Citation
Hydrolytic and Antioxidant Properties of Ellagic Acid and Its Precursors Present in Muscadine Grape

Material Information

Title:
Hydrolytic and Antioxidant Properties of Ellagic Acid and Its Precursors Present in Muscadine Grape
Creator:
Lee, Joonhee ( Dissertant )
Talcott, Stephen ( Thesis advisor )
Publication Date:
Copyright Date:
2004

Subjects

Subjects / Keywords:
Acetates ( jstor )
Antioxidants ( jstor )
Enzymes ( jstor )
Flavonoids ( jstor )
Food ( jstor )
Glycosides ( jstor )
Grapes ( jstor )
Hydrolysis ( jstor )
Juices ( jstor )
Skin ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Lee, Joonhee. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
12/18/2004
Resource Identifier:
57735004 ( OCLC )

Downloads

This item has the following downloads:


Full Text












HYDROLYTIC AND ANTIOXIDANT PROPERTIES OF ELLAGIC ACID AND ITS
PRECURSORS PRESENT IN MUSCADINE GRAPE
















By

JOONHEE LEE


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2004

































Copyright 2004

by

Joonhee Lee















ACKNOWLEDGMENTS

It is a privilege to me to acknowledge the unconditional support of my supervisory

committee chair, Dr. Stephen Talcott, who has been a teacher, mentor and friend during

my years as a graduate student. His advice, wisdom, and constant help have made this

dissertation possible.

I sincerely appreciate the help offered by the members of my supervisory

committee, Dr. Sims, Dr. Marshall, and Dr. Saba. Their cooperation and suggestions have

improved the quality of my dissertation considerably.

Being a member of the Department of Food Science and Human Nutrition has been

a wonderful academic experience and a unique opportunity to meet extraordinary people

who helped me immensely along the way. I would like to give special deep thanks to

current and past lab mates; David, Flor, Youngmok, Chris, Lanier, Kristine, Stacy,

Lisbeth, Janelle, Melanie, Jennifer, Danielle, and Angela.

My deepest recognition goes to my beloved parents, who helped me in any

imaginable way to achieve my objectives and fulfill my dreams. They have been an

inexhaustible source of love and inspiration all my life.

My most special thanks go to my husband Jeongho for his patience, understanding,

and encouragement, without which it would have been impossible to complete my

degree. Finally, I thank to my specially beloved Minsuh, hoping that the effort of these

years may offer him a more plentiful life in the years to come.
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ................................................................................................. iii

LIST OF TABLES ................................................... vii

LIST OF FIGURES ........................................... ............................ viii

A B S T R A C T ........................................................................................................ ........ .. x

CHAPTER

1 IN TR O D U C T IO N ........ .. ......................................... ..........................................1.

Justification ................................................................................................... ...............1
O bjectiv e s ........................................................................ ................................. . .2

2 LITER A TU R E R EV IEW .................................................................... ...............3...

Muscadine Grapes ................................ .........................3
Phytochem icals in M uscadine Grapes ....................... ..........................................4...
E llagic A cid and Its Precursors ........................................................ ...............4...
Ellagic Acid and Quality of Muscadine Products ..........................................10
A n th o cy an in s ....................................................................................................... 12

3 FRUIT MATURITY AND JUICE EXTRACTION INFLUENCES ELLAGIC
ACID DERIVATIVES AND OTHER ANTIOXIDANT POLYPHENOLICS IN
M U SC A D IN E G R A P E S ............................................................................................14

In tro d u ctio n ............................................................................................................... .. 14
M materials and M ethods .. ..................................................................... ............... 15
C hem ical A nalyses .............. .................. .............................................. 15
Statistical A analysis .............. .................. .............................................. 17
R results and D discussion .................................................... .......... .. ............ ............ 17
Identification of Ellagic Acid and Its Precursors ..........................................17
E llagic A cid and Its D erivatives..................................................... ............... 18
A n th o cy an in s ............................... ..................................................................... 2 1
Total Phenolics and Antioxidant Capacity ..................................... ................ 24
C o n c lu sio n ............................................................................................................... .. 2 7









4 IDENTIFICATION OF ELLAGITANNINS AND CONJUGATES OF ELLAGIC
A CID IN M U SCAD IN E GRA PE S ............................................................................28

In tro d u ctio n ................................................................................................................. 2 8
M materials and M ethods .. ..................................................................... ................ 29
Iso latio n s .............................................................................................................. 2 9
A analysis by H PL C -PD A ....................................... ....................... ................ 30
A analysis by H PLC -M Sn ................................... .... ................ 31
Total Polyphenols and Antioxidant Capacity ................................................31
Statistical A naly sis .............. ...... ............ .............................................. 32
R results and D discussion ....................................................................... .......... ............ .......... .. 32
Isolations and Quantification of Polyphenolic Compounds by Solid Phase
E extraction ............... . ... .... ... ..... ... ............................ 32
Identifications of Ellagic Acid Derivatives and Flavonoids by HPLC-PDA
and M Sn ............. .... ........... ................ ............... 36
Ellagic acid derivatives in Isolate I ......................................... ................ 37
Ellagic acid derivatives and flavonoid glycosides in isolate II .................41
C o n c lu sio n s................................................................................................................ 4 5

5 HYDROLYTIC PROPERTIES OF ELLAGIC ACID DERIVATIVES IN
M U SCAD INE GRAPES ................... .............................................................. 46

In tro d u ctio n ................................................................................................................ 4 6
M materials and M ethods .......................................................................... ................ 47
Response Surface Methodology (RSM) and Statistical Analyses....................47
E nzym es P reparation .......................................... ......................... ................ 47
Chem ical A analysis ... ................................................................................ 48
R results and D iscu ssion ................................................ ................. .......... .......... .. 49
Ellagic Acid Derivatives and Flavonoid Glycosides: Effects of Time,
T em perature and pH s .............................................. ...... ........... ................ 49
Antioxidant Capacity of Polyphenolics as Affected by Aglycone vs
G lycosides w ith H ydrolysis ........................................................ ................ 53
Enzymatic Hydrolysis of Ellagic Acid Derivatives.......................................61
C o n c lu sio n s............................................................................................................... .. 6 4

6 HYDROLYTIC AND OXIDATIVE PROPERTIES ON ELLAGIC ACID
DERIVATIVES DURING STORAGE OF MUSCADINE GRAPES JUICES .........65

In tro d u ctio n ................................................................................................................ 6 5
M materials an d M eth od s ............................................................................. ............... 66
Storage of R ed Juices and Isolations.............................................. ................ 66
Storage of W hite Juices and Isolations........................................... ................ 66
Chemical Analysis ........................................................................ 67
R results and D discussion .............................................. ....... ...... .. ........... ............ 67
Changes of Ellagic Acid Derivatives during Storage of Red Juice..................67
Initial Ellagic Acid Derivatives in White Muscadine Juice as Affected by
A scorbic A cid and A ir ....................................... ...................... ................ 69


v









Post-storage Levels of Ellagic Acid Derivatives in White Muscadine Juice as
Affected by Ascorbic Acid and Air .................................................. 76
Conclusions .............................................................................. ........ ......... ............... 81

7 SUM M ARY AND CON CLU SION ...................................................... ................ 83

L IST O F R E F E R E N C E S ...................................................................................................85

BIOGRAPHICAL SKETCH ...................................................... 92















LIST OF TABLES


Table page

3-1 The concentrations of free ellagic acid, two ellagic acid glycosides and total
ellagic acids of muscadine grapes as affected by cultivars and ripening stages ...... 19

3-2 Concentration of six anthocyanidins and total anthocyanidins of red muscadine
grapes as affected by ripening stages .................................................. ................ 22

3-3 Concentrations of total soluble phenolics in methanolic and ethyl acetate extracts
as affected by cultivars and ripening stages ....................................... ................ 25

3-4 Antioxidant capacity of methanolic and ethyl acetate extracts as affected by
cultivars and ripening stages .................................... ...................... ................ 26

4-1 The concentrations of free ellagic acid, ellagic acid glycosides and total ellagic
acids on each fraction from three different cultivars........................... ................ 33

4-2 UVmax and HPLC-ESI(-)-MSn analyses of polyphenols in isolate I from
muscadine grapes. ......... .. ................... .... .............. ...............38

4-3 UVmax and HPLC-ESI(+)/(-)-MSn analyses of ellagitannins, glycosides of
ellagic acid and flavonoids in isolate II from muscadine grapes. ..........................44

5-1 Pearson correlations coefficients of individual ellagic acid derivatives contents
with total soluble phenolics and antioxidant capacity......................... ................ 60

5-2 Concentrations of ellagic acid, ellagic acid glycosides and gallic acid affected
by enzyme treatment in two different fractions from Doreen extracts..................63















LIST OF FIGURES


Figure page

2-1 E llagic acid chem ical structure ............................................................. ...............5...

2-2 Ellagic acid glycosides found in raspberry ........................................... ...............6...

2-3 Ellagitannins (Lambertianin C) found in raspberry fruits....................................7...

2-4 Ellagitannins conversion to ellagic acid via hexahydroxydiphenic acid (HHDP).....7

2-5 Chemical structures of punicalagin found in pomegranate juice ...............................8

2-6 UV spectra of pomegranate juice characteristic compounds. ................................... 8

2-7 Characteristic UV spectra of hydrolysable tannins in birch leaves.........................9...

2-8 Chemical structures of Geraniin and Amariin, containing
dehydrohexahydroxydiphenyl (DHHDP) group ................................... ...............9...

4-1 Fraction scheme and tentative classification of polyphenolics present in
m ethanolic extracts of m uscadine grapes ........................................... ................ 29

4-2 Total soluble phenolics and antioxidant capacities of five fractions from three
different cultivars. .................................. ........ ...... ...............36

4-3 HPLC-PDA chromatogram (280 nm) of Isolate I of muscadine grapes...............37

4-4 UV spectra of ellagic acid derivatives in Isolate I............................... ................ 40

4-5 HPLC-PDA chromatogram (280 and 360 nm) of Isolate II of muscadine grapes...42

5-1 Hydrolysis time and temperature combinations included in central composite
design (C C D ) experim ent ........................................ ........................ ................ 48

5-2 Tridimensional representation of free ellagic acid generated using fold increases
by response surface model in the absence and presence of 2N HCl.....................50

5-3 Tridimensional representation of ellagic acid xyloside generated using fold
increases by response surface model in the absence and presence of 2N HC1. .......51









5-4 Tridimensional representation of ellagic acid rhamnoside generated using fold
increases by response surface model in the absence and presence of 2N HC1. .......52

5-5 Tridimensional representation of myricetin rhamnoside generated using fold
increases by response surface model in the absence and presence of 2N HC1. .......54

5-6 Tridimensional representation of quercetin rhamnoside generated using fold
increases by response surface model in the absence and presence of 2N HC1. .......55

5-7 Tridimensional representation of kaempferol rhamnoside generated using fold
increases by response surface model in the absence and presence of 2N HC1. .......56

5-8 Tridimensional representation of total soluble phenolics generated using fold
increases by response surface model in the absence and presence of 2N HC1. .......58

5-9 Tridimensional representation of antioxidant capacity generated using fold
increases by response surface model in the absence and presence of 2N HC1. ......59

6-1 Changes in whole red juice for ellagic acid derivatives during storage................70

6-2 Changes in water isolate for ellagic acid derivatives during storage ....................71

6-3 Changes in ethyl acetate isolate for ellagic acid derivatives during storage............72

6-4 Changes in methanol isolate for ellagic acid derivatives during storage ..............73

6-5 Concentrations of ellagic acid derivatives depending on treatments of
ascorbic acid and air at Oday of Doreen juice as affected by thermal
pasteurization m ethod ..................................................................... ................ 75

6-6 Concentrations of free ellagic acid depending on ascorbic acid and air during
storage of Doreen juice as affected by thermal pasteurization...............................77

6-7 Concentrations of ellagic acid xyloside depending on ascorbic acid and air
during storage of Doreen juice as affected by thermal pasteurization ..................78

6-8 Concentrations of ellagic acid rhamnoside depending on ascorbic acid and air
during storage of Doreen juice as affected by thermal pasteurization ..................79

6-9 Concentrations of total ellagic acid depending on ascorbic acid and air during
storage of Doreen juice as affected by thermal pasteurization...............................80















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

HYDROLYTIC AND ANTIOXIDANT PROPERTIES OF ELLAGIC ACID AND ITS
PRECURSORS PRESENT IN MUSCADINE GRAPE

By

Joonhee Lee

December 2004

Chair: Stephen T. Talcott
Major Department: Food Science and Human Nutrition

Muscadine grape (Vitis rotundifolia) has potential health-promoting benefits and

antioxidant properties from characteristic phytochemicals such as ellagic acid, ellagic

acid glycosides, and ellagitannins. To provide fundamental information on ellagic acid

derivatives in this commodity, typical polyphenolic compounds were characterized in

eight cultivars. All polyphenolics generally increased as fruit ripened and the highest

concentrations were located in the skins. Hot-pressed juices contained considerably lower

polyphenolic concentrations than were present in whole grapes with actual recovery

varying widely among cultivars. Antioxidant capacity was appreciably influenced by

cultivar, maturity, and location in the fruit with good correlations to soluble phenolics

found in both methanolic and ethyl acetate extracts (r=0.83 and 0.92, respectively).

Glycosidic forms of ellagic acid and major flavonoids were isolated by a series of solid-

phase extractions. By using HPLC-MS/PDA, chemical identities were elucidated as

ellagic acid xyloside, ellagic acid rhamnoside, myricetin rhamnoside, quercetin









rhamnoside and kaempferol rhamnoside. Ellagitannins, a major source for hydrolytic free

ellagic acid, were present in the C 18 non retained fraction and their molecular weights

having never been reported were determined.

Because of the lack of data on hydrolytic properties of ellagic acid precursors,

central composite design was applied to demonstrate the evolution of hydrolytic ellagic

acid depending on various time-temperature combinations in response surface

methodology. Hydrolysis of ellagic acid precursors would be more temperature-

dependent than time-dependent, and the resultant additional free ellagic acid showed

good correlation to antioxidant capacity (r=0.83). Using enzymes, 13-glucosidase (E.C.

3.2.1.21) or tannase (E.C. 3.1.1.20) is an alternative application to hydrolyze ellagic acid

precursors, 13-glucosidase showed better reactivity on ellagic acid derivatives compare to

tannase.

During storage, ellagic acid glycosides were relatively stable with ellagitannin

hydrolysis the major source for evolution of free ellagic acid in a juice system. Free

ellagic acid is partially responsible for the formation of insoluble sediments in addition to

other juice constituents such as metal ions or insoluble pectins. Ascorbic acid fortification

and air sparging did not significantly affect concentrations of ellagic acid derivatives in

stored juices. However, a mild heat treatment influenced ellagitannin hydrolysis and was

likely associated with activation of natural enzymes present in the juice.

Results provided by these studies suggest beneficial reasons to consume muscadine

grapes leading to improve the market value of this crop not only as fresh grapes, juice or

wine but also as health promoting components in various types of processed food.














CHAPTER 1
INTRODUCTION

Justification

Muscadine grapes (Vitis rotundifolia) are commonly cultivated in the southeastern

U.S. as an excellent alternative fruit crop, because many traditional Vitis species are

impossible to survive due to Pierce's disease and typical climatic characteristics in this

region. Muscadine grapes are favorably consumed as fresh fruit, juice, wine or jelly not

only for distinguished aroma and flavor characteristics but also for positive health

benefits from characteristic phytochemicals including ellagic acid/ellagic acid derivatives

and anthocyanins 3,5-diglucosides. Presence of these compounds has been associated

with quality defects such as rapid color deterioration of anthocyanin 3,5-diglucosides and

the formation of insoluble sediments that may be affected by ellagic acid derivatives

during storage. To solve the problem with sediment formation, research has focused on

removing free ellagic acid from the matrix using various processing techniques (1, 2).

However, none of the trials successfully solved the sedimentation problem due to lack of

fundamental information on these compounds in muscadine juice or wine. Therefore, the

primary objective of this work is to selectively identify and determine those compounds

that impact levels of free ellagic acid in efforts to determine their concentration and

stability in fresh and processed muscadine grapes. Mechanisms for their hydrolysis and

relative changes are evaluated that may reveal key components affecting their radical

scavenging properties and the accumulation of free ellagic acid during storage of

muscadine juice. These studies also seek to determine the antioxidant polyphenolics in









muscadine grapes as influenced by their location in the grape, those extracted into juice,

and solubility in various fractionations as a function of cultivar and maturity. These

evaluations are beneficial for exploring biochemical synthesis and/or fate of pro-ellagic

acid compounds. Finally, chemical and physical characteristics of these identified

compounds will lead to possible mechanisms) concerning their hydrolysis and release of

free ellagic acid into insoluble sediments. This information will add market value and

significantly increase marketability for this under-utilized grape by identification of novel

compounds and evaluation of their antioxidant potential.

Objectives

* Objective 1: To quantify the antioxidant polyphenolics in muscadine grapes as
influenced by their location in the grape, juice production, and polyphenolic
fractionation as a function of cultivar and maturity.

* Objective 2: To isolate and identify ellagic acid glycosides and ellagitannins using
advanced HPLC methodologies (PDA, MS) and evaluate their antioxidant capacity.

* Objective 3: To investigate the hydrolytic and oxidative properties of individual
ellagic acid derivatives and other polyphenolics affecting concentrations and
antioxidant capacity.

* Objective 4: To determine the major source of hydrolytic ellagic acid, and to
elucidate the effects of ascorbic acid fortification and oxidation on changes of
ellagic acid derivatives in juice system.














CHAPTER 2
LITERATURE REVIEW

Muscadine Grapes

Muscadine grapes are botanically categorized under the genus Vitis, which consists

of two subgenera: Euvitis ("bunch grapes") and Muscadinia ("berry grapes") (3). Euvitis

includes diverse species of grapes including V. vinifera, which is the most common grape

in the world. Muscadinia can be represented as an American grape species V.

rotundifolia Michaux, muscadine grape, and V. munsoniana Simpson. When muscadine

grapes are grown for commercial harvest, the majority of plantings are V. rotundifolia (4)

and they are indigenously found in the southeastern United States, from Delaware to

central Florida and along the Gulf of Mexico to eastern Texas (3). In these regions,

muscadine grapes are favorably cultivated fruit crops since other grapes are hard to

survive in typical climate like humid summers and warm winters (5, 6). Another benefit

of growing muscadine grapes is that V rotundifolia has excellent resistance to pests and

various diseases, namely Pierce's disease, which is a lethal disease of the grape vine

caused by the bacterium Xylellafastidiosa and the main obstacle for production of V

vinifera in the south and southeastern states (3). Unlike bunch grapes, muscadine grape

fruits develop in round clusters, containing approximately 4-16 grapes. The fruit are

round or oval, larger varieties may reach 26 mm in diameter and weigh from 3-10 g each.

Fruits can range from light-skinned (green, pearly white, or bronze) to dark red in color

and most have a tough, leathery skin with the pulp inside ranging from meaty, to melting,

or juicy (6). At present, home gardening or commercial production is provided by over 70









different cultivars, the oldest of which includes Scuppernong (green), Carlos (green),

Doreen (green), Noble (red), and Magnolia (green). These cultivars are consumed as

fresh fruit, juice, wine, or jelly (7). In the production of either fermented or unfermented

muscadine grape products, the physicochemical composition of fruit is one of the most

important factors affecting overall quality and acceptability. In direct comparisons with

V. vinifera, muscadine grapes generally have a lower soluble solids concentration (SSC)

and titratable acidity (8) ranging from 10-18% (13.2% on average) and 0.39-1.5% (0.84%

on average) expressed as tartrate equivalents, respectively; resulting in a pH range of 2.9-

3.4 (pH 3.14 on average). Muscadine "hulls" (skin plus firmly attached tissue layer) were

found to contain significantly higher organic acid concentrations than the pulp on a per

weight basis. This helps to explain how juices that are immediately pressed have lower

titratable acidity values compared to those for hot-pressed juices or skin fermentation (9,

10).

Phytochemicals in Muscadine Grapes

Ellagic Acid and Its Precursors

Ellagic acid (Figure 2-1) has been found in many woody plants and in diverse fruits

and nuts in various concentrations. Interestingly ellagic acid is present only in Vitis

rotundifolia but not in Vitis vinifera. Ellagic acid is believed to be formed by the

hydrolytic release from ellagic acid derivatives including ellagic acid glycosides (Figure

2-2, A-C) and ellagitannins (Figure 2-3). The presence of ellagic acid in various fruits

and nuts was determined for the purpose of botanical classification, and was identified in

strawberries, blackberries, and walnuts (11). Concentrations of ellagic acid in various

fruits and nuts were determined as total ellagic acid, which was hydrolytic ellagic acid

from ellagic acid precursors followed by complete hydrolysis with acid (12). This was









done because analyses on individual ellagic acid derivatives were difficult with lower

analytical techniques and in the absence of authentic standards. Appreciable

concentrations of ellagic acid were detected after acid hydrolysis in strawberries (630

utg/g), raspberries (1500 utg/g), blackberries (1500 utg/g), walnuts (590 utg/g), pecans (330

[tg/g), and cranberries (120 [tg/g). In strawberries, most ellagic acid (95.7%) was found in

pulp, and the remainder was found in seeds. However, the seeds of raspberries contained

87.5% ellagic acid, while the pulp had 12.2%. Amakura et al. (13) determined ellagic

acid levels in fresh and processed fruits by a simple and rapid high performance liquid

chromatographic (HPLC) method that did not involve a hydrolysis step. Values for

similar commodities (blackberry, 87.6 utg/g; strawberry, 17.7 utg/g; and raspberry, 5.84

utg/g) were appreciably lower without the hydrolysis step; possibly indicating the

presence of ellagic acid precursors including ellagic acid glycosides and ellagitannins.

Ellagic acid glycosides are forms of a single sugar moiety linked to the hydroxyl at

the 4-position of ellagic acid, and five or six different ellagic acid glycosides have been

separated in raspberry (14, 15): ellagic acid 4-acetylxyloside, ellagic acid 4-arabinoside,

and ellagic acid 4-acetylarabinoside (Figure 2-2;). Muscadine grapes also contain two

ellagic acid glycosides, which are characterized by 3-7 nm hypsochromic shifts in UV

spectra (252 and 360 nm) (16). However, free ellagic acid and ellagic acid glycosides

0
0 OH

HO OH


HO


Figure 2-1. Ellagic acid chemical structure









contribute to only part of the total ellagic acid on acid hydrolysis; highly indicating the

presence of ellagitannins in muscadine grapes. Ellagitannins (Figure 2-3) are

characterized as hydrolysable conjugates containing one or more hexahydroxydiphenoyl

(HHDP) groups esterified to a sugar, mainly glucose. HHDP groups are released from the

main structure leading to spontaneous conversion into ellagic acid by hydrolysis (Figure

2-4). Specific information on ellagic acid derivatives are lacking for muscadine grapes,

but the presence of ellagic acid and its derivatives in muscadine grapes may add value

and marketability to the crop due to possible health benefits such as its antioxidant

activity (17, 18), anti-carcinogenic properties influencing cell cycle arrest and apoptosis

(19), and inhibition of tumor formation and growth in mammalian models (20, 21). The


A 0
O0 OH
OH / \_ O

HO 0
0

B 0
COCH3-0 O OH
o / \oOH
HO OH
HO 0-O
0

0
C 0 OH

COCH3 0-OH
HO 0
0

Figure 2-2. Ellagic acid glycosides found in raspberry. A) Ellagic acid 4-arabinoside, B)
Ellagic acid 4-acetylarabinoside, C) Ellagic acid 4-acetylxyloside. (Mullen et.
al. Phytochemistry. 2003, 64, 617-624)










types of ellagitannins are varied with approximately 500 different compounds isolated

and identified in nature (22). Structural diversity in ellagitannins originates from the

number of HHDP units, the location of galloyl ester groups participating in biaryl

linkage, and the conformation of the glucose ring (23). HPLC assisted by mass

spectrometry and diode array detections are the most commonly employed to separate

and identify various ellagitannins from fruits or plants extracts. Mullen et. al. (14)

OH



0OH HOb- C
HO HO Loi 0iT04

C 0O OH-
OI. 10 -_ 0\ l" "0 .... f. -o t- OH h



O- \ P C HO OH l -O OH
NoH5 C-T 11C '()]I


HO OH OH Oil

F0 Oli OH OH


Figure 2-3. Ellagitannins (Lambertianin C) found in raspberry fruits. (Mullen et. al.
Phytochemistry 2003, 64, 617-624)



OH CH



140 Hylysis HO QH
HO H HHO Q HO 0 I0 H3 C4H
ElImtainin H n.stum dIlh.rdc I (-HHFP)

^0 OH

HO H

EIla9c Aidd
Figure 2-4. Ellagitannins conversion to ellagic acid via hexahydroxydiphenic acid
(HHDP). (Quideau and Feldman, Chem. Rev. 1996. 96, 475-503)









reported phytochemical profiles in raspberry including anthocyanins, quercetin

conjugates, ellagic acid glycosides and 3 different ellagitannins, Sanguiin H-10 (3

HHDPs, mass: 1568), Lambertianin C (6 HHDPs, mass: 2804) and Sanguiin H-6 (4

HHDPs, mass: 1870). All of these had observed maximum UV absorbance at 250 nm, but

it could be lower than 250 nm since they scanned from 250 to 700 nm. Punicalagins

(Figure 2-5) are different types of ellagitannins found in fruits, especially pomegranate

(24, 25) and composed of glucose, HHDP and gallagyl acid (ellagic + 2 gallic), which



P0








Figure 2-5. Chemical structures of punicalagin found in pomegranate juice as main
ellagitannins (Gill et. al. J. Agric. Food Chem. 2000, 48, 4581-4589).















Figure 2-6. UV spectra of pomegranate juice characteristic compounds. A;
galloylglucose, C; hydrolyzable tannins, D; punicalagin, G; ellagic acid (Gill
et. al. J. Agric. Food Chem. 2000, 48, 4581-4589).









had a resulting UV spectrum showing maxima around 375 and 265 nm (Figure 2-6, D).

Unidentified hydrolysable tannins (ellagic acid + gallic acid + tertgallic acid + etc.) were

also observed with characteristic UV spectra, maximum at 266 nm (Figure 2-6, C).

Similar UV spectra were reported with birch (Betula pubescens) leaves for bis-HHDP-

glucopyranose isomers (26, Figure 2-7). They also observed that UV spectra of

hydrolysable tannins were very similar to gallic acid spectra with 3-7 nm hypsochromic

2nm gallic acid 19 O 6


o(s 2 na 2m _____________



S17 \ 10 \ 18


__a___w__ __ SB s I____Q a 2!nJ MD fl 2A4
Figure 2-7. Characteristic UV spectra of hydrolysable tannins in birch (Betula pubescens)
leaves. 19:pentagalloylglucopyranose isomers, 6:bis-HHDP-glucopyranose
isomers, 17:galloyl-bis-HHDP-glucopyranose isomers, 10:digalloyl-HHDP-
glucopyranose isomers, 18:trigalloyl-HHDP-glucopyranose. (Gill et. al. J.
Agric. Food Chem. 2000, 48, 4581-4589).













Figure 2-8. Chemical structures of Geraniin and Amariin, containing

dehydrohexahydroxydiphenyl (DHHDP) group, found in Phyllanthus amarus
(Foo. Phytochemistry. 1995, 39, 217-224).









shifts depending on the number of galloyl groups. In some plants, ellagitannins are found

in oxidized form as containing dehydrohexahydroxydi-phenyl (DHHDP) group, and

geraniin and amariin have been identified by NMR data (27). The released DHHDP units

may not be able to convert to ellagic acid because two linked galloyl molecules in HHDP

will be inhibited from making a lactone ring by additional linkage (Figure 2-8).

Ellagic Acid and Quality of Muscadine Products

The presence of ellagic acid in muscadine grape and its products are important not

only because of its potential health benefits but also because of their possible contribution

to form insoluble materials in processed juice and wine. Ellagic acid, hardly soluble in

water, results in a significant defective role in the quality perception of wines and juices.

In general, V. vinifera is not known to contain ellagic acid in its seed, skin or pulp, but

small amounts are usually detected following oak barrel storage and aging. Boyle and

Hsu (28) evaluated ellagic acid concentrations in juices from 11 cultivars of muscadine

grapes and found a range from 1.6-23 [tg/mL, with concentrations influenced by skin

color. Ellagic acid evaluations in muscadine have been frequently conducted with its

products like juice or wine with regard to sediment formation during storage. Boyle and

Hsu (28) reported that ellagic acid is the only compound detectable in sediment, present

as yellowish to red crystals; however, recent quantitative analysis of the collected

sediments revealed that no more than 12% free ellagic acid by weight was actually

present in the sediments. The remaining constituents consist of either unidentified

compounds or conjugated forms of ellagic acid (16). Many efforts to prevent ellagic acid

sedimentation in muscadine juices and wines have been employed, including chemical

and physical remediation procedures, but none these were successful in solving the

sedimentation problem. Lin and Vine (1) treated Magnolia (a white cultivar) juice with









increasing concentrations of fining agents such as polyvinylpolypyrrolidone (PVPP) and

gelatin, and found that the highest concentration of PVPP (1.08 g/L) was most effective

in reducing ellagic acid. Gelatin (0.4 g/L) also decreased ellagic acid concentration in

sediment by 56% in red muscadine juice (2). An ultrafiltration technique involving

passing juice through a 10,000-30,000 dalton molecular weight membrane demonstrated

at most a 50% reduction of ellagic acid in sediment (2). However, these chemical and

physical remediation procedures only lowered the levels of ellagic acid or sediment and

ellagic acid precipitation continued during further storage suggesting that ellagic cid

could be hydrolyzed from larger molecules like ellagic acid glycosides or ellagitannins

(16, 29). According to the work of Garrido et. al. (2), the formation of ellagic acid

sediments in white muscadine juice was accelerated by increased storage temperature and

following thermal pasteurization (1000C for 10 min), which resulted in more sediment

than sterile filtered juices after 8 months storage at 1.50C. Sims and Bates (30)

investigated the effect of skin fermentation time on ellagic acid sedimentation of Noble

muscadine grape wine. Wines fermented with skins for four and six days had greater

amounts of ellagic acid sediment than non-skin and 2-day skin fermented wines.

Muscadine juice is normally manufactured by two extraction techniques depending on

the final intentions for use. Hot-pressed juices with red cultivars are made following

crushing and heating at 700C prior to pressing, while cold-pressed juices are pressed

immediately after crushing white grapes (8). The main purpose of the hot-pressing

procedure is to increase juice yield and to improve overall juice quality including high

intensity of color through extracting more phytochemicals from fruits. Hot-pressing is









likely to influence the extraction of ellagitannins or ellagic acid glycosides resulting in an

increase of ellagic acid precipitation in red muscadine juice after 50 days storage (16).

Anthocyanins

Other than ellagic acid derivatives, muscadine grapes also contain distinguishable

phytochemicals, anthocyanin 3,5-diglucosides, which have been identified as delphinidin,

cyanidin, petunidin, peonidin, and malvidin in non-acylated forms (31, 32). Even though

anthocyanin stability is influenced by several factors during food processing, the presence

of anthocyanin 3,5-diglucosides is the main reason for rapid color loss during storage of

muscadine juice or wine due to low stability of diglucosides form compared to

corresponding monoglucosides forms of anthocyanins (11). It has been found that color

loss by oxidation of anthocyanins was correlated to a decrease in radical scavenging

activity (17). In order to protect the degradation of color in muscadine products,

alternative processing schemes have been recently employed to understand the chemical

nature of 3,5-diglucoside anthocyanins and to consequently lead the economic growth of

this crop (33, 34). Usually anthocyanins are stabilized and develop intense color by

chelation with metal ions or binding with colorless polyphenolics and this is known as

"copigmentation" reactions. Associated with muscadine grape products, copigmentation

can be an alternative strategy to improve quality and market value since it has been

reported that incremental addition of rosemary extract (0- 0.4% v/v) affect the

hyperchromic shift of anthocyanins corresponding to increased antioxidant activity

through copigment complexes with anthocyanins (33). Diverse technological

improvements have been employed to replace the heating process because heat is a prime

source for quality loss during food process. High hydrostatic pressure (HHP), a promising

alternative to traditional pasteurization technologies (35-37), has been employed for






13


muscadine grape products in an effort to preserve thermolabile phytonutrients and

favorable copigmentation between anthocyanins and plant based polyphenolics such as

rosemary and thyme extracts (33, 34). However, HHP can hinder improving juice quality

by the presence and/or activation of residual enzymes such as polyphenol oxidase (PPO)

due to accelerated oxidation of anthocyanins by activated PPO.














CHAPTER 3
FRUIT MATURITY AND JUICE EXTRACTION INFLUENCES ELLAGIC ACID
DERIVATIVES AND OTHER ANTIOXIDANT POLYPHENOLICS IN MUSCADINE
GRAPES

Introduction

Depending on maturity and availability, it is common to blend grape cultivars for

muscadine wine and juice production to obtain the most desirable acidity, color, and

flavor. However, little information is available on the phytochemical and antioxidant

characteristics among cultivars suitable for wine or juice production.

The phytochemistry of muscadine grapes is distinguishable from most other grape

varieties due to its predominance of anthocyanin 3,5-diglucosides and presence of ellagic

acid and ellagic acid precursors (7). The anthocyanins 3,5-diglucosides, which may be

more resistant to degradation during thermal processing compared to monoglucosides,

are typically unstable during storage due to a decreased ability to form polymeric

pigments and are particularly prone to oxidation and browning reactions (38, 39). The

ellagic acid derivatives are the most distinguishing chemical attribute in muscadine grape

since these components have not been found in any Vitis species. Associated with quality

of muscadine juice or wine, ellagic acid has been considered as an undesirable element

even though it has potential health benefits because ellagic acid and its precursors are

believed to develop insoluble sediments during storage. Phenolic contents in different

muscadine cultivars have been reported on only free ellagic acid, resveratrol and other

flavonoids including myricetin, quercetin and kaempferol (17); however, the current

study represents impacts on free ellagic acid as well as ellagic acid glycosides and total









ellagic acid released by all ellagic acid precursors. The objectives of this study were to

quantify the antioxidant polyphenolics in muscadine grapes as influenced by their

location in the grape, juice production, and polyphenolic fractionation as a function of

cultivar and maturity. This information can be used to determine wine or juice blending

schemes to produce higher quality muscadine grape products in terms of phytochemical

composition and antioxidant potential.

Materials and Methods

Muscadine grapes were donated from local grape growers in central Florida and

collected at two maturity stages from the same vines at different time intervals, about 15-

20 days apart depending on variety. Varieties included Carlos, Fry, and Doreen, classified

as either white or more specifically bronze colored fruit, and the red-skinned varieties

Noble, Albemarle, Cowart, Nesbitt, and Georgia Red. Random samplings of 8-15 fruit in

duplicate were manually divided between skin and pulp, while whole grapes were

processed into juice using a hot-break technique (700C for 30 min). Polyphenolics were

extracted from the skin and pulp by homogenizing with 25 mL of 100% methanol,

filtered through Whatman #4 filter paper, and solvent removed at 400C under a stream of

nitrogen. The juice was analyzed directly following centrifugation and filtration. Non-

anthocyanin polyphenolics were subsequently partitioned from each isolate into ethyl

acetate in three sequential extractions after which the solvents were pooled, removed

under reduced pressure at 400C, and residues redissolved in 50% methanol.

Chemical Analyses

Polyphenolics were separated and quantified by HPLC using solvent programs to

identify phenolic acids, free ellagic acid, and ellagic acid derivatives in ethyl acetate

extracts, and total ellagic acid and individual anthocyanidins in methanolic extracts









following acid hydrolysis (2N HCI for 60 min at 950C). Separations were conducted on a

Dionex HPLC system using a PDA-100 photodiode array detector and a 250 mm x 4.6

mm Acclaim 120 C18 column (Dionex, Sunnyvale, CA) with a C18 guard column. Mobile

phases consisted of 100% water (phase A) and 60% methanol (phase B) both adjusted to

pH 2.4 with o-phosphoric acid and run at 1 mL/min according to modified conditions of

Lee and Talcott (16). Free ellagic acid, ellagic acid glycosides and phenolic acids were

separated using a gradient elution program where phase B changed from 0-30% in 3 min;

30-50% in 5 min; 50-70% in 17 min; 70-80% in 5 min; 80-100% in 5 min; and 100% in 9

min for a total run time of 44 min, after which the column was equilibrated to original

conditions in 1 min for the next sample injection. Anthocyanidins and total ellagic acid

were also separated with a gradient program that ran phase B from 30-50% in 3 min; 50-

70% in 2 min; 70-90% in 5 min; and 90-100% in 10 min and returning to original

composition in 1 min for column equilibration. Ellagic acid and its derivatives were

quantified in ellagic acid equivalents, flavonoid glycosides in equivalents of myricetin

(Sigma Chemical, St. Louis, MO), and each anthocyanidin quantified in cyanidin

equivalents (Polyphenols Laboratories AS, Sandnes, Norway).

Total soluble phenolics were analyzed using Folin-Ciocalteu assay (40) and

expressed in gallic acid equivalents (GAE). Antioxidant activity (41) was determined

using the oxygen radical absorbance capacity (ORAC) assay with fluorescein as modified

by Ou et. al. (42) from initial protocol by Cao et. al. (43). Fluorescence loss by reaction

with hydroxyl radical (70 min, 37C) was monitored on a Molecular Devices fmax

96well fluorescent microplate reader (Sunnyvale, CA) following appropriate dilution of









each isolate and data expressed in Trolox equivalents per g of fresh fruit or per mL of

juice.

Statistical Analysis

Data represent the mean of duplicate analyses with analysis of variance and Pearson

correlations conducted using JMP5 software (44); mean separation was conducted using

the LSD test (P<0.05).

Results and Discussion

Identification of Ellagic Acid and Its Precursors

The free (aglycone) form of ellagic acid and two ellagic acid glycosides were found

in all eight muscadine grape cultivars following ethyl acetate extraction and separation by

HPLC. The ellagic acid glycosides were tentatively characterized based on UV spectral

properties (252 and 360 nm) similar to that of free ellagic acid (252 and 365 nm) as was

previously characterized in muscadine grapes (16) indicating that these compounds were

most likely glycosidic forms at the 4-position of ellagic acid rather than HHDP moieties

esterified to glucose (true ellagitannins), with maximum absorption at or near 250 nm

(15, 45). Preliminary work to characterize these compounds has identified the presence of

glucose, xylose, or rhamnose moieties (data not shown). Similar ellagic acid glycosides

were thought to exist in raspberries and were characterized by spectroscopic shifts (4-7

nm hypsochromic) and disappearance of the glycoside after hydrolysis, with a

corresponding increase in free ellagic acid (15, 45- 47). Similarly, the two ellagic acid

glycosides identified in muscadine grapes yielded free ellagic acid upon both acid and

enzyme (13-glucosidase) hydrolysis. True ellagitannins, containing esterified HHDP units

to a carbohydrate, were also believed to be present in the grape isolates, but were not

separated or detected in muscadine grapes using the HPLC methodology employed.









Evidence of these highly polar compounds was established indirectly by passing an

aqueous grape extract through a pre-conditioned Waters C18 Sep Pak cartridge and

evaluating the non-retained fraction. No peaks analogous to ellagic acid were present in

this isolate in the range of 200-400 nm, but following acid hydrolysis free ellagic acid

was one of the hydrolytic products, thus providing evidence for their existence. Total

ellagic acid was subsequently determined from the methanolic extracts following acid

hydrolysis and represented the sum of free ellagic acid and ellagic acid released from

both ellagitannins and ellagic acid glycosides.

Ellagic Acid and Its Derivatives

Concentrations of ellagic acid and its derivatives in muscadine grapes were found

to significantly vary with ripening, in skin and pulp tissue, among cultivars, and

following juice extraction (Table 3-1). Ripening was a critical factor influencing

concentrations since appreciable increases in skin and juice during ripening were

observed. Since muscadine grapes grow in clusters rather than bunches, inconsistent

maturity at harvest is a common occurrence. Changes with ripening were also highly

variable among cultivars for free ellagic acid and its glycosidic forms, and ranged from a

0.3 to 13-fold increase in the skins alone. Differences during ripening were less variable

for total ellagic acid at a 1.7-fold average increase in the skins. The large increases in

ellagic acid and its glycosides observed during ripening may have resulted from various

reasons: amplified hydrolysable tannins synthesis during veraison (8); a chemoprotective

response similar to the formation of resveratrol (48); or accelerated hydrolysis of HHDP

units from ellagitannins that was observed to produce greater quantities of free ellagic

acid in each cultivar. Compared to total ellagic acid, relatively low levels of free ellagic

acid and ellagic acid glycosides were present in the grapes, an indication that












Table 3-1. The concentrations (mg/kg, mg/L) of free ellagic acid (EA), two ellagic acid glycosides (EAG 1 and 2) and total ellagic
acids on skin, pulp and juice of muscadine grapes as affected by cultivars and ripening stages (U: unripe and R: ripe).
Cutars Coor Free EA EAG11 EAG22 Total EA3
Cultivars Color
U R U R U R U R
Carlos White 32.1 b4 8.04 e* 17.4 b 6.76 d* 16.7 d 20.1 d 368 d 879 d*
Fry White 31.3 b 87.4 cd* 13.0 cd 90.3 a* 8.81 e 13.6 d 531cd 879 d*
Doreen White 10.8 d 138 ab* 3.78 f 93.0 a* 29.7 b 115 a* 918 ab 1620 b*
Noble Red 17.5 c 76.4 d* 10.5 de 23.2 c* 31.0 b 41.8 bc 474 d 592 e
Albemarle Red 12.7 cd 110 bc* 24.8 a 23.5 c 29.0 b 53.9 b* 1030 a 1090 c
Cowart Red 27.5 b 162 a* 13.8 c 95.9 a* 24.5 bc 46.1 bc* 732 bc 1900 a*
Nesbitt Red 15.5 cd 136 ab* 7.53 ce 61.7 b* 18.7 cd 39.4 c* 555 cd 1100 c*
Georgia Red Red 42.9 a 74.8 d* 12.8 cd 20.5 c* 38.7 a 10.1 d* 996 a 587 e
Carlos White 4.73 e 2.66 c 3.32 a 1.00 c 2.83 cd 2.90 c 159 b 231 b*
Fry White 6.44 de 1.01 d* 3.30 a ND5 d* 1.57 d ND e* 189 b ND c*
Doreen White 14.1 a 0.93 d* 1.22 cd trace d* 12.1 a 0.66 d* 474 a trace c*
Noble Red 3.51 ef 8.69 b* 0.88 d 2.98 b* 2.82 cd 5.79 b* 208 b 168 b
Pulp Albemarle Red 12.2 ab 24.5 a* 2.06 bc 6.04 a* 9.36 b 12.8 a 203 b 455 a
Cowart Red 8.28 cd 1.24 d* 2.12 b trace d* 5.08 c trace e 232 b ND c*
Nesbitt Red 10.1 bc 0.54 d* 3.53 a trace d* 8.63 b trace e 197 b ND c*
Georgia Red Red 1.13 f 1.00 d trace d ND d 1.13 d ND e 38.2 c ND c*
Carlos White 3.01 cde 4.34 e 1.02 cd 8.60 cd* 4.96 c 5.34 cd* 12.5 e 106 e*
Fry White 3.99 bcd 11.2 bcd* 2.63 a 21.7 a* 2.94 e 3.13 d 59.1 c 105 e*
Doreen White 3.34 cde 14.1b* 0.56 e 7.68 d* 6.02 b 15.7 b* 12.7 e 172 d*
Juice7 Noble Red 8.75 a 20.5 a* trace e 5.78 e* 6.70 ab 15.6 b* 10.1 e 257 b*
Albemarle Red 5.15 b 23.4 a* 0.77 de 9.68 bc* 6.81 a 20.1 a* 14.0 e 322 a*
Cowart Red 4.03 bc 12.5 bc* 1.31 bc 11.2 b* 4.07 d 6.81 c* 81.0 b 219 c*
Nesbitt Red 2.17 e 8.82 d* 1.16 cd 5.19 e* 3.20 e 4.85 cd* 26.1 d 187 cd*
Georgia Red Red 2.66 de 9.77 cd 1.60 b ND f* 3.47 de 3.16 d 88.0 a 198 cd*
1,2Expressed in ellagic acid equivalents. 3The sum of free ellagic acid and ellagic acid released following acid hydrolysis. 4Similar letters
within columns for each fruit part are not significantly different (LSD test. P<0.05). 5Concentrations below detection limit. 6Concentration
below 0.5 ppm. 7Hot-pressed juice. Asterisk (*) indicates significant effects by fruit ripening for each fruit parts (LSD test. P<0.05).









ellagitannins were the major source of ellagic acid following hydrolysis. However, the

actual concentrations of the ellagic acid glycosides were likely influenced by the use of

free ellagic acid as the quantifying standard.

As with most grape varieties, polyphenolic compounds are typically concentrated

in epidermal tissues, which are exceptionally thick in muscadine grapes and often hinder

efficient juice extraction. On average, the skin and pulp tissue constituted 21 and 69% of

the total mass of the grapes respectively, and were similar for both unripe and ripe fruit.

Ellagic acid and its derivatives were generally concentrated in the skin, which contained

51-67% of these compounds in unripe fruit on a fresh weight basis. Upon ripening, these

compounds were even more localized in the skin and accounted for 82-87% of the total.

Doreen and Cowart contained the highest concentrations of ellagic acid and its glycosides

among the cultivars, but no meaningful correlation could be made between free ellagic

acid and/or ellagic acid glycosides and concentrations of total ellagic acid, an observation

that likely reflected the influence of ellagitannins in each isolate. Compared to ellagic

acid concentrations present in the skin and pulp, levels present in juice were considerably

lower and reflected the low solubility of ellagic acid in aqueous systems. A hot break or

"hot-press" technique is commonly used with muscadine grapes to increase juice yields

or add pigmentation to wines or juices, and when combined with macerating enzymes

(49), juice extractions are better facilitated. Additionally, the time and temperature of the

heating process will appreciably influence juice yields and phytochemical concentration

compared to non-heated fruit juice (17) and white or bronze grapes, depending on

cultivar, may not be heated to prevent enzymatic and autoxidative browning reactions

affecting juice quality (50). Textural differences also occur in the grapes during ripening









from action of natural pectinase and may also influence phytochemical solubilization.

Typical juice yields may range from 60-75% by weight for hot-pressed muscadine grape

juices and is influenced by heating conditions, pressing conditions, the use of pressing

aids such as rice hulls, and skin thickness (33). The highest concentrations of total ellagic

acid were found in the juice of ripe Albemarle (322 mg/L) and Noble (257 mg/L), which

reflected a 24% average increase in concentration over juice pressed from unripe grapes.

Concentrations of total ellagic acid present in the juice were not necessarily a reflection

of levels found in whole grapes, since the juice of unripe fruit contained 2-26% of the

amount present in whole grapes compared to 19-78% for ripe fruit. For simplicity, these

data were determined based on a 60% juice yield and accounted for the variable

contributions from skin and pulp tissue (seeds not included) to the total weight of the

grapes. Juice from ripe grapes of Noble, Cowart, Nesbitt, and Georgia Red had the

highest total ellagic acid extractions (>58%), while Carlos, Fry, Doreen, and Albemarle

were considerably lower (<34%). The low recovery of ellagic acid derivatives in the

latter cultivars reflected the difficulty in solubilizing polyphenolics, likely due to physical

barriers associated with their thick skins, which left high concentrations of these

compounds behind in the skin and pulp material. Free ellagic acid itself, sparingly soluble

in water, was also poorly solubilized in all juices, retaining only 27 and 37% on average

of the total present in whole grapes for unripe and ripe fruit, respectively. However, the

ellagic acid glycosides were considerably more soluble in juices with >56% recovery

from whole grapes.

Anthocyanins

Anthocyanidins, quantified only in the red cultivars, were expressed in cyanidin

equivalents (Table 3-2) since the predominant anthocyanins in muscadine grapes were













Table 3-2. Concentration (mg/kg, mg/L in cyanidin equivalents) of six anthocyanidins and total anthocyanidins on skin, pulp and juice
of red muscadine grapes as affected by ripening stages (U: unripe and R: ripe).


Delphinidin1


Cyanidin


Petunidin


Malvidin +
Peonidin


Total2


Skin







Pulp


Noble
Albemarle
Cowart
Nesbitt
Georgia
Red
Noble
Albemarle
Cowart
Nesbitt


Georgia
Red
Noble
Albemarle

Juice6 Cowart
Nesbitt
Georgia
Red


U
ND4 b5
44.2 a


1450 b
424 c


57.6 a 1290 b
66.8 a 3550 a

72.1 a 300 c


ND b
0.84 b
4.54 a


102 a
67.1 b
3.75 c


ND b 19.3 c

1.17b 2.52c


ND b
ND b
ND b
6.98 a


131 a
52.4 c
48.6 c
72.5 b


NDb 10.1 d


U
ND b
28.5 a


R
692 c
291 d


37.5 a 1210 a
35.1 a 860 b

35.4 a 52.5 e


6.95 a
5.85 a
4.00 b


93.9 a
89.0 a
10.6 b


ND c 12.4 b

0.980 c 0.76 b


ND b
ND b
ND b


125 a
86.1 b
94.0 b


2.83 a 49.3 c

NDb 7.16 d


U
159 a
ND b


R
1070 a
ND b


12.0 b 294 b
ND b ND b

ND b 20.3 b


ND b
ND b
0.90 a


78.4 a
29.7 b
0.63 c


ND b 5.03 c

ND b 0.25 c


ND b
ND b
ND b


155 a
25.7 c
21.4 c


3.04 a 44.3 b

ND b 2.61 a


U
ND a
ND a


R
926 a
102 d


ND a 445 c
ND a 825 b

ND a 17.9 e


ND a
ND a
ND a


114 a
21.8 b
ND b


ND a ND b

ND a ND b


ND a
ND a
ND a
ND a


200 a
18.2 b
16.5 b
23.6 b


ND a ND c


U
159 a
72.6 b


R
4140 b
817 d


107 ab 3250 c
102 ab 5230 a

108 ab 390 d


6.95 a
6.70 a
9.44 a


383 a
212b
15.0 c


ND b 36.8 c

2.15 b 3.53 c


ND b
ND b
ND b
12.8 a


610 a
182 b
180 b
190 b


ND b 19.9 c


1Cyanidin equivalents. 2Sum of individual anthocyanidins. 3All anthocyanins are significantly different at ripening stage.
4. Concentrations below detection limit. 5Similar letters within columns for each fruit part are not significantly different (LSD
test.P<0.05). 6Hot-pressed juice. Asterisk (*) indicates significant effects by fruit ripening for each fruit part (LSD test. P<0.05)









previously identified as non-acylated 3,5-diglucosides of six anthocyanidin bases (17). In

the current study, only three anthocyanidins were positively elucidated following acid

hydrolysis using the column and solvent conditions described, due to incomplete

separation of peonidin and malvidin and the absence of pelargonidin. As expected,

anthocyanins appreciably increased in the skin as the fruit ripened with low

concentrations also found in pulp material nearest the skin. Anthocyanidin abundance in

ripe fruit were delphinidin > petunidin > malvidin+peonidin > cyanidin with Nesbitt,

Noble, and Cowart containing the highest overall concentrations. Color instability of

muscadine wine and juice is an established quality defect, and is a consequence of their

high concentrations of monomeric 3,5-diglucosides with o-diphenolic substituents that

include delphinidin, cyanidin, and petunidin (51). Among the cultivars evaluated, these

three anthocyanidins accounted for 78-96% of the total in fresh grapes and from 67-100%

in juice. Ripe Noble grapes, one of the most popular wine and juice cultivars, contained

the highest concentration of malvidin+peonidin among the cultivars evaluated. Malvidin

is generally considered the most stable anthocyanin form and along with peonidin was

present at 22% of the total in the skins compared to 33% in juice. However even with

high malvidin+peonidin concentrations, the juice from Noble grapes is considered highly

susceptible to color degradation due to likely lack of inter- and intra-molecular

copigmentation of 3,5 diglucosidic anthocyanins (51) and provides an indication that the

remaining cultivars would be even less stable to oxidation or other deteriorative reactions

affecting juice or wine pigmentation due to their lower malvidin+peonidin

concentrations. These cultivars, as well as the white/bronze varieties, may be more

suitable for juice blending to take advantage of their high ellagic acid contents. Noble









grape juice also contained the highest total anthocyanin concentration (610 mg/L), while

Georgia Red contained considerably less (20 mg/L) even in relation to the other red

varieties that ranged from 180-190 mg/L. Based on a 60% juice yield, only 12% of the

total anthocyanins present in grape skins were solubilized into the juice of Nesbitt and

Georgia Red, both consumed primarily as table grapes, the former having high

anthocyanin content yet poor anthocyanin solubility characteristics during juicing. Juice

from the remaining cultivars, commonly consumed either fresh or processed, contained

27-32% of the total anthocyanins present in the each grape. The low anthocyanins

recovery values in juice, especially in relation to ellagic derivatives, reflect the degree of

processing necessary to solubilize sufficient anthocyanins to produce a suitable red wine

or juice.

Total Phenolics and Antioxidant Capacity

Measurements of total phenolics by the Folin-Ciocalteu metal reduction assay and

peroxyl radical scavenging activity using the ORAC assay are common index that

provide an overall assessment of the content and chemical activity of compounds present

in fruits and vegetables. These attributes were quantified in methanolic and ethyl acetate

extracts of grape skin, pulp, and juice and following partitioning of phenolic acids and

flavonols into ethyl acetate, into which anthocyanins are not soluble, to differentiate

between major polyphenolic classes (Tables 3-3 and 4). Values for total phenolics, which

varied among cultivars and with fruit ripening, were good predictors of antioxidant

capacity in both methanolic and ethyl acetate extracts (r= 0.83 and 0.92, respectively).

The higher correlation coefficient for ethyl acetate extracts may have reflected the

removal of potentially interfering/prooxidant polar compounds or reflected interactions

between anthocyanins and other polyphenolics in the methanolic extracts (52, 53). Based










Table 3-3. Concentrations (mg/kg, mg/L) of total soluble phenolics (Folin-Ciocalteu
metal reduction assay) in methanolic and ethyl acetate extracts as affected by
cultivars and ripening stages (U: unripe and R: ripe).


Cultivars


Skin


Carlos

Fry

Doreen

Noble

Albemarle

Cowart

Nesbitt

Georgia Red


Carlos

Fry

Doreen

Noble
Pulp
Albemarle

Cowart

Nesbitt

Georgia Red

Carlos

Fry

Doreen

Noble
Juice2
Albemarle

Cowart

Nesbitt

Georgia Red


Color


White

White

White

Red

Red

Red

Red

Red

White

White

White

Red

Red

Red

Red

Red

White

White

White

Red

Red

Red

Red


Methanolic Extract


2430 b1

1440 c

3860 a

2660 b

2580 b

2660 b

2480 b

4220 a

405 de

566 cd

1210b

601 c

1410 a

lll0b

567 c

312 e

1145 c

1069 c

1673 a

1630 a

1460 ab

1200 bc

739 d


R

2530 e

3360 d*

3990 c*

3090 d

2260 e

4370 c*

5030 b*

9470 a*

738 b

276 d*

192 d*

848 b*

1100 a*

200 d*

443 c

467 c*

979 de*

1500 cd*

1293 d

1950 b

1770 bc

1360 cd

1210 d*


1140 c 2860 a*


Ethyl Acetate Extract
U R

428 d 706 f*

459 d 987 e*

1430 a 2280 b*

1020 bc 727 f*

1320 ab 756 ef

1130 ab 1890 c

627 cd 1300 d*

1500 a 2910 a*


128 d

138 d

1300 a

332 cd

622 b

528 bc

502 bc

99.4 d

165 a

90.1 d

161 a

139 abc

147 ab

122 bc

61.2 e

118 c


258 a

102 cd

39.2 de*

120 bc*

274 a*

38.2 de*

16.3 e*

183 b*

66.4 b*

81.0 c

141 a

69.1 c*

120 ab

89.4 bc

70.8 c

139 a


'Similar letters within columns for each fruit part are not significantly different
(LSD test. P<0.05). 2Hot-pressed juices. Asterisk (*) indicates significant effects
by fruit ripening for each fruit part (LSD test. P<0.05).










Table3-4. Antioxidant capacity ([imol trolox equivalents/g or mL) of methanolic and
ethyl acetate extracts as affected by cultivars and ripening stages (U: unripe
and R: ripe).


Cutivars


Skin


Carlos

Fry

Doreen

Noble

Albermerle

Cowart

Nesbitt

Georgia Red


Carlos

Fry

Doreen

Noble
Pulp
Albermerle

Cowart

Nesbitt

Georgia Red

Carlos

Fry

Doreen

Noble
Juice2
Albermerle

Cowart

Nesbitt

Georgia Red


Color


White

White

White

Red

Red

Red

Red

Red

White

White

White

Red

Red

Red

Red

Red

White

White

White

Red

Red

Red

Red

Red


Methanolic Extract


58.0 d1

49.3 d

104 a

97.2 ab

90.8 b

97.7 ab

69.7 c

89.0 b

5.95 de

6.35 de

34.0 a

9.40 cd

14.8 b

11.8 bc

13.9b

4.95 e

20.3 b

14.6 d

25.3 a

24.3 a

19.3 bc

17.4 cd

10.8 e

20.7 b


R

86.2 cd*

72.3 d

90.4 c*

100 c

71.1 d

119b

136 a*

128 ab*

8.75 b*

7.50 bc

2.45 d*

14.3 a

13.1 a

2.60 d*

4.60 cd

6.80 bc

15.5 d*

20.1 bc

19.6 bc

26.7 a

23.3 ab

21.4 bc*

18.3 cd*

26.6 a*


Ethyl Acetate Extract

U R

10.2 b 17.5 b

10.7 b 19.0 b*

25.2 a 25.5 a

22.5 a 12.3 c*

22.6 a 12.0 c*

22.6 a 26.3 a*

12.1 b 25.1 a*

25.4 a 29.1 a*


3.90 e

3.90 e

15.0 a

4.70 e

8.05 c

5.70 d

9.50 b

4.10 e

2.37 b

1.61 cd

2.91 a

2.55 ab

2.89 a

1.99 c

1.27 d

2.50 b


5.05 b

3.70 d

1.60 e*

4.10 cd

6.45 a*

1.55 ef*

1.00 f*

4.45 c

2.06 bc

1.92 bc

2.38 ab

1.51 cd

2.14b

1.91 bc

1.20 d

2.96 a*


1Similar letters within columns for each fruit part are not significantly different
(LSD test. P<0.05). 2Hot-pressed juices. Asterisk (*) indicates significant effects
by fruit ripening for each fruit part (LSD test. P<0.05).









on abundance, anthocyanins were the major antioxidant compounds present in muscadine

grape skin and juice and their concentration was directly related to antioxidant capacity

(r- 0.99). Ethyl acetate soluble compounds also contributed to antioxidant capacity and

ranged from 12-29%, 22-83%, and 5.7-15% of the total present in methanolic extracts of

skin, pulp, and juice, respectively. Other than ellagic acid and its derivatives, many

additional compounds were also identified in the ethyl acetate extract including several

flavonoids glycosides, phenolic acids, and procyanidins that are all known to possess

antioxidant activity (54, 55). In various concentrations, gallic acid, protocatechuic acid,

catechin, and epicatechin were identified in ethyl acetate extracts. Flavonoid glycosides

were tentatively identified based on their spectroscopic similarities to myricetin,

quercetin, and kaempferol with glucose and/or rhamnose moieties. A myricetin glycoside

was the predominant flavonoid present in all cultivars and ranged from 8.7-1,350 mg/kg

in skin, 0-50 mg/kg in pulp, and 1.6-50 mg/L in juice. Among the cultivars, ripe Georgia

Red contained the highest concentrations of total phenolics in both ethyl acetate and

methanolic extracts of both skin and juice, in contrast to its low anthocyanin, ellagic acid,

and ellagic acid glycoside content, which was primarily attributed to its high flavonoid

concentration.

Conclusion

This study demonstrated that ripening, physiology, and juice processing influence

phytochemical composition and antioxidant capacity of muscadine grapes. Data suggest a

diversity of phytochemical compounds that can be used for novel blending schemes for

muscadine grape juice or wine to obtain a desired quality and polyphenolic content

relating to their antioxidant capacity.














CHAPTER 4
IDENTIFICATION OF ELLAGITANNINS AND CONJUGATES OF ELLAGIC ACID
IN MUSCADINE GRAPES

Introduction

V. rotundifolia are differentiated from V vinifera in several points such as lack of

an organized fruit bunch, strong disease resistance, and its unique phytochemical

composition primarily due to presence of ellagic acid derivatives. Ellagic acid derivatives

are not uncommon in plants and are in abundance in raspberry (14, 15, 46, 47),

pomegranate (24, 66, 67), oak (56), birch leaves (68), or medicinal herbs/plants (69, 70);

yet their presence in muscadine grapes is unique among Vitis species. Ellagic acid

derivatives are a broad classification that includes the free acid state of ellagic acid, these

conjugated with various sugars to form simple glycosides or more complex ellagitannins

(46). Ellagitannins are characterized as hydrolysable conjugates containing one or more

hexahydroxydiphenoyl (HHDP) group esterified to a sugar, usually glucose. In

raspberries, the predominant ellagitannins were identified as lambertianin C and sanguiin

H-6 as well as arabinosides, acetylarbinosides, and acetylxylosides of ellagic acid (14)

yet similar compounds, if present, have not been determined in the muscadine grape. The

objective of this study was to elucidate identities and concentrations of ellagic acid

derivatives in muscadine grapes by various extraction and analytical procedures.

Investigations on ellagic acid and ellagic acid derivatives in muscadine grapes will add

value and marketability to the crop due to beneficial health benefits such as its

antioxidant activity (19, 20), anti-carcinogenic properties influencing cell cycle arrest and









apoptosis (20), and the inhibition of tumor formation and growth in mammalian models

(21, 40) previously attributed to these compounds.

Materials and Methods

Muscadine grapes (cvs. Doreen, Noble, Albemarle) were donated from local grape

growers in central Florida and polyphenolics extracted from the skin and pulp with 100%

methanol (0.01% HC1). Extracts were filtered through Whatman #4 filter paper, solvent

removed at 400C under reduced pressure, and polyphenolic residue redisolved in water at

pH 3.5.

Isolations

Due to the diversity of ellagic acid precursors potentially present, it was necessary

to fractionate the polyphenolics based on their affinity to C18 and Sephadex LH-20 and

partition based on their solubility in various organic solvents (Figure 4-1). Initially, grape

extracts were applied to a Sep-Pak C18 cartridge and polyphenolics eluted with various




Grape Extracts


Sep-Pak C18



Unbound Fr. Ethyl acetate Fr. Methanol Fr.

Sephadex LH-20 Sephadex LH-20

Isolate I Isolate II Isolate III Isolate IV
By 100% MeOH :Ellagitannins By 10% MeOH By 100% MeOH
:Ellagitannins :Ellagic acid glycosides : Anthocyanins : Ellagitannins
:Ellagic acid
:Flavonoids

Figure 4-1. Fraction scheme and tentative classification of polyphenolics present in
methanolic extracts of muscadine grapes.









solvents in order of water, ethyl acetate, and then methanol (0.01% HC1) to partially

isolate compounds of interest. Ethyl acetate was removed under reduced pressure and

redissolved in water and methanol (95:5, pH 3.5) and this fraction contained phenolic

acids, flavonoids, and ellagic acid derivatives including free ellagic (aglycones), ellagic

acid glycosides, and ellagitannins. Ellagic acid derivatives that remained in the unbound

(water) fraction and in the final methanol-soluble fraction, following evaporation and

solubilization in water, were further concentrated by partitioning from a mini-cartridge of

Sephadex LH-20 based on their selective adsorption properties.

Analysis by HPLC-PDA

HPLC-PDA analysis was initially employed to tentatively identify and quantify

ellagic acid derivatives in grape extracts and isolated polyphenolic fractions from three

muscadine grape cultivars (Doreen, Noble, and Albemarle). Free ellagic acid, ellagic acid

glycosides, and total ellagic acid (following acid hydrolysis in 2N HCI for 60 min at

950C) were evaluated in equivalents of free ellagic acid. Separations were conducted on a

Dionex HPLC system using a PDA-100 photodiode array detector and a 250 mm x 4.6

mm Acclaim 120 C18 column (Dionex, Sunnyvale, CA) with a Cis guard column.

Identical mobile phases in Chapter 3 were employed to separate polyphenolics with

modified gradient elution program where phase B (60% methanol) changed from 0-60%

in 30 min; 60-80% in 10 min; 80-100% in 10 min; and 100% in 10 min for a total run

time of 60 min, after which the column was equilibrated to original conditions in 2 min

for the next sample injection.

A second HPLC-PDA separation was applied to Isolates I and II in order to obtain

UV absorbance spectral data under identical condition with MSn analysis. Separations

were performed using a Waters Alliance 2695 HPLC system and polyphenolics separated









on Phenomenex (Torrace, CA) Synergi 4u Hydro-RP 80A (2 x 150 mm; 4 um;

S/N= 106273-5) plus C18 guard column (2mm x 4 mm) and a Waters 996 photodiode

array detector recorded UV spectra. Mobile phase and gradient program were identical to

MSn analysis stated below.

Analysis by HPLC-MSn

Mass spectrometric analyses were carried out to achieve structural information

based on molecular masses and fragment ions. Only Isolates I and II were evaluated for

polyphenolics on an Agilent HPLC system (Palo Alto, CA) using an 1100 series binary

pump and separated using the Phenomenex column previously described. Mobile phases

consisted of (A) 0.5% formic acid in water (5 mM ammonium format ) and (B) 0.5%

formic acid in methanol and run at 0.15 mL/min. Polyphenolic compounds were

separated using a gradient elution program where phase B changed from 5-30% in 5 min;

30-65% in 70 min; 65-95% B in 30 min and held for 20 min; and 95-5% B in 10 min to

equilibrate the column and whole system to original conditions for 30 min. Effluents

from the column were passed through the UV detector (Applied Biosystems Model

785A) and then analyzed by mass spectrometer, ThermoFinnigan (San Jose, CA) LCQ

with electrospray ionization (ESI). In order to confirm molecular masses, ionization was

conducted in both negative and positive mode.

Total Polyphenols and Antioxidant Capacity

Total soluble phenolics were analyzed using Folin-Ciocalteu assay (40) and

antioxidant activity was determined using the oxygen radical absorbance capacity

(ORAC) assay as described in Chapter 3.









Statistical Analysis

Data represent the mean of duplicate analyses with analysis of variance conducted

using JMP5 software (44); mean separation was conducted using the LSD test (P<0.05).

Results and Discussion

Isolations and Quantification of Polyphenolic Compounds by Solid Phase Extraction

Numerous polyphenolic compounds were present in muscadine grapes when

analyzed by HPLC, and their separation and identification were enhanced by preparing

fractions using a series of solid phase and liquid-liquid extractions based on polarity and

affinity characteristics of each compound (Figure 4-1). Overall, Isolate II represented the

majority of non-anthocyanin polyphenolics present in muscadine grapes due to the high

affinity of these compounds for ethyl acetate. This fraction was exceptionally high in

both phenolic acids and flavonoids. By contrast, Isolate I contained only the most polar

compounds, not retained on Sep Pak C18 cartridges. This isolate had a strong affinity to

Sephadex LH-20, a cross-linked dextran, which is widely used to isolate tannins from

plant based sample (55). Compounds that remained on the Sep Pak C18 cartridges

following elution with ethyl acetate were subsequently eluted with acidified methanol

and were found to predominantly contain anthocyanins. These compounds also yielded

high concentrations of free ellagic acid after acid hydrolysis indicating the presence of

ellagic acid derivatives. Sephadex LH-20 was employed to separate anthocyanins from

these derivatives, with anthocyanins removed first with up to 10% methanol followed by

ellagic acid derivatives with 100% methanol to give Isolates III and IV, respectively. As a

result, the initial grape extract and four sub-fractions from three cultivars (Doreen, Noble,

and Albemarle) were created for subsequent phytochemical analyses and quantification

of total soluble phenolics and antioxidant capacity (Table 4-1 and Figure 4-2).










In each grape cultivar extract, free ellagic acid, three types of ellagic acid

glycosides (EAG 1-3) and total ellagic acid were evaluated. Previous studies including

Chapter 3 (16) have reported only two ellagic acid glycosides, however changes in the

HPLC gradient program resulted in the separation of a third ellagic acid glycoside that

eluted earlier than the previous two compounds. Extracts from Noble had the highest

concentrations of free ellagic acid (49.7 mg/kg) and total ellagic acid glycosides (86.9

mg/kg), yet following acid hydrolysis, total ellagic acid was higher for Albemarle (912

mg/kg) compared to Noble (686 mg/kg). This difference was likely due to the presence of


Table 4-1. The concentrations (mg/kg) of free ellagic acid, ellagic acid glycosides (EAG
1, 2 and 3) and total ellagic acids on each fraction from three different
cultivars (Doreen, Noble, Albemarle).


Cultivars Isolates'


Free
Ellagic acid
13.5 a2
0.25 b
12.9 a
0.15 b
0.80 b
49.7 a
0.35 d
11.5 b
1.90 d
5.55 c
32.9 a
0.50 c
27.2 b
1.15 c
2.45 c


EAG3

1.60a
0.40b
N.A.
N.D.3
0.55 b
6.05 a
N.D.
N.D.
2.75 b
2.20 c
7.80 a
N.D.
N.D.
N.D.
4.55 b


EAG1

19.5 a
N.A.
9.15 b
N.D.
1.95 c
31.4 a
N.D.
N.D.
9.55 c
13.5 b
20.0 a
N.D.
4.60 b
0.20 c
6.95 b


EG ~Total
EAG2 .
Ellagic acid
22.5 a 360 a
N.A. 13.1 c
9.70 b 58.9 b
N.D. N.D.
1.80 b 2.95 d
49.4 a 686 a
N.D. 16.8 e
1.40 d 102 b
14.0 c 32.7 d
24.0 b 63.2 c
37.6 a 912 a
N.D. 53.7 c
12.1 b 130 b
0.40 c 0.65 c


12.8 b


33.0 c


1 Isolates were prepared by using Sep-Pak C18 and Sephadex LH-20 depending
on different behaviors for various solvents. 2. Similar letters within columns for
each cultivar are not significantly different (LSD test. P<0.05). 3. Not Detected.


Doreen







Noble







Albemarle


Whole
I
II
III
IV
Whole
I
II
III
IV
Whole
I
II
III
IV









ellagitannins. The fractionation scheme successfully separated ellagitannins into Isolate I

since high levels of total ellagic acid was observed in all three cultivars compared to low

amounts (< 1 mg/kg) of either free ellagic acid or ellagic acid glycoside. For example,

extracts of Albemarle resulted in >100-fold increase in free ellagic acid after acid

hydrolysis. Isolate I of Albemarle was applied to further MSn analysis to identify the

ellagitannins responsible for release of HHDP groups and subsequent conversion to free

ellagic acid. Isolate II contained the majority of the free ellagic acid in Doreen (95%) and

Albemarle (83%), but only 23% for Noble, which also contained the highest

concentration of anthocyanins followed by Albemarle. Although not fully elucidated,

these data seem to indicate that anthocyanins may interfere with or inhibit desorption of

free ellagic acid with ethyl acetate from C18 Sep Pak cartridges. Isolate II was also

suspected to contain ellagitannins based on total ellagic acid content following acid

hydrolysis; therefore this fraction from Albemarle was subjected to further analysis by

MSn to investigate chemical structures specific to ellagic acid glycosides, ellagitannins,

as well as flavonoid glycosides. Isolate III contained predominantly anthocyanins, but

low concentrations of free ellagic acid and ellagic acid glycosides were also found.

Subsequent recovery of remaining ellagic acid derivatives in Isolate IV was

accomplished, representing those that were not solubilized by ethyl acetate earlier in the

fractionation process. However, Isolate IV did not show an appreciable increase in total

ellagic acid 0.58, 1.4 and 1.2% for Doreen, Noble, and Albemarle respectively. This

observation demonstrated that ellagitannins were not retained on C18 following elution

with ethyl acetate and that the ellagic acid glycosides likely had a similar detector

response to free ellagic acid since the sum of free ellagic acid and the three ellagic acid









glycosides were similar before and after acid hydrolysis. This is an important

consideration since authentic standards for ellagic acid glycosides are not commercially

available.

Folin-Ciocalteu metal reduction assay and oxygen radical absorbance capacity

(ORAC) assay were applied to the initial grape extracts and four sub-fractions isolates

with the purpose to quantify the total reducing capacity of the samples and to specifically

determine the peroxyl radical scavenging properties, respectively (Figure 4-2). Generally,

total soluble phenolic concentrations are very well correlated with antioxidant capacity

and each fraction showed strong positive relation, at least r=0.93, between the two

attributes. As previously noted, Noble muscadine grapes are a commonly utilized cultivar

for juice or wine making due to their high anthocyanin content and resulted in both high

total soluble phenolic and antioxidant capacity in whole extracts and in Isolate III. The

compounds in Isolate I, mainly ellagitannins, also contributed to antioxidant capacity by

17%, 1.5% and 11% of the total present in whole extracts of Doreen, Noble and

Albemarle, respectively. Most antioxidant compounds present in Doreen and Albemarle

were extracted with ethyl acetate into Isolate II due to relative abundance of non-

anthocyanins compounds and resulted in 41% and 48% antioxidant capacity compare to

whole extracts. Total soluble phenolics (18%) and antioxidant capacity (6.9%) in Isolate

IV of Noble is likely due to residual ellagic acid glycosides and anthocyanins. Due to

potential health benefits related with antioxidant activity, data re-emphasize that

muscadine grapes are an excellent alternative crop to be utilized as value-added

application.











2500
Total Soluble Phenolics
a
2000 Doreen
"m Noble
m Albemarle
1500 b


1000 C
a

500 a a aa
b: a
c c b bib


25 Antioxidant Capacity


20 -
-2
0
S15 -
E




5- b aa

a b a b b b b b

Whole I II III IV
Isolate


Figure 4-2. Total soluble phenolics and antioxidant capacities of five fractions (Whole,
Isolates I, II, III and IV) from three different cultivars (Doreen, Noble, and
Albemarle). Same letters within fractions for each attribute are not
significantly different (LSD test, P<0.05).

Identifications of Ellagic Acid Derivatives and Flavonoids by HPLC-PDA and MS"

To identify compounds associated with the evolution of free ellagic acid and

differentiate these compounds from flavonoids in muscadine grapes, polyphenolic

compounds in Isolates I and II from Albemarle were examined for their UV

spectroscopic properties and molecular mass/charge ratio by HPLC-PDA and MSn,









respectively. Only one cultivar was used for these analyses since the phytochemistry

among cultivars are similar, and a goal of the analyses was to initially elucidate the

identity and concentration of compounds present. HPLC coupled to PDA detector is an

effective tool for tentative identification of the polyphenolic classification and can be

used in combination with an authentic standard to identify unknown compounds in a

plant-based system; however current trends couple this method with HPLC-MSn to

additionally confirm the identity of unknown compounds or compounds where an

authentic standard is not available. By combining these methods, a compound can be

characterized with greater certainty of its identification.

Ellagic acid derivatives in Isolate I

Isolate I was prepared by sequential solid phase extractions with Sep-Pak C 18 and

Sephadex LH-20 resulted in a fraction that contained highly polar polyphenolic

compounds such as gallic acid, epigallocatechin, and hydrolyzable tannins. As discussed

previously, the presence of ellagic acid derivatives was indicated by the release of free

ellagic acid after acid hydrolysis. The sample isolate was employed to obtain the HPLC-

PDA chromatogram and UV spectroscopic information (Figure 4-3), and also analyzed


4
0 .02 2





10.00 20.00 30.00 40.0 0
Minutes

Figure 4-3. HPLC-PDA chromatogram (280 nm) of Isolate I of muscadine grapes.














Table 4-2. UVmax and HPLC-ESI(-)-MSn analyses of polyphenols in isolate I from muscadine grapes.
Peak Rt (min) Compound Amax MW MS1(m/z) MS2 (m/z) MS3 (m/z)
No.


Ellagitannin 1

HHDP-Galloylglucose

Gallic acid

HHDP-Galloylglucose

Ellagitannin 2

Epigallocatechin

Digalloyl glucose

Ellagitannin 3

Digalloyl glucose


228,
262sh

268

272

265

268

291

295

276

272


802

634

170

634

834

306

484

832

484


801

633

169

633

8355a

305

483

833a

483


757, 481 301

301:[M-H]- 332(Gal-Glc) 301

170, 125

301: [M-H]- 332(Gal-Glc) 301

798, 696, 303

261,221,219, 179

331:[M-H]- 152(Gal) 271.

481,303

331:[M-H]- 152(Gal) 271.


1

2

3

4

5

6

7

8

9

* [M+H]


5.7

9.2

12.1

13.6

15.0

17.5

18.8

21.4

30.7


, 193, 169


, 193, 169









by HPLC- ESI(+/-)-MSn to determine molecular masses for compound elucidation (Table

4-2).

Peak 1 (tR =5.7 min; Xmax =228, 262sh nm) was tentatively identified as ellagitannin

1 according to UV spectrum (Figure 4-4, A), which agreed with previous studies reported

by Zafrilla et al. (45). This compound was identified as having at least a single HHDP

unit esterified to glucose. The average molecular weight (m/z) of peak 1 was 802 by both

positive ([M+H]+ = m/z 803) and negative ([M-H]- = m/z 801) ion modes. In negative ion

mode, the base peak produced fragment ions at m/z 481 (M-321), which corresponded to

one glucose (179) and one HHDP (302) unit following MS2. Further fragmentation (MS3)

produced its major ion at m/z 301 which corresponds to the ellagic acid precursor HHDP

and confirmed by free ellagic acid formed following acid hydrolysis.

Peaks 2 (tR=9.2 min; Xmax =268 nm) and 4 (tR=13.6 min; Xmax =265 nm) had the

greatest detector response for the compounds separated and contained similar UV

spectroscopic properties (Figure 4-4, B) and (-)ESI-MS mass spectrums. The most

abundant ions in their mass spectrum was m/z 633 as [M-H]-, which likely corresponded

to HHDP-galloylglucose (MW 634) found in birch leaves (26). The MS2 spectrum also

had fragments at m/z 301 that resulted from the loss of a galloylglucose unit (332) from

its parent ion. HHDP-galloylglucose contains a single HHDP and single galloyl group

conjugated to a glucose unit and is often referred to as strictinin, sanguiin H4 or sanguiin

H5 depending on the location of the HHDP and galloyl group (56).

Peak 3 (tR=12.1 min; Xmax =272 nm) was identified as free gallic acid on the basis

of its retention time and UV absorbance spectrum in relation to an authentic standard.










Identification was confirmed by MS-MS, which yielded [M-H]- at m z 169 and a

predominant fragment at m z 125.

Peak 5 (tR=15 min; Xmax =268 nm), even at very low UV absorbance, was

tentatively identified as an HHDP glucoside (ellagitannin 2) due to the presence of m/z

303 ions after serial MS analysis in positive mode. The parent compound MW 834

produced an m/z 835 [M+H] ion, which underwent MS/MS to produce m z 303 as its

most abundant ion.


A

0.04- 0.004- 5 .
0.002



0.00- 0.000- 0.000
300 400 300 400 300 400

B
2 0.020 4



0.000

0.000 \
300 400 300 400
C


0.002 0.001




300 400 300 400


Figure 4-4. UV spectra of ellagic acid derivatives in Isolate I. A, ellagitannins (peaks 1, 5,
8); B, HHDP-galloylglucose (peaks 2, 4); C, digalloyl glucose (peaks 7, 9).









Peak 6 (tR=17.5 min; Xmax =290 nm) was identified as epigallocatechin on the basis

of retention time and UV absorbance spectrum in relation to an authentic standard. This

was confirmed by the MS-MS, which yielded [M-H]- at m/z 305 and predominant MS2

ion at m/z 261, 221, 219, and 179.

Peaks 7 (tR=18.8 min; Xmax = 295 nm) and 9 (tR =30.7 min; Xmax = 272 nm) were

tentatively elucidated as digalloyl glucose (26) because [M-H]- at m/z 483 yielded m/z

331 by losing a galloyl group (MW 152) from glucose followed by a second galloyl

group with subsequent ionization.

Peak 8 (tR =21.4 min; Xmax =276 nm) was thought to contain co-eluting peaks due

to the resultant molecular weight indicated from MSn analysis. One of the compounds

was determined as MW 832 due to ions at m/z 833 as [M+H] and underwent CID-

MS/MS to produce m/z 481 and 303, indicating presence of an HHDP unit. Therefore,

this compound, ellagitannin 3 had potential to convert into ellagic acid with hydrolysis.

Additional compounds (MW 480 and 818) were also detected, but no evidence that either

was an ellagic acid precursor.

Ellagic acid derivatives and flavonoid glycosides in isolate II

Isolate II was prepared by ethyl acetate elution through Sep-Pak C18 cartridges and

resulted in a fraction that was free of anthocyanins yet rich in ellagic acid derivatives and

flavonoid glycosides with an aglycone base of myricetin, quercetin, kaempferol, which

have been reported as predominant flavonoids in muscadine grape products (17).

Significant increase in ellagic acid by acid hydrolysis indicated the likely presence of

ellagic acid glycosides or ellagitannins in this isolate. In order to elucidate the presence of

ellagitannins, two wavelengths (280 and 360 nm) were monitored for HPLC-PDA










application (Figure 4-5). MSn analysis (Table 4-3) was applied in both (-) and (+) ESI

modes to determine molecular weights and compound identification of each peak

detected.

Peak 1 (tR =58-60 min; Xmaxs =280sh)was detected only at 280 nm similar to the

three ellagitannins observed in Isolate I (Figure 4-4). MSn analysis, however, yielded

single mass spectrum at m/z 800 and produced major fragment ions at m/z 447 and 303,

which matched fragmentation patterns of HHDP-galloylglucose (Peaks 2 and 6) in Table

4-2. Although MS analysis did not clearly explain its molecular identity, this peak was

tentatively categorized as an ellagic acid derivative on the basis of its UV absorbance

spectrum and presence of an ion at m/z 303.

Peak 2 (tR =86 min; Xmax =352 nm) was identified as myricetin-rhamnoside on the

basis of its UV absorbance typical for a flavonoid and MS-MS spectrum. The parent

compound [M+H]+ at m/z 465 ionized to produce the major MS2 fragment at m/z 319,

which is indicative of a rhamnosyl unit (146 amu). Further ionization of m/z 319 ion


280 nm 4 6
S.05 2 .


o .DD -
S36CI nrn 6




0 .0
40.DD 60.01D 80.0 0 1 .00
Minute s


Figure 4-5. HPLC-PDA chromatogram (280 and 360 nm) of Isolate II of muscadine
grapes.









yielded typical myricetin fragment at m/z 301, 273, 255 and 245 (57, 71). Myricetin was

previously identified as the most abundant flavonoid in Noble muscadine grapes (17).

Peak 3 (tR =90.5 min; Xmax =360 nm) has been identified as one of ellagic acid

glycosides due to its typical UV spectrum. MS-MS analysis revealed that [M+H] at m/z

435 lost 132 amu, which corresponds to a xylosyl unit, resulting in an ellagic acid ion at

m/z 303 by MS2 (14, 15). MS3 of ellagic acid produced major ions at m/z 285 and 257.

Consequently, peak 4 is confirmed as ellagic acid xyloside.

Peak 4 (tR =91.3 min; Xmax =361 nm) had a [M+H] at m/z 449 and MS2 produced a

major fragment at m/z 303 (M-146, loss of rhamnosyl group) (72). Therefore, this peak

was identified as ellagic acid rhamnoside due to presence of major ions at m/z 285 and

257 by MS3.

Peak 5 (tR =92.3 min; Xmax =366 nm) is ellagic acid on the basis of identical

retention time and UV spectrum with authentic standard. This was confirmed by the MS-

MS, which yielded a [M-H]- at m/z 301 and prominent MS2 ions at m/z 301, 257 and 229

(14, 15).

Peak 6 (tR =94.2 min; Xmax =351 nm) had [M+H]+ at m/z 449 same as peak 5, which

was identified as ellagic acid rhamnoside and also yield the same aglycone (m/z 303) by

initial ionization. Further ionization produced major fragments at m/z 271, 255, 179, and

151, which is the typical fragmentation pattern for flavonoid compounds. Both ellagic

acid and quercetin have identical molecular weights as 302, however, it has been reported

that MSn fragmentation pattern can be used to discriminate these two compounds. Ellagic

acid seems to have more rigid structure than quercetin because the former produces

bigger ion pieces than the later does with same collision energy (14). Also according to














Table 4-3. UVmax and HPLC-ESI(+)/(-)-MSn analyses of ellagitannins, glycosides of ellagic acid and flavonoids in isolate II from
muscadine grapes.
Peak Rt (min) Compound Amaxa MW MS1(m/z) MS2 (m/z) MS3 (m/z)
No.
261
1 58-60 Ellagitannins 281sh 799 818b 447, 303 277
280sh
2 86.0 Myricetin-Rhamnoside 352 464 465 319: [M+H] -146(Rhamnose) 301,273,255,245

3 90.5 Ellagic acid-Xyloside 360 434 435 303: [M+H] -132(Xylose) 285,257


4 91.3 Ellagic acid-Rhamnoside 361 448 449 303: [M+H] -146(Rhamnose) 285,257


5 92.3 Ellagic acid' 366 302 301 301,257,229

6 94.2 Quercetin-Rhamnoside 351 448 449 303 257, 229, 165

7 97.5 Kaempferol-Rhamnoside 344 432 433 287: [M+H] -146(Rhamnose) 241,213, 165, 133

a. Amax at 2nd band, b. [M+NH4]+, 'characterized by HPLC-ESI(-)-MSn.









UV spectra, this peak is a flavonoid glycoside not an ellagic acid glycoside.

Consequently, peak 7 was identified as quercetin rhamnoside.

Peak 7 (tR =97.5 min; Xmax =344 nm) was kaempferol rhamnoside because [M+H]

at m/z 432 produced aglycone ion at m/z 287 by losing a rhamnosyl (146) unit (71, 72)

and MS3 confirmed kaempferol with fragments at m/z 241, 213, 165, 133 and 121 (73).

Even though PDA was not able to detect certain compound, while MS revealed

trace levels of compounds such as HHDP-galloylglucose ([M+H] ; m/z 635), myricetin-

glucoside ([M+H] ; m/z 481), unknown flavonoid pentosyl conjugate ([M+H]+; m/z 467)

and unknown compounds containing galloyl and acetylrhamnosyl groups ([M+H]+; m/z

923).

Conclusions

Major phytochemicals in muscadine grape were identified by UV spectral

properties and mass-charge ratio followed by extraction with a suitable solid phase

support. The application of multiple MS analysis discovered fragments consistent with

known sugar moieties in ellagic acid glycosides and 4 different ellagitannins in partially

purified extracts of muscadine grape. In the case of ellagitannins, these methods were

able to assess molecular weights of respective fragments, but not exact chemical

identities due to diversity of ellagitannins present with varying functional groups.

Additionally, predominant flavonoids, such as myricetin, quercetin and kaempferol, were

determined as conjugated forms with rhamnose. All identified phytochemicals were

known as excellent antioxidant compounds.














CHAPTER 5
HYDROLYTIC PROPERTIES OF ELLAGIC ACID DERIVATIVES IN MUSCADINE
GRAPES

Introduction

Muscadine grapes have been historically used to produce various food products in

both small and large-scale operations including juice, wine, j am\j elly, and more recently

dried/concentrated products for value-added applications due to their unique and

relatively high antioxidant and anticarcinogenic properties. Since no information

previously existed on the ellagic acid glycoside and ellagitannin content of muscadine

grapes, and likewise, no information is available on ellagic acid conversion from its

precursors associate with heating. Therefore, by evaluating hydrolysis time and

temperature on the relationship of free ellagic acid from its precursors, the functional

properties of these compounds can be evaluated as a result of processing and prolonged

storage. Through the use of acid hydrolysis and evaluation of hydrolyase enzymes,

specific to certain polyphenolics, the release of free ellagic acid was evaluated in

comparison to glycosidic forms for stability characteristics following pasteurization.

Response surface methodology (RSM) is a powerful statistical method for modeling and

analyzing the response of interest within multiple-interrelated parameters in an effort to

optimize this response (58). However in the current study RSM was utilized for a

different purpose as a means to continuously monitor the response as affected by two

independent variables. The objectives of this study were to determine hydrolytic

properties of ellagic acid derivatives and the resultant effects on antioxidant capacity









based on time-temperature combinations using a central composite design for RSM

analysis. Additionally, by exploring the characteristics of ellagic acid glycosides and

ellagitannins in the presence of B-glucosidase and tannase the functional properties of

these compounds can be evaluated in the absence of a heat-catalzyed hydrolytic reaction.

Materials and Methods

Response Surface Methodology (RSM) and Statistical Analyses

Methanolic extracts from the cultivar Doreen were evaluated using various

hydrolytic conditions to monitor changes to ellagic acid derivatives that yield free ellagic

acid in a time-temperature dependent manner. Assessment was conducted based on RSM

with a central composite design (CCD) including 8 treatment combinations and 3 center

points (Figure 5-1). Hydrolysis was performed at pH 3 (control) and at 2N hydrochloric

acid (pH <1) with the design evaluated in duplicate. Time-temperature combinations

were tested at a range of hydrolysis times from 1 min to 2 hrs and temperatures from

20C to 100C in order to generate conditions ranging from partial to complete hydrolysis

at each acid concentration. Data were analyzed using JMP5 software (44) with analysis of

variance and mean separation conducted using the LSD test (p<0.05).

Enzymes Preparation

In an effort to evaluate reactions of polyphenolic-active enzymes on ellagic acid

derivatives, B-glucosidase (E.C. 3.2.1.21) and tannase (E.C. 3.1.1.20) were added to Sep

Pack C 18 unbound and ethyl acetate-soluble fractions, which corresponded to Isolate I

and II, respectively from Chapter 4. B-glucosidase (1.333 units/ml) and tannase (1.738

units/ml) prepared in 0.2M phosphate buffer at pH 5 were added into each fraction at 37

C and reactions were stopped by boiling the solution after 3 hr incubation. Enzyme

treated samples were analyzed compared to controls with no added enzymes. Controls









were prepared at pH 3 and 5 in order to determine the influence of pH on enzyme

reaction and evaluated compounds.

Chemical Analysis

Polyphenolics were quantified with a Waters Alliance 2695 HPLC system

connected to Acclaim 120 C18 column (250 mm x 4.6 mm, Dionex, Sunnyvale, CA)

with a C 18 guard column and to Waters 996 photodiode array detector that recorded UV

spectra from 200-400 nm. Identical mobile phases and gradient elution program were

employed to quantify polyphenol compounds in Chapter 4. Total soluble phenolics were

analyzed using Folin-Ciocalteu assay (40) and antioxidant activity was determined using

the oxygen radical absorbance capacity (ORAC) assay as described in Chapter 3.



Temp. (C)
1000C, 65 min


600C, 1 min


Time (min)
600C, 120min


200C, 65 min

Figure 5-1. Hydrolysis time and temperature combinations included in central composite
design (CCD) experiment.


880C, 20 min 88C, 110 min
I t




600C, 65 min


320C, 20 min 320C, 110 min









Results and Discussion

Ellagic Acid Derivatives and Flavonoid Glycosides: Effects of Time, Temperature
and pHs

Polyphenolic analyses were conducted using the CCD to determine hydrolytic

properties of ellagic acid derivatives including free and glycosidic forms of ellagic acid

and ellagitannins. Also, resultant changes in antioxidant capacity were measured as an

index for functional properties due to changes in chemical composition. The 3-

dimensional representations were developed to show continuous changes in ellagic acid

derivatives as influenced by two independent variables, time and temperature, during

hydrolysis in the presence and absence of hydrochloric acid (Figures 5-2, 3, 4). The data

from hydrolysis at pH 3 was utilized to support the limited information on the influence

of heating on phytochemical compositions and functional properties during thermal

processing of muscadine grape products. Hydrolysis with high acid concentrations (0.5-

2N HC1) at various times and temperature is a common way to evaluate polyphenolic

aglycones from their respective glycosides (59, 60) and has also been used to assess total

ellagic acid following complete hydrolysis of ellagic acid derivatives (12, 14). The

present study attempted to provide a picture for alteration of individual phytochemicals

by statistical methods during hydrolysis in ranges of 1 to 2 hrs and 20 to 100C. Overall,

the levels of ellagic acid derivatives were significantly influenced by increased hydrolysis

time, temperature, and acid concentration that accelerated conversion to free ellagic acid

from ellagic acid precursors. Free ellagic acid (Figure 5-2) changed by 3 and 5-fold with

respect to absence and presence of acid respectively, as compared to initial hydrolysis

conditions (20C and 1 min) through hydrolysis of ellagic acid glycosides and/or

ellagitannins. However, when individually monitored, two ellagic acid glycosides













pH3


6

5I


5 44

S2 3













6

5


3


/ N


0. 120






Te -,,o 4 30 20


-'. ( (C)


2N HCI


7U 60
50
Temp.(0C)


Figure 5-2. Tridimensional representation of free ellagic acid generated using fold
increases by response surface model with central composite design
experiment in the absence and presence of 2N HC1.


I-


\


\


2







51






pH3
2

2. ,. "r.. .




10



1 20 30 20





100
-oo








-]o
2N HCI










-1 30 20
100 40
80 60 50
60 70 40 9





Figure 5-3. Tridimensional representation of ellagic acid xyloside generated using fold
increases by response surface model with central composite design
experiment in the absence and presence of 2N HC1.







52






pH3

2




00





-1 20
30
100 40
80 50 0.
60 060 C
Q 40 80
% 20 90
100









0



















experiment in the absence and presence of2N HCl.
-1 20


80 5040

40 90 80

100

Figure 5-4. Tridimensional representation of ellagic acid rhamnoside generated using fold
increases by response surface model with central composite design
experiment in the absence and presence of 2N HC1.









(xyloside and rhamnoside) (Figures 5-3, 4) did not decrease in concentration at pH 3 that

would indicate glycosidic hydrolysis into free ellagic acid, giving an indication that free

ellagic acid was initially derived from ellagitannins in the extract(s) when hydrolysis

occurred without acid. These ellagic acid glycosides were not completely converted to

free ellagic acid until hydrolysis conditions reached 600C for 65 min, whereby an

appreciable increase in free ellagic acid was observed. Similar observations were made

with cyanidin glycosides in blackberry and quercetin glycosides in onion, cleaving the

sugar moiety in the first hour of acid hydrolysis at 750C (61). The current data were

collected only up to 120 min at fixed acid content (2N HC1), because higher acid contents

and prolonged exposure time to acid might lead to degradation of aglycones following

hydrolysis (61, 62). Additionally, preliminary data observed that the level of ellagic acid

was not significantly changed beyond 120 min at 1000C.

Since flavonoid glycosides also convert into aglycone of flavonoids by liberating a

sugar moiety during hydrolysis, major flavonoid glycosides, myricetin, quercetin, and

kaempferol rhamnosides, were evaluated in ranges of 1.00-2.06, 0.92-1.82 and

1.89-3.57 mg/kg, respectively, as change in time-temperature combinations for the

absence of acid. For 2N HCI hydrolysis, all glycosides were not detected after 65 min.

and 600C, as observed with ellagic acid glycosides. Shown in the 3- dimensional

representations (Figures 5-5, 6, and 7), flavonoid glycosides showed less than 50%

increase after 2hrs at 1000C compare to before hydrolysis at pH 3 indicating that most

flavonoid glycosides might survive through short term heat processing.

Antioxidant Capacity of Polyphenolics as Affected by Aglycone vs Glycosides with
Hydrolysis

Phytochemicals contribute to the functional properties of food systems and their







54







3 pH3



0" 2










40
20
80 80 90 100






2 2NHCI







o 0

2-.

20

60


100 60 70
120 20 30 40 1e cc)


Figure 5-5. Tridimensional representation of myricetin rhamnoside generated using fold
increases by response surface model with central composite design
experiment in the absence and presence of 2N HC1.
















3



2



0
"- 1


Cu-




2
2





Cc-
090
080




20
80 80 0 100
100 60 70

120 20 30 40 e, T G




e 2NHCI










-1

20
A40
0 60
80 80 90 100
100 60 70
50
120 20 30 40 .. x ., )

Figure 5-6. Tridimensional representation of quercetin rhamnoside generated using fold
increases by response surface model with central composite design
experiment in the absence and presence of 2N HC1.







56





3pH3

U,
0
C 2
E



cc 4



10
00



20





e i ia 2N HCIl

U5 2 ,
06







-E I

cc
08
05









20
800
S40

,o\ 60 o
80 90 100
80
100 60 70
50
120 20 30 40 .


Figure 5-7. Tridimensional representation of kaempferol rhamnoside generated using fold
increases by response surface model with central composite design
experiment in the absence and presence of 2N HC1.









roles may diverge depending on concentration, structural differences, and

synergistic/antagonistic responses in foods and biological systems. Many polyphenolics

in fruits and vegetables are found as conjugated forms, with various esterified sugar

moieties. Anthocyanins, the most widely distributed class of flavonoids in plants, are

commonly investigated to explore antioxidant capacity changes among the various

aglycones and glycosidic forms (62-64). Anthocyanidins, the aglycone form of

anthocyanins, tended to have higher radical scavenging activities than those of

corresponding glycosides in ORAC assay (64) or DPPH assay (31), whereas superior

activity was obtained with monoglycosylation of malvidin, pelargonidin, and peonidin in

the P-carotene bleaching method (31). The current experiment was able to investigate the

role of aglycones and sugar conjugated forms including hydrolysable tannins and

glycosides of non-anthocyanins polyphenolics in muscadine grapes to affect metal

reduction and hydroxyl radical scavenging properties as a result of different stages of

hydrolysis. Contour plots were presented for total soluble phenolics (Figure 5-8) and

antioxidant capacity (Figure 5-9) by Folin-Ciocalteu assay and oxygen radical absorbance

capacity (ORAC) assay, respectively. Data on both attributes support that aglycone

polyphenolics have higher ability to reduce metal ions and scavenge hydroxyl radical

compared prior to hydrolysis of polyphenolics. Ellagic acid aglycones showed strong

correlation with both total soluble phenolics and antioxidant capacity, r=0.89 and r=0.58,

respectively, while lower correlations were observed in ellagic acid glycosides, average

52% and 36% for total soluble phenolics and antioxidant capacity, respectivley (Table 5-

1). Usually flavonoids glycosides have been evaluated as containing lower antioxidant







58






pH3
7

6

5

0- 4

003
o, a


N 2
0 2 120
1-u-
100


1090 6 0 "
60 50 40

Ternp(OC) 3020



2N HCI
7


6

0 5




a,,
0- 4
0 8






90 80 70 50 40 30 20

TeMop. (OC) 20




Figure 5-8. Tridimensional representation of total soluble phenolics generated using fold
increases of Folin-Ciocalteu measurements by response surface model with
central composite design experiment in the absence and presence of 2N HCI.





















C.
C-


CZ





90 6O8 0 40 20
70
LL 120


0100
400


300









-60 50 4 3 20
Tenp(o) 20



Acid
5









.212
1 100
80
0 60 ,
90 80 70040
60 504 020

TeMp(0o 20

Figure 5-9. Tridimensional representation of antioxidant capacity generated using fold
increases of ORAC measurement by response surface model with central
composite design experiment in the absence and presence of 2N HC1.









capacity compared to aglycone of flavonoids, since blocking the 3-hydroxyl group in the

heterocyclic ring influenced stability of the aroxyl radical of flavonoids and also a

decrease in number of free hydroxyl group (-OH) might play an important role in

scavenge free radicals (74). The structure-activity relationship for hydrolysable tannins,

ellagitannins and hydrolytic compounds has not been elucidated with pure compounds;

however these data provide evidence that chemical antioxidant capacity increases with

higher concentrations of hydrolytic or aglycone forms. Even though HHDP carries 6

hydroxyl groups compared to 4 in free ellagic acid, their electron donating properties are

likely inhibited by the presence of the glycosidic moiety. However, due to the low water

soluble characteristics of ellagic acid, the tested samples were prepared to contain low

amounts of ellagic acid derivatives (<5 mg/mL as total ellagic acid). Therefore, this

premise on higher radical scavenging activity of free ellagic acid should be limited to the

fact that all free ellagic acid was completely solubilized in solution; however, this is



Table 5-1. Pearson correlations coefficients of individual ellagic acid derivatives contents
with total soluble phenolics and antioxidant capacity.
Variable 1 Variable 2 Partial hydrolysis Complete hydrolysis
at pH 3 with 2N HCI

Total Soluble Phenolics Antioxidant Capacity 0.77 0.81
Ellagic acid aglycone Total Soluble Phenolics 0.89 0.98
Antioxidant Capacity 0.58 0.83
Ellagic acid-xyloside Total Soluble Phenolics 0.42 -0.29
Antioxidant Capacity 0.30 -0.35
Ellagic acid-rhamnoside Total Soluble Phenolics 0.62 -0.23
Antioxidant Capacity 0.42 -0.28









the first trial to compare the structure-activity relationship between glycosides or

hydrolysable tannins and aglycones in crude plant extracts.

Enzymatic Hydrolysis of Ellagic Acid Derivatives

Though considered a less prevalent technique, hydrolysis by enzymes such as

esterase are alternative methods to release aglycones from various polyphenolics through

degradation of carbohydrate linkages (59). B-glucosidase (E.C. 3.2.1.21) as well as other

glycosidases like arabinofuranosidase were considered important enzymes associated

with the quality of foods and beverages and have been used to release aroma components

in musts, wine, and fruit juices (65). Tannase or tannin acyl hydrolase (E.C. 3.1.1.20) are

widely applied in the food industry to produce instant tea, to manufacture gallic acid and

propylgallate, which can be utilized as food preservatives, and are commonly used to

remove undesirable tannins. However, there are no published data on reactivity of these

enzymes on ellagitannins and ellagic acid glycosides in food systems. Muscadine grape

extracts were hydrolyzed by B-glucosidase and tannase in order to provide more

information on the properties of ellagic acid derivatives associated with enzyme reactions.

Overall, the two enzymes showed distinctive responses in each grape isolate evaluated

(Table 5-2) due to their characteristic polyphenolic composition. As described in Chapter

4, an ethyl acetate fraction contained mostly polyphenolic compounds including free

ellagic acid, ellagic acid xyloside, ellagic acid rhamnoside, and ellagitannins (MW 799),

also there is evidence for the presence of gallotannins in this fraction. Compared to the

complex components in the ethyl acetate fraction, the water fraction is relatively simple

containing tannin types of ellagic acid precursors such as three different ellagitannins

(MW 802, 834, and 832) and HHDP-galloyl glucose. Therefore, enzyme reactivity on

ellagic acid glycosides was evaluated in ethyl acetate fraction and any free ellagic acid









evolution in the water fraction was a consequence of enzyme reactions on ellagitannins.

Reactions were performed at pH 5 to satisfy optimal conditions for both enzymes, and

fortunately this elevated pH did not significantly affect the components as compared to

the original pH of grapes. The activities of each enzyme were effectively evaluated by

HPLC by monitoring both products and reactants under hydrolysis conditions. 3-

glucosidase showed higher reactivity on xylose compared to rhamnose moieties as

indicated by a significant decrease in ellagic acid xyloside (3.14 to 1.05 mg/kg), whereas

the rhamnoside was unaltered in the ethyl acetate soluble fraction after 3 hrs incubation at

37C. Despite the observed decrease in ellagic acid xyloside, a corresponding increase in

free ellagic acid was not observed due to the low initial concentrations of the xyloside

present. However, when higher concentrations of 1-glucosidase were used in preliminary

studies, free ellagic acid significantly increased and was also found in the insoluble

sediments. A correspondingly significant decrease in both ellagic acid xyloside and

ellagic acid rhamnoside was observed after 2 hrs incubation at 37C (data not shown).

Reaction of B-glucosidase on each ellagic acid glycoside might vary depending on

the sugar groups attached to ellagic acid, because ellagic acid glycoside was hydrolyzed

with B-glucosidase in a preliminary study. After incubation of each isolate with tannase,

no activity on either ellagic acid glycosides or HHDP units of ellagitannins was detected.

However, free gallic acid was significantly affected compared to control as a 17.8-fold

increase. Considering the composition of ethyl acetate fraction, additional free gallic acid

after incubation with tannase can be a hydrolytic product of both gallotannins and

ellagitannins. However, no significant increase in gallic acid for the water fraction

suggests that tannase has higher reactivity for gallotannins than ellagitannins. The









reactivity of tannase (from Aspergillusficuum) may not depend on sugar specificity as

discussed in B-glucosidase due to no change in ellagic acid glycosides with tannase

incubation. Consequently, tannase seems preferably to cleave off a single gallic acid

rather than oxidative coupling of neighboring gallic acid, HHDP unit, from ellagitannins.

Various studies on tannase activity investigated various types of hydrolysable tannins

such as tannic acid, methylgallate, ethylgallate, or n-propylgalltae (75-77). Using this

form of tannase for muscadine grape extracts did not show promise in hydrolyzing

ellagitannins into free ellagic acid, however, other tannase sources may have different

reactivities to those compounds present in muscadine juice or wine as a means to

alleviate quality defects or to utilize muscadine pomace to produce ellagic acid.


Table 5-2. Concentrations (mg/kg) of ellagic acid, ellagic acid glycosides (xyloside and
rhamnoside) and gallic acid affected by enzyme treatment in two different
fractions from Doreen (bronze) extracts.
Fractions1 pH Enzyme Free ellagic Ellagic acid Ellagic acid Gallic acid
acid xyloside rhamnoside

3 None 31.9 b4 3.92 a 10.5 a 1.80 c

5 None 39.4 ab 3.14 a 11.3 a 1.80 bc
Ethyl acetate
5 B-glucosidase 2 41.4 ab 1.05 b 10.7 a 2.78 b

5 Tannase3 47.2 a 3.86 a 11.4 a 32.1 a

3 None 1.47 a N.D. N.D. 0.589 ab

5 None 1.27 a N.D. N.D. 0.577 ab
Water
5 B-glucosidase 2 1.37 a N.D. N.D. 0.505 b

5 Tannase3 1.29 a N.D. N.D. 0.789 a
prepared by Sep-pak C18. 2. 1.333 units/ml, 3. 1.738 units/ml in final solution and 3hrs
incubation at 37C.4.Similar letters within columns for each fraction are not
significantly different (LSD test. P<0.05).









Conclusions

Properties of ellagic acid derivatives were studied as affected by non-enzymatic

(time-temperature combinations) and enzymatic (8-glucosidase and tannase) hydrolysis.

Elevated time and temperature without acid created the environment for partial

hydrolysis of ellagic acid derivatives including ellagitannins and ellagic acid glycosides;

however most glycosidic components (ellagic acid glycosides and flavonoid glycosides)

remained after the reaction. Additional 2N HCI completely hydrolyzed ellagitannins and

ellagic acid glycosides in lhr and the free ellagic acid produced was likely to scavenge

more hydroxyl radicals than conjugated forms of ellagic acid. This was also indicated by

a significant increase in antioxidant capacity with evolution of ellagic acid after heating

both in the absence and presence of acid. B-glucosidase showed the possibility for

application on muscadine grape juice or other products to hydrolyze ellagic acid

glycosides; however tannase was not a feasible option for ellagic acid precursors.














CHAPTER 6
HYDROLYTIC AND OXIDATIVE PROPERTIES ON ELLAGIC ACID
DERIVATIVES DURING STORAGE OF MUSCADINE GRAPES JUICES

Introduction

Free ellagic acid increases with hydrolysis of its precursors including ellagic acid

glycosides and ellagitannins and contributes to the development of insoluble sediments in

muscadine juice or wine during storage. Prior to investigating chemical or physical

processing options to remediate or accelerate sediments in muscadine products, it is

important to determine key components affecting relative changes of ellagic acid

derivatives during storage.

According to a recent study on phytochemical stability of muscadine grape juice

(33), phytochemical losses following processing with high hydrostatic pressure (HHP)

were presumably due to the activation of residual oxidases after juice extraction and/or

autoxidative mechanisms resulting in co-oxidation of anthocyanins and ascorbic acid.

Ascorbic acid fortification is common for producing juice with additional oxidative

protection while contributing to additional health benefits, quality, and market value. It is

hypothesized that ascorbic acid fortification may protect ellagitannins from oxidative

degradation in non-anthocyanin containing muscadine grape juices as ascorbic acid is

commonly added in plant extracts to protect ellagitannins from oxidation during analysis

(26). However, few studies have investigated the effects of ascorbic acid fortification on

oxidative stability of individual polyphenolics, especially ellagic acid conversion from

ellagic acid derivatives, in food products.









The objective of study was to evaluate individual ellagic acid derivatives in whole

juice and sub-isolates over time in an effort to reveal the main precursors for

producing free ellagic acid during storage of muscadine juice. Additionally, by assessing

ellagic acid derivatives as affected by thermal processing, ascorbic acid fortification, and

sparging with air, conclusions can be drawn concerning the chemical and oxidative

behaviors of ellagic acid derivatives over time in muscadine grape juice.

Materials and Methods

Storage of Red Juices and Isolations

In order to monitor the behaviors of individual ellagic acid derivatives in whole

juice and isolations, hot-pressed red muscadine juice was initially prepared by blending

Noble and Albemarle (1:1) cultivars, which contain high anthocyanins and ellagic acid

derivatives, respectively. Grapes were donated from local grape growers in central

Florida and were frozen until processed. Equal portions of each grape were blended and

pressed following heating the grapes at 70 C for 15 min. Juice was filtered and thermally

pasteurized (90 C, 5 min). Isolations were prepared with Sep-Pak C18 cartridge as

described in Chapter 4. Resulting isolates including whole juice, water (unbound), ethyl

acetate and methanol isolates were stored at 4 and 37C for 5 weeks.

Storage of White Juices and Isolations

Cold-pressed white muscadine grape (Doreen) juice was prepared by simply

crushing and pressing the fruit in an effort to evaluate the influence on ellagic acid

derivatives by thermal pasteurization, ascorbic acid, and air sparging. To compare

thermal processing, a portion of juice was thermally pasteurized (90 C, 5 min) and to the

remaining juice sodium azide was added to retard microbiological growth prior to









treatment application. Portions of the remaining juice (unpasteurized) were then fortified

with ascorbic acid (1,000 mg/L) and equally distributed into flasks and sparged with air.

Air bubbled into samples for 1 hr at room temperature and additional buffer (citric acid

buffer, pH 3.2) was added to recover loss of water during the treatment as needed. Equal

volumes of non-treated and treated samples were kept in test tubes at 4'C and 37C for 5

weeks.

Chemical Analysis

Samples were collected at 0, 1, 3, and 5 weeks from both temperatures and

centrifuged to remove insoluble particles prior to analysis. Ellagic acid derivatives were

then analyzed by HPLC, as described in Chapter 5. Total ellagic acid was evaluated

following acid hydrolysis (2N HCI for 60 min at 95 C) and separation was achieved

using phase B (60% methanol, pH 2.4) changed from 50-70% in 3 min; 70-80% in 2 min;

80-100% in 20 min; and 100% B in 5 min for a total run time of 30 min.

Results and Discussion

Changes of Ellagic Acid Derivatives during Storage of Red Juice

Muscadine grape products are prone to developing insoluble sediments that are

likely created from ellagic acid derived from hydrolysis of its precursors during storage.

Unfortunately, various processing and storage regimes have not been successful for

reducing sediment formation (1, 2, 28-30) due to lack of understanding the behavior of

each ellagic acid derivative during storage. In this study, muscadine juice and 3 isolates

from Sep-Pak C18 were stored for 5 weeks at 4 C and 37 C, and relative changes in

individual ellagic acid derivatives were quantified. Significant changes were observed in

each isolate during storage, and were influenced by storage temperature. The whole juice

prior to fractionation represented intact juice and quantified chemical attributes were









varied for storage time (Figure 6-1). Variations were more significant at 370C, especially

for free ellagic acid concentration with a 143% increase compared to 58% increase at 40C.

Increases in free ellagic acid were the result of ellagitannin degradation rather than

ellagic acid glycosides, since two quantified ellagic acid glycosides decreased only 4%

and 9% on average at 4 and 37C, respectively. Changes in free ellagic acid

concentrations should be considered with formation of sediments, but accurate analyses

on sediments were difficult due to the low juice volume used for evaluation. However,

free ellagic acid loss via sediments were indirectly determined by evaluating total ellagic

acid as sediments removed prior to analysis and resulted in a decrease in total ellagic acid

concentrations. Total ellagic acid observed changes over 5 weeks were not significant

with an 11% increase at 4'C and 11% decrease at 37C. Considering that whole juice

developed the most insoluble components during storage, suggesting free ellagic acid

seems to be a minor contributor and supported by a previous study where 12% of

sediment by weight was free ellagic acid (16).

Water (unbound) isolate was prepared by a Sep-Pak C 18 with water to elute polar

ellagic acid precursors such as ellagitannins, and the stability of ellagitannins were

measured indirectly by evaluating total ellagic acid. Total ellagic acid was considered as

hydrolytic ellagic acid from mainly ellagitannins, because only trace levels (<1 mg/L) of

ellagic acid glycosides were evaluated. Pro-ellagic acid compounds did not significantly

change at 5C but at 37C total ellagic acid decreased 78% (25 -> 5 mg/L) (Figure 6-2)

indicating significant degradation of ellagitannins. It is interesting to note that free ellagic

acid concentrations were not influenced by ellagitannin hydrolysis, indicating that free

ellagic acid may precipitate by other components such as metal ions, soluble pectin, or









organic acids isolated in this fraction. It was shown previously that ellagic acid has the

ability to bind metal ions (16, 78, 79).

Ellagic acid glycosides were evaluated in both ethyl acetate and methanol isolates

(Figures 6-3, 4) and all ellagic acid glycosides including ellagic acid xylosides and ellagic

acid rhamnosides were observed to decrease ranging from 12 to 14% and 13 to 35% at

4C and 370C, respectively. Ellagic acid glycosides such as ellagic acid acetyl-xyloside

and ellagic acid arabinoside were reported to be stable for up to 6 months in raspberry

jam (45). Compared to the slow degradation of ellagic acid glycosides in the ethyl acetate

isolate, total ellagic acid (1 88%) and free ellagic acid ( 37%) showed more

distinguishable changes during storage at 370C. Since the presence of ellagitannins in the

ethyl acetate isolate was confirmed by HPLC-MS/PDA in Chapter 4, ellagitannins

degraded and formed insoluble compounds in solution resulting in a decrease in total

ellagic acid. Consequently, elevated storage temperature significantly accelerated

hydrolysis of ellagic acid precursors during storage time. Among ellagic acid precursors,

ellagitannins are likely to hydrolyze before ellagic acid glycosides, and then stay in

solution or contribute to the formation of insoluble compounds with other juice

constituents. Additionally, consideration of the possible effects of oxidation on

ellagitannins and resultant conversion into free ellagic acid in juice storage was needed,

because ellagitannins are very susceptible to oxidation (27).

Initial Ellagic Acid Derivatives in White Muscadine Juice as Affected by Ascorbic
Acid and Air

Ascorbic acid is commonly fortified into fruit juices or products in efforts to retard

the oxidation and add nutritional value; however this is challenge for red grape juice due

to mutually destructive properties between anthocyanins and ascorbic acid in presence of

















1000 -




800 -

--

E 600 -
'O




0
--
8 4000


200



n-


100




80
-j


S60
.2
0


| 40-
0


20




0


0 1 2 3 4 5 6
Time (weeks)


-0- EA-xylose
-V-- EA-rhamnose
Free Ellagic


I I I I I I
0 1 2 3 4 5 6
Time (Weeks)


1000n


800 -
--


S600 -





0
C-

0 400
-




200



0-


0 1 2 3 4 5 6
Time (weeks)



Total Ellagic


0 1 2 3 4 5 6
Time (Weeks)


Figure 6-1. Changes in whole red juice for ellagic acid derivatives during storage. A) At

40C. B) At 370C.


A 40C


U


B 370C


140


120


100
--


S80
1
0.



o
S 60-

o 40



20
0
0







0


pp p 0














A 40C
1.2


30


25 -


S20-
E
0
S 15


0 10-

5-
5-


1.0


0.8-


1 0.6-


o 0.4-
)


0.2 -


0.0


B 370C
1.2 --


30 -


25 -


20-
E
0)


15-


5 10-
0-


0 1 2 3 4 5 6
Time (weeks)


0 1 2 3 4 5 6
Time (weeks)



Total Ellagic


0 1 2 3 4 5 6
Time (weeks)


Figure 6-2. Changes in water isolate for ellagic acid derivatives during storage. A) At
40C. B) At 370C.


I--I I-
0 1 2 3 4 5 6
Time (weeks)


-0- EA-xylose
-V- EA-rhamnose
Free Ellagic


1.0 -


0.8-
0)


0.6-


c 0.4-
)


0.2 -


00 -














A 40C
10 ---


100



80
--


c 60
O

0)
2O
I 40-
0



20



0


0 1 2 3 4 5 6
Time (Weeks)


-0- EA-xylose
-V- EA-rhamnose
Free Ellagic


B370C


100



80

E 0-

60



8 40
0
0


20


0 1 2 3 4 5 6

Time (weeks)


8 -
-j




0 4-
E 6 -

C4)
0
0


o-




2-



0
-


0 1 2 3 4 5 6
Time (Weeks)


Figure 6-3. Changes in ethyl acetate isolate for ellagic acid derivatives during storage. A)
At 40C. B) At 370C.


0 1 2 3 4 5 6

Time (Weeks)


Total Ellagic Acid


10



8-





0
E 6-

)
0
| 4-

-0
0

2-



0 -



















250 -



200 -



-. 150 -
0


100
0


50


50



40 -
-j


1 30


20

)


10



0


EA-glucoside
-V-- EA-xyloside
--- EA-rhamnoside
Free Ellagic

B 370C


250



200

--

150
0


--
0


50


0 1 2 3 4 5 6
Time (weeks)


0 1 2 3 4 5 6
Time(weeks)



Total Ellagic


0 1 2 3 4 5 6
Time (weeks)


Figure 6-4. Changes in methanol isolate for ellagic acid derivatives during storage. A) At
40C. B) At 370C.


A 40C


0 1 2 3 4 5 6
Time(weeks)


50



40 -
-j


E 30-
--
0


0
0
C--



10



0


-U









oxygen (84, 85). For this reason, the white muscadine grape cultivar Doreen was

investigated for fortification as a means to study stability and the effects of oxygen on

ellagic acid derivatives during storage. Prior to treatment, juices were either pasteurized

or non-pasteurized to evaluate the effect of treatments. Before storage, juices were

analyzed for initial levels of ellagic acid derivatives to determine treatment effects

(Figure 6-5). A comparison of heated and non-heated juice resulted in a significant

increase in free ellagic acid and ellagic acid glycosides, but total ellagic acid remained.

This data supported that ellagitannins break down into free ellagic acid by elevated

temperature prior to hydrolysis of ellagic acid glycosides, as observed by RSM in

Chapter 5. In non-heat treated juices, ascorbic acid seemed to play a role in increasing

only free ellagic acid because 2.5 and 3.4 fold increases were observed in samples with

ascorbic acid fortification, and the combination of ascorbic acid and air sparging,

respectively. Levels of ellagic acid glycosides or total ellagic acid were not significantly

changed by ascorbic acid fortification indicating that ascorbic acid may help to retain free

ellagic acid in solution or to accelerate ellagitannins hydrolysis. This impact of ascorbic

acid on ellagitannins and ellagic acid were likely related to the presence of natural

oxidative enzymes such as polyphenol oxidase or peroxidase since ascorbic acid

fortification did not influence ellagic acid derivatives in thermally processed juices.

Compared to changes by ascorbic acid, excessive amounts of air in the system did not

influence the initial level of free ellagic acid. Air sparging at natural juice pH is not likely

to be at optimal conditions to induce oxidation of ellagic acid derivatives, since the mode

of oxidation is expected to be influenced by the presence of semiquinones that form by

the action of phenolate anions with triplet oxygen (80). Oxidation of free ellagic by air






























Control AsA only Air only AsA+Air


m_1 EA-xylose
SEA-Rham
S Free Ellagic
Total Ellagic


0 IoIm I IImP I I Imm I Imm I
Control AsA only Air only AsA+Air
Figure 6-5. Concentrations (mg/L) of ellagic acid derivatives depending on treatments of
ascorbic acid (1,000 mg/L) and air at Oday of Doreen juice as affected by
thermal pasteurization method (A: Non-heat, B: Heat, indicates significant
differences between non-heat and heat treatments).









sparging could have occurred with an elevated pH due to hydrolysis of its lactone ring or

by alkanine hydrolysis of ellagic acid glycosides and ellagitannins.

Post-storage Levels of Ellagic Acid Derivatives in White Muscadine Juice as
Affected by Ascorbic Acid and Air

The evolution of free ellagic acid is a consequence of hydrolysis of ellagic acid

derivatives during long term storage. In this study, white muscadine juices were stored

for 5 weeks at 370C and relative changes in ellagic acid derivatives quantified to

determine the effects of pasteurization process, ascorbic acid fortification and excessive

amounts of air on each attribute. Significant changes were observed for each ellagic acid

derivatives during storage; however treatments with ascorbic acid fortification and air

sparging for juices were not major factors on relative changes of ellagic acid derivatives

over time. Among different ellagic acid derivatives, only free ellagic acid showed effects

of ascorbic acid fortification (Figure 6-6). Ascorbic acid fortification was initially

evaluated to retain higher concentrations of free ellagic acid, but ascorbic acid fortified

juice retained greater levels of free ellagic acid through storage in both non-heat and heat

treated juices.

Significant changes in chemical composition of stored juices were influenced by

thermal processing prior to individual fortification or air sparging. The heat-treated juice

without additional treatments had initially higher concentrations of free ellagic acid and

changed from 4.02 to 9.35 mg/L during storage, compare to 0.74 to 2.53 mg/L in non-

heat treated juice. It is interesting to note that most of the increase in free ellagic acid for

non-heat treated juice occurred in the first week of storage and the level decreased but not

significantly. However, the heat treated juice increased in the last two weeks (Figure 6-6).

The relative stability of ellagic acid glycosides were confirmed again in heat treated































0 1 2 3 4 5 6
Time (weeks)


0 1 2 3 4 5


Time (weeks)

Figure 6-6. Concentrations (mg/L) of free ellagic acid depending on ascorbic acid (1,000
mg/L) and air during storage (5 weeks, 37C) of Doreen juice as affected by
thermal pasteurization (A: Non-heat treatment, B: Heat treatment).




































0 1 2 3 4 5 6

Time (weeks)


0)
E 0.6





0.0
0
t*-'


00.2


0.0


0 1 2 3 4 5


Time (Weeks)
Figure 6-7. Concentrations (mg/L) of ellagic acid xyloside depending on ascorbic acid
(1,000 mg/L) and air during storage (5 weeks, 37C) of Doreen juice as
affected by thermal pasteurization (A: Non-heat treatment, B: Heat treatment).
































0.2


0.0



1.4


1.2


1.0


0.8
c-
0
o 0.6
e-*
0
a0.4
0

0.2


0.0


0 1 2 3 4 5 6


Time (weeks)


0 1 2 3 4 5 6
Time (weeks)
Figure 6-8. Concentrations (mg/L) of ellagic acid rhamnoside depending on ascorbic acid
(1,000 mg/L) and air during storage (5 weeks, 37C) of Doreen juice as
affected by thermal pasteurization (A: Non-heat treatment, B: Heat treatment).






































0 1 2 3 4 5 6

Time (weeks)


70


60


50
0)

E 40
c
o 30
c-
0
a 20
0


10


0


0 1 2 3 4


5 6
Time (weeks)


Figure 6-9. Concentrations (mg/L) of total ellagic acid depending on ascorbic acid (1,000
mg/L) and air during storage (5 weeks, 37C) of Doreen juice as affected by
thermal pasteurization (A: Non-heat treatment, B: Heat treatment).









juices based on only 27% and 17% decrease for ellagic acid xyloside and ellagic acid

rhamnoside, respectively. This stability of ellagic acid glycosides in heat-treated juices

seems to be affected by thermal processing, since ellagic acid xyloside were no longer

detected after 3 week storage and ellagic acid rhamnoside decreased by 74% after 5

weeks in non-heat treated juices (Figures 6-7, 8). The changes of ellagitannins were

monitored indirectly as total ellagic acid. Total ellagic acid in non-heat treated juices

continuously decreased with storage and retained only 35% of its initial value after 5

weeks (Figure 6-9). This decrease in total ellagic acid was impacts of degradation of

mainly ellagitannins, rather than ellagic acid glycosides due to the low amounts of ellagic

acid glycosides (< 1 mg/L) in solution. Data on ellagic acid derivatives for control

suggested that thermal pasteurization could be important factor on relative changes on

ellagic acid precursors during storage through impacts on natural enzymes in juice.

Ellagic acid derivatives may not be primer targets for oxidative enzymes due to lack of o-

dihydroxyl in the structure (33, 81). However, thermal pasteurization may hinder the

developments of o-quinones or secondary oxidation products from phenolic acids in

grape juice (82, 83), and affect to ellagitannins degradation.

Conclusions

Changes in ellagic acid derivatives of muscadine juices were evaluated initially and

after 5 weeks storage to determine primer precursor for free ellagic acid evolution.

Through evaluating whole juice and each isolates, ellagitannins seemed to more closely

influence on free ellagic acid evolution rather than ellagic acid glycosides. Additionally,

other miscellaneous components such as metal ions, soluble pectin or organic acid were

likely to accelerate the free ellagic acid decrease via formation of sediments. Ascorbic

acid fortification is probably non-harmful options to retain free ellagic acid in white






82


juices. Thermal pasteurization was a significant factor for relative changes in ellagic acid

derivatives during storage likely due to inactivate natural enzymes present in juices, but

additional studies are required to directly evaluate the presence of enzymes and their

activities in juices prepared by different pasteurization schemes.














CHAPTER 7
SUMMARY AND CONCLUSION

Studies were conducted to provide improved information on ellagic acid

derivatives including free ellagic acid, ellagic acid glycosides and ellagitannins in

muscadine grape (Vitis rotundifolia) to understand the role of these compounds

influencing on quality. Overall phytochemicals evaluations in different cultivars of

muscadine grapes demonstrated that ellagic acid derivatives were present in a wide range

of concentrations and were influenced by ripening, physiology, and juice processing,

resulting vary in antioxidant capacity. New blending schemes with Noble and Albemarle

can be suggested for red muscadine grape juice or wine to produce high quality products

in terms of high color intensity and high contents of ellagic acid derivatives with

corresponding high antioxidant capacity.

The main antioxidants were isolated with a series of solid phase extraction and

identified by application of advanced chromatographic techniques, PDA and MS

detectors connecting to HPLC. Predominant ellagic acid glycosides and flavonoid

glycosides were determined their chemical identities; ellagic acid glycoside, ellagic acid

xyloside, ellagic acid rhamnoside, myricetin rhamnoside, quercetin rhamnoside, and

kaempferol rhamnoside in muscadine grape. In the case of ellagitannins, these methods

were able to assess molecular weights of the respective fragments, but not exact chemical

identities due to diversity of ellagitannins present with varying functional groups.

Using response surface methodology with two independent variables, time and

temperature, successfully demonstrated that evolution of ellagic acid was a result of









temperature dependent hydrolysis of ellagic acid glycosides and ellagitannins in both

presence and absence of acid. This additional free ellagic acid in solution was likely to

play a significant role for scavenging hydroxyl radical indicated by high correlation

(r-=0.83) between two attributes. Enzymatic application with 13-glucosidase (E.C.

3.2.1.21) or tannase (E.C. 3.1.1.20) was tested to suggest options for hydrolysis of ellagic

acid glycosides and ellagitannins. B-glucosidase showed promise as a way to hydrolyze

ellagic acid glycosides, resulting in high free ellagic acid content. However, tannase was

not a feasible option for break down of ellagic acid precursors.

Through evaluating whole juice and each isolates, ellagitannins seemed to be the

main precursor for free ellagic acid evolution since ellagic acid glycosides were evaluated

relatively stable during storage. Additionally, it is possible that ellagic acid precipitation

may be aided by binding other components forming insoluble precipitates such as

proteins, short chain pectins, organic acids, or metal ions. Thermal processing for

pasteurization increased free ellagic acid via ellagitannins hydrolysis, and also influenced

the kinetic changes of ellagic acid derivatives during storage, possibly due to inactivation

of natural enzymes present in juices. Additional studies are required to directly evaluate

the presence of enzymes and their activities in juices prepared by different pasteurization

schemes.















LIST OF REFERENCES

1. Lin, T. Y.; Vine, R. P. Identification and reduction of ellagic acid in muscadine
grape juice. J. FoodSci. 1990, 55, 1607-1609, 1613.

2. Garrido, V. M.; Sims, C. A.; Marshall, M. R.; Bates, R.P. Factors influencing
ellagic acid precipitation in muscadine grape juice during storage. J. Food Sci.
1993, 58, 193-196.

3. Olien, W. C. The muscadine grape: Botany, viticulture, history, and current
industry. Hortsci.. 1990, 25, 732-739.

4. Winkler, A. J.; Cook, J. A.; Kliewer, W. M.; Lider, L. A. General Viticulture.
Univ. of California Press; Berkeley; 1974.

5. Poling, E. B.; Mainland, C. M.; Earp, J. B. Muscadine grape production guide for
North Carolina. N.C., Agric. Ext. Serv. AG-94; 1984.

6. USDA (United States Department of Agricultre). Muscadine grapes-a fruit for the
south. Farmers' Bull. 1965; pp 2157.

7. Olien, W. C. Muscadine A classic southeastern fruit. Hortsci. 1990, 25, 726- 831.

8. Carroll, D. E. Muscadine grapes: Factors influencing product quality. Evaluation of
quality of fruits and vegetables. Ed. Pattee, H.E., AVI Publishing Co., Westport,
CT. 1985

9. Woodroof, J. G.; Cecil, S. R.; DuPree, W.E. Processing muscadine grapes. Ga.
Agric. Esp. Stn., Bull. [N.S.] 1956; 17:1-35.

10. Flora, L.F. Processing and quality characteristics of muscadine grapes. J. Food Sci.
1977, 42, 935-938, 952.

11. Haslam, E. Symmetry and promiscuity in procyanidin biochemistry. Phytochem..
1977, 11, 1207-1218.

12. Daniel, E. M.; Krupnick, A. S.; Heur, Y. H.; Blinzler, J. A.; Nims, R. W.; Stoner,
G. D. Extraction, stability, and quantitation of ellagic acid in various fruits and
nuts. J. Food Comp. and Anal. 1989, 2, 338-349.









13. Amakura, Y.; Okada, M.; Tsuji, S.; Tonogai, Y. High-performance liquid
chromatographic determination with photodiode array detection of ellagic acid in
fresh and processed fruits. J. Chrom. A. 2000, 896, 87-93.

14. Mullen, W.; Yokota, T.; Lean, M. E.; Crozier, A. Analysis of ellagitannins and
conjugates of ellagic acid and quercetin in raspberry fruits by LC-MSn Phytochem.
2003, 64, 617-624.

15. Mullen, W.; McGinn, J.; Lean, M. E.; MacLean, M. R.; Gardner, P.; Duthie, G. G.;
Yokota, T.; Crozier, A. Ellagitannins, flavonoids, and other phenolics in red
raspberries and their contribution to antioxidant capacity and vasorelaxation
properties. J. Agric. Food Chem. 2002, 50, 5191-5196.

16. Lee, J. H.; Talcott, S. T. Ellagic acid and ellagitannins affect on sedimentation in
muscadine juice and wine. J. Agric. Food Chem. 2002, 50, 3971-3976.

17. Talcott, S. T.; Lee, J. H. Ellagic acid and flavonoid antioxidant content of
muscadine wine and juice. J. Agric. Food Chem. 2002, 50, 3186-3192.

18. Pastrana-Bonilla, E.; Akoh, C. C.; Sellappan, S.; Krewer, G. Phenolic Content and
Antioxidant Capacity of Muscadine Grapes. J. Agric. Food Chem. 2003, 51, 5497-
5503.

19. Mertens-Talcott, S. U.; Talcott, S. T.; Percival, S. S. Low concentrations of
quercetin and ellagic acid synergistically influence proliferation, cytotoxicity and
apoptosis in MOLT-4 Human Leukemia Cells. J. Nutr. 2003, 133, 2669-2674.

20. Stoner, G. D.; Morse, M. A. Isothiocyanates and plant polyphenols as inhibitors of
lung and esophageal cancer. Cancer Lett. 1997, 114, 113-119.

21. Khanduja, K. L.; Gandhi, R. K.; Pathania, V.; Syal, N. Prevention of N-
nitrosodiethylamine-induced lung tumorigenesis by ellagic acid and quercetin in
mice. Food Chem Toxicol. 1999, 37, 313-318.

22. Helm, R. F.; Zhentian, L.; Ranatunga, T.; Jervis, J.; Elder, T. Toward
understanding monomeric ellagitannin biosynthesis. In Plant polyphenols 2:
Chemistry, biology, pharmacology, ecology. Ed. Gross, G.G., Hemingway, R.W.,
and Yoshida, T., Kluwer Academic/Plenum Publishers, New York, NY. 1999

23. Quideau, S.; Feldman, K. S. Ellagitannin chemistry. Chem. Rev. 1996. 96, 475-503.

24. Gill, M. I.; Tomas-Barberan, F. A.; Hess-Pierce, B.; Holcroft, D. M.; Kader, A. A.
Antioxidant activity of pomegranate juice and its relationship with phenolic
composition and processing. J. Agric. Food Chem. 2000, 48, 4581-4589.

25. Machado, T. B.; Leal, I. C. R.; Amaral, A. C. F.; dos Santos, K. R. N.; da Silva, M.
G.; Kuster, R. M. Antimicrobial ellagitannins ofPunica granatum fruits. J. Braz.
Chem. Soc. 2002, 13, 606-610.









26. Salminen, J. P.; Ossipov, V.; Loponen, J.; Haukioja, E.; Pihlaja, K.
Characterization of hydrolysable tannins from leaves of Betulapubescens by high-
performance liquid chromatography-mass spectrometry. J. Chromatogr. A. 1999,
864, 283-291.

27. Foo, L. Y. Amarinic acid and related ellagitannins from Phyllanthus amarus.
Phytochem. 1995, 39, 217-224.

28. Boyle, J. A.; Hsu, L. Identification and quantitation of ellagic acid in muscadine
grape juice. Am. J Enol. Vitic. 1990, 41, 43-47.

29. Musingo, M. N.; Sims, C. A.; Bates, R. P.; O'keefe, S. F.; Lamikanra, 0. Changes
in ellagic acid and other phenols in muscadine grape (Vitis rotundifolia) juices and
wines. Am. J Enol. Vitic. 2001, 52, 109-114.

30. Sims, C. A.; Bates, R. P. Effects of skin fermentation time on the phenols,
anthocyanins, ellagic acid sediment. Am. J Enol. Vitic. 1994, 45, 56-62.

31. Hukumoto, L. R.; Mazza, G. Assessing antioxidants are affected by colloidal
properties of phenolic compounds. J. Agric. Food Chem. 2000, 48, 3957-3604.

32. Goldy, R. G.; Ballinger, W. E.; Maness, E. P. Fruit anthocyanin content of some
Euvitis X Vitis rotundifolia hybrids. J. Amer. Soc. Hort. Sci. 1986, 111, 955-960.

33. Talcott, S. T.; Brenes, C. H.; Pires, D. M.; Del Pozo-Insfran, D. Phytochemical
stability and color retention of copigmented and processed muscadine grape juice.
J. Agric. Food Chem. 2003, 51, 957-963.

34. Del Pozo-Insfran, D.; Del Follo-Martinez, A.; Brenes, C. H.; Talcott, S. T.
Phytochemical stability of an ascorbic acid fortified and copigmented muscadine
grape juice as affected by high hydrostatic pressure processing. Submitted in J.
Agric. Food Chem. 2004.

35. Gomez, M. R. A.; Ledward, D. A. Effect of high pressure treatment on the activity
of some polyphenoloxidases. Food Chem. 1996. 56, 1-5.

36. Sun, N.; Lee, S.; Kim, B. S. Effect of high-pressure treatment on the molecular
properties of mushroom polyphenoloxidase. Lebens. Wiss. Technol. 2002, 35, 315-
318.

37. Zabetakis, I.; Leclerc, D.; Kajda, P. The effect of high hydrostatic pressure on the
strawberry anthocyanins. J. Agric. Food Chem. 2000, 40, 2749-2754.

38. Timberlake, C. F.; Bridle, P. Anthocyanins. In Developments in food colours 1.
Ed. Walford, J., Applied Science Publishers LTD., London, 1980; pp. 115-143.









39. Sims, C. A.; Morris, J. R. A comparison of the color components and color stability
of red wine from Noble and Cabernet Sauvignon at various pH levels. Am. J. Enol.
Vitic. 1985, 36, 181-184.

40. Swain, T.; Hillis, W. E. The phenolic constituents of Prumus domestic. I. The
quantitative analysis of phenolic constituents. J. Sci. FoodAgric. 1959, 10, 63-68.

41. Talcott, S. T.; Percival, S. S.; Pittet-Moore, J.; Celoria, C. Phytochemical
Composition and Antioxidant Stability of Fortified Yellow Passion Fruit
(Passiflora edulis). J. Agric. Food Chem. 2003, 51, 935 -941.

42. Ou, B.; Hampsch-Woodill, M.; Prior, R. L. Development and validation of an
improves oxygen radical absorbance capacity assay using fluorescein as the
fluorescent probe. J. Agric. Food Chem. 2001, 49, 4619-4626.

43. Cao, G.; Verdon, C. P.; Wu, A. H. B.; Wang, H.; Prior, R. L. Automated oxygen
radical absorbance capacity assay using the COBAS FARA II. Clin Chem. 1995, 41,
1738-1744.

44. SAS Institute, Inc., SAS Campus Drive, Cary, NC, 1996.

45. Zafrilla, P.; Ferreres, F.; Tomas-Barberan, F. A. Effect of processing and storage on
the antioxidant ellagic acid derivatives and flavonoids of red raspberry (Rubus
idaeus) Jams. J. Agric. Food Chem. 2001, 49, 3651-3655.

46. Rommel, A.; Wrolstad, R. E. Ellagic acid content of red raspberry juice as
influenced by cultivar, processing, and environmental factors. J. Agric. Food
Chem. 1993, 41, 1951-1960.

47. Nunan, K. J.; Sims, I. M.; Bacic, A., Robinson, S. P., Fincher G. B. Changes in cell
wall composition during ripening of grape berries. Plant Physiol. 1998, 118, 783-792.

48. Sleas, G. J.; Diamandis, E. P.; Goldberg, D. M. Resveratol; A molecule whose time
has come? And gone? Clin. Biochem. 1997, 30, 91-113.

49. Zoecklein, B. W.; Fugelsang, K. C.; Gump, B. H.; Nury, F. S. Phenolic compounds
and wine color. In Wine analysis and production. Chapmean & Hall, New York,
NY. 1994.

50. Harkness, E. M.; Vine, R. P.; Coign, M. J.; Heitz, J. R. The effect of different must
treatments on muscadine juice quantity and quality. Am. J. Enol. Vitic. Meeting
Rpt. 1987, 10.

51. Gazzani, G.; Papetti, A.; Massolini, G.; Daglia, M. Anti- and prooxidant activity of
water soluble components of some common diet vegetables and the effect of
thermal treatment. J. Agric. Food Chem. 1998, 46, 4118-4122.









52. Meyer, A. S.; Heinone, M.; Frankel, E. N. Antioxidant interactions of catehin,
cyanidin, caffeic acid, quercetin, and ellagic acid on human LDL oxidation. Food
Chem. 1998, 61, 71-75.

53. Adamson, G. E.; Lazarus, S. A.; Mitchell, A. E.; Prior, R. L.; Cao, G.; Jacobs, P.
H.; Kremers, B. G; Hammerstones, J. F.; Rucker, R. B.; Ritter, K. A.; Schmitz. H.
H. HPLC method for the quantification ofprocyanidins in cocoa and chocolate
samples and correlation to total antioxidant capacity. J. Agric. Food Chem. 1999,
47, 4184-4188.

54. McPhail, D. B.; Hartley, R. C.; Gardner, P. T.; Duthie, G. G. Kinetic and
stoichiometric assessment of the antioxidant activity of flavonoids by electron spin
resonance spectroscopy. J. Agric. Food Chem. 2003, 51, 1684-1690.

55. Scalbert, A.; Duval, L.; Peng, S.; Monties, B. Polyphenols of Quercus robur L. II.
Preparative isolation by low-pressure and high-pressure liquid chromatography of
heartwood ellagitannins. J. Chrom. 1990. 502, 107-119.

56. Lei, Z. Monomeric ellagitannins in oaks and sweetgum. Doctoral Dissertation,
2002, Virginia Polytechnic Institute and State University. Blacksburg, Virginia

57. Atoui, A. K.; Mansouri, A.; Boskou, G.; Kefalas, P. Tea and herbal infusions: Their
antioxidant activity and phenolic profile. In Press Food Chem. 2004

58. Montgomery, D. C. Response surface methods and other approaches to process
optimization. In Design and Analysis of Experiments. 5th Ed. John Wiley & Sons,
Inc. US. 2001.

59. Robbins, R. J. Phenolic acids in foods: An overview of analytical methodology. J.
Agric. Food Chem. 2003, 51, 2866-2887.

60. Gao, L.; Mazza, G. Rapid method of complete chemical characterization of simple
and acylated anthocyanins by high-performance liquid chromatography and
capillary gas-liquid chromatography. J. Agric. Food Chem. 1994, 42, 118-125.

61. Merken, H. M.; Merken, C. D.; Beecher, G. R. Kinetics method for the quantitation
of anthocyanidins, flavonols, and flavones in foods. J. Agric. Food Chem. 2001, 49,
2727-2732.

62. Wang, G.; Kuan, S. S.; Francis, 0. J.; Ware, G. M.; Carman, A.S. A simplified
HPLC method for the determination of phytoestrogens in soybean and its processed
products. J. Agric. Food Chem. 1990, 38, 185-190.

63. Kahkonen, M. P.; Heinonen, M. Antioxidant activity of anthocyanins and their
aglycons. J. Agric. Food Chem. 2003, 51, 628-623.

64. Wang, H.; Cao, G.; Pror, R. L. Oxygen radical absorbing capacity of anthocyanins.
J. Agric. Food Chem. 1997, 45, 304-309.




Full Text

PAGE 1

HYDROLYTIC AND ANTIOXIDANT PROPERTI ES OF ELLAGIC ACID AND ITS PRECURSORS PRESENT IN MUSCADINE GRAPE By JOONHEE LEE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Joonhee Lee

PAGE 3

iii ACKNOWLEDGMENTS It is a privilege to me to acknowledge the unconditional support of my supervisory committee chair, Dr. Stephen Talcott, who ha s been a teacher, mentor and friend during my years as a graduate student. His advice, wisdom, and constant help have made this dissertation possible. I sincerely appreciate the help offere d by the members of my supervisory committee, Dr. Sims, Dr. Marshall, and Dr. Sa ba. Their cooperation and suggestions have improved the quality of my dissertation considerably. Being a member of the Department of Food Science and Human Nutrition has been a wonderful academic experience and a unique op portunity to meet extraordinary people who helped me immensely along the way. I woul d like to give special deep thanks to current and past lab mates; David, Flor Youngmok, Chris, Lanier, Kristine, Stacy, Lisbeth, Janelle, Melanie, Jenni fer, Danielle, and Angela. My deepest recognition goes to my bel oved parents, who helped me in any imaginable way to achieve my objectives and fulfill my dreams. They have been an inexhaustible source of love a nd inspiration all my life. My most special thanks go to my husb and Jeongho for his pa tience, understanding, and encouragement, without which it woul d have been impossible to complete my degree. Finally, I thank to my specially bel oved Minsuh, hoping that the effort of these years may offer him a more plentiful life in the years to come.

PAGE 4

iv TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF TABLES............................................................................................................vii LIST OF FIGURES.........................................................................................................viii ABSTRACT....................................................................................................................... ..x CHAPTER 1 INTRODUCTION........................................................................................................1 Justification.................................................................................................................. .1 Objectives..................................................................................................................... 2 2 LITERATURE REVIEW.............................................................................................3 Muscadine Grapes........................................................................................................3 Phytochemicals in Muscadine Grapes..........................................................................4 Ellagic Acid and Its Precursors.............................................................................4 Ellagic Acid and Quality of Muscadine Products...............................................10 Anthocyanins.......................................................................................................12 3 FRUIT MATURITY AND JUICE EX TRACTION INFLUENCES ELLAGIC ACID DERIVATIVES AND OTHER ANTIOXIDANT POLYPHENOLICS IN MUSCADINE GRAPES............................................................................................14 Introduction.................................................................................................................14 Materials and Methods...............................................................................................15 Chemical Analyses..............................................................................................15 Statistical Analysis..............................................................................................17 Results and Discussion...............................................................................................17 Identification of Ellagic Acid and Its Precursors................................................17 Ellagic Acid and Its Derivatives..........................................................................18 Anthocyanins.......................................................................................................21 Total Phenolics and Antioxidant Capacity..........................................................24 Conclusion..................................................................................................................27

PAGE 5

v 4 IDENTIFICATION OF ELLAGITANNINS AND CONJUGATES OF ELLAGIC ACID IN MUSCADINE GRAPES............................................................................28 Introduction.................................................................................................................28 Materials and Methods...............................................................................................29 Isolations..............................................................................................................29 Analysis by HPLC-PDA......................................................................................30 Analysis by HPLC-MSn......................................................................................31 Total Polyphenols and Antioxidant Capacity......................................................31 Statistical Analysis..............................................................................................32 Results and Discussion...............................................................................................32 Isolations and Quantific ation of Polyphenolic Compounds by Solid Phase Extraction.........................................................................................................32 Identifications of Ellagic Acid De rivatives and Flavonoids by HPLC-PDA and MSn............................................................................................................36 Ellagic acid derivatives in Isolate I..............................................................37 Ellagic acid derivatives and fla vonoid glycosides in isolate II....................41 Conclusions.................................................................................................................45 5 HYDROLYTIC PROPERTIES OF ELLAGIC ACID DERIVATIVES IN MUSCADINE GRAPES............................................................................................46 Introduction.................................................................................................................46 Materials and Methods...............................................................................................47 Response Surface Methodology (RSM) and Statistical Analyses.......................47 Enzymes Preparation...........................................................................................47 Chemical Analysis...............................................................................................48 Results and Discussion.. ..............................................................................................49 Ellagic Acid Derivatives and Flavono id Glycosides: Effects of Time, Temperature and pHs.......................................................................................49 Antioxidant Capacity of Polyphenolics as Affected by Aglycone vs Glycosides with Hydrolysis.............................................................................53 Enzymatic Hydrolysis of Ellagic Acid Derivatives.............................................61 Conclusions.................................................................................................................64 6 HYDROLYTIC AND OXIDATIVE PROPERTIES ON ELLAGIC ACID DERIVATIVES DURING STORAGE OF MUSCADINE GRAPES JUICES.........65 Introduction.................................................................................................................65 Materials and Methods...............................................................................................66 Storage of Red Juices and Isolations...................................................................66 Storage of White Juices and Isolations................................................................66 Chemical Analysis...............................................................................................67 Results and Discussion...............................................................................................67 Changes of Ellagic Acid Derivati ves during Storage of Red Juice.....................67 Initial Ellagic Acid Derivatives in White Muscadine Juice as Affected by Ascorbic Acid and Air.....................................................................................69

PAGE 6

vi Post-storage Levels of Ellagic Acid Derivatives in White Muscadine Juice as Affected by Ascorbic Acid and Air.................................................................76 Conclusions.................................................................................................................81 7 SUMMARY AND CONCLUSION...........................................................................83 LIST OF REFERENCES...................................................................................................85 BIOGRAPHICAL SKETCH.............................................................................................92

PAGE 7

vii LIST OF TABLES Table page 3-1 The concentrations of free ellagic aci d, two ellagic acid glycosides and total ellagic acids of muscadine grapes as a ffected by cultivars and ripening stages......19 3-2 Concentration of six anthocyanidins a nd total anthocyanidins of red muscadine grapes as affected by ripening stages.......................................................................22 3-3 Concentrations of total so luble phenolics in methanolic and ethyl acetate extracts as affected by cultivar s and ripening stages ............................................................25 3-4 Antioxidant capacity of methanolic a nd ethyl acetate extracts as affected by cultivars and ripening stages ...................................................................................26 4-1 The concentrations of free ellagic acid, ellagic acid glycosides and total ellagic acids on each fraction from three different cultivars................................................33 4-2 UVmax and HPLC-ESI(-)-MSn analyses of polyphenol s in isolate I from muscadine grapes.....................................................................................................38 4-3 UVmax and HPLC-ESI(+)/(-)-MSn analyses of ellagita nnins, glycosides of ellagic acid and flavonoids in is olate II from muscadine grapes.............................44 5-1 Pearson correlations coefficients of indi vidual ellagic acid de rivatives contents with total soluble phenolics and antioxidant capacity..............................................60 5-2 Concentrations of ellagic acid, ellagic acid glycosides and gallic acid affected by enzyme treatment in two different fractions from Doreen extracts.....................63

PAGE 8

viii LIST OF FIGURES Figure page 2-1 Ellagic acid chemical structure..................................................................................5 2-2 Ellagic acid glycosid es found in raspberry................................................................6 2-3 Ellagitannins (Lambertianin C) found in raspberry fruits..........................................7 2-4 Ellagitannins conversion to ellagic acid via hexahydroxydiphenic acid (HHDP).....7 2-5 Chemical structures of punical agin found in pomegranate juice...............................8 2-6 UV spectra of pomegranate juice characteristic compounds.....................................8 2-7 Characteristic UV spectra of hydrol ysable tannins in birch leaves............................9 2-8 Chemical structures of Ge raniin and Amariin, containing dehydrohexahydroxydiphenyl (DHHDP) group........................................................9 4-1 Fraction scheme and tentative clas sification of polyphenolics present in methanolic extracts of muscadine grapes.................................................................29 4-2 Total soluble phenolics and antioxidant capacities of five fractions from three different cultivars.....................................................................................................36 4-3 HPLC-PDA chromatogram (280 nm) of Isolate I of muscadine grapes..................37 4-4 UV spectra of ellagic acid derivatives in Isolate I....................................................40 4-5 HPLC-PDA chromatogram (280 and 360 nm) of Isolate II of muscadine grapes...42 5-1 Hydrolysis time and te mperature combinations included in central composite design (CCD) experiment.........................................................................................48 5-2 Tridimensional representation of free ella gic acid generated using fold increases by response surface model in the ab sence and presence of 2N HCl........................50 5-3 Tridimensional representation of ella gic acid xyloside generated using fold increases by response surface model in the absence and presence of 2N HCl........51

PAGE 9

ix 5-4 Tridimensional representation of ellagi c acid rhamnoside generated using fold increases by response surface model in the absence and presence of 2N HCl........52 5-5 Tridimensional representation of myri cetin rhamnoside generated using fold increases by response surface model in the absence and presence of 2N HCl........54 5-6 Tridimensional representation of quer cetin rhamnoside generated using fold increases by response surface model in the absence and presence of 2N HCl........55 5-7 Tridimensional representation of kaem pferol rhamnoside generated using fold increases by response surface model in the absence and presence of 2N HCl........56 5-8 Tridimensional representation of tota l soluble phenolics generated using fold increases by response surface model in the absence and presence of 2N HCl........58 5-9 Tridimensional representation of anti oxidant capacity generated using fold increases by response surface model in the absence and presence of 2N HCl.......59 6-1 Changes in whole red juice for el lagic acid derivatives during storage...................70 6-2 Changes in water isolate for ella gic acid derivatives during storage.......................71 6-3 Changes in ethyl acetat e isolate for ellagic acid derivatives during storage............72 6-4 Changes in methanol isolate for e llagic acid derivatives during storage.................73 6-5 Concentrations of ellagic acid de rivatives depending on treatments of ascorbic acid and air at 0day of Do reen juice as affected by thermal pasteurization method..............................................................................................75 6-6 Concentrations of free ellagic acid depending on ascorbic acid and air during storage of Doreen juice as affected by thermal pasteurization.................................77 6-7 Concentrations of ellagic acid xylosi de depending on ascorbic acid and air during storage of Doreen juice as affected by thermal pasteurization.....................78 6-8 Concentrations of ellagic acid rham noside depending on ascorbic acid and air during storage of Doreen juice as affected by thermal pasteurization.....................79 6-9 Concentrations of total ellagic acid depending on ascorbic acid and air during storage of Doreen juice as affected by thermal pasteurization.................................80

PAGE 10

x Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy HYDROLYTIC AND ANTIOXIDANT PROPERTI ES OF ELLAGIC ACID AND ITS PRECURSORS PRESENT IN MUSCADINE GRAPE By Joonhee Lee December 2004 Chair: Stephen T. Talcott Major Department: Food Science and Human Nutrition Muscadine grape ( Vitis rotundifolia ) has potential health-p romoting benefits and antioxidant properties from ch aracteristic phytochemicals such as ellagic acid, ellagic acid glycosides, and ellagitannins. To provi de fundamental information on ellagic acid derivatives in this commodity, typical pol yphenolic compounds were characterized in eight cultivars. All polypheno lics generally increased as fr uit ripened and the highest concentrations were located in the skins. Ho t-pressed juices contai ned considerably lower polyphenolic concentrations than were presen t in whole grapes w ith actual recovery varying widely among cultivars. Antioxidant capacity was appreciably influenced by cultivar, maturity, and location in the fru it with good correlations to soluble phenolics found in both methanolic and ethyl acetat e extracts (r=0.83 and 0.92, respectively). Glycosidic forms of ellagic acid and major fl avonoids were isolated by a series of solidphase extractions. By using HPLC-MS/PDA, ch emical identities were elucidated as ellagic acid xyloside, ellagic acid rham noside, myricetin rhamnoside, quercetin

PAGE 11

xi rhamnoside and kaempferol rhamnoside. Ella gitannins, a major source for hydrolytic free ellagic acid, were present in the C18 non re tained fraction and their molecular weights having never been reported were determined. Because of the lack of data on hydrolytic properties of ellagic acid precursors, central composite design was applied to dem onstrate the evolution of hydrolytic ellagic acid depending on various time-temperature combinations in response surface methodology. Hydrolysis of ellagic acid pr ecursors would be more temperaturedependent than time-dependent, and the result ant additional free ellagic acid showed good correlation to antioxidant capa city (r=0.83). Using enzymes, -glucosidase (E.C. 3.2.1.21) or tannase (E.C. 3.1.1.20) is an alternative application to hydrolyze ellagic acid precursors, -glucosidase showed better reactivity on ellagic acid derivatives compare to tannase. During storage, ellagic acid glycosides we re relatively stable with ellagitannin hydrolysis the major source for evolution of free ellagic acid in a juice system. Free ellagic acid is partially responsible for the fo rmation of insoluble sediments in addition to other juice constituents such as metal ions or insoluble pectins. Ascorbic acid fortification and air sparging did not significan tly affect concentrations of ellagic acid derivatives in stored juices. However, a mild heat treatment influenced el lagitannin hydrolysis and was likely associated with activation of na tural enzymes present in the juice. Results provided by these studies suggest beneficial reasons to consume muscadine grapes leading to improve the market value of this crop not only as fresh grapes, juice or wine but also as health promoting compone nts in various types of processed food.

PAGE 12

1 CHAPTER 1 INTRODUCTION Justification Muscadine grapes ( Vitis rotundifolia ) are commonly cultivated in the southeastern U.S. as an excellent alternativ e fruit crop, because many traditional Vitis species are impossible to survive due to Pierces disease a nd typical climatic characteristics in this region. Muscadine grapes are favorably consumed as fresh fruit, juice, wine or jelly not only for distinguished aroma and flavor char acteristics but also for positive health benefits from characteristic phytochemicals in cluding ellagic acid/ella gic acid derivatives and anthocyanins 3,5-diglucosides. Presence of these compounds has been associated with quality defects such as rapid color deterioration of an thocyanin 3,5-diglucosides and the formation of insoluble sediments that ma y be affected by ellagic acid derivatives during storage. To solve the problem with sediment formation, research has focused on removing free ellagic acid from the matr ix using various processing techniques ( 1 2 ). However, none of the trials successfully solved the sedimentation problem due to lack of fundamental information on these compounds in muscadine juice or wine. Therefore, the primary objective of this work is to selec tively identify and determine those compounds that impact levels of free ellagic acid in efforts to determine their concentration and stability in fresh and processed muscadine grapes. Mechanisms fo r their hydrolysis and relative changes are evaluated that may rev eal key components affecting their radical scavenging properties and the accumulation of free ellagic acid during storage of muscadine juice. These studies also seek to determine the antioxidant polyphenolics in

PAGE 13

2 muscadine grapes as influenced by their locati on in the grape, those extracted into juice, and solubility in various fractionations as a function of cultivar and maturity. These evaluations are beneficial for exploring biochemical synthesis and/or fate of pro-ellagic acid compounds. Finally, chemical and physic al characteristics of these identified compounds will lead to possible mechanism(s) concerning their hydrolysis and release of free ellagic acid into insoluble sediments. This information will add market value and significantly increase marketab ility for this under-utilized grape by identification of novel compounds and evaluation of their antioxidant potential. Objectives Objective 1: To quantify the antioxidant polyphenolics in muscadine grapes as influenced by their location in the grape, juice production, and polyphenolic fractionation as a function of cultivar and maturity. Objective 2: To isolate and identify ellagi c acid glycosides an d ellagitannins using advanced HPLC methodologies (PDA, MS) a nd evaluate their antioxidant capacity. Objective 3: To investigat e the hydrolytic and oxidativ e properties of individual ellagic acid derivatives and other polyphenolics affe cting concentrations and antioxidant capacity. Objective 4: To determine the major s ource of hydrolytic ellagic acid, and to elucidate the effects of ascorbic acid fo rtification and oxidation on changes of ellagic acid derivatives in juice system.

PAGE 14

3 CHAPTER 2 LITERATURE REVIEW Muscadine Grapes Muscadine grapes are botanically categorized under the genus Vitis which consists of two subgenera: Euvitis (bunch grapes) and Muscadinia (berry grapes) ( 3 ). Euvitis includes diverse specie s of grapes including V. vinifera which is the most common grape in the world. Muscadinia can be represented as an American grape species V. rotundifolia Michaux, muscadine grape, and V. munsoniana Simpson. When muscadine grapes are grown for commercial harves t, the majority of plantings are V. rotundifolia ( 4 ) and they are indigenously found in the south eastern United States, from Delaware to central Florida and along the Gulf of Mexico to eastern Texas ( 3 ). In these regions, muscadine grapes are favorably cultivated fruit crops since other grapes are hard to survive in typical climate like humid summers and warm winters ( 5 6 ). Another benefit of growing muscadine grapes is that V. rotundifolia has excellent resistance to pests and various diseases, namely Pierces disease, wh ich is a lethal disease of the grape vine caused by the bacterium Xylella fastidiosa and the main obstacle for production of V. vinifera in the south and southeastern states ( 3 ). Unlike bunch grapes, muscadine grape fruits develop in round clusters, containi ng approximately 4 grapes. The fruit are round or oval, larger varieties may reach 26 mm in diameter and weigh from 3-10 g each. Fruits can range from light-skinned (green, pe arly white, or bronze) to dark red in color and most have a tough, leathery skin with the pulp inside ranging from meaty, to melting, or juicy ( 6 ) At present, home gardening or commerc ial production is provided by over 70

PAGE 15

4 different cultivars, the oldest of which includes Scuppernong (green ), Carlos (green), Doreen (green), Noble (red), and Magnolia (green). These cultivars are consumed as fresh fruit, juice, wine, or jelly ( 7 ). In the production of either fermented or unfermented muscadine grape products, the physicochemical composition of fruit is one of the most important factors affecting overall quality a nd acceptability. In direct comparisons with V. vinifera muscadine grapes generally have a lo wer soluble solids concentration (SSC) and titratable acidity ( 8 ) ranging from 10-18% (13.2% on av erage) and 0.39-1.5% (0.84% on average) expressed as tartra te equivalents, respectively; resulting in a pH range of 2.93.4 (pH 3.14 on average). Muscadine hulls (skin plus firmly attached tissue layer) were found to contain significantly higher organic acid concentra tions than the pulp on a per weight basis. This helps to explain how jui ces that are immediately pressed have lower titratable acidity values compar ed to those for hot-pressed juices or skin fermentation ( 9 10 ). Phytochemicals in Muscadine Grapes Ellagic Acid and Its Precursors Ellagic acid (Figure 2-1) ha s been found in many woody plants and in diverse fruits and nuts in various concentrations. Intere stingly ellagic acid is present only in Vitis rotundifolia but not in Vitis vinifera Ellagic acid is believed to be formed by the hydrolytic release from ellagic acid derivatives including ella gic acid glycosides (Figure 2-2, A-C) and ellagitannins (Figure 2-3). The presence of ellagic acid in various fruits and nuts was determined for the purpose of botan ical classification, a nd was identified in strawberries, blackberries, and walnuts ( 11 ). Concentrations of e llagic acid in various fruits and nuts were determined as total el lagic acid, which was hydr olytic ellagic acid from ellagic acid precursors followed by complete hydrolysis with acid ( 12 ). This was

PAGE 16

5 done because analyses on individual ellagic acid derivatives were difficult with lower analytical techniques and in the absence of authentic standards. Appreciable concentrations of ellagic acid were detected after acid hydrolysis in strawberries (630 g/g), raspberries (1500 g/g), blackberries (1500 g/g), walnuts (590 g/g), pecans (330 g/g), and cranberries (120 g/g). In strawb erries, most ellagic acid (95.7%) was found in pulp, and the remainder was found in seeds. Ho wever, the seeds of raspberries contained 87.5% ellagic acid, while the pulp had 12.2%. Amakura et al. ( 13 ) determined ellagic acid levels in fresh and processed fruits by a simple and rapid high performance liquid chromatographic (HPLC) method that did not involve a hydrolysis step. Values for similar commodities (blackberry, 87.6 g/g; strawberry, 17.7 g/g; and raspberry, 5.84 g/g) were appreciably lower without th e hydrolysis step; pos sibly indicating the presence of ellagic acid precursors including el lagic acid glycosides and ellagitannins. Ellagic acid glycosides are forms of a single sugar moiety linked to the hydroxyl at the 4-position of ellagi c acid, and five or six different ellagic acid gl ycosides have been separated in raspberry ( 14 15 ): ellagic acid 4-acetylxyloside, ellagic acid 4-arabinoside, and ellagic acid 4-acetylarabinoside (Figure 2-2;). Muscadine grapes also contain two ellagic acid glycosides, which are characteri zed by 3-7 nm hypsochromic shifts in UV spectra (252 and 360 nm) ( 16 ). However, free ellagic acid and ellagic acid glycosides OH O H O O O H OH O O Figure 2-1. Ellagic acid chemical structure

PAGE 17

6 contribute to only part of th e total ellagic acid on acid hydr olysis; highly indicating the presence of ellagitannins in muscadine grapes. Ellagitannins (Figure 2-3) are characterized as hydrolysable conjugates containing one or more hexahydroxydiphenoyl (HHDP) groups esterified to a sugar, mainly glucose. HHDP groups are released from the main structure leading to spontaneous conve rsion into ellagic aci d by hydrolysis (Figure 2-4). Specific information on ellagic acid deri vatives are lacking fo r muscadine grapes, but the presence of ellagic acid and its deri vatives in muscadine grapes may add value and marketability to the crop due to possible health benefits such as its antioxidant activity ( 17, 18 ), anti-carcinogenic properties influenc ing cell cycle arre st and apoptosis ( 19 ), and inhibition of tumor formation and growth in mammalian models ( 20 21 ). The O O OH OH O O O H O O O OH O H COCH3 O O OH OH O O O H O O O OH O H COCH3 O O OH OH O O O H OOHO OH O H A B C Figure 2-2. Ellagic acid glycosid es found in raspberry. A) Ella gic acid 4-arabinoside, B) Ellagic acid 4-acetylarabinoside, C) Ella gic acid 4-acetylxyloside. (Mullen et. al. Phytochemistry 2003 64, 617-624)

PAGE 18

7 types of ellagitannins are varied with a pproximately 500 differen t compounds isolated and identified in nature ( 22 ). Structural diversity in ella gitannins originates from the number of HHDP units, the location of gall oyl ester groups partic ipating in biaryl linkage, and the conformati on of the glucose ring ( 23 ). HPLC assisted by mass spectrometry and diode array detections ar e the most commonly employed to separate and identify various ellagitannins from fr uits or plants extracts. Mullen et. al. ( 14 ) Figure 2-3. Ellagitannins (Lambertianin C) found in raspberry fruits. (Mullen et. al. Phytochemistry 2003 64 617-624) Figure 2-4. Ellagitannins conversion to el lagic acid via hexa hydroxydiphenic acid (HHDP). (Quideau and Feldman, Chem. Rev. 1996 96 475-503)

PAGE 19

8 reported phytochemical profiles in ras pberry including anthocyanins, quercetin conjugates, ellagic acid glyc osides and 3 different ellagitannins, Sanguiin H-10 (3 HHDPs, mass: 1568), Lambertianin C (6 HHDP s, mass: 2804) and Sanguiin H-6 (4 HHDPs, mass: 1870). All of these had observed maximum UV absorbance at 250 nm, but it could be lower than 250 nm since they scanned from 250 to 700 nm. Punicalagins (Figure 2-5) are different type s of ellagitannins found in fr uits, especially pomegranate ( 24 25 ) and composed of glucose, HHDP and ga llagyl acid (ellagic + 2 gallic), which Figure 2-5. Chemical structures of punical agin found in pomegranate juice as main ellagitannins (Gill et. al. J. Agric. Food Chem 2000 48 4581-4589). Figure 2-6. UV spectra of pomegranat e juice characteristic compounds. A; galloylglucose, C; hydrolyzable tannins, D; punicalagin, G; ellagic acid (Gill et. al. J. Agric. Food Chem 2000 48 4581-4589).

PAGE 20

9 had a resulting UV spectrum showing maxi ma around 375 and 265 nm (Figure 2-6, D). Unidentified hydrolysable tannins (ellagic acid + gallic acid + tertgallic acid + etc.) were also observed with characteristic UV sp ectra, maximum at 266 nm (Figure 2-6, C). Similar UV spectra were reported with birch ( Betula pubescens ) leaves for bis-HHDPglucopyranose isomers ( 26, Figure 2-7). They also observed that UV spectra of hydrolysable tannins were very similar to ga llic acid spectra with 3-7 nm hypsochromic Figure 2-7. Characteristic UV spectra of hydrolysable tannins in birch ( Betula pubescens ) leaves. 19:pentagalloylglucopyranose isomers, 6:bis-HHDP-glucopyranose isomers, 17:galloyl-bis-HHDP-glucopyr anose isomers, 10:digalloyl-HHDPglucopyranose isomers, 18:trigalloyl-HHDP-glucopyranose.(Gill et. al. J. Agric. Food Chem 2000 48, 4581-4589). Figure 2-8. Chemical structures of Geraniin and Amariin, containing dehydrohexahydroxydiphenyl (DHHDP) group, found in Phyllanthus amarus (Foo. Phytochemistry 1995 39 217-224).

PAGE 21

10 shifts depending on the number of galloyl gr oups. In some plants, ellagitannins are found in oxidized form as containing dehyd rohexahydroxydi-pheny l (DHHDP) group, and geraniin and amariin have been identified by NMR data ( 27 ). The released DHHDP units may not be able to convert to ellagic acid because two linked galloyl molecules in HHDP will be inhibited from making a lactone ring by additional linkage (Figure 2-8). Ellagic Acid and Quality of Muscadine Products The presence of ellagic acid in muscadine grape and its products are important not only because of its potential heal th benefits but also because of their possible contribution to form insoluble materials in processed jui ce and wine. Ellagic acid, hardly soluble in water, results in a significant defective role in the quality percepti on of wines and juices. In general, V. vinifera is not known to contain ellagic acid in its seed, skin or pulp, but small amounts are usually detected followi ng oak barrel storage and aging. Boyle and Hsu ( 28 ) evaluated ellagic acid conc entrations in juices from 11 cultivars of muscadine grapes and found a range from 1.6-23 g/mL, with concentrations influenced by skin color. Ellagic acid evaluations in muscadine have been frequently conducted with its products like juice or wine with regard to sediment formation during storage. Boyle and Hsu ( 28 ) reported that ellagic acid is the onl y compound detectable in sediment, present as yellowish to red crystals; however, recen t quantitative analysis of the collected sediments revealed that no more than 12% free ellagic acid by weight was actually present in the sediments. The remaining constituents consist of either unidentified compounds or conjugated forms of ellagic acid ( 16 ). Many efforts to prevent ellagic acid sedimentation in muscadine juices and wine s have been employed, including chemical and physical remediation procedures, but none these were successf ul in solving the sedimentation problem. Lin and Vine ( 1 ) treated Magnolia (a white cultivar) juice with

PAGE 22

11 increasing concentrations of fining agents such as polyvinylpolypyrrolidone (PVPP) and gelatin, and found that the highe st concentration of PVPP ( 1.08 g/L) was most effective in reducing ellagic acid. Gelatin (0.4 g/L) also decreased el lagic acid con centration in sediment by 56% in red muscadine juice ( 2 ). An ultrafiltration technique involving passing juice through a 10,000-30,000 dalton molecu lar weight membrane demonstrated at most a 50% reduction of e llagic acid in sediment ( 2 ). However, these chemical and physical remediation procedures only lowered the levels of ellagic acid or sediment and ellagic acid precipitation continued during fu rther storage suggesting that ellagic cid could be hydrolyzed from larger molecules li ke ellagic acid glycos ides or ellagitannins ( 16 29 ). According to the wo rk of Garrido et. al. ( 2 ), the formation of ellagic acid sediments in white muscadine juice was acceler ated by increased storage temperature and following thermal pasteurization (100C for 10 min), which resulted in more sediment than sterile filtered juices after 8 mont hs storage at 1.5C. Sims and Bates ( 30 ) investigated the effect of sk in fermentation time on ellagic acid sedimentation of Noble muscadine grape wine. Wines fermented with skins for four and six days had greater amounts of ellagic acid sediment than nonskin and 2-day skin fermented wines. Muscadine juice is normally manufactured by two extraction tec hniques depending on the final intentions for use. Hot-pressed ju ices with red cultivars are made following crushing and heating at 70C prior to pre ssing, while cold-pressed juices are pressed immediately after crushing white grapes ( 8 ). The main purpose of the hot-pressing procedure is to increase juice yield and to improve overall juice quality including high intensity of color through ex tracting more phytochemicals fr om fruits. Hot-pressing is

PAGE 23

12 likely to influence the extraction of ellagitannins or ellagic acid glycosid es resulting in an increase of ellagic acid precipitation in red muscadine juice after 50 days storage ( 16 ). Anthocyanins Other than ellagic acid derivatives, muscad ine grapes also contain distinguishable phytochemicals, anthocyanin 3,5-di glucosides, which have been identified as delphinidin, cyanidin, petunidin, peonidin, and malvidin in non-acylated forms ( 31 32 ). Even though anthocyanin stability is influenced by several factors during food processing, the presence of anthocyanin 3,5-diglucosides is the main reason for rapid color loss during storage of muscadine juice or wine due to low stability of diglucosides form compared to corresponding monoglucosides forms of anthocyanins ( 11 ). It has been found that color loss by oxidation of anthocyanins was correla ted to a decrease in radical scavenging activity ( 17 ). In order to protect the degrada tion of color in muscadine products, alternative processing schemes have been re cently employed to understand the chemical nature of 3,5-diglucoside ant hocyanins and to consequently lead the economic growth of this crop ( 33 34 ). Usually anthocyanins are stabil ized and develop intense color by chelation with metal ions or binding with colorless polyphenolics and this is known as copigmentation reactions. Associated w ith muscadine grape products, copigmentation can be an alternative strategy to improve qua lity and market value since it has been reported that incremental addition of rose mary extract (00.4% v/v) affect the hyperchromic shift of anthocyanins corres ponding to increased antioxidant activity through copigment complexes with anthocyanins ( 33 ). Diverse technological improvements have been employed to replace th e heating process because heat is a prime source for quality loss during food process. Hi gh hydrostatic pressure (HHP), a promising alternative to traditional pa steurization technologies ( 35 37 ), has been employed for

PAGE 24

13 muscadine grape products in an effort to preserve thermolabile phytonutrients and favorable copigmentation between anthocyani ns and plant based polyphenolics such as rosemary and thyme extracts ( 33 3 4 ). However, HHP can hinder improving juice quality by the presence and/or activation of residua l enzymes such as polyphenol oxidase (PPO) due to accelerated oxidation of anthocyanins by activated PPO.

PAGE 25

14 CHAPTER 3 FRUIT MATURITY AND JU ICE EXTRACTION INFLUE NCES ELLAGIC ACID DERIVATIVES AND OTHER ANTIOXIDANT POLYPHENOLICS IN MUSCADINE GRAPES Introduction Depending on maturity and availability, it is common to blend grape cultivars for muscadine wine and juice production to obtai n the most desirable acidity, color, and flavor. However, little info rmation is available on the phytochemical and antioxidant characteristics among cultivars suitabl e for wine or juice production. The phytochemistry of muscadine grapes is distinguishable from most other grape varieties due to its predomin ance of anthocyanin 3,5-diglucos ides and presence of ellagic acid and ellagic ac id precursors ( 7 ). The anthocyanins 3,5-diglucosides, which may be more resistant to degradation during ther mal processing compared to monoglucosides, are typically unstable during storage due to a decreased ability to form polymeric pigments and are particularly prone to oxidation and brow ning reactions ( 38 39 ). The ellagic acid derivatives are the most distingu ishing chemical attribute in muscadine grape since these components have not been found in any Vitis species. Associated with quality of muscadine juice or wine, ellagic acid ha s been considered as an undesirable element even though it has potential h ealth benefits because ellagi c acid and its precursors are believed to develop insoluble sediments during storage. Phen olic contents in different muscadine cultivars have been reported on only free ellagic acid, resveratrol and other flavonoids including myricetin, quercetin and kaempferol ( 17 ); however, the current study represents impacts on free ellagic acid as well as ellagic acid glycosides and total

PAGE 26

15 ellagic acid released by all e llagic acid precursors. The objec tives of this study were to quantify the antioxidant polyphenolics in mu scadine grapes as influenced by their location in the grape, jui ce production, and polyphenolic fr actionation as a function of cultivar and maturity. This information can be used to determine wine or juice blending schemes to produce higher quality muscadin e grape products in terms of phytochemical composition and antioxidant potential. Materials and Methods Muscadine grapes were donated from local grape growers in central Florida and collected at two maturity stages from the same vines at different time intervals, about 1520 days apart depending on variety. Varieties in cluded Carlos, Fry, and Doreen, classified as either white or more specifically bronze colored fruit, and the red-skinned varieties Noble, Albemarle, Cowart, Nesbitt, and Ge orgia Red. Random samplings of 8-15 fruit in duplicate were manually divided between sk in and pulp, while whole grapes were processed into juice using a hot-break tec hnique (70C for 30 min). Polyphenolics were extracted from the skin and pulp by hom ogenizing with 25 mL of 100% methanol, filtered through Whatman #4 filter paper, and solvent removed at 40C under a stream of nitrogen. The juice was analyzed directly following centrifugation and filtration. Nonanthocyanin polyphenolics were subsequently partitioned from each isolate into ethyl acetate in three sequential extractions afte r which the solvents were pooled, removed under reduced pressure at 40C, and re sidues redissolved in 50% methanol. Chemical Analyses Polyphenolics were separated and quantifie d by HPLC using solvent programs to identify phenolic acids, free ellagic acid, and ellagic acid derivati ves in ethyl acetate extracts, and total ellagic acid and individua l anthocyanidins in methanolic extracts

PAGE 27

16 following acid hydrolysis (2N HCl for 60 min at 95C). Separations were conducted on a Dionex HPLC system using a PDA-100 photodi ode array detector and a 250 mm 4.6 mm Acclaim 120 C18 column (Dionex, Sunnyvale, CA) with a C18 guard column. Mobile phases consisted of 100% water (phase A) and 60% methanol (phase B) both adjusted to pH 2.4 with o -phosphoric acid and run at 1 mL/min according to modified conditions of Lee and Talcott ( 16 ). Free ellagic acid, el lagic acid glycosides and phenolic acids were separated using a gradient el ution program where phase B changed from 0-30% in 3 min; 30-50% in 5 min; 50-70% in 17 min; 70-80% in 5 min; 80-100% in 5 min; and 100% in 9 min for a total run time of 44 min, after which the column was equilibrated to original conditions in 1 min for the next sample inject ion. Anthocyanidins and total ellagic acid were also separated with a gradient program that ran phase B from 30-50% in 3 min; 5070% in 2 min; 70-90% in 5 min; and 90-100% in 10 min and retu rning to original composition in 1 min for column equilibrati on. Ellagic acid and its derivatives were quantified in ellagic acid equi valents, flavonoid glycosides in equivalents of myricetin (Sigma Chemical, St. Louis, MO), and each anthocyanidin quantified in cyanidin equivalents (Polyphenols Laborat ories AS, Sandnes, Norway). Total soluble phenolics were analyz ed using Folin-Ciocalteu assay ( 40 ) and expressed in gallic ac id equivalents (GAE). Antioxidant activity ( 41 ) was determined using the oxygen radical absorbance capacity (ORAC) assay with fluorescein as modified by Ou et. al. ( 42 ) from initial protocol by Cao et. al. ( 43 ). Fluorescence loss by reaction with hydroxyl radical (70 min, 37 C) was monitored on a Molecular Devices fmax 96well fluorescent microplate reader (Sunnyva le, CA) following appropriate dilution of

PAGE 28

17 each isolate and data expressed in Trolox equi valents per g of fresh fruit or per mL of juice. Statistical Analysis Data represent the mean of duplicate analyses with analysis of variance and Pearson correlations conducted using JMP5 software ( 44 ); mean separation was conducted using the LSD test ( P <0.05). Results and Discussion Identification of Ellagic Acid and Its Precursors The free (aglycone) form of ellagic acid a nd two ellagic acid gl ycosides were found in all eight muscadine grape cultivars following ethyl acetate extraction and separation by HPLC. The ellagic acid glycosides were te ntatively characterized based on UV spectral properties (252 and 360 nm) similar to that of free ellagic acid (252 and 365 nm) as was previously characterized in muscadine grapes ( 16 ) indicating that these compounds were most likely glycosidic forms at the 4-positi on of ellagic acid rath er than HHDP moieties esterified to glucose (true ellagitannins), with maximum absorption at or near 250 nm ( 15 45 ). Preliminary work to characterize thes e compounds has identified the presence of glucose, xylose, or rhamnose moieties (data not shown). Similar ellagic acid glycosides were thought to exist in raspbe rries and were characterized by spectroscopic shifts (4-7 nm hypsochromic) and disappearance of th e glycoside after hydrolysis, with a corresponding increase in free ellagic acid ( 15, 4547 ). Similarly, the two ellagic acid glycosides identified in muscadine grapes yielded free ellagic acid upon both acid and enzyme (-glucosidase) hydrolysis. True ella gitannins, containing esterified HHDP units to a carbohydrate, were also believed to be present in the grape isolates, but were not separated or detected in muscadine gr apes using the HPLC methodology employed.

PAGE 29

18 Evidence of these highly polar compounds wa s established indirectly by passing an aqueous grape extract through a pre-conditioned Waters C18 Sep Pak cartridge and evaluating the non-retained fraction. No peaks analogous to ellagic acid were present in this isolate in the range of 200-400 nm, but following acid hydrolysis free ellagic acid was one of the hydrolytic products, thus provi ding evidence for their existence. Total ellagic acid was subsequently determined from the methanolic extracts following acid hydrolysis and represented the sum of free e llagic acid and ellagic acid released from both ellagitannins and ellagic acid glycosides. Ellagic Acid and Its Derivatives Concentrations of ellagic acid and its de rivatives in muscadine grapes were found to significantly vary with ripening, in skin and pulp ti ssue, among cultivars, and following juice extraction (Table 3-1). Ripe ning was a critical factor influencing concentrations since appreciable increases in skin and juice during ripening were observed. Since muscadine grapes grow in cl usters rather than bunches, inconsistent maturity at harvest is a common occurrence. Changes with ripening were also highly variable among cultivars for free ellagic acid and its glycosidic forms, and ranged from a 0.3 to 13-fold increase in the skins alone. Di fferences during ripeni ng were less variable for total ellagic acid at a 1.7fold average increase in the skins. The large increases in ellagic acid and its glycosides observed dur ing ripening may have resulted from various reasons: amplified hydrolysable tannins synthesis during veraison ( 8 ); a chemoprotective response similar to the fo rmation of resveratrol ( 48 ); or accelerated hydrolysis of HHDP units from ellagitannins that was observed to produce greater quantities of free ellagic acid in each cultivar. Compared to total ellagic acid, relatively low levels of free ellagic acid and ellagic acid glycos ides were present in the grapes, an indication that

PAGE 30

19Table 3-1. The concentrations (mg/kg, mg/L) of free ellagic acid (EA), two ellagic acid glycosides (EAG 1 and 2) and total ella gic acids on skin, pulp and juice of muscadine gr apes as affected by cultivars and ripeni ng stages (U: unripe and R: ripe). Free EA EAG11 EAG22 Total EA3 Cultivars Color U R U R U R U R Carlos White 32.1 b4 8.04 e* 17.4 b 6.76 d* 16.7 d 20.1 d 368 d 879 d* Fry White 31.3 b 87.4 cd* 13.0 cd 90.3 a* 8.81 e 13.6 d 531cd 879 d* Doreen White 10.8 d 138 ab* 3.78 f 93.0 a* 29.7 b 115 a* 918 ab 1620 b* Noble Red 17.5 c 76.4 d* 10.5 de 23.2 c* 31.0 b 41.8 bc 474 d 592 e Albemarle Red 12.7 cd 110 bc* 24.8 a 23.5 c 29.0 b 53.9 b* 1030 a 1090 c Cowart Red 27.5 b 162 a* 13.8 c 95.9 a* 24.5 bc 46.1 bc* 732 bc 1900 a* Nesbitt Red 15.5 cd 136 ab* 7.53 ce 61.7 b* 18.7 cd 39.4 c* 555 cd 1100 c* Skin Georgia Red Red 42.9 a 74.8 d* 12.8 cd 20.5 c* 38.7 a 10.1 d* 996 a 587 e Carlos White 4.73 e 2.66 c 3.32 a 1.00 c 2.83 cd 2.90 c 159 b 231 b* Fry White 6.44 de 1.01 d* 3.30 a ND5 d* 1.57 d ND e* 189 b ND c* Doreen White 14.1 a 0.93 d* 1.22 cd trace d* 12.1 a 0.66 d* 474 a trace c* Noble Red 3.51 ef 8.69 b* 0.88 d 2.98 b* 2.82 cd 5.79 b* 208 b 168 b Albemarle Red 12.2 ab 24.5 a* 2.06 bc 6.04 a* 9.36 b 12.8 a 203 b 455 a Cowart Red 8.28 cd 1.24 d* 2.12 b trace d* 5.08 c trace e 232 b ND c* Nesbitt Red 10.1 bc 0.54 d* 3.53 a trace d* 8.63 b trace e 197 b ND c* Pulp Georgia Red Red 1.13 f 1.00 d trace6 d ND d 1.13 d ND e 38.2 c ND c* Carlos White 3.01 cde 4.34 e 1.02 cd 8.60 cd* 4.96 c 5.34 cd* 12.5 e 106 e* Fry White 3.99 bcd 11.2 bcd* 2.63 a 21.7 a* 2.94 e 3.13 d 59.1 c 105 e* Doreen White 3.34 cde 14.1b* 0.56 e 7.68 d* 6.02 b 15.7 b* 12.7 e 172 d* Noble Red 8.75 a 20.5 a* trace e 5.78 e* 6.70 ab 15.6 b* 10.1 e 257 b* Albemarle Red 5.15 b 23.4 a* 0.77 de 9.68 bc* 6.81 a 20.1 a* 14.0 e 322 a* Cowart Red 4.03 bc 12.5 bc* 1.31 bc 11.2 b* 4.07 d 6.81 c* 81.0 b 219 c* Nesbitt Red 2.17 e 8.82 d* 1.16 cd 5.19 e* 3.20 e 4.85 cd* 26.1 d 187 cd* Juice7 Georgia Red Red 2.66 de 9.77 cd 1.60 b ND f* 3.47 de 3.16 d 88.0 a 198 cd* 1,2Expressed in ellagic acid equivalents. 3The sum of free ellagic acid and ellagi c acid released following acid hydrolysis. 4Similar letters within columns for each fruit part are not significantly different (LSD test. P <0.05). 5Concentrations below detection limit. 6Concentration below 0.5 ppm. 7Hot-pressed juice. Asterisk (*) indicates significant eff ects by fruit ripening for each fruit parts (LSD test. P<0.05).

PAGE 31

20 ellagitannins were the major source of ella gic acid following hydrolysis. However, the actual concentrations of the ellagic acid glyc osides were likely influenced by the use of free ellagic acid as the quantifying standard. As with most grape varieties, polypheno lic compounds are typically concentrated in epidermal tissues, which are exceptionally th ick in muscadine grap es and often hinder efficient juice extraction. On average, the skin and pulp tissue constituted 21 and 69% of the total mass of the grapes respectively, and were similar for both unripe and ripe fruit. Ellagic acid and its derivatives were generally concentrated in the skin, which contained 51-67% of these compounds in unripe fruit on a fresh weight basis. Upon ripening, these compounds were even more localized in the skin and accounted for 82-87% of the total. Doreen and Cowart contained the highest concen trations of ellagic acid and its glycosides among the cultivars, but no meaningful correla tion could be made between free ellagic acid and/or ellagic acid glycos ides and concentrations of to tal ellagic acid, an observation that likely reflected the influence of ellagita nnins in each isolate. Compared to ellagic acid concentrations present in the skin and pul p, levels present in ju ice were considerably lower and reflected the low solubility of ella gic acid in aqueous systems. A hot break or hot-press technique is commonl y used with muscadine grapes to increase juice yields or add pigmentation to wines or juices, a nd when combined with macerating enzymes ( 49 ), juice extractions are better facilitated. Additionally, the time and temperature of the heating process will appreciably influence juice yields and phytochemical concentration compared to non-heated fruit juice ( 17 ) and white or bronze grapes, depending on cultivar, may not be heated to prevent enzymatic and autoxidative browning reactions affecting juice quality ( 50 ). Textural differences also o ccur in the grapes during ripening

PAGE 32

21 from action of natural pectinase and may al so influence phytochemical solubilization. Typical juice yields may range from 60-75% by weight for hot-pressed muscadine grape juices and is influenced by heating conditions pressing conditions, the use of pressing aids such as rice hulls and skin thickness ( 33 ). The highest concentrat ions of total ellagic acid were found in the juice of ripe Albema rle (322 mg/L) and Noble (257 mg/L), which reflected a 24% average increa se in concentration over jui ce pressed from unripe grapes. Concentrations of total ellagi c acid present in the juice we re not necessarily a reflection of levels found in whole grapes, since the ju ice of unripe fruit contained 2-26% of the amount present in whole grapes compared to 19-78% for ripe fruit. For simplicity, these data were determined based on a 60% juice yield and accounted for the variable contributions from skin and pulp tissue (seed s not included) to the total weight of the grapes. Juice from ripe grap es of Noble, Cowart, Nesbitt, and Georgia Red had the highest total ellagic acid extractions (>58 %), while Carlos, Fry, Doreen, and Albemarle were considerably lower (<34%). The low r ecovery of ellagic aci d derivatives in the latter cultivars reflected the difficulty in solubilizing polyphenolics, likely due to physical barriers associated with their thick skins, which left high concentrations of these compounds behind in the skin and pulp material Free ellagic acid itself, sparingly soluble in water, was also poorly solubilized in al l juices, retaining only 27 and 37% on average of the total present in whole grapes for unrip e and ripe fruit, respectively. However, the ellagic acid glycosides were considerably mo re soluble in juices with >56% recovery from whole grapes. Anthocyanins Anthocyanidins, quantified only in the red cultivars, were expressed in cyanidin equivalents (Table 3-2) since the predominan t anthocyanins in muscadine grapes were

PAGE 33

22Table 3-2. Concentration (mg/kg, mg/L in cyanidin equivalents) of six anthocyanidins and total anthocyanidins on skin, pulp and juice of red muscadine grapes as affected by ri pening stages (U: unripe and R: ripe). Delphinidin1 Cyanidin Petunidin Malvidin + Peonidin Total2 U R3 U R U R U R U R Noble ND4 b5 1450 b ND b 692 c 159 a 1070 a ND a 926 a 159 a 4140 b Albemarle 44.2 a 424 c 28.5 a 291 d ND b ND b ND a 102 d 72.6 b 817 d Cowart 57.6 a 1290 b 37.5 a 1210 a 12.0 b 294 b ND a 445 c 107 ab 3250 c Nesbitt 66.8 a 3550 a 35.1 a 860 b ND b ND b ND a 825 b 102 ab 5230 a Skin Georgia Red 72.1 a 300 c 35.4 a 52.5 e ND b 20.3 b ND a 17.9 e 108 ab 390 d Noble ND b 102 a 6.95 a 93.9 a ND b 78.4 a ND a 114 a 6.95 a 383 a Albemarle 0.84 b 67.1 b 5.85 a 89.0 a ND b 29.7 b ND a 21.8 b 6.70 a 212 b Cowart 4.54 a 3.75 c 4.00 b 10.6 b 0.90 a 0.63 c ND a ND b 9.44 a 15.0 c Nesbitt ND b 19.3 c ND c 12.4 b ND b 5.03 c ND a ND b ND b 36.8 c Pulp Georgia Red 1.17 b 2.52 c 0.980 c 0.76 b ND b 0.25 c ND a ND b 2.15 b 3.53 c Noble ND b 131 a ND b 125 a ND b 155 a ND a 200 a ND b 610 a Albemarle ND b 52.4 c ND b 86.1 b ND b 25.7 c ND a 18.2 b ND b 182 b Cowart ND b 48.6 c ND b 94.0 b ND b 21.4 c ND a 16.5 b ND b 180 b Nesbitt 6.98 a 72.5 b 2.83 a 49.3 c 3.04 a 44.3 b ND a 23.6 b 12.8 a 190 b Juice6 Georgia Red ND b 10.1 d ND b 7.16 d ND b 2.61 a ND a ND c ND b 19.9 c 1Cyanidin equivalents. 2Sum of individual anthocyanidins. 3All anthocyanins are significan tly different at ripening stage. 4. Concentrations below detection limit. 5Similar letters within column s for each fruit part are not significantly different (LSD test. P <0.05). 6Hot-pressed juice. Asterisk (*) indicat es significant effects by fr uit ripening for each fruit part (LSD test. P<0.05)

PAGE 34

23 previously identified as non-acylated 3,5-digl ucosides of six anthocyanidin bases ( 17 ). In the current study, only three anthocyanidins were positively elucidated following acid hydrolysis using the column and solvent c onditions described, due to incomplete separation of peonidin and malvidin and th e absence of pelargonidin. As expected, anthocyanins appreciably increased in th e skin as the fruit ripened with low concentrations also found in pulp material nearest the skin. Anthocyanidin abundance in ripe fruit were delphinidin > petunidin > ma lvidin+peonidin > cyanidin with Nesbitt, Noble, and Cowart containing the highest ove rall concentrations. Color instability of muscadine wine and juice is an established qu ality defect, and is a consequence of their high concentrations of monome ric 3,5-diglucosides with o -diphenolic substituents that include delphinidin, cyan idin, and petunidin ( 51 ). Among the cultivars evaluated, these three anthocyanidins accounted for 78-96% of the total in fresh grapes and from 67-100% in juice. Ripe Noble grapes, one of the most popular wine and jui ce cultivars, contained the highest concentration of malvidin+peonidin among the cultivars evaluated. Malvidin is generally considered the most stable anthocyanin form and along with peonidin was present at 22% of the total in the skins comp ared to 33% in juice. However even with high malvidin+peonidin concentrat ions, the juice from Noble gr apes is considered highly susceptible to color degradation due to likely lack of interand intra-molecular copigmentation of 3,5 digluc osidic anthocyanins ( 51 ) and provides an indication that the remaining cultivars would be even less stable to oxidation or other deteriorative reactions affecting juice or wine pigmentation due to their lower malvidin+peonidin concentrations. These cultivars, as well as the white/bronze varieties, may be more suitable for juice blending to take advantag e of their high ellagic acid contents. Noble

PAGE 35

24 grape juice also contained the highest total anthocyanin conc entration (610 mg/L), while Georgia Red contained consider ably less (20 mg/L) even in relation to the other red varieties that ranged from 180-190 mg/L. Base d on a 60% juice yiel d, only 12% of the total anthocyanins present in grape skins were solubilized into the juice of Nesbitt and Georgia Red, both consumed primarily as table grapes, the former having high anthocyanin content yet poor anthocyanin solu bility characteristics during juicing. Juice from the remaining cultivars, commonly cons umed either fresh or processed, contained 27-32% of the total anthocyanins present in the each grape. The low anthocyanins recovery values in juice, especially in relati on to ellagic derivatives, reflect the degree of processing necessary to solubilize sufficient anthocyanins to produce a suitable red wine or juice. Total Phenolics and Antioxidant Capacity Measurements of total phenolics by the Folin-Ciocalteu metal reduction assay and peroxyl radical scavenging activity using the ORAC assay are common index that provide an overall assessment of the conten t and chemical activity of compounds present in fruits and vegetables. Thes e attributes were quantified in methanolic and ethyl acetate extracts of grape skin, pulp, and juice a nd following partitioning of phenolic acids and flavonols into ethyl acetate, into which ant hocyanins are not soluble, to differentiate between major polyphenolic classes (Tables 3-3 and 4). Values for total phenolics, which varied among cultivars and with fruit ri pening, were good predictors of antioxidant capacity in both methanolic and ethyl acet ate extracts (r= 0.83 and 0.92, respectively). The higher correlation coefficient for ethyl acetate extracts may have reflected the removal of potentially interf ering/prooxidant polar compound s or reflected interactions between anthocyanins and other polyphe nolics in the methanolic extracts ( 52 53 ). Based

PAGE 36

25 Table 3-3. Concentrations (mg/kg, mg/L) of total soluble pheno lics (Folin-Ciocalteu metal reduction assay) in methanolic and ethyl acetate extracts as affected by cultivars and ripening stages (U: unripe and R: ripe). Methanolic Extract Ethyl Acetate Extract Cultivars Color U R U R Carlos White 2430 b1 2530 e 428 d 706 f* Fry White 1440 c 3360 d* 459 d 987 e* Doreen White 3860 a 3990 c* 1430 a 2280 b* Noble Red 2660 b 3090 d 1020 bc 727 f* Albemarle Red 2580 b 2260 e 1320 ab 756 ef Cowart Red 2660 b 4370 c* 1130 ab 1890 c Nesbitt Red 2480 b 5030 b* 627 cd 1300 d* Skin Georgia Red Red 4220 a 9470 a* 1500 a 2910 a* Carlos White 405 de 738 b 128 d 258 a Fry White 566 cd 276 d* 138 d 102 cd Doreen White 1210 b 192 d* 1300 a 39.2 de* Noble Red 601 c 848 b* 332 cd 120 bc* Albemarle Red 1410 a 1100 a* 622 b 274 a* Cowart Red 1110 b 200 d* 528 bc 38.2 de* Nesbitt Red 567 c 443 c 502 bc 16.3 e* Pulp Georgia Red Red 312 e 467 c* 99.4 d 183 b* Carlos White 1145 c 979 de* 165 a 66.4 b* Fry White 1069 c 1500 cd* 90.1 d 81.0 c Doreen White 1673 a 1293 d 161 a 141 a Noble Red 1630 a 1950 b 139 abc 69.1 c* Albemarle Red 1460 ab 1770 bc 147 ab 120 ab Cowart Red 1200 bc 1360 cd 122 bc 89.4 bc Nesbitt Red 739 d 1210 d* 61.2 e 70.8 c Juice2 Georgia Red Red 1140 c 2860 a* 118 c 139 a 1Similar letters within columns for each fr uit part are not significantly different (LSD test. P <0.05). 2Hot-pressed juices. Asterisk (* ) indicates significant effects by fruit ripening for each fruit part (LSD test. P<0.05).

PAGE 37

26 Table3-4. Antioxidant capacity (mol trolox equivalents/g or mL) of methanolic and ethyl acetate extracts as affected by cultivars and ripening stages (U: unripe and R: ripe). Methanolic Extract Ethyl Acetate Extract Cutivars Color U R U R Carlos White 58.0 d1 86.2 cd* 10.2 b 17.5 b Fry White 49.3 d 72.3 d 10.7 b 19.0 b* Doreen White 104 a 90.4 c* 25.2 a 25.5 a Noble Red 97.2 ab 100 c 22.5 a 12.3 c* Albermerle Red 90.8 b 71.1 d 22.6 a 12.0 c* Cowart Red 97.7 ab 119 b 22.6 a 26.3 a* Nesbitt Red 69.7 c 136 a* 12.1 b 25.1 a* Skin Georgia Red Red 89.0 b 128 ab* 25.4 a 29.1 a* Carlos White 5.95 de 8.75 b* 3.90 e 5.05 b Fry White 6.35 de 7.50 bc 3.90 e 3.70 d Doreen White 34.0 a 2.45 d* 15.0 a 1.60 e* Noble Red 9.40 cd 14.3 a 4.70 e 4.10 cd Albermerle Red 14.8 b 13.1 a 8.05 c 6.45 a* Cowart Red 11.8 bc 2.60 d* 5.70 d 1.55 ef* Nesbitt Red 13.9 b 4.60 cd 9.50 b 1.00 f* Pulp Georgia Red Red 4.95 e 6.80 bc 4.10 e 4.45 c Carlos White 20.3 b 15.5 d* 2.37 b 2.06 bc Fry White 14.6 d 20.1 bc 1.61 cd 1.92 bc Doreen White 25.3 a 19.6 bc 2.91 a 2.38 ab Noble Red 24.3 a 26.7 a 2.55 ab 1.51 cd Albermerle Red 19.3 bc 23.3 ab 2.89 a 2.14 b Cowart Red 17.4 cd 21.4 bc* 1.99 c 1.91 bc Nesbitt Red 10.8 e 18.3 cd* 1.27 d 1.20 d Juice2 Georgia Red Red 20.7 b 26.6 a* 2.50 b 2.96 a* 1Similar letters within columns for each fr uit part are not significantly different (LSD test. P <0.05). 2Hot-pressed juices. Asterisk (* ) indicates significant effects by fruit ripening for each fruit part (LSD test. P<0.05).

PAGE 38

27 on abundance, anthocyanins were the major antioxidant compounds present in muscadine grape skin and juice and their concentration was directly rela ted to antioxidant capacity (r= 0.99). Ethyl acetate soluble compounds also contributed to antioxidant capacity and ranged from 12-29%, 22-83%, and 5.7-15% of the total present in methanolic extracts of skin, pulp, and juice, respectively. Other than ellagic acid and its derivatives, many additional compounds were also identified in the ethyl acetate extract including several flavonoids glycosides, pheno lic acids, and procyanidins that are all known to possess antioxidant activity ( 54 55 ). In various concentrations, ga llic acid, protoc atechuic acid, catechin, and epicatechin were identified in ethyl acetate extracts. Flavonoid glycosides were tentatively identified based on their spectroscopic similarities to myricetin, quercetin, and kaempferol with glucose and/ or rhamnose moieties. A myricetin glycoside was the predominant flavonoid present in a ll cultivars and ranged from 8.7-1,350 mg/kg in skin, 0-50 mg/kg in pulp, and 1.6-50 mg/L in juice. Among the cultivars, ripe Georgia Red contained the highest concentrations of total phenolics in bot h ethyl acetate and methanolic extracts of both skin and juice, in contrast to its low an thocyanin, ellagic acid, and ellagic acid glycoside content, which wa s primarily attributed to its high flavonoid concentration. Conclusion This study demonstrated that ripening, physiology, and juice processing influence phytochemical composition and antioxidant capa city of muscadine grapes. Data suggest a diversity of phytochemical compounds that can be used for novel blending schemes for muscadine grape juice or wine to obtain a desired quality and polyphenolic content relating to their antioxidant capacity.

PAGE 39

28 CHAPTER 4 IDENTIFICATION OF ELLAGITANNINS AN D CONJUGATES OF ELLAGIC ACID IN MUSCADINE GRAPES Introduction V. rotundifolia are differentiated from V. vinifera in several points such as lack of an organized fruit bunch, strong disease re sistance, and its unique phytochemical composition primarily due to presence of ellagi c acid derivatives. El lagic acid derivatives are not uncommon in plants and are in abundance in raspberry ( 14 15 46 47 ), pomegranate ( 24 66 67 ), oak ( 56 ), birch leaves ( 68 ), or medicinal herbs/plants ( 69 70 ); yet their presence in muscadine grapes is unique among Vitis species. Ellagic acid derivatives are a broad classifica tion that includes the free acid state of ellagic acid, these conjugated with various sugars to form simple glycosides or more complex ellagitannins ( 46 ). Ellagitannins are characterized as hydrol ysable conjugates containing one or more hexahydroxydiphenoyl (HHDP) gr oup esterified to a sugar, usually glucose. In raspberries, the predominant ellagitannins we re identified as lambertianin C and sanguiin H-6 as well as arabinosides, acetylarbinosid es, and acetylxylosides of ellagic acid ( 14 ) yet similar compounds, if present, have not b een determined in the muscadine grape. The objective of this study was to elucidate iden tities and concentrations of ellagic acid derivatives in muscadine grapes by various extraction and anal ytical procedures. Investigations on ellagic acid and ellagic acid derivatives in muscadine grapes will add value and marketability to the crop due to beneficial health benefits such as its antioxidant activity ( 19 20 ), anti-carcinogenic properties in fluencing cell cy cle arrest and

PAGE 40

29 apoptosis ( 20 ), and the inhibition of tumor formation and growth in mammalian models ( 21 40 ) previously attributed to these compounds. Materials and Methods Muscadine grapes (cvs. Doreen, Noble, Al bemarle) were donated from local grape growers in central Florida a nd polyphenolics extracted from the skin and pulp with 100% methanol (0.01% HCl). Extracts were filte red through Whatman #4 filter paper, solvent removed at 40C under reduced pressure, and po lyphenolic residue redisolved in water at pH 3.5. Isolations Due to the diversity of ellagic acid precurs ors potentially present, it was necessary to fractionate the polyphenolics based on their affinity to C18 and Sephadex LH-20 and partition based on their solubility in various organic solvents (Figure 4-1). Initially, grape extracts were applied to a Sep-Pak C18 cartridge and polyphenolics eluted with various Sep-Pak C18 Sephadex LH-20 Sephadex LH-20 Figure 4-1. Fraction scheme and tentative classification of pol yphenolics present in methanolic extracts of muscadine grapes. Grape Extracts Unbound Fr. Methanol Fr. Isolate I By 100% MeOH :Ellagitannins Isolate III By 10% MeOH : Anthocyanins Eth y l ace t ate Fr. Isolate II :Ellagitannins :Ellagic acid glycosides :Ellagic acid :Flavonoids Isolate IV By 100% MeOH : Ellagitannins

PAGE 41

30 solvents in order of water, ethyl acetate, and th en methanol (0.01% HCl) to partially isolate compounds of interest. Ethyl acetate was removed under reduced pressure and redissolved in water and methanol (95:5, pH 3.5) and this fraction contained phenolic acids, flavonoids, and ellagic acid derivatives including free ellagic (aglycones), ellagic acid glycosides, and ellagitannins. Ellagic acid derivatives that remained in the unbound (water) fraction and in the final methanol -soluble fraction, follow ing evaporation and solubilization in water, were further concentr ated by partitioning from a mini-cartridge of Sephadex LH-20 based on their se lective adsorption properties. Analysis by HPLC-PDA HPLC-PDA analysis was initially employe d to tentatively identify and quantify ellagic acid derivatives in grape extracts a nd isolated polyphenolic fractions from three muscadine grape cultivars (Doreen, Noble, a nd Albemarle). Free ellagic acid, ellagic acid glycosides, and total ellagic acid (following acid hydrolysis in 2N HCl for 60 min at 95C) were evaluated in equiva lents of free ellagic acid. Se parations were conducted on a Dionex HPLC system using a PDA-100 photodi ode array detector and a 250 mm 4.6 mm Acclaim 120 C18 column (Dionex, Sunnyvale, CA) with a C18 guard column. Identical mobile phases in Chapter 3 were employed to separate polyphenolics with modified gradient elution program where pha se B (60% methanol) changed from 0-60% in 30 min; 60-80% in 10 min; 80-100% in 10 min; and 100% in 10 min for a total run time of 60 min, after which the column was e quilibrated to original conditions in 2 min for the next sample injection. A second HPLC-PDA separation was applied to Isolates I and II in order to obtain UV absorbance spectral data unde r identical condition with MSn analysis. Separations were performed using a Waters Alliance 2695 HPLC system and polyphenolics separated

PAGE 42

31 on Phenomenex (Torrace, CA) Synergi 4u Hydro-RP 80A (2 x 150 mm; 4 um; S/N=106273-5) plus C18 guard column (2 mm x 4 mm) and a Waters 996 photodiode array detector recorded UV spectra. Mobile ph ase and gradient program were identical to MSn analysis stated below. Analysis by HPLC-MSn Mass spectrometric analyses were carried out to achieve structural information based on molecular masses and fragment ions. On ly Isolates I and II were evaluated for polyphenolics on an Agilent HPLC system (Pal o Alto, CA) using an 1100 series binary pump and separated using the Phenomenex colu mn previously described. Mobile phases consisted of (A) 0.5% formic acid in wate r (5 mM ammonium formate ) and (B) 0.5% formic acid in methanol and run at 0.15 mL/min. Polyphenolic compounds were separated using a gradient el ution program where phase B changed from 5-30% in 5 min; 30-65% in 70 min; 65-95% B in 30 min and he ld for 20 min; and 95-5% B in 10 min to equilibrate the column and whole system to original conditions for 30 min. Effluents from the column were passed through th e UV detector (Applied Biosystems Model 785A) and then analyzed by mass spectrometer, ThermoFinnigan (San Jose, CA) LCQ with electrospray ionization (ESI). In orde r to confirm molecular masses, ionization was conducted in both negative and positive mode. Total Polyphenols and Antioxidant Capacity Total soluble phenolics were analyz ed using Folin-Ciocalteu assay ( 40 ) and antioxidant activity was determined usi ng the oxygen radical absorbance capacity (ORAC) assay as described in Chapter 3.

PAGE 43

32 Statistical Analysis Data represent the mean of duplicate analys es with analysis of variance conducted using JMP5 software ( 44 ); mean separation was conducted using the LSD test ( P <0.05). Results and Discussion Isolations and Quantification of Polypheno lic Compounds by Solid Phase Extraction Numerous polyphenolic compounds were pr esent in muscadine grapes when analyzed by HPLC, and their separation and identification were enhanced by preparing fractions using a series of solid phase and liquid-liquid extractions based on polarity and affinity characteristics of each compound (Figur e 4-1). Overall, Isolate II represented the majority of non-anthocyanin polyphenolics pres ent in muscadine grapes due to the high affinity of these compounds for ethyl acetate. This fraction was exceptionally high in both phenolic acids and flavonoids By contrast, Isolate I co ntained only the most polar compounds, not retained on Sep Pak C18 cartridges. This isolate had a strong affinity to Sephadex LH-20, a cross-linked dextran, which is widely used to isolate tannins from plant based sample ( 55 ). Compounds that remained on the Sep Pak C18 cartridges following elution with ethyl acetate were subs equently eluted with acidified methanol and were found to predominantly contain an thocyanins. These compounds also yielded high concentrations of free ella gic acid after acid hydrolysis indicating the presence of ellagic acid derivatives. Se phadex LH-20 was employed to separate anthocyanins from these derivatives, with anthocyanins remove d first with up to 10% methanol followed by ellagic acid derivatives with 100% methanol to give Isolates III and IV, respectively. As a result, the initial grape extrac t and four sub-fractions from three cultivars (Doreen, Noble, and Albemarle) were created for subsequent phytochemical analyses and quantification of total soluble phe nolics and antioxidant capacity (Table 4-1 and Figure 4-2).

PAGE 44

33 In each grape cultivar extract, free ella gic acid, three types of ellagic acid glycosides (EAG 1-3) and tota l ellagic acid were evaluate d. Previous studies including Chapter 3 ( 16 ) have reported only two ellagic acid glycosides, however changes in the HPLC gradient program resulted in the separa tion of a third ellagic acid glycoside that eluted earlier than the prev ious two compounds. Extracts from Noble had the highest concentrations of free ellagi c acid (49.7 mg/kg) and total ellagic ac id glycosides (86.9 mg/kg), yet following acid hydrolysis, total el lagic acid was higher for Albemarle (912 mg/kg) compared to Noble (686 mg/kg). This di fference was likely due to the presence of Table 4-1. The concentrations (mg/kg) of free ellagic acid, ellagic ac id glycosides (EAG 1, 2 and 3) and total ellagic acids on each fraction from three different cultivars (Doreen, Noble, Albemarle). Cultivars Isolates1 Free Ellagic acid EAG3 EAG1 EAG2 Total Ellagic acid Whole 13.5 a2 1.60a 19.5 a 22.5 a 360 a I 0.25 b 0.40b N.A. N.A. 13.1 c II 12.9 a N.A. 9.15 b 9.70 b 58.9 b III 0.15 b N.D.3 N.D. N.D. N.D. Doreen IV 0.80 b 0.55 b 1.95 c 1.80 b 2.95 d Whole 49.7 a 6.05 a 31.4 a 49.4 a 686 a I 0.35 d N.D. N.D. N.D. 16.8 e II 11.5 b N.D. N.D. 1.40 d 102 b III 1.90 d 2.75 b 9.55 c 14.0 c 32.7 d Noble IV 5.55 c 2.20 c 13.5 b 24.0 b 63.2 c Whole 32.9 a 7.80 a 20.0 a 37.6 a 912 a I 0.50 c N.D. N.D. N.D. 53.7 c II 27.2 b N.D. 4.60 b 12.1 b 130 b III 1.15 c N.D. 0.20 c 0.40 c 0.65 c Albemarle IV 2.45 c 4.55 b 6.95 b 12.8 b 33.0 c 1. Isolates were prepared by using Se p-Pak C18 and Sephadex LH-20 depending on different behaviors for various solvents. 2. Similar letters within columns for each cultivar are not significantly different (LSD test. P <0.05). 3. Not Detected.

PAGE 45

34 ellagitannins. The fractionation scheme successfully separated ellagitannins into Isolate I since high levels of total ellagic acid was obs erved in all three cultivars compared to low amounts (< 1 mg/kg) of either free ellagic aci d or ellagic acid glycoside. For example, extracts of Albemarle resulted in >100-fold increase in free ella gic acid after acid hydrolysis. Isolate I of Albema rle was applied to further MSn analysis to identify the ellagitannins responsible for re lease of HHDP groups and s ubsequent conversion to free ellagic acid. Isolate II contained the majority of the free ellagic acid in Doreen (95%) and Albemarle (83%), but only 23% for Nobl e, which also contained the highest concentration of anthocyanins followed by Albemarle. Although not fully elucidated, these data seem to indicate that anthocyanins may interfere with or inhibit desorption of free ellagic acid with ethyl acetate from C18 Sep Pak cartridges. Isolate II was also suspected to contain ellagita nnins based on total ellagic acid content following acid hydrolysis; therefore this frac tion from Albemarle was subjec ted to further analysis by MSn to investigate chemical structures specific to ellagic acid glycos ides, ellagitannins, as well as flavonoid glycosides. Isolate III contained predominantly anthocyanins, but low concentrations of free ellagic acid a nd ellagic acid glycosides were also found. Subsequent recovery of remaining ella gic acid derivatives in Isolate IV was accomplished, representing those that were not solubilized by ethyl a cetate earlier in the fractionation process. However, Isolate IV di d not show an appreciable increase in total ellagic acid 0.58, 1.4 and 1.2% for Doreen, N oble, and Albemarle respectively. This observation demonstrated that ella gitannins were not retained on C18 following elution with ethyl acetate and that the ellagic aci d glycosides likely had a similar detector response to free ellagic acid since the sum of free ellagic acid and the three ellagic acid

PAGE 46

35 glycosides were similar before and after acid hydrolysis. This is an important consideration since authentic standards for ellagic acid glycosides are not commercially available. Folin-Ciocalteu metal reduction assay a nd oxygen radical absorbance capacity (ORAC) assay were applied to the initial grap e extracts and four s ub-fractions isolates with the purpose to quantify the total reducing capacity of the samples and to specifically determine the peroxyl radical scavenging properties, respecti vely (Figure 4-2). Generally, total soluble phenolic concentrations are ve ry well correlated with antioxidant capacity and each fraction showed strong positive rela tion, at least r=0.93, between the two attributes. As previously noted, Noble muscad ine grapes are a commonly utilized cultivar for juice or wine making due to their high anthocyanin content and resulted in both high total soluble phenolic and antioxidant capacity in whole extracts and in Isolate III. The compounds in Isolate I, mainly ellagitannins, also contribut ed to antioxidant capacity by 17%, 1.5% and 11% of the total present in whole extracts of Doreen, Noble and Albemarle, respectively. Most antioxidant co mpounds present in Doreen and Albemarle were extracted with ethyl ace tate into Isolate II due to relative abundance of nonanthocyanins compounds and resulted in 41% and 48% antioxidant capacity compare to whole extracts. Total soluble phenolics (18 %) and antioxidant capac ity (6.9%) in Isolate IV of Noble is likely due to residual ellagic acid glycosid es and anthocyanins. Due to potential health benefits related with an tioxidant activity, data re-emphasize that muscadine grapes are an excellent alternat ive crop to be utilized as value-added application.

PAGE 47

36 Figure 4-2. Total soluble phenolic s and antioxidant capacities of five fractions (Whole, Isolates I, II, III and IV) from three different cultivars (Doreen, Noble, and Albemarle).Same letters within fr actions for each attribute are not significantly different (LSD test, P <0.05). Identifications of Ellagic Acid Derivat ives and Flavonoids by HPLC-PDA and MSn To identify compounds associated with th e evolution of free ellagic acid and differentiate these compounds from fla vonoids in muscadine grapes, polyphenolic compounds in Isolates I and II from Al bemarle were examined for their UV spectroscopic properties a nd molecular mass/charge ratio by HPLC-PDA and MSn, Total Soluble Phenolics Concentration (mg/kg GAE) 0 500 1000 1500 2000 2500 Antioxidant Capacity ORAC ( mol Trolox/ml) 0 5 10 15 20 25 Doreen Noble Albemarle Whole I II IIIIV Isolateab c a b c a b c a b a b b a a b b a b a b a a b b a a b b

PAGE 48

37 respectively. Only one cultivar was used for these analyses since the phytochemistry among cultivars are similar, and a goal of th e analyses was to initially elucidate the identity and concentration of compounds present. HPLC coupl ed to PDA detector is an effective tool for tentative identification of the polyphenolic cla ssification and can be used in combination with an authentic standard to identify unknown compounds in a plant-based system; however current tre nds couple this method with HPLC-MSn to additionally confirm the identity of unknown compounds or compounds where an authentic standard is not available. By combining these methods, a compound can be characterized with greater cert ainty of its identification. Ellagic acid derivatives in Isolate I Isolate I was prepared by sequential solid phase extractions with Sep-Pak C18 and Sephadex LH-20 resulted in a fraction th at contained highly polar polyphenolic compounds such as gallic acid, epigallocatech in, and hydrolyzable ta nnins. As discussed previously, the presence of ella gic acid derivatives was indi cated by the release of free ellagic acid after acid hydrol ysis. The sample isolate was employed to obtain the HPLCPDA chromatogram and UV spectroscopic inform ation (Figure 4-3), and also analyzed Figure 4-3. HPLC-PDA chromatogram (280 nm) of Isolate I of muscadine grapes.

PAGE 49

38 Table 4-2. UVmax and HPLC-ESI(-)-MSn analyses of polyphenols in isol ate I from muscadine grapes. Peak No. Rt (min) Compound max MW MS1( m/z ) MS2 ( m/z ) MS3 ( m/z ) 1 5.7 Ellagitannin 1 228, 262sh 802 801 757, 481 301 2 9.2 HHDP-Galloylglucose 268 634 633 301:[M-H]332(Gal-Glc) 301 3 12.1 Gallic acid 272 170 169 170, 125 4 13.6 HHDP-Galloylglucose 265 634 633 301: [M-H]332(Gal-Glc) 301 5 15.0 Ellagitannin 2 268 834 835a 798, 696, 303 6 17.5 Epigallocatechin 291 306 305 261, 221, 219, 179 7 18.8 Digalloyl glucose 295 484 483 331:[M-H]152(Gal) 271, 193, 169 8 21.4 Ellagitannin 3 276 832 833a 481, 303 9 30.7 Digalloyl glucose 272 484 483 331:[M-H]152(Gal) 271, 193, 169 [M+H]+

PAGE 50

39 by HPLCESI(+/-)-MSn to determine molecular masses for compound elucidation (Table 4-2). Peak 1 ( tR =5.7 min; max =228, 262sh nm) was tentativel y identified as ellagitannin 1 according to UV spectrum (Figur e 4-4, A), which agreed with previous studies reported by Zafrilla et al. ( 45 ). This compound was identified as having at least a single HHDP unit esterified to glucose. The average molecular weight ( m / z ) of peak 1 was 802 by both positive ([M+H]+ = m/z 803) and negative ([M-H]= m/z 801) ion modes. In negative ion mode, the base peak produced fragment ions at m/z 481 (M-321), which corresponded to one glucose (179) and one HHDP (302) unit following MS2. Further fragmentation (MS3) produced its major ion at m/z 301 which corresponds to the ellagic acid precursor HHDP and confirmed by free ellagic acid fo rmed following acid hydrolysis. Peaks 2 ( tR =9.2 min; max =268 nm) and 4 ( tR =13.6 min; max =265 nm) had the greatest detector response for the compounds separated and contained similar UV spectroscopic properties (Figure 4-4, B) and (-)ESI-MS mass spectrums. The most abundant ions in their mass spectrum was m/z 633 as [M-H]-, which likely corresponded to HHDP-galloylglucose (MW 634) found in birch leaves ( 26 ). The MS2 spectrum also had fragments at m/z 301 that resulted from the loss of a galloylglucose unit (332) from its parent ion. HHDP-galloyl glucose contains a single HHDP and single galloyl group conjugated to a glucose unit and is often referre d to as strictinin, sanguiin H4 or sanguiin H5 depending on the location of the HHDP and galloyl group ( 56 ). Peak 3 ( tR =12.1 min; max =272 nm) was identified as free gallic acid on the basis of its retention time and UV absorbance spectrum in relation to an authentic standard.

PAGE 51

40 Identification was confirmed by MS-MS, which yielded [M-H]at m/z 169 and a predominant fragment at m/z 125. Peak 5 ( tR =15 min; max =268 nm), even at very low UV absorbance, was tentatively identified as an HHDP glucoside (ellagitannin 2) due to the presence of m/z 303 ions after serial MS analysis in positive mode. The parent compound MW 834 produced an m/z 835 [M+H]+ ion, which underwent MS/MS to produce m/z 303 as its most abundant ion. Figure 4-4. UV spectra of ellagic acid derivatives in Isolate I. A, ellagitannins (peaks 1, 5, 8); B, HHDP-galloylglucose (peaks 2, 4) ; C, digalloyl glucose (peaks 7, 9).

PAGE 52

41 Peak 6 ( tR =17.5 min; max =290 nm) was identified as epigallocatechin on the basis of retention time and UV absorbance spectrum in relation to an authentic standard. This was confirmed by the MS-MS, which yielded [M-H]at m/z 305 and predominant MS2 ion at m/z 261, 221, 219, and 179. Peaks 7 ( tR =18.8 min; max = 295 nm) and 9 ( tR =30.7 min; max = 272 nm) were tentatively elucidated as digalloyl glucose ( 26 ) because [M-H]at m/z 483 yielded m/z 331 by losing a galloyl group (MW 152) from glucose followed by a second galloyl group with subsequent ionization. Peak 8 ( tR =21.4 min; max =276 nm) was thought to contain co-eluting peaks due to the resultant molecular weight indicated from MSn analysis. One of the compounds was determined as MW 832 due to ions at m/z 833 as [M+H]+ and underwent CIDMS/MS to produce m/z 481 and 303, indicating presence of an HHDP unit. Therefore, this compound, ellagitannin 3 ha d potential to convert into ellagic acid with hydrolysis. Additional compounds (MW 480 and 818) were also detected, but no evidence that either was an ellagic acid precursor. Ellagic acid derivatives and fl avonoid glycosides in isolate II Isolate II was prepared by ethyl acetate elution through Sep-Pak C18 cartridges and resulted in a fraction that was free of anthocyanins yet rich in ellagic acid derivatives and flavonoid glycosides with an aglycone base of myricetin, querce tin, kaempferol, which have been reported as predominant fl avonoids in muscadine grape products ( 17 ). Significant increase in ellagic acid by acid hydrolysis indicate d the likely presence of ellagic acid glycosides or ellagi tannins in this isolate. In orde r to elucidate the presence of ellagitannins, two wavelengths (280 a nd 360 nm) were monitored for HPLC-PDA

PAGE 53

42 application (Figure 4-5). MSn analysis (Table 4-3) was applied in both (-) and (+) ESI modes to determine molecular weights a nd compound identification of each peak detected. Peak 1 ( tR =58~60 min; maxs =280sh)was detected only at 280 nm similar to the three ellagitannins observed in Isolate I (Figure 4-4). MSn analysis, however, yielded single mass spectrum at m/z 800 and produced major fragment ions at m/z 447 and 303, which matched fragmentation patterns of HH DP-galloylglucose (Peaks 2 and 6) in Table 4-2. Although MS analysis did not clearly explain its molecu lar identity, this peak was tentatively categorized as an ellagic acid de rivative on the basis of its UV absorbance spectrum and presence of an ion at m/z 303. Peak 2 ( tR =86 min; max =352 nm) was identified as myricetin-rhamnoside on the basis of its UV absorbance typical for a flavonoid and MS-MS spectrum. The parent compound [M+H]+ at m/z 465 ionized to produce the major MS2 fragment at m/z 319, which is indicative of a rhamnosyl unit (146 amu). Furt her ionization of m/z 319 ion Figure 4-5. HPLC-PDA chromatogram (280 an d 360 nm) of Isolate II of muscadine grapes.

PAGE 54

43 yielded typical myri cetin fragment at m/z 301, 273, 255 and 245 ( 57 71 ). Myricetin was previously identified as the most abundant flavonoid in Noble muscadine grapes ( 17 ). Peak 3 ( tR =90.5 min; max =360 nm) has been identifie d as one of ellagic acid glycosides due to its typical UV spectrum MS-MS analysis revealed that [M+H]+ at m/z 435 lost 132 amu, which corresponds to a xylosyl unit, resulting in an ellagic acid ion at m/z 303 by MS2 ( 14 15 ). MS3 of ellagic acid produced major ions at m/z 285 and 257. Consequently, peak 4 is confir med as ellagic acid xyloside. Peak 4 ( tR =91.3 min; max =361 nm) had a [M+H]+ at m/z 449 and MS2 produced a major fragment at m/z 303 (M-146, loss of rhamnosyl group) ( 72 ). Therefore, this peak was identified as ellagic acid rhamnosi de due to presence of major ions at m/z 285 and 257 by MS3. Peak 5 ( tR =92.3 min; max =366 nm) is ellagic acid on the basis of identical retention time and UV spectrum with authentic standard. This was confirmed by the MSMS, which yielded a [M-H]at m/z 301 and prominent MS2 ions at m/z 301, 257 and 229 ( 14 15 ). Peak 6 ( tR =94.2 min; max =351 nm) had [M+H]+ at m/z 449 same as peak 5, which was identified as ellagic acid rhamnoside and also yield the same aglycone ( m/z 303) by initial ionization. Further ioniza tion produced major fragments at m/z 271, 255, 179, and 151, which is the typical fragmentation pa ttern for flavonoid compounds. Both ellagic acid and quercetin have identical molecular we ights as 302, however, it has been reported that MSn fragmentation pattern can be used to discriminate these two compounds. Ellagic acid seems to have more rigid structure than quercetin because the former produces bigger ion pieces than the later does with same collision energy ( 14 ). Also according to

PAGE 55

44 Table 4-3. UVmax and HPLC-ESI(+)/(-)-MSn analyses of ellagitannins, glycosides of ellagic acid a nd flavonoids in isolate II from muscadine grapes. Peak No. Rt (min) Compound max a MW MS1(m/z) MS2 (m/z) MS3 (m/z) 1 58-60 Ellagitannins 261 281sh 280sh 799 818b 447, 303 277 2 86.0 Myricetin-Rhamnoside 352 464 465 319: [M+H]+ -146(Rhamnose) 301, 273, 255, 245 3 90.5 Ellagic acid-Xyloside 360 434 435 303: [M+H]+ -132(Xylose) 285, 257 4 91.3 Ellagic acid-Rhamnoside 361 448 449 303: [M+H]+ -146(Rhamnose) 285, 257 5 92.3 Ellagic acidc 366 302 301 301, 257, 229 6 94.2 Quercetin-Rhamnoside 351 448 449 303 257, 229, 165 7 97.5 Kaempferol-Rhamnoside 344 432 433 287: [M+H]+ -146(Rhamnose) 241, 213, 165, 133 a. max at 2nd band, b. [M+NH4]+, c.characterized by HPLC-ESI(-)-MSn.

PAGE 56

45 UV spectra, this peak is a flavonoid glyc oside not an ellagi c acid glycoside. Consequently, peak 7 was identified as quercetin rhamnoside. Peak 7 ( tR =97.5 min; max =344 nm) was kaempferol rhamnoside because [M+H]+ at m/z 432 produced aglycone ion at m/z 287 by losing a rhamnosyl (146) unit ( 71 72 ) and MS3 confirmed kaempferol with fragments at m/z 241, 213, 165, 133 and 121 ( 73 ). Even though PDA was not able to det ect certain compound, wh ile MS revealed trace levels of compounds such as HHDP-galloylglucose ([M+H]+; m/z 635), myricetinglucoside ([M+H]+; m/z 481), unknown flavonoid pent osyl conjugate ([M+H]+; m/z 467) and unknown compounds containing galloyl and acetylrhamnosyl groups ([M+H]+; m/z 923). Conclusions Major phytochemicals in muscadine grape were identified by UV spectral properties and mass-charge ra tio followed by extraction w ith a suitable solid phase support. The application of mu ltiple MS analysis discovered fragments consistent with known sugar moieties in ellagic ac id glycosides and 4 different ellagitannins in partially purified extracts of muscadine gr ape. In the case of ellagi tannins, these methods were able to assess molecular weights of resp ective fragments, but not exact chemical identities due to diversity of ellagitannins pr esent with varying functional groups. Additionally, predominant flavonoids, such as myricetin, quercetin a nd kaempferol, were determined as conjugated forms with rh amnose. All identifie d phytochemicals were known as excellent antioxidant compounds.

PAGE 57

46 CHAPTER 5 HYDROLYTIC PROPERTIES OF ELLAGIC ACID DERIVATIVES IN MUSCADINE GRAPES Introduction Muscadine grapes have been historically used to produce various food products in both small and large-scale operations including juice, wine, jam\jelly, and more recently dried/concentrated products for value-a dded applications due to their unique and relatively high antioxidant and anticarci nogenic properties. Since no information previously existed on the ellagic acid glycos ide and ellagitannin content of muscadine grapes, and likewise, no information is avai lable on ellagic acid conversion from its precursors associate with heating. Theref ore, by evaluating hydrolysis time and temperature on the relationship of free ellagi c acid from its precursors, the functional properties of these compounds can be evaluate d as a result of processing and prolonged storage. Through the use of acid hydrolysis and evaluation of hydrolyase enzymes, specific to certain polyphenolics, the releas e of free ellagic acid was evaluated in comparison to glycosidic forms for stabi lity characteristics following pasteurization. Response surface methodology (RSM) is a powerful statistical method for modeling and analyzing the response of interest within multip le-interrelated parameters in an effort to optimize this response (58). However in the current study RSM was utilized for a different purpose as a means to continuously monitor the response as affected by two independent variables. The objectives of this study were to determine hydrolytic properties of ellagic acid derivatives and the resultant effects on antioxidant capacity

PAGE 58

47 based on time-temperature combinations us ing a central composite design for RSM analysis. Additionally, by exploring the characteristics of ellagic acid glycosides and ellagitannins in the presence of -glucosid ase and tannase the functional properties of these compounds can be evaluated in the absen ce of a heat-catalzyed hydrolytic reaction. Materials and Methods Response Surface Methodology (RSM ) and Statistical Analyses Methanolic extracts from the cultivar Doreen were evaluated using various hydrolytic conditions to monitor changes to ellagic acid deriva tives that yield free ellagic acid in a time-temperature dependent manne r. Assessment was conducted based on RSM with a central composite design (CCD) includ ing 8 treatment combinations and 3 center points (Figure 5-1). Hydrolysis was performed at pH 3 (cont rol) and at 2N hydrochloric acid (pH <1) with the design evaluated in duplicate. Time-tempera ture combinations were tested at a range of hydr olysis times from 1 min to 2 hrs and temperatures from 20oC to 100oC in order to generate co nditions ranging from partial to complete hydrolysis at each acid concentration. Data were analyzed using JMP5 software ( 44 ) with analysis of variance and mean separation conduct ed using the LSD test (p<0.05). Enzymes Preparation In an effort to evaluate reactions of polyphenolic-active enzy mes on ellagic acid derivatives, -glucosidase (E.C. 3.2.1.21) and tannase (E.C. 3.1.1.20) were added to Sep Pack C18 unbound and ethyl acetate-soluble fr actions, which corresponded to Isolate I and II, respectively from Chapter 4. -gluc osidase (1.333 units/ml) and tannase (1.738 units/ml) prepared in 0.2M phosphate buffer at pH 5 were added in to each fraction at 37 C and reactions were stopped by boiling the solution after 3 hr incubation. Enzyme treated samples were analyzed compared to controls with no added enzymes. Controls

PAGE 59

48 were prepared at pH 3 and 5 in order to determine the influence of pH on enzyme reaction and evaluated compounds. Chemical Analysis Polyphenolics were quantified with a Waters Alliance 2695 HPLC system connected to Acclaim 120 C18 column (250 mm 4.6 mm, Dionex, Sunnyvale, CA) with a C18 guard column and to Waters 996 photodiode array detect or that recorded UV spectra from 200-400 nm. Identical mobile pha ses and gradient elution program were employed to quantify polyphenol compounds in Chapter 4. Total soluble phenolics were analyzed using Folin-Ciocalteu assay (40) a nd antioxidant activity was determined using the oxygen radical absorbance capacity (ORAC) assay as described in Chapter 3. Figure 5-1. Hydrolysis time a nd temperature combinations in cluded in central composite design (CCD) experiment. 60oC, 65 min Time (min) Temp. (oC)32oC, 20 min 88oC, 110 min 60oC, 120min 60oC, 1 min 100oC, 65 min 20oC, 65 min 32oC, 110 min 88oC, 20 min Time (min)

PAGE 60

49 Results and Discussion Ellagic Acid Derivatives and Flavonoid Gly cosides: Effects of Time, Temperature and pHs Polyphenolic analyses were conducted us ing the CCD to determine hydrolytic properties of ellagic acid derivatives includi ng free and glycosidic forms of ellagic acid and ellagitannins. Also, resultan t changes in antioxidant capa city were measured as an index for functional properties due to changes in chemical composition. The 3dimensional representations were developed to show continuous ch anges in ellagic acid derivatives as influenced by two independent variables, time and temperature, during hydrolysis in the presence and absence of hydr ochloric acid (Figures 5-2, 3, 4). The data from hydrolysis at pH 3 was utilized to support the limited information on the influence of heating on phytochemical compositions and functional properties during thermal processing of muscadine grape products. Hydrol ysis with high acid concentrations (0.52N HCl) at various times and temperature is a common way to evaluate polyphenolic aglycones from their respective glycosides ( 59 60 ) and has also been used to assess total ellagic acid following complete hydrolys is of ellagic acid derivatives ( 12 14 ). The present study attempted to pr ovide a picture for alteration of individual phytochemicals by statistical methods duri ng hydrolysis in ranges of 1 to 2 hrs and 20 to 100 C. Overall, the levels of ellagic acid derivatives were significantly influenced by increased hydrolysis time, temperature, and acid concentration that accelerated conversion to free ellagic acid from ellagic acid precursors. Free ellagic acid (Figure 5-2) changed by 3 and 5-fold with respect to absence and presence of acid respectively, as compared to initial hydrolysis conditions (20 C and 1 min) through hydrolysis of ellagic acid glycosides and/or ellagitannins. However, when individually monitored, two ellagic acid glycosides

PAGE 61

50 0 1 2 3 4 5 6 20 40 60 80 100 120 20 30 40 50 60 70 80 90F r e e E l l a g i c A c i d F o l d I n c re a s eT i m e ( m i n )T e m p ( o C ) 0 1 2 3 4 5 6 20 40 60 80 100 120 20 30 40 50 60 70 80 90F r e e E l l a g i c A c i d F o l d I n c r e a s eT i m e ( m i n )T em p .(oC ) pH3 2N HCl Figure 5-2. Tridimensional representation of free ellagic acid generated using fold increases by response surface model with central composite design experiment in the absence and presence of 2N HCl.

PAGE 62

51 -1 0 1 2 20 40 60 80 100 120 20 30 40 50 60 70 80 90 100E l l a g i c A c i d X y l o s i d e F o l d I n c r e a s eT i m e ( m i n )T e m p ( o C ) -1 0 1 2 20 40 60 80 100 20 30 40 50 60 70 80 90 100E l l a g i c A c i d X y l o s i d e F o l d I n c r e a s eT i m e ( m i n )T e m p ( o C ) pH3 2N HCl Figure 5-3. Tridimensional representation of e llagic acid xyloside generated using fold increases by response surface model with central composite design experiment in the absence and presence of 2N HCl.

PAGE 63

52 -1 0 1 2 3 4 20 40 60 80 100 20 30 40 50 60 70 80 90 100E l l a g i c Ac i d R h a m n o s i d e F o l d I n c r e a s eT i m e ( m i n )T e m p ( o C ) -1 0 1 2 20 40 60 80 100 20 30 40 50 60 70 80 90 100E l l a g i c A c i d R h a m n o si d e F o l d I n cr e a seT i m e ( m i n )T e m p (oC ) pH3 2N HCl Figure 5-4. Tridimensional representation of e llagic acid rhamnoside generated using fold increases by response surface model with central composite design experiment in the absence and presence of 2N HCl.

PAGE 64

53 (xyloside and rhamnoside) (Figures 5-3, 4) did not decrease in concentration at pH 3 that would indicate glycosidic hydrol ysis into free ellagi c acid, giving an in dication that free ellagic acid was initially deri ved from ellagitannins in th e extract(s) when hydrolysis occurred without acid. These ellagic acid glyc osides were not completely converted to free ellagic acid until hydrolysis conditi ons reached 60C for 65 min, whereby an appreciable increase in free ellagic acid was observed. Similar observations were made with cyanidin glycosides in blackberry a nd quercetin glycosides in onion, cleaving the sugar moiety in the first hour of acid hydrolysis at 75C ( 61 ). The current data were collected only up to 120 min at fixed acid cont ent (2N HCl), because higher acid contents and prolonged exposure time to acid might lead to degradation of aglycones following hydrolysis ( 61 62 ). Additionally, preliminary data obser ved that the level of ellagic acid was not significantly changed beyond 120 min at 100C. Since flavonoid glycosides al so convert into aglycone of flavonoids by liberating a sugar moiety during hydrolysis, major fla vonoid glycosides, myri cetin, quercetin, and kaempferol rhamnosides, were eval uated in ranges of 1.00~2.06, 0.92~1.82 and 1.89~3.57 mg/kg, respectively, as change in time-temperature combinations for the absence of acid. For 2N HCl hydrolysis, all glycosides were not detected after 65 min. and 60C, as observed with ellagic acid gl ycosides. Shown in the 3dimensional representations (Figures 55, 6, and 7), flavonoid glycosid es showed less than 50% increase after 2hrs at 100C compare to before hydrolysis at pH 3 indicating that most flavonoid glycosides might survive through short term heat processing. Antioxidant Capacity of Polyphenolics as Affected by Aglycone vs Glycosides with Hydrolysis Phytochemicals contribute to the functi onal properties of food systems and their

PAGE 65

54 0 1 2 320 40 60 80 100 120 20 30 40 50 60 70 80 90 100M y r i c e t i n r h a m n o s i d e F o l d I n c r e a s eT i m e ( m i n )T e m p ( o C ) Myricetin-rhamnoside-1 0 1 220 40 60 80 100 120 20 30 40 50 60 70 80 90 100M y r ic e t in r h a mn o s id e F o l d I n c r e a s eT i m e ( m i n )T e m p (oC ) 2N HClpH3 Figure 5-5. Tridimensional representation of myricetin rhamnoside generated using fold increases by response surface model with central composite design experiment in the absence and presence of 2N HCl.

PAGE 66

55 0 1 2 320 40 60 80 100 120 20 30 40 50 60 70 80 90 100Q u e r c e t i n r h a m n o s i d e F o l d I n c r e a s eT i m e ( m i n )T e m p (oC ) -1 0 1 220 40 60 80 100 120 20 30 40 50 60 70 80 90 100Q u e r c e t i n r h a m n o s i d e F o l d I n c r e a s eT i m e (m i n )T e m p (oC ) 2N HClpH3 Figure 5-6. Tridimensional representation of quercetin rhamnoside generated using fold increases by response surface model with central composite design experiment in the absence and presence of 2N HCl.

PAGE 67

56 0 1 2 320 40 60 80 100 12020 30 40 50 60 70 80 90 100K a e m p f e r o l r h a m n o s i d e F o l d I n c r e a s eT i m e ( m i n )T e m p (oC ) -1 0 1 2 320 40 60 80 100 120 20 30 40 50 60 70 80 90 100K a e m p f e r o l r h a m n o s i d e F o l d I n c r e a s eT i m e ( m i n )T e m p (oC ) 2N HCl pH3 Figure 5-7. Tridimensional representation of k aempferol rhamnoside generated using fold increases by response surface model with central composite design experiment in the absence and presence of 2N HCl.

PAGE 68

57 roles may diverge depending on concen tration, structural differences, and synergistic/antagonistic responses in foods and biological systems. Many polyphenolics in fruits and vegetables ar e found as conjugated forms, with various esterified sugar moieties. Anthocyanins, the most widely dist ributed class of flavonoids in plants, are commonly investigated to explore antioxi dant capacity changes among the various aglycones and glycosidic forms ( 62 64 ). Anthocyanidins, the aglycone form of anthocyanins, tended to have higher radi cal scavenging activities than those of corresponding glycosides in ORAC assay ( 64 ) or DPPH assay ( 31 ), whereas superior activity was obtained with monoglycosylation of malvidin, pelargonidin, and peonidin in the -carotene bleaching method ( 31 ). The current experiment was able to investigate the role of aglycones and suga r conjugated forms including hydrolysable tannins and glycosides of non-anthocyanins polyphenolic s in muscadine grapes to affect metal reduction and hydroxyl radical scavenging propert ies as a result of different stages of hydrolysis. Contour plots were presented for total solubl e phenolics (Figure 5-8) and antioxidant capacity (Figure 5-9) by Folin-C iocalteu assay and oxygen radical absorbance capacity (ORAC) assay, respectively. Data on both attributes support that aglycone polyphenolics have higher ability to reduce metal ions and scavenge hydroxyl radical compared prior to hydrolysis of polyphenolic s. Ellagic acid aglycones showed strong correlation with both total so luble phenolics and antioxida nt capacity, r=0.89 and r=0.58, respectively, while lower correlations were observed in ellagic acid glycosides, average 52% and 36% for total soluble phenolics and antioxidant capacity, respectivley (Table 51). Usually flavonoids glycosides have been evaluated as containi ng lower antioxidant

PAGE 69

58 0 1 2 3 4 5 6 720 40 60 80 100 120 20 30 40 50 60 70 80 90 100T o t al S ol ub l e P h en ol i c s F ol d I n c r e as eT i m e ( m i n )T e m p ( o C ) pH30 1 2 3 4 5 6 720 40 60 80 100 120 20 30 40 50 60 70 80 90 100T o t a l S o l u b l e P h e n o l i c s F o l d I n c re a s eT i m e ( m i n )T e m p ( o C ) 2N HCl Figure 5-8. Tridimensional representation of to tal soluble phenolics generated using fold increases of Folin-Ciocalteu measurem ents by response surface model with central composite design experiment in the absence and presence of 2N HCl.

PAGE 70

59 0 1 2 3 4 5 20 40 60 80 100 12020 30 40 50 60 70 80 90 100A n t i o x i d a n t C a p a c i t y F o l d I n c r e a s eT i m e ( m i n )T e m p ( o C ) 0 1 2 3 4 520 40 60 80 100 12020 30 40 50 60 70 80 90A nt i ox i da nt C ap ac i t y F o l d I n c r e a s eT i m e ( m in )T e m p ( o C ) pH3 Acid Figure 5-9. Tridimensional representation of an tioxidant capacity generated using fold increases of ORAC measurment by re sponse surface model with central composite design experiment in the absence and presence of 2N HCl.

PAGE 71

60 capacity compared to aglycone of flavonoids since blocking the 3-hydroxyl group in the heterocyclic ring influenced stability of the aroxyl radica l of flavonoids and also a decrease in number of fr ee hydroxyl group (OH) might play an important role in scavenge free radicals ( 74 ). The structure-activity relati onship for hydrolysable tannins, ellagitannins and hydrolytic compounds has not been elucidated with pure compounds; however these data provide evidence that ch emical antioxidant capacity increases with higher concentrations of hydrolytic or ag lycone forms. Even though HHDP carries 6 hydroxyl groups compared to 4 in free ellagi c acid, their electron donating properties are likely inhibited by the presence of the glycosid ic moiety. However, due to the low water soluble characteristics of ellagic acid, the te sted samples were prepared to contain low amounts of ellagic acid derivatives (<5 mg/m L as total ellagic acid). Therefore, this premise on higher radical scavenging activity of free ellagic acid shou ld be limited to the fact that all free ellagic acid was completely solubilized in solution; however, this is Table 5-1. Pearson correlations coefficients of individual ellagic acid derivatives contents with total soluble phenolics and antioxidant capacity. Variable 1 Variable 2 Partial hydrolysis at pH 3 Complete hydrolysis with 2N HCl Total Soluble Phenolics Antioxidant Capacity 0.77 0.81 Ellagic acid aglycone Tota l Soluble Phenolics 0.89 0.98 Antioxidant Capacity 0.58 0.83 Ellagic acid-xyloside Tota l Soluble Phenolics 0.42 -0.29 Antioxidant Capacity 0.30 -0.35 Ellagic acid-rhamnoside Total Soluble Phenolics 0.62 -0.23 Antioxidant Capacity 0.42 -0.28

PAGE 72

61 the first trial to compare the structure-act ivity relationship between glycosides or hydrolysable tannins and aglycone s in crude plant extracts. Enzymatic Hydrolysis of Ellagic Acid Derivatives Though considered a less prevalent tec hnique, hydrolysis by enzymes such as esterase are alternative met hods to release aglycones from various polyphenolics through degradation of carbohydrate linkages ( 59 ). -glucosidase (E.C. 3.2.1.21) as well as other glycosidases like arabinofuranosidase were considered important enzymes associated with the quality of foods and beverages and ha ve been used to release aroma components in musts, wine, and fruit juices ( 65 ). Tannase or tannin acyl hydrolase (E .C. 3.1.1.20) are widely applied in the food industry to produce instant tea, to manufa cture gallic acid and propylgallate, which can be uti lized as food preservatives, and are commonly used to remove undesirable tannins. However, there are no published data on reactivity of these enzymes on ellagitannins and ellagic acid glyc osides in food systems. Muscadine grape extracts were hydrolyzed by -glucosidase and tannase in order to provide more information on the properties of ellagic acid de rivatives associated with enzyme reactions. Overall, the two enzymes showed distinctive responses in each grape isolate evaluated (Table 5-2) due to their char acteristic polyphenolic composit ion. As described in Chapter 4, an ethyl acetate fracti on contained mostly polypheno lic compounds including free ellagic acid, ellagic acid xyloside, ellagic acid rhamnoside, and e llagitannins (MW 799), also there is evidence for the presence of gallotannins in th is fraction. Compared to the complex components in the ethy l acetate fraction, th e water fraction is relatively simple containing tannin types of ella gic acid precursors such as three different ellagitannins (MW 802, 834, and 832) and HHDP-galloyl glucose. Therefore, enzyme reactivity on ellagic acid glycosides was evaluated in et hyl acetate fraction and any free ellagic acid

PAGE 73

62 evolution in the water fraction was a conseque nce of enzyme reactions on ellagitannins. Reactions were performed at pH 5 to satis fy optimal conditions for both enzymes, and fortunately this elevated pH did not signifi cantly affect the compone nts as compared to the original pH of grapes. Th e activities of each enzyme we re effectively evaluated by HPLC by monitoring both products and r eactants under hydrolysis conditions. glucosidase showed higher reactivity on xylose compared to rhamnose moieties as indicated by a significant d ecrease in ellagic acid xylosid e (3.14 to 1.05 mg/kg), whereas the rhamnoside was unaltered in the ethyl acet ate soluble fraction afte r 3 hrs incubation at 37 C. Despite the observed decrease in ellagi c acid xyloside, a corresponding increase in free ellagic acid was not observed due to the low initial concentrations of the xyloside present. However, when higher concentrations of -glucosidase were used in preliminary studies, free ellagic acid si gnificantly increased and was also found in the insoluble sediments. A correspondingly significant d ecrease in both ellagic acid xyloside and ellagic acid rhamnoside was observe d after 2 hrs incubation at 37 C (data not shown). Reaction of -glucosidase on each ellagic acid glycoside might vary depending on the sugar groups attached to ellagic acid, be cause ellagic acid gl ycoside was hydrolyzed with -glucosidase in a preliminary study. Af ter incubation of each isolate with tannase, no activity on either ellagic ac id glycosides or HHDP units of ellagitannins was detected. However, free gallic acid was significantly a ffected compared to control as a 17.8-fold increase. Considering the composition of ethyl acetate fraction, additi onal free gallic acid after incubation with tannase can be a hydr olytic product of bot h gallotannins and ellagitannins. However, no significant incr ease in gallic acid for the water fraction suggests that tannase has higher reactivity for gallotannins than ellagitannins. The

PAGE 74

63 reactivity of tannase (from Aspergillus ficuum ) may not depend on sugar specificity as discussed in -glucosidase due to no change in ellagic acid glyc osides with tannase incubation. Consequently, tannase seems prefer ably to cleave off a single gallic acid rather than oxidative coupling of neighboring gallic acid, HHDP unit, from ellagitannins. Various studies on tannase activ ity investigated various t ypes of hydrolysable tannins such as tannic acid, met hylgallate, ethylgallate, or n -propylgalltae ( 75 77 ). Using this form of tannase for muscadine grape extrac ts did not show promise in hydrolyzing ellagitannins into free ellagic acid, however, other tannase sources may have different reactivities to those compounds present in muscadine juice or wine as a means to alleviate quality defects or to utilize muscadine pom ace to produce ellagic acid. Table 5-2. Concentrations (mg/kg) of ellagic acid, ellagic acid glyc osides (xyloside and rhamnoside) and gallic acid affected by enzyme treatment in two different fractions from Doreen (bronze) extracts. Fractions1 pH Enzyme Free ellagic acid Ellagic acid xyloside Ellagic acid rhamnoside Gallic acid 3 None 31.9 b4 3.92 a 10.5 a 1.80 c 5 None 39.4 ab 3.14 a 11.3 a 1.80 bc 5 -glucosidase 2 41.4 ab 1.05 b 10.7 a 2.78 b Ethyl acetate 5 Tannase3 47.2 a 3.86 a 11.4 a 32.1 a 3 None 1.47 a N.D. N.D. 0.589 ab 5 None 1.27 a N.D. N.D. 0.577 ab 5 -glucosidase 2 1.37 a N.D. N.D. 0.505 b Water 5 Tannase3 1.29 a N.D. N.D. 0.789 a 1.prepared by Sep-pak C18. 2. 1.333 units/ml, 3. 1.738 units/ml in final solution and 3hrs incubation at 37 C.4.Similar letters with in columns for each fraction are not significantly different (LSD test. P <0.05).

PAGE 75

64 Conclusions Properties of ellagic acid derivatives we re studied as affected by non-enzymatic (time-temperature combinations) and enzyma tic (-glucosidase and tannase) hydrolysis. Elevated time and temperature without aci d created the environment for partial hydrolysis of ellagic acid deriva tives including ella gitannins and ellagic acid glycosides; however most glycosidic components (ellagic acid glycosides and flavonoid glycosides) remained after the reaction. A dditional 2N HCl completely h ydrolyzed ellagitannins and ellagic acid glycosides in 1hr and the free ellagic acid produced was likely to scavenge more hydroxyl radicals than conjugated forms of ellagic acid. This was also indicated by a significant increase in antioxidant capacity with evolution of ella gic acid after heating both in the absence and presence of acid. -glucosidase showed the possibility for application on muscadine grape juice or other products to hydrolyze ellagic acid glycosides; however tannase was not a feasib le option for ellagic acid precursors.

PAGE 76

65 CHAPTER 6 HYDROLYTIC AND OXIDATIVE PROPERTIES ON ELLAGIC ACID DERIVATIVES DURING STORAGE OF MUSCADINE GRAPES JUICES Introduction Free ellagic acid increases w ith hydrolysis of its precurs ors including ellagic acid glycosides and ellagitannins a nd contributes to the developmen t of insoluble sediments in muscadine juice or wine during storage. Pr ior to investigating chemical or physical processing options to remediate or acceler ate sediments in musc adine products, it is important to determine key components aff ecting relative changes of ellagic acid derivatives during storage. According to a recent study on phytochemical stability of muscadine grape juice ( 33 ), phytochemical losses following processi ng with high hydrostatic pressure (HHP) were presumably due to the activation of re sidual oxidases after ju ice extraction and/or autoxidative mechanisms resulting in co-oxi dation of anthocyanins and ascorbic acid. Ascorbic acid fortification is common fo r producing juice with additional oxidative protection while contributing to additional health benefits, qualit y, and market value. It is hypothesized that ascorbic acid fortification may protect ellagitannins from oxidative degradation in non-anthocyanin containing muscadine grape juices as ascorbic acid is commonly added in plant extract s to protect ellagitannins fr om oxidation during analysis ( 26 ). However, few studies have investigated th e effects of ascorbic acid fortification on oxidative stability of individual polyphenolic s, especially ellagic acid conversion from ellagic acid derivatives, in food products.

PAGE 77

66 The objective of study was to evaluate indivi dual ellagic acid derivatives in whole juice and sub-isolates over time in an effort to reveal the main precursors for producing free ellagic acid during storage of muscadine juice. Add itionally, by assessing ellagic acid derivatives as affected by therma l processing, ascorbic ac id fortification, and sparging with air, conclusions can be dr awn concerning the chemical and oxidative behaviors of ellagic acid derivatives over time in muscadine grape juice. Materials and Methods Storage of Red Juices and Isolations In order to monitor the behaviors of indivi dual ellagic acid de rivatives in whole juice and isolations, hot-pressed red muscadin e juice was initially prepared by blending Noble and Albemarle (1:1) cultivars, which contain high anthocyanins and ellagic acid derivatives, respectively. Grapes were dona ted from local grape growers in central Florida and were frozen until processed. Equal portions of each grape were blended and pressed following heating the grapes at 70 C for 15 min. Juice was filtered and thermally pasteurized (90 C, 5 min). Isolations were prepar ed with Sep-Pak C18 cartridge as described in Chapter 4. Resulting isolates including whole juice, water (unbound), ethyl acetate and methanol isolates were stored at 4 and 37 C for 5 weeks. Storage of White Juices and Isolations Cold-pressed white muscadine grape (Doreen) juice was prepared by simply crushing and pressing the fruit in an effort to evaluate the influence on ellagic acid derivatives by thermal pasteurization, asco rbic acid, and air sparging. To compare thermal processing, a portion of ju ice was thermally pasteurized (90 C, 5 min) and to the remaining juice sodium azide was added to retard microbiological growth prior to

PAGE 78

67 treatment application. Portions of the remaining jui ce (unpasteurized) we re then fortified with ascorbic acid (1,000 mg/L) and equally distributed into fl asks and sparged with air. Air bubbled into samples for 1 hr at room te mperature and additional buffer (citric acid buffer, pH 3.2) was added to recover loss of water during the treatment as needed. Equal volumes of non-treated and treated samp les were kept in test tubes at 4 C and 37 C for 5 weeks. Chemical Analysis Samples were collected at 0, 1, 3, and 5 weeks from both temperatures and centrifuged to remove insoluble particles prior to analysis. Ellagic ac id derivatives were then analyzed by HPLC, as described in Ch apter 5. Total ellagic acid was evaluated following acid hydrolysis (2N HCl for 60 mi n at 95 C) and separation was achieved using phase B (60% methanol, pH 2.4) change d from 50-70% in 3 min; 70-80% in 2 min; 80-100% in 20 min; and 100% B in 5 mi n for a total run time of 30 min. Results and Discussion Changes of Ellagic Acid Derivativ es during Storage of Red Juice Muscadine grape products are prone to de veloping insoluble sediments that are likely created from ellagic acid derived from hydrolysis of its precursors during storage. Unfortunately, various processing and storag e regimes have not been successful for reducing sediment formation (1, 2, 28-30) due to lack of understanding the behavior of each ellagic acid derivati ve during storage. In this study, muscadine juice and 3 isolates from Sep-Pak C18 were stored for 5 weeks at 4 C and 37 C, and relative changes in individual ellagic acid deriva tives were quantified. Signifi cant changes were observed in each isolate during storage, and were influen ced by storage temperature. The whole juice prior to fractionation represen ted intact juice and quantifie d chemical attributes were

PAGE 79

68 varied for storage time (Figure 6-1). Variati ons were more significant at 37C, especially for free ellagic acid concentration with a 143 % increase compared to 58% increase at 4C. Increases in free ellagic acid were the result of ellagitannin degradation rather than ellagic acid glycosides, sin ce two quantified ellagic acid gl ycosides decreased only 4% and 9% on average at 4 and 37 C, respectively. Changes in free ellagic acid concentrations should be consid ered with formation of sedime nts, but accurate analyses on sediments were difficult due to the low juice volume used for evaluation. However, free ellagic acid loss via sedime nts were indirectly determin ed by evaluating total ellagic acid as sediments removed prior to analysis and resulted in a decrease in total ellagic acid concentrations. Total ellagic acid observed changes over 5 weeks were not significant with an 11% increase at 4 C and 11% decrease at 37 C. Considering that whole juice developed the most insoluble components durin g storage, suggesting free ellagic acid seems to be a minor contributor and s upported by a previous study where 12% of sediment by weight was free ellagic acid ( 16 ). Water (unbound) isolate was prepared by a Sep-Pak C18 with water to elute polar ellagic acid precursors such as ellagitannins and the stability of ellagitannins were measured indirectly by evaluating total ellagic acid. Total ellagic acid was considered as hydrolytic ellagic acid from mainly ellagitannin s, because only trace levels (<1 mg/L) of ellagic acid glycosides were evaluated. Pro-ellagic acid compounds did not significantly change at 5 C but at 37 C total ellagic acid decreased 78% (25 5 mg/L) (Figure 6-2) indicating significant degradation of ellagitannins. It is interesting to note that free ellagic acid concentrations were not influenced by ellagitannin hydrolysis, indicating that free ellagic acid may precipitate by other component s such as metal ions, soluble pectin, or

PAGE 80

69 organic acids isolated in this fraction. It was shown previ ously that ella gic acid has the ability to bind metal ions ( 16 78 79 ). Ellagic acid glycosides were evaluated in both ethyl acetate and methanol isolates (Figures 6-3, 4) and all ellagic acid glycosid es including ellagic acid xylosides and ellagic acid rhamnosides were observed to decrease ranging from 12 to 14% and 13 to 35% at 4C and 37C, respectively. Ellagic acid glycos ides such as ellagic acid acetyl-xyloside and ellagic acid arabinoside were reported to be stable for up to 6 months in raspberry jam ( 45 ). Compared to the slow degradation of e llagic acid glycosides in the ethyl acetate isolate, total ellagic acid ( 88%) and free ellagic acid ( 37%) showed more distinguishable changes during st orage at 37C. Since the presence of ellagitannins in the ethyl acetate isolate was confirmed by HPLC-MS/PDA in Chapter 4, ellagitannins degraded and formed insoluble compounds in solution resulting in a decrease in total ellagic acid. Consequently, elevated storage temperatur e significantly accelerated hydrolysis of ellagic acid precu rsors during storage time. Am ong ellagic acid precursors, ellagitannins are likely to hydrolyze before ellagic acid glycosides, and then stay in solution or contribute to the formation of insoluble compound s with other juice constituents. Additionally, consideration of the possible eff ects of oxidation on ellagitannins and resultant conversion into fr ee ellagic acid in juice storage was needed, because ellagitannins are very susceptible to oxidation ( 27 ). Initial Ellagic Acid Derivativ es in White Muscadine Juice as Affected by Ascorbic Acid and Air Ascorbic acid is commonly fortified into fr uit juices or products in efforts to retard the oxidation and add nutritional value; however this is chal lenge for red grape juice due to mutually destructive properties between anth ocyanins and ascorbic acid in presence of

PAGE 81

70 Time (Weeks) 0123456 Concentration (mg/L) 0 20 40 60 80 100 120 140 Time (weeks) 0123456 Concentration (mg/L) 0 20 40 60 80 100 EA-xylose EA-rhamnose Free Ellagic Time (weeks) 0123456 Concentration (mg/L) 0 200 400 600 800 1000 Total Ellagic Time (Weeks) 0123456 Concentration (mg/L) 0 200 400 600 800 1000 Figure 6-1. Changes in whole red juice for ella gic acid derivatives du ring storage. A) At 4C. B) At 37C. B37 C A 4 C

PAGE 82

71 Time (weeks) 0123456 Concentration (mg/kg) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Time (weeks) 0123456 Concentration (mg/L) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 EA-xylose EA-rhamnose Free Ellagic 0123456 Concentration (mg/L) 0 5 10 15 20 25 30 Total Ellagic Time (weeks) Time (weeks) 0123456 Concentration (mg/kg) 0 5 10 15 20 25 30 Figure 6-2. Changes in water is olate for ellagic acid deriva tives during storage. A) At 4C. B) At 37C. A 4 C B37 C

PAGE 83

72 Time (weeks) 0123456 Concentration (mg/kg) 0 2 4 6 8 10 Time (Weeks) 0123456 Concentration (mg/L) 0 2 4 6 8 10 EA-xylose EA-rhamnose Free Ellagic Time (Weeks) 0123456 0 20 40 60 80 100 Total Ellagic Acid 0123456 0 20 40 60 80 100 Time (Weeks)Concentration (mg/L) Concentration (mg/kg) Figure 6-3. Changes in ethyl acet ate isolate for ellagic acid de rivatives during storage. A) At 4C. B) At 37C. A 4 C B37 C

PAGE 84

73 Time (weeks) 0123456 Concentration (mg/L) 0 10 20 30 40 50 Time(weeks) 0123456 Concentration (mg/L) 0 10 20 30 40 50 EA-glucoside EA-xyloside EA-rhamnoside Free Ellagic Time(weeks) 0123456 Concentration (mg/L) 0 50 100 150 200 250 Total Ellagic Time (weeks) 0123456 Concentration (mg/L) 0 50 100 150 200 250 Figure 6-4. Changes in methanol isolate for ella gic acid derivatives during storage. A) At 4C. B) At 37C. A 4 C B37 C

PAGE 85

74 oxygen ( 84 85 ). For this reason, the white muscadine grape cultivar Doreen was investigated for fortification as a means to study stability and the effects of oxygen on ellagic acid derivatives during storage. Prior to treatment, ju ices were either pasteurized or non-pasteurized to evaluate the effect of treatments. Before st orage, juices were analyzed for initial levels of ellagic acid derivatives to determine treatment effects (Figure 6-5). A comparison of heated and non-heated juice resu lted in a significant increase in free ellagic acid and ellagic acid glycosides, but total ellagic acid remained. This data supported that ella gitannins break down into fr ee ellagic acid by elevated temperature prior to hydrolysis of ellagic acid glycosid es, as observed by RSM in Chapter 5. In non-heat treated juices, ascorbic acid seemed to play a role in increasing only free ellagic acid because 2.5 and 3.4 fold increases were observed in samples with ascorbic acid fortification, and the comb ination of ascorbic acid and air sparging, respectively. Levels of ellagi c acid glycosides or total ella gic acid were not significantly changed by ascorbic acid fortifi cation indicating that ascorbic acid may help to retain free ellagic acid in solution or to accelerate ella gitannins hydrolysis. This impact of ascorbic acid on ellagitannins and ellagic acid were likely related to the presence of natural oxidative enzymes such as polyphenol oxidase or peroxidase since ascorbic acid fortification did not influen ce ellagic acid deriva tives in thermally processed juices. Compared to changes by ascorbic acid, exce ssive amounts of air in the system did not influence the initial level of free ellagic acid. Air sparging at natural juice pH is not likely to be at optimal conditions to induce oxidation of ellagic acid deriva tives, since the mode of oxidation is expected to be influenced by the presence of semiquinones that form by the action of phenolate ani ons with triplet oxygen ( 80 ). Oxidation of free ellagic by air

PAGE 86

75 A. Non-Heat ControlAsA onlyAir only AsA+Air Concentration(mg/kg) 0 50 60 70 EA-xylose EA-Rham Free Ellagic Total Ellagic B. Heat ControlAsA onlyAir only AsA+Air Concentration(mg/kg) 0 50 60 70 Figure 6-5. Concentrations (mg/L) of ellagic acid derivatives depending on treatments of ascorbic acid (1,000 mg/L) and air at 0day of Doreen juice as affected by thermal pasteurization method (A: Non-h eat, B: Heat, indi cates significant differences between non-heat and heat treatments). * * * *

PAGE 87

76 sparging could have occurred with an elevated pH due to hydrolysis of its lactone ring or by alkanine hydrolysis of ellagic acid glycosides and ellagitannins. Post-storage Levels of Ellagic Acid D erivatives in White Muscadine Juice as Affected by Ascorbic Acid and Air The evolution of free ellagic acid is a consequence of hydrolysis of ellagic acid derivatives during long term storage. In this study, white muscadine juices were stored for 5 weeks at 37C and rela tive changes in ellagic acid derivatives quantified to determine the effects of pasteurization proce ss, ascorbic acid fortification and excessive amounts of air on each a ttribute. Significant changes were observed for each ellagic acid derivatives during storage; however treatments with ascorbic acid fortification and air sparging for juices were not major factors on relative changes of ella gic acid derivatives over time. Among different ellagic acid derivati ves, only free ellagic acid showed effects of ascorbic acid fortificati on (Figure 6-6). Ascorbic acid fortification was initially evaluated to retain higher con centrations of free ellagic aci d, but ascorbic acid fortified juice retained greater levels of free ellagic acid through storage in bot h non-heat and heat treated juices. Significant changes in chemical composition of stored juices were influenced by thermal processing prior to individual fortifi cation or air sparging. Th e heat-treated juice without additional treatments had initially higher concentrations of free ellagic acid and changed from 4.02 to 9.35 mg/L during storage, compare to 0.74 to 2.53 mg/L in nonheat treated juice. It is interesting to note that most of the increase in free ellagic acid for non-heat treated juice occurred in the first week of storage an d the level decreased but not significantly. However, the heat treated juice in creased in the last tw o weeks (Figure 6-6). The relative stability of ellagic acid glycosid es were confirmed again in heat treated

PAGE 88

77 0123456 0 2 4 6 8 10 12 Time (weeks) 0123456 Concentration (mg/kg) 0 2 4 6 8 10 12 Control AsA only Air only AsA+Air A. NonHeatConcentration (mg/kg)Time (weeks)B. Heat Figure 6-6. Concentrations (mg/L) of free e llagic acid depending on ascorbic acid (1,000 mg/L) and air during storage (5 weeks, 37 C) of Doreen juice as affected by thermal pasteurization (A: Non-heat treatment, B: Heat treatment).

PAGE 89

78 Time (weeks) 0123456 Concentration (mg/kg) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Control AsA only Air only AsA+Air A. NonHeat Time (Weeks) 0123456 0.0 0.2 0.4 0.6 0.8 1.0 1.2 B. HeatConcentration (mg/kg) Figure 6-7. Concentrations (mg/L) of ellagi c acid xyloside depending on ascorbic acid (1,000 mg/L) and air during storage (5 weeks, 37 C) of Doreen juice as affected by thermal pasteurization (A: Nonheat treatment, B: Heat treatment).

PAGE 90

79 Time (weeks) 0123456 Concentration (mg/kg) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Control AsA only Air only AsA+Air A. NonHeatConcentration (mg/kg) 0123456 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Time (weeks)B. Heat Figure 6-8. Concentrations (mg/L) of ellagic acid rhamnoside depending on ascorbic acid (1,000 mg/L) and air during storage (5 weeks, 37 C) of Doreen juice as affected by thermal pasteurization (A: Nonheat treatment, B: Heat treatment).

PAGE 91

80 0123456 0 10 20 30 40 50 60 70 Time (weeks) 0123456 Concentration (mg/kg) 10 20 30 40 50 60 70 Control AsA only Air only AsA+Air A. NonHeatConcentration (mg/kg)Time (weeks)B. Heat Figure 6-9. Concentrations (mg/L) of total ellagic acid depending on ascorbic acid (1,000 mg/L) and air during storage (5 weeks, 37 C) of Doreen juice as affected by thermal pasteurization (A: Non-heat treatment, B: Heat treatment).

PAGE 92

81 juices based on only 27% and 17% decrease fo r ellagic acid xyloside and ellagic acid rhamnoside, respectively. This stability of e llagic acid glycosides in heat-treated juices seems to be affected by thermal processing, since ellagic acid xyloside were no longer detected after 3 week storage and ellagi c acid rhamnoside decreased by 74% after 5 weeks in non-heat treated juices (Figures 6-7, 8). The changes of ellagitannins were monitored indirectly as total ellagic acid. Total e llagic acid in non-heat treated juices continuously decreased with storage and reta ined only 35% of its initial value after 5 weeks (Figure 6-9). This decrease in total ellagic acid was impact s of degradation of mainly ellagitannins, rather th an ellagic acid glycosides due to the low amounts of ellagic acid glycosides (< 1 mg/L) in solution. Da ta on ellagic acid deri vatives for control suggested that thermal pasteurization could be important factor on relative changes on ellagic acid precursors during storage thr ough impacts on natural enzymes in juice. Ellagic acid derivatives may not be primer targets for oxidati ve enzymes due to lack of o dihydroxyl in the structure (33, 81 ). However, thermal pasteurization may hinder the developments of o -quinones or secondary oxidation products from phenolic acids in grape juice ( 82 83 ), and affect to ella gitannins degradation. Conclusions Changes in ellagic acid derivatives of mus cadine juices were evaluated initially and after 5 weeks storage to determine primer precursor for free ellagic acid evolution. Through evaluating whole juice and each isolat es, ellagitannins seemed to more closely influence on free ellagic acid evolution rather than ellagic acid gl ycosides. Additionally, other miscellaneous components such as metal ions, soluble pectin or organic acid were likely to accelerate the free ellagic acid d ecrease via formation of sediments. Ascorbic acid fortification is probably non-harmful options to retain free ellagic acid in white

PAGE 93

82 juices. Thermal pasteurization was a significant factor for relative cha nges in ellagic acid derivatives during storage likely due to inactiv ate natural enzymes present in juices, but additional studies are required to directly evaluate the presence of enzymes and their activities in juices prepared by di fferent pasteurization schemes.

PAGE 94

83 CHAPTER 7 SUMMARY AND CONCLUSION Studies were conducted to provide imp roved information on ellagic acid derivatives including free ellagic acid, ella gic acid glycosides and ellagitannins in muscadine grape ( Vitis rotundifolia ) to understand the role of these compounds influencing on quality. Overall phytochemical s evaluations in different cultivars of muscadine grapes demonstrated that ellagic ac id derivatives were present in a wide range of concentrations and were influenced by ripening, physiology, and juice processing, resulting vary in antioxidant capacity. New blending schemes with Noble and Albemarle can be suggested for red muscadine grape jui ce or wine to produce high quality products in terms of high color intens ity and high contents of ella gic acid derivatives with corresponding high antioxidant capacity. The main antioxidants were isolated with a series of solid phase extraction and identified by application of advanced chromatographic techniques, PDA and MS detectors connecting to HPLC. Predominan t ellagic acid glycosides and flavonoid glycosides were determined their chemical id entities; ellagic acid glycoside, ellagic acid xyloside, ellagic acid rhamnoside, myrice tin rhamnoside, quercetin rhamnoside, and kaempferol rhamnoside in muscadine grape. In the case of ellagitannins, these methods were able to assess molecular weights of the respective fragments, but not exact chemical identities due to diversity of ellagitannins pr esent with varying functional groups. Using response surface methodology with tw o independent variables, time and temperature, successfully demonstrated that evolution of ellagic acid was a result of

PAGE 95

84 temperature dependent hydrolysis of ellagic acid glycosides and ellagitannins in both presence and absence of acid. This additional free ellagic acid in solution was likely to play a significant role for scavenging hydroxyl radical indica ted by high correlation (r=0.83) between two attributes. Enzymatic application with -glucosidase (E.C. 3.2.1.21) or tannase (E.C. 3.1.1.20) wa s tested to suggest options for hydrolysis of ellagic acid glycosides and ellagitannins. -glucos idase showed promise as a way to hydrolyze ellagic acid glycosides, result ing in high free ellagic acid content. However, tannase was not a feasible option for break down of ellagic acid precursors. Through evaluating whole juice and each isol ates, ellagitannins seemed to be the main precursor for free ellagic acid evolution since ellagic acid glycosides were evaluated relatively stable during storage. Additionally, it is possible that ellagic acid precipitation may be aided by binding other components forming insoluble precipitates such as proteins, short chain pectins, organic aci ds, or metal ions. Thermal processing for pasteurization increased free ellagic acid via el lagitannins hydrolysis, and also influenced the kinetic changes of ellagic acid derivative s during storage, possibl y due to inactivation of natural enzymes present in juices. Additional studies are re quired to directly evaluate the presence of enzymes and their activities in juices prepared by different pasteurization schemes.

PAGE 96

85 LIST OF REFERENCES 1. Lin, T. Y.; Vine, R. P. Identification and reduction of ellagic acid in muscadine grape juice. J. Food Sci 1990 55 1607-1609, 1613. 2. Garrido, V. M.; Sims, C. A.; Marshall, M. R.; Bates, R.P. Factors influencing ellagic acid precipitation in muscadine grape juice during storage. J. Food Sci. 1993, 58 193-196. 3. Olien, W. C. The muscadine grape: Bo tany, viticulture, history, and current industry. Hortsci. 1990 25 732-739. 4. Winkler, A. J.; Cook, J. A.; Klie wer, W. M.; Lider, L. A. General Viticulture Univ. of California Press; Berkeley; 1974. 5. Poling, E. B.; Mainland, C. M.; Earp, J. B. Muscadine grape production guide for North Carolina N.C., Agric. Ext. Serv. AG-94; 1984. 6. USDA (United States Department of Agricu ltre). Muscadine grapes-a fruit for the south Farmers’ Bull. 1965; pp 2157. 7. Olien, W. C. Muscadine – A cl assic southeastern fruit. Hortsci. 1990 25 726831. 8. Carroll, D. E. Muscadine grapes: Factors influencing product quality. Evaluation of quality of fruits and vegetables Ed. Pattee, H.E., AVI Publishing Co., Westport, CT. 1985 9. Woodroof, J. G.; Cecil, S. R.; DuPree, W.E. Processing muscadine grapes. Ga. Agric. Esp. Stn., Bull. [N.S.] 1956; 17:1-35. 10. Flora, L.F. Processing and quality ch aracteristics of muscadine grapes. J. Food Sci 1977, 42 935-938, 952. 11. Haslam, E. Symmetry and promiscu ity in procyanidin biochemistry. Phytochem. 1977 11 1207-1218. 12. Daniel, E. M.; Krupnick, A. S.; Heur, Y. H. ; Blinzler, J. A.; Nims, R. W.; Stoner, G. D. Extraction, stability, and quantitation of ellagic acid in various fruits and nuts. J. Food Comp. and Anal 1989 2 338-349.

PAGE 97

86 13. Amakura, Y.; Okada, M.; Tsuji, S. ; Tonogai, Y. High-performance liquid chromatographic determination with photodi ode array detection of ellagic acid in fresh and processed fruits. J. Chrom. A 2000 896 87-93. 14. Mullen, W.; Yokota, T.; Lean, M. E.; Crozie r, A. Analysis of ellagitannins and conjugates of ellagic acid and querce tin in raspberry fruits by LC-MSn Phytochem. 2003 64 617-624. 15. Mullen, W.; McGinn, J.; Lean, M. E.; MacLea n, M. R.; Gardner, P.; Duthie, G. G.; Yokota, T.; Crozier, A. Ellagitannins, flavonoids, and other phenolics in red raspberries and their contribution to an tioxidant capacity and vasorelaxation properties. J. Agric. Food Chem 2002 50 5191-5196. 16. Lee, J. H.; Talcott, S. T. Ellagic acid and ellagitannins affect on sedimentation in muscadine juice and wine. J. Agric. Food Chem. 2002 50 3971–3976. 17. Talcott, S. T.; Lee, J. H. Ellagic ac id and flavonoid antioxidant content of muscadine wine and juice. J. Agric. Food Chem. 2002 50 3186–3192. 18. Pastrana-Bonilla, E.; Akoh, C. C.; Sellappa n, S.; Krewer, G. Phenolic Content and Antioxidant Capacity of Muscadine Grapes. J. Agric. Food Chem 2003 51 54975503. 19. Mertens-Talcott, S. U.; Talcott, S. T.; Percival, S. S. Low concentrations of quercetin and ellagic acid synergistically influence proliferation, cytotoxicity and apoptosis in MOLT-4 Human Leukemia Cells. J. Nutr. 2003 133 2669-2674. 20. Stoner, G. D.; Morse, M. A. Isothiocyanate s and plant polyphenol s as inhibitors of lung and esophageal cancer. Cancer Lett. 1997 114 113-119. 21. Khanduja, K. L.; Gandhi, R. K.; Pathania, V.; Syal, N. Prevention of Nnitrosodiethylamine-induced lung tumorigenesis by ellagi c acid and quercetin in mice. Food Chem Toxicol. 1999 37 313-318. 22. Helm, R. F.; Zhentian, L.; Ranatunga, T.; Jervis, J.; Elder, T. Toward understanding monomeric ellagi tannin biosynthesis. In Plant polyphenols 2: Chemistry, biology, pharmacology, ecology. Ed. Gross, G.G., Hemingway, R.W., and Yoshida, T., Kluwer Academic/Plenum Publishers, New York, NY. 1999 23. Quideau, S.; Feldman, K. S. Ellagitannin chemistry. Chem. Rev. 1996 96 475-503. 24. Gill, M. I.; Toms-Barbern, F. A.; Hess-Pi erce, B.; Holcroft, D. M.; Kader, A. A. Antioxidant activity of pomegranate ju ice and its relationship with phenolic composition and processing. J. Agric. Food Chem 2000 48 4581-4589. 25. Machado, T. B.; Leal, I. C. R.; Amaral, A. C. F.; dos Santos, K. R. N.; da Silva, M. G.; Kuster, R. M. Antimicrobial ellagitannins of Punica granatum fruits J. Braz. Chem. Soc 2002 13 606-610.

PAGE 98

87 26. Salminen, J. P.; Ossipov, V.; Loponen, J.; Haukioja, E.; Pihlaja, K. Characterization of hydrolysable tannins from leaves of Betula pubescens by highperformance liquid chromatography-mass spectrometry. J. Chromatogr. A 1999 864 283-291. 27. Foo, L. Y. Amarinic acid and related ellagitannins from Phyllanthus amarus Phytochem 1995 39 217-224. 28. Boyle, J. A.; Hsu, L. Identification and qua ntitation of ellagic acid in muscadine grape juice. Am. J. Enol. Vitic 1990 41 43-47. 29. Musingo, M. N.; Sims, C. A.; Bates, R. P.; O’keefe, S. F.; Lamikanra, O. Changes in ellagic acid and other phenols in muscadine grape ( Vitis rotundifolia ) juices and wines. Am. J. Enol. Vitic 2001 52 109-114. 30. Sims, C. A.; Bates, R. P. Effects of skin fermentation time on the phenols, anthocyanins, ellagic acid sediment. Am. J. Enol. Vitic. 1994 45 56-62. 31. Hukumoto, L. R.; Mazza, G. Assessing antioxidants are affected by colloidal properties of phenolic compounds. J. Agric. Food Chem. 2000 48 3957–3604. 32. Goldy, R. G.; Ballinger, W. E.; Maness, E. P. Fruit anthocyanin content of some Euvitis X Vitis rotundifolia hybrids. J. Amer. Soc. Hort. Sci 1986 111 955-960. 33. Talcott, S. T.; Brenes, C. H.; Pires, D. M.; Del Pozo-Insfran, D. Phytochemical stability and color retention of copigmen ted and processed muscadine grape juice. J. Agric. Food Chem. 2003 51 957–963. 34. Del Pozo-Insfran, D.; Del Follo-Martinez, A.; Brenes, C. H.; Talcott, S. T. Phytochemical stability of an ascorbic acid fortified and copigmented muscadine grape juice as affected by high hydrostatic pressure processing Submitted in J. Agric. Food Chem 2004. 35. Gomez, M. R. A.; Ledward, D. A. Effect of high pressure treatment on the activity of some polyphenoloxidases Food Chem. 1996 56 1-5. 36. Sun, N.; Lee, S.; Kim, B. S. Effect of high-pressure treatment on the molecular properties of mushroom polyphenoloxidase. Lebens. Wiss.Technol. 2002 35 315318. 37. Zabetakis, I.; Leclerc, D.; Kajda, P. Th e effect of high hydrostatic pressure on the strawberry anthocyanins. J. Agric. Food Chem 2000 40 2749-2754. 38. Timberlake, C. F.; Bridle, P. Anthocyanins. In Developments in food colours – 1 Ed. Walford, J., Applied Science Publishers LTD., London, 1980; pp. 115-143.

PAGE 99

88 39. Sims, C. A.; Morris, J. R. A comparison of the color components and color stability of red wine from Noble and Cabern et Sauvignon at various pH levels. Am. J. Enol. Vitic 1985 36 181-184. 40. Swain, T.; Hillis, W. E. Th e phenolic constituents of Prumus domestica I. The quantitative analysis of phenolic constituents. J. Sci. Food Agric. 1959 10 63-68. 41. Talcott, S. T.; Percival, S. S.; Pittet-Moore, J.; Celoria, C. Phytochemical Composition and Antioxidant Stability of Fortified Yellow Passion Fruit ( Passiflora edulis ). J. Agric. Food Chem. 2003 51 935 –941. 42. Ou, B.; Hampsch-Woodill, M.; Prior, R. L. Development and validation of an improves oxygen radical absorbance capaci ty assay using fluorescein as the fluorescent probe. J. Agric. Food Chem. 2001 49 4619-4626. 43. Cao, G.; Verdon, C. P.; Wu, A. H. B.; Wang, H.; Prior, R. L. Automated oxygen radical absorbance capacity assay using the COBAS FARA II. Clin Chem. 1995 41 1738-1744. 44. SAS Institute, Inc., SAS Campus Drive, Cary, NC, 1996. 45. Zafrilla, P.; Ferreres, F.; Toms-Barbern, F. A. Effect of processing and storage on the antioxidant ellagic acid derivative s and flavonoids of red raspberry ( Rubus idaeus ) Jams. J. Agric. Food Chem 2001 49 3651-3655. 46. Rommel, A.; Wrolstad, R. E. Ellagic aci d content of red raspberry juice as influenced by cultivar, proces sing, and environmental factors. J. Agric. Food Chem 1993 41 1951-1960. 47. Nunan, K. J.; Sims, I. M.; Bacic, A., Robinson, S. P., Fincher G. B. Changes in cell wall composition during ripe ning of grape berries. Plant Physiol 1998 118 783–792. 48. Sleas, G. J.; Diamandis, E. P.; Goldberg, D. M. Resveratol; A molecule whose time has come? And gone? Clin. Biochem 1997 30 91-113. 49. Zoecklein, B. W.; Fugelsang, K. C.; Gump, B. H.; Nury, F. S. Phenolic compounds and wine color. In Wine analysis and production Chapmean & Hall, New York, NY. 1994. 50. Harkness, E. M.; Vine, R. P.; Coign, M. J.; He itz, J. R. The effect of different must treatments on muscadine juice quantity and quality Am. J. Enol. Vitic. Meeting Rpt 1987, 10. 51. Gazzani, G.; Papetti, A.; Massolini, G.; Da glia, M. Antiand prooxidant activity of water soluble components of some common diet vegetables and the effect of thermal treatment. J. Agric. Food Chem. 1998 46 4118-4122.

PAGE 100

89 52. Meyer, A. S.; Heinone, M.; Frankel, E. N. Antioxidant interactions of catehin, cyanidin, caffeic acid, quercetin, and ellagic acid on human LDL oxidation. Food Chem 1998 61 71-75. 53. Adamson, G. E.; Lazarus, S. A.; Mitchell, A. E.; Prior, R. L.; Cao, G.; Jacobs, P. H.; Kremers, B. G; Hammerstones, J. F.; Rucker, R. B.; Ritter, K. A.; Schmitz. H. H. HPLC method for the quantification of procyanidins in cocoa and chocolate samples and correlation to total antioxidant capacity. J. Agric. Food Chem. 1999 47 4184–4188. 54. McPhail, D. B.; Hartley, R. C.; Gardne r, P. T.; Duthie, G. G. Kinetic and stoichiometric assessment of the antioxidant activity of flavonoids by electron spin resonance spectroscopy. J. Agric. Food Chem. 2003 51 1684–1690. 55. Scalbert, A.; Duval, L.; Peng, S.; Monties, B. Polyphenols of Quercus robur L. II. Preparative isolation by low-pressure a nd high-pressure liquid chromatography of heartwood ella gitannins. J. Chrom 1990 502 107-119. 56. Lei, Z. Monomeric ellagitannins in oa ks and sweetgum. Doctoral Dissertation, 2002, Virginia Polytechnic Institute and State University. Blacksburg, Virginia 57. Atoui, A. K.; Mansouri, A.; Boskou, G.; Kefala s, P. Tea and herbal infusions: Their antioxidant activity and phe nolic profile. In Press Food Chem 2004 58. Montgomery, D. C. Response surface met hods and other approaches to process optimization. In Design and Analysis of Experiments. 5th Ed John Wiley & Sons, Inc. US. 2001. 59. Robbins, R. J. Phenolic acids in foods : An overview of analytical methodology. J. Agric. Food Chem. 2003 51 2866-2887. 60. Gao, L.; Mazza, G. Rapid method of complete chemical characterization of simple and acylated anthocyanins by high-pe rformance liquid chromatography and capillary gas-liquid chromatography. J. Agric. Food Chem. 1994 42 118-125. 61. Merken, H. M.; Merken, C. D.; Beecher, G. R. Kinetics method for the quantitation of anthocyanidins, flavonols, and flavones in foods. J. Agric. Food Chem. 2001 49 2727-2732. 62. Wang, G.; Kuan, S. S.; Francis, O. J.; Ware, G. M.; Carman, A.S. A simplified HPLC method for the determination of phytoe strogens in soybean and its processed products. J. Agric. Food Chem. 1990 38 185-190. 63. Khknen, M. P.; Heinonen, M. Antioxidant activity of anthoc yanins and their aglycons. J. Agric. Food Chem 2003 51 628-623. 64. Wang, H.; Cao, G.; Pror, R. L. Oxygen radi cal absorbing capacity of anthocyanins. J. Agric. Food Chem 1997 45 304-309.

PAGE 101

90 65. Spagna, G.; Barbagallo, R. N.; Pifferi, P. G.; Blanco, R. M.; Guisan, J. M. Stabilization of a b-glucosidase from Aspergillus niger by binding to an amine agarose gel. J. Mol. Catalysis B: Enzymatic 2000 11 63–69 66. Seeram, N.; Lee, R; Hardy, M.; Heber, D. Rapid large scal e purification of ellagitannins form pomegranate husk, a by-product of th e commercial juice industry. In press Separation and Purification Technology 2004. 67. Cerd, B.; Cern, J. J.; Toms-Barbern, F. A.; Espn, J. C. Repeated oral administration of high doses of the pomegran ate ellagitannin punical agin to rats for 37 days is not toxic. J. Agric. Food Chem. 2003 51 3493-3501. 68. Ossipov, V.; Salminen, J.; Ossipova, S.; Hauki oja, E.; Pihlaja, K. Gallic acid and hydrolysable tannins are formed in birch leaves from an intermediate compound of the shikimate pathway. Biochem Sys. and Ecol. 2003 31 3-16. 69. Luo, H.; Chen, L.; Li, Z.; Ding, Z.; Xu, X. Frontal immunoaffinity chromatography with mass spectrometric detection: A me thod for finding active compounds from traditional Chinese herbs. Anal. Chem. 2003 75 3994-3998. 70. Lee, S.; Tanaka, T.; Nonaka, G.; Nishioka, I. Structure and Biogenesis of Jolkinin, a high oxygenated ellagitannins from Euphorbia jolkinii In press J. Nat. Pro. 2004. 71. Nawwar, M. A. M..; Marzouk, M. S.; Ni gge, W.; Linscheid, M. High-performance liquid chromatographic/elec trospray ionization mass sp ectrometric screening for polyphenolic compounds of Epilobium hirsutum -The structure of the unique ellagitannin Epilobamide-A. J. Mass Spectrometry 1997 32 645-654. 72. Del Rio, D.; Stewart, A. J.; Mullen, W.; Bu rns, J.; Lean, M. E. J.; Brighenti, F.; Crozier, A. HPLC-MSn analysis of phenolic compounds and purine alkaloids in green and black tea. J. Agric. Food Chem. 2004 52 2807-2815. 73. Schieber, A.; Berardini, N.; Carle, R. Identification of fla vonol and xanthone glycosides from mango (Mangifera indi ca L.Cv. “Tommy Atkins”) peels by highperformance liquid chromatography-electr ospray ionization mass spectrometry. J. Agric. Food Chem. 2003 51 5006-5011. 74. Rice-Evans, C. A.; Miller, N. J.; Paganga, G. Structure-antioxidant activity relationships of flavonoids and phenolic acids. Free Redic. Biol. Med. 1996 20 933-956. 75. Belmares, R.; Contreras-Esquivel, J. C. Rodrguez-Herrera, R.; Coronel, A. R. Aguilar, C. N. Microbial production of ta nnase: an enzyme with potential use in food industry. Lebensm.-Wiss. U.-Technol 2004, In press. 76. Bajpai, B.; Patill, S. Introductionof ta nnin acyl hydrolase (EC. 3.1.1.20) activity in some members of fungi imperfecti. Enzyme Microbial Technology 1997, 20 612614.

PAGE 102

91 77. Lekha, P. K.; Lonsane, B. K. Production a nd application of tannin acyl hydrolase: State of the art. Adv. in Appl. Microbio. 1997 44, 215-260. 78. Nelson, P. F.; Hewitt, D. G. Ellagic acid and the pulping of eucalypts II. Complexes of ellagic acid with cation and solvents. Holzforschung 1965 19 102105. 79. Thulstrup, P. W.; Thormann, T.; Spanget-La rsen, J.; Bisgaard, H. C. Interaction between ellagic acid and calf thymus DNA studied with flow linear dichroism UVVIS spectroscopy. Biochem. Biophy. Re. Comm ., 1999 265 416-421. 80. Cilliers, J. J. L.; Singleton, V. L. Char acterization of the products of nonenzymic autoxidative phenolic reactions in a caffeic acid model system. J. Agric. Food Chem. 1991 39 12987-1303. 81. Sarni-Manchado, P.; Cheynier, V.; Mout ounet, M. Reaction of polyphenoloxidase generated caftaric acid o -quinone with malvidin 3O -glucoside. Phytochem. 1997 45 1365 1369. 82. Skrede, G.; Wrolstad, R. E.; Durst, R. W. Changes in anthocyanins and polyphenolics during juice processi ng of highbush blueberries ( Vaccinium corymbosum L.). J. Food Sci. 2000 65 357 364. 83. Kader, F.; Rovel, B.; Girardin, M.; Metc he, M. Mechanism of browning in fresh highbush blueberry fruit ( Vaccinium corymbosum L.). Role of blueberry polyphenol oxidase, chlorogeni c acid and anthocyanins. J. Sci. Food Agric. 1997 74 31 34. 84. Poei-Langston, M.S.; Wrolstad, R.E. Colo r degradation of ascorbic-anthocyanin-flavanol model system J. Food Sci. 1981 46 1218-1222. 85. Calvi, J. P.; Francis, F. J. Stability of Concord grape ( V. labrusca ) anthocyanins in model systems. J. Food Sci. 1978 43 1448 1456.

PAGE 103

92 BIOGRAPHICAL SKETCH Joonhee Lee received her BachelorÂ’s de gree from Songsim WomenÂ’s University, Korea, majoring in Food Science and Huma n Nutrition in 1995. And she earned Master of Science degree with emphasis on food scienc e from Catholic University of Korea in 1997. After 1 year experience as research assist ant at Food Hygiene Institution of Korea, she moved to University of Florida to join Food Science and Human Nutrition Department. She has continued her degr ee of doctor of philosophy emphasizing food science since she earned Ma ster of Science in 2001.