Antioxidants in Food: Content, Measurement, Significance, Action, Cautions, Caveats, and Research...

8/10/2019 Antioxidants in Food: Content, Measurement, Significance, Action, Cautions, Caveats, and Research Needs

1/53

CHAPTER ONE

Antioxidants in Food: Content,Measurement, Significance,Action, Cautions, Caveats, andResearch NeedsIris F.F. Benzie 1 , Siu-Wai ChoiDepartment of Health Technology and Informatics, The Hong Kong Polytechnic University, Kowloon,Hong Kong1 Corresponding author: e-mail address: [email protected]

Contents

1. Introduction: Antioxidants in Food Their Role and Importance 21.1 Basic concepts of oxidation and the role of antioxidants 21.2 Antioxidants: Types and action 31.3 A closer look at ascorbic acid: A key antioxidant 41.4 Human antioxidant defense and dietary influences 6

2. Measuring Total Antioxidant Content of Food 92.1 Basic principles, notes on calibration, units, and confusion 92.2 Limitations of the total antioxidant content approach: Cautions

and caveats 113. The FRAP Assay and Its Modified Form (FRASC) for Ascorbic Acid 13

3.1 Basic principles 133.2 The FRAP assay Procedure in brief 143.3 Application of the FRAP assay in food science 163.4 A note on the nonuric acid FRAP value of blood plasma and its relevance

to nutritional science 163.5 The modified FRAP assay (FRASC) for total antioxidant and ascorbic acid

measurement 324. Why Is the Antioxidant Content of Food of Interest? 33

4.1 Basic concepts 334.2 Antioxidants and health: The evidence and potential impact 34

5. Antioxidants in Food: Enigmas and Evolutionary Aspects 366. Mechanisms of Action: Redox Issues, Phytohormesis, and Colonic Microbiota 38

6.1 Redox tone changes as a driving force for cytoprotection and the biologicalsense of restricted bioavailability of dietary antioxidants: The redox-sensitiveKEAP-1 Nrf2 signaling pathway 39

6.2 Colonic microflora and metabolism of food antioxidants: Moving into newterritory of the microbiome 42

Advances in Food and Nutrition Research, Volume 71 # 2014 Elsevier Inc.ISSN 1043-4526 All rights reserved.http://dx.doi.org/10.1016/B978-0-12-800270-4.00001-8

1
http://dx.doi.org/10.1016/B978-0-12-800270-4.00001-8http://dx.doi.org/10.1016/B978-0-12-800270-4.00001-8


2/53

7. The Story So Far 438. Concluding Remarks and Research Needs 46Acknowledgments 46References 47

Abstract

There are a multitude of antioxidants in foods, especially in foods of plant origin. Higherintake of antioxidant-rich foods is clearly associated with better health and functionallongevity. The specific agents and mechanisms responsible are not yet clear, but thereis convincing evidence that including more plant-based, antioxidant-rich foods, herbs,and beverages in the diet is effective in promoting health and lowering risk of variousage-related diseases. The content of some individual antioxidants, such as vitamin C, in

food can be measured, but it is not feasible to attempt to measure each antioxidantseparately, and methods have been developed to assess the total antioxidant content

of foods. One of the most widely used methods is the ferric reducing/antioxidant power(FRAP) assay, which is relatively simple, quick, sensitive, and inexpensive to perform. There are many published studies that have used the FRAP assay, and these have gen-erated a very large database of total antioxidant content of foods that can help guidefood choices for increased antioxidant intake. The FRAP assay has also been used toassess the bioavailability of antioxidants in foods and to investigate the effects of grow-ing conditions, storage, processing, and cooking method on the total antioxidant con-tent of food. The test can be employed as a quality control check device, and to detectadulteration of food. Furthermore, in a modified form (FRASC), the assay can measureascorbic acid content almost simultaneously with the total antioxidant content of thesample. In this chapter, basic concepts of oxidation and the role of antioxidants, as wellas the types and action of different antioxidants in foods will be reviewed briefly, and theunderpinning concepts and evidence for health benefits of increased intake of dietaryantioxidants will be discussed, with some focus on vitamin C, and also in the context of our evolutionary development. The basic concepts and limitations of measuring totalantioxidant content of food will be presented. The FRAP assay and the modified versionFRASC will be described, and the total antioxidant content (as the FRAP value) of a range

of foods will be presented. Finally, issues of bioavailability and redox balance will bediscussed in relation to the biological significance and molecular action of antioxidantsin foods, some caution and caveats are presented about overcoming biological barriersto absorption of antioxidant phytochemicals, and research needs to further our under-standing in the important area of food, antioxidants, and health will be highlighted.

1. INTRODUCTION: ANTIOXIDANTS IN FOOD THEIRROLE AND IMPORTANCE

1.1. Basic concepts of oxidation and the roleof antioxidants

In chemical terms, oxidation is the addition of oxygen to or the removal of hydrogen or an electron from a molecule. A biological antioxidant can be

2 Iris F.F. Benzie and Siu-Wai Choi


3/53


4/53

intracellular levels of oxygen and ROS and have evolved specialized defensefactors to deal with the high oxidant challenge and protect plant structuresand tissues (Benzie, 2000 ). These factors are referred to as antioxidants.The best-known antioxidant is vitamin C (ascorbic acid), but there arevarious others, and many are unique to the plant kingdom. These includevitamin E, a group of eight tocopherols and tocotrienols, and the verylarge families of flavonoids and carotenoids ( Benzie, 2005; Buettner,1993; Gey, 1998; Tam et al., 2005 ). Indeed, there are thousands of differentantioxidants in plants, and they work in different ways ( Table 1.2 ). Someantioxidants prevent the generation of ROS, some are enzymes that destroyROS, some are small water-soluble molecules that act as reducing (hydrogenor electron-donating) agents to neutralize free radicals, and some absorbelectrons or excess energy from ROS and dissipate this within their complex lipophilic structure ( Ames, 1998; Halliwell & Gutteridge, 2007 ).Overall, the effect is to oppose ROS action and limit oxidative stress.

1.3. A closer look at ascorbic acid: A key antioxidantVitamin C (ascorbic acid) is present in high concentration in many plants

(Benzie & Wachtel-Galor, 2009; Frei et al., 1989; Gey, 1998;Halliwell & Gutteridge, 2007; Halvorsen et al., 2002 ). The content of thiswater-soluble antioxidant is particularly high in fruits, but it is found in var-ious plant parts and aids repair and growth of plants and the ripening of seeds.

Figure 1.1 The concept of antioxidant/oxidant imbalance and the development of oxidative stress.



5/53

Most animals, fish, and reptiles can synthesize vitamin C in liver or kidney,but humans have lost the ability to make ascorbic acid, even though we have

retained an absolute requirement for it ( Benzie, 2000, 2003; Benzie &Strain, 2005; Frei et al., 1989 ). Ascorbic acid is known to be needed for col-lagen synthesis in humans, and severe deficiency results in scurvy, which isnowadays rare but cases are nonetheless still reported ( Doll & Ricou, 2013 ).Humans must obtain a regular and adequate dietary supply to maintainplasma and tissue levels, and the best dietary sources are fruits and vegetables(Frei et al., 1989). In the past, the daily recommended intake of vitaminC was set at a level ( 30 mg/day) that would prevent overt deficiency

(Newton et al., 1983 ). However, accumulated epidemiological and exper-imental evidence indicates that the role of ascorbic acid in the human body isnot restricted to collagen synthesis and that it has an important role in healthmaintenance overall ( Benzie, 1999 ). Ascorbic acid is recognized as a major

Table 1.2 Plant-derived antioxidant types, action, and sourcesAntioxidant Action Mechanism Sources

Ascorbic acid

(vitamin C)

Scavenging

of ROS

Sacrificial interaction

by replaceable or recyclable substrates toscavenge and destroyROS

Fruits and vegetables,

particularlystrawberries, citrus,kiwi, Brussels sprouts,cauliflower, someChinese greenvegetables

Vitamin E ( a , b ,d, g) isomers of tocopherols and

tocotrienols

Quenchingof ROS andchain

breaking

Absorption of electronsand/or energy

Green leafy vegetables(e.g., spinach); nuts,seeds, especially wheat

germ; vegetable oils,especially sunflower

Carotenoids(hundreds)

Chainbreaking

Chain breaking at lowpartial pressures of oxygen, complementsaction of vitamin E

Orange/red fruits andvegetables (e.g., carrot,tomato, apricot, melon,pumpkin), green leafyvegetables, peppers

Flavonoids (large

range of differenttypes)

Scavenging

of ROS

Sacrificial interaction Berries, apples, onions,

tea, red wine, someherbs (parsley, thyme)citrus fruits, grapes,cherries

Source: Halliwell and Gutteridge (2007) .

5Antioxidants in Food


6/53

player in human antioxidant defense ( Benzie, 1999, 2000 ). Current dietaryrecommendations are for at least 70 mg/day for women, 100 mg/day for men, with higher levels advised in pregnancy and for smokers ( Carr &

Frei, 1999 ). These levels of intake are recommended to attain optimal sta-tus for health, though this has not been clearly defined. Human cells, espe-cially white blood cells, and some organs, most notably the eye, containmillimolar concentrations of ascorbic acid ( Choy et al., 2001, 2003 ). Plasmaascorbic acid concentrations in the fasting state range quite widely, from< 10 (worryingly low) to 100 mmol/l in apparently healthy adults ( Choiet al., 2004). Tissue and plasma levels are maintained by, and so reflect, die-tary intake of vitamin C, and the plasma concentration of ascorbic acid can

act as a marker of fruit and vegetable intake (provided no supplements areused) (Choi et al., 2004 ). A concentration < 10 mmol/l indicates very lowintake and severe deficiency, even if signs of scurvy are not clearly present(Levine et al., 2011 ). Plasma levels at 12 h after ingestion of a large dose (1 gor more) of vitamin C can approach 200 mmol/l, but peak levels are limitedby restricted gastrointestinal absorption of a large dose and by urinary loss of ascorbic acid when the plasma concentration exceeds the renal threshold of

100 mmol/l ( Levine et al., 2011 ). Therefore, most of a single, large dose is

not retained in the body, and regular, modest doses (500 mg or less) are likelyto be more effective in enhancing the ascorbic acid status of the body. Inlarge prospective trials, it has been shown that healthy people with higher fasting plasma concentrations of food-derived ascorbic acid (i.e., not fromsupplements) have significantly lower risk of incident diabetes, cancer,stroke, heart failure, and all cause mortality in the ensuing years ( Fig. 1.2;Harding et al., 2008; Pfister et al., 2011 ). It has been estimated that for every20 mmol/l increase in fasting plasma ascorbic acid concentration, there is a

9% relative reduction in risk of incident heart failure and a 29% decrease inrisk of incident type 2 diabetes ( Harding et al., 2008; Pfister et al., 2011 ).Plant-based food is the primary source of ascorbic acid in human plasma.However, it must be noted that the protective agent may not be vitaminC itself. It could be something very closely associated with vitamin C inplant-based foods, or it could be that the health benefits are due to the inter-active effects of the various antioxidants present in plant-based foods.

1.4. Human antioxidant defense and dietary influencesThe human body is exposed to ROS on a continuous basis, and we have aneffective antioxidant defense system that has evolved to help deal with the



7/53

threat of oxidant challenge ( Benzie, 2000, 2003 ). The system is complexand, as with plants, there are different types of antioxidants that prevent gen-eration of, divert, or destroy ROS ( Benzie & Wachtel-Galor, 2009 ). Our intrinsic antioxidant defense is highly effective but not wholly adequate,

and dietary input of antioxidants is needed ( Benzie, 2003; Benzie &Wachtel-Galor, 2009; Norat et al., 2014 ). Plant-based foods and beverages,such as fruits, vegetables, tea, coffee, spices, and herbs, are the main source of antioxidants in the human diet ( Benzie & Wachtel-Galor, 2009; Halvorsenet al., 2006). It is noted that the only dietary-derived antioxidants that havebeen shown to be essential for human health are the water-soluble vitamin Cand the lipophilic vitamin E (which in human tissues and structures is mainlyin the form of a-tocopherol). Still, conceptually at least, the other antiox-

idants that evolved to deal with oxidant challenge in plants could have a roleto play in defending human tissues from this same challenge, augmenting our endogenous antioxidant system in the prevention of oxidative stress. Towhat extent they do this and their contribution to health outcomes are

0 1 2 3 40.0

0.2

0.4

0.6

0.8

1.0

1.2

Plasma vitamin C concentrationSelf-reported dietary fruit andvegetable intake

Quartiles of plasma vitamin C or dietary fruit and vegetable intakeError bars indicate 95% Cl

H a z a r d r a

t i o

f o r

i n c

i d e n

t h e a r t

f a i l u r e

Figure 1.2 Risk for incident heart failure decreases with increased plasma vitamin C insubjects observed over a period of 16 years. Data taken from Pfister et al. (2011).



8/53

not yet clear. Plant-derived antioxidants including the catechins and someanthocyanins, carotenoids, xanthophylls, and other members of the vitaminE family are found in human plasma ( Table 1.3 ), and the antioxidant capacity

of human plasma increases after intake of polyphenol-rich dietary agentssuch as tea and coffee (Benzie et al., 1999; Garc a-Alonso et al., 2006;Hukkanen et al., 2006; Molan et al., 2008; Othman et al., 2007; Seeramet al., 2008). However, the very low (nanomolar) concentrations of thenonessential plant-derived antioxidants have made it difficult to study

Table 1.3 Plant-derived antioxidants and their typical concentrations in fasting plasmaConcentration ( m M) Food source

Ascorbic acid 1060 Fruits, particularly kiwi fruits,strawberries

VitaminE (a -tocopherol)

1640 Wheatgerm, sunflower seeds, nuts

Epicatechin 0.005 Tea, chocolate, wine

Epicatechin gallate 0.041

Epigallocatechingallate

0.016

Epigallocatechin Undetectable

Catechin Undetectable

Quercetin 0.001 Apple, onion, red grape, leafy greenvegetables

Anthocynanin 0.046 Aubergine, blackberry, blackcurrant

Lutein 0.160 Green leafy vegetables, spinach, kaleZeaxanthin 0.050 Wolfberry, egg yolk, corn, paprika

Lycopene 0.377 Tomatoes, guava, grapefruit,watermelon

-carotene 0.354 Sweet potato, carrots, kale, butternutsquash

Phytoene 0.069 Most fruits and vegetables

Phytofluene 0.044 Most fruits and vegetablesIsoflavones 0.428 Soya, chickpea, peanut, alfalfa

Source: Benzie et al. (1993), Halliwell and Gutteridge (2007) , Harding et al. (2008) , and Engelmannet al. (2011) .



9/53

individual nonessential antioxidants in relation to their bioavailability,bioactivity, and metabolism in human experimental trials, and most of theevidence for their health benefits has come from epidemiological, food-

based studies. This is changing with the adoption of hyphenated technologyplatforms, particularly liquid chromatography with tandem mass spectrom-etry (LCMS/MS), which provide highly specific and sensitive measure-ment methods for many individual plant-derived antioxidants in foods,plasma, and other biological materials. Still, the antioxidant fingerprint or profiles of foods are far from complete as it is not feasible to measure eachand every antioxidant in a complex mixture such as food. In the face of thislimitation, a widely adopted approach is to determine the total antioxidant

content of food and of blood plasma following intake of antioxidant-richfood (Bartoz, 2003 ). Several methods have been developed for measuringtotal antioxidant content. In Section 2, basic principles of total antioxidantcontent measurement and limitations of this approach are discussed, anddetails of one of the most widely used methods, the ferric reducing/antiox-idant power (FRAP) assay, along with the total antioxidant content (as theFRAP value) of a range of foods are presented.

2. MEASURING TOTAL ANTIOXIDANT CONTENT OF FOOD

2.1. Basic principles, notes on calibration, units,and confusion

The different methods that have been developed to measure total antioxi-dant content of food use one of two basic principles and can be described as

direct or indirect methods ( Bartoz, 2003 ). The basic principle of adirect method is to measure the reductive action of all redox-active (electrondonating) antioxidants present as they reduce a component, added to thereaction mixture in excess, with the production of a change in an indicator signal. The signal can be a change in absorption of light (spectrophotomet-ric methods) or in an electrochemical signal, such as current flow. The mag-nitude of the change in signal is directly related to the combined reducingaction of the electron-donating antioxidants in the mixture, and the signal

change can be quantified with reference to the signal change induced by acalibrator, such as pure ascorbic acid, run in parallel. This electron transfer principle is applied in, for example, the FRAP assay and in the assay referredto as Cuprac in which cupric ions are reduced to cuprous ions ( Benzie &



10/53

Strain, 1996a ). The basic principle of indirect (lag-phase) methods is to add afree radical generator to the reaction mixture and to time how long it takesbefore the antioxidants present in the reaction mixture are exhausted, and an

indicator molecule begins to be reduced by the radicals that by that stage actunopposed on the indicator molecule. The reduction of the indicator mol-ecule produces a change in signal, such as an absorbance change. The indi-cator molecule can be a dye that changes color when oxidized, but anunsaturated fatty acid can be used, with the change in spectral signal relatedto the formation of conjugated dienes within the oxidized lipid ( Bartoz,2003). The lag-phase approach is most commonly used in the forms of the oxygen radical absorption capacity test and the ABTS radical test

(Bartoz, 2003 ).To translate changes in signal to total antioxidant content in direct

methods, the lag time is translated into a measure of antioxidant action withreference to the lag time of a known amount of a purified antioxidant, mostcommonly Trolox as calibrator (standard). This is a water-soluble ana-logue of a -tocopherol, and the most commonly used unit for reporting totalantioxidant activity is mmol/l (or mmol/l or mmol/g depending on the sam-ple type) of Trolox Equivalents (TE though different terminologies are

used to express results). Solutions of Trolox can be used in direct reduc-tive methods also, but solutions of pure ascorbic acid are often used becauseascorbic acid is inexpensive and highly soluble in water, and so someresearchers use units of ascorbic acid equivalents (AAE). These units(TE and AAE) should be interchangeable because one molecule of Trolox

and ascorbic acid can each donate two electrons, and therefore, in theory,they have antioxidant equivalence. However, while Trolox in solutionis fairly stable, we have found it is far less easy to dissolve than the highly

soluble ascorbic acid. For this reason, ascorbic acid solutions of greater accu-racy can be prepared. Yet, ascorbic acid, while fairly stable in its solid form, isunstable in aqueous solution. Therefore, total antioxidant activity resultsobtained using either Trolox -based or ascorbic acid standards can be over-estimated because of, in one case, poor solubility, leading to lower actualconcentrations of standards than the nominal concentrations and, in theother, degradation of the ascorbic acid in aqueous solution if these arenot prepared immediately before use. Such problems can be detected by

close monitoring of the signal given by the standards, which should be con-sistent across different test runs and batches of standards. A further compli-cation is that in direct reductive methods, the stoichiometric factor (which asnoted is 2.0 in the case of both ascorbic acid and Trolox ) is often applied to



11/53

give a result that reflects the number of electrons transferred during the reac-tion. In this case the total antioxidant activity result is twice that of theresult expressed in either ascorbic acid equivalents or Trolox equiva-

lents. A further complication is that stoichiometric factors of antioxidants,including Trolox and ascorbic acid, vary with concentration in somemethods. Moreover, while results on purified antioxidants such as ascorbicacid can match fairly well across the different methods, results on complexmixtures, such as food extracts or blood plasma, from the different assays donot agree well because different antioxidants react differently in the differenttest systems and reaction conditions. This makes it very difficult to makemeaningful comparison of total antioxidant results from studies that have

used different methods of measurement. It is easier to compare data obtainedusing the same method, but even then results from different studies are oftendifficult to compare directly because of differences in sample handling, stan-dardization procedure, reaction conditions (time, temperature, pH), solventused, and the units used to express results. Various units, such as TEACunits, TAC units, TAA units, TRAP units, TE, AAE, radicalscavenging, or radical absorbance units appear in the literature, andsome results are presented simply as absorbance or fluorescence units,

the lag time or the time taken to change the signal by a particular amount(Bartoz, 2003 ). These variations in measuring principle and methodsettings, calibration, and activity units reported cause confusion and makeit difficult to compare or combine results of different studies unlesssufficient information is given on how the results were obtained in termsof: sample type and preparation; method and calibrator used; reaction timeused; the signal measured, and how that was translated into the resultpresented ( Benzie & Wachtel-Galor, 2013 ).

2.2. Limitations of the total antioxidant content approach:Cautions and caveats

The basic concepts and principles of measuring total antioxidant content aregenerally quite simple, but the underlying chemistry of the various methodsis complex, and as noted there are differences in measuring principles andsetting, calibrators, and reporting units to consider, confound, and confuse.

Each of the methods currently available has its own limitations, for example,thiol groups do not react well or quickly in the FRAP assay so that GSH andprotein make very little contribution to the FRAP value, while results of proteinaceous samples (meat, plasma) are overwhelmingly due to protein



12/53

thiol groups, especially when the reaction time is prolonged to an hour or more (Bartoz, 2003; Benzie & Strain, 1996a ). Currently, there is no univer-sally accepted reference method and even with the same method, reaction

conditions can vary greatly between laboratories. Furthermore, it must beunderstood that the whole concept of measuring total antioxidant activityby a chemical test has serious limitations (Bartoz, 2003 ). A clear understand-ing of what each method measures is needed for the result to have somemeaning for nutrition, food, and health science. None of the methods mea-sures all antioxidants present in a mixture. The tests are limited to measuringthe effect of redox-active antioxidants, and so do not measure the catalyticaction of superoxide dismutase (SOD) and glutathione peroxidase (GPx), or

the preventive action of ferritin. As noted, some are not sensitive to thiols,while others furnish results that are determined largely by the action of thiolgroups in proteins present in the sample ( Bartoz, 2003 ). The tests are allin vitro tests and most if not all the reaction conditions are very far fromthe physiological. Results on an extract of food may have little in the wayof physiological significance, as antioxidants may work in quite a differentmanner when present in a mixture or in a living system, asopposed to in puri-fied form or in vitro. There are also issues of bioavailability and biotransfor-

mation to consider. Dietary antioxidants are often not absorbed well, andwhat is estimated to be contained in a food or the diet overall is unlikelyto reflect what enters the circulation ( Davey et al., 2000 ). Those that areabsorbed may undergo rapid conjugation in the liver and renal clearance.Antioxidants that are not absorbed may be extensively metabolized by thecolonic microbiota. For example, ring scission metabolites of catechins(flavan-3-ols) in green tea are formed by colonic metabolism and are foundin the circulating plasma several hours after ingestion of green tea ( Del Rio

et al., 2010). Therefore, data on the in vitro total antioxidant content of foods are of limited value in relation to physiological relevance. Still,despite its limitations, the measurement of total antioxidant content of foods has become a standard approach in nutritional and food science inregard to the potential health benefits or value of different foods and foodextracts or supplements ( Ashfari et al., 2007; Bartoz, 2003; Benzie & Szeto,1999; Benzie et al., 1999; Chung et al., 2001; De et al., 2008; Dragstedet al., 2004; Duthie et al., 2006; Fernandez-Pachon et al., 2005; Garc a-

Alonso et al., 2006; Lotito & Frei, 2006; Ma et al., 2008; Rabovskyet al., 2006; Serafini et al., 2003; Smet et al., 2006; Wachtel-Galor ,Szeto, Tomlinson, & Benzie, 2004 ). Indeed, a quick search of the literaturewith key words of antioxidant and food reveals 19,677 papers



13/53

published between 2008 and 2013 (Pubmed accessed August 2013). Manypackaged foods now include a measure of total antioxidant content ontheir labels and marketing materials. Therefore, it is important for food sci-

entists and nutritionists to understand the concepts and limitations outlinedabove, and to generate and use data with these in mind.

In summary, different methods exist to measure the total antioxidantcontent of foods, but there is no universally accepted reference method; thereare important caveats to the basic concept of measuring total antioxidantcontent of food. Caution is needed in using the results. To compare resultsfrom different analyses, it is important to understand what each method mea-sures, how it is calibrated (standardized), and what the reported units of con-

tent or activity represent. It is important also to understand the limitations of the method, and where possible to adopt a standard operating procedure for the method. In this way, results obtained on different materials and from dif-ferent research groups can be compared, integrated, and used to aid our understanding of the role and value of total antioxidant activity in nutri-tional, food, and health sciences. In Section 3, the focus is on one widely usedmethod, the FRAP assay. This method has low cost and high sensitivity, isquick and relatively simple to perform, and has been used to generate a rich

database of total antioxidant content of a wide variety of foods. As will bedescribed, the method has been used also to assess bioavailability of reductiveantioxidants in foods and the effect of processing and cooking on the antiox-idant content of foods, and it can be used to monitor batch-to-batch variationand detect adulteration of foods. In addition, the method in a modified formcan be used to measure ascorbic acid in the sample almost simultaneously withits total antioxidant content ( Benzie & Strain, 1997 ).

3. THE FRAP ASSAY AND ITS MODIFIED FORM (FRASC)FOR ASCORBIC ACID

3.1. Basic principlesThe FRAP assay uses the reduction of the pale yellow-colored triazine-complexed ferric (Fe 3 ) to its ferrous (Fe2 ) form as the signal, or indicator, reaction in this direct method to measure total reductive anti-oxidant action. The 2,4,6-tripyridyl- s-triazine (TPTZ)Fe 2 complex is

deep blue in color, and the reduction-driven change in absorbance at593 nm is the signal measured. A wide range of sample types can be testedin the FRAP assay, and the assay can be used successfully in a simple manualversion that requires little in the way of specialized equipment, in a



14/53

semiautomated version using a microplate reader, or adapted to run in fullyautomated mode using a user-defined program in a biochemical analyzer.Reagents are stable and of low toxicity, sensitivity and precision of the

method are high, stoichiometric factors of reacting antioxidants are constant,and the test is robust in that small differences in reaction conditions do notmarkedly affect results. In a modified version known as the FRASC assay,both the total antioxidant activity and the ascorbic acid (vitamin C) concen-tration of the test sample can be measured virtually simultaneously byemploying ascorbic oxidase to destroy ascorbic acid in one of a pair of sam-ples run in parallel (Benzie & Strain, 1997, 1999; Szeto & Benzie, 2002 ). Inaddition, the contribution of uric acid, a possible confounder in the mea-

surement of total antioxidant activity of plasma, can be removed by simplecalculation (if the uric acid concentration is known), providing the nonuricacid FRAP value ( Benzie & Strain, 1996b; Benzie & Szeto, 1999 ). This isuseful when using the FRAP assay with plasma samples to investigateabsorption of antioxidants ( Bartoz, 2003 ).

3.2. The FRAP assay Procedure in brief

For detailed technical procedures, please refer to Benzie and Strain (1996a) .The basic protocol given here can be employed in a simple manual methodand can be translated by the user into a semiautomated microplate method or a program designed for a fully automated method on a biochemical analyzer.In brief, the method simply involves the addition of an aliquot of the sampleof interest to a measured volume of working FRAP assay reagent. Thisreagent contains ferric chloride and TPTZ in acetate buffer. The workingreagent is prepared just before use by mixing 300 mmol/l acetate buffer

(pH 3.6), 10 mmol/l TPTZ in 40 mmol/l hydrochloric acid, and20 mmol/l ferric chloride in a ratio of 10:1:1. The recommended relativeamounts of sample and reagent are 1:30. The sample must be in liquid form.The solvent/diluent is usually water, but can be methanol, ethanol, or hex-ane if lipophilic antioxidants are of interest. After 4 min (if the reaction iscarried out at 37 C), the absorbance at 593 nm is measured. If the reactionmixture is kept at room temperature, a longer reaction time (8 min) is rec-ommended to align results with those obtained with a 4-min reaction time at

37 C values. A reagent blank is used to zero the spectrophotometer at593 nm (or as close to that as possible if wavelength selection is based onfilters). The absorbance of the sample/reagent mixture is used to calculate



15/53

the FRAP value with reference to the absorbance change given by a calibra-tor (standard) run in parallel with the test samples. This can be a freshly pre-pared aqueous solution of pure ascorbic acid. However, because ascorbic

acid in solution is unstable, the recommended calibrator is an aqueous solu-tion of known Fe 2 concentration (prepared from ferrous sulfate). Ferroussulfate is soluble in water up to 2 mmol/l, and the solution is stable for atleast 1 week. The recommended concentration of the calibrator is1000 mmol/l (1 mmol/l). A range of calibrators of different concentrationscan be prepared if a calibration line is desired, but the doseresponse is highlylinear and reproducible (under fixed conditions), and in our laboratory, weuse a single-point calibration at 1000 mmol/l.

It is recommended that results are expressed as the FRAP value, that is,the FRAP value, in mmol/l. This is calculated as follows:

4 min A593 nm of test sample reaction mixture4 min A593 nm of Fe2 standard reaction mixture

Fe2 standard concentration mmol=l

It is noted that the FRAP assay is run as a fixed point assay, but that thereaction is not completed at 4 min. Some antioxidants and their metabolitescontinue to react slowly beyond the 4-min reaction time, and so the absor-bance at 593 nm continues to increase, though the rate of further change isslow. If so desired, the reaction time can be extended for specific samples,but a standard 4 min (at 37 C) reaction time window is recommended tofacilitate comparison of results from different studies. In regard to samples,various biological fluids(plasma, serum,urine,or saliva) canbe testedwithout

extraction. If urine is tested, it should be run neat and diluted 1/5 and 1/10.Foods, herbs, and spices can be extracted in various solvents, such as in coldwater, hot water, methanol, ethanol, acetone, or hexane. It has been shownthat the FRAPassay is less susceptible to solvent effects than other tests of totalantioxidant capacity ( Perez-Jime nez & Saura-Calixto, 2006 ). If the extract isturbid or highly colored, it is recommended that a sample blank is run. This isdone by measuring the absorbance at 593 nm of the extract in workingreagent minus TPTZ, and subtracting this absorbance from that of the

complete reagent/sample mixture. It is noted that the FRAP value of plant-based materials can be very high, and additional predilution of theextract is often needed.



16/53

3.3. Application of the FRAP assay in food scienceThe FRAP assay has been applied widely in nutritional science. Apart frommeasuring the total antioxidant content of various foods, the FRAP assayhas been used also to explore absorption of antioxidants from foods, such assoya milk, cocoa, and tea, and to investigate the effect of processing andcooking on the antioxidant content of foods. It is well recognized that trans-port to market, storage, and cooking practices affect the content of labileantioxidants in foods, and the World Health Organization (WHO) has takenthis information into account in their recommendations for vitamin andmineral requirements in human nutrition. WHO recommendations for cooking foods containing labile antioxidants are to steam or stir fry. If water is used in the cooking of vegetables, it may be advisable to also consume thecooking water, as it contains antioxidants released from the food ( Wachtel-Galor et al., 2008 ). The FRAP assay has also been used as part of a qualitycontrol system in the agri-food industry, and to assess the effect of geneticvariation, season, growing conditions, and storage on the total antioxidantcontent of foods. For example, total antioxidant content of blueberries of the same cultivar grown in the same field can vary by up to 25% dependingon the harvesting year, and variation of up to 47% in total antioxidant con-tent is seen in different cultivars grown in the same area and harvested in thesame year (Dragovic -Uzelac et al., 2010 ). Some representative FRAP valuesof different foods are given in Table 1.4 , and some illustrative data on theeffects of cultivar, season, processing, and cooking on the antioxidant con-tent of some foods are shown in Table 1.5 .

3.4. A note on the nonuric acid FRAP value of blood plasmaand its relevance to nutritional science

The FRAP assay has been used to investigate the changes in plasma antiox-idant content after ingestion of foods and beverages, including apples, fruit juices, and teas, results being used to explore bioavailability of redox-active antioxidants in the diet ( Benzie & Szeto, 1999; Benzie et al., 1999;Duthie et al., 2006; Lotito & Frei, 2004 ). However, it is important to notethat there is an increase in plasma uric acid following ingestion of some die-tary agents. Uric acid is an endogenous compound formed from the break-down of purines and generally contributes 5060% of the total antioxidantactivity of plasma. However, following ingestion of purine-rich red meat or fructose-rich foods and beverages, a postingestion increase in uric acid (andhence antioxidant content) of plasma is seen ( Benzie & Strain, 1996b;



17/53

Table 1.4 Representative FRAP values of various foods

Food and food typeTotalantioxidant

Per average serving(m mol) References

Tea (1% w/v infusion )Green tea 10,000 2500 Benzie and Szeto (1999)

Oolong tea 5000 1250

Black tea 3500 875

Fruit juices

Orange juice 1500 375 Seeram et al. (2008)

Apple juice 1100 275

Blueberry juice 4300 1075

Grapefruit juice 8220 2050 Pellegrini et al. (2003)

Pear juice 7430 1858

Pineapple juice 5160 1209

Tropical juice 6150 1538

Grape juice 120,000 30,000 Carlsen et al. (2010)

Prune juice 10,000 2500

Tomato juice 48,000 12,000

Cranberry juice 92,000 23,000

Wine

Red wine 25,000 6250 Benzie and Strain (1999)

Rose wine 7220 1805 Pellegrini et al. (2003)

White wine 3000 750 Benzie and Strain (1999)

Beer 0.137 0.034 Halvorsen et al. (2006)

Cola 0.046 0.012

Other beverages

Cows milk 300 75

Soya milk 750 187.5

Coffee (expresso) 129,380 15,526 Pellegrini et al. (2003)Coffee(decaffeinated)

93,010 23,253

Continued



18/53

Table 1.4 Representative FRAP values of various foods cont'd



FruitsApple (green) 6.3 1146.6 Szeto and Benzie (2002)

Apple (red) 4.2 764.4

Strawberry 15.9 2337.3

Blackberry 22.0 3168

Blueberry 10.8 1555.2

Orange 9.4 1231.4Pear 4.1 680.6

Kiwifruit 8.2 565.8

Pineapple 6.0 996

Lemons 10.4 72.8 Szeto and Benzie (2002)

Banana 2.28 269.04

Cherries 8.1 1190.7

Peach 1.5 225 Halvorsen et al. (2002)

Papaya 8 1256

Honeydew melon 2.27 304.18

Grapes (black) 11.09 1663.5

Grapes (green) 3.25 487.5

Mango 5.06 834.9

Cranberries 32.9 4836.3

Plums 13.3 2008.3

Grapefruit 8.2 1049.6

Watermelon 0.4 48.8

Guava 30 4950 Patthamakanokpornet al. (2008)

Vegetables (raw )Onion 2.4 168 Halvorsen et al. (2002)

Pepper (green) 2.6 192.4



19/53




Mushrooms(white) 3.9 374.4

Tomato 2.4 436.8

Potato 1.4 298.2

Broccoli 3.9 577.2 Wachtel-Galor et al.(2008)

Cabbage 3.5 311.5

Vegetables (raw )

Lettuce 4.94 439.66 Pellegrini et al. (2003)

Peppers 20.98 1552.52

Carrots 1.06 64.66

Cauliflower 3.5 374.5 Carlsen et al. (2010)

Cucumber 0.71 36.92 Pellegrini et al. (2003)

Courgettes 3.33 273.06Kidney beans (can) 2.7 496.8 Halvorsen et al. (2006)

Peas (can) 1.2 220.8

Baked beans (can) 2.4 441.6

Snack foods

Chocolate (dark) 42 1848

Chocolate (milk) 30.8 1355.2

Walnut 230 26,910 Blomhoff et al. (2006)

Peanut 19.7 2304.9

Pecans 27.4 3205.8

Almond 4.1 479.7

Hazelnut 9.4 1099.8

Macadamia nuts 5.9 690.3

Sesame seeds 0.6 17.4

Pistachio 14.3 1673.1Continued



20/53




Ice cream(chocolate) 5 430

Ice cream (vanilla) 0.5 43

Yoghurt (plain) 0.4 68

Popcorn 4.6 138

Crisps 7.8 234

Pretzels 11 660

Oils

Butter 7.3 109.5 Carlsen et al. (2010)

Margarine 13.8 207

Olive oil 153 2065.5 Cheung et al. (2007)

Sesame oil 803 10,840.5

Peanut oil 133 1795.5

Corn oil 100 1350

Sunflower oil 108 1458

Canola oil 400 5400

Cheddar cheese 0.62 17.36 Halvorsen et al. (2006)

Parmesan cheese 1.02 28.56

Mozarella cheese 0.64 17.92

Herbs and spicesGinger 215.7 431.4 Halvorsen et al. (2002)

Garlic 8.0 16 Carlsen et al. (2010)

Cinnamon 176.5 353 Halvorsen et al. (2002)

Turmeric 156.8 313.6

Mustard seeds 105.3 210.6

Curry powder 99.8 199.6Pepper (black) 44.5 89



21/53




Breakfast itemsAll Bran 15.6 468 Halvorsen et al. (2006)

Bran flakes 42.9 1287

Corn flakes 12.3 369

Shredded wheat 2.3 69

Rice Krispies 8.6 258

Egg muffin 0.9 27Oat puff 21.1 633 Carlsen et al. (2010)

Special K 15.6 468

Weetabix 13 390

Miscellaneous

Oat flour 3.2 NA

Muffin (blueberry) 4.6 519.8 Halvorsen et al. (2006)

Cookie (chocolate) 17.2 430

Doughnut (plain) 1.5 169.5

Egg (whole) 0.4 45.2

Cheeseburger 1.1 125.4

Chicken nugget 2.0 2.0

Fish burger 1.3 182

French fries 3.3 396

Hamburger 1.4 1.4

Milkshake (vanilla) 1.2 360

French bread 1.7 39.1

Wheat bread 3.4 78.2

White bread 1.6 36.8

White rice(cooked)

0.3 55.8

Continued



22/53

Chung et al., 2001; Dragsted et al., 2004; Duthie et al., 2006; Lotito & Frei,2004). In one case, this is due to the degradation of purines in red meat. Inthe case of fructose, the increase in uric acid is due to increased degradationof hepatic AMP. In contrast, a postingestion increase in ascorbic acid canlead to increased renal excretion of uric acid and thereby lower its plasmaconcentration. Therefore, purine- or fructose-driven postingestionincreases in uric acid can be misinterpreted as signaling absorption of anti-oxidants from food, and ascorbic acid absorption can be masked by its effects

on uric acid (and thereby on total antioxidant content). In such cases, itwould be useful to be able to remove the contribution of uric acid from thetotal antioxidant content of plasma. This can be done very simply with theFRAP value. The stoichiometric factor of uric acid in the FRAP assay is 2.Provided the plasma uric acid concentration is known (and this can be mea-sured using commercially available kit methods), the plasma FRAP value canbe corrected for the contribution of uric acid by subtracting twice the uricacid concentration (in mmol/l) from the FRAP value (in the same units)

(Benzie & Strain, 1996a, 1996b, 1997; Benzie et al., 1999 ). By taking thisapproach, nutritional science studies of total antioxidant bioavailability of foods can be performed without being confounded by postingestion changesin plasma uric acid concentration ( Fig. 1.3 ). For bioavailability and biokinetics




Spaghetti (cooked) 0.2 28Oatmeal (cooked) 0.8 120

Honey 1.4 19.6

French saladdressing

4.4 61.6

Italian saladdressing

0.8 11.2

Ranch saladdressing

3.7 51.8

Thousand Islanddressing

0.7 9.8

Tomato ketchup 4.1 57.4

Units given for beverages are mmol/l and mmol/g fresh wet weight for other foods.



23/53

Table 1.5 Effect of different cooking methods, growing season, storage conditions andgenotype on total antioxidant content (as the FRAP value) of plant foods

FoodTotal antioxidantcontent References

(a) Effect of cooking method

Cauliflower Wachtel-Galor et al.(2008)

Raw 2.8

Microwaved 4.9

Boiled 6.8

Steamed 8.7Cabbage

Raw 3.5

Microwaved 1.5

Boiled 2.5

Steamed 3.9

BroccoliRaw 3.9

Microwaved 5.4

Boiled 7.2

Steamed 13.1

Artichoke Ferracane et al.(2008)

Raw 56.9

Boiled 524.2

Steamed 705.0

Fried 430.4

Carrot Miglio et al. (2008)

Raw 6.8

Boiled 14.5

Steamed 12.3

Fried 32.5

Continued



24/53

Table 1.5 Effect of different cooking methods, growing season, storage conditions andgenotype on total antioxidant content (as the FRAP value) of plant foods cont'd


Courgette

Raw 27.9

Boiled 63.2

Steamed 59.2

Fried 79.7

Garlic Gorinstein et al.

(2008)Raw 12.0

Blanched 10.5

Boiled 7.4

Fried 10.9

Onion (white)

Raw 23.2Blanched 22.1

Boiled 16.5

Fried 22.9

Potato Adlia Lemos et al.(2013)

Fresh 670

Baked 510

Boiled 790

Microwaved 610

Steamed 670

(b) Effect of storage

Guava Storage at 5 C Patthamakanokporn

et al. (2008)Fresh 16.2

Day 3 15.2



25/53



Day 6 18

Day 10 22

Currants Storage at 4 C Storage at25 C

Piljac-Z egarac andSamec (2011)

Fresh 3.2 3.2

Day 2 2.75 2.75

Day 4 2.9 2.8Day 7 2.95 3.25

Day 9 2.15 2.25

Day 11 3.0 2.9

Day 14 2.9 3.35

Day 17 2.5 3.25

Day 22 2.15 2.2Day 30 3.75

Strawberry Storage at 4 C Storage at25 C


Fresh 2.5 2.5

Day 2 2.25 1.25

Day 4 2.5 2.25

Day 7 1.4

Day 9 1.5

Day 11 2.75

Raspberry Storage at 4 C Storage at25 C


Fresh 1.8 1.8

Day 2 2.35 2.75Day 4 1.8 3.25

Continued



26/53



Sour cherry Storage at 4 C Storage at25 C


Fresh 1.75 1.75

Day 2 1.5 1.45

Day 4 1.4 1.05

Day 7 1.3 1.1

Day 9 1.5 1.5Day 11 1.2 1.3

Day 14 2.1 1.8

Day 17 1.3 1.7

Day 22 1.6

Day 30 1.5

Cherry Storage at 4

C Storage at25 C Piljac-Z egarac andSamec (2011)

Fresh 1.1 1.1

Day 2 0.75 0.8

Day 4 0.75 0.75

Day 7 0.8

Day 9 0.85

Day 11 0.6

Day 14 1.4

Day 17 1.2

Cherry Storage at 4 C(with cling film)

Storage at25 C (coveredin filmimpermeable to

oxygen)

Giacalone andChiabrando (2013)

Fresh 7.8 7.8

Day 5 10.8 9.6



27/53



Day 10 7.8 9.0

Day 15 8.5 6.2

Mandarin Storage at 10 C(packed in boxes)

Shen et al. (2013)

Fresh 6.14

Day 2 5.74

Day 4 5.80Day 7 5.91

Day 9 5.80

Bayberry Storage at 1 Cand 15 kPa

Storage at 1 Cand 55 kPa

Chen et al. (2013)

Fresh 1.2 2.4

Day 2 1.5 3.52

Day 4 1.5 2.88

Day 7 1.14 3

Day 9 1.5 3.68

Day 11 1.9 4.32

(c) Effect of different cultivars

Tomato species Lenucci et al. (2006)

Cherubino 0.0035

Cherelino 0.003

Coralino 0.0034

Corbus 0.0045

LS203 0.0024

Lycorino 0.00235

Minired 0.0030

Noami 0.0024

Piccadilly 0.0023

Continued



28/53



Rubino top 0.0025

Sakura 0.0026

Salentino 0.0022

Sharon 0.0021

Shiren 0.0032

HLY02 0.0026

HLY13 0.003

HLY18 0.004

Kalvert 0.0029

Durian species Toledo et al. (2008)

Mon Thong 5.22

Chani 4.642

Kan Yao 4.094

Pung Mance 4.498

Kradum 3.948

Raspberrygenotypes

Tosun et al. (2009)

ERZ1 235

ERZ2 176

ERZ4 180

ERZ6 214

ERZ7 210

ERZ10 163

Apple species Wojdyo et al. (2008)

Alwa 1.0

BramleysSeeding

3.6

Cortland 1.12



29/53



Delbarestivale 1.56

Discovery 3.68

Elise 0.736

Fameuse 1.24

Fialka 5.04

Freedom 1.304

Geneva Early 3.06

Getmanskoje 2.268

Jonafree 1.196

Julyred 2.648

Kosztela 5.12

Meris 3.88

Odra 1.292

Piekna zHerrnhut

4.804

Rajka 1.292

Rubin 0.644

Sunrise 2.46

Teremok 1.872

ZimniejeLimonnoje

1.38

(d) Effect of harvest time

Strawberry(harvest month)

Pineli et al. (2012)

May 22

July 39September 30

Continued



30/53



Noni (harvestmonth)

Unripe Mediumripe

Sub-mature

Mature Iloki Assanga et al.(2013)

February March

1600 1750 1450 1550

MayJune 750 2500 2300 2050

November 550 560 3450 3750

Herbs (harvesttime)

Summer savory

Marjoram Thyme Vabkova andNeugebauerova(2012)

July 105 100 112

August 79 104 82

September 66 89 86

Curly Endive(harvest time)

Fresh Stored at4 C for 1 day

Stored at 4 Cfor 1 week

Venneria et al.(2012)

March 3.5 1.5 1.0

May 5.0 1.1 0.9

July 3.0 1.1 1.1

September 3.1 2.0 2.0

Rocket salad(harvest time)




January 10.6 9.6 8.0

March 7.2 7.1 6.9

May 12.0 11.0 9.8

July 13.0 12.0 11.0

September 12.8 12.8 12.4

November 6.0 6.0 5.9



31/53

testing, it is desirable to monitor changes in the plasma and urine concentra-tions of specific antioxidants, such as epigallocatechin gallate, ascorbic acid or

zeaxanthin, or their metabolites. However, this is not always possible, andexploring postabsorption changes in the FRAP or the nonuric acid FRAPvalue offers a useful initial screening approach if specific antioxidants in thefood are not known or for some reason cannot be measured.



Endive (harvesttime)




March 2.1 1.7 1.7

May 4.5 1.2 1.4

July 2.6 1.3 1.3

September 2.6 1.5 1.0

November 1.7 1.7 1.6

Units are mmol/g fresh wet weight, with exception of herbs, which are mmol/g dry weight.

Fasting totalFRAP

After meal totalFRAP

Fastingnon-UA FRAP

After mealnon-UA FRAP

0

250

500

750

1000

1250

Uric acid Ascorbic acidOther antioxidants

T o

t a l a n

t i o x

i d a n

t s ( M

)

Figure 1.3 Concept of nonuric acid FRAP value.



32/53

3.5. The modified FRAP assay (FRASC) for total antioxidantand ascorbic acid measurement

Many foods, especially fruits and fruit juices, are rich in ascorbic acid, and infood science, it is useful to measure this in order to advise consumers onwhich foods and how much of each is needed to reach the recommendeddaily intake of 70100 mg ( Chung et al., 2001; Levine et al., 1999, 2011 ).Strawberries and kiwi fruits (also known as Chinese gooseberry) have highascorbic acid content, and the daily requirement can be met by 45 straw-berries or 12 kiwi fruits. However, some fruits, including apples and pears,have low ascorbic acid content. Therefore, the daily intake of ascorbic acidcan vary greatly depending on the type of fruit eaten, and even people whotake their five-a-day could have lower than recommended intakes.

The FRASC assay represents a simple, one-step modification of the FRAPassay that permits the measurement of ascorbic acid in the same sample and atthe same time as the FRAP assay ( Benzie & Strain, 1996a; Benzie et al., 1999 ).FRASC has been validated against a reference HPLC method for ascorbic acid(Chung et al., 2001 ). In brief, the procedure is to prepare two matching ali-quots of each test sample. To one aliquot is added a small volume of a 4 unit/lsolution of ascorbic oxidase (ao). The recommended volume is 40 ml enzymesolution added to exactly 100 ml of sample. This very quickly destroys allascorbic acid present. To the other aliquot, a matching volume of water isadded. The paired samples are then run in the FRAP assay as described above,but an additional absorbance reading is taken at 1 min for both members of thepair. The 1-min absorbance of the aliquot treated with water represents theaction of all the redox-active antioxidant present, including ascorbic acid,while the 1-min absorbance of the enzyme-treated aliquot represents theaction of all the redox-active antioxidant except for ascorbic acid. The differ-ence between these two absorbances is due to the action of ascorbic acid aloneand is used to obtain the ascorbic acid concentration, which is obtained asfollows:

Ascorbic acid AA concentration of test sample mmol=l

Sample 1 min A593 nm ao A593 nm ao

1 min A593 nm standard

AA concentration of standard mmol=l

The 4-min value of the aliquot to which water (not enzyme) is used toobtain the FRAP value. To compensate for the dilution effect of the



33/53

addition of water to the samples, it is important that the calibrator betreated the same way, that is, 40 ml of water should be added to exactly100 ml of calibrator. If this is not done, the test results will be over-

estimated by 30%. For calibration, ascorbic acid solutions can be used,but the usual Fe 2 calibrator used for the FRAP assay works well, pro-vided its concentration is translated into AAE for calculation of ascorbicacid concentration. Remembering that one ascorbic acid molecule canreduce two Fe 3 to two Fe 2 , it follows that the signal (absorbance) givenby a 1000 mmol/l Fe 2 calibrator in the FRAP assay is equivalent to thesignal of reductive activity that would be given by a 500 mmol/l solutionof ascorbic acid.

In summary, the FRAP assay offers a speedy, inexpensive, and flexibletool to measure the combined activity of redox-active antioxidants in afood, or in plasma or urine following ingestion of a food. The resultscan be corrected easily for the contribution of uric acid. A modified formof the assay, FRASC, can be used to measure ascorbic acid concentration atthe same time as the FRAP value. The FRAP assay is very widely usedworldwide, and there now exists a rich database of FRAP values of thou-sands of foods, beverages, spices, and herbs, and these can be used to guide

dietary choices for enhanced antioxidant intake. In Section 4 , the reasonswhy this is thought to be beneficial and the supporting evidence arereviewed briefly.

4. WHY IS THE ANTIOXIDANT CONTENT OF FOODOF INTEREST?

4.1. Basic conceptsThe concept that enhanced optimal antioxidant defense of the body lowersrisk of disease and slows biological aging is an attractive one and impliesthat increased intake of antioxidant-rich foods can promote healthy agingand lower risk of chronic degenerative disease. The concept derives fromthe observation that key biomolecules undergo deleterious changes inform and function owing to oxidation by ROS, and that antioxidantsoppose such changes, and thereby decrease the oxidative stress that

is believed to be core to the aging process and to the etiology of chronicdegenerative diseases, including cardiovascular disease, cancer, anddementia ( Ames, 2006; Benzie, 2005; Frei, 2004; Gey, 1998; Kalioraet al., 2006; World Cancer Research Fund/American Institute for



34/53

Cancer Research, 2007 ). This is a simplistic view, but optimal or enhancedantioxidant defense achieved by increased intake of antioxidant-rich foodsmay confer benefit in relation to lowering risk of chronic age-related

disease (Ames, 2006; Benzie, 2005; Frei, 2004; Gey, 1998 ; Kaliora &Dedoussis, 2007 ; World Cancer Research Fund/American Institute for Cancer Research, 2007 ).

Many different types of antioxidant supplements are commercially avail-able, but supplementation trials with purified antioxidants, such as ascorbicacid, a -tocopherol, or b -carotene, have not demonstrated beneficial effects,and current recommendations for health promotion are for increased intakeof foods that are rich in antioxidants, not for supplements ( Lampe, 1999;

Packer & Cadenas, 2007 ). The specific agents and mechanisms in foods thatare responsible for health benefits are not yet clear, but there is strong andconvincing evidence that including more plant-based, antioxidant-richfoods, herbs, and beverages in the diet is effective in promoting healthand lowering risk of various age-related diseases ( Benzie & Wachtel-Galor, 2009; Duthie et al., 2006; Ma et al., 2008; Packer & Cadenas,2007; Tam et al., 2005; Thomas et al., 2006; Wachtel-Galor, Tomlinson,Benzie, 2004 ).

4.2. Antioxidants and health: The evidence and potentialimpact

The global population is aging, and the burden of chronic disease is increas-ing, bringing socioeconomic problems across the world. Therefore, there isintense research interest into cost-effective strategies for health promotion.Clearly, one attractive strategy is to increase the antioxidant status of the

body. Endogenous antioxidants such as SOD, glutathione and bilirubin can-not be easily or purposefully modulated, but higher intake of antioxidants inthe diet offers a potential route to higher defense and better protectionagainst oxidative stress. Information on the type, action, content, stability,bioaccessibility, and bioavailability of antioxidants in food are importantand complex issues. The evaluation of foods and the design of new methodsof food production, processing, and storage and the production of func-tional and designer foods for health promotion through enhanced anti-

oxidant content or effects are the focus of much research in the agri-foodsector. Therefore, the impact of research in measuring and modulating anti-oxidants in food has high potential impact on the commercial field as well ason the area of preventive healthcare.



35/53

Vegetarians who consume a well-balanced diet (avoiding iron andB vitamin deficiencies) have less chronic disease than omnivores and leadlonger, healthier lives ( Benzie & Wachtel-Galor, 2009 ). Intake of vitamin

C from food and plasma vitamin C concentrations are inversely correlatedwith risk of all cause mortality, incident heart disease, stroke, type 2 diabetes,and dementia ( Cooper et al., 2012; Harding et al., 2008; Jeurnink et al.,2012; Leenders et al., 2013 ). Also, those whose habitual diet is rich inplant-based, high antioxidant content foods have better glycaemic control,lower inflammation, and less oxidative stress ( Kashyap et al., 2005; Ma et al.,2008; Thomas et al., 2006 ). Still, while high intake of antioxidant-rich foodsis associated with better health, the underlying mechanisms are not yet clear,

and a role for the nonessential dietary-derived antioxidants remains to beestablished. Polyphenols are of special interest because it is known that some,in particular, the catechins from green tea, enter the plasma fairly rapidlyafter ingestion, although in small amounts ( Del Rio et al., 2010; Funget al., 2013 ). Furthermore, supplementation with green tea is associated withless oxidation-induced DNA damage and enhanced repair of these poten-tially mutagenic lesions (Han et al., 2011; Ho et al., 2013 ). This supportsthe large body of observational data showing that those who include green

tea regularly in their diet have less cancer and other diseases associated withoxidative stress (Benzie et al., 1999; Han et al., 2011; Ho et al., 2013; Molanet al., 2008).

Overall, the observational and experimental data are sufficiently strongand consistent to be able to recommend high intake of antioxidant-richfoods as a public health measure to promote healthy aging, even thoughthe mechanisms of action remain unclear. The molecular action of antiox-idants is an area of intense research interest but is very challenging to study.

Cell culture studies allow probing of gene activation and protein expressionin response to exposure to antioxidants, but results, especially when per-formed using immortalized cancer cells and very high doses of purifiedantioxidants, have very limited relevance to the normal physiological settingwithin the human body. Animal studies are useful as they offer a biosystemapproach, but most animals make their own vitamin C and are likely to havequite different absorption and metabolism patterns for food-derived antiox-idants compared to humans, and to have very different colonic microbiota.

Human studies are much more difficult because samples that can be collectedfrom healthy human subjects are very limited in type and volume. For study-ing effects at the biochemical level, blood plasma is most often used. For studying effects in relation to signaling pathways, gene expression, and



36/53

protein production, the most commonly used type of nucleated cell is theperipheral lymphocyte. It is not known how representative these samplesare of other fluids and cells, but recent human studies throw some light

on how dietary-derived antioxidants work at the molecular level, and bio-chemical data are providing insight into the metabolic fate of dietary anti-oxidants. These studies may lead to the answer as to why, given that wehave an endogenous antioxidant defense system, we still require regular input of dietary-derived, plant-based antioxidants for health maintenance.In Section 5, some recent human studies of metabolism and action of anti-oxidants in food are reviewed briefly, and current thinking in regard to thepossible molecular action of redox-active antioxidants or their metabolites is

discussed.

5. ANTIOXIDANTS IN FOOD: ENIGMAS ANDEVOLUTIONARY ASPECTS

The human body expends valuable resources in producing a range of antioxidants, and these are effective against the various ROS to which our

vital biological components are continually and unavoidably exposed. Still,we have an absolute need for at least two antioxidants (vitamins C and E) thatthe human body cannot synthesize and that we must obtain from food inregular and adequate amounts. This reliance on dietary input is an enigma,especially for vitamin C. With very few exceptions, all animals can manu-facture vitamin C. One of these exceptions is homo sapiens. Our ancestorscould synthesize this antioxidant, and human cells still contain the genefor the enzyme ( L-gulono- g-lactone oxidase) that transforms glucose into

ascorbic acid, but it is highly mutated and inactive ( Lachapelle & Drouin,2011 ). Therefore, during human evolutionary development, our ancestorslost the ability to manufacture a substance that was and remains essential for health. Evolutionary theory tells us that the loss of this synthetic ability cau-sed no physiological disadvantage; otherwise, it would not have becomeprevalent. Indeed, because this has become a ubiquitous characteristic of all humans, the initial silencing mutation of the gene must have conferredsome biological advantage. This is likely to have been a metabolic saving

of resources needed to manufacture something that was needed but suppliedin other ways, in this case in the diet. The Paleolithic diet of our hunter-gatherer ancestors is estimated to have provided a rich supply of ascorbicacid, along with other antioxidants ( Benzie, 2003; Cordain et al., 2002 ).



37/53

Therefore, it is likely that the mutation led to a metabolic saving without abiological cost or downside.

A comparison of the estimated antioxidant content of the Paleolithic diet

and the modern-day diet is shown in Fig. 1.4 . It can be seen that the intake of plant-based antioxidants that our biological systems were habituated to dur-ing our evolutionary development was high. This level of intake is likely tostill be what our biological systems have developed to deal with and perhapsrequire for optimal antioxidant defense and biological function. As noted,the evidence is clear and strong that habitual diets that are rich in plant-basedfoods are beneficial for health, and it is suggested that aligning our modern-day antioxidant intake more closely to that of the hunter-gatherer may be an

effective strategy for health, even if the optimal doses, key bioactives, andmolecular action of antioxidants in plant-based foods remain unclear. Still,it does not necessarily follow that health benefits of antioxidants in the dietcontinue to accrue with higher and higher doses. Indeed, there is evidencethat very high doses of vitamin E, vitamin C, and polyphenols in pure form

Paleolithic diet Micronutrient Modern diet604 Vitamin C (mg) 59115

33 Vitamin E (mg) 5.26.0

357 Folate ( mg) 208317

5560 Carotenoids ( mg) 18462048

87 Iron (mg) 9.517.2

? Copper (mg) 11.3

43

Paleolithic diet Modern day diet

Zinc (mg) 7.113.6

Figure 1.4 A closer look at the micronutrient intakes in the Paleolithic and modern-daydiet reveals that hunter gatherers consumed up to fivefoldmore vitamin C and vitamin E,and at least double the amount of carotenoids. Data from Benzie (2003) ; figure courtesy of SL Choi.



38/53

do not lower risk of disease and may in fact be harmful. For example, veryhigh doses of epigallocatechin gallate have been found to damage liver andkidney, and high-dose vitamin E and carotene were found to increase risk of

lung cancer ( Byers, 2008; Lambert et al., 2010; Russell, 2002 ).It is worth noting that the human body treats the polyphenol antioxi-

dants from plant-based foods as xenobiotics. Their absorption is very low,and there is rapid conjugation and excretion of those that are absorbed. Con-sequently, the plasma concentration of all polyphenols (including the cate-chins, quercetin, and the anthocyanins) is only 1 mmol/l, even with a dietthat provides up to 1 g per day of polyphenols from fruits and vegetables(Manach et al., 2004; Packer & Cadenas, 2007 ). This compares to plasma

concentrations of 80 and 40 mmol/l, respectively, for ascorbic acidand a-tocopherol, the daily inputs of which are 200 and 10 mg inhealthy diets. The severely restricted access of polyphenols is intriguingand paradoxical, given the health benefits of polyphenol-rich, plant-baseddiets. Recent evidence in two main areas gives clues as to solve the enigmaand explain why this may be central to the mechanisms by which diets rich inantioxidant phytochemicals exert their beneficial effects. One area of recentresearch is in redox chemistry and molecular effects of antioxidants, and the

other in relation to the symbiotic relationship between the human body andits colonic microbiota.

6. MECHANISMS OF ACTION: REDOX ISSUES,PHYTOHORMESIS, AND COLONIC MICROBIOTA

The mechanisms by which antioxidants in food drive health benefitsare not clearly established. Some effects are likely to be due to antioxidant

action, but the concepts that all effects are due to direct antioxidant effectsand that more is better are overly simplistic. The very limited bioavailabil-ity of most phytochemicals is unlikely to be an accidental physiologicaloccurrence. If more efficient absorption of phytochemicals from food wasof physiological benefit and led to improved reproductive success, thenmutations that enhanced absorption would have been retained and becomeprevalent. The fact they have not implies that either these phytochemicalsare not needed or that they are needed in low levels only. Indeed, our bio-

logical restrictions that effectively limit bioavailability suggest that high levelsare to be avoided. This is paradoxicalbut by considering redox issues,some sense of the biological value of restricted bioavailability of phytochem-icals and the molecular action of these is emerging. In addition, recent



39/53

evidence points to the colonic microflora as playing a potentially importantsymbiotic role in phytochemical metabolism.

6.1. Redox tone changes as a driving force for cytoprotectionand the biological sense of restricted bioavailability of dietary antioxidants: The redox-sensitive KEAP-1 Nrf2signaling pathway

Protein-based antioxidants, such as enzymes, in food are likely to be des-troyed during cooking and the digestive process. The antioxidants most rel-evant to nutritional science are redox-active substances that can destroy or neutralize ROS by quenching or electron donation. These include vitamins

C and E, and the polyphenol family that contains catechins, quercetin, andthe deeply colored anthocyanins. Chocolate, teas, wines, berries and other fruits, and leafy green vegetables are good sources of these redox-active anti-oxidants (Benzie & Wachtel-Galor, 2010; Halvorsen et al., 2006; Serafiniet al., 2003). During their reductive action, these antioxidants becomereduced, and all redox-active antioxidants have the potential to act asprooxidants. Ascorbic acid and epigallocatechin gallate, for example, cangenerate the ROS hydrogen peroxide in vitro, and it has been shown that

this ROS generation is dose dependent ( Ho et al., 2013 ). At high concen-tration, this prooxidant activity leads to in vitro cytotoxic effects of a dietaryantioxidant, such as vitamin C, or an antioxidant-rich food or beverage, suchas tea (Chai et al., 2003 ). Whether this prooxidant effect occurs in vivo is atopic of intense research interest. It is noted that there is no evidence thatantioxidant-rich foods (as opposed to purified supplements) induce oxida-tive stress or have other deleterious effects on human health. Furthermore,if confirmed to occur in vivo, prooxidant effects would be limited by the very

limited bioavailability of most dietary-derived antioxidants and their rapidmetabolism and excretion. However, frequent but subtle prooxidantchanges in intracellular redox tone, or balance, could occur following die-tary input of phytochemicals, and there is growing evidence that these smallprooxidant waves underlie the health protective effects of regular intake of antioxidant-rich foods ( Benzie & Wachtel-Galor, 2010; Han et al., 2011 ).This does not preclude direct antioxidant effects by some food componentsin some situations, but the prooxidant activity could lead to an array of

endogenous redox-sensitive cytoprotective adaptations that have wide-spread beneficial effects for health (Benzie & Wachtel-Galor, 2010; Leeet al., 2013; Surh et al., 2008 ). This concept of prooxidant phytochemicalstriggering redox-sensitive cytoprotective adaptations has been termed



40/53

redox hormesis and phytohormesis ( Davinelli et al., 2012 ). There is agrowing body of evidence from cell culture and animal studies that manyphytochemicals, including allium compounds, sulforaphane, curcumin,

epigallocatechin gallate, lycopene, quercetin, and resveratrol, inducechanges in cellular signaling pathways that trigger these protective adapta-tions (Lee et al., 2013). The phytohormetic effect is believed to be mediatedmainly via the redox-sensitive KEAP-1Nrf2 signaling pathway, but there isemerging evidence that there are phytochemical-induced effects on kinasesand DNA methylation also ( Benzie & Wachtel-Galor, 2010; Lee et al.,2013; Surh et al., 2008 ).

Nuclear factor erythroid-2-related factor-2 (Nrf2) is a transcription

factor that in resting cells is mainly bound in an inactive form tocysteine-rich Kelch-like ECH-associated protein-1 (Keap-1) ( Lee et al.,2013; Surh et al., 2008 ). Keap-1 promotes ubiquitination and proteolysisof bound Nrf2. Release of Nrf2 from its suppressor protein allows Nrf2to translocate to the nucleus. The mechanism by which Nrf2 is releasedfrom Keap-1 is currently unclear, but there is evidence that oxidation or chemical modification of key cysteine residues in Keap-1 leads to stabili-zation of Nrf2, proteolysis of Keap-1, and release of stabilized Nrf2 into

the cytoplasm ( Lee et al., 2013). However, phosphorylation of serine or threonine residues in Nrf2 is suggested to occur, inducing release of Nrf2from its suppressor. It is also suggested that phosphorylation of tyrosine141residue in Keap-1 enhances its stability and Nrf2 binding, while dephos-phorylation promotes degradation of Keap-1, with stabilization and releaseof Nrf2 (Lee et al., 2013). Therefore, oxidative action, direct chemicalmodification, and effects on kinases and phosphorylases may all be involvedin phytochemical action ( Lee et al., 2013).

After release, stabilized Nrf2 translocates to the nucleus and combineswith Maf protein and other transcription factors, and binds to the regulatorycis-element in the 5 0-flanking promoter regions of genes that encode for var-ious cytoprotective and stress opposing factors. The regulatory regions of such genes are referred to as antioxidant response elements (ARE) or elec-trophile response elements. The products of the Nrf2/ARE-activated genesinclude antioxidant and detoxifying enzymes, hOGG1, the enzyme that cat-alyzes the first step in repair of oxidation-induced lesions to DNA, and heme

oxygenase-1, which has antioxidant and antiinflammatory effects ( Bach,2005; Soares & Bach, 2008). This phytohormetic redox-associated conceptis outlined in Fig. 1.5. Overall, the regular, subtle push of mild pro-oxidant effects of antioxidant phytochemicals could induce powerful



41/53


42/53

antioxidant effects of dietary antioxidants but ascribes far more widespreadhealth-related effects to prooxidant changes that are insufficient to causedamage, but that induce a type of oxidative polishing or conditioning

of the cell that enhances its survival and defense. However, it is noted thatthe effects, if confirmed, are driven by small amounts of phytochemicals, andthis brings into question the advisability of overcoming natural barriers thatlimit absorption of phytochemicals. Advances in nanodelivery systems havebeen used to enhance absorption of phytochemicals and nutraceuticals, andmicroencapsulation technology can prevent destruction of antioxidantswithin the gastrointestinal tract, increasing bioaccessibility of antioxidants(Huang & Chang, 2009; Neves et al., 2013; Souto et al., 2013 ). These tech-

nologies apply mainly to production of supplements, but caution is neededin their development and use. Overcoming the physiological barriers thatlimit bioavailability of phytochemicals in the human body may not be help-ful for health, and indeed may do more harm than the good no doubtintended. Increasing absorption of redox-active phytochemicals could leadto large changes in redox tone that the cells cannot adapt to and overcome,with damage and cytotoxic effects ensuing rather than cytoprotection.A further point worth considering is the fate of phytochemicals left

unabsorbed in the gut. Recent findings indicated that these may play animportant role in the health effects of antioxidant-rich diets. This aspectof antioxidants in food is discussed briefly in the next section.

6.2. Colonic microflora and metabolism of food antioxidants:Moving into new territory of the microbiome

The absorption of antioxidant phytochemicals from the human gastrointes-tinal tract is severely restricted. Even in the case of ascorbic acid, which is

the most soluble and best absorbed antioxidant, absorption is limited. If the dose is low, absorption can approach 100%, but as the dose increasesthe relative amount that is absorbed decreases ( Davey et al., 2000 ). Conse-quently, the increase in plasma ascorbic acid following ingestion of severalgrams of vitamin C is not significantly larger than that seen after taking a fewhundred milligrams ( Benzie & Strain, 1997 ).

Many foods and beverages, especially of plant origin, have a high andvaried content of antioxidants, and most of the ingested antioxidants in every

meal or drink are left behind and arrive in large amounts in the large intes-tine. Previously, if considered at all, these were believed to serve no purpose,representing only nonabsorbable and unwanted remnants of food. It wasthen suggested that these unabsorbed phytochemicals could benefit colonic



43/53

health, perhaps by acting as antioxidants in situ or by acting as probiotics,enhancing conditions for the more desirable types of colonic microflora.However, the microbial inhabitants of the colon (the microbiota) are

increasingly recognized as our partners in metabolizing unabsorbed phy-tochemicals into absorbable and possibly bioactive molecules ( Del Rio et al.,2010; Leone et al., 2013 ). The bulk of the unabsorbed phytochemicals arepolyphenolic compounds, which includes anthocyanins, flavonols, andflavanols. In the past few years, many colonic metabolites of flavan-3-ols(catechins) from tea have been identified in human plasma ( Del Rioet al., 2010). Ring scission products of catechins appear in plasma at sevenor more hours after ingestion of tea, and if they are included in the calcu-

lation of overall bioavailability of tea polyphenols, the figure is substantiallyelevated in some individuals ( Crozier et al., 2009; Del Rio et al., 2010 ). It isnot yet known if these colonic metabolites are bioactive. Furthermore,interindividual variation is very high, and these ring scission products of tea polyphenols are not found in all subjects. The variation in response issuggested to be due to differences in the colonic microbiome. The micro-biome is not a constant and can be changed with changes in diet, antibiotics,prebiotics, and probiotics ( Leone et al., 2013 ). Furthermore, the colonic

microbiome of the breast-fed baby is different from that of the formula-fed baby, and the first imprinting of the neonatal microbiome seems tobe difficult to change and to have an impact on risk of diseases later in life(Zivkovic et al., 2011 ). Overall, there is increasing awareness that we have asymbiotic relationship with the microbial inhabitants of our lower intestine,and modifying the microbiome could affect the metabolism, bioactivity, andabsorption of unabsorbed antioxidants from food. This forms a new andchallenging area of research in food, nutrition, and health science.

7. THE STORY SO FAR

In summary, some key points in relation to antioxidants in food andthe concepts and cautions discussed in this chapter are:

1. The health benefits of antioxidant-rich diets are clear, but they arerelated to intake of whole foods and do not apply to supplementationwith pure antioxidants.

2. Plant-based diets contain thousands of different antioxidant phyto-chemicals, but whether there are only few bioactives of importanceor if there is a role for each or if there is important interaction betweenthem is not known.



44/53

3. It is neither possible nor feasible to measure the content of each indi-vidual antioxidant in food, and measurement of the total antioxidantcontent of foods has become a widely adopted approach.

4. There is now a large body of data on the total antioxidant content of foods, and this can be used in dietary planning for increased antioxidantintake.

5. All the currently available tests that measure total antioxidant contentof foods and biological fluids are in vitro tests, and phytochemicals mayact quite differently in an in vivo system.

6. Vitamin C (ascorbic acid) is known to be vital for human health, andvitamin E (mainly a -tocopherol in the human body) plays a key role in

protecting lipid components and structures, but even so, the rec-ommended daily i

Date post:	02-Jun-2018
Category:	Documents
Upload:	andrew-berry
View:	223 times
Download:	0 times

Antioxidants in Food: Content, Measurement, Significance, Action, Cautions, Caveats, and Research...

Documents