EFFECTS OF LOW AND HIGH pH ON LIPID OXIDATION IN SPANISH MACKEREL
By
HOLLY TASHA PETTY
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
2007
1
© 2007 Holly Tasha Petty
2
To all with a passion for knowledge and the compassion to share it.
3
ACKNOWLEDGMENTS
I thank my supervising committee chair (Hordur G. Kristinsson) for the exceptional
privilege of conquering a future world problem, for his mentorship, for his friendship, and for
giving me the courage to realize my potential and dreams. I thank my supervisory committee
(Steve Talcott, Marty Marshall, and Charlie Sims) for their valuable expertise, instruction, and
revolutionary contributions to the academic field of food science. I thank the external member on
my committee (Christian Leeuwenburgh) for helping me realize my passion for applied
physiology and kinesiology and for providing worldly understanding of free radicals, aging, and
life. I especially thank Wei Huo for his statistical expertise and generous help. Others in my
graduate department who especially provided me with support include Maria Plaza, Dr. Ross
Brown, Susan Hillier, Carmen Graham, Bridget Stokes, Maria Ralat, Marianne Mangone, and
Walter Jones.
I would like to thank all of my family and many influential friends for encouraging me to
fulfill this dream. I specifically thank Andrew Piercy for his kindness and patience and Sean
McCoy for his unconditional inspiration and contribution. I thank all of my lab peers for lending
me knowledge, support, tears, and smiles. I thank Captain Dennis Voyles and his family for their
friendship and the Spanish mackerel capture of a lifetime. I thank those I have not specifically
acknowledged or who crossed my graduate path, for their positive influences and support. Lastly,
I am thankful, for the true lessons I have learned, and irreplaceable treasures I have earned.
4
TABLE OF CONTENTS
page
ACKNOWLEDGMENTS ...............................................................................................................4
LIST OF TABLES...........................................................................................................................9
LIST OF FIGURES .......................................................................................................................10
ABSTRACT...................................................................................................................................12
CHAPTER
1
INTRODUCTION ..................................................................................................................14
2
LITERATURE REVEIW .......................................................................................................16
Obtaining a Fish Protein Isolate .............................................................................................16
Conventional Surimi Processing .....................................................................................16
Acid/Alkali-Aided Processing.........................................................................................17
Advantages of the Acid/Alkali-Aided process ................................................................17
Lipids in Fish Muscle .............................................................................................................19
Lipid Composition ..................................................................................................................19
Lipid Oxidation in Fish...........................................................................................................20
Autocatalytic Mechanism................................................................................................20
Transition Metals: Catalysts of Lipid Oxidation .............................................................21
Heme vs. Non Heme Iron and Lipid oxidation ...............................................................21
Heme Proteins and Autoxidation............................................................................................22
Hemoglobin vs Myoglobin and Lipid Oxidation....................................................................23
Lipid Content and Lipid Oxidation.........................................................................................23
Effects of Extreme pH on Lipid Oxidation in Fish Muscle.............................................24
Lipid Oxidation and Antioxidant Implementation ..........................................................25
Lipid oxidation and Proteins ...........................................................................................26
Protein Oxidation....................................................................................................................27
Mechanisms of Protein Oxidation ..........................................................................................28
Protein Scission ......................................................................................................................29
Protein Polymerization ...........................................................................................................29
Protein Carbonylation.............................................................................................................30
Lipid Mediated Protein Oxidation ..........................................................................................31
Protein Oxidation Accumulation ............................................................................................31
Protein Oxidation and Muscle Food Quality ..........................................................................32
Protein oxidation and conformational changes in proteins .............................................33
Functional properties and protein oxidation....................................................................34
Significance ............................................................................................................................36
3
EFFECTS OF ALKALI/ACIDIC ADJUSTMENT ON LIPID OXIDATION IN
SPANISH MACKEREL HOMOGENATE ...........................................................................37
5
Introduction.............................................................................................................................37
Materials and Methods ...........................................................................................................38
Raw Materials.........................................................................................................................38
Chemicals ...............................................................................................................................39
Storage Studies and Preparation of Homogenate ...................................................................39
Antioxidant Implementation...................................................................................................40
Total Lipid Analysis ...............................................................................................................40
Lipid Hydroperoxides.............................................................................................................40
Thiobarbituric Acid Reactive Substances (TBARS) ..............................................................41
Analysis of Moisture and pH..................................................................................................41
Statistical Analysis..................................................................................................................41
Results.....................................................................................................................................42
Changes in Spanish Mackerel Lipid Oxidation ......................................................................42
Water and lipid Content..........................................................................................................42
Lipid Oxidation in Whole and Minced Spanish Mackerel .....................................................43
Lipid Oxidation in pH Treated Spanish Mackerel Homogenates...........................................43
Changes in homogenate lipid oxidation with storage......................................................44
Changes in lipid oxidation with pH and storage .............................................................45
Changes in lipid oxidation with initial quality and pH....................................................45
Changes in lipid oxidation with holding time and pH.....................................................46
Overall pH effects............................................................................................................47
Lipid Oxidation of Homogenates with Antioxidant Implementation.....................................47
Lipid Oxidation of Homogenates after Readjustment to pH 5.5 and 7 ..................................49
Batch and homogenate pH readjustment .........................................................................50
Time and homogenate pH readjustment..........................................................................51
Discussion...............................................................................................................................51
Lipid Oxidation and Storage Stability ....................................................................................51
Lipid Oxidation in Acid/Alkali Adjusted Homogenates ........................................................52
Homogenate pH Readjustment ...............................................................................................58
Antioxidant Implementation...................................................................................................60
EDTA and Citric Acid Implementation at Low pH................................................................64
Conclusion ..............................................................................................................................65
4
THE EFFECTS OF THE ALKALI/ACIDIC AIDED ISOLATION PROCESS ON
LIPID AND PROTEIN OXIDATION ...................................................................................75
Introduction.............................................................................................................................75
Materials and Methods ...........................................................................................................76
Raw Materials..................................................................................................................76
Chemicals ........................................................................................................................76
Storage studies and Preparation of Homogenate and Isolate ..........................................77
Thiobarbituric Acid Reactive Substances (TBARS) .......................................................77
Protein Quantification and Characterization ...................................................................78
Rheology..........................................................................................................................79
Protein Oxidation.............................................................................................................79
Analysis of Moisture and pH...........................................................................................80
Statistical Analysis ..........................................................................................................80
6
Results.....................................................................................................................................81
Oxidative Stability of Acid/Alkali Derived Protein Isolates...........................................81
Gel Forming Ability of Isolates.......................................................................................84
Protein Composition of Homogenates and Recovered Isolate Fractions ........................84
Discussion...............................................................................................................................85
Changes in pH of Mince with Storage ............................................................................85
Lipid Oxidation During Extreme pH Adjustment and Readjustment to 5.5 ...................86
Protein Oxidation.............................................................................................................89
Gel Forming Ability ........................................................................................................92
Textural Changes and Protein and Lipid Oxidation ........................................................95
SDS page .........................................................................................................................97
Conclusion ..............................................................................................................................98
5
EFFECT OF ACID PROCESSING ON LIPID OXIDATION IN A WASHED
SPANISH MACKEREL MUSCLE MODEL SYSTEM .....................................................107
Introduction...........................................................................................................................107
Materials and Methods .........................................................................................................108
Raw Materials................................................................................................................108
Chemicals ......................................................................................................................108
Washed Fish ..................................................................................................................108
Preparation of Spanish Mackerel Hemolysate ..............................................................109
Hemoglobin Quantification ...........................................................................................110
Spanish mackerel Pressed juice.....................................................................................110
Thiobarbituric Acid Reactive Substances (TBARS) .....................................................110
Spanish mackerel Pressed Juice protein Quantification and Characterization..............110
Preparation of Low Moisture Washed Spanish Mackerel Mince Oxidation Model
System........................................................................................................................111
Preparation of Hemoglobin and Pressed Juice Oxidation Model System.....................112
Oxygen Radical Absorbance Capacity (ORAC) ...........................................................113
Total Antioxidant Potential ...........................................................................................114
Moisture Content and pH ..............................................................................................115
Statistical Analysis ........................................................................................................115
Results...................................................................................................................................116
One and Two Way adjustment ......................................................................................117
Two and Three Component Adjustment .......................................................................118
Comparison of one way and multi-component pH adjustment in Washed System ......118
Antioxidant mechanisms of non-substrate components WFM model system ..............119
ORAC Test .............................................................................................................119
TEAC Assay...........................................................................................................120
SDS Page .......................................................................................................................121
Discussion.............................................................................................................................121
No pressed juice treatment (control without pressed juice) ..........................................122
Lipid oxidation in the control samples ..........................................................................123
The Effect of pH Adjustment on Tissue Components...................................................125
The pH effect on hemoglobin.................................................................................125
Hemoglobin adjustment to low pH and readjustment to pH 6.8 ...................................127
7
Pressed Juice adjustment ...............................................................................................129
Lipid oxidation in the adjusted washed fish muscle......................................................131
Antioxidant Capacity as related to oxidation in washed muscle ...................................135
Conclusion ............................................................................................................................139
6
CONCLUSION.....................................................................................................................145
LIST OF REFERENCES.............................................................................................................148
8
LIST OF TABLES
Table
page
3-1
Initial average mince values for moisture content, pH, and lipid content. ........................67
4-1
Moisture content (%) and protein content (%) of isolates .................................................99
9
LIST OF FIGURES
Figure
page
2-1
Conventional surimi processing schematic........................................................................16
2-2
Flow schematic of acid and alkaline process.....................................................................18
3-1
Lipid oxidation in Spanish Mackerel mince ......................................................................67
3-2
Development of lipid oxidation in homogenates, freshly prepared...................................68
3-3
Development of lipid oxidation in homogenates after 1 day of storage ............................69
3-4
Development of lipid oxidation in homogenates after 2 days of storage ..........................70
3-5
The effect of ascorbic acid and tocopherol on pH 3 and pH 6.5 homogenates. ................71
3-6
The effect of EDTA and citric acid on acid homogenates.................................................72
3-7
The effect of pH readjustment in homogenates at pH 3 ....................................................73
3-8
The effect of pH readjustment in homogenates at pH 11. .................................................74
4-1
Schematic of the observed centrifugation layers ...............................................................99
4-2
Lipid oxidation of acid, alkaline, and control isolates after one hour..............................100
4-3
Lipid oxidation acid, alkaline, and control isolates after 8 hours. ...................................100
4-4
Lipid oxidation of acid, alkaline, and neutral isolates, 1 and 8 hours and 3 days ...........101
4-5
Levels of protein oxidation after 1 hour and 8 hrs pH, and 3 days..................................101
4-6
Storage Modulus (G’) of isolate pastes derived from 1 and 8 hrs, and 3 days ................102
4-7
Examples of rheograms from acid (pH 3) protein isolates ..............................................102
4-8
Examples of rheograms from alkaline (pH 11) protein isolates ......................................103
4-9
Examples of rheograms from control (pH 6.5) protein isolates.......................................103
4-10
Protein composition of isolates from adjusting for 1 hour ..............................................104
4-11
Protein composition of isolates from adjusting for 8 hour ..............................................104
4-12
Protein composition of isolates, 8 hours of holding followed by 3 days.........................105
4-13
Protein composition of initial homogenates ....................................................................105
10
4-14
Protein composition of homogenates held for one hour .................................................106
4-15
Protein composition of homogenates held for 8 hours ...................................................106
5-1
The role of acidified (pH 3) and pH6.8 tissue components on lipid oxidation................141
5-2
The role of two and three acidified (pH 3) and pH (6.8) tissue components...................142
5-3
Antioxidant capacity of Spanish mackerel hemoglobin and pressed juice......................143
5-4
Protein composition of untreated Spanish mackerel pressed juice..................................144
11
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
EFFECT OF LOW AND HIGH PH ON LIPID OXIDATION IN SPANISH MACKEREL
By
Holly Tasha Petty
May 2007
Chair: Hordur Kristinsson
Major Department: Food Science and Human Nutrition
Because of over fishing, traditional sources cannot satisfy the demand for fish proteins.
The production of protein isolate from underutilized fish sources using a new high and low pH
process is being investigated. Underutilized fish sources include pelagic species, by-catch, and
fish byproducts. The objective of this study was to investigate how the acid/alkali-aided process
influences lipid oxidation development in homogenates, isolates, and washed fish prepared from
Spanish mackerel muscle.
Studies of homogenates at acid and alkaline pH showed that lipid oxidation increased with
age of raw material, mince cold storage, pH exposure time, and acidification. Lipid oxidation in
the acid homogenates was effectively inhibited by ascorbic acid and tocopherol, but was
unaffected by the addition of chelators. The pH readjustment from acid to neutral pH did not
significantly affect lipid oxidation in Spanish mackerel homogenates; however, the addition of
antioxidants in combination with pH readjustment was problematic.
Isolates prepared from acid and alkaline treatment had comparable levels of lipid
oxidation, regardless of homogenate holding time. High levels of lipid oxidation were observed
in the supernatant fractions from acid, alkaline, and control treatments. Overall, alkaline isolates
contained greater levels of protein oxidation compared to acid and alkaline isolates. Protein
12
degradation occurred during acid and alkali processing of Spanish mackerel. Rheological testing
showed that there was greater gel strength in alkaline isolates, followed by the control and acid
isolates. Protein oxidation could explain differences in gel quality between acid and alkaline
isolates.
The adjustment from pH 3 to pH 6.8 greatly affected all components in a hemoglobinmediated lipid oxidation washed Spanish mackerel model system. The addition of pH treated
hemoglobin to the washed Spanish mackerel system resulted in lipid oxidation levels comparable
to the control. Increases in lipid oxidation compared to the control were observed when pressed
juice was pH treated. The greatest levels of lipid oxidation in the Spanish mackerel model
system occurred when washed fish muscle was pH treated. These results show that lipid
oxidation incurred during acidic processing could be primarily due to phospholipid changes,
secondary to changes in hemoglobin and/or aqueous extracts.
.
13
CHAPTER 1
INTRODUCTION
Because of the large demand for fish protein and inability to satisfy the need by traditional
means, the utilization of fatty pelagic species and by-catch is under current study. By-catch from
shrimp vessels, pelagic species, and by-products are underutilized as human food (1). By-catch
includes low value fish that are caught and discarded back into the sea or that are further used for
fish meal (2). It is estimated that four pounds of by-catch result for every pound of shrimp
caught. In addition, dark muscle fish species compose 40-50% of total catch worldwide (3, 4)
Pelagic species present numerous obstacles if processed conventionally, due to small size, shape,
bone structure, seasonal availability, and most importantly the prevalence of unstable lipids and
pro-oxidants (4-6).
Extracting functional and high quality fish proteins from fish sources by solubilizing the
proteins at low and high pH values and recovering solubilized proteins with isoelectric
precipitation has been conducted (7). From these findings, isolation of functional proteins from
underutilized, low value fish sources was developed (7, 8). These newly developed processes
produce a protein isolate low in lipids. This is of major importance, due to the abundance of
triacylglycerols (neutral storage lipids) and membrane phospholipids in mitochondria of the dark
fish muscle (5). Phospholipids are the primary substrates in lipid oxidation due to their highly
unsaturated nature (9, 10).
Protein isolates made from white muscle using the alkali process have been found to
possess greater oxidative and color stability over unprocessed or acid treated muscle (11).
Studies have shown that high pH inactivates hemoglobin, thereby preventing oxidation of lipids
and formation of an undesired yellow-brown color as well as off odors and flavors. In contrast,
acid treatment denatures hemoglobin, enhancing its pro-oxidative activity (12). Pelagic fish
14
species and fish byproducts are rich in blood and heme proteins. The blood and heme proteins
are used for oxygen transport for production of long-term energy in the dark muscle via
oxidative metabolism. In white muscle, heme proteins are used for more short-term energy (5).
A number of other changes in homogenized fish muscle can occur at low and high pH
values that can contribute to the acceleration or delay of lipid oxidation. Currently there is a
limited understanding of how low and high pH influences oxidative processes in fish muscle.
Further investigation is therefore needed regarding the stability of fish muscle as affected by
extreme pH. Due to the abundance of heme proteins in pelagic fish species, it is imperative to
study their oxidative stability and to control their pro-oxidative effects. These proposed studies
could further the implementation of pelagic fish protein isolates as a food source and functional
food ingredient, and an understanding of oxidation mechanisms in general.
15
CHAPTER 2
LITERATURE REVEIW
Obtaining a Fish Protein Isolate
Conventional Surimi Processing
The production of surimi has proved to be somewhat useful in utilizing fish proteins.
Briefly, fish muscle is ground and washed at approximately 1 part fish to 2-5 parts water. The
washing step can be repeated several times depending upon processing conditions (i.e. type of
fish and plant location). During the washing step, 30% of the total proteins in the form of
sarcoplasmic proteins are lost. The fish muscle is then dewatered using a screw press. Normally,
cryoprotectants are added to prevent protein denaturation during freezing and thawing. This
process (Figure 2-1). Unfortunately, high quality surimi from conventional processing has only
resulted from white flesh fish. Attempts to create surimi from dark pelagic species has resulted in
products having poor color and lipid stability due to pigments, pro-oxidants, and an abundance of
lipids (4, 6).
Ground Fish Muscle
Wash with 2- 5 parts water
(3 times with 0.3% NaCl in last step)
Refine and Dewater
Stabilize
(Add cryoprotectants and freeze)
Surimi
Figure 2-1. Conventional surimi processing schematic. The final surimi is heated to form a gel,
resulting in various seafood analog products (13).
16
Acid/Alkali-Aided Processing
Solubilization and recovery of protein by pH rather than salt adjustment, led to an
economical protein isolate from low value fish sources (7). At high and low pH values, the
proteins become solubilized due to large net charges. It was also discovered that at extreme pH,
the cellular membranes surrounding the myofibrillar proteins are disrupted (13, 14). As a
consequence of these changes in protein solubility and membrane integrity, the viscosity of a
muscle protein homogenate at very low and high pH values was found to decrease significantly.
This viscosity decrease enables soluble proteins to be separated from undesirable materials in the
fish muscle via high speed centrifugation (e.g.bones, scales, connective tissue, microorganisms,
membrane lipids, neutral lipids etc.). The soluble proteins can then be recovered and
precipitated by adjusting the pH into the range of their isoelectric point (~pH 5.3 to 5.6) (Figure
2-2). This process uses a very low (2.5 to 3.0) or high pH (10.8 to 11.5), thus enabling for a
simple and economical recovery of low value fish species (7, 8, 15). Studies have demonstrated
that of the two processes, the alkali-aided process is substantially better than the acid-aided
process since oxidation and color problems are reduced and protein functionality is superior (13,
16, 17).
Advantages of the Acid/Alkali-Aided process
Using the acid/alkali-aided process versus conventional processing is beneficial for various
reasons. Protein recoveries can range from 90-95% using the acid/alkali-aided process versus
conventional surimi recoveries ranging from 55-65% from fish fillets. Conventional processing
only retains the myofibrillar proteins, in contrast, this new process allows for the recovery of
both myofibrillar and sarcoplasmic proteins. Processing is more economical using the
acid/alkali-aided process, reducing labor, expenses, production complexity, and handling
17
(homogenate compatibility in processing). Additionally, minimally processed and lower quality,
less utilized
Ground Fish Material
Homogenization
1 parts fish : 5-9 parts water
pH reduction or increase
pH 2.5 (HCl) / pH 11 (NaOH)
Centrifuge (10,000 xg)
Sediment Layer
Membrane lipids
Solution Phase
Soluble muscle
Upper Layer
Protein aggregation
pH adjusted to 5.5
Centrifuge (10,000 x g)
Supernatant
Mostly water, can be
(a)
Sediment = Protein isolate
Figure 2-2. Acid and alkaline process used to produce protein isolate from low- and high-value
fish protein sources.
sources of fish protein can be implemented. Before processing, fish sources can be of lower
quality and can be previously frozen. Lipids can be less of problem due to the separation of
triacylglycerols and phospsholipids from the protein homogenate, enabling the production of a
more stable product with better color and flavor; however, there is not absolute separation of
heme proteins from the final protein isolate. Due to the abundance of mitochondria in dark
muscle, the acid/alkali process is advantageous in the separation of phospholipids from dark
muscle (5). Functional properties including gelation, water holding capacity, and solubility are
all improved during the acid/alkali-aided process due to a desirable partial denaturation of the
18
fish proteins. Overall, the protein isolate produced is safer due to elimination or reduction of
toxins (PCBs and mercury) and microbial loads, and no pollution due to processing (6, 13, 16,
18, 19).
Lipids in Fish Muscle
Lipid Composition
For dark muscle fish species, fat composition can vary in response to environment,
seasonal variation, migratory behavior, sexual maturation, and feed (20). Fish living in colder
water will have more dark muscle. Dark muscle tends to vary in composition, especially with
season and location; therefore, when sampling dark muscle or fatty fish species, it is important to
consider the possible variations in lipid content (20).
The two types of lipids contained within teleost (bony) fish muscle are phospholipids and
triacyglycerides. Phospholipids or structural lipids are the major components of cell membranes
within the fish tissue (myofibrillar protein); while triaglycerides are found in adipocytes and
serve as storage energy. Phospholipids are found in various membranes including those of the
cell and organelles within the cell.
In general, muscle from lean fish is composed of approximately 1% lipid, primarily
phospholipid. The majority of lipids in lean fish are stored in the liver. Some lean species
include the bottom-dwellers: cod, saithe, and hake. Major lipid deposits in fatty fish species are
found in tissue other than the liver: subcutaneous tissue, in the belly flap, and throughout the
muscle tissue. Fatty species include the pelagic species such as herring, mackerel, and sprat.
Most importantly, lipids are widely distributed within muscle tissue, primarily next to the
connective tissue (myocommata) and between light and dark muscle (21). In pelagic species, the
dark muscle uses triacylglycerides for continuous aerobic energy in migrations, whereas white
19
muscle uses glycogen as the main source of energy and is quickly fatigable due to accumulation
of lactic acid during anaerobic respiration (22).
Lipid Oxidation in Fish
Lipid oxidation serves as a mediator of important processes in living biological systems
(23). However, post-mortem lipid oxidation can only be controlled to a certain extent, before
antioxidants and their systems are exhausted (24). Lipid oxidation is one of the primary causes
of deterioration in cold stored fish muscle (25) and negatively affects color (26), odor and flavor
(27), protein functionality and conformation (28), and overall nutritional content of fish muscle
(29, 30). This quality deterioration is due to the high content of polyunsaturated fatty acids
contained within fish muscle (9, 10, 25) along with highly active pro-oxidants (5).
Major reactants in lipid oxidation are oxygen and unsaturated fatty acids. Exposure of
lipids to atmospheric oxygen results in unstable intermediates (31). Ultimately, these
intermediates then degrade into unfavorable compounds attributing to off flavor and aroma.
Lipid oxidation has many plausible mechanism of initiation, including non-enzymatic and
enzymatic reactions. Non-enzymatic mechanisms include autoxidation (free-radical mechanism)
and photogenic oxidation (singlet oxygen mediated) (31). Enzymatic mechanisms include
actions by lipoxygenase and cylooxygenase. Lipid oxidation most commonly occurs by a free
radical mechanism, involving the formation of a reactive peroxyl radical (31). In addition, it has
been reported that the detrimental effects due to heme (autoxidation mechanisms) are greater
than effects due to lipoxygenase, mainly due to longer-lasting pro-oxidative activity (32).
Autocatalytic Mechanism
The autocatalytic mechanism is deemed the primary cause of lipid oxidation in post
mortem fish (31) and is historically referred to as lipid peroxidation (33). The free radical
mechanism has three classic stages: initiation, propagation, and termination. Initiation begins
20
with hydrogen abstraction from an unsaturated fatty acid, particularly those having a pentadiene
structure (31). The lipid radical (L●) then reacts with molecular oxygen and a peroxyl radical
results (LOO●). Propogation is the phase in which the peroxyl radical (LOO●) abstracts
hydrogen from neighboring lipid molecules, thereby forming more lipid radicals and lipid
hydroperoxide (LOOH) molecules. Lipid hydroperoxides cannot be detected through sensory
analysis, which is the reason why peroxides do not correlate with “off flavor”. The degradation
of these lipid hydroperoxides, due to metal catalysts, results in the unfavorable secondary
intermediates of shorter chain length: ketones, aldehydes, alcohols, and small alkanes. The
secondary products give rise to off aromas, flavors, and yellow color in fish. Aldehydes are
normally associated with off odor. Termination occurs when a radical reaction is quenched by a
reaction with another radical or antioxidant. It is important to note that these reactions are
occurring simultaneously and that the mechanism of lipid oxidations can become quite complex.
(22, 31).
Transition Metals: Catalysts of Lipid Oxidation
Decomposition of lipid hydroperoxides at refrigerator and freezing temperatures is due to
electron transfer from metal ions. Transition metals serve as primary oxidation catalysts in lipid
oxidation, mainly iron. Iron content varies among muscle food sources (31). The main catalysts
in biological systems include heme and non-heme iron (5). Non-heme sources include released
iron from heme proteins, protein complexes (transferritin, ferritin, and metalloprotiens), and
other low molecular weight complexes (diphosphonucleotides, triphosphonucleotide, or citrate
chelates) (34).
Heme vs. Non Heme Iron and Lipid oxidation
White muscle contains less iron than dark muscle, due to less myoglobin and hemoglobin
(5). Through research, it has been shown that heme proteins serve as potent pro-oxidants (7, 35).
21
Further more, it is believed that heme proteins serves as the main pro-oxidant in post mortem
fish muscle (36, 37). It is believed that hemoglobin is the main initiator of oxidation in vitro
systems (12, 38); whereas, hydrogen peroxide mediated metmyoglobin has been shown to
initiate oxidation in vitro-model systems (39, 40). In addition, Kanner et al (41) found that lowmolecular weight iron also initiates lipid oxidation. Nonetheless, studies have shown that
hemoglobin is the leading pro-oxidant (16, 35, 42, 43). The basic mechanism of lipid oxidation is
known; however, the predominant initiators(s) of oxidation are still debated.
Heme Proteins and Autoxidation
If most of the hemoglobin is not removed through bleeding, the fish muscle will become
more susceptible to oxidation. Mincing or mechanical mixing of fish muscle can promote
oxidation through the incorporation of hemoglobin, hemoglobin subunits, free heme, and iron.
(38). The presence of hemoglobin and/or myoglobin during the acid and alkali-aided process and
in the end product (protein isolate) could lead to serious lipid oxidation (42, 44, 45).
Whether oxidized or reduced, heme proteins can serve as potent pro-oxidants (46). Post
mortem, heme proteins are in their ferrous state (Fe+2). In this state, oxygen can be bound or unbound to the heme thus giving oxyhemoglobin/myoglobin and deoxyhemoglobin/myoglobin,
respectively. The heme iron can become oxidized to the ferric state (Fe+3) and result in
metmyoglobin/hemoglobin, with water binding to the 6th coordination site (47). This oxidation
will also give rise to the superoxide anion. Subsequent superoxide dismutation can result in the
highly reactive hydroperoxyl radical, hydroxyl radical, and hydrogen peroxide (6, 24)
Autoxidation of hemoglobin or myoglobin by superoxide can lead to the formation of the
hypervalent ferryl-hemoglobin/myoglobin radical. The formation of this porphyrin cation radical
is deemed the primary decomposer of lipid hydroperoxides, resulting in the secondary
compounds, associated with off-flavors and odors (37, 48).
22
Hemoglobin vs Myoglobin and Lipid Oxidation
Myoglobin is located in the muscle cell itself while hemoglobin is found in the blood cells
of the muscle tissue vascular system (49). Hemoglobins from various fish species are
differentiated from how they react to changes in temperature, pH, etc (50). It has been found that
with decreases below pH 7.6 (43) and as low as pH 1.5 (51), either through physiological
changes post mortem and/or experimental manipulation, a shift from oxy to deoxy-hemoglobin
occurs, called the Bohr shift. The change to the deoxy state is believed to increase hemoglobin’s
pro-oxidative activity due to heme iron exposure (52). Physiologically, this phenomenon may be
advantageous in the release of oxygen in the production of pyruvate and consequently lactate, in
pelagic species due low oxygen affinity (53).
Recently, the comparison of myoglobin and hemoglobin from trout as catalyst for lipid
oxidation was presented by Richards and Dettmann(54) The researches found that hemoglobin
has a greater tendency to induce lipid oxidation due to the mechanism of heme dissociation
hversus myoglobin at neutral pH (55). The oxidative nature of myoglobin versus hemoglobin at
low pH is being investigated.
Lipid Content and Lipid Oxidation
According to Richards and Hultin (56), only low levels of phospholipid (0.01%) are
required for hemoglobin mediated lipid oxidation and for off flavors to develop. According to
Undeland et al. (57), it is the presence and amount of hemoglobin that affects the extent of
oxidation and not the amount of lipid. In a study conducted by Richards and Hultin (58), even a
six-fold increase in membrane phospholipids did not affect the extent of lipid oxidation. Despite
bleeding, hemoglobin still remains within the fish muscle and can still pose problems (59).
23
Effects of Extreme pH on Lipid Oxidation in Fish Muscle
At extreme acidic pH, heme proteins are denatured, increasing the propensity of lipid-prooxidant interactions. Lipid membranes become more susceptible to oxidation when in the
presence of heme proteins adjusted to low pH or readjusted from low pH (3.0) to neutral pH (7)
(18). This is due to unfolding of the heme protein and exposure of the heme iron upon
denaturation, or lack of refolding with pH readjustment (51). In addition, transition metals and
free heme become active at low pH (56). Autoxidation and ferryl-Hb/Mb production are
enhanced at low pH (56, 60). Furthermore, the heme crevice of hemoglobin is unfolded at low
pH, increasing reactions with lipid substrates, and thus oxidation (12). With acidic pH, there is
also a greater likelihood of subunit dissociation of the hemoglobin and thus loss of the heme (51,
61). In contrast, the pro-oxidative activity and autoxidation of hemoglobin is reduced at alkaline
pH due to increased stabilization of the heme group (51, 60).
Studies of the effect of extreme pH on hemoglobin and oxidation have been conducted in
model systems by Kristinsson and Hultin (51). Trout hemoglobin was added to a washed cod
model system. This model muscle system contain virtually all membrane phospholipids and
proteins, but is free of soluble pro- and antioxidants through the washing step. The ability of
hemoglobin to oxidize cod at pH values 2.5, 3.5, 10.5 and 11 was studied. It was found that there
was considerably more oxidation at low pH values compared to higher pH values.
Undeland et al. (62) studied the minimization of oxidation of herring muscle homogenate
at low pH (pH 2.7) by reducing exposure times, using high speed centrifugation, and
implementing antioxidants or chelators. Storage stability of the protein isolates was only
improved by antioxidant and chelator implementation, erythrobate and
(ethylenediaminetetraacetic acid) EDTA, respectively (62).
24
Various techniques to reduce lipid oxidation exist and include heme separation, lipid
membrane centrifugation, pH exposure time, and antioxidant implementation. However, the
quantification of lipid oxidation in the homogenate made from fish of varying quality in certain
conditions requires attention such as extended holding times (12 hours) and an expanded pH
range (2.5 to 11.5).
Lipid Oxidation and Antioxidant Implementation
The greatest defense against lipid oxidation is the use of antioxidants. Inhibition of lipid
oxidation can occur directly or indirectly. Inhibitors can prevent initiation and propagation steps
in the free radical mechanism (31). Antioxidants occur exo- and endogenously. Antioxidants
donate hydrogen to the peroxyl radical and eliminate it as a reactive substance in the radical
mechanism (31).
Past research evaluated the effects of antioxidants on lipid oxidation in mackerel fillets
(63). Exogenous antioxidants included: ascorbic acid, tocopherol, and butylated-hydroxytoluene
(BHT). The antioxidants were applied separately and in combination. At frozen storage
temperatures (-20°C and –30°C), the BHT and BHT combined with ascorbic acid proved to be
the most inhibitive. This implies that there is a synergistic effect between lipid and water-soluble
antioxidants in delaying lipid oxidation in whole fish muscle (63).
Concentration effects of antioxidants are well known in vivo. At low ascorbic acid
concentrations, there is a reduction of transition metals (iron), promoting hydroxyl radical
formation, and at high concentration, ascorbic acid scavenges hydroxyl and superoxide radicals
in vivo (49). According to Witting (64) and Chow (65), tocopherol serves as a pro-oxidant at
high concentration and an antioxidant at low concentrations.
At low pH versus high pH, ascorbic acid and tocopherol can inhibit lipid oxidation.
Together, ascorbic acid, in high concentrations, and tocopherol effectively prevent oxidation in
25
vitro. Tocopherol by itself is especially effective in reducing lipid oxidation in vivo (66). In
herring fish homogenate (pH 3), there was a significant reduction in oxidation as measured by
malondialdehyde when implementing ascorbic acid and tocopherol (62). In addition, when
incorporating antioxidants erythrobate (iso-ascorbic acid) and STPP (sodium tripolyphosphate),
there was a significant reduction in lipid oxidation in the final protein isolate (67). According to
previous studies, it is essential that antioxidants be added early in the acid aided process to obtain
a good lipid stable final product (67-69).
According to Undeland et al. (67), low molecular weight aqueous extracts, also know as
“press juice,” from fish muscle are effective in inhibiting lipid oxidation in cod lipid membranes
caused by hemoglobin. Possible inhibitors in the “press juice” could include nucleotides and
other reducing agents. The role of these compounds at low or high pH is not well understood.
Lipid oxidation and Proteins
The functionality of fish protein isolate preparations relate mostly to the behavior and
interaction between myofibrillar proteins (most notably actin and myosin). Sarcoplasmic and
myofibrillar proteins contribute to varying portions of the overall protein isolate, depending on
whether the alkaline or acidic process is used. Myofibrillar proteins are the primary components
of protein isolates produced from the alkaline process, and sarcoplasmic proteins are the primary
proteins in isolates prepared from acid processing (14).
The effect of lipid oxidation and reactive oxygen species (ROS), on protein stability and
function in muscle foods is poorly understood. Howell and Saeed (70) reported texture changes
in frozen Atlantic mackerel (Scomber scombrus) stored at 20 and 30ºC for up to 2 years, which
could be related to an increase in lipid oxidation. Increases in G’ (elastic) and G’’ (viscous)
moduli were observed. With the implementation of antioxidants in the muscle, a correlation
between the kinetics of lipid oxidation and rheological behavior was observed. In another study
26
by Saeed and Howell (63), it was reported that lipid oxidation with frozen storage caused a
reduction in ATPase activity, solubility, and myosin heavy chains. The incorporation of
antioxidants in Atlantic mackerel prevented/delayed the deleterious effects caused by lipid
oxidation. Hitherto, studies have not addressed the effects of acid and alkali processing on
protein oxidation development.
Protein Oxidation
The relationships between muscle food quality and protein oxidation are of great interest
and have yet to be fully understood. Consequence of oxidative stress, proteins undergo covalent
modification and incur structural changes. Reported protein modifications include and are not
limited to back bone disruption, cross-linking, unfolding, side-chain oxidation, alteration in
hydrophilic/hydrophobic character and conformation, altered response to proteases, and
introduction of newly formed reactive groups, or amino acid conversion (71-74). Although
protein functionality and quality changes are commonly observed with increases in lipid
oxidation, the actual mechanisms by which proteins are oxidized in muscle foods are not well
understood..
Extensive research shows a positive correlation between protein oxidation and diverse
physiologically deteriorative events, such as aging and disease. Biological events are highly
dependant upon radicals and oxidative processes (75). Oxidative stress arises when antioxidant
mechanisms are unable to counteract oxidant loads (76). Oxidative stress is secondary to partial
and/or complete exhaustion or degradation of antioxidant systems (24). In the event of oxidative
stress, generated radicals and non-radical species can oxidize proteins directly or indirectly. Due
to oxidatively reliant metabolism in vivo, direct protein oxidative attack occurs most frequently
in the presence of reactive oxygen species (ROS): hydrogen peroxide, superoxide, nitric oxide,
peroxynitrite, hydroxyl radical, peroxyl, and/or alkoxyl radicals, etc. (77). Similarly, these ROS
27
are produced in muscle foods during mechanical processing and storage (78-80). Indirect protein
oxidation can occur in the presence of secondary lipid and protein oxidation products (81-83).
Oxidizing agents common to living and post mortem muscle could include and are not
limited to transition metals, protein-bound transition metals (Heme), U.V., X, λ irradiation,
ozone, lipid and protein oxidation products, and oxidative enzymes (84, 85). Neutrophils,
macrophages, myloperoxidase (hypochlorous acid), and oxireductase enzymes are just a few
possible mediators specific to in vivo protein oxidation (84, 85). Rational for oxidative
mechanisms post mortem are extrapolated from those which occur physiologically. The
introduction of oxygen during muscle food processing is analogous to reperfusion in vivo (86).
Pro-oxidative environments present themselves in the aging and restructuring of meat
products. With storage, endogenous antioxidant levels decrease in whole muscle; whereas, an
increased exposure to oxygen and endogenous oxidants as well as decreasing endogenous
antioxidants occur in processed muscle. Thus, radical species and derivatives induce greater
levels of oxidative stress with decreases in antioxidant capacity (87, 88).
Mechanisms of Protein Oxidation
A multitude of mechanisms give rise to protein oxidation. Oxidatively susceptible protein
sites in the presence of irradiation and/or metal generated ROS have been studied (89-92).
Structural protein units most prone to attack include the α-carbon and side chains of amino acid
residues as well as the peptide backbone. Amino acid residues most susceptible to oxidation in
vivo and in vitro are cysteine and methionine due to the presence of a readily oxidizable sulfur
atom and vulnerability to all oxidizing sources (76). Oxidation of cysteine can lead to disulfides
(S-S), glutathiolation, and thyl radicals (93, 94), while the oxidation of methionine
predominately leads to methionine sulfoxide (95). Oxidative modifications do not always result
28
in deleterious effects. In fact, negligible effects or antioxidative capabilities may arise with
protein oxidation (96-98).
Reactive oxygen species (ROS) can directly oxidize proteins by abstracting a hydrogen
atom from the α–carbon atom of an amino acid, producing a carbon centered radical. Further
reaction of these radicals with oxygen gives a peroxyl radical. Peroxyl radicals can then abstract
hydrogens or react with hydroperoxyl radicals (protonated superoxide), forming the alkoxyl
radical. A hydroxyl derivative can be produced from the reaction of the hydroperoxyl radical and
the alkoxyl radical (89-92).
In conjunction with protein radical formation, oxidation of proteins, peptides, and amino
acids can give rise to various protein derivatives. Protein oxidative derivatives common to both
physiological and post mortem tissue result from scission, polymerization, carbonylation, and
protein-lipid interaction. Specific oxidation modifications more common in vivo include
hydroxylation, nitrosylation, nitration, sulfoxidation, and/or chlorination of protein residues (99).
Protein Scission
Scission of peptide backbones and/or amino residues can occur with protein oxidation.
Alkoxyl radicals and alkoxyl peroxides can be cleaved through two major mechanisms: the α –
amidation pathway and the α –diamide pathway. Carbon radicals or derivatives are cleaved to
form amide and α-keto-acyl derivatives in the α –amidation pathway; while diamide and
isocyante derivatives form in the diamide cleavage pathway (89). According to Garrison (89) and
Uchida et al. (100) hydroxyl radical mediated cleavage/fragmentation of individual residues such
as glutamate and aspartate (89) and proline (100) can occur.
Protein Polymerization
Along with peptide cleavage, protein and/or amino radicals and derivatives can undergo
polymerization, forming cross-linked protein residues (82-85). Protein cross-linkage can occur
29
through covalent and non-covalent interactions. Non-covalent protein cross-linking is due to
hydrophilic and hydrophobic bonding. Oxidation of proteins can result in greater exposure of
hydrophobic groups, and thus, greater hydrophobic interactions among residues. Increased
interaction of hydrophobic residues can lead to aggregation. Likewise, hydrogen bonding can
facilitate protein and protein-lipid aggregation (101). Covalent intra- and inter-protein
polymerization are due to radical mediated mechanisms. ROS induced protein cross-linking in
vivo and in muscle foods can occur under several primary conditions: (1) homolysis of two
carbon, protein, or tyrosine radicals (89), (2) thio-disulfide exchange of oxidized cysteine
sulfhydryl groups (89, 102), (3) interaction of oxidatively derived protein carbonyl groups with
lysine/arginine residues (87, 103), (4) cross-linking of dialdehydes (malondialdehyde or
dehydroascorbate) with two lysine residues (104, 105)
Protein Carbonylation
Metal catalyzed oxidation (MCO) is a primary mechanism by which proteins are directly
oxidized in vivo to form carbonyl derivatives. Likewise, considerable protein oxidation is due to
MCO in muscle foods (106). In the vicinity of proteins and or amino acid residues, protein bound
transition metals illicit the production of ROS and oxidation of amino acid side-chains.
Hydrogen peroxide is reduced to ROS species (hydroxyl radical and ferryl radical) in the
presence of transition metals (107-110). MCO can lead to the formation of 2-oxo-histidine from
histidine and carbonyl derivatives from lysine, arginine, proline, and threonine (107, 111).
Carbonyl derivatives are frequently investigated in physiological and post mortem systems,
due to convenience and elucidation of overall protein oxidation. Carbonyl groups (aldehydes,
ketones) are indicative of protein oxidization induced by various sources of ROS and oxidative
stress. Carbonyls can be detected in various ways and the assay involves the derivitization of
carbonyl groups with 2,4-dinitropheylhyrazine (DNP), producing a 2,4-dinitrophynl (DNP)
30
hydrazone (112). Following covalent derivatization, various quantitative techniques can be
utilized including and not limited to 2-D page, spectrophotometric methods, and high
performance liquid chromatography (HPLC) (112). Additionally, carbonyl measurement is
applicable to mixed samples: homogenate tissue, plasma, and proteins. Other methods require
purification, but allow for differentiation of the oxidative source (Shacter, 2000).
Lipid Mediated Protein Oxidation
Lipid mediated protein oxidation is believed to occur by various mechanisms (113). One
proposed mechanism describes a protein or amino residue radical and lipid oxidation product
(malondialdhyde) condensation reaction, forming a lipid-protein complex. Proteins could react
with lipid oxidation products (alkyl radicals, hyderoperoxides, and secondary oxidation products)
giving rise to a protein-lipid radicals (63). Another mechanism for lipid-protein oxidation is
described: (1) free radicals react with proteins side chains producing protein free radicals, (2)
protein radicals could then react with oxygen and produce a peroxy radical, and (3) hydrogen
peroxide radicals formation could thus follow and induce formation of carbonyl products. The
extent of protein oxidation could theoretically develop faster than lipid oxidation, due to the
cytosolic associated location of proteins in muscle foods and possible exposure to free radicals
(Srinivasan and Hultin, 1995). Specific amino acids are highly susceptible to lipid induced
oxidation due to the presence of reactive side chains: sulfhydryl, thioether, amino group,
imidazole ring, and indole ring. Corresponding amino acids include cysteine, methionine, lysine,
arginine, histidine, and trypotphan (101, 114).
Protein Oxidation Accumulation
Physiologically, oxidatively modified proteins can accrue over time. Reduced activities
and/or levels of specific proteases, low rates of cell turnover, depletion of antioxidant
mechanisms, or inherently undetected oxidative modifications can lead to protein oxidation
31
accumulation in vivo (115). In instances where an amino acid conversion occurs, the cell is
unable to discriminate between oxidized and un-oxidized residues. MCO can lead to
modifications that go unnoticed including the conversion of proline to hydroxyproline, or the
conversion of glutamate to asparagine or aspartate (76). The cell predominantly relies upon the
protease known as the proteosome, to degrade oxidatively modified proteins (116). Methionine
sulfoxide reductase is one of the few repairing proteases, enabling for the conversion of
methionine sulfoxide to methionine (MET) in vivo (117). A decrease in proteosome activity
occurs with aging and is believed to be predominantly responsible for the accumulation of
oxidized proteins (106, 118, 119). Accumulation of protein oxidation products increases the
likelihood of cross-linking. Furthermore, polymerized proteins can not be degraded by the
proteosome and can prevent degradation of other oxidized proteins (73). Unlike living systems,
reparation measures for protein oxidation in post mortem tissue do not exist. Thus an
understanding of protein oxidation mechanisms in muscle foods and how functionality is
affected by protein oxidation is well warranted.
Protein Oxidation and Muscle Food Quality
Protein oxidation can induce physiochemical changes in proteins. Oxidative modifications
can lead to decreases and or loss in protein activity and stability, negatively impacting cellular
processes in vivo and functionality in vitro (120). In post mortem muscle foods, the intrinsic
property primarily affected by oxidative modification is conformation and/ or structure. The
functional property in muscle foods most affected by this intrinsic change is hydration.
Alteration in hydration capacity can ultimately affect other functionally properties: gelation,
emulsification, viscosity, solubility, digestibility, and water holding capacity (121-125). Due to
the significant contributions to muscle functionality, myosin is extensively studied in protein
oxidation studies (86). Numerous muscle food studies conclude that changes in protein structure
32
and functionality, under oxidizing conditions, are due to protein oxidation. Understanding the
cause of structural or functional protein changes as they relate to protein oxidation is important
to muscle food quality.
Protein oxidation and conformational changes in proteins
Changes in conformational stability in muscle food proteins is attributed to changes in free
energy. As a result of oxidative modification, intra- and intermolecular interactions are altered
within a protein’s structure; thereby, decreasing the thermal energy and conformational stability.
Liu and Xiong (126) studied thermal stability changes in the Fe (II)/ascorbate/H202 induced
oxidation of myosin, derived from chicken pectoralis muscle. Using differential scanning
calorimetry (DSC), the protein oxidation by hydroxyl radicals and effects on myosin
conformational stability were measured. Significant reduction in myosin thermal stability was
observed as indicated by decreases in enthalpy (∆ H). Additionally, decreases in transition
temperature (Tm) were observed (126). The un-oxidized myosin transition temperature (55°C)
designates the myosin conformational change to a randomly coiled, ‘broken net,’ structure upon
heating. Thus, decreases in Tm are indicative of decreased conformation stability (127). Relative
to the native myosin, a new transition (60-80°C) temperature was observed among oxidized
samples upon heating. This newly formed heating transition was attributed to cross-linking
among oxidized cysteine residues as the Tm was not observed among the oxidized samples
containing N-ethyl-maleimide, a thiol blocking reagent(126). According to Srinivasan and Xiong
(128), decreases in conformation stability were observed in bovine cardiac muscle myofibrillar
proteins when subjected to hydroxyl radical attack (128). Similarly, Saeed and Howell (70)
observed decreases in enthalpy (∆H) and Tm in Atlantic mackerel fillets stored at -20°C versus
those stored at -30°C for 2 years. These changes were indicative of aggregation and denaturation
within the mackerel fillets (70). In another study conducted by Saeed and Howell (63),
33
denaturation of myofibrillar proteins and lipid oxidation of docosahexaenoic and
eicosapenaenoic acid were observed in frozen storage of Atlantic Mackerel and were
significantly inhibited by incorporations of antioxidants: namely BHT and ascorbic acid in a
minced system (63).
Increases in hydrophobic character and ultra violet absorption (U.V.) serve as markers of
protein destabilization and are presumed indicators of protein oxidation. With structural change
and denaturation, hydrophobic residues located within the interior of a protein become exposed.
Li and King (129) observed an increase in chicken myosin hydrophobicity in an iron (II)
/ascorbate mediated oxidation system (129). Wang et al.(130) observed an increase in
hydrophobicity in cold stored (-15 or -29°C) beef heart surimi along with increases in lipid
oxidation and loss of sulfhydryl groups. Wang et al.(130) elucidated that increases in lipid
oxidation may cause or be mechanistically related to protein oxidation development. Past studies
have concluded that lipid oxidation induces protein oxidation; however, the mechanisms of
protein oxidation in the presence of lipids remain inconclusive.
Functional properties and protein oxidation
Increased hydrophobicity and cross-linking as a result of protein oxidation give rise to
decreased solubility. According to Buttkus (131) and Dillard and Tappel (132), significant
solubility decreases were attributed to malonaldehyde protein adduct products in muscle foods.
Gel electrophoretic patterns revealed protein aggregation with free-radical induced myosin
oxidation. These patterns were analogous to electrophoretic patterns for processed meat (88, 121,
133). Jarenback and Liljemark (134) observed extensive decreases (90%) in cod myosin
solubility in a linoleic hydroperoxide system. Significant decreases in solubility have been
observed in various other transition/metal mediated myosin (Turkey, beef) oxidation systems
(123, 133). According to Decker et al (133), transition metal/ascorbate induced protein oxidation
34
of turkey myosin resulted in decreased gelling ability, solubility, and hydration. Decreases in
functionality were attributed to increased levels of carbonyls and most likely cross-linking of
oxidized proteins (133).
Both a loss (135) and increase in gelling ability (121, 136) have been observed in the
oxidation of myosin. In a storage study conducted by Saeed and Howell (70), viscoelastic and
elastic modulus were measured in myosin prepared from matched Atlantic Mackerel fillets
stored at -20 ºC and -30ºC, for up to two years. The storage/elastic modulus (G’) and viscous
modulus (G’’) were higher in the -20°C versus the -30°C derived myosin paste and were higher
with increased storage. This is most likely due to covalent and/or other interactions in
actomyosin (70). In this study, protein denaturation and aggregation correlated well with the loss
of ATPase activity and protein solubility shown in previous studies conducted by Saeed and
Howell (63). The elastic modulus was significantly decreased with the incorporation of
tocopherol (500 ppm), this was attributed to effective lipid oxidation prevention. Transfer of free
radicals or protein-lipid interactions could proceed from lipid oxidation (70). It was proposed
that protein G’ and G’’ were reduced due to less lipid oxidation, induced aggregation. The
samples with greater lipid oxidation had greater levels of protein aggregation.. It was inferred
that the use of tocopherol would quench the radicals and reduce the formation of lipid and
protein oxidation product and thus reduce the gelling ability. Cross-linking can occur through
protein free radicals in muscle systems possibly causing aggregation and may improve gel
formation and strength (63).
Liu and Xiong (137) found that in an iron/ascorbate/H202 mediated myosin oxidation
system, gel elasticity was significantly reduced (40%) compared to the non-oxidized myosin.
This decrease in elasticity was attributed to protein oxidation. Myosin cross-linking and scission
35
were exibited by the presence of protein fragments and aggregates (137). In contrast to this
study, Srinivasan and Hultin (121) found that the iron/ascorbate induced oxidation of washed and
minced cod muscle resulted in increased gelling ability and carbonyl content. Hitherto, few
studies have correlated measures of protein oxidation with protein functionality and structure.
Significance
The development of lipid oxidation at low and high pH in fish homogenates and in protein
isolates obtained from the alkali-acid aid process has only been investigated to a limited extent.
This research could be very helpful in improving the stability of the homogenates and fish
protein isolates in processing and utilization of pelagic species, which otherwise are highly prone
to oxidation. In addition, the impact of lipid oxidation on proteins during the acid and alkaliaided process has not been fully investigated. Contradictions exist as to what are the key
catalysts of lipid oxidation and very limited knowledge is available on the role of different key
catalysts or mediators of oxidation at low and high pH in muscle homogenates. This work will
further strengthen the argument that heme proteins are the prime mediators of lipid oxidation in
fish muscle homogenates and protein products thereof. Basic work in this area is expected to
greatly further our knowledge on oxidation processes at low and high pH and allow for effective
controls to minimize oxidation.
36
CHAPTER 3
EFFECTS OF ALKALI/ACIDIC ADJUSTMENT ON LIPID OXIDATION IN SPANISH
MACKEREL HOMOGENATE
Introduction
Current fish protein sources throughout the world dwindle and methods to efficiently use
existing fish supplies are well warranted. A novel acid/alkali-aided process enables for the
isolation of fish proteins from traditional and unconventional (by-catch and pelagic) sources (7,
8). In this process, fish is homogenized with water (1:9). The homogenate is then adjusted to low
pH (2.5 to 3.5) or high pH (10.5- 11.5) to solubilize the muscle proteins. Soluble proteins are
then separated from unwanted fractions (14) using centrifugation at 10,000g. The solubilized
proteins are then subjected to isoelectric conditions (pH 5.5 to 7.0) and are centrifuged at
10,000g. The final protein isolate (138) can then be used for various seafood analog products
(surimi) and ingredient applications (8).
Efforts to access the protein potential of low value fish sources are currently underway.
Conventional processing of pelagic species presents various challenges due to their small size,
shape, bone structure, and high levels of pro-oxidants and unstable lipids (4-7). The novel
alkaline/aided process, versus conventional techniques for the isolation of fish proteins, offers
many advantages including improved yield, functionality, stability, and color (6, 13, 16, 18, 19).
Enhanced protein stability arises from the reduction in lipids and pro-oxidants. Triglyceride and
phospholipids content is reduced through the acid/alkali-aided process; however, it is the
reduction in phospholipids and heme proteins that are attributed to lower levels of lipid oxidation
and increased stability (5, 9, 10).
Throughout the acid/alkali aided process two major reactants in lipid oxidation are most
important: hemoglobin and lipids. Hemoglobin is deemed the primary catalyst of lipid oxidation
in fish (12, 42). Hemoglobin undergoes various changes at low pH that could increase its
37
oxidative capacities (43, 51). Minimal amounts of lipid ( less than 1%) in the presence of
hemoglobin, are sufficient for lipid oxidation to occur (58, 139). Separating all lipid and
hemoglobin from the fish protein isolate is not feasible; thus, lipid oxidation is inevitable during
the acid/alkali-aided process and measures to delay and/or prevent lipid oxidation are necessary.
The effects of the alkali/acid-aided treatment on lipid oxidation in fish homogenate has
received limited attention. Undeland et al. (67) investigated how low pH (2.7) adjusted herring
homogenate was affected by pH exposure, centrifugation, and antioxidant addition. Addition of
erythorbate (0.2%) by itself or in combination with STPP (sodium tripolyphosphate) (0.2%) and
EDTA (0.044%) as well as high speed centrifugation of the homogenate significantly reduced
lipid oxidation. Thus far, lipid oxidation in homogenates at extended exposure times (up to 12
hours) and expanded pH ranges (2.5 to 11.5) have not been studied.
The objectives of this study were to investigate: (1) the effects of low and high pH on the
development of primary and secondary products of lipid oxidation in fish muscle homogenates,
(2) the effect of preformed lipid oxidation products in raw material on lipid oxidation
development in the homogenates at high and low pH, (3) the effectiveness of different
antioxidant preparations on lipid oxidation in fish muscle homogenates at low and high pH and
(4) the relevance of these homogenate findings to lipid oxidation at subsequent steps in the
alkaline/acid aided process.
Materials and Methods
Raw Materials
The first batch of Spanish mackerel (Scomberomorous maculates) was obtained from
Savon Seafood in May 2004 (Tampa, FL). Postmortem age of the fish was approximately 3 days.
Due to limited supplies of Spanish mackerel and inability to control for batch variability, the
second batch of Spanish mackerel was freshly caught in September 2005 by lab mates and
38
myself on Captain Dennis Voyles charter, Reel Therapy (Cedar Key, Florida). Fish were
manually skinned, gutted, and filleted. One half of the fillets were minced using a large scale
grinder (The Hobart MFG. Co., Model 4532, Troy, Ohio). Fillets and ground fish (100 gram
quantities) were vacuum packed and stored at -80ºC until needed.
Chemicals
Tricholoracetic acid, sodium chloride, chloroform (stabilized with ethanol), methanol,
ammonium thiocyanate, iron sulfate, barium chloride dihydtrate, 12 N hydrochloric acid, sodium
hydroxide, tocopherol, L- ascorbic acid, alpha-tocopherol, and citric acid were obtained from
Fischer Scientific (Fair Lawn,New Jersey). Thiobarbituric acid and 1,1, 3,3 tetraethoxypropane
were obtained from Sigma Chemical Co. (St. Louis, MO). Propyl gallate and
(ethlenediaminetetraacetic acid 99%), EDTA were obtained from Acros Organics (New Jersey,
USA).
Storage Studies and Preparation of Homogenate
Pre-frozen fish (-80ºC) were thawed in plastic bags under cold running water and
immediately placed on ice for further preparation. For the aging studies, fillets and/or mince
were held in one gallon zip lock bags at 4ºC. The homogenate was prepared by diluting the
mince with cold deionized water (1:9) and mixing (45 second at 4ºC, speed 3) using a
conventional kitchen blender equipped with a rheostat (Staco® Inc., Dayton, Ohio, 3p 1500B).
The homogenate was aliquoted equally into 250 mL glass beakers. Homogenates were adjusted
to appropriate pH (2.5, 3, 3.5, 6.5, 10.5, 11, and 11.5) through drop-wise additions of 2M HCL or
NaOH. The homogenates were lightly stirred over the 12 hour duration at 4°C. To ensure that pH
values were maintained, homogenates were periodically checked and readjusted with drop-wise
amounts of 2 M NaOH or HCL.
39
Antioxidant Implementation
Antioxidants (tocopherol, ascorbic acid, citric acid, and EDTA) were thoroughly mixed
into the homogenate at 0.2% prior to pH adjustment. When used in combination, antioxidants
were added at 0.1% or 0.2% each prior to pH adjustment.
Total Lipid Analysis
Total lipid was measured in the minced muscle according to Lee et al. (140) using a 1:1
methanol/chloroform ratio. Results are reported as percentages based on wet weight.
Lipid Hydroperoxides
For designated time points, 3 grams of homogenate per each treatment was sampled for
lipid hydroperoxides. Samples were analyzed immediately. Lipid hydroperoxides were
determined according to Shantha et al. (141) with modifications by Undeland et al. (57). Total
lipids were extracted using 9 mL of a chloroform/methanol (1:1) mixture with vortexing. Sodium
chloride, 3 mL (0.5%), was then added to the sample. The sample was then vortexed for 30
seconds. The samples were centrifuged (Ependorf, model 5702, Brinkman Instruments, Inc.,
Westbury, N.Y) at 4 ºC for 10 minutes at 2,000 g. To alleviate phase separation problems due to
pH, homogenates were adjusted to pH 6.5 before extraction with experimentally determined
volumes of 1N NaOH or HCL. Two milliters of the bottom layer was obtained using a 2 mL
glass syringe and needle (Micro-Mate® Interchangeble glass syringe, stainless steel needle, 20G
x18”, Popper and Sons, Inc., New Hyde Park, N.Y). Two milliters of the chloroform layer was
mixed with 1.33 mL of chloroform/methanol (1:1). Ammonium thiocyantate (50 uL) was added
to the sample, and then iron chloride (94), 50 uL, was added to the sample. Following each
reagent addition, the mixture was vortexed for 2-4 seconds. After an incubation period of 5
minutes at room temperature (22.5ºC), samples were read at 500 nm using a spectrophotometer
(Agilent 8453 UV-Visible, Agilent Technologies, Inc., Palo Alto, CA). A blank contained all
40
reagents but the sample. A standard curve was constructed using a solution of iron (III) chloride.
Samples were kept in subdued light and on ice during the assay.
Thiobarbituric Acid Reactive Substances (TBARS)
TBARS were assessed according to Beuge et al. (142) as modified by Richards et al. (35).
A TCA/TBA solution (50%/1.3%) was freshly prepared and heated to 65ºC on the day of
analysis. It is imperative that the temperature not exceed 65ºC, to ensure that the TCA is
completely dissolved and that the TBA is not heat abused. The TCA/TBA solution was added to
0.5 g of homogenate and was heated for one hour at 65ºC. The samples were centrifuged (4ºC)
for 10 minutes at 2,500 g and read at 532 nm using a spectrophotometer (Agilent 8453 UVVisible, Agilent Technologies, Inc., Palo Alto, CA). A standard curve was constructed using 1,1,
3,3 tetraethoxypropane.
Analysis of Moisture and pH
Moisture content was assessed using a moisture balance (CSI Scientific Company, Inc.,
Fairfax, VA). The pH was measured using an electrode ( Thermo Sure-Flow electrode (Electron
Corp., Waltham, MA) and pH meter (Denver Instrument, model 220, Denver, CO). The
homogenate samples were directly measured for pH with continuous mixing.
Statistical Analysis
The SAS system was used to evaluate storage, pH adjustment, and antioxidant
implementation studies over time. All analyses were performed in at least duplicate and all
experiments were replicated twice. Analytical variation was established through first fitting two
way ANOVA models. Then multiple pair-wise comparisons were performed on estimated
measurement means, using Tukey adjustment controlling for Type I error.
Data was reported as mean ± standard deviation. The analysis of variance and least significant
difference tests were conducted to identify differences among means, while a Pearson correlation
41
test was conducted to determine the correlations among means. Statistical significance was
declared at p < 0.05.
Results
Changes in Spanish Mackerel Lipid Oxidation
Water and lipid Content
As it is a rarity to receive fresh and quality consistent raw material for processing, it is
important to understand how pre-formed lipid oxidation products in the raw material affect lipid
oxidation of mackerel muscle homogenates at low and high pH. Upon receiving and/or
harvesting of fish, baseline levels of oxidation were measured to assess similarities and
differences in lipid oxidation of fish freshly caught or at postharvest storage for a few days. To
determine the quality of the Spanish mackerel raw material, lipid oxidation products for fresh
and aged fillets and/or mince were measured. The average moisture content, pH, and lipid
content for Batch 1 and 2 fish muscle are summarized in Table 3-1.
The Spanish mackerel obtained in May (Batch 1) was significantly (p<0.05) higher in lipid
content (7.5 ± 2.12 %) than the September catches (Batch 2), 2.65± 0.07%. Average moisture
contents for May (Batch1) and September catch (Batch 2) were 79±0.251% and 77±1.76%,
respectively. Average pH content of the Spanish mackerel mince for May (Batch 1) and
September (Batch 2) were 6.49±0.045 and 6.49±0.015, which are in agreement with previous
research (67). Depending upon various environmental factors (20), the observed differences
among the batches are to be expected. Addressing compositional difference in catch due to
differences in seasonal variability and postmortem age is important to study the changes
observed in the acidic-alkaline aided process. Industrially, these compositional factors could
have significant bearing on processing parameters and the quality of protein isolates obtained on
a mass processing scale.
42
Lipid Oxidation in Whole and Minced Spanish Mackerel
It was evident that aging of the fish muscle at 4ºC led to an increase in oxidation products
(Figure 3-1). Significant (p<0.05) increases in lipid oxidation for Batch 1 mince and Batch 2
mince were observed for up to 4 days and between 2 and 4 days of storage, respectively. Batch 2
mince after 4 days of storage reached equal levels of oxidation (~30 umol MDA/kg of tissue) as
Batch 1 mince at day 1. To detect changes in lipid oxidation incurred through mincing of raw
material, a comparative analysis of fillets and mince in Batch 2 Spanish mackerel was conducted
(Figure 3-1). Lipid oxidation in the fillets did not significantly increase over time. Batch 2 mince
accumulated overall significantly (p<0.003) higher TBARS levels than its corresponding fillets.
The levels of lipid oxidation for same day storage were not significantly different between Batch
2 fillets and Batch 2 mince; except for day four, (p<0.05). Lipid oxidation levels were
significantly higher in Batch 2 for days 2 and 3 than 0 and 1 days of storage for Batch 2 fillets.
Maximum TBARS values of ~70 and ~34 µmol of malondialdehyde (MDA /kg) for Batch 1
mince and Batch 2 mince were reached after 1 and 4 days, respectively. When comparing lipid
oxidation over specific days, Batch 1 mince was significantly (p<0.05) greater than Batch 2
mince from 2 and 4 days of refrigerated storage.
Increasing the age of the raw material, significantly affected the lag time (P<0.004)
between Batch 1 and 2 mince. The onset of rancidity according to Richards et al. (14) is marked
by a TBARS increase of 20 umol/kg of tissue or more. Rancidity occurred within 1 day and after
2 days of storage in Batch 1 mince and Batch 2 mince, respectively. TBARS were overall
significantly (p<0.001) higher in Batch 1 than in Batch 2 mince.
Lipid Oxidation in pH Treated Spanish Mackerel Homogenates
An initial interest was to understand how preformed oxidation products in raw material
affected lipid oxidation with storage. The next objective was to evaluate the effects of low and
43
high pH on lipid oxidation in fish homogenates with respect to initial fish quality, cold storage,
and pH holding time.
Changes in homogenate lipid oxidation with storage
Figure 3-2 (b,d) and 3-3(b,d) and 3-4 (b,d) show that Batch 1 and 2 homogenates formed
significantly greater levels of TBARS with one and two days of mince storages versus
homogenates made from freshly prepared mince (p<0.001). As shown by Figures 3-2 (b) and 3-4
(d), TBARS levels in Batch 2 homogenates prepared from mince stored for two days were
equivalent to Batch 1 homogenates made from fresh mince.
Initial levels of PV and TBARS (Figures 3-2, 3-3, and 3-4) were equal in homogenates
prepared from mince stored for 0 or 1 days, and were significantly (p<0.001) less than
homogenates from mince stored for two days. Initial Batch 1 PV values for day 0, 1, and 2 were:
0.54± 0.017, 0.60±.018 , 2.52±0.42 mmol LPH/kg homogenate. Initial Batch 2 values at day 0, 1,
and 2 were 0.75±0.01, 0.630±0.09, 0.88±.007 mmol LPH/kg tissue, respectively. Significant
(p<0.001) increases in homogenate PV values for 0 and 1 days of mince storage were observed
after twelve hours of pH exposure. Endpoint levels of primary oxidation were equivalent in
homogenates made from 1 and 2 day stored mince and were significantly greater than
homogenates made from freshly prepared mince.
Batch 1 and 2 homogenates made from either 0 or 1 days storage mince started at
approximately the same TBARS levels (Figure 3-3). Batch 1 initial values on day 0, 1, and 2
were: 3.93 ±1.077, 6.025±3.33, and 13.62±3.38 umol MDA/kg of homogenate. Batch 2 initial
values at day 0, 1, and 2 were 3.22 ± 1.26, 6.09±0.57, and 7.23± 0.94 umol MDA/kg
homogenate. Figures 3-3 (b,d) and 3-4 (b,d) show that with holding times of four or more hours,
day one TBARS values reached equal or greater levels of oxidation compared to initial day 2
44
value (p<0.05). Holding times greater than 8 hours had no affect on TBARS formed in
homogenates made from mince aged for 1 or 2 days.
Changes in lipid oxidation with pH and storage
As Figure 3-2(b) shows, pH 3 and 3.5 homogenates made from freshly prepared Batch 1
mince had significantly greater TBARS levels (P<0.001) than all other pH treatments. The levels
of TBARS in the pH 3.5 homogenates were significantly greater that those at pH 3 (p<0.016).
Maximum levels of lipid oxidation in pH 3.5 and 3.0 homogenates made from freshly prepared
Batch 1 mince were 64.00 ±5.58 and 53.52±4.98 umol MDA/kg of homogenate, respectively
(Figure 3-3b). TBARS levels were significantly greater in homogenates made from 1 day stored
mince versus freshly prepared mince (Figure 3-3b). Overall day 1 levels of lipid oxidation at pH
3 and 3.5 were not significant from day 2 levels (Figure 3-4b). After one day of mince storage,
levels of lipid oxidation at pH 2.5 were significantly (p<0.001) greater than control and alkaline
treatments (Figure 3-4b), with the pH 2.5 homogenate still accruing greater levels of lipid
oxidation after two days of refrigerated storage (Figure 3-4b). After two days of mince storage,
pH 2.5 TBARS were significantly (p= 0.007) less than observed levels at pH 3 (Figure 3-4b). No
significant increases in lipid oxidation were observed among the control or alkaline treatments
with mince storage (Figure 3-2b,3-3b,3-4b).
Changes in lipid oxidation with initial quality and pH
Homogenates 3.5 and 3 had significantly greater TBARS than pH 2.5, while TBARS were
significantly higher in homogenate 2.5 than pH 6.5, 10.5, 11.0, and 11.5 homogenates. For
Batch 1, levels of oxidation at pH 3.5 and pH 3 were not significantly different. Control (pH 6.5)
homogenates in Batch 1 developed significantly greater TBARS than alkaline homogenates. For
Batch 2, pH 3.5 homogenates had significantly greater TBARS than alkaline pH homogenates
(p<0.04). Batch 1 mince acidic homogenates oxidized significantly more than Batch 2 acidic,
45
alkaline, and control homogenates. Batch 2 alkaline (pH 10.5, 11, 11.5) treatments incurred
significantly (p<0.001) less TBARS than control (6.5) homogenate in Batch 1 and equivalent
levels to the alkaline treatments in Batch 2.
Changes in lipid oxidation with holding time and pH
After 4 hrs and up to 12 hrs, Batch 1 incurred significantly greater PV levels than Batch 2
mince (Figures 3-2a,c). PV levels (mmol LPH/kg homogenate) started out significantly higher
(p<0.001) in Batch 1 compared to Batch 2; however, after 12 hours the Batch 1 and 2
homogenates had equivalent PV levels.
As shown in Figure 3-2(a), PV values increased significantly after four hours for
homogenates at pH 3.5. Homogenate 2.5, 3, and 6.5 significantly increased in PV levels after 12
hours. Homogenate 3 and 3.5 reached the same levels of primary oxidation after 12 hours and
were followed by homogenates 6.5, 2.5, and alkaline pH in descending order of primary
oxidation.
From Figures 3-2 (b), homogenate holding times of four or more hours at acidic pH were
sufficient for rancidity to occur in Batch 1 homogenates. Homogenates held for 4 hours at acidic
pH had significantly (p<0.001) greater TBARS than control and alkaline homogenates held for
the same time. A similar observance was found in unadjusted pH turkey homogenates with 2.5
hours of refrigerated storage (143). The pH 3.5 homogenates were not significantly different than
pH 3 homogenates after 4 hours. After eight hours, pH 3 and 3.5 homogenates were
significantly greater than pH 2.5 (p<0.05). After a 12 hour holding time, homogenates at 3.5 had
significantly (p =0.027) greater TBARS than pH 3.0 homogenates, and both pH treatments
proved to be significantly greater (p<0.001) than all other pH treatments at pH 2.5 and higher pH
values (6.5, 10.5, 11, 11.5). The control (pH 6.5) treatment had equal levels of oxidation
compared to pH 2.5 and alkaline values after 12 hours.
46
Overall pH effects
The extent and rate of lipid oxidation (TBARS) was significantly (p<0.001) higher in
homogenates at acidic pH (2.5, 3, 3.5) than at alkaline (10.5, 11,11.5) or control pH (6.5).
Figures 3-2(b), 3-3(b), 3-4(b) shows that there was no significant difference in TBARS
formation among alkaline homogenates. TBARS formation proceeded faster and to a greater
extent in the following homogenate treatment order: acid> control>alkaline. Levels of TBARS
in descending order relative to individual pH treatment were as follows: 3.5, 3, 2.5, 6.5, and
alkaline pH (10.5, 11.0, 11.5). Levels of lipid oxidation among alkaline homogenates were
insignificant.
Lipid Oxidation of Homogenates with Antioxidant Implementation
The effect of antioxidant implementation in homogenates at low and high pH was worth
investigating, as it may decrease overall levels of lipid oxidation in the early stages of the
alkali/acid aided process and improve the quality of the final protein isolate. A commonly used
combination of alpha tocopherol and ascorbic acid in muscle foods was employed.
Batch 1 mince was stored for one day and was subjected to pH 3 as per findings in the
previous section, in which rancidity occurred after one day of storage and the greatest overall
levels of homogenate lipid oxidation developed (3 and 3.5). The levels of lipid oxidation of
homogenates from day old mince at pH 3 and 3.5 were not significantly different. pH 3 was
chosen, as this is close to the acid-aided process pH values used for the isolation of proteins. To
ensure that Batch 2 was of comparable oxidative quality of Batch 1 after one day of storage,
Batch 2 mince was stored for 4 days (See Figure 3-1).
The antioxidants ascorbic acid and/or tocopherol, citric acid/and or EDTA were added at
two concentrations, 0.1% and 0.2% w/w, in numerous combinations. The effects of these
47
antioxidants were also tested under physiological pH conditions (pH 6.5) in the homogenate
system.
Levels of lipid oxidation between batches were comparable, as all antioxidant treatments
at pH 3 for Batch 1 homogenates, with exception of pH 3 (0.1% ascorbic acid + 0.1%
tocopherol) homogenates (Figure 3-5b), were not significantly different from pH 3 homogenates
from Batch 2 (Figure 3-5c). However, pH 6.5 from Batch 1 homogenate (Figure 3-5b) formed
less TBARS than pH 6.5 from Batch 2 (p=0.0002) (Figure 3-5 c). Batch 2 pH 3 (0.1% ascorbic
acid + 0.1% tocopherol) formed greater TBARS compared to the corresponding Batch 1
treatment. The difference between pH 3 (0.1% ascorbic acid + 0.1% tocopherol) samples from
Batch 1 and 2 is negligible regarding quality as both treatments formed total levels of lipid
oxidation below the point of rancidity (>20 umol MDA/kg tissue). Because pH 6.5 was not
comparable among batches, pH 6.5 treatments from Batch 1were only compared to other
homogenate treatments prepared from Batch 1 mince. Homogenate values were compared
among corresponding pH 3 treatments in both batches.
In Batch 1 treatments, various differences were observed. Overall levels of TBARS formed
in control pH 3 homogenates were significantly greater than all Batch 1 and 2 pH 3 treatments.
Homogenate treatment at pH 6.5 (0.1% ascorbic acid + 0.1% tocopherol) and pH 6.5 (0.1%
ascorbic acid) formed the highest TBARS levels (p<0.05) among all other treatments in Batch 1.
The antioxidant treatments were all equally effective in delaying/preventing oxidation in the pH
3 in contrast to pH 6.5.
As shown in Figure 3-6, the addition of EDTA (0.1%), citric acid (0.1%), and ETDA
(0.1%) + citric acid (0.1%), at pH 3 were rather ineffective in reducing lipid oxidation (TBARS,
PV). Over a twelve hour holding time, PV levels at pH 3 were not significantly affected by citric
48
acid (0.1%) and were significantly (p<0.001) reduced by the addition of EDTA (0.1%) and
EDTA (0.1%) + citric acid (0.1%) (Figure 3-6b). As shown in Figure 3-6b, the use of these
antioxidants when evaluated separately or in combination did not significantly reduce overall
levels of TBARS. These results suggest that the sole incorporation of chelators does not protect
against lipid oxidation in Spanish mackerel fish homogenates (51). However, in combination
with other antioxidants, chelators are promising inhibitors of oxidation according to Undeland
(67).
Lipid Oxidation of Homogenates after Readjustment to pH 5.5 and 7
In the acid and alkali-aided processes, homogenate proteins that are solubilized at low or
high pH are recovered by readjusting the pH to values where they aggregate. Consistent with
previous finding, Batch 1 mince was stored for one day and was subjected to pH 3 as per
findings in (Lipid oxidation in whole and minced Spanish Mackerel, Chapter 3) in which
rancidity occurred after one day of storage and the greatest overall levels of homogenate lipid
oxidation with respect to pH (3 and 3.5) homogenates occurred. Batch 2 mince was stored for 4
days as per previously discussed findings (Lipid oxidation in whole and minced Spanish
Mackerel, Chapter 3) to ensure approximate oxidation levels as Batch 1 samples. Homogenates
at pH 3 and 11 were held for one hour and then were readjusted to pH 5.5 and 7. The former pH
is commonly used to recover fish muscle proteins. However, improper folding of heme proteins
when readjusted from low pH to pH 5.5 may cause significant lipid oxidation due to incomplete
refolding (51). More refolding and less pro-oxidative activity is expected of hemoglobin with
readjustment to pH 7 (51).
The extent of lipid oxidation increased over time (p<0.0001) for all samples. Lipid
oxidation was measured at 0, 3, and 12 hours. These time points were chosen as per section,
“Lipid oxidation in whole and minced Spanish Mackerel,”(Chapter 3), in which trends of lipid
49
oxidation were realized. Batch 1 and 2 homogenates proceeded from the same initial levels of
oxidation; however, greater levels of lipid oxidation formed in Batch 1 compared to Batch 2
(p<0.0001).
Batch and homogenate pH readjustment
In Batch 1, greater levels of oxidation (PV,TBARS) occurred with the implementation of
antioxidants and readjustment from pH 3 to either pH 5.5 or 7 (p<0.0002) (Figures 3-7 a,b)
compared to pH 3. When homogenates were readjusted from pH 3 to 5.5 or 7 in the presence of
the antioxidant combination that was effective at pH 3 (0.1% ascorbic acid + 0.1% tocopherol)
(w/w) (Figures 3-5 b,c), a significant (p<0.0001) pro-oxidative effect was noted (Figures 3-7
a,b). This could be explained by the pro-oxidative effect of this combination observed (Figure 35a) at alkaline pH 6.5, which is intermediate to pH 5.5 and 7. Interestingly, the same antioxidant
treatments with readjustment from pH 11 to pH 5.5 or 7 did not exhibit the same pro-oxidative
effect on the homogenates in terms of PV values as it did for the system readjusted from low pH
(Figure 3-8a); however, the significant increase in TBARS with pH readjustment to pH 5.5 or 7
and antioxidant implementation at pH 11 (Figures 3-8b) were consistent with the same
treatments at pH 3 (Figure 3-7b). According to TBARS results, antioxidant treatments with
readjustment from pH 11 were not significantly different from the low levels of oxidation at
control pH 11 in Batch 2 (3-8c). The pro-oxidative effect of readjustment from pH 3 to 7 with
antioxidants was further emphasized, having greater lipid oxidation TBARS and PV values
(p<0.0035) compared to pH 7 alone as exhibited in Batch 1 and 2 (Figure 3-7 a,b,c,d). TBARS
and PV levels of samples with pH readjustment from pH 3 and 11 were less (p<0.02) than
corresponding treatments with antioxidants. Oxidation levels with readjustment to pH and
antioxidants proved to be higher at pH 7 than at pH 5.5 in Batch 1 (PV) p<0.0001.
50
Time and homogenate pH readjustment
After 12 hours, trends among measures of lipid oxidation (TBARS, PV) among pH 3 and
11 readjustment treatments were observed. TBARS levels proved to be greater (p<0.0001) in the
pH 11 homogenate adjusted to 5.5 (0.1% ascorbic acid + 0.1% tocopherol) compared to
homogenates at pH 3, 5.5, 7, 3-5.5, and 3-7. Readjustment from pH 3 to 7 (0.1% ascorbic acid +
0.1% tocopherol) proved to be more pro-oxidative than the same treatment readjusted from pH
11 and all other treatments at pH 3 and 11. Samples at pH 5.5 had greater TBARS than samples
at pH 11 (p < 0.02). The pH readjustment treatments were always surpassed in levels of lipid
oxidation (TBARS) by corresponding treatments with antioxidants in Batch 1 (p<0.02) (Figures
3-7b and 3-8b). In Batches 1 and 2, readjustments to pH 7 with antioxidants had greater TBARS
than treatments adjusted to pH 5.5 with and without antioxidants except for pH (p<0.0001),
except for one sample pH 11-5.5 (Figure 3-8 d). PV values seemed to follow the same trends as
TBARS; however, statistically little differences were observed.
Discussion
Lipid Oxidation and Storage Stability
Aged fish muscle should incur greater and more rapid lipid oxidation, and this was
observed in the fillets and ground muscle (Figure 3-1). This is in agreement with the literature. It
has been reported that if the fish muscle has surpassed the induction phase of lipid oxidation,
which can occur with storage, the rate and extent of lipid peroxidation will be greater (144). In
addition, the levels of lipid oxidation assessed in the ground muscle were significantly higher
than those observed in the fillets. This is evident in the fact that mince from freshly caught fish
(Batch 2) accumulated significantly higher levels of lipid oxidation than the corresponding fillets
(Figure 3-1). Regardless of postmortem age of the fish, there is a significant different in the
extent of lipid oxidation with mechanical processing. Aside from compositional muscle aspects,
51
the differences in lipid oxidation in the mince and fillets are well explained and are consistent
with observed changes in muscle due to extrinsic factors. Various external factors can contribute
to lipid oxidation in muscle food including and not limited too: temperature, oxygen exposure,
light, mechanical processing, packaging atmosphere, introduction of pro-oxidants, and
cooking/heating (145). The extrinsic factor most relative to the storage study is mechanical
processing of meat. Various levels of oxidation will arise through mincing, flaking,
emulsification, and other physical handling. Tissue disruption increases oxygen-lipid interaction,
and thus, increases the occurrence of lipid oxidation. Mincing also releases pro-oxidants such as
hemoglobin and myoglobin and brings them in closer contact with lipids (80, 146). The Spanish
mackerel muscle was maintained at 4ºC during the grinding process; however, a “painty” aroma
was observed and is indicative of increased lipid oxidation due to tissue disruption (30, 147) .
With storage and physical manipulation of tissue, reduced quality is evident in flavor, odor,
color, and nutritional changes (29, 30). This storage study served as comparative assessment of
raw material quality, and thus, serves as an applicable baseline for changes in lipid oxidation
implicated in the alkali/acid-aided protein isolation process.
Lipid Oxidation in Acid/Alkali Adjusted Homogenates
The initially higher PV levels in homogenates from Batch 1 mince is to be expected
(Figures 3-2, 3, 4). Batch 1 was stored for several days prior to arrival. Lipid hydroperoxides
are initial products of lipid oxidation. Lipid hydroperoxides are transient lipid oxidation
products, degrading soon after formation (48). Free radicals more readily react with antioxidants
than lipid components, due to bond dissociation energies. Once antioxidant defenses are
depleted, the rates of lipid hydroperoxides and secondary product formation will proceed faster
(148). According to Petillo et al. (143), antioxidant sources found in the blood plasma of stored
mince Spanish mackerel were exhausted with storage. Ascorbate and glutathione were exhausted
52
first as they serve to reduce the activation of metal containing pro-oxidants (148). Increases in
volatile secondary lipid oxidation products were perceptually detected in the homogenate
headspace; however, further studies are needed to substantiate these findings. The greatest
intensities were observed in the acidic treatments and negligible differences were detected in the
alkaline treatments. These easily detected volatiles are most likely aldehydes and can contribute
to rancidity and alterations in proteins comprising DNA, enzymes, and phospholipids structure
(87, 149, 150). In studies conducted by Alasalvar (151), the most predominant oxidation
products formed from cold storage of mackerel were hexanal, heptanal, octanal, dodecanal, 2pentenal, methyl-2-pentenal E-2-hexanal, and Z-4- heptenal. Thus, these volatiles could
potentially contribute to oxidation of the homogenates.
Comparable TBARS levels in homogenates made from one and two day old Batch 1 mince
is most likely due to the depletion of hydroperoxides, thereby, slowing the formation of
secondary products (148). Greater levels of lipid oxidation were observed among acidified
homogenates (Figures 3-2,3,4). These high levels may be explained in part by the high moisture
content (97-98%) in the homogenates. Water activity is an influential factor in the rate of lipid
oxidation. It is known that with high water activity, the rate and extent of lipid oxidation is
greater (148). Thus, the homogenates would accumulate higher levels of oxidation, not only due
to acidification but due to enhanced incorporation of pro-oxidants and oxygen via the water
phase.
During the homogenization process and pH adjustment, various changes to the
predominant catalysts, heme proteins can occur and affect rates of lipid oxidation. According to
Shikama (60) and Kristinsson (152), heme proteins become oxidatively stable due to increased
heme group preservation at alkaline pH (60, 152). With lowering of pH, subunit dissociation,
53
unfolding, release of heme groups and/or iron, and loss of oxygen can occur with heme proteins
(43, 50, 52, 153-155). Such modifications are more likely when hemoglobin/ myoglobin are
subjected to low pH; thereby, converting hemoglobin to more pro-oxidative forms including
deoxy-, met-, perferryl-, and ferryl-hemoglobin/myoglobin (37). Hemoglobin/ myoglobin Bohr
effect is amplified with decreases in pH below neutrality and leads to deoxygenation (50).
According to Richards et al. (42), deoxyhemoglobin/myglobin are highly pro-oxidative.
Consequentially, activated heme proteins can give rise to the production of reactive oxygen
species and decomposition of lipids and/or lipid oxidation products as observed in the acidified
homogenates.
The effect of pH on hemoglobin is particularly relevant to the processing of migratory
pelagic species. Hemoglobin in dark muscle pelagic species has a lower affinity for oxygen. A
more flexible tertiary structure facilitates quick release of oxygen for metabolic respiration (156).
Richards and Hultin (157) reported significantly greater catalytic activity in mackerel versus
trout hemoglobin (157). Significant pH effects could be realized during the isolation of dark
muscle fish species, due to less stable hemoglobin. Further investigation is required to
understand the effects of pH shifts on heme proteins in a variety of fish species homogenates and
how these implications affect lipid oxidation during protein processing.
The effects of pH in muscle foods, or single or multi-component model systems could
explain the effect of pH in the lipid oxidation of homogenates. Higher levels of lipid oxidation in
the homogenates at low pH could be explained by heme protein changes in aqueous model
systems. A study conducted by Kristinsson (12) revealed that considerable changes in flounder
hemoglobin tertiary structure occurred with lowering of pH. The most significant unfolding
transitions occurred between pH 4 and 3.5 and the greatest unfolding occurred at pH 2.5 (12).
54
Hemoglobin unfolding occurred very quickly (seconds) at pH 1.5-2.5, while unfolding at higher
acidic pH (3-3.5) took over 30 minutes. Alkaline pH (10-11) brought about very little or
negligible changes in heme structure (12). The low levels of lipid oxidation in the alkaline
homogenates can be explained by heme group and oxidative stability (51). The proxidative
activity of hemoglobin/myoglobin at low pH and inactivity at high pH is greatly dependant on
distal and proximal histidine interactions with the heme group and oxygen (158). Under acidic
conditions, interactions with the heme and distal histidine and proximal histidine can be lost
through autoxidation and denaturation (12, 152). Under alkaline conditions, substantial heme
unfolding at low pH would explain the high levels of oxidation observed in the acidic
homogenates (Figures 3-2 ,3 ,4). In a study by Kristinsson and Hultin (51), conformational
changes were reported for trout hemoglobin (0.8 uM in water, 5°C) at pH 3, 7, and 10.5 over
time. According to UV-Vis spectral (300-700), trout hemoglobin conformational changes
occurred in two minutes at pH 3; while changes at pH 7 and 10.5 were observed in 29 and 120
hours, respectively (159). The levels of oxidation in the Spanish mackerel homogenates at
neutral pH were less than acidic homogenates; this is most likely due to less structural flexibility
and hydrophobicity of hemoglobin at neutral pH (51). Additionally, homogenates at pH 6.5 were
more susceptible to autoxidation relative to alkaline pH and significantly less susceptible to autooxidation at low pH. Richards and Hultin (43) found that with lowering of pH from 7.6 to 6.0,
hemoglobin was quickly converted to pro-oxidative forms: deoxyhemoglobin and metmyoglobin
(43). In the same study, Richards and Hultin (43) found that a hemoglobin mediated washed cod
system had shorter lag phases at lower pH (43). Thus, the differing levels of lipid oxidation in
homogenates at neutral pH could be caused by less oxidative stability and considerably more
stability of hemoglobin at alkaline and acidic pH, respectively.
55
Various conformation changes within hemoglobin/myglobin could explain the differences
in oxidation among the pH treated homogenates. Because the extent of unfolding is greater at
low pH versus neutral and alkaline pH, hemoglobin becomes more hydrophobic and thus is
better able to interact with lipid membranes (152). Additionally, Kristinsson (12) found that
flounder hemoglobin existed as a molten globule (partially unfolded) state at low pH. A molten
globule is both hydrophobic and hydrophilic in nature, providing greater accessibility to lipids.
Thus at low pH, hemoglobin would be very pro-oxidative in the presences of membrane lipids in
the homogenates. Access of the protein to lipids is not only enhanced at low pH, but the heme
group becomes significantly exposed at low pH. According to Kristinsson (12), increased
hydrophobicity in flounder hemoglobin was attributed to greater tryptophan residue exposure
with lowering of pH from neutrality (pH 7). Increased rates of met-myoglobin/hemoglobin
formation are known to occur at low pH (60). Autoxidation, unfolding, and dissociation of
myglobin/hemoglobin can give rise to various reactive species such as ferryl hemoglobin. Ferryl
hemoglobin is believed to be a potent catalyst of oxidation in muscle foods, particularly at low
pH (56). These oxidized forms of hemoglobin/myglobin can readily catalyze lipid oxidation at
low pH (160) and could explain greater lipid oxidation in the acidic homogenates.
Results from washed cod muscle systems could help explain the overall levels and rate of
lipid oxidation (TBARS) in Spanish mackerel homogenates. The pH 6.5 homogenates were
lower than observed acidic homogenates and higher than alkaline homogenates in Batch 1
(Figures 3-2b,3b,4b).
Homogenate pH exposure times significantly affected the lipid oxidation development in
homogenates. Acidic homogenates made from freshly prepared Batch 1 mince developed
rancidity after four hours while neutral pH developed rancidity levels after 12 hours and alkaline
56
pH homogenates did not reach rancidity levels. Pazos et al. (161) observed similar trends in a
hemoglobin mediated lipid oxidation system using washed cod. At pH 3, the onset of lipid
oxidation (~5 hours) was less than pH 6.8 (~ 10 hours) and the endpoint oxidation levels were
comparable. Likewise, the pH 3.5 and 3.0 homogenates from freshly prepared Batch 1 mince
developed rancidity levels in approximately 6 hours, while pH 6.5 developed rancidity in
approximately 12 hours. Pazos et al. (161) observed similar endpoint levels of oxidation
(TBARS) in the pH 3 and pH 6.8 adjusted washed systems. The pH 3 and 6.8 homogenates, from
freshly prepared Batch 1 mince, did not reach equivalent levels of oxidation at 12 hours;
however, greater holding times or mince storage may have resulted in equal endpoint levels of
lipid oxidation (Figures 3-2(b), 3-4(b)). As observed in the homogenates, Kristinsson and Hultin
(51) reported similar lipid oxidation trends in hemoglobin mediated washed system samples.
Trout hemoglobin was adjusted to acidic (pH 2.5, 3, 3.5), neutral (pH 7), and alkaline (10.5, 11)
pH values. The greatest extent and rate of lipid oxidation occurred at low pH, followed by
neutral pH, and then alkaline pH. Increased rates and extents of lipid oxidation followed by
longer lag phases and decreased levels of lipid oxidation at neutral (6.5) was observed in the
Spanish mackerel homogenates. The development of TBARS was greatest in a washed cod
system at pH 2.5 and 3 than at pH 3.5 (51). The homogenate samples at pH 3.5 over time had
greater rates and levels of lipid oxidation than pH 2.5 and 3. As Kristinsson and Hultin (51)
observed at pH 10.5 and 11 in the washed cod system, minimal levels of lipid oxidation
developed in pH 10.5, 11, and 11.5 homogenates.
Although quantitative color differences among the homogenates were not reported, it was
evident that acidified homogenates were yellowish white while the alkaline and neutral
homogenates were pink in color. From these color differences, it is evident that significant
57
unfolding and denaturation occurred in the acidified homogenates, while the less unfolded and
native heme proteins in the other treatments retained their color. Richards et al. (162) and Lee et
al. (163) found a significant loss in redness, “a” value was due to conformational changes as
heme proteins were converted from their oxy-to met-forms. Hultin (5) reported that meat color
changes are byproducts of free radical induced oxidation, secondary to lipid hydroperoxide
formation.
Homogenate pH Readjustment
To investigate changes in lipid oxidation in acid/alkali-aid process upon pH readjustment,
Spanish mackerel homogenates were held for one hour at pH 3 and 11 and were then readjusted
to pH 5.5 and 7. The homogenates were held at pH 5.5 and 7 at 4°C for 11 hours with constant
mixing. The readjustment from acidic and alkaline pH to pH 5.5 and 7 mimics the pH
readjustment step in the acid/alkali aid process in which proteins are precipitated and then
centrifuged to obtain a protein isolate.
To date, no studies have investigated changes in lipid oxidation over time in acid/alkali
homogenates upon readjustment, nor has the effectiveness of antioxidants in preventing lipid
oxidation during homogenate readjustment been investigated. Lipid oxidation levels (TBARS) in
pH 3 and 11 homogenates were not significantly affected by pH readjustment to pH 5.5 or 7
(Figures 3-7b, 3-8b). Trout hemoglobin refolding upon readjustment from acidic or alkaline pH
to neutral pH 7 was studied by Kristinsson and Hultin (51). Trout hemoglobin was adjusted to
pH 3 and 10.5 and was then readjusted to pH 7 and added to a washed cod system. The effect of
pH lowering and pH holding times on the pro-oxidative activity of refolded hemoglobin were
investigated. Less refolding occurred for more unfolded acidified hemoglobin (pH 2.5) compared
to hemoglobin at pH 3.5. Longer acidic pH holding times and lower pH values gave
hemoglobins that were incompletely unfolded and most likely more pro-oxidative relative to
58
native hemoglobin (pH 7). Alkaline readjusted hemoglobin did not incur significant changes in
pro-oxidative activity under refolding conditions (51). Similarly, pH readjustment of
homogenates from pH 11 did not significantly change and were minimal (Figures 3-8 b). In
contrast to Kristinsson and Hultin (51) findings in which increased levels of lipid oxidation were
observed upon readjustments from pH 3, levels of lipid oxidation in homogenates at pH 3 with
readjustment to pH 5.5 and 7 remained unchanged. Insignificant changes in homogenate lipid
oxidation levels could be attributed to changes in hemoglobin conformation and thus prooxidative activity. Kristinsson and Hultin (51) observed significant increases in pro-oxidative
activity of hemoglobin at pH 3, when held at pH 3 for 20 minutes versus 90 seconds.
Conformational studies gave proof to greater refolding with shorter holding times at pH 3 (51).
The hemoglobin in homogenates at pH 3 were likely held for a duration long enough (1 hour) to
cause irreversible denaturation and thus inability for structure recovery. Similar to lowering of
pH, incomplete refolding can increase the exposure of hydrophobic sites and heme within the
tertiary structure of hemoglobin. The site exposure makes hemoglobin more pro-oxidative (164).
Due to longer holding times, acidified hemoglobin was unable to refold. Pro-oxidative activity
of the hemoglobin would then be very similar among the pH 3 homogenates with and without pH
readjustment.
Interestingly, the incorporation of tocopherol and ascorbic acid into pH 3 and pH 11
readjustment treatments proved to be pro-oxidative, compared to controls and readjustment
treatments (Figures 3-7b, 3-8b). The pH 3 and 11 homogenates with antioxidants formed greater
levels of lipid oxidation when readjusted to pH 7 compared to 5. These increases in oxidation
with readjustment to neutral pH is most likely impart due to the instability of ascorbic acid at
neutral pH as per discussion in the following section.
59
Antioxidant Implementation
As lipid oxidation could be problematic during acid processing of various dark muscled
fish species (13), antioxidants were added to the homogenates. Ascorbic acid and tocopherol,
either alone or in combination at 0.1% and 0.2%, were equally effective in preventing oxidation
in pH 3 homogenates (Figures 3-5 b,c). In a study conducted by Undeland et al. (67),
incorporation of antioxidants in the pre-wash and/or homogenization step significantly decreased
levels of lipid oxidation in the acid-aided isolates obtained from herring. Incorporation of
erythorbate (0.2% w/w) in combination with EDTA (0.2% w/w) or sodium tripolyphosphate
(STPP) (0.2% w/w) into herring homogenates at pH 2.5, have proven beneficial in maintaining
the quality of dark muscled fish protein isolate. Erythrobate, and iso-form of ascorbic acid were
found to be effective heme-iron reducing agents in herring homogenates. According to other
findings (68, 153), antioxidant implementation early in the modification of mackerel fish muscle
is most efficacious in reducing overall levels of oxidation in the final product. However,
antioxidant interventions prove to be most beneficial in light versus dark mince, most likely due
to significant levels of hemoglobin (68, 153). Antioxidants could not only be implemented in the
homogenates, but in the pre-wash of the Spanish mackerel. Additionally, various procedural
modifications in combination with antioxidants could reduce lipid oxidation in homogenates
made from dark muscle species: washing the mince, removal of dark muscle, and/or reducing the
holding time at extreme pH (67). The effects of how tocopherol and ascorbic acid incorporation
into Spanish mackerel homogenates affect lipid oxidation levels in the protein isolate requires
further investigation.
The incorporation of ascorbic acid (0.1%) alone or in combination with tocopherol (0.1%)
at pH 6.5 (Figure 3-5b) proved to be pro-oxidative in comparison to corresponding treatments at
pH 3 (Figures 3-5b). A decrease in lipid oxidation due to use of lower antioxidant concentration
60
increases was observed in the pH 3 homogenates (Figure 3-7b). Studies have shown a
concentration and pH effect on the ability of ascorbic acid and tocopherol to scavenge free
radicals. It was expected that lower levels of ascorbic acid (0.1%) rather than higher levels
(0.2%) in the homogenates would be pro-oxidative. Previous studies demonstrated a
concentration effect of ascorbic acid and tocopherol at physiological pH (64, 65). It has been
reported that at high and low concentrations, tocopherol and ascorbic acid impart pro-oxidant
and antioxidant effects, respectively. Cillard (165) found that high concentrations ( > 7,600
mg/kg, 0.7%) and low concentration (<3,800 mg/mL, 0.38%)of tocopherol induced prooxidative effects in linoleic acid. Ascorbic acid could be prooxidative (>5,000 mg/kg, 0.5%) in a
hemoglobin containing solution (166). Watts (167) found that ascorbic acid was pro-oxidant at
176 mg/mL in a carotene-linoleate aqueous model, and it was concluded that chelators were
needed to sequester catalytic iron in the presence of these ascorbic acid concentrations.
Ultimately, there is a net prooxidative or antioxidative role played by these antioxidants.
Ascorbic acid is known to promote lipid oxidation by facilitated iron release from heme proteins
(168), on the other hand ascorbate can quench myoglobin ferryl radicals post mortem (169). The
roles that ascorbic acid and tocopherol play in lipid oxidation is determined by net antioxidant
activity. At pH 3, ascorbic acid and tocopherol either alone or in combination at 0.1% and 0.2%
were equally effective in preventing oxidation (Figures 3-5b,c), very similar to the study
conducted by Undeland et al. (67) in which erythorbate (0.2%) was added to herring
homogenates. Hultin and Kelleher (68) found that washing Atlantic mackerel mince with 0.2%
ascorbic solution and 1.5 mM EDTA, affectively prevented lipid oxidation. At high
concentrations (0.2%), ascorbic acid can quench ferryl radicals. In a study conducted by
Richards and Li (170) , incorporation of high (2.2mM) and low levels (20-200 uM) of ascorbate
61
into hemoglobin-mediated lipid oxidation system resulted in inhibition and no pro-oxidative
effect on TBARS, respectively. In the same study conducted by Hultin and Kelleher (68),
ascorbic acid may have prevented lipid oxidation through reduction of heme iron from the ferric
to the ferrous form. The concentration effect of antioxidants could be very different among
studies; due to differences in pH and aqueous system compositions. The change in effectiveness
of these antioxidants with pH and concentration is to be expected.
Ascorbic acid’s structural and oxidative instability at neutral pH may explain the high
degree of oxidation found in 6.5 pH treatments using ascorbic acid at 0.1% concentrations as
well as significant increase in oxidation after pH readjustment to pH 5.5 and 7 (0.1% tocopherol
+ 0.1% ascorbic acid). Because antioxidants have various pH dependant ionic forms, a non-linear
relationship between pH and antioxidant stability exists (171). Changes in pH are likely to cause
changes in antioxidant structural stability and thus function. It has been found that tocopherol
and ascorbic acid have different stabilities at different pH values (172). In general, ascorbic acid
is stable at acidic pH and unstable at neutral and alkaline pH. Tocopherol is stable at neutral,
acidic, and alkaline pH (172). Tocopherol’s stability and thus ability to prevent oxidation exist
over a wide pH and thus it is most likely that ascorbic acid counteracts the positive effects of
tocopherol, when adjusted to neutral pH. It is likely that ascorbic acid is more antioxidative at
low pH compared to pH 6.5. Ascorbic acid is deprotonated with neutralization as it has a pKa
4.04 (173). Ascorbic acid could effectively reduce heme iron from ferric to the active ferrous
form. The ferrous form is known to breakdown lipid hydroperoxides and hydrogen peroxide to
form lipid radicals and ROS, respectively (169, 170, 174). Thus, it is likely that in the pH 6.5
homogenates, ascorbic acid is prooxidative through indirect propagation of lipid oxidation. At
low pH, ascorbic acid served as an effective antioxidant. Similarily, Undeland et al. (67) found
62
that ascorbic acid at 0.2% concentrations effectively reduced lipid oxidation in the protein
isolates made from acidified (pH 2.5) herring homogenates. Due to the fact that ascorbic acid
was not effective in reducing lipid oxidation in the homogenates at pH 6.5, a more stable analog,
erythorbate, may be a better choice with pH fluctuations. It is likely that both concentration
effect and pH influenced the effectiveness of ascorbic acid as an antioxidant.
Due to the key roles antioxidant roles exemplified in various studies, ascorbic acid and
tocopherol were used in the present studies. Post mortem antioxidants are preferentially
exhausted due to one-electron reduction potentials, also known as the “pecking order” (175) and
locations within muscle cells (5). The major antioxidant defenses against lipid oxidation in
muscle foods are the water-soluble inhibitors, ascorbate and glutathione, and the lipid-soluble
inhibitors, tocopherol and coenzyme Q. The tocopherol isomer of greatest abundance in skeletal
muscle is α- tocopherol and accumulates through dietary intake (176). In a paired fillet kinetics
study conducted by Petillo et al.(143), the loss of antioxidants in Atlantic mackerel was as
follows: glutathione and ascorbate, alpha tocopherol, and ubiquinone-10 (oxidized form of
coenzyme Q). The rate of antioxidant loss was significantly higher in the dark muscle than in the
light muscle (177). Similarly, the same sequence of antioxidant depletions was indicative of oneelectron transfers, according to Jia et al. (28) and Pazos et al. (178, 179).
Over time, it is evident that significant increases in TBARS levels occurred after four
hours in the pH 3 homogenate samples (Figure 3-5b). This is most likely due to losses in
tocopherol, which is highly correlated with increases in lipid oxidation and hydroperoxides in
muscle foods (177, 179, 180). The loss of alpha tocopherol is deemed the last defense against
lipid oxidation, in vitro. In turkey breast and thigh homogenates, the loss of α-tocopherol as it
relates to lipid oxidation was observed. Within 2.5 hours at 4ºC, there was significant increase in
63
lipid oxidation ( TBARS and PV) and a substantial loss of α-tocopherol (50%) (181). It is
evident that the loss of tocophereol was prevented by the implementation of ascorbic acid and
tocopherol. Other studies have reported preservation of tocopherol in muscle foods. Propyl
gallate and grape polyphenols were found to significantly preserve tocopherol in horse mackerel
during frozen storage (179). It is commonly known that ascorbate regenerates tocopherol and
serves an important function in preserving the free radical scavenging capacity of tocopherol in
membrane systems (182). Ascorbate is known for its ability to reduce hypervalent hemoglobin, a
highly pro-oxidative hemoglobin species. Ascorbate also possesses the ability to break down
hydroperoxides, which are activators of met-Mb/Hb to the ferryl form (183). It is widely
accepted that in membrane systems, a lipophilic antioxidant (tocopherol) and a hydrophilic
antioxidant (ascorbate) should be combined due to synergistic affects. In the homogenates, a
significant synergistic affect was not evident as all antioxidants treatments, alone or in
combination, had equivalent levels of lipid oxidation inhibition at pH 3 (Figure 3-5 b).
EDTA and Citric Acid Implementation at Low pH
EDTA and citric acid at 0.1% concentrations, together or in combination were ineffective
in inhibiting lipid oxidation (PV, TBARS) in pH 3 homogenates (Figures 3-6a,b). In contrast to
these finding, Undeland et al. (67) found that the implementation of EDTA (1.5 mM; 0.044%)
effectively reduced lipid oxidation in protein isolates obtained from acidified (pH 2.5) herring
homogenates. Similar to pH 3 homogenates results , EDTA (2.2 mM) did not significantly affect
lipid oxidation levels in a hemoglobin mediated lipid oxidation system according to Richard and
Li (170). Due to the ineffectiveness of the chelators, it is evident that free iron or iron released
from heme may not have served as the predominant catalyst in the lipid oxidation of the
homogenates. To fully evaluate the effectives of these chelators, it would be necessary to
evaluate the levels of TBARS formed in the final isolate. The appropriate implementation of
64
antioxidants in different types of muscle systems should be considered, and the effectiveness of
appropriate heme-related and/or low molecular appropriate antioxidants should be assessed. In a
model study conducted by Kanner et al (41), oxidation was generated through two catalytic
mechanisms: hydrogen peroxide activated met-myoglobin and iron chloride/ascorbate; however,
only the oxidation induced by the heme bound iron was inhibited by muscle cytosolic extracts in
washed fish systems (41). Obviously the chelators and or antioxidant mechanism differ for free
transition metals versus those that are heme bound proteins. Additionally, chelators in
combination with other antioxidants could be effective in reducing overall lipid oxidation in the
final product isolate as observed by Undeland et al. (67).
Conclusion
This storage study enables the understanding of how raw material of different quality may
affect lipid oxidation formation during the alkali/acid-aided protein isolation process, particularly
in the initial homogenization and pH adjustment step of pelagic specie protein extraction. Later
research (Chapter 4) will investigate whether the lipid oxidation incurred during the
homogenization steps affects the quality of the actual protein isolate. Lipid oxidation formation
proved to be greatest in Spanish mackerel homogenates at pH 3 and 3.5. The implications of
holding the proteins at an acidic pH followed by pH readjustment and isolation on protein and
lipid oxidation will be addressed in the following chapter.
The implementation of tocopherol and ascorbic acid at low pH significantly decreased
levels of lipid oxidation in homogenates; however, antioxidants proved to be pro-oxidative with
pH readjustment of homogenates from acidic and alkaline pH to neutral pH. Increases in lipid
oxidation in the readjusted homogenates with antioxidants were observed. Readjustment alone
had no affect on the levels of lipid oxidation in homogenates. Longer holding times significantly
increased the levels of lipid oxidation in the readjusted homogenates with antioxidants.
65
Successful implementation of antioxidants into the acid and alkali-aided protein isolation process
requires further investigation of appropriate antioxidants and their effectiveness at various pH
values and holding times.
66
Table 3-1. Initial average mince values for moisture content, pH, and lipid content. Assessments
were conducted on two separate batches of fish: Batch 1 (3 day old raw material) and
Batch 2 (freshly caught raw material). Lipid content means and standard deviations
were calculated from triplicate analysis. Different letters within each row indicate a
significant difference (p<0.05).
Batch 1 May- 3days old
Moisture Content (%)
pH
Batch 2 (September)- Fresh
a
79±0.251
77±1.76a
6.49±0.045a
6.49±0.015a
7.5±2.12a
2.65± 0.07b
Lipid Content (%)
100
TBARS (µmol MDA/ kg tissue)
90
Batch 1 Mince
80
Batch 2 Mince
70
Batch 2 Fillet
60
50
40
30
20
10
0
0
1
2
3
4
Time (days)
Figure 3-1. Lipid oxidation in Spanish Mackerel mince as assessed by thiobarbituric acid
reactive substances (TBARS). Experiments were conducted on two separate batches
of fish: Batch 1 (3 day old raw material received in May) and Batch 2 (freshly caught
raw material). Each sampling point is representative of a TBARS mean with standard
deviation bars from triplicate analysis.
67
(A)
pH 2.5
10
100
TBARS (umol M DA/ kg of homogenate)
PV (mm ol LHP/kg tissue)
12
pH 3.0
pH 3.5
8
pH 6.5
pH 10.5
6
pH 11
4
pH 11.5
2
0
0
2
4
6
8
10
90
pH 2.5
80
pH 3
70
pH 3.5
60
pH 6.5
pH 10.5
50
pH 11
40
pH 11.5
30
20
10
0
12
0
Time (hours)
12
2
4
6
Time (hours)
8
10
12
10
12
10
TBARS (umol M DA/ kg of homogenate)
100
pH 2.5
PV (m mol/kg hom ogenate)
(B)
(C)
pH3.0
pH 3.5
8
pH 6.5
pH 10.5
6
pH 11.0
4
pH 11.5
2
0
0
2
4
6
8
10
pH 2.5
90
(D)
pH 3
80
pH 3.5
70
pH 6.5
60
pH 10.5
50
pH 11
40
pH 11.5
30
20
10
0
12
0
Time (hours)
2
4
6
Time (hours)
8
Figure 3-2. Development of lipid hydroperoxides and thiobarbituric reactive substances
(TBARS) in pH treated homogenates. Panels A and B represent lipid hydroperoxides
and TBARS in homogenates made from Batch 1 mince that was freshly prepared.
Panels C and D represent lipid hydroperoxides and TBARS in homogenates made
from Batch 2 mince that was freshly prepared. Each sampling point is representative
of TBARS and PV mean with standard deviation bars from triplicate analysis.
68
100
pH 2.5
10
TBA RS ( um ol M D A / kg of hom ogenate)
PV (m m ol LPH /kg hom ogenate)
12
(A)
pH 3.0
pH 3.5
8
pH 6.5
pH 10.5
6
pH 11
4
pH 11.5
2
0
0
2
4
6
8
10
70
pH 6.5
60
pH 10.5
50
pH 11
pH 3.0
8
pH 3.5
pH 6.5
6
pH 10.5
4
pH 11
pH 11.5
2
0
0
2
4
6
8
10
pH 11.5
40
30
20
10
0
TBARS (umol M DA/ kg of homogenate)
PV (mmol LHP/kg of homogenate)
(C)
pH 3.5
0
12
12
pH 2.5
(B)
pH 3.0
80
Time (hours)
10
pH 2.5
90
12
4
6
Time (hours)
8
10
12
10
12
100
90
pH 2.5
80
pH 3.0
70
pH 3.5
60
pH 6.5
50
(D)
pH 10.5
40
pH 11
30
pH 11.5
20
10
0
0
Time (hours)
2
2
4
6
8
Time (hours)
Figure 3-3. Development of lipid hydroperoxides and thiobarbituric reactive substances
(TBARS) in pH treated homogenates. Panels A and B represent lipid hydroperoxides
and TBARS in homogenates made from Batch 1 mince, after 1 day of storage. Panels
C and D represent lipid hydroperoxides and TBARS in homogenates made from
Batch 2 mince, after 1 day of storage. Each sampling point is representative of
TBARS and PV mean with standard deviation bars from triplicate analysis.
69
100
pH 2.5
pH 3.0
pH 3.5
pH 6.5
pH 10.5
pH 11.0
pH 11.5
10
8
(A)
TBARS (um ol M D A/ kg of hom ogenate)
PV (m m ol LH P/kg hom ogenate)
12
6
4
2
0
0
2
4
6
8
10
90
pH 2.5
80
pH 3.0
70
pH 3.5
pH 6.5
60
pH 10.5
50
pH 11.0
40
pH 11.5
30
20
10
0
12
0
Time (hours)
2
4
6
Time (hours)
8
10
12
100
pH 2.5
10
TBARS (umol M DA/ kg of homogenate)
12
PV (m m ol LHP/kg hom ogenate)
(B)
(C)
pH 3.0
pH 3.5
8
pH 6.5
6
pH 10.5
pH 11
4
pH 11.5
2
0
0
2
4
6
8
10
90
pH 2.5
80
pH 3.0
70
pH 3.5
60
(D)
pH 6.5
50
pH 10.5
40
pH 11
30
pH 11.5
20
10
0
12
0
Time (hours)
2
4
6
Time (hours)
8
10
12
Figure 3-4. Development of lipid hydroperoxides (PV) and thiobarbituric reactive substances
(TBARS) in pH treated homogenates. Panels A and B represent lipid hydroperoxides
and TBARS in homogenates made from Batch 1 mince, after 2 days of storage.
Panels C and D represent lipid hydroperoxides and TBARS in homogenates made
from Batch 2 mince, after 2 days of storage. Each sampling point is representative of
TBARS and PV mean with standard deviation bars from triplicate analysis.
70
(A)
pH 3
pH 6.5 (control)
pH 6.5 Vit. C (0.1%)
pH 6.5 Vit. E (0.1%)
pH 6.5 Vit. C (0.1%) + Vit. E (0.1%)
pH 6.5 Vit. C (0.2%)
pH 6.5 Vit. E (0.2%)
pH 6.5 Vit.C (0.2%) + Vit. E (0.2%)
90
80
70
60
50
40
TBARS (um ol M DA/ kg of hom ogenate)
TBARS (um ol M DA/ kg of hom ogenate)
100
30
20
10
0
0
2
4
6
8
10
100
90
pH 3.0 (Acid control)
80
pH 6.5 (Control)
70
(B)
pH 3 Vit C. (0.2%)
60
pH 3 Vit E. (0.2%)
50
pH 3 Vit C. (0.1%) + Vit E. (0.1%)
40
30
20
10
0
12
0
2
4
Time (hours)
6
8
10
12
Time (hours)
TBA RS (um ol M D A /kg hom ogenate)
100
90
pH 3.0
80
pH 6.5
(C)
pH 3 0.1% C
70
pH 3 0.1% E
60
pH 3 Vit. C (0.1%)+ Vit E (.1%)
50
pH Vit. C (0.2%)
40
pH 3 Vit. E (0.2%)
30
pH 3 (0.2%E+0.2%C)
20
10
0
0
2
4
6
8
10
12
Time (hours)
Figure 3-5. Effect of ascorbic acid and tocopherol at 0.1% and 0.2% w/w on lipid oxidation in
homogenates. Panel A represents pH 3 homogenate from Batch 1, after 1 day of
storage. Panel B represents pH 6.5 homogenate from Batch 1, after 1 day of storage.
Panel C represents pH 3 homogenates equivalent to Batch 1 quality by using Batch 2,
after 4 days of storage. Each sampling point represents the mean and standard
deviation, from triplicate analysis, repeated twice.
71
PV (umol LPH/ kg of homogenate)
12
11
10
(A)
pH 3
9
pH 3 (0.1% EDTA + 0.1% Citric acid)
8
7
0.1% Citric acid
6
pH 3 EDTA
5
4
3
2
1
0
0
2
4
6
8
10
12
Time (hours)
TBARS (MDA/ kg of homogenate)
100
(B)
90
pH 3
80
pH 3 (0.1% EDTA+ 0.1% Citric acid)
pH 3 Citric acid
70
pH 3 EDTA (0.2%)
60
50
40
30
20
10
0
0
2
4
6
8
10
12
Time (hours)
Figure 3-6. Effect of EDTA and citric acid at 0.1% w/w on lipid oxidation in pH 3 homogenates.
Panel A represents lipid hydroperoxides in Batch 1, after 1 day of cold storage.Panel
B represents TBARS in homogenates. Each sampling represents the mean and
standard deviation, from triplicate analysis from replicate experiments.
72
pH 3
pH 3 to 5.5
pH 3 to 7
pH 3 to 5.5 (0.1% Vit. C+ 0.1% Vit.E)
pH 3 to 7 (0.1% Vit.C +0.1% Vit.E)
15
TBARS (um ol M DA/ kg of hom ogenate)
PV(mmol LPH/ kg of homogenate)
20
(A)
10
5
0
0
2
4
6
8
10
220
pH 3
pH 3 to 5.5
pH 3 to 7
pH 3 to 5.5 (0.1% Vit. C + 0.1% Vit. E)
pH 3 to 7.0 (0.1% Vit.. C +0.1% Vit. E)
200
180
160
140
120
100
80
60
40
20
0
0
12
2
4
TBA R S (um ol M D A / kg of hom ogenate)
PV (m m ol LPH / kg of hom ogenate)
20
(C)
pH 3.0
pH 3.0 to 5.5
pH 3 to 7
pH 3 to 5.5 (0.1% Vit.C+0.1% Vit. E)
pH 3 to 7 (0.1% Vit.C+0.1% Vit. E)
pH 5.5
pH 7
10
5
0
0
2
4
6
6
8
10
12
10
12
Time (hours)
Time (hours)
15
(B)
8
10
12
50
pH 3
pH 3 to 5.5
pH 3 to 7
pH 3 to 5.5 (0.1% Vit. E+0.1% Vit.C)
pH 3 to 7 (0.1% Vit. C+ 0.1% Vit. E)
pH 5.5
pH 7
40
30
(D)
20
10
0
0
2
4
6
8
Time (hours)
Time (hours)
Figure 3-7. Effect of pH readjustment to neutral pH (5.5 and 7.0) antioxidants (0.1% ascorbic
acid +0 .1% tocopherol), on development of lipid hydroperoxides (PV) and
thiobarbituric reactive substances (TBARS) in pH treated homogenates. Panel A and
B represent lipid hydroperoxides and TBARS in homogenates made from Batch 1
mince, after 1 day of storage. Panels C and D represent lipid hydroperoxides and
TBARS in homogenates made from Batch 2 mince, after 1 day of storage (4°C). A
control homogenate at pH 3 is also shown. Homogenates were held at pH 3 for one
hour prior to readjustment. Each sampling point is representative of the mean with
standard deviations from triplicate analysis, repeated twice.
73
(A)
pH 11
pH 11 to 5.5
15
TBARS (um ol M DA/ kg of hom ogenate)
PV(mmol LPH/ kg of homogenate)
20
pH 11 to 7
pH 11 to 5.5 (0.1% Vit. C+0.1%Vit. E)
10
pH 11 to 7 (0.1% Vit. C+0.1% Vit. E)
5
0
0
2
4
6
8
10
200
180
pH 11
pH 11 to 5.5
pH 11 to 7
pH 11 to 5.5 (0.1% Vit. C+0.1% Vit. E)
pH 11 to 7 (0.1%Vit. C+0.1% Vit.E)
160
140
120
100
80
60
40
20
0
12
0
2
4
Time (hours)
15
TBA RS (um ol M D A / kg of hom ogenate)
PV (m m ol LPH / kg of hom ogenate)
(C)
pH 11 to 5.5
pH 11 to 7
pH 11 to 5.5 (0.1% Vit. E+0.1% Vit. E)
pH 11 to 7 (0.1% Vit.C+0.1%Vit. E)
10
pH 5.5
pH 7
5
0
0
2
4
6
6
8
10
12
10
12
Time (hours)
20
pH 11
(B)
8
10
50
pH 11 to 5.5
pH 11 to 7
pH 11 to 5.5 (0.1% Vit. C+0.1% Vit. E)
30
pH 11 to 7 (0.1%Vit.C+0.1% Vit.E)
pH 5.5
20
pH 7
10
0
0
12
(D)
pH 11
40
2
4
6
8
Time (hours)
Time (hours)
Figure 3-8. Effect of pH readjustment to neutral pH (5.5 and 7.0) antioxidants (0.1% ascorbic
acid +0 .1% tocopherol), on development of lipid hydroperoxides (PV) and
thiobarbituric reactive substances (TBARS) in pH treated homogenates. Panel A and
B represent lipid hydroperoxides and TBARS in homogenates made from Batch 1
mince, after 1 day of storage. Panels C and D represent lipid hydroperoxides and
TBARS in homogenates made from Batch 2 mince, after 1 day of storage (4°C). A
control homogenate at pH 11 is also shown. Homogenates were held at pH 11 for one
hour prior to readjustment. Each sampling point is representative of the mean with
standard deviations from triplicate analysis, repeated twice.
74
CHAPTER 4
THE EFFECTS OF THE ALKALI/ACIDIC AIDED ISOLATION PROCESS ON LIPID AND
PROTEIN OXIDATION
Introduction
Oxidative processes inherently occur during handling and/or storage of muscle foods.
Muscle foods are particularly prone to oxidative degradation due to proportionally high levels of
unsaturated fatty acids, metals (heme bound or free), and multitudes of other catalysts and/ or
propagators (80). The actual mechanisms by which oxidative processes change muscle food
proteins are not well understood.
Increasing interests in the acid/alkali aided isolation process warrant study of quality
changes in muscle tissue during the process and the final isolated protein. According to various
studies, protein oxidation is implicated in various protein functionality changes: gelation,
emulsification, viscosity, solubility, and hydration (78, 114, 184-186). Various physical and
chemical protein changes are attributed to the reaction of protein with reactive oxygen species
(ROS) as well as lipid and protein oxidation products, resulting in polymerization, degradation,
and formation of other protein complexes (102, 124, 125, 130, 133, 187-191).
The main intent of the acid/alkali aided isolation process is to obtain quality isolates from
underutilized fish sources, namely pelagic species, by-catch, and fish byproducts. The majority
of these underutilized sources have significant levels of dark muscle (red muscle). Dark muscle
is particularity difficult to work with due to its metabolic nature. Dark muscle is heavily reliant
on oxidative metabolism, and relative to white muscle, depends on higher concentrations of
heme proteins for increased blood flow, mitochondrial function, and lipids for sustained energy
(192). Although, much of these problematic constituents are removed during the isolation
process, residuals can pose difficulties in obtaining a high quality isolate (5). To date, studies
addressing the simultaneous effects of extreme pH and/or readjustment to neutral pH on protein
75
and lipid oxidation have not been conducted. The objectives of this study, using Spanish
mackerel as a model pelagic species, were to 1) investigate how the extent of lipid oxidation in
low or high pH- adjusted homogenates affects lipid oxidation development in the final Spanish
mackerel protein isolate, and 2) investigate how lipid oxidation in acidic/alkali pH adjusted
homogenates and in the final protein isolate relate to protein oxidation and functionality.
Information regarding overall functionality and quality of isolates obtained from the acid/alkali
aided separation is important for improvement and successful implementation of this process
industrially.
Materials and Methods
Raw Materials
Spanish mackerel was freshly caught during the fall season (September) in Cedar Key,
Florida. The freshly caught Spanish mackerel corresponds to Batch 2 (Materials and Methods,
Chapter 3). Fish were manually skinned, degutted, and filleted. One half of the fillets were
minced using a large scale grinder (The Hobart MFG. Co., Model 4532, Troy, Ohio). Fillets and
ground fish (100 gram quantities) were vacuum packed and stored at -80ºC until needed. The
initial average moisture and lipid content and pH were 77±1.76 %, 6.49±0.015, and and 2.65±
0.07, respectively (Table 3-1).
Chemicals
Tricholoracetic acid, sodium chloride, sodium tripoly phosphate chloroform (stabilized
with ethanol), methanol, ammonium thiocyanate, iron sulfate, barium chloride dihydtrate, 12 N
hydrochloric acid were obtained from Fischer Scientific (Fair Lawn,New Jersey). Thiobarbituric
acid, 1,1, 3,3 tetraethoxypropane, streptomycin sulfate SigmaMarker® molecular weight
standard, and EZ-Blue stain solution were obtained from Sigma Chemical Co., (St. Louis, MO).
Propyl gallate and (ethlenediaminetetraacetic acid 99%), EDTA, TRIS, Potassium Chloride
76
(KCL) were obtained from Acros Organics (New Jersey, USA). Pre-cast linear 4-20% Tris-HCL
gradient gel, Laemmli buffer, and B-mercaptoethanol, Power Npac 300 power supply were
obtained from Bio-Rad laboratories (Hercules, CA). The Coomassie Plus™- Better Bradford
Assay kit was obtained form Pierce (Rockford, Il). The Zentech PC Test -Protein Carbonyl
Enzyme Immuno- Assay Kit was obtained from Biocell Corporation. LTD.(New Zealand).
Storage studies and Preparation of Homogenate and Isolate
Frozen, minced Spanish mackerel (-80ºC) was thawed in vacuum bags under cold running
water and was immediately placed in zip-lock bags at 4ºC for 4 days. From preliminary studies,
storing freshly caught fish for four days would mimic industrial fish quality for the isolate
process. In addition, raw material storage would ideally allow for detectable measurement
changes in both lipid and protein quality. Homogenates were prepared by diluting the mince with
cold deionized water (1:9) and mixing (45 second at 4ºC, speed 3) using a conventional kitchen
blender equipped with a rheostat (Staco® Inc., Dayton, Ohio, 3p 1500B). The pH of the
homogenate was approximately 6.2; which is lower than homogenate prepared from fresh mince.
Homogenates were slowly adjusted to pH 3, pH 6.5 (control), or pH 11 through dropwise
addition of 2 N HCL or NaOH. After a specified holding time (1 to 8 hrs) at 4ºC, the
homogenates were adjusted to pH 5.5. The homogenates were centrifuged at 10,000 g for 20
minutes at 4ºC to collect the precipitated proteins. Protein isolates kept on ice during analysis
were assessed for moisture, pH, and protein content prior to analysis. Isolates were then held for
3 days in one gallon Ziplock™ polyethylene bags at 4ºC and were analyzed again.
Thiobarbituric Acid Reactive Substances (TBARS)
TBARS were assessed according to Lemon (193) as modified by Richards and Hultin
(43). Briefly, one gram of sample was homogenized with 6 mL of trichloracetic acid (TCA)
solution composed of 7.5% TCA, 0.1% propyl gallate, and 0.1% EDTA dissolved in deionized
77
water. The homogenate was filtered through Whatman No. 1 filter paper. The TCA extract (2
mL) was added to 2 mL of thiobarbituric acid (TBA) solution (0.23% dissolved in deionized
water). The TCA and TBA solutions were freshly prepared and heated if needed. The TCA/TBA
solution was heated for 40 minutes at 100ºC. The samples were centrifuged for 10 minutes (4ºC)
at 2,500 g and were read at 532 nm using a spectrophotometer (Agilent 8453 UV-Visible,
Agilent Technologies, Inc., Palo Alto, CA). A standard curve was constructed using 1,1, 3,3
tetraethoxypropane (193).
Protein Quantification and Characterization
Protein isolate concentrations were assessed using a modified Biuret method (194). The
Bradford method was used to quantify proteins levels, either in low concentration (<1 mg/mL) or
limited quantity (195) using the Coomassie Plus™- Better Bradford Assay kit. Protein
composition in isolates was analyzed with electrophoresis. SDS-PAGE (sodium dodecyl sulfate
polyacrylamide gel electrophoresis) (196). Protein concentrations were adjusted to 3 mg/ mL.
SDS-PAGE samples were prepared by combining 0.33 µL of protein sample, 0.66 µL of
Laemelli buffer, and 50 µL of B- mercaptoethanol (1mg/ mL protein concentration). Samples
were boiled for 5 minutes in 1.5 mL cryogenic tubes and were then cooled on ice. To elucidate
changes in disulfide bonds, electrophoresis samples were prepared without β-mercaptoethanol
using 50 µL of Laemmli buffer instead. Pre-cast linear gradient 4-20% acrylamide gels were run
for 1 h at 200 V; mA 120 with a Power Npac 300 power supply. Samples were applied to the
gels in 10 and 15 µL aliquots. Molecular weight standards (205,000- 6,500 Mw) were run in 5
and 15 uL quantities. Gels were set with 12% TCA for 1 h and were stained with EZ-Blue stain
solution (minimum of 1 hr with agitation). Gels were then destained with deionized water for 18
hrs with agitation, changing out the water numerous times.
78
Rheology
To compare gel quality among isolates, oscillatory rheology was conducted using an AR
2000-Advanced Rheometer (TA Instruments, New Castle, DE). Viscoelastic behavior and
thermal stability of the protein isolates were evaluated and characterized according to Kristinsson
and Demir (13). Prior to analysis, protein isolates were adjusted to an average moisture content
of 85% through addition of deionized water. Sodium tripolyphoshate (0.3%) and sodium
chloride (600 mM) were added to the moisture adjusted isolates according to Liang and Hultin
(197). Isolate additives were incorporated using a mortar and pestle. Samples were adjusted to
pH 7.2 through dropwise addition of 1 N NaOH. Freshly prepared gels were analyzed in
duplicate using an oscillating acrylic plate (40 mm). Approximately 20 grams of sample were
place on the acrylic plate (5°C). The gap between the head and acrylic plate was set to 500 µm.
After the gap was reached, access sample was carefully removed with a stainless steel spatula.
Two temperature ramps were used to measure gel forming ability. A heating ramp (5-80°C) and
a cooling ramp (80-5°C) at rates of 2°C/ min. Oscillation procedures were conducted using 0.01
maximum strain and an angular frequency 0.6284 rad/s. Changes in the storage modulus (G’) as
a function of temperature were analyzed. The storage modulus is indicative of elasticity (198).
Protein Oxidation
Protein carbonyl content was measured for the pH derived isolates using the Zentech PC
Test (Protein Carbonyl Enzyme Immuno-Assay Kit). Briefly, protein carbonyls are derivatized
with dinitrophylhyrazine (DNP). The DNP bound proteins bind to the Elisa plate. The plate
bound DNP is then reacted with an anti-DNP-biotin antiobody. A stretavidin-linked horseradish
peroxidase is then bound to the complex. The chromatin reagent (contains peroxide) is then
added. The addition of peroxide initiates 3,3’,5,5’-tetramethylbenzidine (TMB) oxidation and
thus a color reaction. The application of the stopping reagent (an acid) terminates the reaction.
79
The yellow chromophore is measured at 450 nm in a microplate reader (Molecular Device
Corporation, Sunnyvale, CA). A standard curve is constructed using standards provided in the
kit, concentration (nmol/g) against absorbance (450 nm). The inter-assay variation was within kit
protocol recommendations (<15%). Prior to analysis, all samples (0.5 g) were homogenized 1:1
with 20 mM Tris-HCL, pH 7.5, 0.5 M KCL buffer and a glass homogenizer. This buffer rather
than water was used to maintain pH and increase isolate solubility according to Kristinsson and
Hultin (199). Connective tissue and/or insoluble components were centrifuged out at 2,000 g
prior to analysis. A glass homogenizer versus a motorized tissue tearer was used to reduce
foaming. All samples were normalized to 1 mg/mL and dilutions were performed when
necessary using the Enzyme Immuno Assay (EIA) kit buffer. All samples and standards were run
in at least triplicate.
Analysis of Moisture and pH
Moisture content was assessed using a moisture balance (CSI Scientific Company, Inc.,
Fairfax, VA). The pH was measured using an electrode ( Thermo Sure-Flow electrode (Electron
Corp., Waltham, MA) and pH meter (Denver Instrument, model 220, Denver, CO). The
homogenate samples were directly measured for pH. Isolates samples were diluted with
deionized water (1:10) to measure pH.
Statistical Analysis
The SAS system was used to evaluate data. A two-way ANOVA model was used to
evaluate statistical significance in lipid oxidation (TBARS, PV), and protein oxidation, with time
and among pH treatments as factors. All analyses were performed in at least duplicate and all
experiments were replicated twice. Least square means were used for post hoc multiple
comparisons with Tukey adjustment controlling for Type I error.
80
Data were reported as mean ± standard deviation. The analysis of variance and least
significant difference tests were conducted to identify differences among means, while a Pearson
correlation test was conducted to determine the correlations among means. Statistical
significance was declared at p < 0.05.
Results
Initially, to ensure that the quality of fish was approximately equivalent to studies in
Chapter 3, Spanish mackerel batch 2 mince was stored at 4°C for four days. Homogenates were
subjected to pH 3 as per findings in Chapter 3, in which the greatest overall levels of homogenate
lipid oxidation with respect to pH occurred at pH 3 and 3.5. Because the overall levels of lipid
oxidation with respect to homogenates 3 and 3.5 were insignificant from each other, and the fact
that pH 3 is an intermediate of pH values (2.5, 3, and 3.5) tested in Chapter 3, pH 3 was chosen
as the primary acid treatment. Likewise, no significant differences in lipid oxidation were
observed among homogenates at alkaline pH of 10.5, 11, or 11.5, so the intermediate pH 11 was
chosen as a treatment for this study. This pH is commonly used in the alkaline process. Prior to
experimentation, moisture content and pH of the starting homogenate were 97.6±0.707% and
6.215±0.077.
Oxidative Stability of Acid/Alkali Derived Protein Isolates
Levels of lipid oxidation were assessed throughout the isolation process using one and
eight hour exposure times at either pH 3 or pH 11. The isolates were obtained through
centrifuation at pH 5.5. The isoelectric point of fish proteins is close to pH 5.5, enabling for
lowest repulsion and highest yield of extracted proteins (mainly of myofibrillar nature)(17, 200,
201). Isolates were assessed for oxidation right after isolation and after 3 days of 4°C storage.
Average moisture and protein contents of acid, alkaline, and control (pH 6.5) isolates from one
and 8 hour hold times were similar and are summarized in Table 4-1.
81
The first centrifugation step (Figure 2-2) was omitted in this experiment to increase protein
isolate yield and retain more lipids (13). Kristinson and Demir (13) observed a gel-like sediment
in the soluble phase following the first centrifugation of acidified mullet. Similarly, a gel-like
sediment was observed after readjusting Spanish mackerel homogenates from extreme and
neutral pH to pH 5.5 (Figure 4-1). This phase was grayish white, soft, slimy, and easily
disrupted. Unlike the sediment observed by Kristinsson and Demir (13), the gel like sediment
was found above and adjacent to the collected proteins. Despite differences in processing, this
sediment was observed. This sediment, mainly neutral lipids, was removed with a plastic spatula.
When comparing the two processes, it is evident that more lipids are retained as exhibited by the
Spanish mackerel sediment layer when the first centrifugation step is excluded.
Holding times (1 or 8 h) did not significantly affect lipid oxidation in homogenates (See
Figures 4-2,3,4). The supernatant from the isolation of homogenate at pH 3 for eight hours had
the greatest level of TBARS (µmol MDA/kg of tissue) compared to all other pH homogenates,
isolates, and supernatants at pH 3, 6.5, and 11 (Figure 4-3).The levels of TBARS in the
supernatant (Figure 4-3) held at pH 3 for 8 h (94) were significantly (p<0.001) greater compared
to the corresponding isolates, or 52.59± 5.48 and 31.17±6.97 umol MDA/kg tissue, respectively.
The level of lipid oxidation in the pH 3 isolate after 1 hour storage was significantly (p <0.03)
less than the corresponding supernatant. The amount of oxidation measured in the pH 11 derived
isolate, after an 8 hour holding time, was significantly less than the same isolate stored for three
days (p<0.0291) (Figure 4-3). Levels of oxidation in pH 11 derived isolate (Figure 4-4), after 8
hours of holding, were significantly lower than the corresponding isolate at pH 6.5 (p<0.0289).
TBARS levels measured in the pH 6.5 supernatant, after an 8 hour hold, were significantly
greater than TBARS levels in isolate made after an 8 hour hold at pH 11 (p<0.0053).
82
Significantly (p<0.0068) higher TBARS formed in the homogenate exposed to pH 3 for one hour
compared to the pH 11 isolate, derived from an 8 hour homogenate holding time (Figure 4-3).
As shown in Figure 4-3 and 4-4, isolate from the pH 11 treatment obtained after 8 hours of
holding had significantly greater levels of oxidation after 3 days of storage than the
corresponding freshly prepared isolate (p <0.03).
Levels of protein oxidation were assessed among the isolates after 1 hour, 8 hours, and 8
hours followed by three days of storage (See Figure 4-5). Overall levels of oxidation were
significantly different (p<0.0001) with respect to pH treatment: pH 11> pH 6.5> pH 3 and
isolate: 1 hr> 8hr> 8hrs+ 3days. Specifically, isolate from the pH 11 treatment (1 h holding)
contained significantly higher levels of protein oxidation compared to all other combinations of
isolate and pH, with exception of these isolates from the 6.5 treatment (1 h holding) and the pH
11 treatment (8 hr holding) (Figure 4-5). The isolate produced from holding the 6.5 homogenate
for one hour, was significantly higher in carbonyls than the corresponding isolate at pH 11 (p<
0.0031). The isolate formed from a holding time of 8 hours at pH 11 had significantly
(p<0.0001) greater levels of protein oxidation than the corresponding pH 11 and pH 6.5 isolates
(p <0.0006) formed from a one hour holding time. The pH 11 isolate derived from an 8 hour
holding time was more oxidized than the corresponding pH 6.5 isolate (Figure 4-5). The pH 6.5
isolate carbonyl levels from the 8 h holding treatment were significantly less than carbonyls after
1 hour of holding. After the isolates from the 8 hr holding treatment were stored for three days, a
significant decrease (p<0.03) in protein carbonyls compared to the fresh 8 hour isolates was
observed among pH 11and 6.5 treatments (Figure 4-5). Mean values were comparable to an
average carbonyl content of 1.5 nmol/mg of protein by Liu and Xiong (189) in red and white
muscle of chicken.
83
Gel Forming Ability of Isolates
In terms of rheological differences, the storage modulus (G’) of isolates was not
significantly affected by holding time or storage; however, the elasticity (G’) of isolates as
significantly (p≤ 0.0002) affected by pH treatment with pH 11 isolate having the highest G’ (Pa),
followed by pH 6.5, and then pH 3 (Figures 4-6).
Examples of rheograms generated during the oscillatory testing of isolate pastes derived
from 1 hour, and isolates derived from 8 hr pH holding times in addition to three days of
refrigerated storage are shown in Figures 4-7,8,9. Initial G’ values were similar in the pH 11 and
6.5 isolates. These initial G’ values were greater than G’ for pH 3 isolates. Initial increase in the
storage modulus (G’) on heating were observed at 30°C for all samples at pH 3.0, 6.5 (control),
and pH 11 (Figures 4-7,8,9). During heating, one noticeable transition temperature (Tm) was
observed at 40°C for the isolates at pH 6.5 (control) (Figure 4-9). The same Tm, although
significantly less noticeable, was also observed for pH 3.0 and pH 11 isolates (Figure 4-7,8).
Significant increases in G’ were observed between 50 and 60°C for isolates derived from pH 3
and 6.5 homogenates and between 40 and 55°C for isolates derived from pH 11 homogenates
(Figures 4-7,8,9). G’ values were stabilized between 70 an 80°C. Significant increases in G’
during the cooling step were observed in all isolates. During cooling, the greatest increase in G’
occurred in the pH 11 isolates, followed by pH 6.5 (control), and pH 3.
Protein Composition of Homogenates and Recovered Isolate Fractions
SDS Page, with and without, mercaptoethanol showed differences in protein composition
among isolate fractions. The inclusion versus the exclusion of mercaptoethanol in samples was
applied to the gels to assess whether differences among protein fractions, if any, were due to
covalent (S-S) bonds or non-covalent cross-linking, possibly attributed to protein oxidation.
Cross-linking was not observed in any of the samples; however, hydrolysis was observed for all
84
pH (11,6.5, 3). Hydrolysis was evident in isolates derived from adjusting and holding
homogenates at acidic (3), alkaline (11) or control (6.5) pH for 1 hour, followed by readjustment
to pH 5.5 in Figure 4-10: Lanes 2-5 (pH 11); Lanes 10 and 11 (pH 6.5); Lanes 14 and 15 (pH 3).
Hydrolysis was also observed in isolates derived from adjusting and holding homogenates at
acidic (3), alkaline (11) or control (6.5) pH for 8 hour, followed by readjustment to pH 5.5 in
Figure 4-11: Lanes 4 and 5 (pH 6.5); Lanes 8 and 9 (pH 3); and Lanes 13 and 14 (pH 11).
Proteolysis was evident in isolates derived from adjusting and holding homogenates at acidic (3),
alkaline (11) or control (6.5) pH for 8 hours, followed by readjustment to pH 5.5, and finally
followed by 3 days of refrigerated storage in Figure 4-12: Lanes 4 and 5 (pH 6.5), Lanes 6 and 7
(pH 11); Lanes 14 and 15 (pH 3). Hydrolysis was also evident in pH 3, 11, and (6.5) control
homogenates at initial pH adjustment (Figure 4-13) and 1 and 8 h holding (Figures 4-14,15), at
pH 3, 6.5 (control), and 11.
The most prevalent proteins fractions were myosin heavy chain (MHC)(~205 Kda) and
actin ( ~ 43kDa), and tropomyosin (35.5 kDa). Other identified proteins included desmin (~56
kDa), troponin T (41 kDa), and topomyosin-beta (~38 kDa).
Discussion
The isolation of fish proteins following pH readjustment from extreme pH is performed for
greater protein recovery and to recover proteins with modified functional properties. With
greater protein recovery, co-extraction of unwanted components (lipids and heme proteins) and
consequentially oxidative degradation could occur, negatively affecting the quality and
performance of the isolated proteins.
Changes in pH of Mince with Storage
The pH (6.2) of mince stored for 4 days was less than initial pH values of unstored mince
homogenate (6.45) as reported in Chapter 3, indicative of metabolite accumulation and cellular
85
degradation post mortem (31). Various postmortem changes can influence oxidative occurrence
within the fish muscle: decreases in ATP and increases in ATP byproducts, loss of reducing
compounds (ascorbate, NAD(P)H, glutathione, etc.), increased levels of low molecular weight
transition metals (copper, iron), oxidation of iron in heme proteins, disruption of cellular
structure, loss of antioxidants, and accumulation of calcium and other ions. With mincing,
postmortem muscle tissue can experience conditions very similar to ischemia-reperfusion.
Consequentially, production of superoxide and hydrogen peroxide can occur, possibly leading to
protein and lipid oxidation (5).
Lipid Oxidation During Extreme pH Adjustment and Readjustment to 5.5
It was observed that through separation of Spanish mackerel proteins by pH readjustment
from extreme pH (3, 6.5, 11) to neutral pH (6.5) that lipids were retained (Figure 4-1).
Kristinsson and Demir (13) reported greater lipid levels in acid mullet isolates when the first
centrifugation step was skipped. This could be explained by the co-precipitation of membrane
and neutral lipids with proteins at isolelectric pH. Co-precipitation is most likely a result of
increased protein hydrophobicity and surface activity with pH adjustment as observed by
Kristinsson and Hultin (18) with cod myosin. Increases in hydrophobicity result in enhanced
emulsification ability and interaction with hydrophobic residues of lipids and heme proteins.
Increases in hydrophobicity is most likely attributed to denaturation of myofibrillar and
sarcoplasmic proteins, as supported by low extractability of rockfish proteins during the
acid/alkali aided process (202). Kristinsson and Hultin (199) have shown that significant
structural changes occur in myosin upon incomplete refolding of cod myosin with the
readjustment to pH 7.5 from acid and alkaline treatment. Partial refolding of myosin’s tertiary
structure could facilitate hydrophobic interactions. Furthermore, low levels of retained
hemoglobin upon aggregation can pose lipid oxidation problems. Undeland et al. (67) found that
86
extensive hemoglobin chelation by 0.2% STPP and 0.2% EDTA during acid isolation of herring
still gave rise to low TBARS levels. These results emphasize that despite significant removal of
hemoglobin in the pH cycle, hemoglobin could still be problematic regardless of concentration
found in the isolate.
The greatest level of TBARS were found in the supernatant from the isolation of pH 3
Spanish mackerel homogenate held for eight hours relative to all other pH homogenates, isolates,
and supernatants at pH 3, 6.5, and 11 (Figure 4-2 and 4-4). Additionally, the pH 3 supernatant,
after 1 h of holding, was greater than the corresponding isolate (Figure 4-2). According to
Undeland (16), lipids were present in the highest amounts in the emulsion (supernatant) layer
followed by the gel sediment in the acidic pH (2.7) extraction of herring. Lower lipid levels were
observed from alkaline extraction (10.8) in which the emulsion layer had 37% lipid and the
isolate had 18% lipid (16). Just as Undeland (67) reported, the Spanish mackerel supernatant
fraction at acidified pH (3) contained the highest TBARS levels (Figure 4-3). During the
conventional production of horse mackerel surimi performed by Eymard et al. (138), significant
TBARS were found in the wash water. Lipid oxidation products formed during acid processing
could be water soluble or at least suspendable, which partially explaines why the supernatant
fractions (Figures 4-2 and 4-3) had such high TBARS.
Overall, lipid oxidation in the pH treated isolates were not significantly different (Figure
4-4). Lack of significant differentiation in TBARS between alkali and acid processing compared
to other published studies is attributed to skipping the first centrifugation step. Kristinsson and
Demir (13) observed significant differences among color values when incorporating the first
centrifugation in the separation of mullet and croaker proteins; however, in absence of
centrifugation both acid and alkaline isolates were negatively affected in color value, suggesting
87
more retention of heme proteins. With both centrifugation steps, the alkaline treatment was
whiter in color (high L value) and was less yellow (low b value) than acid isolate. This was
attributed to increased removal of heme proteins, found in the supernatant of the first
centrifugation step (13). Kristinsson and Hultin (51) concluded that alkaline treatment resulted in
greater hemoglobin solubility and reduced co-precipitation at pH 5.5 as per discussion of
homogenate pH adjustment in Chapter 3. From the lipid oxidation studies, it appears that acidic
and alkaline pH processing of Spanish mackerel with exclusion of the first centrifugation is
equally problematic.
Holding times did not significantly affect the amount of oxidation among Spanish
mackerel isolates within treatments (Figures 4-2,3,4). According to Undeland et al. (203), the
holding time at acidic pH (4, 30, or 70 minutes) prior to acidic isolation of herring isolates did
not affect overall levels of oxidation. In another study, Kristinsson (51) found that the
prooxidative activity of trout hemoglobin at pH 3.0 and 3.5 versus pH 2.5 was unaffected by
holding time prior to pH readjustment. Kristinsson et al. (204) reported no significance
difference in lipid oxidation among the alkaline or acidic pH isolation of catfish proteins;
however, when the first centrifugation was skipped, significantly more lipid oxidation resulted
relative to the other treatments. The lack of significance among the acid and alkali treatment
concurs with some previous findings. The significant difference between alkaline and control pH
isolates, after 8 hours of holding, is interesting (Figure 4-4). Kristinsson and Liang (205),
reported reduced levels of lipid oxidation in an alkaline isolate relative to a surimi control and
acid isolates from Atlantic croaker. As suggested by Kristinsson (12), this could be a result of
hemoglobin adopting a more stable tertiary structure upon alkaline treatment compared to acid
treatment (51). Additionally, oxidative products were most likely removed through
88
centrifugation at 5.5, impart explaining a reduction in lipid oxidation product in the alkaline
isolate made an after 8 h holding (Figure 4-3).
Lipid oxidation was observed among all isolate treatments (Figure 4-4). Similarly, the
greatest increase in lipid oxidation during 7 days of storage of beef heart surimi, washed with
buffer at 5.5 and 7.0, was observed in the 5.5 treatment by Srinivasan et al. (136). It would be
interesting to investigate the effects of readjustment to pH 7, as previous studies suggest it could
pose less lipid oxidation problems (190); however, reduced protein recovery would most likely
result. With the final protein isolate at pH 5.5, it is to be expected that lipid oxidation would
proceed significantly faster than at pH 7 as hemoglobin is more pro-oxidative at pH below
neutrality as observed in a washed cod model systems (43, 51, 206). Lipid oxidation in isolates,
after storage at low pH (5.5) was evident. This can be explained by rapid autoxidation of heme
proteins in the isolates at pH 5.5. According to Richards and Hultin (43), increases in TBARS
were due to decreases in pH from 7.6 to 6.0 in a hemoglobin mediated washed cod system.
Protein Oxidation
Overall levels of protein oxidation (Figure 4-5) were highest at pH 11, followed by pH 6.5,
and pH 3. The effect of pH on lipid and protein oxidation has been studied in various meat
systems. Similar to observed pH affects in this study, Xiong et al. (124) and Wan et al. (123)
have shown that myofibrils from bovine cardiac muscle improved in functionality with decreases
in lipid oxidation. In a study conducted by Srinivasan et al. (190), beef heart was washed with
various pH buffers, ranging from pH 7.0 to 5.5. After seven days of storage, the levels of lipid
and protein oxidation were significantly higher at pH 5.5 and 6.0 than at pH 7.0. Srinivasan (136)
reported a simultaneous increase in markers of lipid (TBARS, conjugated dienes) and protein
oxidation (carbonyls) with lowering of pH in bovine cardiac surimi. Kenney et al. (207) found
that washed beef heart surimi had superior textural properties compared to fish surimi. Similar to
89
dark muscled fish, beef heart contains an abundance of heme proteins, iron, and polyunsaturated
fatty acids (208).
The mince had significantly lower levels of protein oxidation compared to all isolation
treatments at pH 3, 6.5, and 11 (See Figure 4-5). Hultin and Luo (209) reported that protein
modifications will occur with changes in muscle temperature, salt concentration, or pH, leading
to significant degradation and polymerization of proteins due to protein oxidation (209). This
phenomenon could explain why the mince had lower levels of oxidation, as the isolates were
obtained through pH processing. Interestingly, carbonyl levels were higher in the isolates made
after one hour holding at pH 6.5 and 11 compared to pH 3. According to Srinivasan et al. (136),
bovine heart surimi washed with antioxidants above pH 5.7 produced stronger gels than surimi
washed below pH 5.7. The differences in G’ at various washing pH was attributed to oxidative
modification of protein residues and or protein-lipid oxidation interactions, affecting the
antioxidant efficacy. The observations by Srinivasan et al. (136) could explain the differences in
the protein oxidation in Spanish mackerel isolates with pH, as different protein interactions occur
in the presence of oxidized lipids or proteins, more so at high pH. Kristinsson and Hultin (199)
reported partial refolding of cod myosin with pH 7.0 readjustment from alkaline pH. Partial
refolding at high pH could make alkaline treated isolates more susceptible to protein oxidation.
In the pH 11 Spanish mackerel isolates, protein carbonyl content significantly increased
after 8 hours of holding compared to one hour of holding (Figure 4-5); while there was an
apparent decrease in TBARS after 8 hour hold compared to the 1 hour hold. Similar trends
between protein and lipid oxidation were observed by Srinivasan and Hultin (210). In this study,
significant increases in protein and lipid oxidation were observed when washed cod muscle was
subjected to free radical exposure at pH 6.8 for 2 hours. Interesting kinetics between protein and
90
lipid oxidation were observed in this system. With storage, protein oxidation (carbonyl) levels
increased quickly, while there was a lag phase in lipid oxidation. Lipid oxidation began to
accelerate at a point where protein oxidation declined slightly. Protein oxidation levels then
experienced another rapid increase and lipid oxidation levels also increased. Hultin (210)
concluded that perhaps protein or lipid oxidation could propagate each other. Similarly, longer
pH holding times may afford greater production of ROS through lipid oxidation, providing
adequate initiation of protein oxidation (211).
Significant decreases in protein oxidation after 3 days of storage occurred among the pH
11 and 6.5 isolates compared to the freshly prepared isolate (Figures 4-5). This is in contrast to
other storage studies, in which, protein and lipid oxidation increased over a duration equivalent
to 7 days or more (63, 70, 136). It is possible that oxidized, aggregated proteins or insoluble
proteins could not be adequately assessed via the carbonyl assay, as solubilized proteins are
applied to the protein carbonyl kit. With storage, protein degradation was observed in the gels.
Protein aggregation may have possibly occured in the isolates as protein bands were observed at
the top of the SDS-PAGE gels. It is also possible that protein oxidative products served as
intermediary reactive species during the oxidative process, becoming pro or antioxidive with
changes in pH and conformation environment (212, 213). Protein oxidation products could have
served as antioxidants for the alkaline pH and control pH homogenate treatments and as oxidants
in the pH 11 derived isolate, stored at pH 5.5. It has been proposed that protein radical species
are highly dynamic and even transient, depending upon interaction with other molecules and
environmental conditions (213). In a BSA hydrogen peroxide activated protein oxidation model
study conducted by Ostdal et al. (211), BSA successfully oxidized urate, linoleic acid and a
91
complex amino acid residue. Thus, protein radicals could play various roles in the oxidation of
muscle foods either as end-products or intermediates.
Carbonyl measures are readily performed and acceptable measures of complex-muscle
system samples (112, 210). Carbonyls are a global form of oxidation and can be formed as result
of various reactive species (76), unlike other protein oxidation products. Strict reliance on
carbonyl levels as an indicator of protein oxidation is tenuous, as various other compounds form
as result of protein oxidation. A more comprehensive attempt to measure protein oxidation could
include the measure of dinitrotyrosine, 3- dinitrotyrosine, and modified AA residues (214).
Protein carbonyl content in isolates would be best analyzed by a method requiring chemical
digestion. Chemical digestion would eliminate solubility problems and possible underestimation
of protein oxidation levels, upon analysis and/or plating. Kjaersgard et al. (215) have
successfully used two-dimensional immunoassaying followed by mass spectroscopy in the
identification and quantification of oxidized proteins and residues in fish species.
Gel Forming Ability
Noticeable differences in rheograms among the isolates derived from pH 3, 6.5 (control),
and pH 11 homogenates were observed. All isolate pastes showed initial increases in G’ at 30°C.
Yougsawatdigul and Park (200) also observed initial increases in surimi as well as alkaline and
acid isolates at 34°C (200). Unlike the acid (pH 3) and alkaline (pH 11) isolates (Figures 4-7,8),
a significant decrease in G’ was observed during the heating step in the pH 6.5 (control) isolates.
The lowest G’ was observed at approximately 47°C for isolates derived from pH 6.5
homogenates (Figure 4-9). Yongsawatdgul and Park (200), Kristinsson and Liang (205), and
Kristinsson and Hultin (18) observed similar rheology differences among surimi and pH treated
samples of rock fish, Atlantic croaker, and cod muscle proteins, respectively. The Tm (40°C)
observed in the control (6.5) isolate was not observed in the pH treated isolates. Similarly, Hultin
92
and Kristinsson (18) reported absence of this peak in pH treated cod proteins compared to
untreated cod proteins, and this observation was attributed to different protein-protein
interactions among the control and pH treated proteins (18). Ingadottir (216) and Theordore
(217) observed significant rheological differences among surimi and pH treated samples for
tilapia and catfish, respectively. These rheological differences between controls and acid/alkaliaided isolates may be due to protein denaturation with pH treatment (218). In agreement with
findings by Kristinsson and Hultin (18) and Kristinsson and Liang (205) for washed rockfish and
croakers, a large drop in storage modulus occurred in pH 6.5 (control) isolate samples.
According to Kristinsson and Liang (205), a large decrease in G’ could be attributed to
actyomyosin gel formation. Non-covalent protein interactions between actin and myosin are due
to various conformational changes. The Tm observed at 40°C in the pH 6.5 isolates marks the
beginning of gel network formation through various conformational changes including myosin
light chain dissociation from myosin heavy chains (MHC) (219, 220), exposure of hydrophobic
groups, namely tryptophan, and transformation of the heavy myosin rod (tail) from a α- helix to a
random coil. G’ significantly increased on cooling among all isolate samples.
The pH treatment had a significant affect on gel strength, as alkaline pH gave stronger gels
than the control (6.5) and the acid (pH 3) prepared samples (Figure 4-6). This is in disagreement
with previous findings by Kristinsson and Liang, in which croaker alkaline and acid isolates had
greater G’ than surimi. Kristinsson and Hultin (18) reported similar G’ values among acid (pH
2.5), alkaline (ph 11), and untreated (pH 7.5) cod muscle proteins; however, different
mechanisms of gelation among the treatments were observed. Shikha et al (221) studied the
effect of acid pH shifting in walley pollack. Surmi that was acidified and then readjusted to
neutral pH had lower gel strength (g/cm2) than the control (pH 7.2 mince). Various other studies
93
reported difference in G’ values among alkaline, acid, and untreated samples for catfish, tilapia,
mullet, and Spanish mackerel (13, 216, 218). Theodore (217) reported G’ trends comparable to
the Spanish mackerel isolate samples, in which the alkali treated catfish samples were followed
by surimi and then the acid samples in G’ values. In agreement with this study, Kristinsson and
Demir (13) found that alkaline processing of Spanish Mackerel and mullet gave greater G’
values than surimi processing, followed by acid processing. Higher G’ in the alkaline samples
indicates greater protein-protein interactions on cooling, enabling better gel-network formation
upon setting. According to Kristinsson and Hultin (199), proteins can unfold through extreme
pH and upon readjustment only partially refold. Specific changes in protein structure with
unfolding and refolding could explain the differences in G’ among the isolates and control (pH
6.5) treatment. At low and high pH, proteins unfold and denature. Kristinsson and Hultin (199)
reported dissociation of the cod myosin head group with unfolding at extreme pH. With
readjustment to pH 7.5, the alkaline and acid treated myosin partially refolded; however, alkaline
treated cod myosin remained more dissociated. Due to more flexibility than acid isolates upon
refolding, alkaline treated proteins could form a better gel network. Increased exposure of thiol
groups could also explain the difference in G’ among the alkaline (pH 11), control (pH 6.5), and
acid isolates (pH 2.5). Kristinsson and Hultin (199) found an increases in thiol exposure when
cod myosin was adjusted to alkaline pH and readjusted to pH 7.5. This increase in sulfhydryl
group (-SH) reactivity could explain the higher G’ in the Spanish mackerel alkaline isolates
versus the control (pH 6.5) and acid isolates (Figure 4-6). The acid isolates did show an increase
in sulfhydryl reactivity with refolding according to Kristinsson and Hultin (199), but this
increase was less compared to alkaline refolding. A lower G’ in the acid versus alkaline
processing could be due to less thiol interaction among proteins.
94
Textural Changes and Protein and Lipid Oxidation
Greater overall levels of protein oxidation (Figure 4-5) in the alkaline treatments could
explain high G’ (Figure 4-6). Protein oxidation could lead to textural changes in muscle food
products (86), and could explain the differences in gel strength among isolates. As a consequence
of oxidation, proteins unfold and exposes their hydrophobic residues. Srinivasan and Xiong
(130, 136) reported extensive sulfhydryl loss with protein oxidation. This increase in
hydrophobic affinity could promote protein interaction and aggregation, improving gel strength.
Protein oxidation proceeds in a manner comparable to lipid oxidation in the presence of
oxidizing lipids and/or other reactive oxygen species (ROS) (101, 222). Free radicals will react
with either amino acid (21) side chains and/or peptide backbones. Proteins will often fragment
and complex upon free radical attack (86). Hydroxyl groups are predominantly responsible for
protein oxidation in the absence of lipids (106). Certain AA groups are preferentially oxidized,
the most susceptible is cysteine. Myosin, the most abundant and important muscle food protein,
has 40 free thiol groups. (223). The abundance of myosin compared to sarcoplasmic proteins
(224) warrants investigation of how myosin is affected during the acid/alkali aided process (13).
Protein carbonyls are product of: AA residue oxidation, peptide backbone degradation, proteinreducing sugar interaction, and interaction with non-protein carbonyl groups (86). Oxidized
proteins will nucleophilically attack proximal proteins, forming cross-linkages. According to
Xiong and Decker (87), protein cross-linkages result in protein polymerization, aggregation, and
insolubility. Cross-linking may cause protein aggregation and increased G’ values (70). The
quality of muscle foods is greatly impacted by these physiochemical protein changes.
Intermolecular interactions may be weak with protein oxidation; however, oxidatively induced
polymerization may counteract these weak bonds and strengthen non-covalent interactions,
through increased rigidity and low mobility within the gel network.
95
The duration of storage for the Spanish mackerel isolates was short and the development of
carbonyls did not exceed 2 nmol/mg (Figure 4-5). According to Lee et al. (104) and Nishimura
et al. (225) there is a very high, positive correlation (r=0.99) in beef heart surimi, between
protein carbonyls and G’ when there is sufficient protein oxidation and storage. It was observed
that increases in G’ occurred with increases in protein carbonyls to at least (136, 190, 191).
According to other studies (136, 190, 191), protein oxidation improved gel forming ability in
muscle tissue through greater gel network formation. In contrast to these studies, the level of
oxidation was significantly less 15 nmol/mg protein. With increased storage, a greater correlation
between G’ and protein oxidation as well as lipid oxidation may have been observed.
Lipid oxidation in the alkaline isolate after 8 h holding was significantly less than with 1
hour holding (Figure 4-4). Protein oxidation levels also significantly increased after 8 hr of
holding compared to 1 hr of holding (Figure 4-5). A possible correlation between protein
oxidation and lipid oxidation and G’ in the isolates exists. As per discussion in previous “Protein
oxidation” section, protein oxidation could be propagated by lipid oxidation. According to Saeed
and Howell (70), the elastic modulus in Atlantic Mackerel fillets stored at -20ºC and -30ºC was
significantly decreased with the incorporation of vitamin E (500 ppm). Because the transfer of
free radicals or protein-lipid interactions could proceed from lipid oxidation (70), it was inferred
that vitamin E quenched lipid radicals and reduced protein oxidation formation. This study
elucidates that lipid oxidation is an initiator of protein oxidation.
Protein oxidation did not proportionally increase with lipid oxidation (Figure 4-4, 4-5) in
the isolates. This is in contrast to other studies in which increases in protein oxidation are
proportional to increases in lipid oxidation. According to Smith (122), decreases in myofibrillar
gel strength were observed in deboned turkey with increases in TBARS, and decreases in
96
solubility and myosin ATPase activity (122). The authors attributed the decrease in solubility and
ATPase activity to protein oxidation. The oxidative effects of the acid/alkali aided process on
lipid and proteins has received little attention and deserves further investigation.
SDS page
Electrophoretic patterns (Figures 4:7-12) revealed hydrolysis in alkaline and acid samples.
Hydrolysis is a concern as it could negatively affect gel strength due to gel softening (226, 227).
Several studies suggest that proteolysis during pH processing can result in poorer gels.
According to various studies, the alkaline/acid-aided process can initiate protease acitivity (16,
201, 228). Enzymes can be activated at low pH and at the isoelectric pH 5.5 (201, 229). Similar
to this study, hydrolysis was observed in acid/alkali aided samples by Kristinsson and Demir
(13), Choi and Park (16, 201), Undeland et al. (36) and Ingadottir (216) in catfish, spanish
mackerel, mullet, croaker, pacific whiting, herring, and tilapia. A band of approximately 150
KDA was observed in the gels (4-10,11,12, 13, 14, 15), similar to other studies analyzing the
affects of pH processing on fish proteins (205, 230-232). Myosin heavy chain (MHC)
degradation is indicated by this band (230). The observed hydrolysis among the Spanish
mackerel isolates is most likely attributed to protease activation versus alkaline or acid
adjustment as the control (pH 6.5) also contained hydrolysis. Because proteolytic hydrolysis was
observed among all treatments and it did not negatively affect the gel strength of either alkaline
or control treatments, hydrolysis could not explain the differences in gel strength in the alkaline,
acid, or neutral isolates (Figure 4-6). SDS page results do therefore suggest that the differences
in gel strength are impart due to changes in comformation with pH treatment, or these are
chemically induced changes linked to oxidation. With storage, protein aggregation or
degradation was not observed in the gels; however, it is possible that aggregates were too large
to enter the gel.
97
Conclusion
Protein oxidation and lipid oxidation were significantly affected by holding time and pH.
The pH 11 isolate contained the highest overall levels of protein oxidation, followed by pH 6.5,
and pH 3. Overall, levels of lipid oxidation were similar among the pH treatments, and
significant removal of lipid oxidation occurred with centrifugation at pH 3, after an 8 h holding.
A positive correlation between lipid oxidation and protein oxidation was not apparent, as
observed in other studies. G’ was consistently higher in alkaline isolates compared to pH 6.5 and
pH 3. Increases in G’ may be attributed to conformational changes at higher pH (6.5 and 11); and
or protein modifications due to protein oxidation.
As research currently exists, positive correlations between lipid and protein oxidation and
gel forming ability can only be tentatively made. The study of protein oxidation in muscle foods
has only just begun and further findings are necessary to substantiate the influence of lipid and
protein oxidation on protein physiochemistry and functionality.
98
Table 4-1. Moisture content (%) and protein content (%) of isolates obtained after readjustment
from pH 3, 11, or 6.5 to pH 5.5, and held for 1 and 8 hours. Moisture and protein
content means and standard deviations were calculated from triplicate analysis.
Isolate 1 (1 hour hold)
Isolate 2 (8 hour hold)
Control
Acid
Alkaline
Control
Acid
Alkaline
(pH 6.5→5.5)
(pH 3→5.5)
(pH 11→5.5)
(pH6.5→5.5)
(pH 3→5.5)
(pH 11→5.5)
Moisture Content
(%)
81.3 ±0.353
82.5 ±1.41
81.5 ±0.707
80.8 ±0.354
80.05 ±1.20
81.3 ±0.707
Protein Content
(%)
10.9± 2.02
12.3±0.198
12.29±0.505
10.4±2.52
12.3±0.500
11.8±0.940
Supernatant: neutral lipids, saturated fatty
acids, and (sarcoplasmic) proteins
Sediment: Connective tissue,
membrane and neutral lipids
Protein Isolate: primarily myofibrillar
proteins, heme proteins, and other
precipitated/ aggregated proteins
Figure 4-1. Schematic of the observed centrifugation layers with direct readjustment from
extreme pH (3 and 11) and neutral pH (6.5) to pH 5.5. Centrifugation prior to
readjustment was not employed. This sediment layer was skimmed off with a plastic
spatula.
99
TBARS (umol MDA/kg tissue)
70
pH 3
60
pH 6.5
50
pH 11
a
40
abc
30
20
ab
abc abc
abc
bc
abc
abc
abc
abc c
10
0
intial homogenate
homogenate after
one hour
isolate
supernatant
Isolation Treatments
Figure 4-2. Lipid oxidation (TBARS) during the production of acid (pH 3), alkaline (pH 11) and
“control” (pH 6.5) derived isolates. Homogenates were exposed to the given pH
values for one hour prior to obtaining an isolate and corresponding isolates. Each bar
in the graph is representative of TBARS (umol MDA/kg of tissue) with standard
deviations. Different small letters indicate a significant difference.
TBARS (umol MDA/kg tissue)
70
pH 3
60
a
pH 6.5
50
pH 11
40
abc
30
20
bc
bc
b
b
bc
b
bc
bc
bc
bc
b b
c
10
0
intial
homogenate
homogenate
after 8 hours
isolate
supernatant
8 hr + 3 days
Isolation Treatments
Figure 4-3. Lipid oxidation (TBARS) during the product of acid (pH 3), alkaline (pH 11) and
“control” (pH 6.5) derived isolates. Homogenates were exposed to the given pH
values for 8 hours prior to obtaining and an isolate and corresponding isolates. Each
bar in the graph is representative of TBARS (umol MDA/kg of tissue) with standard
deviations. Different small letters indicate a significant difference.
100
TBARS (umol MDA/kg tissue)
70
60
50
40
pH 3
pH 6.5
pH 11
a
4 day storage
ab
30
ab ab ab
ab
a a
ab
20
b
10
0
0 hr
1 hr
8 hr
8 hr + 3 day
storage
Isolation Treatments
Figure 4-4. Lipid oxidation (TBARS) of acid (pH 3), alkaline (pH 11) and neutral (pH 6.5)
protein isolates derived from 1 and 8 hour pH exposure, analyzed right after
preparation and after three days of refrigerated storage. Each bar in the graph is
representative of TBARS (umol MDA/kg of tissue) with standard deviations. The
level of lipid oxidation in mince stored for four days (0 hr) is also shown. Different
small letters indicate a significant difference.
Protein Carbonyl (nmol/ mg protein)
1.8
1.6
1.4
1.2
pH 3
a
pH 6.5
b
pH 11
c
4 day storage
1
0.8
de
0.6
0.4
e
d
de e
de
f
0.2
0
0 hr
1 hr
8 hr
8 hr + 3 day
storage
Isolation Treatments
Figure 4-5. Levels of protein oxidation as indicated by protein carbonyls (nmol/mg protein) in
mince stored for four days (0 hr), isolates derived from 1 hour, and 8 hr pH holding
times in addition to 3 days of refrigerated storage. Different small letters indicate a
significant difference.
101
45000
35000
30000
G'(Pa)
ab
pH 3
pH 6.5
pH 11
40000
ab
b
a
b
ab
25000
20000
15000
10000
5000
c
c
c
0
1 hr
8 hr
Isolation Treatment Time
8 hr + 3 day storage
Figure 4-6. Storage Modulus (G’) expressed in terms of Pascal’s (Pa) for isolate pastes derived
from 1 hour, and isolates derived from 8 hr pH holding times in addition to three days
of refrigerated storage. Isolate pastes (85% moisture) were made through
incorporation of sodium tripolyphoshate (0.3%), sodium chloride (600 mM), and cold
deionized water. Samples were adjusted to a final moisture content and of 85% and
pH 7.2 Oscilation test were conducted using a heating ramp (5-80°C) and cooling
ramp (80-5°C) at rates of 2°C/ min at 1% strain and 0.6284 Hz. Different small letters
indicate a significant difference.
1000
1 hr
900
800
8 hr
700
8 hr + 3 day storage
G' (Pa)
600
Cooling
500
400
300
Heating
200
100
0
0
10
20
30
40
50
60
70
80
90
Temperature (°C)
Figure 4-7. Examples of rheograms from acid (pH 3) protein isolates derived from 1 and 8 hour
pH exposure, analyzed right after preparation and after three days refrigerated
storage. The rheograms show storage modulus (G’) using a heating ramp (5-80°C)
and cooling ramp (80-5°C) at rates of 2°C/ min at 1% strain and 0.6284 Hz. Isolates
pastes (85% moisture) were made through incorporation of sodium tripolyphoshate
(0.3%), sodium chloride (600 mM), and cold deionized. Samples were adjusted to a
final moisture content and of 85% and pH 7.2.
102
45000
1 hr
40000
8 hr
35000
8 hr + 3 day storage
G' (Pa)
30000
25000
Cooling
20000
15000
10000 Heating
5000
0
0
10
20
30
40
50
60
Temperature (°C)
70
80
90
Figure 4-8. Examples of rheograms from alkaline (pH 11) protein isolates derived from 1 and 8
hour pH exposure, analyzed right after preparation and after three days refrigerated
storage. The rheograms show storage modulus (G’) using a heating ramp (5-80°C)
and cooling ramp (80-5°C) at rates of 2°C/ min at 1% strain and 0.6284 Hz. Isolates
pastes (85% moisture) were made through incorporation of sodium tripolyphoshate
(0.3%), sodium chloride (600 mM), and cold deionized. Samples were adjusted to a
final moisture content and of 85% and pH 7.2.
35000
1h
30000
8h
G' (Pa)
25000
8 h +3 day storage
20000
Cooling
15000
10000 Heating
5000
0
0
10
20
30
40
50
60
Temperature (°C)
70
80
90
Figure 4-9. Examples of rheograms from control (pH 6.5) protein isolates derived from 1 and 8
hour pH exposure, analyzed right after preparation and after three days refrigerated
storage. The rheograms show storage modulus (G’) using a heating ramp (5-80°C)
and cooling ramp (80-5°C) at rates of 2°C/ min at 1% strain and 0.6284 Hz. Isolates
pastes (85% moisture) were made through incorporation of sodium tripolyphoshate
(0.3%), sodium chloride (600 mM), and cold deionized. Samples were adjusted to a
final moisture content and of 85% and pH 7.2.
103
205 kDa
116 kDa
97 kDa
84 kDa
66 kDa
55 kDa
45 kDa
36 kDa
29 kDa
24 kDa
20 kDa
14 kDa
1 2
3 4 5 6 7
8
9 10 11
12
13 14 15
Figure 4-10. Protein composition of isolates derived from adjusting and holding homogenates at
acidic (3), alkaline (11) or control (6.5) pH for 1 hour, followed by readjustment to
pH 5.5. Unless stated the sample were prepared and applied to the gel with inclusion
of mercaptoethanol. Lanes 1: molecular weight standards (5 uL). Lanes 2-5 represent
pH 11 isolates (15 uL, 10 uL, 10 uL, and 5 uL). Lane 6 and 7 represent pH 11 protein
isolate (15 and 10 uL) prepared with no mercaptoethanol. Lanes 8 and 9 represent pH
6.5 isolate (15, 10 uL). Lanes 10 and 11 represent pH 6.5 isolate (15, 10 uL) with no
mercaptoethanol. Lanes 12 and 13 represent pH 3 isolate (15, 10 uL). Lane 14 and 15
represent pH 3 isolate with no mercaptoethanol (15, 10 uL).
1
2 3
4
5 6
7
8 9 10 11 12
13 14 15
Figure 4-11. Protein composition of isolatesx derived from adjusting and holding homogenates at
acidic (3), alkaline (11) or control (6.5) pH for 8 hour, followed by readjustment to
pH 5.5. Unless stated the sample were prepared and applied to the gel with inclusion
of mercaptoethanol. Lanes 1,12, 15: molecular weight standards (5 uL). Lanes 2 and
3 represent pH 11 protein isolate (15, 10 uL). Lanes 4 and 5 represent 6.5 protein
isolate with no mercaptoethanol (15, 10 uL). Lanes 6 and 7 represent pH 6.5 isolate
(15 and 10 uL). Lanes 8 and 9 represent pH 3 isolate without mercaptoethanol. Lanes
9 and 10 represent pH 3 isolate (15 and 10 uL). Lanes 13 and 14 represent pH 11
isolate without mercaptoethanol.
104
1
2 3 4 5 6
7
8
9 10 11 12 13 14 15
Figure 4-12. Protein composition of isolates derived from adjusting and holding homogenates at
acidic (3), alkaline (11) or control (6.5) pH for 8 hours, followed by readjustment to
pH 5.5, and finally followed by 3 days of refrigerated storage. Unless stated the
sample were prepared and applied to the gel with inclusion of mercaptoethanol. Lanes
1, 10, and 11: molecular standards (5 uL). Lanes 2 and 3 represent pH 6.5 isolate (15,
10 uL). Lanes 4 and 5 represent pH 6.5 isolate without mercaptoethanol. Lanes 6 and
7 represent pH 11 isolate without mercaptoethanol (15, 10 uL). Lanes 8 and 9
represent pH 11 isolate. Lanes 12 and 13 represent pH 3 isolate (15, 10 uL). Lanes 14
and 15 represent pH 3 isolates without mercaptoethanol.
1
2
3
4
5
6
7
8
9
Figure 4-13. Protein composition of initial homogenates at acidic (3), alkaline (11) or contol(6.5)
pH. Unless stated the sample were prepared and applied to the gel with inclusion of
mercaptoethanol. Lane 9: molecular standards (5 uL). Lanes 1 and 2 represent pH 3
homogenate (15, 10 uL). Lanes 3 and 4 represent pH 6.5 homogenate. Lanes 6 and 7
represent pH 11 homogenate (10, 15 uL). Lanes 7 and 8 represent intial mince
homogenate before pH adjustment.
105
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15
Figure 4-14. Protein composition of homogenates held for one hour at acidic (3), alkaline (11) or
control (6.5) pH. Unless stated the sample were prepared and applied to the gel with
inclusion of mercaptoethanol. Lanes 1,4,15: molecular standards (5 uL). Lanes 2 and
3 represent pH 11 homogenate (15, 10 uL). Lanes 5 and 6 represent pH 3.0
homogenate without mercaptoethanol. Lanes 7 and 8 represent pH 3 homogenate (10,
15 uL). Lanes 9 and 10 represent pH 6.5 homogenate, without mercaptoethanol (15,
10 uL). Lanes 11 and 12 represent pH 6.5 homogenate (10, 15 uL). Lanes 13 and 14
represent pH 11 homogenate without mercaptoethanol.
1 2
3
4 5 6
7
8
9
10 11 12 13 14 15
Figure 4-15. Protein composition of homogenates held for 8 hours at acidic (3), alkaline (11) or
neutral (6.5) pH. Unless stated the sample were prepared and applied to the gel with
inclusion of mercaptoethanol. Lanes 1,10,15: molecular standards (5 uL). Lanes 2 and
3 represent pH 11 homogenate (15, 10 uL). Lanes 4 and 5 represent pH 11
homogenates without mercaptoethanol (10, 15 uL). Lanes 6 and 7 represent pH 6.5
homogenate (10, 15 uL). Lanes 8 and 9 represent pH 6.5 homogenate, without
mercaptoethanol (15, 10 uL). Lanes 11 and 12 represent pH 3 homogenate (10, 15
uL). Lanes 13 and 14 represent pH 11 homogenate without mercaptoethanol (15,10
uL).
106
CHAPTER 5
EFFECT OF ACID PROCESSING ON LIPID OXIDATION IN A WASHED SPANISH
MACKEREL MUSCLE MODEL SYSTEM
Introduction
Lipid oxidation in muscle foods is predominantly detrimental to overall quality and storage
stability (29, 233). An array of tissue components contribute to lipid oxidation in muscle foods.
Integral components in lipid oxidation postmortem as elucidated by studies thus far are heme
proteins, phospholipids, and components in the aqueous phase of the muscle (155, 234). Of the
various heme proteins found in dark muscle, the major promoter of lipid oxidation is hemoglobin
(46, 206, 235). According to various sources (5, 13, 16, 51, 67), acid processing, in contrast to
alkaline processing, results in rapid development of lipid oxidation as well as poor protein
functionality and color. It is known that pro-oxidative activity of hemoglobin is greatly increased
at acidic pH (12, 43, 57, 206) and that cytosolic extracts have both anti and pro-oxidative effects
in vitro and in vivo like systems (41, 155, 236-239). The primary substrates for lipid oxidation
are membrane lipids and/or lipid hydroperoxides according to various sources (36). It is
important to understand the interaction between cellular membrane lipids, hemoglobin, and
inherent cytosolic compounds, as the quality of dark muscle foods could be favorably controlled.
The purpose of this study was to investigate how key components in Spanish mackerel
muscle tissue contribute to oxidation during adjustment to low pH and readjustment to neutral
pH using a washed fish muscle basic model system. In addition, this study was conducted to
investigate whether the effects of lipid oxidation in a Spanish mackerel basic model system are
congruent with the extent of lipid oxidation observed in acid homogenates and isolates made
from the same fish muscle source.
107
Materials and Methods
Raw Materials
Spanish mackerel was freshly caught during the Fall season in Cedar Key, Florida. Within
approximately an hour, fish was transported on ice to the Aquatic Food and Bimolecular
Research Laboratory, University of Florida. Upon arrival, fish were manually skinned, degutted,
and filleted. One half of the fillets were minced using a large scale grinder (The Hobart MFG.
Co., Model 4532, Troy, Ohio). Fillets and ground fish (100 gram quantities) were vacuum
packed and stored at -80ºC until needed. The inital average moisture content and pH of the
prepared mince from both batches was 80% and 6.6.
Chemicals
Tricholoracetic acid was obtained from Fischer Scientific (Fair Lawn,New Jersey).
Thiobarbituric acid (TBA), 1,1, 3,3 tetraethoxypropane, potassium phosphate (K2HPO4),
sodium phosphate (NaH2PO4), fluorescein, SigmaMarker® molecular weight standard, and EZBlue stain solution, methanol, and streptomycin sulfate were obtained from Sigma Chemical Co.
(St. Louis, MO). Propyl gallate and (ethlenediaminetetraacetic acid 99%) EDTA were obtained
from Acros Organics (New Jersey, USA). Pre-cast linear 4-20% Tris-HCL gradient gel, Laemmli
buffer, and B-mercaptoethanol, Power Npac 300 power supply were obtained from Bio-Rad
laboratories (Hercules, CA). The Coomassie Plus™- Better Bradford Assay kit was obtained
form Pierce (Rockford, Il). 2,2’ Azobis (2-amidinopropane) dihydrochloride (AAPH) was
obtained from Waco Chemicals, Richmond, VA.. Trolox (6-hydroxy-2,5,7,8tetramethylchroman-2-carboxylic acid) was obtained from Aldrich Chem, Inc., MI, WI.
Washed Fish
Frozen (-80ºC ) Spanish mackerel was thawed under cold running water and was placed
immediately on ice. Washed fish was prepared according to Undeland et al (155). First, all dark
108
muscle was removed. The white muscle was minced (Oster® Heavy Duty Food Grinder,
Sunbeam Corporation, Inc.) and then washed with 3 volumes of cold (4ºC) distilled water. The
mixture was manually mixed with a plastic spatula for 2 minutes and was held on ice for 15
minutes. The mince was then dewatered with 2 layers of cheese cloth. Two washes with 3
volumes of cold 50 mM NaCl solution (pH 5.5) followed. The mince was dewatered between
each wash. For the first wash with NaCl, the mixture was stirred for 2 minutes and was held for
15 minutes. The second wash with NaCl was followed by 1 minute homogenization and
centrifiguation (15,000 g, 20 minutes, 4 ºC). The final moisture content of the washed mince was
75%. The mince was vacuum packed in 50 g quantities and was stored at -80ºC .
Preparation of Spanish Mackerel Hemolysate
Freshly caught Spanish mackerel were anesthetized (0.5 g ethyl 3- aminoenzoate/ L of
water) and placed on ice. Blood was collected from the caudal vein using syringes (25 guage, 1
inch needles) according to Rowley (240). The syringes were pre-loaded with heparin solution
(150 units/mL). Hemolysate was collected from Spanish mackerel blood according to Fyhn et al.
(241). Four volumes of extraction buffer (1.7% NaCl, 1 mM Tris, pH 8.0) was added to the
collected blood. Samples were centrifuged at 700 g for 10 minutes, 4 ºC, using a tabletop
centrifuge (Eppendorf Centrifuge 5702, Brinkman Instruments. Inc., Westbury, N.Y.). The
supernatant (plasma) was discarded and the sediment (red blood cells) was washed in three
volumes of the extraction buffer and again centrifuged at 700g. The supernatant was again
decanted and the red blood cells (sediment) were lysed through addition of 3 volumes of 1mM
Tris, pH 8.0 buffer for one hour. Then, 1 mM NaCl (1/10 volume) was added to precipitate and
maximize stromal protein removal. The lysed red blood cells were centrifuged for 15 minutes,
28,000g, 4ºC. The hemolysate, primarily hemoglobin, was stored in 1 mL cryogenic tubes at
80ºC.
109
Hemoglobin Quantification
Hemoglobin levels were quantified according to Bradford (195) using the Coomassie
Plus™- Better Bradford Assay kit. An approximate hemoglobin concentration was calculated
using a BSA standard curve and the average molecular weight of hemoglobin (ca. 66,000).
Spanish mackerel Pressed juice
Spanish Mackerel pressed juice was prepared according to Gunnarsson et al. (238). Prefrozen Spanish mackerel mince was thawed under cold running water and was added to
centrifuge tubes (100 g) samples. The samples were spun at 22,000 g for 4 hours at 4 ºC. The
supernatant was drawn off and was filtered through four layers of cheese cloth. The pressed juice
was nitrogen flushed and stored at -80ºC until needed.
Thiobarbituric Acid Reactive Substances (TBARS)
TBARS were assessed according to Lemon (193). One gram of sample was homogenized
with 6 mL of trichloracetic acid (TCA) solution composed of 7.5 %TCA, 0.1% propyl gallate,
and 0.1% EDTA dissolved in deionized water. The homogenate was filtered through Whatman
No. 1 filter paper. The TCA extract (2 mL) was added to 2 mL of thiobarbituric acid (TBA)
solution (0.23% dissolved in deionized water). The TCA and TBA solutions were freshly
prepared and heated if needed. The TCA/TBA solution was heated for 40 minutes at 100ºC. The
samples were centrifuged at 2,500 g for 10 minutes at 4ºC. A standard curve was constructed
using 1,1, 3,3 tetraethoxypropane (193).
Spanish mackerel Pressed Juice protein Quantification and Characterization
Protein isolate concentrations were assessed using a modified Biuret method (194). The
Bradford method was used to quantify proteins levels, either in low concentration (<1 mg/mL) or
limited quantity (195), using the Coomassie Plus™- Better Bradford Assay kit. Protein
composition in isolates were analyzed using electrophoresis (196). SDS-PAGE (sodium dodecyl
110
sulfate polyacrylamide gel electrophoresis). Briefly, protein concentration was adjusted to 3 mg/
mL. A 1 mg/mL protein concentration was obtained by combining 0.33 uL of sample, 0.66 uL of
Laemelli buffer, and 50 uL of B-mercaptoethanol. Samples were boiled for 5 minutes in 1.5 mL
cryogenic tubes and were then cooled on ice. Pre-cast linear 4-20% gels were run for 1 hr at 200
V; mA 120 with the Power Npac 300 power supply. Samples were applied to the gels in 10 and
15 uL volumes. Molecular weight standards (6,500 - 205,000 Mw) were run in 5 and 15 uL
quantities. Gels were then set with 12% TCA for 1 hr and were stained with EZ-Blue stain
solution (minimum of 1 hr with agitation). Gels were then destained with deionized water for 18
hrs with agitation, changing out the water numerous times.
Preparation of Low Moisture Washed Spanish Mackerel Mince Oxidation Model System
The model system was prepared according to Undeland et al. (155). Thawed washed
Spanish mackerel (75% moisture) was divided out into 15-20 gram portions with 200 ppm
streptomycin sulfate added. The raw pressed juice had a protein concentration of approximately
100 mg/mL and was diluted to 30 mg/mL. To test the anti- and pro-oxidative effects of pressed
juice and hemoglobin treated at low pH (3) and readjustment to physiological pH (6.8), pressed
juice and hemoglobin were adjusted to appropriate pH and were held for 1 hour and readjusted to
pH 6.8 if needed. Pressed juice and water were added to the washed muscle to bring the moisture
content of the overall system to 82%. The pressed juice was added to the washed muscle such
that ~ 4 fold dilution was achieved. The final pressed juice protein concentration was ~ 8-10
mg/mL. In control samples, pressed juice was replaced with equal amounts of water. Samples
were adjusted to pH 6.8 through drop wise addition of NaOH and thorough mixing with a mortar
and pestle. The system was periodically checked for pH. Once the desired pH was reached, the
samples were plated in petri dishes. Hemoglobin was added in 6 µM concentrations and was well
incorporated with at least 100 turns of a spatula. Thorough mixing of the hemoglobin was
111
indicated by a homogenous color. Plated samples were stored at 4ºC for several days. The
combinations involved pH adjustment of one, two, or three oxidation model constituents: washed
fish muscle (WFM), pressed juice (103), or hemoglobin (Hb). The various combinations were
conducted together and included:
One way Adjustment:
Control (Hb (pH 6.8)+ PJ (pH 6.8) + WFM ( pH 6.8))
Control without PJ (Hb (pH 6.8)+ WFM (pH 6.8))
Hb (pH 3-6.8) +PJ (pH 6.8) +WFM (pH 6.8)
Hb (pH 6.8) +PJ (pH 3-6.8) + WFM (pH 6.8)
Hb (pH 6.8)+ PJ (pH 6.8) + WFM (pH 3-6.8)
Multiple Adjustment:
Control (Hb (pH 6.8) + PJ (pH 6.8) + WFM (pH 6.8))
Hb (pH 3-6.8)+ PJ (pH 6.8) + WFM (pH 3-6.8)
Hb (pH 3-6.8) + PJ (pH 3-6.8) + WFM (pH 6.8)
Hb (pH 6.8) + PJ (pH 3-6.8) + WFM (pH 3-6.8)
Hb (pH 3-6.8) + PJ (pH 3- 6.8) + WFM (pH 3-6.8)
Samples were sniffed several times day to evaluate relative lipid oxidation progress. Based
upon sensory, the 0.5-1 g plugs were taken once a day until odor intensity began to subside.
After plugs were taken, the remaining samples were remixed and flattened and were returned to
4ºC. Samples were stored in aluminum foil at -80ºC until analysis was to be performed.
Preparation of Hemoglobin and Pressed Juice Oxidation Model System
To mimic the washed system pressed juice and hemoglobin concentration upon addition to
the washed fish muscle, the same volumes were prepared. Typically, 4.3 to 4.5 mL of pressed
juice (30 mg/mL of protein) and 0.391 µL of hemoglobin (6 µM) were added to 11 grams of
112
washed fish (75% moisture). Appropriate aliquots of pressed juice (103) and hemoglobin (Hb)
were prepared and were adjusted to pH 3 or pH 6.8 with addition of 1 N NaOH or HCL (5 µl at a
time). The pH treated samples were held for an hour in glass test tubes for one hour. After one
hour, the pH 3 adjusted pressed juice and/or hemoglobin were readjusted to pH 6.8 with addition
of NaOH or HCL, except for controls, which remained the same pH. The final treatments were
as follows:
Hemoglobin /Pressed juice treatments: Hb (pH 3-6.8)
PJ (pH 6.8)
Hb (pH 3-6.8)
PJ (pH 3-6.8)
Hb (pH 6.8) + PJ (pH 6.8)
Hb (pH 6.8) + PJ (pH 3-6.8)
Hb (pH 3-6.8) + PJ (pH 6.8)
Hb (pH 3-6.8) + PJ (pH 3-6.8).
Oxygen Radical Absorbance Capacity (ORAC)
To investigate oxidative changes in the washed system as result of adding pressed juice
and hemoglobin, antioxidant activities were evaluated. The antioxidant activity of pressed juice
and hemoglobin alone and in combination were assessed using the ORAC assay according to
Cao et al. (242). Florescein is used a marker of oxidation inhibition, induced by AAPH. Briefly,
samples were centrifuged at a low speed (1,000 g, Eppendorf Centrifuge 5415 D, Brinkmann
Instruments. Inc., Westbury, N.Y.) to remove insolubles. Samples were normalized to 1 mg/ mL
with water using the Bradford method (195). The Bradford protocol were conducted according to
the Coomassie Plus™- Better Bradford Assay kit. Sample were then diluted 1:10 with ORAC
buffer (0.75 M K2HPO4 and 0.75 M NaH2PO4, pH 7) buffer. The sample (50 uL) was applied to
113
the 96 well Costar ™ microplate (Fischer Scientific, Fair Lawn, New Jersey). A florescein stock
solution (50 nM in methanol) was diluted with ORAC buffer (1:1250) and was added (100 µl) to
the microwells. Before each run, AAPH was freshly made from an AAPH stock solution (90 mg
/mL ORAC buffer) and was quickly added to the microwells (50 µl). The microplate was
immediately placed in a pre-heated (42°C) 96-well florescent Molecular Device fmax®
microplate reader (Molecular Devices Corporation, Sunnyvale, CA) set to a 15-20 s pre-mix
period, 485 nm excitation and 538 emission). Relative florescent unit (RFU) readings were
recorded every 2 minutes in kinetics mode using SoftMax Pro software (Molecular Devices
Corporation, Sunnyvale, CA). The total assay time is 1 hour and 10 minutes. Trolox standards
were applied in 25-200 µM concentrations. Buffer, trolox standards, and samples were applied in
at least duplicate. The area under the florescein decay curve for samples and standards was
calculated and final values were calculated using a quadratic equation.
AUC (Area under the curve): 0.5 +f5/f4+f6/f4 +f7/f4 +…f34/f4+ f35/f4) x CT
Where: F4 = fluorescent reading at cycle 4
Fi = flouresenct reading at cycle i.
CT = cycle time in minutes.
All final ORAC value were expressed in micromoles of Trolox equivalents (TE)/ mg of
protein. Data was calculated and tabulated in Microsoft Excel (Microsoft, Redmond, WA).
Total Antioxidant Potential
Total antioxidant capacity was assayed using the Cayman Chemical Antioxidant Kit
(Cayman Chemicals, Ann Arbor, MI). In this kit, hydrophilic and lipophilic compounds are
tested for their ability to inhibit metmyoglobin induced oxidation of 2,2’- Azino-di- (3ethylbenzthiazoline sulphonate ABTS®. The presence of the oxidized form of ABTS®, ABTS®●+,
is measured at 750 nm. With successful inhibition of oxidation, lower absorbancies are observed.
114
Antioxidant capacities are calculated relative to standard concentrations of Trolox, a water
soluble analogue of tocopherol. Samples were normalized to 1 mg / mL of protein and were
centrifuged to remove insoluble compounds prior to analysis. Total antioxidant capacity was
expressed in mmol Trolox equivalents/mg of protein.
Moisture Content and pH
Moisture content was assessed using a moisture balance (CSI Scientific Company, Inc.,
Fairfax, VA). The pH was measured using an electrode ( Thermo Sure-Flow electrode (Electron
Corp., Waltham, MA) and pH meter (Denver Instrument, model 220, Denver, CO). The
homogenate samples were directly measured for pH. Isolates samples were diluted with
deionized water (1:10) to measure pH.
Statistical Analysis
The SAS system was used to evaluate data. A two-way ANOVA model was used to
evaluate statistical significance for lipid oxidation (TBARS). The two factors for the TBARS
data were treatment and time and they were analyzed with cross design. Antioxidant capacities
for pressed juice, hemoglobin, and combinations were analyzed with treatment
nested within time.
All analyses were performed at least twice and all experiments were
replicated twice. Least square means were used for post hoc multiple comparisons with Tukey
adjustment controlling for Type I error. Data were reported as mean and standard deviation. The
analysis of variance and least significant difference tests were conducted to identify differences
among means, while a Pearson correlation test was conducted to determine the correlations
among means. Statistical significance was declared at p < 0.05.
115
Results
To gain an understanding of how key components in Spanish mackerel tissue contribute to
oxidation through adjustment to acidic pH and then readjustment to neutral pH, three
experiments were conducted using a washed muscle model system (WFM) in which: (1) One of
two components in a washed Spanish mackerel system were adjusted to pH 3 for one hour and
readjusted to pH 6.8 (2) Two and three components in the washed Spanish mackerel system were
adjusted to pH 3 for one hour and then readjusted to pH 6.8. Lastly, non substrate components,
including hemoglobin and the cytosolic extract, alone or in combination were adjusted to pH 3
and readjusted to pH 6.8.
Aqueous extract (pressed juice (PJ), hemoglobin (Hb), and WFM (phospholipids source)
were prepared from freshly caught Spanish mackerel. PJ (30 mg/mL), Hb (6 uM) and WFM
(75%) moisture were combined. The components were combined resulting in a 4 fold PJ
dilution. Prior to acidic adjustment, the pH of the pressed juice and blood were ~6.08 and ~6.68,
respectively. Moisture contents of the washed muscle prior to experimentation ranged between
74 and 75%. Protein contents of hemoglobin and pressed juice were 186.19±1.71 µg/mL and
102.44± 3.41 mg/mL, respectively. This Spanish mackerel pressed juice protein content was
greater than the pressed juice (51 mg/ mL) obtained by Undeland et al. (155) from cod muscle.
Auto-oxidation was apparent with pH 3 adjustment within minutes, as indicated by a
chromophore change from red to rusty brown and coagulation of hemoglobin. As reported by
Lawrie (243), denaturing agents will oxidize hemoglobin to a globular met or deoxy hemoglobin
state with a brown, and sometimes grayish color. In agreement with the change of color and
aggregation, the globin is most likely detached, but the hematin nucleus mostly likely remains
intact (243). Hemoglobin retained the same color and aggregated appearance with readjustment
to pH 6.8.
116
One and Two Way adjustment
Overall lipid hydroperoxides (LHP) in the one component adjustment experiment (Figure
5-1a) were significantly different (p>0.02) with WFM (pH 3-6.8) and NPJ (pH 6.8) >PJ (pH 36.8)> Control (pH 6.8) and Hb (pH 3-6.8). Hydroperoxide levels began declining in the WFM
(pH 3-6.8) treatment at day 3, or as TBARS increased (Figure 5-1b). Increasing LHP levels were
in agreement with levels of formed TBARS, with LHP (mmol/kg tissue) significantly (p<0.001)
increasing in WFM (pH 3-6.8) and NPJ (pH 3-6.8) at 1 and 2 days, respectively. At day 7, WFM
(pH 3-6.8) and NPJ (pH 3-6.8) reached the same levels of LPH (Figures 5-1a). Similar to
TBARS, Hb (pH 3-6.8) and control PV values were comparable.
In one component adjusted treatments, an overall significant increase (p<0.0001) in
TBARS over 7 days was observed (Figure 5-1b). Overall TBARS formed in the following order:
control without pressed juice (NPJ (pH 6.8))> WFM (pH 3-6.8)>PJ (pH 3-6.8) > control (pH 6.8)
= Hb (pH 3-6.8). Significant increases among treatments were observed over time (p<0.0001);
however, the control (pH 6.8) and Hb (pH 3-6.8) treatments did not significantly increase over
time nor did they reach rancidity levels (20 µmol MDA/kg tissue) (Figure 5-1b). The greatest lag
phase (7 days) with respect to TBARS levels for one component tissue adjustment was observed
in the control and Hb (pH 3-6.8), followed by PJ (pH 3-6.8)>NPJ>WFM (pH 3-6.8). The WFM
(pH 3-6.8) and NPJ (pH 3-6.8) treatments formed rancidity levels of TBARS after one and two
days refrigerated storage (p<0.0001). After one day of storage, TBARS levels in the NPJ (pH
6.8) treatment were greater (p<0.0442) than the PJ (pH 3-6.8) at day 0. Appreciable levels of
lipid oxidation in the PJ (3-6.8) compared to Hb (pH 3-6.8) and the control (pH 6.8) occurred
after three days of storage, and rancidity levels were not reached until approximately, 7 days of
storage (Figure 5-1b). After 7 days of storage, the NPJ (pH 6.8) and PJ (pH 3-6.8) treatments
were still increasing in levels of oxidation (p<0.0388) and (p<0.0054) after 7 days of storage.
117
The PJ (3-6.8) treatment reached equal levels of oxidation as the WF (3-6.8) treatment after 7
days of storage (Figure 5-1b).
Two and Three Component Adjustment
Increases in LHP values among treatments were observed over 7 days. Overall levels
(p<0.01) of LPH were as follows: Hb+ WFM (pH 3-6.8), Hb +PJ (pH 3-6.8), and PJ +WFM (pH
3-6.8)> Control (pH 6.8) = Hb+PJ+WFM (pH 3-6.8). The control (pH 6.8) and Hb +WFM (pH
3-6.8) treatment primary and secondary levels of oxidation appeared to follow the same trends.
TBARS levels (Figure 5-2b), were significantly less in the control (pH 6.8) than in all
other treatments. The treatment with all three components adjusted (Hb+PJ+WFM) from pH 3 to
6.8 had significantly (p<0.0001) greater levels of oxidation compared to (Hb+ WFM) adjusted
from pH 3 to 6.8 with the exclusion of adjusted PJ samples. At day 1, significant (p<0.001)
increases in TBARS for all treatments occurred, excluding the control (pH 6.8), as there was not
a significant lag phase among treatments. At day 7, Hb+PJ (pH 3-6.8) incurred higher levels
(p<0.0151) than the PJ +WFM (pH 3-6.8) and PJA +WFM (pH 3-6.8).
Comparison of one way and multi-component pH adjustment in Washed System
After 7 days of storage, several differences among pH adjusted and readjusted components
were observed (Figure 5-1b,2b). The NPJ (pH 6.8), WFM (pH 3-6.8), PJ (3-6.8), Hb (pH 3-6.8)
+PJ (pH 3-6.8) had significantly greater TBARS levels than the control(s) overall (p<0.0002). Of
the treatments significantly different than the control, the NPJ (pH 6.8) had the greatest peak
levels of TBARS after 7 days (p<0.0001). WFM (pH 3-6.8) TBARS were not significantly
different from Hb (pH 3-6.8) +PJ (pH 3-6.8) or PJ (pH 3-6.8). The Hb+PJ (pH 3-6.8) treatment
was not significantly different from the PJ (pH 3-6.8) treatment, and both had greater TBARS
than Hb (pH 3-6.8). The TBARS trend among one way and multi-component pH treatments was
118
as follows: NPJ (pH 6.8)>WFM (pH 3-6.8)= Hb+PJ (pH 3-6.8) = PJ (pH 3-6.8) > all other
treatments.
The overall greatest rate and extent of oxidation (TBARS and PV) over seven days was
found in the WFM (3-6.8), aside from the NPJ (pH 6.8) treatment (Figure 5-1,2). PJ (pH 3-6.8),
Hb (pH 3-6.8) +PJ (pH 3-6.8), PJ (3-6.8) + WFM (3-6.8) + Hb (3-6.8), PJ (3-6.8) + WF (3-6.8)
were equally pro-oxidative. PV levels corresponded with TBARS levels among the treatments.
Lipid hydroperoxides began increasing (p<0.0001) at the fastest rate and occurred to greatest
extent (aside from the pressed juice) in the WFM (pH 3-6.8) followed by equivalent trends in the
Hb+PJ (3-6.8) and PJ (3-6.8) treatment (Figures 5-1a, 5-2a).
Antioxidant mechanisms of non-substrate components WFM model system
ORAC Test
To better understand the effects of acidic pH and readjustments to pH 6.8 on PJ and Hb,
the same samples from the WFM model system were analyzed for antioxidant activity, using
ORAC and TEAC assay (Figure 5-3 a,b). Relative concentrations of Spanish mackerel PJ (30 mg
protein/ mL) and Hb (6 µM) alone or in combination remained consistent with the WFM model
system in the absence of Spanish mackerel WFM.
The radical scavenging ability of PJ and Hb alone or in combination was measured by the
oxygen radical absorbance capacity (ORAC) assay (Figure 5-3a). To confirm that Spanish
mackerel Hb and PJ had scavenging ability at physiological pH, untreated Hb (pH 6.6) and PJ
(pH 6.08) were tested. ORAC values were significantly different (p<0.0001) among native Hb
and PJ, with values of 0.719±0.123 and 0.221± 0.107 umol Trolox equivalents (TE)/mg protein.
The scavenging capacity of PJ and Hb at “neutral” (pH 6.8) and acidic pH (pH 3) compared to
physiological pH was also tested. The radical scavenging ability of pressed juice and hemoglobin
was reduced (p<0.001) with acidification and “neutral” pH compared to physiological pH
119
(Figure 5-3a). Acidified hemoglobin and pressed juice had undetectable ORAC values.
Readjustment of hemoglobin and pressed juice from pH 3 to “neutral” pH gave (p<0.001) ORAC
values that were significantly less than the untreated hemoglobin, and readjustment of acidified
PJ to pH 6.8 gave ORAC values that were insignificant from untreated pressed juice. Untreated
PJ (pH 6.6) ORAC values were not significantly different from Hb (pH 6.8), nor were Hb (pH
6.8) and PJ (pH 6.8) significantly different from each other (Figure 5-2a).
Similar to the WFM model system, the Hb and PJ were held at acidic pH for one hour and
were then readjusted to neutral pH for ORAC testing, either alone or in combination. As shown
in Figure 5-2 a, one hour of holding for Hb (pH 6.8) gave increased ORAC values, or 0.993±
0.122 µmol TE/ mg protein; while, no change was observed for PJ (pH 6.8). Hemoglobin and PJ
exposure to pH 3 for one hour followed by readjustment to pH 6.8, gave significant (p<0.0001)
increases in scavenging capacity relative to initial values at pH 3; however, they were not
significantly different from each other (Figure 5-3a). Interestingly, an accumulative effect of
ORAC abilities was observed when PJ (pH 3-6.8) + Hb (pH 6.8) and PJ (pH 3-6.8)+ Hb (3-6.8)
were combined. Overall increases in scavenging capacity increased when PJ (3-6.8) was
combined with Hb at pH 6.8 or at pH 3-6.8 (Figure 5-3a). Combinations involving PJ at pH 6.8
with either Hb at pH 6.8 or pH 3-6.8 were not significantly different from each other, nor were
they significantly different from PJ at pH 6.8.
TEAC Assay
Trolox equivalent antioxidant capacity (TEAC) test was used to test the same samples. In
contrast to ORAC, no significant difference was observed among untreated or acidified Hb and
PJ. An adjustment to pH 6.8 gave rise to increases in TEAC for PJ (Figure 5-3b). This increase
in PJ antioxidant capacity following pH readjustment was not observed, using ORAC (Figure 53a). After one hour of holding, Hb and PJ at pH 6.8 both increased (p<0.001) in antioxidant
120
capacity with corresponding values of 0.318 ± 0.052 and 0.378 ± 0.025 mM TE/mg protein.
Combined PJ and Hb treatments were either equal to or less than corresponding PJ and Hb pH
6.8 treatments. Similar to ORAC, PJ (pH 3-6.8)+ Hb (6.8) was significantly greater Hb (pH 36.8) and PJ (3- 6.8), and pJ (pH 6.8)+ Hb (pH 3-6.8) with p<0.0001.
SDS Page
The electrophoretic pattern of Spanish mackerel pressed juice was comparable to cod
pressed juice as found by Undeland et al. (155), with exception of the 8 kDa fraction observed in
the cod pressed juice. Identifiable proteins in the Spanish mackerel pressed juice include αsubunits of myosin (97, 75 kDa, 67 kDa), Desmin and Vimentin (55 and 52 kDa), actin (49 kDa),
γ actin subunit (36 kDa), and troponin T (29 kDa), and myoglobin and/or subunits of hemoglobin
(14 kDa).
Discussion
The pH sequence to which fish muscle components are exposed in the acid/alkali aided
process can activate oxidative mechanisms. In general, the effect of pH can vary depending upon
the complexity of the system and the catalysts involved. The specific tissue components proved
to be most important in the lipid oxidation of muscle foods are still under much debate. To
investigate the roles of tissue components in the lipid oxidation of dark post-mortem fish muscle,
Spanish mackerel hemoglobin, pressed juice, and washed muscle were combined and exposed to
acidic (pH 3) and then pH 6.8 readjustment. The pH 3 adjustment treatment was utilized as per
discussion in Chapter 3, in which the pH 3 homogenates along with pH 3.5 homogenates
incurred the greatest levels of lipid oxidation (PV and TBARS). Furthermore, other studies have
shown significant oxidation in washed cod systems adjusted to pH 3. Kristinsson (51, 152)
observed a rapid increase in TBARS in a hemoglobin mediated washed cod system in which
washed cod and hemoglobin were adjusted to acid pH (2.5-3.5). Similarly, Pazos et al. (161)
121
observed a rapid onset and rate of lipid oxidation in a pH 3.5 adjusted wash cod system with
hemoglobin (161). Because holding times (1 and 8 hours) had no effect on lipid oxidation
development in the acid isolation of Spanish mackerel, a one hour holding time of tissue
components was implemented. Kristinsson and Hultin (51) found that increased holding times
from 90 sec to 20 minutes increased levels of oxidation in a pH 3.0 adjusted washed cod
hemoglobin mediated system; however, at pH 2.5 no noticeable changes were observed with
different holding times. Additionally, pH 6.8 was chosen as the readjustment pH due to findings
made by Pazos et al. (179). In a hemoglobin-catalyzed lipid oxidation system, the extent and rate
at which cod microsomes oxidized fastest occurred at pH 6.8 (179). The optimal pH range for
sarcoplasmic reticulum (SR) oxidation for in vitro fish models has been found to be between pH
6.5 and 6.9 (5). The washed Spanish mackerel muscle system serves as the phospholipid matrix.
No pressed juice treatment (control without pressed juice)
Lipid hydroperoxides and TBARS were the highest in the control treatment with no added
pressed juice (NPJ) in which Spanish mackerel washed fish muscle (WFM) and hemoglobin
(Hb) were combined at pH 6.8 (Figure 5-1b). According to Pazos et al. (161), there were
appreciable levels of lipid oxidation at pH 6.8 in a microsomal cod system to which cod
hemoglobin was added (161). Above pH 6.8, lipid oxidation decreased with increased
microsomal pH adjustments to 8.4. Adjustments below 6.8 reduced rates of lipid oxidation
further at pH 6 and at pH 3.5 compared to pH 8.4. Pazos et al. (161) also investigated the effects
of pH adjustment on the development of lipid oxidation in a washed fish cod matrix with
hemoglobin added, and an inverse correlation between lipid oxidation and pH was observed
compared to the tested microsomal system. Longer lag phases and reduced TBARS were
observed with decreases in pH from 7.4 to 6.8. Decreasing the pH to 3.5, resulted in even greater
extent and rate of oxidation (161). The latter findings in the microsomal system agree with the
122
rapid rate of oxidation seen at pH 6.8 in the NPJ treatment in the presence of 5 µM Spanish
mackerel hemoglobin (Figure 5-1a,b). Kristinsson (152) found that in the absence of
hemoglobin, TBARS did not form in a washed cod system. Thus, the pro-oxidative activity of
hemoglobin exhibited in the study by Kristinsson (152) concurred with the NPJ treatment in
which only hemoglobin and washed fish muscle were combined at pH 6.8. Undeland et al. (155)
found that there was significant lipid oxidation in a system very similar to the current study,
including washed cod and 5 µM trout hemoglobin (155). In a study implementing native sperm
whale myoglobin and washed cod, significant levels of oxidation were observed (244). Washed
cod in the absence of sperm hemoglobin developed very little levels of oxidation (TBARS,
LHP).
Lipid oxidation in the control samples
The fact that in the control treatment, untreated pH hemoglobin did not catalyze lipid
oxidation in the presence of pressed juice compared to the no pressed juice (NPJ) control
treatment (Figure 5-1a,b), confirms the antioxidative capacity of pressed juice. According to
Undeland et al. (155), TBARS and PV formation were completely inhibited in a hemoglobin
mediated washed cod muscle system, in the presence of cod pressed juice (155). In the Spanish
mackerel WFM system, the peak levels of oxidation were highest in the control without pressed
juice (Figure 5-1a,b). The inhibitory activity of pressed juice and the highly pro-oxidative nature
of hemoglobin on lipid oxidation are equally evident from the NPJ (pH 6.8) and control (pH 6.8)
treatments (Figures 5-1a,b). It is evident that the incorporation of aqueous extracts native to fish
cytosol would enable for protection from reactive oxygen species in the phospholipid membrane
post mortem, similar to native tissue. According to Undeland et al. (155), the formation of
hydroperoxides and painty odor was significantly delayed or inhibited in the hemoglobin
mediated cod oxidation system when cod aqueous extracts were added. Similar antioxidative
123
capacities have been observed in other fish species including haddock, dab, and winter flounder
(155). In a study conducted by Decker and Hultin (245), pressed juice from light mackerel
muscle effectively inhibited metal catalyzed oxidation (245).
It was evident from the low levels of lipid oxidation in the control (pH 6.8) compared to
the NPJ (pH 6.8) treatment, that a four fold dilution of the Spanish mackerel pressed juice was
effective in inhibiting lipid oxidation (Figure 5-1a,b). It is well known that the antioxidative
nature of compounds is affected by environmental changes in pH and concentration (64-66).
Undeland et al. (155) found that dilutions up to 6 fold significantly prolonged the lag phase by at
least 4 days; whereas, whole pressed juice delayed the onset of oxidation by >11 days. A < 1
kDa dialysis retentate and whole pressed juice were equally inhibitive of lipid oxidation in the
washed cod system. Thus, the antioxidative nature of the pressed juice is attributed to low
molecular weight compounds rather than antioxidant proteins (superoxide dismutase, catalase,
and glutathione peroxidase). Chicken pressed juice was most effective with low and high
molecular weight compounds present as found by Rong et al. (246). When ascorbate oxidase
(200 µM) was added to the chicken pressed juice in the presence of hemoglobin, significantly
more oxidation occurred. These findings support the importance of low molecular weight
compound including and not limited to ascorbate, glutathione, and urate (246). An antioxidative
balance in the presence of hemoglobin, in the Spanish mackerel pressed juice, affirms the
abundant antioxidative capacity of low molecular weight compounds as observed by Undeland et
al. (66) in cod pressed juice and Takama (146) in trout pressed juice. The reasons for decreased
prooxidative activity of hemoglobin in the presence of pressed juice have yet to be elucidated.
Inhibitory mechanisms may be related to the conversion of hemoglobin to deoxyhemoglobin, or
with autoxidation, i.e. conversion to metmyoglobin. This means as a cause for reduced lipid
124
oxidation were tentatively concluded by Undeland et al. (155) in a hemoglobin/pressed juice
aqueous system. These studies suggest that Spanish mackerel pressed juice may have inhibited
lipid oxidation by preventing conversion of hemoglobin to prooxidative forms. Heme proteins
are believed to promote oxidation through release of hemin from globin, enabling opportunities
for hemin and lipids in muscle tissue to react (247-253). Grunwald and Richards (244) showed
that hemoglobin dissocation into subunits incurred greater levels of lipid oxidation than
undissoaciated hemoglobin. Hemoglobin was genetically altered to achieve non dissociable
subunits. With dissociation, myoglobin subunits result. According to Grunwald and Richards
(244), spermwhale metmyglobin induced greater formation rates and levels of oxidation in a
washed cod muscle system versus ferrous myoglobin due to lower hemin affinity. Hemin is
defined as the poryphrin ring with ferric (Fe3+) iron bound. Grunwald and Richards (244) found
that hemin alone could activate lipid oxidation in the washed cod system, and hemin contributes
to the initiation of lipid oxidation, but not the rate at which lipid oxidation occurs. Metmyoglobin
is believed to be pro-oxidative due to a lower affinity for hemin and greater likelihood of hemin
release. The Spanish mackerel pressed juice may have inhibited lipid oxidation by preventing
hemoglobin oxidation, dissociation, and/or heme loss.
The Effect of pH Adjustment on Tissue Components
The pH effect on hemoglobin
No significant difference was observed between the treatment in which hemoglobin was
adjusted from pH 3 to 6.8 and the control (Figure 5-1a,b). Previous studies would suggest that at
low pH, the Spanish mackerel hemoglobin became more prooxidative. Lowering the pH of
hemoglobin has been shown to activate fish hemoglobin (12, 35, 43, 51, 162). Lowering to pH 6
from neutrality (pH 7) results in increased hemoglobin pro-oxidative activity due to the
formation of deoxyhemoglobin as found by measurement of lipid oxidation in a washed cod
125
model system (206). The process of oxygen loss from hemoglobin as the pH is lowered is known
as the Bohr effect (50). A change in subunit dissociation could increase access and enable for
hydroperoxide decomposition and autooxidation, producing potent ferryl-Hb radicals and even
heme dissociation. Methemoglobin dissociation is 60 times more likely compared to oxy-and
deoxyhemoglobin when exposed to pH, resulting in the release of hemin which can interact with
membrane layers due to its polarity. In addition, the lowering of pH forces the heme-iron outside
the plane of the protein . In the case of heme dissociation into dimers from tetramers, it has been
found that autooxidation occurs 16 times faster (254). The increase of pro-oxidative activity is
due to the partial unfolding of the hemoglobin, giving rise to a molten globular state (12). In
another study, when hemoglobin was exposed to extreme pH (pH 2.5) versus neutral pH (7),
lipid oxidation proceeded very quickly in a washed cod model system. This increase in lipid
oxidation was attributed to various property changes in the hemoglobin that allowed for greater
association with the phospholipids: unfolding, increased hydrophobicity, and detachment from
the proximal histidine group (230). As dissociation of hemoglobin with pH adjustment is likely
to occur, the specific affect of pH on myoglobin should also be considered. At neutral pH (6.2),
metmyoglobin is a potent oxidative catalyist. At acidic pH and in the presence of
hydroperoxides, metmyoglobin is very effective catalyst of lipid oxidation than at neutral pH.
This was evident in increased levels of lipid oxidation in linoleate acid emulsion (255).
Mechanistically, pH lowering can cause oxy-hemoglobin (Fe2+-O2) to auto-oxidize to Met (Fe3+O2), releasing superoxide (256). Further mutation can then result in the formation of hydrogen
peroxide. Hydrogen peroxide activates Met Hb/Mb to two hypervalent species,
perferrylmyoglobin/hemoglobin and ferrylmyglobin/hemoglobin. Perferrylmyoglobin is a very
transient species and its pro-oxidative and initiative activity is negligible at pH values 5.5-6.5
126
(211) and is quickly converted to the most stable hypervalent species, ferryl-Mb/Hb. In fresh
meat (pH 5.5-5.8), the ferryl-Hb/Mb species are very short lived and are quickly autoreduced
back to the deoxy-Mb/Hb; however, the ferryl-Hb/Mb species is deemed very pro-oxidative and
deemed the primary inducer of lipid oxidation. With aged fish, the antioxidative defenses
dwindle, and the ability of this species to initiate lipid oxidation could occur with ease when
subjected to reduced pH and lipid concentration; however, in a model linoleate system is was
found that the met-Hb form was as oxidative as the ferryl-Hb form (47). In contrast to the ferryl
species, it is observed that the oxidative potential of oxy/deoxy/ and met-Hb is dependant upon
hydrogen peroxide concentration (43).
This accumulation of the met-form occurs with drops in post-mortem pH and or acid-aid
processing (43, 257). In the presence of hydroperoxides, Met-Hb catalyzed lipid oxidation occurs
and gives rise to products that can propagate oxidation (46, 258, 259). Studies have found that
met and deoxy- hemoglobin/ myoglobin to be the more pro-oxidant forms versus oxymyoglobin/hemoglobin (43, 47, 206). Oxygen release from the heme protein may provide greater
access to oxidative substrates and or reactive oxygen species (52). It is evident that prooxidative
effects by lowering pH were not evident in Hb (pH 3- 6.8). In contrast to various studies, an
increase in prooxidative activity with pH lowering was not evident in the hemoglobin (pH 3-6.8)
treatment (Figure 5-1a,b). Thus, the readjustment to pH 6.8 or other factors affected the catalytic
activity of hemoglobin.
Hemoglobin adjustment to low pH and readjustment to pH 6.8
In the presence of pressed juice and the substrate (WFM), hemoglobin (pH 3-6.8) proved
ineffective in promoting lipid oxidation (Figure 5-1a,b). The onset of rancidity occurred
significantly earlier in a hemoglobin mediated cod system adjusted to pH 2.5 compared to pH 3
and 3.5 in a study conducted by Kristinsson and Hultin (60). It was found that the rate of lipid
127
oxidation was greater when trout hemoglobin, adjusted to pH 2.5 and then readjusted to pH 7,
was added to washed cod compared to when hemoglobin and washed cod were pH adjusted
together. The main difference between the current study and that which was conducted by
Kristinsson (152) at low pH followed by neutralization of blood, is that the current experiment
was conducted in the presence of pressed juice. The results with Spanish mackerel thus clearly
demonstrate how effectively the PJ functions as an antioxidant.
Kristinsson and Hultin (51) investigated the changes in trout hemoglobin conformation
with pH adjustment. UV-spectra revealed significant heme peak degradation and blue shifting as
well and reduced oxygen peaks at low pH. Hemoglobin readjusted to neutrality from low pH
retained the conformation changes and remained somewhat unfolded. These conformational
changes are in part indicative of autoxidation, but also unfolding of the protein molecule (60).
The autoxidation of hemoglobin with pH lowering provide evidence for the increased prooxidative activity of hemoglobin compared to native hemoglobin (pH 7) in a washed model
system. The increased pro-oxidative activity at low pH and with readjustment could possibly
explain a quicker onset of oxidation (152).
Hemoglobin aggregation was observed at low pH and after pH readjustment. According to
Kristinsson (152), aggregation due to increases in hydrophobicity at low pH, due to unfolding,
could encourage interactions between hemoglobin and phospholipids within the washed fish
matrix. Kristinsson and Hultin (51) found that despite greater rates of lipid oxidation, peak levels
of oxidation with hemoglobin adjusted to low pH followed by neutral pH adjustment did not
exceed oxidation at neutral pH in a washed cod system. This could be due to aggregation of the
hemoglobin upon partial refolding, which decreased access of the hemoglobin to substrates. The
pro-oxidative activity of hemoglobin is negligible at alkaline pH (10.5-11) in a washed muscle
128
system and lipid oxidation was significantly suppressed compared to the neutral pH in the wash
fish Hb-initiated oxidation system (51). Similar to studies conduced by Kristinsson and Hultin
(218), the Spanish mackerel hemoglobin could have indeed been more pro-oxidative with acid
treatment and readjustment to neutral pH; however, either deleterious aggregation due to one
hour in this study versus 20 minutes in the study conducted by Kristinsson and Hultin (51)
prevented interaction with the washed fish matrix. Kristinsson and Hultin (199) have shown that
longer holding times at low pH prior to readjustment significantly reduced hemoglobin’s ability
to refold and leads to reduced solubility. Hemoglobin’s propensity to aggregate is due to
increased exposure of hydrophobic groups as well and increased release of heme/hemin. Heme
can interfere with refolding of heme proteins (260). Heme has a high affinity towards an
aggregative state. The pH treated hemoglobin solution was most likely a mix of species that
easily aggregated (261). A possible species would include the permanently denatured
hemoglobin known as a hemichrome, which may have formed with a one hour holding time at
pH 3. Severe aggregation would inhibit interaction with the washed system. Additionally,
pressed juice may have contributed to the negligible effects elicited by pH treated hemoglobin by
quenching any existent pro-oxidative activity.
Pressed Juice adjustment
To date, published studies on the pH adjustment from extreme pH of a hemoglobin
mediated WFM muscle system, with added PJ, do not exist. It was observed that adding the pH
adjusted PJ (3-6.8) and PJ (3-6.8) + Hb (3-6.8) to the washed fish muscle resulted in equivalent
levels and rates of oxidation (Figures 5-1b, 5-2 b). All treatments containing readjusted pressed
juice had greater overall values of oxidation relative to the control; however, all values were
comparably less than the WFM (3-6.8) and the NPJ (pH 6.8) treatments (Figures 5-1b, 5-2 b).
The observation that treated pressed juice compared to untreated pressed juice resulted in
129
significantly higher levels of oxidation shows that the antioxidative effects are inhibited and less
able to control hemoglobin mediated lipid oxidation. Additionally, pH adjusted hemoglobin did
not significantly effect the overall TBARS levels as the PJ (3-6.8) and PJ (3-6.8) + Hb (3-6.8)
were equivalent. Similarly, all other treatments in which pH adjusted hemoglobin was added
were very low and were comparable to levels found in the control. In the pressed juice
treatments, further evidence of reduced and/or inhibited pro-oxidative status of hemoglobin is
presented.
Undeland et al. (155) found that heating of pressed juice suppressed or destroyed
antioxidative capacities in chicken and herring pressed juice compared to cod pressed juice.
Reduced antioxidative capacities in chicken and herring pressed juice upon heating were
attributed to higher levels of hemoglobin (3.2-11.6 uM) compared to cod pressed juice (1.8 uM).
It is possible that through the heating process, iron catalyst such as heme and iron are released
(262), and thus the antioxidative balance is shifted toward a more pro-oxidative one. Rong et al.
(246) found that heated chicken pressed juice was significantly less effective in inhibiting
oxidation in a cod hemoglobin mediated system. By chelating low molecular weight iron in the
heated pressed juice before addition to the washed muscle, a reduction in TBAR formation by
23% was observed (246). In this study, the pH treatment of pressed juice could activate low
molecular weight iron and hemoglobin inherently present and provide less lipid oxidation
protection as observed in the pressed juice treatments: PJ (3-6.8) and Hb (3-6.8) +PJ (3-6.8) juice
treatment. Undeland et al. (155) investigated the effects of pH in a washed fish system.
Decreasing the pH of the system with the inclusion of pressed juice from pH 6.6 to 6.0 reduced
the effectiveness of the pressed juice. At pH 6.6 versus 6.0, the lipid oxidation lag phase of
hydroperoxides was reduced from 5 to 2 days. Similarly, pH adjustment from pH 3 and
130
readjustment to pH 6.8, significantly reduced the antioxidative properties of Spanish mackerel
pressed juice. It is possible that at low pH the proteins contained within the pressed juice could
undergo protein oxidation and resulting protein radicals could propagate significant increases in
lipid oxidation, relative to the untreated pressed juice constituents. Ostdal et al. (211) found that
oxidized BSA radicals effectively initiated oxidation in the presence of other molecules.
Albumin was effectively oxidized by ferrylmyoglobin at low pH in a model system as indicated
by increases in dityrosine (211). Pressed juice does contain blood components, including
albumin and other proteins. The increases in pro-oxidative activity in the PJ (3-6.8) + Hb (3-6.8)
treatment, as hemoglobin did not effectively promote oxidation, and decreases in antioxidative
capacity in the PJ (3-6.8) treatment could be in part due to the formation of reactive protein
radicals in the pressed juice upon pH adjustment and readjustment.
Lipid oxidation in the adjusted washed fish muscle
It is important to address changes in the main constituents of washed fish with pH. The
washed cod muscle tissue should be virtually free of pro- and antioxidants and is mainly
composed of phospholipids and myofibrillar proteins (244). In various studies, the susceptibility
of the muscle to lipid oxidation is determined by the pro-oxidants and available membrane lipids.
In general, fatty fish store their lipid (triacylglycerol deposits) intra- and intercellularly; while
lean fish store their main sources of lipid in the liver. Unlike triacyglyerol content, phospholipids
maintain a consistent presence (0.5-1% w/w) among fish species (25). According to Pazos et al.
(161), the washed cod system utilized for hemoglobin mediated lipid oxidation was less than one
percent lipid. Phospholipids comprised 86% of the total lipids present (161).
In a washed cod system (0.1% fat), aromas indicative of lipid oxidation were observed.
When increasing the phospholipids content by six times, the rate nor the extent of rancidity was
significantly affected. However, no rancidity developed when hemolysate was added to a myosin
131
preparation, emphasing that there needs to be some lipid for the oxidation to occur (but very low
levels) (58). Furthermore, an enzyme iron-catalyzed system was used to initiate oxidation in
microsomes. The phospholipid oxidation increased the rate of co-suspended triacyglycerol
oxidation (236). In another study (57), triacyglycerols in the form of menhaden oil and
hemoglobin were added to a washed cod model system. No significant differences in the rates of
oxidation were observed with or without menhaden oil. In addition, the levels of hemoglobin
correlated with extents and rates of lipid oxidation. When hemoglobin was not added to the
washed system, lipid oxidation proceeded very slowly; while doubling the amount of
hemoglobin doubled the extent of oxidation, and decreased the lag phase. In fact, the samples
containing only phospholipids oxidized faster than the treatments with the menhaden oil added
(57). Hemoglobin may preferentially oxidize phospholipids due to: 1) the surface area of
phospholipids and 2) the polar nature of phospholipids allowing for water soluble hemoglobin to
acquire better access and interaction (Borst et al, 2000). From these studies, the washed fish
muscle matrix serves as an appropriate substrate for hemoglobin mediated lipid oxidation
studies.
The greatest levels of oxidation, aside from the NPJ treatment formed in the washed fish
muscle (WFM) system that was first adjusted to pH 3 for one hour and then readjusted to pH 6.8.
TBARS levels were negligible and comparable to the control, when hemoglobin was treated with
and without WFM. These results provide evidence for significant changes in washed fish muscle
that make it less resistant to oxidation. Conformational changes occurring in the washed fish
muscle most likely facilitate increased native hemoglobin association with the less resistant
WFM (3-6.8).
132
Pazos et al. (161) also tested the affects of adjusting a microsomal cod system to pH 3.5
(30 minutes) followed by readjustment to pH 6.0 and 8.0 in the presence of 5 µM cod
hemoglobin. Results indicated that there was greater rate and onset of oxidation in the control
samples (pH 6.0 and 8.0) than in the microsomal samples first adjusted to pH 3.5. Microsomal
fractions at pH 6.0 and 8.0 incurred greater levels of oxidation than at pH 3.5 with adjustment to
neutral pH. Less lipid oxidation occurred with pH lowering before readjustment to pH 8.0
compared to pH 6.0. This lowering affect is somewhat analogous to the significantly reduced
levels of oxidation in the WF (3-6.8) treatment, compared to the NPJ treatment (Hb and washed
fish adjusted to pH 6.8).
It was suggested by Pazos et al. (161) that pH effects on the structure of both the blood and
the microsomes play a role in a reduced levels of lipid oxidation, with previous adjustment to pH
3. Additionally, in the same study (161), a washed cod model system was studied with the
addition of 3 µmol of hemoglobin/kg of tissue. Lipid oxidation results from the washed cod
muscle system varied greatly from the microsomal model system. In the instance of washed cod,
oxidation at pH 3.5 proceeded very quickly and to a greater extent compared to pH 6.8. Washed
cod at pH 3.5 and pH 6.8 proceeded faster than systems adjusted to pH 7.8. Interestingly, peak
levels of oxidation were equivalent among pH treatments. These findings by Pazos et al. (161)
concur with results found by Kristinsson and Hultin (51), in which a washed cod muscle system
with trout hemoglobin were adjusted to low pH 2.5 and neutral pH. The washed cod system
developed significantly higher levels of oxidation at pH 2.5 than at pH 7 (51). Interestingly,
Pazos et al. (161) found that microsomes adjusted to pH 3.5 for 6 hours in the absence of
hemoglobin lost significantly more sulfydryl groups than other pH treatments. In the presence of
hemoglobin, TBARS trends and loss of sulfhydryls within the microsome system correlated well
133
with all pH treatments, excluding pH 3.5. The lack of correlation in indicators of lipid and
protein oxidation could be indicative of phospholipid changes that occur at low pH (161).
Changes within the phospholipids structure could explain the high levels of lipid oxidation
observed in the WFM (3-6.8) treatment.
The overall greatest rate and extent of lipid oxidation among all pH adjustments, occurred
in the WFM (3-6.8) treatment (Figure 5-1a,b). Kristinsson and Hultin (18) addressed the affects
of adjusting the washed fish to pH 2.5 and then to pH 7, followed by the addition of untreated
trout hemoglobin. Washed cod that was acidified for 20 minutes followed by pH 7 readjustments
had a shorter lag phase than at pH 7. In the instance of alkaline adjustment and readjustment to
neutrality, reduced susceptibility to oxidation occurred in the washed cod compared to the
control. When hemoglobin was alkaline treated and readjusted to pH 7 and added to washed
muscle, equivalent rates and levels of oxidation occurred, compared to the control (18). Changes
in susceptibility to lipid oxidation when washed fish was treated, substantiates evidence that the
washed fish (phospholipids and myofrillar proteins) play an essential role in lipid oxidation
development with pH adjustment.
Studies on the effects of myofibrillar and myosin proteins may give insight as to why the
pH treated washed fish matrix would result in greater lipid oxidation. Kristinsson and Hultin (18)
investigated the affects of low (2.5) and high pH (11) adjustment and readjustment on myosin
and myofibrillar proteins. The “myofibrillar fraction” is obtained through three washes and
centrifugation to remove soluble compounds (230), including sarcoplasmic proteins. The
“myofibrillar fraction” is equivalent to a washed cod system. Myosin and myofibrillar proteins
experienced improved functionality (gelation, solubility, and emulsification) with pH
readjustment from low and high pH to pH 7.5. Increased functionality was due to conformational
134
changes within the protein structure including increased hydrophobicity and surface activity. In
most cases, the myosin proteins revealed slightly greater increases in functionality with pH
readjustment (18). These results may explain the greater levels of lipid oxidation in the WFM (36.8). Changes in protein conformation may affect the interaction between proteins and lipids.
Conformational changes within the WFM may enable hemoglobin to better interact with the
WFM and effectively promote oxidation.
Grunwald and Richards (244) summarized the heme protein mediated lipid oxidation
mechanism in washed cod muscle. Metmyoglobin has a low affinity for heme relative to ferrous
heme. Hemin, released from the globin, will preferentially interact with phospholipids through
hydrophobic interaction (263). Dissociation of hemoglobin will cause more oxidation, due to low
hemin affinity of subunits compared to intact hemoglobin or myoglobin (252, 264). These
findings substantiate hemoglobins proxidative ability in a WFM model system. Hydrophilic
portions of the hemin can also further the interaction with phospholipids by binding to the polar
head groups (263). Miyazawa (265) has reported the presence of preformed lipid hydroperoxides
in phospholipid membranes, in vivo. Lipid hydroperoxides and hemin will react to form various
reactive oxygen species (ROS). These ROS can further oxidize phospholipids by extracting
hydrogen. Additionally, hemin is known to degrade upon reactions with ROS (244). It is further
concluded that hemin is a reactant versus a catalyst during lipid oxidation. This was concluded
because hemin can be degraded by lipid oxidation products and that lipid oxidation is not
proportional to increases in hemoglobin concentrations, in a washed cod muscle system (154).
Antioxidant Capacity as related to oxidation in washed muscle
Antioxidant potential of the hemoglobin and pressed juice added to the washed Spanish
mackerel were assessed to possibly elicit reasons as to why there are differences in lipid
oxidation among washed Spanish mackerel pH treatments. The oxygen radical absorbance
135
capacity (ORAC) and Trolox equivalent antioxidant capacity (TEAC) test were used to generate
an “antioxidant profile”.
Antioxidant mechanisms generally fit into one of two categories: hydrogen atom transfer
(HAT) or single electron transport (SET). HAT mechanisms measure the ability of antioxidant(s)
to donate a hydrogen atom to free radicals; whereas, SET reactions measure the ability of an
antioxidant to donate an electron and reduce various compounds including, metals, carbonyls,
radicals, etc (266). HAT and SET mechanisms occur simultaneously in oxidative reactions. The
extent to which each occurs is dependant upon reaction environment as well as antioxidant
properties.
The oxygen radical absorbance capacity (ORAC) method is a HAT based protocol. It is an
in vitro assay, modeling antioxidant-lipid interactions in food and physiological environments.
Hydrophilic protocols are used quite extensively, and hydrophobic ORAC methods are currently
being adapted. ORAC protocols involve analysis over long durations (> 1 hour); whereas SET
reactions, such as the Trolox equivalent antioxidant (TEAC) capacity assay, require short
incubation times (4-6 minutes). TEAC, or ABTS (2,2’-azinobis (3-ethylbenzothiazoline-6solfonic acid) assays measure antioxidant scavenging capacity for a long lived radical anion,
ABTS●+. The TEAC method is food and physiologically appropriate and can be used with
varying conditions and media. Specifically, the assay can detect antioxidative capacity with
changes in pH and can be used for both hydrophilic and hydrophobic antioxidant assessment
(267). The accumulated effects of various aqueous and lipid antioxidants are measured
including: vitamins, proteins, lipids, glutathione, uric acid, etc. (268). TEAC reactions may not
concur with slower reactions, such as ORAC. Short incubation times may result in low TEAC
values because the endpoint maybe prematurely measured for certain antioxidants and/or media
136
being assessed. Quantitatively, this assay has been given little merit according to Van den Berg
and Haenen(269), due to lack of standardized endpoints; however, this assay could provide
information for comparative analysis of samples and/or antioxidants (269). Due to the limitations
of each assay, a discussion of general trends observed with each test for Spanish mackerel blood
and pressed juice follows.
Initial hemoglobin and pressed juice samples increased in antioxidant activity with pH
adjustment from physiological pH (6.08 and 6.6) correspondingly, to pH 6.8 (Figures 5-3 a,b).
These results concur with previous findings. As per the previous discussion of the pH affect on
hemoglobin, reduced pro-oxidative activity occurs with increases in pH above pH 6 (12, 42, 51,
161, 206). According to Antonenkov and Sies (270), rat liver cytosol optimally prevented lipid
oxidation in a microsome system between pH 6.2 and 6.8. After a one hour holding period, pH
6.8 hemoglobin and pressed juice increased or retained their antioxidative potential as observed
in the TEAC and ORAC Assays (Figures 5-3a,b). Undeland et al. (155) observed a greater
inhibition of lipid oxidation in a washed cod model system when cytosol was added at pH 6.6
compared to pH 6. When pH was lowered, the ORAC assay (Figure 5-3 a) revealed a significant
reduction in ant-oxidative capacity for the Spanish mackerel hemoglobin and pressed juice. This
reduction is most likely attributed to increase in pro-oxidative activity with decreases in pH and
aggregation of proteins at pH 3. Denaturation and aggregation most likely inhibit mobility and
antioxidative activity of proteins in the pressed juice. Antonenkov and Seis (270) found the
antioxidative assessment of rat cytosol unreliable at pH values below 5.8, due to protein
aggregation. Antioxidative capacity was regained in the hemoglobin and pressed juice with
readjustment to pH 6.8 (Figure 5-3 a, b). It is possible that the denatured proteins could refold
and thus serve as greater radical scavengers with readjustment to pH 6.8 compared to pH 3.
137
The additive effects of pressed juice and hemoglobin combinations were equal to or higher
than treated components alone. The combined mixture of hemoglobin (pH 6.8) and pressed juice
(pH 6.8) possessed high antioxidant potential. This observation confirms the low levels of lipid
oxidation in the control treatment for the washed system. The relatively lower levels of lipid
oxidation observed in the pressed juice (3-6.8) + Hemoglobin (6.8), compared to the WFM (36.8) treatment, parallel a relatively high level of antioxidant potential (Figure 5-2b). There
seemed to be little correlation between the different combinations of pressed juice and
hemoglobin and the level of lipid oxidation in the washed system. Extracts/samples from the
entire washed system is necessary to determine the levels of lipid oxidation. As realized, the state
of the washed system significantly affects interactions with the pressed juice and hemoglobin.
Analysis of the WFM model system was attempted; however, results were undetectable or
inconsistent. The problems were most likely due to dilutions or protein insolubility. Thus,
methods and/or extraction techniques to better assess the antioxidative status in the whole system
are necessary
Antioxidative potential for each treatment did vary among the TEAC and ORAC assay;
however, similarities among the assays was observed. Variability among assays could be due to
various factors regarding assay function: simplicity, required instrumentation, biological
suitability, mechanisms, incubation and endpoints, and lipophilic and/or hydrophilic affinity
(268).
Differences among the aqueous system vs. the washed model system are that protein
aggregates either in the hemoglobin through pH adjustment to pH 6.8 or adjustment from 3 to
6.8, must be centrifuged out prior to assessment. These aggregates could still possibly play a pro
or antioxidative role in WFM hemoglobin mediated lipid oxidation. The aggregation may reduce
138
the overall antioxidative potential of treatments, but at the same time remove prooxidative
compounds from the system. Lastly, addition of the pressed juice and hemoglobin combinations
into the washed system resulted in 4 fold dilution of pressed juice. Components in the aqueous
system were in the same concentration used in the washed system. Pressed juice dilution still
proved to be effective in delaying or preventing oxidation in the washed system; however, the
“antioxidative profiles” of the hemoglobin and pressed juice are most likely reduced in the
washed system compared to the wet system, due to dilution.
Conclusion
In conclusion, the role of hemoglobin as a reactant and pro-oxidant in a washed Spanish
mackerel model system was demonstrated in the NPJ (Hb (6.8) +WFM (6.8)) treatment, having
the highest levels of lipid oxidation (PV, TBARS) compared to the control and pH treatments.
The antioxidant activity of Spanish mackerel pressed juice was evident as the control (Hb (6.8)
+PJ (6.8) +WFM (6.8)) had significantly lower levels of lipid oxidation (PV, TBARS) compared
to all other treatments. The levels of oxidation in the Hb (3-6.8) treatment were not significantly
different from control; thus, it is likely that pro-oxidant activity of hemoglobin observed in the
NPJ treatment was lost with adjustment from pH 3 to 6.8. The WFM (3-6.8) treatment had the
highest levels of oxidations among thepH treatments. This finding pointed to significant changes
within the phospholipids, enabling for greater interaction with hemoglobin. A significant
increase in lipid oxidation with pH adjustment of Spanish mackerel PJ, compared to the control,
provides evidence that the antioxidative activity of pressed juice was lost. Additionally,
treatments containing adjusted pressed juice had equivalent levels of lipid oxidation. This result
shows that the pH adjustment of pressed juice has a significant effect on lipid oxidation
development in a WFM system. Low levels of lipid oxidation in the control agreed with the high
levels of antioxidant activity in the aqueous system, as measured by the TEAC and ORAC
139
methods. The results show that lipid oxidation incurred during acidic processing could be
primarily due to phospholipids changes, secondary to changes in hemoglobin and/or aqueous
extracts. Means to retain the antioxidative capacity of fish cytosolic extract with alkali and acidic
processing needs further investigation. The effects of low pH and readjustment to neutral pH on
pressed juice, hemoglobin, and myofibrillar extract (washed fish) warrant further investigation.
140
160
Control (pH 6.8)
Control (pH 6.8) w/o pressed juice
Hemoglobin (pH 3-6.8)
Pressed juice (pH 3-6.8)
Washed fish muscle (pH 3-6.8)
PV ( mmol LPH/kg tissue)
140
120
(a)
100
80
60
40
20
0
0
1
2
3
4
5
6
7
Time (days)
TBARS ( umol MDA/kg tissue)
100
Control (pH 6.8)
Control (pH 6.8) w/o pressed juice
Hemoglobin (pH 3-6.8)
Pressed juice (pH 3-6.8)
Washed fish muscle (pH 3-6.8)
90
80
70
(b)
60
50
40
30
20
10
0
0
1
2
3
4
5
6
7
Time (days)
Figure 5-1. The role of acified (pH 3) and pH 6.8 tissue components on lipid oxidation (A) Lipid
hydroperoxides and (B) TBARS in a washed model system. Each treatment was a
mixture of hemoglobin, and cytosolic extract (Pressed Juice), and washed muscle
from Spanish mackerel, except for the control without pressed juice. One of three
components for each treatment was first adjusted to pH 3 for one hour, followed by
readjustment to pH 6.8. The control sample components were subject to a pH of 6.8.
Samples were stored for 7 days at 4°C. Each sampling point is representative of the
mean and standard deviation from triplicate analysis.
141
160
Control (pH 6.8)
(Hemoglobin + Washed fish muscle) pH 3-6.8
(Hemoglobin + Pressed juice) pH 3-6.8
(Pressed juice + Washed fish muscle) pH 3-6.8
(Hemoglobin + Pressed juice+ Washed fish muscle) pH 3-6.8
PV ( mmol LPH/kg tissue)
140
120
100
(a)
80
60
40
20
0
0
1
2
3
4
5
6
7
Time (days)
TBARS ( umol MDA/kg tissue)
100
Control (pH 6.8)
(Hemoglobin + Washed fish muscle) pH 3-6.8
(Hemoglobin + Pressed Juice) pH 3-6.8
(Pressed juice + Washed fish muscle) pH 3-6.8
(Hemoglobin + Pressed Juice + Washed fish Muscle) pH 3-6.8
90
80
70
60
(b)
50
40
30
20
10
0
0
1
2
3
4
5
6
7
Time (days)
Figure 5-2. The role of two and three acified (pH 3) and neutralized (6.8) tissue components on
lipid oxidation (a) Lipid hydroperoxides and (b) TBARS in a stored washed model
system. Each treatment was a mixture of hemoglobin, and cytosolic extract (Pressed
Juice), Two to three components for each treatment were adjusted to pH 3 for one
hour, followed by readjustment to pH 6.8. The control sample components were
subjected to a pH of 6.8. Samples were stored for 7 days at 4°C. Each sampling point
is representative of the mean and standard deviation from triplicate analysis.
142
ORAC (umol Trolox equivalents/ mg protein)
H
b
(p
H
6.
08
PJ
)
(p
H
6.
6)
H
b
(p
H
H
3)
b
(p
H
6.
8)
PJ
(p
H
PJ
3)
(p
H
H
6.
b
(
pH 8)
PJ
(p
3PJ
H
6.
8)
PJ
3(
p
6
H
(p
.
8)
H
+H 3-6.
38)
6.
b
8)
(
p
PJ
H
+H
6.
(p
b
8)
H
(p
PJ
6.
H
8
3(p
)+
6.
H
H
8)
6.
b
8)
(
pH
+
H
6.
b
8)
(p
H
36.
8)
1.8
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
a
(A)
0 hour hold
1 hour hold
b
c
c
efg
f f
g
de
d
ef
defg
a
(B)
a
a
0 hour hold
a
1 hour hold
bc
bc
a
b
bcd
e
e
de
e
(p
H
6.
08
PJ
)
(p
H
6.
6)
H
b
(p
H
H
3)
b
(p
H
6.
8)
PJ
(p
H
3)
PJ
(p
H
H
6.
b
8)
(p
PJ
H
3(p
PJ
6.
H
8)
PJ
3(p
6.
H
(p
8)
3H
+H
6.
38)
6.
b
8)
(p
PJ
H
+H
6.
(p
b
8)
H
(p
PJ
6.
H
8)
3(p
+H
6.
H
8)
6.
b
8)
(p
+
H
H
6.
b
8)
(p
H
36.
8)
0.45
0.4
0.35
0.3
0.25
0.2
0.15
0.1 cde
0.05
0
H
b
TEAC (mM Trolox equivalents/ mg protein)
Treatments
Treatments
Figure 5-3. Antioxidant capacity (A) ORAC and (B) TEAC of Spanish mackerel hemoglobin and
pressed juice alone or combination with adjustment to pH 3 for one hour followed by
readjustment. Several combinations were implemented. Each bar is representative of
the mean and standard deviation from triplicate analysis. Different small letters
indicate a significant different (p< 0.05).
143
205 kDa
116 kDa
97 kDa
84 kDa
66 kDa
55 kDa
45 kDa
36 kDa
29 kDa
24 kDa
20 kDa
14 kDa
1
2
3
4
5
6
Figure 5-4. Protein composition of untreated Spanish mackerel pressed juice. Lanes 1 and 6:
molecular standards (5 uL). Lanes 2 and 5 represent Spanish Mackerel pressed juice
(5 uL). Lanes 3 and 4 represent Spanish mackerel homogenates at 10 and 15uL.
144
CHAPTER 6
CONCLUSION
The acid/alkali-aided process promotes utilization of by-catch and dark muscled fish
species for human food. Dark muscle fish species pose various processing challenges due to the
abundance of unstable lipids and heme proteins. To investigate the effects of the acid/alkali aided
processing on lipid oxidation in dark muscle fish species, Spanish mackerel was used as a model.
The effects of acid and alkaline processing on lipid oxidation in Spanish mackerel varied among
homogenates, isolates, and the washed fish muscle system.
Lipid oxidation was prevalent at low pH compared to high pH in the Spanish mackerel
homogenates. The initial quality of the Spanish mackerel, storage, and pH holding times all
significantly affected lipid oxidation development in the homogenates. Addition of ascorbic acid
and tocopherol to the acid homogenates, effectively inhibited lipid oxidation, but was not
affected by the addition of chelators, EDTA and citric acid. The pH readjustment from acidic to
neutral pH did not significantly affect the lipid oxidation in the homogenates; however, the
addition of tocopherol (0.1%) and ascorbic acid (0.1%) in combination with pH readjustment
proved to be highly oxidative. How adding ascorbic acid and tocopherol to the acid and alkali
homogenates affects the final proteins isolate needs further investigation.
Isolates were prepared from acid and alkaline processing of Spanish mackerel. Lipid
oxidation was comparable among the alkaline and acid isolates, with holding time and storage.
High levels of lipid oxidation were observed in the supernatant fractions from the acid, alkaline,
and control treatments. Alkaline isolates contained overall greater protein oxidation than the
acid and control isolates, as determined by carbonyl levels. Protein degredation occurring during
acid and alkali processing. Rheological testing provided information regarding isolate gel
strength. Greater gel strength was observed in the alkaline isolates, followed by the control and
145
acid isolates. Possible correlations between lipid and protein oxidation and gel strength warrant
further investigation. Additionally, the mechanisms by which protein and/or lipid oxidation
develop with pH adjustment during the isolation process should be investigated. The addition of
antioxidants in the final isolate could prevent lipid and protein oxidation, and thus affect isolate
quality.
The adjustment from pH 3 to pH 6.8 greatly affected all components in a washed Spanish
mackerel model system. The pH adjustment significantly inhibited hemoglobin’s catylitic
activity and pressed juice antioxidative activity. The washed fish muscle was significantly
affected by pH adjustment. This was evident in significantly higher levels of lipid oxidation in
WFM treatment (pH 3-6.8) compared to all other pH treatments. The addition of hemoglobin,
adjusted from pH 3 to 6.8 to the washed Spanish mackerel muscle, gave lipid oxidation levels
comparable to the control. Greater levels of oxidation were observed in the pH treatments in
which pressed juice (3-6.8) was added, compared to the control. The affect of pH adjustment
from extreme pH to “neutral” pH on a washed fish muscle system, containing pressed juice, has
not been investigated. Thus, there needs to be further understanding of the pH affects on the key
components in fish muscle and how this relates to lipid oxidation development. Methods to
better access the antioxidative changes in the washed fish muscle require investigation. An
understanding of components contributing to the high antioxidative acitivty in pressed juice is
also needed.
Results from this study point to various changes in lipid and protein oxidation due to
alkaline and acidic treatment. Acid treatment resulted in high levels of lipid oxidation in the
homogenates and in the washed fish model system. Acid and alkaline processing proved to be
equally lipid oxidative in the preparation of Spanish mackerel isolates; however, greater levels of
146
protein oxidation developed with alkaline treatment. Thus, measures of lipid and protein
oxidation are imperative to understanding how fish protein structure and quality is altered.
147
LIST OF REFERENCES
(1) Cushman Jr., J. H., Cuts Sought in Wasteful Fish Kills. New York Times January 13,
1998,p2.
(2) Windsor, M.; Barlow, S., Introduction to Fishery Byproducts. Fishing news (Books) Ltd:
London, 1981.
(3) FAO, The production of Fish Meal and Oil. In FAO Fisheries Technical Paper, No.
142,Rome, 1975.
(4) Okada, M., Utilization of small pelagic species for food. In Proceedings of the Third National
Technical Seminar on Mechanical Recovery and Utilization of Fish Flesh, Martin, R. E.,
Ed. National Fisheries Institute: Washington, D.C., 1980; pp 265-282.
(5) Hultin, H. O., Oxidation of Lipids in Seafoods. In Seafoods: Chemistry, Processing
Technology and Quality, Shahidi, F.; Botta, J. R., Eds. Blacki Academic: Glasgow, 1994;
pp 49-74.
(6) Hultin, H. O.; Kelleher, S. D., Surimi Processing from Dark Muscle Fish In Surimi and
Surimi Seafood, Park, J. W., Ed. Marcel Dekker: New York, 2000; pp 59-77.
(7) Hultin, H. O.; Kelleher, S. D.; Feng, Y.; Kristinsson, H. G.; Richards, M. P.; Undeland, I. A.;
Ke, S. High Efficiency Alkaline Protein Extraction. 60,230,397, 2000.
(8) Hultin, H. O.; Kelleher, S. D. Process for Isolating a Protein Composition from a Muscle
Source and Protein Composition. 6,005,073, 1999.
(9) Shewfelt, R. L., Fish Muscle Lipolysis- A review. J Food Biochem 1981, 5, 79-100.
(10) Gandemer, G., Lipids and Meat Quality: Lipolysis, Oxidation, Maillard Reaction and
Flavour. Sci. Aliment. 1999, 19, 439-458.
(11) Hultin, H. O.; Kristinsson, H. G.; Lanier, T. C., Process for recovery of functional proteins
by pH shifts. In Surimi and Surimi Seafood, 2nd ed.; Marcel Dekker: New York, 2004.
(12) Kristinsson, H. G., Acid unfolding of flounder hemoglobin: Evidence for a molten globular
state with enhanced pro-oxidative activity. J. Agric Food Chem 2002, 50, (26), 7669-7676.
(13) Kristinsson, H. G.; Demir, N., Functional Fish Protein Ingredients from Fish Species of
Warm and Temperate Waters: Comparison of Acid and Alkali-Aided Processing vs.
Conventional Surimi Processing. In Advances in Seafood Byproducts, Alaska Sea Grant
College Program: 2003.
(14) Kelleher, S. D.; Feng, Y. M.; Hultin, H. O.; Livingston, M. B., Role of initial muscle pH on
the solubility of fish muscle proteins in water. Food Biochemistry 2004, 28, (4), 279-292.
(15) Hultin, H. O., Controlling lipid oxidation during processing at high and low pH. In Annual
IFT Meeting, Las Vegas, Nevada, 2004.
(16) Undeland, I. A.; Kelleher, S. D.; Hultin, H. O., Recovery of functional proteins from herring
(Clupea haerangus) light muscle by an acid or alkaline solubilization process. J. Agric.
Food Chem. 2002, 50, 7371-7379.
(17) Kristinsson, H. G.; Theodore, A. E.; Demir, N.; Ingatottir, B., Acid-aided and alkali-aided
process. Food Chemistry and Toxicology 2005, 70, (4), C298-C306.
(18) Kristinsson, H. G.; Hultin, H. O., Effect of low and high pH treatment on the functional
properties of cod muscle proteins. J. Agric Food Chem 2003, 51, 5103-110.
(19) Theodore, A. E.; Kristinsson, H. G., A comparative study between acid- and alkali-aided
processing and surimi processing for the recovery of proteins from channel catfish muscle.
Journal of Food Science 2003, 70, (4), C298-C306.
148
(20) Reinitz, G. L.; Orme, L. E.; Hitzel, F. N., Variations of body composition and growth
among strains of rainbow trout (Salmo gairdneri). Trans. Am. Fish Soc. 1979, 108, 204207.
(21) Kiessling, A. T.; Aasgaard, T.; Storebakken, T.; Johansson, L.; Kiessling, K. H., Changes in
the structure and function of the epaxial muscle of rainbow trout (Oncorhynchus mykiss)
in relation to ration and age. Aquaculture 1991, 93, 373-387.
(22) Huss, H. H., Quality and Quality Changes in Fresh Fish. In Fisheries Technical Papers T-34,
ed.; FAO: Rome, 1994; p 195.
(23) Sevanian, A.; Hochstein, P., Mechanisms and consequences of lipid peroxidation in
biological systems. Annu. Rev. Nutr. 1985, 5, 365-390.
(24) Sies, H., Biochemistry of oxidative stress. Angew Chem. Int. Ed. Engl. 1986, 25,
1058-1071.
(25) Ackman, R. G., Fish Lipids. Part 1. In Advances in Fish Science and Technology, Fishing
News Books Ltd: England, 1980; pp 86-103.
(26) Wasasundara, U. N.; Shahidi, F., Canola extracts as an alternative antioxidant for canola oil.
J. Am. Oil Chem. Soc. 1994, 71, 817-822.
(27) Bateman, L.; Hughs, H.; Morris, A. L., Hydroperoxide decomposition in relations to the
initiation of radical chain reactions. Disc Faraday Soc. 1953, 4, 190-194.
(28) Gutteridge, J. M. C., Lipid peroxidation: Some Problems and Concepts. In Oxygen Radicals
and Tissue Injury, Federation of American Societies of Experimental Biology: Bethesda,
MD, 1988; pp 9-19.
(29) Pearson, A. M.; Gray, J. I.; Wolzak, A. M.; Horenstein, N. A., Safety implications of
oxidized lipids in muscle foods. Food Technology 1983, 37, (7), 121-129.
(30) Gray, J. I.; Pearson, A. M., Rancidity and warmed-over flavor. Meat. Res. 1987, 3, 221-269.
(31) Erickson, M. C., Lipid Oxidation of Muscle Food. In Quality attributes of Muscle Foods,
Kluwer Academic: Boston, 1999; pp 365-368.
(32) Medina, I.; Saeed, S.; N., H., Enzymatic oxidative activity in sardine (Sardina pilchardus)
and herring (Clupea harengus) during chilling and correlation with quality. European Food
Researh and Technology 1999, 210, (1), 34-38.
(33) Mead, J. F., Free Radical in Biology and Medicine. Academic Press: New York, 1976; Vol.
1, p 51.
(34) Floyd, R. A.; Lewis, A., Hydroxyl free radical formation from hydrogen peroxide by ferrous
iron-nucleotide complexes. Biochemistry 1983, 22, 2645.
(35) Richards, M. P.; Ostdal, H.; Andersen, H. J., Deoxyhemoglobin-mediated lipid oxidation in
washed fish muscle. J Agri Food Chem 2002, 50, (5), 1278-1283.
(36) Younathan, M. T.; Watts, B. M., Relationship of meat pigments to lipid oxidation. Food
Res. 1959, 24, 728-733.
(37) Kanner, J.; Harel, S.; Granit, R., Initiation of membranal lipid peroxidation by activated
methemoglobin and metmyoglobin. Arch. Biochem. Biophys. 1985, 289, 130.
(38) Hultin, H. O.; Kelleher, S. D.; Feng, Y.; Kristinsson, H. G.; Richards, M. P.; Undeland, I.
A.; Ke, S. High Efficiency Alkaline Protein Extraction. 2000.
(39) Rhee, K. S., Natural antioxidants for meat products. In Warmed-over flavor of meat, A.J., S.
A.; Bailey, M. E., Eds. Academic Press: Orlando, 1987.
(40) Asgar, A.; Gray, J.; Buckley, D.; Pearson, A. M.; Booren, A. M., Perspectives on warmed
over flavor. Food Technology 1988, 42, (6), 102-108.
149
(41) Kanner, J.; Salon, M. A.; Harel, S.; Shegalovich, I., Lipid peroxidation of muscle food: the
roles of cytosolic fraction. J Agri Food Chem 1991, 39.
(42) Richards, M. P.; Hultin, H. O., Effect of pH on lipid oxidation using trout hemolysate as a
catalyst: a possible role for deoxyhemoglobin. J. Agric Food Chem 2002, 48, 3141-3147.
(43) Richards, M. P.; Hultin, H. O., Effect of pH on lipid oxidation using trout hemolysate as a
catalyst: a possible role for deoxyhemoglobin. J Agri Food Chem 2000, 48, 3141-3147.
(44) Connell, J. J., Control of Fish Quality. Whitefriars Press, Ltd.: London, 1975; p 333.
(45) Wheaton, F. W.; Lawson, T. B., Processing Aquatic Food products. John Wiley and Sons:
New York, 1985.
(46) Everse, J.; Hsia, N., The toxicities of native and modified hemoglobins. Free Rad. Bio. Med
1999, 22, 1075-1099.
(47) Baron, C. P.; Andersen, H. J., Myoblobin-induced lipid oxidation: A review. J Agri Food
Chem 2002, 50, 3887-3897.
(48) Nawar, W., Lipids. In Food Chem., 3 ed.; Fenemma, O. R., Ed. Marcel Dekker Inc.: New
York, 1996; pp 225-320.
(49) Gutteridge, J. M. C.; Halliwell, B., The measurement and mechanism of lipid peroxidation
in biological systems. Trends Biochem. Sci. 1990, 15, 129-135.
(50) Weber, R. E., Adaptions of oxygen transport: Lessons from fish hemoglobins. In
Hemoglobin function in vertebrates, Prisco, G. d.; Giardina, B.; Weber, R. E., Eds.
Springer: Verlag, Italia, 2000; pp 23-27.
(51) Kristinsson, H. G.; Hultin, H. O., The effect of acid and alkali unfolding and subsequent
refolding on the pro-oxidative activity of trout hemoglobin. J Agri Food Chem. 2004, 52,
(17), 5482-5490.
(52) Levy, A.; Rifkind, J. M., Low-temperature formation of distal histidine complex in
hemoglobins: probe for heme pocket flexibility. Biochemistry 1985, 24, 6050-6054.
(53) Astatke, M.; McGee, W. A.; Parkhurst, L. J., A flow procedure to determine oxygen binding
isotherms for low affinity and easily oxidizable hemoglobins. Comp. Biochem. Physiol.
1992, 101 B, 683-688.
(54) Richards, M. P.; Dettmenn, M. A., Comparative analysis of different hemoglobins:
autoxidation, reaction with peroxide, and lipid oxidation. . J Agri Food Chem 2003, 51,
(13), 3886-3891.
(55) Richards, M. P.; Dettmenn, M. A., Comparative analysis of different hemoglobins:
Autoxidation, reaction with peroxide, and lipid oxidation. J Agri Food Chem 2003, 51,
(13), 3886-3891.
(56) Reeder, B. J.; Wilson, M. T., The effects of pH on the mechanism of hydrogen peroxide and
lipid hydroperoxide consumption by myoglobin: A role for the protonated ferryl species.
Free Rad. Bio. Med 2001, 30, (1311-1318), 30.
(57) Undeland, I. A.; Richards, M. P., Added triacylglycerols do not hasten hemoglobinmediated lipid oxidation in washed minced cod muscle. J Agri Food Chem 2002, 50,
6847-6853.
(58) Richards, M. P.; Hultin, H. O., Rancidity development in a fish muscle model system as
affected by phospholipids. J. Food Lipids 2001, 8, 215-230.
(59) Porter, P. J.; Kennish, J. M.; Kramer, D. E., The effects of exsanguination of sockeye
salmon on the changes of lipid composition during frozen storage. In Seafood Science and
Technology, Bligh, E. D., Ed. Fishing News Books: Oxford, 1992; pp 76-83.
150
(60) Shikama, K., The molecular mechanism of autoxidation for myoglobin and hemoglobin: an
enerable puzzle. Experientia 1998, 98, 1357-1373.
(61) Rogers, M. S.; Patel, R. P.; Reeder, B. J.; Sarti, P.; Wilson, M. T.; Alayash, A. I., Prooxidant effects of cross-linked haemoglobins explored using lyposome and cytochrome c
oxidase vesicle model membranes. Biochemistry 1995, 310, 827-833.
(62) Undeland, I. A.; Hall, G.; Karin, W.; Gangby, I.; Rutgersson, A., Preventing lipid oxidation
during Recovery of functional proteins from herring (Clupea harengus) fillets by an aid
solubilization process. J Agri Food Chem 2005, 53, 5625-5634.
(63) Howell, N.; Saeed, S., Effect of lipid oxidation and frozen storage on muscle proteins of
Atlantic Mackerel. j Science of Food and Agriculture 2002, 82, 579-586.
(64) Witting, L. A., Vitamin E and lipid antioxidants in free-radical initiated reactions. In Free
Radicals in Biology, Priyor, A., Ed. Academic Press: New York, 1995; Vol. IV.
(65) Chow, C., Vitamin E and oxidative stress. Free Rad. Bio. Med 1991, 11, 215-21.
(66) Cadenas, E.; Packer, L., Handbook of antioxidants. Marcel Dekker, Inc.: New York, 1996; p
545-91.
(67) Undeland, I. A.; Hall, G.; Wendin, K.; Gangby, I.; Rutgersson, A., Preventing Lipid
Oxidation during Recovery of Functional Proteins from Herring (Clupea harengus) fillets
by an Acid Solubilization Process. J Agri Food Chem 2005, (53), 5625-5634.
(68) Kelleher, S. D.; Silva, L. A.; Hultin, H. O., Inhibition of oxidation during processing of
washed, minced Atlantic mackerel. J. Food Sci 1992, 57, 1103-1119.
(69) Richards, M. P.; Kelleher, S. D.; Hultin, H. O., Effect of washing with or without
antioxidants of quality retention of mackerel fillets during refrigerated and frozen storage.
J Food Sci. 1998, 46, (4363-4371).
(70) Howell, N.; Saeed, S., Rheological and differential scanning calorimetry studies on
structural and textural changes in frozen Atlantic mackerel (Scomber scombrus). J Science
of Food and Agriculture 2004, 82, 579-586.
(71) Dean, R. T.; Fu, S.; Stocker, R.; Davies, M. J., Biochemistry and pathology of radicalmediated protein oxidation. Biochem. J. 1997, 324, 1-18.
(72) Stadtman, E. R., Oxidation of free amino acids and amino acid residues in proteins by
radiolysis and by metal-catalyzed reactions. Annu. Rev. Biochem. 1993, 62, 797-821.
(73) Davies, M. J.; Dean, R. T., Radical-mediated protein oxidation: from chemistry to medicine.
Oxford Univ. Press: Oxford, 1997.
(74) Hawkins, C. L.; Davies, M. J., Generation and propagation of radical reactions on proteins.
Biochem. Biophys. Acta 2001, 1504, 196-219.
(75) Halliwell, B., Free radicals, antioxidants and human disease: curiosity, cause or
consequence? Lancet 1994, 344, 721-724.
(76) Shacter, E., Quantification and significant of protein oxidation in biological samples. In
Drug Metab. Rev., 2000; Vol. 32, pp 307-326.
(77) Hwang, E.; Kim, G.-H., Review: Biomarkers for oxidative stress status of DNA, lipids, and
proteins in vitro and in vivo cancer research. Toxicology 2007, 229, 1-10.
(78) Schaich, K. M., Free radical initiation in proteins and amino acids by ionizing and
ultraviolet radiations and lipid oxidation. III: Free radical transfer from oxidizing lipids.
CRC Crit. Rev. Food Sci. Nutr. 1980, 13, (189-244).
(79) Kanner, J., Oxidative processes in meat and meat products: quality implication. Meat Sci.
1994, 36, 169-180.
151
(80) Johns, A. M.; Birkinshaw, L. H.; Ledward, D. A., Catalysits of lipid oxidation in meat
products. Meat. Sci. 1989, 25, (209-220).
(81) Haberland, M. E.; Fong, D.; Cheng, L., Malondialdehyde-altered protein occurs in atheroma
of watanabe hertable hyperlipidemic rabbits. Science 1988, 241, 215-218.
(82) Kim, J. G.; Sabbagh, F.; Santanam, N.; Wilcox, J. N.; Medford, R. M.; Parthasarathy, S.,
Generation of polyclonal antibody against lipid peroxide-modified proteins. Free Rad. Bio.
Med 1997, 23, 251-259.
(83) Requena, J. R.; Fu, M. X.; Ahmed, M. U.; Jenkins, A. J.; Lyons, T. J.; Baynes, J. W.;
Thorpe, S. R., Quantification of malondialdehyde and 4 hydroxynonenal adducts to lysine
residue in native and oxidized human low density lipoprotein. Biochem. 1997, 322,
317-325.
(84) Dean, R. T.; Fu, S.; Stocker, R.; Davies, M. J., Biochemistry and pathology of radicalmediated protein oxidation. Biochem. 1997, 324, 1-18.
(85) Berlett, B. S.; Stadtman, E. R., Protein oxidation in aging, disease, and oxidative stress. J.
Biol. Chem. 1997, 272, 201313-6.
(86) Xiong, L., Protein Oxidation and Implication for muscle food Quality. In Antioxidants in
Muscle Foods, Decker, E. A.; Faustman, C.; Lopez-Bote, C. J., Eds. John Wiley and Sons,
Inc.: New York, 2000.
(87) Xiong, Y. L.; Decker, E. A., Alternations in muscle protein functionality by oxidative and
antioxidative processes. J. Muscle Foods 1995, 6, 139-160.
(88) Martinaud, A.; Mercier, Y.; Marinova, P.; Tassy, C.; Gatellier, P.; Renerre, M., Comparison
of oxidative processed on myofibrillar proteins from beef during maturation and by
different model oxidation systems. J. Agric. Food Chem. 1997, 45, 2481-2487.
(89) Garrison, W., Reaction mechanisms in radiolysis of peptides, polypeptides, and proteins.
Chem. Rev 1987, 87, 381-398.
(90) Garrison, W.; Jayko, M. E.; W., B., Radiation-induced oxidation of protein in aqueous
solution. Rad Research 1962, 16, 483-502.
(91) Swallow, A., Effect of ionizing radiation on proteins, RCO groups, peptide bond cleavage,
inactivation,- SH oxidation. John Wiley and Sonds: New York, 1960; p 211-224.
(92) Schuessler, H.; Schilling, K., Oxygen effect in radiolysis of proteins. Part 2. Bovine Serum
albumin. Int. J Radiat. Biol 1984, 45, 267-281.
(93) Radi, R.; Bush, K. M.; Cosgrove, T. P.; Freman, B. A., Reaction of xanthine oxidasederived oxidants with lipid and protein of human plasma. Arch. Biochem. Biophys. 1991,
286, 117-125.
(94) Lii, C. K.; Chai, Y. C.; Zhao, W.; Thomas, J. A.; Hendrich, S., S-thiolation and irreversible
oxidation of sulfhydrls on carbonic anhydrase III during oxidative stress: A method for
studying protein modification in intact cells and tissues. Arch. Biochem. Biophys. 1994,
308, 231-239.
(95) Vogt, W., Oxidation of methionyl residues in proteins: tools, targets, and reversal. Free Rad.
Bio. Med 1995, 18, 93-105.
(96) Levine, R. L.; Mosoni, L.; Berlett, B. S.; Stadtman, E. R., Methionine residues as
endogenous antioxidants in proteins. Proc Natl Acad Sci 1996, 93, 15036-15,040.
(97) Reddy, V. Y.; Desorchers, P. E.; Pizzo, S. V.; Gonias, S. L.; Sahakian, J. A.; Levine, R. L.;
Weiss, S. J., Oxidative dissociations of human alpha 2-maroglobulin tetramers in
dysfunctional dimers. J. Biol. Chem. 1994, 269, 4683-4691.
152
(98) Gitlin, G.; Tsarbopoulos, A.; Patel, S. T.; Sydor, W.; Pramanik, B. N.; Jocobs, S., Isolation
and characterization of a monomethioninesulfoxide variant of interferon alpha-2b. Pharm
Res. 1996, 13, 762-769.
(99) Stadtman, E. R.; Levine, R. L., Free radical-mediated oxidation of free amino acids and
amino acid residues in proteins. Amino Acids 2003, 25, 207-218.
(100) Uchida, K.; Kato, Y.; Kawakishi, S., A novel mechanism for oxidative damage of prolyl
peptides induced by hydroxyl radicals. Biochem. Biophys. Res. Commun 1990, 169,
265-271.
(101) Gardner, H. W., Lipid hydroperoxide reactivity with proteins and amino acids: A review. J
Agri Food Chem 1979, 30, 1204-1208.
(102) Srinivasan, S.; Xiong, Y. L., Sulfhydryls in antioxidant-washed beef heart surimi. J.
Muscle foods 1997, 8, 252-263.
(103) Verziji, N.; De Groot, J.; Oldehinkel, E.; Bank, R. A.; Thorpe, S. R.; Baynes, J. W.;
Bayliss, M. T.; Bijlsma, W. J.; Lafeber, F.; Tekoppele, J. M., Age accumulation of Mailard
products in human articular cartilage collagen. Biochem. 2000, 350, 381-387.
(104) Lee, H. G.; Lee, C. M.; Chung, H.; Lavery, S. A., Sodium ascorbate affects surimi-gel
forming properties. . J. Food Sci. 1992, 57, (1343-1347.).
(105) Uschida, K.; Stadtman, E. R., Covalent attachment of 4-hydroxynoneal to glyceraldehyde3-phosphate dehydrogenase. J. Biol. Chem. 1993, 268, 6388-6393.
(106) Stadtman, E. R.; Oliver, C. N., Metal-catalyzed oxidation of proteins. J. Biol. Chem. 1991,
(266), 2005-2008.
(107) Stadtman, E. R., Metal ion-catalyzed oxidation of proteins: Biochemical mechanism and
biological consequences. Free Radical Biol. and Med. 1990, 9, (315-325).
(108) Korashi, R.; Kohen, R.; Katzhendler, J.; Chevion, M., Iron mediates paraquat toxicity in
Escherichia coli. J. Biol. Chem. 1986, 261, 12,472-12,476.
(109) Czapski, G., On the use of OH radical scavengers in biologicals systems. Issrael J. Chem.
1984, 24, 29-32.
(110) Samuni, A.; Aronovitch, J.; Godiner, D.; Chevion, M.; Czapski, G., Cytoxicity of vitamin
C and metal ions. A site-specific Fenton mechanism. Eur. J. Biochm. 1983, 137, 119-124.
(111) Amici, A.; Levine, R. L.; Tsai, L.; Stadtman, E. R., Conversion of amino acid residues in
proteins and amino acid homopolymers to carbonyl derivatives by meta-catalyzed
oxidation reactions. J. Biol. Chem. 1989, 264, 3341-3346.
(112) Levine, R. L.; Garland, D.; Oliver, c. N.; A., A.; Climent, I.; Lenz, A. g.; Ahn, B. W.;
Shaltiel, S.; Stadtman, E. R., Determination of carbonyl content in oxidatively modified
proteins. Methods of Enzym. 1990, 186, (464-477).
(113) Schaich, K.; Karel, M., Free radical reactions of peroxidizing lipid with amino acids and
proteins: and ESR study. Lipids 1975, 11, 392-400.
(114) Roubal, W. T.; A.L., T., Damage to proteins, enzyme, and amino acids by peroxidizing
lipids. Arch. Biochem. Biophys. 1966, 113, 5-8.
(115) Stadtman, E. R., Protein oxidation in aging and age-related diseases. Ann. N.Y. Acad. Sci.
2001, 928, 22-38.
(116) Coux, O.; Tanaka, K.; Goldberg, A., Structure and functions of 20S and 26S proteosomes.
Annu Rev Biochem. 1996, 65, 801-847.
(117) Moskovitz, J.; Berlett, B. S.; Poston, J. M.; Stadtman, E. R., Methionine sulfoxide
reductase in antioxidant defense. Methods Enzymol. 1999, 300, 239-244.
153
(118) Starke-Reed, P. E.; Oliver, C. N., Protein oxidation and proteolysis during aging and
oxidative stress. Arch Biochem. Biophys. 1989, 275, 559-567.
(119) Carney, J. M.; Starke-Reed, P. E.; Oliver, C. N.; Landum, R. W.; Chang, M.; Wu, J.;
Floyd, R. A., Reversal of age-related increase in brain protein oxidation, decrease in
enzyme activity, and loss in temporal and spatial memory by chronic administration of the
spin-trapping compound N-tert-butyl-alpha-phenylnitrone. Proc Natl Acad Sci 1991, 88,
(3633-3636).
(120) Levine, R. L., Carbonyl modified proteins in cellular regulation, aging, and disease. Free
Radical Biol. and Med. 2002, 32, (9), 790-796.
(121) Srinivasan, S.; Hultin, H. O., Chemical, physical, and functional properties of cod proteins
modified by nonenzymatic free-radical generating system. J. Agric. Food Chem. 1997, 45,
310-320.
(122) Smith, D. M., Functional and biochemical changes in deboned turkey due to frozen storage
and lipid oxidation. J. Food Sci 1987, 52, 22-27.
(123) Wan, L.; Xiong, Y. L.; Decker, E. A., Inhibition of oxidation during washing improves the
functionality of bovine cardiac myofibrillar protein. J. Agric Food Chem 1993, 41, 22672271.
(124) Xiong, Y. L.; Decker, E. A.; Robe, G. H.; Moody, W. G., Gelation of crude myofibrillar
protein isolated from beef heart under antioxidative conditions. J. Food Sci 1993, 58,
(1241-1244).
(125) Wang, B. W.; Xiong, Y. L., Functional Stability of antioxidant-washed, cryoprotectanttreated beef heart surimi during frozen storage. J. Agric Food Chem 1998, 41, 2267-2271.
(126) Liu, G.; Xiong, Y. L., Thermal denaturation and in vitro digestibility of oxidized myosin.
1997; Vol. 56-4.
(127) Stone, A. P.; Stanley, D. W., Mechanisms of fish muscle gelation. Food Research
International 1992, 25, 381-388.
(128) Srinivasan, S.; Xiong, L., Thermal stability of oxidized muscle proteins as analyzed by
differential scanning calorimentry. 1996 IFT Annual Meeting: Book of Abstracts Abst. No.
38-5.
(129) Li., S. J.; King, A. J., Lipid oxidation and myosin enaturation in dark chicken meat. J.
Agric. Food Chem. 1996, 44, 3080-3084.
(130) Wang, B. W.; Xiong, Y. L.; Srinivasan, S., Chemical Stability of antioxidant washed beef
heart surimi during frozen storage. J. Food Sci 1997, 62, 939-945, 991.
(131) Buttkus, H., The reaction of myosin with malonaldehyde. J. Food Sci. 1967, 32, 432-434.
(132) Dillard, C. J.; Tappel, A. L., Fluorescent products from the reaction of peroxidizing PUFA
with phophtdyl ethanolamine and phenylalanine. Lipids 1973, 8, 183-189.
(133) Decker, E. A.; Xiong, Y. L.; Clavert, J. T.; Crum, A. D.; Blanchard, S. P., Chemical
physical and functional properties of oxidized turkey white muscle myofibrillar proteins. J
Agri Food Chem 1993, 41, 186-189.
(134) Jarenback, L.; Liljemark, A., Ultrastructural changes during frozen storage of cod. III:
Effects of linoleic acid and linoleic acid hydroperoxides on myofibrillar proteins. J. Food
Technol. 1975, 10, 437-452.
(135) Xiong, Y. L., Oxidative Processes in meat. 49th Annual Reciprocal Meat Conference
1996, pp 79-86.
154
(136) Srinivasan, S.; Xiong, Y. L.; Decker, E. A., Inhibition of Protein and Lipid Oxidation in
Beef heart Surimi-like material by Antioxidants and Combinations of pH, NaCL, and
buffer type in the washing media. J Agri Food Chem 1996, 44, 119-125.
(137) Liu, G.; Xiong, Y. L., Rheological properties of oxidatively modified muscle, whey and
soy protein composite systems. 1998 IFT Annual Meeting: Book of Abstracts, Abst. No.
46C-1.
(138) Eymard, S.; Carcouet, E.; Rochet, M.; Dumay, J.; Chopin, C.; Claude, G., Development of
lipid oxidation during manufacturing of horse mackerel surimi. J Science of Food and
Agriculture 2005, 85, 1750-1756.
(139) Undeland, I. A.; Richards, M. P.; Hultin, H. O., Added triacylglycerols do not hasten
hemoglobin-mediated lipid oxidation in washed minced cod muscle. J. Agric Food Chem
2001, 50, 6847-6853.
(140) Lee, C. M.; Trevino, B.; Chaiyawat, M. A., A simple and rapid solvent extraction method
for determining total lipids in fish tissue. J. AOAC Int. 1996, 79, 487-492.
(141) Shantha, N. C.; Decker, E. A., Rapid, sensitive, iron-based. spectrophotometric methods
for determination of peroxide values of food. lipids. . J. AOAC Int. 1994, 77, (2), 421-424.
(142) Buege, J. A.; Aust, S. D., Microsomal lipid peroxidation. In Methods Enzymol., Fleischer,
S.; Packer, L., Eds. Academic Press: New York, 1978; p 302.
(143) Petillo, D. Antioxidant losses in the light ad dark muscle of Atlantic mackerel (Scomber
scrombrus). Massachusetts, Amherst, 1996.
(144) Hamilton, R. J., The chemistry of the Rancidity of foods. In Rancidity in foods, Allen, J.
C.; Hamilton, R. J., Eds. Chapman and Hall: New York, 1994.
(145) Monahan, F. J., Chapter 1. Oxidation of Lipids in Muscle Foods: Fundamental and
Applied Concerns. In Antioxidants in Muscle Foods: Nutritional Strategies to Improve
Quality, Decker, E. A.; Faustman, C.; Lopez-Bote, C. J., Eds. John Wiley and Songs, Inc.:
New York, 2000; p 499.
(146) Takama, K., Changes in flesh lipids of fish during frozen storage. Bull. Fac. Fish. Hokaido
University. 1974, V, (25), 256-263.
(147) Sato, K.; Hegarty, G. R., Warmed-over flavor in cooked meats. J Food Sci. 1971, 36,
1098-1102.
(148) Decker, E. A.; Livisay, S. A.; Zhou, S., Mechanisms of Endogenous Skeletal Muscle
Antioxidants: Chemical and Physical Aspects. In Antioxidants in Muscle Foods, Decker,
E. A.; Faustman, C.; Lopez-Bote, C. J., Eds. John Wiley and Sons, Inc.: New York, 2000.
(149) Uchida, K.; Stadtman, E. R., Modifications of histidine residues in proteins by reaction
with 4- hydroxynonenal. Biochemistry 1992, 89, 4544-4548.
(150) Eisenbrand, G.; Schuhmacher, J.; Golzer, P., The influence of glutathione and detoxifying
enzymes on DNA damage induced by 2-alkenal in primary rat hempatocytes and hum
lymphhoblastoid cells. Chem. Res Toxicol. 1995, 8, (1), 40-46.
(151) Alsavar, C.; Quantick, P. c.; Grigor, J. M., Aroma compounds of fresh and stored Mackerel
(Scomber scombrus) In Chapter 5 in Flavor and lipid chemistry of Seafoods., Shahidi, F.;
Cadwallader, K. R., Eds. American Chemical Society: Washington, 1997; pp 39-54.
(152) Kristinsson, H. G. Conformation and functional changes of hemoglobin and myosin
induced by pH: functional role in fish quality. University of Massachusetts, Amherst,
2002.
155
(153) Richards, M. P.; Kelleher, S. D.; Hultin, H. O., Effect of washing with or without
antioxidants on quality retention of mackerel fillets during refrigerated and frozen storage.
J Food Sci. 1998, 46, (4363-4371).
(154) Richards, M. P.; Hultin, H. O., Contributions of blood and blood components to lipid
oxidation in fish muscle. J Agri Food Chem 2002, 50, 555-564.
(155) Undeland, I. A.; Hultin, H. O.; Richards, M. P., Aqueous Extracts from Some Muscles
Inhibit Hemoglobin-Mediated Oxidation of Cod Muscle Membrane Lipids. J. Agric Food
Chem 2003, 51, (3111-3119).
(156) Riggs, A., Properties of Hemoglobin. Academic Press: New York, 1970; Vol. IV, p
209-252.
(157) Richards, M. P.; Hultin, H. O., Hemolysates from mackerel, herring and trout promote
lipid oxidation at different rates. Fish Sci. 2003, 69, 1298-1300.
(158) Hargrove, M. S.; Krzywda, S.; Wilkinson, A. J.; Dou, Y.; Ikedasaito, M.; Olson, J. S.,
Stability of myoglobin - A model for the folding of heme-proteins. Biochemistry 1994, 33,
(39), 11767-11775.
(159) Kristinsson, H. G., Changes in trout hemoglobin conformations and solubility after
exposure to acid and alkali pH. J Agri Food Chem 2004, 52, 3633-3643.
(160) Alayash, A. I., Hemoglobin-based blood substitutes: oxygen carriers, pressor agents, or
oxidants? Nat. Biotchnol. 1999, 17, 545-549.
(161) Pazos, M.; Medina, I.; Hultin, H. O., Effect of pH on Hemoglobin-Catalyzed lipid
Oxidation in Cod Muscle Membranes in Vitro and in Situ. J Agri Food Chem 2005, 53,
3605-3612.
(162) Richards, M. P.; Modra, A. M.; Li., R., Role of deoxyhemoglobin in lipid oxidation of
washed cod muscle mediated by trout, poultry and beef hemoglobins. Meat Sci. 2002, 62,
157-163.
(163) Lee, S.; Joo, S.; Phillips, A. L.; Faustman, C. In Oxymyoglobin and lipid oxidation in
yellowfin tune (Thunnus albacares). Annual Meeting and Food Expo., Anaheim, A, 2002;
Anaheim, A, 2002.
(164) Goto, Y.; Fink, A. L., Acid-induced folding of heme proteins. Methods of Enzym. 1994,
232, 3-15.
(165) Cilliard, J.; Cilliard, P.; Cormier, M., Effect of experimental factors on the prooxidant
behaviour of alpha-tocopherol. J. A. Oil Chem. Soc. 1980, 57, 255-261.
(166) Benedict, R. C.; Strange, E. D.; Swift, C. E., Effect of lipid antioxidants on the stability of
meat during storage. J. Agric Food Chem 1975, 23, 167-173.
(167) WATTS, B. M., Oxidative rancidity and discoloration in meat. J. Advances Food Research
1954, 5, (1).
(168) Halliwell, B.; Gutteridge, J. M. C., Oxygen free radicals and iron in relation to biology and
medicine: Some problems and concepts. Arch. Biochem. Biophys. 1986, 246, 501-514.
(169) Kanner, J., Mechanism of nonenzymatic lipid peroxidation in muscle foods. In Lipid
Oxidation in Foods, St. Angelo, A. J., Ed. American Chemical Society Books:
Washington, D.C., 1992; Vol. 500, pp 55-73.
(170) Richards, M. P.; Li., R., Effects of released iron, lipid peroxides, and ascorbate in trout
hemoglobin-mediated lipid oxidation of washed cod muscle. J. Agric. Food Chem. 2004,
52, 4323-4329.
(171) Buettner, G., In the absence of catalytic metals, ascorbate does not autoxidize at pH 7:
Ascorbates as a test of catalytic metals. J. Biochem. Biophys. Methods 1988, 16, 27-40.
156
(172) Harris, R. S., General Discussion on the stability of nutrients. In Nutritional evaluation of
Food Processing, Harris, R. S.; von Loesecke, H., Eds. AVI: Westport, CT., 1971; pp 1-4.
(173) Gregory III, J. F., Vitamins. In Food Chem., 3rd ed.; Fenemma, O. R., Ed. Marcel Dekker,
Inc.: New York, 1996; pp 532-609.
(174) Lee, S. H.; Oe, T.; Blair, I. A., Vitamin C induced-induced decomposition of lipid
hydroperoxides to endogenous genotoxins. Science 2001, 292, 2083-2086.
(175) Buettner, G., The pecking order of free radicals and antioxidants: lipid peroxidation, alphatocopherol, and ascorbate. . Arch. Biochem. Biophys. 1993, 300, 535-543.
(176) Traber, M. G., Determinants in plasma vitamin E concentrations. Free Radical Biol. Med.
1994, 16, (229-239).
(177) Petillo, D.; HO, H.; J, K.; WR, A., Kinetics of antioxidant loss in mackerel light and dark
muscle J Agri Food Chem 1998, 46 (10), 4128-4137
(178) Jia, T.; Kelleher, S.; Hultin, H.; Petillo, D.; Maney, R.; Krzynowek, J., Comparison of
quality loss and changes in the glutathione antioxidant system in stored mackerel and
bluefish muscle J. Agric. Food Chem. 1996, 44, (5), 1195-1201.
(179) Pazos, M.; Gonzalez, M., Preservation of the endogenous antioxidant system of fish
muscle by grape polyphenols during frozen storage. European Food Researh and
Technology 2005, 220, 514-519.
(180) Erickson, M. C., Compositional Parameters and their relationship to oxidative Stability of
Channel Catfish J Agri Food Chem 41, 1213-1218.
(181) Decker, E. A.; Livisay, S. A.; Zhou, S., Mechanisms of Endogenous Skeletal Muscle
Antioxidants: Chemical and Physical Aspects. In Antioxidants in Muscle Foods, Decker,
E. A.; Faustman, C.; Lopez-Bote, C. J., Eds. John Wiley and Sons: New York, 2000.
(182) McCay, P. B., Vitamin E: Interactions with free radicals and ascorbate. Annu. Rev. Nutr.
1985, 5, 323-340.
(183) Kroger-Ohlsen, M.; Andersen, L., Acid-catalyzed autoreduction of ferrymyoglobin in
aqueous solution studied by freeze quenching and ESR spectroscopy. Free Radic. Res.
1999, 30, (4).
(184) Pokorny, J.; Davidek, J.; Chocholata, V.; Panek, J.; Bulantova, H.; Janitz, W.; Valentova,
H.; Vierecklova, M., Interactions of oxidized lipids with protein. XVI: Interactions of
oxidized ethyl linoleate with collagen. Die Nahrung 1990, 34, 159-169.
(185) Kanner, J.; Karel, M., Changes in lysozyme due to reactions with the peroxidizing methyl
linoleate in a dehydrated model system. J. Agric. Food Chem. 1976, 24, 468-472.
(186) Funns, J. A.; Weiss, U.; Karel, M., Effects of reaction conditions and reactant
concentrations on polymerization of lysozyme reacted with peroxidizing lipids. . J Agri
Food Chem 1982, 30, 1204-1208.
(187) Kelleher, S. D.; Hultin, H. O.; Wilhelm, K. A., Stability of mackerel surimi prepared under
lipid-stabilizing processing conditions. J. Food Sci 1994, 59, (269-271).
(188) Smith, D. M., Functional and Biochemical changes is deboned turkey due to frozen storage
and lipid oxidation. J. Food Sci 1987, 52, (22-27).
(189) Lui, G.; Xiong, Y. L., Storage stability of antioxidant-washed myofibrils from chicken
white and red muscle. J. Food Sci 1996, 61, (890-894).
(190) Srinivasan, S.; Xiong, Y. L., Gelation of beef heart surimi as affected by antioxidants. J.
Food Sci 1996, 61, 707-711.
157
(191) Parkington, J. K.; Xiong, Y. L.; Xiong, S.; Blanchard, S. P.; Srinivasan, S.; Froning, G.
W., Chemical and functional properties of oxidatively modified beef heart surimi stored at
2 degrees C. J Food Sci. 2000, 65, (3), 428-433.
(192) Kisia, S. M., Structure of fish locomotoray muscle. In Fish Morphology, Datta-Munshi, J.
S.; Gutta, H. M., Eds. AA Balkema: Rotterdam, 1996; pp 167-178.
(193) Lemon, D. W. An improved TBA test for Rancidity; 51 Halifax, NS, Canada, 1975.
(194) Torten, J.; Whitaker, J. R., Evaluation of Biuret and Dye-binding Methods for Protein
Determination in Meats. J. Food Sci 1964, 29, (2), 168-174.
(195) Bradford, M. M., A rapid and sensitive method for the quantification of microgram
quantities of protein utilizing the principle of protein-dye binding. . Anal. Biochem 1976,
72, 248-254
(196) Laemmli, U. K., Cleavage of structural proteins during the assembly of the head of
bacteriophase T4. . Nature 1970, 227, 680-685.
(197) Liang, Y.; Hultin, H., Functional protein isolates from mechanically deboned turkey by
alkaline solubilization with isoelectric precipitation. J. Muscle foods 2003, 14, 195-206.
(198) Sherman, P., Industrial Rheology: with particular reference to foods, pharmaceuticals, and
cosmetics. Academic Press: New York, 1970.
(199) Kristinsson, H. G.; Hultin, H. O., Changes in Conformation and Subunit Assembly of cod
Myosin at Low and High pH and after Subsequent Refolding. J. Agric Food Chem 2003,
51, 7187-7196.
(200) Yongsawatdigul, J.; Park, J., Effects of alkali and acid solubilation on gelation
characteristics of rockfish muscle proteins. j Food Sci. 2004, 69, C499-505.
(201) Choi, Y. J.; Park, J. W., Acid-aided protein recovery from enzyme-rich Pacific whiting. . J
Food Sci. 2002, 67, 2962-2967.
(202) Yongsawatdigul, J.; Park, J. W.; Esturk, O., Gelation characteristics of alkaline and acid
solubilization of fish muscle proteins. In Institute of Food Technologist (IFT), New
Orleans, 2001.
(203) Undeland, I. A.; Kelleher, S. D.; Hultin, H. O.; McClements, J.; Thongraung, C.,
Consistency and solubility Changes in herring (Clupea harengus) Light muscle
Homogenates as a function of pH. J. Agric Food Chem 2003, 51, 3992-3998.
(204) Kristinsson, H. G.; Theodore, A. E.; Necla, D.; Ingadottir, B., A Comparative Study
between Acid and Alkali-aided Processing and Surimi Processing for the Recovery of
Proteins from Channel Catfish Muscle. Food Chemistry and Toxicology 2005, 70, (4),
C298-C306.
(205) Kristinsson, H. G.; Liang, Y., Effect of pH-shift processing and surimi processing on
Atlantic Croaker (Micropogonias undulates) Muscle Proteins. Food Chemistry and
Toxicology 2006, 71, (5), C304-C312.
(206) Undeland, I. A.; Kristinsson, H. G.; Hultin, H. O., Hemoglobin catalyzed oxidation of
washed cod muscle phospholipids-effect from pH and hemoglobin source. J Agri Food
Chem 2004, 52, 4444-4451.
(207) Kenney, P. B.; Kastner, C. L.; Kropf, D. H., Muscle washing and raw material source
affect quality and physiochemical properties of low-salt, low-fat, restructured beef. J. Food
Sci 1992 57, 545-550.
(208) USDA, Beef products: Raw, Processed and Prepared. In U.S. Department of Agriculture
Handbook 8-13, U.S. Government Printing Office: Washington, D.C., 1990.
158
(209) Luo, S.; Hultin, H. O., In vitro lipid oxidation modifies proteins and functional properties
of sarcoplasmic reticulum. Bioenergetics and biomembranes 1986, 18, (4), 315-323.
(210) Srinivasan, S.; Hultin, H. O., Hydroxyl radical modification of fish muscle proteins.
Journal of Food Biochemistry 1995, 18, 402-425.
(211) Ostdal, H.; Davies, M. J.; Andersen, H. J., Reaction between protein radicals and other
biomolecules. Free Rad. Bio. Med 2002, 33, (2), 201-209.
(212) Olinescu, R. M.; Kummerow, F. A., Fibrinogen is an efficient antioxidant. J. Nutr.
Biochem. 2001, 12, 162-169.
(213) Luxford, C.; Morin, B.; Dean, R. T.; Davies, M. J., Histone H1- and other protein-and
amino acid hydroperoxides can give rise to free radicals which oxidize DNA. Biochem. J.
1999, 344, 125-134.
(214) Stadtman, E. R.; Berlett, B. S., Reactive oxygen-mediate protein oxidation in aging and
disease. Chem. Res Toxicol. 1997, 10, 484-494.
(215) Kjaersgard, I. V. H.; Norrelykke, M. R.; Baron, C. P.; Jessen, F., Identification of
carbonylatd protein in frozen rainbow trout (Oncorhynchus mykiss) fillets and
development of Protein Oxidation during frozen storage. J Agri Food Chem 2006, 54,
9437-9446.
(216) Ingadottir, B. The use of acid and alkali-aided protein solubilation the precipidation
methods to produce functional protein ingredients from tilapia. University of Florida,
Gainesville, 2005.
(217) Theodore, A. E. Bioactive and functional properties of catfish protein hydrolysates and
catfish protein isolates. University of Florida, Gainesville, 2005.
(218) Davenport, M.; Kristinsson, H. G., Effect of different acid and alkali-treatments on the
molecular and functional properties of catfish muscle proteins. Abstract: 49G-14. In IFT
Annual Meeting, Las Vegas, Nevada,2004.
(219) Samejima, K.; Takahashi, K.; Yasui, T., Heat induced denaturation of myosin total rod.
Agric. Biol. Chem. 1976, 40, (12), 2455.
(220) Samejima, K.; Ishioroshi, M.; Yasui, T., Relative roles of the head and tail portions of the
molecule in the heat-induced gelation of myosin. J. Food Sci. 1981, 46, 1412.
(221) Shikha, F. H.; Hossain, M. I.; Morioka, K.; Kubota, S.; Yoshiaki, I., Effect of pH-shifting
on the gel forming characteristics of salt-ground meat from walleye pollock. Fisheries
Science 2006, 72, 870-876.
(222) Schaich, K. Alterations in protein functionality due to oxidative reactions. University of
Kentucky, Lexington, 1993.
(223) Strehler, E. E.; Strehler-Page, M. A.; Perriad, J. C.; M., P.; Nadal-Ginard, B., Complete
nucleotide and encodes amino acid sequences of a malian myosin heavy chain gene:
Evidence again intron-dependent evolution of the rod. J. Mol. Biol 1986, 190, 291-317.
(224) Belitz, H. D.; Grosch, W.; Schieberle, P., Meat. 3rd ed.; Springer-Verlag: Berlin,
Germany, 2004.
(225) Nishimura, K.; Ohishi, N.; Tanaka, Y.; Saskura, C., Participation of radicals in
polymerization by ascorbic acid of crude actomyosin from frozen surimi of Alaska pollock
during a 40C incubation. Biosci Biotech Biochem 1992, 56, 24-28.
(226) Lanier, T. C.; Lin, T. S.; Hamann, D. D.; Thomas, F. B., Effects of alkaline protease in
minced fish on texture of heat-processed gels. J. Food Sci. 1981, 46, 1643.
(227) Lee, C. M., Surimi Process Technology. Food Tech. 1984, 38, 69.
159
(228) An, H.; Weerasinghe, V.; Seymore, T. A.; Morrissey, M. T., Cathepsin degradation of
Pacific whiting surimi proteins. J Food Sci. 1994, 59, (1013-1017, 1033).
(229) Kristinsson, H. G.; Rasco, B. A., Hydrolysis of salmon muscle proteins by an enzyme
mixture extracted from Atlantic salmon (Salmo salar) pyloric caeca. J. Food Biochem.
2000, 24, 177-187.
(230) Kristinsson, H. G., Evaluation of different methods to isolate cod (Gadus morhua) muscle
myosin. J. Food Biochem. 2001, 48, 3141-3147.
(231) Takeda, H.; Seki, N., Enzyme catalyzed cross-linking and degradation of myosin heavy
chain in Walleyed pollock surimi paste during setting. Fisheries Sci. 1996, 62, 462-7.
(232) Kelleher, S. D.; Hultin, H. O., Surimi processing from dark muscled fish. In Surimi and
Surimi Seafood, Park, J. W., Ed. Marcel Dekker, Inc.: New York, 2000; pp 59-77.
(233) Gray, J. I.; Pearson, A. M., Rancidity and warmed over flavour. Adv. Meat Res. 1987, 3,
221.
(234) Erickson, M. C.; Hultin, H. O., Nonenzymatic catalysis of lipid oxidation by the soluble
fractions of mackerel muscle. J. Food Sci 1990, 55, 947-950, 953.
(235) Richards, M. P.; Hultin, H. O., Hemolysate from mackerel herring and trout promote lipid
oxidation at different rates. Fish Sci.
(236) Slabyj, B. M.; Hultin, H. O., Lipid-peroxidation by microsomal fractions-isolated from
light and dark muscles of herring (Clupea-harengus). J. Food Sci 1982, 47, 1395-1398.
(237) Han, H. T.; Liston, J., Lipid-peroxidation protection factors in rainbow-trout (SalmoGairdnerii) muscle cytosol. J. Food Sci 1989, 54, (4), 809-813.
(238) Gunnarsson, G.; Undeland, I. A.; Thippeswamy, S.; Sanberg, A.; Lingard, A.; MattssonHulten, L.; Soussi, B., Inhibitory Effect of Known Antioxidants and of Press Juice from
Herring (Clupea harengus) Light Muscle on the Generation of Free Radicals in Human
Monocytes. J Agri Food Chem 2006, 54, 8212-8221.
(239) Rong, L., Characterization of aqueous components in chicken breast muscle as inhibitors
of hemoglobin-mediated lipid oxidation. J. Agric Food Chem 2005, 53, 767-775.
(240) Rowley, A. F., Collection, separation and identification of fish leukocytes. In Techniques
in Fish Immunology, Stolen, J. S.; Fletcher, T. C.; Andersen, D. P.; Roberson, B. S.; Van
Muuiswinkel, W. B., Eds. SOS Publications: New Jersey, 1990; pp 113-135.
(241) Fyhn, U. E.; Fyhnm, H. J.; Davis, B. J.; Powers, D. A.; Fink, W. L.; Garlick, R. L.,
Hemoglobin heterogenity in Amazonian fishes. Comp. Biochem. Physiol. 1979, 62A, 3966.
(242) Prior, R. L.; Cao, G., In vivo total antioxidant capacity: comparison of different analytical
methods. Free Radical Biol. Med. 1999, 27, (1173-1181).
(243) Lawrie, R. A., Meat Science. 4th ed.; Pergamon Press: New York, 1985.
(244) Grunwald, E., Mechanisms of Heme Protein-Mediated lipid oxidation using hemoglobin
and myoglobin variants in raw and heated washed muscle. Agriculture and Food
Chemistry 2006, 54, 8271-8280.
(245) Decker, E. A.; Hultin, H. O., Factors influencing catalysis of lipid oxidation by the soluble
faction of mackerel muscle. J. Food Sci 1990, 55, 947-950,953.
(246) Rong, L.; Richards, M. P.; Undeland, I. A., Characterization of aqueous components in
chicken breast muscle as inhibitors of hemoglobin-mediated lipid oxidation. Agriculture
and Food Chemistry 2005, 53, 767-775.
160
(247) Chiu, D. T.; Van den Berg, J.; Kuypers, F. A.; Hung, I. J.; Wei, J. S.; Lui, T. Z.,
Correlation of membrane lipid peroxidation with oxidation of hemoglobin variants:
possibly related to hemin release. Free Rad. Bio. Med 1996, 21, 89-95.
(248) Bamm, V. V.; Tsemakhovich, V. A.; Shaklai, M.; Shaklai, N., Haptoglobin phenotypes
differ in their ability to inhibit heme transfer from hemoglobin to LDL. Biochem. 2004, 43,
3899-3906.
(249) Balla, G.; Jacob, H. S.; Eaton, J. W.; Belchler, J. D.; G.M., V., Hemin: a possible
physiological mediator of low density lipoprotein oxidation in endothelial injury.
Arterioscler. Throm. 1991, 11, 1700-1711.
(250) Grinshtein, N.; Bamm, B. B.; Tsemakhovich, V. A.; Shaklai, N., Mechanism of low
density lipoprotein oxidation by haemoglobin derived iron. Biochemistry 2003, 42,
6977-6985.
(251) Jorolim, P.; Lahav, M.; Liu, S. C.; Palek, J., Effect of hemoglobin oxidation products on
the stability of red cell membrane skeletons and the association of skeletal proteins:
correlations with release of heme. Blood 1990, 76, 2125-2131.
(252) Richards, M. P.; Detmann, M. A.; Grunwald, E., Pro-oxidative characteristics of trout
hemoglobin and myoglobin: A role for released heme in oxidation of lipids. J Agri Food
Chem 2005, 53, 10231-10238.
(253) Vincent, S. H.; Grady, R. W.; Shaklai, N.; Snider, J. M.; Muller-Eberhard, U., The
influence of heme-binding proteins in heme catalyzed oxidations. Arch. Biochem.
Biophys. 1988, 265, 539-550.
(254) Griffonm, N.; Baudin, V.; Dieryck, W.; Dumoulin, A.; Pagnier, J.; Poyart, C.; Marden, M.
C., Tetramer-dimer equilibrium of oxyhemoglobin mutant determined from auto-oxidation
rates. Protein Sci. 1988, 7, 673-680.
(255) Baron, C. P.; Skibsted, L. H.; Andersen, H. J., Prooxidative activity of myoglobin species
in linoleic acid emulsions. J. Agric. Food Chem. 1997, 45, (5), 1704-1710.
(256) Storage, M.; Matsuoka, A.; Highmore, A.; Seaward, Y.; Shikama, K., The molecular
mechanism of autoauxidation for human oxyhemoglobin. J. Biol. Chem. 1998, 273, 86078615.
(257) Misra, H. P.; Fridovich, I., Generation of superoxide during autoxidation of hemoglobin. J.
Biol. Chem. 1972, 247, (6960-6962).
(258) Yoshida, Y.; Kashiba, K.; Niki, E., Free-radical mediated oxidation of lipid induced by
Hemoglobin in aqueous dispersions. Biochem. Biophs. Act 1994, 1201, 165-172.
(259) Harel, S.; Kanner, J., Hydrogen peroxide-activated methae-myoglobin and other
haemoproteins as initiators of membranal lipid peroxidation. In Superoxide and
Superoxide Dismutase in Chemistry, Rotilo, G., Ed. Elsevier Science B.V. (Biomedical
Division): New York, 1986; pp 25-28.
(260) Shen, L.; Hermans, J., Kinetics of conformation change of sperm-whale myoglobin.
Biochem. 1972, III. Folding and unfolding of apomyoglobins and the suggested overall
mechanism, 1845-1849.
(261) Kibald, M.; Luck, T.; Holley, W.; Klostermeyer, H., Decolorization of slaughterhouse
thick blood for the recovery of a colourless protein preparation and an iron-containing
heme fraction. In Food proteins: structure and functionality, Schwenke, K. D.; Mothes, R.,
Eds. VCH: Weinheim, Germany, 1991; pp 103-105.
(262) Ericksson, C. E.; Olsson, P. A.; S.G., S., Denatured hemoproteins as catalysts in lipid
oxidation. JAOCS 1971, 48, 442-447.
161
(263) Cannon, J. B.; Kuo, F. S.; Paternack, R. F.; Wong, N. M.; Muller-Eerhard, U., Kinetics of
the interaction of hemin liposomes with heme binding proteins. Biochem. 1984, 23, 37153721.
(264) Hargrove, M. S.; Whitaker, T.; Olson, J. S.; Vali, R. J.; Mathews, A. J., Quaternary
structure regulates hemin dissociation from human hemoglobin. J. Biol. Chem. 1997, 272,
17385-17389.
(265) Miyawawa, T.; Suzuki, T.; Fujimoto, K.; Kinoshita, M., Age related change of
phosphatiylchonline hydroperoxide and phosphotidylethanolamine hydroperoxide levels in
normal human red blood cells. Mech. Ageing Dev. 1996, 86, 145-150.
(266) Wright, J. S.; Johnson, E. R.; Dilabio, G. A., Predicting the activity of phenolic
antioxidants: theoretical method, analysis of substituent effects, and application to major
families of antioxidants. J. Am. Chem. Soc. 2001, 123, 1173-1183.
(267) Awiki, J. M.; Rooney, L. W.; Wu, X.; Prior, R. L.; Cisneros-Zevallos, L., Screening
methods to measure antioxidant activity of sorghum (Sorghum bicolor) and sorghum
products. J. Agric. Food Chem. 2003, 51, 6657-6662.
(268) Prior, R. L.; Wu, X.; Schaich, K., Standardized Methods for the Determination of
Antioxidant Capacity and Phenolics in Foods and Dietary Supplements. J. Agric. Food
Chem. 2005, 53, 4290-4302.
(269) Van den Berg, R.; Haenen, G. R. M. M.; Van den Berg, H.; Bast, A., Applicability of an
improved Trolox equivalent antioxidant capacity (TEAC) assay for evaluation of
antioxidant capacity measurements of mixtures. Food Chemistry 1999, 66, 511-517.
(270) Antonenkov, V. D.; Sies, H., Some basic properties of ascorbate-dependent antioxidativedefense factors from rat liver cytosol. Chem.-Biol. Interact. 1995, 94, 169-181.
162
BIOGRAPHICAL SKETCH
Ms. Holly Tasha Petty was born in Anchorage Alaska, and was raised in Nikiski, Alaska
for the first 15 years of her life. She then lived in Oviedo, Florida, and received her high school
diploma from Oviedo High school. She received her Bachelor of Science degree in food science
and human nutrition, with honors. In Fall 2002, Holly began her doctoral graduate studies in
food science and human nutrition, and minor in applied physiology and kinesiology. As a
graduate student, she served as the Florida Section Institute of Food Technologists student
representative from, 2005-2006; and she served as the Food Science Graduate Representative,
2005-2006. During her graduate schooling, Holly was awarded the Agricultural Woman’s
Scholarship, 2006; Florida Section Institute of Food Technologists Student Fellowship, 2006;
Aylesworth Foundation for the Advancement of Marine Sciences fellowship, 2005-2007; and the
Institute of Food Technologists National graduate scholarship, 2004-2005. She is the main
inventor of the, “Novel Low-Allergenic Food Bar, U.S. Provisional Patent No. 60/844,431,” as
well and co-inventor of “Liquid Nutrient Composition for Improving Performance, UF.
Disclosure No. 12237.” Holly has contributed to various metabolic and health-related studies as
a participant and trainer through the Applied Physiology and Kinesiology Department,
University of Florida. Holly has been a member of the National Strength and Conditioning
Association, International Association for Food Protection, American Oil Chemist Society,
Institute of Food Technologists, Phi Tau Sigma Honor Society, and the University of Florida
Women’s Cycling Team. Upon graduation, Holly hopes to contribute to industry and academia,
promoting health initiatives through her knowledge of food science, nutrition, and exercise
physiology; and her passion for the vitality of people and the environment.
163