rate of elimination of querce- they can enter an enterohepatic cycle when deconjugated tin metabolites was relatively low and that high plasma by the action of the colonic microflora and reabsorbed. concentrations can be easily maintained with a regular Alternatively, they can be fully metabolized and converted supply of flavonoids in the diet. into simple phenolic acids after hydrolysis of their fla- Van het Hof et a1.6' studied the kinetics of absorption vone structure (opening of the heterocycle) mediated by and elimination of tea catechins. These investigators found bacterial enzymes. The hydroxylation pattern will deter- that maximum blood levels occurred 2 hours after tea in- mine the susceptibility of polyphenols to bacterial degra- gestion and that elimination half-life varied between 4.8 dation, with the absence of hydroxyl groups preventing and 6.9 hours for green and black tea catechins, respec- ring cleavage.28,s'The phenolic acids formed as fission tively. These values differed from those reported by products, such as free soluble phenolics, are absrorbed Hollman et al.,63who found that maximum plasma querce- through the intestinal mucosa and excreted in the tin concentrations after the ingestion of , which are .28,50-52,58 rich in quercetin, occurred after 3.3 hours and that the Although evidence of the absorption and metabo- elimination half-life was 16.8 hours. Therefore, there seem lism of polyphenols in the gut exists, less is known about to be important differences in the rate and extent of ab- the efficiency of such uptake and the permanence of phe- sorption and elimination of dietary polyphenols, depend- nolic compounds or their conjugates and derivatives in ing on their ch>mical structure. the body. Animal studies with I4C-labeledphenolics indi- cate that only partial absorption takes place. Thus, only Fermentation of Polyphenolic Compounds 20% of the 14C-quercetin administered to rats was ab- sorbed, 30% was excreted, and the remaining 50% was As mentioned previously, fermentative microflora play metabolized, yielding phenolic acids and C02.s9Similarly, a crucial role in the metabolism of some polyphenolic King et al.47reported an absorption of about 20% of the compounds. However, not all phenolics are equally sus- soy isoflavones administered to rats, with a fecal excre- ceptible to bacterial degradation, because certain com- tion of approximately 2 1% of the ingested dose. No differ- pounds, such as insoluble condensed tannins, are ex- ences between the aglycone and the glycoside were ob- creted in feces apparently without being affected by served in this experiment. Conversely, Buchanan et al.45 the colonic bacteria. In addition to their susceptibility showed that 19% of the gavaged 14C-phenolics(p-coumaric to microbial degradation, dietary polyphenols also can and ferulic acids) bound to cell wall polysaccharides were influence intestinal microflora and their fermentative excreted in rat feces. About 20% ofthe dose was excreted capacity toward other food components. in urine and more than 34% of the label was incorporated In vitro fermentation of highly polymerized con- into body tissues 18 hours after gavage; long-term perma- densed tannins from carob pod using rat cecal con- nence of phenolic compounds was not studied. tents as the inoculum showed that these condensed Human experiments, although restricted to the study tannins were not affected by fermentative mi~roflora.~'*~~ of flavonoids, also show only partial absorption of Also, the levels of short-chain fatty acids (SCFA), such polyphenols. Absorption of orally administered quercetin as acetic, propionic, and butyric acid, which are the in healthy ileostomized individuals varied between 24% primary end products of colonic fermentation, were not and 52% of the ingested aglycone and glycoside, respec- affected by the presence of carob pod condensed ti~ely.~~ isoflavones administered to humans were tannins in the incubation system. This suggests that absorbed by healthy volunteers in a range varying from highly polymerized phenols do not affect intestinal mi- 9% to 21%, depending on the .60Blood concen- croflora. Similarly, cecal contents of rats fed diets con- trations of total catechins of 0.17 pmol/L after ingestion of taining condensed tannins from grape pomace did not black tea and up to 0.55 pmol/L after green tea were re- affect normal in vitro fermentation of apple used ported recently;6' these data agree with those reported by as ino~ulurn.~~This suggests that condensed tannins Lee et a1.,62who estimated that the absorption of tea cat- do not affect the colonic microflora or their fermenta-

Nutrition Reviews, Vol. 56, No. 11 325 tive capacity. Nutritional and Physiologic Effects of Conversely, the analysis of cecal contents of rats Polyphenols fed water-soluble condensed tannins from quebracho (Schinopsis quebracho-colorado), composed of mono- Influence of Polyphenols on the Digestibility of mers, dimers, trimers, and oligomers, showed depoly- Macronutrients merization of polymers, degradation of monomers, and Probably one of the best-known properties of polyphe- accumulation of simple phenolic compounds;66this in- nolic compounds is their capacity to bind and precipitate dicates bacterial degradation of soluble condensed protein. Although this protein-binding capacity is com- tannins of a low degree of polymerization. Contrary to mon to most polyphenols, thanks to their high degree of highly polymerized compounds, quebracho soluble hydroxylation, low-molecular-weightphenols are unable tannins depressed the production of SCFAs, showing to precipitate protein, and it has been shown that oligo- a bacteriostatic effect. mers must contain at least three flavonol subunits to ef- In vitro fermentation of the flavonoid quercetin for fectively precipitate protein.75Highly polymerized tannins 72 hours yielded a reduced production of propionate are the most effective precipitators of protein. Tannin- and butyrate but a high production of acetic acid; this protein complexes are usually established through hydro- suggests that quercetin is fermented by bacterial mi- gen bonds and hydrophobic interactions, without the con- croflora with the opening of the aromatic ring, which tribution of covalent or ionic bond^.^',^^ would explain the high production of acetic acid origi- With regard to nutrition, tannins traditionally have nating from the complete hydrolysis of the flavonoid been considered antinutrients because the presence of and the absence of other SCFAs typically produced tannins in plant foods is usually accompanied by a re- during fermentati~n.~~ e duced digestibility of protein and a subsequent increase Similarly, in vitro fermentation of the flavonoid cat- in fecal nitrogen Similarly, in vitro protein echin and of tannic acid (with gallic acid as the con- digestibility also is reduced in the presence of condensed stituent monomeric phenol) also was characterized by a tannins.30*35*82,83However, the ability of polymeric high relative production of acetic acid after 72 hours of proanthocyanidins (condensed tannins) to form insoluble fermentation, but not after 24 hours, suggesting that protein-polyphenolcomplexes is limited to those molecules the proposed hydrolysis of the aromatic ring needs long physically accessible to soluble proteins. Highly poly- fermentation time^.^^*^' Tannic acid caused a reduced merized tannins (nonextractablepolyphenols) are insoluble production of total SCFAs, indicating an inhibitory ef- compoundsthat usually form part of a complex matrix with fect of this type of phenolic structure on the fermenta- cell wall polysa~charides,~~or insoluble tannin granules,I6 tive microflora. This effect also was observed by Arri- which greatly reduces their protein-binding ability. These goni et al.," who reported that soluble polyphenols (both effects are more pronounced with soluble oligomeric flavonoid and gallic acid structures), but not nonex- proanthocyanidins or hydrolyzable tannins,*' since the tractable polyphenols (condensed tannins), slow down simple phenols have no ability to precipitate pr0tein.7~ the fermentability of polysaccharides. The increased fecal nitrogen excretion after ingestion On the other hand, in vitro fermentation assays of tannin-containing diets is likely caused by an enhanced using rumen microbes showed different degradation of elimination of endogenous protein rather than by a re- quebracho tannins: soluble and extractable proantho- duced digestibility of dietary protein. This was confirmed cyanidins were partly fermented, whereas nonextract- by Shahkhalili et a1.86and Mole et a1.87in experiments of able condensed tannins bound to protein were not de- rats fed I4C- and I'N-labeled proteins, respectively. Con- graded.68,69Both quebracho and tannic acid decreased versely, fecal protein excreted after tannin ingestion is the production of SCFAS.~~Similarly, certain phenolic very rich in proline.88Salivary proline-rich proteins have a acids (ferulic,p-coumaric, and cinnamic acids) have been very high affinity for tannins. Secretion of these tannin- reported to inhibit the growth of rumen microorgan- binding proteins, which is thought to be a mechanism of ism~~~and their fermentative effect on carbohydrates adaptation by consuming high-tannin dietsYs9 and pr~tein,~'.~~although they are metabolized by the is induced by the presence of tannins in the diet.g0 rumen mi~roflora.~~,~~ Furthermore, tannins can bind other endogenous pro- In summary, the degradation and absorption of teins in the intestinal tract, such as digestive enzymes, polyphenols within the gastrointestinal tract depend and inhibit them.85~91-95This causes a reduction in the di- on the nature not only of the phenolic compound but gestibility not only of proteing3but of other macronutri- also of the intestinal microflora, which fermentative ef- ents, such as starch and lipids.%%Inhibition of amylolytic fect on other dietary components will be affected, con- enzymes and the subsequent reduction of dietary carbo- versely, by the type of polyphenolic compound. hydrate hydrolysis can decrease the postprandial glyce-

326 Nutrition Reviews, Vol. 56, No. 11 mic response.97Likewise, polyphenols also can form com- have been shown to reduce zinc absorption in rats,l17 and plexes with polysaccharides other than those that form negative effects of polyphenols also have been observed the plant cell wall (i.e., star~h)~~,~~and affect the glycemic on the bioavailability of sodium118and aluminum1I9but and insulinemic responses as well. not manganese,1z0calcium, or magne~ium."~ The effect of food polyphenols on lipid metabolism has not been extensively studied. Both soluble polyphe- Antioxidant Activity of Food Polyphenols nols and condensed tannins have been shown to increase Recent interest in food phenolics has increased owing to fecal fat ex~retion.~~~~J~~O~In addition, hypocholesterol- their roles as , antimutagens, and scavengers emic effects have been reported in animals fed diets con- of free radicals and their implication in the prevention of taining grape tannic acid,IMand tea cat- pathologies such as cancer and cardiovascular disease. echins,lo5with increased plasma levels of high-density Epidemiologic studies have shown a correlation between lipoprotein (HDL) and reduced concentrations an increased consumption of phenolic antioxidants and a of low-density lipoprotein (LDL) cholesterol. This reduced risk of cardiovascular disease121-1uand certain hypocholesterolemic action of dietary polyphenols is types of Similarly, moderate consumption of mediated by an enhanced reverse-cholesterol transport red wine, which is rich in polyphenols, has been associ- and by reduced intestinal cholesterol absorption and in- ated with a low risk of coronary heart disea~e.~~~J~~ creased bile acid excretion.'01J02The exact mechanism of Phenolic antioxidants function as terminators of free action, however, is not known. radicals and chelators of metal ions that are capable of catalyzing lipid peroxidation. Phenolic antioxidants inter- Influence of Polyphenols on Bioavailability of fere with the oxidation of lipids and other molecules by Minerals rapid donation of a hydrogen atom to radicals, as illus- Polyphenols can form complexes with meTa1 cations trated in the following reactions: through their carboxylic and hydroxylic groups, and thus ROO*+ PPH +ROOH + PP* interfere with the intestinal absorption of minerals. Nu- RO*+ PPH +ROH + PP* merous experiments in both humans and animals have Moreover, the phenoxy radical intermediatesare relatively shown that polyphenols strongly inhibit iron absorp- stable; therefore, a new chain reaction is not easily tion.Iob"' This action has been attributed to the galloyl inititated.-The phenoxy radical intermediates also act as and groups of polyphenolic compounds.Io6Mo- terminators of the propagation route by reacting with other nomeric flavonoids in green and herb (catechins),lo" free radicals:1z7 108~110phenolic acids in coffee (chlorogenic acid),IMpoly- ROO- + PP*+ ROOPP merized products in black tea and cocoa,III and wine RO* + PP*+ ROPP polyphenol~~~~have been shown to reduce iron However, under certain conditions (high concentrations bioavailability. Conversely, tannins from soybean protein, of phenolic antioxidants, high pH, presence of iron), phe- chickpeas, and red kidney beans had no significant effect nolic antioxidants can initiate an autooxidation process on iron absorption,"2 suggesting a lack of effect of con- and behave like pro oxidant^.^^^ densed tannins (nonextractable polyphenols). This find- The efficiency of polyphenols as antioxidant com- ing was not, however, confirmed by Jansman et al.,Il3who pounds greatly depends on their chemical structure. Phe- reported a reduced absorption of iron and copper in pigs nol itself is inactive as an antioxidant, but ortho- andpuru- fed condensed tannins from fava beans. diphenolics have antioxidant capacity, which increases Reduced copper absorption after consumption of tea with the substitution of hydrogen atoms by ethyl or n- also has been observed in humans,114but contrary results butyl groups.'27Flavonoids are among the most potent were reported by Vaquero et al.,I15 who observed an in- plant antioxidants because they possess one or more of creased absorption of 'Wu and an enhanced retention of the following structural elements involved in the antiradi- copper in the liver of rats fed tea. cal activity (Figure 3): (1) an o-diphenolic group (in ring Although the chelating action of polyphenols on B), (2) a 2-3 double bond conjugated with the 4-0x0 func- metals such as copper and iron can have negative effects tion, and (3) hydroxyl groups in positions 3 and 5.12"130 by reducing their bioavailability, this action can be benefi- Quercetin, a flavonol that combines all of these character- cial in certain circumstances. In the native state, copper istics, is one of the most potent natural antioxidants. Also, and iron can be the initiators of hydroxyl radical produc- the antioxidant efficiency of flavonoids is directly corre- tion by the Fenton and Haber-Weiss reactions.Il6 Chela- lated with their degree of hydroxylation and decreases tion of these metals is one of the ways polyphenols exert with the presence of a sugar moiety (glycosides are not their antioxidant activity. antioxidants, whereas their corresponding aglycones are With regard to the effect of polyphenols on the avail- antioxidant^).'^^ ability of other minerals, chlorogenic and caffeic acids Flavonoids are very effective scavengers of hydroxyl

Nutrition Reviews, Vol. 56, No. 11 32 7 OH tion ofmetal ions (for areview see reference 145). Through these antioxidant actions, polyphenols exert their protec- tive effect against cardiovascular disease. In addition, fla- vonoids have antithrombotic and vasoprotective effects as well as hypolipidemic effects, as discussed previously. Different types of polyphenols (phenolic acids, hy- drolyzable tannins, and flavonoids) also have been shown to have anticarcinogenic Polyphenols might interfere in several of the steps that lead to the develop- ment of malignant tumors, thereby protecting DNA from oxidative damage, inactivating carcinogens, inhibiting the expression of mutant genes and the activity of enzymes involved in the activation of procarcinogens, and activat- Figure 3. Structure of quercetin showing the structural charac- ing enzymatic systems involved in the detoxification of teristics related to its antioxidant capacity. xenobiotics (for a review see reference 129). Some polyphe- nols, however, also have been shown to have mutagenic and peroxyl radicals, although their efficiency as scaven- activity in microbial assays, although contradictory re- gers of the superoxide anion is not yet clear.L29As men- sults, depending on the type of assay used and type of tioned previously, polyphenols are chelators of metals and phenolic studied, have been reported, as reviewed by inhibit the Fenton and Haber-Weiss reactions, which are ~rown.149 important sources of active oxygen radicals.lz6In-addi- tion, flavonoids retain their free radical-scavenging ca- Conclusion pacity after forming complexes with metal ions.13' Although the number and variability of food phenolics Although antioxidant activity traditionally has been make the study of this immense group of metabolites dif- attributed only to soluble phenolic compounds (extract- ficult, their nutritional significance as well as their poten- able polyphenols), a recent report suggests that tially beneficial health effects call for detailed studies. nonextractable polyphenols (polymeric proanthocyanidins Polyphenols may have important applications in the pre- and high-molecular-weight hydrolyzable tannins) are 15 vention and treatment of highly prevalent human diseases, to 30 times more effective at quenching peroxyl radicals such as cardiovascular disease and cancer, as well as gas- than are simple phen01s.I~~Because these compounds are tric and duodenal ulcer, allergy, vascular fragility, viral and not absorbed, they could exert their antioxidant activity bacterial infections, etc. To fully understand the actual within the digestive tract and protect lipids, proteins, and significance of food phenolics, it is necessary to investi- carbohydrates from oxidative damage during digestion, gate not only their bioavailability but also their mecha- and spare soluble antioxidants. nisms of action and their possible synergism with other Most studies have shown the antioxidant activity of constituents either in the diet or within the human body, polyphenols using different in vitro model^,'^^'^^ and sub- as well as the polyphenolic content and composition of sequently, phenolic compounds are classified according foods. These factors constitute the body of future re- to their antioxidant capacity or antiradical effi- search. ~iency.'*~J~'J~*The role of polyphenols in vivo is not clear. The antioxidant efficiency of polyphenols depends on the extent of absorption and metabolism of these compounds, 1. Kuhnau J. The flavonoids: a class of semi-essen- tial food components: their role in . as well as the activity of methoxylated and conjugated World Rev Nutr Diet 1976;24:117-91 forms circulating in plasma. As mentioned before, only 2. Singleton VL. Naturally occurring food toxicants: partial amounts of food polyphenols are absorbed in phenolic substances of plant origin common in V~VO,~~~and only very low levels of tea catechins were foods. Adv Food Res 1981;27:149-242 detected in plasma after tea ingestion.61.62Nevertheless, 3. Saito M, Hosoyama H, ArigaT, et al. Antiulcer activ- ity of grape seed extract and procyanidins. J Agric these low concentrations seem sufficient to exert a potent Food Chem 1998;46: 1460-4 antioxidant action in vivo, as observed in human stud- 4. HarborneJB. The flavonoids: advances in research and as suggested by epidemiologic data.12'-124 since 1986. London: Chapman and Hall, 1993 Antioxidant polyphenols, mainly flavonoids, are po- 5. Harborne JB. Methods in plant biochemistry, I: plant tent inhibitors of LDL o~idation.'*~J~~J~~J~~Several mecha- phenolics. London: Academic Press, 1989 HarborneJB. The flavonoids: advances in research nisms by which flavonoids exert their protective effect 6. since 1980. London: Chapman and Hall, 1988 have been proposed: (1) reduction of free radical forma- 7. Hemingway RW, Karchesy JJ. Chemistry and sig- tion, (2) protection of a- in LDL from oxidation, nificance of condensed tannins. New York: Plenum (3) regeneration of oxidized a-tocopherol, and (4) chela- Press, 1989

328 Nutrition Reviews, Vol. 56, No. 1 I 8. Markham KR. Techniques of flavonoid identifica- vegetables and beverages by HPLC-UV and LC- tion. London: Academic Press, 1982 MS and estimation of the total daily flavonoid in- 9. Haslam E. Vegetable tannins. London: Academic take in Denmark. In: Amado R, Andersson H, Press, 1966 Bardocz S, Serra F, eds. Polyphenols in food. Lux- 10. Scalbert A. Polyphenolic phenomena. Paris: INRA embourg: Office for Official Publications of the Eu- Editions, 1993 ropean Communities, 1998;39-40 11. Jung HG. Forage lignins and their effects on fiber 30. Saura-Calixto F. Effect of condensed tannins in the digestibility. Agron J 1989;81:33-8 analysis of dietary fiber in carob pods. J Food Sci 12. Wallace G, Chesson A, Lomax JA, Jarvis MC. Lig- 1988;53:1769-71 nin-carbohydrate complexes in graminaceous cell 31. Bravo L. Propiedades de 10s compuestos walls in relation to digestibility. Anim Feed Sci Tech polifenolicos como constituyentes de la dieta: 1991;32:193-9 comparacion con componentes de lafibra dietetica 13. Harborne JB, Mabry TJ. The flavonoids: advances [Doctoral dissertation]. Madrid: Universidad in research. London: Chapman and Hall, 1982 Autonoma de Madrid, 1993 14. Herrmann H. On the occurrence of flavonol and 32. Saura-Calixto F. Dietary fibre complex in a sample flavone glycosides in vegetables. Z Lebensm rich in condensed tannins and uronic acids. Food Unters Forsch 1988;186:1-5 Chem 1987;23:95-103 15. Mazza G. Anthocyanins in and grape prod- 33. Gofii I, Torre M, Saura-Calixto F. Determination of ucts. Crit Rev Food Sci Nutr 1995;35:341-71 dietary fibre in cider wastes: comparison of meth- 16. Wursch p del Vedovo S, Rosset J, Smiley M. The ods. Food Chem 1989;33:151-9 tannin granules from ripe carob pod. Lebens Wiss 34. Saura-Calixto F, Gofii I, Mafias E, Abia R. Klason Techno1 1984;17:351-4 lignin, condensed tannins, and resistant protein as 17. Ragan MA, Glombitza K. Phlorotannins: brown al- dietary fibre constituents: determination in grape gal polyphenols. Prog Phycol Res 1986;4:177-241 pomaces. Food Chem 1991;39:299-309 18. Porter LW. Tannins. In: Harborne JB, ed. Methods 35. Bravo L, Saura-Calixto F. Characterization of di- in plant biochemistry, I: plant phenolics. London: etary fiber and the in vitro indigestible fraction of Academic Press, 1989;389-419 grape pomace. Am J Enol Vitic 1998;49:135-41 19. Peleg H, Naim M, Rouseff RL, Zehavi U. Distribu- 36. Bravo L, Saura-Calixto F, Gofii 1. Effects of dietary tion of bound and free in oranges fibre and tannins from apple pulp on the composi- (Citrus sinensis) and grapefruits (Citrus paradisi). tion of faeces in rats. Br J Nutr 1992;67:463-73 J Sci Food Agric 1991;57:417-26 37. Ma@-Carron N, Garcia-Alonso A, GoAi I, Saura- 20. Ho C-T, Lee CY, Huang M-T. Phenolic compounds Calixto F. Nutritional and physiological properties in food and their effects on health, I: analysis, oc- of grape pomace as a potential food ingredient. currence, and chemistry. ACS Symposium Series Am J Enol Vitic 1997;48:328-32 506. Washington, DC: American Chemical Soci- 38. Bravo L, Mafias E, Saura-Calixto F. Dietary ety, 1992 nonextractable condensed tannins as indigestible 21. Shahidi F, Naczk M. Food phenolics: sources, chem- compounds: effects on faecal weight, and protein istry, effects, applications. Lancaster, PA: Technomic and fat excretion. J Sci Food Agric 1993;63:63-8 Publishing Co, 1995 39. Bravo L, Abia R, Eastwood MA, Saura-Calixto F. 22. Hertog MGL, Hollman PCH, Katan MB. Content of Degradation of polyphenols (catechin and tannic potentially anticarcinogenic flavonoids of 28 veg- acid) in the rat intestinal tract: effect on colonic fer- etables and 9 fruits commonly consumed in The mentation and faecal output. Br J Nutr Netherlands. J Agric Food Chem 1992;40:2379- 1994;71:933-46 83 40. Bravo L, Abia R, Saura-Calixto F. In vivo and in vitro 23. Visioli F, Galli C. The effect of minor constituents of degradation of tannic acid. in: Malkki Y, Cummings on cardiovascular disease: new findings. JH, eds. Dietary fibre and fermentation in the co- Nutr Rev 1998;56:142-7 lon. COST Action 92. Luxembourg: Office for Offi- 24. Hertog MGL, Hollman PCH, van der Putte B. Con- cial Publications of the European Communities, tent of potentially anticarcinogenic flavonoids of 1996;181-5 tea infusions, wines, and fruits juices. J Agric Food 41. Bravo L, Abia R, Saura-Calixto F. Polyphenols as Chem 1993;41:I 242-6 dietary fiber associated compounds: comparative 25. Shao W, Powell C, Clifford MN. The analysis by study on in vivo and in vitro properties. J Agric Food HPLC of green, black, and Pu’er teas produced in Chem 1994;42:1481-7 Yunnan. J Sci Food Agric 1995;57:417-26 42. Saura-Calixto F, Bravo L. Intestinal degradation of 26. Bakker J, Preston NW, Timberlake CF. The deter- polyphenols. In: Malkki Y, Cummings JH, eds. Di- mination of anthocyanins in aging red wines: com- etary fibre and fermentation in the colon. COST parison of HPLC and spectral methods. Am J Enol Action 92. Luxembourg: Office for Official Publica- Vitic 1986;37:121-6 tions of the European Communities, 1996;87-92 27. Hertog MGL, Hollman PCH, Katan MB, Kromhout 43. Degen AA, Becker K, Makkar HPS, Borowy N. Aca- D. Intake of potentially anticarcinogenic flavonoids cia saligna as a fodder tree for desert livestock and and their determinants in adults in The Nether- the interaction of its tannins with fibre fractions. J lands. Nutr Cancer 1993;20:21-9 Sci Food Agric 1995;68:65-71 28. Hollman PCH. Bioavailability of flavonoids. Eur J 44. Jimenez-Ramsey LM, Rogler JC, Housley TL, et al. Clin Nutr 1997;51 (suppl 1):S66-9 Absorption and distribution of 14C-labeledcon- 29. Leth T, Justesen U. Analysis of flavonoids in fruits, densed tannins and related sorghum phenolics in

Nutrition Reviews, Vol. 56, No. I1 329 chickens. J Agric Food Chem 1994;42:963-7 63. Hollman PCH, Van der Gaag M, Mengelers MJB, et 45. Buchanan CJ, Wallace G, Fry SC, Eastwood MA. In at. Absorption and disposition kinetics of the di- vivo release of 14C-labeled phenolic groups from etary antioxidant quercetin in man. Free Radic Biol intact dietary spinach cell walls during passage Med 1996;21:703-7 through the rat intestine. J Sci Food Agric 64. Arrigoni E, Balzer Amado R. Influenceof polyphe- 1996;71:459-69 nols on the fermentability of polysaccharides. In: 46. Manach C, Morand C, Demigne C, et al. Amado R, Andersson H, Bardocz S, Serra F, eds. Bioavailability of rutin and quercetin in rats. FEBS Polyphenols in food. Luxembourg: Office for Offi- Lett 1997;409:12-6 cial Publications of the European Communities, 47. King RA, Broadbent JL, Head RJ. Absorption of the 1998; 1 87-92 soy isoflavone genistein in rats. J Nutr 65. Martin-Carron N, Gofii I. Prior exposure of cecal 1 996; 126: 176-82 microflora to grape pomaces does not inhibit in 48. Jung HJG, Fahey GC. Effects of phenolic mono- vitro fermentation of pectin. J Agric Food Chem mers on rat performance and metabolism. J Nutr 1998;46:1064-70 1983;1 13:546-56 66. Levrat A-M, Texier 0, Regerat F, et at. Comparison 49. Wolffram S,Weber T, Grenacher B, Scharrer E. A of the effects of and pectin on Na+-dependentmechanism is involved in mucosal cecal fermentations and lipid metabolism in the rat. uptake of cinnamic acid across the jejunal brush Nutr Res 1993;13:427-33 border in rats. J Nutr 1995;125:1300-8 67. Mafias E, Lebet V, Arrigoni E, Amado R. In vitro 50. Williams RT. Metabolism of phenolics in animals. studies of polyphenol fermentability. In: Malkki Y, In: Harborne JB, ed. Biochemistryof phenolic com- Cummings JH, eds. Dietary fibre and fermentation pounds. London: Academic Press, 1964;205-48 in the colon. COST Action 92. Luxembourg: Office 51. Griffiths LA. Mammalian metabolism of flavonoids. for Official Publications of the European Commu- In: Harborne JB, Mabry TJ, eds. The flavonoids: nities, 1996;186-90 advances in research. London: Chapman and Hall, 68. Makkar HPS, Blummel M, Becker K. In vitro effects 1982;681-718 of and interactions between tannins and 52. Hackett AM. The metabolism of flavonoid com- and fate of tannins in the rumen. J Sci Food Agric pounds in mammals. In: Cody V, Middleton E, 1995;69:481-93 Harborne JB, eds. Plant flavonoids in biology and 69. Makkar HPS, Becker K, Abet H, Szegletti C. Degra- medicine: biochemical,pharmacological, structure- dation of condensed tannins by rumen microbes activity relationships. New York: Alan R. Liss, Inc, exppsed to quebracho tannins (QT) in rumen simu- 1986; 1 77-97 lation technique (RUSITEC) and effects of QT on 53. Hollman PCH, de Vries JHM, van Leeuwen SD, et fermentative processes in the RUSITEC. J Sci Food al. Absorption of dietary quercetin glycosides and Agric 1995;69:495-500 quercetin in healthy ileostomyvolunteers. Am J Clin 70. Chesson A, Stewart CS, Wallace RJ. Influence of Nutr 1995;62:1276-82 plant phenolic acids on growth and cellulolytic ac- 54. Griffiths LA, Barrow A. Metabolism of flavonoid com- tivity of rumen bacteria. Appl Environ Microbiol pounds in germ-free rats. Biochem J 1982;44:597-603 1972;130:1161-2 71. Akin DE. Forage cell wall degradation and p- 55. Manach C, Morand C, Texier 0, et al. Quercetin coumaric, ferulic, and sinapic acids. Agron J metabolites in plasma of rats fed diets containing 1982;74:424-8 rutin or quercetin. J Nutr 1995;125:1911-22 72. Jung HJG. Inhibition of structural carbohydrate fer- 56. Manach C, Texier 0, Regerat F, et at. Dietary quer- mentation by forage phenolics. J Sci Food Agric cetin is recovered in rat plasma as conjugated de- 1985;36:74-80 rivatives of isorhamnetin and quercetin. J Nutr 73. Jung HJG, Fahey GC. Interactions among phenolic Biochem 1996;7:375-80 monomers and in vitro fermentation. J Dairy Sci 57. Sfakianos J, Coward L, Kirk M, Barnes S. Intestinal 1983;66:1255-63 uptake and biliary excretion of the isoflavone 74. Grabber JH, Jung GA. In vitro disappearance of genistein in rats. J Nutr 1997;127:1260-8 carbohydrates, phenolic acids, and lignin from pa- 58. Martin AK. The origin of urinary aromatic com- renchyma and sclerenchyma cell walls isolated pounds excreted by , 2: the metabolism from cocksfoot. J Sci Food Agric 1991;57:315-23 of phenolic cinnamic acids to benzoic acid. Br J 75. Hagerman AE. Chemistry of tannin-protein com- Nutr 1982;47:155-64 plexation. In: Hemingway RW, Karchesy JJ, eds. 59. Ueno I, Nakano N, Hirono 1. Metabolic fate of “*C] Chemistry and significance of condensed tannins. quercetin in the ACI rat. Jpn J Exp Med New York: Plenum Press, 1989;323-33 1983;53:41-50 76. Hagerman AE. Tannin-protein interaction. In: Ho 60. Xu X, Wang HJ, Murphy PA, et at. Daidzein is a CT, Lee CY, Huang MT, eds. Phenolic compounds more available isoflavone than is genistein in adult in food and their effects on health, I: analysis, oc- women. J Nutr 1994;124:825-32 currence, and chemistry. ACS Symposium Series 61. Van het Hof KH, Kivits GAA, Westtrate JA, Tijburg 506. Washington: American Chemical Society, LBM. Bioavailability of catechins from tea: the ef- 1992;236-47 fect of milk. Eur J Clin Nutr 1998;52:35&9 77. Bressani R, Hernandez E, Braham JE. Relation- 62. Lee M-J, Wang ZV, Li H, et al. Analysis of plasma ship between content and intake of bean and urinary tea polyphenols in human subjects. polyphenolics and protein digestibility in humans. Cancer Epidemiol Prev 1995;4:393-9 Plant Foods Hum Nutr 1988;38:5-21

330 Nutrition Reviews, Vol. 56, No. 11 78. Butler LG. Effects of condensed tannins on animal 2.4.21.4) and chymotrypsin (EC 2.4.21.1) in digesta nutrition. In: Hemingway RW, Karchesy JJ, eds. collected from the small intestine of pigs. Br J Nutr Chemistry and significance of condensed tannins. 1994;71:627-41 New York: Plenum Press, 1989;391-402 94. Longstaff MA, McNab JM. The effect of concentra- 79. Longstaff MA, McNab JM. The inhibitory effects of tion of tannin-rich bean hulls (Vicia faba L.) on ac- hull polysaccharides and tannins of field beans tivities of lipase (EC 3.1.1.3) and a-amylase (EC (Vicia faba L.) on the digestion of amino acids, 3.2.1 .I) in digesta and pancreas and on the diges- starch, and lipid, and on digestiveactivities in young tion of lipid and starch by young chicks. Br J Nutr chicks. Br J Nutr 1991;65:199-216 1991;66:139-47 80. Alzueta C, Treviiio J, Ortiz L. Effect of tannin from 95. Carmona A, Seidl DS, Jaffe WG. Comparison of fava beans on protein utilization in rats. J Sci Food extraction methods and assay procedures for the Agric 1992;59:551-3 determination of the apparent tannin content of 81. Tanner GJ, Moore AE, Larkin, PJ. Proantho- common beans. J Sci Food Agric 1991;56:291- cyanidins inhibit hydrolysis of leaf proteins by ru- 301 men microflora in vitro. Br J Nutr 1994;71:947-58 96. Longstaff MA, Feverstein D, McNab JM, 82. Aw T-L, Swanson BG. Influence of tannin on McCorquodale C. The influence of proantho- Phaseolus vulgaris protein digestibility and qual- cyanidin-rich bean hulls and level of dietary pro- ity. J Food Sci 1985;50:67-71 tein on energy metabolizability and nutrient digest- 83. Barroga CF, Laurena AC, Mendoza EMT. Effects of ibility by adult cockerels. Br J Nutr 1993;70:355- condensed tannins on the in vitro protein digest- 67 ibility of mung bean (Vigna radiata (L.) Wilzeck). J 97. Thompson LU, Yoon JH, Jenkins DJA, et al. Rela- Agric Food Chem 1985;33:1157-9 tionship between polyphenol intake and blood glu- 84. Andary C, Mondolot-Cosson L. Histolocalizationof cose response of normal and diabetic individuals. plant polyphenols in tissue and cell walls: some Am J Clin Nutr 1984;39:745-51 applications. In: Amado R, Andersson H, Baydocz 98. Desphande SS, Shalunke DK. Interactions of tan- S, Serra F, eds. Polyphenols in food. Luxembourg: nic acid and catechin with starches. J Food Office for Official Publications of the European Com- Sci 1982;47:2080-1,3 munities, 1998;41-7 99. Ya C, Gaffney SH, Lilley TH, Haslam E. Carbohy- 85. Quesada C, Bartolome B, Nieto 0, et al. Phenolic drate-polyphenol complexation. In: Hemingway inhibitors of a-amylase and trypsin enzymes by RW, Karchesy JJ, eds. Chemistry and significance extracts from pears, lentils, and cocoa. J Food Prot of condpsed tannins. New York: Plenum Press, 1996;59:185-92 1989;307-22 86. Shahkhalili V, Finot PA, Hurrell R, Fern E. Effects of 100. Wursch P. Influence of tannin-rich carob pod fiber foods rich in polyphenols on nitrogen excretion in on the cholesterol metabolism in the rat. J Nutr rats. J Nutr 1990;120:346-52 1979;109:685-92 87. Mole S, Rogler JC, Butler LG. Use of 15N-labeled 101, Regerat F, Remesy C, Tixier 0, et al. Effects of con- protein to determine the effect of dietary tannin on densed tannins and pectin on cecal fermentations the relative abundance of endogenous and dietary and lipid metabolism in the rat. Bull Liaison Groupe protein in feces. Proceedings of the Groupe Polyphenols 1992;16:201-4 Polyphenols XVth International Conference, 1990 102. Tebib K, Bitri L, Besanqon e Rouanet J-M. Poly- 88. Mole S, Rogler JC, Morel1 C, Butler LG. meric grape seed tannins prevent plasma choles- growth reduction by tannins: use of Waldbauer ra- terol changes in high-cholesterol-fed rats. Food tio techniques and manipulation of salivary protein Chem 1994;49:403-6 production to elucidate mechanisms of action. 103. Tebib K, BesanGon Rouanet J-M. Dietary grape Biochem System Ecol 1990;18:183-97 seed tannins affect lipoproteins, lipoprotein lipases, 89. Charlton AJ, Baxter NJ, Haslam E, Williamson MF! and tissue lipids in rats fed hypercholesterolemic Salivary proteins as a defense against dietary diets. J Nutr 1994;124:2451-7 tannins. In: Amado R, Andersson H, Bardocz S, 104. Yugarani T, Tan BKH, Das NP. The effects of tannic Serra F, eds. Polyphenols in food. Luxembourg: acid on serum and liver lipids of RAlF and RlCO Office for Official Publications of the European Com- rats fed on high fat diet. Comp Biochem Physiol munities, 1998;179-85 1993;104:339-43 90. Mehanso H, Asquith TN, Butler LG, et al. Tannin- 105. Muramatsu K, Fukuyu M, HaraY Effect of green tea mediated induction of proline-richproteins in mam- catechins on plasma cholesterol level in choles- mals. J Agric Food Chem 1992;40:93-7 terol-fed rats. J Nutr Sci Vitaminol 1986;32:613-22 91. Oh HI, Hoff JE. Effect of condensed grape tannins 106. Brune M, Rossander L, Hallberg L. Iron absorption on the in vitro activity of digestive proteases and and phenolic compounds: importance of different activation of their zymogens. J Food Sci phenolic structures. Eur J Clin Nutr 1989;43:547- 1986;51:577-80 58 92. Ahmed AE, Smithard R, Ellis M. Activity of enzymes 107. Gillooly M, Bothwell TH, Torrance JD, et al. The of the pancreas, and the lumen and mucosa of the effects of organic acids, phytates, and polyphenols small intestine in growing broiler cockerels fed on on the absorption of iron from vegetables. Br J Nutr tannin-containing diets. Br J Nutr 1991;65:189-97 1983;49:331-42 93. Jansman AJM, Enting H, Verstegen MWA, Huisman 108. Reddy MB, Cook JD. Assessment of dietary deter- J. Effect of condensedtannins in hulls of faba beans minants of nonheme iron absorption in humans (Vicia faba L.) on the activities of trypsin (EC and rats. Am J Clin Nutr 1991;54:723-8

Nutrition Reviews, Vol. 56, No. 11 33 1 109. Cook JD, Reddy MB, Hurrell RF. The effect of red Elderly Study. Nutr Cancer 1994;22:175-84 and white wines on nonheme iron absorption in 125. Renaud S, De Lorgeril M. Wine, alcohol, platelets, humans. Am J Clin Nutr 1995;61:800-4 and the French paradox for coronary heart disease. 110. South PK, House WA, Miller DD. Tea consumption Lancet 1992;339:1523-6 does not affect iron absorption in rats unless tea 126. Fuhrman B, Lavy A, Aviram M. Consumption of red and iron are consumed together. Nutr Res wine with meals reduces the susceptibility of hu- 1997;17:1303-10 man plasma and LDL to lipid peroxidation. Am J 11 1. Hurrell RF, Reddy M, Cook JD. Influence of polyphe- Clin Nutr 1995;61:549-54 nol-containing beverages on iron absorption. In: 127. Shahidi F, Wanasundara PKJPD. Phenolic antioxi- Amado R, Andersson H, Bardocz S, Serra F, eds. dants. Crit Rev Food Sci Nutr 1992;32:67-103 Polyphenols in food. Luxembourg: Office for Offi- 128. Bors W, Heller W, Michel C, Saran M. Flavonoids cial Publications of the European Communities, as antioxidants: determination of radical-scaveng- 1998; 169-72 ing efficiencies. Methods Enzymol 1990;186:343- 112. Garcia-Lopez S, Erdman JW, Sherman AR. Iron re- 55 tention by rats from casein-legume test meals: ef- 129. Manach C, Regerat F, Texier 0, et al. Bioavailability, fect of tannin level and previous diet. J Nutr metabolism, and physiological impact of 4-0x0-fla- 1990;120:760-6 vonoids. Nutr Res 1996;16:517-44 113. Jansman AJM, Houdijk JGM, Verstegen MWA. Ef- 130. Ratty AK, Das NF! Effects of flavonoids on nonenzy- fects of condensed tannins in faba beans (Vicia mic lipid peroxidation: structure activity relation- faba L.) on the availability of minerals in pigs. In: ship. Biochem Med Metab Biol 1988;39:69-79 Schlemmer U, ed. Bioavailability ‘93: Nutritional, 131. Afanas’ev IB, Dorozhko Al, Brodskii AV, et al. Chelat- Chemical and Food Processing Implications of ing and free radical scavenging mechanisms of Nutrient Availability. Karlsruhe: Federation of Eu- inhibitory action of rutin and quercetin in lipid ropean Chemical Societies, 1993 peroxidation. Biochem Pharmacol 1989;38: I 763- 114. Kies C, Umoren, J. Inhibitors of ‘Eopper 9 bioutilization:fiber, lead, phytate, and tannins. Adv 132. Hagerman AE, Riedl KM, Jones GA, et al. High- Exp Med Biol 1989;258:81-93 molecular-weight plant polyphenolics (tannins) as 115. Vaquero Me Veldhuizen M, Van Dokkum W, et al. biological antioxidants. J Agric Food Chem Copper bioavailability from breakfasts containing 1998;46:1887-92 tea: influence of the addition of milk. J Sci Food 133. Tore1 J, Cillard J, Cillard F! Antioxidant activity of Agric 1994;64:475-81 flavoneids and reactivity with peroxyl radical. Phy- 116. Thompson M, Williams CR. Stability of flavonoid tochemistry 1986;25:91-100 complexes of copper (11) and flavonoid antioxidant 134. Yuting C, Rongliang Z, Zhongjian J, Yong J. Fla- activity. Anal Chim Acta 1976;1976:375-81 vonoids as superoxide scavengers and antioxi- 117. Coudray C, Bousset C, Pepin D, et al. Effect of acute dants. Free Radic Biol Med 1990;9:19-21 ingestion of polyphenol compounds on zinc and 135. Morel I, Lescoat G, Cogrel P, et al. Antioxidant and copper absorption in the rat: utilisation of stable iron-chelatingactivities of the flavonoids catechin, isotopes and ICP/MS technique. In: Amado R, quercetin and diosmetin on iron-loaded rat hepa- Andersson H, Bardocz S, Serra F, eds. Polyphe- tocyte cultures. Biochem Pharmacol 1993;45:13- nols in food. Luxembourg: Office for Official Publi- 9 cations of the European Communities, 1998;173- 136. Kanner J, Frankel E, Granit R, et al. Natural antioxi- 7 dants in grapes and wines. J Agric Food Chem 118. Freeland WJ, Calcott PH, Geiss DF! Allelochemicals, 1994;42:64-9 minerals, and hervibore population size. Biochem 137. Yen G-C, Chen H-Y. Antioxidant activity of various System Ecol 19853 3:195-206 tea extracts in relation to their antimutagenicity. J 119. Fairweather-Tait SJ, Piper Z, Fatemin SJA, Moore Agric Food Chem 1995;43:27-32 GR. The effect of tea on iron and aluminum me- 138. Cao G, Sofic E, Prior RL. Antioxidant capacity of tea tabolism in the rat. Br J Nutr 1991;65:61-8 and common vegetables. J Agric Food Chem 120. Fraile AL, Flynn A. The absorption of manganese 1996;44:3426-31 from polyphenol-containingbeverages in suckling 139. Teissedre PL, Frankel EN, Waterhouse AL, et al. rats. Int J Food Sci Nutr 1992;43:163-8 Inhibition of in vitro human LDL oxidation by phe- 121. Hertog MGL, Feskens EJM, Hollman PCH, et al. nolic antioxidants from grapes and wines. J Sci Dietary antioxidant flavonoids and risk of coronary Food Agric 1996;70:55-61 heart disease: the Zutphen Elderly Study. Lancet 140. Frankel EN, Kanner J, German JB, et al. Inhibition 1993;342:1007-11 of oxidation of human low-density lipoprotein by 122. Hertog MGL, Sweetnam PM, Fehily AM, et al. Anti- phenolic substances in red wine. Lancet oxidant flavonols and ischemic heart disease in a 1993;341:454-7 Welsh population of men: the Caerphilly Study. Am 141. Brand-Williams W, Cuvelier ME, Berset C. Use of a J Clin Nutr 1997;65:1489-94 free radical method to evaluate antioxidant activity. 123. Hertog MGL, Kromhout D, Aravanis C, et al. Fla- Lebensm Wiss Techno1 1995;28:25-30 vonoid intake and long-term risk of coronary heart 142. Sanchez-Moreno C, Larrauri JA, Saura-Calixto F. disease and cancer in the Seven Countries Study. A procedure to measure the antiradical efficiency Arch Intern Med 1995;155:381-6 of polyphenols. J Sci Food Agric 1998;76:270-6 124. Hertog MGL, Feskens EJM, Hollman PCH, et al. 143. Serafini M, Ghiselli A, Ferro-Luzzi A. Red wine, tea Dietary flavonoids and cancer risk in the Zutphen and antioxidants. Lancet 1994;344:626

332 Nutrition Reviews, Vol. 56, No. 11 144. Serafini M, Ghiselli A, Ferro-Luzzi A. In vivo antioxi- dietary anticancer agent ellagic acid is a potent dant effect of green and black tea in man. Eur J inhibitor of DNA topoisomerases in vitro. Nutr Can- Clin Nutr 1996;50:28-32 cer 1995;23:121-30 145. Cook NC, Samman S. Flavonoids: chemistry, me- 148. Noroozi M, Angerson WJ, Lean MEJ. Effects of fla- tabolism, cardioprotective effects, and dietary vonoids and on oxidative DNA damage sources. J Nutr Biochem 1996;7:66-76 to human lymphocytes. Am J Clin Nutr 146. Gensler HL, Gerrish KE, Williams T, et al. Preven- 1998;67:1210-8 tion of photocarcinogenesis and UV-induced im- 149. Brown JF! A review of the genetic effects of natu- munosuppression in mice by topical tannic acid. rally occurring flavonoids, , and Nutr Cancer 1994;22:121-30 related compounds. Mutat Res 1980;75:243-77 147. Constantinou A, Stoner GD, Mehta R, et al. The

Nutrition Reviews, Vol. 56, No. 11 333