Citation
Emerging Technologies and Strategies to Enhance Anthocyanin Stability

Material Information

Title:
Emerging Technologies and Strategies to Enhance Anthocyanin Stability
Copyright Date:
2008

Subjects

Subjects / Keywords:
Antioxidants ( jstor )
Carbon dioxide ( jstor )
Enzymes ( jstor )
Food ( jstor )
Grape juice ( jstor )
Grapes ( jstor )
Juices ( jstor )
Pigments ( jstor )
Rosemary ( jstor )
Thyme ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Embargo Date:
7/24/2006

Downloads

This item has the following downloads:


Full Text











EMERGING TECHNOLOGIES AND STRATEGIES TO ENHANCE
ANTHOCYANIN STABILITY














By

DAVID DEL POZO-INSFRAN


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


2006

































Copyright 2006

by

David Del Pozo-Insfran
















ACKNOWLEDGMENTS

My deepest recognition and gratitude goes to my role models in life: my beloved parents

(Dr. Myriam D. Insfran and Dr. David Del Pozo) and my dearest sister (Piu). They have helped

me in any imaginable way to achieve everything I have in life and to fulfill all my dreams. They

have been an inexhaustible source of love, inspiration, and encouragement all my life. They were

there to catch me when I fell, support me when I needed it, and cheer me up all the step of the

way. Just hearing their words-"How can help you?"- made my day. Words can not express all the

gratitude and love for my family. The present work could have not been possible without them.

I would like to thank Elisa Del Pozo R., Teresa Rendon, and Robin K. Minor for their

unconditional support and love throughout my life. These wonderful persons have a very special

place in my heart.

I would also like to acknowledge the unconditional support of my supervisory committee

chair, Dr. Stephen Talcott, and my mentors, Dr. Carmen Hernandez Brenes and Dr. Ronald H.

Schmidt. Through their guidance, wisdom, and never-ending care they have helped me to

achieve all my goals and accomplishments in my professional career. I sincerely appreciate the

help offered by the members of my supervising committee: Dr. Murat O. Balaban, Dr. Bala

Rathinasabapathi, Dr. Susan S. Percival, and Dr. Jesse Gregory.

Finally, I would also like to give special thanks to Flor Nunez, Rena Schonbrun, Minna

Schuster, Joon H. Lee, Janna Underhill, Gillian Folkes, Asli Odabasi, Sibel Damar, Youngmok

Kim, Lanier Fender, Chris Duncan, Lisbeth Pacheco, and Angela Lounds, for all their support

and love.


















TABLE OF CONTENTS


page

ACKNOWLEDGMENT S ................. ................. iii.._._. ....


LI ST OF T ABLE S ........._..... .......... ............... vii...

LIST OF FIGURES .............. .................... ix


AB STRAC T ................ .............. xi


CHAPTER


1 INTRODUCTION ................. ...............1.......... ......

Ju sti fi cati on ................. ...............1............ ...
Obj ectives ................. ...............2.......... ......

2 LITERATURE REVIEW .............. ...............4.....


Anthocyanins ................... ........ ..... .. .. ...........5
Anthocyanins and Intermolecular Copigmentation Reactions............__ ............_9
Anthocyanin Stability .............. ..... ... ..._.._ .............1
Phytonutrient Stability and Intermolecular Copigmentation.............................1
Sensory Attributes of Anthocyanin-containing Beverages as Affected by the
Addition of Polyphenolic Cofactors. ................... .... ....... ........._.............16
Novel Processing Technologies That May Enhance Anthocyanin Stability ..............18
High Hydrostatic Pressure (HHP) Processing ....._____ .........__ ..............19
HHP and microbial inactivation ...._ ......_____ ...... ...___..........2
The effect of HHP in food components .............. ...............22....
Dense Phase-CO2 Pasteurization............... .............2
DP-CO2 and microbial inactivation............... ..............2
DP-CO2 and enzymatic inactivation............... ..............3

3 PHYTOCHEMICAL COMPOSITION AND PIGMENT STABILITY OF ACAI
(EUTERPE OLERACEA MART.)............... ...............36.

Introducti on ........._._ ...... .___ ...............36....
Materials and Methods .............. ...............38....
M material s ........._._ ..... ............... 8....
Color Stability .............. ...............39....
Effect of Copigmentation .............. ...............39....












Phytochemical Analyses............... ...............40
Statistical Analysis .............. ...............42....
Results and Discussion .............. .. ........... ..........4
Anthocyanin and Polyphenolic Characterization ................. .......................42
Antioxidant Capacity ................... .... .._._. .. ....___ ... .... ........4
Color Stability as Affected by Hydrogen Peroxide and Temperature ........._......46
Color Stability in the Presence of Ascorbic Acid and Natural Cofactors ...........49
Conclusion ............ _...... ._ ...............55....


4 STABILITY OF COPIGMENTED ANTHOCYANINTS AND ASCORBIC ACID
INT MUSCADINE GRAPE JUICE PROCESS SED BY HIGH HYDROSTATIC
PRE S SURE ................. ...............57........... ....


Introducti on ................. ...............57.................
M materials and M ethods .............. ...............59....
Materials and Processing ................. ........... ...............59......
PPO activity during juice extraction .............. ...............59....
Juice extraction and processing ................. ...............60................
Chemical Analyses .............. ...............61....
Statistical Analysis .............. ...............62....
Results and Discussion .................. ........ ... .. ........6
Initial Effects of Copigmentation in Muscadine Grape Juice .............................63
PPO Activity as Affected by HHP Processing ............__......._ ..............64
Phytochemical Stability Following HHP Processing ............_.. .........__.....68
Phytochemical Retention During Storage .............. ...............74....
Conclusions............... ..............7


5 PASTEURIZATION AND QUALITY RETENTION OF DENSE PHASE-CO2
PROCES SED MUSCADINE GRAPE JUICE ......____ ..... ... ._ ............. ..78


Introducti on ............ ..... ._ ...............78....
Material and Methods ............ ..... ._ ...............79...
M material s .............. .. ..._ ...............79....
Processing Equipment .............. ...............79....
Microbial Inactivation Study ............_.......____ ......__ ............8
Phytochemical and Microbial Stability Study ................. ....._._ ...............81
Physicochemical and Microbial Analyses............... ...............82
Sensory Evaluation ................. ...............83.................
Statistical Analysis .............. ...............83....
Results and Discussion ................. ...............84........... ....
Microbial Inactivation Study ................. ............. ...............84......
Phytochemical and Microbial Stability Study ........................... ...............86
Sensory evaluation............... ...............9
Conclusions............... ..............9











6 ENHANCING THE RETENTION OF PHYTOCHEMICALS AND
ORGANOLEPTIC ATTRIBUTES IN MUSCADINE GRAPE JUICE BY
DENSE PHASE-CO2 PROCESS SING AND COPIGMENTATION ................... .......95

Introducti on ................. ...............95.................
M materials and M ethods .............. ...............97....
Materials and Processing ................. .......... ...............97......
Physicochemical and Microbial Analyses............... ...............98
Sensory Evaluation ................. ...............98.................
Statistical Analysis .............. ...............99....
Results and Discussion ................... ......._. ... ... ... ... ... ...... ...........10
Initial Effects of Copigmentation and Ascorbic Acid Fortification ..................1 00
Phytochemical Changes Due to Thermal and DP-CO2 Processing ................... 101
Organoleptic Changes Due to Addition of Thyme Polyphenolic Cofactors .....102
Phytochemical and Microbial Changes During Refrigerated Storage ..............103
Conclusions............... ..............11


7 INACTIVATION OF POLYPHENOL OXIDASE IN MUSCADINE GRAPE
JUICE B Y DEN SE PHA SE- CO2 PRO CE S SING ................. ................. ...... 11 6

Introducti on ................. ...............116................
M materials and M ethods ................. ...............117...............
M materials .................. ......... ....... .. ....... ............11
Effect of DP-CO2 Processing on PPO activity .................. ....... .... ..... ............118
Storage Stability of Muscadine Juice with Residual PPO Activity ................... 118
Chemical Analyses ................ ...............119................
Statistical Analysis ................. ...............119......... ......
Results and Discussion ................ ........ ......... ..... ...............119...
Effect of DP-CO2 Processing on PPO Activity .................. ....... .... ....... ..........1 19
Storage Stability of Muscadine Juice with Residual PPO Activity ................... 125
Conclusions............... ..............12


8 SUMMARY AND CONCLUSIONS ................ ...............130...............

LI ST OF REFERENCE S ................. ...............13. 1......... ....


BIOGRAPHICAL SKETCH ................. ...............144......... ......

















LIST OF TABLES


Table pg

2-1 Effect of rosemary extract (0, 0.1i, 0.2, 0.4% v/v) on color, aroma, and flavor
attributes of a commercial strawberry cocktail juice ................. ............ .........18

3-1 Anthocyanin and polyphenolic content (mg/L fresh pulp) of agai (Euterpe
oleracea M art.). ........._ ...... .... ...............45....

3-2 The effect of hydrogen peroxide (30 mmol/L) and temperature on kinetic
parameters of color degradation for different anthocyanin sources. ........................48

3-3 Percent monomeric anthocyanins and CIE color attributes of a juice model
system (pH 3.5, 100 mg/L sucrose) prepared with different pigment sources.........51

3-4 The effect of ascorbic acid and naturally occurring polyphenolic cofactors on
kinetic parameters of anthocyanin degradation during storage at 37 oC of in vitro
models systems prepared with different pigment sources............... .................5

4-1 The effect of rosemary and thyme cofactors and ascorbic acid fortification on
the anthocyanin content of unprocessed and high hydrostatic pressure processed
muscadine grape j uice. ............. ...............69.....

4-2 The effect of rosemary and thyme cofactors and ascorbic acid fortification on
the anthocyanin content and antioxidant capacity of high hydrostatic pressure
processed muscadine grape juice after 21 days of storage at 240C. ................... ......77

5-1 The effect of heat or DP- CO2 paSteurization on the total anthocyanin, soluble
phenolic, and antioxidant content of unprocessed muscadine grape juice............_...88

6-1 The effect of thyme cofactors and ascorbic acid fortification on the total
anthocyanin, soluble phenolic and antioxidant content of unprocessed, heat, and
DP-CO2 paSteurized muscadine grape juice ................. ................ ......... .104

6-2 Effect of thyme cofactors and ascorbic acid fortification on first-order
degradation kinetic parameters of anthocyanins present in heat or DP- CO2
processed muscadine grape juice during storage at 4 oC. ................ ................. 105

6-3 Effect of thyme cofactors and ascorbic acid on first-order degradation kinetic
parameters of soluble phenolics in heat or DP- CO2 prOcessed muscadine grape
juice during storage at 4 oC. ............. ...............106....










6-4 Effect of thyme cofactors and ascorbic acid fortification on first-order
degradation kinetic parameters of antioxidant capacity in heat or DP- CO2
processed muscadine grape juice during storage at 4 oC. ................ ..................107

6-5 Effect of thyme cofactors on first-order degradation kinetic parameters of total
ascorbic acid present in heat or DP- CO2 prOcessed muscadine grape juice
during storage at 4 oC............... ...............108..

7-1 Individual and total anthocyanin content of unprocessed muscadine grape juice
as affected by DP-CO2 prOcessing pressure and CO2 COntent ............ ................120

















LIST OF FIGURES


figure pg

2-1 Structure of the six basic anthocyanindins (A), along with their different
positional glycosides (3-glycosides, B; 3,5-glycosides, C) ................. ................. .7

2-2 Schematic representing a simple and an acylated cyanidin glycoside. ......................8

2-3 Schematic representation of anthocyanin self-association and copigmentation
reactions. ............. ...............9.....

3-1 HPLC chromatogram of anthocyanin 3 -glucosides (A) monitored at 520 nm and
their aglycones (B) present in agai (Euterpe oleracea Mart.) ................ ...............43

3-2 HPLC chromatogram of phenolic acids and flavonoids present in agai (Euterpe
oleracea M art.). ................. ...............43.__._......

3-3 Antioxidant capacity of different phytochemical fractions of agai (Euterpe
oleracea M art.). ........._ ...... .... ...............47....

4-1 Antioxidant capacity of muscadine grape juice as affected by HHP processing
and copigmentation with rosemary or thyme cofactors in the absence or
presence of ascorbic acid (450 mg/L). ............. ...............65.....

4-2 Polyphenoloxidase activity (A), and browning index (B) of muscadine grape
juice as influenced by preheating time and temperature prior to juice extraction. ..66

4-3 Polyphenoloxidase activity in muscadine grape juice as affected by HHP
processing and copigmentation in the absence (A) or presence (B) of ascorbic
acid. ............. ...............67.....

4-4 Total anthocyanin content of muscadine grape juice as affected by HHP
processing and copigmentation with rosemary or thyme cofactors in the absence
or presence of ascorbic acid. ............. ...............70.....

4-5 Total ascorbic acid content of muscadine grape juice as affected by HHP
processing and copigmentation with rosemary or thyme polyphenolic cofactors. ..71

5-1 Schematic diagram of the DP-CO2 prOcessing equipment ................. ................. 80










5-2 Inactivation of yeast/molds and total aerobic microorganisms after DP-CO2
pasteurization of muscadine juice as influenced by processing pressure and CO2
content. ............. ...............87.....

5-3 Scanning electron micrographs of naturally occurring yeast cells in muscadine
juice before (A) and after DP-CO2 at 34.5 MPa and 16% CO2 (B). ........................87

5-4 Total anthocyanin (A) and antioxidant content (B) of heat and DP-CO2
pasteurized muscadine juice during refrigerated storage (4 oC). ............. ................93

5-5 Total soluble phenolic content of heat and DP-CO2 paSteurized muscadine juice
during refrigerated storage (1-10 weeks at 4 oC). ............. .....................9

5-6 Yeast/mold counts of heat and DP-CO2 paSteurized muscadine juice during
refrigerated storage (4 oC). ............. ...............94.....

6-1 Total anthocyanin content of muscadine grape juice without and with ascorbic
acid during refrigerated storage as affected by heat and pasteurization and the
addition of thyme cofactors. ................. ...._._ ...._._ ............1

6-2 Total soluble phenolic content of muscadine grape juice without and with
ascorbic acid during refrigerated storage as affected by heat and DP-CO2
pasteurization, and the addition of thyme cofactors ........._._ ...... .._._ ...........112

6-3 Antioxidant capacity of muscadine grape juice without and with ascorbic acid
during refrigerated storage as affected by heat and DP-CO2 paSteurization, and
the addition of thyme cofactors. .............. ...............113...._.__ ....

6-4 Total ascorbic acid content of muscadine grape juice during refrigerated storage
as affected by heat and DP-CO2 paSteurization, and the addition of thyme
cofactors. ........._ ..... ._ ...............114....

7-1 Effect of DP-CO2 at different processing pressures and CO2 leVOIS on residual
PPO activity (A) and resultant anthocyanin losses (B) in muscadine grape juice. 121

7-2 Effect of DP-CO2 prOcessing pressures and CO2 leVOIS on PPO-induced losses
in soluble phenolics (A) and antioxidant capacity (B) in muscadine grape juice..122

7-3 Total anthocyanin, soluble phenolics, and antioxidant capacity content of DP-
CO2 prOcessed muscadine grape juice during refrigerated storage as affected by
processing CO2 COntent and initial PPO activity ................. .........................116
















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

EMERGING TECHNOLOGIES AND STRATEGIES TO ENHANCE
ANTHOCYANIN STABILITY

By

David Del Pozo-Insfran

May 2006

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

Anthocyanins are polyphenolic compounds that are responsible for the bright blue

and red colors of many foods and act as phytochemical antioxidants with potential health-

related benefits. Recent shifts in consumer preference for natural pigments have focused

on applications of anthocyanins as suitable replacements for certified colorants. However

their relative high cost and poor stability are factors that limit their commercial

application. Due to these limitations, the food industry is constantly looking for novel,

inexpensive and stable sources of these natural colorants. Therefore, this study first

determined the phytochemical composition and stability of agai anthocyanins as a new

source of anthocyanins.

Of identical need for the food industry is the exploration of strategies and/or

technologies that can serve to alleviate the limitations of anthocyanins. High hydrostatic

pressure processing (HHP) and dense phase-carbon dioxide pasteurization (DP-CO2) are

non-thermal processing technologies that may lessen detrimental effects to anthocyanins









and other phytochemicals. However, a downside of these technologies is the presence

and/or activation of enzymes following processing that may be responsible for oxidative

degradation. This study investigated the role of polyphenol oxidase (PPO) in

phytochemical degradation during HHP processing of muscadine grape juice and

established a potential remediation strategy using polyphenolic cofactors from two plant

sources: rosemary and thyme. Cofactor addition not only increased juice color and

antioxidant activity but also reduced anthocyanin, polyphenolic, and ascorbic acid losses.

DP-CO2 WAS also evaluated as a potential non-thermal pasteurization technology

with results concluding that this process served to protect anthocyanins and antioxidant

levels without comprising juice sensory attributes. However, microbial stability of DP-

CO2 juicOS was only comparable to heat-pasteurized counterparts for >5 weeks at 4 oC.

Due to the preceding evidence, the addition of thyme polyphenolic cofactors along

with the DP-CO2 prOcess was evaluated as a combined strategy to decrease

phytochemical and antioxidant losses that occur in anthocyanin-containing beverages.

Results concluded that DP-CO2 and addition of thyme cofactors served to increase

phytochemical stability of muscadine juice without affecting juice sensory attributes.

Cofactor addition also masked the detrimental color fading that occurred during storage.

This study also showed that partial inactivation of PPO can be obtained by DP-

CO2 and that processing CO2 leVOIS was the main processing variable influencing PPO

activity as well as polyphenolic and antioxidant retention in muscadine juice.















CHAPTER 1
INTTRODUCTION

Justification

Anthocyanins are polyphenolic compounds that are responsible for the bright blue

to red colors of foods and act as phytochemical antioxidants imparting important health-

related benefits and nutraceutical properties. Due to current trends in consumer

preferences for natural pigments, these compounds are considered as potential

replacements for certain certified and banned dyes. However, many limitations exist for

their commercial application due to high raw material costs and their poor stability during

processing and storage. Developing strategies and technologies that serve to alleviate

these limitations is thus vital for economic growth of the U. S. food industry, not only

improving quality attributes and phytonutrient stability of anthocyanin-containing

products, but also possibly contributing to improve public health through increasing

phytonutrient and health promoting agents intake in food products.

High hydrostatic pressure processing (HHP) and dense phase carbon dioxide

pasteurization (DP-CO2) are pfOmising alternatives to traditional thermal pasteurization

technologies and may lessen detrimental effects to anthocyanins and other thermolabile

phytonutrients during processing and storage (Gomez and Ledward, 1996; Zabetakis et

al., 2000; Sun et al., 2002). However, a downside of these technologies is the presence

and/or activation of residual enzymes following processing that may be partially

responsible for oxidative degradation. Therefore, associated changes in phytonutrient

stability due to processing are of interest and will be investigated in the present study.









Furthermore, comparisons between non-thermal and thermal processes are one way to

assess the benefits offered by these novel processing technologies.

Previous studies have shown that formation of intermolecular copigmentation

complexes between anthocyanins and exogenously added polyphenolic cofactors could

assert a protective effect against anthocyanin, antioxidant capacity, and ascorbic acid

degradation in both model and juice systems (Talcott et al., 2003; Brenes et al., 2004;

Eiro and Heinonen, 2002; Malien-Aubert et al., 2001; Del Pozo-Insfran et al., 2004). In

addition to preventing quality and nutritional losses, copigmentation also increases

anthocyanin color intensity and antioxidant content of beverages, and masks the

detrimental color changes that take place during processing and storage. Consequently,

intermolecular copigmentation could be used as an important remediation strategy which

attenuates phytonutrient degradation that takes place in anthocyanin containing juice

systems during processing and storage. Therefore, the present study will also evaluate the

phytonutrient stability and sensory properties of fortified juice and beverage systems

containing copigmented anthocyanins in an effort to add economic value, reduce

oxidation, and maintain nutrient stability.

Obj ectives

The obj ectives of the present research work were:

1. To characterize the maj or anthocyanin and polyphenolic compounds present in
agai and to determine the potential usage of agai as a novel anthocyanin source.
2. To assess muscadine grape juice (Vitis rotundifolia) for phytochemical stability
following HHP processing and ascorbic acid fortification, and to investigate the
effect of exogenously added polyphenolic cofactors purified from rosemary
(Rosmarinus officinalis) and thyme (Thymus vulgaris) as a means to improve
overall phytochemical stability.
3. To determine the microbial destruction, phytochemical stability, and sensory
attributes of DP-CO2 prOcessed muscadine grape juice.









4. To determine the phytochemical retention and sensory properties of an ascorbic
acid fortified and copigmented muscadine grape juice following DP-CO2 and
subsequent storage.
5. To determine the effect of DP-CO2 On PPO activity and its consequent effect on
polyphenolic and antioxidant changes in muscadine juice.















CHAPTER 2
LITERATURE REVIEW

A steady increase in the development of natural food colorants and functional food

sources has been observed in recent years, not only due to consumer preferences for

natural pigments but also for their health-related benefits and nutraceutical properties

(Frankel et al., 1995; Meyer et al., 1997; Skrede et al., 2000). Anthocyanins are a viable

replacement for synthetic colorants due to their bright, attractive colors and water

solubility, which allows their incorporation into a variety of food systems (Rodriguez-

Saona et al., 1999). However, many limitations exist for the commercial application of

anthocyanins due to high raw material costs and their poor stability. Pigment stability

may be affected by chemical structure, concentration, pH, temperature, oxygen, light,

presence of cofactors, and polymeric forms. Furthermore, beverages containing

anthocyanins and ascorbic acid are known to be mutually destructive in the presence of

oxygen, which limits fortification in fruit juices and in products containing anthocyanins

(Frankel et al., 1995; Meyer et al., 1997; Rodriguez-Saona et al., 1999; Skrede et al.,

2000). Developing strategies and technologies that serve to alleviate these limitations is

thus vital for economic growth of the U. S. food industry, not only improving quality

attributes and phytonutrient stability of anthocyanin-containing products, but as well as

contributing to improve public health through increasing phytonutrient and health

promoting agents intake in food products.

Exploration of a system to stabilize anthocyanins from color loss and oxidation,

and the effects of ascorbic acid fortification, are important if these pigments present in









juices and beverages are to compete in a market dominated by certified dyes such FD&C

Red #40 and Red #3. To alleviate some of these problems, the present research proj ect

suggests a strategy by which the addition of exogenous anthocyanin cofactors (water-

soluble thyme and rosemary extracts), to not only attenuate degradation of anthocyanins

but also provide a means by which phytonutrient degradation can be prevented.

Furthermore, novel technologies such as high hydrostatic pressure and dense phase-CO2

pasteurization are non-thermal processing methods which ensure microbial destruction

and may extend shelf life of products without having adverse effects on their quality

attributes such as flavor, color, and phytonutrient retention. These research studies are

important due to the high demand for fruit juice, of which sales reached nearing $4.5

billion in 2001 of which over a quarter were likely to contain anthocyanins from various

fruit and vegetable sources. Moreover, there is an interest on the phytonutrient stability of

fortified beverages systems which have been consumed with a substantial increase during

the last 24 months By monitoring the chemoprotective and sensory aspects of

exogenously added anthocyanin cofactors and the effects of novel processing methods,

the juice and beverage industry can more effectively retain both quality and nutritional

aspects of their products, increasing overall consistency and adding value to this multi-

billion dollar industry.

Anthocyanins

Anthocyanins are best known for their brilliant red and purple colors, and as

polyphenolic compounds their antioxidant and antiradical capacity have been firmly

established (Abuja et al., 1998; Frankel et al., 1998; Ghiselli et al., 1998; Heinonen et al.,

1998; Lapidot et al., 1999; Espin et al., 2000). These pigments are considered as potential

replacements for certain certified and banned dyes because of their bright, attractive









colors and water solubility, which allows their incorporation into aqueous food systems

(Rodriguez-Saona et al., 1999; Boulton, 2001). Moreover, among the various food

products currently available, those containing anthocyanins tend to consume the visual

sense on retail shelves due to their diverse and bright array of colors exhibited. However,

the use of these natural pigments can result in inconsistent color and nutrient degradative

reactions during processing and storage, which limits both usage and fortification efforts.

Anthocyanins are the glycoside forms of anthocyanidins (flavonoids) which have a

C6-C3-C6 skeleton. Although over 300 different anthocyanins are present in nature only

six basic anthocyanin skeletons exist (Figure 2-1A) and which vary in the number and

position of hydroxyl and methoxy substituents. These aglycon forms (anthocyanidins) are

rarely found in the nature but with some exceptions which include their 3-deoxy forms

and which have been reported to be present in red-skinned bananas, sorghum, black tea.

Anthocyanins also differ due to the number, position and type of glycoside moieties

attached to their aglycon moieties (Figure 2-1B and IC). Glucose, galactose, rhamnose,

and arabinose are the most common glycoside moieties attached; however, other

complex glycosides (rutinoside, sophoroside, sambubioside) also occur naturally

(Clifford, 2000).

Anthocyanins moieties can also vary according to the extent and the type of

acylating constituents which are attached to their glycoside chains. These acylated

anthocyanins are the result of an enzyme-catalyzed transfer of an aliphatic or aromatic

organic acid onto a sugar moiety (via an acyl linkage) of an anthocyanin glycoside. The

enzyme system responsible for acylation was identified as 3-O-glucoside-6"-O-

hydroxycinnamoyltransferase. This enzyme catalyzes the transfer of cinnamic acid










coenzyme A esters to the glycosyl anthocyanin moieties. The more common acylating

constituents include cinnamic acids (p-coumaric, caffeic, ferulic, sinapic, and

chlorogenic) flavonoids, flavan-3-ols, and tannins, which may themselves bear glycosidic

chains and aliphatic acids (acetic, malic, maloni, oxalic, and succinic). These compounds

are typically bound to C-4 of a sugar attached to position C-3 of an anthocyanin

molecule, but can also be found on other sugar moieties of the anthocyanin (Figure 2-2).

The location, number of acylating compounds, and diversity of these anthocyanins is

highly dependent on plant type, but those found in fruits and vegetables tend to be

structurally simpler compared to flowering plants.


R1 R1 R1
OH OH AOH

HO O /HO O HO ,O
/ R / 1R2 Y/ Y \ R2

OH V O-Gly Y" O-Gly
OH OH O-Gly

A B C


Anthocyanin R1 R2

Pelargonidin H H

Cyanidin OH H

Delphinidin OH OH
Peonidin O-CH3 H

Petunidin O-CH3 OH

Malvidin O-CH3 O-CH3



Figure 2-1. Structure of the six basic anthocyanindins (A) showing the different
substituents for the six common anthocyanins found in food systems, along
with their different positional glycosides (3-glycosides, B; 3,5-glycosides, C).

























Cyanidin-3-P-D-glu co side Cyanidin-3-(6-O-p-coumaryl)-P-D-glucoside

Figure 2-2. Schematic representing a simple and an acylated cyanidin glycoside.


Acylation significantly improves the stability of anthocyanins through

intramolecular copigmentation. The aromatic residues of the acyl groups stack

hydrophobically with the pyrilium ring of the flavylium cation and consequently greatly

decrease their susceptibility to the nucleophilic attack of water (Figure 2-3). As the result

of the presence of these acylated constituents, anthocyanins also exhibit a significant

bathochromic shift (1-8 nm increase) from that of the parent anthocyanin. Moreover,

acylated anthocyanins have superior color intensity and stability over identical 3-

glucosides and maintain a desirable pigmentation in low acid or neutral conditions (Asen

et al., 1972). As a general rule, the degree of anthocyanin acylation can be estimated by

their color intensities at a decreasing acid content. Highly acylated anthocyanins will

retain greater red or blue colors at pH 5, whereas low levels of acylation will be nearly

colorless at this same pH. The source of the anthocyanin pigment significantly affects the

color and stability characteristics of these pigments. Red cabbage is often used as a

standard by which anthocyanin color stability is compared due to its high degree of










acylation. Color enhancement of this nature is difficult to augment on a commercial scale

due to a need to purify these enzymes, but occurs very efficiently in the vacuole of plant

cells and better explains much of the color diversity in plant systems (Asen et al., 1972).








Intermolecular Intramolecular self-association
(Sandwich Type)













Anthocyanin Co-pigment Acyl group Sugar



Figure 2-3. Schematic representation of anthocyanin self-association and copigmentation
reactions.


Anthocyanins and Intermolecular Copigmentation Reactions

Anthocyanin intermolecular copigmentation reactions are common in nature and

result from association between pigments and cofactors such as polyphenolics and/or

metal ions, or other anthocyanins (self-association) (Figure 2-3). Preferably formed under

acidic conditions, these weak chemical associations can augment anthocyanin stability

and increase antioxidant properties (Mazza and Brouillard, 1990; Boulton, 2001; Malien-

Aubert et al., 2001). Studies have suggested that the copigmentation phenomenon is the









main anthocyanin stabilizing mechanism in plants (Mazza and Brouillard, 1990; Boulton,

2001). Polyphenolics are the predominant cofactors present in anthocyanin-containing

fruits and vegetables, and increased anthocyanin stability has been attributed to their high

concentrations in foods. In general, this type of copigmentation was originally interpreted

as a weak complex formed between an anthocyanin and a cofactor agent (Robinson and

Robinson, 1931), which is still held as the most popular and recognized mechanism

(Markakis, 1982). The current understanding of copigmentation consists of a "stacking"

of a cofactor on the planar polarized nuclei of an anthocyanin in its flavylium ion form.

The hydrophobic complexation reactions between anthocyanins and cofactors may also

effectively protect anthocyanins against the nucleophilic water attack at position 2 of the

pyrilium nucleus, thus displacing the equilibra towards the flavilium form (colored)

rather than that of the less-colored hemiketal or chalcone forms (Boulton, 2001; Malien-

Aubert et al., 2001; Es-Safi et al, 2002). The effectiveness of this stabilizing effect will

depend on the same variables that affect copigmentation. Intermolecular copigmentation

also exerts a protective effect on anthocyanin degradation as cofactors compete with

anthocyanins and preferentially react in several condensation reactions involving a wide

variety of carbonyl compounds (Es-Safl et al, 1999; Malien-Aubert et al., 2001; Es-Safl

et al, 2002). As discussed by Boulton (2001), the color enhancement and stabilizing

effect conferred by copigmentation is different for a given anthocyanin-cofactor pair and

depends on the concentration of pigment, the molar ratio of cofactor to pigment, pH, the

extent of non-aqueous conditions, and presence of anions in solution. The increased

protection observed for a specific pigment source due to the presence of cofactors is most

likely related to the type, and content of polyphenolics present, as a higher










copigment/pigment molar ratio could occur for a determined source. Moreover, specific

polyphenolics or classifications of polyphenolics are more likely to form stable

intermolecular complexes with anthocyanins than others (Boulton, 2001; Malien-Aubert

et al., 2001; Eiro and Heinonen, 2002). Copigmentation reactions are typically

concentration dependent (both for anthocyanins and cofactor) and may be dissociated by

heating or by the addition of alcohol, since hydrogen bonding may link the compounds

together (Asen et al., 1972; Boulton, 2001).

Anthocyanin Stability

Anthocyanin color and stability is influenced by chemical structure, concentration,

pH, temperature, oxygen, light, polymeric forms, ascorbic acid, and the presence of

natural or exogenously added cofactors. Anthocyanins and ascorbic acid have long been

shown to be mutually destructive in the presence of oxygen, which causes a decrease in

color, functional properties, and nutritional quality of a food product (Calvi and Francis,

1978; Poei-Langston and Wrolstad, 1981). Mechanisms for their mutual degradation have

included direct condensation between anthocyanins and ascorbic acid, or the formation of

free radicals that induce oxidative deterioration of each compound (Garcia-Viguera et al.,

1999). Several studies (Calvi and Francis, 1978; Poei-Langston and Wrolstad, 1981;

Garcia-Viguera et al., 1999) have implicated a complex chemical interaction possibly

involving co-oxidative reactions between ascorbic acid and anthocyanins, the effects of

which may lead to co-oxidation of other fortified nutrients. Model systems containing

anthocyanins and ascorbic acid have demonstrated destruction under both aerobic and

anaerobic conditions, therefore exclusion of oxygen during processing would not be

sufficient to prevent nutrient degradation of anthocyanin-containing beverages. Studies

involving the degradation of these phytonutrients have also been ambiguous since Kaack









and Austed (1998) reported a protective effect on anthocyanins containing ascorbic acid

or when sparged with nitrogen in elderberry juice, while ascorbic acid alone was

responsible for anthocyanin decreases in Concord grape juice (Calvi and Francis, 1978).

Color degradation of blood orange juice, which naturally contains cyanidin-3 -glucoside

and ascorbic acid, was found to correlate with ascorbic acid concentrations and resulted

in juice discoloration and loss of fortified ascorbic acid (Choi et al., 2002). Therefore,

degradative reactions of this nature seem to be commodity specific depending on the

phytochemical or anthocyanin composition.

Anthocyanin 3,5-diglucosides were reported as less stable to oxidation and heat

compared to corresponding 3-glucosides (Markakis, 1982) and may result in rapid color

loss during wine or juice storage. The 3,5-diglucosides reported to be most unstable in

muscadine grape juice were delphinidin and petunidin (Flora, 1978; Goldy et al., 1986)

and their oxidation during storage were correlated to decrease radical scavenging activity

(Talcott and Lee, 2001). Both delphinidin and petunidin, along with cyanidin, contain at

least one o-dihydroxy group, making them more susceptible to oxidation than the other

anthocyanin forms. Flora (1978) reported large reductions in delphinidin, cyanidin, and

petunidin-3,5-diglucosides in muscadine grapes after severe heat treatments when

analyzed by thin-layer chromatography. Malvidin 3,5-diglucoside was found to be less

stable than acylated forms of malvidin present in red cabbage (Hrazdina et al., 1970), but

in model systems the stability of malvidin 3,5-diglucoside was greater than malvidin 3-

glusoside both with and without added ascorbic acid (Hrazdina et al., 1970; Garcia-

Viguera and Bridle, 1999). The relative stability of a particular source of anthocyanins is









likely a function of a complex chemical matrix, structural features, and the combined

effects of processing and storage.

Phytonutrient Stability and Intermolecular Copigmentation

Preventing anthocyanin and color loss in beverages can be accomplished by strict

oxygen control during processing but also by a physical stabilization of anthocyanins

through the addition of exogenous anthocyanin cofactors (intermolecular

copigmentation), rather than through antioxidant addition via ascorbic acid. However,

current trends in beverage production dictate that juice and beverage products be fortified

with ascorbic acid, which is intentionally done at the expense of anthocyanin, ascorbic

acid, and quality deterioration of the product. Therefore, exploration of a system to

stabilize anthocyanins from color loss, the effects of ascorbic a fortification, and other

deleterious reactions under condensation conditions or a free radical mechanism is

important if anthocyanins present in juices and beverages are to compete in a market

dominated by certified colors such FD&C Red #40 and Red #3. Copigmentation may be

useful in enhancing the value of foods containing anthocyanins; serving as a functional

food ingredient with antioxidant properties, producing greater visual color perception,

and increasing stability to oxidation and heat.

In previous studies (Malien-Aubert et al., 2001; Eiro and Heinonen, 2002; Talcott

et al., 2003; Brenes et al., 2005) the addition of polyphenolic-based anthocyanin cofactors

was shown to significantly reduce the mutual destruction of anthocyanins and ascorbic

acid as well as appreciably augment visual color and antioxidant capacity. These studies

have shown the success in slowing the kinetics of mutual destruction of anthocyanins and

ascorbic acid through cofactor addition, and have led to the proposed obj ectives herein.

The implications of which will have a profound impact on both fruit juice fortification









and in other applications in which anthocyanins are utilized as a natural color source.

Additionally, a maj or advantage of using exogenously added cofactors is an ability to

control their concentration through standardization, which would allow for maximal

increases in color, thermal stability, nutrient protection, and organoleptic properties as

influenced by processing and storage. In addition to preventing quality and nutritional

losses by copigmentation, the present remediation strategy increases oxidative and

thermal stability, which adds economic value to fruit juices and beverages containing

isolated anthocyanins as colorants.

The evidence for copigmentation and phytonutrient retention has been previously

investigated using a diversity of copigment sources ranging from polyphenolics isolated

from natural sources (rosemary, thyme, sage, red clover, grape seeds, grape skins,

orange/grapefruit peels, etc.), from purified compounds (various cinnamic acids and

flavonoids), and from metal ions (calcium, magnesium, aluminum, zinc, copper, iron,

metal ascorbates) against a diversity of anthocyanin sources (cabernet grape, muscadine

grape, black carrot, purple sweet potatoes, red cabbage, red radish, hibiscus, elderberry,

and acai fruit) (Covarrubias, 2002; Kemmerer, 2002; McGuinness, 2002; Talcott et al.,

2002a; Talcott et al., 2002b; Talcott et al., 2003a; Brenes et al. 2004; Del Pozo-Insfran et

al., 2004; Brenes et al., 2005). Overall, the outcomes of these studies demonstrated that

the addition of polyphenolic cofactors can augment visual and instrumental color

properties, increase heat and process stability of anthocyanins and ascorbic acid, increase

overall radical scavenging properties, serve to enhance antimicrobial properties over that

exhibited by anthocyanins and cofactors alone, serve to inhibit lipoxygenase and

polyphenol oxidase when highly purified cofactors are used.









For instance, Talcott et al. (2003a) demonstrated that addition of concentrated

polyphenolic cofactors in the range of 0. 1-0.4% v/v (100-1,500 mg/L gallic acid

equivalents) readily forms copigment complexes with anthocyanins and results in

concentration-dependent hyperchromic shifts from 10-50% depending on source, which

also corresponds with increased antioxidant activity. Brenes et al. (2005) also showed

significant improvements in phytonutrient retention, when model anthocyanin systems

copigmented with a purified water-soluble extract of rosemary with and without fortified

ascorbic acid were studied. In a temperature dependent reaction, the addition of cofactors

was able to retain 14 and 24% more ascorbic acid at 0.2 and 0.4% (v/v) cofactor addition;

preserving both anthocyanin and ascorbic acid from oxidative and/or free radical damage

in their mutual presence. Cofactor addition appreciably affected monomeric

anthocyanins, overall color, and extended anthocyanin half-life from 11 days without the

rosemary extract to 15 and 19 days at 0.2% and 0.4% v/v rosemary extract, respectively.

Other studies (Covarrubias, 2002; Kemmerer, 2002; McGuinness, 2002; Talcott et

al., 2003a; Brenes et al. 2005) also have demonstrated that systems where these

copigments are to be used should be free of residual oxidase enzymes, as polyphenolic

cofactors can serve as enzyme substrates and destroy phytochemical and alter quality

characteristics. In the presence of residual enzymes, it was found that ascorbic acid

fortification of copigmented grape anthocyanins was highly detrimental to fruit juice

quality and resulted in rapid anthocyanin, ascorbic acid, and antioxidant activity losses

during processing, especially under high hydrostatic pressure conditions where the

enzymes may be more active (Talcott et al., 2003a). However, ascorbic acid fortification

in the absence of cofactors demonstrated enzyme inhibition and served to protect grape









anthocyanins prior to thermal and high pressure processing. It was concluded that

although physicochemical attributes were enhanced by copigmentation with rosemary

extract, methods to inactive residual enzymes should be addressed prior to

copigmentation to prevent degradation of anthocyanins in the presence of ascorbic acid.

Due to the outcome of several studies in which the protective effect of different

cofactors was demonstrated to prevent phytonutrients loss, the present research proj ect

suggests a strategy by which the addition of exogenous anthocyanin cofactors (water-

soluble extracts from thyme and rosemary) will alleviate many of the physical and

chemical degradative reactions impacting anthocyanin-containing fruit juice and

beverage fortification. Cofactors isolated from spices are an economical and food-grade

source of polyphenolics for use in the beverage industry. Their polyphenolic

concentrations are naturally very high, so concentration efforts are easily accomplished

(Zheng and Wang, 2001). Based on present scientific knowledge, it is hypothesized that

anthocyanin copigmentation can offer a protective effect for fortified phytonutrients

commonly used in the beverage industry. Success of this work will generate knowledge

for strategies to manufacture highly stable food colorants and to minimize phytonutrients

loss.

Sensory Attributes of Anthocyanin-containing Beverages as Affected by the
Addition of Polyphenolic Cofactors.

Other than empirical evidence from informal evaluations, no formal evaluations on

the organoleptic properties of copigmented anthocyanins have been conducted. Attributes

such as color are usually measured using instrumental techniques rather than with human

subjects. More importantly, the taste attributes imparted by added copigments are an

important consideration affecting their use in food systems. Since polyphenolic cofactors










may impart bitter or astringent flavors to a food system, or may impart aroma from

volatile components, evaluating their sensory thresholds is vital to determining an

effective use level. Taste or aroma carry-over from the use of natural cofactor sources

such as thyme or rosemary extracts is also a concern, and use levels may be affected by

the intensity of compounds that do not serve as anthocyanin cofactors (i.e. aroma-active

compounds).

A previous study evaluated the effect of a commercially available rosemary

polyphenolic extract in the sensory attributes of a strawberry juice cocktail via a rating

test using a 9-point hedonic scale evaluating color, aroma and flavor. Several rosemary

extract concentrations (0, 0.1i, 0.2 and 0.4% v/v) were added to the strawberry juice as

anthocyanin cofactors. The use of these concentrations was based on preliminary studies

(Talcott et al., 2003a; Del Pozo-Insfran et al., 2004; Brenes et al., 2005) which showed

that the addition of polyphenolic-based anthocyanin cofactors, at these chosen

concentrations, significantly reduced the mutual destruction of anthocyanins and ascorbic

acid as well as appreciably augmented visual color and antioxidant capacity of model

beverage systems. Results of the sensory evaluations (Table 1) concluded that the

addition of rosemary cofactors significantly increased the color of the strawberry juice

with respect to that of the control treatment (P<0.01) independently of the cofactor

concentration (0. 1-0.4% v/v). Addition of rosemary cofactors did not significantly affect

either the aroma or flavor of the juices. Results of both the sensory and stability studies

give evidence that the addition of exogenous anthocyanin cofactors, at levels high enough

to prevent phytonutrient degradation, exceed the physicochemical barriers influencing

sensory perception.











Table 2-1. Effect of rosemary extract (0, 0.1i, 0.2, 0.4% v/v) on color, aroma, and flavor
attributes of a commercial strawberry cocktail juice.


Rosemary
extract (% v/v) Color Aroma Flavor
0 4.4 bl 5.4 a 4.7 a
0.1 6.1 a 5.6 a 5.1 a
0.2 6.4 a 5.7 a 5.2 a
0.4 6.6 a 6.2 a 5.2 a

SValues with similar letters within columns of each sensor attribute are not significantly
different (LSD test, P>0.05), and indicate the effect of rosemary extract addition.


Since there is evidence that the addition of polyphenolic cofactors may exert a

protective effect against oxidation of anthocyanins and ascorbic acid, their effect on the

organoleptic properties of copigmented beverage systems at levels in which they are

effective are of importance and will be investigated in the present study.

Novel Processing Technologies That May Enhance Anthocyanin Stability

The food industry is continuously searching for novel processing technologies that

ensure microbial destruction and extend the shelf life of products without having adverse

effects on their quality attributes such as flavor, color, and phytonutrient retention (Butz

and Tauscher, 2002; Matser et al., 2004). Moreover, current trends in food marketing and

show that consumers desire high-quality foods with "fresh-like" characteristics and

enhanced shelf life that require only a minimum amount of effort and time for preparation

(Butz and Tauscher, 2002; Krebbers et al., 2003). At the present time, a wide variety of

emerging non-thermal processing technologies are available to process food and

beverages and include high hydrostatic pressure, irradiation, ultrasound, pulsed electric

fields, light pulses, and oscillating magnetic fields.










High hydrostatic pressure processing (HHP) and dense phase carbon dioxide

pasteurization (DP-CO2) are promising alternatives to traditional pasteurization

technologies and may lessen detrimental effects on thermolabile phytonutrients (Gomez

and Ledward, 1996; Zabetakis et al., 2000; Sun et al., 2002). In hydrostatic pressure

applications, the food or beverage system is pressurized uniformly throughout the product

in which the generation of pressure can be accomplished by direct compression, indirect

compression using an intensifier pump, or by heating of the pressure medium. One of the

main differences between these processing technologies is that a wide variety of food

liquids and solids can be processed by HHP whereas DP-CO2 pasteurization only can

process foods in a liquid form only. Another difference is that HHP utilizes batch and

semi-continuous systems whereas DP-CO2 pasteurization uses continuous systems.

Although both of these processing techniques are potential alternatives to thermal

processing and may in theory reduce phytonutrient losses and quality characteristic of

foods, studies have shown that a downside of their use is the presence and/or activation of

residual enzymes following processing consequently resulting in extensive oxidative

degradation. Due to these factors associated changes in phytonutrient stability due to

processing are of interest and will be investigated in the present study. Furthermore,

comparisons between non-thermal and thermal processes are one way to assess the

benefits offered by these novel processing technologies.

High Hydrostatic Pressure (HHP) Processing

High hydrostatic pressure (HHP) processing, also referred as ultra high pressure

(UHP) or high pressure processing (HPP), is a preservation technique currently used on a

commercial scale in Japan, France, Spain, USA and Mexico for the pasteurization of a

variety of food products that include fruit juices, guacamole, oysters and ham. This










processing technology subj ects foods with or without packaging to pressures between

100 and >1,000 MPa over a range of temperature (0-100 oC) and time (2 sec and to >20

min) conditions. HHP acts instantaneously and uniformly throughout a food mass

independently of its size, shape and composition. Therefore, adiabatic heating of the

product occurs in homogeneous manner compared to that of conventional heat

sterilization where a temperature profile occurs. Generally, the compression process will

increase the temperature of foods approximately 3 OC per 100 MPa. It may be expected

that product characteristics that are dependent on the heat liability of certain components,

are less significantly changed by high pressure sterilization compared to conventional

heat sterilization. The effect of high pressure sterilization on product quality is strongly

depended on the chosen food product. Some of quality parameters that may be affected

are texture, color, production of off-flavors, and phytochemical degradation. The results

of research on high pressure sterilized products have shown that the effects are product

dependent and that careful selection of the appropriate process conditions is necessary

(Barbosa-Canovas et al., 1998; Hendrickx et al., 1998; Butz and Tauscher, 2002; Butz et

al., 2003; Matser et al., 2004).

HHP and microbial inactivation

The effects of HHP on several food constituents have been studied over the last

decade in order to evaluate the effect of these new technologies on food safety and

quality. The outcomes of several published studies (Hendrickx et al., 1998; Smelt, 1998;

Butz and Tauscher, 2002; Lado and Yousef, 2002; Butz et al., 2003; Matser et al., 2004)

have shown that HHP processing can produce commercially sterile products where

kinetics of microbial inactivation are highly dependent on processing parameters

(pressure and temperature), acid content of the product and the type of microbial flora.









For instance vegetative cells, including yeasts and molds, are rather pressure sensitive

and can be inactivated by pressures between 300 and 600 MPa, whereas bacterial spores

are highly pressure resistant and need pressures >1,200 MPa for complete inactivation.

Therefore, preservation of acid foods (pH < 4.6) is the most likely application of HHP.

On the other hand, sterilization of low-acid foods (pH > 4.6) requires a combination of

pressure and mild-temperature treatments (Hendrickx et al., 1998; Butz and Tauscher,

2002; Butz et al., 2003). Microbial inactivation is highly dependent not only on

processing parameters (pressure, temperature, time, number of pulses) and by food

composition, but also by the types of microorganisms present in a food matrix. For

instance, in order to obtain a 4 log reduction of E coli the medium needs to processed at

100 MPa at 300C for 720 min, while for L. monocytogenes the applied pressure needs to

be increased to 238-340 MPa over 20 min in order to obtain a similar reduction in

microbial load (Barbosa-Canovas et al., 1998).

At ambient temperatures, application of pressures in the range of 400-600 MPa

inactivate vegetative micro-organisms and reduce the activity of enzymes resulting in a

pasteurized product, which can be stored for a considerable time at 4-6 oC. The

inactivation of vegetative micro-organisms and enzymes, combined with retention of

phytochemicals and low molecular weight food molecules that are responsible for taste

and color, results in HHP-pasteurized products with a prolonged shelf-life and fresh-alike

characteristics. For sterilization of HHP products a combined process where both

pressure and temperature (60-1100C) is needed to achieve the complete inactivation of

spores and enzymes. The result of the later process is a shelf stable product, which in

many cases has a higher degree of quality than those products obtained using









conventional processing. High-pressure inactivation of vegetative micro-organisms is

caused by membrane damage, protein denaturation and decrease of intracellular pH,

suggesting that pressure results in deactivation of membrane-bound enzymes associated

with efflux of protons. Water activity and pH are critical process factors in the

inactivation of microbes by HHP. Temperatures in the range of 45-50 oC appear to

increase the rate of inactivation of food pathogens and spoilage microbes. Temperatures

ranging from 90 to 110 OC in conjunction with pressures of 500-700 MPa have been used

to inactivate spore-forming bacteria such as Clostridium botulinum (Hendrickx et al.,

1998; Butz and Tauscher, 2002; Butz et al., 2003).

The effect of HHP in food components

An advantage of HHP processing is that food quality characteristics, sensory

attributes, and phytonutrient retention are either unaffected or only minimally altered by

processing at room temperature, except when some type of enzymatic activity is

involved. For instance, a recent study investigated the effect of different high pressure

treatments on odor and aroma of an orange-lemon-carrot juice mixture and its subsequent

storage (21 days at 4 OC) (Butz et al., 2003). Results indicated that HHP treated juices

(e.g. 500 MPa for 5 min) presented only minor changes in odor and flavor after

processing when compared to pasteurized juices. Moreover, the HHP juices did not

present significant changes in odor, flavor and overall quality after storage whereas

attributes were significantly decreased for control juices. However, several studies have

also demonstrated that besides microbial destruction there are other pressure-induced

effects on food components such as protein denaturation, enzyme activation or

inactivation, changes in enzyme-substrate interactions, changes in the properties of

polysaccharides and fats, protein gelation, etc (Tauscher, 1998; Messens et al., 2002;









Butz et al., 2003). The physiochemical changes induced by HHP have also open up the

possibility of producing foods with novel texture (e.g. meat, fish, dairy products) as well

as the modification of food protein functionality (Messens et al., 2002). Several chemical

changes have been reported for food macromolecules that have been HHP, such as the

stability of aspartame present in milk, TRIS-buffers and water during different treatments

(600 MPa at 60 oC for 3-30 min) (Butz et al., 1997). Results of this study indicated that

after HHP with a holding-time of <3 min only about 50% of the original content of

aspartame was detectable in milk (pH 6.8). The degradation by-products were identified

as aspartylphenylalanine and a diketopiperazine. However, the stability of aspartame was

insignificantly affected when present in model acid systems resembling fruit

preparations, juices, or carbonated drinks. Another example of pressure-induced chemical

changes is that observed for p-carotene in model solutions and in sliced carrots under

different pressure and temperatures regimes (Tauscher, 1998). This study showed that the

content of p -carotene in ethanolic model solutions after HHP for 20 min at 75 OC was

reduced by more than 50%. However, its content was not reduced significantly when

processed for 40 min at 600 MPa and 75 oC. Authors proposed that the carotenoids were

well protected against pressure/temperature degradation in the last instance since these

compounds are buried in lipophilic environments. Both of these studies demonstrate the

importance of the food matrix as a beneficial protective action on the retention of

nutrients and quality attributes after HHP.

The balance of intramolecular food components and solvent-protein interactions is

greatly affected by the HHP parameters (pressure and temperature). Therefore the extent

of unfolding of the polypeptide chain is strongly dependent on the processing conditions









and consequently one can observe different enzymatic activity on the quality attributes of

a food matrix. Structural rearrangements taking place in the protein upon pressurization

are governed by the principle of Le Chatelier, which states that processes associated with

a volume decrease are encouraged by pressure increases, whereas processes involving a

volume increase are inhibited by pressure increases. The volume decrease accompanying

denaturation arises from the formation or rupture of noncovalent bonds and from the

rearrangements of solvent molecules (Hendrickx et al., 1998; Butz and Tauscher, 2002;

Butz et al., 2003). Pressure induced activation of enzymes and/or their residual activity

can significantly affect the quality of food products. Due to the pressure stability of some

of these food quality-related enzymes, combined technologies involving pressure and

temperature are necessary for complete enzymatic inactivation. Such effect is specifically

related to the type of enzymes and the processing conditions of foods. Studies have

shown that some enzymes can be deactivated using pressures < 200 MPa, while others

can withstand pressures over 1,000 MPa (Barbosa-Canovas et al., 1998; Hendrickx et al.,

1998; Butz and Tauscher, 2002; Butz et al., 2003; Matser et al., 2004). For example,

several studies have investigated the effect of HHP on different pectin methyl esterases

(PME) and its consequent effect on cloud destabilization, gelation and loss of consistency

of several food products (Seyderhelm et al., 1996; Basak and Ramaswamy, 1996; Cano et

al., 1997; Stoforos and Taoukis; 2003; Irwe and Olsson, 1994). PME is usually

inactivated by conventional thermal processes that have detrimental effects on flavor,

color and nutritional quality. Studies have shown that HHP processing of orange juice

results in a commercial stable product with higher quality attributes when compared to

that of thermally processed orange juice (Irwe and Olsson, 1994; Basak and Ramaswamy,









1996; Stoforos and Taoukis; 2003). HHP treatments of > 600 MPa have been shown to

irreversibly inactivate (> 90%) PME. However, tomato PME seems to be more pressure

resistant and it seems to be activated during HHP at < 400 MPa especially in the presence

of calcium ions and in acidic media (pH 3.5-4.5) (Seyderhelm et al., 1996). Pressure-

induced activation of orange juice PME was also noted by Cano et al. (1997) in the case

of treatments at room temperature and 200-400 MPa. Krebbers et al. (2003) showed the

combined effect of HHP and thermal treatments on the quality attributes and microbial

stability of tomato puree. Their results showed that HHP alone caused partial inactivation

of PG (~70%) yet activation of PME was observed. The use of combined treatments (e.g.

700 MPa and 900C for 30 s yield a commercial sterile product (> 4.5 log reduction) that

had > 99% enzyme inactivation (both PME and PG). The obtained shelf stable product

had superior sensory and quality attributes (increased color, increased water binding,

lower viscosity, and higher lycopene retention) when compared to a thermally sterilized

product.

Similar trends have been observed for food products in the case where

polyphenoloxidase (PPO) is present. Mushroom and potato PPO are very pressure stable,

since treatments between 800 and 900 MPa are required for activity reduction (Eshtiaghi

et al., 1994; Gomez and Ledward, 1996; Weemates et al., 1997). A similar trend was

observed for avocado PPO by Hendrickx~ et al. (1998) who investigated the combined

effect of pressure (0. 1-900 MPa) and temperature (25-77.5 OC). Results of this study

demonstrated that PPO inactivation at 21 OC was only achieved for pressures > 900 MPa.

Pressure induced activation has also been reported for apple (Anese et al., 1995), onion

(Butz et al., 1994), pear (Asaka et al., 1994) and strawberry PPO (Cano et al., 1997).










Grape, strawberry, apricot and apple PPO seem to be more pressure sensitive than other

PPO's. Apricot, strawberry and grape PPO can be inactivated by pressures exceeding

100, 400 and 600 MPa, respectively. Inactivation of apple PPO varies between HHP at

100-700 MPa in function of the pH of the matrix. For several PPO enzymes, it has been

reported that pressure-induced inactivation proceeds faster at lower pH, however the

inactivation is also influenced by the addition of salts, sugars or other compounds. For

example, the pressure inactivation of apple PPO is enhanced by the addition of calcium

chloride whereas for mushroom PPO is enhanced in the presence of 50 mM benzoic acid

or 5 mM glutathione. The sensitizing effect of glutathione was suggested to be due to an

interaction with a disulphide bond of the enzyme (Anese et al., 1995).

Since HHP processing is a promising alternative to traditional pasteurization

technologies and may lessen detrimental effects to thermolabile compounds associated

changes in phytonutrient stability due to processing are of interest and will be

investigated in the present study. Furthermore, comparisons between non-thermal and

thermal processes are one way to assess the benefits offered by these novel processing

technologies.

Dense Phase-CO2 PaSteurization

As previously mentioned, HHP is currently used in the U.S., Europe and Japan to

produce a variety of commercial products. However, in order to ensure the safety of

foods against some pressure-resistant microorganisms and bacterial endospores, HHP

needs to be used in combination with other thermal and non-thermal processes.

Moreover, processing techniques that along with improving the efficacy of HHP

microbial inactivation decrease processing costs (i.e operating pressures and

temperatures, dwell time) are desired. Dense phase-CO2 pasteurization (DP-CO2), also









known as high-pressure carbon dioxide processing, is therefore a potential candidate as a

non-thermal processing due to its ability to inactivate both microbes and enzymes under

more cost-effective processing conditions. The principle of the microbial inactivation of

DP-CO2 is based on gas dissolution in a microbial cell by pressurization that, when

rapidly decompressed to atmospheric pressure, causes fatal functional damage and

explosive decompression of the cell (Balaban et al., 1991; Park et al., 2002). It is

noteworthy to mention that CO2 preSsurization does not always lead to cell burst but in

some cases only leads to leakage of cellular components and changes in the cell

membrane permeability which are responsible for cell damage and microbial inactivation

(Lin et al., 1993; Isenschmid et al., 1995; Park et al., 2003). DP-CO2 affects biological

systems by causing protein denaturation, lipid phase changes, and rupture of cell walls

and membranes.

DP-CO2 and microbial inactivation

Carbon dioxide under both atmospheric and supercritical pressures has

demonstrated antimicrobial effects in foods. It is used successfully in modified

atmosphere packaging to reduce horticultural respiration and reduce microbial growth

(Corwin and Shellhammer, 2002). Similarly, it extends the shelf-life of dairy products

through inhibition of microbial growth. In addition to its use at atmospheric conditions,

supercritical carbon dioxide can inactivate a wide range of microbiota (Kamihira et al.,

1987; Lin et al., 1993; Ballestra et al., 1996) and recent work indicates that it can also

inactivate bacterial spores (Kamihara et al., 1987; Haas et al.; 1989 Enomoto et al.,

1997). The inhibitory effects of DP-CO2 and supercritical carbon dioxide (SC-CO2) on

microorganisms for food preservation have recently received a great deal of attention and

have been extensively investigated over the last decade. Fraser (1951) first tried gas









pressurization with N2, NO2, Ar, and CO2, and reported that CO2 COuld inactivate 95-99%

ofE. coli at 3.40 MPa and 37 oC. Nakamura et al. (1994) found that 108 cells/ml of

baker' s yeast could be inactivated after CO2 Saturation at 4.05 MPa and 400C for 3 h, and

proposed this processing as a novel method for sterilization of food microorganisms.

Kamihira et al. (1987) demonstrated that complete inactivation of baker' s yeast, E. coli,

S. aureus, and A. niger, can be accomplished by contact with SC-CO2 at 20.26 MPa and

35 OC for 2 h. More recently, Park et al. (2002) observed that DP-CO2 exerted a

relatively large effect on total aerobes present in a carrot juice at 0.98 MPa and that its

bactericidal effect gradually increased to achieve a 4-log reduction at 4.9 MPa for 10 min.

Ballestra et al. (1996) found a decrease in the survivors of E. coli at 5 MPa using CO2 at

temperatures > 35 oC, with the most effective inactivation found after 5 MPa at 45 oC for

20 min. However, Corwin and Shellhammer (2002) found that under the studied

conditions (0, 365 and 455 MPa) the amount of dissolved CO2 in the medium had only a

slight significant effect in the inactivation of E coli Kl2 and that the processing pressure

was responsible for the 4 and 6-log reduction in microbial reduction, respectively for 365

and 455 MPa. Similar results were obtained by Park et al. (2003) as carbonation itself did

not have a significant effect on B. subtilits inactivation rates yet a synergistic effect was

observed when combined with HHP and resulted in a 5-log reduction at 600 MPa.

Enomoto et al. (1997) examined the lethal effect of DP-CO2 on spore cells of

Bacillus megaterium and observed that the bactericidal effect of CO2 WAS enhanced with

increasing temperature and treatment time, and that processing at 5.9 MPa and 60 oC for

30 h could reduced the survival ratio of the spores to about 10-7. Kamihara et al. (1987)

also observed that under low pH and heating conditions, SC-CO2 was an effective









method to inactivate endospores of B. subtilis and B. \iarlstherr~lI~limespibi The later

observation (low pH and high temperature conditions) was confirmed by Haas et al.

(1989) who achieved a high reduction of the survival ratio for spore cells of C.

sporogenes 3679 under SC-CO2 at 5.5 MPa at 70 oC for 2 h.

During DP-CO2 the carbon dioxide solubility increases directly proportional with

increments of processing pressure. Therefore a pH change of the medium is caused due to

carbonic acid formation. Although several studies have attributed the later fact as the

main explanation for the antimicrobial activity of SC-CO2, several studies have shown

that extraction of essential intracellular substances such as phospholipids and

hydrophobic compounds from cells or cell membranes, and enzyme inactivation also play

important roles as mechanisms of microbial inactivation (Kamihira et al., 1987; Haas et

al., 1989; Lin and Lin, 1993; Ballestra et al., 1996). For instance, Haas et al. (1989)

showed that a 6 log bacterial number reduction was obtained following CO2

pressurization compared to only a 2 log reduction for a control both in which its pH was

reduced from 5.3 to 3.2. These results were subsequently confirmed by Ballestra et al.

(1996) who also observed that the microbial destruction occurred in two stages during

processing at 5 MPa for 15 min at 35 oC. In the first step, cells undergo a stress by

pressurized carbon dioxide and in which a slower inactivation rate occurs when compared

to the second step of microbial destruction where rapid inactivation occurs as a critical

level of the gas is reached. The authors also noticed selective enzyme inactivation which

was attributed as a result of the internal drop of pH during processing. Lin et al. (1994)

also observed two inactivation stages when studying the effect of pressurized carbon

dioxide on the viability of L. monocytogenes. These authors also observed that the pH










drop was not the only reason for microbial inactivation, but that the conversion of

bicarbonate, formed from the dissociation of carbonic acid, into carbonate precipitates

intracellular calcium and other ions resulting in cell malfunction and damage.

The temperature, as well as the processing pressure, at which a food or beverage is

processed significantly affects the efficiency of carbon dioxide processing as a non-

thermal pasteurization technique, since both parameters control the solubilization rate of

CO2 and its solubility in a suspending medium (Erkman, 2000a; Erkman,2000b). For

instance, Arreola et al. (1991b) achieved a 2-log decrease in total plate count during the

pressurization of a single strength orange juice at 33 MPa and 35 oC for 1 h, while

processing at 45 and 60 oC achieved the same reduction at 45 and 15 min, respectively.

Authors also observed a decrease in D-values for microbial reduction at the same

temperature when pressure was increased, and results were attributed due to the

combination of high pressure, shear rate during depressurization of the juice, and the

larger extent of carbon dioxide solubility. Erkmen (2000a, 2000b) also observed that the

time to achieve complete inactivation of L. monocytogenes at 6.08 MPa CO2 and 25 oC

was reduced from 115 min to 75 and 60 min at 35 and 45 OC, respectively. Similar results

were observed by Ballestra et al. (1996) while studying the survival rates of E. coli in

Ringers solutions and in which results were attributed to increments on the amounts of

dissolved CO2. The later effect can be attributed to a higher CO2 absorption as the

processing temperatures increases. Hong et al. (1999) reported that microbial inactivation

by DP-CO2 is governed essentially by the transfer rate and the penetration of carbon

dioxide into cells, the effectiveness of which can be improved by increasing pressure,

decreasing the pH of the suspension, and increasing the processing temperature.









Temperature has a close relation with the characteristics of CO2 maSs transfer and most

likely higher temperatures stimulate the CO2 diffusivity into the microbial cell and could

also increase the fluidity of the cell membrane to make the CO2 penetration more easy

(Erkmen, 2001).

Lin et al. (1994) along with Erkmen (2000a) and Wei et al. (1991) reported that not

only processing conditions play an important role in microbial stability but also food

components (i.e. fats, proteins), since they might reduce the bactereostatic effect of CO2

by delaying its penetration into the cells. For instance, Wei et al. (1991) observed that

SC-CO2 (13.7 MPa, 35 oC for 2 h) inactivated completely Salmonella in egg yolks, while

processing of whole eggs only resulted in a 64% bacterial inactivation. Similar

observations were observed by Erkman (2000a; 2000b; 2000c) in which inactivation of

Staphylococcus attreus suspended in broth was achieved at lower pressures and shorter

processing times when compared to that of raw milk. This effect was further

demonstrated when comparing the protecting effect of certain food components on

microbial survival rates in raw milk when compared to those of orange, peach and carrot

juices. Other authors also have stated that the water content of foods significantly affects

the efficacy of CO2 to inactivate microorganisms. Kamihira et al. (1987) observed that

the sterilization of Koji, which contained baker' s yeast, E. coli, Staphylococcus attreus

and conidia of Aspergilhts niger, was achieved at 20 MPa and 35 oC when the water

content of each microorganism was 70-90% (wet cells). However, when the water

content was decreased to 2-10% (dry cells) an incomplete microbial inactivation was

observed under the same processing conditions. The outcome of this study demonstrated

that product sterilization depends on several factors including type of microorganism,









water content, and addition of co-solvent. Authors concluded that when the water content

of a microbial suspension increases, the walls of the cell swell and thus carbon dioxide

has a larger surface area of penetration and consequently has a larger effect on the

microbial cell. In addition, when combined with water, carbon dioxide produces carbonic

acid that affects cell permeability (Lin et al., 1994). Kumagai et al. (1997) also observed

that a higher yeast inactivation is accomplished with higher water activities of foods and

higher processing pressures due to their effects on CO2 adsorption in the yeast cells.

Besides processing parameters (pressure, temperature, time, CO2 COncentrations)

and food composition, the presence of co-solvents (i.e. ethanol) also significantly affects

microbial inactivation in the pressurized medium. The presence of these compounds

modifies the rate of CO2 Solubility and adsorption by modifying the critical temperature

and pressure of the medium(Taylor, 1996). For instance, the addition of ethanol increases

CO2 adsorption on surface sites during extraction of components and thus prevents the re-

adsorption of certain compounds (Clifford and Williams, 2000).

Although the exact mechanism of the bacteriostatic action exerted by pressurized

CO2 is not known several possible mechanisms have been reported and include (Daniels

et al., 1984; Lin and Lin, 1993; Lin et al.; 1994; Wei et al., 1991; Ballestra et al., 1996;

Erkmen, 1997;):

Reduced growth rate of aerobic bacteria due to the replacement of oxygen by
CO2.
Formation of carbonic acid resulting on decreases in the cell's pH and
consequently affecting metabolic activities.

CO2 penetration into the cell which may enhance its chemical activity on the
internal metabolic processes of the cell









Increase in the cell membrane permeability by high pressure treatment causing
cell leakage and damaging cell function
Protein denaturation stops the uptake of amino acids which are essential for cell

growth and also affects the enzymatic systems of the cell.
DP-CO2 and enzymatic inactivation

Several studies have shown the effect of SC-CO2 and DP-CO2 on PlVE, PE,

lipoxygenase (LOX), peroxidase (POD), and PPO in model and real food systems (Chen

et al., 1992; Park et al., 2002; Corwin and Shellhammer, 2002; Boff et al., 2003;). For

instance, Taniguchi et al. (1987) studied the effect of SC-CO2 on nine different enzymes

at 20.3 IVPa and 35 OC for 1 h, and authors showed that > 90% of the enzymatic activity

was retained when the water content of the enzyme preparations was 5-7% wt. Chen et al.

(1992) reported that PPO can be inactivated at low temperatures with SC-CO2; however,

the degree of inhibition was dependent on the source of the enzyme. In this study, spiny

lobster PPO was greatly inactivated followed by shrimp, potato, and lastly apple juice

PPO. The circular dichroism spectra at far UV-range showed that SC-CO2 treatment

caused conformational changes in the secondary structure of the enzymes, being source

of marines enzymes (lobster and shrimp) the ones that underwent the most evident

conformational changes. In the same study, the authors showed that SC-CO2 also

inhibited orange juice PE where thermal inactivation was insignificant. Authors

concluded that the extent of PE inactivation depended on pressure, temperature and time

of processing. Overall results of the study showed that SC-CO2 processing was an

effective non-thermal technology to reduce microbial loads and enzyme activity. Arreola

(1990) studied the effect of SC-CO2 and HHP on the microbial stability and quality

attributes of a single strength orange juice. The author used a batch system where the gas

was allowed to mix in a closed vessel under high pressure at temperatures between 35 to









60 oC. Results of this study showed that this process was effective in destroying

microorganisms and obtaining an acceptable product with improved cloud retention,

despite residual PE activity (50%) was present following processing. Arreola et al.

(1991b) also investigated PE activity and showed that its inactivation was affected by

temperature, pressure and process time. Complete PE inactivation was achieved at 26.9

MPa and 56 OC for 145 min. Boff et al. (2003) also investigated the effect of HHP and

DP-CO2 on PME activity and the physiochemical properties of a single-strength

Valencia orange juice following processing and during 4 months of storage at 4 and 30

oC. Authors observed that although 28% of PME activity remained after processing this

product had enhanced cloud stability and higher ascorbic acid retention when compared

to HHP and thermally pasteurized samples. DP-CO2 produced a cloud-stable orange

juice with more ascorbic acid and flavor volatiles than the thermally processed juice.

Corwin and Shellhammer (2002) also compared the inactivation of PME and PPO by

HHP and DP-CO2 at 25 and 50 oC. Authors observed that in the inactivation of PME,

pressure was a significant factor at both processing temperatures and that CO2 was a

significant factor in further inactivating PME beyond that which pressure would achieve

alone. Authors observed the same trend but for PPO as HHP processing only reduced this

enzymatic activity by 8-21% when compared to 44-79% inactivation for DP-CO2

processed treatments. Park et al. (2002) observed that a combined treatment of 4.90 MPa

of SC-CO2 and 600 MPa-HHP effectively inactivated enzymes in a carrot juice. The

residual activities of PPO, LOX, PME in this study were less than 1 1.3%, 8.8%, and

35.1%, respectively. The effect of SC-CO2 in LOX and POD activity was also been

investigated by Tedj o et al. (2000) in 30% sucrose solutions. Authors observed that










application of SC-CO2 at 3 5.2 MPa and 40 oC for 15 min inactivated 3 5% of LOX

activity, while pressurization at 62. 1 MPa and 55 oC for 15 min inactivated 65% of POD

activity. Total inactivation of LOX (10.3 MPa, 50 oC and 15 min) and POD (62.1 MPa,

55 OC and 15 min) was achieved through SC-CO2 for unbuffered solutions. These

authors also observed that by increasing the concentration of sucrose and buffering (pH

range 4 to 9) the working solutions the enzymes increased their resistance for SC-CO2

inactivation.

Since the use of DP-CO2 processing has been shown to be a promising non-thermal

process to inactivate microorganisms and enzymes it might be used as a novel technology

to enhance phytochemical stability. However, its effect on anthocyanin and ascorbic acid

stability is not widely known and, therefore, will be investigated in the present study.

Moreover, the effect of DP-CO2 in the sensory attributes and quality retention of

anthocyanin-containing juices following processing and during storage has not been

investigated.















CHAPTER 3
PHYTOCHEMICAL COMPOSITION AND PIGMENT STABILITY OF ACAI
(EUTERPE OLERACEA MART.)

Introduction

Agai (Euterpe oleracea Mart.) is a palm plant widely distributed in northern South

America with its greatest occurrence and economic importance in the floodplains of the

Brazilian Amazonian state of Para (Muniz-Miriet et al., 1996; Silva et al., 1996; Murrieta

et al., 1999;). Agai is a slender, multi-stemmed, monoecious palm that can reach a height

of over 30 meters. A wide variety of marketable products are produced from this palm,

but the spherical fruits that are mainly harvested from July to December are its most

important edible product. Each palm tree produces from 3 to 4 bunches of fruit, each

bunch having from 3-6 kg of fruit. The round-shaped fruits appear in green clusters when

immature and ripen to a dark, purple colored fruit that ranges from 1.0-1.5 cm in

diameter. The seed accounts for most of the fruit size and is covered by thin fibrous fibers

under which is a small edible layer. A viscous juice is typically prepared by macerating

the edible pulp that is approximately 2.4% protein and 5.9% lipid (Silva, 1996). The juice

is currently used to produce energetic snack beverages, ice cream, jelly, liqueur, and is

commonly blended with a variety of other juices.

A steady increase in the development of natural food colorants and functional food

sources has been observed in recent years, not only due to consumer preferences for

natural pigments but also for their health-related benefits and nutraceutical properties

(Frankel et al., 1995; Skrede et al., 2000; Meyer et al., 1997;). Anthocyanins are a viable










replacement for synthetic colorants due to their bright, attractive colors and water

solubility, which allows their incorporation into a variety of food systems (Rodriguez-

Saona et al., 1999). However, limitations exist for their commercial application due to

high raw material costs and their poor stability that is affected by their chemical structure,

environmental factors, and the presence of additional phytochemicals in solution. Due to

these constraints, a need exists to find stable, inexpensive anthocyanin pigments with a

diverse array of functional properties food and nutraceutical applications.

Anthocyanin intermolecular copigmentation reactions are common in nature and

result from association between pigments and cofactors such as polyphenolics and/or

metal ions, or other anthocyanins (self-association). Preferably formed under acidic

conditions, these weak chemical associations can augment anthocyanin stability and

increase antioxidant properties (Mazza and Brouillard, 1990; Boulton, 2001; Malien-

Aubert et al., 2001). Studies have suggested that the copigmentation phenomenon is the

main anthocyanin stabilizing mechanism in plants (Mazza and Brouillard, 1990; Boulton,

2001). Polyphenolics are the predominant cofactors present in anthocyanin-containing

fruits and vegetables, and increased anthocyanin stability has been attributed to their high

concentrations in foods (Mazza and Brouillard, 1990; Boulton, 2001; Malien-Aubert et

al., 2001). Malien-Aubert et al. (2001) described how the diversity of polyphenolic

compounds among different anthocyanins sources might affect anthocyanin stability, yet

additional research on how these polyphenolics influence anthocyanin stability via

copigmentation reactions has not been conducted.

The obj ective of this study was to characterize the major polyphenolics and

anthocyanins present in agai pulp, and to determine their contribution to the overall










antioxidant capacity of this palm fruit. Color and pigment stability against hydrogen

peroxide, ascorbic acid, and the presence/absence of naturally occurring cofactors was

also determined and compared to other commercially available anthocyanin sources.

Results of these studies can be used to determine application and functional properties of

agai polyphenolics in a variety of food products.

Materials and Methods

Materials

Pasteurized, frozen agai pulp was kindly donated by Amazon Energy, LLC

(Greeley, CO) and was shipped overnight to the Department of Food Science and Human

Nutrition at the University of Florida. The pulp was thawed, centrifuged (2,000 x g) at 4

oC for 15 min to separate lipids from the aqueous slurry, and subsequently filtered

through Whatman #1 filter paper. The aqueous portion was then partitioned into

lipophilic and hydrophilic extracts by the addition of petroleum ether and acetone,

respectively. The upper petroleum ether phase was removed and evaporated under a

gentle stream of nitrogen and re-dissolved in a known volume of acetone and ethanol

(1:1). Acetone was removed from the lower aqueous phase under reduced pressure at

temperatures <40 oC, and the resultant fraction containing hydrophilic compounds was

diluted to a known volume with acidified water (0.1% HC1). Polyphenolics from the

aqueous phase were subsequently concentrated using Cls Sep-Pak Vac 20 cc mini-

columns (Waters Corporation, Mass. U.S.A.). Residual sugars and organic acids were

removed with water (0.01% HC1), and polyphenolic compounds recovered with acidified

methanol (0.01% HC1). Methanol was removed from the polyphenolic fraction using

vacuum evaporation at <40 oC, and the resulting isolate was re-dissolved in a known

volume of acidified water.









Commercially available anthocyanin extracts from black carrot (Daucus carota;

Exberry, Tarrytown, NY), red cabbage (Bra~ssica oleracea) (Exberry), red grape (Vitis

vinifera) (San Joaquin Valley Concentrates, Fresno, CA), purple sweet potato (Ipomea

batata) (Food Ingredients Solutions, New York, NY), and a non-commercial extract from

red hibiscus flowers (Hibiscus sabdariffa)ddd~~~ddd~~~dd were used for color stability evaluation. Each

pigment source was dissolved in citric acid buffer (pH 3.5; 0.1 M), and polar compounds

removed with Cls columns as previously described. Color and anthocyanin stability were

then assessed against agai for comparison.

Color Stability

Anthocyanin color stability of each pigment source was assessed in the presence of

hydrogen peroxide (0 and 30 mmol/L) at 10, 20, and 30 oC, respectively. Stock solutions

of each anthocyanin source were diluted with citric acid buffer (pH 3.5) to give a final

absorbance value of 1.5 at their respective wavelength of maximum absorbance. Samples

were placed into a water bath or refrigerated storage and allowed to reach the desired

temperature at which a hydrogen peroxide solution was added. Loss of absorbance was

measured periodically over time and percent color retention calculated as a percentage of

the initial absorbance reading. Insignificant changes in absorbance values were observed

for control treatments (no hydrogen peroxide) over 360 minutes of incubation.

Effect of Copigmentation

The effect of naturally occurring intermolecular copigmentation on anthocyanin

stability in the presence and absence of ascorbic acid was also evaluated using in vitro

model systems. Naturally occurring cofactors were removed by additionally loading each

anthocyanin source onto Cls cartridges as previously described. Following elution of

polar compounds with water, the cartridge was first washed with ethyl acetate to elute









phenolic acids and flavonoids, followed by acidified methanol to remove anthocyanins.

Ethyl acetate and methanol isolates were then evaporated under vacuum at <40 OC, and

re-dissolved in a known volume of citric acid buffer. Anthocyanin recovery was >96%

for all sources. Anthocyanin color stability was evaluated using an in vitro model

simulating a soft drink beverage system that contained anthocyanins absorbancee value of

1.5) dissolved in citric acid buffer, sucrose (100 g/L), and sodium azide (50 mg/L) to

control microbial growth. Stock solutions were sub-divided and evaluated with and

without polyphenolic cofactors, and again sub-divided for evaluation with and without

ascorbic acid (450 mg/L). Data were compared to a control that contained an equivalent

volume of citric acid buffer. Each treatment was individually sealed into screw-cap vials

(10 mL), and stored in the dark at 37 OC for 30 days. Samples were collected every day

during the first 8 days of the study, and subsequently every other day until the end of the

study .

Phytochemical Analyses

Individual anthocyanin 3-glycosides present in agai were quantified according to

the HPLC conditions of Skrede et al. (2000) using a Dionex HPLC system and a PDA

100 detector. Compounds were separated on a 250 x 4.6 mm Supelcosil LC-18 column

(Supelco, Bellefonte, PA) and quantified using a cyanidin standard (Polyphenols

Laboratories AS, Sandnes, Norway). Anthocyanins were also characterized based on

PDA spectral interpretation from 200-600 nm, and identification additionally confirmed

following acid hydrolysis into their respective aglycones with 2N HCI in 50% v/v

methanol for 60 min at 90 oC.

Maj or flavonoids and phenolic acids present in agai were separated by HPLC using

modified chromatographic conditions of Talcott et al. (2001). Separations were










performed on a 250 mm X 4.6 mm i.d. Acclaim 120-Cls column (Dionex, Sunnyvale,

CA) with a Cls guard column. Mobile phases consisted of water (phase A) and 60%

methanol (phase B) both adjusted to pH 2.4 with o-phosphoric acid. A gradient solvent

program ran phase B from 0 to 30% in 3 min; 30-50% in 5 min; 50-70% in 17 min; 70-

80% in 5 min; and 80-100% in 5 min, and held for 10 min at a flow rate of 0.8 mL/min.

Polyphenolics were identified by spectroscopic interpretation, retention time, and

comparison to authentic standards (Sigma Chemical Co., St. Louis, MO).

Six isolates were obtained from the extraction of agai pulp that included whole

pulp, lipophilic fraction, Cls non-retained, Cls bound phenolics and anthocyanins, ethyl-

acetate soluble polyphenolics, and anthocyanins. Each fraction was evaluated for

antioxidant capacity using the oxygen radical absorbance capacity assay against a

standard of Trolox as described by Talcott et al. (2003b). Each isolate was appropriately

diluted in pH 7.0 phosphate buffer prior to pipetting into a 96-well microplate with

corrections made for background interference due to phosphate buffer and/or extraction

solvents.

Anthocyanin content in each in vivo model system was determined with the pH

differential method of Wrolstad (1976) and quantified using equivalents of the

predominant anthocyanin present (cyanidin 3-glucoside for agai and hibiscus; cyanidin 3-

sophoroside for black carrot and red cabbage; malvidin 3-glucoside for red grape;

pelargonidin 3-rutinoside for purple sweet potato) (Malien-Aubert et al., 2001; Wrolstad,

1976; Hong and Wrolstad, 1990). Percentage of polymeric anthocyanins was determined

based on color retention in the presence of potassium metabisulfite (Wrolstad, 1976),









while instrumental CIE color characteristics (chroma, and hue angle) were measured

using a Minolta Chroma Meter CR-300 Series (Minolta Co., Ltd., Osaka, Japan).

Statistical Analysis

Anthocyanin stability against hydrogen peroxide was designed as a 6 x 2 x 3 full

factorial that included six anthocyanin sources, two hydrogen peroxide concentrations,

evaluated at three temperatures. Anthocyanin stability in the presence of cofactors and

ascorbic acid was designed as a 6 x 2 x 2 full factorial that included six anthocyanin

sources, two ascorbic acid levels, in the presence or absence of native cofactors. Data for

these evaluations and those for agai characterization represent the mean of three

replicates at each sampling point. Multiple linear regression, analysis of variance, and

Pearson correlations were conducted using JMP software (SAS, Cary, NC) and mean

separation using the LSD test (P < 0.05).

Results and Discussion

Anthocyanin and Polyphenolic Characterization

Due to recurrent issues associated with the instability of anthocyanins during

processing and storage, the food industry is constantly looking for novel, inexpensive and

stable sources of pigments. Anthocyanins present in agai may offer a new source of these

pigments, however their stability has yet to be determined. Furthermore, the

characterization of the maj or polyphenolic compounds in agai and their overall

contribution to the antioxidant capacity has not been previously investigated. Therefore,

this study examined the polyphenolic composition and the anthocyanin stability of agai

under a variety of experimental conditions as compared to other commercially available

anthocyanin sources.
























8.0 9.0 10.0 11.0 12.0 13.0


14.0 15.0 16.0 17.0 18.0 19.0 20.0 21.1


Time (min)

Figure 3-1. HPLC chromatogram of A. anthocyanin 3-glucosides monitored at 520 nm
(Peak assignments: 1. cyanidin 3-glucoside; 2. pelargonidin 3-glucoside) and
their B. aglycones (Peak assignments: 3. cyanidin; 4. pelargonidin) present in
agai (Euterpe oleracea Mart.).


Figure 3-2. HPLC chromatogram of A. phenolic acids monitored at 280 nm and B.
flavonoids monitored at 360 nm present in agai (Euterpe oleracea Mart.).
Peak assignments: 1. gallic acid; 2. p-coumaric acid; 3. protocatechuic acid; 4.
(+)-catechin; 5. p-hydroxybenzoic acid; 6. vanillic acid; 7. gallic acid
derivative-2; 8. gallic acid derivative-5; 9. gallic acid derivative-3; 10. gallic
acid derivative-1; 11. ferulic acid; 12. (-)-epicatechin; 13. gallic acid
derivative-4; 14. ellagic acid; 15. ellagic acid derivative.


Time (min)









Figure 3-1 shows a typical HPLC chromatogram of anthocyanin 3-glycosides

extracted from agai that when hydrolyzed yielded cyanidin (1,040 mg/L pulp) and

pelargonidin (74 mg/L pulp) as the only compounds detected. Spectroscopic analysis

before and after acid hydrolysis confirmed the presence of each anthocyanidin and

tentative identification of a monoglycoside attached to the C-3 position, presumably a

glucose derivative, was made based on A4 4n'Anus,, ratios (~3 3%) as described by Hong and

Wrolstad (1990). Presence of hydroxy-substituted aromatic acids attached to the

glycoside (acylated moieties) was not found for either compound, as shown by the

absence of their typical absorption spectrum in the 3 10-340 nm range.

The predominant polyphenolics present in agai pulp were ferulic acid > epicatechin

> p-hydroxy benzoic acid > gallic acid > protocatechuic acid > (+)-catechin > ellagic acid

> vanillic acid > p-coumaric acid at concentrations that ranged from 17 to 212 mg/L as

reported in Table 3-1. Additionally, five compounds were identified with spectroscopic

characteristics comparable with gallic acid and were tentatively identified as gallotannins,

while one compound shared spectroscopy similarities with ellagic acid and was

tentatively identified as an ellagic acid glycoside (Table 3-1; Figure 3-2). Additional

confirmation of these compounds was made following acid hydrolysis, as these

compounds were no longer detected and a corresponding increase in either gallic acid or

ellagic acid concentrations was observed.


Antioxidant Capacity

Agai pulp was found to have a relatively high antioxidant content (48.6 Clmol

Trolox equivalents/mL) with respect to other anthocyanin-rich fruits such as highbush

blueberries (4.6- 31.1 Clmol TE/g) (Ehlenfeldt and Prior, 2001), strawberries (18.3-22.9)










(Kalt et al., 1999), raspberries (19.2-22.6) (Kalt et al., 1999), blackberries (13.7-25.1)

(Wang and Lin, 2000), cranberries (8.20-145) (Wang and Stretch, 2001), and muscadine

grape juice (18.2-26.7) (Talcott et al., 2003).


Table 3-1. Anthocyanin and polyphenolic content (mg/L fresh pulp) of agai (Euterpe
oleracea Mart.).


Content
(mg/ L fresh pulp)
1,040 & 58.2
74.4 & 2.90
212 & 5.29
129 & 3.28
80.5 & 2.00
64.5 A 1.64
64.4 & 1.64
60.8 & 0.98
55.4 & 1.39
33.2 & 1.39
17.1 + 1.23
47.3 A 1.40
18.4 & 0.89
17.3 A 1.25
13.3 & 0.96
3.9 & 0.18
19.5 & 0.40


Polyphenohic

Cyanidin 3-glucoside
Pelargonidin 3-glucoside
Ferulic acid

(-)-Epicatechin
p-Hydroxy benzoic acid
Gallic acid
Protocatechuic acid

(+)-Catechin
Ellagic acid
Vanillic acid

p-Coumaric acid
Gallic acid derivative-1
Gallic acid derivative-2
Gallic acid derivative-3
Gallic acid derivative-4
Gallic acid derivative-5

Ellagic acid derivative


Fractionation of agai phytochemicals based on solubility and affinity characteristics

was conducted to determine the distribution of antioxidant compounds among the

isolates. Similar antioxidant content was observed for the whole pulp, Cls retained

phenolics (phenolic acids and anthocyanins), and the anthocyanins alone while ethyl

acetate-soluble phenolics, the liphophilic, and Cls non-retained isolates had appreciably









lower contributions to the total antioxidant content (44, 8, and 1.2%, respectively) (Figure

3-3). Results indicated that when ethyl acetate-soluble phenolics and anthocyanin

fractions were evaluated alone for antioxidant capacity, their sum was higher that values

obtained for the total Cls bound polyphenolics. Although these fractions were not

recombined again for analysis, there is indication that physical and/or chemical

interactions among constituents in these fractions unfavorably impacted radical-

scavenging properties. Previous studies have demonstrated antagonistic interactions

between polyphenolics such as quercetin and caffeic acid (Howard et al., 2000), or

cyanidin in combination with catechin and ellagic acid (Meyer et al., 1998) all of which

are present in agai. However, the effectiveness of an antioxidant compound is generally

dependent on the polarity of the testing system, the nature of the radical, and type of

substrate protected by the antioxidant (Prior et al., 2003). The diversity of antioxidant

polyphenolics present in agai create a complex matrix from which evaluations can be

made, but it was apparent that anthocyanins were the predominate contributors to the

antioxidant capacity and their presence with other polyphenolics resulted in an

underestimation of the overall antioxidant capacity of agai pulp.



Color Stability as Affected by Hydrogen Peroxide and Temperature

The anthocyanin color stability of agai was assessed spectrophotometrically in the

presence of hydrogen peroxide (0 and 30 mmol/L) at 10, 20, and 30 oC, and compared to

the five other anthocyanins sources. Regression analysis was used to determine adequacy

of the model describing kinetics of color degradation over time, and confirmed that

degradation rates followed first order kinetics (P<0.05) in agreement with previous

reports (Ozkan et al., 2002; Taoukis et al., 1997). Degradation rate constants (P1) and half






47


life (tl/2) ValUeS of anthocyanin color were calculated according to Taoukis et al. (1997):

In At / Ao = pi time, and tl/2 = In 0.5 / P1; where Ao is the initial absorbance value, and

At is the absorbance value at a given time. Increments in storage temperature allowed for

calculation of a temperature quotient (Qlo) for each anthocyanin source (Ozkan et al.,

2002), which is presented in Table 3-2.


Whole pulp Lypophilic Clg bound Ethyl acetate Anthocyanins Clg non-retained
phenolics soluble phenolics

Fraction

Figure 3-3. Antioxidant capacity of different phytochemical fractions (whole pulp,
liphophilic extract, Cls bound polyphenolics, ethyl acetate-soluble phenolics,
anthocyanins, and Cls non-retained) of agai (Euterpe oleracea Mart.). Bars
represent standard error of the mean (n=6). Antioxidant capacity quantified
using Trolox equivalents (TE).









Table 3-2. The effect of hydrogen peroxide (30 mmol/L) and temperature on kinetic
parameters of color degradation for different anthocyanin sources.


P1 1 tl/2 2 10o 3
Pigment 10 oC 20 oC 30 oC 10 oC 20 oC 30 oC 10-20 oC 20-30 oC
Agai 7.7 11.3 13.9 90 c4 61 c 50 c 1.5 1.2
Hibiscus 6.3 9.8 11.7 110 b 71 b 59 b 1.6 1.2
Purple Potato 5.8 9.5 12.4 120 b 73 b 56 b 1.6 1.3
Black Carrot 8.7 14.4 18.7 80 d 48 d 37 d 1.7 1.3
Red Cabbage 8.4 12.5 15.9 83 d 55 c 44 d 1.5 1.3
Red Grape 2.2 4.2 5.6 315 a 165 a 124 a 1.9 1.3

1Reaction rate constant (pl 103, min-1. 2Half-life (min) of initial absorbance value for
each pigment source. 3Temperature dependence quotients of color degradation as affected
by increments in reaction temperature from 10 to 20 oC, and 20 to 30 oC, respectively.
4Values with similar letters within columns of each reaction temperature are not
significantly different (LSD test, P<0.05).

Compared to agai and the other anthocyanin sources, greater color stability (tl/2)

was observed for red grape anthocyanins, results that were attributed to their high

polymeric anthocyanin content (Table 3-3). The predominantly acylated anthocyanins

from black carrot and red cabbage displayed reduced color stability at each temperature

when compared to the non-acylated agai and hibiscus anthocyanins and to the acylated

anthocyanins from purple sweet potato. Differences in half-life values (Y) between red

grape and other anthocyanin sources increased linearly with reaction temperature (Y=

m*Temperature, R2=0.99), with similar values obtained for these differences for hibiscus,

purple potato, and agai (m=0.135), and more pronounced for red cabbage and black carrot

(m=0.2). Increasing the reaction temperature from 10 to 20 OC significantly increased

color degradation (Qlo ~ 1.6) for all sources except red grape, were a 1.9-fold increase

was observed. This was in contrast to the relatively slower rate of color loss (Qlo = 1.3)









observed for all the anthocyanin sources when the reaction temperature was increased

from 20 to 30 oC.

Rates of anthocyanin degradation during storage significantly varied among

sources and likely occurred due to factors such as varying molar ratios between reactants

(anthocyanins and/or polyphenolics with peroxide), non-anthocyanin polyphenolic

concentration, secondary free radical formation, or other oxidative reactions such as o-

quinone formation involving phenolics and anthocyanins (Boulton, 2001; Ozkan et al.,

2002; Talcott et al., 2003a). Results of this study indicate that acylated anthocyanins were

not more stable than their non-acylated counterparts in the presence of hydrogen

peroxide. This observation may have been influenced by the presence of additional non-

anthocyanin polyphenolics in solution, emphasizing the importance of conducting color

stability evaluations with pigment sources used industrially. These polyphenolics also

form copigment complexes with anthocyanins, resulting in a more intense color that may

be several folds higher in color intensity due to hyperchromic and bathochromic

spectroscopic shifts. Therefore, color comparisons among diverse pigments sources are

difficult since molar ratios between reactants (hydrogen peroxide and anthocyanins) vary

between sources for a given color intensity. Despite these varying ratios, industrial use of

anthocyanins is based on color shade and intensity and their relative color stability under

oxidizing conditions is very important for many food and beverage applications.

Color Stability in the Presence of Ascorbic Acid and Natural Cofactors

A primary concern regarding the use of anthocyanins in the food industry is their

inherent instability during processing and storage. Moreover, a growing trend in the food

industry is to fortify juices with various phytonutrients for both quality and health-

promoting benefits. Ascorbic acid is among the most common fortificants used for this










purpose; however, when present together with anthocyanins, their combination will lead

to mutual degradation that causes the loss of nutrients and color stability during

processing and storage. Therefore a need exists to find an inexpensive and stable

anthocyanin pigment that possesses a diversity of functional properties for food and

nutraceutical applications. The stability of agai anthocyanins was evaluated in the

presence of ascorbic acid (0 and 450 mg/L) under accelerated storage conditions (370g)

using an in vitro model system as compared to those of other common anthocyanin

sources (hibiscus, black carrot, red cabbage, red grape, and purple sweet potato). A

further examination of how naturally occurring cofactors affect color stability within a

given pigment source was also investigated.

Differences in spectroscopic properties and color attributes among in vitro juice

model systems prepared with the six anthocyanin sources were initially observed (Table

3-3). Despite model systems with the same initial color value absorbancee value of 1.5),

color differences were apparent and due to the diversity of ring substitutions (hydroxy,

sugar, or acyl-linked organic acids) among sources. As previously discussed, the nature

of polyphenolic cofactors and their relative molar ratio to anthocyanin concentration were

also influential on color characteristics of each source. Isolation of polyphenolic cofactors

revealed not only the appreciable difference in color exhibited by each pigment, but also

their specific role in anthocyanin stability. Red grape anthocyanins had the largest

hyperchromic shift (49%), followed by purple potato (35%), hibiscus and black carrot

(19.5% on average), and agai and red cabbage (7% on average) due to the presence of

these native cofactors with a slight bathochromic shift in wavelength observed for agai









and red grape anthocyanins. These spectroscopic features translated into a more intense

colored solution and were influential on overall color stability.


Table 3-3. Percent monomeric anthocyanins and CIE color attributes of a juice model
system (pH 3.5, 100 mg/L sucrose) prepared with different pigment sources,
along with their correspondent hyperchromic and bathochromic shifts due to
the presence of naturally occurring polyphenolic cofactors.

% Monomeric Hyperchromic Bathochromic
Pigent anthocyanins Chroma HLue Laxa shift shift
Acai 76.2 c3 20.1 18.2 515 nm 6% 1 nm
Hibiscus 80.3 b 31.2 35.2 521 nm 19% 0 nm
Purple Potato 77.5 c 23.1 13.6 526 nm 35% 0 nm
Black Carrot 77.8 c 22.9 10.2 521 nm 20% 0 nm
Red Cabbage 92.2 a 19.8 -13.9 526 nm 8% 0 nm
Red Grape 58.1 d 17.2 6.1 528 nm 49% 2 nm

1Wavelength of maximum absorption for each pigment source. 2Difference in absorbance
between anthocyanin solutions with and without naturally occurring polyphenolic
cofactors. 3Values with similar letters within columns of each reaction temperature are
not significantly different (LSD test, P<0.05).


Results from obj ective color analysis concluded that chroma values only differed

slightly within anthocyanin sources in accordance with those observed in previous studies

(Stintzing et al., 2002; Giusti and Wrolstad, 2003), except for hibiscus, which had an

appreciably higher value than other sources. Hue angles significantly differed among

pigment sources due to various ring substitutions and were generally lower for acylated

anthocyanins (Table 3-3), giving the later anthocyanins a characteristic intense purple

color in solution that corresponded to their longer wavelength of maximum absorbance.

Red grape anthocyanins were a notable exception due to its high polymeric anthocyanin

content in relation to the other sources. Polymeric anthocyanins typically have greater

color stability over their monomeric counterparts (Es-Safl et al., 2002; Malien-Aubert et









al., 2002; Mateus et al., 2003) and the high content in red grape (58%) appreciably

influenced its color stability during storage. The red grape extract used in this study was

obtained as a by-product of the wine industry, and may contain anthocyanins

polymerized with oligomeric flavanols and/or acetaldehyde (Es-Safi et al., 1999; Eiro and

Heinonen, 2002; Es-Safi et al., 2002; Mateus et al., 2003;) which gives this extract

remarkable color and storage stability.


Table 3-4. The effect of ascorbic acid (0 and 450 mg/L) and naturally occurring
polyphenolic cofactors (presence or absence) on kinetic parameters of
anthocyanin degradation during storage at 37 oC of in vitro models systems
(pH 3.5, 100 mg/L sucrose) prepared with different pigment sources.

No Ascorbic Acid Ascorbic Acid (450 mg/L)
With Cofactors No Cofactors With Cofactors No Cofactors
Pigment j11 tl/2 2 P1 tl/2 P1 tl/2 P1 tl/2
Acai 1.8 385 d3 2.2 319 d*4 55 13 d 49 14 c*
Hibiscus 2.2 315 e 5.3 131 f* 19 37 c 60 11 c*
Purple Potato 0.8 866 b 2.0 355 c* 18 38 c 52 13 c*
Black Carrot 1.3 533 c 1.4 486 b* 52 13 d 60 12 e*
Red Cabbage 0.3 2,450 a 0.6 1,150 a* 14 50 b 34 20 b*
Red Grape
1.3 533 c 2.9 243 e* 11 62 a 23 30 a*
1Reaction rate constants (p3 103, hours- ). 2Half-life (hours) of initial anthocyanin
content. 3Values with similar letters within columns are not significantly different (LSD
test, P<0.05). 4MLeans with an asterisk (*) for each pigment source indicate a significant
effect (LSD test, P<0.05) due to presence of naturally occurring cofactors when
compared to the same treatment with an equivalent ascorbic acid content.

Regression analysis found that anthocyanins under the accelerated storage

conditions of the in vitro models, with and without native cofactors, followed first order

kinetics (P<0.05). Kinetic parameters were calculated as previously described, with

anthocyanin content used as the independent variable. Acylated anthocyanin sources

along with those from red grape were found to be more stable than their non-acylated









counterparts, independent of ascorbic acid content. Naturally occurring cofactors were

shown to be key elements to decrease anthocyanin degradation during storage, an effect

that was more pronounced for non-acylated anthocyanin sources.

Half-life evaluation of pigment stability revealed that acylated anthocyanin sources

generally had increased stability (tl/2 >823 h) with respect to non-acylated sources in the

absence of ascorbic acid. A notable exception was black carrot anthocyanins (tl/2=-515 h),

which showed reduced stability with respect to that of non-acylated red grape

anthocyanins (tl/2 =540 h). By comparison, the red grape anthocyanins had reduced

stability in the absence of ascorbic acid, especially in relation to the high stability

observed against hydrogen peroxide, yet in the presence of ascorbic acid the stability was

again the highest among anthocyanin sources. Red grape anthocyanins (tl/2 = 62 h) were

the most stable compounds in the presence of ascorbic acid followed by red cabbage (tl/2

=50 h), hibiscus and purple potato (tl/2 =37 h), and lastly agai and black carrot (tl/2=-13 h).

Overall, anthocyanin degradation was significantly increased in the presence of

ascorbic acid as compared to non-fortified controls, generally having a more pronounced

effect on acylated anthocyanin sources (40 to 46-fold) than for non-acylated sources (8.4

to 30-fold). A notable exception was purple potato anthocyanins, where ascorbic acid

increased color degradation by 23-fold. Red grape and hibiscus anthocyanins exhibited

the smallest change with only a 8-fold increase in degradation rates. Naturally occurring

polyphenolic cofactors were found to significantly increase anthocyanin retention by up

to 2.4-fold in the absence of ascorbic acid, an effect that was less pronounced for agai

(1.2-fold) and black carrot (1.1-fold) anthocyanins. A similar protective effect conferred

by intermolecular copigmentation was observed in the presence of ascorbic acid for black









carrot, agai and red grape, yet additional increments in this protective effect was observed

for hibiscus (+0.9-fold) and both purple potato and red cabbage (+0.4-fold).

The increased stability of acylated anthocyanins with respect to non-acylated

pigment sources was likely related to the natural synthesis of acylated organic acids and

diversity of glycosidic linkages in relation to these acylated moieties (Rodriguez-Saona et

al., 1999; Boulton, 2001; Giusti and Wrolstad, 2003;). The aromatic or aliphatic acyl

groups covalently bound to these anthocyanins were shown to stack on the planar,

polarizable nuclei of the anthocyanin, protecting the pyrylium nucleus from the

nucleophilic attack of water at carbon 2 (Rodriguez-Saona et al., 1999; Boulton, 2001).

Red cabbage and purple potato extracts typically contain cinnamic acid derivatives

diacylated to their anthocyanins that can simultaneously stack on both faces of the

anthocyanin chromophore in a sandwich-type complex and thus offer greater color

stability, while black carrots contain only monoacylated moieties that can only protect

one face of the pyrylium ring (Mazza and Brouillard, 1990; Rodriguez-Saona et al., 1999;

Boulton, 2001; Malien-Aubert et al., 2001; Stintzing et al., 2002; Es-Safi et al., 2002;

Giusti and Wrolstad, 2003). The observed differences in stability between the various

sources of acylated anthocyanins in this study were likely related to the nature, number,

and position of these substitutions.

For a given set of pH conditions, intramolecular copigmentation exerts a protective

effect against anthocyanin degradation by keeping a larger proportion in their flavylium

ion forms. Consequently, formation of intermolecular complexes will also take place with

these acylated anthocyanins and thus give an additional protective effect against color

degradation. Results of this study also demonstrated and confirmed that both forms of










copigmentation (intra- and intermolecular) cooperatively acted to prevent anthocyanin

color degradation, as demonstrated by similar pigment half-life values (12.5 days) in

black carrot and purple potato after removal of naturally occurring cofactors.

The stabilization effect conferred by intermolecular copigmentation has been

attributed to hydrophobic interactions between anthocyanins and polyphenolic

compounds, consequently protecting the pigment from further polymerization and

degradation reactions (Mazza and Brouillard, 1990; Rodriguez-Saona et al., 1999; Es-

Safi et al., 1999; Boulton, 2001; Eiro and Heinonen, 2002; Es-Safi et al., 2002;). Previous

studies have shown that not only ascorbic acid but also its degradation by-products,

including those from carbohydrates such as furfural and other aldehydes, can participate

in anthocyanin degradation during processing or storage (Eiro and Heinonen, 2002).

Intermolecular copigmentation exerts a protective effect on anthocyanin degradation as

cofactors compete with anthocyanins and preferentially react in the condensation

reactions (Es-Safi et al., 1999; Malien-Aubert et al., 2001; Es-Safi et al., 2002). The

increased protection observed for a specific pigment source due to the presence of

cofactors is most likely related to the type, and content of polyphenolics present, as a

higher copigment/pigment molar ratio could have occurred for a determined source.

Moreover, specific polyphenolics or classifications of polyphenolics are more likely to

form stable intermolecular complexes with anthocyanins than others (Boulton, 2001;

Malien-Aubert et al., 2001;; Eiro and Heinonen, 2002).

Conclusion

Characterization of the maj or polyphenolic compounds present in agai and their

contribution to the antioxidant capacity was determined for the first time. The effect of

exogenously added cofactors on color enhancement and stability was previously









evaluated in many food systems containing isolated anthocyanins, model juices, and

wine, yet the effect of naturally occurring cofactors on color stability was not previously

investigated prior to this study. The stability of agai anthocyanins as a new source of

anthocyanin pigments was also established and can be used to determine application and

functional properties of agai in a variety of food and nutraceutical products.















CHAPTER 4
STABILITY OF COPIGMENTED ANTHOCYANINTS AND ASCORBIC ACID INT
MUSCADINE GRAPE JUICE PROCESSED BY HIGH HYDROSTATIC PRESSURE

Introduction

Muscadine grapes (Vitis rotundifolia) are the predominant grape variety grown in

the southern U.S. with excellent potential for commercial expansion and value-added

development. Deleterious changes in color and phytochemicals appreciably affect

muscadine grape products during processing and storage, as they do with other

anthocyanin containing juices, and are an impediment to future market development.

Processing technologies and/or strategies that could substantially improve quality

attributes of these products are consequently vital for the economic growth of this crop.

High hydrostatic pressure (HHP) is a promising alternative to traditional thermal

pasteurization technologies and may lessen detrimental effects to thermolabile

phytonutrients (Gomez et al., 1996; Sun et al., 2002; Poei-Langston and Wrolstad, 1981).

However, a downside of this technology is the presence and/or activation of residual

enzymes, such as polyphenol oxidase (PPO), lipoxygenase, and peroxidase, during

processing and storage, which may be partially responsible for oxidative degradation

reactions. Quality and phytochemical deterioration due to enzyme action may be further

complicated due to interactions between anthocyanins and ascorbic acid, when the latter

compound is present in the juice or is externally added, resulting in their mutual

destruction (Poei-Langston and Wrolstad, 1981; Garcia-Viguera and Bridle, 1999 Garzon

and Wrolstad, 2002).









A previous study with HHP and muscadine grape juice demonstrated that

phytochemical losses caused by processing were presumably due to the activation of

residual oxidases during juice extraction and/or autoxidative mechanisms resulting in co-

oxidation of anthocyanins and ascorbic acid (Talcott et al., 2003a). Their study utilized a

commercially available polyphenolic extract from rosemary aimed to reduce

phytonutrient degradation through copigmentation, yet the overall quality of the juice was

adversely impacted presumably due to copious amounts of additional polyphenolics

present in the extract that were substrates for oxidative enzymes. A goal of the current

study was to confirm the role of enzymes in phytonutrient degradation during HHP

processing and to establish a potential remediation strategy using partially purified

anthocyanin cofactors from two plant sources. Addition of individual polyphenolic

cofactors has been reported to increase anthocyanin stability during processing and

storage (Malien-Aubert et al., 2001; Eiro and Heinonen, 2002), and their effectiveness in

forming intermolecular linkages with anthocyanins has been linked to their specific

structure and concentration. However, the use of individual polyphenolic cofactors may

not be a feasible option for the food industry and thus there is a need for a concentrated

source of mixed anthocyanin stabilizing agents from natural sources. Evaluation of the

effect of copigment addition during processing and storage regimes, especially in the

presence of residual oxidase enzymes, is important for determining their efficacy in

preventing phytonutrient degradation and their interaction with other food components.

The obj ective of this study was to assess the phytochemical stability of muscadine

grape juice (Vitis rotundifolia) processed by HHP and fortified with ascorbic acid. The

effect of exogenously added polyphenolic cofactors purified from rosemary (Rosmarinus









officinalis) and thyme (Thymus vulgaris) was also investigated as a means to improve

overall phytochemical stability. The role of residual PPO activity was also investigated to

gain knowledge on the mode of deterioration and potential solutions for increased storage

stability of fruit juices containing anthocyanins.

Materials and Methods

Materials and Processing

PPO activity during juice extraction

Muscadine grapes (cv. Noble) were obtained from a local grower in central Florida

and held frozen (-20oC) until needed. Grapes were rapidly thawed by placing them under

running tap water and hand-sorted for uniformity of ripeness. Response Surface

Methodology (RSM) was used to determine the initial PPO activity of muscadine grape

juice under different manual juice extraction procedures (0-24 min, 46-74 OC). PPO

activity and browning index (BI) were used as the dependent variables in the

experimental design that was repeated two times, and each study required 11 experiments

with 4 factorial points and 4 star points to form a central composite design, and 3 center

points for replication (Kim et al., 2001). Experimental data were analyzed by regression

analysis to determine the adequacy of the mathematical models. The RSM models were

used to select the juice extraction conditions for the subsequent study that investigated

the HHP-induced PPO activation. Residual PPO activity in the juice was determined

according to a modified polarographic method described by Kader et al. (1997) using a

YSI 5300 oxygen monitor (Yellow Springs, OH) equipped with a Clark-type electrode in

a 3.1 mL jacketed cell at 35 OC. The reaction was started when 0.2 mL of 0. 12M catechin

was added to 2.8 mL of grape juice mixed with 1 mL of 0. 1M phosphate buffer at pH 3.5.

The assay was carried out in air-saturated solutions agitated with a magnetic stirrer and









the electrode calibrated using air-saturated water (230 nmol 02 / ml H20). Enzymatic

activity was determined from the linear portion of the oxygen consumption curve,

reported as nmoles of oxygen consumed per second (nkat), and expressed as a percentage

of the control juice (100% activity) that was extracted at 25 oC without a heating time-

temperature regime. BI was used as an indirect method to monitor anthocyanin

degradation during the different juice extraction procedures and was calculated as the

ratio of absorbance values obtained at 420 and 520 nm (Buglione and Lozano, 2001).

Juice extraction and processing

Rosemary (Rosmarinus officinalis L.) and thyme (Thymus vulgaris L. ) were

obtained from a local market and the biomass exhaustively extracted with water at 90 OC

for 8 hours. The resulting dark brown liquid was adjusted to pH 2.0 with IM HCI and

centrifuged at 17,000 rpm for 15 min to remove insoluble matter. Polyphenolics were

subsequently concentrated and purified using Cls Sep-Pak Vac 20 cc mini-columns

(Water Corporation, Mass., USA). Polar constituents were removed with acidified water

(0.01% v/v HC1) and polyphenolic compounds subsequently eluted with methanol

(0.01% v/v HC1), solvent that was later evaporated under reduced pressure at < 40 OC.

The resulting polyphenolic extract was re-dissolved in a known volume of 0. 1M citric

acid solution.

Based on the initial RSM evaluations, muscadine grapes were crushed and heated

in an open steam kettle to 46 OC for 11 min to retain enzymatic activity (115% PPO

activity) and facilitate juice extraction during pressing in a hydraulic basket press

(Prospero's Equipment, Cort, NY). Juice was immediately filtered first through

cheesecloth followed by vacuum filtration through a 1cm bed of diatomaceous earth. The

juice was then divided into two portions for copigmentation (0 and 100 cofactor-to-









anthocyanin molar ratio) with either rosemary or thyme extracts. Ratios were adjusted

considering the molar concentration of total phenolics in rosemary and thyme extracts

(0.52 and 0.72 M gallic acid equivalents, respectively) divided by the total anthocyanins

in muscadine grape juice (8.08 mM cyanidin 3-glucoside equivalents). Juices at each

cofactor concentration were again divided and half fortified with ca. 450 mg/L of

ascorbic acid and compared to an equivalent volume of citric acid buffer (pH 3.5, 0. 1M)

as the control. Sodium azide (50 mg/L) was added to retard microbial growth throughout

the analytical determinations. Treatments were prepared for HHP by placing 8 mL juice

portions into heat sealed plastic ampules and processed at 400 and 550 MPa for 15 min

(Stansted Fluid Power, UK). Following HHP processing, ampules were subdivided and

half stored under refrigerated conditions and analyzed within 48 hr of processing. The

remaining half was stored in the dark at 25 OC for 21 days.

Chemical Analyses

Initial anthocyanin content in the juice was determined by the pH differential

spectrophotometric method of Wrolstad (1976) and quantified as cyanidin 3-glucoside

equivalents. Total soluble phenolic concentration in each cofactor source was measured

using the Folin-Ciocalteu assay (Talcott et al., 2000), and quantified as gallic acid

equivalents. Individual anthocyanin 3,5-diglycosides were quantified according to the

HPLC conditions of Skrede et al. (2000) using a Dionex HPLC system and a PDA 100

diode array detector (Dionex Co., Sunnyvale, CA). Compounds were separated on a 250

X 4.6 mm Supelcosil LC-18 column (Supelco, Bellefonte, PA) and quantified using

standards of their respective 3-glucoside forms (Polyphenols Laboratories AS, Sandnes,

Norway). Total ascorbic acid (the sum of L- and dehydro- ascorbic acid) was quantified

by reverse phase HPLC using modified chromatographic conditions described by









Goikmen et al. (2000). Separation was performed on 3.9 x 150 mm Nova-Pak Cls column

(Waters, Milford, MA), using KH2PO4 (0.2M, pH 2.4) as the mobile phase at a flow rate

of 0.5 mL/min with UV detection at 254 nm. Prior to ascorbic acid analysis, all samples

were passed thorough pre-conditioned Waters Cls Sep-Pak cartridges (Waters, Milford,

MA) to remove neutral polyphenolics. After discarding the first mL, samples were

collected, and dithiothreitol (8 mM) subsequently added as a pre-column reductant.

Samples were then stored in the dark for 120 min to convert dehydroascorbic acid to L-

ascorbic acid. After complete conversion, samples were filtered through a 0.45Clm PTFE

filter (Millipore, Bedford, MA) and analyzed for total ascorbic acid. Antioxidant capacity

was determined using the oxygen radical absorbance capacity (ORAC) assay against a

standard of Trolox as described by Talcott et al. (2003b).

Statistical Analysis

Data represents the mean and standard error of juices analyzed as a 3 x 3 x 2

factorial comparing three processing conditions (unprocessed, 400 MPa, and 550 MPa),

three copigmentation treatments (none, rosemary or thyme extracts), and the presence or

absence of ascorbic acid (0 or 450 mg/L). Phytonutrient and antioxidant losses were also

monitored after 21 days of storage following HHP processing. Linear regression, Pearson

correlations and analysis of variance were conducted using JMP software (SAS, Cary,

NC), with mean separation performed using the LSD test (P<0.05). All experiments were

randomized and conducted in triplicate.

Results and Discussion

Muscadine grape juice was previously established to demonstrate appreciable

phytochemical losses following HHP processing apparently due to the activity of residual

oxidases, presumably PPO, and/or other autoxidative reactions (Talcott et al., 2003a). In









that study the addition of anthocyanin cofactors from a commercial rosemary extract was

proposed as an approach to reduce phytonutrient degradation, yet negatively impacted

juice quality characteristics. In the present study, water-soluble polyphenolic extracts

from rosemary and thyme (partially purified by reverse phase chromatography) were

evaluated as a means to stabilize the color and phytonutrient content of ascorbic acid

fortified grape juice. Utilization of the proposed cofactors and extraction/purification

regime was selected as a strategy to favorably enhance the process and storage stability of

anthocyanin-containing fruit juices with residual PPO activity.

Initial Effects of Copigmentation in Muscadine Grape Juice

Copigmentation increased visual color of the juice as evidenced in a decline in hue

angle (data not shown), which appear to the eye as a more intense red color of the grape

juice. A preliminary study indicated that the color intensity of muscadine juice (measured

as hyperchromic shift) could be increased up to 377 and 490% by the addition of thyme

and rosemary extracts respectively at a 400 copigmentation ratio. However, this level was

deemed impractical for commercial use due to potentially adverse flavor characteristics

and increased oxidative susceptibility. Consequently, a 1:100 ratio was selected for each

copigment source in the processing studies as a means to increase phytochemical

stability. At this ratio both treatments presented similar hyperchromic shifts, however

thyme extracts presented a significantly higher bathochromic shift in absorbance (25 nm)

and also resulted in better anthocyanin stability after HHP processing and ascorbic acid

fortification. Copigmentation also served to mask detrimental color changes that occurred

during HHP processing, as only slight changes were subj ectively observed for

copigmented juices when compared to appreciable losses in control juices. Cofactor

addition also increased initial antioxidant capacity of the juices by an average of 33 CIM









Trolox equiv/mL (Figure 4-1), independently of the cofactor polyphenolic source and

ascorbic acid content.

PPO Activity as Affected by HHP Processing

Residual PPO and/or autoxidative reactions following HHP of muscadine grape

juice have been proposed as potential mechanisms by which decreases in anthocyanin,

ascorbic acid, and antioxidant capacity took place (Talcott et al., 2003a). Additionally,

increased oxidation may also occur under conditions of decreased volume such as

pressurization, according to the Le Chatelier principle (Butz and Tauscher, 2002). The

action of oxidase enzymes in contributing to quality and anthocyanin deterioration has

been demonstrated for several fruit systems (Wesche-Ebeling and Montgomery, 1990;

Laminkara, 1995; Kader et al., 1997; Kader et al., 1999), and for muscadine grape PPO

has specifically been shown to be a significant factor influencing phytochemical

degradation (Kader et al., 1999). These findings justify additional studies evaluating the

influence of HHP processing conditions on PPO activity, as well as their effects on

phytochemical stability following pressurization and throughout storage.

Response Surface Methodology (RSM) was used to determine the pre-processing

PPO activity under different hot-pressed times (0-24 min) and temperatures (46-74 OC)

for juice extraction. Therefore, under a known set of extraction conditions the residual

PPO activity could be estimated, and subsequently monitored following HHP processing.

In the present study, muscadine grape juice was extracted at 460C for 11 min (PPO

activity of 115%; Figure 4-2) and used for subsequent experiments with the purpose of

evaluating the copigmentation treatments with initial PPO enzyme activity and during

HHP induced activation.



















c~ I I Rosemary
80- a mm Thym 80 -1 a a



-1 0. b 60 b b b b


40 -40-








Unprocessed 400 MPa 550 MPa Unprocessed 400 MPa 550 MPa





Figure 4-1. Antioxidant capacity of muscadine grape juice as affected by HHP processing and copigmentation with rosemary or thyme
cofactors in the absence (A) or presence (B) of ascorbic acid (450 mg/L). Bars with different letters for each processing
treatment are significantly different (LSD test, P<0.05).






















100 A \B


I ~V~Y i OE60
S60
450
<( 0 *C40
8,, 20 30

020 ~ 20
2070 15
15 65


50 6 F ) 5 05 5 0 0 5





Figure 4-2. Polyphenoloxidase activity (A), and browning index (B) of muscadine grape juice as influenced by preheating time (0-25
min) and temperature (46-74oC) prior to juice extraction.















1 I d
2001 2001 9


li I


Unprocessed 400 MPa 550 MPa Unprocessed 400 MPa 550 MPa

Figure 4-3. Polyphenoloxidase activity in muscadine grape juice as affected by HHP processing and copigmentation with rosemary or
thyme cofactors in the absence (A) or presence (B) of ascorbic acid (450 mg/L). Bars with different letters are significantly
different (LSD test, P<0.05).









For the control treatment, PPO activity was significantly increased following

processing at 400 (3-fold) and 550 MPa (2.5-fold) as compared to the initial juice activity

(Figure 4-3). Results were similar to those previously observed for PPO using both model

and actual food systems (Poei-Langston and Wrolstad, 1981; Gomez et al., 1996; Sun et

al., 2002), where pressure-induced enzyme activation took place during processing.

Possible explanations for enzyme activation have been attributed to the effect of HHP on

the hydrophobic and electrostatic bonds of proteins, which affects their secondary,

tertiary, and quaternary structures. Such conformational changes can cause enzyme

activation by uncovering active sites and consequently facilitating the interaction with

their substrates. Copigmentation aided to decrease PPO pressure-induced activation by

>1.5-fold, with cofactors from thyme generally being more effective than those from

rosemary. A small increase in PPO activity was observed in the presence of ascorbic acid

for both the control (~10%) and copigmented juices (~16%), an effect that was

independent of the processing pressure (Figure 4-3B). These increases may not have

occurred due to actual PPO activation, but potentially occurred due to increased oxygen

consumption caused by ascorbic acid oxidation during the enzyme assay conditions.

Phytochemical Stability Following HHP Processing

Two pressures (400 and 550 MPa) were selected for the HHP processing of

muscadine grape juice, and its phytochemical content was compared to an unprocessed

control following copigment and/or ascorbic acid addition (Table 4-1, Figures 4-4 and 4-

5). In general, treatments processed at 400 MPa had greater phytonutrient losses due to

the highly oxidative conditions that resulted from PPO activation during pressurization.

Higher anthocyanin, ascorbic acid and antioxidant capacity retention was observed for










juices containing thyme cofactors, followed by rosemary cofactors, and lastly the control

juices.

Table 4-1. The effect of rosemary and thyme cofactors at different anthocyanin-to-
cofactor molar ratios (1:0, 1:100), and ascorbic acid fortification (0, 450
mg/L) on the anthocyanin content of unprocessed (control) and high
hydrostatic pressure processed (400, and 550 MPa) muscadine grape juice.

No Ascorbic Added Ascorbic (450 mg/L)
Molar HHP HHP HHP HHP
Copigment Ratio' Unprocessed 40 a50Ma Unprocessed40Ma5Ma

Dephniin Control 0 569 c2 163 c 303 a 605 c 196 b*3 363 b*
3,5-diglucoside Rosemary 100 707 b 728 b 701 a 1012 a 874 a* 841 a*
(mg/L) Thyme 100 1187 a 801 a 794 a 858 b* 905 a* 896 a*

Cyniin Control 0 227 a 68 c 127 c 253 a 82 c* 152 c
3,5-diglucoside Rosemary 100 226 a 227 b 429 a 157 b* 273 b 516 a
(mg/L) Thyme 100 168 a 400 a 367 b 311 a* 451 a 501 b*

Petunidin Control 0 542 c 163 b 290 c 580 b 195 b* 348 c
3,5-diglucoside Rosemary 100 726 b 646 a 626 b 899 a* 777 a 631 b*
(mg/L) Thyme 100 885 a 670 a 734 a 778 b* 755 a 828 a*

Peniin Control 0 392 b 115 b 210 b 414 b 138 b* 251 b
3,5-diglucoside Rosemary 100 524 a 481 a 531 a 579 a 577 a* 637 a*
(mg/L) Thyme 100 649 a 467 a 569 a 547 a 526 a 642 a

Maviin Control 0 397 b 123 b 223 c 444 a 148 b* 267 c
3,5-diglucoside Rosemary 100 536 a 487 a 579 b 547 a 583 a 694 b*
(mg/L) Thyme 100 424 b 498 a 744 a 481 a 561 a 839 a*

1 Indicates the ratio between the molar concentration of total anthocyanins in muscadine
grape juice (expressed as cyanidin 3-glucoside equivalents), and the molar concentration
of each added polyphenolic cofactor (expressed in gallic acid equivalents). 2 Values with
similar letters within columns of each added cofactor are not significantly different (LSD
test, P>0.05), and indicate the effect of an increase in the molar concentration of each
cofactor. 3Means with an asterisk (*) indicate a significant effect (LSD test, P<0.05) due
to addition of ascorbic acid when compared to the same treatment without ascorbic acid.

















S4000 -1 n luo 111U o M uIlU; n~lu I IRosemary 4000 (lI} ASCOTIDIC AGIG
m Thyme a ab

ab ~cde bdcdef bcdab
S3000 -e ,, I 3000-
8 gh
a hi

S 2000 -1 II III 2000


1000 -1000-

1 II I

Unprocessed 400 MPa 550 MPa Unprocessed 400 MPa 550 MPa




Figure 4-4. Total anthocyanin content of muscadine grape juice as affected by HHP processing and copigmentation with rosemary or
thyme cofactors in the absence (A) or presence (B) of ascorbic acid (450 mg/L). Bars with different letters for each
processing treatment are significantly different (LSD test, P<0.05).




































Unprocessed


Figure 4-5. Total ascorbic acid content of muscadine grape juice as affected by HHP processing (400, and 550 MPa), and
copigmentation with rosemary or thyme polyphenolic cofactors. Bars with different letters for each processing treatment
are significantly different (LSD test, P<0.05).


400 MPa


550 MPa









Anthocyanin degradation was observed at both processing pressures but was

appreciably higher at 400 MPa (70% loss) compared to 550 MPa (46% loss), and these

decreases correlated to losses of antioxidant capacity (r=0.89). Copigmentation was

instrumental for improving anthocyanin retention, maintaining on average 2,200 and

1,500 mg/L more total anthocyanins for treatments copigmented with thyme and

rosemary, respectively (Figure 4-4). Individual anthocyanins were also quantified (Table

1). Differences in their structures, mainly in the B-ring, influenced their reaction rates and

consequently individual anthocyanin losses. Previous studies have demonstrated that o-

diphenolic anthocyanins are more susceptible to degradation than the non o-diphenolic.

This due to the presence of hydroxyl group substitutions in the B-ring of the o-diphenolic

anthocyanins, which are more susceptible to enzymatic degradation reactions than the

methoxy groups of non o-diphenolic anthocyanins (Sarni-Machado et al., 1997). The

degradation trends observed in this study are in accordance with previous studies with

muscadine juice (Talcott et al., 2003a), where the 3,5-diglucosides of delphinidin and

petunidin showed the greatest rates of degradation followed by those of cyanidin, and

finally peonidin and malvidin. These results support the idea that intermolecular

copigmentation has a greater protective effect on non o-diphenolic anthocyanins rather

than the o-diphenolic, which has been attributed to the presence of methoxy groups in the

B-ring which facilitate copigmentation (Mazza and Brouillard, 1990; Jackman and Smith,

1996).

Due to the mutual destruction of ascorbic acid and anthocyanins when present

together in foods (Poei-Langston and Wrolstad, 1981; Garcia-Viguera and Bridle, 1999;

Garzon and Wrolstad, 2002; Talcott et al., 2003a), it was anticipated that higher









anthocyanin destruction might occur in the presence of ascorbic acid. However, a

combined anthocyanin protective effect due to copigment and ascorbic acid addition was

observed prior and following HHP processing (Figure 4). Isolating the effect of ascorbic

acid, an early oxidative protection was observed in the unprocessed rosemary-

copigmented grape juices, as indicated by higher anthocyanin retention (474 mg/L, when

compared to the same treatment without ascorbic; Figures 4-4A and B). Following HHP

processing at both pressures, anthocyanin degradation was decreased by 20% in

comparison with treatments without ascorbic for both the control and rosemary-

copigmented juices (Figure 4B). This protective effect was less pronounced for juices

containing thyme cofactors as anthocyanin degradation was only decreased by 13%.

Kader et al. (1997, 1999) also observed that ascorbic acid could offer initial protection

against anthocyanin oxidation in the presence of oxidative enzymes by reducing o-

quinones to their original phenolic moiety and preventing secondary reactions affecting

phytochemical stability and quality deterioration.

Copigmentation with polyphenolic cofactors was also effective in preventing initial

ascorbic acid oxidation (Figure 4-5) and helped to increase the initial antioxidant capacity

(Figure 4-1) of the juices, an effect that was observed prior and following HHP.

However, ascorbic acid retention was appreciably influenced by the conditions of HHP

processing with losses of 84% at 400 MPa compared to 18% at 550 MPa for control

juices. On average, the polyphenolic copigments reduced ascorbic acid degradation after

processing by 32% at 400 MPa and by 20% at 550 MPa, with thyme cofactors again

conferring the greatest protection.









As previously mentioned copigment addition increased antioxidant capacity of

unprocessed treatments by an average of 43% (~33 CIM Trolox equiv/mL), when

compared to control juices (Figure 4-1). After pressurization, juices containing

copigments and processed at 400 MPa and 550 MPa presented similar antioxidant

capacity losses (19 C1M Trolox equiv/mL) and were not impacted by ascorbic acid

addition. Control treatments presented losses of 45% and 21% at 400 and 550 MPa,

respectively, values that were decreased to 26% and 15% in the juices containing

ascorbic acid. Observed losses in antioxidant capacity were greater for treatments that

presented higher rates of PPO activation, likely indicating that phenolic compounds are

being consumed as enzyme substrates or being destroyed by oxidation and thus lowering

the ORAC values.

Phytochemical Retention During Storage

Most studies looking at the effects of HHP processing on phytonutrient stability

only include evaluations after processing (Corwin and Shellhammer, 2002; Park et al.,

2002; Boff et al., 2003), and do not consider their stability during the shelf-life of the

product. The present study did not include a complete shelf-life evaluation due to the

large number of treatments evaluated, however it included a one point evaluation at 21

days after processing selected based on previous studies reporting the shelf-life stability

of ascorbic acid-anthocyanin systems (Garcia-Viguera and Bridle, 1999; Boulton, 2001).

Ascorbic acid was not detected in the juices after 21 days of storage independently of

processing pressure regimes and presence of polyphenolic cofactors. Anthocyanin

content also decreased by 28-34% throughout storage for all treatments and

pressurization conditions (Table 4-2). Oxidation of individual anthocyanins followed









similar rates during storage, despite their different B-ring substitutions, which was in

agreement with Garzon and Wrolstad (2002). Other studies (Sarni-Machado et al., 1997;

Kader et al., 2000; Boulton, 2001) have demonstrated that anthocyanin degradation can

be influenced by structural differences, but the rapid rate of anthocyanin degradation that

occurred due to the highly oxidative conditions created by HHP processing likely

contributed to these observations. Antioxidant capacity was likewise appreciably

decreased following storage (Table 4-2), losses were more pronounced for control

treatments (28%) than copigmented juices (~13%). However, the rates antioxidant

capacity degradation varied insignificantly with pressure processing and ascorbic acid

fortification.

The protective effect exerted by anthocyanin-copigment complexes following

pressurization was appreciably reduced during storage, observations that can be attributed

to the increased rates and complexity of degradative reactions occurring simultaneously

to anthocyanins, ascorbic acid, and cofactor polyphenolics. Although rate constants can

not be calculated on a single storage point, copigmentation did not appear to slow down

the fast rates of anthocyanin degradation during storage. However it aided to retain

greater anthocyanin content when compared to control treatments as higher

concentrations were observed for copigmented treatments initially and immediately after

HHP processing. Ascorbic acid oxidation may have also been a contributing factor

promoting phytochemical degradation due to hydrogen peroxide formation, leading to

additional oxidative and polymeric degradative reactions. Furthermore, peroxide

formation could contribute to activation of residual peroxidase that may further degrade

phytochemicals (Garzon and Wrolstad, 2002). Additionally, by-products from the










degradation of ascorbic acid and/or monosaccharides, such as aldehydes, may contribute

to anthocyanin degradation during storage. Interactions between anthocyanins and

furfural derivatives have been previously investigated (Kader et al., 2000; Boulton, 2001;

Es-Safi et al., 2002; Es-Safl et al., 1999) and were hypothesized to participate in

condensation reactions yielding brown, polymerized pigments.

Conclusions

Blending a commercially available rosemary extract with muscadine grape juice

was previously reported to be deleterious to juice quality in the presence of ascorbic acid,

yet by the use of the purification protocol used in this study an inverse effect was

observed. The phytochemical extraction and isolation procedures utilized in this study

resulted in concentrates that were lower in enzyme substrates and remained efficient as

anthocyanin cofactors. Commercially available botanical extracts may contain a variety

of PPO substrates following harsh extraction procedures, solubilizing compounds such as

cinnamic acid derivatives or tannins that may accentuate enzymatic oxidation within the

food system. Results of this study demonstrated that using an aqueous extract of

rosemary and thyme followed by partial purification with Cls columns could decrease the

destruction of both anthocyanins and ascorbic acid during HHP processing, which created

a highly oxidative environment due to residual PPO. Copigmentation was found to

effectively stabilize anthocyanins and provided additional understanding of the

mechanisms involved in phytochemical losses during pressurization and storage of fruit

juices processed by HHP. Addition of polyphenolic cofactors also increases visual color

and antioxidant capacity, important factors affecting consumer acceptability and potential

health benefits of grape juice consumption.










Table 4-2. The effect of rosemary and thyme cofactors at different molar ratios (1:0,
1:100), and ascorbic acid fortification (0 and 450 mg/L) on the anthocyanin
content and antioxidant capacity of high hydrostatic pressure processed (400
and 550 MPa) muscadine grape juice after 21 days of storage at 240C.
No Ascorbic Added Ascorbic
(450 mg/L)
Molar HHP HHP HHP HHP
Copigment Ratio' 400MPa 550MPa 400MPa 550MPa

Dephniin Control 0 97.8 b2 207 c 134 b 248 b

3,5-diglucoside Rosemary 100 497 a 458 b 610 a 574 a
(mg/L) Thyme 100 491 a 496 a 591 a 612 a

Cyniin Control 0 42.9 c 86 c 54.8 c 106 c

3,5-diglucoside Rosemary 100 155.3 b 293 a 182 b 352 a
(mg/L) Thyme 100 251.2 a 186 b 302 a 210 b
Petunidin Control 0 107 b 193 c 127 b 243 c
3,5-diglucoside Rosemary 100 441 a 368 b 496 a 441 b
(mg/L) Thyme 100 437 a 459 a 492 a 578 a

Peondin Control 0 73.4 b 144 b 90.1 b 172 b

3,5-diglucoside Rosemary 100 283 a 347 a 361 a 435 a
(mg/L) Thyme 100 281 a 380 a 329 a 439 a

Malvdin Control 0 78.8 b 159 c 96.5 b 187 c

3,5-diglucoside Rosemary 100 304 a 386 b 398 a 474 b
(mg/L) Thyme 100 313 a 519 a 391 a 573 a

TotalControl 0 399 b 790 c 503 b 956 c

Anthocyanins Rosemary 100 1679 a 1851 b 2047 a 2275 b
(mg/L) Thyme 100 1773 a 2040 a 2105 a 2411 a

Anioidnt Control 0 16.5 b 25.0 b 23.4 b 32.8 b

Capacity Rosemary 100 51.3 a 48.8 a 50.6 a 50.7 a
(CrM TE/ mL)3 Thyme 100 51.2 a 49.3 a 49.5 a 51.3 a
1 Indicates the ratio between the molar concentration of total anthocyanins in muscadine
grape juice (expressed as cyanidin 3-glucoside equivalents), and the molar concentration
of each added polyphenolic cofactor (expressed in gallic acid equivalents). 2 Values with
similar letters within columns of each added cofactor are not significantly different (LSD
test, P>0.05), and indicate the effect of an increase in the molar concentration of each
cofactor. 3 Expressed as Trolox equivalents per mL of muscadine grape juice.















CHAPTER 5
PASTEURIZATION AND QUALITY RETENTION OF DENSE PHASE-CO2
PROCESSED MUSCADINE GRAPE JUICE

Introduction

Dense phase-CO2 prOcessing (DP-CO2) is a continuous, non-thermal processing

system for liquid foods that utilizes pressure (< 90 MPa) in combination with carbon

dioxide (CO2) to destroy microorganisms as a means of food preservation. Numerous

studies have investigated the efficacy of pressurized CO2 to inactivate microorganisms

and enzymes in batch or semi-continuous systems (Balaban et al., 1991; Lin and Lin,

1993; Isenschmid et al., 1995; Ballestra et al., 1996; Wouters et al., 1998; Butz and

Tauscher, 2002; Corwin and Shellhammer, 2002; Park et al., 2003). However,

information relating to deleterious changes in color and phytochemicals during

processing and storage are generally lacking, especially for continuous processing

systems. Therefore to prove the effectiveness of DP-CO2 prOcessing as a novel food

processing technology, the microbial destruction, phytochemical stability, and sensory

attributes of DP-CO2 prOcessed muscadine grape juice was compared to a thermally

pasteurized juice (75 oC, 15 sec). Treatments were additionally evaluated following

storage for 10 weeks at 4 oC. A central composite design was initially conducted to

determine the DP-CO2 prOcessing parameters which achieved > 5 log reduction of

aerobic microorganisms and yeast/molds. Results of this study demonstrated differences

between microbial, phytochemical, and sensory attributes of DP-CO2 and thermal










processing, parameters that are of significant importance to assess the benefits offered by

novel processing technologies.

Material and Methods

Materials

Muscadine grapes (cv. Noble) were obtained from a local grower in central Florida

and held frozen (-20oC) until needed. Fruit was rapidly thawed by placing them under

running tap water and hand-sorted for uniformity of ripeness. Grapes were then crushed,

heated to 75 OC in an open steam kettle, and held for 2 min prior to juice extraction in a

hydraulic basket press (Prospero's Equipment, Cort, NY). Preliminary investigations

demonstrated that this juice extraction method was sufficient to inactivate oxidase

enzymes. The juice was immediately filtered through cheesecloth followed by vacuum

filtration through a 1cm bed of diatomaceous earth.

Processing Equipment

The DP-CO2 system was constructed by APV (Chicago, IL) for Praxair (Chicago,

IL) and provided as a gift to the University of Florida (Gainesville, FL). The equipment is

capable of continuously treating liquid foods with CO2 at pressures up to 69 MPa. The

system mixes cooled, pressurized liquid CO2 with a juice feed pressurized by its own

pump (Figure 5-1). The mixture is then pressurized by a reciprocating intensifier pump

and subsequently fed to a holding tube (79.2 m, 0.635 cm ID) for the specified residence

time, which is modified by changing the flow rate of the mixture. An external heater and

insulation electronically controls the temperature of the system and upon exiting the

holding tube the juice is depressurized by passing through a backpressure valve and was

finally collected into a holding tank.










I LILPump CO2

1 Pressurization
Chilleri I chamber




oHeating
Expasio




streamn


Processed
juice
Figure 5-1. Schematic diagram of the DP-CO2 prOcessing equipment.

For thermal processing, juice was pumped by a peristaltic pump (Cole Parmer,

Chicago, 1L) through a stainless steel tube into a temperature controlled water bath (Hart

Scientific, American Fork, UT) were it was held at 75 OC for 15 sec (HTST). The juice

was then passed through a cooling tube and chilled to 10 oC whereby it was collected into

sterile glass containers.

Microbial Inactivation Study

Preliminary investigations were conducted to determine the DP-CO2 parameters

that could achieve >5 log reduction of aerobic microorganisms and yeasts/molds using

Response Surface Methodology. Microbial counts (yeast and molds, and total aerobic

microorganisms) were used as the dependent variables in the experimental design that

was conducted in duplicate. Each study required 11 experiments with 4 factorial points, 4

star points and 3 center points for replication. A high initial microbial load in the juice









(8.1 x 106 CFU/mL of yeasts/molds, 1 x 105 CFU/mL of total aerobic microorganisms)

was required and obtained by incubating the filtered juice for 4 days at 21 OC. Juice was

then subj ected to DP-CO2 using different pressures (1.2 to 40.2 MPa) and CO2 levels (0

to 15.7%) using a constant residence time (6.25 min) and temperature (30 oC). Microbial

inactivation was evaluated immediately after processing.

Microbial counts were made from triplicate samples of each processing treatment

serially diluted (1 x 10-1 to 1 x 10-6) in duplicate by mixing 1 mL of each juice with 9 mL

of sterile Butterfield's buffer. Total plate counts were determined on aerobic count plates

and yeast/mold plates (3M Petrifilm Microbiology Products, St. Paul, MN) by plating 0.1

mL of the dilutions onto the agar in triplicate and enumerated after 48 hr at 3 5 oC and 72

hr at 24 oC, respectively, according to the manufacturers guidelines. Experimental data

were analyzed by regression analysis using JMP software (SAS, Cary, NC), fit to

quadratic polynomial equations, and results used to select two DP-CO2 COnditions for

assessment of phytochemical stability and sensory evaluation: (i) D-1 (34.5 MPa, 8%

CO2) and (ii) D-2 (34.5 MPa, 16% CO2).

Scanning electron microscopy was used to investigate changes in yeast

microstructure due to DP-CO2. YeaSt cells present in the grape juice before and after

processing were treated according to the conditions described by Park et al. (2003) before

being observed in the scanning electron microscope (Hitachi S-4000, Pleasanton, CA).

Phytochemical and Microbial Stability Study

Muscadine grape juice was divided into three equal portions for subsequent

processing by the two DP-CO2 COnditions (34.5 MPa at 8 or 16% CO2) and thermal

pasteurization at 75 oC for 15 sec. After processing, each juice was again divided into 3

proportions for assessment of microbial, phytochemical and sensory characteristics.









Samples for microbial and phytochemical analysis were immediately transferred into 20

mL screwed cap vials and stored at 4 oC for 10 weeks, whereas samples for sensory

analysis were transferred to sterile 4 L glass containers. Sodium azide (50 mg/L) was

added to the samples used for phytochemical analysis in order to retard microbial growth.

Physicochemical and Microbial Analyses

Individual anthocyanin 3,5-diglycosides were quantified by reverse phase HPLC

using modified chromatographic conditions described in chapter 4. Compounds were

separated on a 250 X 4.6 mm Supelcosil LC-18 column (Supelco, Bellefonte, PA) and

quantified using standards of their respective 3-glucoside forms (Polyphenols

Laboratories AS, Sandnes, Norway). Mobile phases consisted of 100% acetonitrile

(Phase A) and water containing 10% acetic acid, 5% acetonitrile, 1% phosphoric acid

(Phase B). A gradient solvent program ran phase B from 100 to 88% in 8 min; 88-50% in

2 min, and held for 12 min at a flow rate of 1.8 mL/min. Anthocyanins were

characterized based on UV-VIS spectral interpretation from 200-600 nm, comparison to

authentic standards (Polyphenols Laboratories AS, Sandnes, Norway), and identification

additionally confirmed following acid hydrolysis into their respective aglycones with 2N

HCI in 50% v/v methanol for 60 min at 90 OC.

Antioxidant capacity was determined using the oxygen radical absorbance capacity

(ORAC) assay and quantified using Trolox equivalents (TE) as described in chapter 3.

Total soluble phenolic levels were measured using the Folin-Ciocalteu assay (Talcott et

al., 2003a), and quantified as gallic acid equivalents. pH was measured using a Thermo

Orion Model 720 pH meter (Thermo Electron Corp., New Haven, CT). Total titratable

acidity was determined by potentiometric titration against 0. 1N NaOH to pH 8.2 using an

automatic titrator (Fisher Titrimeter II, Pittsburgh, PA) and expressed in tartaric acid










equivalents. CO2 COntent in the juices was determined using a Orion CO2 electrode

(Thermo Electron Corp., New Haven, CT). Microbial counts throughout storage were

determined as previously described.

Sensory Evaluation

Flavor, aroma, and color intensity of fresh and processed juices were compared

using a difference-from-control test. A randomized complete block design was used and

difference from control measurements were recorded on a line scale with anchors at 0 and

10 that represented "no difference" to "extremely different" in sensory attributes.

Panelists compared the sensory attributes of the reference (fresh/unprocessed juice) with

those presented by the hidden reference (fresh juice) and the thermally or DP-CO2

processed juices. A 9-point hedonic scale was also conducted in order to compare the

overall likeability of fresh (hidden reference) and processed juices.

Before sensory analysis, all juices (fresh, DP-CO2 and thermally processed) were

degassed in order to equalize carbonation levels by placing them in a 4 L sterile glass

container on a hot plate with continuous stirring for 4 h at 20 oC. Juices were then served

on a tray at room temperature in randomly numbered plastic cups with the reference

sample placed at the center of the tray. A cup of deionized water and non-salted crackers

were also provided to the panelists between evaluations. All sensory tests were performed

at the University of Florida' s taste panel facility using sixty untrained panelists (3 1

females, 95% in the 18-44 age range).

Statistical Analysis

Data represents the mean and standard error of juices analyzed as a 3 x 9 factorial

comparing three processing conditions (DP-CO2 at 8% or 16% both at 34.5 MPa, or

thermally pasteurized) evaluated at nine sampling points (unprocessed, processed, week










1, 2, 3, 4 6 8 and 10). Linear regression, Pearson correlations and analysis of variance

were conducted using JMP software (SAS, Cary, NC), with mean separation performed

using the LSD test (P<0.05). All experiments were randomized and conducted in

triplicate. Sensory data was recorded and analyzed using Compusense five (Compusense,

Guelph, Ontario, Canada), and analysis of variance was conducted by using the Tukey's

multiple comparisons method at the 5% significance level.

Results and Discussion

Microbial Inactivation Study

The effects of DP-CO2 at various processing pressures (0-40 MPa) and CO2 leVOIS

(0-18%) on the inactivation of yeasts/molds and total aerobic microorganisms can be

observed in Figure 5-1. Results showed that although processing pressure was a

significant factor affecting microbial inactivation, CO2 COntent was the processing

parameter that had the maj or influence in microbial log reduction. This trend was also

observed by Hong et al. (1999) which reported that microbial inactivation by DP-CO2 is

governed essentially by the transfer rate and the penetration of carbon dioxide into cells,

the effectiveness of which can be improved by increasing pressure, decreasing the pH of

the suspension, and increasing the processing temperature. Studies investigating high

hydrostatic pressure (HHP) processing and super critical-CO2 batch systems have

reported that microbial inactivation is also highly dependant on other processing

parameters such as residence time and number of pulse cycles as well as the composition

of the food (Balaban et al., 1991; Lin and Lin, 1993; Isenschmid et al., 1995; Ballestra et

al., 1996; Wouters et al., 1998; Butz and Tauscher, 2002; Park et al., 2003). Results also

demonstrated that under identical processing conditions, yeasts/molds were destroyed at

significantly higher rates than aerobic microorganisms. Moreover, the synergistic effect









between pressure and CO2 WAS only observed for the inactivation of yeast/molds.

Microbial inactivation is highly dependent on the type of microorganisms present in the

food matrix due to distinct microbial cell microstructure and the diffusion of CO2 into the

microbial cell (Ballestra et al., 1996 Wouters et al., 1998; Corwin and Shellhammer,

2002; Park et al., 2003). For instance vegetative cells, including yeasts and molds, are

rather pressure and CO2 Sensitive, whereas bacterial spores are more pressure resistant

and thus need higher pressures for complete inactivation. Park et al. (2003) showed that a

combined treatment of carbonation and HHP at 500 MPa yielded an 8-log reduction

Staphylococcus aureus, Fusarium oxysporum, and F. sporotrichioides while only a 4-log

reduction was obtained for Bacillus subtilis. Overall, microbial reduction is attributed to

the fact that CO2 Solubility increases directly proportional with increments of processing

pressure (Balaban et al., 1991; Park et al., 2003) which consequently affects the diffusion

of CO2 into the microbial cell as well as the explosive decompression that occurs during

DP-CO2 prOcessing. Results of this optimization study were used to determine those DP-

CO2 COnditions that achieved > 5 log reduction of aerobic microorganisms and

yeast/molds that set the processing conditions of 34.5 MPa with 8% CO2 (D-1) and 16%

CO2 (D-2).

Micrographic observations aided in elucidating the mechanism of yeast destruction

and concluded that explosive decompression of the microbial cell along with changes in

cell membrane structure occurred during DP-CO2 (Figure 5-2). Conversely, heat

pasteurized yeast cells still appeared round and pert but with slightly textured surfaces.

Results also indicated that the number of decompressed cells was directly related to

increments in processing CO2 leVOIS. Previous investigations have also demonstrated









shown that microbial destruction by pressurized CO2 Systems was based on gas

dissolution inside a microbial cell that when rapidly decompressed to atmospheric

pressure caused fatal damage to cell functioning and explosive decompression of the cell

(Balaban et al., 1991; Lin and Lin, 1993; Ballestra et al., 1996; Park et al., 2003). Other

theories concerning bacterial death by CO2 preSsurization have indicated that the

depressurization leads to leakage of cellular components and changes in the cell

membrane permeability which is responsible for cell damage and eventual microbial

death (Lin and Lin, 1993; Isenschmid et al., 1995; Park et al., 2003). Related studies have

shown that removal of essential intracellular substances such as phospholipids and

hydrophobic compounds from cells or cell membranes play important roles as

mechanisms of microbial inactivation (Lin and Lin, 1993; Ballestra et al., 1996; Butz and

Tauscher, 2002). Additionally, DP-CO2 effects biological systems by causing protein

denaturation, lipid phase changes, and rupture of membranes inside the microbial cell

(Lin and Lin, 1993; Ballestra et al., 1996; Park et al., 2003).


Phytochemical and Microbial Stability Study

Differences in phytochemical and antioxidant levels were observed in muscadine

grape juice as affected by processing methods and storage. Thermal pasteurization was

found to be more detrimental to anthocyanins, soluble phenolics, and antioxidant capacity

as compared with DP-CO2 and unprocessed juices. Moreover, enhanced oxidative

stability and retention of antioxidant compounds was observed for DP-CO2 prOcessed

juices throughout storage. However, microbial stability was only comparable to heat-

pasteurized juices for the first five weeks of storage.






87





8 AtB





O O

14 20

12 1 01 2 1 8 1 0

44 20 0 00 4 2
Fiue52. Inciaino es/ols(& ;A n oaarbcmcoraim
(TM B) afe PC2pseuiaino ucdiegaejiea
inlene bypoesn rsue(-0M0)adC 2cnet(-57 )


I-m I-
Figure 5-3. Scanning electron micrographs of naturally occurring yeast cells in
muscadine juice before (A) and after DP-CO2 at 34.5 MPa and 16% CO2 (B).










Table 5-1. The effect of heat (75 oC for 15 sec) or DP- CO2 (D-1: 34.5 MPa, 8% CO2; D-
2: 34.5 MPa, 16% CO2) pasteurization on the total anthocyanin, soluble
phenolic, and antioxidant content of unprocessed muscadine grape juice.

Total Soluble Antioxidant
Treatment anthocyanins phenolics capacity
(mg/L) (mg/L) (CLM TE/ mL)

Unprocessed 1,105 a' 2,211 a 22.1 a

DP-1
(345 Ma,8% O21,077 a 2,213 a 20.7 a

DP-2
(345 Ma, 6% O21,102 a 2,157 b 21.7 a

HTST
(75oC,15 ec)866 b 1,859 c 18.2 b

SValues with similar letters within columns are not significantly different (LSD test,
P>0.05).

Thermal pasteurization decreased total anthocyanins by 16%, total soluble

phenolics by 26%, and antioxidant capacity by 10% whereas no significant changes were

observed for either DP-CO2 prOcesses (Table 5-1). Individually quantified anthocyanins

followed a similar trend with greater losses occurring for o-dihydroxy substituted

anthocyanins (delphinidin and cyanidin) with respect to the methoxylated anthocyanins

(peonidin and malvidin) as previously observed in muscadine grape juice (Talcott et al.,

2003). Losses ranged from 8-16% following thermal pasteurization for delphinidin,

cyanidin and petunidin, while peonidin and malvidin remained stable (< 4% losses).

Anthocyanin degradation during processing and storage was highly correlated to total

soluble phenolics (r-0.86) and antioxidant capacity (r=0.82). Insignificant changes in

juice pH (3.2) or titratable acidity (0.56 meq tartaric acid/mL) were observed between




Full Text

PAGE 1

EMERGING TECHNOLOGIES AND STRATEGIES TO ENHANCE ANTHOCYANIN STABILITY By DAVID DEL POZO-INSFRAN 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 2006

PAGE 2

Copyright 2006 by David Del Pozo-Insfran

PAGE 3

iii ACKNOWLEDGMENTS My deepest recognition and gratitude goes to my role models in life: my beloved parents (Dr. Myriam D. Insfran and Dr. David Del Pozo) a nd my dearest sister (Piu). They have helped me in any imaginable way to achieve everything I have in life and to fulfill all my dreams. They have been an inexhaustible source of love, insp iration, and encouragement all my life. They were there to catch me when I fell, support me when I needed it, and cheer me up all the step of the way. Just hearing their words-“How can help you ?”made my day. Words can not express all the gratitude and love for my family. The present work could have not been possible without them. I would like to thank Elisa Del Pozo R., Teresa Rendon, and Robin K. Minor for their unconditional support and love thro ughout my life. These wonderful persons have a very special place in my heart. I would also like to acknowledge the uncondi tional support of my supervisory committee chair, Dr. Stephen Talcott, and my mentors, Dr Carmen Hernandez Brenes and Dr. Ronald H. Schmidt. Through their guidance, wisdom, and never-ending care they have helped me to achieve all my goals and accomplishments in my pr ofessional career. I sincerely appreciate the help offered by the members of my supervisi ng committee: Dr. Murat O. Balaban, Dr. Bala Rathinasabapathi, Dr. Susan S. Pe rcival, and Dr. Jesse Gregory. Finally, I would also like to give special thanks to Flor Nunez, Rena Schonbrun, Minna Schuster, Joon H. Lee, Janna Underhill, Gillia n Folkes, Asli Odabasi, Sibel Damar, Youngmok Kim, Lanier Fender, Chris Duncan, Lisbeth Pa checo, and Angela Lounds, for all their support and love.

PAGE 4

iv TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF TABLES............................................................................................................vii LIST OF FIGURES...........................................................................................................ix ABSTRACT....................................................................................................................... xi CHAPTER 1 INTRODUCTION........................................................................................................1 Justification.................................................................................................................. .1 Objectives..................................................................................................................... 2 2 LITERATURE REVIEW.............................................................................................4 Anthocyanins................................................................................................................5 Anthocyanins and Intermolecular Copigmentation Reactions..............................9 Anthocyanin Stability..........................................................................................11 Phytonutrient Stability and Intermolecular Copigmentation...............................13 Sensory Attributes of Anthocyanin-c ontaining Beverages as Affected by the Addition of Polyphenolic Cofactors................................................................16 Novel Processing Technologies That May Enhance Anthocyanin Stability..............18 High Hydrostatic Pressure (HHP) Processing.....................................................19 HHP and microbial inactivation...................................................................20 The effect of HHP in food components.......................................................22 Dense Phase-CO2 Pasteurization.........................................................................26 DP-CO2 and microbial inactivation.............................................................27 DP-CO2 and enzymatic inactivation............................................................33 3 PHYTOCHEMICAL COMPOSITION AND PIGMENT STABILITY OF AAI ( EUTERPE OLERACEA MART.)..............................................................................36 Introduction.................................................................................................................36 Materials and Methods...............................................................................................38 Materials..............................................................................................................38 Color Stability.....................................................................................................39 Effect of Copigmentation....................................................................................39

PAGE 5

v Phytochemical Analyses......................................................................................40 Statistical Analysis..............................................................................................42 Results and Discussion...............................................................................................42 Anthocyanin and Polyphenolic Characterization................................................42 Antioxidant Capacity...........................................................................................44 Color Stability as Affected by H ydrogen Peroxide and Temperature.................46 Color Stability in the Presence of As corbic Acid and Natural Cofactors...........49 Conclusion..................................................................................................................55 4 STABILITY OF COPIGMENTED ANTHOCYANINS AND ASCORBIC ACID IN MUSCADINE GRAPE JUICE PROC ESSED BY HIGH HYDROSTATIC PRESSURE.................................................................................................................57 Introduction.................................................................................................................57 Materials and Methods...............................................................................................59 Materials and Processing.....................................................................................59 PPO activity during juice extraction............................................................59 Juice extraction and processing....................................................................60 Chemical Analyses..............................................................................................61 Statistical Analysis..............................................................................................62 Results and Discussion...............................................................................................62 Initial Effects of Copigmentati on in Muscadine Grape Juice.............................63 PPO Activity as Affected by HHP Processing....................................................64 Phytochemical Stability Following HHP Processing..........................................68 Phytochemical Retention During Storage...........................................................74 Conclusions.................................................................................................................76 5 PASTEURIZATION AND QUALITY RETENTION OF DENSE PHASE-CO2 PROCESSED MUSCADINE GRAPE JUICE...........................................................78 Introduction.................................................................................................................78 Material and Methods.................................................................................................79 Materials..............................................................................................................79 Processing Equipment.........................................................................................79 Microbial Inactivation Study...............................................................................80 Phytochemical and Micr obial Stability Study.....................................................81 Physicochemical and Microbial Analyses...........................................................82 Sensory Evaluation..............................................................................................83 Statistical Analysis..............................................................................................83 Results and Discussion...............................................................................................84 Microbial Inactivation Study...............................................................................84 Phytochemical and Micr obial Stability Study.....................................................86 Sensory evaluation...............................................................................................92 Conclusions.................................................................................................................92

PAGE 6

vi 6 ENHANCING THE RETENTION OF PHYTOCHEMICALS AND ORGANOLEPTIC ATTRIBUTES IN MUSCADINE GRAPE JUICE BY DENSE PHASE-CO2 PROCESSING AND COPIGMENTATION..........................95 Introduction.................................................................................................................95 Materials and Methods...............................................................................................97 Materials and Processing.....................................................................................97 Physicochemical and Microbial Analyses...........................................................98 Sensory Evaluation..............................................................................................98 Statistical Analysis..............................................................................................99 Results and Discussion.............................................................................................100 Initial Effects of Copigmentation an d Ascorbic Acid Fortification..................100 Phytochemical Changes Due to Thermal and DP-CO2 Processing...................101 Organoleptic Changes Due to Addition of Thyme Polyphenolic Cofactors.....102 Phytochemical and Microbial Change s During Refrigerated Storage..............103 Conclusions...............................................................................................................114 7 INACTIVATION OF POLYPHENOL OXIDASE IN MUSCADINE GRAPE JUICE BY DENSE PHASE-CO2 PROCESSING....................................................116 Introduction...............................................................................................................116 Materials and Methods.............................................................................................117 Materials............................................................................................................117 Effect of DP-CO2 Processing on PPO activity..................................................118 Storage Stability of Muscadine Ju ice with Residual PPO Activity...................118 Chemical Analyses............................................................................................119 Statistical Analysis............................................................................................119 Results and Discussion.............................................................................................119 Effect of DP-CO2 Processing on PPO Activity.................................................119 Storage Stability of Muscadine Ju ice with Residual PPO Activity...................125 Conclusions...............................................................................................................128 8 SUMMARY AND CONCLUSIONS.......................................................................130 LIST OF REFERENCES.................................................................................................131 BIOGRAPHICAL SKETCH...........................................................................................144

PAGE 7

vii LIST OF TABLES Table page 2-1 Effect of rosemary extract (0, 0.1, 0. 2, 0.4% v/v) on color, aroma, and flavor attributes of a commercial strawberry cocktail juice................................................18 3-1 Anthocyanin and polyphenolic c ontent (mg/L fresh pulp) of aai ( Euterpe oleracea Mart.).........................................................................................................45 3-2 The effect of hydrogen peroxide (30 mmol/L) and temperature on kinetic parameters of color degradation fo r different anthocyanin sources.........................48 3-3 Percent monomeric anthocyanins and CIE color attributes of a juice model system (pH 3.5, 100 mg/L sucrose) prepar ed with different pigment sources.........51 3-4 The effect of ascorbic acid and natu rally occurring polyphenolic cofactors on kinetic parameters of anthocyani n degradation during storage at 37 C of in vitro models systems prepared with different pigment sources........................................52 4-1 The effect of rosemary and thyme co factors and ascorbic acid fortification on the anthocyanin content of unprocessed and high hydrostatic pressure processed muscadine grape juice..............................................................................................69 4-2 The effect of rosemary and thyme co factors and ascorbic acid fortification on the anthocyanin content and antioxidant capacity of high hydrostatic pressure processed muscadine grape juice afte r 21 days of storage at 24C..........................77 5-1 The effect of heat or DPCO2 pasteurization on the to tal anthocyanin, soluble phenolic, and antioxidant content of unprocessed muscadine grape juice...............88 6-1 The effect of thyme cofactors and ascorbic acid fortification on the total anthocyanin, soluble phenolic and anti oxidant content of unprocessed, heat, and DP-CO2 pasteurized muscadine grape juice...........................................................104 6-2 Effect of thyme cofactors and asco rbic acid fortification on first-order degradation kinetic parameters of ant hocyanins present in heat or DPCO2 processed muscadine grape juice during storage at 4 C.......................................105 6-3 Effect of thyme cofactors and ascorbic acid on first-order degradation kinetic parameters of soluble phenolics in heat or DPCO2 processed muscadine grape juice during storage at 4 C....................................................................................106

PAGE 8

viii 6-4 Effect of thyme cofactors and asco rbic acid fortification on first-order degradation kinetic parameters of anti oxidant capacity in heat or DPCO2 processed muscadine grape juice during storage at 4 C.......................................107 6-5 Effect of thyme cofactors on first-orde r degradation kinetic parameters of total ascorbic acid present in heat or DPCO2 processed muscadine grape juice during storage at 4 C.............................................................................................108 7-1 Individual and total anthocyanin cont ent of unprocessed muscadine grape juice as affected by DP-CO2 processing pressure and CO2 content ..............................120

PAGE 9

ix LIST OF FIGURES Figure page 2-1 Structure of the six basic anthocya nindins (A), along with their different positional glycosides (3-glycosides, B; 3,5-glycosides, C)........................................7 2-2 Schematic representing a simple a nd an acylated cyanidin glycoside.......................8 2-3 Schematic representation of anthoc yanin self-association and copigmentation reactions.....................................................................................................................9 3-1 HPLC chromatogram of anthocyanin 3glucosides (A) monitored at 520 nm and their aglycones (B) present in aai ( Euterpe oleracea Mart.)..................................43 3-2 HPLC chromatogram of phenolic ac ids and flavonoids present in aai ( Euterpe oleracea Mart.).........................................................................................................43 3-3 Antioxidant capacity of different phytochemical fractions of aai ( Euterpe oleracea Mart.).........................................................................................................47 4-1 Antioxidant capacity of muscadine gr ape juice as affected by HHP processing and copigmentation with rosemary or thyme cofactors in the absence or presence of ascorbic acid (450 mg/L)......................................................................65 4-2 Polyphenoloxidase activity (A), and browning index (B) of muscadine grape juice as influenced by preheating time and temperature prior to juice extraction...66 4-3 Polyphenoloxidase activity in muscadine grape juice as affected by HHP processing and copigmentation in the abse nce (A) or presence (B) of ascorbic acid........................................................................................................................... 67 4-4 Total anthocyanin content of mus cadine grape juice as affected by HHP processing and copigmentation with rosema ry or thyme cofactors in the absence or presence of ascorbic acid.....................................................................................70 4-5 Total ascorbic acid c ontent of muscadine grape juice as affected by HHP processing and copigmentation with rose mary or thyme polyphenolic cofactors...71 5-1 Schematic diagram of the DP-CO2 processing equipment.......................................80

PAGE 10

x 5-2 Inactivation of yeast/molds and to tal aerobic microorganisms after DP-CO2 pasteurization of muscadine juice as infl uenced by processing pressure and CO2 content......................................................................................................................87 5-3 Scanning electron micrographs of natura lly occurring yeast cells in muscadine juice before (A) and after DP-CO2 at 34.5 MPa and 16% CO2 (B).........................87 5-4 Total anthocyanin (A) and antioxida nt content (B) of heat and DP-CO2 pasteurized muscadine juice dur ing refrigerated storage (4 C)..............................93 5-5 Total soluble phenolic c ontent of heat and DP-CO2 pasteurized muscadine juice during refrigerated storage (1-10 weeks at 4 C).....................................................94 5-6 Yeast/mold counts of heat and DP-CO2 pasteurized muscadine juice during refrigerated storage (4 C)........................................................................................94 6-1 Total anthocyanin conten t of muscadine grape juice without and with ascorbic acid during refrigerated storage as aff ected by heat and pasteurization and the addition of thyme cofactors....................................................................................110 6-2 Total soluble phenolic content of mu scadine grape juice without and with ascorbic acid during refrigerated stor age as affected by heat and DP-CO2 pasteurization, and the add ition of thyme cofactors...............................................112 6-3 Antioxidant capacity of muscadine grape juice without and with ascorbic acid during refrigerated storage as affected by heat and DP-CO2 pasteurization, and the addition of th yme cof actors..............................................................................113 6-4 Total ascorbic acid content of muscadin e grape juice during refrigerated storage as affected by heat and DP-CO2 pasteurization, and th e addition of thyme cofactors.................................................................................................................114 7-1 Effect of DP-CO2 at different processing pressures and CO2 levels on residual PPO activity (A) and resultant anthocyanin losses (B) in muscadine grape juice.121 7-2 Effect of DP-CO2 processing pressures and CO2 levels on PPO-induced losses in soluble phenolics (A) and antioxidant capacity (B) in muscadine grape juice..122 7-3 Total anthocyanin, soluble phenolics, and antioxidant capacity content of DPCO2 processed muscadine grape juice duri ng refrigerated storage as affected by processing CO2 content and initial PPO activity....................................................116

PAGE 11

xi 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 EMERGING TECHNOLOGIES AND STRATEGIES TO ENHANCE ANTHOCYANIN STABILITY By David Del Pozo-Insfran May 2006 Chair: Stephen T. Talcott Major Department: Food Science and Human Nutrition Anthocyanins are polyphenolic compounds that are responsible for the bright blue and red colors of many foods and act as phytochemical antioxi dants with potential healthrelated benefits. Recent shifts in consumer preference for natural pigments have focused on applications of anthocyanins as suitable re placements for certified colorants. However their relative high cost an d poor stability are factors that limit their commercial application. Due to these limitations, the food industry is constantly looking for novel, inexpensive and stable sources of these natural colorants. Therefore, this study first determined the phytochemical composition and st ability of aai anthocyanins as a new source of anthocyanins. Of identical need for the food industry is the exploration of strategies and/or technologies that can serve to alleviate th e limitations of anthoc yanins. High hydrostatic pressure processing (HHP) and dense pha se-carbon dioxide pa steurization (DP-CO2) are non-thermal processing technologies that may le ssen detrimental effects to anthocyanins

PAGE 12

xii and other phytochemicals. However, a downsid e of these technologies is the presence and/or activation of enzymes following proces sing that may be responsible for oxidative degradation. This study inve stigated the role of po lyphenol oxidase (PPO) in phytochemical degradation during HHP pro cessing of muscadine grape juice and established a potential remedi ation strategy using polyphenolic cofactors from two plant sources: rosemary and thyme. Cofactor ad dition not only increas ed juice color and antioxidant activity but also reduced anthocyanin, polyphenolic, and ascorbic acid losses. DP-CO2 was also evaluated as a potentia l non-thermal pasteurization technology with results concluding that th is process served to protect anthocyanins and antioxidant levels without comprising juice sensory attr ibutes. However, microbial stability of DPCO2 juices was only comparable to heat-paste urized counterparts for >5 weeks at 4 C. Due to the preceding evidence, the add ition of thyme polyphenolic cofactors along with the DP-CO2 process was evaluated as a combined strategy to decrease phytochemical and antioxidant losses that o ccur in anthocyanin-containing beverages. Results concluded that DP-CO2 and addition of thyme cof actors served to increase phytochemical stability of mus cadine juice without affectin g juice sensory attributes. Cofactor addition also masked the detrimental color fading that occurred during storage. This study also showed that partial inac tivation of PPO can be obtained by DPCO2 and that processing CO2 levels was the main processi ng variable influencing PPO activity as well as polyphenolic and antioxi dant retention in muscadine juice.

PAGE 13

1 CHAPTER 1 INTRODUCTION Justification Anthocyanins are polyphenolic compounds that are responsible for the bright blue to red colors of foods and act as phytochemical antioxida nts imparting important healthrelated benefits and nutraceu tical properties. Due to current trends in consumer preferences for natural pigments, these compounds are considered as potential replacements for certain certified and banne d dyes. However, many limitations exist for their commercial application due to high raw material costs and thei r poor stability during processing and storage. Devel oping strategies and technologie s that serve to alleviate these limitations is thus vital for economic growth of the U.S. food industry, not only improving quality attributes and phytonutri ent stability of an thocyanin-containing products, but also possibly contributing to improve public health through increasing phytonutrient and health promoting ag ents intake in food products. High hydrostatic pressure processing (HHP) and dense phase carbon dioxide pasteurization (DP-CO2) are promising alternatives to traditional thermal pasteurization technologies and may lessen detrimental effect s to anthocyanins and other thermolabile phytonutrients during processi ng and storage (Gomez and Le dward, 1996; Zabetakis et al., 2000; Sun et al., 2002). However, a downsid e of these technologies is the presence and/or activation of residual enzymes fo llowing processing that may be partially responsible for oxidative degr adation. Therefore, associat ed changes in phytonutrient stability due to processing are of interest and will be investigated in the present study.

PAGE 14

2 Furthermore, comparisons between non-therma l and thermal processes are one way to assess the benefits offered by thes e novel processing technologies. Previous studies have shown that form ation of intermolecular copigmentation complexes between anthocyanins and exoge nously added polyphenolic cofactors could assert a protective effect against anthocya nin, antioxidant capac ity, and ascorbic acid degradation in both model and juice systems (Talcott et al., 2003; Brenes et al., 2004; Eiro and Heinonen, 2002; Malien-Aubert et al ., 2001; Del Pozo-Insfran et al., 2004). In addition to preventing quality and nutritiona l losses, copigmentation also increases anthocyanin color intensity and antioxidant content of beverages, and masks the detrimental color changes that take place du ring processing and storage. Consequently, intermolecular copigmentation could be used as an important remediation strategy which attenuates phytonutrient degradation that takes place in anthocyanin containing juice systems during processing and storage. Therefor e, the present study will also evaluate the phytonutrient stability and sens ory properties of fortified juice and beverage systems containing copigmented anthocyanins in an effort to add economic value, reduce oxidation, and maintain nutrient stability. Objectives The objectives of the present research work were: 1. To characterize the major anthocyanin and polyphenolic compounds present in aai and to determine the potential usage of aai as a novel anthocyanin source. 2. To assess muscadine grape juice ( Vitis rotundifolia ) for phytochemical stability following HHP processing and ascorbic acid fortification, and to investigate the effect of exogenously added polyphenolic cofactors purified from rosemary ( Rosmarinus officinalis ) and thyme ( Thymus vulgaris ) as a means to improve overall phytochemical stability. 3. To determine the microbial destructi on, phytochemical stability, and sensory attributes of DP-CO2 processed muscadine grape juice.

PAGE 15

3 4. To determine the phytochemical retention and sensory properties of an ascorbic acid fortified and copigmented mus cadine grape juice following DP-CO2 and subsequent storage. 5. To determine the effect of DP-CO2 on PPO activity and its consequent effect on polyphenolic and antioxidant cha nges in muscadine juice.

PAGE 16

4 CHAPTER 2 LITERATURE REVIEW A steady increase in the development of natural food colorants and functional food sources has been observed in recent years, not only due to consumer preferences for natural pigments but also fo r their health-related benefits and nutraceutical properties (Frankel et al., 1995; Meyer et al., 1997; Skrede et al., 2000). Anthocyanins are a viable replacement for synthetic colorants due to their bright, attracti ve colors and water solubility, which allows their incorporati on into a variety of food systems (RodriguezSaona et al., 1999). However, many limitations exist for the commercial application of anthocyanins due to high raw material cost s and their poor stability. Pigment stability may be affected by chemical structure, concentration, pH, temperature, oxygen, light, presence of cofactors, and polymeric fo rms. Furthermore, beverages containing anthocyanins and ascorbic acid are known to be mutually destructiv e in the presence of oxygen, which limits fortification in fruit juices and in products containing anthocyanins (Frankel et al., 1995; Meyer et al., 1997; Rodriguez-Saona et al., 1999; Skrede et al., 2000). Developing strategies and technologies that serve to al leviate these limitations is thus vital for economic growth of the U. S. food industry, not only improving quality attributes and phytonutrient stab ility of anthocyanin-containi ng products, but as well as contributing to improve public health th rough increasing phytonut rient and health promoting agents intake in food products. Exploration of a system to stabilize an thocyanins from color loss and oxidation, and the effects of ascorbic ac id fortification, are important if these pigments present in

PAGE 17

5 juices and beverages are to compete in a mark et dominated by certified dyes such FD&C Red #40 and Red #3. To alleviate some of thes e problems, the presen t research project suggests a strategy by which the addition of exogenous anthocyanin cofactors (watersoluble thyme and rosemary extracts), to not only attenuate degradation of anthocyanins but also provide a means by which phytonut rient degradation can be prevented. Furthermore, novel technologies such as hi gh hydrostatic pressure and dense phase-CO2 pasteurization are non-thermal processing methods which ensu re microbial destruction and may extend shelf life of products without having adverse effects on their quality attributes such as flavor, color, and phyt onutrient retention. These research studies are important due to the high demand for fruit juice, of which sales reached nearing $4.5 billion in 2001 of which over a quarter were likely to contain anthocyanins from various fruit and vegetable sources. Moreover, there is an interest on the phyt onutrient stability of fortified beverages systems which have been consumed with a substantial increase during the last 24 months By monitoring the chemoprotective and sensory aspects of exogenously added anthocyanin cofactors a nd the effects of novel processing methods, the juice and beverage industr y can more effectively retain both quality and nutritional aspects of their products, increasing overall consistency and adding value to this multibillion dollar industry. Anthocyanins Anthocyanins are best known for their brilliant red and purple colors, and as polyphenolic compounds their antioxidant and an tiradical capacity have been firmly established (Abuja et al., 1998; Frankel et al., 1998; Ghisel li et al., 1998; Heinonen et al., 1998; Lapidot et al., 1999; Esp n et al., 2000). These pigments are considered as potential replacements for certain certified and banne d dyes because of their bright, attractive

PAGE 18

6 colors and water solubility, which allows th eir incorporation into aqueous food systems (Rodriguez-Saona et al., 1999; Boult on, 2001). Moreover, among the various food products currently available, those containing anthocyanins tend to consume the visual sense on retail shelves due to their diverse and bright array of colors exhibited. However, the use of these natural pigments can result in inconsistent color and nutrient degradative reactions during processing and storage, which li mits both usage and fortification efforts. Anthocyanins are the glycoside forms of anthocyanidins (flavonoids) which have a C6-C3-C6 skeleton. Although over 300 different an thocyanins are present in nature only six basic anthocyanin skeleton s exist (Figure 2-1A) and whic h vary in the number and position of hydroxyl and methoxy substituents. These aglycon forms (anthocyanidins) are rarely found in the nature but with some exceptions which include their 3-deoxy forms and which have been reported to be present in red-skinned bananas, sorghum, black tea. Anthocyanins also differ due to the numb er, position and type of glycoside moieties attached to their aglycon moie ties (Figure 2-1B and 1C). Gl ucose, galactose, rhamnose, and arabinose are the most commong glyc oside moieties attached; however, other complex glycosides (rutinoside, sophorosid e, sambubioside) also occur naturally (Clifford, 2000). Anthocyanins moieties can also vary acco rding to the extent and the type of acylating constituents which are attached to their glycoside chains. These acylated anthocyanins are the result of an enzyme-catal yzed transfer of an aliphatic or aromatic organic acid onto a sugar moiety (via an acyl linkage) of an anthoc yanin glycoside. The enzyme system responsible for acylation was identified as 3O -glucoside-6”-Ohydroxycinnamoyltransferase. This enzyme catalyzes the transf er of cinnamic acid

PAGE 19

7 coenzyme A esters to the glycosyl anthoc yanin moieties. The more common acylating constituents include cinnamic acids (p-c oumaric, caffeic, ferulic, sinapic, and chlorogenic) flavonoids, flavan-3-ols, and tanni ns, which may themselves bear glycosidic chains and aliphatic acids ( acetic, malic, maloni, oxalic, a nd succinic). These compounds are typically bound to C-4 of a sugar attach ed to position C-3 of an anthocyanin molecule, but can also be found on other sugar moieties of the anthocyanin (Figure 2-2). The location, number of acy lating compounds, and diversity of these anthocyanins is highly dependent on plant type, but those f ound in fruits and vegetables tend to be structurally simpler compar ed to flowering plants. O O H OH OH R1 OH R2 O O H OH R1 R1 OH R2 O O H OH R1 R1 OH R2 + + OH O-Gly + OH O-Gly O-Gly A C B Anthocyanin R1 R2 Pelargonidin H H Cyanidin OH H Delphinidin OH OH Peonidin O-CH3 H Petunidin O-CH3 OH Malvidin O-CH3 O-CH3 Figure 2-1. Structure of the six basic an thocyanindins (A) showing the different substituents for the six common ant hocyanins found in food systems, along with their different positional glycosides (3-glycosides, B; 3,5-glycosides, C).

PAGE 20

8 Cyanidin-3-D-glucoside Cyanidin-3-(6O p -coumaryl)-D-glucoside Figure 2-2. Schematic representing a simple and an acylated cyanidin glycoside. Acylation significantly improves the stability of anthocyanins through intramolecular copigmentation. The aromatic residues of the acyl groups stack hydrophobically with the pyrilium ring of the fl avylium cation and consequently greatly decrease their susceptibility to the nucleophilic attack of wa ter (Figure 2-3). As the result of the presence of these acylated constituents anthocyanins also exhibit a significant bathochromic shift (1-8 nm increase) from that of the parent anthocyanin. Moreover, acylated anthocyanins have superior color intensity and stability over identical 3glucosides and maintain a desirable pigmenta tion in low acid or ne utral conditions (Asen et al., 1972). As a general rule, the degree of anthocyanin acylation can be estimated by their color intensities at a decreasing acid content. Highly acylate d anthocyanins will retain greater red or blue colors at pH 5, whereas low levels of acylation will be nearly colorless at this same pH. The source of th e anthocyanin pigment significantly affects the color and stability characteristics of these pigments. Red cabbage is often used as a standard by which anthocyanin color stabilit y is compared due to its high degree of +OH OH OH O OH O H O O H O OH OH + OH OH OH OH O OH O H O O H O OH O O

PAGE 21

9 acylation. Color enhancement of this nature is difficult to augment on a commercial scale due to a need to purify these enzymes, but occu rs very efficiently in the vacuole of plant cells and better explains much of the color diversity in plant systems (Asen et al., 1972). Intermolecular Intramolecular Self-association (Sandwich Type) Anthocyanin Co-pigment Acyl group Sugar Figure 2-3. Schematic representation of ant hocyanin self-association and copigmentation reactions. Anthocyanins and Intermolecular Copigmentation Reactions Anthocyanin intermolecular copigmentati on reactions are common in nature and result from association between pigments and cofactors such as polyphenolics and/or metal ions, or other anthocyanins (self-associ ation) (Figure 2-3). Pr eferably formed under acidic conditions, these weak ch emical associations can augment anthocyanin stability and increase antioxidant properties (Mazza and Brouillard, 1990; Boulton, 2001; MalienAubert et al., 2001). Studies have suggested that the c opigmentation phenomenon is the

PAGE 22

10 main anthocyanin stabilizing mechanism in plants (Mazza and Brouillard, 1990; Boulton, 2001). Polyphenolics are the predominant cofa ctors present in an thocyanin-containing fruits and vegetables, and incr eased anthocyanin stab ility has been attributed to their high concentrations in foods. In general, this type of copigmentation was originally interpreted as a weak complex formed between an ant hocyanin and a cofactor agent (Robinson and Robinson, 1931), which is still held as th e most popular and recognized mechanism (Markakis, 1982). The current understanding of copigmentation consists of a “stacking” of a cofactor on the planar pol arized nuclei of an anthocyani n in its flavylium ion form. The hydrophobic complexation reactions between anthocyanins and cofactors may also effectively protect anthocyanins against the nu cleophilic water attack at position 2 of the pyrilium nucleus, thus displa cing the equilibra towards th e flavilium form (colored) rather than that of the less-colored hemi ketal or chalcone forms (Boulton, 2001; MalienAubert et al., 2001; Es-Safi et al, 2002). The effectiveness of this stabilizing effect will depend on the same variables that affect copigmentation. Intermolecular copigmentation also exerts a protective eff ect on anthocyanin degradation as cofactors compete with anthocyanins and preferentially react in se veral condensation reac tions involving a wide variety of carbonyl compounds (Es-Safi et al 1999; Malien-Aubert et al., 2001; Es-Safi et al, 2002). As discussed by Boulton (2001), the color e nhancement and stabilizing effect conferred by copigmentation is different for a given anthocyanin-cofactor pair and depends on the concentration of pigment, the mo lar ratio of cofactor to pigment, pH, the extent of non-aqueous conditions, and presen ce of anions in solution. The increased protection observed for a specific pigment source due to the presence of cofactors is most likely related to the type and content of polyphenolics present, as a higher

PAGE 23

11 copigment/pigment molar ratio could occur for a determined source. Moreover, specific polyphenolics or classifications of polyphenol ics are more likely to form stable intermolecular complexes with anthocyanins than others (Boulton, 2001; Malien-Aubert et al., 2001; Eiro and Heinonen, 2002). C opigmentation reactions are typically concentration dependent (both for anthocyani ns and cofactor) and may be dissociated by heating or by the addition of alcohol, since hydrogen bonding may link the compounds together (Asen et al., 1972; Boulton, 2001). Anthocyanin Stability Anthocyanin color and stabilit y is influenced by chemical structure, concentration, pH, temperature, oxygen, light, polymeric fo rms, ascorbic acid, and the presence of natural or exogenously added cofactors. Ant hocyanins and ascorbic acid have long been shown to be mutually destructive in the presence of oxygen, which causes a decrease in color, functional properties, and nutritional qua lity of a food product (Calvi and Francis, 1978; Poei-Langston and Wrolstad, 1981). Mechan isms for their mutual degradation have included direct condensation between anthocyani ns and ascorbic acid, or the formation of free radicals that induce oxidati ve deterioration of each compound (Garcia-Viguera et al., 1999). Several studies (Calvi and Francis, 1978; Poei-Langston and Wrolstad, 1981; Garcia-Viguera et al., 1999) have implicat ed a complex chemical interaction possibly involving co-oxidative reactions between ascorbic acid and an thocyanins, the effects of which may lead to co-oxidation of other fo rtified nutrients. Mode l systems containing anthocyanins and ascorbic acid have de monstrated destruction under both aerobic and anaerobic conditions, therefore exclusion of oxygen during processing would not be sufficient to prevent nutrient degradation of anthocyanin-containing beverages. Studies involving the degradation of these phytonutrien ts have also been ambiguous since Kaack

PAGE 24

12 and Austed (1998) reported a protective effect on anthocyanins cont aining ascorbic acid or when sparged with nitrogen in elderber ry juice, while ascorbic acid alone was responsible for anthocyanin decreases in Conc ord grape juice (Calvi and Francis, 1978). Color degradation of blood orange juice, wh ich naturally contains cyanidin-3-glucoside and ascorbic acid, was found to correlate with ascorbic acid concentrations and resulted in juice discoloration and loss of fortified ascorbic acid (C hoi et al., 2002). Therefore, degradative reactions of this nature seem to be commod ity specific depending on the phytochemical or anthocyanin composition. Anthocyanin 3,5-diglucosides were reported as less stab le to oxidation and heat compared to corresponding 3-glucosides (Mar kakis, 1982) and may result in rapid color loss during wine or juice storag e. The 3,5-diglucosides report ed to be most unstable in muscadine grape juice were delphinidin and petunidin (Flora, 1978; Goldy et al., 1986) and their oxidation during storage were correl ated to decrease radical scavenging activity (Talcott and Lee, 2001). Both delphinidin and petunidin, along with cyanidin, contain at least one o -dihydroxy group, making them more suscep tible to oxidation than the other anthocyanin forms. Flora (1978) reported large reductions in delphinidin, cyanidin, and petunidin-3,5-diglucosides in muscadine gr apes after severe heat treatments when analyzed by thin-layer chromatography. Malv idin 3,5-diglucoside was found to be less stable than acylated forms of malvidin pres ent in red cabbage (Hr azdina et al., 1970), but in model systems the stability of malvidin 3,5-diglucoside was grea ter than malvidin 3glusoside both with and without added asco rbic acid (Hrazdina et al., 1970; GarciaViguera and Bridle, 1999). The relative stability of a particular source of anthocyanins is

PAGE 25

13 likely a function of a complex chemical matr ix, structural features, and the combined effects of processing and storage. Phytonutrient Stability and In termolecular Copigmentation Preventing anthocyanin and color loss in beverages can be accomplished by strict oxygen control during processing but also by a physical stabilization of anthocyanins through the addition of exogenous ant hocyanin cofactors (intermolecular copigmentation), rather than through antioxida nt addition via ascorbic acid. However, current trends in beverage production dictate th at juice and beverage products be fortified with ascorbic acid, which is intentionally done at the expense of anthocyanin, ascorbic acid, and quality deterioration of the product. Therefore, exploration of a system to stabilize anthocyanins from color loss, the e ffects of ascorbic a fo rtification, and other deleterious reactions under condensation c onditions or a free radical mechanism is important if anthocyanins present in juices and beverages are to compete in a market dominated by certified colors such FD&C Red #40 and Red #3. Copigmentation may be useful in enhancing the value of foods containing anthocya nins; serving as a functional food ingredient with antioxidant properties producing greater visu al color perception, and increasing stability to oxidation and heat. In previous studies (Malie n-Aubert et al., 2001; Eiro and Heinonen, 2002; Talcott et al., 2003; Brenes et al., 2005) the addition of polyphenolic-based anthocyanin cofactors was shown to significantly reduce the mutual destruction of anthoc yanins and ascorbic acid as well as appreciably augment visual color and antioxidant capacity. These studies have shown the success in slowing the kinetics of mutual destruction of anthocyanins and ascorbic acid through cofactor addition, and have led to the proposed objectives herein. The implications of which will have a prof ound impact on both fruit juice fortification

PAGE 26

14 and in other applications in which anthocyani ns are utilized as a natural color source. Additionally, a major advantage of using exoge nously added cofactors is an ability to control their concentration through standa rdization, which would allow for maximal increases in color, thermal stability, nutri ent protection, and orga noleptic properties as influenced by processing and storage. In addition to preventing quality and nutritional losses by copigmentation, the present remediation strate gy increases oxidative and thermal stability, which adds economic value to fruit juices and beverages containing isolated anthocyanins as colorants. The evidence for copigmentation and phytonut rient retention has been previously investigated using a divers ity of copigment sources ra nging from polyphenolics isolated from natural sources (rosemary, thyme, sage red clover, grape seeds, grape skins, orange/grapefruit peels, etc.), from pur ified compounds (various cinnamic acids and flavonoids), and from metal ions (calcium magnesium, aluminum, zinc, copper, iron, metal ascorbates) against a diversity of ant hocyanin sources (cabernet grape, muscadine grape, black carrot, purple sweet potatoes, re d cabbage, red radish, hibiscus, elderberry, and acai fruit) (Covarrubias, 2002; Kemmerer, 2002; McGuinness, 2002; Talcott et al., 2002a; Talcott et al., 2002b; Talcott et al., 2003 a; Brenes et al. 2004; Del Pozo-Insfran et al., 2004; Brenes et al., 2005). Overall, the ou tcomes of these studies demonstrated that the addition of polyphenolic cofactors can augment visual and instrumental color properties, increase heat and process stability of anthocyanins and ascorbic acid, increase overall radical scavenging properties, serve to enhance antimicrobial properties over that exhibited by anthocyanins and cofactor s alone, serve to inhibit lipoxygenase and polyphenol oxidase when highly pur ified cofactors are used.

PAGE 27

15 For instance, Talcott et al (2003a) demonstrated that addition of concentrated polyphenolic cofactors in the range of 0.1-0.4% v/v (100-1,500 mg/L gallic acid equivalents) readily forms copigment comp lexes with anthocyanins and results in concentration-dependent hyperchromic shifts from 10-50% depending on source, which also corresponds with increased antioxidant activity. Brenes et al. (2005) also showed significant improvements in phytonutrient re tention, when model anthocyanin systems copigmented with a purified water-soluble extr act of rosemary with and without fortified ascorbic acid were studied. In a temperatur e dependent reaction, the addition of cofactors was able to retain 14 and 24% more ascorbic acid at 0.2 and 0.4% (v/v ) cofactor addition; preserving both anthocyanin and ascorbic acid from oxidative and/or free radical damage in their mutual presence. Cofactor a ddition appreciably affected monomeric anthocyanins, overall color, and extended ant hocyanin half-life from 11 days without the rosemary extract to 15 and 19 days at 0.2% and 0.4% v/v rosemary extract, respectively. Other studies (Covarrubias, 2002; Kemmere r, 2002; McGuinness, 2002; Talcott et al., 2003a; Brenes et al. 2005) also have demonstrated that systems where these copigments are to be used should be free of residual oxidase enzymes, as polyphenolic cofactors can serve as enzyme substrates and destroy phytochemical and alter quality characteristics. In the presence of residua l enzymes, it was found that ascorbic acid fortification of copigmented grape anthocyanins was highly detrimental to fruit juice quality and resulted in rapid anthocyanin, ascorbic acid, and anti oxidant activity losses during processing, especially under high hydr ostatic pressure conditions where the enzymes may be more active (Talcott et al., 2003a). However, ascorbic acid fortification in the absence of cofactors demonstrated en zyme inhibition and serv ed to protect grape

PAGE 28

16 anthocyanins prior to thermal and high pr essure processing. It was concluded that although physicochemical attributes were e nhanced by copigmentation with rosemary extract, methods to inactive residual enzymes should be addressed prior to copigmentation to prevent degradation of anthoc yanins in the presence of ascorbic acid. Due to the outcome of several studies in which the protective e ffect of different cofactors was demonstrated to prevent phytonutrients loss, th e present research project suggests a strategy by which the addition of exogenous anthocyanin cofactors (watersoluble extracts from thyme and rosemary ) will alleviate many of the physical and chemical degradative reactions impacti ng anthocyanin-containing fruit juice and beverage fortification. Cofactors isolated fr om spices are an economical and food-grade source of polyphenolics for use in the beverage industry. Their polyphenolic concentrations are naturally ve ry high, so concentration e fforts are easily accomplished (Zheng and Wang, 2001). Based on present scientific knowledge it is hypothesized that anthocyanin copigmentation can offer a prot ective effect for for tified phytonutrients commonly used in the beverage industry. Su ccess of this work w ill generate knowledge for strategies to manufacture highly stable food colorants and to minimize phytonutrients loss. Sensory Attributes of Anthocyanin-conta ining Beverages as Affected by the Addition of Polyphenolic Cofactors. Other than empirical evidence from inform al evaluations, no formal evaluations on the organoleptic properties of copigmented an thocyanins have been conducted. Attributes such as color are usually measured using inst rumental techniques rath er than with human subjects. More importantly, th e taste attributes imparted by added copigments are an important consideration affec ting their use in food systems. Since polyphenolic cofactors

PAGE 29

17 may impart bitter or astringent flavors to a food system, or may impart aroma from volatile components, evaluating their sensory thresholds is vital to determining an effective use level. Taste or aroma carry-ove r from the use of natural cofactor sources such as thyme or rosemary extracts is also a concern, and use leve ls may be affected by the intensity of compounds that do not serve as anthocyanin cofactors (i.e. aroma-active compounds). A previous study evaluated the effect of a commercially available rosemary polyphenolic extract in the sensory attributes of a strawberry juice cocktail via a rating test using a 9-point hedonic scale evaluating color, aroma and flavor. Several rosemary extract concentrations (0, 0.1, 0.2 and 0.4% v/v) were added to the strawberry juice as anthocyanin cofactors. The use of these con centrations was based on preliminary studies (Talcott et al., 2003a; Del Pozo-Insfran et al ., 2004; Brenes et al., 2005) which showed that the addition of polyphe nolic-based anthocyanin cofactors, at these chosen concentrations, significantly reduced the mutual destruction of anthocyanins and ascorbic acid as well as appreciably augmented visual color and antioxidant capacity of model beverage systems. Results of the sensory evaluations (Table 1) concluded that the addition of rosemary cofactors significantly increased the color of the strawberry juice with respect to that of th e control treatment (P<0.01) i ndependently of the cofactor concentration (0.1-0.4% v/v). A ddition of rosemary cofactors did not significantly affect either the aroma or flavor of the juices. Re sults of both the sensory and stability studies give evidence that the addition of exogenous anthocyanin cofactors, at levels high enough to prevent phytonutrient degradation, exceed the physicochemical barriers influencing sensory perception.

PAGE 30

18 Table 2-1. Effect of rosemary extract (0, 0.1, 0.2, 0.4% v/v) on color, aroma, and flavor attributes of a commercial strawberry cocktail juice. Rosemary extract (% v/v) Color AromaFlavor 0 4.4 b1 5.4 a 4.7 a 0.1 6.1 a 5.6 a 5.1 a 0.2 6.4 a 5.7 a 5.2 a 0.4 6.6 a 6.2 a 5.2 a 1 Values with similar letters within columns of each sensor attribut e are not significantly different (LSD test, P>0.05), and indicate th e effect of rosemary extract addition. Since there is evidence that the addi tion of polyphenolic cofactors may exert a protective effect against oxidati on of anthocyanins and ascorbic acid, their effect on the organoleptic properties of copigmented bevera ge systems at levels in which they are effective are of importance and will be investigated in the present study. Novel Processing Technologies That May Enhance Anthocyanin Stability The food industry is continuously searchi ng for novel processing technologies that ensure microbial destruction and extend the shelf life of products without having adverse effects on their quality attribut es such as flavor, color, a nd phytonutrient retention (Butz and Tauscher, 2002; Matser et al., 2004). More over, current trends in food marketing and show that consumers desire high-quality f oods with "fresh-like" characteristics and enhanced shelf life that require only a mini mum amount of effort and time for preparation (Butz and Tauscher, 2002; Krebbers et al., 2003 ). At the present time, a wide variety of emerging non-thermal processing technologies are available to process food and beverages and include high hydrostatic pre ssure, irradiation, ultrasound, pulsed electric fields, light pulses, and os cillating magnetic fields.

PAGE 31

19 High hydrostatic pressure processing (HHP) and dense phase carbon dioxide pasteurization (DP-CO2) are promising alte rnatives to traditional pasteurization technologies and may lessen detrimental eff ects on thermolabile phytonutrients (Gomez and Ledward, 1996; Zabetakis et al., 2000; S un et al., 2002). In hydrostatic pressure applications, the food or beverage system is pressurized uniformly throughout the product in which the generation of pressure can be accomplished by direct compression, indirect compression using an intensifier pump, or by h eating of the pressure medium. One of the main differences between these processing tech nologies is that a wi de variety of food liquids and solids can be pr ocessed by HHP whereas DP-CO2 pasteurization only can process foods in a liquid form only. Another difference is that HHP utilizes batch and semi-continuous systems whereas DP-CO2 pa steurization uses continuous systems. Although both of these processing techniques are potential alternatives to thermal processing and may in theory reduce phytonutri ent losses and quality characteristic of foods, studies have shown that a downside of th eir use is the presence and/or activation of residual enzymes following processing conseque ntly resulting in extensive oxidative degradation. Due to these factors associated changes in phytonutri ent stability due to processing are of interest and will be inve stigated in the present study. Furthermore, comparisons between non-thermal and ther mal processes are one way to assess the benefits offered by these novel processing technologies. High Hydrostatic Pressure (HHP) Processing High hydrostatic pressure (HHP) processing, also referred as ultra high pressure (UHP) or high pressure proces sing (HPP), is a preservation te chnique currently used on a commercial scale in Japan, France, Spain, US A and Mexico for the pasteurization of a variety of food products that include fruit juices, guacamole, oysters and ham. This

PAGE 32

20 processing technology subjects foods with or without packaging to pressures between 100 and >1,000 MPa over a range of temperatur e (0-100 C) and time (2 sec and to >20 min) conditions. HHP acts instantaneous ly and uniformly throughout a food mass independently of its size, shape and compos ition. Therefore, adiabatic heating of the product occurs in homogeneous manner compared to that of conventional heat sterilization where a temperature profile o ccurs. Generally, the compression process will increase the temperature of foods approximately 3 C per 100 MPa. It may be expected that product characteristics that are dependent on the heat liab ility of certain components, are less significantly changed by high pressu re sterilization compared to conventional heat sterilization. The effect of high pressu re sterilization on produc t quality is strongly depended on the chosen food product. Some of quality parameters that may be affected are texture, color, producti on of off-flavors, and phytoche mical degradation. The results of research on high pressure sterilized produc ts have shown that the effects are product dependent and that careful selection of the appropriate process conditions is necessary (Barbosa-Canovas et al., 1998; He ndrickx et al., 1998; Butz an d Tauscher, 2002; Butz et al., 2003; Matser et al., 2004). HHP and microbial inactivation The effects of HHP on several food constitu ents have been studied over the last decade in order to evaluate the effect of these new technologies on food safety and quality. The outcomes of several published stud ies (Hendrickx et al., 1998; Smelt, 1998; Butz and Tauscher, 2002; Lado and Yousef, 2002; Butz et al., 2003; Matser et al., 2004) have shown that HHP processing can pr oduce commercially sterile products where kinetics of microbial inac tivation are highly dependent on processing parameters (pressure and temperature), acid content of the product and the type of microbial flora.

PAGE 33

21 For instance vegetative cells, including yeasts and molds, are rather pressure sensitive and can be inactivated by pressures between 300 and 600 MPa, whereas bacterial spores are highly pressure resistan t and need pressures >1,200 MPa for complete inactivation. Therefore, preservation of acid foods (pH 4.6) is the most likely application of HHP. On the other hand, sterilization of low-acid foods (pH > 4.6) requires a combination of pressure and mild-temperature treatments (H endrickx et al., 1998; Butz and Tauscher, 2002; Butz et al., 2003). Mi crobial inactivation is hi ghly dependent not only on processing parameters (pressure, temperat ure, time, number of pulses) and by food composition, but also by the types of micr oorganisms present in a food matrix. For instance, in order to ob tain a 4 log reduction of E. coli the medium needs to processed at 100 MPa at 30C for 720 min, while for L. monocytogenes the applied pressure needs to be increased to 238-340 MPa over 20 min in order to obtain a similar reduction in microbial load (Barbos a-Canovas et al., 1998). At ambient temperatures, application of pressures in the range of 400-600 MPa inactivate vegetative micro-organisms and re duce the activity of enzymes resulting in a pasteurized product, which can be stored for a considerable time at 4-6 C. The inactivation of vegetative micr o-organisms and enzymes, combined with retention of phytochemicals and low molecular weight food molecules that are re sponsible for taste and color, results in HHP-pasteurized products with a prolonged shelf-life and fresh-alike characteristics. For sterilization of HHP products a combined process where both pressure and temperature (60110C) is needed to achieve the complete inactivation of spores and enzymes. The result of the later process is a shelf stable product, which in many cases has a higher degree of qualit y than those products obtained using

PAGE 34

22 conventional processing. High-pressure inac tivation of vegetative micro-organisms is caused by membrane damage, protein denatura tion and decrease of intracellular pH, suggesting that pressure results in deactiv ation of membrane-bound enzymes associated with efflux of protons. Water activity and pH are critical pro cess factors in the inactivation of microbes by HHP. Temperatur es in the range of 45–50 C appear to increase the rate of inactiva tion of food pathogens and spoilage microbes. Temperatures ranging from 90 to 110 C in conjunction with pressures of 500–700 MPa have been used to inactivate spore-forming bacteria such as Clostridium botulinum (Hendrickx et al., 1998; Butz and Tauscher, 2002; Butz et al., 2003). The effect of HHP in food components An advantage of HHP processing is that food quality characteristics, sensory attributes, and phytonutrient rete ntion are either unaffected or only minimally altered by processing at room temperature, except when some type of enzymatic activity is involved. For instance, a recent study investigated the effect of different high pressure treatments on odor and aroma of an orange-lem on-carrot juice mixtur e and its subsequent storage (21 days at 4 C) (But z et al., 2003). Results indicated that HHP treated juices (e.g. 500 MPa for 5 min) presented only minor changes in odor and flavor after processing when compared to pasteurized juices. Moreover, the HHP juices did not present significant changes in odor, flavor and overall quality after storage whereas attributes were significantly decreased for c ontrol juices. However, several studies have also demonstrated that besides microbial destruction there are other pressure-induced effects on food components such as prot ein denaturation, enzyme activation or inactivation, changes in enzyme-substrate in teractions, changes in the properties of polysaccharides and fats, protein gelation, etc (Tauscher, 1998; Messens et al., 2002;

PAGE 35

23 Butz et al., 2003). The physiochemical changes induced by HHP have also open up the possibility of producing foods with novel text ure (e.g. meat, fish, dairy products) as well as the modification of food protein functionali ty (Messens et al., 2002). Several chemical changes have been reported for food macrom olecules that have been HHP, such as the stability of aspartame present in milk, TRISbuffers and water during different treatments (600 MPa at 60 C for 3-30 min) (Butz et al., 1997). Results of this study indicated that after HHP with a holding-time of <3 min onl y about 50% of the original content of aspartame was detectable in milk (pH 6.8). The degradation by-products were identified as aspartylphenylalanine and a diketopiperazi ne. However, the stability of aspartame was insignificantly affected when present in model acid systems resembling fruit preparations, juices, or carbonated drinks. A nother example of pressure-induced chemical changes is that observed for -carotene in model solutions and in sliced carrots under different pressure and temper atures regimes (Tauscher, 1998) This study showed that the content of -carotene in ethanolic model solutio ns after HHP for 20 min at 75 C was reduced by more than 50%. However, its content was not reduced significantly when processed for 40 min at 600 MPa and 75 C. Au thors proposed that the carotenoids were well protected against pressure/temperature de gradation in the last instance since these compounds are buried in lipophilic environments. Both of th ese studies demonstrate the importance of the food matrix as a benefi cial protective action on the retention of nutrients and quality at tributes after HHP. The balance of intramolecular food component s and solvent-protein interactions is greatly affected by the HHP parameters (pressu re and temperature). Therefore the extent of unfolding of the polypeptid e chain is strongly dependent on the processing conditions

PAGE 36

24 and consequently one can observe different en zymatic activity on the quality attributes of a food matrix. Structural rearrangements ta king place in the protein upon pressurization are governed by the principle of Le Chatelier, which states that proc esses associated with a volume decrease are encouraged by pressure increases, whereas processes involving a volume increase are inhibited by pressure increases. The volume decrease accompanying denaturation arises from the formation or rupture of noncovalent bonds and from the rearrangements of solvent mo lecules (Hendrickx et al., 1998 ; Butz and Tauscher, 2002; Butz et al., 2003). Pressure induced activati on of enzymes and/or their residual activity can significantly affect the qua lity of food products. Due to th e pressure stability of some of these food quality-related enzymes, co mbined technologies involving pressure and temperature are necessary for complete enzymatic inactivation. Such e ffect is specifically related to the type of enzymes and the processing conditions of foods. Studies have shown that some enzymes can be deactivated using pressures < 200 MPa, while others can withstand pressures over 1,000 MPa (Barbo sa-Canovas et al., 1998; Hendrickx et al., 1998; Butz and Tauscher, 2002; Butz et al ., 2003; Matser et al., 2004). For example, several studies have investigat ed the effect of HHP on differe nt pectin methyl esterases (PME) and its consequent effect on cloud dest abilization, gelation and loss of consistency of several food products (Seyderhelm et al., 1996; Basak and Ramaswamy, 1996; Cano et al., 1997; Stoforos and Taoukis; 2003; Irwe and Olsson, 1994). PME is usually inactivated by conventional thermal processes that have detrimental effects on flavor, color and nutritional quality. St udies have shown that HHP processing of orange juice results in a commercial stable product with higher quality at tributes when compared to that of thermally processed orange jui ce (Irwe and Olsson, 1994; Basak and Ramaswamy,

PAGE 37

25 1996; Stoforos and Taoukis; 2003). HHP treatments of > 600 MPa have been shown to irreversibly inactivate (> 90%) PME. However, tomato PME seems to be more pressure resistant and it seems to be activated during HHP at < 400 MPa especially in the presence of calcium ions and in acidic media (pH 3.5-4.5) (Seyderhelm et al., 1996). Pressureinduced activation of orange juice PME was al so noted by Cano et al (1997) in the case of treatments at room temperature and 200–400 MP a. Krebbers et al. (2003) showed the combined effect of HHP and thermal treatmen ts on the quality attr ibutes and microbial stability of tomato puree. Their results show ed that HHP alone cause d partial inactivation of PG (~70%) yet activ ation of PME was observed. The use of combined treatments (e.g. 700 MPa and 90C for 30 s yield a commercial sterile product (> 4.5 log reduction) that had > 99% enzyme inactivation (both PME a nd PG). The obtained sh elf stable product had superior sensory and quality attributes (increased color, increased water binding, lower viscosity, and higher lycopene retention) when compared to a thermally sterilized product. Similar trends have been observed for food products in the case where polyphenoloxidase (PPO) is present. Mushroom and potato PPO are very pressure stable, since treatments betw een 800 and 900 MPa are required for activity reduction (Eshtiaghi et al., 1994; Gomez and Ledward, 1996; Weemat es et al., 1997). A similar trend was observed for avocado PPO by Hendrickx et al (1998) who investig ated the combined effect of pressure (0.1-900 MPa) and temperature (25-77.5 C ). Results of this study demonstrated that PPO inactivation at 21 C was only achieved for pressures > 900 MPa. Pressure induced activation has also been re ported for apple (Anese et al., 1995), onion (Butz et al., 1994), pear (Asa ka et al., 1994) and strawb erry PPO (Cano et al., 1997).

PAGE 38

26 Grape, strawberry, apricot and apple PPO seem to be more pressure sensitive than other PPOÂ’s. Apricot, strawberry and grape PPO can be inactivated by pressures exceeding 100, 400 and 600 MPa, respectively. Inactivatio n of apple PPO varies between HHP at 100-700 MPa in function of the pH of the matr ix. For several PPO enzymes, it has been reported that pressure-induced inactivation proceeds faster at lower pH, however the inactivation is also influenced by the addi tion of salts, sugars or other compounds. For example, the pressure inactivation of apple PPO is enhanced by the addition of calcium chloride whereas for mushroom PPO is enhanc ed in the presence of 50 mM benzoic acid or 5 mM glutathione. The sensitizing effect of glutathione was suggested to be due to an interaction with a disul phide bond of the enzyme (Anese et al., 1995). Since HHP processing is a promising alte rnative to traditional pasteurization technologies and may lessen detrimental effect s to thermolabile compounds associated changes in phytonutrient stability due to processing are of interest and will be investigated in the present study. Furtherm ore, comparisons between non-thermal and thermal processes are one way to assess the benefits offered by these novel processing technologies. Dense Phase-CO2 Pasteurization As previously mentioned, HHP is currently used in the U.S., Europe and Japan to produce a variety of commercial products. Howe ver, in order to ensure the safety of foods against some pressure -resistant microorganisms a nd bacterial endospores, HHP needs to be used in combination with other thermal and non-thermal processes. Moreover, processing techniques that al ong with improving the efficacy of HHP microbial inactivation decrease processi ng costs (i.e operating pressures and temperatures, dwell time) are desired. Dens e phase-CO2 pasteurization (DP-CO2), also

PAGE 39

27 known as high-pressure carbon diox ide processing, is therefor e a potential candidate as a non-thermal processing due to its ability to inactivate both microbes and enzymes under more cost-effective processing conditions. The principle of the microbial inactivation of DP-CO2 is based on gas dissolution in a mi crobial cell by pressu rization that, when rapidly decompressed to atmospheric pressu re, causes fatal functional damage and explosive decompression of th e cell (Balaban et al., 1991; Park et al., 2002). It is noteworthy to mention that CO2 pressurization does not always lead to cell burst but in some cases only leads to leakage of ce llular components and changes in the cell membrane permeability which are responsible for cell damage and microbial inactivation (Lin et al., 1993; Isenschmid et al., 1995; Pa rk et al., 2003). DP-CO2 affects biological systems by causing protein denaturation, lipid phase changes, and r upture of cell walls and membranes. DP-CO2 and microbial inactivation Carbon dioxide under both atmospheri c and supercritical pressures has demonstrated antimicrobial effects in foods It is used successfully in modified atmosphere packaging to reduce horticultura l respiration and reduce microbial growth (Corwin and Shellhammer, 2002). Similarly, it extends the shelf-life of dairy products through inhibition of microbial growth. In add ition to its use at atmospheric conditions, supercritical carbon dioxide can inactivate a wi de range of microbiota (Kamihira et al., 1987; Lin et al., 1993; Ballestra et al., 1996) and recent work indicat es that it can also inactivate bacterial spores (Kamihara et al., 1987; Haas et al.; 1989 Enomoto et al., 1997). The inhibitory effects of DP-CO2 a nd supercritical carbon dioxide (SC-CO2) on microorganisms for food preserva tion have recently received a great deal of attention and have been extensively investigated over the last decade. Fraser (1951) first tried gas

PAGE 40

28 pressurization with N2, NO2, Ar, and CO2, and reported that CO2 could inactivate 95-99% of E. coli at 3.40 MPa and 37 C. Nakamura et al. (1994) found that 108 cells/ml of bakerÂ’s yeast could be inactivated after CO2 saturation at 4.05 MPa and 40C for 3 h, and proposed this processing as a novel method for sterilization of food microorganisms. Kamihira et al. (1987) demons trated that complete inac tivation of bakerÂ’s yeast, E. coli S. aureus and A. niger can be accomplished by contact with SC-CO2 at 20.26 MPa and 35 C for 2 h. More recently, Park et al (2002) observed that DP-CO2 exerted a relatively large effect on tota l aerobes present in a carrot juice at 0.98 MPa and that its bactericidal effect gradually increased to achieve a 4-log reduction at 4.9 MPa for 10 min. Ballestra et al. (1996) found a decrease in the survivors of E. coli at 5 MPa using CO2 at temperatures > 35 C, with the most effective inactivation found after 5 MPa at 45 C for 20 min. However, Corwin and Shellham mer (2002) found that under the studied conditions (0, 365 and 455 MPa) the amount of dissolved CO2 in the medium had only a slight significant effect in the inactivation of E. coli K12 and that the processing pressure was responsible for the 4 and 6-log reducti on in microbial reduction, respectively for 365 and 455 MPa. Similar results were obtained by Pa rk et al. (2003) as carbonation itself did not have a significant effect on B. subtilits inactivation rates yet a synergistic effect was observed when combined with HHP and resu lted in a 5-log reduction at 600 MPa. Enomoto et al. (1997) examined the lethal effect of DP-CO2 on spore cells of Bacillus megaterium and observed that the bactericidal effect of CO2 was enhanced with increasing temperature and tr eatment time, and that processing at 5.9 MPa and 60 C for 30 h could reduced the survival ratio of the s pores to about 10-7. Ka mihara et al. (1987) also observed that under low pH and hea ting conditions, SC-CO2 was an effective

PAGE 41

29 method to inactivate endospores of B. subtilis and B. stearothermophilus The later observation (low pH and high temperature conditions) was confirmed by Haas et al. (1989) who achieved a high reduction of th e survival ratio for spore cells of C. sporogenes 3679 under SC-CO2 at 5.5 MPa at 70 C for 2 h. During DP-CO2 the carbon dioxide solubility increases directly proportional with increments of processing pressure. Therefore a pH change of the medium is caused due to carbonic acid formation. Although several studies have attributed the later fact as the main explanation for the antimicrobial activ ity of SC-CO2, several studies have shown that extraction of essential intracellula r substances such as phospholipids and hydrophobic compounds from cells or cell membrane s, and enzyme inactivation also play important roles as mechanisms of microbial inactivation (Kamihira et al., 1987; Haas et al., 1989; Lin and Lin, 1993; Ballestra et al ., 1996). For instance, Haas et al. (1989) showed that a 6 log bacterial num ber reduction was obtained following CO2 pressurization compared to only a 2 log reducti on for a control both in which its pH was reduced from 5.3 to 3.2. These results were s ubsequently confirmed by Ballestra et al. (1996) who also observed that the microbial destruction occurred in two stages during processing at 5 MPa for 15 min at 35 C. In the first step, cells undergo a stress by pressurized carbon dioxide and in which a slower inactivation rate o ccurs when compared to the second step of microbial destruction where rapid inactivation occurs as a critical level of the gas is reached. The authors also noticed selective enzyme inactivation which was attributed as a result of the internal drop of pH during proces sing. Lin et al. (1994) also observed two inactivation stages when studying the effect of pressurized carbon dioxide on the viability of L. monocytogenes These authors also observed that the pH

PAGE 42

30 drop was not the only reason for microbial inactivation, but that the conversion of bicarbonate, formed from the dissociation of carbonic acid, into car bonate precipitates intracellular calcium and other ions re sulting in cell malfunction and damage. The temperature, as well as the processing pressure, at which a food or beverage is processed significantly affects the effici ency of carbon dioxide processing as a nonthermal pasteurization technique since both parameters contro l the solubilization rate of CO2 and its solubility in a suspending medium (Erkman, 2000a; Erkman,2000b). For instance, Arreola et al. (1991b) achieved a 2-log decrease in total plate count during the pressurization of a single strength orange juice at 33 MPa and 35 C for 1 h, while processing at 45 and 60 C achieved the same reduction at 45 and 15 min, respectively. Authors also observed a decrease in D-va lues for microbial reduction at the same temperature when pressure was increased, and results were attributed due to the combination of high pressure, shear rate dur ing depressurization of the juice, and the larger extent of carbon dioxide solubility. Er kmen (2000a, 2000b) also observed that the time to achieve complete inactivation of L. monocytogenes at 6.08 MPa CO2 and 25 C was reduced from 115 min to 75 and 60 min at 35 and 45 C, respectiv ely. Similar results were observed by Ballestra et al. (1996) while studying the survival rates of E. coli in Ringers solutions and in which results were attributed to increments on the amounts of dissolved CO2. The later effect can be attributed to a higher CO2 absorption as the processing temperatures increases. Hong et al (1999) reported that microbial inactivation by DP-CO2 is governed essentially by the tr ansfer rate and the penetration of carbon dioxide into cells, the effectiveness of wh ich can be improved by increasing pressure, decreasing the pH of the suspension, a nd increasing the processing temperature.

PAGE 43

31 Temperature has a close relation with the characteristics of CO2 mass transfer and most likely higher temperatures stimulate the CO2 diffusivity into the microbial cell and could also increase the fluidity of the cell membrane to make the CO2 penetration more easy (Erkmen, 2001). Lin et al. (1994) along with Erkmen (2000a) and Wei et al. (1991) reported that not only processing conditions play an important role in microbi al stability but also food components (i.e. fats, proteins), since they might reduce the bactereo static effect of CO2 by delaying its penetration into the cells. For instance, Wei et al. (1991) observed that SC-CO2 (13.7 MPa, 35 C for 2 h) inactivated completely Salmonella in egg yolks, while processing of whole eggs only resulted in a 64% bacterial in activation. Similar observations were observed by Erkman (2000a ; 2000b; 2000c) in which inactivation of Staphylococcus aureus suspended in broth was achieved at lower pressures and shorter processing times when compared to that of raw milk. This effect was further demonstrated when comparing the protecti ng effect of certain food components on microbial survival rates in raw milk when co mpared to those of orange, peach and carrot juices. Other authors also have stated that the water conten t of foods significantly affects the efficacy of CO2 to inactivate microorganisms. Kami hira et al. (1987) observed that the sterilization of Koji, wh ich contained bakerÂ’s yeast, E. coli Staphylococcus aureus and conidia of Aspergillus niger was achieved at 20 MPa and 35 C when the water content of each microorganism was 70-90% (wet cells). However, when the water content was decreased to 2-10% (dry cells) an incomplete microbial inactivation was observed under the same processing conditions. The outcome of this study demonstrated that product sterilization depends on severa l factors including type of microorganism,

PAGE 44

32 water content, and addition of co-solvent. Au thors concluded that when the water content of a microbial suspension increases, the wa lls of the cell swell a nd thus carbon dioxide has a larger surface ar ea of penetration and consequen tly has a larger effect on the microbial cell. In addition, when combined with water, carbon di oxide produces carbonic acid that affects cell permeability (Lin et al ., 1994). Kumagai et al. (1997) also observed that a higher yeast inactivation is accomplishe d with higher water ac tivities of foods and higher processing pressures due to their effects on CO2 adsorption in the yeast cells. Besides processing parameters (p ressure, temperature, time, CO2 concentrations) and food composition, the presence of co-solvents (i.e. ethanol) also significantly affects microbial inactivati on in the pressurized medium. The presence of these compounds modifies the rate of CO2 solubility and adsorption by m odifying the critical temperature and pressure of the medium(T aylor, 1996). For instance, the addition of ethanol increases CO2 adsorption on surface sites dur ing extraction of components and thus prevents the readsorption of certain compounds (Clifford and Williams, 2000). Although the exact mechanism of the bacter iostatic action exerted by pressurized CO2 is not known several possible mechanisms have been reported and include (Daniels et al., 1984; Lin and Lin, 1993; Lin et al.; 1994; Wei et al., 1991; Ballestra et al., 1996; Erkmen, 1997;): • Reduced growth rate of aerobic bacteria due to the replacement of oxygen by CO2. • Formation of carbonic acid resultin g on decreases in the cell’s pH and consequently affecting metabolic activities. • CO2 penetration into the cell which may enhance its chemical activity on the internal metabolic processes of the cell

PAGE 45

33 • Increase in the cell membrane permeabi lity by high pressure treatment causing cell leakage and damaging cell function • Protein denaturation stops the uptake of amino acids which are essential for cell growth and also affects the en zymatic systems of the cell. DP-CO2 and enzymatic inactivation Several studies have shown the effect of SC-CO2 and DP-CO2 on PME, PE, lipoxygenase (LOX), peroxidase (POD), and PPO in model and real food systems (Chen et al., 1992; Park et al., 2002; Corwin and Shellhammer, 2002; Boff et al., 2003;). For instance, Taniguchi et al. (1987) studied the effect of SC-CO2 on ni ne different enzymes at 20.3 MPa and 35 C for 1 h, and authors show ed that > 90% of th e enzymatic activity was retained when the water content of the en zyme preparations was 5-7% wt. Chen et al. (1992) reported that PPO can be inactivated at low temperatures with SC-CO2; however, the degree of inhibition was dependent on the source of the enzyme. In this study, spiny lobster PPO was greatly inactivated followe d by shrimp, potato, and lastly apple juice PPO. The circular dichroism spectra at fa r UV-range showed that SC-CO2 treatment caused conformational changes in the secondary structure of the enzymes, being source of marines enzymes (lobster and shrimp) th e ones that underwent the most evident conformational changes. In the same study, the authors showed that SC-CO2 also inhibited orange juice PE where therma l inactivation was insignificant. Authors concluded that the extent of PE inactivati on depended on pressure, temperature and time of processing. Overall results of the study showed that SC-CO2 processing was an effective non-thermal technology to reduce mi crobial loads and enzy me activity. Arreola (1990) studied the effect of SC-CO2 and HHP on the microbial stability and quality attributes of a single strength orange juice. The author used a batch system where the gas was allowed to mix in a closed vessel under high pressure at temper atures between 35 to

PAGE 46

34 60 C. Results of this study showed that this process was effective in destroying microorganisms and obtaining an acceptable product with improve d cloud retention, despite residual PE activity (50%) was pr esent following processing. Arreola et al. (1991b) also investigated PE activity and showed that its inactivation was affected by temperature, pressure and process time. Co mplete PE inactivation was achieved at 26.9 MPa and 56 C for 145 min. Boff et al. (2003) also investigated the effect of HHP and DP-CO2 on PME activity and the physiochem ical properties of a single-strength Valencia orange juice following processing a nd during 4 months of storage at 4 and 30 C. Authors observed that although 28% of PM E activity remained af ter processing this product had enhanced cloud stability and higher ascorbic acid retention when compared to HHP and thermally pasteurized samples. DP-CO2 produced a cloud-stable orange juice with more ascorbic acid and flavor vol atiles than the thermally processed juice. Corwin and Shellhammer (2002) also comp ared the inactivation of PME and PPO by HHP and DP-CO2 at 25 and 50 C. Authors observed that in the inactivation of PME, pressure was a significant factor at both processing temperatures and that CO2 was a significant factor in further inactivating PME beyond that which pressure would achieve alone. Authors observed the same trend but fo r PPO as HHP processing only reduced this enzymatic activity by 8-21% when compar ed to 44-79% inactivation for DP-CO2 processed treatments. Park et al. (2002) obs erved that a combined treatment of 4.90 MPa of SC-CO2 and 600 MPa-HHP effectively inac tivated enzymes in a carrot juice. The residual activities of PPO, LOX, PME in this study were less than 11.3%, 8.8%, and 35.1%, respectively. The effect of SC-CO2 in LOX and POD activity was also been investigated by Tedjo et al (2000) in 30% sucrose solu tions. Authors observed that

PAGE 47

35 application of SC-CO2 at 35.2 MPa and 40 C for 15 min inactivated 35% of LOX activity, while pressurization at 62.1 MPa and 55 C for 15 min inactivated 65% of POD activity. Total inactivation of LOX (10.3 MP a, 50 C and 15 min) and POD (62.1 MPa, 55 C and 15 min) was achieved through SC-CO2 for unbuffered solutions. These authors also observed that by increasing the concentration of sucrose and buffering (pH range 4 to 9) the working solutions the en zymes increased their resistance for SC-CO2 inactivation. Since the use of DP-CO2 processing has been shown to be a promising non-thermal process to inactivate microorga nisms and enzymes it might be used as a novel technology to enhance phytochemical stability. However, its effect on anthocyanin and ascorbic acid stability is not widely known and, therefore, will be investigated in the present study. Moreover, the effect of DP-CO2 in the se nsory attributes and quality retention of anthocyanin-containing juices following processing and during storage has not been investigated.

PAGE 48

36 CHAPTER 3 PHYTOCHEMICAL COMPOSITION AND PIGMENT STABILITY OF AAI ( EUTERPE OLERACEA MART.) Introduction Aai ( Euterpe oleracea Mart.) is a palm plant widely distributed in northern South America with its greatest occurrence and econom ic importance in the floodplains of the Brazilian Amazonian state of Par (Muniz-Miri et et al., 1996; Silva et al., 1996; Murrieta et al., 1999;). Aai is a slender, multi-stemm ed, monoecious palm that can reach a height of over 30 meters. A wide variety of market able products are produced from this palm, but the spherical fruits that are mainly ha rvested from July to December are its most important edible product. Each palm tree produces from 3 to 4 bunches of fruit, each bunch having from 3-6 kg of fruit. The round-shap ed fruits appear in green clusters when immature and ripen to a dark, purple colo red fruit that ranges from 1.0-1.5 cm in diameter. The seed accounts for most of the fr uit size and is covered by thin fibrous fibers under which is a small edible layer. A viscous juice is typically prepared by macerating the edible pulp that is approximately 2.4% protein and 5.9% lipid (Silva, 1996). The juice is currently used to produce energetic snack beverages, ice cream, jelly, liqueur, and is commonly blended with a vari ety of other juices. A steady increase in the development of natural food colorants and functional food sources has been observed in recent years, not only due to consumer preferences for natural pigments but also fo r their health-related benefits and nutraceutical properties (Frankel et al., 1995; Skrede et al., 2000; Meyer et al., 1997;). Anthocyanins are a viable

PAGE 49

37 replacement for synthetic colorants due to their bright, attracti ve colors and water solubility, which allows their incorporati on into a variety of food systems (RodriguezSaona et al., 1999). However, limitations ex ist for their commercial application due to high raw material costs and their poor stability that is affected by their chemical structure, environmental factors, and the presence of additional phytochemicals in solution. Due to these constraints, a need exists to find st able, inexpensive anthoc yanin pigments with a diverse array of functiona l properties food and nutrace utical applications. Anthocyanin intermolecular copigmentati on reactions are common in nature and result from association between pigments and cofactors such as polyphenolics and/or metal ions, or other anthocyanins (self-asso ciation). Preferably formed under acidic conditions, these weak chemical associati ons can augment anthocyanin stability and increase antioxidant proper ties (Mazza and Brouillard, 1990; Boulton, 2001; MalienAubert et al., 2001). Studies have suggested that the c opigmentation phenomenon is the main anthocyanin stabilizing mechanism in plants (Mazza and Brouillard, 1990; Boulton, 2001). Polyphenolics are the predominant cofa ctors present in an thocyanin-containing fruits and vegetables, and incr eased anthocyanin stab ility has been attributed to their high concentrations in foods (Mazza and Brouill ard, 1990; Boulton, 2001; Malien-Aubert et al., 2001). Malien-Aubert et al. (2001) de scribed how the diversity of polyphenolic compounds among different anthocyanins source s might affect anthocyanin stability, yet additional research on how these polypheno lics influence anthoc yanin stability via copigmentation reactions has not been conducted. The objective of this study was to characterize the ma jor polyphenolics and anthocyanins present in aai pulp, and to determine their contri bution to the overall

PAGE 50

38 antioxidant capacity of this palm fruit. Color and pigment stability against hydrogen peroxide, ascorbic acid, and the presence/abse nce of naturally occu rring cofactors was also determined and compared to other commercially available anthocyanin sources. Results of these studies can be used to dete rmine application and f unctional properties of aai polyphenolics in a variety of food products. Materials and Methods Materials Pasteurized, frozen aai pulp was kindly donated by Amazon Energy, LLC (Greeley, CO) and was shipped overnight to the Department of Food Science and Human Nutrition at the University of Florida. The pulp was thawed, centrifuged (2,000 x g ) at 4 C for 15 min to separate lipids from the aqueous slurry, and subsequently filtered through Whatman #1 filter paper. The aqueous portion was then partitioned into lipophilic and hydrophilic extract s by the addition of petr oleum ether and acetone, respectively. The upper petroleum ether pha se was removed and evaporated under a gentle stream of nitrogen and re-dissolved in a known volume of acetone and ethanol (1:1). Acetone was removed from the lower aqueous phase under re duced pressure at temperatures <40 C, and the resultant fraction containing hydrophilic compounds was diluted to a known volume with acidified water (0.1% HCl). Polyphenolics from the aqueous phase were subseque ntly concentrated using C18 Sep-Pak Vac 20 cc minicolumns (Waters Corporation, Mass. U.S.A.). Residual sugars and organic acids were removed with water (0.01% HCl), and polyphe nolic compounds recovered with acidified methanol (0.01% HCl). Methanol was rem oved from the polyphenolic fraction using vacuum evaporation at <40 C, and the resu lting isolate was re-dissolved in a known volume of acidified water.

PAGE 51

39 Commercially available anthocya nin extracts from black carrot ( Daucus carota ; Exberry, Tarrytown, NY), red cabbage ( Brassica oleracea ) (Exberry), red grape ( Vitis vinifera) (San Joaquin Valley Concentrates, Fresno, CA), purple sweet potato ( Ipomea batata ) (Food Ingredients Solutions, New York, NY), and a non-commercial extract from red hibiscus flowers ( Hibiscus sabdariffa ) were used for color stability evaluation. Each pigment source was dissolved in citric aci d buffer (pH 3.5; 0.1 M), and polar compounds removed with C18 columns as previously described. Co lor and anthocyanin stability were then assessed against aai for comparison. Color Stability Anthocyanin color stability of each pigmen t source was assessed in the presence of hydrogen peroxide (0 and 30 mmol/L) at 10, 20, and 30 C, respectively. Stock solutions of each anthocyanin source were diluted with citric acid buffer (pH 3.5) to give a final absorbance value of 1.5 at th eir respective wavelength of maximum absorbance. Samples were placed into a water bath or refrigerat ed storage and allowed to reach the desired temperature at which a hydrogen peroxide solution was added. Loss of absorbance was measured periodically over time and percent color retention ca lculated as a percentage of the initial absorbance reading. Insignificant changes in abso rbance values were observed for control treatments (no hydrogen peroxi de) over 360 minutes of incubation. Effect of Copigmentation The effect of naturally occurring inte rmolecular copigmentation on anthocyanin stability in the presence and absence of ascorb ic acid was also evaluated using in vitro model systems. Naturally occurring cofactor s were removed by additionally loading each anthocyanin source onto C18 cartridges as previously de scribed. Following elution of polar compounds with water, the cartridge was first washed with ethyl acetate to elute

PAGE 52

40 phenolic acids and flavonoids, followed by acidi fied methanol to remove anthocyanins. Ethyl acetate and methanol isolates were th en evaporated under vacuum at <40 C, and re-dissolved in a known volume of citric aci d buffer. Anthocyanin recovery was >96% for all sources. Anthocyanin color stability was evaluated using an in vitro model simulating a soft drink beverage system that contained anthocyanins (absorbance value of 1.5) dissolved in citric acid buffer, sucros e (100 g/L), and sodium azide (50 mg/L) to control microbial growth. Stock solutions were sub-divided and evaluated with and without polyphenolic cofactors, and again sub-divided for evaluation with and without ascorbic acid (450 mg/L). Data were compared to a control that contained an equivalent volume of citric acid buffer. Each treatment was individually sealed into screw-cap vials (10 mL), and stored in the dark at 37 C for 30 days. Samples were collected every day during the first 8 days of the study, and subse quently every other day until the end of the study. Phytochemical Analyses Individual anthocyanin 3-glyc osides present in aai we re quantified according to the HPLC conditions of Skrede et al. (2000 ) using a Dionex HPLC system and a PDA 100 detector. Compounds were separated on a 250 x 4.6 mm Supelcosil LC-18 column (Supelco, Bellefonte, PA) and quantified using a cyanidin st andard (Polyphenols Laboratories AS, Sandnes, Norway). Anthocya nins were also characterized based on PDA spectral interpretation from 200-600 nm, and identification additionally confirmed following acid hydrolysis into their respect ive aglycones with 2N HCl in 50% v/v methanol for 60 min at 90 C. Major flavonoids and phenolic acids presen t in aai were separated by HPLC using modified chromatographic conditions of Talcott et al. (2001) Separations were

PAGE 53

41 performed on a 250 mm X 4.6 mm i.d. Acclaim 120-C18 column (Dionex, Sunnyvale, CA) with a C18 guard column. Mobile phases consisted of water (phase A) and 60% methanol (phase B) both adjusted to pH 2.4 with o -phosphoric acid. A gradient solvent program ran phase B from 0 to 30% in 3 mi n; 30-50% in 5 min; 50-70% in 17 min; 7080% in 5 min; and 80-100% in 5 min, and held for 10 min at a flow rate of 0.8 mL/min. Polyphenolics were identified by spectrosco pic interpretation, retention time, and comparison to authentic standards (Sig ma Chemical Co., St. Louis, MO). Six isolates were obtained from the extr action of aai pulp that included whole pulp, lipophilic fraction, C18 non-retained, C18 bound phenolics and anthocyanins, ethylacetate soluble polyphenolics, and anthocya nins. Each fraction was evaluated for antioxidant capacity using the oxygen radi cal absorbance capac ity assay against a standard of Trolox as described by Talcott et al. (2003b). Each isol ate was appropriately diluted in pH 7.0 phosphate buffer prior to pi petting into a 96-we ll microplate with corrections made for background interference due to phosphate buffe r and/or extraction solvents. Anthocyanin content in each in vivo mode l system was determined with the pH differential method of Wrolstad (1976) a nd quantified using e quivalents of the predominant anthocyanin present (cyanidin 3glucoside for aai and hibiscus; cyanidin 3sophoroside for black carrot and red cabbage ; malvidin 3-glucoside for red grape; pelargonidin 3-rutinoside for purple sweet potat o) (Malien-Aubert et al., 2001; Wrolstad, 1976; Hong and Wrolstad, 1990). Percentage of polymeric anthocyanins was determined based on color retention in the presence of potassium metabisulfite (Wrolstad, 1976),

PAGE 54

42 while instrumental CIE color characteristic s (chroma, and hue angle) were measured using a Minolta Chroma Meter CR-300 Seri es (Minolta Co., Ltd., Osaka, Japan). Statistical Analysis Anthocyanin stability against hydrogen per oxide was designed as a 6 x 2 x 3 full factorial that included six anthocyanin s ources, two hydrogen peroxide concentrations, evaluated at three temperatures. Anthocyanin stability in the presence of cofactors and ascorbic acid was designed as a 6 x 2 x 2 full factorial that included six anthocyanin sources, two ascorbic acid levels, in the pres ence or absence of native cofactors. Data for these evaluations and those for aai charac terization represent the mean of three replicates at each sampling point. Multiple li near regression, analysis of variance, and Pearson correlations were conducted using JMP software (SAS, Cary, NC) and mean separation using the LSD test ( P < 0.05). Results and Discussion Anthocyanin and Polyphenolic Characterization Due to recurrent issues associated with the instability of anthocyanins during processing and storage, the food industry is c onstantly looking for novel, inexpensive and stable sources of pigments. Anthocyanins pres ent in aai may offer a new source of these pigments, however their stability has ye t to be determined. Furthermore, the characterization of the major polypheno lic compounds in aai and their overall contribution to the antioxidant capacity has no t been previously inve stigated. Therefore, this study examined the polyphenolic compositi on and the anthocyanin stability of aai under a variety of experimental conditions as compared to other commercially available anthocyanin sources.

PAGE 55

43 Figure 3-1. HPLC chromatogram of A. anthocyani n 3-glucosides monitored at 520 nm (Peak assignments: 1. cyanidin 3-glucos ide; 2. pelargonidin 3-glucoside) and their B. aglycones (Peak assignments: 3. cyanidin; 4. pelargonidin) present in aai ( Euterpe oleracea Mart.). Figure 3-2. HPLC chromatogram of A. phenolic acids monitored at 280 nm and B. flavonoids monitored at 360 nm present in aai ( Euterpe oleracea Mart.). Peak assignments: 1. gallic acid; 2. p -coumaric acid; 3. protocatechuic acid; 4. (+)-catechin; 5. p -hydroxybenzoic acid; 6. vanillic acid; 7. gallic acid derivative-2; 8. gallic acid derivative-5; 9. gallic acid derivative-3; 10. gallic acid derivative-1; 11. ferulic acid ; 12. (-)-epicatechin; 13. gallic acid derivative-4; 14. ellagic acid; 15. ellagic acid derivative. 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 30.0 32.5 35.0 37.5 40.0 42.5 45.2 103 -50 0 50 100 150 200 251 1 Phenolics-OK #3 [modified by Florida U Acai-2 X UV_VIS 2 Phenolics-O K Acai-2 X UV_VIS mAU min 2 1 WVL:360 nm A B10 7 15 14 11 8 2 1 9 3 4 5 6 12 13Time (min) 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 30.0 32.5 35.0 37.5 40.0 42.5 45.2 103 -50 0 50 100 150 200 251 1 Phenolics-OK #3 [modified by Florida U Acai-2 X UV_VIS 2 Phenolics-O K Acai-2 X UV_VIS mAU min 2 1 WVL:360 nm A B10 7 15 14 11 8 2 1 9 3 4 5 6 12 13Time (min) 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 16.0 17.0 18.0 19.0 20.0 21. 1 67 0 25 2 50 3 70 1 ACAI ANTHOCYANINS #3 [mod Acai UV_ V 2 Acai-Anthocyanidins #5 [modifi e Acai UV_ V mAU min 2 1 WVL:520 nm A B3 1 2 4 Time (min) 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 16.0 17.0 18.0 19.0 20.0 21. 1 67 0 25 2 50 3 70 1 ACAI ANTHOCYANINS #3 [mod Acai UV_ V 2 Acai-Anthocyanidins #5 [modifi e Acai UV_ V mAU min 2 1 WVL:520 nm A B3 1 2 4 Time (min)

PAGE 56

44 Figure 3-1 shows a typical HPLC chroma togram of anthocyanin 3-glycosides extracted from aai that when hydrolyzed yielded cyanidin (1,040 mg/L pulp) and pelargonidin (74 mg/L pulp) as the only compounds detected. Spectroscopic analysis before and after acid hydrolysis confirme d the presence of each anthocyanidin and tentative identification of a monoglycoside attached to the C-3 position, presumably a glucose derivative, was made based on A440/ Amax ratios (~33%) as described by Hong and Wrolstad (1990). Presence of hydroxy-substitu ted aromatic acids attached to the glycoside (acylated moieties) was not found for either compound, as shown by the absence of their typical absorption spectrum in the 310-340 nm range. The predominant polyphenolics present in aai pulp were ferulic acid > epicatechin > p -hydroxy benzoic acid > gallic acid > protoc atechuic acid > (+)-cat echin > ellagic acid > vanillic acid > p -coumaric acid at concentrations th at ranged from 17 to 212 mg/L as reported in Table 3-1. Additionally, five com pounds were identified with spectroscopic characteristics comparable with gallic acid a nd were tentatively identified as gallotannins, while one compound shared spectroscopy sim ilarities with ellagic acid and was tentatively identified as an ellagic acid gl ycoside (Table 3-1; Figure 3-2). Additional confirmation of these compounds was made following acid hydrolysis, as these compounds were no longer detected and a corr esponding increase in eith er gallic acid or ellagic acid concentrations was observed. Antioxidant Capacity Aai pulp was found to have a rela tively high antioxidant content (48.6 mol Trolox equivalents/mL) with respect to othe r anthocyanin-rich fruits such as highbush blueberries (4.631.1 mol TE/g) (Ehlenfeldt and Prior, 2001), strawberries (18.3-22.9)

PAGE 57

45 (Kalt et al., 1999), raspberries (19.2-22.6) (Kalt et al., 1999), blackberries (13.7-25.1) (Wang and Lin, 2000), cranberries (8.20-145) (Wang and Stretch, 2001), and muscadine grape juice (18.2-26.7) (Talcott et al., 2003). Table 3-1. Anthocyanin and polyphenolic co ntent (mg/L fresh pulp) of aai ( Euterpe oleracea Mart.). Fractionation of aai phytochemicals based on solubility and affinity characteristics was conducted to determine the distribu tion of antioxidant compounds among the isolates. Similar antioxidant conten t was observed for the whole pulp, C18 retained phenolics (phenolic acids and anthocyanins), and the anthocyanins alone while ethyl acetate-soluble phenolics, the liphophilic, and C18 non-retained isolates had appreciably Polyphenolic Content (mg/ L fresh pulp) Cyanidin 3-glucoside 1,04058.2 Pelargonidin 3-glucoside 74.42.90 Ferulic acid 2125.29 (-)-Epicatechin 1293.28 p -Hydroxy benzoic acid 80.52.00 Gallic acid 64.51.64 Protocatechuic acid 64.41.64 (+)-Catechin 60.80.98 Ellagic acid 55.41.39 Vanillic acid 33.21.39 p -Coumaric acid 17.11.23 Gallic acid derivative-1 47.31.40 Gallic acid derivative-2 18.40.89 Gallic acid derivative-3 17.31.25 Gallic acid derivative-4 13.30.96 Gallic acid derivative-5 3.90.18 Ellagic acid derivative 19.50.40

PAGE 58

46 lower contributions to the to tal antioxidant content (44, 8, and 1.2%, respectiv ely) (Figure 3-3). Results indicated that when ethy l acetate-soluble phenolics and anthocyanin fractions were evaluated alone for antioxidant capacity, thei r sum was higher that values obtained for the total C18 bound polyphenolics. Although th ese fractions were not recombined again for analysis, there is indication that physical and/or chemical interactions among constituents in these fractions unfavorably impacted radicalscavenging properties. Previous studies have demonstrated antagonistic interactions between polyphenolics such as quercetin a nd caffeic acid (How ard et al., 2000), or cyanidin in combination with catechin and el lagic acid (Meyer et al., 1998) all of which are present in aai. However, the effectivene ss of an antioxidant compound is generally dependent on the polarity of the testing system the nature of the radical, and type of substrate protected by the antioxidant (Prior et al., 2003). The diversity of antioxidant polyphenolics present in aai create a comple x matrix from which evaluations can be made, but it was apparent that anthocyanins were the predominate contributors to the antioxidant capacity and their presence w ith other polyphenolics resulted in an underestimation of the overall antioxidant capacity of aai pulp. Color Stability as Affected by Hydrogen Peroxide and Temperature The anthocyanin color stability of aai wa s assessed spectrophotometrically in the presence of hydrogen peroxide (0 and 30 mmo l/L) at 10, 20, and 30 C, and compared to the five other anthocyanins sources. Regressi on analysis was used to determine adequacy of the model describing kinetics of color degradation over time, and confirmed that degradation rates followed first order kine tics (P<0.05) in agreement with previous reports (Ozkan et al., 2002; Taoukis et al., 1997). Degradatio n rate constants ( 1) and half

PAGE 59

47 life (t1/2) values of anthocyanin color were calcu lated according to Taoukis et al. (1997): ln At / Ao = 1 time, and t1/2 = ln 0.5 / 1; where Ao is the initial absorbance value, and At is the absorbance value at a given time. Incr ements in storage temperature allowed for calculation of a temperature quotient (Q10) for each anthocyanin source (Ozkan et al., 2002), which is presented in Table 3-2. Figure 3-3. Antioxidant capaci ty of different phytoche mical fractions (whole pulp, liphophilic extract, C18 bound polyphenolics, ethyl acetate-soluble phenolics, anthocyanins, and C18 non-retained) of aai ( Euterpe oleracea Mart.). Bars represent standard error of the mean (n=6). Antioxi dant capacity quantified using Trolox equivalents (TE). Fraction Antioxidant capacity (mol TE/mL fresh pulp) 0 10 20 30 40 50 60 C 18 non-retained Whole pulp LypophilicC 18 bound phenolics Ethyl acetate soluble phenolics Anthocyanins

PAGE 60

48 Table 3-2. The effect of hydrogen peroxide (30 mmol/L) and temperature on kinetic parameters of color degradation fo r different anthocyanin sources. 1 1 t1/2 2 Q10 3 Pigment 10 C 20 C 30 C 10 C 20 C 30 C 10-20 C 20-30 C Aai 7.7 11.3 13.9 90 c4 61 c 50 c 1.5 1.2 Hibiscus 6.3 9.8 11.7 110 b 71 b 59 b 1.6 1.2 Purple Potato 5.8 9.5 12.4 120 b 73 b 56 b 1.6 1.3 Black Carrot 8.7 14.4 18.7 80 d 48 d 37 d 1.7 1.3 Red Cabbage 8.4 12.5 15.9 83 d 55 c 44 d 1.5 1.3 Red Grape 2.2 4.2 5.6 315 a 165 a 124 a 1.9 1.3 1Reaction rate constant ( 1 103, min-1). 2Half-life (min) of initial absorbance value for each pigment source. 3Temperature dependence quotients of color degradation as affected by increments in reaction temperature from 10 to 20 C, and 20 to 30 C, respectively. 4Values with similar letters within colu mns of each reaction temperature are not significantly different (LSD test, P<0.05). Compared to aai and the other anthocya nin sources, greater color stability (t1/2) was observed for red grape anthocyanins, resu lts that were attributed to their high polymeric anthocyanin content (Table 3-3) The predominantly acylated anthocyanins from black carrot and red cabbage displayed re duced color stability at each temperature when compared to the non-acylated aai and hibiscus anthocyanins and to the acylated anthocyanins from purple sweet potato. Differe nces in half-life values (Y) between red grape and other anthocyanin sources increased linearly with reaction temperature (Y= m*Temperature, R2=0.99), with similar values obtained for these differences for hibiscus, purple potato, and aai (m=0.135), and more pronounced for red cabbage and black carrot (m=0.2). Increasing the reaction temperature from 10 to 20 C significantly increased color degradation (Q10 ~ 1.6) for all sources except re d grape, were a 1.9-fold increase was observed. This was in contrast to the relatively slower ra te of color loss (Q10 = 1.3)

PAGE 61

49 observed for all the anthocyanin sources wh en the reaction temperature was increased from 20 to 30 C. Rates of anthocyanin degradation duri ng storage significan tly varied among sources and likely occurred due to factors such as varying molar ratios between reactants (anthocyanins and/or polyphenolics with peroxide), non-anthocyanin polyphenolic concentration, secondary free ra dical formation, or other ox idative reactions such as o quinone formation involving phenolics and anth ocyanins (Boulton, 2001; Ozkan et al., 2002; Talcott et al., 2003a). Results of this st udy indicate that acylated anthocyanins were not more stable than their non-acylated counterparts in the presence of hydrogen peroxide. This observation may have been influenced by the presence of additional nonanthocyanin polyphenolics in solution, emphasi zing the importance of conducting color stability evaluations with pigment sources used industrially. Thes e polyphenolics also form copigment complexes with anthocyanins, resulting in a more in tense color that may be several folds higher in color intensity due to hyperchromic and bathochromic spectroscopic shifts. Therefore, color comp arisons among diverse pigments sources are difficult since molar ratios between reactants (hydrogen peroxide and anthocyanins) vary between sources for a given color intensity. De spite these varying ratio s, industrial use of anthocyanins is based on color shade and inte nsity and their relative color stability under oxidizing conditions is very important for many food and beverage applications. Color Stability in the Presence of As corbic Acid and Natural Cofactors A primary concern regarding the use of an thocyanins in the food industry is their inherent instability during processing and st orage. Moreover, a grow ing trend in the food industry is to fortify juices with various phytonutrients for both quality and healthpromoting benefits. Ascorbic acid is among th e most common fortificants used for this

PAGE 62

50 purpose; however, when present together with anthocyanins, their combination will lead to mutual degradation that causes the lo ss of nutrients and color stability during processing and storage. Therefore a need ex ists to find an inexpensive and stable anthocyanin pigment that possesses a dive rsity of functional properties for food and nutraceutical applications. The stability of aai anthocyanins was evaluated in the presence of ascorbic acid (0 and 450 mg /L) under accelerated storage conditions (37C) using an in vitro model system as compar ed to those of other common anthocyanin sources (hibiscus, black carrot, red cabbage red grape, and purple sweet potato). A further examination of how naturally occurri ng cofactors affect color stability within a given pigment source was also investigated. Differences in spectroscopic properties a nd color attributes among in vitro juice model systems prepared with the six anthocya nin sources were initially observed (Table 3-3). Despite model systems with the same initial color value (abs orbance value of 1.5), color differences were apparent and due to the diversity of ring substitutions (hydroxy, sugar, or acyl-linked organic acids) among s ources. As previously discussed, the nature of polyphenolic cofactors and their relative mo lar ratio to anthocyanin concentration were also influential on color characteristics of each source. Isolation of polyphenolic cofactors revealed not only the appreciable difference in color exhibited by each pigment, but also their specific role in anthocyanin stabili ty. Red grape anthocyanins had the largest hyperchromic shift (49%), followed by purple potato (35%), hibisc us and black carrot (19.5% on average), and aai and red cabbage (7% on average) due to the presence of these native cofactors with a slight bathochr omic shift in wavelength observed for aai

PAGE 63

51 and red grape anthocyanins. These spectroscopic features translated into a more intense colored solution and were influential on overall color stability. Table 3-3. Percent monomeric anthocyanins and CI E color attributes of a juice model system (pH 3.5, 100 mg/L sucrose) prepar ed with different pigment sources, along with their correspondent hyperchromic and bathochromic shifts due to the presence of naturally occurr ing polyphenolic cofactors. Pigment % Monomeric anthocyanins Chroma Huemax 1 Hyperchromic shift 2 Bathochromic shift 2 Acai 76.2 c3 20.1 18.2515 nm6% 1 nm Hibiscus 80.3 b 31.2 35.2521 nm19% 0 nm Purple Potato 77.5 c 23.1 13.6526 nm35% 0 nm Black Carrot 77.8 c 22.9 10.2521 nm20% 0 nm Red Cabbage 92.2 a 19.8 -13.9526 nm8% 0 nm Red Grape 58.1 d 17.2 6.1 528 nm49% 2 nm 1Wavelength of maximum absorption for each pigment source. 2Difference in absorbance between anthocyanin solutions with and without naturally oc curring polyphenolic cofactors. 3Values with similar letters within co lumns of each reaction temperature are not significantly differe nt (LSD test, P<0.05). Results from objective color analysis conc luded that chroma values only differed slightly within anthocyanin sources in accordan ce with those observed in previous studies (Stintzing et al., 2002; Giusti and Wrolstad, 2003), except for hibiscus, which had an appreciably higher value than other sources. Hue angles significantly differed among pigment sources due to various ring substitutions and were generally lower for acylated anthocyanins (Table 3-3), gi ving the later anthocyanins a characteristic intense purple color in solution that corresponded to their longer wave length of maximum absorbance. Red grape anthocyanins were a notable excep tion due to its high polymeric anthocyanin content in relation to the othe r sources. Polymeric anthocya nins typically have greater color stability over their mono meric counterparts (Es-Safi et al., 2002; Malien-Aubert et

PAGE 64

52 al., 2002; Mateus et al., 2003) and the high content in red grape (58%) appreciably influenced its color stability during storage. The red grape extract used in this study was obtained as a by-product of the wine i ndustry, and may contain anthocyanins polymerized with oligomeric flavanols and/or acetaldehyde (Es-Safi et al., 1999; Eiro and Heinonen, 2002; Es-Safi et al., 2002; Mateus et al., 2003;) which gives this extract remarkable color and storage stability. Table 3-4. The effect of ascorbic acid (0 and 450 mg/L) a nd naturally occurring polyphenolic cofactors (presence or ab sence) on kinetic parameters of anthocyanin degradatio n during storage at 37 C of in vitro models systems (pH 3.5, 100 mg/L sucrose) prepared with different pigment sources. No Ascorbic Acid Ascorbic Acid (450 mg/L) With Cofactors No Cofactors With Cofactors No Cofactors Pigment 1 1 t1/2 2 1 t1/2 1 t1/2 1 t1/2 Acai 1.8 385 d3 2.2 319 d*4 55 13 d 49 14 c* Hibiscus 2.2 315 e 5.3 131 f* 19 37 c 60 11 c* Purple Potato 0.8 866 b 2.0 355 c* 18 38 c 52 13 c* Black Carrot 1.3 533 c 1.4 486 b* 52 13 d 60 12 e* Red Cabbage 0.3 2,450 a 0.6 1,150 a* 14 50 b 34 20 b* Red Grape 1.3 533 c 2.9 243 e* 11 62 a 23 30 a* 1Reaction rate constants ( 1 103, hours-1). 2Half-life (hours) of initial anthocyanin content. 3Values with similar letters within colu mns are not significantly different (LSD test, P<0.05). 4Means with an asterisk (*) for each pigment source indicate a significant effect (LSD test, P<0.05) due to presence of naturally occurr ing cofactors when compared to the same treatment with an equivalent ascorbic acid content. Regression analysis found that anthoc yanins under the accelerated storage conditions of the in vitro mode ls, with and without native cofactors, followed first order kinetics (P<0.05). Kinetic parameters were calculated as previously described, with anthocyanin content used as the independent variable. Ac ylated anthocyanin sources along with those from red grape were found to be more stable than their non-acylated

PAGE 65

53 counterparts, independent of ascorbic acid co ntent. Naturally occurring cofactors were shown to be key elements to decrease anthoc yanin degradation during storage, an effect that was more pronounced for non-acylated anthocyanin sources. Half-life evaluation of pigment stability re vealed that acylated anthocyanin sources generally had increased stability (t1/2 >823 h) with respect to non-acylated sources in the absence of ascorbic acid. A notable ex ception was black carrot anthocyanins (t1/2 =515 h), which showed reduced stability with resp ect to that of nonacylated red grape anthocyanins (t1/2 =540 h). By comparison, the red gr ape anthocyanins had reduced stability in the absence of ascorbic acid, especially in relation to the high stability observed against hydrogen peroxide, yet in the presence of ascorbic acid the stability was again the highest among anthocyanin sources. Red grape anthocyanins (t1/2 = 62 h) were the most stable compounds in the presence of ascorbic acid followed by red cabbage (t1/2 =50 h), hibiscus and purple potato (t1/2 =37 h), and lastly aai and black carrot (t1/2 =13 h). Overall, anthocyanin degradation was si gnificantly increased in the presence of ascorbic acid as compared to non-fortifie d controls, generally having a more pronounced effect on acylated anthocyanin sources (40 to 46-fold) than for non-acylated sources (8.4 to 30-fold). A notable exception was purple potato anthocyanins, where ascorbic acid increased color degradation by 23-fold. Red gr ape and hibiscus anthocyanins exhibited the smallest change with only a 8-fold increa se in degradation rates. Naturally occurring polyphenolic cofactors were found to signifi cantly increase anthoc yanin retention by up to 2.4-fold in the absence of ascorbic acid, an effect that was less pronounced for aai (1.2-fold) and black carrot (1.1-fold) anthocya nins. A similar protecti ve effect conferred by intermolecular copigmentation was observed in the presence of ascorbic acid for black

PAGE 66

54 carrot, aai and red grap e, yet additional increments in th is protective effect was observed for hibiscus (+0.9-fold) and both purple potato and red cabbage (+0.4-fold). The increased stability of acylated anthocyanins with respect to non-acylated pigment sources was likely related to the na tural synthesis of acyla ted organic acids and diversity of glycosidic linkage s in relation to these acylated moieties (Rodriguez-Saona et al., 1999; Boulton, 2001; Giusti and Wrolstad, 2003;). The aromatic or aliphatic acyl groups covalently bound to these anthocyani ns were shown to stack on the planar, polarizable nuclei of the anthocyanin, pr otecting the pyrylium nucleus from the nucleophilic attack of water at carbon 2 (Rodriguez-Saona et al., 1999; Boulton, 2001). Red cabbage and purple potato extracts typi cally contain cinnam ic acid derivatives diacylated to their anthocyanins that can simultaneously stack on both faces of the anthocyanin chromophore in a sandwich-type complex and thus offer greater color stability, while black carrot s contain only monoacylated moieties that can only protect one face of the pyrylium ring (Mazza and Broui llard, 1990; Rodriguez-Saona et al., 1999; Boulton, 2001; Malien-Aubert et al., 2001; Stintzing et al ., 2002; Es-Safi et al., 2002; Giusti and Wrolstad, 2003). The observed differe nces in stability between the various sources of acylated anthocyanins in this st udy were likely related to the nature, number, and position of these substitutions. For a given set of pH conditions, intramol ecular copigmentation exerts a protective effect against anthocyanin degradation by k eeping a larger proporti on in their flavylium ion forms. Consequently, formation of interm olecular complexes will also take place with these acylated anthocyanins and thus give an additional protective effect against color degradation. Results of this study also dem onstrated and confirmed that both forms of

PAGE 67

55 copigmentation (intraand intermolecular) c ooperatively acted to prevent anthocyanin color degradation, as demonstrated by simila r pigment half-life va lues (12.5 days) in black carrot and purple potato after removal of naturall y occurring cofactors. The stabilization effect conferred by intermolecular copigmentation has been attributed to hydrophobic interactions between anthocyanins and polyphenolic compounds, consequently protecting the pi gment from further polymerization and degradation reactions (Mazza and Brouillard, 1990; Rodri guez-Saona et al., 1999; EsSafi et al., 1999; Boulton, 2001; Eiro and He inonen, 2002; Es-Safi et al., 2002;). Previous studies have shown that not only ascorbic acid but also its degradation by-products, including those from carbohydrates such as fu rfural and other alde hydes, can participate in anthocyanin degradation during proces sing or storage (Eir o and Heinonen, 2002). Intermolecular copigmentation exerts a protec tive effect on anthocyanin degradation as cofactors compete with anthocyanins and preferentially react in the condensation reactions (Es-Safi et al., 1999; Malien-Aubert et al., 2001; Es-Safi et al., 2002). The increased protection observed for a specific pigment source due to the presence of cofactors is most likely related to the type and content of polyphe nolics present, as a higher copigment/pigment molar ratio could have occurred for a determined source. Moreover, specific polyphenolics or classifi cations of polyphenolics are more likely to form stable intermolecular complexes with anthocyanins than others (Boulton, 2001; Malien-Aubert et al., 2001;; Eiro and Heinonen, 2002). Conclusion Characterization of the major polypheno lic compounds present in aai and their contribution to the antioxidant capacity was determined for the first time. The effect of exogenously added cofactors on color enha ncement and stability was previously

PAGE 68

56 evaluated in many food systems containing is olated anthocyanins, model juices, and wine, yet the effect of naturally occurring co factors on color stability was not previously investigated prior to this st udy. The stability of aai anthocyanins as a new source of anthocyanin pigments was also established a nd can be used to determine application and functional properties of aai in a variety of food and nutraceutical products.

PAGE 69

57 CHAPTER 4 STABILITY OF COPIGMENTED ANTHOCYANINS AND ASCORBIC ACID IN MUSCADINE GRAPE JUICE PROCESSED BY HIGH HYDROSTATIC PRESSURE Introduction Muscadine grapes ( Vitis rotundifolia ) are the predominant gr ape variety grown in the southern U.S. with excellent potentia l for commercial expansion and value-added development. Deleterious changes in colo r and phytochemicals appreciably affect muscadine grape products during processing and storage, as they do with other anthocyanin containing juices, and are an impediment to future market development. Processing technologies and/or strategies that could s ubstantially improve quality attributes of these products ar e consequently vital for the eco nomic growth of this crop. High hydrostatic pressure (HHP) is a prom ising alternative to traditional thermal pasteurization technologies and may lessen detrimental effects to thermolabile phytonutrients (Gomez et al., 1996; Sun et al ., 2002; Poei-Langston and Wrolstad, 1981). However, a downside of this technology is the presence and/or activation of residual enzymes, such as polyphenol oxidase ( PPO), lipoxygenase, and peroxidase, during processing and storage, which may be part ially responsible for oxidative degradation reactions. Quality and phytochemical deterior ation due to enzyme action may be further complicated due to interactions between ant hocyanins and ascorbic acid, when the latter compound is present in the juice or is exte rnally added, resulti ng in their mutual destruction (Poei-Langston and Wrolstad, 1981; Garcia-Viguera and Bridle, 1999 Garzon and Wrolstad, 2002).

PAGE 70

58 A previous study with HHP and muscad ine grape juice demonstrated that phytochemical losses caused by processing were presumably due to the activation of residual oxidases during juice ex traction and/or autoxidative m echanisms resulting in cooxidation of anthocyanins and ascorbic acid (Talcott et al., 2003a). Their study utilized a commercially available polyphenolic extr act from rosemary aimed to reduce phytonutrient degradation through copigmentation, yet the overal l quality of the juice was adversely impacted presumably due to copious amounts of additional polyphenolics present in the extract that were substrates for oxidative enzymes. A goal of the current study was to confirm the role of enzyme s in phytonutrient degradation during HHP processing and to establish a potential reme diation strategy usi ng partially purified anthocyanin cofactors from two plant sour ces. Addition of in dividual polyphenolic cofactors has been reported to increase anthocyanin stability during processing and storage (Malien-Aubert et al ., 2001; Eiro and Heinonen, 2002), and their effectiveness in forming intermolecular linkages with anthocya nins has been linked to their specific structure and concentration. However, the us e of individual pol yphenolic cofactors may not be a feasible option for th e food industry and thus there is a need for a concentrated source of mixed anthocyanin stabilizing agen ts from natural sources. Evaluation of the effect of copigment addition during processi ng and storage regimes, especially in the presence of residual oxidase enzymes, is important for determining their efficacy in preventing phytonutrient degradation and their interaction with other food components. The objective of this study was to assess the phytochemical stability of muscadine grape juice ( Vitis rotundifolia ) processed by HHP and fortifie d with ascorbic acid. The effect of exogenously added polyphenolic cofactors purified from rosemary ( Rosmarinus

PAGE 71

59 officinalis ) and thyme ( Thymus vulgaris ) was also investigated as a means to improve overall phytochemical stability. The role of resi dual PPO activity was also investigated to gain knowledge on the mode of deterioration and potential so lutions for increased storage stability of fruit juices containing anthocyanins. Materials and Methods Materials and Processing PPO activity during juice extraction Muscadine grapes (cv. Noble) were obtained from a local grower in central Florida and held frozen (-20C) until needed. Grapes were rapidly thawed by placing them under running tap water and hand-sorted for uni formity of ripeness. Response Surface Methodology (RSM) was used to determine th e initial PPO activity of muscadine grape juice under different manual juice extrac tion procedures (0-24 min, 46-74 C). PPO activity and browning index (BI) were used as the dependent variables in the experimental design that was repeated tw o times, and each study required 11 experiments with 4 factorial points and 4 star points to form a central composite design, and 3 center points for replication (Kim et al., 2001). Experimental data were analyzed by regression analysis to determine the adequacy of the mathematical models. The RSM models were used to select the juice extr action conditions for the subseq uent study that investigated the HHP-induced PPO activation. Residual PPO activity in the juice was determined according to a modified polarographic method described by Kader et al. (1997) using a YSI 5300 oxygen monitor (Yellow Springs, OH) e quipped with a Clark-type electrode in a 3.1 mL jacketed cell at 35 C. The reaction was started when 0.2 mL of 0.12M catechin was added to 2.8 mL of grape juice mixed w ith 1 mL of 0.1M phosphate buffer at pH 3.5. The assay was carried out in air-saturated solu tions agitated with a magnetic stirrer and

PAGE 72

60 the electrode calibrated using ai r-saturated water (230 nmol O2 / ml H2O). Enzymatic activity was determined from the linear portion of the oxygen consumption curve, reported as nmoles of oxygen consumed per s econd (nkat), and expressed as a percentage of the control juice (100% activity) that was extracted at 25 C without a heating timetemperature regime. BI was used as an indirect method to monitor anthocyanin degradation during the different juice extraction procedures and was calculated as the ratio of absorbance values obtained at 420 and 520 nm (Buglione and Lozano, 2001). Juice extraction and processing Rosemary ( Rosmarinus officinalis L. ) and thyme ( Thymus vulgaris L. ) were obtained from a local market and the biomass exhaustively extracted with water at 90 C for 8 hours. The resulting dark brown liquid was adjusted to pH 2.0 with 1M HCl and centrifuged at 17,000 rpm for 15 min to remove insoluble matter. Polyphenolics were subsequently concentrated and purified using C18 Sep-Pak Vac 20 cc mini-columns (Water Corporation, Mass., USA). Polar consti tuents were removed with acidified water (0.01% v/v HCl) and polyphenol ic compounds subsequently eluted with methanol (0.01% v/v HCl), solvent that was later evap orated under reduced pressure at < 40 C. The resulting polyphenolic extr act was re-dissolved in a known volume of 0.1M citric acid solution. Based on the initial RSM evaluations, mus cadine grapes were crushed and heated in an open steam kettle to 46 C for 11 mi n to retain enzymatic activity (115% PPO activity) and facilitate juice extraction dur ing pressing in a hydraulic basket press (ProsperoÂ’s Equipment, Cort, NY). Juice was immediately filtered first through cheesecloth followed by vacuum filtration thr ough a 1cm bed of diatomaceous earth. The juice was then divided into two portions for copigmentation (0 and 100 cofactor-to-

PAGE 73

61 anthocyanin molar ratio) with either rosemary or thyme extracts. Ratios were adjusted considering the molar concentration of tota l phenolics in rosemary and thyme extracts (0.52 and 0.72 M gallic acid equivalents, resp ectively) divided by th e total anthocyanins in muscadine grape juice (8.08 mM cyanidin 3-glucoside equivalents). Juices at each cofactor concentration were again divided and half fortified w ith ca. 450 mg/L of ascorbic acid and compared to an equiva lent volume of citric acid buffer (pH 3.5, 0.1M) as the control. Sodium azide (50 mg/L) was added to retard microbi al growth throughout the analytical determinations. Treatments we re prepared for HHP by placing 8 mL juice portions into heat sealed plastic ampules and processed at 400 and 550 MPa for 15 min (Stansted Fluid Power, UK). Following HHP processing, ampules were subdivided and half stored under refrigerated conditions a nd analyzed within 48 hr of processing. The remaining half was stored in the dark at 25 C for 21 days. Chemical Analyses Initial anthocyanin content in the juic e was determined by the pH differential spectrophotometric method of Wrolstad (1976 ) and quantified as cyanidin 3-glucoside equivalents. Total soluble phenolic concentr ation in each cofactor source was measured using the Folin-Ciocalteu assa y (Talcott et al., 2000), and quantified as gallic acid equivalents. Individual ant hocyanin 3,5-diglycosides were quantified according to the HPLC conditions of Skrede et al. (2000) using a Dionex HPLC system and a PDA 100 diode array detector (Dionex Co., Sunnyvale CA). Compounds were separated on a 250 X 4.6 mm Supelcosil LC-18 column (Supelco, Bellefonte, PA) and quantified using standards of their re spective 3-glucoside forms (Pol yphenols Laboratories AS, Sandnes, Norway). Total ascorbic acid (the sum of Land dehydroascorbic acid) was quantified by reverse phase HPLC using modified chromatographic conditions described by

PAGE 74

62 Gkmen et al. (2000). Separation was performed on 3.9 x 150 mm Nova-Pak C18 column (Waters, Milford, MA), using KH2PO4 (0.2M, pH 2.4) as the mob ile phase at a flow rate of 0.5 mL/min with UV detection at 254 nm. Pr ior to ascorbic acid analysis, all samples were passed thorough pre-conditioned Waters C18 Sep-Pak cartridges (Waters, Milford, MA) to remove neutral polyphenolics. Afte r discarding the first mL, samples were collected, and dithiothreitol (8 mM) subsequently added as a pre-column reductant. Samples were then stored in the dark for 120 min to convert dehydroascorbic acid to Lascorbic acid. After complete conversion, samples were filtered through a 0.45m PTFE filter (Millipore, Bedford, MA) and analyzed for total ascorbic acid. Antioxidant capacity was determined using the oxygen radical ab sorbance capacity (ORAC) assay against a standard of Trolox as descri bed by Talcott et al. (2003b). Statistical Analysis Data represents the mean and standard e rror of juices analy zed as a 3 x 3 x 2 factorial comparing three pr ocessing conditions (unprocessed, 400 MPa, and 550 MPa), three copigmentation treatments (none, rosemary or thyme extracts), and the presence or absence of ascorbic acid (0 or 450 mg/L). P hytonutrient and antioxidant losses were also monitored after 21 days of storage following HHP processing. Linear regression, Pearson correlations and analysis of variance were conducted usi ng JMP software (SAS, Cary, NC), with mean separation pe rformed using the LSD test (P <0.05). All experiments were randomized and conducted in triplicate. Results and Discussion Muscadine grape juice was previously es tablished to demonstrate appreciable phytochemical losses following HHP processing a pparently due to the activity of residual oxidases, presumably PPO, and/or other aut oxidative reactions (Tal cott et al., 2003a). In

PAGE 75

63 that study the addition of anthocyanin cofactor s from a commercial rosemary extract was proposed as an approach to reduce phytonutri ent degradation, yet negatively impacted juice quality characteristics. In the pr esent study, water-soluble polyphenolic extracts from rosemary and thyme (partially purifi ed by reverse phase chromatography) were evaluated as a means to stab ilize the color and phytonutrien t content of ascorbic acid fortified grape juice. Utiliz ation of the proposed cofactor s and extraction/purification regime was selected as a strategy to favorably enhance the process and storage stability of anthocyanin-containing fruit jui ces with residual PPO activity. Initial Effects of Copigmentati on in Muscadine Grape Juice Copigmentation increased visual color of the juice as evidenced in a decline in hue angle (data not shown), which a ppear to the eye as a more intense red color of the grape juice. A preliminary study indicated that the color intensity of muscadine juice (measured as hyperchromic shift) could be increased up to 377 and 490% by the addition of thyme and rosemary extracts respectively at a 400 c opigmentation ratio. However, this level was deemed impractical for commercial use due to potentially adverse flavor characteristics and increased oxidative susceptibility. Cons equently, a 1:100 ratio was selected for each copigment source in the processing studies as a means to increase phytochemical stability. At this ratio both treatments pres ented similar hyperchromic shifts, however thyme extracts presented a significantly higher bathochromic shift in absorbance (25 nm) and also resulted in better anthocyanin stabil ity after HHP processing and ascorbic acid fortification. Copigmentation also served to mask detrimental color changes that occurred during HHP processing, as only slight ch anges were subjectively observed for copigmented juices when compared to apprec iable losses in control juices. Cofactor addition also increased initial antioxidant capacity of the juices by an average of 33 M

PAGE 76

64 Trolox equiv/mL (Figure 4-1), independently of the cofactor polyphenolic source and ascorbic acid content. PPO Activity as Affected by HHP Processing Residual PPO and/or autoxidative reac tions following HHP of muscadine grape juice have been proposed as potential mech anisms by which decreases in anthocyanin, ascorbic acid, and antioxidant capacity t ook place (Talcott et al ., 2003a). Additionally, increased oxidation may also occur under conditions of decreased volume such as pressurization, according to the Le Chatelie r principle (Butz and Tauscher, 2002). The action of oxidase enzymes in contributing to quality and anthocya nin deterioration has been demonstrated for several fruit syst ems (Wesche-Ebeling and Montgomery, 1990; Laminkara, 1995; Kader et al., 1997; Kader et al., 1999), and for muscadine grape PPO has specifically been shown to be a si gnificant factor infl uencing phytochemical degradation (Kader et al., 1999). These findi ngs justify additional studies evaluating the influence of HHP processing conditions on PPO activity, as well as their effects on phytochemical stability following pressu rization and throughout storage. Response Surface Methodology (RSM) was used to determine the pre-processing PPO activity under different hot-pressed times (0-24 min) and temperatures (46-74 C) for juice extraction. Therefore, under a known set of extraction c onditions the residual PPO activity could be estimated, and subse quently monitored following HHP processing. In the present study, muscadine grape juic e was extracted at 46C for 11 min (PPO activity of 115%; Figure 4-2) and used for subsequent expe riments with the purpose of evaluating the copigmentation treatments with initial PPO enzyme activity and during HHP induced activation.

PAGE 77

65 0 20 40 60 80 100 Unprocessed Antioxidant capacity ( M equiv/mL) 0 20 40 60 80 100 Control Rosemary Thyme (A) No Ascorbic AcidUnprocessed 550 MPa(B) Ascorbic Acid400 MPa 550 MPa 400 MPac a a f b e b b b c a a b d b b b e Figure 4-1. Antioxidant capacity of muscadine grape juice as affected by HHP processing and copigmentation with rosemary or thy me cofactors in the absence (A) or presence (B) of ascorbic acid (450 mg/L). Bars w ith different letters for each processing treatment are significantly different (LSD test, P<0.05).

PAGE 78

66 Figure 4-2. Polyphenoloxidase activity (A), a nd browning index (B) of muscadine grape ju ice as influenced by preheating time (0 -25 min) and temperature (46-74 C) prior to juice extraction. 20 30 40 50 60 70 80 0 5 10 15 20 50 55 60 65 70B r o w n i n g I n d e x ( % )T i m e ( m i n )T e m p e r a t u r e ( C ) 0 20 40 60 80 100 120 50 55 60 65 70 0 5 10 15 20P P O A c t i v i t y ( % )T e m p e r a t u r e ( C ) T i m e ( m i n ) A B

PAGE 79

67 Unprocessed PPO activity ( ) 0 100 200 300 Control Rosemary Thyme (A) No Ascorbic AcidUnprocessed 0 100 200 300 550 MPa(B) Ascorbic Acid400 MPa 550 MPa 400 MPaij jk k b f d h g i h ij jk a e c g f i Figure 4-3. Polyphenoloxidase activity in mus cadine grape juice as affected by HHP pro cessing and copigmentation with rosemary or thyme cofactors in the absence (A) or presence (B) of ascorbic acid (450 mg/L). Bars with diffe rent letters are significantly different (LSD test, P<0.05).

PAGE 80

68 For the control treatment, PPO activity was significantly increased following processing at 400 (3-fold) and 550 MPa (2.5-fold) as compared to the initial juice activity (Figure 4-3). Results were similar to those previously observed for PPO using both model and actual food systems (Poei-Langston and Wrolstad, 1981; Gomez et al., 1996; Sun et al., 2002), where pressure-induced enzyme activation took place during processing. Possible explanations for enzyme activation have been attributed to the effect of HHP on the hydrophobic and electrostatic bonds of proteins, which affects their secondary, tertiary, and quaternary stru ctures. Such conformational changes can cause enzyme activation by uncovering active sites and cons equently facilitating the interaction with their substrates. Copigmentation aided to decrease PPO pressure-induced activation by >1.5-fold, with cofactors from thyme genera lly being more effective than those from rosemary. A small increase in PPO activity was observed in the presence of ascorbic acid for both the control (~10%) a nd copigmented juices (~16%), an effect that was independent of the processing pressure (Figure 4-3B). Th ese increases may not have occurred due to actual PPO activation, but potentially occu rred due to increased oxygen consumption caused by ascorbic acid oxida tion during the enzyme assay conditions. Phytochemical Stability Following HHP Processing Two pressures (400 and 550 MPa) were selected for the HHP processing of muscadine grape juice, and its phytochemical content was compared to an unprocessed control following copigment and/or ascorbic acid addition (Table 4-1, Figures 4-4 and 45). In general, treatments processed at 400 MPa had greater phytonutrient losses due to the highly oxidative conditions that resulted from PPO activation during pressurization. Higher anthocyanin, ascorbic acid and antio xidant capacity retent ion was observed for

PAGE 81

69 juices containing thyme cofactors, followed by rosemary cofactors, a nd lastly the control juices. Table 4-1. The effect of rosemary and thym e cofactors at different anthocyanin-tocofactor molar ratios (1:0, 1:100), and ascorbic acid fortification (0, 450 mg/L) on the anthocyanin content of unprocessed (control) and high hydrostatic pressure processed (400, and 550 MPa) muscadine grape juice. No Ascorbic Added Ascorbic (450 mg/L) Copigment Molar Ratio1 Unprocessed HHP 400MPa HHP 550MPa Unprocessed HHP 400MPa HHP 550MPa Delphinidin Control 0 569c2 163c 303a 605 c 196b*3 363b* 3,5-diglucoside Rosemary 100 707b 728b 701a 1012 a 874a* 841a* (mg/L) Thyme 100 1187a 801a 794a 858 b* 905a* 896a* Cyanidin Control 0 227a 68c 127c 253 a 82c* 152c 3,5-diglucoside Rosemary 100 226a 227b 429a 157 b* 273b 516a (mg/L) Thyme 100 168a 400a 367b 311 a* 451a 501b* Petunidin Control 0 542c 163b 290c 580 b 195b* 348c 3,5-diglucoside Rosemary 100 726b 646a 626b 899 a* 777a 631b* (mg/L) Thyme 100 885a 670a 734a 778 b* 755a 828a* Peonidin Control 0 392b 115b 210b 414 b 138b* 251b 3,5-diglucoside Rosemary 100 524a 481a 531a 579 a 577a* 637a* (mg/L) Thyme 100 649a 467a 569a 547 a 526a 642a Malvidin Control 0 397b 123b 223c 444 a 148b* 267c 3,5-diglucoside Rosemary 100 536a 487a 579b 547 a 583a 694b* (mg/L) Thyme 100 424b 498a 744a 481 a 561a 839a* 1 Indicates the ratio between the molar concentr ation of total anthocyanins in muscadine grape juice (expressed as cyan idin 3-glucoside equivalents) and the molar concentration of each added polyphenolic cofactor (expr essed in gallic acid equivalents). 2 Values with similar letters within columns of each added cofactor are not significantly different (LSD test, P>0.05), and indicate the effect of an increase in the molar concentration of each cofactor. 3Means with an asterisk (* ) indicate a significant eff ect (LSD test, P<0.05) due to addition of ascorbic acid when compared to the same treatment without ascorbic acid.

PAGE 82

70 Total anthocyanin content (mg/L) 0 1000 2000 3000 4000 Control Rosemary Thyme (A) No Ascorbic AcidUnprocessed 400 MPa 550 MPai fg ab k gh j defg efg cde 0 1000 2000 3000 4000 Unprocessed 400 MPa 550 MPa(B) Ascorbic Acidhi bcd a k cdef bcd j abc ab Figure 4-4. Total anthocyanin cont ent of muscadine grape juice as affected by HHP processing and copigmentation with rosemary o r thyme cofactors in the absence (A) or presence (B) of ascorb ic acid (450 mg/L). Bars w ith different letters for each processing treatment are significan tly different (LSD test, P<0.05).

PAGE 83

71 Ascorbic acid content (mg/L) 0 100 200 300 400 Control Rosemary Thyme Unprocessed 400 MPa 550 MPab a a e d c d b a Figure 4-5. Total ascorbic acid content of muscadine grape juice as affected by HHP processing (400, and 550 MPa), and copigmentation with rosemary or thyme polyphenolic cofactors. Bars with different letters for each processing treatment are significantly differe nt (LSD test, P<0.05).

PAGE 84

72 Anthocyanin degradation was observed at both processing pressures but was appreciably higher at 400 MPa (70% loss) compared to 550 MPa (46% loss), and these decreases correlated to losse s of antioxidant capacity (r=0.89). Copigmentation was instrumental for improving anthocyanin re tention, maintaining on average 2,200 and 1,500 mg/L more total anthocyanins for tr eatments copigmented with thyme and rosemary, respectively (Figure 44). Individual ant hocyanins were also quantified (Table 1). Differences in their structures, mainly in the B-ring, influenced their reaction rates and consequently individual ant hocyanin losses. Previous stud ies have demonstrated that o diphenolic anthocyanins are more sus ceptible to degradation than the non o -diphenolic. This due to the presence of hydroxyl gr oup substitutions in the B-ring of the o -diphenolic anthocyanins, which are more susceptible to enzymatic degradati on reactions than the methoxy groups of non o -diphenolic anthocyanins (S arni-Machado et al., 1997). The degradation trends observed in this study are in accordance with previous studies with muscadine juice (Talcott et al., 2003a), wher e the 3,5-diglucosides of delphinidin and petunidin showed the greatest rates of degradation followe d by those of cyanidin, and finally peonidin and malvidin. These resu lts support the idea th at intermolecular copigmentation has a greater protective effect on non o -diphenolic anthocyanins rather than the o -diphenolic, which has been attributed to the presence of methoxy groups in the B-ring which facilitate copigmentation (M azza and Brouillard, 1990; Jackman and Smith, 1996). Due to the mutual destruction of ascorb ic acid and anthocyanins when present together in foods (Poei-Langston and Wrolst ad, 1981; Garcia-Viguera and Bridle, 1999; Garzon and Wrolstad, 2002; Talcott et al ., 2003a), it was anticipated that higher

PAGE 85

73 anthocyanin destruction might occur in th e presence of ascorbic acid. However, a combined anthocyanin protective effect due to copigment and ascorbic acid addition was observed prior and following HHP processing (Fi gure 4). Isolating the effect of ascorbic acid, an early oxidative protection was observed in the unp rocessed rosemarycopigmented grape juices, as indicated by hi gher anthocyanin reten tion (474 mg/L, when compared to the same treatment without ascorbic; Figures 4-4A and B). Following HHP processing at both pressures, anthocyani n degradation was decreased by 20% in comparison with treatments without ascorb ic for both the control and rosemarycopigmented juices (Figure 4B). This prot ective effect was less pronounced for juices containing thyme cofactors as anthocyani n degradation was only decreased by 13%. Kader et al. (1997, 1999) also observed that ascorbic acid could offer initial protection against anthocyanin oxidation in the pres ence of oxidative enzymes by reducing o quinones to their original phenolic moiety a nd preventing secondary reactions affecting phytochemical stability and quality deterioration. Copigmentation with polypheno lic cofactors was also eff ective in preventing initial ascorbic acid oxidation (Figure 45) and helped to increase th e initial antioxidant capacity (Figure 4-1) of the juices, an effect that was observed prior and following HHP. However, ascorbic acid retent ion was appreciably influen ced by the conditions of HHP processing with losses of 84% at 400 MPa compared to 18% at 550 MPa for control juices. On average, the polyphenolic copigmen ts reduced ascorbic acid degradation after processing by 32% at 400 MPa and by 20% at 550 MPa, with thyme cofactors again conferring the greatest protection.

PAGE 86

74 As previously mentioned copigment addi tion increased antioxidant capacity of unprocessed treatments by an average of 43% (~33 M Trolox equiv/mL), when compared to control juices (Figure 4-1) After pressurization, juices containing copigments and processed at 400 MPa a nd 550 MPa presented similar antioxidant capacity losses (19 M Trolox equiv/mL) and were not impacted by ascorbic acid addition. Control treatments presented lo sses of 45% and 21% at 400 and 550 MPa, respectively, values that were decreased to 26% and 15% in the juices containing ascorbic acid. Observed losses in antioxidant capacity were greater for treatments that presented higher rates of PPO activation, lik ely indicating that phenolic compounds are being consumed as enzyme substrates or being destroyed by oxida tion and thus lowering the ORAC values. Phytochemical Retention During Storage Most studies looking at the effects of HHP processing on phytonutrient stability only include evaluations after processing (C orwin and Shellhammer, 2002; Park et al., 2002; Boff et al., 2003), and do not consider their stability during the shelf-life of the product. The present study did not include a complete shelf-life evaluation due to the large number of treatments evaluated, however it included a one poi nt evaluation at 21 days after processing selected based on previ ous studies reporting th e shelf-life stability of ascorbic acid-anthocyanin systems (Gar cia-Viguera and Bridle, 1999; Boulton, 2001). Ascorbic acid was not detected in the juices after 21 days of storage independently of processing pressure regimes and presence of polyphenolic cofactors. Anthocyanin content also decreased by 28-34% th roughout storage for all treatments and pressurization conditions (Table 4-2). Oxid ation of individual an thocyanins followed

PAGE 87

75 similar rates during storage, despite their different B-ring substitutions, which was in agreement with Garzon and Wrolstad (2002). Other studies (Sarni-Machado et al., 1997; Kader et al., 2000; Boulton, 2001) have demons trated that anthocya nin degradation can be influenced by structural differences, but th e rapid rate of anthoc yanin degradation that occurred due to the high ly oxidative conditions created by HHP processing likely contributed to these observa tions. Antioxidant capacity was likewise appreciably decreased following storage (Table 4-2), losses were more pronounced for control treatments (28%) than copigmented juices (~13%). However, the rates antioxidant capacity degradation varied insignificantly with pressure processing and ascorbic acid fortification. The protective effect exerted by anth ocyanin-copigment complexes following pressurization was appreciably re duced during storage, observati ons that can be attributed to the increased rates and complexity of de gradative reactions occurring simultaneously to anthocyanins, ascorbic acid, and cofactor polyphenolics. Although rate constants can not be calculated on a single storage point, copigmentation did not appear to slow down the fast rates of anthocyanin degradation dur ing storage. However it aided to retain greater anthocyanin content when compar ed to control treatments as higher concentrations were observed for copigmente d treatments initially and immediately after HHP processing. Ascorbic acid oxidation may have also been a contributing factor promoting phytochemical degradation due to hydrogen peroxide fo rmation, leading to additional oxidative and polymeric degrad ative reactions. Furthermore, peroxide formation could contribute to activation of re sidual peroxidase that may further degrade phytochemicals (Garzon and Wrolstad, 2002) Additionally, by-products from the

PAGE 88

76 degradation of ascorbic acid and/or monos accharides, such as aldehydes, may contribute to anthocyanin degradation during storage. Interactions between anthocyanins and furfural derivatives have been previously investig ated (Kader et al ., 2000; Boulton, 2001; Es-Safi et al., 2002; Es-Safi et al., 1999) and were hypothesized to participate in condensation reactions yielding brown, polymerized pigments. Conclusions Blending a commercially available rosema ry extract with muscadine grape juice was previously reported to be deleterious to juice quality in the presence of ascorbic acid, yet by the use of the purification protocol used in this study an inverse effect was observed. The phytochemical extraction and isola tion procedures utili zed in this study resulted in concentrates that were lower in enzyme substrates and remained efficient as anthocyanin cofactors. Commercially availabl e botanical extracts may contain a variety of PPO substrates following harsh extraction procedures, solubilizi ng compounds such as cinnamic acid derivatives or tannins that may accentuate enzymatic oxidation within the food system. Results of this study demonstr ated that using an aqueous extract of rosemary and thyme followed by partial purification with C18 columns could decrease the destruction of both anthocyani ns and ascorbic acid during HHP processing, which created a highly oxidative environment due to re sidual PPO. Copigmentation was found to effectively stabilize anthocyanins and provided additional understanding of the mechanisms involved in phytochemical losses during pressurization a nd storage of fruit juices processed by HHP. Addition of polyphenolic cofactors also increases visual color and antioxidant capacity, important factors a ffecting consumer acceptability and potential health benefits of grape juice consumption.

PAGE 89

77 Table 4-2. The effect of rosemary and thyme cofactors at different molar ratios (1:0, 1:100), and ascorbic acid fortificatio n (0 and 450 mg/L) on the anthocyanin content and antioxidant capacity of hi gh hydrostatic pressure processed (400 and 550 MPa) muscadine grape juice af ter 21 days of storage at 24C. No Ascorbic Added Ascorbic (450 mg/L) Copigment Molar Ratio1 HHP 400MPa HHP 550MPa HHP 400MPa HHP 550MPa Delphinidin Control 0 97.8b2 207c 134 b 248 b 3,5-diglucoside Rosemary 100 497a 458b 610 a 574 a (mg/L) Thyme 100 491a 496a 591 a 612 a Cyanidin Control 0 42.9c 86c 54.8 c 106c 3,5-diglucoside Rosemary 100 155.3b 293a 182 b 352a (mg/L) Thyme 100 251.2a 186b 302 a 210b Petunidin Control 0 107b 193c 127 b 243c 3,5-diglucoside Rosemary 100 441a 368b 496 a 441b (mg/L) Thyme 100 437a 459a 492 a 578a Peonidin Control 0 73.4b 144b 90.1 b 172b 3,5-diglucoside Rosemary 100 283a 347a 361 a 435a (mg/L) Thyme 100 281a 380a 329 a 439a Malvidin Control 0 78.8b 159c 96.5 b 187c 3,5-diglucoside Rosemary 100 304a 386b 398 a 474b (mg/L) Thyme 100 313a 519a 391 a 573a Total Control 0 399b 790c 503 b 956c Anthocyanins Rosemary 100 1679a 1851b 2047 a 2275b (mg/L) Thyme 100 1773a 2040a 2105 a 2411a Antioxidant Control 0 16.5b 25.0b 23.4 b 32.8b Capacity Rosemary 100 51.3a 48.8a 50.6 a 50.7a ( M TE/ mL)3 Thyme 100 51.2a 49.3a 49.5 a 51.3a 1 Indicates the ratio between the molar concentr ation of total anthocyanins in muscadine grape juice (expressed as cyan idin 3-glucoside equivalents) and the molar concentration of each added polyphenolic cofactor (expr essed in gallic acid equivalents). 2 Values with similar letters within columns of each added cofactor are not significantly different (LSD test, P>0.05), and indicate the effect of an increase in the molar concentration of each cofactor. 3 Expressed as Trolox equivalents pe r mL of muscadine grape juice.

PAGE 90

78 CHAPTER 5 PASTEURIZATION AND QUALITY R ETENTION OF DENSE PHASE-CO2 PROCESSED MUSCADINE GRAPE JUICE Introduction Dense phase-CO2 processing (DP-CO2) is a continuous, non-thermal processing system for liquid foods that utilizes pressu re (< 90 MPa) in combination with carbon dioxide (CO2) to destroy microorganisms as a m eans of food preservation. Numerous studies have investigated the efficacy of pressurized CO2 to inactivate microorganisms and enzymes in batch or semi-continuous sy stems (Balaban et al ., 1991; Lin and Lin, 1993; Isenschmid et al., 1995; Ballestra et al., 1996; Wouters et al., 1998; Butz and Tauscher, 2002; Corwin and Shellhamme r, 2002; Park et al., 2003). However, information relating to deleterious ch anges in color and phytochemicals during processing and storage are generally lack ing, especially for continuous processing systems. Therefore to prove the effectiveness of DP-CO2 processing as a novel food processing technology, the microbial destruction, phytochemical stability, and sensory attributes of DP-CO2 processed muscadine grape jui ce was compared to a thermally pasteurized juice (75 C, 15 sec). Treatments were additionally evaluated following storage for 10 weeks at 4 C. A central composite design was initially conducted to determine the DP-CO2 processing parameters which achieved > 5 log reduction of aerobic microorganisms and yeast/molds. Result s of this study demonstrated differences between microbial, phytochemical, a nd sensory attributes of DP-CO2 and thermal

PAGE 91

79 processing, parameters that are of significant importance to assess th e benefits offered by novel processing technologies. Material and Methods Materials Muscadine grapes (cv. Noble) were obtained from a local grower in central Florida and held frozen (-20C) until needed. Fru it was rapidly thawed by placing them under running tap water and hand-sorted for uniformity of ripeness. Grapes were then crushed, heated to 75 C in an open steam kettle, and held for 2 min prior to juice extraction in a hydraulic basket press (ProsperoÂ’s Equipmen t, Cort, NY). Preliminary investigations demonstrated that this ju ice extraction method was suffi cient to inactivate oxidase enzymes. The juice was immediately filtere d through cheesecloth followed by vacuum filtration through a 1cm bed of diatomaceous earth. Processing Equipment The DP-CO2 system was constructed by APV (Chicago, IL) for Praxair (Chicago, IL) and provided as a gift to the University of Florida (Gaine sville, FL). The equipment is capable of continuously treati ng liquid foods with CO2 at pressures up to 69 MPa. The system mixes cooled, pressurized liquid CO 2 with a juice feed pressurized by its own pump (Figure 5-1). The mixture is then pr essurized by a reciprocating intensifier pump and subsequently fed to a holding tube (79.2 m, 0.635 cm ID) for the specified residence time, which is modified by changing the flow ra te of the mixture. An external heater and insulation electronically cont rols the temperature of th e system and upon exiting the holding tube the juice is depressurized by pa ssing through a backpre ssure valve and was finally collected into a holding tank.

PAGE 92

80 P Main Pump Juice stream Vacuum Heating system Pressurization chamber Processed juice CO2CO2tank Expansion valve Pump Pump Chiller Figure 5-1. Schematic diagram of the DP-CO2 processing equipment. For thermal processing, juice was pumped by a peristaltic pump (Cole Parmer, Chicago, IL) through a stainless steel tube into a temperatur e controlled water bath (Hart Scientific, American Fork, UT) were it was held at 75 C for 15 sec (HTST). The juice was then passed through a coo ling tube and chilled to 10 C whereby it was collected into sterile glass containers. Microbial Inactivation Study Preliminary investigations were conducted to determine the DP-CO2 parameters that could achieve >5 log reduction of aer obic microorganisms and yeasts/molds using Response Surface Methodology. Microbial counts (yeast and molds, and total aerobic microorganisms) were used as the dependent variables in the experimental design that was conducted in duplicate. Each study require d 11 experiments with 4 factorial points, 4 star points and 3 cente r points for replication. A high ini tial microbial load in the juice

PAGE 93

81 (8.1 x 106 CFU/mL of yeasts/molds, 1 x 105 CF U/mL of total aerobic microorganisms) was required and obtained by incubating the filtered juice for 4 days at 21 C. Juice was then subjected to DP-CO2 using different pr essures (1.2 to 40.2 MPa) and CO2 levels (0 to 15.7%) using a constant residence time (6.25 min) and temperature (30 C). Microbial inactivation was evaluated im mediately after processing. Microbial counts were made from triplica te samples of each processing treatment serially diluted (1 x 10-1 to 1 x 10-6) in duplicate by mixing 1 mL of each juice with 9 mL of sterile ButterfieldÂ’s buffer. Total plate c ounts were determined on aerobic count plates and yeast/mold plates (3M Petrifilm Microb iology Products, St. Paul, MN) by plating 0.1 mL of the dilutions onto the agar in triplicate and enumerated after 48 hr at 35 C and 72 hr at 24 C, respectively, according to the manufact urers guidelines. Experimental data were analyzed by regression analysis using JMP software (SAS, Cary, NC), fit to quadratic polynomial equations, and re sults used to select two DP-CO2 conditions for assessment of phytochemical stabi lity and sensory evaluation: ( i ) D-1 (34.5 MPa, 8% CO2) and ( ii ) D-2 (34.5 MPa, 16% CO2). Scanning electron microscopy was used to investigate changes in yeast microstructure due to DP-CO2. Yeast cells present in the grape juice before and after processing were treated according to the conditi ons described by Park et al. (2003) before being observed in the scanning electron mi croscope (Hitachi S-4000, Pleasanton, CA). Phytochemical and Microbial Stability Study Muscadine grape juice was divided into three equal portions for subsequent processing by the two DP-CO2 conditions (34.5 MPa at 8 or 16% CO2) and thermal pasteurization at 75 C for 15 sec. After processing, each juice was again divided into 3 proportions for assessment of microbial, phyto chemical and sensory characteristics.

PAGE 94

82 Samples for microbial and phytochemical analys is were immediately transferred into 20 mL screwed cap vials and stored at 4 C for 10 weeks, whereas samples for sensory analysis were transferred to sterile 4 L gl ass containers. Sodium azide (50 mg/L) was added to the samples used for phytochemical anal ysis in order to retard microbial growth. Physicochemical and Microbial Analyses Individual anthocyanin 3,5diglycosides were quantif ied by reverse phase HPLC using modified chromatographic conditions described in chapter 4. Compounds were separated on a 250 X 4.6 mm Supelcosil LC18 column (Supelco, Bellefonte, PA) and quantified using standards of their re spective 3-glucoside forms (Polyphenols Laboratories AS, Sandnes, Norway). Mobile phases consisted of 100% acetonitrile (Phase A) and water containing 10% acetic acid, 5% acetonitrile, 1% phosphoric acid (Phase B). A gradient solvent program ran pha se B from 100 to 88% in 8 min; 88-50% in 2 min, and held for 12 min at a flow ra te of 1.8 mL/min. Anthocyanins were characterized based on UV-VI S spectral interpretation fr om 200-600 nm, comparison to authentic standards (Polyphenols Laboratories AS, Sandnes, Norway), and identification additionally confirmed following acid hydrolysis into their re spective aglycones with 2N HCl in 50% v/v methanol for 60 min at 90 C. Antioxidant capacity was determined us ing the oxygen radical absorbance capacity (ORAC) assay and quantified using Trolox equi valents (TE) as desc ribed in chapter 3. Total soluble phenolic levels were measured using the Folin-Ciocalteu assay (Talcott et al., 2003a), and quantified as gallic acid equi valents. pH was measured using a Thermo Orion Model 720 pH meter (Thermo Electron Corp., New Haven, CT). Total titratable acidity was determined by potentiometric titra tion against 0.1N NaOH to pH 8.2 using an automatic titrator (Fisher Titrimeter II, Pitt sburgh, PA) and expressed in tartaric acid

PAGE 95

83 equivalents. CO2 content in the juices was determined using a Orion CO2 electrode (Thermo Electron Corp., New Haven, CT). Microbial counts throughout storage were determined as previously described. Sensory Evaluation Flavor, aroma, and color intensity of fr esh and processed juices were compared using a difference-from-control test. A randomiz ed complete block design was used and difference from control measurements were reco rded on a line scale with anchors at 0 and 10 that represented “no difference” to “extr emely different” in sensory attributes. Panelists compared the sensory attributes of the reference (f resh/unprocessed juice) with those presented by the hidden reference (fre sh juice) and the thermally or DP-CO2 processed juices. A 9-point he donic scale was also conducte d in order to compare the overall likeability of fresh (hidden reference) and processed juices. Before sensory analysis, all juices (fres h, DP-CO2 and thermally processed) were degassed in order to equalize carbonation leve ls by placing them in a 4 L sterile glass container on a hot plate with c ontinuous stirring for 4 h at 20 C. Juices were then served on a tray at room temperature in randomly numbered plastic cups with the reference sample placed at the center of the tray. A c up of deionized water and non-salted crackers were also provided to the panelists between evaluations. All sensory tests were performed at the University of Florida’s taste panel facility using sixty untrained panelists (31 females, 95% in the 18-44 age range). Statistical Analysis Data represents the mean and standard erro r of juices analyzed as a 3 x 9 factorial comparing three processing conditions (DP-CO 2 at 8% or 16% both at 34.5 MPa, or thermally pasteurized) evaluated at nine sampling points (unprocessed, processed, week

PAGE 96

84 1, 2, 3, 4 6 8 and 10). Linear regression, Pear son correlations and analysis of variance were conducted using JMP software (SAS, Ca ry, NC), with mean separation performed using the LSD test (P<0.05). All experiments were randomized and conducted in triplicate. Sensory data was recorded and analyzed using Compusense five (Compusense, Guelph, Ontario, Canada), and analysis of variance was conducted by using the TukeyÂ’s multiple comparisons method at the 5% significance level. Results and Discussion Microbial Inactivation Study The effects of DP-CO2 at various processing pr essures (0-40 MPa) and CO2 levels (0-18%) on the inactivation of yeasts/molds and total aerobic microorganisms can be observed in Figure 5-1. Results showed that although processing pressure was a significant factor affecting microbial inactivation, CO2 content was the processing parameter that had the major influence in microbial log re duction. This trend was also observed by Hong et al. (1999) which reporte d that microbial inactivation by DP-CO2 is governed essentially by the transfer rate and the penetration of car bon dioxide into cells, the effectiveness of which can be improved by increasing pressure, decreasing the pH of the suspension, and increasing the processing temperature. Studies investigating high hydrostatic pressure (HHP) pro cessing and super critical-CO2 batch systems have reported that microbial inac tivation is also highly de pendant on other processing parameters such as residence time and number of pulse cycles as well as the composition of the food (Balaban et al., 1991; Lin and Lin, 1993; Isenschmid et al., 1995; Ballestra et al., 1996; Wouters et al., 1998; Butz and Tauscher, 2002; Park et al., 2003). Results also demonstrated that under identical processing conditions, yeasts/molds were destroyed at significantly higher rates than aerobic microorganisms. Moreove r, the synergistic effect

PAGE 97

85 between pressure and CO2 was only observed for the in activation of yeast/molds. Microbial inactivation is highl y dependent on the type of microorganisms present in the food matrix due to distinct microbial ce ll microstructure and the diffusion of CO2 into the microbial cell (Ballestra et al., 1996 Wouters et al., 1998; Corwin and Shellhammer, 2002; Park et al., 2003). For instance vegeta tive cells, including yeasts and molds, are rather pressure and CO2 sensitive, whereas bacterial spores are more pressure resistant and thus need higher pressures for complete in activation. Park et al. (2003) showed that a combined treatment of carbonation and HHP at 500 MPa yielded an 8-log reduction Staphylococcus aureus Fusarium oxysporum and F. sporotrichioides while only a 4-log reduction was obtained for Bacillus subtilis Overall, microbial reduc tion is attributed to the fact that CO2 solubility increases directly propor tional with increments of processing pressure (Balaban et al., 1991; Park et al., 2003) which consequently affects the diffusion of CO2 into the microbial cell as well as the explosive decompression that occurs during DP-CO2 processing. Results of this optimization study were used to determine those DPCO2 conditions that achieved > 5 log re duction of aerobic microorganisms and yeast/molds that set the processing conditions of 34.5 MPa with 8% CO2 (D-1) and 16% CO2 (D-2). Micrographic observations aide d in elucidating the mechan ism of yeast destruction and concluded that explosive decompression of the microbial cell along with changes in cell membrane structure occurred during DP-CO2 (Figure 5-2). Conversely, heat pasteurized yeast cells still app eared round and pert but with slightly textured surfaces. Results also indicated that the number of decompressed cells was directly related to increments in processing CO2 levels. Previous investigatio ns have also demonstrated

PAGE 98

86 shown that microbial dest ruction by pressurized CO2 systems was based on gas dissolution inside a microbial cell that when rapidly decompressed to atmospheric pressure caused fatal damage to cell functi oning and explosive decompression of the cell (Balaban et al., 1991; Lin and Lin, 1993; Ballest ra et al., 1996; Park et al., 2003). Other theories concerning bacterial death by CO2 pressurization have indicated that the depressurization leads to l eakage of cellular components and changes in the cell membrane permeability which is responsible for cell damage and eventual microbial death (Lin and Lin, 1993; Isenschmid et al., 199 5; Park et al., 2003). Related studies have shown that removal of essential intrace llular substances such as phospholipids and hydrophobic compounds from cells or cell me mbranes play important roles as mechanisms of microbial inac tivation (Lin and Lin, 1993; Balle stra et al., 1996; Butz and Tauscher, 2002). Additionally, DP-CO2 effects biological systems by causing protein denaturation, lipid phase changes, and ruptur e of membranes inside the microbial cell (Lin and Lin, 1993; Ballestra et al., 1996; Park et al., 2003). Phytochemical and Microbial Stability Study Differences in phytochemical and antioxida nt levels were observed in muscadine grape juice as affected by processing methods and storage. Thermal pasteurization was found to be more detrimental to anthocyanins, soluble phenoli cs, and antioxidant capacity as compared with DP-CO2 and unprocessed juices. More over, enhanced oxidative stability and retention of antioxida nt compounds was observed for DP-CO2 processed juices throughout storage. However, microbial stability was only comparable to heatpasteurized juices for the first five weeks of storage.

PAGE 99

87 Figure 5-2. Inactivation of yeast/molds (Y&M ; A) and total aerobic microorganisms (TAM; B) after DP-CO2 pasteuriza tion of muscadine grape juice as influenced by processing pressure (0-40 MPa) and CO2 content (0-15.7%). B A Figure 5-3. Scanning electron micrographs of naturally occurring yeast cells in muscadine juice before (A) and after DP-CO2 at 34.5 MPa and 16% CO2 (B). 5 6 7 810 20 30 40 0 2 4 6 8 10 12 14L o g r ed u cti o n o f Y & MP r e s s u r e ( M P a ) % C O2 A 2 3 4 5 610 20 30 40 0 2 4 6 8 10 12 14L o g r e d u c t i o n o f T A MP r e s s u r e ( M P a ) % C O 2 B

PAGE 100

88 Table 5-1. The effect of heat (75 C for 15 sec) or DPCO2 (D-1: 34.5 MPa, 8% CO2; D2: 34.5 MPa, 16% CO2) pasteurizatio n on the total anthocyanin, soluble phenolic, and antioxidant content of unprocessed muscadine grape juice. Treatment Total anthocyanins (mg/L) Soluble phenolics (mg/L) Antioxidant capacity ( M TE/ mL) Unprocessed 1,105a1 2,211a 22.1 a DP-1 (34.5 MPa, 8% CO2) 1,077a 2,213a 20.7 a DP-2 (34.5 MPa, 16% CO2) 1,102a 2,157b 21.7 a HTST (75 C, 15 sec) 866b 1,859c 18.2 b 1 Values with similar letters within columns are not significantly different (LSD test, P>0.05). Thermal pasteurization decreased tota l anthocyanins by 16%, total soluble phenolics by 26%, and antioxidant capacity by 10% whereas no significant changes were observed for either DP-CO2 processes (Table 5-1). Indivi dually quantified anthocyanins followed a similar trend with greater losses occurring for o -dihydroxy substituted anthocyanins (delphinidin and cyanidin) with respect to the met hoxylated anthocyanins (peonidin and malvidin) as previously observe d in muscadine grape juice (Talcott et al., 2003). Losses ranged from 8-16% following th ermal pasteurization for delphinidin, cyanidin and petunidin, while peonidin and ma lvidin remained stable (< 4% losses). Anthocyanin degradation during processing a nd storage was highly correlated to total soluble phenolics (r=0.86) and antioxidant ca pacity (r=0.82). Insignificant changes in juice pH (3.2) or titratable acidity (0.56 meq tartaric acid/mL) were observed between

PAGE 101

89 treatments after processi ng or storage. Initial CO2 content after DP-CO2 was 6.70 and 9.81 mM for juices pressurized at 8 and 16% CO2 levels, respectively. Trends for polyphenolic and antioxidant ch anges during storage were similar to those observed after processing, where DP-CO2 processed juices presented reduced oxidative degradation when compared to th ermally-pasteurized juices, especially for anthocyanins. Regression analysis conclude d that polyphenolic and antioxidant losses throughout storage followed first order degrad ation kinetics, in accordance with other anthocyanin-containing juices (Del Pozo-Insfran et al., 2004; Brenes et al., 2005; Skrede et al. 2000; Garzon and Wrolstad, 2002), and that greater losses in polyphenolics and antioxidant capacity (1.4-fold) were observe d for thermally pasteurized juices when compared to both DP-CO2 processes. Independent of CO2 concentration, the DP-CO2 juices retained higher total anthocyanins and antioxidant capacity (335 mg/L and 10.9 M Trolox equivalents/mL, re spectively; Figure 5-4A a nd 5-4B), and 473 mg/L higher total soluble phenolics (Figure 5-5) than ther mally-pasteurized juic es after 10 weeks of storage at 4 C. Increased an thocyanin and polyphenolic de gradation presented by heatpasteurized juices presumably occurred due to formation of by-products from carbohydrate and organic acids degradation du ring thermal processing and storage such as furfurals and other carbonyl compounds th at can form condensation products with anthocyanins and polyphenolics (Es-Safi et al., 2002; Malien-Aubert et al., 2001; Boulton, 2001). Previous studies have obser ved that the formation and occurrence of these compounds, accelerated by heat and acid, promotes polyphenolic degradation to yield brown or polymerized pigments that nega tively impact juice quality (Es-Safi et al., 2002; Es-Safi et al., 1999; Dufour and Sauva itre, 2000). Comparable degradation rates

PAGE 102

90 were observed among individual anthocya nins (38%) present in both DP-CO2 juices, while again o -dihydroxylated anthocyanins present in heat-pasteurized juices showed higher degradation rates than their methoxylat ed counterparts. Del phinidin and cyanidin showed the greatest losses ( 69%) after thermal processing followed by petunidin (48%), peonidin (45%) and malvidin (24%). These re sults are in accordance to earlier reports that investigated the stability of different anthocyanins that demonstrated malvidin as the most stable anthocyanidin (Talcott et al., 2003a; Garcia -Viguera and Bridle, 1999; Iacobucci and Sweeny, 1983; Poei-Langston and Wrolstad, 1981). However in actual food systems, the relativ e stability of an anthocyanin is likely a function of its matrix, structural features, and the combined condi tions of processing and storage (Boulton, 2001; Es-Safi et al., 2002; Dufour and Sa uvaitre, 2000; Del Pozo-Insfran et al., 2004; Talcott et al., 2003a; Garcia-Viguera a nd Bridle, 1999; Iacobucci and Sweeny, 1983). The concentration of CO2 utilized during DP-CO2 processing insignificantly affected anthocyanin stability dur ing storage, while increasing CO2 from 8 to 16% offered enhanced storage stability for tota l soluble phenolics and antioxidant capacity (Figure 5-4b and 5-5). These results may s uggest that anthocyanin destruction occurs independently of oxygen content in the juic e matrix while polyphenolic degradation is directly linked to the presence of oxygen. Poei-Langston and Wrolstad (1981) also observed the destruction of an thocyanins in nitrogen sparged systems and proposed a condensation mechanism for their destru ction that did not involve oxygen. Model systems containing anthocyanins and ascorb ic acid have also demonstrated the destruction of these phytochemicals under bot h aerobic and anaerobic conditions (Garzon and Wrolstad, 2002; Garcia-Viguera and Bridle, 1999; Iacobucci and Sweeny, 1983;

PAGE 103

91 Poei-Langston and Wrolstad, 1981; Dufour a nd Sauvaitre, 2000) and therefore exclusion of oxygen during processing woul d not be sufficient to prev ent anthocyanin degradation. However, the prevention and/or reduction of furfurals formation during processing and storage might be an important approach to attenuate ant hocyanin degradation. Consequently, DP-CO2 could be used as a strategy to reduce the degradation of these phytochemical compounds. Similar microbial stability was observed between DP-CO2 and thermallypasteurized juices during the first 5 weeks of storage at 4 C; however significant differences were observed subsequently through 10 weeks (Fi gure 5-6). Yeast/mold counts for both DP-CO2 juices continuously increased throughout subsequent storage whereas no changes were observed for heat-pas teurized juices. Increasing the processing CO2 content from 8 to 16% served to delay microbial growth afte r the sixth week of storage, an effect that might be attributed to the specifi c oxygen content of the juices and/or the inactivation of mi crobial spores during processi ng. Previous investigations have shown that a combined approach between pressure, temperature and CO2 is needed to completely inactivate bacterial spores (Corwin and Shellhammer, 2002; Butz and Tauscher, 2002; Isenschmid et al., 1995; Lin and Lin, 1993). For example, Enomoto et al. (1997) examined the leth al effect of DP-CO2 on spore cells of Bacillus megaterium and observed that the bacter icidal effect of CO2 was found to be enhanced with increasing temperature and treatment time. Similarly, Kami hara et al. (1987) and Haas et al. (1989) observed that temperatures above 70 C were needed to inactivate endospores of B. subtilis B. stearothermophilus and C. sporogenes 3679. Independently of pasteurization

PAGE 104

92 techniques, insignificant changes in the num ber of total aerobic microorganisms were observed during storag e (data not shown). Sensory evaluation Insignificant differences in flavor, arom a, and color intensity were observed between the reference (unprocessed), the hidden reference, and both DP-CO2 processed juices. However, significant differences in fl avor and aroma were detected by panelists ( p < 0.012) between the reference and the heat-pas teurized juice. Alt hough not specifically quantified, the formation of cooked and burnt fl avors is often associat ed with off-flavor development in heat pasteurized juices (EsSafi et al., 2002). Pane lists ranked both DPCO2 juices higher in overall likeability than the heat-pasteurized juice, whereas no difference was observed between DP-CO2 juices and the unprocesse d juice. Panel scores for overall likeability were 6.2 for the hidden reference and both DP-CO2 juices compared to 4.0 for the heat-pasteurized juice (a higher numbe r indicates a higher preference for the juice). Changes in microbial counts were read ily perceived in informal sensory evaluations by the presence of gas formation and the appearance of a yeasty-like aroma that increased throughout storage. Conclusions DP-CO2 served to protect polyphenolic and antioxidant levels throughout processing and storage without comprising sensory attributes of the juices. However, microbial stability of DP-CO2 juices was only comparable to heat-pasteurized counterparts for the first five weeks of st orage. This technology was proven to be a feasible pasteurization tec hnique especially for juices containing heat labile phytochemical, antioxidant, and flavor compounds.

PAGE 105

93 Figure 5-4. Total anthocyanin (A) and antioxidant conten t (B) of heat (HTST; 75 C, 15 sec) and DP-CO2 pasteurized (D-1: 34.5 MPa, 8% CO2; D-2: 34.5 MPa, 16% CO2) musc adine juice during refrigerated storage (4 C). 0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 Week 1 Week 2 Week 3 Week 4 Week 6 Week 8 Week 10 Total anthocyanin content (mg/L)HTST D-1 D-2 0 5 10 15 20 Antioxidant capacity (M TE/mL)HTST D-1 D-2A B

PAGE 106

94 0 500 1000 1500 2000 Total soluble phenolics (mg/mL)HTST D-1 D-2 Week 1 Week 2 Week 3 Week 4 Week 6 Week 8 Week 10 Figure 5-5. Total soluble phenolic content of heat (HTST; 75 C, 15 sec) and DP-CO2 pasteurized (D-1: 34.5 MPa, 8% CO2; D-2: 34.5 MPa, 16% CO2) muscadine juice during refrigerated storage (1-10 weeks at 4 C). Storage time (weeks) 0246810 Log (CFU/mL) 0 1 2 3 4 HTST D-1 D-2 Figure 5-6. Yeast/mold counts of heat (HTST; 75 C, 15 sec) and DP-CO2 pasteurized (D-1: 34.5 MPa, 8% CO2; D-2: 34.5 MPa, 16% CO2) muscadine juice during refrigerated storage (4 C).

PAGE 107

95 CHAPTER 6 ENHANCING THE RETENTION OF PHYT OCHEMICALS AND ORGANOLEPTIC ATTRIBUTES IN MUSCADINE GRAPE JUICE BY DENSE PHASE-CO2 PROCESSING AND COPIGMENTATION Introduction Anthocyanins are polyphenolic compounds that are responsible for the bright blue and red colors of many foods and act as phytochemical antioxi dants with potential healthrelated benefits (Frankel et al., 1995; Meyer et al., 1997; Skrede et al., 2000). Recent shifts in consumer preference for natural pigments have focused on applications of anthocyanins as suitable replacements fo r certified colorants used in juices and beverages. However their relative high cost a nd generally poor stab ility during processing and storage are factors that limit their co mmercial application (Mazza and Brouillard, 1990; Frankel et al., 1995; Meyer et al., 1997; Skrede et al., 2000; Boulton, 2001; Malien-Aubert et al., 2001). Overall, any st rategy or technology that may serve to alleviate these limitations and improve the qua lity attributes of an thocyanin-containing products is of significant impor tance to the food industry. Previous investigations have demonstrated that the formation of intermolecular copigmentation complexes (copigmentation) between anthocyanins and exogenously added polyphenolic cofactors o ffered a protective effect against anthocyanin, ascorbic acid, and antioxidant capacity degradation in both model a nd juice systems (Eiro and Heinonen, 2002; Talcott et al., 2003 a; Brenes et al., 2005; Talc ott et al., 2005). Moreover, previous investigations (Del Pozo-Insfran et al., 2005) revealed that water-soluble polyphenolic cofactors from thyme ( Thymus vulgaris L. ) were most efficacious for

PAGE 108

96 stabilizing anthocyanins under the highly oxi dative conditions crea ted by activation of residual polyphenoloxidase during high pressu re processing. However, the stabilizing effects have not been evaluated in the abse nce of oxidase enzymes or throughout storage. Additionally, the sensory attributes imparted by polyphenolic cofactors, at levels where they are effective for phytochemical retenti on, are an important consideration affecting their use in food systems. A promising non-thermal processing technol ogy that may help with phytochemical stability is the continuous dense phase-CO2 process (DP-CO2). Without heat, a reduction in the formation of carbohydrate and asco rbic acid by-products such as carbonyl compounds is realized, which have been id entified as a key factor for preventing anthocyanin degradation in fruit juices (EsSafi et al., 1999; Malien -Aubert et al., 2001; Es-Safi et al., 2002;), especia lly those containing ascorbic acid (Poei-Langston and Wrolstad, 1981; Garcia-Viguera and Bridle, 1999; Garzon and Wrolst ad, 2002; Talcott et al., 2003). DP-CO2, also known as carbon dioxide-assisted high pressure processing, is a continuous pasteurization tec hnology that uses pressures 90 MPa in combination with dissolved carbon dioxide to inactivate microorganisms and presumably protect thermolabile phytochemical and flavor compounds. A secondary benefit of the processing system is the removal of disso lved oxygen, which is instrumental in preventing degradation to antioxidant phytochemicals. This cumulative knowledge suggests that ad dition of polyphenolic cofactors from thyme along with the DP-CO2 process may prove to be an effective strategy to decrease phytochemical and antioxidant losses that o ccur during storage of juices containing anthocyanins and ascorbic acid. Therefore, this study evaluated phytochemical stability

PAGE 109

97 and organoleptic properties of a DP-CO2 processed muscadine grape juice as affected by the addition of thyme cofactors and ascorbic acid. Results were compared to juices processed by a thermal pasteurization in or der to assess the differences between the processing methods and to observe if phytoc hemical degradation was increased by the formation of compounds created during thermal processing. Materials and Methods Materials and Processing Polyphenolics from dried thyme leaves (McCormick & Co., Inc., Hunt Valley, MD) were exhaustively extracted with hot water, purified us ing reverse phase C18 SepPak Vac 20 cc mini-columns (Water Corporati on, Mass., USA), and re-dissolved in 0.1M citric acid solution according to the conditions described in Chapter 4. Purified spring water (Publix, Lakeland, FL) and 100% food-gr ade ethanol (McCormick Distillery Co., Weston, MO) were used to purify and/or solubi lize the isolated polyphenolic cofactors. Red muscadine grapes (cv. Noble) were obtained from a local grower and handsorted for uniformity. Fruit was crushed and heated to 75 C for 2 min in an open steam kettle and the juice extracted using a hydraulic basket press (ProsperoÂ’s Equipment, Cort, NY). Preliminary investigations concluded that this heat treatment inactivated oxidase enzymes. Juice was subsequently filtered first through cheesecloth followed by vacuum filtration through a 1 cm bed of diatomaceous earth. The resultant juice was then divided into two portions for additi on of the polyphenolic cofactors from thyme (0 and 1:100 anthocyanin-to-cofactor molar ra tio). The ratio corresponds to the molar concentration of total anthocyanins present in the juice and the molar concentration of total polyphenolics present in the thyme extract. Juices at each cofactor concentration were again divided into two portions and half fortified with ca. 450 mg/L ascorbic acid as opposed to an

PAGE 110

98 equivalent volume of the citric acid buffer as the control. The four treatments were then divided for heat (HTST; 75 C for 15 sec) or DP-CO2 pasteurization ( 34.5 MPa at 8 and 16% CO2). The DP-CO2 regimes were confirmed to impart > 5 log reduction of aerobic microorganisms and yeast/mold according to the results observed in Chapter 5. After pasteurization, each juice treatment was divided into respective portions for microbial, phytochemical, and organoleptic evaluations. Samples for microbial and phytochemical analysis were immediately tran sferred into 20 mL screwed cap vials and stored at 4 C for 10 weeks, whereas samples for sensory analysis were transferred to sterile 4 L glass containers. Samples used exclusively for phytochemical assessment were dosed with sodium azide (50 mg/L) to retard microbial growth. Physicochemical and Microbial Analyses Individual anthocyanin 3,5diglycosides, total solubl e phenolics, antioxidant capacity, pH, total titratable acidity, and residual CO2 content in the juices were determined as described in Chapter 5. Total ascorbic acid (the sum of Land dehydroascorbic acid) was quantified by reverse phase HPLC using the conditions described in Chapter 4. Microbial counts throughout processing and storage were determined on aerobic count plates and yeast & mold plates (3M Petrifilm Microbiology Products, St. Paul, MN) as described in Chapter 5. Sensory Evaluation Flavor, aroma, and color intensity of ju ices with and without added polyphenolic cofactors were compared using a differenc e-from-control test for each of the three pausterization treatments. Panelists compared the sensory attributes of the reference (no added cofactors) with that of a hidden reference and the copigmented juice. A randomized complete block design was used and difference from control measurements

PAGE 111

99 were recorded on a line scale with anchors at 0 and 10 that represente d “no difference” to “extremely different” in juice sensory attri butes. Each panelist evaluated all three combinations of processed juice (DP-CO2 at 8 and 16% CO2, and HTST). A 9-point hedonic scale was also conducted in order to compare the overall likeability of the reference and copigmented juices pr ocessed by each processing regime. Before sensory analysis, all juices were degassed in order to equalize carbonation levels by placing them in a 4 L sterile gl ass container on a hot plate with continuous stirring for 4 h at 20 C. Juices were then served at room temperature in randomly numbered plastic cups. A tray with a cup of water and non-salted crac kers was also given to each of the panelists. All sensory tests we re performed in the University of Florida’s taste panel facility using sixty untrained panelists (33 females, 18-29 age range). Statistical Analysis Data represents the mean and standard erro r of juices analyzed as a 3 x 2 x 2 x 9 factorial comparing three pr ocessing conditions (DP-CO2 at 8% or 16% both at 34.5 MPa, or thermally pasteurized), with and without thyme cofactors, in the presence/absence of ascorbic acid, and eval uated at nine sampli ng points (unprocessed, processed, week 1, 2, 3, 4 6 8 and 10). Li near regression, Pear son correlations and analysis of variance were conducted using JMP software (SAS, Cary, NC), with mean separation performed using the LSD test (P< 0.05). All experiments were randomized and conducted in triplicate. Sensor y data was recorded and analyzed using Compusense five (Compusense, Guelph, Ontario, Canada), and analysis of variance was conducted by using the Tukey’s multiple comparisons method (P<0.05).

PAGE 112

100 Results and Discussion This study investigated phytochemical a nd organoleptic changes in muscadine grape juice associated with DP-CO2 processing and the addi tion of thyme polyphenolic cofactors as a means to reduce the oxidativ e phytochemical degradation that occurs in thermally processed, anthocyanin-containing be verages that are commonly fortified with L-ascorbic acid. Significant differences in anthocyanins, soluble phenolics, antioxidant capacity, and organoleptic attributes were observed by processing methods, copigmentation, and ascorbic acid fortification. Thermal pasteurization was more detrimental to polyphenolics, antioxidant cap acity, and organoleptic attributes as compared to the DP-CO2 processes. Moreover, enhanced storage stability was observed for DP-CO2 processed juices in rela tion to thermal pasteurization. In addition to reducing phytochemical and antioxidant losses, copi gmentation increased anthocyanin color intensity and antioxidant content of the juices, and also masked the detrimental color fading that took place during storage. L-ascorb ic acid fortification increased the initial polyphenolic and antioxidant con centrations of the juices likely due to its reducing and antioxidant properties, however the addition of this phytochemical resulted in increased phytochemical and antioxidant de gradation during storage. Initial Effects of Copigmentation and Ascorbic Acid Fortification Preliminary investigations indicated that the hyperchromic intensity (red coloration) of muscadine juice could be increased up to 5-fold by the addition of polyphenolic cofactors isolated from thyme in a 1:400 ratio. However due to adverse bitter and astringent flavors at high cofactor concentrations, this study evaluated juices at a 1:100 ratio based on informal sensory eval uations. The anthocyanin content of the juices was not affected by the addition of th e cofactors but was initially protected by the

PAGE 113

101 addition of L-ascorbic acid (Table 6-1) w ith 100 mg/L higher con centrations than the control juices, likely due to its antioxidant protection. Cofactor addition increased the initial content of soluble phenolics by 660 mg/L and antioxidant capacity by 30 M TE/mL, concentrations that were additionally increased by 175 mg/L and 9 M TE/mL following ascorbic acid fortific ation, respectively (Table 1), due to the metal reducing and antioxidant properties of this phytonutrien t. Cofactors also increased visual color intensity of the juice as evidenced in a decline in hue angle from 19.2 to 10.2, which appear to the naked eye as a more intense red color. Phytochemical Changes Due to Thermal and DP-CO2 Processing The DP-CO2 processes insignificantly altered juice phytochemical and antioxidant content (Table 6-1), while thermal pasteu rization reduced anthocyanins by 263 mg/L, soluble phenolics by 366 mg/L, L-ascorbic acid by 42 mg/L and antioxidant capacity by 6 mol TE/mL. Similarly to the resu lts observed in chapter 5, DP-CO2 did not affect juice pH (3.2) or titratable acidity (0.57 meq tartaric acid/mL). Residual CO2 content was 6.74 and 9.75 mM for juices pressurized at 8 and 16% CO2 levels, respectively. Copigmentation and ascorbic acid fortif ication did not reduce phytochemical and antioxidant losses, but generally helped to re tain higher levels of anthocyanins, phenolics, and antioxidant capacity when compared to control treatments. Copigmented treatments contained higher soluble phenolic (249 mg/L), antioxidant capacity (18 M TE/mL), and ascorbic acid (17 mg/L) content than contro l treatments, while ascorbic acid fortified juices presented higher anthocyanin (154 mg/L ), phenolic (214 mg/L), and antioxidant (8 M TE/mL) content than heat-pasteurized control juices. The combined addition of ascorbic acid and thyme cofactors synergis tically acted to protect phytochemicals and

PAGE 114

102 antioxidant levels of thermally pasteuri zed juices, as evidenced by the additional retention of these compounds following thermal processing. Organoleptic Changes Due to Addition of Thyme Polyphenolic Cofactors For each processing treatment (DP-CO2 or HTST), panelists compared the sensory attributes of the reference juice (no added co factors) with those pr esented by the hidden reference and the copigmented juice. The hi dden control was used to determine if the consumer was in fact detecting a difference be tween the juices and if so to determine the extent of organoleptic differences between the juices. Generally, addition of polyphenolic cofactors insignificantly affected the flavor and aroma of DP-CO2 and HTST processed muscadine grape juices. The only significant di fference (< 1 unit on a 10 point scale) that was detected by the panelists was between the aroma of the reference and the copigmented juice processed by DP-CO2 at 34.5 MPa and 8% CO2; however, these juices had similar ratings for overall likeability. Copigmented juices r eceived higher panel scores for overall likeability than cont rol juices processed by heat and DP-CO2 at 16% CO2. Juice color intensity was the only organoleptic trait in which panelists were able to detect a difference between control and copi gmented juices. This difference was likely attributed to the color enhancing proper ties of thyme polyphenolic cofactors, as copigmented juices had a more intense red color when compared to their corresponding control treatments. In addition, copigmentati on served to mask the detrimental color changes that occurred during thermal processing, as insignif icant changes in hue values were observed for copigmented juices when compared to a appreciable color fade in control juices. Although panelist s did not evaluate in paralle l the organoleptic attributes of juices processed by the different pasteuri zation regimes, results indicated that DP-CO2

PAGE 115

103 juices received higher panel scores for overall acceptability (>2 units) than heat pasteurized juices indicating a higher preference for DP-CO2 processed juices. Phytochemical and Microbial Changes During Refrigerated Storage Thermal processing and ascorbic acid for tification were the variables that most affected the polyphenolic and antioxidant cap acity levels of muscadine juice throughout storage at 4 C, while the DP-CO2 processes and copigmentation helped to increase phytochemical and antioxidant re tention of the juices. In or der to investigate if each independent variable (processing regime, copigm entation, or ascorbic acid fortification) delayed phytochemical degradation over time or if the protective effects were only observed during processing, degradation rates we re calculated according to Taoukis et al. (1997). Regression analysis concluded that rates of anthocyani n, soluble phenolics, antioxidant capacity and ascorbic acid de gradation over time followed first order degradation kinetics (Tables 62 to 6-5) in accordance to pr evious studies (Skrede et al., 2000; Garzon and Wrolstad, 2002; Del Pozo-Insf ran et al., 2004; Bren es et al., 2005). Independently of processing parameters, DP-CO2 juices retained 386 mg/L higher anthocyanin content (Figure 61) and presented 2-fold lowe r degradation rates (Table 62) than thermally pasteurized juic es after 10 weeks of storage at 4 C. This presumably occurred due to formation of carbohydrate a nd/or ascorbic acid degradation by-products during thermal processing and subsequent storage that accelerated anthocyanin degradation yielding brown polymerized pigmen ts that negatively impacted juice quality (Es-Safi et al., 1999; Bradshaw et al., 2001; Dufour and Sa uvaitre, 2000; Malien-Aubert et al., 2001; Es-Safi et al., 2002).

PAGE 116

104Table 6-1. The effect of thyme cofactors (0 and 100:1 anthocyanin-to-cof actor ratio) and ascorbic ac id fortifica tion (0, 450 mg /L) on the total anthocyanin, soluble phe nolic and antioxidant content of unprocessed, heat (HTST; 75 C, 15 sec), and DP-CO2 (34.5 MPa at 8% or 16% CO2) pasteurized muscadine grape juice. 1 Indicates the ratio between the molar concentration of total an thocyanins in muscadine grape juice (expressed as malvidin 3glucoside equivalents) and the molar con centration of thyme polyphenolic cofactors (e xpressed in gallic acid equivalents). 2 Means with similar letters within columns are not significantly different (LSD test, P>0.05). 3 Means with an asterisk (*) for each response variable indicate a significant e ffect (LSD test, P<0.05) due to addition of ascorbic acid. No Ascorbic acid Added Ascorbic acid (450 mg/L) Treatment Cofactor ratio1 Total anthocyanins (mg/L) Soluble phenolics (mg/L) Antioxidant capacity ( M TE/mL) Total anthocyanins (mg/L) Soluble phenolics (mg/L) Antioxidant capacity ( M TE/mL) Total ascorbic acid (mg/L) Unprocessed 0 1,110a2 2,121c 23.9a 1,210a*3 2,579c*31.9c* 441a 100:1 1,175a 2,780a 50.9a 1,219a 2,955a*59.4a* 445a 34.5 MPa, 8% CO2 0 1,090a 2,118c 23.2c 1,120a 2,569c*31.3c* 428a 100:1 1,078a 2,661b 48.0a 1,185a* 2,774b 58.1a* 434a 34.5 MPa, 16% CO2 0 1,103a 2,121c 24.7c 1,194a* 2,531c* 35.3c* 434a 100:1 1,101a 2,753a 46.4a 1,208a* 2,862a 61.1a* 432a HTST 0 843b 1,754d 17.7d 997c* 1,968e*26.1d* 399c (75 C, 15 sec) 100:1 865b 2,003c 35.9b 1,101b* 2,136d*50.1b* 416b

PAGE 117

105 Table 6-2. Effect of thyme cofactors (0 a nd 100:1 anthocyanin-to -cofactor ratio) and ascorbic acid fortification (0, 450 mg /L) on first-order degradation kinetic parameters of anthocyanins present in heat (HTST; 75 C for 15 sec) or DPCO2 (34.5 MPa at 8 or 16% CO2) processed muscadine grape juice during storage at 4 C. No Ascorbic acid Added Ascorbic acid (450 mg/L) Treatment Cofactor ratio1 1 2 t1/2 3 R2 1 t1/2 R2 34.5 MPa, 8% CO2 0 7.07 a4 98.1 c0.99 17.8 c*5 39.0 c*0.95 1:100 6.49 a 107 b0.96 12.6 b 55.0 b*0.98 34.5 MPa, 16% CO2 0 7.37 a 94.1 c0.96 13.9 b* 50.0 b*0.98 1:100 5.50 b 126 a0.98 10.6 a* 65.3 a* 0.98 HTST 0 15.8 d 43.9 e0.90 21.3 d* 32.5 c*0.95 1:100 10.3 c 67.2 d0.99 12.7 b* 54.4 b* 0.98 1 Indicates the ratio between the molar co ncentration of total anthocyanins in muscadine grape juice (expressed as malvidin 3-glucoside equivalents) and the molar concentration of thyme polyphenol ic cofactors (expressed in gallic acid equivalents). 2 Indicates the degradation rate (slope, 1) of anthocyanins (days-1). 3 Indicates the half life (days) of initial anthocyanin content. 4 Means with similar letters within columns are not significantly different (LSD test, P>0.05). 5 Means with an asterisk (*) for each kinetic parameter indicate a significant effect (LSD test, P<0.05) due to addition of ascorbic acid when compared to the same tr eatment without ascorbic acid.

PAGE 118

106 Table 6-3. Effect of thyme cofactors (0 a nd 100:1 anthocyanin-to -cofactor ratio) and ascorbic acid (0, 450 mg/L) on first-orde r degradation kineti c parameters of soluble phenolics in heat (HTST; 75 C for 15 sec) or DPCO2 (34.5 MPa at 8 or 16% CO2) processed muscadine grape juice during storage at 4 C. No Ascorbic acid Added Ascorbic acid (450 mg/L) Treatment Cofactor ratio1 1 2 t1/2 3 R2 1 t1/2 R2 34.5 MPa, 8% CO2 0 25.1 c4 27.6 c 0.91 38.2 c*5 18.1 c*0.99 1:100 10.9 a 63.6 a 0.98 24.3 ab* 28.5 ab* 0.98 34.5 MPa, 16% CO2 0 19.2 b 36.1 b 0.97 21.9 a 31.7 a*0.86 1:100 10.3 a 67.2 a 0.92 26.5 b* 26.2 b*0.97 HTST 0 36.2 d 19.1 d 0.97 52.6 d* 13.2 d 0.98 1:100 14.5 b 47.9 b 0.87 25.0 b* 27.8 b*0.97 1 Indicates the ratio between the molar co ncentration of total anthocyanins in muscadine grape juice (expressed as malvidin 3-glucoside equivalents) and the molar concentration of thyme polyphenol ic cofactors (expressed in gallic acid equivalents). 2 Indicates the degradation rate (slope, 1) of soluble phenolics (days-1). 3 Indicates the half life (days) of initial soluble phenolics content. 4 Means with similar letters within columns are not significantly different (LSD test, P>0.05). 5 Means with an asterisk (*) for each kinetic parameter indicate a significan t effect (LSD test, P<0.05) due to addition of ascorbic acid when compared to the same treatment without cofactors.

PAGE 119

107 Table 6-4. Effect of thyme cofactors (0 a nd 100:1 anthocyanin-to -cofactor ratio) and ascorbic acid fortification (0, 450 mg /L) on first-order degradation kinetic parameters of antioxidant capacity in heat (HTST; 75 C for 15 sec) or DPCO2 (34.5 MPa at 8 or 16% CO2) processed muscadine grape juice during storage at 4 C. No Ascorbic acid Added Ascorbic acid (450 mg/L) Treatment Cofactor ratio1 1 2 t1/2 3 R2 1 t1/2 R2 34.5 MPa, 8% CO2 0 9.55 c472.6 c 0.93 21.2 c*5 32.6 d*0.98 1:100 8.54 b81.1 b 0.99 7.64 a 90.7 a 0.97 34.5 MPa, 16% CO2 0 7.37 a94.0 a 0.97 15.4 b* 45.0 c*0.95 1:100 7.70 a90.3 a 0.98 7.90 a 87.6 a 0.98 HTST 0 13.0 d53.4 d 0.94 31.6 d* 21.9 e*0.95 1:100 14.2 d49.0 d 0.87 11.0 b 62.8 b 0.98 1 Indicates the ratio between the molar concentr ation of total anthocyanins in muscadine grape juice (expressed as malv idin 3-glucoside equivalents) and the molar concentration of thyme polyphenolic cofactors (expre ssed in gallic acid equivalents). 2 Indicates the degradation rate (slope, 1) of antioxidant capacity (days-1). 3 Indicates the half life (days) of initial antioxidant capacity content. 4 Means with similar letters within columns are not significantly different (LSD test, P>0.05). 5 Means with an asterisk (*) for each kinetic parameter indicate a significant effect (LSD test, P<0.05) due to addition of ascorbic acid when compared to the same tr eatment without ascorbic acid.

PAGE 120

108 Table 6-5. Effect of thyme cofactors (0 and 100:1 anthocyanin-to-c ofactor ratio) on firstorder degradation kinetic pa rameters of total ascorbic acid present in heat (HTST; 75 C for 15 sec) or DPCO2 (34.5 MPa at 8 or 16% CO2) processed muscadine grape juice during storage at 4 C. Without thyme cofactors With thyme cofactors 1 Treatment 1 2 t1/2 3 R2 1 t1/2 R2 34.5 MPa, 8% CO2 57.8 b412.0 b 0.98 42.9 a*516.2 a* 0.95 34.5 MPa, 16% CO2 46.8 a 14.8 a 0.96 43.1 a 16.1 c* 0.96 HTST 152 c 4.62 c 0.83 92.5 b*7.51 b* 0.97 1 Indicates the ratio between the molar concen tration of total anthocyanins in muscadine grape juice (expressed as malv idin 3-glucoside equivalents) and the molar concentration of thyme polyphenolic cofactors (expre ssed in gallic acid equivalents). 2 Indicates the degradation rates ( 1) of total ascorbic acid (days-1). 3 t1/2 indicates the half life (days) of initial total ascorbic acid content (450 mg/L). 4 Means with similar letters within columns are not significantly diff erent (LSD test, P>0.05). 5 Means with an asterisk (*) for each kinetic parameter indicate a significant effect (LSD test, P<0.05) due to addition of thyme cofactors when compared to the same treatment without ascorbic acid. Comparison of kinetic parameters am ong treatments indicated that although ascorbic acid fortification increased an thocyanin degradation for all processing treatments, its addition had a more pronounced effect for DP-CO2 juices (3.5-fold increase in degradation rates) than for the he at-pasteurized juices (1.4-fold). This likely occurred due to the slower rates of de gradation initially observed for DP-CO2 juices when compared to the significantly higher degradation rates and oxidative conditions presented by thermally processed controls. As expected, ascorbic acid-fortified juices retained less anthocyanins than control ju ices confirming the destructive interaction between these compounds in a food system (Poei-Langston and Wr olstad, 1981; Garcia-

PAGE 121

109 Viguera and Bridle, 1999; Garzon and Wrolst ad, 2002; Brenes et al., 2005), and again this effect was more marked for DP-CO2 processed juices. Addition of thyme cofactors reduced anthocyanin degradation rates by 1.5fold (Table 6-2), ex tended anthocyanin half-life values from 9 to 32 days, and reta ined higher anthocyanin concentrations than controls from 57 up to 278 mg/L after 10 weeks storage (Figure 6-1), in a manner independent of processing me thod or L-ascorbic acid for tification. For non-copigmented juices processed by DP-CO2, the CO2 concentration had no effect on anthocyanin storage stability. However, in copigmented juices an increase in processing CO2 from 8 to 16% delayed anthocyanin degradation by 1.4-fold and increased anthocya nin retention by > 63 mg/L at the end of the storage (Figure 6-1) This trend was also observed for ascorbic acid-fortified juices and likely occurred due to the protection of polyphenolic compounds, present both in the juice and cofactors, by oxygen excl usion. Dissolved oxygen is known to significantly increase the rate and seque nce of polyphenolic and/or ascorbic acid degradation that produce quinones or carbonyl compounds that re act and accelerate anthocyanin degradation (Es-Sa fi et al., 1999; Malien-Aubert et al., 2001; Es-Safi et al., 2002). Results also suggested th at the prevention and/or redu ction of furfural formation during thermal processing and storage might be an important approach to attenuate anthocyanin degradation that can be obtained by non-therma l processing such as DPCO2. Anthocyanin losses were correlated to in strumental color evaluations (r=0.84) that showed slight changes in hue values for DP-CO2 processed juices during storage while thermally processed juices showed a prominent loss of red color. As previously discussed, results from color analysis indicate d that cofactor addition served to mask the detrimental color changes that occurred throughout storage.

PAGE 122

110 Figure 6-1. Total anthocyanin content of muscadine grape juice without (A) and with ascorbic acid (B; 450 mg/L) dur ing refrigerated storage (4 C) as affected by heat (HTST; 75 C, 15 sec) and DP-CO2 (34.5 MPa at 8% or 16% CO2) pasteurization and the addition of thyme cofactors (Thy; 0 and 1:100 anthocyanin-to-cofactor ratio). Copigmented juices markedly presented hi gher concentrations of soluble phenolics and antioxidant capacity than control juices dur ing storage (Figures 62 and 6-3), even in the presence of L-ascorbic acid which was detected as an interference in the FolinCiocalteu assay. Similar to results for anthoc yanins, the rate of degradation for soluble phenolics and antioxidant capacity (Table s 6-3 and 6-4, respectively) was more pronounced for heat-pasteurized juices followed by the DP-CO2 process at 8% CO2 and lastly at 16% CO2. Increases in CO2 concentrations from 8 to 16% during DP-CO2 offered enhanced storage stability for antio xidant capacity, soluble phenolics, and total ascorbic acid (Figures 6-2 to 6-4), indicat ing that greater exclusion of dissolved oxygen Storage time (weeks) 0246810 Total anthocyanin content (mg/L) 200 400 600 800 1000 1200 A. No ascorbic acid 8% CO2 8% CO2, Thy 16% CO2 16% CO2, Thy HTST HTST, Thy Storage time (weeks) 0246810 200 400 600 800 1000 1200 B. Ascorbic acid (450 mg/L)

PAGE 123

111 was instrumental for phytochemical retention during storage. Addition of thyme cofactors reduced rates of soluble phenolic degradation by 2-fold for all treatments independent of ascorbic acid concentration (Table 6-3), a protective effect that was also observed for losses in antioxidant capacity for L-ascorbic aci d-fortified juices (Tab le 6-4). Addition of ascorbic acid generally increased the rates of soluble phenolics and antioxidant capacity degradation by 2-fold when compared to unf ortified control trea tments. Addition of ascorbic acid also presented a negative eff ect for copigmented treatments, however juices with added cofactors showed enhanced stab ility that their respective non-copigmented counterparts as evidenced by their smaller de gradation rates and hi gher half-life values. Similar to trends observed for anthocyanins, soluble phenolics, and antioxidant capacity heat pasteurized juices demonstrated 3-fold faster rates of total ascorbic acid degradation when compared to DP-CO2 counterparts, rates that were reduced to 2-fold by the addition of thyme cofactors (Table 65). Copigmentation also reduced ascorbic degradation for the juice pr ocessed at 34.5 MPa and 8% CO2. Increasing processing CO2 levels from 8 to 16% offered enhanced storag e stability for ascorbic acid, suggesting that losses of this phytonutrient can be preven ted by reduction of oxygen content in food systems as those reported by Poei-Langst on and Wrolstad (1981) in anthocyanincontaining juices (1981). These results al ong with the trends observed for soluble phenolics, suggest that anthocyanin destruc tion occurs independently of oxygen content in the juice matrix while polyphenolic and ascorb ic acid degradation is directly linked to the presence of oxygen. Copigmented DP-CO2 juices contained significantly higher ascorbic acid content than heat-pasteuriz ed controls after six weeks of storage,

PAGE 124

112 independently of processing parameters. Howe ver, no ascorbic acid was detected in any of the juices after the se venth week of storage. Figure 6-2. Total soluble phenolic content of muscadine grape juice without (A) and with ascorbic acid (B; 450 mg/L) dur ing refrigerated storage (4 C) as affected by heat (HTST; 75 C, 15 sec) and DP-CO2 pasteu rization (34.5 MPa at 8% or 16% CO2), and the addition of t hyme cofactors (Thy; 0 and 1:100 anthocyanin-to-cofactor ratio). Stora g e time ( weeks ) 0246810 0 500 1000 1500 2000 2500 3000 B. Ascorbic acid (450 mg/L) Storage time (weeks) 0246810 Total soluble phenolics (mg/mL) 0 500 1000 1500 2000 2500 3000 A. No ascorbic acid 8% CO2 8% CO2, Thy 16% CO2 16% CO2, Thy HTST HTST, Thy

PAGE 125

113 Figure 6-3. Antioxidant capacity of muscadine grape juice without (A ) and with ascorbic acid (B; 450 mg/L) during re frigerated storage (4 C) as affected by heat (HTST; 75 C, 15 sec) and DP-CO2 pasteuri zation (34.5 MPa at 8% or 16% CO2), and the addition of thyme cofact ors (Thy; 0 and 1: 100 anthocyanin-tocofactor ratio). Similarly to the trends observed in chap ter 5, thermally-pasteurized juices did not present significant yeast/mold growth (< 10 CFU/mL) during the entire shelf-life, while DP-CO2 juices exhibited a gr adual increase in microbial growth after the 6th week of storage that was depe ndent of processing CO2 content. Results also concluded that microbial stability was comparable between control and copigmented treatments even though previous investigations have reported antimicrobial propertie s of thyme extracts (Nychas, 1995; Rauha, 2000; Sagdic, 2003). The used polyphenolic level of thyme extract was likely not effective in prev enting microbial growth. Independently of Storage time (weeks) 0246810 0 10 20 30 40 50 60 B. Ascorbic acid (450 mg/L) Storage time (weeks) 0246810 Antioxidant capacity (M TE/mL) 0 10 20 30 40 50 60 A. No ascorbic acid 8% CO2 8% CO2, Thy 16% CO2 16% CO2, Thy HTST HTST, Thy

PAGE 126

114 processing regimes and copigmentation, to tal aerobic counts were < 10 CFU/mL throughout refrigerated storage. Figure 6-4. Total ascorbic acid content of muscadine grape juice during refrigerated storage (4 C) as affected by heat (HTST; 75 C, 15 sec) and DP-CO2 pasteurization (34.5 MPa at 8% or 16 % CO2), and the addition of thyme cofactors (Thy; 0 and 1:100 ant hocyanin-to-cofactor ratio). Conclusions Results of this study showed that DP-CO2 and addition of thyme polyphenolic cofactors served to protect phytochemical a nd antioxidant levels in muscadine juice throughout storage without comprising the organole ptic attributes of th e juice. In addition to preventing anthocyanin and ascorbic acid losses, copigmentation was shown to be an effective strategy to increase the color in tensity of muscadine juice and mask the Storage time (weeks) 01234567 Total ascorbic acid (mg/L) 0 100 200 300 400 8% CO2 8% CO2, Thy 16% CO2 16% CO2, Thy HTST HTST, Thy

PAGE 127

115 detrimental color fading that occurred duri ng storage, parameters that along with antioxidant content are significant factor s affecting consumer acceptability and preference.

PAGE 128

116 CHAPTER 7 INACTIVATION OF POLYPHENOL OXIDAS E IN MUSCADINE GRAPE JUICE BY DENSE PHASE-CO2 PROCESSING Introduction Dense phase-CO2 processing (DP-CO2) is a continuous, non-thermal processing system for liquid foods that utilizes pressu re (< 90 MPa) in combination with carbon dioxide (CO2) to inactivate microorganisms. This emerging processing technology is a promising alternative to traditional therma l pasteurization technol ogies and may lessen detrimental effects to thermo labile phytonutrients (Corwin and Shellhammer, 2002; Park et al., 2002; Boff et al., 2003). However, a potential downside of this technology is the presence and/or activation of residual enzymes, such as polyphenol oxidase (PPO), lipoxygenase, and peroxidase, during processing and storage, which may be partially responsible for oxidative degrad ation reactions. Seve ral studies have shown the effect of SC-CO2 and DP-CO2 on pectin methyl esterase, li poxygenase, peroxidase, and PPO in model and real food systems (Chen et al., 1992 ; Corwin and Shellhammer, 2002; Park et al., 2002; Boff et al., 2003;). For instance, Tani guchi et al. (1987) st udied the effect of SC-CO2 on nine different enzymes at 20.3 MPa a nd 35 C for 1 h and showed that > 90% of the enzymatic activity was retained when the water content of the enzyme preparations was 5-7 %. Chen et al. ( 1992) also reported that PPO can be inactivated at low temperatures with SC-CO2; however, the degree of inhi bition was dependent on the source of the enzyme. Arreola et al. (1991a) investigated pe ctin esterase activity and showed that its inactivation was affected by temperature, pressure and process time, and

PAGE 129

117 that complete enzyme inactivation was achieved at 26.9 MPa and 56 C for 145 min. Although several reports have re ported the pressure induced activation or inactivation of pectin esterease in both m odel and real systems, and PPO in model systems, limited information is currently available in regard s to PPO present in muscadine grape juice, especially using a continuous CO2 processing system. This information is of relevance to the food industry due to inherent instability of anthocyanin-containi ng products to both chemical and enzymatic degradation and its consequent deleterious effects on quality attributes and phytochemical deterioration. Th erefore, the present study investigated the effect of DP-CO2 on PPO activity and its conse quent effect on polyphenolic and antioxidant changes in muscadine juice unde r a variety of processing pressures (27.6, 38.3, and 48.3 MPa) and CO2 levels (0, 7.5, and 15%). Since the majority of the studies looking at the effects of HHP or DP-CO2 on enzyme inactivation do not take into account the shelf-life stability of the product, this study also investigated the polyphenolic and antioxidant changes that occurred in muscad ine juice during its refrigerated storage (4 C) and which were associated with residual PPO activity following DP-CO2 at 48.3 MPa and 0 and 15% CO2. Materials and Methods Materials Muscadine grapes (cv. Noble) were obtained from a local grower in central Florida and held frozen (-20C) until needed. Fru it was rapidly thawed by placing them under running tap water and hand-sorted for uniformity of ripeness. Grapes were then crushed, heated to 46 C in an open steam kettle, and held for 11 min prior to juice extraction in a hydraulic basket press (ProsperoÂ’s Equipment, Cort, NY). Previous studies demonstrated that the maximum PPO activity in muscadine juice could be obtained by using this juice

PAGE 130

118 extraction procedure. The juice was immedi ately filtered through cheesecloth followed by vacuum filtration through a 1 cm bed of diatomaceous earth. Sodium azide (50 mg/L) was then added to the juice in order to retard microbial growth throughout the study. Effect of DP-CO2 Processing on PPO activity Residual PPO activity in muscadine juice following different DP-CO2 treatments varying in processing pressure (27.6, 38.3, and 48.3 MPa) and CO2 content (0, 7.5, and 15%) at a constant proc essing temperature (30 C) and flow rate (500 mL/min) was investigated and determined according to a modified polarographic method described in Chapter 4. The reaction was started when 0.9 mL of 0.12 M catechin was added to 2.8 mL of muscadine juice mixe d with 1 mL of 0.1 M phosphate buffer at pH 3.5. The assay was carried out in air-saturated solutions agitated with a magn etic stirrer and the electrode calibrated using airsaturated water (245 nmol O2 / ml H2O). Enzymatic activity was determined from the linear portion of the oxygen consumption curve, reported as nmoles of oxygen consumed per second (nkat) and expressed as a percentage of the unprocessed juice (100% activity, 0.12 nkat) Polyphenolic and antioxidant changes associated with enzyme inactivation were also investigated and compared among the different processing treatments. Storage Stability of Muscadine Juice with Residual PPO Activity Two DP-CO2 treatments (48.3 MPa at 0 or 15% CO2) were chosen to investigate the phytochemical changes associated with residual PPO activity during refrigerated storage (4 C) of muscadine juice. Treatments were additionally compared to control juices containing no enzyme activit y and processed at equal DP-CO2 processing conditions. Previous studies determined th at complete PPO inactivation could be obtained by extracting the juice at 75 C for 2 min. Following DP-CO2 processing, all

PAGE 131

119 juices were immediately transferred into 20 mL screwed cap vials, and stored at 4 C for 4 weeks. Samples were collected every week and were evaluated for anthocyanin, soluble phenolics and antioxidant capacity levels. Chemical Analyses Individual anthocyanin 3,5diglycosides, total solubl e phenolics, antioxidant capacity, pH, total titratable acidity, and residual CO2 content in the juices were determined as described in Chapter 5. Statistical Analysis Data for the DP-CO2 inactivation study represents th e mean and standard error of juices analyzed as a 3 x 3 factorial comparing three DP-CO2 processing pressures (27.6, 38.3, and 48.3 MPa) and three CO2 levels (0, 7.5, and 15%). Data for the storage study represents the mean and standard error of juices analyzed as a 2 x 2 x 5 factorial comparing two juices enzyme activities (con trol and residual PPO ) processed at two different CO2 processing levels (0 and 15%) with constant pressure (48.3 MPa), and evaluated at 5 sampling points (processed, week 1, 2, 3, and 4). Linear regression, Pearson correlations and analysis of varian ce were conducted using JMP software (SAS, Cary, NC), with mean separation perfor med using the LSD test (P<0.05). All experiments were randomized and conducted in triplicate. Results and Discussion Effect of DP-CO2 Processing on PPO Activity Differences in PPO activity, anthocyanins polyphenolics, and antioxidant levels were observed in muscadine grape juice as affected by processing pressure and CO2 levels (Figures 7-1 and 7-2). Pressure alone was responsible for a ~40% decrease in PPO activity and resulted in 16-40% polyphenolic and antioxidant losses, while increasing

PAGE 132

120 CO2 processing levels from 0 to 7.5% was respon sible for an additional ~35% decrease in enzyme activity and ~2-fold higher polypheno lic and antioxidant retention. Further increases of CO2 processing levels to 15% did no t serve to reduce PPO activity nor prevented polyphenolic and antioxidant losses. Table 7-1. Individual an d total anthocyanin content of unprocessed muscadine grape juice as affected by DP-CO2 processing pressure (27.6, 38.3, and 48.3 MPa) and CO2 content (0, 7.5, and 15%). 1 Sum of individual 3,5anthocyani n diglucosides quantified by HPLC. 2 Means with similar letters within columns are not si gnificantly different (LSD test, P>0.05). Treatment % CO2 Delphinidin 3,5-glucoside (mg/L) Cyanidin 3,5-glucoside (mg/L) Petunidin 3,5-glucoside (mg/L) Peonidin 3,5-glucoside (mg/L) Malvidin 3,5-glucoside (mg/L) Total anthocyanins 1 (mg/L) Unprocessed 410 a2 201a 386a 159a 118 a 1275a 27.6 MPa 0% 295 cd 146ef 278cd117f 87.5 b 924d 7.5% 349 b 172bc 328b 136bc 102 ab1087b 15% 345 b 170bcd325b 135d 101 ab1075b 38.3 MPa 0% 319 bc 158de 300bc125e 93.8 b 996c 7.5% 351 b 173b 330b 137b 102 ab1093b 15% 321 bc 158cde302bc126e 94.3 b 1002c 48.3 MPa 0% 271 b 135f 256d 108 g 81.2 b 851e 7.5% 351 b 173b 331b 137 b 102 ab1095b 15% 346 b 170bcd326b 135 cd 101 ab1078b

PAGE 133

121 Figure 7-1. Effect of DP-CO2 at differ ent processing pressures (27.6, 38.3, and 48.3 MP a) and CO2 levels (0, 7.5%, and 15%) on residual PPO activity (A) and resultant anthocyanin losses (B) in muscadine grape juice. 27.6 MPa38.3 MPa48.3 MPa Residual PPO activity (%) 0 20 40 60 80 0% CO 2 7.5% CO 2 15% CO 2 ab a abc cd abc cd d d cd 27.6 MPa38.3 MPa48.3 MPa Total anthocyanin losses (%) 0 10 20 30 40 0% CO 2 7.5% CO 2 15% CO 2 b a c d c d d d dA B

PAGE 134

122 Figure 7-2. Effect of DP-CO2 at different processing pressu res (27.6, 38.3, and 48.3 MPa) and CO2 levels (0, 7.5%, and 15%) on PPOinduced losses in soluble phenolic s (A) and antioxidant capacity (B) in muscadine grape juice. 27.6 MPa38.3 MPa48.3 MPa Antioxidant capacity losses (%) 0 10 20 30 40 50 0% CO 2 7.5% CO 2 15% CO 2 a a b b b b b b b 27.6 MPa38.3 MPa48.3 MPa Total soluble phenolic losses (%) 0 10 20 30 40 50 0% CO 2 7.5% CO 2 15% CO 2 b a c e c de d de deA B

PAGE 135

Processing juices at 27.6 MPa and 0% CO2 resulted in a 40% decrease on initial PPO activity (Figure 7-1A), and this inactivat ion rate was insignificantly affected when processing pressures were increased to 38.3 and 48.3 MPa. Increasing processing CO2 levels from 0 to 7.5% served to additi onally reduce enzymatic activity by 35%, yet further increases in CO2 levels did not serve to re duce residual PPO activity. These changes in enzymatic activity were not related to variations in the pH of the system since insignificant changes in juice pH values (3.2) were observed among treatments right after DP-CO2 processing. Previous studies (Arreola, 1991a; Chen et al., 1992; Fadiloglu and Erkmen, 2002; Park et al., 2002; Tisi, 2004) have reported pH changes due to the formation of carbonic acid in aqueous systems pressurized with CO2. The insignificant changes in pH observed in this study might be attributed to the low concentrations of CO2 dissolved in the juices that re sulted when this gas was stripped from the juice by vacuum during the last stage of pro cessing (Figure 5-1). Juice CO2 content was 6.25 and 13.2 mM for treatments pressu rized at 7.5 and 15% CO2, respectively, and these levels varied insignificantly with processing pressure. In addition to the pH lowering mechanistic inactivation presented by CO2, previous studies have showed that CO2 can also react with enzyme-bound arginine to form a bicarbonate-pro tein complex that is responsible for the loss of enzymatic activity (Weder, 1984; Chen et al., 1992; Weder et al., 1992). Changes in PPO activity could also be attributed to conformational changes in enzyme structure, enzyme stability, and/or disruption of en zyme-substrate interactions during DP-CO2 and SC-CO2 as those described in previous studies (Chen et al., 1992; Ishi kawa et al., 1995; Fadiloglu and Erkmen, 2002; Tisi, 2004; Gui et al., 2005). For example, Chen et al. (1992) investigated changes in PPO activity associ ated with different high pressure CO2

PAGE 136

conditions (pressure and residence time) and concluded that pressure-induced inactivation was caused by changes in enzy me secondary structure and that the magnitude of conformational changes was de pendant in the source of PPO (lobster > shrimp > potato). Ishikawa et al. (1996) and Gui et al. (2005) also observed that decreases in enzyme activity caused by SC-CO2 were directly related to changes in enzyme -helix (secondary structure) as evidenced by sp ectroscopic, UV-circular dichroism, and tryptophan florescence analysis. Furthermore, Chen et al. (1992) showed that pressuredinduced PPO activation can also occur due to the ionization of peptide groups during processing. Although juices processed with diffe rent pressure levels at 0% CO2 presented similar enzymatic activity, losses of total an thocyanins (Figure 7-1B), soluble phenolics (Figure 7-2A), and antioxidant capacity (Figure 7-2B) differed among treatments and followed a bell curve shape that was dependant on processing pressure. When compared to their initial total anth ocyanin (1,275 mg/L), solubl e phenolics (2,183 mg/L), and antioxidant capacity (31.3 M TE/mL) content, higher losses were observed when juices were processed at 48.3 MPa followed by juices processed at 27.6 MPa and lastly juices processed at 38.3 MPa. Generally, increasing processing CO2 levels from 0 to 7.5% served to reduce losses in total anthocyanins (2-fold), soluble phenolics (1.6 to 2-fold), and antioxidant capacity (1.7 to 2-fold) for juices processed at 27.6 and 48.3 MPa. However, this protective effect was either mi nor or insignificant for juices processed at 38.3 MPa. Similarly than those trends observed for PPO activity, increasing the processing CO2 content to 15% did not offer additional polyphenolic and antioxidant retention. Within comparable pro cessing conditions (pressure and CO2 levels), similar

PAGE 137

degradation rates were observed among individu ally quantified anthoc yanins (Table 7-1) which contrasted with those trends of non-en zymatic degradation observed in chapters 5 and 6 where greater losses occurred for o -dihydroxy substituted anthocyanins with respect to their methoxylated c ounterparts. These differences could be attributed due to the specific mechanism of anthocyanin degradation when by-products formed by PPO oxidation are present in the food system and/or due to rate, sequence, and multiplicity of degradative reactions occurring simultane ously. Anthocyanin (r=0.78), polyphenolic (r=0.93), and antioxidant capacity (r=0.84) de gradation was highly correlated to PPO activity. Storage Stability of Muscadine Juice with Residual PPO Activity Most studies looking at the effects of HHP or DP-CO2 on phytonutrient stability only include evaluations after processing (Boff et al., 2003; Pa rk et al., 2002; Corwin and Shellhammer, 2002), and do not consider thei r stability during the shelf-life of the product. Therefore, this study investigated the polyphenolic and antioxidant changes associated with residual PPO activity during the refrigerated storage (4 C) of muscadine juice. Two DP-CO2 treatments (48.3 MPa at 0 or 15% CO2) were chosen for this aim and were selected because each presented the hi ghest and lowest inactivation rates following DP-CO2 processing and because the effect of CO2 processing levels can be best evaluated by maintaining a constant processing pressu re. Furthermore according to the results obtained in Chapter 5, both processing trea tments can achieve a > 5 log reduction of aerobic microorganisms and yeasts/molds and thus are more likely to be used in a commercial scale.

PAGE 138

Results demonstrated that independently of CO2 processing levels, juices that had residual PPO activity showed higher anthocya nin (8-fold), soluble phenolics (10-fold), and antioxidant (4-fold) degrad ation rates than control juices with no enzyme activity. Furthermore, control treatments retained hi gher total anthocyanins (620 mg/L), soluble phenolics (1,590), and antioxidant capacity (18.1 M Trolox equivalents/mL respectively; Figure 7-3) than counterpa rts containing residual enzymatic activity following DP-CO2 at the end of the storage. The rate of anthocyanin degradation considerably contrasted with those trends presented by soluble phenolics and antioxidant capacity degradation (Figure 7-3). Anthocyani ns presented gradual losses during the first two weeks of storage at 4 C, just like soluble phenolic s and antioxidant capacity throughout the entire shelf-life, while the rate of degradation of anthocyanins significantly increased after the second week of storage. Comparable degradation rates were observed among individual anthoc yanins (80%) present in DP-CO2 juices with residual PPO-activity, while the o -dihydroxylated anthocyanins, delphinidin and cyanidin, present in control juices showed greater losses (35%) than its methoxylated counterparts petunidin (20%), peonidin (23%) and malvidin (12%). Similarly to those trends observed in chapter 5 and 6, the concentration of CO2 utilized during DP-CO2 processing insignificantly affected anthocyani n stability during storage, while increasing CO2 from 0 to 15% offered enhanced storage stability for total soluble phenolics and antioxidant capacity. However, increasing the CO2 processing levels from 0 to 15% did offer enhanced anthocyanin stability for juices with residual PPO activity that was likely associated to decreases in enzymatic ac tivity and/or protection of polyphenolic compounds during refrigerated storage.

PAGE 139

These results clearly dem onstrate that residual enzymatic activity and/or the production of PPO degradation by-products were responsible for the enhanced oxidative degradation of these juices. However, ther e was not a clear relationship between the residual activity following DP-CO2 and the storage stability of the juices. Based on residual activity after DP-CO2, it was expected that juices processed at 48.3 MPa and 0% CO2 presented higher rates of polyphenolic and an tioxidant degradation than counterparts processed at 15% CO2. However, similar degradati on rates were observed for both treatments. One factor that could have contri buted to this particul ar situation is the reversibility of enzymatic activity after a certain time following DP-CO2 processing. Previous studies have shown that reversible changes in protein structure can occur during and following DP-CO2 and thus enzyme activity can vary depending on processing conditions such as pressure, CO2 levels, temperature and length of processing and storage (Fadiloglu and Erkmen, 2002). For example, Gu i et al. (2005) showed that horseradish peroxidase treated with SC-CO2 (< 30 MPa) recovered its initial activity as well as its initial conformational structur e after a 7 day storage at 4 C. Other studies have also corroborated that pressures 310 MPa generally produce changes in enzymatic conformation that are revers ible upon pressurization af ter a short storage period. However, the degree of this reversibility is highly dependant on the type of enzyme, enzyme source, nature of food matrix, temper ature, time, pH of the food system, etc. (Arreola, 1991b; Chen et al., 1992; Fadiloglu and Erkmen, 2002; Park et al., 2002; Tisi, 2004). Another factor that could have contributed to the reduc tion of enzyme activity, is the temperature used for the shelf-life of the juices, since temperatures < 10 C are known

PAGE 140

to negatively impact the stability and ac tivity of an enzyme (Ishikawa et al., 1995; Fadiloglu and Erkmen, 2002; Pa rk et al., 2002; Tisi, 2004). Conclusions The present study showed that partial in activation of PPO can be obtained by DPCO2. Processing CO2 levels was the main processing va riable influencing PPO activity as well as polyphenolic and antioxida nt retention in muscadine ju ice. Results also concluded that PPO residual activity during the refrigerated storage of th e juices was responsible for their enhanced polyphenolic and antioxidant degradation and that again processing CO2 levels was instrumental in decreasing these losses.

PAGE 141

Figure 7-3. Total anthocyanin (A), soluble phenolics (B), and antioxidant capacity (C) content of DP-CO2 processed (48.3 MPa) musc adine grape juice during refrigerated storage (4 C) as affected by processing CO2 content (0 or 15%) and initial PPO activity (No Activity=Control, and residual activity following DP-CO2). 01234 Total anthocyanin content (mg/L) 200 400 600 800 1000 1200 0% CO2, Control 0% CO2, PP0 15% CO2, Control 15% CO2, PPO 01234 Total soluble phenolics (mg/L) 0 500 1000 1500 2000 Storage time (weeks) 01234 Antioxidant capacity ( M TE/mL) 0 10 20 30 A B C

PAGE 142

130 CHAPTER 8 SUMMARY AND CONCLUSIONS Due to the need for novel, inexpensive, and stable sources of anthocyanin pigments, the stability of aai anthocyanins as a new source of these pigments was established in the present study and can be us ed to determine application and functional properties of aai in a variety of food and nutra ceutical products. The effect of naturally occurring cofactors on anthocyanin co lor stability was also determined. Addition of cofactors isolated from thym e was proven to not only increased juice color and antioxidant activity but also reduced anthocyanin, polyphenolic, and ascorbic acid losses. Copigmentation did not influence the sensory pr operties of the juices yet increased the red color intensity and preven ted the color fading that occurred during storage. Likewise, the DP-CO2 process served to protect anthocyanins and antioxidant levels without comprising the sensory attri butes of a muscadine juice. However, the microbial stability of DP-CO2 juices was significantly lower when compared to heatpasteurized counterparts. This study also showed that the main DP-CO2 processing variable influencing microbial and enzyma tic inactivation, as well reducing polyphenolic and antioxidant losses, was processing CO2 levels. The combination of DP-CO2 and copigmentation was proven to be instrument al in increasing the phytochemical stability of muscadine juice and can be used by the f ood and nutraceutical indus try as a strategy to reduce phytochemical losses and increase the commercial application of these pigments.

PAGE 143

131 LIST OF REFERENCES Abuja, P.M., Murkovic, M., Pfannhauser, W. Antioxidant and prooxi dant activities of elderberry ( Sambucus nigra ) extract in low-density lipop rotein oxidation. J. Agric. Food Chem., 1998, 46, 4091-4096. Anese, M., Nicoli, M.C., Dallaglio, G., Lerici, C. Effect of high pressure treatments on peroxidase and polyphenoloxidase activ ities. J. Food Biochem. 1995, 18, 285-293. Arreola, A. G. Effects of s upercritical carbon dioxide on some quality attributes of single strength orange juice. 1990. MasterÂ’s Thesis. University of Florida. Arreola, A G, Balaban, M. O., Marshall, M. R., Pepl ow, A. J., Wei, C. I., Cornell, J. A. Supercritical carbon dioxide effects on so me quality attributes of single strength orange juice. J. of Food Sci. 1991a, 56, 1030-1033. Arreola A. G., Balaban, M. O., Wei, C. I., Peplow, A., Marshall, M. Effect of supercritical carbon dioxide on microbial popu lations in single strength orange juice. J. of Food Qual. 1991b, 14, 275-284. Asaka, M., Aoyama, Y., Ritsuko, N., Hayashi, R. Purification of a latent form of polyphenoloxidase from La France pear frui t and its pressure-activation. Biosci. Biotech. Biochem. 1994, 58, 1486-1489. Asen, S., Stewart, R.N., Norris, K.H. Copigmen tation of anthocyanins in plant tissues and its effects on color. Phytochem. 1972, 11, 1139-1144. Balaban, M. O., Arreola, A. G., Marshall, M. R., Peplow, A., Wei, C. I., Cornell, J. Inactivation of pectin esterase in orange juice by s upercritical carbon dioxide. J. Food Sci. 1991, 56, 743-746. Ballestra, P., Abreu Da Silva, A., Cuq, J. L. Inactivation of Escherichia coli by carbon dioxide under pressure. J. Food Sci. 1996, 61, 829-831. Barbosa-Canovas, G. V. B., Pothakamury, U. R., Palou, E., Swanson, B. G. Nonthermal preservation of foods Fennema, O. R., Karel, M., Sanderson, G. W., Tannenbaum, S. R., Walstra, P., Whitaker, J. R.. Marcel Dekker Inc., New York, NY, 1998, 1203. Bartolem, B., Bengoeshea, M.L., Glvez, M.C., Prez-Ilzarbe, F.J., Hernndez, T. Photodiode array detection for elucidati on of the structure of phenolic compounds. J. Chrom. 1993, 655, 119-125.

PAGE 144

132 Bartolem, B., Hernndez, T., Bengoeshea, Qu esada, C., Gmez-Cordovs, C., Estrella, I. Determination of some structural feat ures of procyanidins and related compounds by photodiode-array detection. J. Chrom. 1996, 723, 19-26. Basak, S.,Ramaswamy, H. S. Ultra high pressu re treatment of ora nge juice: a kinetic study on inactivation of pectin methyl esterase. Food Res. Int. 1996, 29, 601-607. Basile, A., Jimenez-Carmona, M.M., Clifford, A.A. Extraction of rosemary by superheated water. J. Agric. Food Chem. 1998, 46 5205-5209. Boff, M., Truong, T.T, Min, D.B., Shellhammer, T.H. Effect of thermal processing carbon dioxide-assisted highprocessing on pectinmethyles terase chemical changes in orange juice. J. Food Sci. 2003, 68, 1179-1184. Boulton, R. The copigmentation of anthocyanins and its role in the color of red wine: a critical review. Am. J. Enol. Viticult. 2001, 52, 67-86. Bradshaw, M. P., Prenzler, P. D., Scollary, G. R. Ascorbic acid-induced browning of (+)catechin in a model wine system. J. Agric. Food Chem., 2001, 49, 934-939. Brenes, C.H., Talcott, S.T., Covarrubias, A.F., Del Pozo-Insfran, D. Stability of copigmented grape anthocyanins ( Vitis vinifera ) in the presence of polyphenoloxidase. To be submitted to the J. Agric. Food Chem., 2006. Brenes, C.H., Del Pozo-Insfran, D., Talcott, S.T. Stability of copigmented anthocyanins and ascorbic acid in a gr ape juice model system. J. Agric. Food Chem. 2005, 53, 49-56. Buglione, M., Lozano, J. Nonenzymatic brow ning and chemical changes during grape juice storage. J. Food Sci. 2001, 67, 1538-1543. Butz P., Tauscher B. Emerging technologies : chemical aspects. Food Res. Int. 2002, 35, 279-284. Butz P., Fernndez Garca A., Lindauer R ., Dieterich, S., Bognr, A., Tauscher, B. Influence of ultra high pressure processi ng on fruit and vegetable products J. Food Eng. 2003, 56, 233-236. Butz, P., Fernandez A., Fister H, Tauscher b. Influence of high hydrostatic pressure on aspartame: instability at neutral pH J. Agric. Food Chem. 1997, 45, 2302-2303. Butz, P., Koller, W.D., Tauscher, B., Wolf, S. Ultra-high pressure processing of onions: chemical and sensory changes. Le bensm. -Wiss u. Technol. 1994, 27, 463-467. Calvi, J.P., Francis, F. J. Stability of Concord grape ( V. labrusca ) anthocyanins in model systems. J. Food Sci 1978 43, 1448-1456.

PAGE 145

133 Cano, M.P., Hernandez, A., De Ancos, B. High pressure and temperature effects on enzyme inactivation in strawberry and orange products. J. Food Sci. 1997, 62, 8588. Castellari, M., Matricardi, L., Arfelli, G., R overe, P., Amati, A. Effects of high pressure processing on polyphenoloxida se enzyme activity of grape musts. Food Chem. 1997, 60, 647-649. Chen, J.S., Balaban, M.O., Wei, C., Mars hall, M.R., Hsu, W.Y. Inactivation of polyphenol oxidase by high-pressure carbon dioxide. J. Agric. Food Chem., 1992, 40, 2345-2349. Chen, J.S., Balaban, M.O., Wei, C., Mars hall, M.R., Hsu, W.Y. Inactivation of polyphenol oxidase by high-pressure carbon dioxide. J. Agric. Food Chem. 1992, 40, 2345-2349. Choi, M.H., Kim, G.H., Lee, H.S. Effects of ascorbic acid retention on juice color and pigment stability in blood orange ( Citrus sinensis ) juice during refrigerated storage. Food Res. Intl. 2002, 35, 753-759. Chowdhury, M. A. R., Ravel, J. Hill, R. T. Huq, A., Colwell, R. R. Physiology and molecular genetics of viable but non-culturable microorganisms. In: Biotechnology risk assessment: USEPA/USDA environment Canada Levin, M., Grim, C., Angle, J.S. University of Maryland, College Park, MD, 1994. Clegg, M.K., Morton, A.D. Phenolic com pounds of black curran t juice and their protective effect on ascorbic acid. J. Food Tech. 1968, 3, 227. Clifford, M. N. Review: Anthocyanins – natu re, occurrence and dietary burden. J. Sci. Food Agric. 2000, 80, 1063-1072. Clifford, J. R., Williams, A.A. Supercritical fluid methods and protocols Humana Press, Totowa, NJ, 2000. Colwell1, R.R., Brayton, P. R., Grimes, D. J.; Roszak, D. B.; Huq, S. A., Palmer, S. A. Viable but non-culturable Vibrio cholerae and related pathogens in the environment: implications for release of genetically engineered microorganisms. Biotech. 1985, 3, 817-820. Corwin, H., Shellhammer, T.H. Combined ca rbon dioxide and high pressure inactivation of pectin methyleste rase, polyphenol oxdiase, Lactobacillus plantarum and Escherichia coli J. Food Sci. 2002, 67, 697-701. Covarrubias, A.F. Evaluation of the stability of anthocyanins copigmented with rosemary ( Rosmarinus officinalis ) phenolic extracts in the pr esence of polyphenoloxidase. Mater’s Thesis, 2002, ITESM, Campus Monterrey, Mexico.

PAGE 146

134 Darnton-Hill, I. Technical aspe cts of food fortification. In Food and Nutrition Bulletin. United Nations University Press, Boston, MA, 1998. Daniels, J. A. Krishnamurthi, R., Rizvi, S. S. H. A review of effects of carbon dioxide on microbial growth and food quality. J. Food Prot. 1984, 48, 532-537. Del Pozo-Insfran, D., Brenes C.H., Talcot t, S.T., Cantu, H. Determination of antimicrobial properties of anthocyanins naturally stabilized by flavonoids. 2002. Presented at the Annual Meeting of the In stitute of Food Technologists, Anaheim, CA. Del Pozo-Insfran, D., Brenes, C.H., Talcott, S.T., Covarrubias A.F. Stability of copigmented anthocyanins and ascorbic acid in juice model systems. 2002. Presented at the Annual Meeting of the In stitute of Food Technologists, Anaheim, CA. Del Pozo-Insfran, D., Talcott, S.T., Bren es, C.H. Phytochemical composition and pigment stability of Aai ( Euterpe oleracea ). J. Agric. Food Chem. 2004, 54, 15391545. Del Pozo-Insfran D., Del Follo-Martinez A., Br enes C. H., Talcott S. T. Stability of Fortified and Copigmented Muscadine Grape Juice as Affected by High Hydrostatic Pressure (HHP) Processing. Accepted for publication in Food Res. Intnl. 2006. Dufour, C., Sauvaitre, I. Inte ractions between anthocyanins and aroma substances in a model system: effect on the flavor of gr ape-derived beverages. J. Agric. Food Chem. 2000, 48, 1784-1788. Ehlenfeldt, M. K., Prior, R. L. Oxygen ra dical absorbance capacity (ORAC) and phenolic and anthocyanin concentrations in fruit a nd leaf tissues of highbush blueberry. J. Agric. Food Chem. 2001, 49, 2222-2227. Eiro, M. J., Heinonen, M. Anthocyanin color be havior and stability dur ing storage: effect of intermolecular copigmentation. J. Agric. Food Chem. 2002, 50, 7461-7466. Enomoto, A., Nakamura, K., Hakoda, M., Amaya, N. Lethal effect of high-pressure carbon dioxide on a bacterial spore. J. Ferment Bioeng, 1997, 83, 305-307. Erkman, O. Predictive Modeling of Listeria Monocytogenes inactivation under high pressure carbon dioxide. Lebensm.-Wiss. u.-Technol. 2000a. 33, 514-519. Erkman, O. Effect of carbon dioxide pressure on Listeria Monocytogenes in physiological saline and foods Food Microb. 2000b, 17, 589-596. Erkman, O. Antimicrobial effect of pressurized carbon dioxide on Enterococcus faecalis in physiological saline and foods. J. Sci. Food Agric. 2000c, 80, 465-470.

PAGE 147

135 Espn, J.C., Soler-Rivas, C., Wichers, H.J ., Garca-Viguera, C. Anthocyanin-based natural colorants: a new sour ce of antiradical activity fo r foodstuff. J. Agric. Food Chem. 2000, 48, 1588-1592. Es-Safi, N., Cheynier, V.R., Moutounet, M. Role of aldehydic derivatives in the condensation of phenolic compounds with em phasis on the sensorial properties of fruit-derived foods. J. Ag ric. Food Chem. 2002, 50, 5571-5585. Es-Safi, N., Fulcrand, H., Cheynier, V., Moutounet, M. Studies on the acetaldehydeinduced condensation of (-)-epicatechin and malvidin 3O -glucoside in a model solution system. J. Agric. Food Chem. 1999, 47, 2096-2102. Eshtiaghi, M.N., Stute, R., Knorr, D. High pressure and freezing pretreatment effects on drying, rehydration, texture a nd color of green beans, car rots and potatoes. J. Food Sci. 1994, 59, 1168-1169. Fad loglu, S., Erkmen, O.Inactivation of lipase by carbon dioxide under atmospheric pressure. J. Food Eng. 2002, 52, 331–335. Flora, L.F. Processing and quality characteri stics of muscadine grapes. J. Food Sci. 1978, 42, 935-938, 952. Frankel, E.N., Bosanek, C.A., Meyer, A.S., Silliman, K., Kirk, L.L. Commercial grape juices inhibit the in vitro oxidation of human low-density lipoproteins. J. Agric. Food Chem. 1998, 46, 834-838. Frankel, E. N., Waterhouse, A.L., Teissedre, P.L. Principal pheno lic phytochemicals in selected California wines and their anti oxidant activity in in hibiting oxidation of human low-density lipoproteins. J. Agric. Food Chem. 1995 43, 890-894. Fraser D. Bursting bacteria by releas e of gas pressure. Nature. 1951, 167, 33-37. Garcia-Viguera, C., Bridle, P. Influence of structure on color stabil ity of anthocyanins and flavylium salts with ascorb ic acid. Food Chem. 1999, 64, 21-26. Garzon, G.A., Wrolstad, R.E. Comparison of the stability of pelargonidin-based anthocyanins in strawberry jui ce and concentrate. J. Food Sci 2002 67, 12881299. Ghiselli, A., Nardini, M., Baldi, A., Scacci ni, C. Antioxidant activity of different phenolic fractions separated from an It alian Red wine. J. Agric. Food Chem. 1998, 46, 361-367. Giusti, M. M., Wrolstad, R. E. Acylated anthocyanins fr om edible sources and their applications in food systems. Biochem. Eng. J. 2003, 14, 217-225.

PAGE 148

136 Glassen, W.E., Rose, A., Madlung, J., Kock, W ., Gleitz, J., Seitz, H.U. Regulation of enzymes involved in anthocyanin biosynthesi s in carrot cells cultu res in response to treatment with ultraviolet light and fungal elicitors. Planta. 1998, 204, 490-498. Gokmen, V., Kahraman, N., Demir, N., A car, J. Enzymatically validated liquid chromatographatic method for the determin ation of ascorbic and dehydroascorbic acids in fruit and vegetabl es. J. Chrom. A. 2000, 881, 309-316. 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. Gomez, M.R.A., Ledward, D.A. Effect of high pressure treatments on the activity of some polyphenoloxidases. Food Chem. 1996, 56, 1-5. Gould, G. W. Methods for preservation and ex tension of shelf life. Int. J. Food Microb. 1996, 33, 51-64. Haas, G. J., Prescott, H. E., Dudley, E., Dik, R., Hintlian, C., Keane, L. Inactivation of microorganisms by carbon dioxide under pressure. J. Food Safety. 1989, 9, 253265. Heinonen, I.M., Meyer, A.S., Frankel, E.N. Antioxidant activity of berry phenolics on human low-density lipoprotein and lipos ome oxidation. J. Agric. Food Chem. 1998, 46, 4107-4112. Hendrickx M., Ludikhuyze L., Van den Broeck I ., Weemaes C. Effects of high pressure on enzymes related to food. Trends Food Sci. Tech. 1998, 5, 197-203. Hong, V., Wrolstad, R.E. Use of HPLC se paration/photodiode array detection for characterization of anthocyanins. J. Agric. Food Chem. 1990, 38, 708-715. Howard, L. R., Talcott, S. T., Brenes, C. H., Villalon, B. Changes in phytochemical and antioxidant activity of sele cted pepper cultivars ( Capsicum Species) as influenced by maturity. J. Agric. Food Chem. 2000, 48, 1713-1720. Hrazdina, G.;,Borzell, A. J., Robinson, W. B. Studies on the stability of the anthocyanin3,5-diglucosides. Am. J. Enol. Vitic. 1970, 21, 201. Iacobucci, G.A., Sweeny, J.G. The chemistry of anthocyanins and related flavylium salts. Tetrahedron. 1983, 23, 1057-1064 Irwe, S., Olsson, I. Reduction of pectinesterase activity in orange ju ice by high pressure treatment. In Minimal processing of foods and pr ocess optimizati on: an interface. CRC Press, Boca Raton, FL, 1994, 35-42. Isenschmid, A., Marison I. W., von Stockar, U. The influence of pressure and temperature of compressed CO2 on the survival of yeast cells. J. Biotech. 1995, 39, 229-237.

PAGE 149

137 Ishikawa, H., Shimoda, M., Shiratsuchi, H., Os ajima, Y. Sterilization of microorganisms by the supercritical carbon dioxide micro-bubble method. Biosci. Biotech. Biochem. 1995, 59, 1949-1950. Jackman, R.L., Smith, J.L. Anthocyanins and betalains. In: Natural food colorants Hendry, G.A.F., Houghton, J.D. Blackie A&P Press, London, England, 1996, 245280. Jurd, L. Some advances in the chemistry of anthocyanins-type pigments. In: The Chemistry of Plant Pigments Chichester, C.O. Academic Press, New York:, NY, 1972, 123-142. Kaack, K., Austed, T. Interaction of vitamin C and flavonoids in elderberry ( Sambucus nigra L .) during juice processing. Plant Foods Hum. Nutr. 1998, 52, 187-198. Kader, F., Irmouli, M., Nicolas, Metche, M. Involvement of blueberry peroxidase in the mechanism of anthocyanin degradation in blueberry juice. J. Food Sci. 2000, 67, 910-915. Kader, F.,Nicolas, J.P., Metche, M. Degrad ation of pelargonidin 3-glucoside in the presence of chlorogenic acid and blueberry polyphenol oxidase. J. Sci. Food Agric. 1999, 79, 517-522. Kader, F., Rovel, B., Girardin, M., Metche, M. Mechanism of brow ning in fresh highbush blueberry fruit ( Vaccinium corymbosum L.). Role of blueberry polyphenol oxidase, chlorogenic acid and anthocyanins. J. Sci. Food Agric. 1997 74, 31-34. Kalt, W., Forney, C. F., Martin, A., Prior, R. L. Antioxidant capacity, vitamin C, phenolics, and anthocyanins after fresh st orage of small fruits. J. Agric. Food Chem. 1999, 47, 4638-4644. Kamihira, M., Taniguchi, M., Kobayasbi, T. Sterilization of microorganisms with supercritical carbon dioxide. Ag ric. Biol. Chem. 1987, 51, 407-412. Kemmerer, M. Copigmentation and lipoxygenase inhibiting propertie s of anthocyanins. MasterÂ’s Thesis, 2002, Univ ersity of Florida. Kim, Y., Park, S., Park, J. Effects of combin ed treatment of high hydr ostatic pressure and mild heat on the quality of carro t juice. J. Food Sci. 2001, 66, 1355-1360. Krebbers B., Matser A. M., Hoogerwerf S. W., Moezelaar R., Tomassen M. M. M., van den Berg R. W. Combined high-pressure and thermal treatments for processing of tomato puree: evaluation of microbial inactivation and quality parameters. Innov. Food Sci. Emerg. Tech. 2003, 4, 377-385. Kumagai, H., Hata, C., Nakamura, K. CO2 sorption by microbial cells and sterilization by high-pressure CO2. Biosci. Biotech. Biochem. 1997, 61, 931-935.

PAGE 150

138 Lado B. H., Yousef A. E. Alternative f ood-preservation technologies: efficacy and mechanisms Microbes and Infection. 2002, 4, 433-440. Lamikanra, O. Enzymatic browning of muscadine grape products. In: Enzymatic browning and its prevention Lee, H, Whitaker,L. American Chemical Society, Washington, DC, 1995, 166-177. Lapidot, T., Harel, S., Akiri, B., Granit, R., Kanner, J. pH-dependent forms of red wine anthocyanins as antioxidants. J. Agric. Food Chem. 1999, 47, 67-70. Lin, H. M. and Lin, E. T. Inactivations of organisms with supe rcritical fluids. In: Science for the food industry of the 21st century, biotechnology, supercritical fluids, membranes and other advanced technologi es for low calorie, healthy food alternatives Yalpani, M. ATL Press Inc ., Mount Pleasant, MA, 1993, 254-264. Lin, H. M., Cao, N., Chen, L. Antimicrobial effect of pressurized carbon dioxide on Listeria monocytogenes J. Food Sci. 1994, 59, 657-659. Lopez-Sebastian, S., Ramos, E., Ibanez, E., Bueno, H.M., Ballester, L., Tabera, J., Reglero, G. Dearomatization of Antioxida nt Rosemary Extracts by Treatment with Supercritical Carbon Dioxide. J. Agric. Food Chem. 1998, 46 13-19. Malien-Aubert, C., Dangles, O., Amiot, M. J. Color stability of commercial anthocyaninbased extracts in relation to the phenolic composition: Protective effects by intraand intermolecular copigmentation. J. Agric. Food Chem. 2001, 49 170-176. Malien-Aubert, C., Dangles, O., Amiot, M. J. Influence of procyanidins on the color stability of oenin solutions. J. Agric. Food Chem. 2002, 50, 3299-3305. Mateus, N., Carvalho, E., Carvalho, A. R. F., Melo, A., Gonzalez-Paramas, A. M., Santos-Buelga, C., Silva, A. M. S., de Freitas, V. Isolation and structural characterization of new acylated anthocyani n-vinyl-flavanol pigments occurring in aging red wines. J. Agric. Food Chem. 2003, 51, 277-282. Markakis, P. Stability of anthocyanins in foods. In Anthocyanins as Food Colors Markakis, P. Academic Press, Inc, New York, NY, 1982, 163-180. Matser A.M., Krebbers B., van den Berg R.W., Bartels P. V. Advantages of high pressure sterilization on quality of food products. Trends Food Sci. Tech. 2004, 15, 79-85. Mazza, G., Brouillard, R. The m echanism of co-pigmentation of anthocyanins in aqueous solutions. Phytochem.1990, 29, 1097-1102. McGuinness, J. Isoflavonoid copigmentation reactions and the functional properties of muscadine wine. MasterÂ’s Thesis 2002, University of Florida.

PAGE 151

139 Meyer, A.S., Heinonen, M., Frankel, E. N. Antioxidant interactions of cathechin, cyandin, caffeic acid, quercetin, and e llagic acid on human LDL oxidation. Food Chem.1998, 861, 71-75. Meyer, A.S., Yi, O. Pearson, D.A., Waterhouse, A.L., Frankel, E.N. Inhibition of human low-density lipoprotein in relation to composition of phenolic antioxidants in grapes ( Vitis vinifera ). J. Agric. Food Chem. 1997 45, 1638-1643. Muniz-Miret, N., Vamos, R., Hiraoka, M., Montagnini, F., Medelsohn, R.O. The economic value of managing the aai palm ( Euterpe oleracea Mart.) in the floodplains of the Amazon estuary, Para, Brazil. Forest Ecol. Manage. 1996, 87, 163-173. Murrieta, R. S. S., Dufour, D.L., Siqueira, A. D. Food consumption and subsistence in three caboclo populations on Marajo Island, Amazonia, Brazil. Human Ecol. 1999, 27, 455-475. Nakamura, K., Enomoto, A., Fukushima, H., Nagai, K., and Hakoda, M. Disruption of microbial cells by the flash discharge of high-pressure car bon dioxide. Biosci. Biotech. Biothem., 1994, 58, 1297-1301. Nychas, G. J. E. Natural antimicrobials from plants. In: New methods of food preservation Gould, G. W. Blackie Academic & Professional, London, U.K., 1995, 58-89. Ottaway, P.B. Stability of vitamins in food. In The technology of vitamins in food Ottaway, P.B. Blackie Academic & Pr ofessional, Glasgow, Scotland, 1993, 90-112. Ou, B., Hampsch-Woodill, M., Prior, R.L. Development and validation of an improved oxygen radical absorbance capac ity assay using fluoresce in as the fluorescent probe. J. Agric. Food Chem 2001, 49 4619-4626. zkan, M., Yemenicioglu A., Asefi N., Cemeroglu B. Degradation kinetics of anthocyanins from sour cherry, pomegra nate, and strawberry juices by hydrogen peroxide. J. Food Sci. 2002, 67, 525-529. Park, S. J., Lee, J. I., Park, J. Effects of a combined process of high pressure carbon dioxide and high hydrostatic pressure on the quality of carrot juice. J. Food Sci. 2002, 67, 1827-1834. Park, S.J., Lee, J.I., Park, J. Inactivati on kinetic of food poisoning microorganisms by carbon dioxide and high hydrostatic pr essure. J. Food Sci. 2003, 48, 896-981. Poei-Langston, M.S., Wrolstad, R.E. Color de gradation of ascorbic-anthocyanin-flavanol model system J. Food Sci 1981 46, 1218-1222.

PAGE 152

140 Polydera A. C., Stoforos N. G., Taoukis P. S. Comparative shelf life study and vitamin C loss kinetics in pasteurized and high pressure processed reconstituted orange juice. J. Food Eng. 2003, 60, 21-29. Prior, R. L., Hoang, H., Gu, L., Wu, X., Bacchiocca, M., Howard, L., Hampsch-Woodill, M., Huang, D., Ou, B., Jacob, R. Assays for hydrophilic and lipophilic antioxidant capacity (oxygen radical ab sorbance capacity (ORACFL)) of plasma and other biological and food samples. J. Agric. Food Chem. 2003, 51, 3273-3279. Rauha, J., Remes, S., Heinonen, M., Hopia, A., Kahkonen, M., Kujala, T., Pihlaja, K., Vourela, P. Antimicrobial effects of Fi nnish plant extracts containing flavonoids and other phenolic compounds. Int. J. Food Microb. 2000, 56, 3-12. Richardson, D.P. Food fortification. In: Technology of vitamins in foods Ottaway, P.B. Blackie Academic and Profe ssional, Glasgow, U.K. 1993, 233-45. Robinson, G.M., Robinson, R. A survey of anthocyanins. Intl. Biochem. J. 1931, 25, 1687-1705. Rodriguez-Saona, L. E., Giusti, M. M., Wrol stad. Color and pigment stability of red radish anthocyanins and redfleshed potato anthocyanins in juice model systems. J. Food Sci. 1999, 61, 688-694. Sagdic, O. Sensitivity of four pathogeni c bacteria to Turkish thyme and oregano hydrosols. Lebensm.-Wiss. u.-Technol. 2003, 36, 467–473. Sanchez-Moreno C., Satue-Gracia C., Frankel, E.N. Antioxidant activity of selected Spanish wines in corn oil emulsions. J. Agric. Food Chem 2000, 48 5581-5587. 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. Seyderhelm I., Boguslawski S., Michaelis G., Knorr D. Pressure induced inactivation of selected food enzymes. J. Food Sci. 1996, 61, 308-310. Shrinkhande, A.J., Francis, F.J. Effect of flavonols on ascorbic acid and anthocyanin stability in model systems. J. Food Sci. 1979, 39, 904-906. Silva, S. Fruit in Brazil Dados Internacionais de Catal ogacao na Publicao, Sao Paulo, Brazil, 1996, 14-17. Skrede, G., Wrolstad, R.E., Durst, R.W. Changes in anthocyanins and polyphenolics during juice processing of highbush blueberries ( Vaccinium corymbosum L.). J. Food Sci. 2000, 65, 357-364.

PAGE 153

141 Simonne, A.H., Eitenmiller, E.E. Retention of vitamin E and added retinyl palmitate in selected vegetable oils during deep-fat frying and in fried breaded products. J. Agric. Food Chem. 1998, 46, 5273-5277. Smelt, J. P. P. M. Recent advances in the microbiology of high pressure processing. Trends Food Sci. Tech. 1998, 9, 152-158. Spilimbergo, S., Elvassore, N., Bertucco, A. Mi crobial inactivation by high-pressure. J. Supercritical Fluids. 2003a, 22, 55-63. Spilimbergo, S., Elvassore, N., Bertucco, A. Inactivation of microorganisms by supercritical CO2 by a semi-continuous process. Italian J. Food Sci. 2003b, 1, 115124. Stintzing, F. C., Stintzing, A. S ., Carle, R., Frei, B., Wrolstad, R. E. Color and antioxidant properties of cyanidin-based anthocyanin pigments. J. Agric. Food Chem. 2002, 50, 6172-6181. Stoforos, N.G., Taoukis, P.S. A theoretical procedure for using multiple response timetemperature integrators for the design a nd evaluation of thermal processes. Food Control. 1998, 9, 279-287. Sun, N., Lee, S., Bin Song, K. Effect of high-pressure treatment on the molecular properties of mushroom polyphenoloxidase. Lebens. Wiss. Tech nol. 2002, 35, 315318. Talcott, S.T., Howard, L.R., Brenes, C.H. An tioxidant changes and sensory properties of carrot puree processed with and without periderm tissue. J. Agric. Food Chem. 2000, 48, 1315-1321. Talcott, S. T., Lee, J.H. Ellagic acid and flavonoid antioxidant content of muscadine wine and juice. J. Agric. Food Chem. 2001, 50, 3186-3192. Talcott, S.T., Brenes, C.H., Pires, D.M., Del Pozo-Insfran, D. Phytochemical and color retention of copigmented and processed muscadine grape juice. J. Agric. Food Chem. 2003, 51, 957-963. Talcott, S.T., Percival, S.S., Pittet-Moore, J., Celoria, C. Phytochemical composition and antioxidant stability of for tified yellow passion fruit ( Passiflora edulis ). J. Agric. Food Chem. 2003, 51, 935–941. Talcott, S.T., McGuinness, J.E., Brenes, C.H. Red clover isoflavonoids act as copigments to muscadine grape anthocyani ns. Food Res. Intnl. 2005. 38, 1205-1212. Taoukis, P.S., Labuza, T.P., Saguy, I.S. Kine tics of food deteriora tion and shelf-life prediction. In: Handbook of food engineering practice Valentas, K.J., Rotstein, E., Singh, R.P. CRC Press, NewYork, NY, 1997, 361-403.

PAGE 154

142 Tauscher, B. Effect of high pressure treat ment to nutritive substances and natural pigments. In: Fresh novel foods by high pressure Autio, K. VTT Technical Research Centre of Finland, Espoo, Finland, 1998, 83-95. Taylor, L.T. Supercritical flu id extraction John Wiley and Sons, Inc., New York, NY, 1996. Tedjo, W., Eshtiaghi, M. N., Knorr, D. Im pact of supercritical carbon dioxide and high pressure on lipoxygenase and peroxida se activity. J. Food Sci. 2000, 65, 12841287. Tisi, D.A. Effect of dense phase carbon dioxide on enzyme activity and whey proteins in raw milk. MasterÂ’s Thesis, 2004, Cornell University, Ithaca, NY. Vahteristo, L.T., Ollilainen, V., Koivistoinen, P.E., Varo, P. Improvements in the analysis of reduced folate monoglutamates and folic acid in food by high-performance liquid chromatography. J. Ag ric. Food Chem. 1996, 44, 477-482. Wang, S. Y., Lin, H. S. Antioxida nt activity in fruits and l eaves of blackberry, raspberry, and strawberry varies with cultivar and developmental stage. J. Agric. Food Chem. 2000, 48, 140-146. Wang, S. Y., Stretch, A. W. Antioxidant capacity in cranberry is influenced by cultivar and storage temperature. J. Agric. Food Chem. 2001, 49, 969-974. Weder, J. K. P. Studies on proteins and amino acids exposed to supercritical carbon dioxide extraction conditi ons. Food Chem. 1984, 15, 175-190. Weder, J. K. P., Bokor, M. V., Hegarty, M. P. Effect of supercritical carbon dioxide on arginine. Food Chem. 1992, 44, 281-290. Weemaes, C.A., De Cordt, S.V., Ludikhuyze, L.R., Van den Broeck, I., Hendrickx, M.E., Tobback, P.P. Influence of pH, benzoi c Acid, EDTA, and glutathione on the pressure and or temperature inactivati on kinetics of mushroom polyphenoloxidase. Biotechnol. Progr. 1997, 13, 25-32. Wei, C. I., Balaban, M. O., Fernando, S. Y ., Peplow, A. J. Bacterial effect of high pressure carbon dioxide treatment on foods spiked with Listeria or Salmonella J. Food Prot. 1990, 54,189-193. Wesche-Ebeling, P., Montgomery, M.W. Stra wberry polyphenoloxidase: its role in anthocyanin degradation. J. Food Sci. 1990 55, 731-734. Wouters, P.C., Glaasker E., Smelt J. Effects of high pressure on inac tivation kinetics and events related to proton efflux in Lactobacillus plantarum Appl. Environ. Microb. 1998, 64, 509-514.

PAGE 155

143 Wrolstad, R.E. Color and pigment analysis in fruit products. Oregon Agric. Exp. Stn. Corvallis. 1976, Bulletin 624. 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. Zheng, W., Wang, S.Y. Antioxidant activity and phenolic compounds in selected herbs. J. Agric. Food Chem. 2001, 49 5165-5170.

PAGE 156

144 BIOGRAPHICAL SKETCH David Del Pozo Insfran received his first BachelorÂ’s degree from ITESMMonterrey Tech., majoring in Chemical Engineering with a minor in Business Administration. Two years later he received his second bachelorÂ’s degree majoring in Food Science and Technology with honor s and a minor in Food Marketing and Agribusiness Management from ITESM-Monterrey Tech. He then came to the University of Florida in the Fall of 2002 after r eceiving a post-gradua te degree in Food Biotechnology from ITESM-Monterrey Tech in conjunction with the University of British Columbia (Vancouver, Canada). David will graduate in the Spring of 2006 with a Ph.D. in Food Science and Human Nutrition and also with a MasterÂ’s in Business Administration with Concen tration in Management.