ISOLATION OF MIXTURES FROM AND VEGETABLES AND EVALUATION OF THEIR STABILITY, AVAILABILITY AND BIOTRANSFORMATION IN THE

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of

Philosophy in the Graduate School of The Ohio State University

By

Jian He, M.S.

* * * * *

The Ohio State University 2008

Dissertation Committee:

Approved by

Assistant Professor M. Mónica Giusti, Adviser

Professor Steven J. Schwartz ______

Professor Mark L. Failla Adviser

Professor W. James Harper Science and Nutrition Graduate Program

Associate Professor Timothy A. Schroeder

© Copyright by JIAN HE 2008 ABSTRACT

Anthocyanins are among the most abundant in fruits and vegetables

and exhibit potent activity. Our previous studies suggested that

present in the gastrointestinal tract (GIT) of rats exert chemopreventive effect on colon

. Since such effect was dose dependent, it is important to examine the stability and

transformation of anthocyanins in the GIT. Based on some in vitro studies and limited in

vivo evidence, we hypothesized that anthocyanins were moderately stable under the

influence of digestive enzymes and physiological conditions in the GIT, and their

chemical structures determine the stability, accessibility, biotransformation, and health-

promoting effects.

To validate our hypothesis a series of studies were conducted. We first examined

the transit, stability, transformation, and uptake of black anthocyanins in rat

GIT. Our results suggest that ingested anthocyanin remained relatively stable in the GIT lumen and were efficiently taken up into the GIT tissues with limited transfer to the plasma compartment. We also observed selective degradation of -3- glucoside in the small intestine lumen, which was tentatively attributed to β-glucosidase activity in the small intestine.

ii In order to eliminate interferencing non-anthocyanin compounds in bioassays, a novel mixed mode cation-exchange/reversed-phase (MCX) solid-phase extraction (SPE) technique was developed for high purity isolation of anthocyanins. The new methodology drastically increased purity and efficiency as compared to three widely used SPE techniques – C18, HLB, and LH-20 columns. For the majority of plant materials evaluated,

the new technique achieved above 99% anthocyanin purity while maintaining excellent

yield (93.6 ± 0.55%) and low cost. The resulted high purity anthocyanins were later used

in the enzymatic assays.

To elucidate the effect of intestinal enzymes on anthocyanin digestion, an in vitro

model was employed. Cell-free extract of pig small intestinal mucosa and crude extract of

lactase containing /supplement were examined with respect to their β-glycosidase

activities on highly purified anthocyanins. Selective degradation of anthocyanin

glucosides and galactosides was observed in the presence of small intestinal mucosa cell-

free extract and lactase supplement extract, respectively. The type of aglycone also substantially affected the resistance of anthocyanins to enzymatic degradation.

The outcome of this dissertation demonstrates that chemical structure of

anthocyanins greatly affects their stability in the GIT. Such information will contribute to

the exploration of anthocyanin’s health benefits in vivo, and may eventually facilitate the

development of value added foods/nutraceuticals that promote the healthiness of people.

iii

The dissertation is dedicated to my mother Zonghui Hu, my father Shide He, and my dear

wife Weishu Xue for their love and ultimate support to me.

iv ACKNOWLEDGMENTS

I would like to thank my advisor Dr M. Mónica Giusti for her intelligent advices, warm encouragement, and ultimate patience throughout my graduate study. My success in academia is not possible without her guidance.

I wish to thank Dr Mark L. Failla for sharing his incredible knowledge in nutrition to help me designing my experiments and fulfilling my academic goals.

I also thank the other members of my advisory committee, Dr Steve J. Schwartz and Dr W. James Harper for their invaluable opinions to my research.

I acknowledge Dr Luis Rodriguez-Saona for advising me of the FT-IR work.

I appreciate the chance of collaborating with Kristin E. Keatley who successfully graduated from the Department of Human Nutrition last year.

Great thanks to my colleagues and friends, especially all my lab mates Fer Pólit,

Nuryati Pangestu, Fabiola Gutierrez, Taylor Wallace, Dante Vargas, Christian Sweeney,

Jennifer Willig, Jodee Johnson, Pu Jing, and Matthew Bunce. All of these lovely people helped me as much as they could and I really enjoyed the happy days spent with them.

I am grateful to USDA for funding my doctoral study. My doctoral research was supported by the USDA NRI Competitive Grants Program 2004-35503-15190.

v Very special thanks to my mother Zonghui Hu, my parents in law Zhiguang Xue and Aiping Zheng, and my wife Weishu Xue. Without their unselfish support I could never achieve so much.

vi VITA

Feb 19, 1980·······································Born – Hefei, China

1998 – 2002········································B.E. Food Science and Engineering, Hefei University of Technology, China

2002 – 2004········································M.S. Nutrition and Food Science University of Maryland, College Park, MD

2004 – 2008········································Graduate Research and Teaching Assistant The Ohio State University, Columbus, OH

PUBLICATIONS

1. He, J.; Rodriguez-Saona, L. E.; Giusti, M. M., Mid-infrared spectroscopy for juice authentication - rapid differentiation of commercial juices. J Agric Food Chem 2007, 55, 4443-4452.

2. He, J.; Magnuson, B. A.; Lala, G.; Tian, Q.; Schwartz, S. J.; Giusti, M. M., Intact anthocyanins and metabolites in rat urine and plasma after 3 months of anthocyanin supplementation. Nutr Cancer 2006, 54, 3-12.

vii 3. Lala, G.; Malik, M.; Zhao, C.; He, J.; Kwon, Y.; Giusti, M. M.; Magnuson, B. A., Anthocyanin-rich extracts inhibit multiple biomarkers of colon cancer in rats. Nutr Cancer 2006, 54, 84-93.

4. He, J.; Magnuson, B. A.; Giusti, M. M., Analysis of anthocyanins in rat intestinal contents – impact of anthocyanin chemical structure on fecal excretion. J Agric Food Chem 2005, 53, 2859-2866.

FIELDS OF STUDY

Major Field: Food Science and Nutrition

viii TABLE OF CONTENTS

Page

ABSTRACT······················································································································ii

ACKNOWLEDGMENTS·································································································v

VITA·······························································································································vii

TABLE OF CONTENTS ·································································································ix

LIST OF TABLES ·········································································································xvi

LIST OF FIGURES·······································································································xvii

LIST OF ABBREVIATIONS ·························································································xx

1. INTRODUCTION····································································································1

2. LITERATURE REVIEW·························································································3

2.1 ANTHOCYANINS AS NATURAL PIGMENTS ···········································3

2.2 CHEMICAL STRUCTURE OF ANTHOCYANINS ······································4

2.2.1 Chemical structure of ·························································4

2.2.2 Anthocyanin aglycones········································································5

2.2.3 Glycosylation and acylation·································································7

ix 2.2.4 The influence of pH on anthocyanin chemical structure ······················8

2.2.5 ···········································································11

2.3 ANTHOCYANINS IN HUMAN DIET·························································13

2.3.1 Occurrence of anthocyanins in plant materials···································13

2.3.2 Anthocyanins in foods and beverages················································14

2.3.3 Toxicity of anthocyanins····································································15

2.4 STABILITY OF ANTHOCYANINS ····························································16

2.4.1 The effect of pH·················································································16

2.4.2 The effect of temperature···································································17

2.4.3 The effect of light ··············································································18

2.4.4 The effect of enzymes········································································18

2.4.5 The influence of oxygen, peroxide and ascorbic acid·························19

2.4.6 Self-association, acylation and co-pigmentation ································21

2.5 PUTATIVE HEALTH-PROMOTING EFFECTS OF ANTHOCYANINS···22

2.5.1 Relief of oxidative stress····································································23

2.5.2 Prevention of cardiovascular diseases················································24

2.5.3 Anti-inflammatory activity ································································26

x 2.5.4 Anti-carcinogenic activity··································································27

2.5.5 Prevention of obesity ·········································································32

2.5.6 Control of diabetes·············································································33

2.5.7 Improvement of eye vision ································································34

2.5.8 Antimicrobial activity ········································································34

2.6 BIOAVAILABILITY AND OF ANTHOCYANINS ·········35

2.6.1 Overview ···························································································36

2.6.2 Gastric absorption··············································································37

2.6.3 Direct absorption in the small intestine··············································40

2.6.4 Deconjugation of carbohydrate moieties············································43

2.6.5 The influence of colonic microflora···················································46

2.6.6 Metabolism in intestinal mucosa and tissues······································47

2.6.7 Tissue distribution··············································································48

2.6.8 Excretion····························································································49

3. BLACK RASPBERRY ANTHOCYANINS ARE STABLE IN THE DIGESTIVE TRACT LUMEN AND EFFICIENTLY TRANSPORTED INTO GASTRIC AND SMALL INTESTINAL TISSUES IN RAT·····································································52

3.1. ABSTRACT ··································································································52

xi 3.2. INTRODUCTION·························································································53

3.3. MATERIALS AND METHODS···································································55

3.3.1 Chemicals and materials ····································································55

3.3.2 Black raspberry extract ······································································55

3.3.3 Animals and experimental design ······················································56

3.3.4 Sample preparation. ···········································································57

3.3.5 Recovery····························································································59

3.3.6 HPLC-MS analysis of anthocyanins and anthocyanin metabolites ····59

3.3.7 Calibration curve ···············································································60

3.3.8 Statistical analysis··············································································61

3.4. RESULTS AND DISCUSSION ····································································61

3.4.1 Flux of anthocyanins in GIT lumen ···················································61

3.4.2 Anthocyanins in GIT tissues······························································67

3.4.3 Anthocyanin transformation in the GIT ·············································69

3.4.4 Metabolism and excretion of absorbed anthocyanins·························75

3.5. CONCLUSIONS ···························································································77

3.6. ACKNOWLEDGEMENTS···········································································77

xii 4. HIGH-PURITY ISOLATION OF ANTHOCYANIN MIXTURES FROM FRUITS AND VEGETABLES – A NOVEL SOLID-PHASE EXTRACTION METHOD ··········79

4.1. ABSTRACT ··································································································79

4.2. INTRODUCTION·························································································80

4.3. MATERIALS AND METHODS···································································83

4.3.1 Reagents and standards······································································83

4.3.2 Anthocyanin sources and sample preparation ····································84

4.3.3 Fractionation methods comparison ····················································84

4.3.4 Extended evaluation on a variety of anthocyanin sources ··················89

4.3.5 Optimization for food application······················································89

4.3.6 Statistical analysis··············································································89

4.4. RESULTS AND DISCUSSION ····································································90

4.4.1 MCX method development································································90

4.4.2 Fractionation methods comparison ····················································91

4.4.3 Extended evaluation on a variety of anthocyanin sources ················102

4.4.4 Optimization for food application····················································104

4.5. CONCLUSIONS ·························································································104

4.6. ACKNOWLEDGEMENTS·········································································105 xiii 5. IMPACT OF SMALL INTESTINAL β-GLUCOSIDASE AND DIETARY LACTASE ON ANTHOCYANIN DIGESTION ·························································106

5.1. ABSTRACT ································································································106

5.2. INTRODUCTION·······················································································107

5.3. MATERIALS AND METHODS·································································108

5.3.1 Chemicals and materials ··································································109

5.3.2 Purification of anthocyanins ····························································109

5.3.3 β-glucosidase activity in porcine small intestinal mucosa················111

5.3.4 Lactase from foods···········································································114

5.3.5 HPLC-MS analysis of anthocyanins ················································118

5.3.6 Calibration curve ·············································································119

5.3.7 Statistical analysis············································································119

5.4. RESULTS AND DISCUSSION ··································································120

5.4.1 Lactase activity of pig small intestinal mucosa extract ····················120

5.4.2 β-glucosidase activity on anthocyanins············································120

5.4.3 Calculation of apparent Km and Vmax for intestinal β-glucosidase····124

5.4.4 Substrate specificity of small intestinal β-glucosidase ·····················127

5.4.5 Activity of lactase supplement on anthocyanins ······························131 xiv 5.4.6 The influence of lactase residual in foods on anthocyanins··············131

5.4.7 Substrate specificity of lactase from supplement ·····························135

5.4.8 Calculation of apparent Km and Vmax for dietary lactase ··················135

5.4.9 Anti-proliferative effect on colon cancer cell line····························137

5.5. CONCLUSIONS ·························································································141

5.6. ACKNOWLEDGEMENTS·········································································141

BIBLIOGRAPHY ·········································································································142

xv LIST OF TABLES

Table Page

Table 2.1 Differences on chemical structure, color and λmax of most commonly found in nature...... 6

Table 2.2 Km of quercetin 4'-glu and genistein 7-glu by β-glucosidase from human and animal intestine and liver...... 45

Table 3.1 Peak assignments of anthocyanins extracted from the gastric and intestinal contents of rats administered black raspberry extract...... 63

Table 3.2 Black raspberry anthocyanins in urine collected from rat bladder ...... 76

Table 4.1 Mobile phases used to elute compounds of interest from the SPE cartridges .. 85

Table 4.2 Visually observed anthocyanin color loss during purification of anthocyanin mixtures by the different SPE procedures ...... 94

Table 5.1 Composition of mixtures used to test the effect of pig small intestinal enzymes on black raspberry anthocyanins...... 113

Table 5.2 Compositions of buffered solutions used to investigate the effect of lactase supplement on anthocyanins...... 116

Table 5.3 Kinetic constants for the hydrolysis of LPH substrates...... 124

Table 5.4 Profile change of anthocyanins after incubation with LPH for 30 min 129

xvi LIST OF FIGURES

Figure Page

Figure 2.1 Representative aglycone structures of the common sub-classes...... 5

Figure 2.2 Basic chemical structure of a common ...... 6

Figure 2.3 Chemical structure of an acylated anthocyanin (Mv-3-(p-coumaryl)glu) in grape skin...... 8

Figure 2.4 Scheme of the pH-dependent structural interconversion between dominant forms of mono-glycosylated anthocyanins in aqueous phase...... 9

Figure 2.5 A representative chemical structure of pyranoanthocyanins...... 12

Figure 2.6 Effect of feeding anthocyanin rich diets on total anthocyanin concentration in rat urine, colonic cell proliferation index, large ACF multiplicity, total anthocyanin concentration in feces, fecal moisture content, and fecal bile acids concentration ...... 31

Figure 2.7 Mv-3-glu concentration in plasma sampled from either the portal vein or the heart of rats administered with grape anthocyanins...... 39

Figure 2.8 The ratio of mean 24h cumulative urinary excretion: intake of quercetin ...... 42

Figure 2.9 Intergrated putative pathways of dietary flavonoids absorption, metabolism, distribution, and excretion ...... 51

xvii Figure 3.1 Chromatographic profiles of Cy standard, black raspberry anthocyanins in the administered crude extract, and representative samples of luminal contents from stomach and small intestine from a rat killed 120 min after gavage...... 64

Figure 3.2 Black raspberry anthocyanins in the gastric and small intestinal contents and in the small intestinal tissue ...... 65

Figure 3.3 UV-Vis spectra of the anthocyanin- complex extracted from the stomach tissue ...... 68

Figure 3.4 Relative abundance of black raspberry anthocyanins in the gastric and small intestinal lumens ...... 71

Figure 3.5 Relative abundance of Cy-3-glu is decreased in the small intestinal content . 73

Figure 4.1 The chemical structure of MCX sorbent and an anthocyanin molecule at varied pH conditions...... 83

Figure 4.2 Percentage of individual chokeberry anthocyanin peaks after purification .... 94

Figure 4.3 HPLC-PDA chromatograms of chokeberry crude extract and purified anthocyanin fractions...... 97

Figure 4.4 The purity and recovery of anthocyanins and other-phenols based on the AUC of HPLC-PDA chromatograms...... 98

Figure 4.5 Impurities in chokeberry anthocyanin fractions as indicated by the appearance of noise peaks in the MS chromatogram...... 100

Figure 4.6 HCA dendrograms of chokeberry and corn extracts...... 101

Figure 4.7 The purity of anthocyanins purified from 10 additional commodities...... 103 xviii Figure 5.1 Degradation of purified black raspberry anthocyanins during incubation with pig small intestinal cell-free extract in the presence and absence of β-glucosidase inhibitors ...... 123

Figure 5.2 Saturation curve for the small intestinal mucosal β-glucosidase showing the relationship between Cy-3-glu substrate concentration and rate of Cy-3-glu hydrolysis125

Figure 5.3 HPLC chromatograms of purified grape anthocyanins incubated with LPH for 0 min and 30 min in the presence or absence of D-gluconolactone ...... 128

Figure 5.4 Chemical structures of the common anthocyanidins glycosylated on the C-3 position...... 130

Figure 5.5 HPLC chromatograms of purified blueberry anthocyanins incubated with either heat-inactivated lactase from supplement, active lactase from supplement extract, or lactase from supplement in the presence of 584 mM lactose for 15 min ...... 133

Figure 5.6 Profile change of anthocyanin galactosides in blueberry after incubation with the lactase supplement extract for 15 min...... 134

Figure 5.7 Lactase supplement extract substrate specificity on the 5 anthocyanin galactosides in blueberry...... 136

Figure 5.8 Saturation curve for the lactase supplement extract showing the relationship between Cy-3-gal substrate concentration and rate of Cy-3-gal hydrolysis...... 137

Figure 5.9 Intact chokeberry anthocyanins remaining after incubation with buffer or dietary lactase extract for 45 min...... 138

Figure 5.10 Growth inhibition (48 h) of human colon cancer cell line HT-29 by purified chokeberry anthocyanins and lactase treated chokeberry anthocyanins...... 140 xix LIST OF ABBREVIATIONS

ACF Aberrant crypt foci

ADI Acceptable daily intake

ANOVA Analysis of Variance

AOM Azoxymethane

Ara Arabinoside

CBG Cytosolic β-glucosidase

COMT Catechol-O-methyltransferase

COX Cyclooxygenases

Cy Cyanidin

DD Double distilled

Gal Galactoside

GIT Gastrointestinal tract

Glc Glucuronide

Glu Glucoside

Gly

FBS Fetal bovine serum

HDL High-density lipoprotein

HPLC High-pressure liquid chromatography

xx JECFA Joint FAO/WHO Expert Committee on Food Additives

LDL Low-density lipoprotein

LPH Lactase-phlorizin hydrolase

MPLC Medium-pressure liquid chromatography

MS Mass spectroscopy

Mv

NOEL No-observed-effect-level

ORAC Oxygen radical absorbance capacity

PC Protocatechuic acid

Pg

POD proxidase

PPO Polyphenol oxidase

Pn

Pt

Rha Rhamnose

ROS

Rut Rutinoside

Sam Sambubioside

SAOC Serum antioxidant capacity

SD Standard deviation

xxi SE Standard error

SGLT1 Sodium-dependent glucose transporter

SRB Sulforhodamine B

SULT Phenol sulfotransferases

TAC Total antioxidant capacity

TCA Trichloroacetic acid

UDPGD Uridine diphosphoglucose glucose dehydrogenase

UDPGT Uridine diphosphoglucose glucuronosyl transferase

Xyl Xyloside

xxii

CHAPTER 1

1. INTRODUCTION

Anthocyanins are a group of polyphenols widely distributed in fruits, vegetables,

and based beverages as natural pigments. Among all the recognized polyphenols,

anthocyanins are especially important because of their potent antioxidant property and

high consumption, probably the highest among all flavonoids (1-3). Although cell culture

studies, animal studies and epidemiological studies have suggested association between

anthocyanin intake and disease prevention, the bioactive form of anthocyanins or

anthocyanin metabolites is still controversial. Studies have repeatedly shown that only

very limited amount of intact anthocyanins are absorbed (4). Amounts of glucuronidated,

sulfoconjugated, and methylated anthocyanin metabolites also appear to be limited (4-7).

Our research group has suggested that the unabsorbed intact anthocyanins

remaining in the gastrointestinal tract (GIT) may exert chemopreventive activity through

direct contact with epithelial cells, in that a positive correlation between anthocyanins

concentration in the colon content and the reduced incidence of colon lesion was

observed in our previous study (8). Thus anthocyanin stability in the GIT appears to be a determination factor of anthocyanin’s health benefit on GIT related organs. In addition, 1 the way that anthocyanins are present in the GIT absorption sites likely determines the

absorption of anthocyanins or their metabolites into the blood stream and consequently

being exposed to other organs (5, 6, 9).

Above all, understanding the stability and transformation of anthocyanins in the

GIT is a prerequisite for the evaluation of their putative health benefits. Based on some in

vitro studies about anthocyanin’s stability during simulated digestion (10-12) and limited in vivo evidence (7, 13, 14), we hypothesized that anthocyanins were moderately stable under the influence of digestive enzymes and neutral pH in the GIT and their chemical structures determine the stability, accessibility, biotransformation, and health-promoting effects. The overall objective of this dissertation research was to increase understanding of the relationship between anthocyanin chemical structure and their existance in the GIT, which potentially affects the health-promoting value of anthocyanin-rich diet. First we

investigated the in vivo stability, transformation and uptake of semi-purified black

raspberry anthocyanins in the GIT using a rat model (Chapter 3). Then in order to better

understand the effect of the GIT enzymes on anthocyanins we developed a simple and

efficient methodology to remove co-extracted interference plant components from

anthocyanin mixtures (chapter 4). The highly purified anthocyanins were subsequently

tested in vitro with extracted β-glydosidases (chapter 5). Two types of β-glycosidases, the

endogenous intestinal brush border β-glucosidase and the exogenous lactase, were

evaluated with respect to selective degradation of various anthocyanins.

2

CHAPTER 2

2. LITERATURE REVIEW

2.1 ANTHOCYANINS AS NATURAL PIGMENTS

Anthocyanins constitute the largest and probably the most important group of

water soluble natural pigments (15). To date there has been at least 539 anthocyanins

identified in nature and such a versatile group is responsible for the vivid blue, purple,

and color of many fruits, vegetables and (16). In fact the word anthocyanin

was derived from two Greek words: anthos and kyanos, meaning and dark blue

respectively (17). Anthocyanins are believed to be important to plants as their color

attracts animals, leading to dispersal and pollination. Due to strong absorption of

light, they may also be important in protecting plants from ultraviolet-induced damage

(18).

Anthocyanins are primarily used as colorants in beverages in the food industry.

As public concern about synthetic food dyes has increased recently, consumers and food manufacturers desire colorants from natural sources. The commonly used synthetic dyes in the food industry have been suspected to cause adverse behavioral and neurological

3 effects in the past. A recent trial involving 153 3-year-old and 144 8/9-year-old children

concluded that artificial colorants combined with sodium benzoate preservative in the

diet resulted in statistically significant increase of hyperactivity in children (19). As

promising alternatives to the most widely used synthetic food dye FD&C Red #40 (called

Allura red in Europe), anthocyanins have recently attracted great interest.

2.2 CHEMICAL STRUCTURE OF ANTHOCYANINS

2.2.1 Chemical structure of flavonoids

Anthocyanins belong to a large group of polyphenolics named flavonoids, which

are secondary metabolites synthesized by higher plants. Based on the characteristic of

aglycones, flavonoids are divided into sub-classes (Figure 2.1). Their aglycones share a

C-6 (A-ring)-C-3 (C-ring)-C-6 (B-ring) carbon skeleton (20). The presence and absence of the double bond and carbonyl group on the C ring are the major differences among sub-classes, whereas a shift of B ring substitution from C-2 to C-3 position separates isoflavones from others. Because quercetin, catechin, and isoflavones have similar structure to anthocyanins, their extensively studied bioactivities can provide implication for the evaluation of anthocyanins.

4 OH OH OH OH OH OH

+ HO O HO O HO O

OH OH OH OH O OH OH quercetin catechin cyanidin (flavonol) (flavanol) (anthocyanidin) OH OCH3 OH HO O HO O OH HO O

OH O O OH OH O hesperetin luteolin daidzein () (flavone) (isoflavone)

Figure 2.1 Representative aglycone structures of the common flavonoid sub-classes.

2.2.2 Anthocyanin aglycones

Due to the long of eight conjugated double bonds carrying a positive charge (Figure 2.2), anthocyanins are intensely colored under acidic conditions.

The maximum absorption in the visible range is usually between 465 and 550 nm, while the other maximum absorption band falls in the UV range between 270 and 280 nm (21).

Differing in the number of hydroxyl and methoxyl groups, 6 common anthocyanidins

(aglycones) exist in nature - cyanidin (Cy), peonidin (Pn), pelargonidin (Pg), malvidin

(Mv), (Dp), and petunidin (Pt) (21, 22). The color varies among aglycones

(Table 2.1) with the B ring possessing more hydroxyl groups being more blue and more methoxyl groups being more red (17, 23).

5 R1 OH Chromophore 3' 2' 4' B + 1' OH O 6' 5' 8 1 7 2 R2 A C 3 6 5 4 OH

OH

Figure 2.2 Basic chemical structure of a common anthocyanidin.

Substitution λ (nm) in HCl Name Color max acidified MeOH R1 R2

Cy OH H magenta 535

Pn OCH3 H magenta 532

Pg H H red 520

Mv OCH3 OCH3 purple 542

Dp OH OH purple 546

Pt OCH3 OH purple 543

Table 2.1 Differences on chemical structure, color and λmax of anthocyanidins most commonly found in nature.

6 2.2.3 Glycosylation and acylation

The hydroxyl groups on the aglycone may be substituted by carbohydrate

moieties and the carbohydrate moieties may further be linked to other through

glycosidic bond or acylated with organic acids (cinnamic acid, melonic acid, and acetic

acid, to name a few) through ester bond (Figure 2.3). Both glycosylation and acylation

affect the physical-chemical properties of anthocyanins in that they modify the molecular

size and polarity. Glycosylation increases water and acylation decreases water

solubility. Perhaps due to increased molecular size, acylated anthocyanins are much less

efficiently absorbed than their counterparts without the acylation (24, 25). The aglycone form of anthocyanins is rarely found in nature due to the poor stability. Glycosylation improves anthocyanin stability by forming intramolecular H-bonding network within the anthocyanin molecule (26). Glucose (glu) and rhamnose (rha) are common moieties attached to the aglycone, but galactose (gal), arabinose (ara), xylose (xyl), (rut), sambubiose (sam) and other sugars are also frequently found. Acylated organic acids comprise a broad range of compounds as well, which are normally classified into two categorizes, aliphatic acids and cinnamic acids. Due to the various types of substitutions, many varieties of anthocyanin derivatives exist.

7 OCH3 OH

HO O+ OCH3 O

O O O

OH HO OH OH OH

Figure 2.3 Chemical structure of an acylated anthocyanin (Mv-3-(p-coumaryl)glu) in grape skin.

2.2.4 The influence of pH on anthocyanin chemical structure

Anthocyanins are unique among flavonoids as their structures reversibly undergo pH-dependent transformation in aqueous solution (Figure 2.4). Four major anthocyanin forms exist in equilibria: the red flavylium cation; the blue quinonoidal base; the colorless carbinol pseudobase; and, the colorless chalcone (27). At pH below 2, anthocyanins exist predominantly in the red flavylium cation form. Rapid hydration of the flavylium cation occurs at the C-2 position to generate the colorless carbinol pseudobase at pH values ranging from 3 to 6. As red color is bleached out in this transformation, the mechanism of reaction has been extensively investigated.

8 R1 R1 R1 OH OH O

Neutral and O O HO O OH O slightly R2 R2 R2 acidic pH R3 R3 R3 OH O OH Neutral quinonoidal bases (purple to violet color)

+ + + +H -H +H -H+

-H+ +H+ R1 OH

HO O+ Flavium cation R2 pH < 2 (red to orange color) R3 OH +H+ R1; R2 = H, OH, or OCH3 - H2O Kh R3 = O-glycoside -H+ + H2O R1 R1 OH OH OH OH O HO O HO R pH from 2 R2

3 to 6 R3 R3 OH OH Carbinol pseudobase Chalcone (colorless) (colorless)

Figure 2.4 Scheme of the pH-dependent structural interconversion between dominant forms of mono-glycosylated anthocyanins in aqueous phase. (Source: Houbiers, Lima et al. 1998)

9 The fundamental work conducted by Brouillard and Dubois (28) demonstrated

that the hydration process is fairly rapid and, depending on the extent of pH change, can

take between 30 and ~103 s to reach equilibrium, whereas the reverse transition from

carbinol pseudobase to flavylium cation is almost instant upon acidification. The pKh for

the hydration reaction has been well studied with Mv-3-glu, a major anthocyanin in grape

and using different methodologies (27, 29, 30). The reported pKh was 2.60, 2.80,

or 1.76 using UV/Vis spectroscopy, 1H NMR spectroscopy, or electrophoresis

respectively. It is noteworthy that under the same conditions the 3, 5-di-glu has less

proportion in cation form than the corresponding 3-mono-glu, while acylation leads to

noticeably increased cations especially at above pH 4 (31). For example, larger number of acylated cinnamic acids results in higher pKh, and thereby more red color is retained at

low acidic conditions. This characteristic of acylated anthocyanins makes them preferable

food colorant in moderately acidic foods like yogurt.

The carbinol form can further equilibrate to an open ring form, the colorless

chalcone pseudobase, at a slow rate, and the reaction is favored by increased temperature

(27). However, at any pH condition the chalcone form exists in a much smaller

proportion as compared to the carbinol form. Unlike the carbinol form, reconversion of chalcone to flavylium cation is a very slow process taking hours to reach completion (32).

Deprotonation of flavylium cation to generate quinonoidal base occurs at slightly acidic to neutral condition and the reaction is extremely fast (28). At such a condition, kinetic competition between the deprotonation and hydration reactions predominantly favors

10 deprotonation. As the pH increase to above 8, the quinonoidal base can be ionized to carry one or two negative charges (30, 33).

2.2.5 Pyranoanthocyanins

Pyranoanthocyanins are a group of color-stable anthocyanins first identified in

aged red wines (34-36). The characteristic structure of pyranoanthocyanins is an

additional pyran ring linking C-4 position and C-5 OH group on the backbone (Figure

2.5). This pyrano structure stabilizes the parent anthocyanin’s color markedly in that the

flavylium cation form become more resistant to pH increase and SO2 bleaching (34). It has been proposed that the new ring protects pyranoanthocyanins from nucleophilic attack by water, delaying the hydrolysis of flavylium cations at low acid condition. The

resistance to SO2, a compound commonly used in winery as antioxidant and

antimicrobial agent, is attributed to the occupied C-4 position. A free C-4 position can rapidly form covalent bond with hydrogen sulfite anion and result in rupture of the

conjugated system in the chromophore.

11 R1 OH

HO O+ R2

O O OH

O HO

OH OH R3

Figure 2.5 A representative chemical structure of pyranoanthocyanins. R3 is usually a hydrogen, a carbonyl group, a phenol, or a flavonoid molecule (Source: Vivar-Quintana, Santos-Buelga et al. 2002; Alcalde-Eon, Escribano-Bailon et al. 2004).

It is known that anthocyanins react with pyruvic acid, acetaldehyde, acetone, 4-

vinylphenol, vinylflavanols, or cinnamic acids to generate different classes of

pyranoanthocyanins (35, 37, 38). Nucleophilic attack on the C-4 position of anthocyanin

aglycone is the initial step of reaction, and subsequent cyclic reaction requires a free OH

group on the C-5 position. Therefore 3, 5-diglycosylated anthocyanins are not substrate in

this type of reaction. The generation of pyranoantohcyanins requires weeks for

completion during aging, but can be significantly accelerated by increased

temperature. Schwarz and Winterhalter (37) found that the reaction between Mv-3-glu

and p-dimethylamino cinnamic acid required 6 wk to complete at 15°C, but only 3 days

at 35°C. However, high temperature may also cause more browning via condensation

12 reaction. Appropriate percentage of ethanol in the solvent also accelerates generation as ethanol improves solubility of organic substrates.

However, high concentrations of ethanol deter covalent binding, leading to slower reactions (39). Fermentation during wine making was previously thought to be necessary

for the generation of pyranoanthocyanins, but later it was discovered that juices naturally

containing appropriate precursors could also produce pyranoanthocyanins during storage

(40). For the purpose of enhanced color stability, artificially added cinnamic acids had

been evaluated in some juices and pyranoanthocyanins were successfully produced

(41). Synthesis of novel pyranoanthocyanins with to blue hues has also been

investigated in the attempt to provide stable and novel pigments for application in food

industry (37, 42).

2.3 ANTHOCYANINS IN HUMAN DIET

2.3.1 Occurrence of anthocyanins in plant materials

Anthocyanins are water-soluble vacuolar pigments found in many plant tissues

(43). Edible anthocyanin sources in nature include colored fruits such as , ,

, , and as well as many dark colored vegetables such

as black currant, red , red , black bean, , , ,

and purple sweet (21, 44). Although most commonly accumulated in flowers and

fruits, they are also present in , stems and storage organs (17). Total anthocyanin

content varies substantially across plant species or even (44). Available data

show very wide range of anthocyanin content in plant material with berries usually

13 providing the most anthocyanins per serving. Environmental factors such as light, temperature, and altitude also affect anthocyanin concentration considerably (43).

Abundance of the six common anthocyanidins in the edible parts of plants varies

considerably. Some commodities like contain a limited number of

anthocyanin pigments, whereas others like low bush blueberry may contain a complex mixture. In a previous review Kong et al. (22) estimated the following abundance order:

Cy (50%), Pg (12%), Pn (12%), Dp (12%), pt (7%), and Mv (7%). In a later published

summary an updated evaluation was probably more accurate as more anthocyanins were

included (16). The abundance order was estimated to be Cy (30%), Dp (22%), Pg (18%),

Pn (7.5%), Mv (7.5%), and Pt (5%). In both reports the three non-methylated

anthocyanidins (Cy, Dp and Pg) were shown to be more wide spread than the three

methylated anthocyanins (Pn, Mv and Pt). Considering that glucoside sugar moiety

occurs in over 90% of the identified anthocyanins (16), it is not surprising that Cy-3-glu

is the most widespread anthocyanin in nature (22).

2.3.2 Anthocyanins in foods and beverages

Dietary anthocyanin sources include many colored fruits and vegetables, as well

as fruit based processed foods and beverages like jelly, juices, and red wine. The global anthocyanin consumption from black grapes alone was estimated to be 10,000 tons annually (45). With regard to mass consumed, anthocyanins constitute perhaps the most important sub-class of flavonoids. Daily intake of anthocyanins had previously been overestimated to be 185-215 mg/day/person in the US, and according to a recent report by USDA evaluating more than 100 common foods the more accurate estimation is 12.5

14 mg/day/person (44). Still this is a significant number comparing to other with known or proposed health-promoting benefits. It has to be noted that dietary habits and choices have great impact on anthocyanin consumption. For example, one serving of blueberry increases anthocyanin consumption to greater than 500 mg. Likewise one serving of provides about 200 mg, and one serving of elderberry can supply 2000 mg anthocyanins. Regular red wine drinkers or juice drinkers can also benefit more from anthocyanins as one bottle can readily provide more than 200 mg of anthocyanins (45). Nowadays as the consumers are increasingly concerned about the adverse health effect of synthetic food dyes, more and more food manufacturers are attempting to use anthocyanins as substitutes for FD&C red #40, the most widely used synthetic colorant. For instance, in fruit yogurt and many types of fruit flavored dry mix anthocyanins are gradually gaining application. Indeed synthetic dyes are not allowed in the rapidly growing whole foods market, where anthocyanins become highly important.

Acylated anthocyanins are usually used as food colorants due to superior stability than non-acylated anthocyanins (46). However, certain commodities such as elderberry, barberry, and purple corn can provide exceedingly high level of non-acylated anthocyanins at relatively low cost, thus they also have potential use in the food industry

(47, 48).

2.3.3 Toxicity of anthocyanins

Animals and human have consumed anthocyanins since ancient time. No adverse impact on health has been reported with oral consumption of anthocyanins in foods (49).

The use of natural anthocyanins as additives in foods and beverages is widely permitted

15 within Europe (E163), Japan, the United States, and many other countries (21). Based on

early toxicological studies including mutagenicity, reproductive toxicity, teratogenicity, as well as acute and short-term toxicity evaluations, the Joint FAO/WHO Expert

Committee on Food Additives (JECFA) concluded that anthocyanin-containing extracts had a very low toxicity (50). The no-observed-effect-level (NOEL) for young rats was

determined to be approximately 225 mg/kg body weight in a two-generation reproduction

study. Based on the above result, the estimated acceptable daily intake (ADI) for human

was estimated to be 2.5 mg/kg body weight per day in 1982, using the equation of ADI =

NOEL/100 (45).

2.4 STABILITY OF ANTHOCYANINS

2.4.1 The effect of pH

At pH lower than 3, anthocyanins exist primarily in the red colored flavylium

cation form, which is the most stable form of anthocyanins in the presence of oxygen (32).

As pH increases above the pKh, the majority of anthocyanin molecules exist in the

hemiacetal (chalcone) form. As pointed out by Brouillard, the hemiacetal form is most

likely the target for attacking at slightly acidic pH (28). This speculation was supported by a later discovery of anthocyanin-quinone adducts in the hemiacetal form using LC-MS

(51). A systematic study of the two anthocyanins, Cy-3-glu and Pt-3-(p-coumaryl)rha,5- glu, revealed that pH is a critical factor for anthocyanin stability (52). Cy-3-glu stability slightly decreased from pH 1 to 3, and plummeted at pH above 4 at 23°C in the dark, whereas Pt-3-(p-coumaryl)rha,5-glu stability dramatically decreased above pH 5. At

16 alkaline pH anthocyanins exist in the quinonoidal bases form, which participate more readily in chemical reactions than the flavylium cations. The o-quinones are also known for their tendency to polymerize (53).

2.4.2 The effect of temperature

A logarithmic relationship has been established between temperature and

anthocyanin degradation rate (53). Possible degradation mechanism is that higher

temperature favors transition to the unstable chalcone form (Figure 2.4), while the opened C ring of chalcone is further degraded to brown products (32). Therefore, refrigeration was recommended as an efficient means of preserving anthocyanins. When stored at 4°C for 9 wk, approximately 40% of anthocyanins in juice (pH 2.0) werelost, while no detectable level was found at 37°C (53). The reported half life for grape pomace extract color in carbonated beverage was 1536 days at 3.5°C as compared to 769 days at 10°C, 416 days at 20°C, and 80 days at 38°C (54). Thermal processing procedures like canning and hot fill adversely affect anthocyanin color. Strawberry pigment at 100°C only has a half-life of 1 h. But a short time high temperature treatment may improve anthocyanin stability by facilitating inactivation of native enzymes that are detrimental to anthocyanins. Heat aided drying processes decompose anthocyanins considerably, not only because of the elevated temperature, but also because the drying

process increases the concentration of reactants in the solution. In contrast, the freeze

drying method commonly employed in high quality food preparation has the advantage

that thermal sensitive phytochemicals, including anthocyanins, can be well preserved during the drying process.

17 2.4.3 The effect of light

Light is another deteriorating factor with respect to anthocyanin stability. Due to

the intrinsic light absorption property, anthocyanin pigments absorb energy from photons

efficiently. In daylight versus dark storage, the half-life of grape pomace color extract at

20°C can be shortened by more than 50% (54). Therefore, it is recommended that

anthocyanin colored foods and beverages are stored in the dark. Opaque package material

helps to preserve anthocyanins owing to blocked incoming light. Similarly, acylated

anthocyanins are more stable than non-acylated anthocyanins when exposed to light

under low acid and high temperature conditions (32). Other factors such as the second

sugar moiety on the C-5 position and co-existence of other polyhydroxylated flavonoids in the solution also stabilize anthocyanin molecules. Plausible mechanisms of such stabilizing effects include the established H-bonding network and intramolecular/intermolecular co-pigmentation.

2.4.4 The effect of enzymes

Native plant enzymes, particularly polyphenol oxidase (PPO) and proxidase

(POD), are usually released from cell compartments during anthocyanin extraction,

accelerating anthocyanin degradation. Enzymatic browning along with color loss can

occur very fast after crushing. Studies have shown that PPO does not metabolize

anthocyanins directly, but instead generate quinones from phenolic acids that

subsequently react with anthocyanins through condensation or coupled oxidation (51, 55).

In the presence of H2O2, POD oxidize phenolic acids to their corresponding quinones,

18 which polymerize with anthocyanins to increase color degradation and browning (56).

Both the flavylium cation form and the hydrated forms can form adduct with quinones, although the hydrated forms are more reactive. Prompt inactivation of PPO by means of increased extraction temperature and addition of SO2 efficiently preserves anthocyanins during extraction (57).

Glycosidases are another group of enzymes that affect anthocyanin stability.

Fungal enzymes are widely used to facilitate macerating and pressing of fruits to increase juice yield by breaking cell structure. However crude enzyme extract usually contains β- glycosidases, which are capable of hydrolyzing the glycosidic linkage between anthocyanidins and sugar moieties (58-60). As mentioned above, anthocyanidins lacking the sugar moiety are extremely unstable and decompose spontaneously. Therefore, caution must be taken when choosing an enzyme preparation for use in the food industry

(60). In addition to the fungal β-glycosidases, endogenous β-glycosidases in the animal digestive tract may also have impact on anthocyanin stability and consequently affect the availability of anthocyanins in vivo.

2.4.5 The influence of oxygen, peroxide and ascorbic acid

Oxygen has been well known to accelerate anthocyanin decomposition. As the ultimate electron accepter, oxygen participates directly or indirectly in multiple mechanisms of anthocyanin degradation, including H2O2 bleaching. Bottled juice without head space oxygen can have four times higher anthocyanin concentration remained than that with 2 mL of headspace oxygen after 32 wk of storage (32). In an

19 ascorbic acid-anthocyanin-flavanol model system, bottled samples of purified Pg-3-glu

stored in a nitrogen atmosphere was preserved as compared to samples stored in an

oxygen atmosphere (61). H2O2, the most commonly used packaging sterilant in the food

industry, is a particularly reactive form of oxygen causing anthocyanin degradation.

Residual amount of H2O2 is likely to be present in anthocyanin rich product. Millimolar

levels of H2O2 still accelerate sour and anthocyanin degradation in a concentration dependent manner (62).

The effect of ascorbic acid is complex in that it is matrix dependent. In the

presence of H2O2, ascorbic acid at both 60 and 80 mg/L concentrations protected

anthocyanins in pomegranate juice, whereas 80 mg/L of ascorbic acid dramatically

accelerated anthocyanin degradation in sour cherry juice (62). In the ascorbic acid-

anthocyanin-flavanol model system mentioned above, addition of approximately 0.42

mg/mL of ascorbic acid adversely affected anthocyanin stability compared to addition of

0.64 mg/mL of catechin or sparging the head space with oxygen (61). Two mechanisms

have been proposed for the adverse effect of ascorbic acid, i.e. 1) oxidation of ascorbic

acid generates H2O2 that oxidizes anthocyanins and 2) ascorbic acid directly reacts with

anthocyanins causing condensation. The investigation of dehydroascorbic acid (the

oxidized form of ascorbic acid) level throughout the storage period suggested that direct

condensation was likely to be the predominant mechanism.

20 2.4.6 Self-association, acylation and co-pigmentation

It has been established that anthocyanins at high concentration do not follow

Beer’s Law due to self-association effect (32). For example, a 100-fold increase of Cy-

3,5-di-glu concentration from 0.1 mM to 10 mM at pH 3.16 could result in a 300-fold increase in absorbance. The reason lies in that when the anthocyanin molecules are close to each other at high concentration, they associate with each other creating a stacking effect that enhances absorbance. Generally anthocyanin color is expected to undergo a second order increase in response to greater concentration (39). The stacking effect was believed to protect the anthocyanidins from the nucleophilic attack of the water molecule or other reactants

Another two types of stacking effect that occur with anthocyanins are intramolecular co-pigmentation and intermolecular co-pigmentation. Intramolecular co- pigmentation occurs when the acyl group in acylated anthocyanins interacts with the aglycone. Acylation, particularly with cinnamic acids, will render a bathochromic shift

(from more reddish hue to more bluish) and greater stability to the parent anthocyanins.

Intramolecular has been proposed to be responsible for this stabilization effect (46). It is suggested that acids will fold over the anthocyanidin aglycone avoiding formation of the hydrated species (17), and there is evidence of proximity between hydrogens from acids and the C-4 position (18, 63). Presence of two or more acyl groups further increases the color stability of anthocyanins in aqueous solution (64). , purple , red cabbage, and purple extracts are being commercialized as colorants in the food industry due to their increased stability attributed to the presence of

21 two or more acyl groups. Maraschino cherries colored with radish anthocyanins have a

shelf life of at least 6 months at 25 °C, and the excellent stability of the major pigements

has been associated with the presence of two acyl groups on the pelargonidin derivatives

(65, 66).

Anthocyanins also interact with other phenols and flavonoids to produce a

hyperchromic effect (increase in color intensity) and a bathochromic effect. Such a phenomenon is called intermolecular co-pigmentation, which can occur in acidic, neutral

and even slightly alkaline aqueous solution (18, 67). The occurrence of intermolecular

co-pigmentation relies on at least two effects (68). First, the formation of the π- π complex causes changes in the spectral properties of the molecules in the flavylium form, resulting in hyperchromic shift and bathochromic shift (69). Second, the stabilization of the flavylium form by the π complex shifts the equilibrium to better favor the flavylium, thus boosting the proportion of anthocyanin molecules in the red-colored form.

2.5 PUTATIVE HEALTH-PROMOTING EFFECTS OF ANTHOCYANINS

Interests in dietary polyphenols including anthocyanins drastically intensified

after the recognition of their potential health benefits (70). Epidemiological studies have

suggested a reverse association between high consumption of polyphenols and incidence

of some chronic diseases. For example, drinking red wine regularly has been associated

with the relatively low incidence of coronary heart disease in French people despite a

high fat diet, the well known “French Paradox” (71). Since then a vast number of studies have been carried out on the biological effects of polyphenols using in vitro and in vivo

22 models. Anthocyanins are among the most abundant polyphenols in fruits and vegetables and possess potent antioxidant activity. In vitro models have the merits of low cost and high efficiency, thus they have been widely employed. Animal and human clinical studies on health benefits of anthocyanins are still in the early stage. To date, the suggested health benefits of anthocyanins are more or less related to their antioxidant activity (22).

It has to be noted that not a single class of compounds may explain most of the health promoting effects of consuming fruits and vegetables. Apparently the phytochemicals contained in fruits and vegetables work collaboratively to benefit our body (72, 73). .

2.5.1 Relief of oxidative stress

Reactive oxygen species (ROS) including free radicals, singlet oxygen and peroxides are generated in the body. They are important to the immune system, cell signaling, and many other normal body functions. However, if ROS are overly produced they can elicit cellular damage, leading to degenerative diseases such as inflammation, cardiovascular disease, cancer, and aging (74).

Anthocyanins are potent in vitro. They effectively quench free radicals and terminate the chain reaction that is responsible for the oxidative damage.

Because pH in the human body is generally neutral except in the stomach, the antioxidant activity of anthocyanins at neutral pH is of particular importance. Using a widely accepted antioxidant assessment method, the oxygen radical absorbance capacity (ORAC) assay, antioxidant activity of 14 anthocyanins including Dp, Cy, Pg, Mv, Pn and their glycosylated derivatives was determined in aqueous phase at neutral pH (75). Among

23 these anthocyanins, Cy-3-glu had the highest ORAC value, 3.5 times as potent as Trolox

(water soluble vitamin E analogue). Pg had the lowest ORAC value of tested

anthocyanins, but still as potent as Trolox. As compared to the natural form of vitamin E

(α-tocopherol), Cy-3-glu and its aglycone Cy were shown to have similar antioxidant

potency in linoleic acid autoxidation, liposome, rabbit erythrocyte membrane, and rat

liver microsomal systems (76). Such potent antioxidant activity from anthocyanins may

have protective effects in the biological environment. An in vitro study using human

erythrocytes treated with H2O2 as an oxidative model revealed that red wine fractions rich in anthocyanins significantly lowered ROS in human red blood cells (77).

Protective effect of anthocyanins on oxidative stress induced damage is promising

as shown using in vivo models. In a rat study utilizing hepatic ischemia-reperfusion as an

oxidative stress model, Cy-3-glu efficiently attenuated changes of biomarkers in liver

injury (78). In another rat study the animals were fed vitamin E deficient diets for 12 wk

followed by supplementation with purified anthocyanin-rich extracts. The anthocyanin

diet significantly improved plasma antioxidant capacity and lowered the level of

hydroperoxides and 8-Oxo-deoxyguanosine, indicating significant reductions of the

vitamin E deficiency induced peroxidation and DNA damage, respectively (79).

2.5.2 Prevention of cardiovascular diseases

Oxidation of low-density lipoprotein (LDL) triggers accumulation of macrophage

white blood cells in the artery wall. Rapture of the plaque deposits oxidized cholesterol

into the artery wall, leading to atherosclerosis and eventually cardiovascular diseases (80,

24 81). Dietary antioxidants including anthocyanins have the potential to increase serum

antioxidant capacity and thereby protect against LDL oxidation and prevent

cardiovascular diseases. Research has initially focused on anthocyanin-rich red wine

because of the famous “French paradox” (71). Using a chemiluminescent assay of serum

antioxidant capacity (SAOC), the effects in normal human subjects ingesting 300 mL of

red wine, , or high dose (1000 mg) of vitamin C were studied. In subjects who

ingested red wine the mean SAOC was increased by 18% and 11% after 1 and 2 h, both

higher than that in the white wine group although not as high as that in the vitamin C

group (82).

Following the pioneering studies on red wines, more and more attention had been

given to anthocyanins and other polyphenols present in red wines. A trial involving 7

human subjects demonstrated that daily consumption of 125 mL of concentrated red

grape juice markedly raised serum total antioxidant capacity (TAC) as compared to the baseline. The susceptibility of LDL to oxidation was also reduced. Therefore the nonalcoholic red grape extract was suggested to have similar beneficial effects to red wine (83). Other anthocyanin-rich foods were also extensively studied. Monitoring of the a chemiluminescenct emission intensity of human blood plasma for 8 h following oral administration of black currant anthocyanins demonstrated a rapid increase of plasma antioxidant capacity until 2 h (84). Spray-dried elderberry juice containing high anthocyanin content was investigated with respect to the protective effect on human LDL in vitro (85). A concentration-dependent prolongation of the lag phase was observed in

copper-induced oxidation. Meanwhile a similar prolongation effect was also observed in

25 peroxyl-radical-driven LDL oxidation together with a reduction of maximum oxidation

rate. It is important to note that anthocyanins may not explain all of the protective effects observed in these foods, but likely contributed to some extent. In a UV light radiation

induced lipid peroxidation model, three purified anthocyanins (Pg-3-glu, Cy-3-glu, and

Dp-3-glu), as well as their aglycones all demonstrated strong inhibition of lipid

peroxidation and acted as active oxygen radical scavenging agents (86).

2.5.3 Anti-inflammatory activity

Inflammation is a complex biological response in response to tissue injury. Many

occur at sites of inflammation because inflammatory cells provide

microenvironment favorable for tumor development (87). Anti-inflammatory therapy has

the potential to prevent early neoplastic progression and malignant conversion. Because

cyclooxygenases (COX) convert arachidonic acid to prostaglandins that stimulate

inflammation, inhibitory effect on COX enzymes, especially selective inhibition on one

of the isoenzymes COX-2, is highly desirable. Cy aglycone was reported to possess better

anti-inflammatory activity than the positive control aspirin in the COX activities assays

(88). Purified anthocyanin fractions from tart cherries, sweet cherries, ,

blackberries, , , elderberries, , and were

evaluated using cyclooxygenase-inhibitory assays (89). All the anthocyanin fractions

demonstrated inhibitory effect on COX-1 and COX-2 enzymes, while strawberry,

and raspberry showed the highest activity, comparable to that of the positive

controls ibuprofen and naproxen at 10 μM concentrations. In an in vivo study the

therapeutic efficacy of blackberry anthocyanins (Cy-3-glu accounts for 80%) was

26 investigated in rats with carrageenan induced lung inflammation (90). All parameters of

inflammation were effectively reduced in a dose-dependent manner by anthocyanins.

2.5.4 Anti-carcinogenic activity

Anti-cancer activity of anthocyanins has been established largely based on in vitro

evidence. Anthocyanins extracted from flower petals were found to be more potent than

combined non-anthocyanin flavonoids regarding cell growth inhibition in a human

malignant intestinal carcinoma derived HCT-15 cell line (91). Dose required for 50%

inhibition ranged from 0.5 to 5 µg/mL for representative individual anthocyanins and

anthocyanidins, whereas higher concentrations of other flavonoid evaluated in this study

were required to exhibit the same effect. Similarly, anthocyanin fraction isolated from red wine was also discovered to be significantly more effective than non-anthocyanin flavonoids in red wine or white wine using HCT-15 cell line and AGS cell line, which was derived from human gastric cancer (92). The anti-proliferative effect of anthocyanin fraction from 4 cultivars of muscadine grapes were evaluated using two human colon cancer derived cell liens, HT-29 and Caco-2 (93). In all cultivars and both cell lines, greater inhibitory activity was observed from the anthocyanin fraction than from the phenolic acids fraction or the crude extract. Zhao et al. (94) demonstrated that anthocyanin fractions from commercially available , chokeberry, and grape extracts all exerted anti-proliferative effect in the HT-29 cell line. Dose of 25 µg/mL chokeberry anthocyanins provided 50% inhibition but not affecting the growth of normal colonic NCM460 cells. More in-depth investigation revealed that the chokeberry anthocyanins arrested the cell cycle of HT-29 cells by blocking at the G1/G0 and G2/M

27 phases (95). Highly purified anthocyanins had also been evaluated. Four anthocyanins

isolated from strawberry by means of medium-pressure liquid chromatography (MPLC)

were all shown to reduced cell viability of human oral (CAL-27, KB), colon (HT29,

HCT-116), and prostate (LNCaP, DU145) cancer cells at 100 μg/mL dose level, although

different sensitivity was recorded for each individual compound (72).

Anthocyanins are shown to be promising phytochemicals responsible for, at least part of, the anti-cancer property of many fruits and vegetables, but it is more than likely that anthocyanins work collaboratively with other phytochemicals to help the body defense. Seeram et al. (73) evaluated the antiproliferative effects of total cranberry extract versus its flavonol glycosides (gly), anthocyanins, , and organic acids fractions using human oral (KB, CAL27), colon (HT-29, HCT116, SW480, SW620), and prostate (RWPE-1, RWPE-2, 22Rv1) cancer cell lines. Both the anthocyanin fraction and the fraction exhibited substantial inhibitory effect on all but the SW480 cell lines. However, the combination of these two fractions was the most active against all cell lines.

In animal studies the growing body of data has demonstrated chemopreventive effect of anthocyanins in multiple types of cancers. Nevertheless, the observed preventive effects were primarily limited to the GIT related organs including the oral cavity, the esophagus (96, 97), and the colon (8, 98-100). In the GIT lumen, anthocyanins are largely available and can contact directly with the epithelial layer (13). In contrast, strawberry anthocyanins failed to inhibit 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone-and benzo[a]pyrene-induced lung cancer in a mice model (101).

28 Anti-cancer activity of anthocyanins may be attributed to the collaborative effect of multiple mechanisms (8, 102, 103). Possible mechanisms that have been suggested include antimutagenic activity (104-106), inhibition of oxidative DNA damage (107), inhibition of carcinogen activation and induction of phase II enzymes for detoxification

(108, 109), cell cycle arrest (110), inhibition of COX-2 enzymes, induction of apoptosis

(93, 111), and anti-angiogenesis (112). In the particular case of GIT related cancer, the influence of anthocyanins on the GIT luminal condition is of great importance too. Bruce et al. (113) suggested two mechanisms that initiate colon cancer development, one involving a local irritation that produces a local inflammatory response, and the other relating to an electrolyte imbalance. Both mechanisms result from a defect in the epithelial barrier and both lead to elevated ROS and COX-2 levels in epithelial cells.

Therefore, agents that can improve colon luminal condition hence reduce epithelial barrier damage, can inhibit expression of COX-2 and inflammation, or can quench ROS in local cells have the potential to prevent colon cancer. Anthocyanins have been shown to be powerful antioxidants and selective COX-2 inhibitor, as discussed previously in this section. However, Lala et al. (8) suggested that the inhibitory effect of dietary anthocyanins in a colon carcinogen azoxymethane (AOM) induced rat colon cancer model was primarily attributed to the direct effect on improving colon luminal condition.

The patterns of inhibition on colonic cell proliferation and large aberrant crypt foci (ACF)

multiplicity (Figure 2.6) were not correlated with the total antioxidant capacity in the diet,

with anthocyanin absorption as accessed by urinary anthocyanin concentration, or with

the colonic mucosa COX-2 mRNA levels. The highest correlation was between colon

cancer growth and the total anthocyanin content in the colonic lumen as represented by

29 fecal anthocyanin concentration. Luminal anthocyanins appeared to promote fecal moisture content and excretion of bile acids, a group of endogenous tumor-promoting compounds (Figure 2.6). In addition, luminal anthocyanins may benefit the colon health by protecting the epithelial cells against oxidative damage and microbial infection.

30 70 52.5 a control chokeberry

60 bilberry 44.6 a grape 49.3 40 48.9 a a a 50 45.1 b

39.2 c 33.2 40 b

30 23.6 22.1 a a 17.3 b 16 15.3 15 b b 20 a 12.1 b 11 a 7.7 6 a 10 b 4 2.66 b 1.67 b b 1.35 b ND ND 0 Urinary Colonic cell Large ACF Fecal anthocyanin Fecal moisture Fecal bile acids anthocyanin proliferation index multiplicity (nmol/100g wet content (%) Concentration (nmol/L) feces) (μmol/g dry feces)

Figure 2.6 Effect of feeding anthocyanin rich diets on total anthocyanin concentration in rat urine, colonic cell proliferation index, large ACF multiplicity, total anthocyanin concentration in feces, fecal moisture content, and fecal bile acids concentration (Source: Lala, Malik et al. 2006). Data are means ± SE. Values with a different letter differ significantly (P<0.05) within a same category.

31 2.5.5 Prevention of obesity

Obesity is the result of accumulated excessive adipose tissue caused by the imbalance of energy intake and expenditure. It is usually associated with various metabolic disorders. Consumption of anthocyanins can possibly ameliorate the function of adipocytes, thus may prevent metabolic syndrome and obesity (114). In a fundamental study conducted by Tsuda et al. (115), 24 male mice were fed control, purple corn color, high-fat, or high-fat plus purple corn extract diet for 12 wk. Supplementation with purple corn color suppressed the high-fat diet induced gain of body weight and white/brown adipose tissue weights. Down-regulation of the mRNA levels of enzymes involved in fatty acid and triacylglycerol synthesis was suggested to contribute to this anti-obesity effect. Two additional in vivo studies supported anthocyanin’s anti-obesity effect on high- fat diet. In one of the studies black anthocyanins were found to effectively reverse the weight gain of high-fat diet group rats to the same as that in the control group

(116). Serum composition was also improved by the addition of black soybean anthocyanins into high-fat diet. Serum triglyceride and cholesterol levels were significantly reduced, whereas the high-density lipoprotein (HDL)-cholesterol concentration markedly increased. In the second study male mice were fed high-fat diet for 8 wk with or without supplementation of blueberry anthocyanins in drinking water

(117). Both the whole blueberry and the purified blueberry anthocyanins were evaluated.

The purified anthocyanins resulted in lower body weight gains and body fat than the controls while the whole blueberry with the same level of anthocyanins actually increased obesity, probably due to added calorie intake from sugar. A further study of

32 anthocyanin’s effect on gene expression of adipocytes employed an in vitro model using

isolated rat adipocytes (118). The total RNA isolated from the adipocytes was analyzed using GeneChip microarray. After the treatment of adipocytes with 100 µM of Cy-3-glu or Cy, 633 and 427 genes, respectively, were up-regulated by greater than1.5-fold. Based on the gene expression profile, the up-regulation of hormone sensitive lipase and enhancement of the lipolytic activity were suggested to be the result of anthocyanin treatment on adipocytes.

2.5.6 Control of diabetes

Type-2 diabetes is a metabolic disorder associated with insulin resistance. Insulin

secreted from the β-cells of the pancreas is responsible for stimulation of blood glucose

transportation into skeletal muscle and adipose tissue as well as suppression of hepatic

glucose production (119). Obesity and excessive intake of high fat or high glycemic

index foods are possible reasons for the relative inadequacy of insulin in type-2 diabetes.

Anthocyanins have the potential to control obesity, and consequently may contribute to

the prevention of type-2 diabetes. Furthermore, the antioxidant activity of anthocyanins

may protect β-cells from glucose induced oxidative stress (120). Sugimoto et al. (121)

examined the protective effects of boysenberry anthocyanins against oxidative stress in

streptozotocin induced diabetic rats. Increased plasma and liver biomarkers oxidation was

observed in diabetic rats as compared to control rats. Administration of diet with

boysenberry anthocyanins restored or tended to restore the biomarkers to the level of the

control rats. The results indicated that anthocyanins are effective in preventing the

development of in vivo oxidation that may lead to diabetes. More details about the role of

33 anthocyanins in diabetes prevention can be found in the comprehensive review by Ghosh and Konishi (2007).

2.5.7 Improvement of eye vision

Anecdotal evidence suggests that consumption of anthocyanins can improve eye vision (122). In a double-blind, placebo-controlled, crossover study with healthy human subjects, feeding black currant anthocyanin concentrate at 12.5, 20, and 50 mg/subject resulted in dose-dependent lowering of the dark adaptation threshold (123). The effect with the highest dose (50 mg/subject) was statistically significant effect (P = 0.011).

However, a systematic review of placebo-controlled trials revealed conflicting evidence in the use of anthocyanins to improve night vision (124). The negative outcomes reported may be associated with low doses tested in some trials, the different methodologies used for evaluation, the variation of subjects, and the source of anthocyanins (119). Recently, a study on blueberry anthocyanin distribution in pig tissues confirmed that anthocyanins accumulated in pig eyes after feeding blueberry diet for 4 wk (125). Although the detected concentrations in eye tissue were extremely low (pmol/g), the concentrations were comparable to that in other evaluated tissues including liver.

2.5.8 Antimicrobial activity

Plant phenolics are well known to play an important role in defence against pathogens. Thus, their effects on human intestinal bacteria, both beneficial and pathogenic, have been extensively investigated (126). In a study of the phenolic compounds in 8 common Finnish berries, the berry extracts, as well as the representative

34 individual phenolic compounds contained in the berries, were evaluated against human intestinal bacteria (127). All four anthocyanins tested including Pg chloride, Cy chloride,

Dp chloride, and Cy-3-glu were found to be effective inhibitor of Gram-negative

Escherichia coli strain CM 871, a DNA repair-deficient strain, but not inhibiting regular

E. coli and the beneficial Gram-positive probiotic bacteria. Therefore the antimicrobial activity of anthocyanins was speculated to involve reactions related to DNA. In another study evaluating berry phenolics against severe human pathogens, anthocyanin fraction was the most potent phenolic fraction in berries for reducing viability of Salmonella enterica serovar Typhimurium (126). Such an effect was attributed to the ability of anthocyanins to induce the release of lipopolysaccharide molecules from the outer membrane of the Gram-negative bacteria.

2.6 BIOAVAILABILITY AND METABOLISM OF ANTHOCYANINS

To validate the prominent health-promoting effects revealed in many in vitro models it is important to consider the bioavailability in vivo. The doses reported in some in vitro studies might be inappropriate since the level of intact anthocyanins exposed to tissues (except GIT luminal side tissues) could be very limited due to the observed low concentration in blood (128). The other important issue is the form of metabolites that are generated in the tissues. Some metabolites of flavonoids have comparable to or even more potent bioactivity than the precursors (129). Therefore, to truly evaluate the bioactivity of anthocyanins, it is critical to understand their bioavailability and metabolism.

35 Regarding absorption and metabolic pathways, anthocyanins had been thought to

differ from the common flavonoids, since only intact anthocyanin glycosides were

detected in urine and plasma (130). However, improved analytical techniques have

revealed that anthocyanins are also methylated, sufated and glucuronidated (4, 131-134).

It is now believed that the absorption, metabolism and excretion of anthocyanins share

many similarities with structurally similar flavonoids. In this section both anthocyanins and several well-documented flavonoids are discussed together as a whole.

2.6.1 Overview

After consumption of flavonoid containing foods, the flavonoids are released from

the food matrix by chewing. Absorption could start at the stomach. Flavonoids absorption

by the stomach would appear in the blood extremely rapidly after ingestion (135). Small

intestine is the major site for flavonoid absorption. Endogenic β-glucosidases are

involved at this stage to release aglycones from primarily flavonoid- glu and to a lesser extent gal, xyl, and ara. Free aglycones are more hydrophobic and have smaller size than the glycosides, thus are more likely to penetrate the epithelial layer passively. In contrast, intact glycosides are also absorbed by the small intestine, either by inefficient passive diffusion or by the sodium-dependent glucose transporter (SGLT1). Acylated flavonoids are generally recognized as non-absorbable in the small intestine due to their larger molecular size and lack of a free sugar moiety for transporter binding. However, recent evidence suggests that acylated anthocyanins are slightly bioavailable as the intact form

(13, 136). Unabsorbed flavonoids travel down to the colon, where tremendous amount of microorganisms (~1012/ cm3) reside to provide catalytic and hydrolytic potential (137).

36 Glycosidic and ester bonds are cleaved by colonic microflora (138). Aglycones then

undergo spontaneous ring fission to some extent to generate simple compounds such as

phenolic acids. The released aglycones and generated phenolic acids could be absorbed

by the colon, yet only marginal absorption is expected since the colon is much less

efficient than the small intestine with respect to absorption. For this reason it is

anticipated that the sugar moiety of flavonoids glycosides governs the absorption and bioavailability of the aglycones of many flavonoids. But so far little is known about the effect of enzymatic deglycosylation on anthocyanin absorption. Within the same sub- class, the aglycones have not been reported to be a determining factor for bioavailability

(139).

Flavonoid aglycones taken up fromGIT lumen are subsequently metabolized by

phase II -metabolizing enzymes to glucuronides, sulfates and methylates in the

intestine epithelium, liver and kidney (4, 128). The conjugated metabolites may be

excreted into the jejunum via bile, and later recycled in the intestine/colon by the process

refered to as the enterohepatic circulation pathway.

2.6.2 Gastric absorption

In a study involving 32 Wistar male rats (135), isoflavone aglycones (daidzein

and genistein) and glucosides (daidzin and genistin) were administered to rats at 7.9

µmol/kg body weight. Analysis of plasma samples collected following administration showed that isoflavone aglycones were quickly absorbed (in <3 min) while glycosides were not detected within 5 minutes. This indicated that isoflavone glycosides were not

37 absorbed until reaching the small intestine. Further experiment conducted on absorption

site restricted (pylorus ligated and abdominal wall sutured) rats revealed that only

aglycones but not glycosides were absorbed in the stomach.

Nevertheless, the rule does not apply to all the flavonoids. Two research groups

used similar approaches to demonstrate that anthocyanidin glycosides were efficiently

absorbed in the stomach (140, 141). Passamonti et al. (140) injected grape anthocyanins

into the stomach of 19 Wistar male rats, which had surgically blocked cardias, and

collected blood from both the portal vein and the heart at 6 min intervals. Quantification

of the anthocyanins by HPLC-MS using single ion monitoring revealed that Mv-3-glu

was present in both portal and systemic plasma (0.789 ± 0.491 µM and 0.098 ± 0.078 µM,

respectively; n = 19). Importantly, Mv 3-glu appeared in the plasma even within 6 min

(Figure 2.7), presenting an evidence of stomach absorption. Pn-3-glu, Pt-3-glu and Mv-

3-glu-acetic were inconsistently detected, perhaps due to animal variability. Neither

anthocyanin aglycones nor conjugated derivatives were detected in the plasma.

38 1.4 Portal plasma Systemic plasma 1.2

1

0.838 0.8 0.789 0.47 0.6 0.468

0.4

0.3 0.263 0.2 0.307 0.098 Malvidin 3-glucoside in plasma (uM) 0 0 5 10 15 20 25 30 35 40 Time (min)

Figure 2.7 Mv-3-glu concentration in plasma sampled from either the portal vein or the heart of rats administered with grape anthocyanins. n= 5, 5, 4, 5 for the time points 6, 12, 24, 36 min respectively (Source: Passamonti, Vrhovsek et al. 2003).

Talavera et al. (141) infused anthocyanin standards as well as bilberry and

blackberry extract into the stomach of pylorus and sphincter ligated rats. Gastric contents

and blood were collected from the gastric vein and abdominal aorta 30 min after the

administration. HPLC analysis utilizing an internal standard PEG revealed that a high

proportion (~25%) of anthocyanin mono-glycosides, including glucoside and galactoside,

was absorbed from the stomach, whereas the rutinoside was poorly absorbed. It was

suggested that gastric absorption of anthocyanins involves bilitranslocase (TC

2.A.65.1.1), an organic anion membrane carrier in the gastric mucosa (142).

39 2.6.3 Direct absorption in the small intestine

Small intestine is generally regarded as the most important site for absorption of nutrients. Absorption of anthocyanins in rat small intestine has been evaluated using an in situ perfusion method (143). Intestinal perfusion of anthocyanin supplemented in physiological buffers was conducted on anesthetized rats. Anthocyanin amount remaining in the effluent was used to estimate the rate of anthocyanin absorption in the small intestine. Depending on structure, the absorption rate of supplemented anthocyanins ranged from 22.4 ± 2.0% (Cy-3-glu) to 10.7 ± 1.1% (Mv-3-glu). Such high absorption rates seemed to be contradict the very low levels of anthocyanins observed in the blood

(3). However, it has to be noted that these absorption rates were calculated based on the disappeared amount in the effluent, thus they could indicate the portion of anthocyanins being taken up into the small intestine tissue but not necessarily transferred into the blood.

Interestingly, the disappearance of Cy-3-glu was significantly higher than Cy-3-gal and

Cy-3-rut. Two mechanisms are likely responsible for such difference: preferable uptake or prominent degradation of anthocyanin glucoside in the small intestine lumen. The former is supported by evidence with quercetin glycosides, and the latter is probably due to the β-glucosidase activity in the small intestine, which will be discussed in the next section.

An early human pharmacokinetic study of quercetin rich foods indicated that absorption of quercetin was affected by the type of glycosides (144). Five women and four men received either quercetin-glu rich, quercetin-gal rich, or quercetin-rut rich diet for breakfast in three separated periods. Blood was collected every 6 h over 36 h, and

40 urine was collected continuously for 24 h. Total quercetin concentration was measured after being hydrolyzed to the aglycone form. Urinary quercetin equivalents indicated that the proportion excreted was significantly higher for the quercetin-glu rich diet than for the quercetin-gal rich and quercetin-rut rich diets (Figure 2.8), suggesting preferable absorption of quercetin-glu in the small intestine. The urinary data were also in agreement with the kinetics analysis of plasma quercetin. Plasma quercetin reached peak level 0.7 h after ingestion of quercetin-glu rich diet, 2.5 h after ingestion of quercetin-gal rich diet, and 9 h after ingestion of quercetin-rut rich diet. Almost no quercetin was detected in the plasma for 4 h with the rut diet. The authors suggested that quercetin-glu was absorbed efficiently in the stomach and small intestine, while quercetin-rut must be transported to the colon to be hydrolyzed by the colonic microflora before the aglycone could be absorbed.

41 2 1.39a 1.8 n=9 1.6 1.4 1.2 1 0.33b 0.8 0.44b 0.6 0.4 0.2

Proportion of excretion : intake (%) intake : excretion of Proportion 0 glucoside-rich diet galactoside-rich diet rutinoside-rich diet

Figure 2.8 The ratio of mean 24h cumulative urinary excretion: intake of quercetin (Source: Hollman, van Trijp et al. 1997). Data are means ± SD. Results with a different letter differ significantly (P<0.001).

The hypothesis was supported by Gee et al. (145) with an in situ stimulated efflux model. Jejunal sacs were obtained from male Wister rats, and then everted, ligatured at one end and tied on to syringes to load Krebs bicarbonate buffer. Subsequently, sacs were pre-loaded with 14C radio labeled galactose solution by incubation. Next, rinsed tissue was submerged into buffer bath with or without test compounds (quercetin mono- or di- glu and rut). In the presence or absence of sodium ions, the initial and cumulative efflux rates of labeled galactose were measured. The results showed potent counterflow stimulation by quercetin mono- and di-glu, as opposed to no flux of quercetin- gal and rha. Furthermore, the presence of sodium ions stimulated the rate of efflux of quercetin

42 glycosides. The data suggested that quercetin mono- and di-glu interacted with the sodium-dependent glucose/galactose transport pathway of the small intestine brush- border mucosal cells. A follow up study carried out by the same group of researchers strengthened the conclusion by demonstrating that the presence of quercetin-3-glu stimulated the efflux of preloaded galactose and competitively suppressed the uptake of labeled galactose via rat jejunum (146).

2.6.4 Deconjugation of carbohydrate moieties

Glycosylated flavonoids are more hydrophilic than the corresponding aglycones.

For instance, quercetin has a partition coefficient (log value of concentrations in octanol/water) of 1.2 ± 0.1, while quercetin-3-rut has only 0.37 ± 0.06 (70). With smaller molecular size and better lipid solubility, aglycones are anticipated to penetrate the lipid bi-layer of cell membranes, possibly leading to passive diffusion across the small intestine brush border. This pathway necessitates deglycosylation of ingested anthocyanins. Nonenzymatic deglycosylation is unlikely to play an important role, despite the acidic condition in the stomach. Deglycosylation of quercetin glycosides (145) or anthocyanidins glycosides (10) was not noted after pepsin-HCl digestion at pH 2.0 and

37°C for 2h. Therefore the question is left to the possibility of enzymatic deglycosylation in vivo.

Recent advances in the study of small intestinal β-glucosidases support the hypothesis that they deglycosylate some flavonoids, and play an important role in the digestion of dietary flavonoids. Day et al. (147) were the first to investigate the effect of

43 human β-glucosidases on flavonoids. Crude enzyme extract was obtained from homogenized intestine or liver specimens surgically removed from human patients and incubated with a broad range of flavonoids at 37°C for up to 90 min. The Km for hydrolysis of each compound was estimated. Most of the mono-glucosides, except for those glu at the 3-position on B-ring, were successfully deglycosylated by both small intestine and liver β-glucosidases, regardless of the type of aglycone (quercetin, kaempferol, , apigenin, genistein, and daidzein). In contrast, rutinosides remained intact. The results agreed with a number of previous in situ or in vivo studies specifying the absorption site difference between mono-glu and rut (144, 145, 148). The authors accentuated the importance of the glycosylation position. The glucosylation at C-

3 position clearly hampered catalytic activity. Although the small intestinal β-glucosidase slowly hydrolyzed quercetin-3-glu, hepatic β-glucosidase did not. Whether this inhibition was likely due to steric effect or other reasons was not discussed. Cytosolic β-glucosidase

(CBG) was purified from pig liver (149) and the lactase-phlorizin hydrolase (LPH; EC

3.2.1.62) from lamb small intestine brush border (150) using chromatographic and gel separation techniques. Hydrolytic activity with flavonoids confirmed the previous human tissue study. Most of the mono-glu, except 3-glu, were substrates for the pig hepatic CBG while all mono-glu were substrates for the lamb LPH. Rutinosides were not a substrate for either enzymes. Notably, the Km values of human and animals originated enzymes were different but comparable for both enzymes (Table 2.2). These studies suggested the possibility of using animal models for future research. In situ or ex vivo rat jejunum perfusion model, two research groups independently demonstrated that LPH was capable of hydrolyzing quercetin-glu efficiently, and influencing the transport of quercetin across

44 the epithelial membrane (151, 152). Both studies employed N-

butyldeoxygalactonojirimycin, a LPH inhibitor, to control the activity of LPH, in the

attempt to establish the association between LPH activity and the degree of quercetin deglycosylation and absorption.

Km (µM)

Substrate Human Pig Lamb

Small Small Liver Liver intestine intestine

Quercetin 4’-glu 27 ± 13 37 ± 12 65 44 ± 7

Genistein 7-glu 13 ± 1 14 ± 5 35 85 ± 11

Table 2.2 Km of quercetin 4'-glu and genistein 7-glu by β-glucosidase from human and animal intestine and liver (Source: Day, DuPont et al. 1998; Lambert, Kroon et al. 1999; Day, Canada et al. 2000).

Unlike the well-documented flavonoid glucosides and rutinosides, other

glycosides such as gal, ara, and xyl are not well studied. It is unclear if their absorption

may partly depend on β-glucosidase but there is a lack of evidence. Limited information

from our recent research suggested that anthocyanin- xyl and ara were better retained in

the cecal content and feces as opposed to anthocyanin- gal and glu (13). Further research

is needed to elucidate the fate of such glycosides.

45 Evidence of enzymatic deglycosylation of anthocyanins is still very limited.

Examination of pig and rat GIT content indicated selective degradation of anthocyanin

glucoside in the small intestine (5, 7, 13), but further characterization of anthocyanin

deglycosylation pattern under the effect of isolated small intestinal β-glucosidases is needed.

2.6.5 The influence of colonic microflora

The enzymes present in the small intestine, including β-glucosidase, cannot

account for hydrolysis of all glycosidic bonds, and hence flavonoid- rha and rut can

survive through the small intestine and reach the colon (70). There are no endogenous

esterases in human to release phenolic acids either. Thus, the esterase activity of colonic

microflora is required for the metabolism of acylated flavonoids (153).

Using in vitro anaerobic fecal fermentation model, Aura et al. (154) demonstrated

that human fecal flora readily deconjugates quercetin- rut, glu, and glucuronide (glc). The

deglycosylated quercetin undergoes ring fission to generate simple phenolics such as 3,4-

dihydroxyphenylacetic acid and its derivatives. One of the microorganisms responsible

for the degradation of flavonoids may be Eubacterium ramulus, as addressed by

Schneider and Blaut (155). Anaerobic incubation with a broad range of flavonoids was

performed after inoculating the media with exponentially growing culture of Eubacterium ramulus, which had been previously isolated. The fermentation end products included hydroxyphenylacetic acids and hydroxyphenylpropionic acids. These degradation

46 products, as well as the deglycosylated aglycones, may be absorbed by the colon, and consequently contribute to the bioactivity of ingested flavonoids.

Fermentation of Cy-3-rut and Cy-3-glu in the presence of human fecal slurry revealed that anthocyanins could also be converted by gut microflora (9). Hydrolysis of

Cy-3-glu was almost complete after 2 h of incubation, and less than 1/3 of the Cy-3-rut remained. Protocatechuic acid (PC), a ring fission product of Cy aglycone, was the major metabolite. In another study Cy-3,5-di-glu was incubated with human fecal suspension

(6). Over 90% of the Cy-3,5-glu was degraded after 2 h and partial hydrolysis generated

Cy-mono-glu as a degradation intermediate, which also underwent degradation in the mean time. Corresponding generation and accumulation of PC was again observed.

Further examination of di-acylated anthocyanins from red radish revealed that the acyl group could be cleaved by fecal microflora and the released acids were relatively stable.

Deacylated anthocyanins would then follow the same pathway of degradation as discussed above.

2.6.6 Metabolism in intestinal mucosa and tissues

Several phase II drug detoxification enzymes involved in xenobiotic conjugation appear to be the key enzymes for flavonoid metabolism after absorption. Catechol-O- methyltransferase (COMT; EC 2.1.1.6), which occurs in various tissues may transfer a methyl group to the flavonoid aglycone (156, 157). Uridine diphosphoglucose glucuronosyl transferase (UDPGT; EC 2.4.1.17) and uridine diphosphoglucose glucose dehydrogenase (UDPGD; EC 1.1.11.22), both abundant in liver and intestine, were

47 proposed to catalyze the glucuronidation of flavonoid aglycones (158). Cytosolic

enzymes phenol sulfotransferases (SULT; EC 2.8.2.1) are widely distributed throughout the body. They are likely to sulfate flavonoids (70).

Some of the metabolites contribute to the bioactivity of flavonoids. For instance,

methylated Cy-3-glu is converted to Pn-3-glu (131). Benzoic acid generated by the

metabolism of quercetin-3-rut may provide antioxidant activity or even anticancer effect

(159). Equol as a colonic metabolite of daidzein is more estrogenic than daidzein and the

other metabolites of isoflavones. Hence the production of this metabolite has attracted great interest (129).

2.6.7 Tissues distribution

The protective effects of flavonoids have been associated with diseases occurring

in various tissues, but such claims are mainly based on in vitro evidence using different

types of cell lines. Knowledge about their availability to target tissues is quite limited.

Quercetin is one of the well investigated flavonoids regarding distribution into tissues.

For example, 2 groups of rats fed either 0.1% or 1% quercetin diet for 11 wk demonstrated the same pattern of tissue distribution (160). The combined concentration of quercetin and its metabolites was high in lung, testes and kidney, moderate in thymus, heart and liver, low in brown fat, muscle and bone, and extremely low in white fat, brain and spleen. The highest tissue concentrations were 3.98 and 15.3 nmol/g in lung for diets with 0.1 and 1% quercetin, respectively. The authors also reported that liver (5.87 nmol/g tissue) and kidney (2.51 nmol/g tissue) contained high concentrations of quercetin in pigs

48 fed 3 d of 500 mg quercetin/kg body wt diet, while brain, heart, and spleen had much

lower concentrations (160).

Anthocyanin distribution in tissues has recently been evaluated in rat and pig

models. Male Wistar rats were fed blackberry extract (370 nmol anthocyanin/d) for 15 d

and killed at 3 h after the beginning of the last meal. Total anthocyanins averaged 605 nmol/g in jejunum, 68.6 nmol/g in stomach, 3.27 nmol/g in kidney, 0.38 nmol/g in liver, and 0.25 nmol/g in brain (161). In pigs fed diets supplemented with 0, 1, 2, or 4% w/w blueberries for 4 wk and fasted for 18-21 h before euthanasia, 1.30 pmol/g of anthocyanins were identified in the liver, 1.58 pmol/g in eyes, 0.878 pmol/g in cortex,

and 0.664 pmol/g in cerebellum (125). The results suggested that anthocyanins crossed

the blood-brain barrier and the blood-retinal barrier to potentially provide protection for

brain and eye tissues.

2.6.8 Excretion

Unabsorbed flavonoids are excreted through feces (13, 162, 163). The absorbed

intact anthocyanins and flavonoid aglycones are largely excreted in urine (130, 164).

Conjugated flavonoid metabolites are likely excreted in urine as well (131), but

alternatively a portion of them may re-enter the jejunum with the bile, and later either

being absorbed by the colon entering the enterohepatic circulation (157, 165), or being

excreted with feces. Lung has been reported as a major excretion site for many

phytochemicals including quercetin (166). Over 50% of the orally ingested 14C-labeled

49 14 quercetin were found exhaled as CO2 in humans. There are no data regarding the respiratory excretion of anthocyanins.

Understanding the bioavailability and metabolism pathway is important to the

health benefits evaluation of anthocyanins. Such knowledge is also necessary for the

screening of suitable anthocyanins from numerous sources to facilitate development of

functional foods/supplement that promote human health. In the last decade our

knowledge of the bioavailability and metabolism of anthocyanins has steadily increased.

The pathways reviewed in this section are summarized in Figure 2.9.

50 Dietary flavonoids stomach

Bi litr Aglycone-Gly ans loc Aglycone ase Pa ssi ve mouth dif fus small intestine ion Released from food matrix by Aglycone-Glu chewing Flavonoid conjugates return with bile S GL SULT T1 Aglycone Sulfoconjugates LPH/CBG Aglycone UDPGT Passive diffusion Aglycone-Glc Aglycone-Gly Aglycone-Gly Acylated anthocyanin UDPGD Aglycone-Rut COMT Aglycone-Glu Methylconjugates liver Phenolic acids

other

Fermentation tissues Aglycone colon Ring fission Phenolic acids kidney

Feces Urine

Figure 2.9 Intergrated putative pathways of dietary flavonoids absorption, metabolism, distribution, and excretion.

51

CHAPTER 3

3. BLACK RASPBERRY ANTHOCYANINS ARE STABLE IN THE DIGESTIVE TRACT LUMEN AND EFFICIENTLY TRANSPORTED INTO GASTRIC AND SMALL INTESTINAL TISSUES IN RAT a

3.1. ABSTRACT

Anthocyanins are abundantly consumed natural pigments with reported health

promoting activities. However information regarding their fate in the gastrointestinal tract

(GIT) is scarce. Kinetic flux and absorption of black raspberry anthocyanins throughout

rat stomach and small intestine were evaluated. Fasted male rats were administered either

vehicle only (n = 6) and killed at 30 min (control) or 12 ± 3 mg of anthocyanins (n = 24)

and euthanatized at 30, 60, 120, and 180 min to collect bladder urine, GIT contents,

stomach and small intestine. HPLC-MS analysis showed that anthocyanin content in the

gastric lumen decreased linearly over time with t½ = 30 min. Anthocyanin content in

small intestinal tissue and lumen peaked at 120 min. Uptake by the small intestine tissue

reached 7.5% of the administered dose, a much higher percentage than the previously

reported absorption of these pigments (normally <0.1%). The relatively high recovery

(75-79%) of administered anthocyanins from GIT contents demonstrated their overall

a Jian He1, Taylor Wallace1, Kristin E. Keatley2, Mark L. Failla2, and M. Mónica Giusti1, 1Dept. of Food Science & Technology, 110 Parker Food Science Bldg., 2015 Fyffe Ct., 2Dept. of Human Nutrition, 325 Campbell Hall, 1787 Neil Ave., The Ohio State Univ., Columbus, OH 43210 52 stability at physiological conditions. Selective decrease of the concentration of cyanidin-

3-glucoside in the small intestinal content resulted from β-glucosidase activity. Urine

anthocyanin analysis indicated that profile of absorbed anthocyanins reflected that

presernt in the GIT at the time of absorption. Our results suggest that ingested

anthocyanin glycosides remain relatively stable in the GIT lumen and are efficiently

taken up into the GIT tissues with limited transfer to the plasma compartment.

Key words: Anthocyanin; black raspberry; digestion; gastrointestinal tract; tissue uptake;

absorption; β-glucosidase; stability; metabolism

3.2. INTRODUCTION

Anthocyanins are a group of natural pigments that belong to the large family of

flavonoids and are responsible for the red, purple and blue colors of many plant materials.

Consumption of anthocyanins is high compared to other flavonoids (70) with the daily

intake estimated to be 12.5 mg/day/person in the USA (44). Berries are generally high in

anthocyanin content and contribute a major portion of anthocyanin consumed (167).

Increasing evidence shows that anthocyanins are potent antioxidants and are associated with protective effects against many chronic diseases such as cancer, cardiovascular diseases, and even obesity (3). However, such postulated health- promoting effects are largely based on in vitro evidence, while data related to the in vivo absorption and metabolism of anthocyanins remains limited (168). Recent animal and clinical studies have focused on the bioavailability of anthocyanins by monitoring the 53 concentration in plasma and urine, as well as distribution in organs (133, 161, 169). In contrast, information about the fate of anthocyanins in the GIT is scarce. The lack of such information impairs our understanding of their dietary value. First, since stomach and small intestine have been considered as potential absorption sites of anthocyanins (140,

141, 143), the availability of anthocyanins in these segments would likely be a major factor influencing absorption. Existing data suggest rapid absorption and elimination of a small percentage of ingested anthocyanins (169). Therefore, the emptying time in the stomach and transit rate in the small intestine is expected to determine the duration of mucosal cell uptake and circulation in the blood. Second, intact anthocyanins in the GIT lumen may directly interact with GIT tissues, where these antioxidants may protect GIT against acute distress and chronic diseases associated with oxidative stress (8, 13, 113,

170). Third, metabolism of anthocyanin glycosides has been proposed to be possible before absorption (171). It was suggested that less hydrophilic aglycones could be produced in the small intestine by the action of β-glucosidase with the resulting aglycone being transported across the mucosal epithelium via passive diffusion in a manner analogous to some other flavonoids (168). Further derivatives from the ring fission of aglycones may also be absorbed and contribute to the health-promoting effects (6, 171).

Direct evidence of anthocyanin deglycosylation in the GIT is required to support such speculations.

Black raspberry ( occidentalis) can provide 845 mg of anthocyanins in a single serving and was selected as a representative anthocyanin-rich source for the present study (44). Its phenolic content, primarily anthocyanins, has been shown to

54 effectively inhibit proliferation of human oral, breast, prostate, and colon cancer cells in

vitro (172). Here we investigated the kinetics of black raspberry anthocyanin flux in

stomach and intestine of fasted rats, focusing on the stability, rate of emptying, uptake

into GIT tissues, and selective deglycosylation by intestinal β-glucosidase activity.

3.3. MATERIALS AND METHODS

3.3.1 Chemicals and materials

Reagent grade tri-fluoroacetic acid (TFA), high performance liquid

chromatography (HPLC) grade acetone and methanol for extraction, and HPLC- mass

spectrometer (MS) grade water and acetonitrile for chromatography were purchased from

Fisher Scientific (Fair Lawn, NJ). Formic acid (99%) was from Acros Organics (Morris

Plains, NJ). Cyanidin (Cy) aglycone was prepared from Cy-3-glu standard (ChromaDex

Inc., Irvine, CA) by acid hydrolysis (173). Protocatechuic acid standard was from MP

Biomedicals (Solon, OH).

3.3.2 Black raspberry extract

Lyophilized black raspberry powder was a generous gift from Dr. Laura Kresty,

OSU Comprehensive Cancer Center. The powder was extracted following a modified procedure by Jing et al. (174). One part of the dry powder was mixed with three parts of acetone/water (70:30, v/v) containing 0.1% formic acid, briefly homogenized by a handheld tissuemiser (Fisher Scientific, Fair Lawn, NJ), and sonicated in a Fisher FS30

55 ultrasonic bath for 15 min. The slurry was then centrifuged at 1800×g for 10 min and the

supernatant was collected. The pellet was re-extracted twice for further removal of

pigmentation from the matrix. The pooled supernatant was mixed (1:1, v/v) with

chloroform and centrifuged at 1800×g for 15 min. The red aqueous fraction was

transferred to a round-bottomed flask and evaporated with a Büchii rotovaporator at 35

°C to remove organic solvents. The extract was lyophilized to powder form and stored at

-80°C under N2.

3.3.3 Animals and experimental design

Male Fischer 344 rats (8 wk of age; ~200 g) were purchased (Taconic Farms,

Germantown, NY) and maintained in the University Laboratory Animal Resources

facility on the Ohio State University campus in accordance with NIH Laboratory Animal

Use Guidelines. The Institutional Animal Care and Use Committee at the Ohio State

University approved the experimental protocol (OSU Protocol #2004A0139). Animals

were group housed with two animals per cage and maintained under standard conditions

(22 ± 2 °C, 50 ± 10% relative humidity, 12 h light/dark cycle). Commercial diet and

water were available ad libitum prior to the experiment.

Rats (n=30) at 11 wk of age were randomly assigned to one of five groups. After

being fasted overnight, animals were lightly anesthetized for the delivery of 1.2 ± 0.3 mL

water either containing 0.1% citric acid alone (control) or including black raspberry

extract (treatment) by stomach tube. The precise volume delivered was determined by

weighing the syringe and intubation tube before and after gavage. The administered

56 extract contained 10 mg Cy-3-glu equivalent/mL as measured by pH differential method

(175). Treatment groups were asphyxiated with CO2 after 30 (n = 6), 60 (n = 6), 120 (n =

6), and 180 min (n = 6). Control group (n = 6) was asphyxiated at 30 min after intubation.

Contents from stomach and small intestine were separately collected by perfusing the lumen with 25 mL of ice-cold 20% formic acid in buffered saline (PBS). Urine samples (when present) were collected from bladder and acidified immediately with 20%

(v/v) formic acid. Stomach and small intestine were then removed and immediately frozen by liquid nitrogen. All samples were stored at -80°C until analysis.

3.3.4 Sample preparation.

Urine samples were thawed and weighted with the container. The weight of

container was later recorded and subtracted from the total weight. Urine volume was

calculated based on weight assuming a density of 1.0 g/mL. A 50 μL aliquot was added

to 550 μL of 0.1% TFA and centrifuged at 16000×g for 15 min prior to HPLC analysis of

the supernatant (143). The remaining samples were pooled and semi-purified by a 1cc

® Sep-Pak C18 cartridge (Waters, Milford, MA) for HPLC-MS analysis to confirm peak

identities.

Gastric and small intestinal contents were thawed, sonicated for 15 min in ice

water bath, and centrifuged at 1800×g for 10 min. An aliquot of the supernatant was

filtered (0.45 μm Whatman® polypropylene filter) for HPLC analysis. A second aliquot

® of the remaining sample was pooled by group and semi-purified by a 6cc Sep-Pak C18 cartridge for HPLC-MS analysis to confirm peak identities. The C18 semi-purification

57 method was modified from our previous studies (13, 25). Briefly, sample was loaded onto a pre-conditioned C18 cartridge and washed with 2 vol of 0.1% TFA acidified water before eluting anthocyanins with 1 vol of 0.1% TFA acidified methanol. The methanolic eluate of urine was dried under N2 and redissolved in an aliquot of acidified water for

HPLC analyses. In order to preserve the labile anthocyanidin aglycone in gastrointestinal samples, the methanolic eluate was not evaporated but instead diluted in acidified water

(1:9, v/v) to maintain adsorption of anthocyanin peaks on the reverse-phase HPLC column.

Extraction of anthocyanins from the stomach (1.43 ± 0.032 g, n = 30) and small intestine tissues (7.1 ± 0.12 g, n = 30) were carried out separately. The samples and solutions were kept on ice during the entire process to prevent enzymatic and/or heat degradation. Mucus and epithelial layer were removed from each tissue because dark food particles adhered to the surface. The remaining muscular tissue was homogenized at high speed for approximately 3 min in 10 mL (stomach) or 15 mL (intestine) of methanol

(with 0.1% formic acid) using a hand-hold tissuemiser followed by sonication in iced water-bath for 20 s. The methanolic extract was then centrifuged at 1800×g for 10 min.

Supernatants were collected, pellets were re-extracted and centrifuged at 1800×g for 10 min, and the pooled supernatants were diluted 6 fold using water containing 0.1% formic

® acid. Diluted supernatants were then passed through a Sep-Pak C18 (5 g, 20 cc) cartridge and recovered with 10mL of 0.1% formic acid in methanol which was evaporated using a

Büchii rotary evaporator. The remaining anthocyanins were redissolved to 5mL using

58 10% formic acid prior to centrifugation at 16000×g for 10 min, filtration with 0.45μL polypropylene syringe filters, and HPLC-MS analysis.

3.3.5 Recovery

Black raspberry extract was spiked into effluents of gastric and small intestinal contents from control animals (n = 4) to simulate the concentration found in the treatment groups. The slurry was sonicated in ice water bath for 15 min to facilitate anthocyanin penetration into particulate materials. Spiked samples were frozen and analyzed on the following day according to the procedures described above. Recovery was calculated by total amount recovered/total amount added × 100%. Profiles of recovered anthocyanins from the spiked samples were compared to that from the gastrointestinal contents of treatment groups to identify specific changes not attributable to non-specific binding.

3.3.6 HPLC-MS analysis of anthocyanins and anthocyanin metabolites

Samples were analyzed using a Shimadzu LCMS-2010 EV HPLC-MS (Shimadzu

Scientific Instruments, Inc., Columbia, MD) equipped with an SPD-M20A photodiode array (PDA) detector and a single quadrupole electron spray ionization (ESI) MS detector.

® Separation was accomplished on a Symmetry C18 column (3.5 μm, 4.6×150 mm; Waters

Corp., Milford, MA) with a flow rate of 0.8 mL/min. Conditions for urine samples were as follows: injection volume: 200 μL; mobile phase: A, 4.5% formic acid in HPLC-MS grade water; B, HPLC-MS grade acetonitrile; gradient: 0-5 min, 2-7% B; 5-20 min, 7-

12% B; 20-25 min, 12-17% B; 25-40 min, 17-25% B. Conditions for gastric and small intestinal contents/tissues samples were the following: injection volume: 20 μL; mobile 59 phase: A, 10% formic acid in HPLC-MS grade water; B, HPLC-MS grade acetonitrile; gradient: 0-6 min, 5-8% B; 6-15 min, 8-10% B; 15-20 min, 10-25% B; 20-25 min, 25% B.

After each run, the column was equilibrated for 5 min under the initial condition. Spectral data (250-700 nm) were collected during the entire separation procedure. When MS was coupled to the HPLC, spectra were obtained under positive ion condition using SCAN

(from m/z 200 to 1200) and Selective Ion Monitoring (SIM) modes. Six channels including m/z 287 for Cy and m/z ratios for other common anthocyanin aglycones were monitored in the SIM mode. Anthocyanin concentrations in the urine, perfusate of gastrointestinal contents and tissues were calculated using areas under curve (AUC) of

HPLC chromatograms at 520 nm and a standard calibration curve. Anthocyanin amount was calculated by multiplying the concentration in the aliquot with the corresponding volume.

3.3.7 Calibration curve

Commercially available Cy-3-glu standard (10 mg) was dissolved in 10 mL of double distilled water containing 0.1% TFA, and a series of dilutions were prepared to generate a standard curve (R2 > 0.99). All anthocyanins analyzed fell within the range of the standard curve and were expressed as Cy-3-glu equivalents by weight. The total amount of recovered anthocyanins was calculated by summing up the AUC of individual anthocyanin peaks at 520 nm and using this calibration curve.

60 3.3.8 Statistical analysis

One-way ANOVA or its non-parametric analogue Kruskal-Wallis Test (equal variance not assumed) was conducted using SPSS (version 13, 2004, SPSS Inc., Chicago,

IL) to compare amounts of total and individual anthocyanins in different treatment groups, and data are given as means ± SE. When appropriate, significance of differences between means was determined by Tukey's HSD or Dunnett’s T3 (equal variance not assumed).

Differences of P<0.05 were considered significant.

3.4. RESULTS AND DISCUSSION

3.4.1 Flux of anthocyanins in GIT lumen

Kinetic flux of anthocyanins in the GIT is critical for understanding anthocyanin absorption. Recent studies have suggested that the stomach is a potential site for the absorption of anthocyanins (140, 141), in addition to the small intestine which is a known common absorption site for other structurally related flavonoids (143). Therefore, absorption may begin immediately after ingested anthocyanins reach the stomach. The rate of gastric emptying, transit through the small intestine, as well as anthocyanin stability, will determine the duration and extent of anthocyanin accessibility to the mucosal epithelia for uptake and possible transfer to circulation. Combining kinetic flux through the GIT with pharmacokinetic data is necessary to determine the primary site(s) of anthocyanin absorption.

61 A preliminary examination using spiked perfusate samples from fasted rats

demonstrated recovery rates of 93 ± 2% (n = 4) and 84 ± 10% (n = 4) for black raspberry

anthocyanins added to gastric and small intestinal contents, respectively. We observed

that contents from the small intestine had a gel-like property causing difficulty with

passage through the 0.45 μm polypropylene filter. The extent of pigmentation on filters

was greater for samples with more viscous or gel-like consistency. The viscosity

variability of luminal contents from the small intestine samples contributed to the

relatively large variation in recovery of anthocyanins from the small intestinal contents.

Anthocyanins in the black raspberry extract and collected biological materials

were identified by comparing retention times and UV-visible spectra to known standards,

and by comparing m/z ratios of whole molecular ions and fragments to established values

(176). The four major anthocyanins identified in the crude extract of black raspberry, as

well as gastric and intestinal contents (Table 3.1, Figure 3.1) included Cy-3-

sambubioside (Cy-3-sam), Cy-3-glu, Cy-3-xylosylrutinoside (Cy-3-(xyl)-rut), and Cy-3- rutinoside (Cy-3-rut). Anthocyanin total concentrations in the gastric and small intestinal contents were adjusted for the efficiency of extraction and are presented in Figure 3.2.

Approximately 52% of the total amount of black raspberry anthocyanins delivered to the

stomach (12 mg) was present in gastric luminal contents 30 min after administration of

the bolus. Anthocyanin content in the gastric lumen decreased linearly (R2 = 0.67) during

the 180 min study. The estimated time to deplete one half of the anthocyanin content in

the gastric lumen (t½) of the fasted rat was approximately 120 min, suggesting that

minimal amounts of anthocyanins would be present in the stomach after 4 h. Our result

62 agrees with a previous study by Borges et al. (14) who reported almost no raspberry

anthocyanins remained in the gastric lumen of fasted rats 4 h after gavaging raspberry juice. However, these investigators observed a non-linear decrease of the anthocyanin

content in the stomach, with a more rapid decrease during the first hour. This difference

in the two studies may be due to the administration of a much larger dose of anthocyanins

than that in our study (920 nmol vs. 26.7 nmol, respectively).

m/z Peak number λmax (nm) Intact Aglycone Peak identity molecule fragment 1 520 581 287 Cy-3-sam

2 516 449 287 Cy-3-gluc

3 522 727 287 Cy-3-(xyl)-rut

4 522 595 287 Cy-3-rut

5 520 287 287 Cy

Table 3.1 Peak assignments of anthocyanins extracted from the gastric and intestinal contents of rats administered black raspberry extract.

63 are presentedinTable1. Chromatograms aremonitored at520nm forde stomach (C)andsmall intestine(D)fr the administered crudeextract(B),andrepr Figure 3.1 Absorbance (at 520 nm) 0 Chromatographic profilesofCystandard 5 1 1 1 2 2 2 3 3 3 10 Time (min) 4 4 4 B) Blackraspberry extract A) Cystandard D) Smallintestinalcontent C) Gastriccontent om aratkilled120min aftergavage. 15 esentative samples ofluminal contentsfrom 64 tection ofanthocyanins.Peakidentities (A),blackraspberryanthocyaninsin 5 5 5 20 25

12 ab Stomach ab a 10 Intestine Combined 8 a a b ab 6 ab 4 bab bc

d 2 A) Gastrointestinal content 0 1.2 0 30 60 90 120 150 180 210 Time (min)a Intestine a 1

0.8 a a 0.6 Total Anthocyanin Content (mg cy-3-glu (mg Content Anthocyanin Total eq.)

0.4

0.2 B) Small intestine tissue 0 0 30 60 90 120 150 180 210 Time (min)

Figure 3.2 Black raspberry anthocyanins in the gastric and small intestinal contents (A) and in the small intestinal tissue (B). Rats were administered approximately 12 mg (Cy-3- glu equivalent) of anthocyanins in acidic solution by stomach tube. Data are means ± SE for 6 rats. Within each line, means with different letters are significantly different (P < 0.05).

65 The amount of anthocyanins in the small intestinal lumen increased between 30

and 120 min after administering the black raspberry extract, and then decreased by 180

min (Figure 3.2). This suggests that the amount entering the small intestine prior to 120

min exceeded the total amount of small intestinal mucosa uptake, exiting to the lower

intestine, and perhaps degradation. Combining such information with the fact that plasma

anthocyanin concentration usually is maximum within 0.25-0.5 h in rats after gastric

intubation of anthocyanin extract (169, 177, 178), it is reasonable to speculate that the

stomach has a central role in anthocyanin absorption. Otherwise, plasma concentration of

anthocyanins would be maintained at the same or a higher concentration after 0.5 h as

anthocyanin content in the small intestine continues to increase from 30-120 min.

The total amount of administered anthocyanins recovered from gastric and small

intestinal contents was 75-79% (~ 9 mg Cy-3-glu equivalents) at 30-120 min after

delivery (Figure 3.2), demonstrating that the anthocyanins delivered in the absence of a

food/meal are stable within the lumen of the upper GIT of fasted rats. In contrast, losses of greater than 50% of anthocyanins have been reported during in vitro digestion (6, 10,

11). Anthocyanin are not stable at neutral pH, forming pseudobases, quinoidal bases, and chalcones that subject to nucleophilic attack by water (11). However, we hypothesize that binding of anthocyanins to mucus, secretions and food residues in vivo may increase stability (47). Knowing the stability of anthocyanins in the GIT lumen is important not only because the accessibility affects absorption but also because some suspected health benefits depend on continuous exposure to anthocyanins. For example, Malik et al. reported that exposure of colon cancer cells to anthocyanins caused cell cycle arrest at

66 G1/G0 and G2/M phases, yet after the anthocyanins were removed the cells could recover

from both the G1/G0 and G2/M blocks (95). By 180 min the anthocyanin content in the

combined gastric and small intestinal contents decreased significantly as anthocyanins presumably entered the large intestine (Figure 3.2A).

3.4.2 Anthocyanins in GIT tissues

Intense red color was present in the acidic extract of all gastric and small intestine

tissue samples. The intensity of the red color in the gastric tissue extract steadily

decreased over time, coinciding with the linear decrease of anthocyanins in the gastric

lumen. However, anthocyanins appeared to bind to unidentified protein in the stomach tissue thus could not be quantified as free anthocyanins by HPLC. The presence of anthocyanins in the gastric tissue extract was confirmed by observing spectral change at the pH of 1.0, 4.5, and 10.0 (Figure 3.3). Such binding may be attributed to non-specific binding or perhaps specific binding to transporter protein in the stomach (142). There was

a parallel trend of anthocyanin concentrations in the small intestinal tissue and luminal contents (Figures 3.1 A and B). This correlation suggests that some portion of the newly acquired anthocyanins within GIT mucosa was quickly degraded and/or effluxed across the apical or basolateral membranes. Otherwise, anthocyanins would be expected to continue to accumulate in the tissues with increasing time.

67 Bound anthocyanins pH 1 Bound anthocyanins pH 4.5 Free anthocyanins pH1 Absorbance

300 400 500 600 700 Wavelength (nm)

Figure 3.3 UV-Vis spectra of the anthocyanin-protein complex extracted from the stomach tissue. Reversible decrease of absorbance near 520nm is unique for anthocyanins when pH changes from 1 to 4.5.

The total amount of anthocyanins in the small intestine tissue reached 7.5% of the

administered dose (120 min), a much higher percentage than the reported anthocyanin

absorption (normally <0.1%), which was based on plasma and urine anthocyanin

concentrations (3). Considering the fact that anthocyanins would degrade or exit GIT

tissue rapidly, the actual uptake by the small intestine tissue may be much larger than

7.5%. In the current study when 26.7 nmol of black raspberry anthocyanins was administered as a bolus, the highest concentration in the small intestine tissue (282 pmol

Cy-3-glu equivalent/g, n = 6) was observed at 120 min. This observation agrees with a recent report by Talavera et al. in which 605 pmol Cy-3-glu equivalent/g jejunum tissue

68 was found in rats (n = 6) fed ~370 nmol of blackberry anthocyanins for 15 days (161).

The exceptionally high anthocyanin concentration in GIT tissues as compared to the much lower concentrations in blood or other tissues (125, 161) suggests that intact anthocyanins may be taken up by GIT tissues efficiently but not effectively transported into circulation. Our speculation is supported by a recent study evaluating absorption of black currant anthocyanins by monolayers of human intestinal epithelial Caco-2 cells

(179). This in vitro model revealed that transport across the apical membrane occurs to a much larger extent than further translocation across the basolateral membrane. Despite inefficient transfer across the basolateral membrane into plasma, anthocyanins in GIT tissues may have protective effects in situ. Bruce et al. (113) suggested that colon cancer development might be inhibited by agents that prevent epithelial barrier damage, inhibit inflammation, or quench reactive oxygen species in local epithelial cells. Anthocyanins have been shown to be potent antioxidant and anti-inflammatory agents in vitro (89). In vivo studies examining the ability of anthocyanins to protect GIT against oxidative stress are merited.

3.4.3 Anthocyanin transformation in the GIT

Besides the 4 major anthocyanins identified in the black raspberry diet, a minor peak of Cy (anthocyanidin) was observed in both stomach and intestinal contents (Peak 5 in Figure 3.1 C and D). Anthocyanidins are degradation intermediate of glycosylated anthocyanins when deglycosylation occurs. As anthocyanidins degrade quickly at room temperature, in order to preserve them we shortened sample preparation to avoid using rotary evaporation or a stream of N2 to remove organic solvents. Slight change of the

69 anthocyanin profile over time occurred in the gastric content, probably due to acid hydrolysis. Anthocyanidins are usually produced from glycosylated anthocyanins via acid hydrolysis in the laboratory (173). Boiling and 2N HCl ensure nearly complete cleavage of glycosidic bonds within 30 min. The pH in the gastric lumen after fasting can be as low as 1 and this acidic environment may cause limited hydrolysis. It has to be noted that the vehicle for delivery of the black raspberry extract to rat stomach was slightly acidic and thus expected to stabilize anthocyanins. In contrast, introduction of food matrix into the gastric lumen leads to rapid elevation of gastric pH (180), resulting in perhaps no acid hydrolysis and decreased overall stability of anthocyanins. The presence of Cy in the gastric content (Figure 3.1C) suggested cleavage of glycosidic bond between anthocyanidin and sugar moieties. As cleavage may also occur at glycosidic bond within a multi-glycoside moiety, hydrolysis of tri-glycoside (e.g. between rutinose and xylose) or di-glycoside (e.g. between glucose and rhamnose) to the simpler glycoside was also observed (Figure 3.4). Such hydrolysis explained the slight but steady decrease of Cy-3-

(xyl)-rut with proportional increases in Cy-3-rut and Cy-3-glu in the stomach over time.

The small intestinal contents are neutral so further acid hydrolysis is not expected.

Changes in the anthocyanin profile in the GIT are likely to influence the profile observed in plasma, tissues and urine (7). Failure to consider this factor may overestimate the absorption of mono-glycosides and underestimate the absorption of multi-glycosides present in foods, beverages and formulations.

70 70 A) Gastric content 30min (n=6) 60 60min (n=5) 50 120min (n=6) 40 180min (n=6)

30

20

10

0 70 B) Small intestinal content 60 Percent Peak Area 50

40

30

20

10

0 cy-3-sam cy-3-glu cy-3-(xyl)-rut cy-3-rut pg-3-rut

OH OH OH OH OH OH HO O+ HO O+ HO O+ rhamnoside rhamnoside O glucoside O glucoside O glucoside OH xyloside OH OH cy-3-(xyl)-rut cy-3-rut cy-3-glu

Figure 3.4 Relative abundance of black raspberry anthocyanins in the gastric and small intestinal lumens. ANOVA for linear regression model indicated significant (P < 0.05) decrease in the relative amount of Cy-3-(xyl)-rut and a significant increase in the relative amounts of Cy-3-rut and Cy-3-glu in gastric lumen. Pattern of anthocyanin profile in the small intestine is similar to that in the stomach, indicating no further metabolism with the exception of the decrease in Cy-3-glu.

71 The anthocyanin profile of small intestinal contents was similar to that in the

stomach (Figure 3.4) with one exception. The relative amount of Cy-3-glu in the small

intestinal contents was significantly (p<0.001) decreased in comparison to that in the

administered extract and gastric contents (Figure 3.5). Such selective degradation of this

anthocyanin glucoside was reported in our previous animal study using rats (13) and by

Wu et al. using a pig model (5, 7). We observed additional supporting evidence for the

hydrolysis of anthocyanins as Cy appeared in the intestinal contents (Figure 3.1D). But

interpretation of Cy generation must be cautious, as it is possible that the observed Cy

was carried over from the stomach. LPH, which is located in the brush border membrane

of mature small intestine enterocytes (181), and/or glycosidase activity in microorganisms residing in the small intestine are likely the basis for this hydrolysis.

72 10 8.8a 8.9a 8.5a 8

6.3b 6 n=4 n=4

4 n=23 n=23 2 Percent Peak Area of Cy-3-glu Peak of Area Percent

0 Spiked Trt group Spiked Trt group stomach stomach intestine intestine content content content content

Figure 3.5 Relative abundance of Cy-3-glu is decreased in the small intestinal content. Due to negligible change with time, all data from the 4 treatment groups are pooled. Means with different letters are significantly different (P < 0.05) as determined by Dunnett’s T3 test following Kruskal-Wallis Test.

A preliminary test using the small intestine mucosal layer of 8-month old pigs has

also provided supporting evidence of LPH activity by showing the selective hydrolysis of

Cy-3-glu in black raspberry anthocyanins after incubation (unpublished data). An

alternative explanation to the selective decrease of Cy-3-glu in the intestinal lumen is

high uptake. However, the Cy-3-glu recovered from small intestine tissue represented 7.5

± 0.16% (n = 23) of the total anthocyanins extracted, which was significantly (P<0.001)

lower than that in the black raspberry extract and in the spiked control samples. Such

evidence confirmed that the selective decrease was due to degradation rather than uptake.

73 In a previous study using in situ intestine perfusion to evaluate anthocyanins uptake,

Talavera et al. stated that Cy-3-glu was preferably absorbed as compared to other Cy glycosides based on the evidence that Cy-3-glu disappeared in a large extent in the intestinal efflux (143). We now suggest that their observation might instead be attributed

to degradation of Cy-3-glu during intestine perfusion. In the present study, degradation of

anthocyanin glucoside did not appear to be a major factor for overall anthocyanin

stability. However, it is noteworthy that >90% anthocyanins identified in nature contain

glucoside moiety (16), and in weaning mammals (150) or some adults (182) LPH activity

remains remarkably high, potentially causing extensive biotransformation of ingested anthocyanins.

Hydrolysis of the glycosidic bond is considered critical for flavonoids digestion

since their aglycones rather than the ingested glycosides are readily absorbed (147, 151,

152, 183). Anthocyanins differ from other flavonoids in that anthocyanidins (aglycones)

are considered to be relatively unstable and rarely detected in plasma or urine in many

studies (169). However, even if only present for a relatively short time, anthocyanidins

may have the potential to be taken up by epithelial cells lining the gastric and small

intestinal mucosa where they are metabolized to become conjugated metabolites with

some being absorbed (184). The anthocyanidins may also undergo spontaneous ring

fission to generate smaller phenolic compounds that may be absorbed to have health

promoting benefits (6, 171). In the case of the Cy aglycone, protocatechuic acid (PC) is

expected to be one of the ring fission products. Although we have observed the presence

of PC when genuine Cy standard was degraded in buffer, PC was not detected in either

74 the gastric or small intestinal contents. In another study Cy and PC were particularly monitored but not found in rat plasma after gavage feeding anthocyanin-rich extract from wild mulberry (185). Perhaps Cy and PC were promptly degraded during digestion in gastric and small intestinal lumens, or only primarily produced in the large intestine where absorption is much less efficient than the small intestine (6, 9).

3.4.4 Metabolism and excretion of absorbed anthocyanins

Black raspberry anthocyanins appeared in the urine within 30 min after stomach intubation (Table 3.2). At 120 min total anthocyanins in the urine collected directly from bladder accounted for 0.045% of that administered. Although urine collection was incomplete, this low amount agrees with other studies showing the low bioavailability of anthocyanins (186). Anthocyanin content in the urine collected from bladder at 180 min post-administration declined suggesting excretion and further metabolism. Methylated anthocyanin metabolites, primarily from the most abundant black raspberry anthocyanin

Cy-3-rut, were identified in urine and accounted for 18.2% (n = 15) of total anthocyanins recovered. Nevertheless, such methylated metabolites were not detected in the GIT contents. Enterohepatic cycle extends the plasma elimination half-life of many flavonoids due to re-absorption (168), but with such low concentrations, if any, of anthocyanin metabolites found in the GIT lumen, enterohepatic cycle (184) appeared not capable of influencing anthocyanin excretion half-life (161, 187).

75 60 min 120 min 180 min 30 min (n=3) (n=5) (n=5) (n=2)

Recovery of 0.024 0.013 0.007 0.045 administered (0.007- (0.010- (0.003-0.010)b (0.011-0.10) anthocyanins (%) 0.042) 0.017)

Relative abundance of 9.3 8.6 7.4 6.4 co-eluted Cy-3-glu and (8.9-9.7) (7.3-9.7) (6.7-8.1) (6.3-6.5) Cy-3-sam (%)

a: Data presented as mean values of all available samples in each group.

b: Numbers in parentheses indicate the range for each group.

Table 3.2 Black raspberry anthocyanins in urine collected from rat bladder.a

Relative abundance of individual anthocyanins in urine was similar to that in the

GIT. Wu et al. also reported a positive relationship between the anthocyanin profile in the

GIT and that in the urine of weaned pigs (7). Similarly, we found that hydrolysis of Cy-3- glu in the small intestine (~2.5% of the total anthocyanins, Figure 3.5) was associated

with a relative decrease of Cy-3-glu in urine at 180 min (~2.9% drop comparing to 30

min, Table 3.2). At 180 min the stomach was almost empty, thus gastric absorption was

negligible and the small intestinal content profile determined the urine profile. At 30 min

gastric absorption appeared to be predominant as the urinary profile of anthocyanins was

quite similar to that in the gastric lumen (Table 3.2).

76 3.5. CONCLUSIONS

In conclusion, black raspberry anthocyanins were relatively stable in the gastric and small intestinal lumens of fasted rats in contrast to reports from some in vitro studies.

Total anthocyanins in gastric lumen and tissue steadily decreased during the 180 min period following administration, whereas anthocyanin contents in the small intestinal lumen and tissue were maximum at 120 min before decreasing. A significant portion of administrated anthocyanins was taken up into the GIT tissues, but neither extensively delivered into the blood nor cumulatively retained. The profile of anthocyanins changed in the GIT with some hydrolysis of glycosidic linkages occurring in the acidic gastric lumen, as well as selective hydrolysis of Cy-3-glu in the small intestinal content likely due to endogenous β-glucosidase activity. Unlike some other flavonoids, the stomach is an important site for absorption of anthocyanins. As anthocyanins transit the GIT lumen, the predominant absorption site changed gradually from the stomach to the small intestine. The present study provides novel information about the kinetics of the gastrointestinal flux of anthocyanins in vivo and is expected to contribute to a better understanding of the metabolism, absorption, tissue uptake, and health-promoting activities of these water-soluble plant pigments.

3.6. ACKNOWLEDGEMENTS

We thank the USDA NRI Competitive Grants Program 2004-35503-15190 and the Ohio Agriculture Research and Development Center Project OHOA1192 for

77 supporting this study. We also thank Dr Laura Kresty for providing lyophilized black raspberry powder.

78

CHAPTER 4

4. HIGH-PURITY ISOLATION OF ANTHOCYANIN MIXTURES FROM FRUITS AND VEGETABLES – A NOVEL SOLID-PHASE EXTRACTION METHOD a b

4.1. ABSTRACT

Research on biological activity of anthocyanins requires the availability of high

purity materials. However, current methods to isolate anthocyanins or anthocyanin

mixtures are insufficient for complete removal of chemically related compounds. We

employed a novel cation-exchange/reversed-phase combination technique, the Oasis®

MCX solid-phase extraction (SPE) cartridge, and optimized the use of water/organic

buffer mobile phases to selectively separate anthocyanins. Crude extracts of various

representative anthocyanin sources were purified with this technique and compared to 3

commonly used SPE sorbents: C18, HLB, and LH-20. Purified anthocyanin fractions were

analyzed with high performance liquid chromatography (HPLC) coupled to photodiode

array (PDA) and mass spectrometry (MS) detectors and by Fourier transform infrared

(FT-IR) spectroscopy. The UV-visible chromatograms quantitatively demonstrated that

a Jian He and M. Mónica Giusti, Dept. of Food Science & Technology, 110 Parker Food Science Bldg., 2015 Fyffe Ct., The Ohio State Univ., Columbus, OH 43210 b Patent pending. 79 our novel technique achieved significantly higher (P<0.05) anthocyanin purity than the

C18 cartridge, the next best method, for 11 of the 12 anthocyanin sources tested. Among

the 12 sources, eight were purified to greater than 99% purity (based on UV-Visible

chromatograms). The new method efficiently removed non-anthocyanin phenolics. MS

and FT-IR semi-quantitatively confirmed extensive reduction of impurities. Due to strong

ionic interaction, the MCX sorbent capacity was superior to others, resulting in the

highest throughput and least use of organic solvents. This new methodology for isolation

of anthocyanin mixtures drastically increased purity and efficiency while maintaining

excellent yield (93.6 ± 0.55%) and low cost. The availability of high purity anthocyanins

facilitates expansion of scientific studies, as well as applications in the food and

nutraceutical industries.

Key words: cation-exchange; anthocyanin; flavylium cation; purification; solid-phase

extraction; polyphenols

4.2. INTRODUCTION

Anthocyanins, a class of polyphenols, are responsible for the red, purple and blue

color of most fruits and vegetables. Increasing evidence shows that anthocyanins are potent antioxidants and anthocyanin rich fruits and vegetables, as well as anthocyanin

rich extracts, have been associated with protective effects against many chronic diseases

such as cancer, cardiovascular diseases, and even obesity (3). Interest in the use of

anthocyanins as alternatives to synthetic food colorants has increased, and many

80 researchers are investigating their potential health benefits. Obtaining high-purity

anthocyanins is essential for such research (188). Many bioassays on anthocyanin-rich

commodities are not feasible without eliminating bioactive impurities that confound

interpretation of results. To date there have been 637 naturally occurring anthocyanins

identified (189). Unfortunately, there are only a limited number of pure standards

commercially available and the cost is high. Therefore, many biological studies are

performed using crude anthocyanin extracts from fruits and vegetables.

Isolation methods range from simple solvent extraction to various forms of chromatography (15). SPE methods have gained popularity recently, due to a balance of efficiency and cost (190, 191). However, SPE sorbents normally adsorb analytes via non- selective interactions, which would inevitably allow for a broad spectrum of plant constituents to contaminate the anthocyanin fraction. The impurities, mostly phenolics, are likely to have biological effects. and therefore become confounding factors in bioassays. Thus, explanation of anthocyanin bioactivity could be vague (73), and results from different labs are difficult to compare given the different isolation methods employed. Removal of undesirable compounds from anthocyanins is also of great importance in the food colorant and nutraceutical industries. Sugars, phenolic compounds, amino acids, and metals accelerate degradation of anthocyanins and therefore high purity is necessary for improved stability (192). Some potential low-cost sources of anthocyanins are not being commercialized because of co-extracted adverse flavor/aroma or even natural toxins (65). Removal of undesirable odors or flavors can only be achieved with multiple purification steps and tedious manipulations, increasing cost (15, 193).

81 In this study we explored a low-cost and high-throughput isolation technique for anthocyanin based on a unique property of anthocyanins. Anthocyanin molecules have the unique property that they acquire a positive charge to become flavylium cations at highly acidic pH while losing the positive charge or even obtaining one or two negative charges at basic pH to become ionized quinonoidal bases (Figure 4.1) (30, 33). Such transformation does not occur with most other plant compounds. We hypothesized that the positively charged anthocyanin molecules (pKh ~2.8) (29) would interact with strong cation-exchange sorbent. Meanwhile most other compounds without positive charges could be easily removed. Dissociation of anthocyanins from the solid phase was carried out by increasing the elution solvent pH. A recently introduced sorbent combining both cation-exchange and reversed-phase interaction mechanisms was employed to develop such a method (194), and evaluated against three established common fractionation methods of anthocyanins.

82 Anthocyanin

Flavylium cation R1 OH MCX + pH 2 HO O R2 - pK Hydrophobic backbone SO3 O sugar h R OH 1 OH Hydrophobic backbone OH HO O R2 O sugar OH

R1 Hemiacetal Quinonoidal base OH OO R 2 pH 10 O sugar O -

Figure 4.1 The chemical structure of MCX sorbent and an anthocyanin molecule at varied pH conditions.

4.3. MATERIALS AND METHODS

4.3.1 Reagents and standards

Reagent grade tri-fluoroacetic acid (TFA), certified ACS grade HCl (12 N),

HPLC grade methanol, ethyl acetate and acetone for extraction, and HPLC-MS grade water and acetonitrile for chromatography were purchased from Fisher Scientific (Fair

Lawn, NJ). Formic acid (99%) was from Acros Organics (Morris Plains, NJ). Cyanidin-

3-glucoside standard was from ChromaDex Inc. (Irvine, CA). All other chemicals were certified ACS grade from Fisher Scientific.

83 4.3.2 Anthocyanin sources and sample preparation

Different anthocyanin-rich fruit and vegetable sources were tested. Fresh

blueberry and radish were purchased from a local whole food grocery store. Whole

blueberry and peeled radish skin were quickly frozen and fine powdered in a stainless

blender in the presence of liquid nitrogen (195). The powder was extracted three times by

acidified methanol and centrifuged. The combined supernatant was evaporated in a

Büchii rotovaporator at 35 °C and redissolved in 0.1% trifluoroacetic acid (TFA)

acidified water. Black raspberry and strawberry as freeze-dried powders were extracted in

the same manner. Commercially available crude extracts of red cabbage extract (Food

Ingredient Solutions, Llc., Blauvelt, NY) in liquid, as well as bilberry, black currant,

chokeberry, elderberry (Artemis International, Inc., Fort Wayne, IN), grape, and purple

(Polyphenols, Lab., Sandnes, Norway) in spray-dried powder containing ~10% anthocyanins, were diluted in acidified water. All aqueous solutions were prepared to a concentration of ~1 mM anthocyanin.

4.3.3 Fractionation methods comparison

4.3.3.1 Cation-exchange SPE

Two vol of chokeberry and purple corn crude extracts in aqueous solution were

applied to a strong cation-exchange Oasis® MCX SPE cartridge (6cc, 1g sorbent; Waters

Corp., Milford, MA). After washing with 2 vol of 0.1% TFA, the other-phenols fraction

was collected by elution with 2 vol of methanol (0.1% TFA). Subsequently anthocyanins

were eluted with 1 vol of methanol and 1 vol of water/methanol (40:60, v/v), both

84 containing 1% NH4OH. The combined alkaline eluate was immediately mixed with an

aliquot (250 µL) of formic acid (99%) to lower the pH to < 2. Table 4.1 summarizes the purification conditions used in the present study. For each step, 2 vol of specified eluting solvent was used.

® ® ® ® Cartridges Oasis MCX Sep-pak C18 Sephadex LH-20 Oasis HLB

Washing H2O H2O 20% MeOH 15% MeOH

Other-phenols MeOH EtOAc N/Ab EtOAc fraction

60%-100% Anthocyanin fraction MeOH with 1% MeOH 70% MeOH MeOH c NH4OH

Other-phenols MeOH and 70% N/A N/A N/A fraction acetone

a: 0.1% TFA was added to all acidified solvents except in the original C18 method where 0.01% HCl was used as described in the literature. TFA as a volatile acid effectively reduced the accumulation of acid after evaporation. In each step, 2 vol of mobile phase was passed through the cartridges.

b: N/A, not available. This fraction was not designed to be collected.

c: This alkaline eluate was immediately mixed with an aliquot of formic acid to bring the pH to <2. Acidification must be prompt to prevent anthocyanin degradation.

Table 4.1 Mobile phasesa used to elute compounds of interest from the SPE cartridges.

85 4.3.3.2 C18 and HLB SPE

Two vol of chokeberry and purple corn crude extracts in aqueous solution were

® ® loaded onto Sep-pak C18 (6cc, 1g sorbent; Waters Corp., Milford, MA) and Oasis HLB

(3cc, 60mg sorbent; Waters Corp., Milford, MA) SPE cartridges. Purification conditions

were adapted from established procedures (195-198) and presented in Table 4.1. Purified anthocyanin and other-phenol fractions were obtained for HPLC analysis.

4.3.3.3 LH-20 SPE

The Sephadex® LH-20 (Sigma-Aldrich Corp., St. Louis, MO) sorbent is widely

used in open column chromatography, but SPE cartridge with LH-20 sorbent was not

commercially available. LH-20 cartridges were made by inserting 1g of LH-20 material

in between two polyethylene frits in a polypropylene cartridge (6cc) and then soaking in

20% methanol overnight for activation. The conditions for purification are presented in

Table 4.1. Both other-phenols and anthocyanin fractions were obtained for HPLC analysis.

4.3.3.4 HPLC-PDA-MS analysis

All the fractions with 8 duplicates were briefly evaporated in a Büchii

rotovaporator at 35˚C to remove organic solvent, redissolved in 0.1% TFA acidified

water, filtered through 0.45 μm Whatman polypropylene filter, and then analyzed using a

Shimadzu LCMS-2010 EV Liquid Chromatograph Mass Spectrometer (Shimadzu

Scientific Instruments, Inc., Columbia, MD) equipped with a SPD-M20A PDA detector

86 and a single quadrupole electron spray ionization (ESI) MS detector. Separation was

® accomplished on a Symmetry C18 column (3.5 um, 4.6×150 mm; Waters Corp., Milford,

MA). Other chromatographic conditions were as follows: flow rate: 0.8 mL/min; mobile phase: A, 4.5% formic acid in LCMS grade water; B, LCMS grade acetonitrile; gradient:

0-5 min, 2-7% B; 5-20 min, 7-12% B; 20-25 min, 12-17% B; 25-40 min, 17-25% B; 40-

45 min, 25-30%; 45-55 min, 30% (the gradient stopped at 40 min if all anthocyanins in the sample were non-acylated); and after each run 5 min was given to equilibrate the column to initial condition; injection vol: 20 μL. Spectral data (250-700 nm) was collected throughout the run. Elution of anthocyanins was monitored at wavelength 510-

530 nm. Mass spectra were obtained under positive ion condition using SCAN (from m/z

200 to 1000) and Selective Ion Monitoring (SIM) modes. Six channels including m/z 271

(Pelargonidin), m/z 287 (Cyanidin), m/z 301 (Peonidin), m/z 303 (Delphinidin), m/z 317

(Petunidin) and m/z 331 (Malvidin) were monitored in the SIM mode to detect the 6 common anthocyanin aglycone fragments.

4.3.3.5 Purity and recovery evaluation

Concentrations of anthocyanins and total phenols were represented by area under the curve (AUC) in the 510-530 nm and the 250-700 nm max-plots respectively.

Polyphenols usually have absorbance within 260-400 nm range whereas most anthocyanins, including all anthocyanins tested in this study, have max absorbance in the

510-530 nm range (red color). Since the emphasis of the present study was to compare the extent of purity improvement by different methods, converting AUC to absolute concentration is not necessary, and perhaps not possible for all the detectable peaks.

87 Anthocyanin purity was calculated by dividing the AUC of anthocyanin peaks by the

AUC of all peaks in the 250-700 nm max-plot. The current method for computing anthocyanin purity based upon UV-visible chromatogram was adapted from a previous report in which AUC280 nm was used to represent total phenolics (69). It has to be noted that the purities calculated in this study are not the true purities based on weight.

Recovery was calculated by ratio of eluted total anthocyanin/initially loaded total anthocyanin.

4.3.3.6 FT-IR

FT-IR spectrometer analysis of anthocyanin fractions for composition profiling were conducted following a protocol previously developed in our lab (199). The acid- base neutralization product ammonium formate interfered with anthocyanin IR signal, and therefore a desalting procedure was necessary. Aqueous MCX anthocyanin fractions were loaded onto a C18 cartridge, washed with water to remove the salt, and recovered with acidified methanol.

4.3.3.7 Sorbent capacity

Sorbent capacity was determined by 2 means, a visual observation of breaking- through volume and instrumental measurement of total anthocyanins (175) recovered from saturated cartridges. Chokeberry extract was used as representative source of anthocyanins to determine sorbent capacity. The mass of total anthocyanins recovered was calculated based on a Cy-3-glu standard calibration curve (R2 > 0.99).

88 4.3.4 Extended evaluation on a variety of anthocyanin sources

Aqueous solution of bilberry, black currant, blueberry, black raspberry, elderberry, grape, purple carrot, radish, red cabbage and strawberry crude extracts were loaded onto the MCX cartridge or the C18 cartridge (0.1% TFA) and purified as described in Table

4.1 (4 replications). Purity and recovery were calculated with AUC of HPLC-PDA chromatograms as described above.

4.3.5 Optimization for food application

Crude extract of radish skin was purified by the MCX and C18 methods (Table 4.1)

in the attempt to evaluate the removal of adverse aroma. Purified anthocyanin fractions

were rotoevaporated prior to redissolving in water with 0.01% HCl. Due to the adverse

aroma of residual ammonium, the MCX method was conducted with 1% Na2CO3 to substitute NH4OH in eluting solvents. Anthocyanin fractions along with the initial crude extract were filled into separated glass vials with caps and blindly evaluated for adverse aroma in random order.

4.3.6 Statistical analysis

One-way ANOVA or its non-parametric analogue Kruskal-Wallis Test (equal

variance not assumed) was conducted using SPSS (version 13, 2004, SPSS Inc., Chicago,

IL) to compare amounts of total and individual anthocyanins in different treatment groups,

and data are given as means ± SE. When appropriate, significance of differences between

89 means was determined by Tukey's HSD or Dunnett’s T3 (equal variance not assumed).

Differences of P<0.05 were considered significant.

4.4. RESULTS AND DISCUSSION

4.4.1 MCX method development

A novel means for anthocyanin separation based on cation-exchange mechanism

was developed and optimized. Both our preliminary study with an Alltech® SCX cation-

exchange SPE cartridge (Fisher Scientific, Fair Lawn, NJ) and reported literature

revealed that sorbents relying on solely strong cation-exchange mechanism tend to retain

anthocyanins irreversibly and the column capacity was very low (190, 200). Here, we

employed the MCX cartridge, which was packed with a mixed-mode resin combining

both strong cation-exchange mechanism and reversed-phase interaction, a mechanism

commonly employed for isolation of anthocyanins (Figure 4.1). This sorbent is a modified divinylbenzene-vinylpyrrolidone copolymer with a hydrogen atom on benzene substituted by a sulfonic group (201). The divinylbenzene units provide retention via hydrophobic interaction. The benzene structure forms strong π-π interaction with the ring structure of anthocyanins as well as other phenolic compounds. To elute phenolics other

than anthocyanins, acidified methanol was used. Due to the additional ionic interaction

with sulfonic group, positively charged anthocyanin flavylium cations remain adsorbed to

the sorbent. To subsequently elute anthocyanins from the sorbent (194, 201), NH4OH was added to the mobile phase to raise the pH to above 9.5 and deprotonate anthocyanins

90 to quinonoidal base anion or di-anion (30), which repels the negatively charged sulfonic

group on the sorbent to dissociate anthocyanins. According to Brouillard and Dubois (28),

this pH-dependent structural conversion occurs without delay, therefore equilibration

time is not necessary to ensure complete recovery of anthocyanins from the sorbent.

Excessive NH4+ could also compete with cation binding sites on the resin to facilitate dissociation of flavylium cations remaining in the equilibrium. Eluted alkaline anthocyanin solution was immediately acidified to prevent degradation, as anthocyanins are unstable at high pH (52). Due to the neutralization of alkaline and acid, a trace amount of salt was formed. The residual salt is generally not a concern in chemical and biological studies. However, in situation that salts need to be removed, an additional SPE technique such as C18 can be easily employed. We also investigated neutral pH mobile

phase for eluting anthocyanins, but elution was much slower and incomplete because the

hydration from flavylium cation form to hemiacetal form (Figure 4.1) can take up to 15

min to reach equilibrium (28). It is noteworthy that when acylated anthocyanins are

eluted, the alkaline pH (9.5-10) may cause slight saponification, although the eluted

solution is immediately acidified. In this situation neutral or slightly alkaline pH is

recommended to preserve anthocyanin profile.

4.4.2 Fractionation methods comparison

4.4.2.1 SPE sorbent selection and condition optimization

Sixteen SPE materials were explored by Karemer-Schafhalter et al. (191) for

anthocyanin purification, and two sorbents (Amberlite XAD-7 and RP C18) based on

reversed-phase interaction mechanism achieved the best separation and recovery. 91 Currently the often employed SPE purification method for anthocyanins is based on C18, an octadecyl group bonded silica gel (202). Anthocyanins and other-phenols are adsorbed onto the gel by reversed-phase interaction. Water was used to elute hydrophilic compounds including sugar, salt, and simple acids. Acidified methanol was used to elute anthocyanins and other polyphenolic compounds. Oszmianski et al. (196) improved this method by adding an additional cleanup step to remove less polar polyphenolics using ethyl acetate elution. Giusti et al. (195) followed this method and used HCl at low concentration (0.01%) in eluting solvents to prevent acid hydrolysis of anthocyanin, which might occur later when methanol was removed by vacuum evaporation leaving more concentrated acid residue. However, considering the pKh of ~2.8 (29), 0.01% 12N

HCl (pH=3) was not sufficient to maintain the majority of anthocyanins as flavylium

cations, the most stable form for anthocyanins. Insufficient acidity may not ensure

protonation of some low pKa phenolic acids, resulting in incomplete elution of such

interfering compounds with non-polar ethyl acetate. We used 0.1% TFA instead of 0.01%

HCl to optimize the C18 method. The 0.1% TFA aqueous solution provided an acidic

environment approximately 1 pH unit below 0.01% HCl. Yet volatility (bp 72.4°C) of

TFA prevented increase in concentration during vacuum evaporation (15).

The HLB cartridge explored in this study was packed with a recently developed

porous polymeric sorbent sharing the same divinylbenzene-vinylpyrrolidone structure

with the MCX, but without the sulfonic group. Such polymers provide stronger reversed-

phase interaction and they maintain water wettability better than the C18 gel due to the

hydrophilic N-vinylpyrrolidone group (191). Also, the polymer structure is more durable

92 and sustains wider pH range than silica based gel. Therefore HLB was promoted as a

replacement of the C18 gel for polyphenolics purification.

The hydroxypropylated dextran based LH-20 material is a beaded and cross-

linked gel often used in open column chromatography (15). It is a size exclusion gel also

with polar-polar interaction to adsorb analytes (203). LH-20 has been successfully applied to fractionate polymeric proanthocyanins (197, 204, 205); these compounds are difficult to separate with reversed-phase SPE methods. Aqueous methanol was used to elute anthocyanins and aqueous acetone with its carbonyl oxygen as a strong acceptor for

hydrogen bonding was used to subsequently remove polymers (203).

4.4.2.2 SPE methods comparison

The MCX and C18 cartridges appeared to have unbiased selection on all

anthocyanins (Figure 4.2), whereas the HLB and LH-20 cartridges recovered the

hydrophilic anthocyanin (cyanidin-3-galactoside and cyaniding-3-glucoside) less

efficiently. Visual observation (Table 4.2) revealed leaching of color into the aqueous

washing solution in the case of HLB and LH-20 cartridges, explaining the loss of

hydrophilic anthocyanins.

93 Cyanidin-3-galactoside Cyanidin-3-glucoside Cyanidin-3-arabinoside Cyanidin-3-xyloside

HLB

LH20

Original C18

Optimized C18

MCX

Crude Extract

0% 20% 40% 60% 80% 100%

Figure 4.2 Percentage of individual chokeberry anthocyanin peaks after purification.

® ® ® ® Cartridges Oasis MCX Sep-Pak C18 Sephadex LH-20 Oasis HLB

Loading & washing − − a + + +

Phenol/tannin − + + + fraction

Residue on sorbent + − + + + +

a: −, not observable; +, slightly observed; ++, obvious.

Table 4.2 Visually observed anthocyanin color loss during purification of anthocyanin mixtures by the different SPE procedures.

94 Visual observation revealed strong retention of anthocyanins (high break-through

volume) by the MCX and C18 sorbents (Table 4.2). Therefore, capacity was further

determined for these two sorbents quantitatively (n=3). Presumably due to the additional

ionic interaction, MCX had stronger retention of anthocyanins (45.8 mg cyanidin-3-

glucoside equivalent/g sorbent) than C18 (19.0 mg cyanidin-3-glucoside equivalent/g

sorbent). This high capacity of MCX resulted in less organic waste per gram of product.

Another advantage of this strong retention is that even methanol extracts could be

directly loaded and anthocyanins would bind to the MCX sorbent as long as the

environment is acidic. All other methods required evaporation of organic solvents, an

extra step causing delays and possibly resulting in degradation. Enhanced dissociation of

anthocyanins from MCX was observed as compared to the HLB method (Table 4.2).

This was probably attributed to the repelling between negatively charged anthocyanin

molecule and sulfonic group on sorbent (Figure 4.1).

HPLC-PDA chromatograms clearly show that chokeberry anthocyanins (Figure

4.3 A) purified by all methods (Figure 4.3 B-F) contained less impurities than the crude extract (Figure 4.3 G). The MCX method produced almost pure anthocyanins as in the

250-700 nm max-plot besides the 4 anthocyanin peaks the baseline was flat. C18 methods were apparently the next best with respect to purification. Quantitative analysis of chokeberry and purple corn purity and recovery data was conducted by Kruskal-Wallis test, a non-parametric analogue of ANOVA, using SPSS (version 13). The unequal variances indicated varied reproducibility of different cartridges (Figure 4.4). The MCX and C18 cartridges notably generated highly reproducible results. Post-hoc Dunnett’s T3

95 multiple mean comparisons determined that the MCX method resulted in significantly higher purity (P<0.05) of the anthocyanin fraction (Figure 4.4 A) and significantly lower

residue amount anthocyanin (P<0.05) in the other-phenols fraction (Figure 4.4 B) than all other methods. The recovery rates of the MCX method for both anthocyanins (Figure

4.4 C) and other-phenols (Figure 4.4 C) were also the highest or in par with the highest.

The other-phenols represent an important class of compounds of interests for research and food applications. Removal of anthocyanins is favored in certain situations, such as to remove color. With the MCX method, the phenolics fraction was recovered essentially free of anthocyanins.

96 anthocyanin fractions.Anthoc Figure 4.3 compositions areobservedinthe250-700 nm plots, andduetohighsimilarity onlyonechromatogram isshownhere(A). Overall 4, cyanidin-3-xyloside. identities: 1, cyanidin-3-gal Absorbance HPLC-PDA chromatograms ofchokeb . 001. 002. 0035.0 30.0 25.0 20.0 15.0 10.0 5.0 actoside; 2,cyanidin-3-glucoside; 3,cyanidin-3-arabinoside; 1 1 1 1 1 1 1 1 Time (min) 1 1 1 1 1 1 yanins exclusivelyareobser 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 4 4 4 4 4 97 max-plots (B, C, D, E, F and G).Peak max-plots (B,C,D,E,F and erry crudeextractandpurified 250-700nm C Original D) 250-700nm C) Optimized C 250-700nm B) MCX A) Anthocyanins 510-530nm 250-700nm Crude Extract G) 250-700nm LH-20 F) 250-700nm E) HLB ved inthe510-530nm max- 18 18

A) Anthocyanin fraction purity B) Anthocyanin residue in other-phenols 99.4a99.8a 100 50 86.6b 88.5b 88.3b Chokeberry 83.0c 39.3a 80 40 Purple Corn 33.8a 67.9c

57.7d 60 30 51.1e 24.3b 45.7d

Percent 18.4b 40 20

9.3c 20 10 5.8d 3.2c 2.0e 2.2c 0.4d 0 0 MCX Optimized C18 Original C18 LH20 HLB MCX Optimized C18 Original C18 LH20 HLB C) Anthocyanin recovery rate D) Other-phenols recovery rate

95.3a 95.0a 94.8a 100 92.4b 91.4b 100 87.8b

75.8c 78.0c 77.8a 80 80 72.7a 68.9d 70.5b 66.1c

60 55.4d 60 55.9b 54.5b

43.2d

Percent 37.5c 40 40

20 20 9.3d 1.2e 0 0 MCX Optimized C18 Original C18 LH20 HLB MCX Optimized C18 Original C18 LH20 HLB

Figure 4.4 The purity and recovery of anthocyanins and other-phenols based on the AUC of HPLC-PDA chromatograms. Values are the mean ± SE (n=8). Within the same category means with different letters are significantly different at P<0.05.

98 Our optimized C18 method achieved significantly (P<0.05) higher anthocyanin purity from chokeberry and significantly (P<0.05) higher anthocyanin recovery from purple corn than the original C18 method. As shown in Figure 4.3 C and D, at least one major impurity peak was not completely removed by the original procedure but cleared by the optimized procedure. LH-20 and HLB methods lacked adequate capability to separate anthocyanins from other-phenols (Figure 4.3 A), and the recovery of anthocyanins were also not satisfactory (Figure 4.3 C). In the next experiment testing extended anthocyanin sources, only the MCX and optimized C18 methods were evaluated.

4.4.2.3 Confirmation of purity improvement with supplementary analytical methods

Purity was quantified based on absorbance in the 250-700 nm range. Phenolic compounds and many other organic compounds belong to this class. However, this method has a limitation in that compounds without absorbance in this UV-visible range are not detectable (16). Therefore, two additional analytical methods were employed to confirm the purity. Total ion concentration recorded by a MS detector showed only anthocyanin peaks for the MCX anthocyanin fraction (Figure 4.5), indicating absence of impurities; yet, multiple noise peaks were present with other isolation methods.

Hierarchical cluster analysis (HCA), an unsupervised clustering analysis was carried out on the infrared spectra to determine compositional similarity (199). Anthocyanin fractions (de-salted) with similar compositions were grouped close together. The decreasing similarity from the crude extract to the HLB, LH-20, C18, and MCX eluents indicated increasing purity (Figure 4.6).

99 Total Ion Concentration 3.0 Blank Cyanidin-pentoside 2.0 Cyanidin-hexoside 1.0 0.0 3.0 MCX 2.0

1.0

0.0 3.0 Optimized C18 2.0

1.0

0.0

3.0 Original C18 2.0

1.0

0.0

Ion Concentration (million) Ion Concentration HLB 2.0

1.0

0.0 LH20 2.0

1.0

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 Time (min)

Figure 4.5 Impurities in chokeberry anthocyanin fractions as indicated by the appearance of noise peaks in the MS chromatogram. The MS was operated at positive mode.

100 A) Chokeberry anthocyanin MCX

Optimized C18

Original C18

LH20

HLB

Crude Extract B) Purple corn anthocyanin MCX

Optimized C18

Original C18

LH20

HLB

Crude Extract

1 0 Similarity

Figure 4.6 HCA dendrograms of chokeberry and purple corn extracts. Compositionally similar samples are clustered together. Similarity of 1 indicates identical composition and 0 indicates the largest difference in the sample set.

101 4.4.3 Extended evaluation on a variety of anthocyanin sources

Besides the chokeberry and purple corn being elaborated above, we further tested

(n=4) other common anthocyanin-rich fruits and vegetables. We successfully purified bilberry, black currant, blueberry, elderberry, purple carrot and red cabbage anthocyanins to greater than 99% purity (Figure 4.7). Black raspberry, strawberry, grape and radish anthocyanin purities were also significantly (P<0.05) improved as compared to the optimized C18 method using Mann-Whitney U test (non-parametric analogue of Student’s t-test). It is reasonable to speculate that our method may also be applicable for many other anthocyanin-rich fruits and vegetables as well. Overall recovery of anthocyanins from all 12 commodities (n=56) tested in this study was compared between the MCX and optimized C18 methods using Student’s t-test. The recovery by the MCX method

(93.6±0.55%) was not significantly different (P>0.05) from that by the C18 method

(93.8±0.36%), and both were exceptionally satisfactory.

Sequential coupling of more than one type of resin was explored. As mentioned above, anthocyanins present in organic solvents can be directly loaded onto the MCX resin, and this makes it easy to directly load organic eluents from other resins. When black raspberry extract was sequentially purified by C18 and MCX columns, the anthocyanin purity was successfully increased to almost 100%.

102 purity in all commodities except purplecarrot (P>0.05). purity except allcommodities in Figure 4.7 MCX Optimized C18 The purity (based on the AUC of HPLC-PDA chromatograms) of anthocyanins purified from 10 additional commodities. Values arethe commodities. Values purified ofadditional from anthocyanins 10 purityThe AUC of onchromatograms) theHPLC-PDA (based

99.7 99.9 99.8 99.4 99.0 97.9 97.4 99.2 99.2 100 94.4 95.5 93.8 94.2 90.9 91.7 85.6 81.1 80 76.8

66.5

60 47.0

40 103

20 Anthocyanin Fraction Purity (%) (%) Purity Purity Fraction Fraction Anthocyanin Anthocyanin

0

mean ± SE (n=4). MCX achieved significantly higher (P<0.05) (P<0.05) higher significantly achieved MCX (n=4). SE ± mean bilberry black currant blueberry black raspberry elderberry grape purple carrot radish red cabbage strawberry

Figure 4.7 The purity (based on the AUC of HPLC-PDA chromatograms) of anthocyanins purified from 10 additional commodities. Values are the mean ± SE (n=4). MCX achieved significantly higher (P<0.05) purity in all commodities except purple carrot (P>0.05).

4.4.4 Optimization for food application

Radish extract, a mixture providing stable color and representative hue to synthetic food colorant Red #40, was particularly studied with respect to removal of its strong adverse aroma (173). Our preliminary test determined that the crude extract had high intensity odor whereas the C18 and MCX purified fractions only had moderate and slight odor, respectively. Higher radish anthocyanin purity (85.6%) obtained by the MCX method (Figure 4.7) than the C18 method (47.0%) correlated well with the removal of odor. We propose that by optimizing the pH of eluting solvents anthocyanins could be even further separated from positively charged interfering compounds. For instance, nitro-containing odorants have positive charge at acidic condition but their pKa are higher than the pKh anthocyanin flavylium cations. Adjusting the polarity of eluting solvents, for example, using ethyl acetate, acetone and hexane, may also help to remove more non- polar aromatic compounds extracted from radish.

For the purpose of food application, less toxic solvents and chemicals are proposed. We found ethanol to work similarly to methanol. Na2CO3 will be neutralized by citric acid instead of formic acid. The resulted residual amount of sodium citrate is a common ingredient found in foods.

4.5. CONCLUSIONS

An innovative mixed-mode cation-exchange anthocyanin isolation method was successfully developed. This method was superior to the commonly used SPE methods 104 regarding purity, recovery, sorbent capacity, low cost, simplicity of manipulations, and

organic waste generated per gram of product. Therefore it could become a rapid, low cost,

and high throughput method to provide high-purity anthocyanins in research labs for

minimized interference from other bioactive compounds. A scale-up production based on

this new technique may provide the food colorant industry and nutraceutical industry a

practical way to separate high quality anthocyanins from fruits and vegetables, and even from industry by-products.

4.6. ACKNOWLEDGEMENTS

This work was supported by Grant 2004-35503-15190 from the USDA NRI

Competitive Grants Program. We also thank Artemis International, Inc. (Fort Wayne, IN),

Polyphenols, Lab. (Sandnes, Norway) and Food Ingredient Solutions, Llc. (Blauvelt, NY)

for providing the anthocyanin crude extracts.

105

CHAPTER 5

5. IMPACT OF SMALL INTESTINAL β-GLUCOSIDASE AND DIETARY LACTASE ON ANTHOCYANIN DIGESTION a

5.1. ABSTRACT

Anthocyanins are natural pigments with high antioxidant capacity and potential

health benefits. Studies suggest that their stability on the GIT may greatly impact their

bioactivity. Previous studies have shown selectively degradation of anthocyanin

glucosides in animal small intestinal lumen. Such evidence points to the possibility of

enzymatic deglycosylation by digestive enzymes. In this study we attempted to evaluate

the impact of β-glycosidase that can be present in the small intestine epithelium on

anthocyanin stability. Cell-free extract of pig small intestinal mucosa and crude extract of

lactase containing foods/supplement were incubated with highly purified anthocyanins at

physiological termperature and pH. Selective degradation of anthocyanin glucosides was

observed in the presence of small intestinal mucosa cell-free extract, wherease selective

degradation of anthocyanin galactosides occurred in the presence of lactase supplement

a Jian He1, Mark L. Failla2, and M. Mónica Giusti1, 1Dept. of Food Science & Technology, 110 Parker Food Science Bldg., 2015 Fyffe Ct., 2Dept. of Human Nutrition, 325 Campbell Hall, 1787 Neil Ave., The Ohio State Univ., Columbus, OH 43210 106 extract. In both cases cyanidin glycosides were hydrolyzed the fastest among all the five common anthocyanidins examined while malvidin glycosides were the most resistant to enzymatic degradation. Kinetic parameters Vmax and apparent Km were determined for

both enzyme extracts using their highest affinity substrates, respectively. Our results

demonstrate that small intestinal mucosal enzyme and lactase supplement affect

anthocyanin stability. Bioactivity and bioavailability studies of anthocyanins should not

overlook the influence of enzymatic deglycosylation, which may considerably alter

anthocyanin profiles in the GIT.

Key words: anthocyanin; deglycosylation; β-glucosidase; β-galactosidase; lactase;

stability; biotransformation; small intestine

5.2. INTRODUCTION

Cellular, animal and epidemiological studies have suggested association between

anthocyanin intake and chronic diseases prevention. However, the bioactive forms of

anthocyanins or anthocyanin metabolites is still in dispute. Studies have repeatedly

shown that intact anthocyanins undergo metabolism after absorption (4), but limited

information is available on anthocyanin transformation in the GIT lumen before uptake

into epithelial cells. It is critical to understand the stability of anthocyanins and possible

transformation in the GIT for several reasons. First, the concentration of accessible anthocyanins in the GIT lumen is positively associated with absorption (7). Second,

metabolism of anthocyanins in the GIT may produce bioactive phenolic derivatives that

107 have health benefits (4-7). Third, stable anthocyanins remaining intact in the GIT may

exert antioxidant, anti-inflammary, and chemopreventive activity through direct contact

with GIT epithelial cells (8, 13).

Our previous studies on anthocyanins in rat GIT content (13), as well as extensive

literature on the metabolism of flavonoids (150), revealed that LPH, a β-glucosidase

residing on the brush boarder membrane of small intestine, may hydrolyze anthocyanins

containing glucose sugar moiety and result in loss of anthocyanins in the GIT. Further

investigation using isolated small intestinal enzyme extract and highly purified

anthocyanin substrate is merited to better explain the proposed enzymatic activity

towards anthocyanins. In the present study, pig small intestinal mucosa extract was used

to examine β-glycosidase activity with various anthocyanins.

Exogeneous enzymes responsible for hydrolysis of β-glycosidic linkage are also

likely to affect anthocyanin stability. As the deficiency of LPH causes lactose intolerance

(206), many individuals with low LPH activity rely on exogeneous lactases to hydrolyze

lactose. Fungal lactase in supplement form is widely used. Dairy foods rich in lactose can

be fermented or treated with fungal lactase to substantially reduce lactose content. Here

we also investigated the effect of both, the lactase supplement and the residual dietary

lactase on cleavage of anthocyanin 3-O-β-glycosidic bonds in the digestive tract.

5.3. MATERIALS AND METHODS

108 5.3.1 Chemicals and materials

Molecular biology grade dithiothreitol (DTT) was purchased from Promega

Corporation (Madison, WI). Protease inhibitor cocktail was from Sigma (St Louis, MI).

Reagent grade tri-fluoroacetic acid (TFA), high performance liquid chromatography

(HPLC) grade acetone and methanol for extraction, and HPLC-mass spectrometer (MS) grade water and acetonitrile for chromatography were purchased from Fisher Scientific

(Fair Lawn, NJ). HEPES buffer (>99%), D-gluconolactone (>99%) and formic acid (99%) were from Acros Organics (Morris Plains, NJ). Cyanidin (Cy) aglycone was prepared from Cy-3-glu standard (ChromaDex Inc., Irvine, CA) by acid hydrolysis (19).

Protocatechuic acid standard was from MP Biomedicals (Solon, OH). α-D-lactose monohydrate and all other chemicals were ACS grade from Fisher Scientific. The human colon cancer cell line HT-29 was purchased from the American Type Culture Collection

(Rockville, MD).

5.3.2 Purification of anthocyanins

Blueberry and grape anthocyanins were selected as substrates in the present study because of their representativeness. Blueberry contains 5 common anthocyanidins each linked to 3 common glycosides. Grape contains 5 common anthocyanidins each linked to a glucoside. Highbush blueberry and crimson seedless red grape were purchased from a local whole food grocery store. Blueberry or red grape (28 g, equivalent to 1/5 serving) was crushed by a spatula and then mixed with 20 mL of methanol (0.1% TFA). The mixture was then briefly homogenized using a handheld tissuemiser (Fisher Scientific,

109 Fair Lawn, NJ) briefly followed by sonication in a Fisher FS30 ultrasonic bath for 20 min.

The slurry was then centrifuged at 1800×g for 10 min and the supernatant was collected.

The pellet was re-extracted three times with 25 mL of methanol (0.1% TFA) for further

removal of pigmentation from the matrix. The pooled supernatant was directly applied

onto a strong cation-exchange Oasis® MCX SPE cartridge (6cc, 1g sorbent; Waters Corp.,

Milford, MA) and subsequently purified according to the procedure described in Chapter

4. Briefly, the cartridge was washed with 2 vol of DD water and methanol, both

containing 0.1% TFA, and anthocyanins were eluted by 2 vol of aqueous methanol

containing 1% NH4OH. The alkaline eluent was immediately mixed with an aliquot of formic acid to decrease pH to acidic region. This aqueous methanol mixture containing highly purified anthocyanins, salts and residual formic acid was then diluted with water

® to decrease methanol concentration to <10%, and applied to a Sep-pak C18 cartridge

(20cc, 5g sorbent; Waters Corp., Milford, MA) for further removal of salts. After

washing with 2 vol of DD water (0.01% HCl), the anthocyanins were eluted with 1 vol of methanol (0.01% HCl), evaporated in a Büchii rotovaporator at 35˚C to almost dryness, and then diluted to 20 mL with DD water. As such, 1mL of the extract contained anthocyanins equivalent to 1/100 serving of the corresponding fruit.

Black raspberry was studied to compare with the in vivo results obtained in

Chapter 3. Lyophilized black raspberry powder was a gift from Dr. Laura Kresty, OSU

Comprehensive Cancer Center. Acetone/water extraction of the powder has been

described in Chapter 3. An aliquot of the extract containing 25 mg of total anthocyanins

as determined by pH differential method (175) was further purified using MCX and C18

110 cartridges with the same procedure described above. Eventually the black raspberry

anthocyanins were dissolved in 25mL of DD water.

Chokeberry extract was selected for lactase treatment and further evaluation in

cell culture because of its high Cy-3-gal content, a discovered substrate for dietary lactase.

Commercially available chokeberry extract powder (containing 10% anthocyanin) was

donated by Artemis International, Inc. (Fort Wayne, IN). Following the same MCX/C18 purification procedure above, 540 mg of chokeberry powder dissolved in acidified water was purified and eventually redissolved in 10 mL of DD water.

5.3.3 β-glucosidase activity in porcine small intestinal mucosa

5.3.3.1 Preparation of cell-free extract

Fresh mid-jejunum samples from two 8-month old pigs were generous gifts from

Dr Henry N. Zerby, OSU Department of Animal Sciences. The samples were opened longitudinally on an ice-cold glass plate and washed with ice-cold Phosphate Buffered

Saline (PBS, 50mM, pH 7.0). The mucosa layer was scraped using glass slides and transferred to a tube containing 10 mL of ice-cold HEPES buffer (25 mM HEPES buffer, pH 7.0, containing 154 mM NaCl, 1 mM EDTA, 5 mM DTT, and a 400 fold diluted protease inhibitor cocktail). Mucosal cells were disrupted using a handheld glass homogenizer (15 mL, up and down 15 ×) followed by further homogenization with a tissuemizer for 10 s, and then diluted to 25mL with ice-cold HEPES buffer. Aliquots (1 mL) of the crude extract was transferred and distributed into 5 micro-centrifuge tubes.

The remainder was centrifuged at 10,000×g, 4°C for 40 min, and the resultant as cell-free

111 supernatant was aliquoted into 12 micro- centrifuge tubes (1 mL each), pellets saved.

Samples were frozen at –80°C until enzymatic assays. Frozen samples were thawed in

cold water bath and inverted a few times to mix before use.

5.3.3.2 Protein assay

The protein contents of crude enzyme extract and cell-free extract were measured

in triplicate using a Bio-Rad protein assay kit (Hercules, CA) following the

recommended procedures. Bovine serum albumin (BSA) was used as protein standard.

5.3.3.3 Reference velocity

One hundred µL of the cell-free supernatant prepared from homogenized mucosal

epithelium was diluted in 1 mL of PBS (50 mM, pH 6.5) containing 584 mM lactose

(close to saturation concentration). Initial rate of lactase hydrolysis was measured based

on the velocity of glucose released. Samples in micro-centrifuge tubes were incubated at

37°C in water bath. After 45min, the tubes were immerged in boiling water bath for 2.5

min to stop the reaction, followed by centrifugation at 16,000×g for 5 min (4°C).

Aliquots (0.5 mL) of the supernatant were analyzed using a GAGO-20 glucose kit (Sigma,

St Louis, MI) following instructed procedures. The reactions were conducted in triplicate.

5.3.3.4 Confirmation of enzymatic activity

One micro-centrifuge tube of thawed cell-free supernatant was heated in boiling

water for 2.5 min to inactivate enzymes prior to centrifugation (16,000×g for 5 min at

4°C) for removing protein precipitate. Heat-inactivated and non-heated enzyme extracts 112 (300 µL) were added to polystyrene tubes (15 mL size) containing 10 µL of purified black raspberry anthocyanins (2.10 mM) and 2.19 mL of PBS (50mM, pH 6.5) alone or

PBS with LPH inhibitors (584 mM α-D-lactose or 0.1M D-gluconolactone) (Table 5.1).

These compounds were used to confirmed that hydrolysis of anthocyanins to the aglycone was due to LPH activity. D-gluconolactone is a known potent inhibitor of the lactase active site of LPH (147), whereas α-D-lactose is a substrate of LPH.

Black raspberry Heat- Active Buffer anthocyanins inactivated enzyme pH 6.5 (µL) extract (µL) extract (µL) (2.190 mL) Control 10 300 ⎯ PBS

Active enzyme 10 ⎯ 300 PBS 200 g/L Competitive inhibition 10 ⎯ 300 α-D-lactose in PBS 0.1 M Complete inhibition 10 ⎯ 300 D-gluconolactone in PBS

Table 5.1 Composition of mixtures used to test the effect of pig small intestinal enzymes on black raspberry anthocyanins.

Triplicate samples of each mixture were incubated in a water bath at 37°C avoiding light. After 0, 15, 30, 60, and 90 min, the tubes were gently mixed and an aliquot (0.5 mL) was removed from each tube and immediately mixed with 1 mL of ice- cold 10% formic acid to terminate enzymatic reaction. Samples were then centrifuged at

16,000×g for 5 min (4°C) and the supernatants were directly analyzed by HPLC. 113 5.3.3.5 Determination of apparent Km and Vmax

Tubes containing 100 µL of cell-free supernatant and 390 µL of PBS (50 mM pH

6.5) were pre-equilibrated in water bath to 37°C. Various concentrations (50-2000 mg/L)

of Cy-3-glu (10 µL) were added into each tube, mixed, and placed in 37°C water bath.

After designated times, 1 mL of 10% formic acid was added to terminate the reaction and

samples were centrifuged at 16,000×g for 5 min (4°C). Supernatants were directly

analyzed by HPLC. The experiment was repeated three times.

5.3.3.6 Substrate specificity

An assay was carried out on purified grape anthocyanins to determine substrate

specificity. Five µL of the purified grape anthocyanins (1.37 mM) were mixed with 0.5

mL of the cell-free supernatant and 0.995 mL of PBS (50mM, pH 6.5) and incubated in water bathed at 37°C avoiding light. The assay was also conducted in parallel with 0.1M

D-gluconolactone in PBS. Aliquots (0.5 mL) were removed from each tube after 0, 30,

and 60 min, and mixed with 1mL of ice-cold 10% formic acid to terminate enzymatic

activity. Samples were centrifuged at 16,000×g for 5 min (4°C) and the supernatants

were directly analyzed by HPLC. The assay was conducted with duplicate samples.

5.3.4 Lactase from foods

5.3.4.1 Extraction of crude enzyme

An aliquot (1/10 serving) of either a lactase supplement containing 9000 FCC

lactase units/serving, yogurt with live-culture (22.5 g), or lactase treated skim milk (24 114 mL) was mixed with PBS (50mM, pH 6.5) to ~70 mL followed by hand-shaking the containers for 1 min. The mixtures were centrifuged at 10,000×g for 40 min (4°C) and the supernatants were diluted to 100 mL with PBS. As such, 1 mL aliquots represented

1/1000 serving of the lactase suppliment, yogurt or milk. All samples were either used fresh or frozen in separate tubes at –18°C. Prior to enzymatic assay, aliquots of the frozen extract were thawed in cold water bath and inverted several times to mix.

5.3.4.2 Confirmation of enzymatic activity

One micro-centrifuge tube of thawed lactase supplement extract was heated in boiling water for 2.5 min to inactivate the enzymes prior to centrifugation at 16,000×g for

5 min (4°C) to remove protein precipitate. The heat-inactivated and non-heated extracts

(1 mL) were subsequently mixed in polystyrene tubes (15 mL size) with 100 µL of purified blueberry anthocyanins (2.52 mM) and 1.1 mL of PBS (50mM, pH 6.5) with or without α-D-lactose, a known substrate of lastase (Table 5.2). The mixtures were incubated in a water bath at 37°C avoiding light in triplicate. An aliquot (0.5 mL) was removed from each tube at 15 min and added to 1 mL of ice-cold 10% formic acid to terminate enzymatic activity. Samples were centrifuged (16,000×g for 5 min at 4°C) and the supernatants were analyzed by HPLC.

115 Blueberry Heat- Active Buffer anthocyanins inactivated enzyme pH 6.5 (µL) extract (mL) extract (mL) (1.1 mL) Control 100 1 ⎯ PBS

Active enzyme 100 ⎯ 1 PBS 200 g/L Competitive inhibition 100 ⎯ 1 α-D-lactose in PBS

Table 5.2 Compositions of buffered solutions used to investigate the effect of lactase supplement on blueberry anthocyanins.

5.3.4.3 Substrate specificity assays

Five hundred µL of the purified blueberry anthocyanins (equivalent to 1/200 serving) were mixed with 5 mL (equivalent to 1/200 serving) of the thawed lactase

supplement extract or freshly prepared yogurt and milk extract and incubated at 37°C avoiding light. An aliquot (0.5 mL) was removed from each tube after 1, 5, 10, 20, 30, 60,

90, and 150min, and immediately mixed with 1mL of ice-cold 10% formic acid to terminate enzymatic activity. Samples were then centrifuged (16,000×g for 5 min at 4°C) and the supernatants were directly analyzed by HPLC. The assays were repeated in triplicate.

5.3.4.4 Determination of apparent Km and Vmax

Six micro-centrifuge tubes each containing 10 µL of freshly prepared lactase

supplement extract and 390 µL of PBS (50 mM pH 6.5) were pre-equilibrated in water

116 bath to 37°C. A series dilution of purified chokeberry anthocyanins (100 µL) were pipetted into each tube to give a range of 38.4 – 570.3 mg/L of cy-3-gal concentrations as determined by HPLC. Samples were incubated in 37°C water bath for 3 min prior to addition of 1 mL of 10% formic acid to terminate enzymatic reaction. Samples were centrifuged (16,000×g for 5 min at 4°C) and the supernatants were directly analyzed by

HPLC. The experiment was repeated in triplicate.

5.3.4.5 Anti-proliferative effect on colon cancer cell line

Two mL of purified chokeberry extract (~10 mg anthocyanins) were mixed with

10mL of freshly prepared lactase supplement extract (treatment) or 50mM, pH 6.5 PBS

(control). The mixed solutions were incubated in water bath at 37°C avoiding light for 45 min. An aliquot (0.5 mL) was removed from each solution and immediately mixed with

1mL of ice-cold 10% formic acid to be analyzed by HPLC. The rest of the solutions were cooled in ice-water bath and frozen at -80°C until being used for cell culture.

HT-29 cells were cultured in McCoy’s 5A medium (GIBCO Laboratories, Gran

Island, NY) supplemented with 10% fetal bovine serum (FBS) and incubated in humidified atmosphere (5% CO2, 95% O2) at 37°C. Purified chokeberry was dissolved in medium to give a final concentration of 100 μg/mL and the enzyme treated chokeberry extract was used at equivalent volume. Two fold dilutions were made. Negative control wells contained cells in medium without anthocyanins while blank wells contained only media.

117 HT-29 cells were plated in 24-well cell microtiter plates at a density of 3×104 cells/mL. Cells were incubated for 24 h in culture medium. Growth medium was aspirated and cells were treated for 48 h with the different test media. After incubation, growth inhibition was measured using a modified Sulforhodamine B (SRB) assay. Briefly, test media was removed and cells were fixed by adding 500μL of 10% trichloroacetic acid (TCA) and plates were incubated for 1 h at 4°C. After that, plates were washed with water and air-dried. 500μL of 0.4% SRB solution was applied to each well and plates were incubated for 30 min. Plates were washed using 1% acetic acid and allowed to dry.

Bound SRB was solubilized by adding 1mL of 10mM Tris Base Solution. Optical density

(OD) was read at 565nm. The cell culture experiment was repeated in duplicate.

Growth inhibition was calculated as follow:

OD OD − OD G rowth Cell Growth 100%Inhibition −= ( sample day 0 × )100 ODnegative control − ODday 0

5.3.5 HPLC-MS analysis of anthocyanins

Samples were analyzed using a Shimadzu LCMS-2010 EV HPLC-MS (Shimadzu

Scientific Instruments, Inc., Columbia, MD) equipped with an SPD-M20A photodiode

array (PDA) detector and a single quadrupole electron spray ionization (ESI) MS detector.

® Separation was accomplished on a Symmetry C18 column (3.5 μm, 4.6×150 mm; Waters

Corp., Milford, MA) with a flow rate of 0.8 mL/min. Mobile phase conditions were as

follows: A, 4.5% formic acid in HPLC-MS grade water; B, HPLC-MS grade acetonitrile;

gradient: 0-5 min, 2-7% B; 5-20 min, 7-12% B; 20-25 min, 12-17% B; 25-40 min, 17-

118 25% B. After each run, the column was equilibrated for 5 min under the initial condition.

Spectral data (250-700 nm) were collected during the entire separation procedure. When

MS was coupled to the HPLC for identification, spectra were obtained under positive ion condition using total ion scan (from m/z 200 to 1200) and selective ion monitoring (SIM) modes. Six channels including m/z 287 for Cy and m/z ratios for other common anthocyanin aglycones were monitored in the SIM mode. Anthocyanin concentrations were calculated using areas under curve (AUC) of HPLC chromatograms at 520 nm and a standard calibration curve.

5.3.6 Calibration curve

Commercially available Cy-3-glu standard (10 mg) was dissolved in 10 mL of double distilled water containing 0.1% TFA, and a series of dilutions were prepared to generate a standard curve (R2 > 0.99). All anthocyanins analyzed fell within the range of the standard curve and were expressed as Cy-3-glu equivalents by weight. The amount of anthocyanins was calculated using the AUC of individual anthocyanin peaks at 520 nm and this calibration curve.

5.3.7 Statistical analysis

One-way ANOVA was conducted using SPSS (version 13, 2004) to compare amounts of individual anthocyanins at different time points, and data are given as means

± SE. When appropriate, significance of differences between means was determined by

Tukey's HSD. Differences of P<0.05 were considered significant. Kinetic parameters Km and Vmax were calculated using SigmaPlot (version 10, 2006).

119 5.4. RESULTS AND DISCUSSION

5.4.1 Lactase activity of pig small intestinal mucosa extract

Homogenate prepared from porcine small intestinal mucosa contained 4.97 ± 0.06 g protein/L, while the cell-free extract contained 4.53 ± 0.12 g protein/L. Therefore, the majority of the protein was present in the supernatant but not in the pellet. This is not surprising because centrifugation of homogenized cells at 10,000×g for 40 min was expected to remove nucleus and mitochondria, leaving most of the cytosolic and membrane bound in the aqueous supernatant. Similarly, enzymatic assay using lactose as substrate revealed 22.7 ± 0.15 nmol/min·mg protein lactase activity in the crude extract and 24.0 ± 0.08 nmol/min·mL activity in the cell-free extract. Thus, the supernatant maintained the majority of enzymatic activity in the crude extract. We observed that repeated freezing and thawing significantly lower enzymatic activity. Thus, each thawed aliquot of enzyme extract was discarded after single use. It is noteworthy that the β-glucosidase activity measured in the present study may be underestimated because samples were analyzed after thawing.

5.4.2 β-glucosidase activity on anthocyanins

Stability of black raspberry anthocyanins in rat small intestine was examined and selective degradation of Cy-3-glu was observed (Chapter 3). LPH residing on the brush boarder of small intestine was proposed to be responsible for the selective hydrolysis of

Cy-3-glu. In the present study, cell-free extract of isolated pig small intestinal mucosa layer was investigated with highly purified black raspberry anthocyanins as substrates.

120 Purified anthocyanins eliminate the possibility of interference from numerous other

compounds present in the plant extracts, and therefore improved the reliability of the enzymatic assay. For the four major anthocyanins present in black raspberry (Cy-3-sam,

Cy-3-glu, Cy-3-(xyl)-rut, and Cy-3-rut), only Cy-3-glu was significantly (P<0.05) degraded (by 55.3% in 90 min) by the cell-free extract (Figure 5.1 A). This degradation

was attributed to enzymatic activity rather than chemical instability of Cy-3-glu, as the

amount of Cy-3-glu was not affected during incubation with heat-inactivated cell-free

extract (Figure 5.1 B). To confirm that the enzymatic activity was due to β-glucosidase, a

membrane bound β-glucosidase inhibitor D-gluconolactone was used. Addition of 0.1 M

D-gluconolactone in the PBS buffer completely ceased selective degradation (P>0.05) of

Cy-3-glu (Figure 5.1 C), confirming that β-glucosidase was the catalytic entity for the

hyfrolysis.

Similarly, some flavonoid glucosides could be substrates for β-glucosidases

present in the small intestine cell-free extract (150), while the other three types of

glycosides were not substrates for these enzymes. It is noteworthy that Cy-3-sam, Cy-3-

(xyl)-rut and Cy-3-rut all have a glucose moiety on the C-3 position of aglycone, but

other molecules linked to the glucose moiety likely prevente approximate binding to the

active site of β-glucosidase. Nemeth et al. (181) purified human small intestinal β- glucosidases and determined that in the epithelial cell the membrane bound LPH was responsible for the observed β-glucosidase activity towards flavonoid glucosides.

Although no attempt was made to identify the specific enzyme responsible for anthocyanin hydrolysis in the current study, we speculated that LPH was responsible for

121 deglucosylation of Cy-3-glu. Lactose, a common substrate for LPH, was investigated at

nearly saturated concentration for competitive inhibition on Cy-3-glu hydrolysis. Slight

but significant (P<0.05) decrease (16.3%) of Cy-3-glu was observed in 90 min (Figure

5.1 D). Clearly lactose competed with Cy-3-glu for the active site of β-glucosidase.

However, caution must be taken in that part of the β-glucosidase activity might come

from luminal microbial β-glucosidase adhering to the mucosal layer and cytosolic β-

glucosidase present in mucosal cells.

In all assays, anthocyanin deglycosylation was accompanied by the appearance of

anthocyanidins. However, the concentrations of the anthocyanidins were dependent on

the delay between incubation and HPLC analysis due to the labile characteristic of

anthocyanidins. As a result quantification of the anthocyanidins was not attempted.

122 Cy-3-sam Cy-3-glu Cy-3-(xyl)-rut Cy-3-rut 100 100

90 90

80 80

70 70

60 60

50 50 A) LPH B) Inactive LPH 40 40 100 03060910000306090

90 90 Anthocyanin amount % amount Anthocyanin 80 80

70 70

60 60

50 50 C) LPH + D-gluconolactone D) LPH + lactose 40 40 03060900306090 Time (min)

Figure 5.1 Degradation of purified black raspberry anthocyanins during incubation with pig small intestinal cell-free extract in the presence and absence of β-glucosidase inhibitors (n = 3). The initial amount of each anthocyanin was considered as 100%.

123 5.4.3 Calculation of apparent Km and Vmax for intestinal β-glucosidase

β-glucosidase mediated degradation of Cy-3-glu with Michaelis-Menten kinetics

(Figure 5.2). Apparent Km and maximum velocity (Vmax) were calculated from the

substrate saturation curve for comparison with values reported in literature (Table 5.3).

Km is equivalent to the substrate concentration that reaches half velocity of Vmax, and smaller Km represents higher affinity of enzyme towards its substrate. As summarized in

Table 5.3, Cy-3-glu has comparable apparent Km with structurally similar flavonoid

glucosides, yet much lower than lactose. Therefore, Cy-3-glu is expected to be the

preferred substrate for the small intestinal epithelial β-glucosidases in the presence of

lactose.

V Apparent K Substrate max m (µmol/min·g protein) (µM)

Cy-3-glu 1.21 ± 0.14 29.3 ± 9.0

Quercetin-4’-glu (147) ⎯ 37 ± 12

Genistein-7-glu (147) ⎯ 14 ± 5

Lactose (206) 23.96 8800 ± 300

Table 5.3 Kinetic constants for the hydrolysis of LPH substrates. 124 1.2

1.0

0.8

0.6

0.4 Velocity (umol/min*g protein) Velocity

0.2

0.0 0 20406080100

Cy-3-glu (uM)

Figure 5.2 Saturation curve for the small intestinal mucosal β-glucosidase showing the relationship between Cy-3-glu substrate concentration and rate of Cy-3-glu hydrolysis. Markers with error bars represent means ± SE (n = 3).

125 The concentration of Cy-3-glu required in the reaction to reach half of Vmax was

29.3 µM, which can be easily exceeded by a single serving of red grapes (44). Based on the Vmax and the total mucosal protein extracted from the segment (~20 cm in length) of

weaned pig small intestine, we estimate that in 30 min approximately 0.5-1 mg of Cy-3-

glu could be hydrolyzed within this segment. Although the estimated efficiency of

anthocyanin glucoside hydrolysis may appear to be insignificant considering an average

daily consumption of 12.5 mg anthocyanins, the impact of this enzymatic reaction should

not be underestimated. As discussed above, the cell-free extract was thawed and had lost

a significant part of its enzymatic activity. Therefore we probably underestimated the

maximum rate of degradation. Also, tissue was obtained from 8-month old pigs. Pigs at this age are not expected to express LPH at high level. Pre-weaned animals and humans

express high level of LPH to digest lactose in milk (150). Hence, the degradation of

anthocyanins is expected to be much greater for newly weaned populations. Some

populations, particularly individuals of Northern European origin, maintain high level of

LPH throughout adult life (182). In those people perhaps metabolism of anthocyanin

glucosides differs greatly with normal subjects. High LPH activity may be induced by

diet (182). For example, milk-drinking population have higher LPH activity than non-

milk-drinking population, and high α-saccharides diet effectively elevates LPH in rats

(207). In our previous study, rats after weaning were fed starch-rich diet, thus little

anthocyanin glucosides remained in the GIT segments below the small intestine, causing

over 50% loss of total anthocyanins (13).

126 5.4.4 Substrate specificity of small intestinal β-glucosidase

The impact of anthocyanin structure on the extent of enzymatic hydrolysis was evaluated. A substrate specificity assay was performed with purified grape anthocyanins

(Figure 5.3 A). Grape contains 5 of the most common anthocyanidins conjugated to glucose at C-3 position (Table 5.4). The identities of grape anthocyanins were determined by comparing retention times and UV-visible spectra to known standards, and by comparing m/z ratios of whole molecular ions and fragments to established values

(208). The profile of grape anthocyanins was not altered when incubated with cell free extract in the presence of D-gluconolactone (Figure 5.3 B). Incubation with LPH alone for 30 min resulted in loss of 38.1% total anthocyanins (Figure 5.3 C, Table 5.4) and 20-

50% loss of Cy-3-glu, Pn-3-glu and Dp-3-glu, while Pt-3-glu and Mv-3-glu were quite stable. Further incubation to 60 min resulted in 19% loss of Pt-3-glu, while Mv-3-glu remained almost unchanged. Such substrate specificity may be explained by the B ring structure of the anthocyanidins (Figure 5.4). Mv is the only anthocyanidin that has a methoxyl group on the 5’ position of B ring. Due to the proximity between the 5’ position and the sugar moiety on the C-3 position, the bulk sized hydrophobic methoxyl group may limit binding to the enzyme and consequently prevent hydrolysis of Mv-3-glu.

127 25 4 A) Grape extract 20

15

10

5 5 2 6 1 3 7 0

25 4 B) Grape extract incubated with LPH 20 and D-gluconolactone for 30min

15

10

5 5 2 6 1 3 7 0 Absorbance at520 nm (mAU) 25 C) Grape extract incubated with LPH 20 for 30min

15 4

10

5 5 2 6 1 3 7 0

5.0 10.0 15.0 20.0 25.0 30.0 35.0 Time (min)

Figure 5.3 HPLC chromatograms of purified grape anthocyanins incubated with LPH for 0 min (A) and 30 min in the presence (B) or absence of D-gluconolactone (C). Peak identities are listed in Table 5.4.

128 m/z AUCtreatment/AUCcontrol% Peak# Peak identity after 30 min incubation Intact Aglycone (n = 2) molecule fragment

1 Dp-3-glu 465 303 75.6

2 Cy-3-glu 449 287 50.1

3 Pt-3-glu 479 317 94.3

4 Pn-3-glu 463 301 50.1

5 Mv-3-glu 493 331 101.6

6 Pn-3-(p-coumaryl)-glu 609 301 105.3

7 Mv-3-( p-coumaryl)-glu 639 331 98.3

Total 61.6

Table 5.4 Profile change of grape anthocyanins after incubation with LPH for 30 min. D- gluconolactone was added to the buffer as negative control.

129 OH OCH3 Cy-3-gly OH Pn-3-gly OH

HO O+ HO O+

O Gly O Gly OH OH

OH OCH3 Dp-3-gly OH Pt-3-gly OH

HO O+ HO O+ OH OH

O Gly O Gly OH OH

OCH3 Mv-3-gly OH 2’ 3’ 4’ HO O+ 1’ 6’ 5’ 8 1 7 2 OCH3 6 5 4 3 O Gly

OH

Figure 5.4 Chemical structures of the common anthocyanidins glycosylated on the C-3 position.

130 In this assay Cy-3-glu and Pn-3-glu were hydrolyzed at the highest velocity. The absence of a bulky group on the 5’ position of the B ring (Figure 5.4) is speculated to give advantage to Cy-3-glu and Pn-3-glu for accessing the binding site of β-glucosidase.

Our hypothesis is also supported by the data published by Nemeth et al. (181): purified sheep LPH hydrolyzed kaempferol-3-glu 16% faster than quercetin-3-glu. The only difference between these two substrates is that kaempferol-3-glu has one less hydroxyl group on the B ring.

5.4.5 Activity of lactase supplement on anthocyanins

The effect of lactase from the supplement on blueberry anthocyanins was investigated. Because lactose is a disaccharide composed of glucose and galactose, both

β-glucosidase and β-galactosidase may hydrolyze lactose. Unlike LPH, which is a β- glucosidase, fungal lactase selectively hydrolyzed anthocyanin galactosides (Figure 5.5), whereas other glycosides were intact. In the presence of lactose, hydrolysis of anthocyanin galactosides was reduced due to competitive inhibition (Figure 5.5 and 5.6).

5.4.6 The influence of lactase residual in foods on anthocyanins

Because of the concern about anthocyanin degradation under the effect of food source lactase, yogurt with live-culture (as an example of fermented foods) and lactase treated skim milk were also examined for lactase activity towards blueberry anthocyanins.

However, neither of the crude enzyme extracts (at consumable levels) demonstrated significant enzymatic activity on purified blueberry anthocyanins after 30 min of incubation (data not shown). Lactase activity assay using a GAGO-20 glucose kit

131 revealed that yogurt with live-culture (one serving) had about 1/40 of the activity as

lactase supplement (one serving) using lactose as substrate. We speculate that due to the low enzymatic activity and rich sugar content co-extracted with lactase, anthocyanin hydrolysis was competitively inhibited thus no significant degradation was observed.

Enzymatic activity of the milk extract towards lactose was not detected. This suggested that the processing of lactase treated milk might not leave active lactase residual that could affect anthocyanin stability when ingested together.

132 A) Inactive lactase + blueberry 150 Dp-3-gal 100 Pt-3-gal Cy-3-gal Pn-3-gal Mv-3-gal 50

0

100 B) Lactase + blueberry

75

50

25

0 Absorbance at520 nm (mAU) 100 C) Lactase + blueberry + lactose

75

50

25

0

0.0 5.0 10.0 15.0 20.0 25.0 30.0

Time (min)

Figure 5.5 HPLC chromatograms of purified blueberry anthocyanins incubated with either heat-inactivated lactase from supplement (A), active lactase from supplement extract (B), or lactase from supplement in the presence of 584 mM lactose (C) for 15 min.

133 18 Heat-inactivated 16 Lactase + lactose Active lactase

14

12

10

8 Area% Pear

6

4

2

0 Dp-3-gal Cy-3-gal Pt-3-gal Pn-3-gal Mv-3-gal

Figure 5.6 Profile change of anthocyanin galactosides in blueberry after incubation with the lactase supplement extract for 15 min (n = 3).

134 5.4.7 Substrate specificity of lactase from supplement

Lactase substrate specificity assay was carried out on purified blueberry

anthocyanins. Similar to the observations with intestinal β-glucosidase, Mv-3-gal was the most resistant to hydrolysis among all the five anthocyanin-3-gly (Figure 5.7). Such an effect was again attributed to the steric effect of the methoxyl group on 5’ position of B ring.

5.4.8 Calculation of apparent Km and Vmax for dietary lactase

For kinetic parameters determination, lactase supplement (9000 FCC lactase

units/serving) was extracted prior to assay without being frozen. Purified chokeberry

extract containing 70% Cy-3-gal served as source of substrate. The reaction exhibited

Michaelis-Menten kinetics (Figure 5.8). The Vmax was 2.38 ± 0.14 mmol/min·serving

and the apparent Km was 414 ± 58 µM. It is noteworthy that the apparent Km is about 14

times of that for intestinal β-glucosidase. Thus, fungal lactase from the supplement has

much lower affinity towards Cy-3-gal as compared to the affinity between β-glucosidase

and Cy-3-glu. However, owing to the high level of lactase contained in one serving of the

supplement, hydrolysis of anthocyanin galactosides was extremely rapid. Based on the

calculated Vmax we estimate that 1 g of Cy-3-gal can be hydrolyzed in 2 min giving the

half maximum velocity. Therefore it appears that any person who takes lactase

supplement regularly may benefit less from the intact anthocyanins that he/she consumes.

135 18 Dp-3-gal Cy-3-gal 16 Pt-3-gal Pn-3-gal 14 Mv-3-gal

12

10

8 Peak Area% Peak

6

4

2

0 0 5 10 15 20 25 30 Time (min)

Figure 5.7 Lactase supplement extract substrate specificity on the 5 anthocyanin galactosides in blueberry.

136 2.0

1.8

1.6

1.4

1.2

1.0

0.8

V (mmol/min*serving) 0.6

0.4

0.2

0.0 0 200 400 600 800 1000 1200 1400

Cy-3-gal (uM)

Figure 5.8 Saturation curve for the lactase supplement extract showing the relationship between Cy-3-gal substrate concentration and rate of Cy-3-gal hydrolysis. Markers with error bars represent means ± SE (n = 3).

5.4.9 Anti-proliferative effect on colon cancer cell line

After incubation with dietary lactase for 45 min most Cy-3-gal in purified chokeberry extract disappeared, Cy-3-arab was lost by 46%, while Cy-3-glu and Cy-3- xyl were in comparable levels with that in the control (incubated with PBS buffer). It appears that dietary lactase had slight enzymatic activity on Cy-3-arab but such activity 137 was suppressed when Cy-3-gal was abundant. Overall about 80% of the chokeberry

anthocyanins were hydrolyzed at the end of incubation (Figure 5.9). Large amount of Cy

aglycone was produced as degradation intermediate but a significant portion of it rapidly

disappeared.

1.2

1

0.8

0.6

0.4 Concentration (mg/mL) Concentration

0.2

0 chokeberry - control chokeberry - lactase treated

Figure 5.9 Intact chokeberry anthocyanins remaining after incubation with buffer or dietary lactase extract for 45 min (n=3).

138 The chokeberry anthocyanins (control) inhibited HT-29 colon cancer cell growth by 17% at 100µg/mL concentration in medium, whereas at equivalent concentration

(equal initial anthocyanin concentration before lactase treatment) the dietary lactase treated chokeberry anthocyanins only inhibited cell growth by 10% (Figure 5.10). At diluted concentrations the lactase treated anthocyanins also demonstrated lower anti- proliferative effect. Obviously the exposure to lactase decreased the total amount of chokeberry anthocyanins and as a result reduced their potency of colon cancer prevention.

The mechanism of anthocyanin’s chemoprevention mechanism in HT-29 cells had been researched before (95). The cells could be blocked at G1/G0 and G2/M phases of the cell cycle when exposed to anthocyanins in the medium, but upon removal of anthocyanins from the medium the cells could recover from both the G1/G0 and G2/M blocks.

Therefore both, the concentration of anthcoyanins remaining in the colon and the duration of exposure to the colon epethilial cells could be important factors in anthocyanin’s health benefits. According to our present study, anthocyanins with specific chemical structures such as Mv aglyone and non-glucoside/galactoside sugar moieties

may be more resistant to enzymatic hydrolysis in the GIT than other anthocyanins and

consequently have a better chance to protect the GIT tissues from chronic diseases

including colon cancer.

139 18 chokeberry - control 16 chokeberry - lactase treated

14 . 12

10

8

6 Growth inhibition %

4

2

0 0 20406080100120

Equivalent Concentration (µg/mL)

Figure 5.10 Growth inhibition (48 h) of human colon cancer cell line HT-29 by purified chokeberry anthocyanins and lactase treated chokeberry anthocyanins. Markers represent means of 4 replicates. The experiment was repeated twice independently and data from one experiment is shown.

140 5.5. CONCLUSIONS

Deglycosylation of various anthocyanins by pig small intestinal mucosal extract

and lactase supplement extract was observed in vitro. The intestinal mucosal extract

specifically hydrolyzed anthocyanin glucosides while the lactase supplement extract

specifically decomposed anthocyanin galactosides. The type of aglycone also influenced

the velocity of enzymatic hydrolysis with Mv being the slowest and Cy being the fastest.

Deglycosylation of anthocyanins under the effect of endogenous and exogenous β-

glycosidases may significantly affect the accessibility and health-promoting effects of

intact anthocyanin in the digestive tract. The large variations of β-glucosidase activity

among individuals, as well as the ingestion of exogenous lactase are likely to influence the ability of individuals to metabolize anthocyanins. Further investigation of the effect of β-glycosidases on digestion, metabolism and health benefit of anthocyanins warrants attention.

5.6. ACKNOWLEDGEMENTS

This work was supported by Grant 2004-35503-15190 from the USDA NRI

Competitive Grants Program. We thank Dr Henry N. Zerby from the OSU Department of

Animal Sciences for providing the pig small intestine sample.

141 BIBLIOGRAPHY

1. Hertog, M. G.; Hollman, P. C.; Katan, M. B.; Kromhout, D., Intake of potentially anticarcinogenic flavonoids and their determinants in adults in The Netherlands. Nutr Cancer 1993, 20, (1), 21-29.

2. Galvano, F.; La Fauci, L.; Lazzarino, G.; Fogliano, V.; Ritieni, A.; Ciappellano, S.; Battistini, N. C.; Tavazzi, B.; Galvano, G., : metabolism and biological properties. J Nutr Biochem 2004, 15, (1), 2-11.

3. Prior, R. L., Absorption and metabolism of anthocyanins: potential health effects. In Phytochemicals: mechanism of action, Meskin, M. S.; Bidlack, W. R.; Davies, A. J.; D.S., L.; R.K., R., Eds. CRC Press: Boca Boton, 2004; pp 1-19.

4. Felgines, C.; Talavera, S.; Gonthier, M. P.; Texier, O.; Scalbert, A.; Lamaison, J. L.; Remesy, C., Strawberry anthocyanins are recovered in urine as glucuro- and sulfoconjugates in humans. J Nutr 2003, 133, (5), 1296-1301.

5. Wu, X.; Pittman, H. E., 3rd; McKay, S.; Prior, R. L., Aglycones and sugar moieties alter anthocyanin absorption and metabolism after berry consumption in weanling pigs. J Nutr 2005, 135, (10), 2417-2424.

6. Fleschhut, J.; Kratzer, F.; Rechkemmer, G.; Kulling, S. E., Stability and biotransformation of various dietary anthocyanins in vitro. Eur J Nutr 2006, 45, (1), 7-18.

7. Wu, X.; Pittman, H. E., 3rd; Prior, R. L., Fate of anthocyanins and antioxidant capacity in contents of the gastrointestinal tract of weanling pigs following black raspberry consumption. J Agric Food Chem 2006, 54, (2), 583-589.

8. Lala, G.; Malik, M.; Zhao, C.; He, J.; Kwon, Y.; Giusti, M. M.; Magnuson, B. A., Anthocyanin-rich extracts inhibit multiple biomarkers of colon cancer in rats. Nutr Cancer 2006, 54, (1), 84-93.

9. Aura, A. M.; Martin-Lopez, P.; O'Leary K, A.; Williamson, G.; Oksman- Caldentey, K. M.; Poutanen, K.; Santos-Buelga, C., In vitro metabolism of anthocyanins by human gut microflora. Eur J Nutr 2005, 44, (3), 133-142.

10. Perez-Vicente, A.; Gil-Izquierdo, A.; Garcia-Viguera, C., In vitro gastrointestinal digestion study of pomegranate juice phenolic compounds, anthocyanins, and vitamin C. J Agric Food Chem 2002, 50, (8), 2308-2312.

142 11. Bermudez-Soto, M. J.; Tomas-Barberan, F. A.; Garcia-Conesa, M. T., Stability of polyphenols in chokeberry ( melanocarpa) subjected to in vitro gastric and pancreatic digestion. Food Chemistry 2007, 102, (3), 865-874.

12. McDougall, G. J.; Fyffe, S.; Dobson, P.; Stewart, D., Anthocyanins from red cabbage--stability to simulated gastrointestinal digestion. 2007, 68, (9), 1285-1294.

13. He, J.; Magnuson, B. A.; Giusti, M. M., Analysis of anthocyanins in rat intestinal contents - Impact of anthocyanin chemical structure on fecal excretion. J Agric Food Chem 2005, 53, (8), 2859-2866.

14. Borges, G.; Roowi, S.; Rouanet, J. M.; Duthie, G. G.; Lean, M. E.; Crozier, A., The bioavailability of raspberry anthocyanins and ellagitannins in rats. Mol Nutr Food Res 2007, 51, (6), 714-725.

15. Takeoka, G.; Dao, L., Anthocyanins. In Methods of Analysis for Functional Foods and Nutraceuticals, 1 ed.; Hurst, W. J., Ed. CRC: Boca Boton, 2002; pp 219-241.

16. Andersen, O. M.; Jordheim, M., The Anthocyanins. In Flavonoids: Chemistry, and Applications, Andersen, O. M.; Markham, K. R., Eds. CRC Press: Boca Raton, FL, 2006; pp 471-552.

17. Delgado-Vargas, F.; O., P.-L., Anthocyanins and . In Natural Colorants for Food and Nutraceutical Uses, CRC Press: Boca Raton, 2003; pp 167-219.

18. Mazza, G.; Miniati, E., Grapes. In Anthocyanins in fruits, vegetables and grains, CRC Press: Boca Raton, 1993; pp 149-199.

19. McCann, D.; Barrett, A.; Cooper, A.; Crumpler, D.; Dalen, L.; Grimshaw, K.; Kitchin, E.; Lok, K.; Porteous, L.; Prince, E.; Sonuga-Barke, E.; Warner, J. O.; Stevenson, J., Food additives and hyperactive behaviour in 3-year-old and 8/9-year-old children in the community: a randomised, double-blinded, placebo-controlled trial. Lancet 2007, 370, (9598), 1560-1567.

20. Harborne, J. B., Phenolic Compounds. In Methods - a guide to modern techniques of plant analysis, 3, Ed. Chapman & Hall: New York, 1998; pp 66-74.

21. Eder, R., Pigments. In Food Analysis by HPLC, 2 ed.; NOLLET, L. M. L., Ed. Marcel Dekker Inc.: Monticello, NY, 2000; pp 845-880.

143 22. Kong, J. M.; Chia, L. S.; Goh, N. K.; Chia, T. F.; Brouillard, R., Analysis and biological activities of anthocyanins. Phytochemistry 2003, 64, (5), 923-933.

23. Heredia, F. J.; Francia-Aricha, E. M.; Rivas-Gonzalo, J. C.; Vicario, I. M.; Santos-Buelga, C., Chromatic characterization of anthocyanins from red grapes. I. pH effects. Food Chem 1998, 63, (4), 491-498.

24. Mazza, G.; Kay, C. D.; Cottrell, T.; Holub, B. J., Absorption of anthocyanins from blueberries and serum antioxidant status in human subjects. J Agric Food Chem 2002, 50, (26), 7731-7737.

25. He, J.; Magnuson, B. A.; Lala, G.; Tian, Q.; Schwartz, S. J.; Giusti, M. M., Intact anthocyanins and metabolites in rat urine and plasma after 3 months of anthocyanin supplementation. Nutr Cancer 2006, 54, (1), 3-12.

26. Borkowski, T.; Szymusiak, H.; Gliszczynska-Swiglo, A.; Tyrakowska, B., The effect of 3-O-beta -glycosylation on structural transformations of anthocyanins. Food Res Int 2005, 38, (8-9), 1031-1037.

27. Brouillard, R.; Delaporte, B., Chemistry of anthocyanin pigments. 2. Kinetic and thermodynamic study of proton transfer, hydration, and tautomeric reactions of malvidin 3-glucoside. J Am Chem Soc 1977, 99, (26), 8461-8468.

28. Brouillard, R.; Dubois, J.-E., Mechanism of the structural transformations of anthocyanins in acidic media. J Am Chem Soc 1977, 99, (5), 1359-1364.

29. Houbiers, C.; Lima, J. C.; Macanita, A. L.; Santos, H., Color Stabilization of Malvidin 3-Glucoside: Self-Aggregation of the Flavylium Cation and Copigmentation with the Z-Chalcone Form. J Phys Chem B 1998, 102, (18), 3578-3585.

30. Asenstorfer, R. E.; Iland, P. G.; Tate, M. E.; Jones, G. P., Charge equilibria and pKa of malvidin-3-glucoside by electrophoresis. Anal Biochem 2003, 318, (2), 291-299.

31. Dangles, O.; Saito, N.; Brouillard, R., Anthocyanin intramolecular copigment effect. Phytochemistry 1993, 34, (1), 119-124.

32. Francis, F. J., Food colorants: anthocyanins. Crit Rev Food Sci Nutr 1989, 28, (4), 273-314.

33. Chen, L. J.; Hrazdina, G., Structural transformation reactions of anthocyanins. Cell Mol Life Sci 1982, 38, (9), 1030-1032.

144 34. Bakker, J.; Timberlake, C. F., Isolation, Identification, and Characterization of New Color-Stable Anthocyanins Occurring in Some Red Wines. J Agric Food Chem 1997, 45, (1), 35-43.

35. Vivar-Quintana, A. M.; Santos-Buelga, C.; Rivas-Gonzalo, J. C., Anthocyanin- derived pigments and color of red wines. Anal Chim Acta 2002, 458, (1), 147-155.

36. Alcalde-Eon, C.; Escribano-Bailon, M. T.; Santos-Buelga, C.; Rivas-Gonzalo, J. C., Separation of pyranoanthocyanins from red wine by column chromatography. Anal Chim Acta 2004, 513, (1), 305-318.

37. Schwarz, M.; Winterhalter, P., A novel synthetic route to substituted pyranoanthocyanins with unique color properties. Tetrahedron Lett 2003, 44, (41), 7583- 7587.

38. Lu, Y.; Foo, L. Y., Unusual anthocyanin reaction with acetone leading to pyranoanthocyanin formation. Tetrahedron Lett 2001, 42, (7), 1371-1373.

39. Boulton, R., The Copigmentation of Anthocyanins and Its Role in the Color of Red Wine: A Critical Review. Am J Enol Vitic 2001, 52, (2), 67-87.

40. Schwarz, M.; Wray, V.; Winterhalter, P., Isolation and Identification of Novel Pyranoanthocyanins from Black Carrot ( L.) Juice. J Agric Food Chem 2004, 52, (16), 5095-5101.

41. Rein, M. J.; Ollilainen, V.; Vahermo, M.; Yli-Kauhaluoma, J.; Heinonen, M., Identification of novel pyranoanthocyanins in berry juices. Eur Food Res Technol 2005, 220, (3-4), 239-244.

42. Mateus, N.; Oliveira, J.; Haettich-Motta, M.; de Freitas, V., New family of bluish pyranoanthocyanins. J Biomed Biotechnol 2004, (5), 299-305.

43. Shahidi, F.; Naczk, M., Contribution of phenolic compounds to flavor and color characteristics of foods. In Phenolics in food and nutraceuticals, CRC press: Boca Boton, 2004; pp 443-461.

44. Wu, X.; Beecher, G. R.; Holden, J. M.; Haytowitz, D. B.; Gebhardt, S. E.; Prior, R. L., Concentrations of anthocyanins in common foods in the United States and estimation of normal consumption. J Agric Food Chem 2006, 54, (11), 4069-4075.

45. Clifford, M. N., Anthocyanins - nature, occurrence and dietary burden. J Sci Food Agric 2000, 80, (7), 1063-1072. 145 46. Giusti, M. M.; Wrolstad, R. E., Acylated anthocyanins from edible sources and their applications in food systems. Biochem Eng J 2003, 14, 217-225.

47. Jing, P.; Giusti, M. M., Characterization of anthocyanin-rich waste from purple corncobs (Zea mays L.) and its application to color milk. J Agric Food Chem 2005, 53, (22), 8775-81.

48. Wallace, T. C.; Giusti, M. M., Determination of Color, Pigment, and Phenolic Stability in Yogurt Systems Colored with Nonacylated Anthocyanins from Berberis boliviana L. as Compared to Other Natural/Synthetic Colorants. J Food Sci 2008, 73, (4), C241-C248.

49. Brouillard, R., Chemical structure of anthocyanins. In Anthocyanins as food colors, Markakis, P., Ed. Academic Press: New York, 1982; pp 1-40.

50. WHO In Toxicological evaluation of certain food additives, The 26th meeting of the Joint FAO/WHO Expert Committee on Food Additives Geneva, 1982; Geneva, 1982.

51. Sarni-Manchado, P.; Cheynier, V.; Moutounet, M., Reactions of polyphenoloxidase generated caftaric acid o-quinone with malvidin 3-O-glucoside. Phytochemistry 1997, 45, (7), 1365-1369.

52. Fossen, T.; Cabrita, L.; Andersen, O. M., Color and stability of pure anthocyanins influenced by pH including the alkaline region. Food Chem 1998, 63, (4), 435-440.

53. Macheix, J.-J.; Fleuriet, A.; Billot, J., Phenolic Compounds in Fruit Processing. In Fruit Phenolics, 1 ed.; CRC Press: Boca Raton, Florida, 1990; pp 295-358.

54. Palamidis, N.; Markakis, P., Stability of Grape Anthocyanin in A Carbonated Beverage. J Food Sci 1975, 40, (5), 1047-1049.

55. Kader, F.; Haluk, J.-P.; Nicolas, J.-P.; Metche, M., Degradation of cyanidin 3- glucoside by blueberry polyphenol oxidase: kinetic studies and mechanisms. J Agric Food Chem 1998, 46, 3060-3065.

56. Kader, F.; Irmouli, M.; Nicolas, J. P.; Metche, M., Involvement of Blueberry Peroxidase in the Mechanisms of Anthocyanin Degradation in Blueberry Juice. J Food Sci 2002, 67, (3), 910-915.

57. Lee, J.; Wrolstad, R. E., Extraction of anthocyanins and polyphenolics from blueberry-processing waste. J Food Sci 2004, 69, (7), C564-C573.

146 58. Blom, H.; Thomassen, M. S., Kinetic studies on strawberry anthocyanin hydrolysis by a thermostable anthocyanin-beta -glycosidase from Aspergillus niger. Food Chem 1985, 17, (3), 157-168.

59. Wightman, J. D.; Wrolstad, R. E., Anthocyanin analysis as a measure of glycosidase activity in enzymes for juice processing. J Food Sci 1995, 60, (4), 862-867.

60. Wightman, J. D.; Wrolstad, R. E., beta -Glucosidase activity in juice-processing enzymes based on anthocyanin analysis. J Food Sci 1996, 61, (3), 544-547, 552.

61. Poei-Langston, M. S.; Wrolstad, R. E., Color degradation in an ascorbic acid- anthocyanin-flavanol model system. J Food Sci 1981, 46, (4), 1218-1222, 1236.

62. Ozkan, M., Degradation of anthocyanins in sour cherry and pomegranate juices by hydrogen peroxide in the presence of added ascorbic acid. Food Chem 2002, 78, (4), 499-504.

63. Giusti, M. M.; Ghanadan, H.; Wrolstad, R. E., Elucidation of the Structure and Conformation of Red Radish (Raphanus sativus) Anthocyanins Using One- and Two- Dimensional Nuclear Magnetic Resonance Techniques. J Agric Food Chem 1998, 46, (12), 4858-4863.

64. Brouillard, R., Chemical structure of anthocyanins. In Anthocyanins as food colors, Markakis, P., Ed. Academic Press: New York, NY, 1982; pp 1-40.

65. Giusti, M. M.; Wrolstad, R. E., Radish anthocyanin extract as a natural red colorant for maraschino cherries. J Food Sci 1996, 61, (4), 688-694.

66. Rodriguez-Saona, L. E.; Giusti, M. M.; Durst, R. W.; Wrolstad, R. E., Development and process optimization of red radish concentrate extract as potential natural red colorant. J Food Process Pres 2001, 25, 165-182.

67. Brouillard, R.; Dangles, O., Anthocyanin molecular interactions: the first step in the formation of new pigments during wine aging? Food Chem 1994, 51, (4), 365-371.

68. Waterhouse, A. L., Wine phenolics. Ann N Y Acad Sci 2002, 957, (Alcohol and Wine in Health and Disease), 21-36.

69. Giusti, M. M.; Rodriguez-Saona, L. E.; Wrolstad, R. E., Molar absorptivity and color characteristics of acylated and non-acylated pelargonidin-based anthocyanins. J Agric Food Chem 1999, 47, (11), 4631-4637.

147 70. Scalbert, A.; Williamson, G., Dietary intake and bioavailability of polyphenols. J Nutr 2000, 130, (8S Suppl), 2073S-2085S.

71. Renaud, S.; de Lorgeril, M., Wine, alcohol, platelets, and the French paradox for coronary heart disease. Lancet 1992, 339, (8808), 1523-1526.

72. Zhang, Y.; Seeram, N. P.; Lee, R.; Feng, L.; Heber, D., Isolation and identification of strawberry phenolics with antioxidant and human cancer cell antiproliferative properties. J Agric Food Chem 2008, 56, (3), 670-675.

73. Seeram, N. P.; Adams, L. S.; Hardy, M. L.; Heber, D., Total Cranberry Extract versus Its Phytochemical Constituents: Antiproliferative and Synergistic Effects against Human Tumor Cell Lines. J Agric Food Chem 2004, 52, (9), 2512-2517.

74. Allen, R. G.; Tresini, M., Oxidative stress and gene regulation. Free Radical Biol Med 2000, 28, (3), 463-499.

75. Wang, H.; Cao, G.; Prior, R. L., Oxygen Radical Absorbing Capacity of Anthocyanins. J Agric Food Chem 1997, 45, (2), 304-309.

76. Tsuda, T.; Watanabe, M.; Ohshima, K.; Norinobu, S.; Choi, S.-W.; Kawakishi, S.; Osawa, T., Antioxidative Activity of the Anthocyanin Pigments Cyanidin 3-O-beta -D- Glucoside and Cyanidin. J Agric Food Chem 1994, 42, (11), 2407-2410.

77. Tedesco, I.; Luigi Russo, G.; Nazzaro, F.; Russo, M.; Palumbo, R., Antioxidant effect of red wine anthocyanins in normal and catalase-inactive human erythrocytes. J Nutr Biochem 2001, 12, (9), 505-511.

78. Tsuda, T.; Horio, F.; Osawa, T., The role of anthocyanins as an antioxidant under oxidative stress in rats. Biofactors 2000, 13, (1-4), 133-139.

79. Ramirez-Tortosa, C.; Andersen, O. M.; Gardner, P. T.; Morrice, P. C.; Wood, S. G.; Duthie, S. J.; Collins, A. R.; Duthie, G. G., Anthocyanin-rich extract decreases indices of lipid peroxidation and DNA damage in vitamin E-depleted rats. Free Radical Biol Med 2001, 31, (9), 1033-1037.

80. Aviram, M., Review of human studies on oxidative damage and antioxidant protection related to cardiovascular diseases. Free Radical Res 2000, 33, (Suppl.), S85- S97.

148 81. Aviram, M.; Kaplan, M.; Rosenblat, M.; Fuhrman, B., Dietary antioxidants and paraoxonases against LDL oxidation and atherosclerosis development. Handb Exp Pharmacol 2005, 170, (Atherosclerosis), 263-300.

82. Whitehead, T. P.; Robinson, D.; Allaway, S.; Syms, J.; Hale, A., Effect of red wine ingestion on the antioxidant capacity of serum. Clin Chem (Washington, D. C.) 1995, 41, (1), 32-35.

83. Day, A. P.; Kemp, H. J.; Bolton, C.; Hartog, M.; Stansbie, D., Effect of concentrated red grape juice consumption on serum antioxidant capacity and low-density lipoprotein oxidation. Ann Nutr Metab 1997, 41, (6), 353-357.

84. Matsumoto, H.; Nakamura, Y.; Hirayama, M.; Yoshiki, Y.; Okubo, K., Antioxidant Activity of Black Currant Anthocyanin Aglycons and Their Glycosides Measured by Chemiluminescence in a Neutral pH Region and in Human Plasma. J Agric Food Chem 2002, 50, (18), 5034-5037.

85. Abuja, P. M.; Murkovic, M.; Pfannhauser, W., Antioxidant and Prooxidant Activities of Elderberry ( nigra) Extract in Low-Density Lipoprotein Oxidation. J Agric Food Chem 1998, 46, (10), 4091-4096.

86. Tsuda, T.; Shiga, K.; Ohshima, K.; Kawakishi, S.; Osawa, T., Inhibition of lipid peroxidation and the active oxygen radical scavenging effect of anthocyanin pigments isolated from Phaseolus vulgaris L. Biochem Pharmacol 1996, 52, (7), 1033-1039.

87. Coussens, L. M.; Werb, Z., Inflammation and cancer. Nature 2002, 420, (6917), 860-867.

88. Wang, H.; Nair, M. G.; Strasburg, G. M.; Chang, Y. C.; Booren, A. M.; Gray, J. I.; DeWitt, D. L., Antioxidant and antiinflammatory activities of anthocyanins and their aglycon, cyanidin, from tart cherries. J Nat Prod 1999, 62, (2), 294-296.

89. Seeram, N. P.; Momin, R. A.; Nair, M. G.; Bourquin, L. D., Cyclooxygenase inhibitory and antioxidant cyanidin glycosides in cherries and berries. Phytomedicine 2001, 8, (5), 362-369.

90. Rossi, A.; Serraino, I.; Dugo, P.; Di Paola, R.; Mondello, L.; Genovese, T.; Morabito, D.; Dugo, G.; Sautebin, L.; Caputi, A. P.; Cuzzocrea, S., Protective effects of anthocyanins from blackberry in a rat model of acute lung inflammation. Free Radical Res 2003, 37, (8), 891-900.

149 91. Kamei, H.; Kojima, T.; Hasegawa, M.; Koide, T.; Umeda, T.; Yukawa, T.; Terabe, K., Suppression of tumor cell growth by anthocyanins in vitro. Cancer Invest 1995, 13, (6), 590-594.

92. Kamei, H.; Hashimoto, Y.; Koide, T.; Kojima, T.; Hasegawa, M., Anti-tumor effect of methanol extracts from red and white wines. Cancer Biother Radiopharm 1998, 13, (6), 447-452.

93. Yi, W.; Fischer, J.; Akoh, C. C., Study of Anticancer Activities of Muscadine Grape Phenolics in Vitro. J Agric Food Chem 2005, 53, (22), 8804-8812.

94. Zhao, C.; Giusti, M. M.; Malik, M.; Moyer, M. P.; Magnuson, B. A., Effects of commercial anthocyanin-rich extracts on colonic cancer and nontumorigenic colonic cell growth. J Agric Food Chem 2004, 52, (20), 6122-6128.

95. Malik, M.; Zhao, C.; Schoene, N.; Guisti, M. M.; Moyer, M. P.; Magnuson, B. A., Anthocyanin-rich extract from Aronia meloncarpa E induces a cell cycle block in colon cancer but not normal colonic cells. Nutr Cancer 2003, 46, (2), 186-196.

96. Stoner, G. D.; Kresty, L. A.; Carlton, P. S.; Siglin, J. C.; Morse, M. A., Isothiocyanates and freeze-dried strawberries as inhibitors of esophageal cancer. Toxicol Sci 1999, 52, (2, Suppl.), 95-100.

97. Reen, R. K.; Nines, R.; Stoner, G. D., Modulation of N- nitrosomethylbenzylamine metabolism by black raspberries in the esophagus and liver of Fischer 344 rats. Nutr Cancer 2006, 54, (1), 47-57.

98. Harris, G. K.; Gupta, A.; Nines, R. G.; Kresty, L. A.; Habib, S. G.; Frankel, W. L.; LaPerle, K.; Gallaher, D. D.; Schwartz, S. J.; Stoner, G. D., Effects of lyophilized black raspberries on azoxymethane-induced colon cancer and 8-hydroxy-2'-deoxyguanosine levels in the Fischer 344 rat. Nutr Cancer 2001, 40, (2), 125-133.

99. Hagiwara, A.; Miyashita, K.; Nakanishi, T.; Sano, M.; Tamano, S.; Kadota, T.; Koda, T.; Nakamura, M.; Imaida, K.; Ito, N.; Shirai, T., Pronounced inhibition by a natural anthocyanin, purple corn color, of 2-amino-1-methyl-6-phenylimidazo[4,5- b]pyridine (PhIP)-associated colorectal carcinogenesis in male F344 rats pretreated with 1,2-dimethylhydrazine. Cancer Lett 2001, 171, (1), 17-25.

100. Hagiwara, A.; Yoshino, H.; Ichihara, T.; Kawabe, M.; Tamano, S.; Aoki, H.; Koda, T.; Nakamura, M.; Imaida, K.; Ito, N.; Shirai, T., Prevention by natural food anthocyanins, purple sweet potato color and red cabbage color, of 2-amino-1-methyl-6- phenylimidazo[4,5-b]pyridine (PhIP)-associated colorectal carcinogenesis in rats initiated with 1,2-dimethylhydrazine. J Toxicol Sci 2002, 27, (1), 57-68. 150 101. Carlton, P. S.; Kresty, L. A.; Stoner, G. D., Failure of dietary lyophilized strawberries to inhibit 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone- and benzo[a]pyrene-induced lung tumorigenesis in strain A/J mice. Cancer Lett (Shannon, Irel) 2000, 159, (2), 113-117.

102. Hou, D. X., Potential mechanisms of cancer chemoprevention by anthocyanins. Curr Mol Med 2003, 3, (2), 149-159.

103. Duthie, S. J., Berry phytochemicals, genomic stability and cancer: evidence for chemoprotection at several stages in the carcinogenic process. Mol Nutr Food Res 2007, 51, (6), 665-674.

104. Ohara, A.; Matsuhisa, T.; Hosokawa, K.; Mori, K., Antimutagenicity of anthocyanins against various mutagens in the Ames test. ITE Lett Batteries, New Technol Med 2004, 5, (2), 172-178.

105. Yoshimoto, M.; Okuno, S.; Yamaguchi, M.; Yamakawa, O., Antimutagenicity of deacylated anthocyanins in purple-fleshed sweetpotato. Biosci Biotechnol Biochem 2001, 65, (7), 1652-1655.

106. Gasiorowski, K.; Szyba, K.; Brokos, B.; Kolaczynska, B.; Jankowiak-Wlodarczyk, M.; Oszmianski, J., Antimutagenic activity of anthocyanins isolated from Aronia melanocarpa fruits. Cancer Lett (Shannon, Irel) 1997, 119, (1), 37-46.

107. Singletary, K. W.; Jung, K.-J.; Giusti, M., Anthocyanin-Rich Grape Extract Blocks Breast Cell DNA Damage. J Med Food 2007, 10, (2), 244-251.

108. Shih, P.-H.; Yeh, C.-T.; Yen, G.-C., Anthocyanins Induce the Activation of Phase II Enzymes through the Antioxidant Response Element Pathway against Oxidative Stress-Induced Apoptosis. J Agric Food Chem 2007, 55, (23), 9427-9435.

109. Srivastava, A.; Akoh, C. C.; Fischer, J.; Krewer, G., Effect of Anthocyanin Fractions from Selected Cultivars of Georgia-Grown Blueberries on Apoptosis and Phase II Enzymes. J Agric Food Chem 2007, 55, (8), 3180-3185.

110. Renis, M.; Calandra, L.; Scifo, C.; Tomasello, B.; Cardile, V.; Vanella, L.; Bei, R.; La Fauci, L.; Galvano, F., Response of cell cycle/stress-related protein expression and DNA damage upon treatment of CaCo2 cells with anthocyanins. Br J Nutr 2008, 100, (1), 27-35.

151 111. Yi, W.; Fischer, J.; Krewer, G.; Akoh Casimir, C., Phenolic compounds from blueberries can inhibit colon cancer cell proliferation and induce apoptosis. J Agric Food Chem 2005, 53, (18), 7320-7329.

112. Bagchi, M.; Zafra-Stone, S.; Losso, J. N.; Sen, C. K.; Roy, S.; Hazra, S.; Bagchi, D., Role of edible berry anthocyanins in angiogenesis. Nutraceutical Sci Technol 2007, 6, (Anti-Angiogenic Functional and Medicinal Foods), 527-548.

113. Bruce, W. R.; Giacca, A.; Medline, A., Possible mechanisms relating diet and risk of colon cancer. Cancer Epidemiol Biomarkers Prev 2000, 9, (12), 1271-1279.

114. Tsuda, T., Regulation of adipocyte function by anthocyanins; possibility of preventing the metabolic syndrome. J Agric Food Chem 2008, 56, (3), 642-646.

115. Tsuda, T.; Horio, F.; Uchida, K.; Aoki, H.; Osawa, T., Dietary cyanidin 3-O-beta -D-glucoside-rich purple corn color prevents obesity and ameliorates hyperglycemia in mice. J Nutr 2003, 133, (7), 2125-2130.

116. Kwon, S.-H.; Ahn, I.-S.; Kim, S.-O.; Kong, C.-S.; Chung, H.-Y.; Do, M.-S.; Park, K.-Y., Anti-obesity and hypolipidemic effects of black soybean anthocyanins. J Med Food 2007, 10, (3), 552-556.

117. Prior, R. L.; Wu, X.; Gu, L.; Hager, T. J.; Hager, A.; Howard, L. R., Whole berries versus berry anthocyanins: interactions with dietary fat levels in the C57BL/6J mouse model of obesity. J Agric Food Chem 2008, 56, (3), 647-653.

118. Tsuda, T.; Ueno, Y.; Kojo, H.; Yoshikawa, T.; Osawa, T., Gene expression profile of isolated rat adipocytes treated with anthocyanins. Biochim Biophys Acta 2005, 1733, (2-3), 137-147.

119. Ghosh, D.; Konishi, T., Anthocyanins and anthocyanin-rich extracts: role in diabetes and eye function. Asia Pac J Clin Nutr 2007, 16, (2), 200-208.

120. Al-Awwadi, N. A.; Araiz, C.; Bornet, A.; Delbosc, S.; Cristol, J. P.; Linck, N.; Azay, J.; Teissedre, P. L.; Cros, G., Extracts enriched in different polyphenolic families normalize increased cardiac NADPH oxidase expression while having differential effects on insulin resistance, hypertension, and cardiac hypertrophy in high-fructose-fed rats. J Agric Food Chem 2005, 53, (1), 151-157.

121. Sugimoto, E.; Igarashi, K.; Kubo, K.; Molyneux, J.; Kubomura, K., Protective effects of boysenberry anthocyanins on oxidative stress in diabetic rats. Food Sci Technol Res 2003, 9, (4), 345-349.

152 122. Kramer Joel, H., Anthocyanosides of myrtillus (bilberry) for night vision--a systematic review of placebo-controlled trials. Surv Ophthalmol 2004, 49, (6), 618.

123. Nakaishi, H.; Matsumoto, H.; Tominaga, S.; Hirayama, M., Effects of black current anthocyanoside intake on dark adaptation and VDT work-induced transient refractive alteration in healthy humans. Altern Med Rev 2000, 5, (6), 553-562.

124. Canter Peter, H.; Ernst, E., Anthocyanosides of (bilberry) for night vision--a systematic review of placebo-controlled trials. Surv Ophthalmol 2004, 49, (1), 38-50.

125. Kalt, W.; Blumberg, J. B.; McDonald, J. E.; Vinqvist-Tymchuk, M. R.; Fillmore, S. A. E.; Graf, B. A.; O'Leary, J. M.; Milbury, P. E., Identification of Anthocyanins in the Liver, Eye, and Brain of Blueberry-Fed Pigs. J Agric Food Chem 2008, 56, (3), 705-712.

126. Nohynek Liisa, J.; Alakomi, H.-L.; Kahkonen Marja, P.; Heinonen, M.; Helander Ilkka, M.; Oksman-Caldentey, K.-M.; Puupponen-Pimia Riitta, H., Berry phenolics: antimicrobial properties and mechanisms of action against severe human pathogens. Nutr Cancer 2006, 54, (1), 18-32.

127. Puupponen-Pimia, R.; Nohynek, L.; Meier, C.; Kahkonen, M.; Heinonen, M.; Hopia, A.; Oksman-Caldentey, K. M., Antimicrobial properties of phenolic compounds from berries. J Appl Microbiol 2001, 90, (4), 494-507.

128. Kroon, P. A.; Clifford, M. N.; Crozier, A.; Day, A. J.; Donovan, J. L.; Manach, C.; Williamson, G., How should we assess the effects of exposure to dietary polyphenols in vitro? Am J Clin Nutr 2004, 80, (1), 15-21.

129. Setchell, K. D.; Brown, N. M.; Lydeking-Olsen, E., The clinical importance of the metabolite equol-a clue to the effectiveness of soy and its isoflavones. J Nutr 2002, 132, (12), 3577-3584.

130. Felgines, C.; Texier, O.; Besson, C.; Fraisse, D.; Lamaison, J. L.; Remesy, C., Blackberry anthocyanins are slightly bioavailable in rats. J Nutr 2002, 132, (6), 1249- 1253.

131. Wu, X.; Cao, G.; Prior, R. L., Absorption and metabolism of anthocyanins in elderly women after consumption of elderberry or blueberry. J Nutr 2002, 132, (7), 1865- 1871.

153 132. Felgines, C.; Talavera, S.; Texier, O.; Gil-Izquierdo, A.; Lamaison, J. L.; Remesy, C., Blackberry anthocyanins are mainly recovered from urine as methylated and glucuronidated conjugates in humans. J Agric Food Chem 2005, 53, (20), 7721-7727.

133. Kay, C. D.; Mazza, G. J.; Holub, B. J., Anthocyanins exist in the circulation primarily as metabolites in adult men. J Nutr 2005, 135, (11), 2582-2588.

134. Felgines, C.; Texier, O.; Besson, C.; Lyan, B.; Lamaison, J. L.; Scalbert, A., Strawberry pelargonidin glycosides are excreted in urine as intact glycosides and glucuronidated pelargonidin derivatives in rats. Br J Nutr 2007, 98, (6), 1126-1131.

135. Piskula, M. K.; Yamakoshi, J.; Iwai, Y., Daidzein and genistein but not their glucosides are absorbed from the rat stomach. FEBS Lett 1999, 447, (2-3), 287-291.

136. Harada, K.; Kano, M.; Takayanagi, T.; Yamakawa, O.; Ishikawa, F., Absorption of acylated anthocyanins in rats and humans after ingesting and extract of Ipomoea batatas purple sweet potato tuber. Biosci Biotech Biochem 2004, 68, (7), 1500-1507.

137. Scheline, R. R., Metabolism of foreign compounds by gastrointestinal microorganisms. Pharmacol Rev 1973, 25, (4), 451-523.

138. Bokkenheuser, V. D.; Shackleton, C. H.; Winter, J., Hydrolysis of dietary flavonoid glycosides by strains of intestinal Bacteroides from humans. Biochem J 1987, 248, (3), 953-956.

139. Hollman, P. C. H., Absorption, bioavailability, and metabolism of flavonoids. Pharm Biol (Lisse, Netherlands) 2004, 42, (Suppl.), 74-83.

140. Passamonti, S.; Vrhovsek, U.; Vanzo, A.; Mattivi, F., The stomach as a site for anthocyanins absorption from food. FEBS Lett 2003, 544, (1-3), 210-213.

141. Talavera, S.; Felgines, C.; Texier, O.; Besson, C.; Lamaison, J. L.; Remesy, C., Anthocyanins are efficiently absorbed from the stomach in anesthetized rats. J Nutr 2003, 133, (12), 4178-4182.

142. Passamonti, S.; Vrhovsek, U.; Mattivi, F., The interaction of anthocyanins with bilitranslocase. Biochem Biophys Res Commun 2002, 296, (3), 631-636.

143. Talavera, S.; Felgines, C.; Texier, O.; Besson, C.; Manach, C.; Lamaison, J. L.; Remesy, C., Anthocyanins are efficiently absorbed from the small intestine in rats. J Nutr 2004, 134, (9), 2275-2279.

154 144. Hollman, P. C.; van Trijp, J. M.; Buysman, M. N.; van der Gaag, M. S.; Mengelers, M. J.; de Vries, J. H.; Katan, M. B., Relative bioavailability of the antioxidant flavonoid quercetin from various foods in man. FEBS Lett 1997, 418, (1-2), 152-156.

145. Gee, J. M.; Dupont, M. S.; Rhodes, M. J. C.; Johnson, I. T., Quercetin glucosides interact with the intestinal glucose transport pathway. Free Radic Biol Med 1998, 25, (1), 19-25.

146. Gee, J. M.; DuPont, M. S.; Day, A. J.; Plumb, G. W.; Williamson, G.; Johnson, I. T., Intestinal transport of quercetin glycosides in rats involves both deglycosylation and interaction with the hexose transport pathway. J Nutr 2000, 130, (11), 2765-2771.

147. Day, A. J.; DuPont, M. S.; Ridley, S.; Rhodes, M.; Rhodes, M. J.; Morgan, M. R.; Williamson, G., Deglycosylation of flavonoid and glycosides by human small intestine and liver beta-glucosidase activity. FEBS Lett 1998, 436, (1), 71-75.

148. Hollman, P. C.; Katan, M. B., Absorption, metabolism and health effects of dietary flavonoids in man. Biomed Pharmacother 1997, 51, (8), 305-310.

149. Lambert, N.; Kroon, P. A.; Faulds, C. B.; Plumb, G. W.; McLauchlan, W. R.; Day, A. J.; Williamson, G., Purification of cytosolic beta-glucosidase from pig liver and its reactivity towards flavonoid glycosides. Biochim Biophys Acta 1999, 1435, (1-2), 110- 116.

150. Day, A. J.; Canada, F. J.; Diaz, J. C.; Kroon, P. A.; McLauchlan, R.; Faulds, C. B.; Plumb, G. W.; Morgan, M. R.; Williamson, G., Dietary flavonoid and isoflavone glycosides are hydrolysed by the lactase site of lactase phlorizin hydrolase. FEBS Lett 2000, 468, (2-3), 166-170.

151. Day, A. J.; Gee, J. M.; DuPont, M. S.; Johnson, I. T.; Williamson, G., Absorption of quercetin-3-glucoside and quercetin-4'-glucoside in the rat small intestine: the role of lactase phlorizin hydrolase and the sodium-dependent glucose transporter. Biochem Pharmacol 2003, 65, (7), 1199-1206.

152. Sesink, A. L.; Arts, I. C.; Faassen-Peters, M.; Hollman, P. C., Intestinal uptake of quercetin-3-glucoside in rats involves hydrolysis by lactase phlorizin hydrolase. J Nutr 2003, 133, (3), 773-776.

153. Plumb, G. W.; Garcia-Conesa, M. T.; Kroon, P. A.; Rhodes, M.; Ridley, S.; Williamson, G., Metabolism of chlorogenic acid by human plasma, liver, intestine and gut microflora. J Sci Food Agric 1999, 79, (3), 390-392.

155 154. Aura, A. M.; O'Leary, K. A.; Williamson, G.; Ojala, M.; Bailey, M.; Puupponen- Pimia, R.; Nuutila, A. M.; Oksman-Caldentey, K. M.; Poutanen, K., Quercetin Derivatives Are Deconjugated and Converted to Hydroxyphenylacetic Acids but Not Methylated by Human Fecal Flora in Vitro. J Agric Food Chem 2002, 50, (6), 1725-1730.

155. Schneider, H.; Blaut, M., Anaerobic degradation of flavonoids by Eubacterium ramulus. Arch Microbiol 2000, 173, (1), 71-75.

156. Kuhnle, G.; Spencer, J. P.; Schroeter, H.; Shenoy, B.; Debnam, E. S.; Srai, S. K.; Rice-Evans, C.; Hahn, U., Epicatechin and catechin are O-methylated and glucuronidated in the small intestine. Biochem Biophys Res Commun 2000, 277, (2), 507-512.

157. Ichiyanagi, T.; Shida, Y.; Rahman, M. M.; Hatano, Y.; Matsumoto, H.; Hirayama, M.; Konishi, T., Metabolic pathway of cyanidin 3-O-beta-D-glucopyranoside in rats. J Agric Food Chem 2005, 53, (1), 145-150.

158. Yang, C. S.; Chen, L.; Lee, M. J.; Balentine, D.; Kuo, M. C.; Schantz, S. P., Blood and urine levels of tea catechins after ingestion of different amounts of green tea by human volunteers. Cancer Epidemiol Biomarkers Prev 1998, 7, (4), 351-354.

159. Olthof, M. R.; Hollman, P. C.; Buijsman, M. N.; van Amelsvoort, J. M.; Katan, M. B., Chlorogenic acid, quercetin-3-rutinoside and black tea phenols are extensively metabolized in humans. J Nutr 2003, 133, (6), 1806-1814.

160. de Boer, V. C.; Dihal, A. A.; van der Woude, H.; Arts, I. C.; Wolffram, S.; Alink, G. M.; Rietjens, I. M.; Keijer, J.; Hollman, P. C., Tissue distribution of quercetin in rats and pigs. J Nutr 2005, 135, (7), 1718-1725.

161. Talavera, S.; Felgines, C.; Texier, O.; Besson, C.; Gil-Izquierdo, A.; Lamaison, J. L.; Remesy, C., Anthocyanin metabolism in rats and their distribution to digestive area, kidney, and brain. J Agric Food Chem 2005, 53, (10), 3902-3908.

162. Wiseman, H.; Casey, K.; Bowey, E. A.; Duffy, R.; Davies, M.; Rowland, I. R.; Lloyd, A. S.; Murray, A.; Thompson, R.; Clarke, D. B., Influence of 10 wk of soy consumption on plasma concentrations and excretion of and on gut microflora metabolism in healthy adults. Am J Clin Nutr 2004, 80, (3), 692-699.

163. Griffiths, L. A.; Barrow, A., Metabolism of flavonoid compounds in germ-free rats. Biochem J 1972, 130, (4), 1161-1162.

156 164. McGhie, T. K.; Ainge, G. D.; Barnett, L. E.; Cooney, J. M.; Jensen, D. J., Anthocyanin glycosides from berry fruit are absorbed and excreted unmetabolized by both humans and rats. J Agric Food Chem 2003, 51, (16), 4539-4548.

165. Ichiyanagi, T.; Shida, Y.; Rahman, M. M.; Hatano, Y.; Konishi, T., Bioavailability and tissue distribution of anthocyanins in bilberry (Vaccinium myrtillus L.) extract in rats. J Agric Food Chem 2006, 54, (18), 6578-6587.

166. Walle, T.; Walle, U. K.; Halushka, P. V., Carbon dioxide is the major metabolite of quercetin in humans. J Nutr 2001, 131, (10), 2648-2652.

167. Beattie, J.; Crozier, A.; Duthie, G. G., Potential health benefits of berries. Curr Nutr Food Sci 2005, 1, (1), 71-86.

168. Kay, C. D., Aspects of anthocyanin absorption, metabolism and pharmacokinetics in humans. Nutr Res Rev 2006, 19, (1), 137-146.

169. McGhie, T. K.; Walton, M. C., The bioavailability and absorption of anthocyanins: Towards a better understanding. Mol NutrFood Res 2007, 51, (6), 702-713.

170. Youdim, K. A.; Martin, A.; Joseph, J. A., Incorporation of the elderberry anthocyanins by endothelial cells increases protection against oxidative stress. Free Radic Biol Med 2000, 29, (1), 51-60.

171. Tsuda, T.; Horio, F.; Osawa, T., Absorption and metabolism of cyanidin 3-O- beta-D-glucoside in rats. FEBS Lett 1999, 449, (2-3), 179-182.

172. Seeram, N. P.; Adams, L. S.; Zhang, Y.; Lee, R.; Sand, D.; Scheuller, H. S.; Heber, D., Blackberry, Black Raspberry, Blueberry, Cranberry, Red Raspberry, and Strawberry Extracts Inhibit Growth and Stimulate Apoptosis of Human Cancer Cells In Vitro. J Agric Food Chem 2006, 54, (25), 9329-9339.

173. Giusti, M. M.; Wrolstad, R. E., Characterization of red radish anthocyanins. J Food Sci 1996, 61, (2), 322-326.

174. Jing, P.; Noriega, V.; Schwartz, S. J.; Giusti, M. M., Effects of Growing Conditions on Purple Corncob (Zea mays L.) Anthocyanins. J Agric Food Chem 2007, 55, (21), 8625-8629.

175. Giusti, M. M.; Wrolstad, R. E., Characterization and measurement of anthocyanins by UV-Visible spectroscopy. In Handbook of Food Analytical Chemistry, Pigments, Colorants, Flavors, Texture, and Bioactive Food Components Wrolstad, R. E.; 157 Acree, T. E.; Decker, E. A.; Penner, M. H.; Reid, D. S.; Schwartz, S. J.; Shoemaker, C. F.; Smith, D.; Sporns, P., Eds. Wiley-Interscience: Hoboken, NJ, 2004; pp 19-31.

176. Tian, Q.; Aziz, R. M.; Stoner, G. D.; Schwartz, S. J., Anthocyanin determination in black raspberry () and biological specimens using liquid chromatography-electrospray ionization tandem mass spectrometry. J Food Sci 2005, 70, (1), C43-C47.

177. Miyazawa, T.; Nakagawa, K.; Kudo, M.; Muraishi, K.; Someya, K., Direct intestinal absorption of red fruit anthocyanins, cyanidin-3-glucoside and cyanidin-3,5- diglucoside, into rats and humans. J Agric Food Chem 1999, 47, (3), 1083-1091.

178. Suda, I.; Oki, T.; Masuda, M.; Nishiba, Y.; Furuta, S.; Matsugano, K.; Sugita, K.; Terahara, N., Direct absorption of acylated anthocyanin in purple-fleshed sweet potato into rats. J Agric Food Chem 2002, 50, (6), 1672-1676.

179. Steinert, R. E.; Ditscheid, B.; Netzel, M.; Jahreis, G., Absorption of black currant anthocyanins by monolayers of human intestinal epithelial Caco-2 cells mounted in Ussing type chambers. J Agric Food Chem 2008, 56, (13), 4995-5001.

180. Tyssandier, V.; Reboul, E.; Dumas, J.-F.; Bouteloup-Demange, C.; Armand, M.; Marcand, J.; Sallas, M.; Borel, P., Processing of vegetable-borne in the human stomach and duodenum. Am J Physiol Gastrointest Liver Physiol 2003, 284, (6), G913-923.

181. Nemeth, K.; Plumb, G. W.; Berrin, J. G.; Juge, N.; Jacob, R.; Naim, H. Y.; Williamson, G.; Swallow, D. M.; Kroon, P. A., Deglycosylation by small intestinal epithelial cell beta-glucosidases is a critical step in the absorption and metabolism of dietary flavonoid glycosides in humans. Eur J Nutr 2003, 42, (1), 29-42.

182. Swallow, D. M., Genetics of lactase persistence and lactose intolerance. Annu Rev Genet 2003, 37, 197-219.

183. Wilkinson, A. P.; Gee, J. M.; Dupont, M. S.; Needs, P. W.; Mellon, F. A.; Williamson, G.; Johnson, I. T., Hydrolysis by lactase phlorizin hydrolase is the first step in the uptake of daidzein glucosides by rat small intestine in vitro. Xenobiotica 2003, 33, (3), 255-64.

184. Ichiyanagi, T.; Rahman, M. M.; Kashiwada, Y.; Ikeshiro, Y.; Shida, Y.; Hatano, Y.; Matsumoto, H.; Hirayama, M.; Tsuda, T.; Konishi, T., Absorption and metabolism of delphinidin 3-O-beta-D-glucopyranoside in rats. Free Radic Biol Med 2004, 36, (7), 930- 937.

158 185. Hassimotto, N. M. A.; Genovese, M. I.; Lajolo, F. M., Absorption and metabolism of cyanidin-3-glucoside and cyanidin-3-rutinoside extracted from wild mulberry ( nigra L.) in rats. Nutr Res 2008, 28, (3), 198-207.

186. Bub, A.; Watzl, B.; Heeb, D.; Rechkemmer, G.; Briviba, K., Malvidin-3- glucoside bioavailability in humans after ingestion of red wine, dealcoholized red wine and red grape juice. Eur J Nutr 2001, 40, (3), 113-20.

187. Matsumoto, H.; Ichiyanagi, T.; Iida, H.; Ito, K.; Tsuda, T.; Hirayama, M.; Konishi, T., Ingested delphinidin-3-rutinoside is primarily excreted to urine as the intact form and to bile as the methylated form in rats. J Agric Food Chem 2006, 54, (2), 578-582.

188. Jordheim, M.; Aaby, K.; Fossen, T.; Skrede, G.; Andersen, O. M., Molar Absorptivities and Reducing Capacity of Pyranoanthocyanins and Other Anthocyanins. J Agric Food Chem 2007, 55, (26), 10591-10598.

189. Andersen, Ø. M.; Jordheim, M. In Anthocyanins - food applications, Proceedings of the 5th International Congress on Pigments in Foods: For quality and Health, Helsinki, Finland, August 14-16, 2008; Helsinki, Finland, 2008.

190. Kraemer-Schafhalter, A.; Fuchs, H.; Pfannhauser, W., Solid-phase extraction (SPE) - a comparison of 16 materials for the purification of anthocyanins from aronia melanocarpa var Nero. J Sci Food Agric 1998, 78, (3), 435-440.

191. Masque, N.; Marce, R. M.; Borrull, F., New polymeric and other types of sorbents for solid-phase extraction of polar organic micropollutants from environmental water. Trends Anal Chem 1998, 17, (6), 384-394.

192. Chiriboga, C. D.; Francis, F. J., Ion exchange purified anthocyanin pigments as a colorant for cocktail. J Food Sci 1973, 38, (3), 464-467.

193. Andersen, O. M., Semipreparative isolation and structure determination of pelargonidin 3-O-a-L-rhamnopyranosyl-(1 -> 2)-b-D-glucopyranoside and other anthocyanins from the tree Dacrycarpus dacrydioides. Acta Chem Scand, Series B: Org Chem Biochem 1988, B42, (7), 462-468.

194. Cheng, Y.-F.; Neue, U. D.; Bonin, R.; Block, E.; Bean, L., Simplified procedure for the determination of codeine and its metabolite in urine and plasma by LC/UV and LC/MS using mixed-mode cation exchange for sample preparation. J Liq Chromatogr R T 2001, 24, (9), 1353-1364.

159 195. Giusti, M. M.; Rodriguez-Saona, L. E.; Griffin, D.; Wrolstad, R. E., Electrospray and tandem mass spectroscopy as tools for anthocyanin characterization. J Agric Food Chem 1999, 47, (11), 4657-4664.

196. Oszmianski, J.; Lee, C. Y., Isolation and HPLC determination of phenolic compounds in red grapes. Am J Enol Vitic 1990, 41, (3), 204-206.

197. Prior, R. L.; Lazarus, S. A.; Cao, G.; Muccitelli, H.; Hammerstone, J. F., Identification of and anthocyanins in blueberries and cranberries (Vaccinium spp.) using high-performance liquid chromatography/mass spectrometry. J Agric Food Chem 2001, 49, (3), 1270-1276.

198. Yi, W.; Akoh, C. C.; Fischer, J.; Krewer, G., Absorption of anthocyanins from blueberry extracts by caco-2 human intestinal cell monolayers. J Agric Food Chem 2006, 54, (15), 5651-5658.

199. He, J.; Rodriguez-Saona, L. E.; Giusti, M. M., Midinfrared spectroscopy for juice authentication-rapid differentiation of commercial juices. J Agric Food Chem 2007, 55, (11), 4443-4452.

200. Spagna, G.; Pifferi, P. G., Purification and separation of oenocyanin anthocyanins on sulfoxyethylcellulose. Food Chem 1992, 44, (3), 185-188.

201. Dobrev, P. I.; Kaminek, M., Fast and efficient separation of cytokinins from auxin and abscisic acid and their purification using mixed-mode solid-phase extraction. J Chromatogr A 2002, 950, (1-2), 21-29.

202. Mazza, G.; Cacace, J. E.; Kay, C. D., Methods of analysis for anthocyanins in plants and biological fluids. J AOAC Int 2004, 87, (1), 129-145.

203. Escribano-Bailon, M. T.; Santos-Buelga, C., Polyphenol extraction from foods. In Methods in Polyphenol Analysis, Santos-Buelga, C.; Williamson, G., Eds. Royal Society of Chemistry: Cambridge, 2003; pp 1-16.

204. Gu, L.; Kelm, M.; Hammerstone, J. F.; Beecher, G.; Cunningham, D.; Vannozzi, S.; Prior, R. L., Fractionation of polymeric procyanidins from lowbush blueberry and quantification of procyanidins in selected foods with an optimized normal-phase HPLC- MS fluorescent detection method. J Agric Food Chem 2002, 50, (17), 4852-4860.

205. Svedstrom, U.; Vuorela, H.; Kostiainen, R.; Huovinen, K.; Laakso, I.; Hiltunen, R., High-performance liquid chromatographic determination of oligomeric procyanidins from dimers up to the hexamer in hawthorn. J Chromatogr A 2002, 968, (1-2), 53-60.

160 206. Rivera-Sagredo, A.; Canada, F. J.; Nieto, O.; Jimenez-Barbero, J.; Martin-Lomas, M., Substrate specificity of small-intestinal lactase. Assessment of the role of the substrate hydroxyl groups. Eur J Biochem 1992, 209, (1), 415-422.

207. Goda, T., Regulation of the expression of carbohydrate digestion/absorption- related genes. Br J Nutr 2000, 84 Suppl 2, S245-S248.

208. Pierluca Mazzuca, P. F. G. P. L. C. F. A., Mass spectrometry in the study of anthocyanins and their derivatives: differentiation of and hybrid grapes by liquid chromatography/electrospray ionization mass spectrometry and tandem mass spectrometry. J Mass Spectrom 2005, 40, (1), 83-90.

161