x... a
.b.m.k.....m.

, gafmn. N
«we. fig ..
, in” 3..

3.73.

 

t5
sIV I...
. ,. :.p..g.am.m a.
. uHfi: . 33:7...
4 . uW1.
a w
‘ 'Im
. .. I!“ V
hr“.

f.

5
355‘
I,

 

 

 

 

 

 

fififihfi; ... NVsi..:a.

 

i kit,“
- I.

U‘
~t‘

kid

fl

\

~ 199?

 

LlBRARY
Michigan State
University

 

 

This is to certify that the

 

thesis entitled

Antioxidant and antimicrobial activities of 1.2.3.4-
tetrahydroxybenzene and 2,3,4,5-tetrahydroxybenzoic acid

presented by

Andrea Sue Molengraft

has been accepted towards fulfillment
of the requirements for the

Master of degree in Food Science and Human
Science Nutrition

 

 

tilt/n mm

Major Professor’s Signé’ture
(0/30 / 1003

Date

 

MSU is an Affirmative Action/Equal Opportunity Institution

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRCJDatoDuepas-sz

ANT IOXIDANT AND ANTIMICROBIAL ACTIVITIES OF l,2,3,4-
TETRAHYDROXYBENZENE AND 2,3,4,5-TETRAHYDROXYBENZOIC ACH)
By
Andrea Sue Molengraft

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Department of Food Science and Human Nutrition

2003

ABSTRACT
ANTIOXIDANT AND ANTIMICROBIAL ACTIVITIES OF 1,2,3,4-
TETRAHYDROXYBENZENE AND 2,3,4,5-TETRAHYDROXYBENZOIC ACID
By

Andrea Sue Molengraft

The objectives of this study were to compare the antioxidant activities of 1,2,3,4,-
tetrahydroxybenzene (THB) and 2,3,4,5,-tetrahydroxybenzoic acid (THBA) to
commercial antioxidants currently used in the food industry and to determine whether
these compounds display antimicrobial activity. The compounds were tested for
antioxidant activity in bulk oils, liposomes and ground beef using peroxide value,
fluorescence and the thiobarbituric acid test, respectively. THBA displayed high
antioxidant activity in each system, comparable or superior to all of the commercial
antioxidants tested. THB was superior to the antioxidants tested in bulk oil and in cooked
ground beef, but showed little activity in the liposome assay. A disk diffusion screening
assay using Escherichia coli 01571H7, Listeria monocytogenes, and Salmonella spp,
showed that only THB possessed antimicrobial activity. Minimum inhibitory
concentrations (MIC) of THB were determined against three bacteria in typticase soy
broth (0.6% yeast extract) media for 24 and 48 h at 35°C. Mle for THB ranged from
2.8-8.4 mM at 24 h and 5.6-11.2 mM at 48 h. Given both antioxidant and antimicrobial
activity, THB may serve as a multifunctional food ingredient, whereas THBA may serve

as an alternative antioxidant.

This thesis is dedicated to my family for all their love and support

iii

ACKNOWLEDGMENTS

I would like to express my gratitude to my major professor, Dr. Gale Strasburg,
for his knowledge, guidance, and support through my studies at Michigan State
University.

I am also grateful to my committee members for their prolonged support. I would
like to acknowledge Dr. John Frost for all his support and enthusiasm for this project, and
especially for providing the antioxidant samples used in this research. I would also like
to thank Dr. Elliot Ryser for his guidance with the project, for use of the equipment in his
laboratory, and help with data analyses and interpretation. I also want to express my
thanks to Dr. Arzu Cargi for her guidance in the laboratory as well.

I am also very grateful to Dr. Enayat Gomaa for her assistance with training and
TBARS analysis. I would also like to thank Dr. Alden Booren for his suggestions with
the project and for providing me with more than enough lard than one could hope for! I
would also like to thank past and present colleagues in my laboratory, especially John
Rodgers for his insight and friendship. A special thanks to my friends Vareemon
Tuntivanich, Hyo-Jung Yoon, Jin-Shan Shie and Dr. Wen Chiang for all their support and

long days in the lab.

TABLE OF CONTENTS

 

 

 

 

LIST OF FIGURES - -- - - -- -- - vii
LIST OF TABLES--- ..... ix
INTRODUCTION - - - .... - ........... - - 1
CHAPTER ONE LITERATURE REVIEW ............. 5
1.1 Lipid Autoxidation ................................... . ..................................................... 5

1.1.2 Mechanism of Lipid Autoxidation ......................................................... 7

1.2. Role of Antioxidants .................................................................................. 1 1

1.2.1. Free Radical Terminators .................................................................... 11

1.2.2. Oxygen Scavengers ............................................................................. 19

1.2.3. Chelating Agents ................................................................................ 19

1.2.4. Antioxidants in Food Systems ............................................................ 21

1.2.5. Regulation of Antioxidant Use ........................................................... 22

1.3. Methodologies for Evaluation of Lipid Autoxidation ............................... 23

1.3.1. Acceleration Methods ......................................................................... 23
1.3.1.1.Heat .............................................................................................. 23

1.3.1.2. Light ............................................................................................. 24

1.3.1.3. Transition metals .......................................................................... 25

1.3.2. Methods for Measurement of Lipid Autoxidation .............................. 25

1.3.2.1. AOM and Rancimat ..................................................................... 25

1.3.2.2. Schaal oven test ............................................................................ 26

1.3.2.3. Analysis of Carbonyl Compounds by Gas Chromatography ....... 28

1.3.2.4. Anisidine value ............................................................................ 29

1.3.2.5. Thiobarbituric acid ....................................................................... 29

1.3.2.6. Liposome Screening Assay for Lipid Oxidation ......................... 33

1.4. Natural vs. Synthetic Antioxidants ............................................................ 37

1.5. Antioxidants as Multi-functional Food Preservatives ................................ 41

1.5.1. Antimicrobial Evaluation of Phenolic Compounds ............................ 42

1.5.2. Bacteria Cell Wall Composition and Susceptibility to Antimicrobial

Agents ........................................................................................................... 43

1.5.3. Bacterial Pathogens ............................................................................. 44

1.5.3.1. Listeria monocytogenes ............................................................... 44

1.5.3.2. Escherichia coli 0157:1-17 ........................................................... 45

1.5.3.3. Salmonella spp. ............................................................................ 47

1.6. Methodology for Evaluating Antimicrobial Activity ................................ 49

CHAPTER TWO ANTIOXIDANT ACTIVITY OF 1,2,3,4-
TETRAHYDROXYBENENE AND 2,3,4,5-TETRAHYDROXYBENZOIC ACID. 52

 

 

 

2.1 INTRODUCTION , -- 52
2.2 EXPERIMENTAL PROCEDURES 56
2.2.1. Materials ............................................................................................. 56
2.2.2. Methods ............................................................................................... 57
2.3 RESULTS -- - - - ....... 64
2.3.1. Evaluation of Structurally Related Antioxidants in a Bulk Lipid
System ........................................................................................................... 68
2.3.2. Evaluation of Antioxidants in 3 Meat System .................................... 72
2.3.3.Eva1uation of Antioxidants in a Liposome System .............................. 72
2.4 DISCUSSION - - -_ ...... 80

 

CHAPTER THREE ANTIMICROBIAL ACTIVITY OF 1,2,3,4-
TETRAHYDROXYBENZENE AND 2,3,4,5-TETRAHYDROXYBENZOIC ACID

 

 

 

 

 

 

 

 

- 86

3.1 INTRODUCTION — -- 86
3.2 EXPERIMENTAL PROCEDURES 87
3.2.1. Materials ............................................................................................. 87

3.2.2. Methods ............................................................................................... 87

3.3 RESULTS -- A- 90
3.3.1. Disk Diffusion Assay .......................................................................... 90

3.3.2. Minimum Inhibitory Concentration .................................................... 90
CHAPTER FOUR ................ - - - ...... -- - ..... 96
SUMMARY AND CONCLUSIONS - _ - ............. - _ - ........... 96
CHAPTER FIVE - - ....... -_ A 98
FUTURE RESEARCH - -- _ - 98
LIST OF REFERENCES- - A- ..... - - - - . 100

 

vi

LIST OF FIGURES

Figure 1.1 Scheme of Lipid Autoxidation (adapted from Schmidt 2000; Hsieh and

Kinsella 1989) ......................................................................................................... 8
Figure 1.2 Resonance Stabilization of a Phenolic Radical ............................................... 12
Figure 1.3 Quinone formation by a para-substituted phenolic compound upon reaction
with free radicals ................................................................................................... 13
Figure 1.4 Stabilization of a phenolic free radical by hydrogen bonding of an ortho
hydroxyl substituent .............................................................................................. 13
Figure 1.5 Structures of common antioxidants: Free radical terminators ......................... 15
Figure 1.6 Structures of some hydroxycinnamic acid compounds ................................... 16
Figure 1.7 Structures of commonly occurring flavonoids ................................................ 18
Figure 1.8 Oxygen scavenging activity of ascorbic acid .................................................. 19
Figure 1.9 Chelating agents .............................................................................................. 20
Figure 1.10 Peroxide formation and decomposition as a function of time (adapted from
DeMan 1999) ........................................................................................................ 27
Figure 1.11 Malondialdehyde formation .......................................................................... 30
Figure 1.12 Chromophore formed by the condensation of MDA with TBA .................... 31
Figure 1.13 Structure of liposome bilayer ........................................................................ 34

Figure 1.14 Structures of molecules used in liposomal assay : 1-stearoy1-2-1inoleoyl-sn-
glycero-3-phosphocholine and 3-(p-(6-pheny1)-1,3,5-hexatrieny1)phenylpropionic

acid (DPH-PA) ...................................................................................................... 35
Figure 1.15 Phytic Acid .................................................................................................... 38
Figure 1.16 (3) Synthesis of THB and THBA from phytic acid (Hanson and others 1999)

(b) Synthesis of THBA from THB (Stueben 2002) .............................................. 40
Figure 1.17 Structures of common antimicrobial agents .................................................. 43
Figure 2.1 1,2,3,4-tetrahydroxybcnzene (THB) and 2,3,4,5-tetrahydroxybenzoic acid

(THBA) ................................................................................................................. 54

Figure 2.2 Antioxidant activities of alpha-tocopherol, propyl gallate (PG), 2-tert-
butylhydroquinone (TBHQ) and 1,2,3,4-tetrahydroxybenzene (THB) (0.01%
wt/wt) in (A) lard and (B) stripped corn oil. ......................................................... 65
Figure 2.3 Antioxidant activities of gallic acid (GA), protocatechuic acid (PCA), alpha-
tocopherol, 1,2,3,4-tetrahydroxybenzene (THB), and 2,3,4,5-tetrahydroxybenzoic
acid (THBA) in lard. Concentrations of antioxidants were: (A) 0.01%, (B)
0.005% and (C) 0.001% (wt/wt). .......................................................................... 69
Figure 2.4 Antioxidant activities of gallic acid (GA), protocatechuic acid (PCA), alpha-
tocopherol (aT), 1,2,3,4-tetrahydroxybenzene (THB), and 2,3,4,5-
tetrahydroxybenzoic acid (THBA) in cooked ground beef. Concentration of
antioxidants were 0.01% based on fat content. ..................................................... 73
Figure 2.5 Antioxidant activity of gallic acid (GA), pr0py1 gallate (PG), protocatechuic
acid (PCA), 2,3,4,5-tetrahydroxybenzoic acid (THBA), alpha-tocopherol (aT),
and 1,2,3,4-tetrahydroxybenzene (THB), lOuM using Fe (ID-induced
peroxidation of LUVs. .......................................................................................... 76
Figure 2.6 Concentration dependence of antioxidant activity of 2,3,4,5-
tetrahydroxybenzoic acid (THBA) using Fe (ID-induced peroxidation of LUVs.
THBA was tested at various concentrations as indicated in the figure legend. 80

vii

Figure 2.9 Predicted reaction scheme of THB with free radicals ..................................... 83
Figure 2.10 Alternative predicted reaction scheme of THB with free radicals ................ 84
Figure 3.1 Inhibition of S. Typhimurium DT 104 (25-2D), E. coli 01572H7 (AD 305) and

L. monocytogenes 1503 by THB in a disk diffusion assay. .................................. 91

viii

LIST OF TABLES

Table 2.1 Protection Factors of Antioxidant Compounds in Bulk Oil ............................. 71
Table 3.1 Minimum inhibitory concentrations of THB for E. coli 0157:H7, L.
monocytogenes and Salmonella spp ...................................................................... 92

INTRODUCTION

Naturally occurring antioxidants, including rosemary and sage extracts,
polyphenolic catechins of green tea, flavonoids and other phenolic compounds are often
superior to synthetic antioxidants in terms of activity, safety, and consumer acceptance
(Mukhopadhyay 2000). In addition, significant human health benefits such as anti-
cancer, anti-mutagenic and anti-inflammatory activities have been attributed to
consumption of plant-derived compounds (Duthie 1991). The prospect of using naturally
derived compounds to improve human health while stabilizing lipid-containing foods,
offers opportunities to enhance both the diets of consumers and the quality of food
products.

Even though natural antioxidants are widely accepted by consumers, their
adoption by the food industry has been limited. Fluctuations in both supply and cost of
the antioxidant product are major concerns for food processors. Problems during harvest,
such as low yield or contamination of crops, may also result in a substandard product.
Plant material extracts rich in antioxidants are frequently used rather than more costly
purified antioxidants. However, these extracts contain variable concentrations of
antioxidants, undesirable colors, odors and flavors. In addition, natural antioxidants are
subjected to less scrutiny and scientific evaluation than synthetic antioxidants. These
natural antioxidants must be used cautiously due to the potential presence of mutagens,
carcinogens and/or other pathogens (Pratt 1996). Therefore, synthetic antioxidants are

usually preferred due to their lower cost, higher effectiveness and consistent availability.

An alternative approach to use of natural-product extracts could involve the use of
abundant, low-value, by-products of food processing as raw material for the synthesis of
novel antioxidant compounds. Disposal of steeping liquor from cereal grain processing is
problematic due to the presence of phytic acid, an anti-nutritive oligophosphate
compound with strong metal-Chelating activity (Wood 2000). While steeping liquor can
be concentrated and sold for animal feed, environmental concerns raised by high levels of
phosphorus in animal excrement limit the value of steeping liquor. Removal of phytic
acid from the steeping liquor for use in synthesis of novel antioxidants could enhance its
value as well as that of the steeping liquor, which could be used as phytate-free animal
feed.

Phytic acid can be isolated from the steeping liquor of grain processing and
dephosphorylated by phytase to produce myo-insitol. In a simple two- or three-step
conversion, myo-inositol is converted to 1,2,3,4-tetrahydroxybenzene (THB) (Hanson and
others 1999) and 2,3,4,5-tetrahydroxybenzoic acid (THBA) (Stueben 2002), respectively.
The polyhydroxylated aromatic structures of these compounds suggest that they may
function as highly effective, free-radical-scavenging antioxidants.

A phenolic compound may function as an antioxidant by donating a hydrogen
atom to a free radical; the resultant phenolic radical remains relatively stable due to
delocalization of the unpaired electron. Previous studies suggest that compounds with
multiple hydroxyl groups attached to the aromatic nucleus will be potent antioxidants
(Cuvelier and others 2000). We hypothesized that the polyphenolic compounds, THB and
THBA, will show equivalent or greater antioxidant activity compared to commercial

antioxidants used currently in the food industry. As with all antioxidants, activity will

vary depending on the nature of the lipid substrate and type of food system to which they
are added. Thus, the first objective of this study was to evaluate the antioxidant activities
of THB and THBA in three different experimental systems; bulk oils, liposomes and
ground beef. Antioxidant activities of these compounds were compared with activities of
structurally related compounds.

The phytic acid derivatives also share structural similarities with the antimicrobial
compounds fumigatin and aurantiogliocladin. We hypothesized that like the former
compounds, THB and THBA would possess antimicrobial activity. To test this
hypothesis, the second objective of this study was to evaluate the antimicrobial activity of
THB and THBA using a disk diffusion assay as a screening tool. If a compound shows
activity in this assay, it is analyzed further in broth culture to determine its minimum
inhibitory concentration against common foodbome pathogens.

Compounds displaying both strong antioxidant and strong antimicrobial activity
with negligible human toxicity have an enormous potential for use as multifunctional
food ingredients. They could be used in multistage or combination spray washing of
animal carcasses or fruits and vegetables, thereby enhancing the safety and quality of the
final product. Utilization of these compounds by the food industry may also reduce water
and energy costs, minimize color degradation, and maintain structural integrity of the
food. Moreover, these compounds may be safer alternatives to more toxic compounds
such as hydrogen peroxide and chlorine.

This thesis is organized into five chapters. The following chapter provides an
overview of the literature for this study. The proceeding chapters each cover a specific

objective and were prepared as manuscripts for publication. Chapter 2 is devoted to the

first objective, i.e., evaluation of antioxidant activity and Chapter 3 describes the
antimicrobial activity of THB and THBA. Overall conclusions and suggestions for future

research are also provided in Chapters 4 and 5.

CHAPTER ONE

LITERATURE REVIEW

1.1 Lipid Autoxidation

Lipid autoxidation is a term that describes a series of free-radical mediated
reactions, which include the direct reaction of atmospheric oxygen with lipids.
Autoxidation products are responsible for off-flavors and odors in lipid-containing foods
as well as a reduction of the nutritional value of the lipid, loss of fat-soluble vitamins,
carotenoids or proteins, or bleaching of food color (Jacobson 1999; Mukhopadhyay
2000). These deteriorative reactions decrease shelf-life and consumer acceptance of food
products. In addition, there may be consequences of autoxidation which are less readily
recognized. Products of lipid autoxidation, including cholesterol oxides and 4-hydroxy-
2-nonenal (Nawar 1996), may lead to pathological processes including premature aging,
tumorgenesis and atherosclerosis (Ladikos and Lougovois 1990; Ramarathnam and
others 1995). Free-radical mediated human diseases may be prevented or delayed by
stabilizing food products with antioxidants as well as by consumption of diets rich in
natural antioxidants. Thus, through the judicious use of antioxidants, human health may
be enhanced in conjunction with the production of a more stable, desirable and
wholesome, healthy product.

Among the various classes of lipid molecules, the two types most commonly
associated with autoxidation are triglycerides and phospholipids. Triglycerides are
neutral lipids typically found in adipose tissue of animals or as storage oils in plant seeds.

They consist of a glycerol backbone to which fatty acids are esterified. Fatty acids are

hydrocarbon chains with a terminal carboxylic acid moiety. Saturated fatty acids are
fully reduced or saturated with hydrogen, whereas unsaturated fatty acids found in foods
typically contain one to six double bonds between the carbon atoms.

Autoxidation of fatty acids generally occurs alpha to the double bonds, and the
degree of unsaturation affects the kinetics of autoxidation. Fatty acids with higher
numbers of double bonds on the fatty acyl chain are more susceptible to lipid
autoxidation. Oxidation rates for oleic acid (18:1), linoleic acid (18:2) and linolenic acid
(18:3) relative to stearic acid (18:0), which contain no double bonds, are 100, 1,200 and
2,500, respectively (Belitz and Grosch 1987). Arachidonic acid (20:4) was reported to
oxidize 2.9 times faster than linoleic acid (Porter and others 1981). Based On these data,
fish oils, soy oil, and other lipids enriched in polyunsaturated fatty acids will oxidize
faster than less unsaturated lipids, such as beef fat or tropical oils.

Heating of oils, during frying of foods, accelerates autoxidation of lipids,
promoting the dominant major reactions, which include both polymerization and
fragmentation of oxidized lipids. These reactions are responsible for physical changes of
the lipid including darker color and increases in average molecular weight, viscosity,
foam and refractive index. Concomitantly, there are decreases in polyunsaturated fatty
acid content and smoke point (Bennion and Park 1968; Nawar 1996).

Phospholipids are lipid molecules that consist of glycerol, fatty acids, phosphate
and a polyhydroxyl compound or organic base. The parent compound of the
phospholipid is phosphatidic acid, which consists of fatty acids esterfied to the sn-1 and
sn-2 hydroxyl groups of the glycerol backbone, and a phosphate at the sn-3 position. In

more complex phospholipid systems, the phosphate group is usually linked through a

diester bond to the basic or polyhydroxy moiety. Examples of common phospholipids
include phosphatidyl serine, phosphatidyl ethanolamine, phosphatidyl choline and
phosphatidyl inositol. The amphiphilic nature of these compounds enables them to form
the bilayer structure of cellular membranes. The high surface-to-volume ratio of
membrane structures makes this lipid class highly susceptible to oxidation. Thus,
phospholipids are the dominant lipid substrate for autoxidation in products having intact
membranes, such as muscle foods and whole plant tissue.

Products of autoxidation in lipid-containing foods are responsible for the rancid
off-flavors, odors and undesirable color changes in foods that have undergone lipid
autoxidation. Thus, it is important to understand the nature of the chemical reaction
involved in autoxidation of these foods and the mechanisms of preventative measures that

may be taken to ensure a more desirable product.

1.1.2 Mechanism of Lipid Autoxidation
Autoxidation is an irreversible, free-radical-mediated series of reactions (Figure
1.1) which occurs in three stages: initiation, propagation, and termination. During the
initiation step, a hydrogen atom (H) is abstracted from an unsaturated fatty acyl group
(RH) by an initiator to form a fatty acyl radical (R') (Equation 1):
RH + initiator R- + H- (Eq. 1)
Initiators include hydroxyl free radicals (OH), superoxide (02"), hydrogen
peroxide (H202), ionizing radiation, singlet oxygen ('02), transition metal ions (MM) or

other reactive species. Transition metals react with H202, forming

Initiation

RH
‘ ;\
R° ROO‘

F1 RH

ROOH
Termination

ROOR, <——- RO- ——> keto, hydroxy, and
ROR dimers \L epoxy compounds

Cleavage

l

Aldehydes,
Alkyl radicals,
Semi-aldehydes,
or oxo-esters

Figure 1.1 Scheme of Lipid Autoxidation (adapted from Schmidt 2000; Hsieh and
Kinsella 1989)

strong oxidizing agents, which are believed to be -OH, MOW'” or MOHW'” (Equations
2 and 3):
M’” + H202 —+ M‘"*”*' + on + OH' (Eq. 2)
MOH‘W'” <—> M0n+ + H* (Eq. 3)
The reaction of H202 with transition metals is referred to as F enton’s Reaction (Equations
4 and 5) (Aust and others 1985; Hsieh and Kinsella 1989), in which a hydroxyl radical is

produced.

og- + M" —> 02 + M”" (Eq. 4)

M"" + H202 —> M" + -OH + OH' (Eq. 5)

Trace amounts of metals are often found in edible oils, originating from the soil in
which the plant was grown, leaching or flaking of metal from processing equipment, or as
natural components of biological tissue. The oxidation state of the metal is important in
determining the rate of induced autoxidation. Reduced iron (Fey), for example, is a
much more effective pro-oxidant than the oxidized form (Fey) (Graf and Empson 1987;
Arora and Strasburg 1997). Oxidized metal ions become effective pro-oxidants in the
presence of a reducing agent (e.g. ascorbic acid), which regenerates the reduced form of
the metal.

Initiation of lipid oxidation requires an active form of oxygen. The direct reaction
of oxygen with unsaturated fatty acids is thermodynamically unfavored due to differences
in spin state of oxygen and the fatty acid. The ground state of unsaturated fatty acids
corresponds to the singlet state, whereas the ground state of oxygen is the triplet state.
Therefore, the reaction of unsaturated fatty acids with triplet oxygen (30;) is forbidden
(Harel and Kanner 1985; Hsieh and Kinsella 1989). Spin restriction of molecular oxygen
is overcome by sufficient activation of energy to form 102. This can be achieved by the
reaction of 0;“ with H202 or the reaction of 3O; with 011 or sensitizers, such as plant
pigments, to form '02.

In propagation, the fatty acyl radical reacts with diatomic oxygen (02) to form a
peroxy radical (ROO-) (Equation 6) which can then initiate autoxidation of another fatty
acid by abstracting an allylic hydrogen atom (rate-limiting step) to form a hydroperoxide
(ROOH) (Equation 7):

R. + 02 R00 (1346)

ROO- +R H ROOH+R - (Eq. 7)

The kinetics of hydroperoxide formation is proportional to the number of double
bonds present in the fatty acid. Polyunsaturated fatty acids are more sensitive to
oxidative damage than monounsaturated fatty acids due to lowered bond dissociation
energy of the allylic hydrogen (Aust and others 1985). The peroxy radical can abstract a
hydrogen more readily from a 1,4—pentadiene system than from a single allyl group due
to resonance stablization of the 1,4-diene radical generated (Belitz and Grosch 1987).

Hydroperoxides are common intermediates of autoxidation that are unstable and
spontaneously decompose, especially in the presence of metal ions, into acids, aldehydes,
alcohols, carbonyls and ketones. These low-molecular-weight, volatile compounds,
commonly referred to as secondary oxidation products, are responsible for the rancid off-
flavors, odors and subsequent undesirable color changes in food products that have
undergone lipid autoxidation.

Hydroperoxides break down through a series of decomposition steps (Nawar
1996). The hydroperoxide first undergoes scission at the oxygen-oxygen bond, resulting
in an alkoxy radical and a hydroxyl radical (Equation 8):

ROOH R0- + OH (Eq. 8)

In the second step the carbon-carbon bond is cleaved on either side of the alkoxyl
group. Cleavage on the carboxyl side results in the formation of an aldehyde and an acid,
whereas scission on the methyl side produces a hydrocarbon and an oxoacid.

Termination of autoxidation, shown in equations 9, 10, and 11, is defined as the
combination of radicals to form non-radical, stable compounds:

ROO- + R- ROOR (Eq. 9)

10

R0- + R- ROR (Eq. 10)

R00 + ROO- ROOR + 02 (Eq. 11)

1.2. Role of Antioxidants

An antioxidant is defined as ‘any substance that when present in low
concentrations, compared with those of the oxidizable substrate, significantly delays or
inhibits autoxidation of that substrate’ (Gutteridge 1994). Antioxidants operate by
different mechanisms depending on their structure, and are classified as free radical

terminators, oxygen scavengers (or reducing agents), or Chelating agents (Giese 1996).

1.2.1. Free Radical Terminators
Free radical terminators interfere with the free-radical chain of oxidative reactions
by donating a hydrogen atom from phenolic hydroxyl groups. Effective free radical
terminators (AH) compete with the unsaturated lipid substrate (Hsieh and Kinsella 1989)
by inhibiting free radical formation in the initiation step or by interrupting propagation
(Equations 12 and 13):
R- + AH RH + A- (Eq. 12)

Red + AH ROOH + A- (Eq. 13)

The most effective free radical terminators are phenolic compounds. When a
phenolic compound donates a hydrogen atom to a lipid free radical formed during
autoxidation, a phenolic radical is formed, which is stabilized'by electron delocalization
over the aromatic nucleus (Figure 1.2). The resulting antioxidant radical (A-) is less

reactive than the fatty acid radical due to this stabilization.

ll

OH / o‘ o O o \
CD‘DD
k ' J

Figure 1.2 Resonance Stabilization of a Phenolic Radical

 

 

The reactivity of an antioxidant compound and the stability of the phenolic radical
are influenced by the presence or absence of substituents on the phenolic nucleus.
Depending on their nature, number, and location, these substituents may either enhance
or suppress antioxidant activity. A second hydroxyl group in the ortho or para position
of a phenolic compound increases antioxidant activity by further electron delocalization
of the aromatic ring (Cuvelier and others 1992). Compounds with ortho or para
substitution are stabilized because the initial semiquinone radical produced by reaction
with a lipid radical can further oxidize to a quinone by reaction with another lipid radical
(Figure 1.3 (a)) or by reacting with another semiquinone radical, thereby regenerating the
antioxidant (Figure 1.3 (b)). The presence of a hydroxyl group at the ortho position, such
as 1,2-dihydroxybenzene derivative, is believed to also stabilize the radical through

intramolecular hydrogen bonding (Figure 1.4) (Baum and Perun 1962).

12

ROOH
OH

ROO ROOH
OH ONO

oO

 

Figure 1.3 Quinone formation by a para-substituted phenolic compound upon reaction
with free radicals

d ----- r
C)

Figure 1.4 Stabilization of a phenolic free radical by hydrogen bonding of an ortho
hydroxyl substituent

l3

Substitution of phenolic compounds with three hydroxyl groups generally
increases antioxidant activity compared to compounds with only two hydroxyl groups.
For example, gallic acid, with three hydroxyl groups, is more effective than
protocatechuic acid with only two hydroxyl groups (Pokomy 1987; Cuvelier and others
1992). Protocatechuic acid and caffeic acid (o-diphenols) have better antioxidant activity
than their respective monophenols, p-hydroxybenzoic acid and p-coumaric acid (Cuvelier
and others 1992).

Alkyl substitution also contributes to increased antioxidant activity. Phenolic
compounds with methyl groups attached to the benzene ring have been shown to increase
antioxidant activity by electron donation to the ring. Thus, the phenolic hydrogen atom is
more labile for donation to a lipid free radical. Steric effects of alkyl substituents also act
to stabilize the antioxidant radical (Baum and Perun 1962). Substitution of a tert-butyl
group ortho to the hydoxyl group makes butylated hydroxytoluene (BHT) and butylated
hydroxyanisole (BHA) effective antioxidants. Steric hindrance of the bulky tert-butyl
group decreases reactivity of the radical. The most effective free radical terminators are
typically aromatic compounds including BHT, BHA, 2-tert-butylhydroquinone (TBHQ),
propyl gallate, Coenzyme Q10, and the tocopherols (Figure 1.5).

Hydroxycinnamic acid compounds, which are widely distributed in nature, show
similar structure-activity relationships to the phenolic compounds discussed above
(Figure 1.6). Methoxy substitution ortho to a hydroxy group has shown to decrease
antioxidant effectiveness for most acid phenols. For example, caffeic acid, with ortho

substitution of a hydroxy group, possesses greater stability of the phenoxy radical and

14

    

 

 

"lulu-o-
"HIM!“-

Eo< 0:36
:0
o
:0
o:
:0
SO oégmou
: e
:0
/ ._.
o\

 

  

803556. .868 poi “mecmExoccm :oEEoo .«o mocaozbm mg DSME

mu~o< omafiouumoOuOhnm
O

.ocosmoooncq—m

IO

I’lrlvoo o o .

  

OI

Che ones—oixouam Bum—baa 3.5m: Boflcmxxoca: ova—55m

OI

IO

2230 7305 Aommb ococScoeafibsnéofim

IO
ol\|

OI

IO

IO

15

HO

OH

Caffeic Acid

HO

OCH3

Ferulic Acid

Figure 1.6 Structures of some hydroxycinnamic acid compounds

16

increased antioxidant activity compared to ferulic acid with methoxy ortho substitution
(Cuvelier and others 1992). Rosmarinic acid, with four hydroxyl groups, shows greater
antioxidant activity than compounds with only one or two hydroxyl groups, such as
ferulic acid and caffeic acid (Chen and Ho 1997). This is believed to be due to the
increased resonance stabilization and/or o-quinone or p-quinone formation (Chen and Ho
1997; Frankel 1998).

Flavonoids are another class of widely distributed plant compounds, some of
which possess potent antioxidant activity (Figure 1.7). The basic structure of a flavonoid
molecule consists of two aromatic rings, identified as rings A and B, linked by a three-
carbon aliphatic chain which condenses to form a pyran ring, C. Substitution with
hydroxyl groups, especially on the B-ring, increases antioxidant activity, whereas
substitution with methoxy groups decreases antioxidant activity (Arora and others 1998).
For example, quercetin, rutin and luteolin, each having 3 ,4 -dihydroxy phenyl
substitution, shows stronger antioxidant activity than naringenin and hesperetin. This is
likely due to increased electron delocalization by the presence of the o-dihydroxy
substitution on the B-ring.

The antioxidant radical may further react with another peroxyl radical, convert
back to AH by reacting with another molecule, or dimerize to A2 (Gutteridge and

Halliwell 1990). Equations 14, 15, and 16 show the termination steps with the addition

of antioxidants.
ROO- + A- ROOA (Eq. 14)
A-+RH AH+R~ (Eq. 15)
A- + A- A3 (Eq. 16)

17

 

R= OH Quercetin R,=H, R2=OH Naringenin
R=O-rutinose Rutin RI=OH, R2=OCH3 Hesperetin
R=H Luteolin

Figure 1.7 Structures of commonly occurring flavonoids

18

1 . 2. 2. Oxygen Scavengers

Oxygen scavengers reduce the partial pressure of oxygen, an essential component
of lipid autoxidation, by transferring hydrogen atoms. Examples of oxygen scavengers
include ascorbic acid and erythorbic acid, which reduce molecular oxygen producing two

molecules of water (Figure 1.8) (Giese 1996).

0: 0:1
OH 3 0 4H. CH E o
02 + 2 o a 2 o + ZHZO
\
HO O
\_/ OH O
V
Ascorbic Acid Dehydroascorbic acid.

Figure 1.8 Oxygen scavenging activity of ascorbic acid

When used in combination with other antioxidants, they function as synergists by
promoting their antioxidative effects. For example, ascorbic acid and its sodium salt,
sodium ascorbate, regenerate phenolic antioxidants by donating hydrogen atoms to

phenoxyl radicals produced by lipid oxidation (Frankel 1998).

1.2.3. Chelating Agents

Chelating agents [e.g. citric acid or ethylenediaminetetraacetic acid (EDTA)] are
often added to a food product to form complexes with endogenous pro-oxidant metal
ions, such as iron or copper, to delay autoxidation (Figure 1.9). A chelate is a complex

formed by combining a metal ion with a multidentate ligand.

19

A monodentate ligand, forms only one bond with the metal while a polydentate
ligand forms multiple bonds, which increases the stability of the complex. For example,
EDTA has six donor atoms, in which it encloses a metal ion, making the metal less
available for autoxidation reactions (Gordon 1990). In addition, Chelating agents lower
the MZVM3+ redox potential, thereby limiting the ability of the metal to act as an
oxidizing agent. In contrast, large bulky ligands tend to be less stable exposing the

metal to solution giving rise to prooxidant effects (Miller 1996)

OH
03g 0 OH 0
N O
Ho—-\<’ L\ fort
o N HO OH
3..
HO HO
Ethylenediaminetetraacetic acid (EDTA) Citric Acid

Figure 1.9 Chelating agents

20

1.2. 4. Antioxidants in Food Systems

In addition to recognizing the structure-activity relationship of antioxidants, it is
important to understand that an antioxidant’s efficacy varies depending on the type of
food system to which it is added. According to Porter and others (1989), lipophilic
antioxidants such as BHT and BHA are more effective in emulsion and liposome systems
than in bulk, dry oils when compared to more hydrophilic antioxidants, such as THBQ.
These findings led to the formation of the “polar paradox” principle, which states that
non-polar antioxidants function best in polar lipid emulsions and membranes, whereas
polar antioxidants are relatively more effective in non-polar lipids such as bulk oils.

The mechanism underlying this paradox is based on differences in distribution or
partitioning of antioxidants with respect to the aqueous and lipid phases (Frankel 1998).
For an oil/water emulsion, hydrophilic antioxidants dissolve in the water phase and
become diluted, while the lipophilic antioxidants partition into the oil-water interface. In
a bulk oil system, hydrophilic antioxidants are oriented at the air-oil interface, where
oxidation of lipids is most likely to occur, while lipophilic antioxidants are dispersed
throughout the oil phase. Antioxidants that are polar, hydrophilic or amphiphilic with
high hydrophile-lipophile balance (HLB) are more effective in low surface-to-volume
ratio (LSV) products, such as bulk oils and fats (either vegetable or animal), or synthetic
esters of fatty acids (Porter and others 1989). In contrast, antioxidants that are non-polar,
lipophilic or amphiphilic with low HLB are more effective in high surface-to-volume
ratio (HSV) lipid structures. HSV matrices include emulsions, micelles and membranes

of whole tissue foods.

21

1.2.5. Regulation of Antioxidant Use

Food preservation strategies were developed long before the science of
antioxidants was understood. Many natural spices and herbs are classified as GRAS
(Generally Recognized as Safe) substances because of their historical use in food
preservation. Examples of GRAS substances possessing antioxidant activity include
rosemary and sage extracts (21 CFR 182.10). Newly discovered antioxidants and
extracts from plant sources without a long history of consumption as foods must be
approved by either the Food and Drug Administration (FDA) or the United States
Department of Agriculture (USDA), depending on the food to which it is added. The
FDA oversees all domestic and imported foods sold in interstate commerce, including
shell eggs, bottled water, and wine beverages with less than 7 percent alcohol. In
contrast, domestic and imported meat, and poultry and related products, such as meat- or
poultry- containing stews, pizzas and frozen foods, and processed egg products (generally
liquid, frozen and dried pasteurized egg products) are under jurisdication of the USDA.
(U. S. Food and Drug Administration 1998).

Based on weight of the fat or oil content of a food, the FDA allows the addition of
antioxidants (BHT, BHA, for example), either alone or in combination at concentrations
up to 0.02% w/w (200 ppm) when used in emulsion stabilizers for shortenings (21 CFR
170) and the USDA allows the addition of BHA, BHT, propyl gallate and TBHQ, for
example, up to 0.01% w/w (100 ppm) or 0.02% in combination, with no individual
antioxidant exceeding 0.01% (9 CFR 424.21). Tocopherols can be added up to 0.03%,

when added to rendered animal fat or vegetable oils (9 CF R 424.21 ).

22

1.3. Methodologies for Evaluation of Lipid A utoxidation

Antioxidant efficacy of new compounds is determined by comparing their effect
on the rate of lipid oxidation to those of antioxidants used presently in a food or model
system. Lipid autoxidation and antioxidant efficacy are typically determined by using an
acceleration method such as heat, light or addition of metal catalysts, followed by

measurement protocol.

1.3.1. Acceleration Methods
1.3.1.]. Heat

Heating is the most common and effective means of accelerating autoxidation in
order to determine an antioxidant’s efficacy in a relatively short time. The mechanism of
autoxidation changes with temperature as described by the Arrhenius equation (Equation

17):

k = kAexp[-I_E_A_] (Eq. 17)
RT

where k = rate constant, kA = Arrhenius equation constant, EA = activation energy (J),
T = absolute temperature (K), and R = 8.3144 (J/mol-K).

When comparing the rate constants of autoxidation with and without the addition
of an antioxidant, an Arrhenius plot of log induction period versus temperature shows
that with the addition of an antioxidant, the log induction period decreases with an
increase in temperature (Taoukis and Labuza 1996). The induction period at ambient
storage temperatures conditions can then be predicted by extrapolating the results from
elevated temperatures to lower temperatures. However, in more complex food systems,

the rate of autoxidation is a composite of multiple dominant reactions. making high

23

temperature testing less reliable. Different changes in rate constants for one series of
reactions in a food may result in their dominance above a certain temperature, while other
reactions may predominate at lower temperatures (Taoukis and Labuza 1996).

At ambient temperature the rate of autoxidation is independent of oxygen partial
pressure. As the temperature increases the rate of autoxidation becomes dependent on
oxygen partial pressure due to a decrease in oxygen solubility (Frankel 1998). In
addition, peroxides begin to decompose at higher temperatures with the scission products
distilling off (Frankel 1993). For example, at 100°C primary hydroperoxides rapidly

decompose, making high temperature tests less reliable (Frankel 1993).

1.3.1.2. Light
Lipid autoxidation also can be accelerated by the exposure to light. Initiation by
photooxidation results from the presence of sensitizers, classified as Type I and Type 11

(Equations 18 and 19, respectively) (Frankel 1998).

hv

3 Sens + RH —> [Intermediate] + O; —> ROOH + Sens (Eq. 18)
hv

3 Sens + O; —> [Intermediate] + lSens —-) '02 + RH —> ROOH (Eq. 19)

Type I sensitizers (e.g. riboflavin) are in the triplet state (3 Sens) and react with the lipid
substrate by abstracting hydrogen atoms to form decomposition products of autoxidation
(e.g. hydroperoxides). Type II sensitizers (e.g. chlorophyll) are also in the triplet state,
but react with oxygen to produce 102, which in turn reacts directly with the lipid

substrate.

24

Photooxidation of lipid-containing food products is usually not a concern since
the reaction only occurs when the lipid is exposed to direct sunlight or fluoresent light
(Frankel 1998). Therefore, photooxidation is often not used as an acceleration method

for lipid oxidation studies.

1.3.1.3. Transition metals
Transition metals increase the rate of autoxidation by activating hydroperoxide
decomposition, reacting with an unoxidized substrate or by activating molecular oxygen

to form 102 , 02', or R0- (Equations 20, 21 and 22):

M"+ + ROOH -—> M‘"*'>* + OH- + Ro- (Eq. 20)
M"*+RH—>M‘"'”*+H++R~ (Eq.21)
M'” + 02 —+ MW”+ + 02' —) 102 + H02. (Eq. 22)

Acceleration methods yield valuable information on shelf-life and antioxidant
activity in a much shorter time frame than experiments performed under ambient
conditions. Careful consideration of the acceleration method and the type of evaluation

for antioxidant activity must be done in order to have valid results.

1 .3 .2. Methods for Measurement of Lipid Autoxidation
1.3.2.1. AOM and Rancimat

In the active oxygen method purified air is bubbled through an oil or fat sample at
110°C-130°C, and with acidic volatile products are passed into a deionized water trap
(Pike 1998). The induction period is determined by periodic measurements of either

peroxide value or sensory evaluation of the rancid odor. The Rancimat method is based

25

on increased electrical conductivity of small molecules formed as a consequence of lipid
autoxidation (Pike 1998). These tests suffer from the disadvantage that at elevated
temperatures, the reaction rate of autoxidation changes (e.g. decrease in oxygen

solubility).

1.3.2.2. Schaal oven test

The Schaal oven test is an example of a low- temperature test in which results
correlate most closely with ambient temperature storage (Pike 1998). The milder
temperatures used in this test do not substantially change the mechanism of autoxidation
relative to that of room temperature storage. In this test a known amount of oil or fat is
heated at 60-70°C with autoxidation determined by quantifying peroxides present in the
sample. The peroxide value, commonly used to measure primary autoxidation products
in bulk lipid substrates (Frankel 1993), serves to predict the formation of secondary
products that are responsible for undesirable off-flavors, odors and undesirable color
changes in food products that undergo lipid autoxidation.

Lipid peroxides are measured by reaction with potassium iodide (K1) in acid
solution; iodine is generated in direct proportion to the amount of peroxides in the sample
and is quantified by reaction with sodium thiosulfate. Starch is used as the indicator for

the presence of iodine (Equations 23 and 24) (Pike 1998).

H+
ROOH + KT ROH + ICOH' + 12 (Eq. 23)
12 + starch + 2 Na2S203 2Nal + starch + NaZS406 (Eq.24)
(blue) (colorless)

26

Peroxide value is usually reported as milliequivalents (mEq) of lipid peroxide per kg
lipid.

Lipid autoxidation can be evaluated by determining the induction period (I.P.),
which is the time required to produce a detectable level of rancidity or for onset of a
sudden increase in the rate of autoxidation (Frankel 1993). Figure 1.10 shows peroxide
formation and decomposition as a function of time. The LP. of an antioxidant-treated
substrate is compared to that of a control (no antioxidant added), with longer induction
periods indicating better antioxidant activity. The protection factor is defined as the ratio
of the LP. of the antioxidant treated sample compared to the control.

The Schaal oven test is highly empirical and subject to some laboratory-to-
laboratory variation. Determination of the endpoint of titration is subjective and thus,
may lead to variations in data. Iodine also can react with carbon-carbon double bonds as
well as with peroxides, which may result in an apparent decrease in peroxide value.
Nevertheless, peroxide value is widely used because it is applicable to most fats and oils

and easy to perform.

| l
“o I I n ’, “ a ,, “ O I ,,
Initiation | propagation | termrnation

. I . . . . .
peroxrde peroxrde stabilization peroxrde

formation L decom osition
I P
| l \

induction period I
A |
[

 

Peroxide Value

 

 

Time

Figure 1.10 Peroxide formation and decomposition as a function of time (adapted from
DeMan 1999)

27

1.3.2.3. Analysis of Carbonyl Compounds by Gas Chromatography

Gas chromatography (GC) has been used to quantify volatile, secondary oxidation
products of autoxidation. Carbonyl products correspond to the decomposition of
hydroperoxides in which acids, aldehydes, ketones or other carbonyl products are formed.
Carbonyl content is often used to evaluate autoxidation because these compounds
contribute to flavor deterioration of a lipid-containing food, thus correlating to sensory
evaluation.

Headspace analysis is commonly used to monitor lipid autoxidation by the
chromatographic analysis of the sample held in a closed container (Frankel 1998). After
the sample is heated and vaporized into the gas phase, an aliquot of the gas phase
headspace is injected directly into a GC column. The concentration of volatile
compounds in the headspace is proportional to the concentration of volatiles in the
sample. Identity of compounds is tentatively based on the co-elution with standards.

The benefits of GC are generally enhanced when coupled with mass spectrometry
(GC-MS). In GC-MS, the identity of compounds within peaks is confirmed on the basis
of mass. Mass spectral analysis of peaks also provides supporting evidence for purity of
the material within a peak. Gas chromatography can be used to assess a wide variety of
food samples without altering the product by extraction. One disadvantage of GC,
however, is that volatiles may be lost during heating. Additionally, GC analysis of

carbonyls requires costly equipment which must be maintained by a trained operator.

28

1.3.2.4. Anisidine value

Anisidine value is a spectrophotometric technique that determines the amount of
aldehydes and related carbonyl compounds formed during autoxidation of fats and oils
(Pike 1998). The aldehydes react with p-anisidine to form a chromagen, the
concentration of which is determined by absorbance at 350 nm. Anisidine values are
advantageous because non-volatile carbonyl compounds can be detected. Anisidine
value and peroxide value are often used together to give a totox value (anisidine value +
(2 x peroxide value)) as an indicator of total autoxidation of a sample by evaluating both
the primary and secondary oxidation products. Anisidine and totox values are widely

used throughout Europe, whereas peroxide value is used most commonly in the US.

(Stauffer 1996).

1.3.2. 5. Thiobarbituric acid

The 2-thiobarbituric acid (TBA) test is one of the oldest and most commonly used
tests for measuring lipid autoxidation, especially in determinating warmed-over-flavor of
cooked meats (Melton 1983; Crackel and others 1988; Hoyland and Taylor 1991).
Cooked meats oxidize at an accelerated rate compared to raw products, and primary
autoxidation products (lipid hydroperoxides) rapidly decompose into secondary oxidation
products (e.g. aldehydes) associated with rancidity.

The formation of malondialdehyde (MDA) in cooked meat products is often used
as in indicator of lipid autoxidation. The hydrogen B to the peroxide group of unsaturated
peroxides is believed to undergo cyclization to a five-membered ring, which then

decomposes to form MDA (Figure 1.11) (Dahle and others 1962). In later studies, it was

29

reported that other primary and secondary lipid oxidation products, such as five-
membered hydroperoxy epidioxides and 1,3-dihydroperoxides, are precursors of MDA

(Frankel and Neff 1983).

c-c—c=c-c—c—c=c—c=c— _. c—c—c-c—c-c—cc—cr—c—
O l I
(l C>§-O

l

c-c—c§ c—c—c§ c=c—c:c—
o. o,

l

C-C-C+ HO—C-SZH 'l' C=C-C=C—
O
malondialdehyde

Figure 1.11 Malondialdehyde formation

One molecule of MDA reacts with two molecules of TBA with the elimination of two
molecules of water (Figure 1.12) to yield a pink compound, which is quantified
spectrophotometrically at 532-535 nm (Antolovich 2001) and used as an indicator of
autoxidation.

In addition to MDA, there are many other secondary autoxidation products that
can react with TBA, including ketones and other carbonyls, which are responsible for
rancid off-flavors and odors (Tarladgis and others 1960; Crackel and others 1988;

Hoyland and Taylor 1991). Therefore, the results of this reaction are often reported as

30

OH

HN

2 S=< +
HN M
OH O 0

TBA MDA

0“ HO
HN _N
s=< _ >_SH + 2HgO
I. — ,,/
OH HO

Figure 1.12 Chromophore formed by the condensation of MDA with TBA

31

TBA reactive substances (TBARS) (mg MDA/kg sample). TBARS are calculated as

shown in Equation 25:

 

 

TBARS = A532nmx K (Eq. 25)
where, K = conc. in moles/SmL of distillate x M.W. MDA x 107 x 100
Absorbance wt. of sample % recovery

The distillation constant, K, is calculated from a standard calibration curve using
tetramethoxypropane (TEP), a MDA precursor.

The whole sample, or less frequently, a lipid extract is examined using the TBA
test. By using the whole sample, oxidation products of protein-bound lipids and
phospholipids can be measured. While the whole sample method is the most
quantitative, it requires numerous solvent extractions which can be very time consuming
(Sinnhuber and Yu 1958). In the past, volatiles from the oxidized sample were collected
by passing steam through the slurry. Since then, the distillation technique has been
simplified by directly heating the samples in Kjeldahl distillation flasks with
simultaneous collection of the volatile compounds, including MDA. A clear, aqueous
distillate of the sample is collected, which is reacted with TBA and the product is
measured spectrophotometrically. With this procedure, interference mechanisms/products
are minimized and the accuracy of the test is improved.

Advantages of the TBA test include sensitivity, accuracy and reproducibility.
However, the TBA test has limited specificity (e.g. MDA may'react with other oxidation
products), requires a large sample size and can be very labor-intensive. Numerous
reports have correlated TBA numbers with sensory evaluation, many of which contradict

each other (Hoyland and Taylor 1991). This may be due to different training practices of

32

the panel. Additionally, variation in TBARS value may also be due to different sources
of meat, storage conditions, processing and packaging. Therefore, the TBA test should
only be used in relative terms when comparing antioxidant activity between laboratories
(Tarladgis and others 1960; Gray and Monahan 1992). The TBA distillation procedure is
still the most commonly used method for evaluation of lipid autoxidation in ground beef

and has been chosen for this study.

1.3.2. 6. Liposome Screening Assay for Lipid Oxidation

Evaluating large numbers of antioxidants in food products can be quite time
consuming. By using a screening method coupled with an appropriate model system,
analysis time can be reduced. Numerous model systems have been used including
linoleic acid and methyl linoleate micelles and liposmes. The liposome screening assay
used in this study is a rapid and sensitive method to evaluate antioxidant compounds in a
model system which is representative of biological membranes (Arora and others 1997).

Other model systems, such as linoleic acid and methyl linoleate micelles, are not
favored as food models because they are not structurally representative of membrane
phospholipids or triglycerides (Frankel 1993). Liposomes, however, mimic membrane
lipid structures dispersed in foodstuffs. Liposomes are a type of emulsion system in
which “phospholipids, when dispersed in water, form multi-layers consisting of bilayers
between the water phase and the innermost layer next to the oil phase” (Frankel 1998)
(Figure 1.13). They mimic biological membranes without the presence of pro-oxidant
enzymes or endogenous antioxidants that could confound interpretation of results (Arora
and others 1997). In addition, their composition and structure can be carefully controlled.

Liposomes, therefore, are often used as models when determining antioxidant efficiency.

33

Waler afar

M)
W . .4”

Bilayer

Figure 1.13 Structure of liposome bilayer

Liposomes are classified by size and number of bilayers: small unilamellar
vesicles (SUVs) (diameter from 20-50 nm) and large unilamellar vesicles (LUVs)
(diameter 60 nm) (Chapman 1984). In addition, multilamellar vesicles (MLVs) form
spontaneously when dry phospholipid films swell in water or buffer resulting in the
formation of both SUVs and LUVs. Due to the heterogeneity in size and the small
proportion of lipid exposed to the aqueous medium of the internal bilayers, the MLV
system is a poor model for lipid autoxidation in membranes. In contrast, approximately
70% of the total lipid is located in the outer leaflet of SUVs (Chapman 1984). Due to
less ordered packing of the SUV, the structure is constrained between the phospholipid
head groups, thus, allowing penetration of pro-oxidants. Therefore, LUVs consisting of
phospholipids found in biological membranes are the preferred membrane model for lipid
autoxidation studies. The lipid substrate used in this model system is l-stearoyl-Z-
linoleoyl-sn-glycero-3-phosphocholine (SLPC), with the saturated fatty acid at the sn-1
position, an unsaturated fatty acid at the sn-2 position and a phosphate containing polar

group at the sn—3 position (Figure 1.14).

34

O O
\ \\ 0
7%“ P\/ O

1-stearoy1-2-linoleoyl-sn-glycero-3-phosphocholine

OH

\\\

3-(p-(6-phenyl)- l ,3,5-hexatrienyl)phenylpropionic acid (DPH-PA)

Figure 1.14 Structures of molecules used in liposomal assay : 1-stearoyl-2-linoleoyl-sn-
glycero-3-phosphocholine and 3-(p-(6-phenyl)-1,3,5-hexatrienyl)phenylpropionic acid
(DPH-PA)

35

Oxidation in the liposomal suspension is determined by fluorescence
spectroscopy which offers great advantages as an analytical tool when combined with a
defined lipid substrate. The analysis can be run using <1 umol of pure lipid, and a small
concentration of fluorescent probe to lipid (1:350), allowing for high sensitivity and
minimal membrane perturbation. In addition, fluorescence spectroscopy is 100-1000
times more sensitive than other spectrophotometric techniques (Arora and others 1997;
Strasburg and Ludescher 1995). The speed of the assay is also beneficial in that iron or
other free-radical generators can be used to induce lipid autoxidation, thus avoiding high-
temperature conditions.

In an aqueous phase, lipophilic and amphilic membrane probes are generally non-
fluorescent, but when embedded in the hydrophobic environment of a lipid bilayer, they
are highly fluorescent. Arora and Strasburg (1997) incorporated the probe, 3-(p-(6-
phenyl)-1,3,5—hexatrienyl) phenylpropionic acid (DPH-PA), (Figure 1.14), into liposomes
to serve as a reporter of lipid autoxidation. The conjugated double bonds of this probe
are highly susceptible to react with free radical intermediates of lipid autoxidation. The
polar substituent of the probe is localized at the lipid-water interface of the membrane
and the hydrophobic tail portion lies parallel to the acyl chains of the membrane. In
addition, DPH-PA is relatively photostable and exhibits strong fluorescence in lipid
environments. Reaction of the probe with free radicals produced by membrane
autoxidation results in a loss of conjugation and consequencly, a decrease in
fluorescence. Presence of an antioxidant decreases the rate of fluorescence decay with
the rate of fluorescence decay reflecting the rate of autoxidation as well as antioxidant

efficacy.

36

1.4. Natural vs. Synthetic Antioxidants

Demand for extended-shelf—life foods in which antioxidants are the basis for
protecting flavor and color in lipid containing foods has grown (Langen 2002). Natural
antioxidants such as rosemary and sage extracts, polyphenolic catechins of green tea,
tocopherols and flavonoids are more accepted by consumers in that perception of the risk
of potential toxicity issues is minimized. (Mukhopadhyay 2000). Even though natural
antioxidants are more widely accepted than synthetic antioxidants by consumers and the
food industry, their adoption has been limited. Inconsistent availability of the plant
source and cost of the crop are major concerns. Problems during harvest, such as
contamination or taints in the finished product may also result in a substandard product.
Extracts of plant materials rich in antioxidants are frequently used rather than more costly
purified antioxidants. However, extracts may have variable concentrations of
antioxidants, and also suffer from off-colors or odors. Caution must be taken when using
extracts for antioxidants in that potential mutagens, carcinogens and/or pathogens may be
present (Pratt 1996). Therefore, synthetic antioxidants are used more commonly
throughout the food industry because they’re relatively inexpensive, highly effective,
readily available and have a well defined toxicity profile.

Despite the advantages of synthetic antioxidants, heightened concerns have been
raised regarding the environmental consequences of using petroleum-based products as
starting materials for chemical synthesis. Use of petroleum-based products in organic
synthesis produces volatile organic compounds, which can contribute to ground level

smog and depletion of the ozone layer. An environmentally friendly alternative to these

37

petroleum-based products involves the use of plant-derived ingredients for the synthesis
of novel antioxidants.

Phytic acid (inositol hexaphosphate) (Figure 1.15) is a naturally occurring
component of plant fiber with antioxidant properties. It chelates iron and/or other pro-
oxidant metals, thus retarding hydroxyl radical formation via the Fenton reaction in lipid-
containing foods (Lee and Hendricks 1982). It is found in highest concentration in corn
(6%), followed by sesame (5%), wheat (3%) and rice (2%) (Quillin and Quillin 2001 ).

Phytic acid is also found in high abundance in the steeping liquor from grain
processing, which presents a disposal problem for the industry (Hoseney 1986; Wood
2000). Steeping liquor may be concentrated and sold for animal feed, but environmental
concerns about high levels of phosphorus derived from phytic acid in animal excrement
limit the value of this product. High levels of phosphorus in water run-off from fields
enhances biological oxygen demand in lakes, rivers and streams leading to a decline in

water activity.

 

Figure 1.15 Phytic Acid

38

One option for limiting the level of phytic acid in the environment is to extract it
from the steeping liquor and convert it to a compound or compounds with increased
value. Phytic acid can be isolated from the grain, dephosphorylated, and hydrolyzed to
myo-inositol from which 1,2,3,4-tetrahydroxybenzene (THB) (Hanson and others 1999)
and 2,3,4,5-tetrahydroxybenzoic acid (THBA) (Stueben 2002) are formed. Figures 1.16
(a) and (b) show the reaction scheme from which THB and THBA are synthesized. This
synthesis can be done relatively quickly, giving consistent results at low cost.

The characteristics of an effective antioxidant are found in the compounds derived
from phytic acid. Ideally, effective antioxidants are considered to have good solubility in
the fat or water phase, good dispersion throughout the food, no harmful physiological
effects, and no objectionable flavors, colors or odors. In addition, they should be
effective in low concentrations, readily available, stable throughout processing and
economical (Dugan 1976; Giese 1996).

Phenolic compounds donate a hydrogen atom to free radicals and remain
relatively stable due to delocalization of the unpaired electron. Previous studies suggest
that compounds with multiple hydroxyl groups attached to the aromatic nucleus will be
potent antioxidants (Cuvelier and others 2000). We hypothesized that the polyphenolic
compounds, THB and THBA, will show equivalent or greater antioxidant activity
compared to commercial antioxidants used currently in the food industry. In this study,
THB and THBA are compared to alpha-tocopherol, gallic acid, protocatechuic acid,
propyl gallate and TBHQ for antioxidant activity in bulk lipids, ground beef and

liposome systems.

39

(a)

 

   

 
   

 

on
Gluconbacter
oxydans ......mon
D
95%
on
myo-inositol myo-2-inosose
0.5 N H2SO4
H2O, A
66%
on
HO
HO
OH
1,2,3,4-tetrahydroxybenzene (THB)
(b)
O” OH 0
HO 1) KHCO3, HO
H2O, OH
>
2) HCl
HO HO

OH OH

1,2,3,4-tetrahydroxybenzene (THB) 2,3,4,5-tetrahydroxybenzoic acid (THBA)

Figure 1.16 (a) Synthesis of THB and THBA from phytic acid (Hanson and others 1999)
(b) Synthesis of THBA from THB (Stueben 2002)

40

1.5. Antioxidants as Multifunctional Food Preservatives.

An estimated 76 million cases foodbome disease results in illnesses, including
325,000 hospitalizations and 5,200 deaths occurring in the United States each year (Mead
and others 1999). According to the Economic Research Service, the estimated cost
(medical costs, productivity losses from missed work, and losses from premature deaths)
of foodbome diseases caused by Campylobacter, Salmonella (nontyphoidal serotypes
only), Escherichia coli OlS7:H7, E. coli non-0157zH7 STEC, and Listeria
monocytogenes. is $6.9 billion annually (US. Department of Agriculture 2002). Based
on heightened interest from consumers and the food industry, we have chosen to test the
antimicrobial activity of THB and THBA against E. coli 0157:H7, Salmonella and L.
monocytogenes.

Food preservatives include food additives that are antimicrobial agents, anti-
browning agents and antioxidants (Davidson 2001). Antimicrobial agents are chemical
compounds, either added to or already in food, that inhibit microbial grth or inactivate
microorganisms, thus preserving safety and quality of the product.

Most antimicrobial agents are bacteriostatic or fungistatic rather than bactericidal
or fungicidal. Many microbial factors, including inherent resistance, growth rate,
antagonism and cellular composition must be considered when using an antimicrobial
agent. In addition, many intrinsic and extrinsic factors associated with the food are
responsible for an antimicrobial agent’s effectiveness. There are many interacting
reactions which take place simultaneously, making it difficult to pinpoint the exact target
of the antimicrobial agent in a food system (Davidson 2001). It is thought, however, that

the main targets of antimicrobial agents are the cell wall, cell membrane, metabolic

41

enzymes, protein synthesis, and genetic systems. For example, organic acids have been
shown to be most effective in their undissociated form, in which they penetrate the lipid

bilayer of the cell membrane to disrupt cell integrity (Davidson 2001).

1.5.1. Antimicrobial Evaluation of Phenolic Compounds

Phenolic compounds were first recognized for their antimicrobial activity in 1867,
when Joseph Lister used phenol for sanitizing surgical equipment (Davidson 1993).
Phenol is currently used as a disinfectant and antiseptic and is also a component of
mouthwash and throat lozenges. However, skin exposure to high levels of phenol can
lead to liver damage, diarrhea, dark urine and hemolytic anemia (Agency for Toxic
Substances and Disease Registry 1999). More recently, phenolic compounds that are
commonly used as antioxidants have been assessed for possible antimicrobial activity.
Parabens including p-hydroxybenzoic acid are the most common examples of
antimicrobial phenolic compounds. They are believed to increase membrane
permeability (Davidson 2001). Phenolic antioxidant compounds including BHT, BHA,
propyl gallate and TBHQ also show antimicrobial activity (Davidson 1993). For
example, Tuncel and Nergiz (1993) found that BHA at a concentration of 400 ug/mL
inhibited Staphylococcus aureus, Bacillus cereus, E. coli, and S. typhimurium when
grown in trypticase soy broth at 37°C for 18 h. BHA also showed antifungal activity at
200 ppm against Aspergillus, Penicillium, and Geotrichum in glucose salt broth when
incubated at 32°C, for 85 h (Ahmad and Branen 1981).

We hypothesized that THB and THBA would have antimicrobial activity, based

on structural similarity to the other phenolics and to the antimicrobial compounds

42

firmigatin and aurantiogliocladin (Figure 1.17). The latter compounds consist of quinone

structures similar to those that could be formed by phenolic compounds THB and THBA.

O O .
HO CH3 HO CH3
H3CO H3CO CH3
O . O
Fumigatin Aurantiogliocladin

Figure 1.17 Structures of common antimicrobial agents

1.5.2. Bacteria Cell Wall Composition and Susceptibility to Antimicrobial Agents

Antimicrobial activity is partly related to polarity. An antioxidant compound
must be sufficiently lipophilic to pass through the cell membrane and once inside, it must
also be sufficiently water-soluble to interact with cell components. Parabens have been
shown to effectively disrupt cell integrity. The mechanism by which phenolic
antioxidant compounds inhibit microbial growth is likely similar to that of parabens
(Davidson 2001).

Gram positive bacteria are generally more sensitive to antioxidants than Gram
negative bacteria (Tuncel and Nergiz 1993). Davidson (1993) reported that hydrophobic
compounds are more effective antimicrobial agents against Gram positive bacteria
whereas Gram negative bacteria are more susceptible to hydrophilic compounds. These
differences between Gram positive and Gram negative bacteria result from differences in

cell wall composition.

43

The cell wall in Gram positive bacteria contains a thick layer of peptidoglycan
which is tightly bound to teichoic acid and lipoteichoic acid. The cell wall is pressed
tightly against the cell membrane with very little space between, making such organisms
easy to kill. In contrast, Gram negative bacteria have a layer of lipid polysaccharide as
part of their cell wall. The outer membrane is a thin shell of peptidoglycan, lipoproteins
and liposaccharides.

Unlike Gram positive bacteria, Gram negative organisms have extensive space
between the peptidoglycan layer and the cell membrane. The outer membrane serves as a
partial chemical sieve by allowing only relatively small molecules to penetrate through
the porins. As a result, Gram negative bacteria are more resistant to antibiotics which

can’t penetrate the lipid polysaccharide layer as readily.

1.5.3. Bacterial Pathogens
1.5. 3. 1 . Listeria monocytogenes

L. monocytogenes is a Gram positive, psychrotrophic, facultative anaerobe that is
motile by means of flagella at ambient temperatures. It is of special concern to the food
industry because of its ability to grow at refrigeration temperatures (2-4°C), although
optimum grth occurs between 30-37°C (Seeliger and Jones 1984; Swaminathan 2001).
As a result, ready-to-eat foods such as soft cheeses, frankfurters, and delicatessen meats
are among the most common foods implicated in outbreaks of listerious. L.
monocytogenes can also multiply and survive in or on foods for extended periods of time
under adverse conditions including low pH and high salt concentrations. L.

monocytogenes is unique in that it is more acid-tolerant than many other foodbome

44

pathogens, with growth occurring at pH values as low as pH 4.3 (Lou and Youscf 1999),
although most strains grow best at neutral to slightly alkaline pH (Seeliger and Jones
1984). It is also salt-tolerant, capable of growing at salt concentrations up to 10% and
surviving in 23% (aw 0.83) brine solutions for extended periods (Shahamat and others
1980). Finally, L. monocytogenes is unique in that it resists the effects of freezing, drying
and heating without forming spores (Center for Food Safety and Applied Nutrition 2003).

Manifestations of listeriosis, the foodbome disease caused by ingestion of L.
monocytogenes, are serious and include septicemia, meningitis and spontaneous abortion.
Complications include encephalitis, endocarditis, abscesses and local purulent lesions
(Seeliger and Jones 1984). The onset time for systemic listeriosis varies, from as few
days to 70 days. Less commonly, this organism causes gastroenteritis with an onset time
of 20-27 hours (Swaminathan 2001). The infectious dose of L. monocytogenes varies
with bacterial strain and with susceptibility of the individual. It is believed, however, that
as few as 100 CFU/g may cause disease in susceptible persons such as pregnant women,
the elderly and immunocompromised adults (Swaminathan 2001).

Cases of listeriosis are usually sporadic with at least 2,518 illnesses, 2,322

hospitalizations, and 504 deaths occurring each year in the US. (Swaminathan 2001).

1.5.3.2. Escherichia coli 0157:H7

Escherichia coli is a Gram negative, motile, oxidase-negative, non-spore forming
lactose ferrnenter. There are five types of E.coli isolates which can cause disease. They
are differentiated based on three major surface antigens, which permit serotyping: the O

(somatic), H (flagella), and K (capsule) antigens. Among these isolates,

45

enterohemorrhagic E. coli (EHEC) is the most significant group based on severity of
disease, and will be the focus of this discussion.

E. coli 0157:H7 predominates in the United States and many other countries
(Meng and others 2001). It has been reported that E. coli 0157:H7 produces large
quantities of one or more related potent toxins: the verotoxin and the shiga-like toxin
(Meng and others 2001), which may cause a severe and sometimes fatal condition known
as hemolytic uremic syndrome (HUS) (kidney failure). E. coli 0157:H7 is unique
compared to other stains of E. coli in that is unable to grow at temperatures 445°C in E.
coli broth, unable to ferment sorbitol within 24 hrs, and is unable to produce [3-
glucuronidase (Meng and others 2001). Unlike many E.coli strains, E.coli 0157:H7 is
acid tolerant (growth at pH minimum 4.0-4.5), and thus is able to grow in fermented and
other low acid foods.

Vehicles most commonly associated with E.coli 0157:H7 infections are
undercooked ground beef and less frequently, fresh produce, unpasteurized milk and
juice (e.g. apple cider). The organism is commonly found in the intestinal tract of cattle;
thus, slaughter practices must be carefully monitored so that the meat does not become
contaminated. Raw milk may become contaminated if fecal matter containing E.coli is
present on the udders at the time of milking (Center for Food Safety and Applied
Nutrition 2003). Produce may become contaminated by exposure to manure used as
fertilizer, and juice prepared from fruit which has fallen to the ground may likewise
become contaminated by fecal matter. The infectious dose of E.coli 0157:H7 is
unknown, but it is believed that as few as 10 cells (dose similar to Shigella) (Center for

Food Safety and Applied Nutrition 2003) may cause disease, with an onset time of 1-2

46

days. Symptoms include severe cramping and bloody diarrhea lasting approximately
eight days. All individuals are at risk, but children and the elderly are most susceptible.
Up to 15% of infected children develop HUS whereas the elderly usually develop
thrombotic thrombocytopenic purpura (TTP) (abnormal decrease in platelets), fever and
neurologic symptoms. Hemolytic uremic syndrome is the primary cause of kidney
failure in children in the United States with most cases caused by E. coli 0157:H7 (Center
for Food Safety and Applied Nutrition 2003).

The CDC has reported that approximately 73,000 cases of infection and 61 deaths
occur each year in the US (Center for Food Safety and Applied Nutrition 2003).
Prevention of E.coli 0157:H7 infection is based on its sensitivity to heat. Cooking meat
products to an internal temperature of at least 74°C (165°F) for several seconds will
inactivate the pathogen (Meng and others 2001). Pasteurization of milk and juices is
sufficient to eliminate E.coli from these products. However, some food products such as
lettuce or alfalfa spouts, will lose their acceptability upon excessive heating. Therefore, it
is important to understand the grth and development of the organism and to find ways

to reduce contamination.

1.5.3.3. Salmonella spp.

Salmonella spp. are Gram-negative, motile, non-sporeforming, facultative
anaerobic, catalase-positive, oxidase-negative rods belonging to the family
Enterobacteriaceae. The ability of Salmonella to metabolize nutrients by both
respiratory and fermentative pathways makes this pathogen a concern in modified-

atmosphere and vacuum-packaged foods, and fermented raw milk products (high acid

47

tolerance). Salmonella spp. are widespread in animals, especially in poultry and swine.
Food sources include eggs, raw meats, and poultry. Infection occurs primarily through
the fecal/oral route (Center for Food Safety and Applied Nutrition 2003). S. Enteritidis is
unique in that the organism is present inside of the egg (in the yolk) suggesting vertical
transmission from chicken to egg (Center for Food Safety and Applied Nutrition 2003).
The CDC estimates 75% of S. Enteritidis outbreaks are associated with consumption of
raw or inadequately cooked Grade A whole shell eggs (Center for Food Safety and
Applied Nutrition 2003).

Though the organism grows optimally at 37°C, some strains grow at elevated
temperatures ( 54°C). The pH range for grth of Salmonella extends from 4.5-9.5,
with an optimum pH for grth of 6.5-7.5 (D’Aoust 1989). Although Salmonella growth
is inhibited in 3-4% brine solutions, some salt tolerance has been observed at increasing
temperatures (D’Aoust 1989). Salmonella also survive for prolonged periods of time in
foods stored at freezing and ambient temperatures (D’Aoust 1989). Thus, the organism
can adapt to and survive under many conditions.

Typhoid and enteric fevers as well as gastroenteritis are common diseases caused
by Salmonella. The infectious dose of Salmonella is as few as 15-20 cells with usual
onset time of 48 hours. Acute symptoms include nausea, vomiting, abdominal cramps,
bloody diarrhea, fever and headache. More chronic consequences include reactive
arthritis 3-4 weeks after onset of acute symptoms. Acute symptoms may last 1-2 days or
longer, depending on dose and the individual. Disease is caused by the penetration and
passage of Salmonella from the gut lumen into the epithelium of the small intestine where

inflammation occurs. All age groups are susceptible, but symptoms are most severe in

48

children and the elderly. Among the elderly in hospitals/nursing homes, Salmonella
exhibits a mortality rate of approximately 3.6% (Center for Food Safety and Applied

Nutrition 2003).

l . 6. Methodology for Evaluating Antimicrobial Activity

The rise in foodbome illnesses is due partly to changes in food preparation (e.g.
microwave ovens), increases in meals prepared outside the home (e.g. fast food), more
susceptible populations, and increased antibiotic resistance. These changes have led to
increased consumer concerns for the safety of food; in response, microbial analysis of
food samples is studied more readily. Various measures have been implemented to
reduce risk of contracting foodbome illness. For example, the development of Hazard
Analysis and Critical Control Point (HACCP) programs and guidelines for safe handling
and food preparation minimize potential contamination.

Conventional methods for evaluating microbial growth include isolating and
enumerating cultures on selective media. More advanced methods include antibody-based
and nucleic acid-based detection assays, many of which are now semi-automated for
more rapid detection.

Methods for testing antimicrobial resistance include agar screening, disk
diffusion, broth dilution and the E test® (Horseheath, Cambridge CB1 6RG, United
Kingdom). Agar screening assesses the growth of bacteria when the antimicrobial agent
is added to the agar. In the disk diffusion method, a paper disk containing a known
amount of the antimicrobial agent is added to a seeded agar plate. After incubation, the

zone of bacterial inhibition around the disc is measured. The broth dilution test is

49

commonly used in laboratories to determine the minimum inhibitory concentration (MIC)
(the lowest concentration that prevents bacterial growth at 35°C for 24 and 48 hours).
Some factors which influence diffusion testing include inoculum density, agar
composition and depth, time and temperature of incubation, and concentration of the
antimicrobial agent.

The E test® is a newer method that combines dilution and diffusion. A plastic
strip with various dilutions of the antimicrobial agent is placed on a seeded agar plate.
After incubation, the antimicrobial agent from the strip that inhibits bacterial growth will
form an elliptical zone of inhibition. The E test® generates an accurate and reproducible
MIC. Testing parameters, such as different bacterial strains, diluents, incubation
conditions and media formulations all account for lab to lab variability. Therefore,
results from different labs need to be carefully interpreted.

In this portion of our study, our objective was to determine if THB and THBA
possess antimicrobial activity. They share structural similarities with the antimicrobial
compounds fumigatin and aurantiogliocladin; therefore, we hypothesize they will also
have antimicrobial properties. The diffusion disk assay was used initially to screen for
activity, followed by MIC determination.

In this study we illustrate how phytic acid, a by-product of grain processing, could
be used to synthesize active antioxidant and antimicrobial compounds. These compounds
can stabilize the shelf life of lipid-containing foods, and give rise to a value-added
product. By expressing both antimicrobial activity and antioxidant activity, THB and

THBA offer several advantages to the food industry. These antioxidants derived from

50

by-products of grain processing provide an inexpensive and efficient tool for food

preservation.

51

CHAPTER TWO
ANTIOXIDANT ACTIVITY OF 1,2,3,4-TETRAHYDROXYBEN ENE AND

2,3,4,5-TETRAHYDROXYBENZOIC ACID

2.1 INTRODUCTION

Lipid autoxidation is the primary mechanism responsible for deterioration of low
moisture foods and is a major contributor to deterioration of other food products
including pre-cooked convenience foods. Free radicals generated by lipid autoxidation
give rise to off-flavors and odors in lipid-containing foods as well as a reduction of the
nutritional value of the lipid, and loss of fat—soluble vitamins, carotenoids, or proteins
(Jacobson 1999; Mukhopadhyay 2000). In some cases, free radical-mediated lipid
oxidation may yield toxic autoxidation products (Nawar 1996). The onset of perceptible
rancidity resulting from autoxidation may be delayed by judicious use of antioxidants.

Demand for extended-shelf-life foods in which antioxidants are the basis for
protecting flavor and color in lipid containing foods has grown (Langen 2002). Natural
antioxidants are preferred by the food industry over synthetic antioxidants in terms of
safety and consumer acceptance (Mukhopadhyay 2000). Even though natural
antioxidants are more widely accepted than synthetic antioxidants by consumers, their
adoption by the food industry has been limited. Volatility in supply of the plant source
and cost of the crop are major concerns. Problems during harvest, such as contamination
or taints in the finished product may also be responsible for a substandard product.
Extracts of plant materials rich in antioxidants are frequently used rather than more costly

purified antioxidants. However, extracts may have variable concentrations of

52

antioxidants, or they may suffer from off-colors or odors. Caution must be taken when
using extracts for antioxidants in that potential mutagens, carcinogens and/or pathogens
may be present (Pratt 1996). Therefore, synthetic antioxidants are used more commonly
throughout the food industry because they’re relatively inexpensive, highly effective,
consistently available and have a well defined safety/toxicity profile.

Despite the advantages of synthetic antioxidants, there has been rising concern
with respect to environmental consequences of using petroleum-based products as
starting material for chemical synthesis. Use of petroleum-based products in organic
synthesis results in production of volatile organic compounds, which have been found to
contribute to ground level smog as well as to the depletion of the ozone layer. In
addition, chronic exposure of workers in the chemical industry to these compounds
places their health in jeopardy. An environmentally friendly alternative to this use of
petroleum-based products involves the use of plant-derived ingredients for the synthesis
of novel antioxidants.

Phytic acid is a naturally occurring component of plant fiber with antioxidant
properties. It chelates iron and or other pro-oxidant metals, thus retarding hydroxyl
radical formation via the Fenton reaction in lipid-containing foods (Lee and Hendricks
1995). It is found in highest concentration in corn (6%), followed by sesame (5%), wheat
(3%) and rice (2%) (Quillin and Quillin 2001).

Phytic acid is also found in high abundance in the steeping liquor from grain
processing, which presents a disposal problem for the industry (Wood 2000). Steeping
liquor may be concentrated and sold for animal feed, but environmental concerns about

high levels of phosphorus derived from phytic acid in animal excrement limit the value of

53

this product. High levels of phosphorus in water run-off from fields enhances biological
oxygen demand of lakes, rivers and streams and contributes to deterioration of water
quality.

In this study, we show the potential for use of phytic acid as starting material for
the synthesis of the novel compounds 1,2,3,4-tetrahydroxybenzene (THB) and 2,3,4,5-
tetrahydroxybenzoic acid (THBA) (Figure 2.1). Previous studies suggest that increasing
the number of hydroxyl groups attached to the aromatic nucleus increases antioxidant
activity (Cuvelier and others 2000). Substituting phenolic compounds with three
hydroxyl groups generally increases antioxidant activity compared to those having only
two hydroxyl groups. Gallic acid with three hydroxyl groups is more effective than
protocatechuic acid (PCA) with only two hydroxyl groups (Pokomy 1987; Cuvelier and
others 1991). The polyhydroxylated aromatic structures of THB and THBA suggest

that they may function as highly effective free-radical scavenging antioxidants.

OH OH 0
HO
HO O OH
HO HO
OH OH
THB THBA

Figure 2.1 1,2,3,4-tetrahydroxybenzene (THB) and 2,3,4,5-tetrahydroxybenzoic acid
(THBA)

54

The objectives of this study were to compare antioxidant efficacy of THB and
THBA with several commonly used commercial antioxidants, and to evaluate activity
with structurally related compounds. Antioxidant activity was evaluated in three lipid
systems: bulk oil, ground beef, and liposomes. The results suggest that both compounds
possess high potential for use as antioxidants in food products, provided they have a

favorable safety profile.

55

2.2 EXPERIMENTAL PROCEDURES
2.2.]. Materials

1,2,3,4-tetrahydoxybenzene (THB) and 2,3,4,5-tetrahydroxybenzoic acid (THBA)
were obtained from the laboratory of Dr. John Frost, Department of Chemistry, Michigan
State University (East Lansing, MI). Fresh ground beef was purchased from the
Michigan State University Meat Lab (East Lansing, MI) immediately after grinding. The
cattle were harvested 2 days prior to boning. Stripped corn oil was purchased from Acros
Organic (Pittsburgh, PA). Lard was purchased from Hasselbach Meats Inc. (Fremont,
OH). Synthetic 1-stearoyl-2-linoleoyl-sn-glycero-3-phosphocholine (SLPC) and
polycarbonate membranes (diameter 0.75", pore diameter 100 nm) were obtained from
Avanti Polar Lipids (Alabaster, AL). The fluorescent probe, 3-(p-(6-phenyl)-l,3,5-
hexatrienyl) phenylpropionic acid (DPH-PA) was obtained from Molecular Probes
(Eugene, OR).

Propyl gallate, protocatechuic acid, gallic acid, alpha-tocopherol,
ethylenediaminetetraacetic acid (EDTA), 2-thiobarbituric acid (TBA), celite (acid-
washed), calcium phosphate, (N-[2-hydroxyethyl]piperazine-N’-[2-ethanesulfonic acid])
(HEPES), (3-[N-morpholine]propanesulfonic acid) (MOPS), butylated hydroxyanisole
(BHA), butylated hydroxytoluene (BHT), iminodiacetic acid Chelating resin and 1%
starch indicator solution were purchased from Sigma Chemical Company (St. Louis,
MO). Ten-butylhydroquinone (TBHQ) was purchased from Eastman Chemical Company
(Rochester, NY). Dimethylsulfoxide, (DMSO), sodium sulfate, sodium chloride,
potassium iodide (KI) (A.C.S. grade) and methanol were purchased from J .T. Baker Co.

(Phillipsburg, NJ)). Glacial acetic acid (reagent grade), dichloromethane, hydrochloric

56

acid, and chloroform (U.S.P. grade) were purchased from EM Science, (Gibbstown, NJ).
A saturated solution of potassium iodide was prepared daily using freshly boiled, distilled
water, and was kept out of direct light. A 0.1 N standardized sodium thiosulfate solution
was purchased from Aldrich Chemical Co. (Milwaukee, W1). A 0.01N sodium thiosulfate
solution was prepared by making a ten-fold dilution of the 0.1N standard solution using
distilled water. Tris(hydroxymethyl)aminomethane (Tris) base was purchased from
GibcoBRL Life Technologies (Grand Island, NY).

Antifoam and glass wool were purchased from Thomas Scientific (Swedesboro,
NJ). Ethyl alcohol (200 proof) was purchased from Pharmco Products (Brookfield, CT).
Boileezers (non-volatile granules) were purchased from Fisher Scientific (Pittsburgh,

PA).

2.2.2. Methods

Fat Determination of Ground Beef: Fat content of the meat was determined using
the method of Marmer and Maxwell (1981). Briefly, a dry column was prepared by
placing a glass wool plug at the bottom of a glass column (diameter 1.25"), followed by
the addition of 10 g celite/calcium phosphate (9:1), which was packed firmly into the
column. The sample (5 g) was ground with 20 g sodium sulfate, followed by 15 g celite,
using a mortar and pestle. This mixture was then added to the column slowly, packing
firmly. The mortar and pestle were rinsed with dichloromethane/methanol (9:1).
Approximately 50-60 mL of dichloromethane/methanol (9:1)was added to the column.
After the first drop was eluted, an additional 150 mL dichloromethane/methanol (9:1)
was added to the column. The filtrate, containing total lipids, was filtered into a 500-mL
round bottom flask of predetermined weight and was evaporated using a Bi’rchi

57

Rotoavapor-R (Flawil, Switzerland) until completely dried before weighing. Percent
lipid was determined by the following equation:

% Lipid = dry lipid weight/sample weight x 100 (Eq. 1)

Schaal Oven Test

Peroxide value: Peroxide value measurements were calculated using A.O.C.S.
Official method Cd 8-53 with the modification that the starch indicator (1.0 mL) was
added prior to titration. All samples were tested in triplicate. Peroxide value
(milliequivalents/kg lipid) was calculated using the following equation:

PV = (S-BlfN) (1000) (Eq. 2)
weight of sample

where: B = titration of blank (mL), S = titration of sample (mL) and
N = normality sodium thiosulfate (equivalents/L).

Each antioxidant (TBHQ, propyl gallate, alpha-tocopherol, protocatechuic acid,
gallic acid, THB, or THBA) was dissolved in 500 uL DMSO and then mixed into either
stripped corn oil or lard to yield final concentrations of 0.01%, 0.005% or 0.001%
(wt/wt). The control treatment consisted of 500 uL DMSO without antioxidant. All
treatments and controls were incubated in SOO-mL beakers at 60° +/- 3°C in the dark using
a Blue M Electric Company Oven (Blue Island, Illinois). Each data point is reported as
the mean, 3: 1 SD, of triplicate measurements.

Antioxidant Evaluation: The induction period (I.P.) corresponds to either the time
required to reach a level of detectable rancidity or a sudden increase in the rate of
autoxidation (Frankel 1993) and indicates antioxidant efficacy. Antioxidant treated

samples were compared to the control, in which nothing was added. The protection

58

factor, defined as the ratio of the LP. of the antioxidant-treated sample compared to the

control) was also determined and used for evaluation.

Thiobarbituric Acid (TBA) Test

Sample preparation: The ground beef was thawed at 4°C for 24 h at which time
the sample were prepared. Antioxidant stock treatments (1%) were prepared by
dissolving compounds in 2.1 mL ethyl alcohol. Solutions were then mixed into the meat
to yield a final antioxidant concentration of 0.01% based on the fat content of the meat
(17.5%). The antioxidant stock solutions were first mixed with approximately 5-10
grams of ground beef, which was, in turn, thoroughly mixed with the rest of the sample
(total 210 g) by hand. The meat was divided into patties (105 g) that were stored 1-2
hours at 4°C in Petri dishes (9 cm diameter x 1.5 cm thickness) until frying. The patties
were cooked on a West Bend electric grill (West Bend, WI) at the highest heat setting to
an internal temperature of 72°C i 2°C (flying time 3 minutes/side). The cooked patties
were stored at 4°C covered with plastic wrap and aluminum foil.

T BARS Analysis: Lipid oxidation was determined using the TBA test as described
by Tarladgis (1960) and modified by Crackel and others (1988). A IO-g portion of each
cooked patty was analyzed daily for TBARS over 5 days, beginning with the zero-time
value taken immediately after frying. The sample was homogenized in 10 mL H20 and
10 mL 1% antioxidant solution (0.5% propyl gallate plus 0.5% EDTA) using a Polytron
PCU-2-110 (Brinkman Instruments, Westbury, NY) at speed 5 for 10 seconds. The
resulting homogenate was transferred quantitatively to a 250-mL Kjeldahl distillation

flask containing 67.5 mL H20 and 2.5 mL 4 N hydrochloric acid. Boiling beads and

59

antifoam spray (5 pumps) were added to the distillation flask, and the samples were
distilled using a Labconco distillation apparatus (Kansas City, MO) adjusted to a heat
setting that kept all flasks at a constant boiling level. A 50-mL aliquot of distillate was
collected.

Five mL of the distillate and 5 mL of freshly prepared TBA solution (0.02 M)
were mixed in a screw capped test tube. The blank was prepared using 5 mL H2O + 5
mL TBA (0.02 M). Samples were capped, boiled for 30 minutes, and cooled to room
temperature for 10 minutes. Absorbance was then read at 532 nm using a Cary
l3E I UV-Visible spectrophotometer (Varian Analytical Instruments, Walnut Creek, CA).
TBARS were calculated using the following equation:
TBARS=A532nm*K (mg MDA/kg sample) (Eq. 3)

where, K = conc. in moles_/5mL of distillate x M.W. MDA x 107 x 100
Absorbance wt. of sample % recovery

 

The distillation constant used for this experiment was determined by previous

experiments performed in our.lab: K=6.8.

Statistical Analysis: Statistical analysis was performed using the pairwise multiple
comparision procedure, Dunnett’s method (SAS® Version 7, Cary, NC). Treatment
samples were analyzed for significant differences (p< 0.05) from the control, in which no
antioxidant was added, and to each other. Results are expressed from triplicate

experiments.

60

Linosorrgl Assay

Large unilamellar vesicles (LUVs) containing the probe DPH-PA were prepared
each day, using the extrusion procedure of MacDonald and others (1991) as modified by
Arora and others (1997).

Preparation of LU Vs: The liposomes [780 uL of 5mg/mL SLPC chloroform
stock + 4.5 uL probe of 1 mg/mL stock] were prepared by drying under vacuum using a
Biichi Rotoavapor-R (Flawil, Switzerland). The dried lipid was then resuspended in 500
uL MOPS buffer (0.15 M sodium chloride, 0.1 mM EDTA, and 0.01 M MOPS; pH 6.8).
After 10 freeze-thaw cycles using a dry ice/ethanol bath, the suspension was passed 29
times through a polycarbonate filter (100 nm pore size) using a Liposofast extruder
apparatus (Avestin, Ottawa, Canada). The suspension was transferred to a
microcentrifuge tube and kept on ice, out of direct light during analysis.

Sample preparation: A control sample was prepared by mixing 100 uL HEPES-
Tris (50 mM, pH 7), 200 uL sodium chloride (1M), and 1.68 mL nitrogen-sparged water,
to which a 20-uL aliquot of the LUV suspension was added. The suspension was mixed
continuously using a magnetic stir bar for 5 minutes at room temperature, out of direct
light, then transferred to a thermostated cuvette holder (24°C) in the spectrofluometer and
incubated for 5 minutes. The suspension was then mixed by inversion and fluorescence
was read (0 time). The fluorescence intensity was read at 1 minute and thereafter at 3
minute intervals up to 21 minutes.

A Fe (ID-containing pro-oxidant sample, in which no antioxidant was added, was
prepared by mixing 100 uL HEPES-Tris (50 mM, pH 7), 200 uL sodium chloride (1 M),

and 1.66 mL nitrogen-sparged water, to which a 20-uL aliquot of the LUV suspension

61

was added. The suspension was mixed continuously using a magnetic stir bar for 5
minutes at room temperature out of light, then transferred to a therrnostated cuvette
holder (24°C) in the spectrofluometer and incubated for 5 minutes. Oxidation was
initiated by the addition of 20 uL of 0.5 mM FeCl2 to yield a final concentration of 5 uM.
The suspension was immediately mixed by inversion and fluorescence was read (0 time).
The fluorescence intensity was read 1 minute after addition of F eC12 and every 3 minutes
thereafter up to 21 minutes.

Antioxidant treatment samples were prepared by mixing 100 uL HEPES-Tris (50
mM, pH 7), 200 DL sodium chloride (1 M), and 1.64 mL nitrogen-sparged water, to
which a 20-uL aliquot of the LUV suspension was added. After 5 minutes of
continuous stirring of the sample at room temperature, out of light, an aliquot of
antioxidant stock solution (1 mM) was added to yield the desired final antioxidant
concentration. The suspension was then transferred to a thermostated cuvette holder (24
°C) in the spectrofluometer and incubated for 5 minutes. Oxidation was initiated by the
addition of 20 uL of 0.5 mM FeCl2 to yield a final concentration of 5 uM. The
suspension was immediately mixed by inversion and fluorescence was read (0 time).
Fluorescence intensity was read 1 minute after addition of FeCl2 and every 3 minutes
thereafter up to 21 minutes.

The final concentrations of components in the cuvette were 2.5 mM HEPES-Tris
(pH 7), 100 mM sodium chloride, 2.5 uM lipid and 7.4 nM probe. The fluorescence
signal was followed with an SLM instruments (Model 4800) spectrofluorometer (Urbana,
IL) equipped with data acquisition hardware and sofiware from On-Line Instrument

Systems (Bogart, GA). The excitation wavelength used was 384 nm with a slit width of 2

62

nm. The emitted light was passed through optical filters (KV 418, Schott, Duryea, PA),
prior to detection.

Antioxidant Evaluation: The rate of fluorescence decay is an indication of rate of
autoxidation. Therefore, the rate in which fluorescence decay is inhibited by the addition
of a particular compound is an indicator of its antioxidant efficacy. Relative fluorescence

(Frel) was calculated by the following equation:

(Frej) = _F_ = fluorescence intensity at a given time (Eq. 4)
F o fluorescence intensity at 0 min

63

2.3 RESULTS

Preliminary experiments were conducted to determine antioxidant activity of
THB in bulk oil using the Schaal oven test. The antioxidant activity of THB was
compared to alpha-tocopherol, TBHQ and propyl gallate (0.01% w/w) using lard and
stripped corn oil as the lipid substrates (Figure 2.2). Antioxidant activity was evaluated
by determining peroxide value of the lard or oil as a function of time.

In lard (Figure 2.2 A), the control had an induction period of approximately 10
days, followed by alpha-tocopherol at 18 days, thus giving a protection factor of 1.8 for
alpha-tocopherol. TBHQ, propyl gallate and THB did not reach the end of the induction
period within 26 days, at which time the substrate was totally consumed. Therefore, we
concluded that the antioxidant activity from greatest to weakest in lard is TBHQ THB
propyl gallate > alpha-tocopherol > control.

These compounds were also evaluated for antioxidant activity in corn oil stripped
of endogenous antioxidants (Figure 2.2 B). In this lipid substrate, the control and alpha-
tocopherol samples both had induction periods of approximately 8 days compared to the
experiments using lard. As with the lard experiments, TBHQ, propyl gallate and THB
did not reach the end of the induction period within 28 days. Therefore, we concluded
that the antioxidant activity in stripped corn oil from greatest to weakest is TBHQ THB

propyl gallate > alpha-tocopherol control at 28 days.

64

Figure 2.2 Antioxidant activities of alpha-tocopherol, propyl gallate (PG), 2-tert-
butylhydroquinone (TBHQ) and l,2,3,4-tetrahydroxybenzene (THB) (0.01% wt/wt) in
(A) lard and (B) stripped corn oil.

Each data point is the mean of triplicate measurements. Error bars correspond to :1: 1 SD.
Error bars not shown are smaller than the symbols.

65

Peroxide Value
(mEq hydroperoxide/kg lard

Peroxide Value
(mEq hydroperoxide/kg oil)

\

' 7'350

I

09
O
O

250

_5
0|
0

.5
o
o

50

500”

450

400

350

250

200

150

100

50

0

200 -

 

 

 

 

 

 

 

 

 

 

+Control
+alpha—
tocopherol
At - PG
A TBHQ
+TI-IB
(A)
+Control
+alpha-
tocopherol
+TBHQ
+THB
' , +PG
l ’_/- ‘
024681012141618202224262830
Time (days)
(B)
Figure 2.2

66

The purpose of these experiments was to do a preliminary screening of THB
antioxidant activity in comparison to antioxidants commonly added to foodstuffs. The
results indicated that THB possesses antioxidant activity in bulk lipid that is comparable,
as far as could be measured within the limits of these experiments, to that of TBHQ and
propyl gallate. In order to distinguish the antioxidant properties of THB and THBA from
the other commonly used antioxidants, a series of experiments was designed with the aim
of determining structure activity relationships within a group of phenolic compounds
comprising two, three, or four hydroxyl groups, and with or without a carboxyl group.
Protocatechuic acid and gallic acid were chosen for these studies based on their
structures, each comprising a carboxylic acid group and two or three hydroxyl groups,
respectively, attached to a benzene ring. THBA also comprises a carboxylic acid group
and four hydroxyl groups attached to a benzene ring.

Three experimental systems were chosen for subsequent detailed analysis of
antioxidant activity of these compounds: bulk oil (lard), ground beef, and liposomes.
Selection of the ground beef and lard systems was based on the observation of Porter and
others (1989) that antioxidants display different activities depending on the food matrix.
Polar antioxidant compounds are generally more effective in bulk oils than non-polar
compounds, whereas non-polar compounds are more effective than polar compounds in
emulsions and membrane-containing food products such as muscle foods. The
compounds were also tested in the liposomal screening assay developed previously in our
laboratory (Arora and Strasburg 1997) to fiirther evaluate the assay’s efficacy as a rapid

screening tool for antioxidant activity of novel compounds.

67

2.3.]. Evaluation of Structurally Related Antioxidants in (1 Bulk Lipid System

Our preliminary studies suggested that within the limited time frame of the
experiments, THB possessed antioxidant activity similar to that of TBHQ and propyl
gallate. In order to distinguish the antioxidant properties of the phenolic compounds, we
conducted subsequent experiments at 0.01%, 0.005% and 0.001% (w/w) using the Schaal
oven test (Figure 2.3). Antioxidant efficacies of the compounds were determined by
comparison of protection factors (Table 2.1).

At the 0.01% level, THB showed the highest antioxidant activity followed by
THBA, gallic acid, alpha-tocopherol, and protocatechuic acid. After 96 days, the end of
the induction period for the THB sample still was not reached. THBA and gallic acid had
induction periods of 70 and 42 days, respectively, yielding protection factors of
approximately 10 and 7, respectively. Alpha-tocopherol showed similar activity to the
control, whereas protocatechuic acid showed slight pro-oxidant activity.

At the 0.005% level, all samples oxidized more readily. THB had an induction
period of 18 days, followed by THBA with 15 days, alpha-tocopherol with 12 days, and
gallic acid with 6 days. Protocatechuic acid was indistinguishable from the control at
approximately 3 days. Based on the protection factors of the antioxidants at 0.005%, the
antioxidant efficacy beginning with the highest activity was TBA followed by, in
decreasing order, THBA, alpha-tocopherol, gallic acid and protocatechuic acid,

respectively.

68

Figure 2.3 Antioxidant activities of gallic acid (GA), protocatechuic acid (PCA), alpha-
tocopherol, 1,2,3,4-tetrahydroxybenzene (THB), and 2,3,4,5-tetrahydroxybenzoic acid
(THBA) in lard. Concentrations of antioxidants were: (A) 0.01%, (B) 0.005% and (C)
0.001% (wt/wt).

Each data point is the mean of triplicate measurements. Error bars correspond to :t 1 SD.
Error bars not observed are smaller than the symbols.

69

Peroxide Value
(mEq hydroperoxide/kg lard)

Peroxide Value

(mEq hydroperoxide/kg lard)

PeroxideValue
(mEq hydroperoxidck/g lard)
8

wt
kl.
o.

250 .

160
140

120 A

s

c888

u
C)
O

N
M
C

§

150

s

so!

200
150 '
100 -

50"

 

35 42 49
Time (days)

(A)

 

Time (days)
(B)

 

 

 

 

DJ

9 12

Time (days)
(C)

Figure 2.3

+ PCA

+ Control

‘ -&— alpha

tocopherol
—0— GA

-9— THBA

+THB

' +PCA

+ Control

—0—GA

-- A alpha
tocopherol

-D- THBA

+ THB
+ Control
+ PCA
—B— THBA
+ GA

-A— alpha
tocopherol

Table 2.] Protection Factors of Antioxidant Compounds in Bulk Oil

 

 

 

 

 

 

 

 

Protection Factor”
Compound 0.01%(wt/wt) 0.005%(wt/wt) 0.001 %(wt/wt)
Gallic Acid 7 2 ~2
Protocatechuic Acid <1 1 ~1
Alpha-tocopherol 1 4 ~2
THB >14 6 l
THBA 10 5 ~1

 

 

 

 

 

a ratio of the induction period of the antioxidant-treated sample compared to the control

71

At the 0.001% level, THB showed slight pro-oxidant effects, whereas THBA and
protocatechuic acid showed no activity, and gallic acid and alpha-tocopherol showed the
highest antioxidant activity. Based on these data, we conclude that antioxidant activity of
THB and THBA is concentration-dependent in a bulk lipid substrate, and are superior to

the other compounds tested over the range of 0.005% to 0.01%.

2.3.2. Evaluation of Antioxidants in a Meat System

Antioxidant activities of THB and THBA were compared with the other phenolic
compounds in a cooked ground beef system using the 2-thiobarbituric acid (TBA) test.
Antioxidant activity was evaluated by calculating TBARS at 0, 24, 48, 72, and 96 hours.
As shown in Figure 2.4, antioxidant activities of THB and THBA were both significantly
greater (p<0.05) than the control at each time point, except 0 time, when only THB had
statistically significant activity (p<0.05). The antioxidant activity of THB was
significantly greater than alpha-tocopherol, protocatechuic acid and gallic acid at each
time point, and was significantly greater than THBA at 48, 72 and 96 hours. Based on
the overall TBARS values, the antioxidant activity of the compounds tested were: THB

> THBA > gallic acid > alpha-tocopherol > protocatechuic acid.

2.3.3.Evaluation of A ntioxidants in a Liposome System

Antioxidant activity of the phenolic compounds in a suspension of liposomes was
evaluated using fluorescence spectrophotometry. Lipid autoxidation was initiated by
adding Fe (II) with the resulting free radicals reacting with the fluorescent probe, which

is localized within the lipid bilayer. The rate of decay of

72

Figure 2.4 Antioxidant activities of gallic acid (GA), protocatechuic acid (PCA), alpha-
tocopherol (aT), 1,2,3,4-tetrahydroxybenzene (THB), and 2,3,4,5-tetrahydroxybenzoic
acid (THBA) in cooked ground beef. Concentration of antioxidants were 0.01% based on
fat content.

Treatments with different superscripts within a specific time point are significantly
different (p < 0.05). Results are expressed as averages of triplicate measurements. Error
bars correspond to :l: 1 SD.

73

N PCA
IIIGA
ETHB
DTHBA

.Control
E aT

 

 

zoic n
d on m

 

 
   

.........................

72

   
 
  

     

         

  

 

   

 

oooooooooooooooooooooooo
uuuuuuuuuuuuuuuuuuuuuuuu
uuuuuuuuuuuuuuuuuuuuuuu

-------------------------

g

48

     

    

      

.' . ,'.“,.. .". . .. fl"; '3' 7 ‘.':~’.:."“‘ 3,35,“:

  

 

>7

 

 

 

V).
N

3.5

(91cmS 3’1/VCIW 3w) saver.

74

Time after heating (hours)
Figure 2.4

fluorescence intensity is an indication of the rate of autoxidation. Therefore, the rate in
which fluorescence decay is inhibited by the addition of a particular compound is an
indicator of its antioxidant efficacy.

When comparing relative antioxidant effectiveness, gallic acid > propyl gallate
protocatechuic acid > THBA > THB > alpha-tocopherol (Figure 2.5). Fluorescence in
the more lipophilic compounds remained just below that of the control indicating high
antioxidant activity. THB, a polar compound and alpha-tocopherol showed minimal
activity. THBA shows biphasic decay in fluorescence with the signal first dropping
slowly and more rapidly at approximately 9 minutes. From this observation, we decided
to vary the concentration of THBA and re-evaluate its antioxidant activity in the
liposome system. Variability of these results is uncertain, but it is speculated THBA is
not as stable at lower concentrations. Further studies will need to be carried out in order
to verify this.

Figure 2.6 shows the concentration dependence of THBA in this system. At the
highest level tested, 50 uM, THBA showed a steady rate of fluorescence equivalent to
that of gallic acid (10 uM). At 20 uM THBA, fluorescence did not decrease until

approximately 9 minutes. The lower concentrations tested, 0.1 and 1 uM, showed an

immediate decay of fluorescence similar to that of THB (10 pM) and the Fe (II)

prooxidant.

75

Figure 2.5 Antioxidant activity of gallic acid (GA), propyl gallate (PG), protocatechuic
acid (PCA), 2,3,4,5-tetrahydroxybenzoic acid (THBA), alpha-tocopherol (aT), and
1,2,3,4-tetrahydroxybenzene (THB), lOuM using Fe (ID-induced peroxidation of LUVs.

Relative fluorescence represents the fluorescence intensity at a given time point divided
by the fluorescence intensity at time = 0 min. Results are expressed as averages of
triplicate measurements, except aT, n=1. Error bars correspond to :t 1 SD.

76

 

 

Relative Fluorescence F/Fo

1.2

 

 

 

 

 

 

12 15 18 21 24

Time (min)

+ Control
-0— GA
+ PG
+ PCA
-B— THBA
- A ~ aT
+ THB

+Fe (II)

 

Figure 2.5
77

 

Figure 2.6 Concentration dependence of antioxidant activity of 2,3,4,5-
tetrahydroxybenzoic acid (THBA) using Fe (ID-induced peroxidation of LUVs. THBA
was tested at various concentrations as indicated in the figure legend.

Relative fluorescence represents the fluorescence intensity at a given time point divided
by the fluorescence intensity at time = 0 min. n=1.

78

 

 

 

 

 

 

 

 

 

+ Control
1 ,'
, : = = : :——-, .‘9‘ 50 uM THBA
0
LL- .
n- 0-8 =+ 20 uM THBA
8 I
C
O
8 I
i: 0 -D- IO uM THBA
g 6
E
D
.2 "
E —x— Fe(ll)
g; 0.4
I
~+ 1 uM THBA
\K I
j—F-0.l uM
‘ \ \I‘ ‘ THBA
0
o 3 6 9 i2 15 18 2| 24
Time (min)
Figure 2.6

79

 

2.4 DISCUSSION

Previous studies suggest that compounds with multiple hydroxyl groups attached
to the aromatic nucleus will be potent antioxidants. We hypothesized that the
polyphenolic compounds, THB and THBA, would show equivalent if not greater
antioxidant activity compared to commercial antioxidants currently used in the food
industry.

Our results support the structure-activity relationships reported by other
investigators. Cuvelier and others (1992) and Chen and Ho (1997) showed that
polyphenols are more active antioxidants than monophenols in a bulk lipid system.
Gallic acid, having three hydroxyl groups, showed greater antioxidant activity than
protocatechuic acid, having only two hydroxyl groups, at each of the concentrations used
in this study. Baum and Perun (1962) showed that 1,2,3,-trihydroxybenzene was a
stronger antioxidant than 1,2-dihydroxybenzene, which in turn, was more effective than
phenoL

Similar trends in antioxidant activity were found in each substrate (corn oil versus
lard) when comparing induction periods. However, the induction period for alpha-
tocopherol was longer in lard than in stripped corn oil. This is similar to what Chen and
Ho (1997) found in their studies, but the reason for this is not clear. For all antioxidants
tested, the induction period was shorter in corn 011 than lard. This was expected because
there are more unsaturated fatty acids in corn oil than in lard, making corn oil more
susceptible to oxidation (Love 1995; Orthoefer 1995).

In ground beef, our results showed that THB and THBA were the most effective

antioxidants of those tested. Overall antioxidant activity increased as the number of

80

hydroxyl groups increased. Alpha-tocopherol was relatively ineffective in this system as
reported in other studies. Benedict and others (1975) also found alpha-tocopherol to be
an ineffective antioxidant when added at 0.005% (wt/wt) to ground beef.

The liposome system was designed as a screening assay to rapidly evaluate
extracts and pure compounds before testing in food products, which is much more time-
and labor-intensive. While tested and validated for several types of structurally related
compounds, this assay may still require further modifications for screening different
types of compounds. THB and THBA were compared in the liposomal assay with other
well-characterized antioxidant compounds. Results obtained for propyl gallate, alpha-
tocopherol, and protocatechuic acid compare favorably with those of Richman and others
(1996) and Chang and others (2003).

The THBA results from the liposomal assay also compare favorably with those
obtained in ground beef. However, based on the results from the liposomal assay, THB is
predicted to be a poor antioxidant in ground beef, whereas the results in cooked ground
beef show it is highly effective. One possible explanation for the low apparent
antioxidant activity of THB in the liposome system is that THB could be regenerating Fe
(II) from Fe (III) in a manner similar to that of ascorbic acid, and thus accelerating
oxidation.

Diphenols have greatest antioxidant activity when hydroxyl groups are attached at
the ortho or para positions. The second hydroxyl group in the ortho or para position of
phenolic compounds increases antioxidant activity by further electron delocalization of

the aromatic ring (Cuvelier and others 1992). Adding a hydroxyl group at the ortho

81

position, such as a 1,2-dihydroxybenzene derivative, is believed to stabilize the radical
through intramolecular hydrogen bonding (Figure 1.4) (Baum and Perun 1962).

Compounds with para substitution, such as 1,4-dihydroxybenzene derivatives, are
also believed to have increased antioxidant activities. This is due to production of a
semiquinone radical that can be further oxidized to form a quinone by reaction with a
lipid radical (Figure 1.3 (a)). Alternatively, the semiquinone radical may react with
another semiquinone radical to regenerate a molecule of antioxidant, and form a quinone
(Figure 1.3 (b)).

THB and THBA share structural similarity with other phenolic antioxidants such
as alpha-tocopherol, BHT, BHA, THBQ, propyl gallate, gallic acid, and protocatechuic
acid, and thus share a similar mechanism of action. THB and THBA each have four
hydroxyl groups, which suggest they may function as effective free radical scavenging
antioxidants because of their ability to stabilize the antioxidant radical by electron
delocalization. Figures 2.9 and 2.10 show the possible resonance structures of THB upon
reaction with a lipid free radical. The order in which the hydrogen atoms are donated is
unknown at this point. Baum and Perun (1962) suggest that a hydroxyl group ortho to a
phenolic radical may stabilize the radical through hydrogen bonding. We hypothesize
that the hydrogen atoms in the 2 or 3 position are more labile because of dual
stabilization of the resultant radical through hydrogen bonding by the two ortho hydroxyl
groups. Thus, the increased antioxidant activity of THB may result from dual
stabilization. of the free radicals by intramoleculer hydrogen bonding and increased

electron delocalization via the ring (Figure 2 .9).

82

IO

IO

IO
Im

.mvkx

ow

to

7:0

I”. cm

0

IO

IO

£868 RE £3, 9:. me 2:28 5:82 @2285 QN 82w?—

waif

IO

OI

OI

 

IO

83

O

\

me
o

OI

OI

OI
cm I”.

, o o
\I o
0:
CI om IN.— «0
”Vikr rill. ”Vin.
on) r.m o o
o o

i/yo
. o
o

o

2868 out 5;» m2# mo oEocom cocoa“: 88:55 o>canz< E .m oSmE

O

o /

IO IO IO

OI OI OI

I
so .70
ATILV A .AIIW.
I . o: o: m.m o:
\i/

O “my IO

84

An alternate mechanism is suggested by studies on the fate of propyl gallate
(Dziedzic and others 1986). Ellagic acid is formed from two molecules of propyl gallate,
suggesting that the hydrogen atom at the para-hydroxyl position is not lost, and therefore
may by less labile than the hydrogen atoms on the meta-hydroxyl groups. Figure 2.10
shows an alternative reaction pathway of THB with lipid free radicals with resonance
stabilization in which the hydrogen atoms are donated initially at the para positions.

Our findings indicate that THBA and THB are the most effective of the
compounds tested contradicting Pokomy (1987) who states that, pyrogallol is the most
active trihydroxybenzene, and that the presence of more than three hydroxyl groups does
not improve activity. No supporting data was provided in his paper.

In this study we have demonstrated the potential for use of low-value, by-products
of food processing as a source of potent antioxidant compounds, while enhancing the
value of the steeping liquor, which can be used as phytate-free animal feed. These
antioxidants could provide an inexpensive and efficient tool for food preservation

provided they have a favorable safety/toxicity profile.

85

CHAPTER THREE
ANTIMICROBIAL ACTIVITY OF 1,2,3,4-TETRAHYDROXYBENZENE

AND 2,3,4,5-TETRAHYDROXYBENZOIC ACID

3.1 INTRODUCTION

Food safety concerns in the United States have risen over the past ten years, in
response to several recently emerged pathogens that can cause serious illnesses and death
(National Center for Infectious Diseases). In the United States, it is estimated that 76
million cases of foodbome illnesses, including 325,000 hospitalizations and 5,200 deaths
occur each year (Mead and others 1999). According to the Economic Research Service,
the estimated annual costs of foodbome disease caused by Salmonella, Listeria
monocytogenes, and Escherichia coli 0157:H7 are $2.4 billion, $2.3 billion and $659.1
million, respectively (US. Department of Agriculture 2002).

Phenolic antioxidants, such as BHA, BHT and TBHQ, have been studied as
antimicrobial agents to inactivate, kill, or inhibit the grth of both pathogenic and
spoilage organisms in food products (Branen 1993). Use of approved antioxidants that
also express antimicrobial activity would reduce the amount of food additives introduced
into a food system and would eliminate the need for further toxicological studies.
Antioxidants with antimicrobial activity may also be used as rinses or sprays to lower
contamination on food surfaces.

In our previous work, THB and THBA, derived from phytic acid, possessed high
antioxidant activity, and thus showed significant promise as food additives. Since these

compounds are structurally related to the antibiotics aurantiogliocladin and fumigatin, we

86

 

hypothesized that THB or THBA would also have antimicrobial activity. The objective
of this study was to screen these compounds for inhibitory activity against E. coli

0157:H7, L. monocytogenes, and Salmonella spp.

3.2 EXPERIMENTAL PROCEDURES
3.2.]. Materials

Listeria monocytogenes (CWD 95, CWD 246, CWD 201 and CWD 1503),
Escherichia coli 0157:H7 (AR, AD 305, and AD 317) Salmonella Typhimurium DT
104 (H3380 and 25-2D), S. Enteritidis (clinical isolate H3527, phage type 13A) and S.
Heidelberg (clinical isolate F5038BG1) were obtained from Dr. Elliot Ryser (Michigan
State University, East Lansing, MI). All strains were stored at
-70°C in trypticase soy broth (TSB) (Difco Laboratories, Detroit, MI) containing 10%
(v/v) glycerol. Each strain was subcultured three times in TSB containing 0.6% (w/v)
yeast extract (TSB-YE) (Difco) at 35°C, 18 to 24 h before use. The two phenolic
antioxidant compounds, 1,2,3,4-tetrahydroxybenzene (THB) and 2,3,4,5-
tetrahydroxybenzoic acid (THBA) obtained from Dr. John Frost (Michigan State

University, East Lansing, M1), were synthesized as described in Figure 1.16.

3.2.2. Methods

Disc Diffusion Assay:
Treatment preparation: Saturated solutions of THB and THBA were prepared at

room temperature in deionized water; the approximate concentrations of the saturated

87

solutions were 1.4 M. Dilutions of 1/ 10, 1/ 100, 1/ 1000, and 1/ 10000 were prepared from

the saturated solutions.

Assay: Twenty mL of trypticase soy agar containing 0.6% yeast extract (TSA-
YE) (Difco) was seeded with 200 uL of an 18-24 h-old culture of S. Typhimurium DT
104 (25-2D) (108 chL), L. monocytogenes 1503 (109 cfu/mL) or E. coli 0157:H7 (AD
305) (109 cfu/mL) and transferred to a Petri dish (100 mm diameter x 15 mm thickness).
After solidification, a 16-mm diameter Whatman #4 filter disk (Whatman Co.,
Maidstone, England) containing 50 uL of THB or THBA of the desired concentration
was placed aseptically in the center of each agar plate. Disks were prepared by soaking
them in 50 DL of treatment. After 24 h of incubation at 35°C, the inhibition zone
diameters were measured perpendicularly to the nearest m using a metric ruler. All tests
were run in duplicate with duplicate measurements averaged. Filter disks containing only

50 uL of sterile distilled water served as negative controls.

Minimum Inhibitogy Concentration (MIC)

All strains were evaluated for MIC. In addition, the following three-strain
cocktails were prepared for MIC determination:

l. L. monocytogenes (CWD 95, CWD 246, and CWD 201)

2. E. coli 0157:H7 (AR, AD 305, and AD 317)

3. Salmonella Typhimurium DT 104 (H3380), S. Enteritidis and S. Heidelberg

Culture preparation: All strains were cultured as described above.

88

Cocktails were prepared by combining 3-mL aliquots of each strain within each
species. After centrifugation at 9740 x g for 15 minutes at 4°C (Sorvall Super T21,
Newtown, CT), the pellet was resuspended in 9 mL peptone broth (0.1%).

Tubes of TSB-YE containing 0.0014, 0.0028, 0.0056, 0.0070, 0.0084, 0.0098,
0.0112, 0.0126, 0.0140 M THB were inoculated with 10 uL of each test organism (109
cfu/mL) or cocktail (10IO cfu/mL) to give a final count of 10° cfu/mL and 107 cfu/mL,
respectively, in each tube. All tubes were incubated at 35°C and observed for growth
after 24 and 48 h. All tests were run in triplicate with results reported as the minimum
inhibitory concentration (MIC) at which no growth was visibly present. Tubes without

THB served as negative controls.

Statistical Analysis: Statistical analysis was performed using a complete block
design, where block = time (which has two levels 24 hours and 48 hours) and treatment is
the main effect. The GLM generalized linear model (SAS® Version 7, Cary, NC) was
used to perform the statistical analysis. Also LSMEANS was used in the GLM procedure

to obtain pairwise comparison of the treatments.

89

 

3.3 RESULTS
3.3.]. Disk Diffusion Assay

THB and THBA were initially screened for antimicrobial activity using a disk
diffusion assay. The 100% saturated solution of THB yielded inhibition zones of 28, 25,
and 33 mm for E. coli 0157:H7 (AD 305), S. Typhimurium DT 104 (25-2D) and L.
monocytogenes (1503), respectively, and 19 mm for each of these strains when using a
10% saturated solution of THB (Figure 3.1). Concentrations less than 10% saturated
solution of THB did not show antimicrobial activity. THBA was noninhibitory at all
concentrations tested, even at 100% saturation.

Based on these results, the MIC of THB was evaluated for L. monocytogenes
(CWD 95, CWD 246, and CWD 201), E. coli 0157:H7 (AR, AD 305, and AD 317) and
Salmonella Typhimurium DT 104 (H3380), S. Enteritidis and S. Heidelberg using a

broth assay.

3.3.2. Minimum Inhibitory Concentration

In TSB-YE, E. coli 0157:H7 (AR) and L. monocytogenes (CWD 95) were
inhibited at 0.0042 M THB and 0.0056 M THB at 24 and 48 h, respectively (Table 3.1).
S. Enteritidis and S. Heidelberg were both inhibited at 0.0056 M THB after 24 and 48
hours. The Salmonella cocktail was inhibited by 0.0028 M THB at 24 h, whereas the E.
coli 0157:H7 (AD 305, AD 317 and AR) and L. monocytogenes (CWD 201, CWD 246
and CWD 95) cocktails were inhibited by 0.0056 M THB at 24 h and 0.0084 and 0.0056
M at 48 h, respectively. The concentration and time effect of the L. monocytogenes

(CWD 201, CWD 246 and CWD 95) cocktail is significantly different (p<0.05) from the

90

 

ITHB 100 %
saturation '

E THB 10%
saturation

Inhibition Zone (mm:

 

 

 

Salmonella E. coli 015 7:H7 L. monocytogenes

Figure 3.1 Inhibition of S. Typhimurium DT 104 (25-2D), E. coli 0157:H7 (AD 305) and
L. monocytogenes 1503 by THB in a disk diffusion assay.

Error bars correspond to i 1 SD. Results are expressed as the average of duplicate
experiments.

91

 

 

 

 

 

Table 3.1
Organism MIC 1 (M x 10°)
24 h 48 h

E.coli 0157:H7

AD 3053 7.0 11.2

AD 317b 8.4 9.8

AR° 4.2 5.6

Cocktail° 5,5 34
Salmonella

Typhimurium a 5.6 7.0

Enteritidisc 5.6 5.6

Heidelberg° 5.6 5.6

Cocktail° 2.8 4_2
L. monocytogenes

CWD 201c 5.6 5.6

CWD 246c 5.6 5.6

2CWD 95b 42 -

Cocktailc 5.6 5.6

 

I Minimum inhibitory concentration- the lowest
concentration at which complete inhibition of growth was
observed.

2 No inhibition at 48 h

Table 3.1 Minimum inhibitory concentrations of THB for E. coli 0157:H7, L.
monocytogenes and Salmonella spp.

Different letters show significant difference (p <0.05) of a complete block design, where
block = time (which has two level 24 hours and 48 hours) and treatment is the main
effect.

92

 

cocktails for Salmonella and E. coli.

3.4 DISCUSSION

The aim of this study was to determine if the antioxidant compounds THB and
THBA possess antimicrobial activity. Using the disk diffusion assay as a screening
method, THB showed significant inhibitory activity against all three pathogens, whereas
THBA showed no activity. In subsequent experiments the minimum inhibitory
concentration (MIC), defined as the lowest concentration of which complete inhibition is
observed, was determined for THB. The results indicate that THB is inhibitory for
growth of the pathogens tested over a range of 2.8-8.4 mM at 24 h and 5.6-11.2 mM at 48
h.

While direct comparison of MIC for THB with other compounds is difficult, some
generalizations can be made. The MIC found in this study using THB against the
Salmonella cocktail was considerably lower to BHA, which is most recognized for its
antioxidant activity, but is also used for its antimicrobial properties. Tuncel and Nergiz
(1993) found that BHA at a concentration of 400 ug/mL inhibited Staphylococcus
aureus, Bacillus cereus, E. coli, and S. Typhimurium when grown in trypticase soy broth
at 37°C for 18 h. The Salmonella cocktail (107 Cfu/mL) in our study was inhibited by
THB with MICs of 2.8 and 4.2 mM THB after 24 and 48 h, respectively. The L.
monocytogenes cocktail was inhibited by 5.6 mM THB at both 24 and 48 h. While the
Mle in each of these studies are quite similar, our incubation times were longer. Other
phenolic antioxidants with well documented antimicrobial activity include BHT and

TBHQ (Raccach 1984; Davidson 2001).

93

The mechanism by which antioxidants inhibit microbial growth is not well
defined (Davidson 2001). However, an antioxidant must be sufficiently lipophilic to
pass through the cell membrane and once inside, it must also be sufficiently water-soluble
to interact with cell components. Hydrophobic compounds are more effective
antimicrobial agents against Gram positive bacteria, whereas Gram negative bacteria are
more susceptible to hydrophilic compounds (Davidson and Branen 1980; Raccah 1984;
Davidson 1993). THB was generally more effective against L. monocytogenes and
Salmonella than E. coli 0157:H7.

Steric effects also influence the antimicrobial activity of phenolic antioxidants.
When several n-alkyl BHT deriviatives were studied by Kabara (1980) substitution of a
bulky tert-butyl group meta to the phenolic group instead of ortho, as in BHT, lowered
the MIC from 250 ug/mL to 25 ug/mL against Streptococcus mutans in trypticase soy
broth. They concluded that BHT, with the hindered phenolic group, was far less effective
than BHA, in which the hydoxyl group is much less hindered.

Even though the exact mode of action for antimicrobial agents is poorly
understood, numerous studies suggest the antimicrobial mechanism of phenolic
compounds is attributed to their action on the cellular membrane, as well as altered
synthesis of nucleic acids, proteins and lipids (Eklund 1980; Raccach 1984; Davidson
2001). Parabens effectively disrupt cell membrane integrity leading to increased
permeability and leakage of cell contents that can lead to death of the organism
(Davidson 2001). The antimicrobial mechanism of THB may be similar to that of the

parabens.

94

THB is a potent antioxidant with good antimicrobial activity, thus making the
compound an attractive multifunctional food ingredient. Possible THB applications in
the food industry include multistage or combination spray washing of meats, fruits and
vegetables. Currently used sanitizers include organic acids, hydrogen peroxide, and
hypochlorite compounds but their use is quite limited. Hydrogen peroxide is an effective
antimicrobial agent, but a decrease in organoleptic quality of some produce is a
consequence (Sapers and Simmons 1998). Chlorine is also often used, but dissipates
quickly in water with high levels of organic residue in addition to creating high levels of
hazardous fumes (Anon 1999).

Results from these studies suggest that THB has significant promise for use as a
food ingredient. Further studies are warranted to determine more precisely its

antimicrobial activity in food processing applications.

95

CHAPTER FOUR

SUMMARY AND CONCLUSIONS

In this study we hypothesized that the polyphenolic compounds, THB and THBA,
would show equivalent or greater antioxidant activity compared to other commercial
antioxidants used currently in the food industry. THB and THBA share structural
similarities with the antimicrobial compounds fumigatin and aurantiogliocladin as well.
Thus, our second hypothesis was that THB and THBA would also possess antimicrobial
activity.

The first objective of this study was to evaluate the antioxidant activities of THB
and THBA in three different experimental systems; bulk oils, ground beef and liposomes.
THB and THBA displayed high antioxidant activity in the bulk oil, comparable or
superior to all of the commercial antioxidants tested. THB and THBA also showed
significant antioxidant activity in ground beef. THB did not show activity in the
liposome system at the concentration tested. THBA showed moderate activity at the
same concentration of the other antioxidants tested and its activity increased at higher
concentrations. The antimicrobial activity of THB was comparable to other phenolic
antioxidants used for antimicrobial application.

In this study we have shown an alternative use for natural-product extracts.
Abundant, low-value by-products of food processing could be used as raw materials for
the synthesis of novel antioxidant compounds. The antioxidant activity of THB and
THBA offers an advantage to the food processing industry and these compounds are

potential value-added products for cereal manufacture. Removal of phytic acid from the

96

steeping liquor for use in synthesis of novel antioxidants enhances its value as well as
that of the steeping liquor, which can be used as phytate-free animal feed. By expressing
both antioxidant and antimicrobial activity, THB may serve as a multifunctional food
ingredient. Reduction in the amount and types of food additives introduced into a food
system and the elimination of toxicological studies is reduced by using a compound

which expresses both antioxidant and antimicrobial properties.

97

 

CHAPTER FIVE

FUTURE RESEARCH
Some suggestions for future research are listed below:

THB and THBA showed potent antioxidant activity in both bulk oil and meat
systems. To demonstrate their potential as active food ingredients, their activities
should be tested in other food products, including processed meats, emulsions and

cereals.

1,2-dihydroxyquinone, which is formed from THB upon reaction with a lipid free
radical should be evaluated for potential antioxidant activity to help in

clarification of the mechanism of THB activity.

In the liposome system, iron-induced oxidation was used to evaluate antioxidant
effectiveness of THB and THBA. The results indicated that THB would not be
an effective antioxidant in contrast to our findings in the oil and meat systems.
One explanation for this is that upon oxidation of iron, THB might be
regenerating Fe (11), thus accelerating oxidation. The effect of THB with iron
may be better understood, if other prooxidant agents, such as a hydrophilic free
radical initiator, e.g. 2,2’-azobis (2-amidinopropane) dihydrochloride (AAPH),
are used to initiate autoxidation in this same system. In this study, liposomes

were formed using an electrically neutral lipid substrate. A negatively charged

98

substrate, (e.g. phosphatidyl serine) when used in combination with SLPC, would
be more representative of a biological membrane and may help to improve

validity of this assay.

We have shown that using fluorescence spectroscopy is a fast and easy way to
evaluate lipid oxidation in liposomes. Development of a screening assay using a
bulk lipid coupled with fluorescence would be very beneficial for evaluating

antioxidant compounds.

THB shows promising antimicrobial activity. A more in depth study will help to
fully understand its mechanism of action. By comparing THB to other phenolic
antioxidants, a better comparison as its relative effectiveness would be useful.
BHA has antifungal activity at 200 ppm against Aspergillus, Penicillium, and
Geotrichum in glucose salt broth when incubated at 32°C, for 85 h (Ahmad and
Branen 1981). The structural similarity of THB and BHA suggests that
compounds, should be tested against other organisms such as spore & non-spore
forming, spoilage bacteria, pathogenic bacteria and molds & yeasts. Evaluation
at different temperatures may also help in understanding the mechanism in which

THB expresses antimicrobial activity.

99

LIST OF REFERENCES

Agency for Toxic Substances and Disease Registry. 1999. Tox FAQs for Phenol.
CAS#108-95-2.

Ahmad S, Branen AL. 1981. Inhibition of mold growth by butylated hydroxyanisole. J
Food Sci 46: 1059-1063.

Antolovich M, Prenzler PD, Patsalides E, McDonald S, Robards K. 2002. Methods for
testing antioxidant activity. Analyst 127: 183-198.

Arora A, Nair MG, Strasburg GM. 1997. A novel fluorescence assay to evaluate
antioxidant efficacy: Application to flavonoids and isoflavonoids. In: Aruoma O,
Cuppett S. editors. Antioxidant Methodology. Champaign, Illinois: AOAC Press.
p 205-222.

Arora A, Nair MG, Strasburg GM. 1998. Structure-Activity relationships for antioxidant

activities of a series of flavonoids in a liposome System. J Free Radic Biol Med
24: 1355-1363.

Arora A, Strasburg GM. 1997. Development and validation of fluorescence
spectroscopic assays to evaluate antioxidant efficacy. Application to metal
chelators. J Am Oil Chem Soc 74: 1031-1040.

Aust SD, Morehouse LA, Thomas CE. 1985. Role of metals in oxygen radical reactions.
J Free Radic Biol. Med 1: 3-25.

Baum BO, Perun AL. 1962. Antioxidant efficiency versus structure. Soc Plast Engrs
Trans 2: 250-257.

Belitz HD, Grosch W. 1987. Lipids. In: Belitz HD, Grosch W, authors. Food Chemistry.
Translation from the second German edition by D. Hadziyev. Berlin, Germany:
Springer Verlag. p 127-200.

Benedict RC, Strange ED, Swift CE. 1975. Effect of lipid antioxidants on the stability of
meat during storage. J Agric Food Chem 23: 167-173.

Bennion M,'Park RL. 1968. Changes in frying fats with different foods. I Am Diet Assoc
52: 308-312.

Branen AL. 1993. Introduction to use of antimicrobials. In: Davidson PM, Branen AL,
editors. Antimicrobials in Foods. New York: Marcel Dekker Inc. p l-10.

100

Center for Food Safety and Applied Nutrition. 2003. US. Foodbome Pathogenic
Microorganisms and Natural Toxins Handbook. “Bad Bug Book”. Food and Drug
Administration: Washington, DC. http://www.cfsan.fda.gov/~mow/intro.html

Chang Y, Almy E, Blamer GA, Gray JI, Frost JW, Strasburg GM. 2003. Antioxidant
activity of 3-dehydroshikimic acid in liposomes, emulsions, and bulk oil. J Agric
Food Chem 51: 2753-2757.

Chapman D. 1984. Physicochemical properties of phospholipids and lipid-water systems.
In: Gregoriadis G, editor. Liposome Technology. 9th ed.,Vol. I. Boca Raton, FL:
CRC Press. p 1-18.

Chen JH, Ho C. 1997. Antioxidant activity of caffeic acid and its related
hydroxycinnamic acid compounds. J Agric Food Chem 45: 2374-23 78.

Code of Federal Regulations. 2002. Title 21. US. Government Printing Office via GPO
Access.

Coupland JN, McClements DJ. 1996. Lipid oxidation and food emulsions. Trends Food
Sci Tech 3: 83-91.

Cuvelier ME, Richard H, Berset C. 1992. Comparison of the antioxidative activity of
some acid-phenols: Structure-activity relationship. Biosci Biotech Biochem 56:
324-325.

Cuvelier ME, Bondet V, Berset C. 2000. Behavior of phenolic antioxidants in a
partitioned medium: Structure-activity Relationship. J Am Oil Chem Soc 77: 819-
823.

Crackel RL, Gray J1, Pearson AM, Booren AM, Buckley DJ. 1988. Some further
observations on the TBA test as an index of lipid oxidation in meats. Food Chem
28: 187-196.

D’Aoust J, Maurer J, Bailey JS. 2001. Salmonella Species. In: Doyle MP, Beuchat LR,
Montville TJ, editors. Food Microbiology: fundamentals and frontiers. 2nd
Edition. New York: Marcel Dekker Inc. p 141-178.

Dahle LK, Hill EG, Holman RT, 1962. The thiobarbituric acid reaction and the
autoxidation of polyunsaturated fatty acid methyl esters. Arch. Biochem. Biophys
98: 253-261.

Davidson PM, Branen AL 1980. Antimicrobial mechanisms of butylated hydroxyanisole
against two Pseudomonas species. J Food Sci 45: 1607-1613.

Davidson PM. 1993. Parabens and phenolic compunds. In: Davidson PM, Branen AL,
editors. Antimicrobials in Foods. New York: Marcel Dekker Inc. 263-306.

101

Davidson, PM. 2001. Chemical preservatives and natural antimicrobial compounds. In:
Doyle, M.P., Beuchat, L.R., Montville, T.J., editors. Food microbiology-
fundamentals and frontiers. 2"d Edition. New York: Marcel Dekker Inc. p 593-
627.

DeMan JM. 1999. Lipids. In: DeMan JM, editor. Principles of Food Chemistry. MD:
Aspen Publishers.

Dugan L. 1976. Lipids. In: Fennema OR, editor. Food Chemistry. 1"” Edition. New York:
Marcel Dekker. p 139-203.

Duthie GG. 1991. Antioxidant hypothesis of cardivascular disease. Trends Food Sci
Tech 2: 205-207.

Dziedzic SZ, Robinson JL, Hudson JFB. 1986. Fate of propyl gallate and
diphosphatidylethanolamine in lard during autoxidation at 120°C. J Agric Food
Chem 34: 1027-1029.

Eklund T. 1980. Inhibition of grth and uptake processes in bacteria by some chemical
food preservatives. The Society for Applied Bacteriology 48: 423-432.

Frankel EN. 1998. Lipid Oxidation. Scotland: Oily Press Ltd. 303 p.

Frankel EN. 1993. In search of better methods to evaluate natural antioxidants and
oxidative stability in food lipids. Trends Food Sci Tech 4: 220-225.

Frankel EN. 1991. Review: Recent advances in lipid oxidation. J Food Sci Agric 54:
495-511.

Frankel EN, Neff WE. 1983. Formation of malonaldehyde from lipid oxidation products.
Biochim Biophys Acta 754: 264-270.

Giese J. 1996. Antioxidants: Tools for preventing lipid oxidation. Food Tech 50: 73-80.

Gordon. 1990. In: Chan HWS, editor. Autoxidation of unsaturated lipids. New York:
Academic Press. 296 p.

Graf E, Empson KL. 1987. Phytic acid-a natural antioxidant. J Biol Chem 262: 11647-
11650.

Gray JI, Monahan FJ. 1992. Measurement of lipid oxidation in meat and meat products.
Trends Food Sci Tech 4: 315-319.

Gutteridge JMC, Halliwell B. 1990. The measurement and mechanism of lipid
peroxidation in biological systems. Trends Biochem Sci 15: 129-135.

102

Hands E. 1996. Lipid composition of selected foods. In: Hui YH, editor. Bailey’s
industrial oil and fat products. 5th Edition: Volume 1 New York: John Wiley &
Sons, Inc. p 441-505.

Hansen C A, Dean AB, Draths KM, Frost JW. 1999. Synthesis of 1,2,3,4-
tetrahydroxybenzene from D-glucose: Exploiting myo-inositol as a prescursor to
aromatic chemicals. J Am Chem Soc 121: 3799-3800.

Hare] S, Kanner J. 1985. Muscle membranal lipid peroxidation initiated by H202-
activated metrnyoglobin. J Agric Food Chem 33: 1188-1192.

Hoseney RC, 1986. Wet Milling: Production of starch, oil and protein. In: Principles of
cereal science and technology. 2“° Edition. St. Paul, MN: American association of
cereal chemists Inc. p 153-166.

Hoyland DV, Taylor A J. 1991. A review of the methodology of the 2-thiobarbituric acid
test. Food Chem 40: 271-291.

Hsieh R J, Kinsella JE. 1989. Advances in food and nutrition research. San Diego,
California: Academic Press. 33: 233-247.

Jacobson C. 1999. Sensory Impact of Lipid Oxidation in Complex Food Systems. Fett :
Zeitschrift fur Wissenschatt und Technologie der F ette, Ole und Wachse = Lipid :
journal for science and technology of fats, oils, and waxes 101: 484-492.

Kabara JJ. 1980. GRAS antimicrobial agents for cosmetic products. J Soc Cosmet Chem
31: 1-10.

Ladikos D, Lougovois V. 1990. Lipid oxidation of muscle foods: A review. Food Chem
295-314.

Langen S. 2002. Danico begins antioxidant production in US. Food Tech 56: 10.

Lee BJ, Hendricks DG. 1982. Nutritional bioavailability of iron. MD: American
Chemical Society.

Lou Y, Yousef. 1999. Charactoristics of Listeria monocytogenes impantant to food
processors. In: Ryser ET, Maith EH, editors. Listeria, listeriosis and food safety.
2nd Ed. New York: Marcel Dekker, Inc. p 131-224.

Love J. 1995. Animal fats. In: Hui YH, editor. Bailey’s industrial oil and fat products. 5’h
Edition: Volume 1 New York: John Wiley & Sons, Inc. p 1-18.

103

 

l...

 

 

MacDonald RC, MacDonald RI, Menco B, Takeshita K, Subbarao NK, Hu L, 1991.
Small-volume extrusion apparatus for preparation of large, unilamellar vesicles.
Biochimica Biophysica Acta 1061: 297-303.

Marmer WN, Maxwell RJ. 1981. Dry column method for the quantitative extraction and
simultaneous class separation of lipids from muscle tissues. Lipids 16: 365.

Mead PS, Slutsker L, Dietz V, McCaig LF, Bresee J S, Shapiro C‘, Griffin PM, Tauxe RV.
1999. Food-Related Illness and Death in the United States. Emerging Infectious
Diseases 5: 607-625.

Meng J, Doyle MP, Zhao T, Zhao S. 2001. Enterohemorrhagic Escherichia coli. In:
Doyle MP, Beuchat, LR, Montville TJ, editors. Food Microbiology-
fundamentals and frontiers. 2nd Edition. New York: Marcel Dekker Inc. p 193-
214.

Melton SL. 1983. Methodology for following lipid oxidation in muscle foods. Food
Tech.37: 105-111.

Miller DD. 1996. Minerals. In: Fennema OR, editor. Food Chemistry. 3rd Edition. New
York: Marcel Dekker. p 617-650.

Mukhopadhyay M. 2000. Natural Antioxidants. In: Mukhopadhyay M. editor. Natural
extracts using supercritical carbon dioxide. Boca Raton FL: CRC Press. p 225-
249.

National Center for Infectious Diseases: Center for Disease Control
http://www.cdc.gov/noidod/dbmd/diseaseinfo/foodborneinfections_g.htin

Nawar WW. 1996. Lipids. In: Fennema OR, editor. Food Chemistry. 3rd Edition. New
York: Marcel Dekker. p 225-231.

Orthoefer F. 1995. Vegetable oils. In: Hui YH, editor. Bailey’s industrial oil and fat
products. 5th Edition: Volume 1. New York: John Wiley & Sons, Inc. p 19-43.

Payne KD, Rico-Munoz B, Davidson PM. 1989. The antimicrobial activity of phenolic
compounds against Listeria moncytogenes and their effectiveness in a model milk
system. J Food Prot 52: 151-153.

Pike 0A. 1998. Fat characterization. In: Nielsen S, editor. Food Analysis. 2nd Edition.
Gaithesburg, MD: Aspen Publishers. Inc. p 217-235.

Pokomy J. 1987. Major factors affecting the autoxidation of lipids. In: Chen HWS,

editor. Autoxidation of unsaturated lipids. London, England: Academic Press,
Inc. p 141-206.

104

 

Porter NA, Lehman LS, Weber BA, Smith KJ. 1981. Unified mechanism for
polyunsaturated fatty acid autoxidation. Competition of peroxy radical hydrogen
atom abstraction, B-scission, and cyclization. J Am Chem Soc 103: 6447.

Porter WL, Black ED, Drolet AM. 1989. Use of polyamide oxidative fluorescence test on
lipid emulsions: Contrast in relative effectiveness of antioxidants in bulk verses
dispersed systems. J Agric Food Chem 37: 615-624.

Pratt D. 1996. Antioxidants: Technical and regulatory considerations. In: Hui YH, editor.
Bailey’s industrial oil and fat products. 5th Edition: Volume 3. New York John
Wiley & Sons, Inc. p 523-545.

Quillin P, Quillin N. 2001. Beating Cancer by Nutrition. Oklahoma: Nutrition Times
Press Inc. 352 p.

Raccach M. 1984. The antimicrobial activity of phenolic antioxidants in foods: A review.
J Food Safety 6: 141-170.

Ramarathnam N, Osawa T, Ochi H, Kawakishi S. 1995. The contribution of plant food
antioxidants to human health. Trends Food Sci Tech 6: 75-82.

Richman J E, Chang Y, Kambourakis S, Draths KM, Almy E, Snell K, Strasburg GM,
Frost JW. 1996. Reaction of 3-dehydroshikimic acid with molecular oxygen and

hydrogen peroxide: products, mechanism, and associated antioxidant activity. Am
Chem Soc 118: 11587-11591.

Sapers GM, Simmons GF. 1998. Hydrogen peroxide disinfection of minimally processed
fruits and vegetables. Food Tech 52: 48-52.

Schmidt K. 2000. Lipids: Functional Properties. In: Christen GL, Smith J S, editors. Food
Chemistry: Principles and applications. West Sacramento, California: Science
Technology System. p 97-113.

Seeliger HPR, Jones D. 1984. Listeria. In: Sneath, P HA, editor. Bergey’s manual of
determinative bacteriology. Baltimore, MD: Williams & Wilkins. 1235-1245.

Shahamat M, Seaman A, Woodbine M. 1980. Survival of Listeria monocytogenes in
high salt concentrations. Zentralblatt fur Bakteriologie, Mikrobiologie und
Hygiene. 1. Abt. Originale A, Medizinische Mikrobiologie, lnfektionskrankheiten
und Parasitologie = Medical microbiology, infectious diseases, parasitology 246:
506-511.

Sinnhuber RO, Yu TC. 1958. Characterization of the red pigment formed in the 2-
thiobarbituric acid determination of oxidative rancidity. Food Res 23: 626-634.

105

Strasburg GM, Ludescher RD. 1995. Theory and applications of fluorescence
spectroscopy in food research. Trends Food Sci Tech 6: 69-75.

Stauffer CE. 1996. Fats and oils. St. Paul, MN: Eagan Press Handbook Series. American
association of cereal chemists. 149 p.

Stueben H. 2002. Personal Interview. 5 Aug.

Swaminathan B. 2001. Listeria monocytogenes. In: Doyle MP, Beuchat LR, Montville
TJ, editors. Food microbiology- fundamentals and frontiers. 2"° Edition. New
York: Marcel Dekker Inc. p 383-410.

Tarladgis BG, Watts BM, Younathan MT. 1960. Distillation method for the quantitative
determination of malonaldehyde in rancid foods. J Am Oil Chem Soc 1: 44-48.

Taoukis P, Labuza T. 1996. Summary: Integrative concepts. In: Fennema OR, editor.

Food chemistry: Principles and applications. 3'° Edition. New York: Marcel
Dekker Inc. p 1013-1042.

Tuncel G, Nergiz C. 1993. Antimicrobial effect of some olive phenols in a laboratory
medium. Letters in Applied Microbiology 17: 300-302.

US. Department of Agriculture: Economic Research Service. 2002. Economics of
foodbome disease. http://wwwers.usda.gov/briefing/FoodbomeDisease/

U. S. Food and Drug Administration: 1998: Food Safety: A Team Approach. FDA
Backgrounder. http://www.fda.gov/opacom/backgrounders/foodteam.litml

Vijayakumar C, Woff-Hall CE. 2002. Minimum bacteriostatic and bactericidal
concentrations of household sanitizers for Escherichia coli stains in tryptic soy
broth. Food Microbiology 19: 383-388.

Wood M. 2000. Feeling weak? Try the tortillas! Agricultural Research 48: 13.

106