<%BANNER%>

The Pro-Oxidative Properties of Tilapia Oxy-, Carboxy-and Met-Hemoglobin in Washed Minced Tilapia Muscle

Permanent Link: http://ufdc.ufl.edu/UFE0041582/00001

Material Information

Title: The Pro-Oxidative Properties of Tilapia Oxy-, Carboxy-and Met-Hemoglobin in Washed Minced Tilapia Muscle
Physical Description: 1 online resource (243 p.)
Language: english
Creator: Aldaous, Sara
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: carbon, hemoglobin, lipid, met, oxidation, oxy, sensory, tilapia
Food Science and Human Nutrition -- Dissertations, Academic -- UF
Genre: Food Science and Human Nutrition thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: THE PRO-OXIDATIVE PROPERTIES OF TILAPIA OXY-, CARBOXY-AND MET-HEMOGLOBIN IN WASHED MINCED TILAPIA MUSCLE World consumption of fish has increased over the past 50 years. The stability of seafood and the extension of the shelf life of its products has become a vital issue for the aquaculture industry. Foremost is the red color found in the dark muscle of fish which is due to the oxygenated and reduced form of hemoglobin and myoglobin. The presence of this red color influences the consumer and is interpreted as freshness , thereby increasing its market value. The brown color of fish due to oxidation suggests lack of freshness. Carbon monoxide has been used to stabilize this red color. This study aimed to investigate the pro-oxidative activity of different forms of hemoglobin in the development of oxidation of lipids and proteins in muscle seafood. The results of this study suggest that oxidation of lipids and proteins could be delayed by stabilizing hemoglobin through the binding of carbon monoxide.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Sara Aldaous.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Sims, Charles A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041582:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041582/00001

Material Information

Title: The Pro-Oxidative Properties of Tilapia Oxy-, Carboxy-and Met-Hemoglobin in Washed Minced Tilapia Muscle
Physical Description: 1 online resource (243 p.)
Language: english
Creator: Aldaous, Sara
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: carbon, hemoglobin, lipid, met, oxidation, oxy, sensory, tilapia
Food Science and Human Nutrition -- Dissertations, Academic -- UF
Genre: Food Science and Human Nutrition thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: THE PRO-OXIDATIVE PROPERTIES OF TILAPIA OXY-, CARBOXY-AND MET-HEMOGLOBIN IN WASHED MINCED TILAPIA MUSCLE World consumption of fish has increased over the past 50 years. The stability of seafood and the extension of the shelf life of its products has become a vital issue for the aquaculture industry. Foremost is the red color found in the dark muscle of fish which is due to the oxygenated and reduced form of hemoglobin and myoglobin. The presence of this red color influences the consumer and is interpreted as freshness , thereby increasing its market value. The brown color of fish due to oxidation suggests lack of freshness. Carbon monoxide has been used to stabilize this red color. This study aimed to investigate the pro-oxidative activity of different forms of hemoglobin in the development of oxidation of lipids and proteins in muscle seafood. The results of this study suggest that oxidation of lipids and proteins could be delayed by stabilizing hemoglobin through the binding of carbon monoxide.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Sara Aldaous.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Sims, Charles A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041582:00001


This item has the following downloads:


Full Text

PAGE 1

1 THE PRO OXIDATIVE PROPERTIES OF TILAPIA OXY, CARBOXYAND MET HEMOGLOBIN IN WASHED MINCED TILAPIA MUSCLE By SARA ALDAOUS 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 2010

PAGE 2

2 2010 Sara Aldaous

PAGE 3

3 To my devoted parents, my siblings, my best friend and bro ther, Dr. Mohammed Aldaous

PAGE 4

4 ACKNOWLEDG MENTS I would like to thank King Abdulaziz University, Je ddah, Saudi Arabia, for the scholarship and giving me this lifetime opportunity to achieve my Ph.D. I am greatful to Dr. Hordur G. Kristinsson for recruiting me, and for his guidance, mentorship, and friendship. I sincerely appreciate my major advisor Dr. Charles Sims for his direction and unconditional help in the completion of this research. I am also gr ate ful to my supervising committee, Dr. Marty Marshall Dr. Liwei Gu, and Dr. Ch ristian Leeuwenburgh for the valuable time devot ed to this project, their suggestions, and words of encouragement during this research. A special thank s to Dr. Wei Huo for his statistical expertise and kind help in inter p r e ting my data and to Shane Ruessler for the capturing of the liv e Tilapia. I would also like to thank my family for their endless love and support especially my b rother Dr. Mohammed Aldaous who believd in me, and supported me with gui dance and understanding throughout this journey in the U.S I am thankful to Dr. Patience Dirkx, my American Mom for her unconditional friendship, inspiration and for all the tears of joy and pain that we shared together I would like to thank my infl uential friends, Kateryna Clark Maria Plaza and Stefan Crynen for encouraging me to fulfill this dream and for their friendship throughout these years. I would like to acknowledge Carmen Graham, Bridget Stokes, Marianne Mangone, and Julie Barber for their help and services in our department. I thank t he Office of International Students at the University of Florida. Finally, I thank those who crossed my graduate path, for their positiv e influences and I am thankful, for the experiences I have gained

PAGE 5

5 TABLE OF CONTENTS ACKNOWLEDGMENTS .................................................................................................. 4 page TABLE OF CONTENTS .................................................................................................. 5 LIST OF TABLES ............................................................................................................ 9 LIST OF FIGURES ........................................................................................................ 11 ABSTRACT ................................................................................................................... 18 CHAPTER 1 INTRODUCTION .................................................................................................... 21 Research Objectives ............................................................................................... 22 Research Significance ............................................................................................ 23 2 LITERATURE REVIEW .......................................................................................... 24 Lipid Oxidat ion in Fish Muscle ................................................................................ 24 Lipid Composition in Fish ................................................................................. 26 Inhibitors of Lipid Oxidation .............................................................................. 27 Changes in Post mortem Fish .......................................................................... 27 Prot ein Oxidation .................................................................................................... 28 Role of Heme Proteins in Oxidation ................................................................. 31 Loss of Hemin .................................................................................................. 36 Met Hemoglobin Reduction ..................................................................................... 39 Role of Carbon Monoxide (CO) .............................................................................. 40 Modified Atmosphere Packaging (MAP) Using CO ................................................. 43 Effect of pH on Oxidation ........................................................................................ 44 Role of Sodium Chloride (NaCl) in Oxidation .......................................................... 46 3 CARBOXY, OXY AND MET HEMOGLOBIN: A COMPAR ATIVE STUDY OF HEMOGLOBIN MEDIATED LIPID/PROTEIN OXIDAT ION IN WASHED MINC ED TILAPIA MUSCLE AT TWO DIFFERENT STORAG E TEMPERATURES .................................................................................................. 49 Introduction ............................................................................................................. 49 Materials and Methods ............................................................................................ 50 Preparation of Washed, Minced Tilapia Muscle (WMTM)................................. 50 Collection of Fish Blood .................................................................................... 51 Preparation of Hemolysate ............................................................................... 51 Quantification of Hemoglobin Levels in Hemolysate ......................................... 52 Oxy CO and Met hemoglobin Preparation .................................................... 52

PAGE 6

6 Sample Preparation: Addition of Hb and NaCl ................................................. 53 Determination of Protein Content ..................................................................... 54 Determination of Total Lipids ............................................................................ 55 Determination of Phospholipids Content .......................................................... 55 Determination of Peroxide Value (PV) .............................................................. 56 Determination of Thiobarbituric Acid Reactive Substances (TBAR S) ............... 57 Determination of Carbonyl Groups ................................................................... 57 Heme Group Autoxidation ................................................................................ 58 Color Analysis .................................................................................................. 59 Results .................................................................................................................... 59 Lipid Oxidation Analysis ................................................................................... 60 Protein Oxidation Analysis ................................................................................ 61 Color Analysis .................................................................................................. 62 Heme Group Autoxidation ................................................................................ 63 Discussion .............................................................................................................. 65 Conclusion .............................................................................................................. 71 4 EFFECT OF LOW AND HI GH CONCENTRATIONS OF HEMOGLOBIN ON THE PRO OXIDATIVE ACTIVITY O F OXY, CO AND MET HEMOGLOBIN IN A WASHED MINCED TILAPIA MUSCLE SYSTEM AT TWO DIFFERENT STORAGE TEMPERATURES ................................................................................ 84 Introduction ............................................................................................................. 84 Materials and Methods ............................................................................................ 85 CO Calibra tion Curve by the Gas Chromatography Method ............................. 85 Sample Analysis by the Gas Chromatography (GC) Method ........................... 85 Gas Chromatography Conditions for CO Analysis ........................................... 86 Results .................................................................................................................... 86 Lipid Oxidation Analysis ................................................................................... 86 Protein Oxidation Analysis ................................................................................ 88 CO Release ...................................................................................................... 90 Color Analysis .................................................................................................. 90 Heme Group Autoxidation ................................................................................ 91 Discussi on .............................................................................................................. 94 Conclusion .............................................................................................................. 97 5 EFFECT OF PH ON THE PRO OXIDATIVE ACTIVITY O F OXY, CO AND MET HEMOGLOBIN IN A WASH ED MINCED TILAPIA MU SCLE SYSTEM AT TWO DIFFERENT STORAG E TEMPERATURES ............................................... 116 Introduction ........................................................................................................... 116 Materials and Methods .......................................................................................... 118 Determinat ion of Hemin Loss ......................................................................... 118 Preparation of buffer ....................................................................................... 119 Preparation of FPLC (fat protein liquid chromatography) ............................... 119 Sample preparation ........................................................................................ 119 Verification and analysis of extracted samples ............................................... 120

PAGE 7

7 The rate of hemin loss .................................................................................... 120 Calculation of Dissociation Rate ..................................................................... 120 Results .................................................................................................................. 121 Lipid Oxidation Analysis ................................................................................. 121 Protein Oxidation Analysis .............................................................................. 123 Hemin Loss Rate ............................................................................................ 124 CO Release .................................................................................................... 124 Color Analysis ................................................................................................ 125 Heme Group Autox idation .............................................................................. 126 Discussion ............................................................................................................ 127 Conclusion ............................................................................................................ 135 6 THE ROLE OF SODIUM CLORIDE ON THE PRO OXIDATIVE ACTIVITY O F OXY, CO AND MET HEMOGLOBIN ON IN A WASHED MINCED T ILAPIA WASHED SYSTEM .............................................................................................. 156 Introduction ........................................................................................................... 156 Materials and Methods .......................................................................................... 157 Results .................................................................................................................. 157 Lipid Oxidation Analysis ................................................................................. 157 Protei n Oxidation Analysis .............................................................................. 159 CO Release .................................................................................................... 160 Color Analysis ................................................................................................ 161 Heme Group Autoxidation .............................................................................. 162 Discussion ............................................................................................................ 164 Conclusion ............................................................................................................ 168 7 EFFECT OF THE OXIDAT ION OF OXY CO AND MET HEMOGLOBIN ON THE QUALITY OF REFRI GERATED WASHED MINCED TILAPIA MUSCLE ..... 187 Introduction ........................................................................................................... 187 Materials and Methods .......................................................................................... 188 Sample Pre paration: Addition of Hb and NaCl ............................................... 188 Descriptive Sensory Analysis ......................................................................... 190 Color Analysis ................................................................................................ 191 Statistical Analysis .......................................................................................... 192 Results .................................................................................................................. 192 Sensory Evaluation ........................................................................................ 192 Lipid Oxidation Analysis ................................................................................. 193 Correlation between Sensory Scores and TBARS ......................................... 195 Color Analysis ................................................................................................ 195 Correlation betw een Sensory Scores and a*Value ......................................... 196 Correlation between TBARS and a*Value ...................................................... 196 Discussion ............................................................................................................ 197 Conclusion ............................................................................................................ 202 8 SUMMARY AND CONCLUSI ON .......................................................................... 214

PAGE 8

8 APPENDIX A RESEARCH SCHEME .......................................................................................... 216 B CHANGES IN L* VALUE AND B* VALUE ............................................................ 217 C PEARSON CORRELATION COEFFICIENT ......................................................... 229 D TRIANGLE T EST BALLOT ................................................................................... 231 E DESCRIPTIVE ANALYSIS BALLOT ..................................................................... 232 LIST OF REFERENCES ............................................................................................. 233 BIOGRAPHICAL SKETCH .......................................................................................... 243

PAGE 9

9 LIST OF TABLES Table page 3 1 Sample preparation of the three forms of hemoglobin ........................................ 72 3 2 Nikon D200 camera settings used for measurement of change in color ............. 73 3 3 Composition of tilapia muscle and washed tilapia muscle model system. .......... 74 3 4 Composition of tilapia muscle and washed tilapia muscle model system on dry weight basis. ................................................................................................. 75 3 5 Changes in a* value, L* value and b* value in washed tilapia muscle containing different forms of Hb at 3.7 C ............................................................ 76 3 6 Changes in a* value, L* value and b* value in washed tilapia muscle containing different forms of Hb at 25C ............................................................ 77 4 1 Changes in a*value in washed tilapia muscle containing different forms of Hb at a concentration of 6, 9, and 12mol/kg at 3.7C ............................................. 98 4 2 Changes in a*value in washed tilapia muscle containing different forms of Hb at a concentration of 6, 9, and 12mol/kg at 25C ............................................. 99 5 1 Hemin loss rate for met Hb (tilapia and trout) ................................................... 136 5 2 Changes in a* value in washed tilapia muscle containing different forms of Hb at pH 6.3, 6.8, and 7.3 at 3.7C ................................................................... 137 5 3 Changes in a* value in washed tilapia muscle containing different forms of Hb at pH 6.3, 6.8, and 7.3 at 25C .................................................................. 138 6 1 Changes in a*value in washed tilapia muscle containing different forms of Hb at concentrations of added NaCl (0, 150, and 450 mM) at 3.7C ..................... 169 6 2 Changes in a* value in washed tilapia muscle containing different forms of Hb at concentrations of added NaCl (0, 150, and 450 mM) at 25C. ............... 170 7 1 Treatments used in the study. Hemoglobin was added to washed tilapia muscle at a concentration of 12 mol Hb/kg muscle ........................................ 203 7 2 Reference samples used to train panelists for the descriptive analysis ............ 204 7 3 Combination of treatments used by panelists to rate the formation of painty/rancid odor as an indicator of lipid oxidation .......................................... 205 7 4 Nikon D200 Settings used for measurement of change in color ....................... 206

PAGE 10

10 7 5 Changes in a*value of tilapia muscle after the addition of the Hb and NaCl at three different pH levels .................................................................................... 207 7 6 Fishers z transformation correlation comparison (R, p<0.05) between sensory scores, TBARS, and a*value ............................................................... 208 B 1 Changes in L*value in washed tilapia muscle containing different forms of Hb at a concentration of 6, 9, and 12mol/kg muscle at 3.7C .............................. 217 B 2 Changes in L*value in washed tilapia muscle containing different forms of Hb at a concentration of 6, 9, and 12mol/kg muscle at 25C ............................. 218 B 3 Changes in b*value in washed tilapia muscle containing different forms of Hb at a concentration of 6, 9, and 12mol/kg muscle at 3.7C .............................. 219 B 4 Changes in b*value in washed tilapia muscle containing different forms of Hb at a concentration of 6, 9, and 12mol/kg muscle at 3.7C .............................. 220 B 5 Changes in L*value in washed tilapia muscle containing different forms of Hb at pH 6.3, 6.8, and 7.3 at 3.7C. ....................................................................... 221 B 6 Changes in L*value in washed tilapia muscle containing different forms of Hb at pH 6.3, 6.8, and 7.3 at 2 5C ....................................................................... 222 B 7 Changes in b*value in washed tilapia muscle containing different forms of Hb at pH 6.3, 6. 8, and 7.3 at 3.7C ....................................................................... 223 B 8 Changes in b*value in washed tilapia muscle containing different forms of Hb at pH 6.3, 6. 8, and 7.3 at 25C ....................................................................... 224 B 9 Changes in L*value in washed tilapia muscle containing different forms of Hb at concentrations of added NaCl (0, 150, and 450mM) at 3.7C ...................... 225 B 10 Changes in L*value in washed tilapia muscle containing different forms of Hb at concentrations of added NaCl (0, 150, and 450mM) at 25C ...................... 226 B 11 Changes in b*value in washed tilapia muscle containing different forms of Hb at concentrations of added NaCl (0, 150, and 450mM) at 3.7C ...................... 227 B 12 Changes in b*value in washed tilapia muscle containing different forms of Hb at concentrations of added NaCl (0, 150, and 450mM) at 25C ...................... 228 C 1 Pearson correlation c oefficient for samples stored at 3.7C ............................. 229 C 2 Pearson correlation coefficient for sampl es stored at 25C ............................. 230

PAGE 11

11 LIST OF FIGURES Figure page 2 1 The dynamic conversion between different Hb forms. ........................................ 48 3 1 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at 3.7C ........................................................................................... 78 3 2 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at 25C ........................................................................................... 78 3 3 TBARS values in washed tilapia muscle containing different forms of Hb at 3.7C ................................................................................................................... 79 3 4 TBARS values in washed tilapia muscle containing different forms of Hb at 25C .................................................................................................................... 79 3 5 Carbonyl values in washed tilapia muscle containing different forms of Hb at 3.7C ................................................................................................................... 80 3 6 Carbonyl values in washed tilapia muscle containing different forms of Hb at 25C .................................................................................................................... 80 3 7 % of a) Oxy b) Met and c) Deoxy Hb in washed tilapia muscle containing different forms of Hb at 3.7C. ............................................................................ 81 3 8 % of a) Oxy b) Met and c) Deoxy Hb in washed tilapia muscle containing different forms of Hb at 25C ............................................................................. 82 3 9 Absorption spectra of met Hb, oxy Hb, and CO Hb solutions containing equivalent hemoglobin concentrations ............................................................... 83 4 1 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at a concentration of 6 mol/kg at 3.7C ....................................... 100 4 2 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at a concentration of 6 mol/kg at 25C ....................................... 100 4 3 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at a concentration of 9 mol/kg at 3.7C ....................................... 101 4 4 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at a concentration of 9 mol/kg at 25C ....................................... 101 4 5 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at a concentration of 12 mol/kg at 3.7C ..................................... 102

PAGE 12

12 4 6 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at a concentration of 12 mol/kg at 25C ..................................... 102 4 7 TBARS in washed tilapia muscle containing different forms of Hb at a concentration of 6 mol/kg at 3.7C .................................................................. 103 4 8 TBARS in washed tilapia muscle containing different forms of Hb at a concentration of 6 mol/kg at 25C ................................................................. 103 4 9 TBARS in washed tilapia muscle containing different forms of Hb at a concentration of 9 mol/kg at 3.7C .................................................................. 104 4 10 TBARS in washed tilapia muscle containing different forms of Hb at a concentration of 9 mol/kg at 25C ................................................................. 104 4 11 TBARS in washed tilapia muscle containing different forms of Hb at a concentration of 12 mol/kg at 3.7C ................................................................ 105 4 12 TBARS in washed tilapia muscle containing different forms of Hb at a concentration of 12 mol/kg at 25C ............................................................... 105 4 13 Carbonyl values in washed tilapia muscle containing different forms of Hb at a concentration of 6 mol/kg at 3.7C ............................................................... 106 4 14 Carbonyl values in washed tilapia muscle containing different forms of Hb at a concentration of 6 mol/kg at 25C .............................................................. 106 4 15 Carbonyl values in washed tilapia muscle containing different forms of Hb at a concentration of 9 mol/kg at 3.7C ............................................................... 107 4 16 Carbonyl values in washed tilapia muscle containing different forms of Hb at a concentration of 9 mol/kg at 25C .............................................................. 107 4 17 Carbonyl values in washed tilapia muscle containing different forms of Hb at a concentration of 12 mol/kg at 3.7C ............................................................. 108 4 18 Carbonyl values in washed tilapia muscle containing different forms of Hb at a concentration of 12 mol/kg at 25C ............................................................ 108 4 19 %CO released during 3.7C storage in washed tilapia muscle containing CO Hb at a concentration of 6, 9, and 12mol/kg muscle ................................ 109 4 20 %CO released during 25C storage in washed tilapia muscle containing CO Hb at a concentration of 6, 9, and 12 mol/kg muscle ...................................... 109 4 21 %Oxy Hb in washed tilapia muscle containing different forms of Hb at a concentration of a) 6, b) 9, and c) 12 mol/kg at 3.7C .................................... 110

PAGE 13

13 4 22 %Oxy Hb in washed tilapia muscle containing different forms of Hb at a concentration of a) 6, b) 9, and c) 12 mol/kg at 25C .................................... 111 4 23 %Met Hb in washed tilapia muscle containing different forms of Hb at a concentration of a) 6, b) 9, and c) 12 m ol/kg at 3.7C .................................... 112 4 24 %Met Hb in washed tilapia muscle containing different forms of Hb at a concentration of a) 6, b) 9, and c) 12 mol/kg at 25C .................................... 113 4 25 %Deoxy Hb in washed tilapia muscle containing different forms of Hb at a concentration of a) 6, b) 9, and c) 12 mol/kg at 3.7C .................................... 114 4 26 %Deoxy Hb in washed tilapia muscle containing different forms of Hb at a concentration of a) 6, b) 9, and c) 12 mol/kg at 25C .................................... 115 5 1 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at pH 6.3 at 3.7C .......................................................................... 139 5 2 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at pH 6.3 at 25C ......................................................................... 139 5 3 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at pH 6.8 at 3.7C .......................................................................... 140 5 4 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at pH 6.8 at 25C. ........................................................................ 140 5 5 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at pH 7.3 at 3.7C .......................................................................... 141 5 6 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at pH 7.3 at 25C ......................................................................... 141 5 7 TBARS values in washed tilapia muscle containing different forms of Hb at pH 6.3 at 3.7C ................................................................................................. 142 5 8 TBARS values in washed tilapia muscle containing different forms of Hb at pH 6.3 at 25C ................................................................................................. 142 5 9 TBARS values in washed tilapia muscle containing different forms of Hb at pH 6.8 at 3.7C ................................................................................................. 143 5 10 TBARS values in washed tilapia muscle containing different forms of Hb at pH 6.8 at 25C ................................................................................................. 143 5 11 TBARS values in washed t ilapia muscle containing different forms of Hb at pH 7.3 at 3.7C ................................................................................................. 1 44

PAGE 14

14 5 12 TBARS values in washed tilapia muscle containing different forms of Hb at pH 7.3 at 25C ................................................................................................. 144 5 13 Carbonyl values in washed tilapia muscle containing different forms of Hb at p H 6.3 at 3.7C ................................................................................................. 145 5 14 Carbonyl values in washed tilapia muscle containing different forms of Hb at pH 6.3 at 25C ................................................................................................. 145 5 15 Carbonyl values in washed tilapia muscle containing different forms of Hb at pH 6.8 at 3.7C ................................................................................................. 146 5 16 Carbonyl values in washed tilapia muscle containing different forms of Hb at pH 6.8 at 25C ................................................................................................. 146 5 17 Carbonyl values in washed tilapia muscle containing different forms of Hb at pH 7.3 at 3.7C ................................................................................................. 147 5 18 Carbonyl values in washed tilapia muscle containing different forms of Hb at pH 7.3 at 25C ................................................................................................. 147 5 19 %CO released during 3.7C storage at different pH in washed tilapia muscle containing CO Hb ............................................................................................. 148 5 20 %CO released during 25C storage at different pH in washed tilapia muscle containing CO Hb ............................................................................................. 148 5 21 %Oxy Hb in washed tilapia muscle containing different forms of Hb at pH a) 6.3, b) 6.8, and c) 7.3 at 3.7C .......................................................................... 149 5 22 %Oxy Hb in washed tilapia muscle containing different forms of Hb at pH a) 6.3, b) 6.8, and c) 7.3 at 25C ......................................................................... 150 5 23 %Met Hb in washed tilapia muscle containing different forms of Hb at pH a) 6.3, b) 6.8, and c) 7.3 at 3.7C .......................................................................... 151 5 24 %Met Hb in washed tilapia muscle containing different forms of Hb at pH a) 6.3, b) 6.8, and c) 7.3 at 25C ......................................................................... 152 5 25 %Deoxy Hb in washed tilapia muscle containing different forms of Hb at pH a) 6.3, b) 6.8, and c) 7.3 at 3.7C ..................................................................... 153 5 26 %Deoxy Hb in washed tilapia muscle containing different forms of Hb at pH a) 6.3, b) 6.8, and c) 7.3 at 25C ..................................................................... 154 5 27 Images of washed tilapia muscle containing oxy CO and met Hb at a concentration of 12mol Hb/kg muscle ............................................................. 155

PAGE 15

15 6 1 Lipid hydroperoxide values in washed tilapia muscle at 3.7C with no added NaCl and containing different forms of Hb ........................................................ 171 6 2 Lipid hydroperoxide values in washed tilapia muscle at 25C with no added NaCl and containing different forms of Hb ........................................................ 171 6 3 Lipid hydroperoxide values in washed tilapia muscle at 3.7C with 150 mM NaCl added and containing different forms of Hb ............................................. 172 6 4 Lipid hydroperoxide values in washed tilapia muscle at 25C with 150 mM NaCl added and containing different forms of Hb ............................................. 172 6 5 Lipid hydroperoxide values in washed tilapia muscle at 3.7C with 450 mM NaCl added and containing different forms of Hb ............................................. 173 6 6 Lipid hydroperoxide values in washed tilapia muscle at 25C with 450 mM NaCl added and containing different forms of Hb ............................................. 173 6 7 TBARS values in washed tilapia muscle at 3.7C with 150 mM NaCl added and containing different forms of Hb ................................................................. 174 6 8 TBARS values in washed tilapia muscle at 25C with no added NaCl and containing different forms of Hb ........................................................................ 174 6 9 TBARS values in washed tilapia muscle at 3.7C with 150 mM NaCl added and containing different forms of Hb ................................................................. 175 6 10 TBARS values in washed tilapia muscle at 25C with 150 mM NaCl added and containing different forms of Hb ................................................................. 175 6 11 TBARS values in washed tilapia muscle at 3.7C with 450 mM NaCl added and containing different forms of Hb ................................................................. 176 6 12 TBARS values in washed tilapia muscle at 25C with 450 mM NaCl added and containing different forms of Hb ................................................................. 176 6 13 Carbonyl values in washed tilapia muscle at 3.7C with 150 mM NaCl added and containing different forms of Hb ................................................................. 177 6 14 Carbonyl values in washed tilapia muscle at 25C with no added NaCl and containing different forms of Hb ........................................................................ 177 6 15 Carbonyl values in washed tilapia muscle at 3.7C with 150 mM NaCl added and containing different forms of Hb ................................................................. 178 6 16 Carbonyl values in washed tilapia muscle at 25C with 150 mM NaCl added and containing different forms of Hb ................................................................. 178

PAGE 16

16 6 17 Carbonyl values in washed tilapia muscle at 3.7C with 450 mM NaCl added and containing different forms of Hb ................................................................. 179 6 18 Carbonyl values in washed tilapia muscle at 25C with 450 mM NaCl added and containing different forms of Hb ................................................................. 179 6 19 %CO released during 3.7C storage of washed tilapia muscle containing CO Hb with 0, 150, and 450 mM NaCl added to the system ................................... 180 6 20 %CO released during 25C storage of washed tilapia muscle containing CO Hb with 0, 150, and 450 mM NaCl added to the system ................................... 180 6 21 %Oxy Hb formed in washed tilapia muscle containing different forms of Hb at concentrations of added NaCl a) 0, b) 150, and c) 450 mM at 3.7C ............... 181 6 22 %Oxy Hb formed in washed tilapia muscle containing different forms of Hb at concentrations of added NaCl a) 0, b) 150, and c) 450 mM at 25C ............... 182 6 23 %Met Hb formed in washed tilapia muscle containing different forms of Hb at concentrations of added NaCl a) 0, b) 150, and c) 450 mM at 3.7C ............... 183 6 24 %Met Hb formed in washed tilapia muscle containing different forms of Hb at concentrations of added NaCl a) 0, b) 150, and c) 450 mM at 25C ............... 184 6 25 %Deoxy Hb formed in washed tilapia muscle containing different forms of Hb at concentrations of added NaCl a) 0, b) 150, and c) 450 mM at 3.7C ........... 185 6 26 %Deoxy Hb formed in washed tilapia muscle containing different forms of Hb at concentrations of added NaCl a) 0, b) 150, and c) 450 mM at 25C ........... 186 7 1 Hb form by storage interaction. Change in painty/rancid off odor of tilapia washed muscle samples with different Hb forms added ................................... 209 7 2 pH by storage interaction. Change in painty/rancid off odor of tilapia washed muscle sample s with different Hb forms added ................................................ 209 7 3 NaCl concentration by storage interaction. Change in painty/rancid off odor at different NaCl concentrations ....................................................................... 210 7 4 TBARS by storage interaction. Development of TBARS at different Hbforms 210 7 5 pH by storage interaction. Development of TBARS at differ ent pH levels ........ 211 7 6 NaCl by storage interaction. Development of TBARS at different NaCl concentrations .................................................................................................. 211 7 7 Correlation (p<0.0001) between the development of TBARS and sensory scores ............................................................................................................... 212

PAGE 17

17 7 8 Correlation (p<0.0001) between change in a*value and sensory scores ........ 212 7 9 Correlation (p<0.0001) between the development of TBARS and change in a*value. ............................................................................................................ 213

PAGE 18

18 Abstract of Disse rtation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE PRO OXIDATIVE PROPERTIES OF TILAPIA OXY, CARBOXYAND MET HEMOGLOBIN IN WASHED MINCED TILA PIA MUSCLE By Sara Aldaous May 2010 Chair: Charles A. Sims Major: Food Science and Human Nutrition Hemoglobin (Hb) plays an important role in quality deterioration of fish, affecting odor, flavor, and nutritional value. Autoxidation of oxy Hb produces a very reactive form; met Hb. Increasing stability of Hb can be done by binding carbon monoxide (CO) to the heme porphyrin group. The objective was to compare the prooxidative activity of oxy CO and met Hb as a function of pH, Hb concentration, NaCl concentration, and storage temperature. Washed minced tilapia muscle (WMTM ) was prepared and adjusted to pH 6.3, 6.8 and 7.3. Oxy Hb was isolated from fresh tilapia blood. CO Hb was prepared by flushing oxy Hb with 100% CO for 2 min. Met Hb was prepared by reacting oxy Hb with K3Fe(CN)6. Concentrations of 6, 9 and 12 M Hb forms and 150 and 450 mM NaCl were added to WMTM. Samples were stored at 3.7C for 8 days and 25C for 24 weeks. Lipid oxidation was monitored by following thiobarbituric acid reactive substances (TBARS), lipid hydroperoxides (LOOH), and sensory analysis. Protein oxidation was measured by the DNPH method. Change in color was measured using Color V ision Machine System. Hb oxidation state was monitored spectrophotometrically. Los s of CO from the muscle system was detected using gas

PAGE 19

19 chromatography. Development of off odor (rancidity) during storage at 3.7C was evaluated by trained panelists and correlated with TBARS a*values Dissociation rate of the hemin group from met Hb was measur ed spectrophotometrically at 4 and 25C. Statistical significance was reported as p the General Linear Model and L east S quare M eans (SAS software 9.2). The effect of pH superseded Hb form, Hb concentration, NaCl concentration, or temperat ure in influencing oxidation. Significant lipid, protein, and Hb oxidation was found to increase for all forms of Hb as pH decreased. More deoxy (7%) and met Hb (14%) forms were found at pH 6.3 compared to pH 7.3. TBARS and LOOH were significantly (p ) lower for CO Hb samples compared to other Hb forms regardless of all other factors evaluated. CO Hb samples had significantly higher red color (p after 24 weeks at 25C. As pH decreased the amount of protein carbonyls increased significantly. Low pH also resulted in loss of CO. As the concentration of NaCl increased, lipid/protein oxidation significantly (p and 25C. Met Hb formation increased significantly (p<0.05) with increasing concentration of NaCl at both temperatures. Treatment with CO preserved the red color significantly at 25 C regardless of concentration of NaCl. Low pH (6.3) increased (p both temperatures regardless of all other factors evaluated. CO Hb had significantly higher red color (p 25 C. Freezing temperatures significantly (p 3.7 C were significantly higher than that at 25 C for 24 weeks. Panelists rating of the intensity of the off odor developed during storage was strongly correlated (r=0.95) with the oxidation measured with TBARS. This work suggests that effect of pH supplanted Hb form and

PAGE 20

20 concentration, NaCl concentration or temperature in influencing oxidation. Low pH increased the suscepti bility of WMTM to oxidation, perhaps due in part to the release of hemin at lower pH. Treatment of WMTM with CO significantly decreases lipid/protein oxidation.

PAGE 21

21 CHAPTER 1 INTRODUCTION Hemoglobin (Hb) plays an important role in quality deterioration of fish muscle by promoting lipid oxidation and leading to color and flavor changes. Hb possesses catalytic abilities that allow it to produce peroxy radicals, alkyl radicals and other produc ts that make it a powerful catalyst of lipid oxidation in fish muscle causing oxidative rancid ity ( 1 ) Additionally, the oxygentransport proteins undergo autoxidation producing the met form which is believed to be far more reactive than the reduced form, since it is capable of reacting with peroxides to form compounds that cause oxidation. To delay formation of the met form, it is important to incr ease the stability of Hb. This can be done by the bi nding of specific ligands such as carbon monoxide (CO) to the heme porphyrin group, which is a common practice in the seafood industry to stabilize red color. Little is k nown how this form of Hb (carboxy Hb) in fish compares in function and structural properties to the other common forms, oxy and met Hb Lipids are chemically unstable food components, and can undergo free radical chain reactions ( 2, 3 ) resulting in rancidity ( 2, 3 ) and degradation of the food product. Lipid oxidation in muscle foods is a complex process which is referred to as lipid peroxidation ( 4, 5 ) Unsaturated fatty acids react with molecular oxygen to undergo autoxidation. The exact mechanism of this reaction is disputed. The direct reaction of unsaturated fatty acids with molecular oxygen is not thermodynamically favorable. Simic and others ( 6 ) report that autoxidation may be initiated by H2O2, or the free radicals or exogenous initiators such as UV, ionizing radiation, and heat. Another source of lipid peroxidation within muscle involves metal ions including Fe++, which can abstract a proton from unsaturated fatty

PAGE 22

22 acids ( 7 ) Two major sources of Fe within both meat and fis h muscle are found in Hb within red cells and myoglobin within the tissues. Research O bjectives This research was to elucidate a better understanding of the pro oxidative activity of oxy carboxy and met hemoglobin on the stability of fish muscle and the development of lipid oxidation. I nformation obtained can be used to develop more effective str ate gies for preserving the quality of fish during storage. Specific objectives are: 1. To investigate the ef fect of different forms of Hb ( oxy CO, and m e t Hb ) on the development of primary and secondary products of oxidation in tilapia washed muscle system. 2. To investigate the effect of differ ent concentrations of added oxy CO and met Hb (6, 9, and 12 mol/kg muscle) on the development of primary and secondary products of oxidation in tilapia washed muscle system. 3. To i nvestigate the effect of different pH values (6.3, 6.8, and 7.3) on the development of primary and secondary products of oxidation in tilapia washed muscle system and tilapia muscle 4. To i nvestigate the effect of different concentrations of added NaCl (150 and 450 mM) on the development of prim ary and secondary products of oxidation in tilapia washed muscle system and tilapia. 5. To i nvestigate the effect of storage at diff erent temperatures (3.7 versus 25C) on the development of primary and secondary products of oxidation in tilapia washed muscl e system 6. To i nvestigate the effect of oxidation on the development of odor and color changes during storage of samples at 3.7C.

PAGE 23

23 Research Significance In recent years CO has been used to increase the stability of foods against spoilage and loss of co lor. When CO binds to Hb and myoglobin ( Mb ) in fish muscle, it increases its stability by slowing its autoxidation reaction, and decreases its ability to participate in lipid oxidation. The binding of CO to the heme molecule is also affected by pH. It is a lso known that the presence of NaCl, which is commonly used as an additive during food processing, can promote lipid oxidation. The significance of this study could give a new direction for the use of CO in the prevention of lipid oxidation in fish muscle during storage. It will further shed light on how CO suppresses the oxidation of oxy Hb to met Hb. In addition, the interaction between lipid oxidation and heme oxidation will be evaluated in relationship to protein oxidation. Further, the effect of pH and salt concentration will be examined in relationship to lipid oxidation. An investigation into these conditions is expected to give a better understanding of oxidation processes mediated by Hb in fish muscle and thus allow for better control of lipid oxi dation.

PAGE 24

24 CHAPTER 2 LITERATURE REVIEW Lipid Oxidation in Fish Muscle Lipi d oxidation is one factor contributing to food spoilage, affecting many aspects of food quality including nutritional value, odor, flavor, functionality and appearance ( 1 ) Muscle lipids and proteins are substrates for oxidation. The two major groups of fish muscle lipids are triacylglycerols and phospholipids ( 8 ) Triacylglycerols are found both within muscle cells, especially in fatty species of fish and outside the muscle cells, but the amount varies with species and environmental conditions. Triacylglycerols are composed of fatty acids with straight chains and 16 or 18 carbon atoms ( 9 ) Phospholipids give structure to the cell membrane and are found in white muscle of fish ~ 0.5% 1 % (w/w) and in dark muscle of fish at higher levels due to the presence of more mitochondria. Phospholipids contain more C 20 and C 22 unsaturated fatty acids. The polyunsaturation is about 15 times greater than that of triacylglycerols, making them more susceptible to oxidation because oxygen attacks the double bonds of fatty acids to form peroxide linkages. Lipid oxidation in muscle foods is a complex proces s which is referred to as lipid peroxidation ( 4, 5 ) Unsaturated fatty acids react with molecular oxygen to undergo autoxidation. The exact mechanism of this reaction is disputed but is believed to occur in three stages: initiation, propagation, and termination. The direct reaction of unsaturated fatty acids with molecular oxygen is not thermodynamically favorable. Simic and others ( 6 ) report that autoxidation may be initiated by H2O2, or the free radicals endogenous catalystsHb, Mb low molecular weight transition metals, lipoxygenases, microsomal

PAGE 25

25 enzymes and mitochondrial enzymes ( 8 ) or exogenous initiators such as UV, ionizing radiation, and heat. Another source of lipid peroxidation within muscle involves metal ions including copper and Fe++, which can abstract a pr oton from unsaturated fatty acids ( 7 ) Two major sources of Fe within both m eat and fish muscle are Hb within red cells and Mb within the tissues. Hb transports oxygen, CO2, and hydrogen. It is present in the blood of muscle in most meats and fish, and is considered responsible for lipid oxidation in muscle products ( 10, 11 ) Mb is more predominant in the dark muscle of fish, and can also lead to undesirable qualities in fish muscle when the heme iron is oxidized ( 7, 11 13) Cytochromes, which are also heme proteins, are associated with membrane systems, especially mitochondria. At post mortem, if H2O2 is high, iron m ay be released from heme proteins, and can initiate lipid peroxidation ( 14 ) A number of protein complexes with sulphydryl groups also bind iron such as transferritin and ferritin. Ferritin is a nonheme iron that stores iron unti l it is needed by various processes within the cell. It can store up to 4500 iron atoms as Fe(O)OH ( 15 ) and this can be released by chelators. Superoxide anion, ascorbate and thiols release Fe(O)OH by reduction processes. The two atoms of iron per transferrin molecule is difficult to release ( 14) Lipoxygenases and cyclooxygenases may play a role in lipid oxidation in seafood. Hultin ( 14) suggests that no direct evidence exists that these enzymes are important catalysts in post mortem fish. Rather, he suggests that they may contribute to lipid oxidation by contamination of the muscle components from the skin of fish, occurring during deboning and deskinning of minced tissue. However, Frankel ( 16) says

PAGE 26

26 lipoxgenases are recognized catalysts of lipid oxidation in fish, in which lower temperatures promote enzymatic oxidation. The lipoxygenase found in gill and skin tissues of fish produce hydroperoxides from polyunsaturated fatty acids in fish. This enzyme can be inactivated by heating above 60C. Lipid Composition in Fish Two types of lipids are found in fish: phospholipids and triacyglycerides. Phospholipids, which are highly unsaturated, account for 1% (w/w) of lipids contained in the muscle of lean fish and are the maj or components of cell membranes ( 8, 17) Approximately 75% of the fatty acids within fish are mono and polyunsaturated fatty acids ( 18 ) Triacyglycerides, the other major lipid in fish, are found in adipocytes and act as a s ource of storage energy. Total lipids of fish vary not only among species, but also from one part of the fish to another. Lean fish species such as cod and hake store the majority of lipids in the liver. The light muscle of lean fish contains approximately 1% phospholipid ( 19) U nsaturated fatty acids of fatty fish species ( 20 ) can undergo oxidative degradation, particularly the polyunsaturation found in membranes of muscle tissue ( 21) The amount of lipids present is not the controlling factor in lipid oxidation but rather the presence and amount of Hb ( 8, 22) Less than 0.1% levels of phospholip ids are necessary for H b mediated oxidation to occur ( 23) Although the percentage of phospholipids is low in washed mince muscle systems, phospholipids are considered the primary substrate for lipid oxidation becau se they have 100 times more surface area than tryacylglycerols and the fatty acids of phospholipids are more unsaturated than those in tryacylglycerols ( 24 ) Richards and Hultin ( 23) found that increasing membrane phospholipids six fold did not affect the extent of lipid oxidation or the extent of rancidity

PAGE 27

27 development during storage of washed cod containing added blood. When trout hemolysate (5.8 mol Hb/kg washed cod) was added to cod myosin preparation at 2C, rancidity and TBARS formation did not develop during six days of storage suggesting that at least trace amounts of lipid are required for rancidity to occur. Inhibitors of Lipid Oxidation Initiation inhibitors remove active reduction products of oxygen or convert transition me tals to inactive forms ( 14) Catalase and perox idases remove hydrogen peroxide, while s uperoxide dismutase removes superoxide anion. Phospholipases inhibit lipid oxidation by inhibiting the ability of the membranes to oxidize i ts lipids. Propagation inhibitors convert free radicals to stable compounds. These inhibitors are either water soluble or lipidsoluble. Ascorbate and the glutathione peroxidase system are water soluble propagation inhibitors. Tocopherols in fish muscle ar e lipid soluble antioxidants and are found in the unsaturated fatty acids membranes. Both the initiation inhibitors and propagation inhibitors act by donating electrons and will be eventually oxidized. When this happens, rapid oxidation will occur. Changes in Post mortem Fish Changes that take place in post mortem storage of fish muscle make it more susceptible to lipid oxidation ( 14 ) During storage at refrigerator temperature, l ow molecular weight iron in light muscle almost doubled from 155 ppb to 225 ppb ( 25) most likely coming from ferritin. Within 10 days post mortem, the reduced forms of Hb and Mb were oxidized to met Hb and met Mb Loss of reducing compounds over time such as ascorbate ( 25) NAD(P)H ( 26 ) and glutathione ( 27 ) decreased the reducing capacity of fish muscle. Muscle cells lose the ability to maintain ion grad ients, especially calcium ions which can activate enzymes such as lipases, phospholipases, and

PAGE 28

28 proteases. Mincing of fish muscle exposes the tissue to more oxygen than the muscle receives with the living fish. Removal of oxygen will inhibit or stop lipid oxidation ( 28 ) ( 29) There is also loss of energy sources such as ATP. The post mortem pH of fish muscle can be 7.0 or higher in white fish such as cod to as low as 5.5 in some red meat fish. Lower pH is associated with increased lipid oxidati on due to higher deoxy Hb content. Protein Oxidation One of the common ways to increase the shelf life of fish is frozen storage. However, frozen storage and fluctuating temperatures can have a negative effect on the quality of fish ( 30) Some quality changes which cannot be attributed to lipid oxidation solely during frozen storage are toughness, changes in texture, loss of juiciness, and loss of protein functional properties ( 31 ) Little is known about the interaction between protein and lipid oxidation. It has been shown that the products of lipid oxidation can interact with proteins to produce oxi dation ( 32 ) Dalle Donne and others ( 33) describe the production of protein carbonyl derivatives (aldehydes and ketones) as occurring in four ways: (1) direct oxidation of amidation pathway or through oxidation of glutam unsaturated aldehydes from lipid peroxidation and (4) reaction of reducing sugars or their oxidation products to the amino group of lysine residues (glycation and glycoxidation). Studies of protein ca rbonyl derivatives cannot distinguish how they have been formed, and protein carbonyl derivatives are therefore considered a broad marker of oxidation. Ostdal and others ( 34 ) have demonstrated that proteins can transfer fr ee radicals to other molecules such as lipids. Baron and others ( 31) investigated prolonged frozen

PAGE 29

29 storage of rai nbow trout to determine if oxidative changes in the protein fraction correlated with changes observed in the lipid fraction. The authors measured lipid oxidation using peroxide values, and protein oxidation was assessed with UV spectroscopy of protein carbonyl groups and SDS PAGE and immunoblotting. No significant levels of protein carbonyls were found in rainbow trout stored for 13 months at 80C and 30C, indicating that little protein oxidation had occurred. However, fish stored at 20C showed a signi ficant increase in protein carbonyls (7.7 nmol/mg of protein) after 8 months of storage, demonstrating that protein oxidation had occurred. Immunoblotting of fish stored at 20C revealed oxidation in bands that were identified as myosin and actin. The thr ee storage temperatures ( 20C, 30C, 80C) showed small differences in protein oxidation patterns and levels with samples at 20C slightly more oxidized compared to the other two temperatures. Levels of lipid hydroperoxides after 8 months of storage at 20C were significantly increased with an even greater increase after 13 months of storage. Samples stored at 80 and 30C did not show any significant increases in peroxides. Comparison of samples stored at 20C indicated that lipid and protein oxidat ion followed the same pattern and suggests that protein and lipid oxidations were simultaneous. However, quantification of secondary volatile oxidation products using headspace GC MS revealed that lipid oxidation proceeded differently in fish stored at 30 C compared to fish stored at 80C. After 8 months of storage at 30C significant increases in volatile oxidation products were detected. At this temperature, similar increases in protein carbonyls were not detected. Eymard and others ( 35 ) investigated the link between lipid oxidation and protein oxidation during processing and storage of horse mackerel. Using fish minces with

PAGE 30

30 differences in lipid and protein fractions and different oxidative levels, the authors were able to compare lipid and protein oxidation development. Protein carbonyls were measured as described by Levine and others ( 36) Horse mackerel fillets were processed into mince samples. The four samples included (MO) mince after the grinding, (M1) mince after the first dewatering, (M2,) mince obtained after the sec ond dewatering, and (M3) mince obtained after the third dewatering. The authors found that the protein carbonyl was lowest in MO, with protein oxidation developing very rapidly during storage at 5C and reaching the maximum after 12 hours of storage. For t he washed minces, high carbonyl levels were detected at times zero and did not increase significantly. The authors a lso demonstrated with that at time zero, myosin and other high molecular weight proteins were already oxidized in all four products including the unwashed mince. The authors further demonstr ated that the loss of Hb and Mb was most pronounced after the first washing. Some iron remained after the third washing and it was speculated that heme, which is very hydrophobic, may have been incorporated into the membranes. Since iron and heme proteins are prooxidative, their removal is a crucial step in obtaining stability against oxidation. Further, the authors found little lipid oxidation at times zero but significant oxidation in the washed samples. The authors also found that washing removed storage proteins more easily than the membrane phospholipids which are very sensitive to oxidation, which increased significantly after the second wash. The authors conclude that lipid and protein oxidation devel oped simultaneously but it was difficult to determine how they are linked. Lipid and protein oxidation share the same catalysts, and they can develop independently of one another, or in parallel, or they can interact with each other.

PAGE 31

31 Kjrsgrd and others ( 30) sought an understanding of the nature of protein oxidation by conducting proteome analysis of rainbow trout muscle proteins fractionated as low salt (LS) and hi ghsalt (HS) soluble, and identified oxidized proteins by LC MS/MS. Analyses of fish, fresh or stored for 3 years at 80C, showed no difference in the total level of carbonyl protein. Comparison of storage temperatures at 20C versus 80C over a twoyear period revealed that frozen storage at 20C produced two times higher levels of protein carbonylation and 10 times higher levels of peroxides. Immunoblots of proteins soluble in LS and HS buffer showed no clear differences in carbonylation at 80 and 20C. High levels of carbonyls were found in muscle proteins fractionated as LS and HS. In the HS soluble protein fraction, three prot eins with a size around 16 kDa, all identified as nucleoside diphosphate kinase (NDPK) showed alterations in concentration with concentrations 10 times higher for fish stored at 20C compared to 80C. The authors conclude that frozen storage at 20C of rainbow trout results in increased protein carbonylation and induces changes in protein solubility. Role of Heme Proteins in Oxidation Hb distributes oxygen to the tissues of the body and MB gives muscle meat its red pigment (along with Hb) and is locate d in the muscle cell ( 5 ) Both of these proteins bind oxygen and contain Fe(II) in the center of the heme group ( 37) Mb contains one polypeptide chain with one porphyrin ring containing one iron atom in the heme pocket ( 38) ; Hb is made up of four polypeptide chains, each containing a heme group. (Heme refers to the porphyrin ring with ferrous [Fe2+], hemin refers to the porphyrin ring containing ferric [Fe3+]). Fe is bound to four nitrogen molecules in the heme pocket. The bright red color of muscle tissue is due to iron in its reduced form (MbFe+2O2). When Fe+2 (ferrous) is converted to Fe+3, autoxidation occurs. Cutting, freezing and

PAGE 32

32 thawing of fish dark muscle makes it more susceptible to autoxidation, resulting in the formation of met Mb (Mb Fe+3H2O), which is brown in color ( 37 ) During autoxidation, oxygen is released from oxy Hb to form the superoxide anion radical, O2 and ferric m et Hb which is converted to hydrogen peroxide which increases the ability of heme proteins to promote lipid oxidation. A s demonstrated in Diagram 1, an adaptation of Baron and Andersens ( 5 ) dynamic of Mb whic h reacts similarly to Hb, two active forms of Hb are present. Both d eoxy Hb (HbFe(II)) and oxy Hb (HbO2Fe(II)) exist in the reduced state and thus are susceptible to autoxidation. As can be seen from the diagram these two reduced forms are oxidized to the m et Hb (Hb Fe(III)) which in turn is reduced again to the d eoxy Hb (Hb Fe(II)). Met Hb in the presence of H2O2, can also be converted to the very prooxidative forms, perferryl Hb (H b Fe(IV)=O) and ferryl Hb (Hb Fe(IV)=O). However, Richards and Hultin ( 4 ) reported that in the presence of 3.4 M of li pid peroxide, reduced Hbs (oxy/deoxy Hb s) produced high levels of lipid peroxide formation, but m et Hb caused little peroxidation of linoleic acid. A reason for this is m et Hb cannot autox idize but the reduced Hbs form superoxide anion radicals during autoxidation which can form hydrogen peroxide. The hydrogen peroxide will then activate m et Hb ( 39) If m et Hb is the initial reactant there is no source of oxygen to form hydrogen peroxide. In addition, reduced Hbs but not m et Hb can produce hydroxyl radicals ( 40 ) Met Hb forms hemichromes which are not considered catalysts of lipid peroxidation ( 41 ) If deoxy oxy or m et Hb is denatured or their structures are altered, low spin iron(II) hemochr omes or iron(III) hemichromes may form which can be

PAGE 33

33 reversible or irreversible reactions but their role in lipid oxidation has received little attention ( 5, 42 ) Deoxy Hb content affects heme protein autoxidation. Deoxy Hb in the presence of O2, is susceptible to rapid oxidation ( 1, 4 ) However, fully oxygenated Hb was found to react slower with hydrogen peroxide because the accessibility of the heme iron is inhibited by the O2 ligand in oxygenated molecules ( 43 ) Oxy Hb is also more compact compared to d eoxy Hb which has greater flexibility in the heme pocket ( 43, 44 ) However, despite the low d eoxy Hb co ntent found for perch Hb, Richards and Dittman ( 1 ) found that perch Hb had a high autoxidation rate. This could be explained by the amino acid sequences near the heme crevice. It has been found that disrupting the hydrogenbonding network of His97 created easier accessibility of H2O into the heme crevice ( 5, 45) This would then accelerate the formation of Met Hb which would also increase lipid oxidation. Met Hb studies have not demonstrated a prooxidative activity at physiological pH ( 4648 ) Sato and Hegarty ( 49) were not able to demonstrate the catalytic activity of m et Hb in meat and other studies have supported this ( 5052) However, these studies used washed muscle systems which could underesti mate the pro oxidative activity of m et Hb by washing away compounds such as hydroperoxides which could be important for the prooxidative activity of m et Hb ( 7 ) Recent research has found that m et Hb is a potential prooxidant at the pH f ound in fresh meat (between 5.3 and 6.2) which emphasizes the importance of hydroperoxides in m et Hb initiated lipid oxidation ( 53 ) Electrostatic and hydrophobic interactions are inv olved when Hb binds to phospholipids. Met Hb affects the structural and physiochemical parameters of the lipidwater interface

PAGE 34

34 ( 54) This results in the formation of hemichrome, a poor initiator of lipid oxidation ( 41 ) at physiological pH in model systems containing longchain free fatty acids, resulting in noncatalytic activity. At lower pH values, the electrostatic and hydrophobic interactions are not involved, most likely due to the different charge distribution on both the fatty acid and the heme protein ( 5 ) Thus, in the presence of lipids, m et Hb at physiological pH, due to the formation of the noncatalytic hemichrome, can undergo rapid neutralization. However, in a high lipophilic environment, denaturation of the hem e protein may result in exposure of the heme group to the surrounding lipids and induce lipid peroxidation. However, at lower pH values, m et Hb is able to initiate lipid oxidation in a lipid hydroperoxide dependent mechanism. Richards and Hultin ( 11 ) found that increasing the concentration of Hb in washed mince cod increased lipid oxidation. Hemolysate containi ng 0.06 and 0.50 mol Hb/kg tissue (pH 6.3) did not develop rancidity during storage at 2C as measured by sensory scores and TBARS. However, rancidity developed at day 1 and day 2 for hemolysate containing 5.8 and 1.8 mol, respectively. These authors als o found that although Hb concentration was higher in whole trout muscle than in light muscle of mackerel, more oxi dation occurred in the light muscle during storage on ice. It was postulated that this difference in oxidation susceptibility is based upon different types of Hb. At p ost mortem pH, anodic Hbs bind oxygen poorly while cathodic Hbs retain a high affinity for oxygen. This is important because d eoxy Hb is a more effective catalyst of lipid oxidati on as compared to oxy Hb and m et Hb Richard and others ( 55) demonstrated that d eoxy Hb is more prooxidativ e than oxy Hb in washed cod muscle. From their data, the reason for this could not be determined but the authors speculated that it c ould be due to ferryl -

PAGE 35

35 Hb forming more readily, or the release of hemin, or better access of lipid hydroperoxides to the hem e crevice in the oxy Hb or a combination of these mechanisms. Richards and others ( 56 ) compared the oxidative ch aracteristics of trout M b versus trout Hb and f ound that Mb was a weaker promoter of lipid oxidation. This was explained in part, by the fact that the reactive heme group in Met Hb is more loosely anchored in the globin and thus more capable of initiating lipid oxidation. Additionally, the m et Hb prote in can react with hydrogen peroxide or lipid peroxides and form the ferryl protein cation radical which in turn, can initiate lipid oxidation. Further, the authors measured heme dissociation from trout Mb and Hb and found that heme dissociated much fas ter from anodic trout Hb than from trout Mb, suggesting that heme dissociation play s a role for various Hbs to promote lipid oxidation. In an earlier study Richards and others ( 55) found a 14fold dif ference in the rate of TBARS formation during 1.5 days of storage (pH 6.3, 2C) for anodic Hb as compared to cathodic Hb. This provided further evidence that heme dissociation is a major contributing factor for different Hb s to promote lipid oxidation. The heme group within both Hb and Mb contains iron in its reduced state when muscle is fresh. When iron is oxidized, it contributes to the brown color of fish muscle ( 57) This oxidation also produces more active forms of Hb and Mb, the met, ferryl and perferryl forms, which leads to more lipid oxidation ( 11 ) One way to reduce the prooxidative activity of these heme proteins is to maintain the proteins in their reduced state. For example, Kristinsson and others ( 57 ) investigated the effect CO gas treatment had on tilapia Hb and found that treatment with CO greatly stabilized the protein with

PAGE 36

36 respect to its pro oxidative potential (in a model linoleic acid system) since it was maintained in the reduced state. Loss of Hemin Mechanisms of hemin loss using Hb and Mb variants were investigated by G runwald and Richards ( 58 ) Comparing human Hb, which was genetically cross linked to prevent tetrameric Hb from dissociating into subunits and a Mb variant from sperm whale in which the native valine residue was substituted with threonine (68th site) to provide high hemin affinity versus the native variants of both Hb and Mb, the authors found that higher hemin affinity in the genetically altered variants resulted in decreased lipid oxidation. TBARS formation occurred more rapidly with the native Hb at pH 5.7 in washed cod muscle during storage at 2C. Similarly, t he genetically altered Mb was less effective in promoting lipid o xidation than the native Mb as measured by TBARS formation in washed cod. To further determine the effect of hemin on lipid oxidation, addition of hemin and hemin with bovine albumin to washed cod was examined during storage at 2C. TBARS reached a maximu m formation after two days of storage with hemin alone. The addition of albumin to hemin increased the onset of TBARS formation. The authors suggest albumin delivers hemin into the lipid phases which increases the ability of hemin to oxidize the lipids. Ha rgrove and others ( 59) develop an assay for hemin dissociation. His64 in sperm whale myoglobin was replaced by Tyr, producing a holoprotein with the discrete green color. Val68 was replaced with Phe in the same protein to increase its stability while retaining high affinity for hemin. This protein can then be used for complete ext raction of hemin from H b and Mb, giving absorbance changes to allow reactions at low hemin concentrations. When this protein (apoprotein Tyr64Val68) is mixed in excess with

PAGE 37

37 mutant Hbs, the solution turns from brown to green, and the absorbance changes can be used to measure the rate of dissociation of hemin. Mutant hybrids of human Hb were prepared by substituting Gly at His64 (E7) and oxidized with ferricyanide and then reacted quickly with apoprotein Tyr64Val68. Hemin dissociation from the Gly64 mutants and subsequent uptake by apoprotein Tyr64Val68 resulted in rapid absorbance increases. Dissociation of hemin from Tyr64 mutant Mb at pH 5.0, 37C is attributed to a protonation of the proximal His93 imidazole side chain, which disrupts the Fe3+ His93 bond. The authors conclude that the results presented indicate that apoprotein Tyr64Val68 can be used reliably for measuring hemin loss. Hemin loss from oxi dized trout and perch Hb and bovine Hb were measured by mixing them with an apoglobin form of H64Y sperm whale Mb at 25C and pH 5.7, 6.3, and 8.0 ( 60) The aut hors found that perch Hb had approximately 50 fold higher hemin loss rates at pH 5.7 and 6.3 compared with bovine Hb Hemin loss rates from trout Hb were 10 to 30 fold faster than bovi ne Hb at the same low pH values. Ara nda and others ( 60 ) suggests that there are four mechanisms by which auto oxidation and hemin loss occur. These include steric displacement of bound ligands, weak anchoring of the heme propionate to the globin, larger channels for solvent entry into the heme pocket, and weakened interactions with the distal histidine. The rate of autoxidation is increased 15 fold when Val (E11 side chain) is replaced by Il e (E11) in sperm whale Mb. located closer to the heme iron atom in the heme pocket than Val (E11). When the secbutyl side chain and the bound ligand is shorter (1.8 2.6 ), strong steric clashes occur with the bound ligand. Thus, the

PAGE 38

38 pr esence of Ile in fish Hb subunits hinders the bound ligand, and enhances the rate of OOH dissociation and autoxidation in perc h and trout Hbs ( 60 ) In perch and trout Hbs, anchoring of the heme propionate. Thr E10 in the subunits of both fish are too short to hydrogen bond through the water molecule to the 7propionate so there are no favorable electrostatic interactions wit h the heme group. Hemin loss increases when this type of bonding is not present ( 61 ) Trout I V met Hb alpha chains at pH 5.7 and 6.3 have a much shorter distance (2.7 2.8) betwee O atom than perch met Hb (3.2 and 4.7 ) at pH 6.3 and 5.7, which indicates little or no electrostatic stabilization of the heme6 propionate. Previous studies have shown that if the heme 6 propionate cannot be stabilized, a higher rate of hemin loss from human beta subunits occurred ( 62) Liong and others ( 45 ) have shown that increased access of water molecules to the heme pocket increases hemin release in m utants of Mb. The amino acid residue at the CD3 position is not able to hydrogen bond to the heme6 propionate O atoms. The further away this residue is from the heme pocket, the easier it is for water to enter the heme pocket increasing auto oxidation an d hemin loss. These distances will vary with p H. Compared to bovine Hb distance between the CD3 side chain and the heme6 proprionate O atom. This increase in distance increases autoxidation and hem in loss ( 60 ) The electrostatic interaction of the hydrogen bonding of the distal His to bound O2 inhibits auto oxidation by limiting access of solvent to the heme crevice ( 63) The rate of hemin loss will also

PAGE 39

39 be decreased if hydrogen bonding of the distal His to water in oxidized Hb occurs ( 45, 61) Auto oxidation and hemin loss for Perch, trout IV, and bovine Hb are pH dependent. As the pH decreases, auto oxidation and h emin loss increase. Ar a nda and others ( 60) suggested that the reason for the pH dependence may involve the protonation of distal His resulting in disruption of the bound O2. This disruption might cause a rotation of the side chain. Another reason is the dissociation of OOH by the bound O2, which result in the formation of met Hb It could also be due to the interruption of the electrostatic forces with amino acids at E10 and CD3 positions Higher rates of auto oxidation and hemi n loss occur at higher pH, for trout and perch Hb as compared to bovine Hb indicating higher rates ( pKa values) for protonation of HisE7, bound O2, HisF8, and the heme propionates. Met Hemoglobin Reduction Hb and Mb share structural properties and physiolog ical functions ( 64 ) However, the mechanism for reducing m et Hb is not t he same for both of these proteins. Work by Arihara and others ( 64 ) has shown that met Mb reduction occurs in the mitochondrial fraction skeletal muscle and at the sarcoplasmic reticulum. OM cytochrome b or cytochrome b5, both of which act as electron transfer mediators, are required for NADH cytochrome b5 reductase to reduce met Mb the former at the mitochondria, and the latter at the sarcoplasmic reticulum. However, in humans, it has been shown that NADHcytochrome b5 reductase uses cytochrome b5 to reduce met Hb The absence of the enzyme NADH cytochrome b5 reductase, causes a n accumulation of met Hb (met hemoglobinemia) in humans. NADH cytochrome b5 reductase pathway has been the only one found as an electron transfer mediator in the erythrocytes of humans as an

PAGE 40

40 electronic transfer mediator ( 64) OM cytochrome b has not been found in the erythrocytes. Although the reduction pathways for met M b and met Hb are not same, these two heme proteins share some factors that affect their reducing activity. In their review article (Bekhit and Faustman ( 65) ) several stu dies indicate that met Mb reducing activity during assay is temperature dependent for different species. For example, the optim al temperature for met Mb reductase in the muscle of tuna is 25C whereas that for bovine cardiac muscle is 37C. Further, at low pH temperature effects are small, but Reddy and Carpenter ( 66 ) reported that met Mb reducing activity increased markedly with increasing the temperature from 4 to 30C. Similar to the effect of temperature at time of assay, met Mb reducing activity is pH dependent and also related to purified preparations versus extr acts and assay conditions ( 65) Conflicting stu dies report optimal pH levels. For example, using methylene blue and NADH, Taylor and Hochstein ( 67 ) reported that the optimum pH for bovine cardiac m et Hb reductase activity was 7.0, but Hagler and others ( 68) reported the m et Hb reductase activity with puri fied bovine cardiac muscle was greatest at pH 6.5. Differing techniques and methodologies from different researchers results in different conclusions. Bekhit and Fauustman ( 65) further suggest m et Hb reducing a ctivity during storage time gives var ying results due to different methodologies. Additionally, the actual storage time was different between studies presented so no general conclusion can be drawn about how storage time affects met Hb reduction. Role of Carbon Monoxide (CO) The autoxidation and the resultant brown coloring can be prevented by treatment of the fish muscle with CO which has a greater affinity (>240 times more than oxygen)

PAGE 41

41 for the Fe(II) binding site, forming carboxy Hb (CO Hb) and carboxy Mb (CO Mb), resulting in a cherry red c olor that is stable over longer periods. When C O is bound to heme in Hb, it gives stability to the protein, and the protein will resist autoxidation on heating, freezing, and thawing ( 69, 70 ) T o improve the quality and preserve the attractive red c olor of dark muscle, CO or tasteless smoke (TS) which contains CO gas have been used by the seafood industry ( 37 ) to preserve the red "fresh" appearance of seafood products. Kristinsson and others ( 69) have shown that treating yellowfin tuna significantly enhanced the red color, reduced lipid oxidation, and lowered aerobic bacterial growth. Anderson and Wu ( 37) used GC /MS to determine CO in tuna and mahi mahi tissues in vacuum packaged products and CO treated frozen mahi mahi samples. Samples of vacuum packaged tuna showed CO l ev els about 1 mcg/g while the CO treated frozen m ahi mahi samples had 500 ng/g. This quantitative determination of CO can be useful for regulatory purposes in determining exposure to CO. Replacing oxygenated Hb and Mb with CO Hb and CO Mb in the dark muscl e of fish may offer some protection against lipid oxidation ( 69, 70 ) High concentrations of unsaturated fatty acids in dark muscle of fish, in the presence of oxy Hb and d eoxy Hb may result in increased levels of lipid peroxides. This is due mainl y to the autoxidation and deoxygenation of Hb as pH was reduced. Kristinsson ( 69 ) re ported that, at lower pH, the protein is partly unfolded, giving the heme portion greater ability to participate in oxidation. Autoxidat ion of the heme protein to the m et form is a critical step in lipid oxidation. CO treatment re tards autoxidation of Hb and Mb to the ferric form, and thus plays a role in decreasing lipid oxidation. Danyali ( 71 ) found that CO and filtered smoke (FS)

PAGE 42

42 treatment can be effective in retarding lipid oxidation. FS which contains phenolics that have the potential to serve as antioxidants, was not more effective than the other gas treatments in retarding oxidation development. After tuna steaks were treated with FS, 4%, 18% and 100% CO, Kristinss on and others ( 72 ) measur ed lipid oxidation using TBARS (thiobarbituric acid reactive substances). They found that the 4% CO treated tuna had higher TBARS than untreated tuna. The authors theorized that the higher levels of O2 and CO2 in the 4% CO may have promoted oxidation of heme protein. The 18% CO and FS (containing approximately 18% CO) may have protected the heme molecule against oxidation by CO2. After freezing and cold storage (4C) for 4 days, TBARS gradually decreased for all treated samples. The authors attributed the r eduction in TBARS to muscle proteins binding the MDA (malondialdehyde) making less available in the TBARS assay. Garner and others ( 73 ) described decreased lipid oxidation in the white and dark muscle of Spanish mackerel that had been treated with CO and FS Additionally, Garner and others ( 73) reported that 100% CO was more effective than 100% nitrogen in retarding lipid oxidation, which suggests that the effect is not due solely to the absence of oxygen. Kristinsson and others ( 72 ) proposed that treating yellowfin tuna with medium to high levels of CO may stabilize the heme protein molecule against lipid oxidation and preserve the red color of muscle foods by binding with Fe+2 of muscle heme, forming a cherry re d carboxy Mb/Hb. Levels of CO found in fish muscle after treatment with CO are not considered hazardous to the health of humans. During cooking, approximately 85% of CO is lost ( 74) However, the stable red color produced by CO can mask spoilage beyond the

PAGE 43

43 r ecommended shelf life of meat ( 75) In United States, exact determination of the l egality of using CO to preserve fish is yet to be ascertained. Modified Atmosphere Packaging (MAP) Using CO Modified atmosphere packaging (MAP) replaces air with a single gas or a mixture gases. These gases include oxygen, nitrogen, carbon dioxide and carbon monoxide, depending upon color stability requirement, growth of microbial agents and the sensitivity of the product to the particular gases that are used ( 76) FDA has reviewed the use of CO in packaging for meat produc ts and has classified it as Generally Recognized as Safe (GRAS ). Low levels of CO in MAP cause no risk to humans. Further, the USDA Food Safety and Inspection Service (FSIS) does not require labeling for modified atmosphere gases, including CO. The low lev els of CO (0.4%) in MAP stabilize the natural red color of fish and meat Caution must be taken by consumers because the red color which indicates freshness to the consumer may last longer than the shelf life of the product. Thus it may mask the spoil age of the product. Srheim and others ( 74) reviewed the use of CO in the modifiedatmosphere packaging of meat and found that for the past 10 years the Norwegian meat industry has been using a combination of gases in MAP that included 0.3 0.4% CO which increased the microbial shelf life and maintained the cherry red color. Evaluation of the stability of this color ( 74 ) in ground beef patties indicated that a mixture of air and CO in MAP stabilized the color for 15 d compared with 5 d for air. Recently, the FDA has declared ( 77, 78) that the use of CO in MAP is a processing aid and does not require product labeling. The 0.4% of CO used in MAP for meats met the legal definition of a processing aid and there would not be significant amounts of CO in the finished product. This declaration paves the way for the use of CO in MAP of fresh fish ( 79 ) Consumers should not rely

PAGE 44

44 exclusively upon the color of the meat to determent its suitability to consume, but should also check the use by date, strong spoilage odor, slippery texture, or packaging that has begun to swell Effect of pH on Oxidation Most post mortem fish muscle and muscle systems have pH values below 7.0. The role of pH under thes e conditions is important in understanding lipid oxidation. Richards and Hultin ( 4 ) investigated the effect of pH (7.6, 7.2, and 6.0) on lipid oxidation using trout Hb as a catalyst in a washed minced cod muscle system and found that at lower pH values formation of m et Hb occurred more rapidly and the level of d eoxy Hb increased sharply. Lipid oxidation of the washed cod muscle induced by trout hemolysate occurred more rapidly at pH 6.0 compared to pH 7.2, using TBARS and sensory scores to measure oxidation. Lowering the pH below neutrality decreas es the oxygenation of Hb. This is known as the Bohr effect ( 43 ) The Bohr effect can explain the high levels of d eoxy Hb at postmortem due to the low pH levels. When pH levels are decreased below 6.5, further deoxygenation occurs and this is known as the Root effect. The authors proposed that these results suggest that d eoxy Hb may play the role of a catalyst in lipid oxidat ion Undeland and others ( 22) studied the effect of pH in a washed, minced cod model system using menhaden, mackerel, flounder, and Pollock Hb as catalysts. At pH 6 .0 TBARS and painty odor showed that at all four Hb s were equally active proxi dants However, at pH 7.2, the proxidative activity of Hb was slowed except that from p ollock. The authors found that the higher rate of oxidation at pH 6.0 corresponded to a greater f ormati on of deoxy and met Hb Richards and others ( 55 ) added Hbs obtained from beef, chicken, and trout to washed cod minced muscle to give a final concentration of 5.8 mol/kg at pH 6.3. These

PAGE 45

45 H b s were used because there was a high variation in the oxygenation of each. At postmortem pH values, compared to bov ine Hb, trout Hb is largely deoxygenated and chicken has an intermediate level of d eoxy Hb content. The authors speculated that the poorly oxygenated levels of trout Hb would be a better catalyst for lipid oxidation then e ither chicken or beef Hb. During sevenday storage at 2C TBARS developed mor e quickly using trout Hb in the cod mince as opposed t o the chicken or beef Hb. The same was al so true for lipid peroxidation development of washed cod muscle. Richards and others ( 80 ) a dded separately trout Hb and tilapia Hb to washed cod muscle at pH 6.3 and 7.4 to assess the onset of lipid oxidation of samples that were stored at 2C. Indicators of lipid oxidation included formation of TBARS, painty odor, and lipid peroxides. At day 2, trout Hb containing samples (pH 6.3) had higher TBARS values compared to those containing tilapia Hb (p < 0.001). At pH 7.4, trout Hb containing samples did not show an increase in TBARS until day 8 of storage while TBARS for tilapia Hb containing samples remained low during the 14 days of storage. The authors also added trout and tilapia Hb to washed tilapia at pH 6.3 and 7.4. They found no difference in TBARS values at pH 6.3, but at pH 7.4 TBARS were significantly depressed for both Hb samples. The trout and tilapia Hbs (pH 6.3) in the tilapia muscle system had higher TBARS compared to the washed cod. During storage at 2C, pH 6.3, trout Hb samples developed a painty odor by day 2, and by day 3 tilapia Hb samples. A t pH 7.4, no painty odor was detected in tilapia Hb samples at day 14 but trout Hb samples had moderate rancid odor at day 12. Lipid peroxide values obtained during storage at 2C (pH 6.3) showed that tilapia Hb samples in washed cod had lower peroxide val ues on day 2 than trout Hb samples. At pH 7.4, low lipid peroxide values

PAGE 46

46 were found for both trout and tilapia Hb samples on day 8 in the washed cod system. Using the same Hb samples in the tilapia washed system showed rapid formation of lipid peroxides at pH 6.3 similar to that observed for TBARS. Both types of Hb in the tilapia system at pH 7.4 showed significantly slower and less peroxide formation. At pH 6.3 and 7.4, each Hb formed lipid peroxide more extensively in washed tilapia compared to washed cod. In another study, using washed cod muscle and cod Hb (3mol Hb/kg of mince), decreasing the pH from 7.8 to 6.3 greatly decreased the lag phase and increased the rate of lipid oxidation ( 81 ) Phospholipids constituted approximately 80% of the total lipids. The lag phase at pH 7.8 was ~40 h and the oxidation rate was slower compared to pH 6.8 and 3.5 which had faster rates of oxidation and lag times of 6 and 3 h respectively. At pH values above 7.8, slower rates of oxidation were observed. The authors conclude that the rate of oxidation as a function of pH was determined by a decrease Hbcatalyzed activity at high pH and decreased susceptibility of the membranes to oxidation at low pH values Role of Sodium Chloride (NaCl) in Oxidation Sodium chloride ions are found in all living cells and can act as a prooxidant of muscle lipid peroxidation ( 82 ) Kanner and others ( 83 ) found that presence of NaCl initiates reactions producing superoxide anion radical (O2 -) which results in formation of hydroxyl radical. The prooxidative effect and the increase of lipid peroxidation by NaCl in model systems were studied and the elevati on of free iron in tissues was ascribed to NaCl ( 82) Wallace and others ( 84 ) found that the stability of oxy Hb and its oxidation to m et Hb may have been affected by the presence of NaCl by shifting ferrous ions from

PAGE 47

47 interaction with oxygen to reaction wi th hydroperoxides and decomposing these compounds to free radicals, accelerating the peroxidation process. Harel ( 85 ) investigated the effect of NaCl on autoxidation of ferrous and cuprous ions in the presence of ascorbic ac id and iron chelators. The generation of hydroxy radicals by ascorbic acid and metal ion's was inhibited by NaCl. NaCl also inhibited the oxidation of ascorbic acid by preventing the interaction of Fe or Cu with oxygen. The chloride anion interacts with the iron ion inhibiting the ferrous ion oxidation. Calcium chloride, magnesium chloride, and Lithium chloride showed similar results Wallace and others ( 84) demonstrated that NaCl accelerates the decomposition of oxy Hb to met Hb Harel ( 85 ) postul ating that NaCl prevents or disturbs the interaction between heme iron and oxygen in the same way. Thus, a review of the literature revels that lipid/protein oxidation are complex mechanisms in determining the fish muscle. Additionally, the role of heme has been more carefully clarified. Use of CO treatment has had a significant effec t on the stability of Hb thereby reducing oxidative processes. It has also been shown that the effect of pH on oxidation supersedes all other factors when controlling for deter ioration of fish muscle Finally, concentrations of NaCl also affect oxidation negatively. The interaction of these various factors contributes considerably to the ov erall stability of fish muscle.

PAGE 48

48 Figure 2 1. The dynamic convers ion between different H b forms ( adapted from Baron and Andersen (2002)).

PAGE 49

49 CHAPTER 3 CARBOXY, OXY AND MET HEMOGLOBIN: A COMPAR ATIVE STUDY OF HEMOGLOBIN MEDIATED LIPID/PROTEIN OXIDA TION IN WASHED MINCED TILAPIA MUSCLE AT TW O DIFFERENT STORAGE TEMPERATURES Introduction Quality deterioration of seafood products caused by lipid oxidation results in off odors, off flavors, and color defects. Two heme proteins play an important role in the quality of seafood products; hemoglobin (Hb), located in red blood cells and myoglobin (Mb) located in muscle cells ( 5 ) Both of these proteins bind oxygen and contain Fe(II) in the center of the heme group (four in Hb and one in Mb) ( 37 ) which can bind to different ligands. The bright red color of muscle tissue is due to iron in its reduced form (Mb Fe+2O2). When Fe+2 (ferrous) is converted to Fe+3, a utoxidation occurs, resulting in the formation of met Mb (Mb Fe+3H2O), producing an undesirable brown color ( 3 7 ) and ultimately, quality deterioration of the seafood product. Both d eoxy Hb (HbFe(II)) and oxy Hb (HbO2Fe(II)) exist in the reduced state and thus are susceptible to autoxidation. Reduc ed Hb s form superoxide anion radicals during autoxidation which c an form hydrogen peroxide (H2O2). Met Hb in the presence of H2O2, can be activated ( 39 ) and converted to the very pro oxidative forms, perferryl Hb (Hb Fe(IV)=O) and ferryl Hb (Hb Fe(IV)=O). Deoxy Hb content affects heme protei n autoxidation. Deoxy Hb in the presence of oxygen, is susceptible to rapid oxidation ( 1, 4 ) Howe ver, fully oxygenated Hb was found to react slower with H2O2 because the accessibility of the heme iron is inhibited by the O2 ligand in oxygenated molecules ( 43) Oxy Hb is also more compact compared to d eoxy Hb which has greater flexibility in the heme pocket ( 43, 44) It has been found that disrupting the hydrogen bonding network of His97 created easier accessibility of

PAGE 50

50 H2O into the heme crevice ( 5, 45 ) This would then accelerate the formation of m et Hb which would also increase lipid oxidation. Richards and Hultin ( 11 ) found that increasing the concentration of Hb in washed mince d cod increased lipid oxidation. Hemolysate containi ng 0.06 and 0.50 mol Hb/kg tissue (pH 6.3) did not develop rancidity during storage at 2C as measured by sensory scores and TBARS. However, rancidity developed at day 1 and day 2 for hem olysate containing 5.8 and 1.8 mol, respectively. One way to reduce the prooxidative activity of these heme proteins is to maintain the proteins in their reduced state. Kristinsson and others ( 57) investigated the effect of CO gas treatment on tilapia Hb and found that treatment with CO greatly stabilized the protein with respect to its prooxidative potential (in a model linoleic acid system). CO binds to heme with greater affinity and easily replaces oxygen from the heme pocket. The CO binding produces a cherry red color whi ch is stable during refrigeration and frozen storage temperatures. The objective of this study was to compare the proox idative activ ity of oxy, CO, and m et Hb in a washed model system It is hypo thesized that tilapia Hb treated with CO will be less pro oxidative than oxy or m et Hb during storage at 3.7C and 25C in a washed minced model system. Materials and Methods Preparation of W ashed, Min ced Tilapia Muscle (WMTM ) Fresh skinless tilapia fillets were obtained locally from Rainforest Aquaculture (Sunrise, FL). All dark muscle, fat, blood spots and excess connective tissue were removed. The remaining fillets were ground (Oster Heavy Duty Food Grinder, Sunbeam Corporation, Inc.) and washed twice with distilled deionized water at 1:3

PAGE 51

51 mince to water ratio (at <5C). The m ince was stirred for 2 min with a glass rod, and allowed to stand for 15 min on ice. The minced muscle was then dewat ered on a nylon screen. Finally, the washed muscle was mixed with a 50 mM sodium di phosphate buffer at the pH of interest (6.3, 6.8, and 7.3) at a ratio of 1:3 mince t o buffer and stirred for 2 min with a glass rod, then allowed to stand for 15 min on ice. The washed muscle was cent rifuged at 15000g for 20 min at 3.7C using an Eppendorf 5702 centrifuge (Eppendorf North America Inc., New York, NY). Streptomycin (200 ppm) was added to prevent microbial growth and samples were vacuum packed (~ 100 g) and stor ed at 80C until used. The final moisture content of the samples was ~83% measured using a moisture balance (CSI Scientific Company, Inc., Fairfax, VA). Collection of Fish Blood Live tilapia were obtained from a pond at the UF Department of Fisheries and Aquatic Sciences and transported in an aerated water tank to the Aquatic Food Pilot Plant, University of Florida. Upon arrival, the fish were placed on ice for 1 minute with the pectoral side facing up. The bleeding procedure was conducted according to Kristinsson and others ( 57 ) Blood was drawn through the caudal vein using syringes (25 gauge, 1 inch needles). Syringes were preloaded with heparin solution (150 units/mL) according to Fyhn and others ( 86) Preparation of Hemolysate The hemolysate was prepared according to a modified procedure of Fyhn and others ( 86) Heparinized blood was washed with four volumes (1:4) of ice cold 1.7% NaCl in 1mM Tris buffer, then centrifuged using a tabletop centrifuge (Eppendorf Centrifuge 5702, Brinkman Instruments. Inc., Westbury, NY ) at 1000g for 10 min at 3.7 C. Plasma was removed and discarded. Red cells were washed three times with 10

PAGE 52

52 volumes (1:10) of ice cold 1.7% NaCl in 1mM Tris buffer and centrifuged at 1000g for 10 min Supernatant was removed and discarded. The red cells were lysed in Tris HCl buffer at pH 8.0 for one hour. Onetenth volume of 1M NaCl was added to aid in stromal removal (pellets) before centrif ugation at 30000g for 15 min at 3.7 C. The hemolysate, obtained as the supernatant was stored in 1.5 mL cryogenic tubes in a freezer at 80 C until used. Quantification of Hemoglobin Levels in Hemolysate Hb levels were quantified spectrophotometrically according to the Bradford procedure( 87 ) using Coomassie Plus Protein Assay Reagent Kit ( Pierce Technology, Rockford, IL) An approximate Hb concentration level was calculated using a BSA standard curve and the average molecular weight of Hb (ca. 66,000). Oxy CO and Met hemoglobin Preparatio n Oxy Hb was obtained from the lysis of red cells (hemolysate) and purified according to Fyhn and others ( 86 ) as modified by Richards and Hultin ( 4 ) described in the preparation of the hemolysate. Oxy Hb was identified using UV Vis spectra of heme proteins according to the method of Kristinsso n and others ( 57) A diluted aliquot of the hemolysate was used to determine the presence of oxy Hb. A peak at 414 nm was observed indicating that the hem e pr otein was bound to oxygen. CO Hb was prepared according to Kristinsson and others ( 57 ) by placing a 5 mL aliquot of oxy Hb in 50 mL Falcon tubes on ice and pas sing a stream of 100% CO over the solution for 2 min. The tube was capped immediately until used. The CO Hb protein aliquot showed a peak at 419 nm suggesting that the muscle sample was bound to CO Met Hb was prepared according to DeYoung and others ( 88 ) by reacting oxy Hb with potassium ferricyanide (K3Fe(CN)6). K3Fe(CN)6 was added at three times the molar

PAGE 53

53 concentration of the Hb and the mixt ure allowed to stand for 30 min on ice. Unreacted ferricyanide was removed using Centricon centrifugal filter devices using a tabletop centrifuge (Eppendorf Centrifuge 5702, Brinkman I nstruments. Inc., Westbury, NY ) at 4000g for 30 min at 3.7C. Sodium phosphate buffer (20 mM) was used twice to remove potassium ferricyanide and centrifuged agai n for 30 min. Changes in Hb levels after preparation of m et Hb were monitored spectrophotometrically according to Bradford ( 87) The s ample peaked at 406 nm and thus was identified as fully oxidized m et Hb. Sample Preparation : Addition of Hb and NaCl Samples of tilapia washed model system previously prepared and stored at 80C were thawed rapidly under running water (20C) and kept on ice. The pH was adjusted to the desired pH of (6.3, 6.8, or 7.3) using 2N NaOH or 2N HCl. After the desired pH was established, moisture content was determined using a moisture balance (CSI Scientific Company, Inc., Fairfax, VA). Oxy Hb previously prepare d was thawed under running water and added to the muscle system to give a final concentration of ~6, 9, and 12 mol/kg washed muscle. Hb was mixed manually into the WMTM system using a plastic spatula. The homogenous color of the minced muscle indicated adequate mixing. Each of these Hb concentrations was added to the system at three pH levels (6.3, 6.8, and 7.3). NaCl was added to the s amples at a concentration of 150, and 450 mM with constant manual mixing. Samples were plated in Petri dishes (~25 g) c ov ered with a lid and stored at 3.7C for 8 days and at 25C for 24 weeks. All samples were stored in duplicate at both temperatures. The various combinat ions were conducted as presented in Table 31. CO Hb samples were prepared by gassing the WMTM with 100% CO for two hours on ice. The WMTM was placed in gastight vacuum bags equipped with a silicon

PAGE 54

54 septum valve obtained from LabPure Instruments. CO Hb was added to the muscle after removal from the bags to give a final concentration of ~6, 9, and 12mol/kg WMTM CO Hb was mixed manually using a plastic spatula. Each of these CO Hb concentrations was added to the system at three pH levels (6.3, 6.8, and 7.3). Samples were placed in Petri dishes (~25 g) covered with a lid and stored at 3.7C for 8 days and at 25C for 24 weeks. All samples were stored in duplicate at both temperatures (Table 31). Met Hb, previously prepared, was mixed manually with the washed system using a plastic spatula to final concentrations of ~6, 9, and 12 m ol/ k g washed muscle. Each of these m et Hb concentrations was added to the system at three pH levels (6.3, 6.8, and 7.3) Samples were plated in Petri dishes (~25 g) covered with a lid and stored at 3.7C for 8 days and at 25C for 24 weeks. All samples were stored in duplicate at both temperatures (Table 3 1). One g of sample was taken from each Petri dish every other day from samples at pH 6.3, 6.8 and 7.3. T he remainder of the samples were returned to 3.7C. For samples stored at 25C, 1g was taken every 4 weeks and analyzed while frozen. The remainder of the samples was returned to the freezer. Due to the difficulty of mixing, the frozen samples and the samples containing 450 mM NaCl, samples were taken from the core and side of the dishes. These sam ples were then stored in aluminum foil at 80C until analysis was performed. Determination of Protein Content Protein content was quantified spectrophotometrically using the Biuret method described by Robinson and Hodgen ( 89) A bovine serum albumin (BSA) standard curve was constructed to calculate protein concentr ation. One g of sample was

PAGE 55

55 homogenized with 10 mL of 0.1N NaOH for 1 min. 100L of the homogenate in a tube (culture tubes) was added to 900 L 0.1 N NaOH. To this, Biuret reagent was added (4 mL) and incubated at room temperature for 30 min. Absorbance was measured at 540 nm against a blank containing 1 mL of 0.1 N NaOH + 4 mL Biuret reagent. Protein concentration was calculated using the BSA standard curve. Determination of Total Lipids Total lipid content was determined using the method of Lee ( 90 ) Ten grams of tilapia muscle system was homogenized with 50 mL of 1:1 (v/v) chl oroform: methanol for 60 s using a hand held homogenizer (Biospec products Inc., Bartlesville, OK). The homogenate was filtered using Whatman filter paper number 4 into separ ating funnels. Twenty mL of 0.5% NaCl solution was added to increase phase separation and the solution gently shaken manually. The mixture was stored at 4C for 3 h. A known volume of the chloroform layer was separated into three aliquots of 4 mL each in previously weigh ed beakers. The total lipid content was obtained gravimetrically by evaporation of the solvent using a hot plate at low temperature (65 70C). Determination of Phospholipids Content Measurement of organic phosphorous in the extracted lipid phase (from previous total lipid determination step) was used to estimate total phospholipid content. Lipid phosphorous was converted to phospholipid based on the assumption that 31 g atoms of phosphorous is equivalent to 750 g atoms of phospholipid. According to a modified procedure of Anderson and others ( 91) a 100 L sample of the chloroform layer was obtained from the lipid analysis step and placed in test tubes. These were heated at 105C to evaporate the chloroform using the metal blocks. Samples were cooled to room temperature, 1 00 L of concentrated sulfuric acid

PAGE 56

56 (H2SO4) were added, and heated again in the metal blocks at 155C for 10 min. After cooling to room temperature, 200 L of 6% H2O2 (hydrogen peroxide) were added and samples were heated in the metal blocks at 155C for 4 0 min Two mL of distilled water were added and vortexed well. To this was added 0.8 mL of a mix of 10.1 mM ammonium molybdate and freshly prepared 0.28 M ascorbate solution and vortexed well. Samples were finally heat ed in a water bath for 7 min then cool ed to room temperature. Absorbance was recorded at 797 nm against sodium phosphate buffer. A standard of sodium phosphate (NaH2PO4) was used to determine the amount of phospholipids. Determination of Peroxide Value (PV) Lipid hydroperoxides (primary products of oxidation) were determined according to Shantha and Decker ( 92) with modifications by Undeland and others ( 8 ) S amples were prepared in duplicate. Total lipids were extracted using 10 mL of a chloroform/methanol (1:1) mixture with vortexing. Sodium chloride, 3 mL (0.5%), was added to the sample, and the sample was vortexed for 30 seconds, and centrifuged (Ependorf, model 5702, Brinkman Instrum ents, Inc., Westbury, N Y) at 4 C for 10 min at 2,000 g. Two mL of the bottom layer were obtained using a 2 mL glass syringe and needle (MicroMate Interchangeable glass syringe, stainless steel needle, 20G x18, Popper and Sons Inc., New Hyde Park, N Y). Two mL of the chloroform layer were mixed with 2 mL of chloroform/methanol (1:1). Fifty L of ammonium thiocyantate (3.94M) were added to the sam ple, followed by addition of 50 L iron chloride (prepared by adding equal amounts of 0.144 M FeSO4 and 0.132 M BaCl2). Following each reagent addition, the mixture was vortexed for 24 seconds. After a n incubation period of 5 min at room temperature, samples were read at 500 nm using a spectrophotometer (Agilent 8453

PAGE 57

57 UV Visible, Agilent Technologies, Inc., Palo Alto, CA). A blank contained all reagents except the sample. A standard curve was prepared using cumene peroxide. Determination of Thiobarbituric Acid Reactive Substances (TBARS) TBARS, which measures secondary oxidation products, were determined according to a modified procedure of Lemon ( 93) All samples were prepared in duplicate. One gram of sample was homogenized with 6 mL of trichloracetic acid (TCA) so lution composed of 7.5 %TCA, 0.1% propyl gallate, and 0.1% EDTA (ethlene diaminetetraacetic aci d 99%) 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) (0.23% dissolved in deionized water). The TCA and TBA solutions were freshly prepared and heated as needed. The TCA/TBA solution was heated for 40 min in boiling water at 100C. Samples were cooled immediately on ice and absorbance measured at 530 nm using a spectrophotometer (Agilent 8453 UV Visible, Agilent Technologies, Inc., Palo Alto, CA). Amount of oxidation was determined from a standard curve of 1, 1, 3, 3 tetraethoxypropane. Determination of Carbonyl Groups Protein carbonyl content was determined using 2, 4 Dinitrophenylhydrazine (DNPH) following a procedure by Levine and others ( 36) as modified by Kj and others ( 94) The 2, 4 Dinitrophenylhydrazine reacts with the carbonyl group of the protein, giving a hydrazone, which is quantified spectrophometrically. Carbonyl groups, which hav e formed a Schiff base with amino acids, also react with DNPH. This method is based on the reaction of hydrazyne and carbonyl groups forming a hydrazone. The hydrazone formed can be measured spectrophotometrically at 370 nm and total protein at 280 nm. One g sample was homognized with 12 mL 50 mM phosphate buffer of a pH

PAGE 58

58 corresponding to the sample pH using a hand held homogenizer. The final concentrat ion was 1 mg protein/mL sample. To three 1.5 mL cryogenic (amber) tubes, 1 mL protein solution was added (2 samples and 1 blank). Protein was precipitated with 100% trichloroacetic acid (TCA) to a final 10% 20% TCA (v/v). Samples were cent rifuged for 3 min at 12, 000 rpm and the supernatant was discarded. One mL of 2 M HCl was added to the protein blank and 1 mL DNPH (2, 4 dinitrophenylhydrazine in 2 M HCl) was added to the samples. Samples were kept for 1 hour at room temperature in the dar k and then precipitated with an additional 50 L TCA. Samples were centrifuged for 3 min at 12 000 rpm. T he supernatant was discarded and the pellet was washed with 1 mL (1:1) Ethanol/Ethyl acetate. Samples were cent rifuged for 3 min at 11,000 rpm the sup ernatant was discarded and this step was repeated twice. The pellets were redissolved in 1 mL of 6 M guanidine HCl. Samples were incubated at 4C overnight. Samples were left at room temperature for 1 hour and centrifuged for 6 min at 12,000 rpm to remove any insoluble material. Absorbance was recorded at 370 nm and 280 nm against guanidi neHCl as a blank. Heme Group Autoxidation Relative oxygenation of heme group within the Hb was monitored during oxidation. H b was extracted from the washed system (2 g) with 4 mL 50 mM phosphate buffer having the same pH (6.3, 6.8, and 7.3) as the samples. The final ratio of buffer to muscle in extracts was 2:1. Samples were homogenized for 10 s using a hand held homogenizer (Biospec products Inc., Bartlesville, OK) and c entrifuged at 4,000 rpm for 20 min using a table top centrifuge (Eppendorf Nort h America Inc., New York, NY). The supernatant was scanned in the visible range from 350 to 700 nm using a spectrophotometer (Agilent Technologies, Palo Alto, CA) according to t he procedure of

PAGE 59

59 Krzywicki ( 95 ) Changes in the heme peak environment indicates how oxygenated the heme group was. Autoxidation of the heme proteins was calculated using Krzywickis modified equation ( 96 ) Color Analysis Deterioration of color during development of oxidation processes in the washed muscle samples was measured using a digital Color Machine Vision System (CMVS) ( 97) The CMVS measures the average li ghtness (L*value), redness (a* value), and yellowness (b* value) for each sample. Pictures of samples were taken at the sam e intervals that samples were taken for chemical and sensory analysis. A 10 g sample was placed i n a closed chamber (impermeable to stray light) A digital camera (Nikon D200 Digital Camera, Nikon Corp., Japan) facing bottom of the chamber was used to capt ur e the pictures of the samples. Two fluorescent lights (top of the chamber), each to simulate illumination by noonday summer sun (D65 illumination ), were used to obtain uniformity of light. The Nikon D200 Settings used are shown in Table 2. A red referenc e tile was placed with each picture and used as a standard for redness (a* value) (Figure 5 27) Results Tilapia was chosen as the raw material for the oxidation system because of its fresh availability all year round. Washing the minced tilapia white musc le removes much of the endogenous proand anti oxidants, retaining the muscle proteins and cellular membranes. This model system therefore allows for controlled and detailed manipulation of pH values, sodium chloride (NaCl) concentrations, and amount of H b added. The composition of tilapia whole muscle was tested and contrasted with the washed muscle at three levels of pH (Table 33, Table 34). Whole muscle had

PAGE 60

60 significantly less H2O, and significantly (p<0.05) less protein (dry WT basis, Table 3 4). The % H2O significantly (p<0.05) increased as the pH increased in minced muscle (Table 33). If the pH drops quickly at post mortem, the net surface charge is reduced, causing the protein to denature. When denaturation of sarcoplasmic (contractile) protein occurs ( 98 ) this de creases the ability of contractile proteins to bind water. As shown is Table 3 3, the lower the pH, significantly (pH<0.05) less % H2O is present in minced muscle. This i s associated with poor water holding capacity and pale color ( 98) The lower the pH, the greater the toughness of texture due to denaturation of myosin at lower pH causing shrinkage. The % of total lipid s was significantly (pH<0.05) lower in the tilapia minced muscle at onl y pH 7.3 compared to tilapia mus cle. The % phospholipid was also significantly (pH<0.05) lower at higher pH levels compared to tilapia muscle. At higher pH levels, the % H20 is hig her, which ultimately results in less protein and phospholipids. Lipid Oxidation Analysis Lipid hydroperoxide values (Figure 3 1) revealed significant differences for the effect of storage at 3.7 C regardless of pH, Hb, and NaCl concentrations. At day 2 and 4 of storage, oxy Hb had significantly higher PV val ues than either CO or m et Hb. By day 8, CO Hb had significantly (p<0.05) lower PV values than oxy and m et Hb which did not differ significantly (p<0.05) from each other. Lipid hydro peroxide values for the effect of storage at 25C (Figure 32) revealed no s ignificant differences between oxy and m et Hb. CO Hb had however significantly (p<0.05) lower lipid peroxide values at week 20 and 24 compared to oxy and m et Hb consistent with the observations at 3.7C. The formation of TBA RS was used to assess lipid oxidation during st orage at 3.7C for 8 days and 25C for 24 weeks Regardless of pH, Hb and NaCl

PAGE 61

61 concentration, the effect of storage for 8 days at 3.7C showed e xten sive formation of TBARS by day 6 for met Hb, but not for o xy and CO Hb forms (Figure 3 3 ). By day 8 of storage, m et Hb continued to have significantly (p<0.05) higher TBARS than oxy and CO Hb. On day 8, although oxy Hb had slightly lower TBARS than C O Hb, these differences were not significant (pH<0.05) On the other hand, at 25C, oxy Hb had significantly (p<0.05) higher TBAR S by week 12 than CO and m et Hb which did not differ from each other (Figure 34 ). After 20 weeks of storage at 25C o xy Hb had significantly (p<0.05) higher TBARS than CO and m et Hb. Throughout the 24 weeks of storage, CO and m et Hb did not differ significantly (p<0.05) from each other but continued to have significantly (p<0.05) lower TBARS formation than oxy Hb. Prot ein Oxidation Analysis The effect of storage for 8 days at 3.7 C regardless of pH, Hb and N aCl concentrations showed that oxy Hb had significantly (p<0.05) higher forma tion of carbonyls than CO and m et Hb throughout the 8 days of storage (Figure 35 ). The lower carbonyl values for m et Hb on day 0 differed significantly (p<0.05) from CO and o xy Hb which did not differ from each other significantly (p<0.05) on day 0. Carbonyl values (Figure 36 ) for the effe ct of storage for 24 weeks at 25C were similar to, although higher in value, to that of those found at 3.7C. oxy Hb had significantly (p<0.05) highe r carbonyl values than CO and m et Hb through week 16 of storage at 25C. At week 24, oxy Hb had significantly (p<0.05) lower hydroperoxide values than CO and m et Hb. CO Hb had a slight but significantly (p<0.05) higher carbonyl values than m et Hb at week 24.

PAGE 62

62 Color Analysis Color analysis of t he effect of storage for 8 days at 3.7C regardless of pH, Hb and NaCl concentrations reveal ed that the three for ms of Hb (oxy CO and m et Hb) differed significantly (p<0.05) from each other. During the 8 day storage period, a*values (redness) declined significantly (p<0.05) ( Table 3 5) for all three forms. Met Hb had a slight increase in a* value on day 2 which di ffered significantly (p<0.05) from the rest o f the storage days for m et Hb. CO Hb had significantly (p<0.05) higher a*values during the first four days of storage compared to oxy and met Hb. By day 6, oxy Hb had significantly (p<0.05) higher a*value than CO and m et Hb. At 25C storage (Table 3 6 ) the a* values differed significantly (p<0.05) for the three forms of Hb. CO Hb maintained higher values than oxy and m et Hb with a significa nt increase in a* value at week 4. Met Hb also had a significantly (p<0.05) higher a* value at week 4 of storage but the remain ing 24 week storage period for m et Hb showed no significant change in a* value. The a* value s for o xy Hb began to decline at week 4 and by week 12 no further significant decline in a*value was noted. L* values (lightness) (Table 3 5 ) differed significantly (p<0.05) for the three forms of Hb at 3.7 C, with m et Hb having significantly (p<0.05) h igher L* values than CO and o xy Hb throughout the 8 days of storage with its highest L*value evident at day 8. The L* values for CO Hb did not differ significantly (p<0.05) from day to day during the 8 day storage period. Oxy Hb had slightly but significant higher L* values beginning on day 4. At 25C storage, L *values for m et Hb differed significantly (p<0.05) from o xy and CO Hb (Table 36) However, there was no significant difference in L*values for m et Hb beginning with week 4 and throughout the remaining 24 week storage period. CO and m et Hb had significan tly (p<0.05) lower L* values after 24 weeks of storage whereas

PAGE 63

63 oxy Hb had higher L* values at week 24 but significantly (p<0.05) stable L* values from week 4 to week 16. b* values (yellowness) (Table 3 5 ) reveal ed that the three forms of Hb (oxy CO and m et Hb) differed significantly (p<0.05) from each other regardless of pH, Hb and NaCl concentrations. All three fo rms of Hb increased in yellowness over time, with day 8 having significantly (p<0.05) higher b* values compared to day 0. b*values for ox yHb were significantly (p<0.05) higher by day 6 and remained similar throughout the rest of the 8 day storage period. A t 25C b*values differed significantly (p<0.05) during storage for the three forms of Hb. All three forms had significantly (p<0.05) hig her b*va lues at week 24 compared to week 0 CO Hb had significantly (p<0.05) lower b*values than oxy and m et Hb throughout the 24 weeks of storage. Heme Group A utoxidation The absorption spectra (Figure 311) iden tified the absorption peak for met Hb at 503 nm, oxy Hb at 582 nm, and CO Hb at 542 nm. The isobestic point was identified at 582 nm. These results are similar to that of Tang ( 96 ) Mb coefficients are quite similar to Hb Some of t he traditional methods ( 95) to estimate m et Hb have not included CO Hb. They are based on the maxima wavelengths for oxy deoxy and m et Hb which does not account for the extinction coefficients and wavelength maxima CO Hb and calculations for the three forms rarely came close to 100%. Tang and others ( 96) modification yielded more consistent and logical results. The % o xy Hb formed (Figure 39 ) during storage at 3.7C for 8 days differed si gnificantly (p<0.05) for the three forms of Hb regardless of pH, Hb and NaCl concentrations. There was a significant decrease in the percentage of oxy Hb during storage for all three forms of Hb. CO Hb had a significantly (p<0.05) higher % o xy Hb

PAGE 64

64 thro ughout the storage period than oxy and m et Hb. The Tangs formula ( 96 ) dose not take % CO Hb into consideration; however, when calculating % o xy Hb in CO Hb samples, % o xy Hb is used to report relative % CO Hb in the sample. At 25C % o xy Hb formed was significantly (p<0.05) different ( Figure 310) for all three forms of Hb with decreasing % o xy Hb over the 24 week storage period. CO Hb had significantly (p<0.05) higher % o xy Hb than met and oxy Hb. Met Hb had the lowest % o xy Hb and differed significantly (p<0.05) from o xy Hb on week 16, 20, and 24. As expected, m et Hb had significantly (p<0.05) higher % m et Hb during storage at 3.7C for 8 days than oxy and CO Hb. However, during the 8 day storage period both o xy and CO Hb had significant increases in the percentage of m et Hb At 25C (Figure 3 10) the three forms of Hb differed significantly (p<0.05) in % m et Hb, with increasing amounts of % m et Hb with increased storage time As would be expected, the m et Hb sample had significantly (p<0.05) higher percent of its Hb in m et form compared to CO and oxy Hb throughout the 24 week storage period at 25 C Percent d eoxy Hb formed during storage at 3.7C for 8 days differed si gnificantly (p<0.05) for the three forms of Hb. All three forms showed a significant increase in % d eoxy Hb formed through day 4. CO and oxy Hb did not differ in % d eoxy Hb formed on day 6 and 8. At 25C, the three forms of Hb differed significantly (p<0.05) over the storage period (Figure 3 10 ). The % d eoxy Hb did not differ significantly (p<0.05) for m et Hb up to week 12, but significant differences were found on week 12 and throughout the remaining 24 weeks of storage. The % d eoxy Hb increased significantly (p<0.05) over the 24 weeks for the three forms of Hb with CO Hb having the greatest

PAGE 65

65 significant increase. By week 16 and for t he remaining weeks of storage, oxy and CO Hb did not differ in the % d eoxy Hb formed. Discussion In the pr esent study pro oxidative activity of three Hb (oxy CO and m et Hb ) in a WMTM system were examined at two storage temperatures 3.7C, and 25C. Oxidative activities of every sample were determined by quantifying products of lipid oxidation ( hydroperox ide value s and TBARS values ), as well as protein oxidation (carbony l groups). Color changes and oxidation state of the three forms of Hb were also monitored and recorded throughout the experiment to establish possible correlation with the oxidative process. H eme pr oteins are considered to be potent catalysts of lipid oxidation in muscle food systems ( 1, 22) From the hydro peroxide value res ults it is evident that o xy Hb oxidizes lipids significantly m ore rapidly than either CO or m et Hb, irrespective of environmental factors, reaching the peak and plateau by day 4 of refrigerated storage. In contrast, lipid oxidati on by CO and m et Hb follows somewhat similar patterns, oxidizing lipids at a slower rate, and reaching the peak and plateau by day 6. Interestingly, by day 8 both oxy and m et Hb exhibited comparable oxidative values, with CO Hb being the slowest catalyst of lipid oxidation (Figure 3 1 ). It is well established that Hb can exist in different redox states. It was proposed that maintaining Hb in the reduced state will delay/decrease autoxidation and lipid oxidation, thus preserving freshness and improving the shelf life of fish products ( 57 ) Oxy Hb is believed to be less prooxidative, possibly due to its reduced state. Despite that, previous research studies suggest that oxy Hb might in fact be a strong promoter of lipid oxidation under certain conditions ( 57) It may be explained, in part, by increased rate of

PAGE 66

66 deoxygenation and deoxy Hb format ion in the post mortem muscle system. It has been demonstrated in numerous previous studies that increased deoxygenation can enhance prooxidative activity of Hb, promote Hb autoxidation and hemin loss, which in turn le ads to a formation of m et Hb and incr eased lipid oxidation ( 22, 56) Formation of m et Hb occurs by the conversion of ferrous iron (Fe2+) of the Hb heme group to the ferric (Fe3+) form which is less stable, readily reactive, and susceptible to oxidation. Furthermore, in the presence of preformed lipid hydroperoxides m et Hb can oxidize to two extremely reactive species or perferryl Hb (Fe5+) and ferr yl Hb (Fe4+) ( 99, 100 ) The presence of CO bound to Hb may considerably delay the deoxygenation process due to its higher affinity to heme proteins, increasing its stability which, in turn, may retard lipid oxidation. CO is extensively used in the food industry to stabilize the red color and preserve freshness of seafood due to its ability to maintain heme proteins in their reduced state for relatively long periods of time. It is thought that CO exerts its action by displacing O2 from the heme, apparently making it more compact and possibly changing conformation of the molecule, hence stabili zing the structure of Hb making it less susceptible to autoxidation ( 57) To determine the rate and pattern of autoxidation of different Hb we examined relative deoxygenation, expressed as the formation of d eoxy and m et Hb in all samples using Tangs equation ( 96) O xygen molecules are replaced by CO in CO Hb therefore, the above method is considered useful in determining the relative rates of heme protein autoxidation. It appears that at storage temperatures (3.7C) the process of deoxygenation, formation of deoxy and m et Hb or autoxidation was significantly retarded in CO Hb and oxy Hb treated samples throughout the storage period as compared to m et Hb. A lthough both CO and

PAGE 67

67 o xy Hb exhibited similar tr ends and rates of formation of deoxy a nd m et Hb CO Hb was significantly (p<0.05) les s oxidized until day 6 (Figure 37 ). During frozen storage ( 25C) slower lipid oxidation with lesser hydroperoxide formation was obs erved f or all three types of Hb. However, CO Hb exhibited greater ox idative stability than oxy and m et Hb (Figure 34 ). This correlated with a slower rate of deoxygenation at 25C. This may be attributed to the ability of low temperatures to preserve muscle systems from lipid oxidation which is in agreement with previous studies ( 31 ) The results of the different Hb forms were different when TBARS values were examined. While o xy Hb demonstrated higher oxidation activity at the beginning, it reached peak TBARS formation sooner than CO and m et Hb with m et Hb reaching the significantly (p<0.05) highest score by day 8 at 3.7C (Figure 3 3 ). On the other hand, fro zen storage conditions retard TB ARS formation in m et and CO Hb systems, but not in the oxy Hb system (Table 35 ). Met Hb requires preformed hydroperoxides for oxidative activity. Under low storage temperatures the formation of hydroperoxides is retarded, which may, to a degree, suppress the oxidative activity of m et Hb. It is possible that continuous deoxygenation and autoxidation of oxy Hb even under frozen storage conditions c ould provide enough substrate for continuous oxidation in this Hb system. Recent studies have associated the quality deterioration of muscle food systems during processing and storage with protein oxidation ( 31, 94 ) However, very few studies investigating protein oxidation in muscle food systems have been conducted to date. In this study comparing different forms of Hb, protein oxidation in WMTM was

PAGE 68

68 monitored by formation of carbonyls Prev ious observations demonstrated great variability in basal levels of protein carbonyls in different tissues ( 33) Furthermore, certain processing manipulations and storage conditions may alter protein structures making them more susceptible to oxidation, and thus affect their functionality and changes in quality ( 94 ) Additionally, prior experiments pointed out the possibility that carbonyl quantification might be affected by hemecon taining compounds, including Hb, since these compounds exhibit high absorbance at the same length as DNPH ( 101 ) All of this may explain the fact that increased levels of carbonylation was detected on day 0 in all tested samples in the present study. Moreover, initial carbonyl content was similar for all Hb systems. Striking similarity was observed in carbonyls formation in o xy Hb system at both storage temper atures, reaching the peak mid way throughout storage, and then subsiding (Fig ure 3 5, 3 6 ). On the contrary, additional formation of the carbony ls was not detected in CO and m et Hb systems at 3.7C, while at 25C steady increase in carbonyls development was observed in CO and m et Hb systems. This finding is consistent with previous studies ( 31 ) Frozen storage of rainbow trout fillets at lower freezing temperatures ( 20C) was found to increase protein carbonylation ( 94 ) Mincing the fish tested in this study could potentially accelerate carbonyl formation duri ng frozen storage due to possibly more protein denaturation and more exposure of proteins by catalysts and oxygen. Nevertheless, it is unclear why protein carbonylation in the oxy Hb system was increased at first and subsequently subsided to the initial level. It may be speculated that increased oxygenation of the oxy Hb could initially accelerate protein oxidation, with subsequent reduction in protein oxidation following o xy Hb deoxygenation.

PAGE 69

69 To determine deterioration of the color in all tested Hb systems during storage, a*values, L*values, and b*values were obtained and monit ored throughout the study. A t 3.7C CO Hb treated samples maintained higher a*value until day 4 of the study as compared to oxy and m et Hb, with subsequent deteriorat ion in color by day 8 in all Hb. Mantilla ( 102 ) found that treating the red muscle of tilapia fillets post mortem with CO increased the red color which remained significantly higher until day 9 compared to untreated fillets and fillets obtained from fish euthanized with CO treatments. Huo ( 103) found that a* values declined after 9 days of storage for untreated samples compared to samples of tuna treated with CO. Based on peak wavelengths, Huo found that there was some CO still bound to Hb. Danyali ( 71) and Garner and Kristinsson ( 73 ) found that the decrease in red color was associated with a decrease of CO binding to heme proteins. In contrast, at 25C red color deterioration expressed as reduction in a*value was alm ost undetected in CO Hb, while oxy and m et Hb sustained significant decrease in a*value. Interestingly, red color appeared to be enhanced for the first 8 weeks, and by week 24 a*value of CO Hb syst em was comparable with that of oxy Hb at day 0. Mantilla ( 102 ) found that a* values decreased significantly after freezing. However, he fou nd no significant difference between a*values of fresh fish and that of tuna after 2 months of storage, suggesting that freezing preserved the red color. After 4 months of storage, a* values of CO euthanized fillets were not significantly different from initial, fresh controls, d esirable red color of meat muscle systems was linked with higher levels of heme proteins. Ability of CO to maintain Hb in the reduced state longer, preventing heme loss, and thus stabilizing red color was previously reported and in agreement with

PAGE 70

70 current findings ( 69, 73) The increase in a*value of CO Hb after 8 weeks of frozen storage found in this study is consistent with Mantilla ( 102 ) findings. Danyali ( 71 ) also found an increase in a* values of yellowfin tuna steaks that were treated with CO and stored at 25C for 30 days. Huo ( 103 ) noted that tuna steaks treated with CO at 20C had higher a* values and CO levels than those treated at 4C indicating more CO binding to heme protein ( 79 )

PAGE 71

71 Conclusion The results of this chapter suggest that tilapia minced muscle treated with CO could have greater stabilities to oxidation, thus resulting in increased color stability and oxid ative stability. Oxy Hb maintained its catalytic effect and is believed to catalyze oxidation by the breakdown of preformed lipid hydroperoxides. The greater the concentration of preformed hydroperoxides, the more the prooxid ativ e activity of oxy Hb met Hb may also form ferryl Hb which may initiate lipid peroxidation. However, low prooxidative activity of the oxidized form (met Hb) could be a result of the absence of oxygen to form hydrogen peroxide required for met Hb oxidation. Washing fish muscle could reduce the amount of hydroperoxides which are important for the prooxidative activity of metHb. The low prooxidative activity of CO Hb is due in part to CO increasing the stability of heme protein structure and slowing down autoxidation. The reduced reactivity of CO Hb with hydrogen peroxides may be due to the str ong affinity of CO to Hb. M ore studies on how CO affects fish Hb will assist in the advancement of CO bas ed methods to stabilize seafood products.

PAGE 72

72 Table 3 1. Sample preparation of the three forms of hemoglobin (Oxy CO and Met Hb) at three different pH values Hb Concentrations, NaCl added, and two storage temperatures*. Temperature pH Hb Concentration (mol/kg) NaCl added (mM/Kg) 3.7C 6.3 6 0 150 450 9 0 150 450 12 0 150 450 6.8 6 0 150 450 9 0 150 450 12 0 150 450 7.3 6 0 150 450 9 0 150 450 12 0 150 450 *Storage temperatures were 3.7 and 25C

PAGE 73

73 Table 3 2. Nikon D200 camera s ettings used for measurement of change in color during storage period. Setting Specification Lens Focal length Sensitivity Optimize image High ISO NR Exposure mode Metering mode Shutter speed and aperture Exposure compensati on (in camera) Focus mode Long exposure NR Exposure compensation (by capture NX) Sharpening Tone compensation Color mode Saturation Hue adjustment White balance Zoom VR 18 200 mm F 3.5 5.6 G 36 mm ISO 100 Custom Off Manual Multi pattern 1/3s F/11 1.0 EV Manual Off 0 EV Normal Normal Mode I Normal 0 Direct sunlight Manual

PAGE 74

74 Table 3 3. Composition of tilapia muscle and washed tilapia muscle model system. Sample %H 2 O %Protein %TL a %PL b %PL/TL Tilapia muscle 78.730.3d 13.83.8a 0.210.04a 0.20.02a 93.50.154a WMTM pH6.3 83.10 .1 c 13.49 .3 ab 0.19 0.001 a 0.18 .001 a 93.3 0.070 a pH6.8 84.270.1 b 12.86 .2 bc 0.18 0.002 a 0.110.04 b 61.7 0.251 c pH7.3 85.80 .2 a 12.14 .6 c 0.16 0.29 b 0.110.04 c 68.7 0.043 b a TL, total lipids. b PL, phospholipid. Values are Means standard deviations (n=3). Different letters within the same columns indicate significant differences at P < 0.05 separated by Tukeys HSD.

PAGE 75

75 Table 3 4. Composition of tilapia muscle and washed til apia muscle model system on dry weight basis. Sample %Protein %TLa %PLb Tilapia muscle 65.020.023d 1.000.053a 0.930.200b WMTM pH6.3 79.790.043 c 1.160.062 a 1.080.052 a pH6.8 81.740.004 b 1.170.112 a 0.720.021 c pH7.3 85.480.006 a 1.120.074 a 0.770.004 b a TL, total lipids. b PL, phospholipid. Values are Means standard deviations (n=3). Different letters within the same columns indicate significant differences at P < 0.05 separated by Tukeys HSD.

PAGE 76

76 Table 3 5. Changes in a* value, L* value and b* value in washed tilapia muscle containing di fferent forms of Hb at 3.7 C, averaged across pH, Hb concentration and added NaCl a* value Storage time (day) Oxy Hb b CO Hb a Met Hb c 0 19.94.7 b1 21.94.1 a1 8.11.7 c2 2 16.67.9 b2 20.85.0 a1 10.22.2 c1 4 15.29.3 b2 17.27.6 a2 8.12.3 c2 6 13.510.3 a3 11.38.2 b3 6.12.7 c3 8 12.19.8 a3 9.87.3 b3 4.82.7 c3 L* value Storage time (day) Oxy Hb c CO Hb b Met Hb a 0 71.23.4 c2 73.33.0 b1 74.93.1 a2 2 71.43.3 c2 72.93.3 b1 73.93.3 a3 4 72.33.5 b1 72.93.3 b1 73.53.1 a3 6 72.64.0 b1 73.23.5 b1 73.93.1 a3 8 72.64.3 b1 72.84.0 b1 75.13.3 a1 b* value Storage time (day) Oxy Hb b CO Hb c Met Hb a 0 3.32.2 ab3 2.91.8 b4 3.81.6 a5 2 4.42.3 b2 4.41.9 b3 5.31.5 a4 4 6.22.4 a1 5.02.4 b3 6.12.0 a3 6 6.82.7 a1 6.12.8 b2 7.22.5 a2 8 6.23.4 c1 7.03.1 b1 8.92.4 a1 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n=54). Letters within the same rows indicate statistically (<0.05) significant differences separated by Tukeys HSD. Numbers within the same columns indicate statistically (<0.05) significant differences separated by Tukeys HSD.

PAGE 77

77 Table 3 6 Changes in a* va lue, L* value and b* value in washed tilapia muscle contain ing different forms of Hb at 25 C averaged across pH, Hb concentration and added NaCl a* value Storage time (week) Oxy Hb b CO Hb a Met Hb c 0 19.94.7 b1 21.94.1 a2 8.11.7 c3 4 17.26.4 b2 24.94.9 a1 12.32.8 c1 8 15.86.3 b23 23.05.9 a2 11.12.4 c12 12 13.57.3 b34 21.45.7 a2 10.32.4 c2 16 14.57.6 b4 20.66.0 a2 10.22.7 c2 20 15.28.9 b4 19.26.2 a3 9.82.8 c2 24 13.49.5 b34 18.46.7 a3 9.53.2 c2 L* value Storage time (week) Oxy Hb b CO Hb b Met Hb a 0 71.23.4 c2 73.33.0 b1 74.93.2 a1 4 69.43.3 b3 69.73.5 b234 71.93.2 a2 8 69.83.4 b3 70.03.6 b234 71.33.2 a2 12 69.03.4 b3 70.73.1 a2 71.53.4 a2 16 69.73.6 b3 70.13.4 b234 72.13.3 a2 20 71.53.8 a2 70.03.5 b234 72.13.5 a2 24 72.84.4 a1 69.53.6 b4 71.93.7 a2 b* value Storage time (week) Oxy Hb a CO Hb c Met Hb b 0 3.32.2 ab7 2.91.8 b7 3.81.6 a7 4 11.12.7 a3 6.72.2 c2 7.82.1 b6 8 12.12.7 a2 6.72.2 c1 8.62.5 b54 12 12.94.3 a125 5.82.2 c3 8.83.0 b4523 16 9.94.6 a4256 4.62.4 b4 9.43.0 a24 20 8.43.0 b5246 3.72.4 c6 9.33.0 a34 24 8.43.3 b645 4.21.9 c456 9.93.3 a1 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n=54). Letters within the same rows indicate statistically (<0.05) significant differences separated by Tukeys HSD. Numbers within the same columns indicate statis tically (<0.05) significant differences separated by Tukeys HSD.

PAGE 78

78 Figure 31 Lipid hydroperoxide values in washed tilapia muscle contain ing different forms of Hb at 3.7C averaged across pH, Hb concentration and added NaCl Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 32 Lipid hydroperoxide values in washed tilapia muscle contain ing different forms of Hb at 25C averaged across pH, Hb concentration and added NaCl Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. -100 0 100 200 300 400 500 0 2 4 6 8Lipid hydroperoxide mol /kg muscleDay (c) CO Hb (b) met Hb (a) Oxy Hb -100 0 100 200 300 400 500 0 4 8 12 16 20 24Lipid hydroperoxide mol/kg muscleWeek (b) CO Hb (a) Met Hb (a) Oxy Hb

PAGE 79

79 Figure 33 TBARS values in washed tilapia muscle contain ing different forms of Hb at 3.7 C averaged across pH, Hb concentrat ion and added NaCl Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 34 TBARS values in washed tilapia muscle contain ing different forms of Hb at 25C averaged across pH Hb concentration and added NaCl Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. -5 0 5 10 15 20 25 30 0 2 4 6 8mol TBARS/kg muscleDay (b) COHb (a) met Hb (b) Oxy Hb -5 0 5 10 15 20 25 30 0 4 8 12 16 20 24mol TBARS/kg muscleWeeks (b) CO Hb (b) Met Hb (a) Oxy Hb

PAGE 80

80 Figure 35 Carbonyl values in washed tilapia muscle containing different forms of Hb at 3.7 C averaged across pH, Hb concentration and added NaCl Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 36 Carbonyl values in washed tilapia muscle containing different form s of Hb at 25C averaged across pH, Hb concentration and added NaCl Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 0 2 4 6 8nM carbonyls/mg Day (b) CO Hb (c) met Hb (a) Oxy Hb 0 10 20 30 40 50 60 70 0 4 8 12 16 20 24nM carbonyls/mg muscleWeek (b) CO Hb (b) Met Hb (a) Oxy Hb

PAGE 81

81 Figure 37 % of a) Oxy b) Met and c) Deoxy Hb in was hed tilapia muscle containi ng different forms of Hb at 3.7 C averaged across pH, Hb concentration and added NaCl Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Oxy -HbDay (c) Met Hb (a) CO Hb (b) Oxy Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8% Met -HbDay (a) Met Hb (c) CO Hb (b) Oxy Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Deoxy -HbDay (a) Met Hb (c) CO Hb (b) Oxy Hb (c)

PAGE 82

82 Figure 38 % of a) Oxy b) Met and c) Deoxy Hb in washed tilapia muscle containing different forms of Hb at 25C, averaged across pH, Hb concentration and added NaCl Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Oxy -HbWeek (c) Met Hb (a) CO Hb (b) Oxy Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Met -HbWeek (a) Met Hb (c) CO Hb (b) Oxy Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24% Deoxy -HbWeek (a) Met Hb (c) CO Hb (b) Oxy Hb (c)

PAGE 83

83 Figure 39 Ab sorption spectra of met Hb o xy Hb, and CO Hb solutions containing equivalent hemoglobin concentrations. The arrows indicate the isobestic point at 525 nm, Met Hb absorption peak at 503 nm, Oxy Hb absorption peak at 582 nm, and CO Hb absorption peak at 542 nm. -0.05 0 0.05 0.1 0.15 0.2 0.25 480 500 520 540 560 580 600 620 640ABS (nm)Wavelength Oxy Hb Met -Hb CO Hb

PAGE 84

84 CHAPTER 4 EFFECT OF LOW AND HI GH CONCENTRATIONS OF HE MOGLOBIN ON THE PRO OXIDATIVE ACTIVITY O F OXY, CO AND MET HEMOGLOBIN IN A WASH ED MINCED TILAPIA MUSCL E SYSTEM AT TWO DIFFERENT STO RAGE TEMPERATURES Introduction Fish muscle contains several components that can serve as possible catalysts for lipi d oxidation including Hb, Mb, low molecular weight transition metal complexes and lipoxygenase ( 81) resulting in undesirable sensory characteristics. Of interest to this experi ment is the amount of Hb present in fish muscle and the concentration levels that are conducive to lipid oxidation. Richards and Hultin ( 11) found that Hb concentrations were higher in trout muscle that had not been tail or gill bled versus those that had been bled. To further investigate the contributions of Hb to lipid oxidation, the authors used a washed cod model system to which varying concentrations of hemolysate (0.06, 0.50, 1.8, and 5.8 mol/kg washed cod) was added in the presence or absence of pl asma. Plasma delayed development of TBARS at all four concentrations of hemolysate. It also delayed rancidity development at the two higher concentrations of hemolysate. No rancidity occurred during storage (2C /6 days) in the two lower levels of hemolysate. TBARS increased with increasing concentrations of hemolysate (R2= 0.99). An unexpected finding was the h igher Hb concentration in trout whole muscle. Yet, mackerel light muscle was more disposed to lipid oxidation. The authors considered t he different types of Hb i.e., anodic or cathodic to explain the differences between the behavior of trout whole muscle and mackerel light muscle. Anodic Hbs have low oxygen affinity at low pH post mortem while cathodic Hbs retain high oxygen affinity. Mackerel Hbs are known to have lower oxygen affinity ( 104 )

PAGE 85

85 Low oxygen affinity results in greater formati on of deoxy Hb which is a potent catalyst for lipid oxidation. The purpose of this study is to investigate the effect of varying concentr ations of tilapia oxy, CO, and m et Hb on oxidation in a WMTM. Materials and Methods The following methods have been described in detail in chapter 3: Preparation of Washed Minced Tilapia Muscle (MTWM) Collection of Fish Blood, Preparation of Hemolysa te, Quantification of Hb Levels in Hemolysate, Oxy CO and Met Hb Preparation, Sample Preparation: Addition of Hb and NaCl, Determination of Peroxide Value (PV) Determination of Thiobarbituric Acid Reactive Substances (TBARS) Determination of Carbonyl Groups Heme Group Autox idation, and Color Analysis. CO Calibration Curve by the Gas Chromatography Method A CO calibration curve was established by injecting 100% CO (CP Grade obtained from Airgas) in various concentrations (0.21.0 L). The areas under the resultant peaks were plotted against the injected quantity of gas (L) to obtain a linear regression equation. The Ideal Gas Law was used to determine the actual CO mass. PV=nRT where P (atm) is the barometric pressure, V (L) is the volume of CO gas, n (mol) is the number of moles of CO, R (0.08206 LatmKmol) is a universal gas constant, and T is the temperature in Kelvin (298K). Sample Analysis by the Gas Chromatography (GC) Method CO loss from the system during oxidation was determined using gas chromatography. Thr ee g of sample (CO Hb mixed with washed muscle) were transferred into a headspace 20 mL vials (Fisher Scientific, Fair Lawn, NJ), and 2 drops of octanol (antifoaming) were added. To this mix, 6 mL of 10% sulfuric acid (H2SO4) was added and the vials were s ealed immediately with a Teflonfluorocarbonresin/silicone

PAGE 86

86 septum lid. The mix was hand shaken for 1 min and incubated for 20 min at 40C. Vials were shaken using a table top shaker (American Optical CO. Scientific Instrument Division, US.) for 15 min at room temperature. Then 500 L of the headspace was taken for GC injection. Gas Chromatography Conditions for CO Analysis An Agilent 6890N GC system (Agilent Technologies, Palo Alto, CA) was used equipped with a flame ionization detector (FID) (Agilent Technologies, Palo Alto, CA), a Porapak Q column, 80/100 mesh, 6 FT x 1/8 IN (Supelco, Bellefonte, PA), and a hydrogen aided Nickel Catalyst (Agilent Technologies, Palo Alto, CA) (to convert CO and CO2 into 42 methane) was used for analysis. The samples were manually injected using a gastight syringe (Hamilton CO., Reno, Nevada). The oven temperature was set at 35C for 2 min. Nitrogen was used as carrier gas and held at a constant flow of 30 ml/min. T he injector temperature, nickel catalyst temperature and FID temperature were held at 100C, 375C and 200C respectively. Results Lipid Oxidation Analysis Washed tilapia muscle containing Oxy Hb at 6 mol/kg muscle and stored at 3.7C developed significantly (p 0.05) higher lipid hydroper oxide values than samples with met and CO Hb, which did not differ significantly from each other (Figure 4 1). Oxy Hb had significantly (p 0.05) higher hydroperoxide values than CO Hb on day 2, 4, and 8, and si gnificantly higher values than m e t Hb on day 2 and 4. CO Hb and m et Hb did not differ significantly on any day during storage. The hydroperoxide value for CO Hb was lower than m et Hb on day 4, 6, and 8 but this did not reach statistical significance (p 0.05) For lipid hydroperoxide values obtained at 25C and 6 mol/kg of

PAGE 87

87 CO oxy and m et Hb (Figure 4 2) Oxy Hb differed signific antly from met Hb but not CO Hb. CO H b and m et Hb did not differ significantly from each other. Washed tilapia muscle containing 9mol/kg muscle of Hb and stored at 3.7C (Figure 43) CO Hb had significantly l ower hydroperoxide values than met and oxy Hb. Oxy Hb had significantly higher hydr operoxide values than CO and m et Hb on day 2 and 4 but did not differ significantly from m et Hb on day 6 and 8. CO Hb had significantly (p 0.05) l ower hydroperoxide values than oxy and m et Hb on day 6 and 8. CO Hb developed the least amount of lipid hydr operoxide at 25C (Figure 4 4). At 12mol/kg muscle, oxy Hb developed significantly (p 0.05) higher hydr operoxides values than CO and m et Hb at 3.7C. CO Hb had significantly (p< 0.05) l ower hydroperoxide values than met and oxy Hb on day 6 and 8. Lipid hydroperoxide values obtained at 25C and 12 mol/kg of CO oxy and met Hb (Figure 46) showed that met and oxy Hb had significantly (p 0.05) higher hydroperoxide values than CO Hb. Washed tilapia muscle containing 6 mol/kg muscle of either oxy C O and m et Hb and stored at 3.7C (Figure 47) did not develop significantly different TBARS values (pH< 0.05). Examination of each storage day also did not show significant differences. At the end of storage on day 8 oxy Hb appeared to be less pro oxidative but these findings were not significant. TBARS values obtained at 25C and 6mol/kg of CO oxy and m et Hb (Figure 48) revealed that oxy Hb w as most prooxidative. CO and m et Hb did not differ from each other. However, oxy Hb was more pr o oxidative than CO Hb at week 8, 12, and 16. Oxy Hb was m ore prooxidative than CO and m et Hb beginning at week 12 and continuing throughout the remaining weeks of storage.

PAGE 88

88 Washed tilapia containing m et Hb at 9mol/kg muscle and stored at 3.7C developed significantly (p< 0.05) higher TBARS values than oxy and CO Hb (Figure 49) which did not differ significantly from each other. At days 4, 6, and 8 CO Hb was less prooxidative than m et Hb. Oxy Hb had significantly (p 0.05) higher TBARS va lues than m et Hb on day 6 and 8. TBARS values obtained at 25 C and 9 mol/kg muscle of CO oxy and m et Hb (Figure 410). Oxy Hb was significantly (p 0.05) m ore prooxidative than CO and m et Hb at week 20 and 24. Met Hb had significantly higher TBARS than CO Hb at week 20 only. Washed tilapia containing m et Hb at 12mol/kg muscle and stored at 3.7C developed significantly (p< 0.05) higher TBARS values than oxy and CO Hb (Figure 411) which did not differ significantly from each other. On day 2 oxy Hb was signifi cantly (p 0.05) more pro oxidative than m et Hb and CO Hb but remained level throughout the rest of the storage period. TBARS values obtained at 25C and 12 mol/kg of CO oxy and m et H b (Figure 412) indicated that oxy Hb was m ore prooxidative than CO and m et Hb. CO Hb wa s significantly (p 0.05) different from m et Hb at week 2 only. Oxy Hb was signif icantly different from CO and m et Hb at week 20 and 24. Protein Oxidation Analysis Wa shed tilapia muscle containing oxy Hb at 6 mol/kg muscle developed signi ficantly (p 0.05) more carbonyl s at 3.7C compared to CO and m et Hb, which did not differ significantly from each other throughout the 8 days of storage (Figure 413). Oxy Hb was signifi cantly more prooxidative than m et and CO Hb on day 2, 4, and 6 but not day 8 where there was no significant di fference among the three forms. Carbonyl values obtained at 25C and 6 mol/kg of CO oxy and m et Hb (Figure 414) showed CO Hb was less pro oxidative than met and oxy Hb which did not differ

PAGE 89

89 significantly (p 0.05) from each other. Oxy Hb was signifi cantly more prooxidative than m et and CO Hb at week 12 but at week 20 and 24 m et Hb was more prooxidative than oxy Hb. CO and m et Hb were both significantly (p 0.05) more pro ox idative at week 20 and 24 than ox yHb. Washed tilapia muscle containing oxy Hb at 9 mol/kg muscle developed significantly more carbonyl s at 3.7C compared to CO and m et Hb, which did not differ significantly (p 0.05) from each other throughout the 8 days of storage (Figure 415). There was no significant (p 0.05) difference on day 0 for the three forms but for the remaining storage days oxy Hb had significantly higher carbonyl values than CO and m et Hb which did not differ on any day. Carbonyl values obtained at 25C and 9 mol/kg of C O oxy and m et Hb (Figure 416) showed a pattern similar to that at 3.7C. CO Hb was signifi cantly (p 0.05) less prooxidative than m et and CO Hb. Wa shed tilapia muscle containing oxy Hb at 12 mol/kg muscle developed significantly (p 0.05) more carbonyl s at 3.7C compared to CO and m et Hb, which did not differ significantly from each other throughout the 8 days of storage (Figure 417). These significant differences were evident for each day of storage. Carbonyl values obtained a t 25C and 12 mol/kg of CO oxy and m et Hb (Figure 4 18) found that all three forms of Hb differed significantly. Oxy Hb had significantly (p 0.05) higher carbonyl values at week 0, 4, 12, and 24 than m et Hb. CO Hb had significantly l ower carbonyl values than both oxy and m et Hb at week 4, 8, 12, and 16. At week 20 there was no significant (p 0.05) difference found for the three forms. However, at week 24 CO Hb was signifi cantly more pro oxidative than met and oxy Hb.

PAGE 90

90 CO Release Results from washed tilapia muscle with C O Hb show that all concentrations tested released significantly greater % CO at 3.7C on day 0 than day 2, 4, and 6 (Figure 419). The sample containing 6 mol/kg of CO Hb released significantly less than the 9 and 12 mol/kg samples, on day 0. At day 8 all three Hb concentrations tested showed an increase in release of % CO. At 25C samples had released significant % CO by week 16 compared to earlier weeks, irrespective of Hb concentration (Figure 420). Unlike the findings at 3.7C, the samples with 6 mol/kg CO Hb released more CO compared to 9 and 12 mol/kg muscle. Color Analysis Samples with Hb at 6 mol/kg muscle, stored at 3.7C, showed a significant (p 0.05) difference in a* values for the three forms of Hb tested (Table 41). CO Hb had si gnifica ntly higher values than m et Hb throughout the 8 days of storage. CO Hb also had significantly higher a*v alues than oxy Hb on day 0, 2, and 6 only. Oxy Hb had signi ficantly higher a* values than m et Hb throughout the 8 days of storage. At 25C, a* values dif fered significantly for all three forms of Hb (Table 4 2). CO Hb had significantly higher a* values th an oxy and m et Hb throughout each of the 24 weeks of storage. Oxy Hb had signi ficantly higher a*values than m et Hb for week 0, 12, and 20. Samples with H b at 9 mol/kg muscle, stored at 3.7C, showed a signific ant difference (p 0.05) between CO and oxy Hb versu s m et Hb (Table 41), w hich had significantly lower a* values throughout each of the 8 days of storage. CO Hb had significantly (p 0.05) higher a*v alues than oxy Hb on day 0, 2, and 4. At 25C storage, a* values differed significantly for all three forms of Hb (Table 4 2). CO Hb had significantly (p 0.05) higher a*values than oxy and m et Hb throughout eac h of the 24

PAGE 91

91 storage weeks. Oxy Hb had signifi cantly (p 0.05) higher a*values than m et Hb for each storage week, except the last where it did not differ significantly. Samples with Hb at 12 mol/kg muscle, stored at 3.7C, showed a significant (p 0.05) difference between the three forms of Hb (Table 41). CO Hb had significantly higher a*values than oxy Hb on day 0, 2, and 4, and significantly (p 0.05) higher than m et Hb on each of the storage days except day 8. Met Hb had significantly lower a*values than oxy Hb throughout each of the 8 days of storage. At 25C, a* values differed significantly (p 0.05) for all three forms of Hb (Table 4 2). CO Hb had significantly (p 0.05) higher a*values than m et Hb for each week during the storage period. Ox yHb had significantly lower a* values than CO Hb at week 14 only. Heme Group Autoxidation At 3.7C storage, the % o xy Hb in washed tilapia muscle containing 6mol/kg muscle of Hb was significantly (p 0.05) different for the three forms of Hb (Figure 421a). CO Hb had significantly greater % oxy Hb than oxy and m et Hb. On day 0, 2, and 4 CO Hb differed significantly (p 0.05) from o xy Hb in % o xy Hb but not on day 6 and 8. The % oxy Hb in m et Hb samples was significantly different from o xy and CO Hb throughout the 8 days of storage. At 25C, all three forms diff ered significantly (p 0.05) in % oxy Hb (Figure 422a) Samples containing 6 mol/kg muscle of CO Hb had significantly greater % oxy Hb than m et Hb throughout each week of storage. Met Hb had significantly (p 0.05) less % oxy Hb than CO and oxy Hb. At 3 .7C storage, the % o xy Hb in washed tilapia muscle containing 9mol/kg muscle of Hb was was significantly (p 0.05) different for the three forms of Hb (Figure 4 21b), the same pattern of % o xy Hb as that seen with 6 mol/kg muscle of Hb. Although the % o x yHb in samples with 9 mol/kg muscle was greater, days 0, 2, and 4

PAGE 92

92 showed the same significant difference (p 0.05) between oxy and CO Hb with m et Hb having significantly less % o xy Hb throughout the 8 days of storage. At 25C, all three forms differed s ignificantly in % oxy Hb (Figure 422b). For each week of storage, except week 24, CO Hb had significantly greater % oxy Hb than oxy and m et Hb. At 3.7C storage, the % o xy Hb in washed tilapia muscle containing 12mol/kg muscle of Hb was was significantl y (p 0.05) different for the three forms of Hb (Figure 4 21c), and the significant differences were similar to those described for 6 and 9 mol/kg muscle. At 25C, all three forms differed significantly (p 0.05) in % oxy Hb (Figure 422c). For each week of storage, except week 24, CO Hb had significantly greater % oxy Hb than oxy and m et Hb, the pattern being similar to that at 6 mol/kg muscle with CO Hb having slightly higher % o xy Hb at 12 mol/kg muscle. At 3.7C storage, the % m et Hb in washed tilapia muscle containing 6mol/kg muscle Hb was was significantly (p 0.05) different for the three forms of Hb (Figure 4 23a). of Met Hb had, as expected, significantly higher % met Hb than CO and oxy Hb on each day of storage. Oxy Hb had significantly (p 0.0 5) greater % m et Hb than CO Hb. At 25C, all three forms differed significantly in % oxy Hb (Figure 424a). Me t Hb had significantly (p 0.05) higher % met Hb than oxy and m et Hb throughout each day of the 24 week storage period. Oxy Hb had greater % m et Hb than CO Hb throughout the 24 weeks. At 3.7C storage, the % m et Hb in washed tilapia muscle containing 9mol/kg muscle Hb was was significantly (p 0.05) different for the three forms of Hb (Figure 4 23b). Met Hb had significantly greater % m et Hb on each day of storage than oxy and CO Hb. Oxy Hb had significantly (p 0.05) greater % m et Hb than CO Hb on day 0, 2,

PAGE 93

93 and 4 but was not significantly (p 0.05) different on day 6 and 8. At 25C, all three forms differed significantly in % o xy Hb formed (Figure 4 24b). Me t Hb had significantly (p 0.05) higher % met Hb than oxy a nd m et Hb throughout each day of the 24 week storage period. Oxy Hb had significantly greater % m et Hb than CO Hb at week 0, 4, 8, 12, 16, and 20; they did not differ significantly week 20. At 3.7C storage, the % m et Hb in washed tilapia muscle containing 12mol/kg muscle Hb was was significantly (p 0.05) different for the three forms of Hb (Figu re 4 23c). Met Hb had greater % met Hb than oxy and CO Hb on each st orage day. Oxy Hb had greater % m et Hb than CO Hb day 0, 2, and 4 but these did not differ significantly (p 0.05) on day 6 and 8. At 25C, all three fo rms differed significantly in % deoxy Hb (Figure 424c). Met Hb had significantly higher % met Hb than oxy and m et Hb through out each day of the 24 week storage period. Oxy Hb had significantly greater % m et Hb than CO Hb a t weeks 0, 4, 8, 12, and 16 but not at weeks 20 and 24. At 3.7C storage, the % d eox yHb in washed tilapia muscle containing 6mol/kg muscle Hb showed that met and oxy Hb had significantly (p 0.05) greater % d eoxy Hb than oxy Hb (Figure 425a). CO Hb had significantly less % deoxy Hb compared to met Hb and oxy Hb on day 0, 2, and 4, but they were not significantly different day 6 and 8. At 25C CO Hb had lower % deoxy Hb had than met and ox yHb (Figure 4 26a). CO Hb and m et Hb differed significantly in % deoxy Hb at week 0 only, while CO Hb and o xy Hb differed significantly (p 0.05) at each week of storage except week 16. At 3.7C storage, the % d eox Hb in washed tilapia muscle containing 9mol/kg muscle Hb showed that met and oxy Hb had significantly greater % d eoxy Hb (Figure 4 25b) than CO Hb day 0, 2, 4, and 6. CO Hb had significantly (p 0.05) less % deoxy

PAGE 94

94 than oxy Hb day 0, 2, 4, and 6. On day 8 there were no significant differences (p 0.05) for the three forms of Hb. At 25C CO Hb had significantly less % deoxy Hb than oxy Hb at week 0, 4, and 8, and significantly (p 0.05) less than m et Hb at week 0 and 8 (Figure 426b). The remaining days of storage showed no significant differences (p 0.05) for the three forms of Hb. At 3.7C storage, the % d eox yHb in washed tilapia muscle containing 9mol/kg muscle Hb showed that met and oxy Hb had significantly (p 0.05) greater % d eoxy Hb (Figure 425c). CO Hb had significantly lower % deoxy Hb than oxy and m et Hb on each day of storage, except day 8 where there were no significant differences for the three forms. At 25 CO Hb had si gnificantly (p 0.05) less % d eoxy than met and oxy Hb at week 0 and less than oxy Hb at week 8 (Figure 426c). There were no significant differences (p 0.05) for the remaining weeks of storage. Discussion The prooxidative activity of fish Hb has been well established in numerous previous studies ( 56 ) The form of Hb present in muscle meat was shown to be one of the important determinants of the rate and magnitude of oxidation during the storage ( 1, 105) Based on accumulated evidence, it appears reasonable to hypothesize that the concentration of the Hb present in fish muscle system may also affect its oxidation pattern. However, to our knowledge, littl e research is available determining possible associations between Hb concentrations and their catalytic capacity to oxidize lipids and proteins in a WMTM system. In this study t he prooxidative activities of oxy CO and m et Hb as a function of different concentrations were examined under two storage conditions. Formation of TBARS, lipid hydroperoxides, and protein carbonyl was used as indicators of oxidative

PAGE 95

95 activities. At 3.7C the concentration of m et Hb significantly influenced production of TBARS wi th lower TBARS correlating with lower m et Hb concentrations (Figure 47, 4 9, 4 11). Increased oxidation with increasing Hb concentrations is consistent with the findings of Richards and Hultin ( 11) who found increasing the mol of Hb added to cod washed muscle produced higher levels of TBARS earlier and rancidity was also detectable earlier in the higher concentrations (1.8 and 5.8 mol) versus the lower concentrations (0.06 and 0.50 mol). In contrast to metHb, concentrations of o xy and CO Hb did not appear to affect TBARS formation to any considerable extent (Figure 47, 4 9, and 4 11). Previous reports have established that m et Hb will readily react with preformed hydrohydroperoxides to generate further oxidation ( 38 ) Increased availability of m et Hb for reaction with hydroperoxides with increased concentration may partially explain the observed res ults. Very different trends were seen for oxidation measured by lipid hydroperoxides. Eac h increase in concentration of oxy Hb showed a significant rise in lipid hydroperoxides (Figure 4 1, 4 3, and 45). Rapid deoxygenation as o xy Hb concentrantion are i ncreased could potentially increase lipid oxidation. Of note, at all concentrations, the rate and pattern of lipid oxidation for each form of Hb was similar, differing only in magnitude (Figure 4 1, 4 3, and 45). Results at 25C were different than thos e at 3.7C. Overall TBARS values were not significantly affected by Hb concentration (Figure 48, 4 10, 412). Interestingly, during frozen storage oxy Hb led to the highest TBARS value, regardless of concentration. Lipid oxidation increased significantly for all forms of Hb as their concentration increased (Figure 4 2, 4 4). However, CO Hb led to less li pid

PAGE 96

96 hydroperoxides compared to oxy and m et Hb regardless of concentration (Figure 4 2, 4 4, 4 6). TBARS and lipid hydroperoxide data support the hypothesi s that COHb is least pro oxidative of the three forms, regardless of Hb concentrations in the washed system. Protein oxidation (carbonyls) appeared not to be affected by Hb concentration at 3.7C (Figure 4 13, 4 15, 417), but was affected more by temperature. Although oxy Hb remained the most prooxidative with respect to protein oxidation of the three forms at both temperatures, at 25C a significant difference in protein oxidation emerged between CO Hb and m et Hb. There was no significant difference between the two forms at 3.7C but at 25C a significant difference emerged with CO Hb being the least prooxidative, i.e. leading to the least protein oxidation. Effects of different Hb concentrations on red color deterioration were also evaluated. At 3.7 C, the higher concentration of oxy Hb was associated with more stable red color (Table 41). Contrary to what was expected, increasing the concentration of CO Hb did not lead to a more stable red color. On the contrary, the higher concentration of CO Hb l ed to lower a*value by the end of the storage time (Table 41). In contrast, at 25C, red color deterioration was significantly retarded by higher concentrations of all of the forms of Hb (Table 42). CO Hb samples showed the most red color stability at 25C. Unfortunately, accumulated research data is insufficient to explain these obtained results. More studies need to be conducted to further explore the effects of concentrations on Hb pro oxidative activities.

PAGE 97

97 Conclusion CO Hb was significantly less pr o oxidative compared to other forms of Hb regardless of concentration used. This work suggests fish with CO bound to Hb may be more stable with respect to lipid oxidation. Temperature significantly affected the stability of CO Hb and its oxidation to M et H b.

PAGE 98

98 Table 4 1 Changes in a*value in washed tilapia muscle containing different forms of Hb at a concentration of 6, 9, and 12mol/kg at 3.7C averaged across pH level and NaCl added. a* value 6mol Hb/kg muscle Storage time (day) Oxy Hb b CO Hb a Met Hb c 0 15.42.2 19.94.5 6.51.1 2 12.95.5 19.45.7 8.31.4 4 11.97.3 18.17.3 6.91.4 6 10.77.7 13.37.9 5.01.8 8 9.58.2 12.48.3 3.82.1 a* value 9mol Hb/kg muscle Storage time (day) Oxy Hb a CO Hb a Met Hb b 0 20.43.1 22.03.5 8.21.4 2 16.87.5 20.74.7 10.21.7 4 15.39.0 16.58.1 8.02.2 6 13.510.4 9.67.7 6.02.5 8 12.410.0 9.37.1 4.82.7 a* value 12mol Hb/kg muscle Storage time (day) Oxy Hb a CO Hb b Met Hb c 0 24.03.8 23.83.5 9.51.2 2 20.19.0 22.14.5 12.21.1 4 18.310.6 17.07.8 9.32.6 6 16.212.0 11.09.1 7.43.2 8 14.311.0 7.85.8 5.83.1 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n= 18 ). Letters within the same rows indicate statistically (<0.05) significant differences separated by Tukeys HSD.

PAGE 99

99 Table 4 2 Changes in a*value in washed tilapia muscle containing different forms of Hb at a concentration of 6, 9, and 12mol/kg at 25 C averaged across pH level and NaCl added. a* value 6mol Hb/kg muscle Storage time (day) Oxy Hb b CO Hb a Met Hb c 0 15.42.2 19.94.5 6.51.1 4 12.83.9 24.05.3 9.51.6 8 11.73.8 22.17.0 8.71.2 12 12.55.5 21.06.6 7.81.3 16 11.36.0 19.67.3 7.61.4 20 11.36.8 19.07.1 7.21.6 24 9.97.3 19.08.4 6.31.6 a* value 9mol Hb/kg muscle Storage time (day) Oxy Hb b CO Hb a Met Hb c 0 20.43.1 22.03.5 8.21.4 4 17.04.9 25.74.8 12.61.8 8 15.85.4 24.75.0 11.21.3 12 13.55.9 22.65.3 10.31.4 16 15.37.9 21.55.5 10.21.7 20 15.48.6 20.45.6 9.71.8 24 13.59.3 19.35.9 9.71.9 a* value 12mol Hb/kg muscle Storage time (day) Oxy Hb b CO Hb a Met Hb c 0 24.03.8 23.83.5 9.51.2 4 21.86.6 25.14.8 14.81.7 8 19.96.6 22.25.4 13.41.9 12 14.610.0 20.65.4 12.71.6 16 17.17.9 20.85.9 12.81.9 20 18.710.0 18.35.2 12.41.9 24 16.810.8 16.85.5 12.42.5 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n=18). Letters within the same rows indicate statistically (<0.05) significant differences separated by Tukeys HSD.

PAGE 100

100 Figure 41 Lipid hydroperoxide v alues in washed tilapia muscle containing different forms of Hb at a concentration of 6 mol/kg at 3.7C averaged across pH level and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 42 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at a concentration of 6 mol/kg at 25C averaged across pH level and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. -100 0 100 200 300 400 500 600 0 2 4 6 8Lipid hydroperoxide mol/kg muscleDay (b) CO Hb (b) Met Hb (a) Oxy Hb -50 0 50 100 150 200 250 300 350 400 450 0 4 8 12 16 20 24Lipid hydroperoxide mol/kg muscleWeek (ab) CO Hb (b) Met Hb (a) Oxy Hb

PAGE 101

101 Figure 43 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at a concentration of 9 mol/kg at 3.7C averaged across pH level and NaCl added. L etters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 44 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at a concentration of 9 mol/kg at 25C averaged across pH level and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. -100 0 100 200 300 400 500 600 0 2 4 6 8Lipid hydroperoxide mol/kg muscleDay (c) CO Hb (b) Met Hb (a) Oxy Hb -50 0 50 100 150 200 250 300 350 400 450 0 4 8 12 16 20 24Lipid hydroperoxides mol/kg muscleWeek (b) CO Hb (a) Met Hb (a) Oxy Hb

PAGE 102

102 Figure 45 Lipid hydroperoxide values in washed tilapia muscle containing different f orms of Hb at a concentration of 12 mol/kg at 3.7C averaged across pH level and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 46 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at a concentration of 12 mol/kg at 25C averaged across pH level and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant diff erences separated by Tukeys HSD. -100 0 100 200 300 400 500 600 0 2 4 6 8Lipid hydroperoxide mol/kg muscleDay (c) CO Hb (b) Met Hb (a) Oxy Hb -50 0 50 100 150 200 250 300 350 400 450 0 4 8 12 16 20 24Lipid hydroperoxides mol/kg muscleWeek (b) CO -Hb (a) Met -Hb (a) Oxy-Hb

PAGE 103

103 Figure 47 TBARS in washed tilapia muscle containing different forms of Hb at a concentration of 6 mol/kg at 3.7 C averaged across pH level and NaCl added. Letters within the legend of the figure indicate statisticall y (<0.05) significant differences separated by Tukeys HSD. Figure 48 TBARS in washed tilapia muscle containing different forms of Hb at a concentration of 6 mol/kg at 25C averaged across pH level and NaCl added. Letters within the legend of the f igure indicate statistically (<0.05) significant differences separated by Tukeys HSD. -5 0 5 10 15 20 25 30 35 0 2 4 6 8mol TBARS/kg muscleDay (a) CO Hb (a) Met Hb (a) Oxy Hb -5 0 5 10 15 20 25 30 0 4 8 12 16 20 24mol TBARS/kg muscleWeeks (b) CO Hb (b) Met Hb (a) Oxy Hb

PAGE 104

104 Figure 49 TBARS in washed tilapia muscle containing different forms of Hb at a concentration of 9 mol/kg at 3.7 C averaged across pH level and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 41 0 TBARS in washed tilapia muscle containing different forms of Hb at a concentration of 9 mol/kg at 25C averaged across pH level and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. -5 0 5 10 15 20 25 30 35 0 2 4 6 8mol TBARS/kg muscleDay (b) CO Hb (a) Met Hb (b) Oxy Hb -5 0 5 10 15 20 25 30 0 4 8 12 16 20 24mol TBARS/kg muscleWeeks (b) CO Hb (b) Met Hb (a) Oxy Hb

PAGE 105

105 Figure 411. TBARS in washed tilapia muscle containing different forms of Hb at a co ncentration of 12 mol/kg at 3.7 C averaged across pH level and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 41 2 TBARS in washed tilapia muscle containing different forms of Hb at a con centration of 12 mol/kg at 25C averaged across pH level and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. -5 0 5 10 15 20 25 30 35 0 2 4 6 8mol TBARS/kg muscleDay (b) CO Hb (a) Met Hb (b) Oxy Hb -5 0 5 10 15 20 25 30 0 4 8 12 16 20 24mol TBARS/kg muscleWeeks (b) CO Hb (b) Met Hb (a) Oxy Hb

PAGE 106

106 Figure 413. Carbonyl value s in washed tilapia muscle containing different forms of Hb at a concentration of 6 mol/kg at 3.7C averaged across pH level and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukey s HSD. Figure 414. Carbonyl values in washed tilapia muscle containing different forms of Hb at a concentration of 6 mol/kg at 25 C averaged across pH level and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 0 2 4 6 8nM carbonyls/mgDay (b) CO Hb (c) Met Hb (a) Oxy Hb 0 10 20 30 40 50 60 70 0 4 8 12 16 20 24nM carbonyls/mg muscleWeek (b) CO Hb (a) Met Hb (a) Oxy Hb

PAGE 107

107 Figure 415. Carbonyl values in washed tilapia muscle containing different forms of Hb at a conc entration of 9 mol/kg at 3.7C averaged across pH level and NaCl added. Letters within the legend of the f igure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 416. Carbonyl values in washed tilapia muscle containing different forms of Hb at a conc entration of 9 mol/kg at 25 C averaged across pH level and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 0 2 4 6 8nM carbonyls/mgDay (b) CO Hb (b) Met Hb (a) Oxy Hb 0 10 20 30 40 50 60 70 0 4 8 12 16 20 24nM carbonyls/mg muscleWeek (b) CO Hb (a) Met Hb (a) Oxy Hb

PAGE 108

108 Figure 417. Carbonyl values in washed tilapia muscle containing different forms of Hb at a conc entration of 12 mol/kg at 3.7C averaged across pH level and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 418. Carbonyl values in washed tilapia muscle containing different forms of Hb at a conc entration of 12 mol/kg at 25C averaged across pH level and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 0 2 4 6 8nM carbonyls/mgDay (b) CO Hb (b) Met Hb (a) Oxy Hb 0 10 20 30 40 50 60 70 0 4 8 12 16 20 24nM carbonyls/mg muscleWeek (c) CO Hb (b) Met Hb (a) Oxy Hb

PAGE 109

109 Figure 41 9 %CO released during 3.7C storage in w ashed tilapia muscle containing CO Hb at a concentration of 6, 9, and 12mol/kg muscle. The effect of forms of Hb (Oxy CO and Met Hb) by storage interaction averaged across pH level and NaCl added. Letters within the legend of the figure indicate stati stically (<0.05) significant differences separated by Tukeys HSD. Figure 42 0 %CO released during 25C storage in washed tilapia muscle containing CO Hb at a concentration of 6, 9, and 12 mol/kg muscle. The effect of forms of Hb (Oxy CO and Met Hb) by storage interaction averaged across pH level and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 5 10 15 20 25 30 35 40 45 50 0 2 4 6 8% CODay (b) 6mol (a) 9mol (a) 12mol 0 5 10 15 20 25 30 35 40 45 50 0 4 8 12 16 20 24 % COWeek (a) 6mol (b) 9mol (c) 12mol

PAGE 110

110 Figure 421. %Oxy Hb in washed tilapia muscle c ontaining different forms of Hb at a concentration of a) 6, b) 9, and c) 12 mol/kg at 3.7C averaged across pH level and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant diff erences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8% Oxy -HbDay (c) Met -Hb (a) CO -Hb (b) Oxy-Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Oxy -HbDay (c) Met Hb (a) CO Hb (b) Oxy Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Oxy -HbDay (c) Met Hb (a) CO Hb (b) Oxy Hb (c)

PAGE 111

1 11 Figure 42 2 %Oxy Hb in washed tilapia muscle containing different forms of Hb at a concentration of a) 6, b) 9, and c) 12 mol/kg at 25C averaged across pH level and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24% Oxy -HbWeek (c) Met Hb (a) CO Hb (b) Oxy Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Oxy -HbWeek (c) Met Hb (a) CO Hb (b) Oxy Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24% Oxy -HbWeek (c) Met Hb (a) CO Hb (b) Oxy Hb (c)

PAGE 112

112 Figure 423. %Met Hb in washed tilapia muscle containing different forms of Hb at a concentration of a) 6, b) 9, and c) 12 mol/kg at 3.7C averaged across pH leve l and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Met -HbDay (a) Met Hb (c) CO Hb (b) Oxy Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Met -HbDay (a) Met Hb (c) CO Hb (b) Oxy Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Met -HbDay (a) Met Hb (c) CO Hb (b) Oxy Hb (c)

PAGE 113

113 Figure 424. %Met Hb in washed tilapia muscle containing different forms of Hb at a concentration of a) 6, b) 9, and c) 12 mol/kg at 25C averaged across pH level and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24% Met HbWeek (a) Met -Hb (c) CO -Hb (b) Oxy-Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Met -HbWeek (a) Met Hb (c) CO Hb (b) Oxy Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24% Met -Hb Week (a) Met Hb (c) CO Hb (b) Oxy Hb (c)

PAGE 114

114 Figure 425. %Deoxy Hb in washed tilapia muscle containing different forms of Hb at a concentration of a) 6, b) 9, and c) 12 mol/kg at 3.7C averaged across pH level and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Deoxy -HbDay (a) Met Hb (b) CO Hb (a) Oxy Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Deoxy -HbDay (a) Met Hb (b) CO Hb (a) Oxy Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Deoxy -HbDay (a) Met Hb (b) CO Hb (a) Oxy Hb (c)

PAGE 115

115 Figure 42 6 %Deoxy Hb in washed tilapia muscle containing different forms of Hb at a concentration of a) 6, b) 9, and c) 12 mol/kg at 25C averaged across pH level and NaCl added. Letters within the legend of the figure i ndicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24% Deoxy -HbWeek (a) Met Hb (b) CO Hb (a) Oxy Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Deoxy -HbWeek (a) Met Hb (b) CO Hb (a) Oxy Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24% Deoxy -HbWeek (a) Met Hb (b) CO Hb (a) Oxy Hb (c)

PAGE 116

116 CHAPTER 5 EFFECT OF p H ON THE PRO OXIDATIVE ACTIVITY O F OXY, CO AND MET HEMOGLOBIN IN A WASH ED MINCED TILAPIA MU SCLE SYSTEM AT TWO DIFFERENT STORAGE TEMPERATURES Introduction Most post mortem fish muscle and muscle systems have pH values below 7.0. The role of pH on lipid oxidation under post mortem conditions is important to understand. At lower pH values formation of Met Hb occurs more rapidly and the level of d eoxy Hb increases sharply ( 4 ) Researchers have demonstrated t hat trout and tilapia Hb added to washed cod muscle at 2C are more pro oxidative at pH 6.3 compared to 7.4. Washed cod muscle containing trout Hb showed more prooxidative activity than tilapia Hb at both pH levels ( 80) Howev er, trout and tilapia Hb added to washed tilapia developed significant levels of TBARS at pH 6.3, while at pH 7.4 TBARS values were significantly depressed for both Hb samples. Using washed cod muscle and cod Hb, Pazos ( 81) established that decreasing the pH from 7.8 to 6.3 greatly decreased the lag phase and increased the rate of lipid oxidation ( 81) The lag phase at pH 7.8 was ~40 h and the oxidation rate was slower compared to pH 6.8 and 3.5 which had faster rates of oxidation and lag times of 6 and 3 h, respectively. At pH values above 7.8, slower rates of oxidation were observed. Lowering the pH below neutrality decr eases the oxygenation of Hb. This is known as the Bohr Effect ( 43 ) .The Bohr Effect can explain the high levels of d eoxy Hb at postmortem. When pH levels are decreased below 6.5, further deoxygenation occurs and this is known as the Root effect. Richards and Hultin ( 4 ) propose d that deoxy Hb may play the role of a catalyst in lipid oxidation. Because its heme crevice is more

PAGE 117

117 accessible, d eoxy Hb acts as a stronger oxidation catalyst than o xy Hb ( 44 ) Undeland and others ( 22 ) found that higher rates of oxidation at pH 6.0 c orresponded to a greater formation of d eoxy and m et Hb when studying Hb from menhaden, mackerel, flounder, and pollock. At pH 6.0 all the Hbs were equally active pro oxidants ; however, at pH 7.2, the pro oxidative activity of Hb was reduced except Hb fr om pollock. At both pH levels, pollock had greater formation of deoxy and m et Hb. The acceleration of lipid oxidation at pH 6 compared to 7.2 may be related to lower oxygenation (Bohr or Root Effect) or to increased oxidation of these Hbs ( 11 ) Maintaining heme proteins in the reduced state has important implications for the quality of seafood products. Treating seafood with carbon monoxide at post mortem pH levels (pH 6.5) maintains the heme protein in the reduced state because car bon monoxide (CO) binds with the heme protein more readily than O2, replacing it from the heme. It is hypothesized that CO Hb will be less pro oxidative at lower pH levels than o xy or m et Hb. As pH is decreased, hemin loss rates increase dramatically ( 60 ) Aranda and others ( 60 ) suggested that there are four mechanisms by which auto oxidation and hemin loss occur. Thes e include steric displacement of bound ligands, weak anchoring of the heme propionate to the globin, larger channels for solvent entry into the heme pocket, and weakened interactions with the distal histidine. Hargrove and others ( 59) developed an assay for hemin dissociation where His64 in sperm whale Mb was replaced by Tyr, producing a holoprotein wi th a discrete green color. Phe replaced Val68 in the same protein to increase its stability, while retaining high affinity for hemin. This protein can then be used for complete ext raction of hemin from Hb and Mb, giving absorbance changes to allow reactions at low hemin

PAGE 118

118 concentrations. When this protein (apoprotein Tyr64Val68) is mixed i n excess with mutant Hb, the solution turns from brown to green, and the absorbance changes can be used to measure the rate of dissociation of hemin. Mutant hybrids of human Hb were prepared by substituting Gly at His64 (E7) and oxidized with ferricyanide and then reacted quickly with apoprotein Tyr64Val68. Hemin dissociation from the Gly64 mutants and subsequent uptake by apoprotein Tyr64Val68 resulted in rapid absorbance increases. The authors conclude d that the results presented indicate that apoprotein Tyr64Val68 can be used reliably for measuring hemin loss. Materials and Methods The following methods have been described in detail in chapter 3: Preparation of Washed Minced Tilapia Muscle (MTWM) Collection of Fish Blood, Preparation of Hemolysa te, Quantification of HB Levels in Hemolysate, oxy CO and m et Hb Preparation, Sample Preparation: Addition of Hb and NaCl, Determi nation of Peroxide Value (PV), Determination of Thiobarbituric Acid Reactive Substances (TBARS), Determination of Carbonyl Groups, Heme Group Autoxidation, and Color Analysis. The Gas Chromatography (GC) Method was described in Chapter 4. Determination of Hemin Loss The dissociation of met Hb was determined according to the method of Hargrove and others ( 106 ) modified by Richards and Grunwald ( 58 ) In this procedure, previously prepared and frozen at 80C holoprotien H64Y with distinct green color was used to prepare the apomyoglobin. The determination of hemin loss was conducted in the following manner.

PAGE 119

119 Prepa ration of buffer In a 1000 ml beaker, 150mM of Bis tris was dissolved in distilled water containing 450mM sucrose. Once dissolved, the pH of the solution was adjusted to the desired pH value of 6.3, 6.8, or 7.3. The solution was then transferred to a grad uate cylinder, diluted to a final volume of 500 ml with distilled water, filtered by vacuum filtration using 0.45 m, Millipore filter paper (Fisher Scientific), and then transferred to a bottle and stored at 24 C overnight. Preparation of FPLC ( fat protein liquid chromatography) Before using the desalting column of the FPLC, it was cleaned of any potential bacteria growth and conditioned by pumping initially 20% ethanol followed by degassed distilled water, and then the first buffer of pH 7.3 at a rate of 5 ml/min. Sample preparation Fro zen Holoprotein H64Y (Mb of sperm whale) was thawed under running water. In an ice cold culture tube (tube 1), 1 ml of the thawed Holoprotein was mixed with 1 ml maleic acid (0.2M) and stirred gently for 1 min till the co lor changed from browngreen to brownred. The sample was incubated in an ice bath for 10 min before 2 ml of ice cold extraction solvent (methyl ethyl ketone) was added, followed by vortexing for 10 seconds, and then incubated in ice for another 10 min. Th e lower layer in tube 1, a goldgreen phase, was transferred to another tube (tube 2) where 2 ml of ice cold extraction solvent was added, vortexed for 20 sec, and then incubated for 10 min in an ice bath. The lower layer in tube 2, a golden phase, was t ransferred to another tube (tube 3). Using a 5 ml syringe, the golden phase in tube 3 was injected into the desalting column and run for 25 min in order to collect the protein fraction (Apoprotein) in a buffer

PAGE 120

120 at pH 7.3. After the 25 min run, sample collec tion began with the first appearance of the chromatogram peak, and stopped before the endpoint of the peak in order to avoid any residual ketone fraction. This same procedure was followed for pH 6.8 and 6.3. Verification and analysis of extracted samples UV vis absorbance spectroscopy in the range of 200700 nm was used to ascertain the success of the preparation and to determine the concentration of the Apo sample prepared. The rate of hemin loss The dissociation rate of hemin was determined spectrophotometrically. In a quartz cuvtte, 1 mL mixture of 10 M met Hb (heme base) was added to 40 M ApoMb with 150 mM buffer of the desired pH and 450 mM sucrose. The absorbance was recorded at 600700 nm against a blank. The blank contained 1 mL 150 mM buffer wit h 450 mM sucrose and 12.5 l 10 mM Tris buffer at pH 8.0. Concentrations of met Hb and ApoMb were calculated using a previously prepared standard curve in Dr. Mark Richardss lab in the Department of Animal Science at the University of Wisconsin, Madison ( 58 ) The cuvettes were transferred to a 6 cell holder and the absorbance was recorded for 24 hours. This procedure was performed at two different temperatures (4 and 25C). Calculation of Dissociation R ate Heme that was released from the samples was gathered by the globin since globin with tyrosine replacement has a strong affinity for heme. Heme binds to the mutant globin upon release, resulting in a green color formation that has a strong absorbance at 600 nm. Rate of hemin loss was calculated using the Igor Pro software (WaveMetrics Inc., Portland, OR) and the exponential function ( 58 ) This procedure was conducted in collaboration with Dr. Richardss lab.

PAGE 121

121 Results Lipid Oxidation Analysis Lipid hydroperoxide values obtained at 3.7C and pH 6.3 (Figure 5 1) showed that o xy Hb oxidized quickly and more (p<0.05) than m et and CO H b, reaching a peak at day 4 and then decreasing until day 8. CO Hb and m et Hb did not differ significantly in hydroperoxide values until day 8, where CO Hb was significantly less pro oxidative than m et and o xy Hb. At pH 6.3 and 25C (Figure 52), o xy and m et Hb were significantly more prooxidative than CO Hb but were not significantly different (p from each other. On week 4 m et Hb was significantly (p less pro oxidative than CO Hb, but by week 20 and 24 CO Hb was significantly less prooxidative than m et and o xy Hb. At 3.7C and pH 6.8 (Figure 53), lipid hydroperoxides differed significantly (p 0.05) for all three forms of Hb, with m et Hb being sign ificantly more prooxidative Less lipid hydroperoxides formed at pH 6.8 compared to pH 6. 3. CO Hb was significantly less pro oxidative than m et Hb on day 4 and day 6, but by day 8, there were no significant differences between them. At pH 6.8 and 25C (Figure 5 4), m et Hb was significantly more prooxidative than CO and o xy Hb. On week 24, CO Hb was significantly (p less pro oxidative than m et and o xy Hb. Lipid hydroperoxide values at 3.7C and pH 7.3 (Figure 5 5) were very low and did not show significant differences (p between the three forms until day 6 when m et Hb had higher values (p<0.05) than o xy and CO Hb. Throughout the 8 days of storage CO and m et Hb did not differ in lipid hydroperoxide values. At pH 7.3 and 25C (Figure 5 6), there were no significant differences for the three forms of Hb and very low levels of lipid hy droperoxides formed.

PAGE 122

122 TBARS development at 3.7C (Figure 5 7) and pH 6.3 showed CO Hb was the least pro oxidative of all forms tested. Oxy Hb oxidized significantly (p 0.05) faster than CO Hb. Met Hb did not differ significantly from o xy Hb. By day 6, m et H b was significantly more prooxidative than o xy Hb, as samples with o xy Hb were on a decline. By day 8 there were no significant differences (p 0.05) for the three forms. At 25C and pH 6.3, o xy Hb (Figure 58) was also significantly more pro oxidative than m et and CO Hb. Although CO Hb appeared to be the least prooxidative it did not differ significantly from m et Hb until week 16 and on. TBARS development was reduced at pH 6. 8 at 3.7C compared to pH 6.3. Significant oxidation of samples did not oc cur u ntil day 4 for all forms. Contrary to results at pH 6.3, where o xy Hb was most pro oxidative, m et Hb was overall significantly (p 0.05) more prooxidative at pH 6.8 than o xy and CO Hb (Figure 5 9), reaching maximum oxidation on day 6. CO and o xy Hb were, however not significantly different from each other, except for day 8. Results for samples stored at 25C and pH 6.8 (Figure 510) showed less TBARS formation compared to pH 6.3. Not much differenc e was found between the samples. H owever at the end of the study, o xy Hb had the highest oxidation. At week 16, 20, and 24 significant differences (p 0.05) were seen between the three forms, with CO Hb being least prooxidative. Samples at pH 7.3 and 3.7C (Figure 511) developed even less TBARS than at pH 6.8. Like at pH 6.8, met Hb was most prooxidative, however significant (p 0.05) oxidation didnt occur until day 6 and on. Oxy and CO Hb caused very little oxidation during the 8 day storage period and did not differ significantly from each other. At 25C and pH 7.3 (Figure 5 12) even lower levels of TBARS were seen than at ph 6.8,

PAGE 123

123 although oxy Hb developed similar levels of TBARS at weeks 20 and 24, significantly more than m et and CO Hb. Throughout the 24 weeks of storage, CO and m et Hb led to very little TBARS formation and did not differ significantly from each other. Protein Oxidation Analysis Carbonyl values obtained at 3.7C and pH 6.3 (Figure 513) revealed that CO Hb was significantly less prooxidative, with respect to protein oxidation, than met and o xy Hb. CO Hb did however have double the carbonyl values (p<0.05) on day 0 compared to the other forms. Protein oxidation was reduced in CO Hb containing samples and at days 4 8 it had significantly lower carbonyl values than o xy Hb. On day 6 and 8, CO and m et Hb did not differ significantly (p 0.05) Oxy Hb developed the highest leve l of carbonyl groups at day 4. Oxy and m et Hb differed significantly on day 2 and 4 only. At 25C and pH 6.3 (Figure 5 14), the three forms of Hb differed significantly. CO Hb had the lowest carbonyl values overall throughout frozen storage, while o xy Hb led to the highest carbonyl value at week 16. All three forms of Hb differed significantly (p 0.05) at each week of storage, except for week 24, which showed no significant difference. At pH 6.8 and 37C (Figure 515), the three forms of Hb differed significantly (p 0.05) in carbonyl values, w ith o xy Hb again being the most prooxidative, and CO Hb having the lowest carbonyl value (until day 6, where Met Hb has equal values). CO and m et Hb did not differ significantly (p 0.05) on days 48. Samples containing CO and m et Hb had lower carbonyl v alues at pH 6.8 compared to pH 6.3, while values were similar for o xy Hb. A different protein oxidation trend was seen at 25C and pH 6.8 (Figure 516) compared to pH 6.3 for all samples. Oxy Hb was significantly more prooxidative than the other forms until week 12, where carbonyls peaked and then decreased rapidly until week 24 where values were similar to values at week 0. CO and

PAGE 124

124 m et Hb did not differ significantly on week, 4, 8, 12, and 20, but differed significantly (p 0.05) on week 0, 16, and 24, wi th CO Hb being less prooxidative. Samples at pH 7.3 and 37C (Figure 517) developed fewer carbonyl g roups than at ph 6.8 and 6.3. However, like those seen at the lower pH values, o xy Hb was the most prooxidative of all forms, while m et Hb led to the low est carbonyl values. CO and m et Hb differed significantly in carbonyl values on day 4 only. A similar trend in the development of carbonyls was seen at 25C and pH 7.3 (Figure 5 18) as was seen at pH 6.8, where o xy Hb samples peaked at week 12, m et Hb at week 20 and CO Hb wa s still increasing at week 24. Overall there was little significance difference (p 0.05) between the three Hb forms. However, at week 0 and 4, o xy Hb had significantly higher carbonyl levels than CO and m et Hb. At week 8, m et Hb had significantly (p 0.05) higher carbonyl values than o xy Hb but on week 24 CO Hb had significantly higher values than both m et and o xy Hb. Hemin Loss Rate Analysis of the rate of hemin loss (Table 51) showed hemin loss occurred at a faster rate at 25C compared to 4C. The rate of hemin loss was also affected by pH; the higher the pH, the slower the rat e of hemin loss. Tilapia Hb, which was used in the current study, was compared to trout Hb and it was found that at pH 6.3 and both temperatures (25C and 4C), the rate of hemin loss was faster for trout Hb, suggesting a stronger hemeprotein interaction of the ti lapia Hb. CO Release Amount of CO loss from CO Hb containing samples varied by storage day and pH (Figure 519). Samples at pH 6.3 and 6.8 releas ed significantly greater amount of CO on day 0 than any other day, and released significantly more than samples at pH 7.3.

PAGE 125

125 Samples at pH 7.3 overall had CO more strongly bound to Hb, while at day 8 they showed a significant amount released. As storage time increased, the amount of CO released per day decreased, until day 6 and 8 when increasing amounts of CO were released. Samples stored at 25C for 24 weeks (Figure 520) overall showed significant amounts of CO released on each week of storage. Samples at pH 6.3 and 6.8 lost more CO than samples at ph 7.3 and showed a sharp increase in CO loss at week 16. All samples lost significantly more CO at weeks 1624 compared to weeks 012. Color Analysis Significant differences in a* value were seen in samples co ntaining CO o xy and m et Hb at 3.7C and pH 6.3 (Table 5 2). Samples containing CO Hb had significantly (p 0.05) higher a*values than o xy and m et Hb on days 0, 2, and 4. As expected, on day 0, o xy and CO Hb had significantly (p 0.05) higher a*values than m et Hb, while on on day 2 and 4 Met Hb had significantly (p 0.05) higher a* values. At 25C and pH 6.3 (Table 53) CO Hb had significantly higher a*values than both o xy and m et Hb during the entire 24 weeks. Only at week 0 and 24 did o xy Hb have significantly different (p 0.05) a*values than m et Hb, higher at week 0 and lower at week 24. At pH 6.8 and 3.7C (Table 52) CO Hb and o xy Hb did not have significantly different (p 0.05) a*values on days 0, 2, and 4 but both were significantly different from m et Hb. Both CO and o xy Hb had higher a*values at pH 6.8 compared to pH 6.3. Met and CO Hb did not differ in a*values on day 6 and 8, but o xy and CO Hb differed significantly (p 0.05) with oxy Hb having higher a* values. At 25C and pH 6.8 (Table 5 3), the three forms of Hb differed significantly (p 0.05) Throughout the 24 weeks CO Hb had the highest a*values of all forms, except for week 0 where it did not differ

PAGE 126

126 significantly from o xy Hb. Oxy Hb had significantly (p 0.05) higher a* value than m et Hb on weeks 0, 4, 8, 20, and 24. At pH 7.3 and 3.7C (Table 52), samples with CO and o xy Hb differed significantly (p 0.05) from m et Hb but did not differ significantly from each other. Throughout the 8 day storage period, CO Hb differed significantly from m et Hb but only differed significantly (p 0.05) with o xy Hb on day 8 where its a*values were lower. At 25C and pH 7.3 (Table 53), both o xy and CO Hb had higher a* values than at the other pH values tested, and also demonstrated more color stability On week 0, 16, 20, and 24, CO and o xy Hb did not significantly (p 0.05) differ in a* values, but were significantly different at weeks 4, 8, and 12, where CO Hb samples had higher a*values. Both o xy and CO Hb differed significantly from m et Hb throughout the 24 week storage period. The results for the change in the L*value (lightness) and b* value (yellowness) in washed tilapia muscle containing different forms of Hb at pH 6.3, 6.8, and 7.3 during 3.7C and 25 C storage are presented in Appendix B. Heme Group Autoxidation Results at 3.7C showed that as pH levels increased, more o xy Hb was present in all samples (Figure 5 21). There were also significant (p 0.05) differences for the three forms of Hb. At all pH levels tested, CO Hb had a greater amount of o xy Hb compared to the other forms. All pH forms tested showed a decline in % o xy Hb during storage, but had the most stability to autoxidation at pH 7.3. Samples stored at 25C (Figure 522) demonstrated a very similar trend in % o xy Hb as the samples at 3.7C Level of o xy Hb decreased over time for all samples and was more stable as pH increased. Under all three pH conditions, CO Hb had higher % o xy Hb. Oxy Hb had significantly (p 0.05) less % o xy Hb than CO Hb but significantly more than m et Hb.

PAGE 127

127 As expected, the level of m et Hb at 3.7C was highest for samples containing m et Hb (Figure 5 23). Met Hb samples showed a slight increase in % m et Hb over the storage period at pH 6.3 and 6.8, while at pH 7.3 they remained stable. Both CO and o xy Hb showed a significant (p 0.05) increase in % m et Hb over the 8 day storage period, regardless of pH, however levels of m et Hb were lower as pH was higher. Very similar trends were seen in samples stored at 25C (Figure 524) with % m et Hb at all three pH levels being highest for m et Hb samples, and CO Hb having the least % m et Hb. % m et Hb significantly (p 0.05) increased for CO and o xy Hb during the 24 weeks of frozen storage, while m et Hb samples remained stable. Since samples were kept aerobically relatively low levels of d eoxy Hb were seen for all samples at either storage temperatures. At 3.7C and pH 6.3, there were no significant (p 0.05) diff erences between o xy CO and m et Hb in % d eoxy Hb (Figure 5 25). At pH 6.8 and 7.3, m et Hb had significantly more % d eoxy Hb than o xy and CO Hb, while CO Hb samples had the lowest levels. At both pH 6.3 and pH 6.8, significant increases in % d eoxy were seen for CO and o xy Hb over time. At pH 7.3, m et Hb samples showed significant (p 0.05) increases in % d eoxy Hb throughout the storage period. At 25C and pH 6.3 there were no significant differences found between m et Hb and CO Hb in % d eoxyHb (Figure 526). Oxy Hb had significantly (p 0.05) greater % d eoxy Hb than m et and CO Hb. At pH 6.8 and 7.3, CO and o xy Hb did not differ significantly in % d eoxy Hb. M et Hb had greater % d eoxy Hb than both CO and o xy Hb at pH 6.8 and 7.3. Discussion Ther e is a sufficient body of evidence accumulated to date suggesting that lowering pH below neutrality may enhance prooxidative activities of Hbs present in

PAGE 128

128 muscle foods including fish ( 10, 22, 107) This phenomenon is attributed to increased rate of deoxygenation of Hbs in acidic environment ( e.g., the post mortem system) that promotes Hb autoxidation and subsequently, lipid oxidation, leading to muscle food deterioration and loss of quality. The change in oxygen affinity of Hb with changes in pH was first discovered and described at the beginning of 20th century by Bohr, and in later years was established as a concept named Bohr Effect. Extensive research has been done since to determine possible mechanism(s) responsible for this effect. It is believed that alterations in Hb structure at different pH levels may, at least partially, be accountable for the Bohr Effect. It was generally proposed that with lowering of pH, Hb unfolds that accelerates release of hemin. Rapid release of hemin in turn can stimulate lipid peroxidation promoting food spoilage ( 58 ) In the current research study we investigated effects of various physiological pH levels on prooxidative activities of different types of Hb ( o xy CO and m et Hb) under two storage condit ions (3.7C and 25C). Three pH levels (6.3, 6.8, and 7.3) were chosen to address common levels of pH seen in post mortem fish muscle systems and fish products. A t pH 6.3, rapid oxidation of all types of Hb was observed. As % o xy Hb was depleated at day 8 (Figure 521a), formation of m et Hb (Figure 5 23a) and % deoxygenation (Figure 525a) increased significantly reach ing virtually full oxidation by the end of experiment (day 8) ; this is in accordance with the Bohr Effect This is further exemplified by t he negative correlations between % o xy Hb and % d eoxy Hb (r= 0.85, Appendix C Table C 1 ). Comparable results were obtained by previous investigators for different Hbs ( 22 ) As expected, the rate of autoxidation of CO Hb was notably slower

PAGE 129

129 than that of other forms of Hb, suggesting that the CO molecule stabilized the heme group by replacing oxygenated Hb with CO Hb ( 69, 70) During autoxidation, oxygen is rele ased from oxy Hb to form the superoxide anion radical, O2 et Hb which is converted to hydrogen peroxide which increases the ability of heme proteins to promote lipid oxidation. Both d eoxy Hb(HbFe(II)) and oxy Hb (HbO2Fe(II)) exist in the reduced state and thus are sus ceptible to autoxidation. These two reduced forms can oxidize to the m et Hb(Hb Fe(III)). MetHb in the presence of H2O2, can also be converted to the very p ro oxidative forms, perferryl Hb (Hb Fe(IV)=O) and ferryl Hb (Hb Fe(IV)= O) ( 5 ) The autoxidation and the resultant brown coloring can be prevented by treatment of the fish muscle with CO which has a greater affini ty for the Fe(II) binding site, forming CO Hb. When CO is bound to heme in Hb, it gives stability to the protein, and the protein will resist autoxidation on heating, freezing, and thawing ( 69, 70) Autoxidation was also highly influenced by pH, with higher pH resulting in less autoxidation. These results are consistent with the findings of Richards and Hultin ( 4 ) who investigated the effect of pH (7.6, 7.2, and 6.0) on lipid oxidation using trout Hb as a catalyst in a washed minced cod muscle system The authors found that at lower pH va lues formation of m et Hb occurred more rapidly and the level of d eoxy Hb increased sharply. Lowering the pH below neutrality decreas es the oxygenation of Hb ( 43 ) .The Bohr effect can explain the high levels of d eoxy Hb at postmortem due to the low pH levels ( 43) When pH levels are decreased below 6.5, further deoxygenation occurs ( Root effect ). This further suggests that d eoxy Hb may play the role of a catalyst in lipid oxidation ( 43) Undeland and others ( 22) studied the effect of pH in a washed, minced

PAGE 130

130 cod model system and found that the higher rate of oxidation at pH 6.0 corresponded to a greater formation of d eoxy an d m et H b. Both lipid hydroperoxide and TBARS values followed a similar trend as a function of pH, where more rapid and in general higher levels of oxidation were seen as pH w as lower (Figures 51 5 12). H igher levels of % m et Hb and lower levels of % o xy Hb were seen as pH was lower (Figures 521 5 26) suggesting a connection between autoxidation and lipid oxidation. Hydrogen peroxide formed during autoxidation of reduced Hbs, increases the ability of heme proteins to promote lipid oxidation ( 5 ) Richards and Dittman ( 1 ) found a high autoxidation rate in perch Hb despite the low d eoxy Hb content. This could be explained by the amino acid sequences near the heme crevice. It has been found that disrupting the hydrogen bonding network of His97 created easier accessibility of H2O into the heme crevice ( 5, 45 ) This would then accelerate the formation of m et Hb which would also increase lipid oxidation. It was of interest to see that at 3.7C and pH 6.3, o xy Hb was the most prooxidative, while m et Hb was the most prooxidative at pH 6.8 and 7.3. Richards and Hultin ( 4 ) reported that in the presence of lipid peroxide, reduced Hbs (oxy/deoxy Hbs) produced high levels of lipid peroxide formation, but m et Hb caused litt le peroxidation. Met Hb cannot autoxidize but the reduced Hbs form superoxide anion radicals during autoxidation, which can form hydrogen peroxide. The hydrogen peroxide will then activate m et Hb ( 39) If m et Hb is the initial reactant there is no source of oxygen to form hydrogen peroxide. In addition, reduced Hbs but not m et Hb can produce hydroxyl radicals ( 40 ) Met Hb forms hemichromes which are not considered catalysts of lip id peroxidation ( 41 )

PAGE 131

131 CO Hb exhibited a similar lipid oxidation pattern as m et Hb at ph 6.8, but showed significantly slower rate for TBARS formation (Figure 57 and 5 1). Of note, by day 8 TBARS scores and carbonyl values (Figure 513) were similar for all forms of Hb, while lipid hydroperoxides formation (Figure 5 1) was hi ghest for o xy and lowest for CO Hb. The production of carbonyls actually diminished in CO Hb samples by day 8 of the experiment (Figure 513). That confirms the hypothesis that treatment with CO may to some extent stabilize Hb structure, thus retarding deterioration of muscle food even under unfavorable acidic conditions. From the earlier reports it is evident that Hb has an ability to stimulate lipid oxidation at lower (more acidic pH), and this ability is in correlation with increasing in deoxy Hb for mation ( 55) These results are in agreement with present findings. Frozen storage ( 25C) in general led to lower TBARS (Figure 58) and lipid hydroperoxide formation (F igure 5 2) in all forms of Hb. However under the same conditions enhanced carbonyl formation was observed for o xy and m et Hb (Figure 514). The latter finding may be attributed to muscle protein denaturation due to a low temperature. The lag time prior to development of TBARS (Figure 5 9) and lipid hydroperoxide formation (Figure 53) for all Hb forms at pH 6.8 and 3.7C increased compared to pH 6.3. However, CO Hb was found to have the most reduced lag time for lipid hydroperoxidation of all forms (Fi gure 5 3). These results are in a greement with autoxidation (Figure 5 21b) and deoxygenation (Fig 525b) observed for the Hbs. The correlations between lipid hydroperoxides and TBARS formation with % o xy % d eoxy and % m et Hb, also reflect that as % d eoxy and % m et Hb increase d lipid oxidation increased ( Appendix C, Table C 1). Although the increase in pH from 6.3 to 6.8 led to

PAGE 132

132 lower levels of carbonyls for m et and CO Hb it did not for o xy Hb (Figure 515). It is speculated that oxy Hb (reduced form of Hb) is a main promoter of lipi d oxidation. Richards and Hultin ( 4 ) reported that reduced Hbs (oxy/deoxy Hbs) produced high levels of lipid peroxide formation, whereas m et Hb caused little peroxidation. These findings might indicate that oxy Hb could also play an important role as a catalyst in promoting protein oxidation. Eymard and others ( 35 ) investigat ed the link between lipid oxidation and protein oxidation during processing and storage of horse mackerel using fish minces with differences in lipid and protein fractions and different oxidative levels. The authors concluded that lipid and protein oxidation developed simultaneously but it was difficult to determine how they are linked. Lipid and protein oxidation share the same catalysts, and they can develop independently of one another, or in parallel, or they can interact with each other ( 35) Lower levels of oxidation at pH 6.8 compared to pH 6.3 were also seen during frozen storage (Figures 5 4 and 5 8) regardless of Hb form, and also correlated (Appendic C, Table C 2) with a drop in % o xy Hb and increase in % m et Hb (Figure 522b and 524b). Formation of carbonyl groups of all samples (Figure 516) was in fact enhanced at 25C (Figure 5 16). There was not a direct relationship seen with lipid oxidati on levels and carbonyl levels. This implies that some other mechanisms might be present in fish muscle system under freezing conditions which leads to more protein oxidation, for example increased protein denaturation and cross linking due to the activity of the t rimethylamineN o xide (TMAO) or other enzymes such as l ipases and phospholipases These enzymes can cause denaturation of fish muscle during froze n storage ( 98) Free fatty acides released by Lipases and phospholipases activity can

PAGE 133

133 react with proteins. The reactio n of oxidized lipids with proteins may result in toughening of fish muscle during frozen storage ( 98) A very significant retardation of lipid oxidation (Figure 5 5, 5 6, 5 11 and 512) was seen for all Hb forms at pH 7.3, regardless of storage temperature. These findings were correlated (Appendix C, Table C 1 and C 2) with a decreased rate of autoxidation at pH 7.3 (Figure 521, 522 ). Carbonyl levels still increased at pH 7.3, unlike lipid oxidation (Figure 517). The fact that carbonyl formation was not affected by pH, unlike the Hb catalysts, implies that there m ay be some other mechanism(s) involved in protein oxidation in the WMTM system. It may be speculated that manual manipulations with tilapia muscles (such as mincing) induces certain protein denaturation that predispose to enhanced protein oxidation independently of pH levels. Eymard and others ( 35 ) investigated the effect of mincing and washing horse mackerel muscle on protein ox idation and found that at time zero, myosin and other high molecular weight proteins w ere already oxidized in all products including the unwashed mince. It may be further proposed that higher rate of o xy Hb autoxidation may to a degree promote protein oxid ation. Detailed mechanism(s) of protein oxidation should be elucidated in future studies. The effects of the three pH levels on the rate of hemin release in tilapia met Hb samples were tested at both 4C and 25C (Table 51). This was of interest since hem e release from Hb is believed to play an important role in the pro oxidative activity of Hb. The rate of hemin loss in trout Hb under both temperatures at pH 6.3 was also measured to provide a relevant comparison. The results show that lower pH may acceler ate hemin loss, which is in agreement with the hypothesis that acidic

PAGE 134

134 environment may affect the structure of Hb promoting its rapid unfolding that in turn triggers rapid hemin loss and increased lipid oxidation ( 38, 58 ) As the pH decreases, auto oxidation and hemin loss increase. Aranda and others ( 60) suggested that the reason for the pH dependence may involve the protonation of distal His resulting in disruption of the bound O2. This disruption might cause a rotation of the side chain. Another reason is the dissociation of OOH by the bound O2, which result in the formation of met Hb. It could also be due to the interruption of the electrostatic forces with amino acids at E10 and CD3 positions ( 60 ) A faster hemin loss rate was noted at 25C compared to 4C. Moreover, at pH 6.3 trout Hb released hemin more rapidly than tilapia Hb, suggesting a stronger hemeglobin interation for tilapia Hb (Table 51). It may indicate that different species may respond in different ways to changes in acidity/alkalinity. Precise mechanism(s) of the observed events should be determined in future more det ailed investigations. Deterioration of desirable red color of fish samples (Figure 5 27) represented as a*value into undesirable brown color was followed throughout the storage period The results show that the red color was highly influenced by pH, being mo re unstable as pH levels decreased. Treatment of fish muscle with CO improved the stability of red color in all CO Hb samples at the three pH levels tested (Table 52 and 53) compared to oxy and m et Hb This suggest s that the autoxidation and the res ultant brown coloring can be prevented by treatment of the fish muscle with CO forming CO Hb, resulting in a cherry red color that is stable over longer periods ( 69, 70) The stability of the red color may als o be a result of r ep lacing oxygenated Hb CO Hb that may offer some protection against lipid oxidation ( 69, 70 ) Moreover, under freezer conditions a*value in CO Hb

PAGE 135

135 samples was found to increase significantly (p 0.05) after one week of storage (Table 5 3). The pH 7.3, h owever, delayed red color deterioration in both o xy and CO Hbs at both tested temperatures (Table 52, 5 3) compared to m et Hb. It has been reported that under alkaline pH conditions, deoxygenation of Hbs is maintained in the reduced state, thus slowing down the autoxidation and deterioration of fish muscle system. This might, in turn, delay formaton of undesirable color ( 4, 81) Conclusion Low pH increased the susceptibil ity of the washed minced tilapia muscle to oxidation and promoted loss of hemin. Fish muscle treated with CO may have better stability to oxidation conditions, thus resulting in red color stability. Treatement with CO delayed the onset of lipid and protein oxidation during cold and freezer storage temperatures.

PAGE 136

136 Table 5 1 Hemin loss rate for met Hb (tilapia and trout) Hb pH T (C) K dissociation /h 1 Tilapia 6.3 25 2.870.32 6.8 25 0.540.07 7.3 25 0.170.01 6.3 4 0.590.01 6.8 4 .020.001 7.3 4 0.00360.0006 Trout 6.3 25 5.550.31 6.3 4 0.760.07 Values are means Standa rd Deviations for all samples (n= 3 ).

PAGE 137

137 Table 5 2 Changes in a* value in washed tilapia muscle containing different forms of Hb at pH 6.3, 6.8, and 7.3 at 3.7C averaged across Hb concentration and NaCl added. a* value pH 6.3 Storage time (day) Oxy Hb b CO Hb a Met Hb c 0 16.33.4 19.63.8 9.11.5 2 6.93.4 16.22.9 11.01.5 4 3.00.9 8.74.6 6.92.3 6 1.60.7 4.62.0 3.81.2 8 1.40.9 4.01.6 2.40.9 a* value pH 6.8 Storage time (day) Oxy Hb b CO Hb a Met Hb c 0 21.34.3 22.94.0 8.41.4 2 21.04.3 21.84.0 11.21.8 4 20.13.9 19.03.9 9.21.7 6 16.28.0 8.13.2 6.52.3 8 13.58.6 8.33.5 5.82.3 a* value pH 7.3 Storage time (day) Oxy Hb a CO Hb a Met Hb b 0 22.24.0 23.23.7 6.71.4 2 21.94.1 24.34.2 8.52.0 4 22.33.9 23.94.1 8.12.4 6 22.64.1 21.35.8 8.12.4 8 21.33.5 17.17.5 6.22.9 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n=18). Letters within the same rows indicate statistically (< 0.05) significant differences separated by Tukeys HSD.

PAGE 138

138 Table 5 3 Changes in a* value in washed tilapia muscle containing different forms of H b at pH 6.3, 6.8, and 7.3 at 25C averaged across Hb concentration and NaCl added. a* value pH 6.3 Storage time (day) Oxy Hb b CO Hb a Met Hb b 0 16.33.4 19.63.8 9.11.5 4 11.63.5 21.64.6 13.02.1 8 10.53.7 19.55.5 10.42.0 12 9.03.7 17.24.5 9.62.1 16 8.73.2 15.85.3 9.22.5 20 6.02.6 14.54.4 8.42.1 24 2.92.9 12.64.9 7.61.9 a* value pH 6.8 Storage time (day) Oxy Hb b CO Hb a Met Hb c 0 21.34.3 22.94.0 8.41.4 4 19.24.5 25.84.3 12.72.8 8 16.05.2 23.25.9 10.62.4 12 10.55.7 21.55.5 10.92.8 16 12.45.4 22.16.0 10.62.8 20 14.75.0 19.75.7 9.82.9 24 13.84.5 19.36.4 9.43.3 a* value pH 7.3 Storage time (day) Oxy Hb b CO Hb a Met Hb c 0 22.24.0 23.23.7 6.71.4 4 20.94.9 27.44.2 11.03.0 8 20.85.0 26.34.1 12.22.6 12 21.05.5 25.54.1 10.42.3 16 22.55.6 24.14.1 10.92.6 20 24.85.2 23.53.9 11.12.7 24 23.65.1 23.23.8 11.43.1 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n=18). Letters within the same rows indicate statistically (<0.05) significant differences separated by Tukeys HSD.

PAGE 139

139 Figure 51 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at pH 6.3 at 3.7 C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 52 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at pH 6.3 at 25C averaged across Hb conc entration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. -100 100 300 500 700 900 1100 0 2 4 6 8Lipid hydroperoxide mol/kg muscleDay (b) CO Hb (b) Met Hb (a) Oxy Hb -100 100 300 500 700 900 1100 0 4 8 12 16 20 24Lipid hydroperoxide mol/kg muscleWeek (b) CO -Hb (a) Met -Hb (a) Oxy-Hb

PAGE 140

140 Figure 53 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at pH 6. 8 at 3.7C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 54 Lipid hydroperoxide values in washed tilapia muscle containi ng different forms of Hb at pH 6. 8 at 25C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. -100 100 300 500 700 900 1100 0 2 4 6 8Lipid hydroperoxide mol/kg muscleDay (c) CO Hb (a) Met Hb (b) Oxy Hb -100 100 300 500 700 900 1100 0 4 8 12 16 20 24Lipid hydroperoxide mol/kg muscleWeek (b) CO Hb (a) Met Hb (b) Oxy Hb

PAGE 141

141 Figure 55 Lipid hydroperoxide valu es in washed tilapia muscle containing different forms of Hb at pH 7 .3 at 3.7C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Fig ure 56 Lipid hydroperoxide values in washed tilapia muscle containing different forms of Hb at pH 7 .3 at 25C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. -100 100 300 500 700 900 1100 0 2 4 6 8Lipid hydroperoxide mol/kg muscleDay (b) CO Hb (a) Met Hb (b) Oxy Hb -100 100 300 500 700 900 1100 0 4 8 12 16 20 24Lipid hydroperoxide mol/kg muscleWeek (a) CO Hb (a) Met Hb (a) Oxy Hb

PAGE 142

142 Figure 57 TBARS values in washed tilapia muscle containing different forms of Hb at pH 6.3 at 3.7C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 58 TBARS values in washed tilapia muscle containing differ ent forms of Hb at pH 6.3 at 25C averaged across Hb concentration and NaCl added. Letters within the legend of the figure ind icate statistically (<0.05) significant differences separated by Tukeys HSD. -10 0 10 20 30 40 50 0 2 4 6 8mol TBARS/kg muscleDay (b) CO -Hb (a) Met -Hb (a) Oxy-Hb -10 0 10 20 30 40 50 0 4 8 12 16 20 24mol TBARS/kg muscleWeeks (b) CO HB (b) Met Hb (a) Oxy Hb

PAGE 143

143 Figure 59 TBARS values in washed tilapia muscle containing different forms of Hb at pH 6.8 at 3.7C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 510. TBARS values in washed tilapia muscle containing different forms of Hb at pH 6.8 at 25C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. -10 0 10 20 30 40 50 0 2 4 6 8mol TBARS/kg muscleDay (b) CO -Hb (a) Met -Hb (b) Oxy-Hb -10 0 10 20 30 40 50 0 4 8 12 16 20 24mol TBARS/kg muscleWeeks (b) CO Hb (b) Met Hb (a) Oxy Hb

PAGE 144

144 Figure 511. TBARS values in washed tilapia muscle containing different forms of Hb at pH 7.3 at 3.7C averaged across Hb concent ration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 51 2 TBARS values in washed tilapia muscle containing different forms of Hb at pH 7.3 at 25C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. -10 0 10 20 30 40 50 0 2 4 6 8mol TBARS/kg muscleDay (b) CO -Hb (a) Met -Hb (b) Oxy-Hb -10 0 10 20 30 40 50 0 4 8 12 16 20 24mol TBARS/kg muscleWeeks (b) CO Hb (b) Met Hb (a) Oxy Hb

PAGE 145

145 Figure 513. Carbonyl values in washed tilapia muscle containing differ ent forms of Hb at pH 6.3 at 3.7C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 514. Carbonyl values in washed tilapia muscle containing different forms of Hb at pH 6.3 at 25C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 0 2 4 6 8nM carbonyls/mgDay (a) CO Hb (b) Met Hb (b) Oxy Hb 0 10 20 30 40 50 60 70 80 90 0 4 8 12 16 20 24nM carbonyls/mg muscleWeek (c) CO Hb (b) Met Hb (a) Oxy Hb

PAGE 146

146 Figure 515. Carbonyl values in washed tilapia muscle containing different forms of Hb at pH 6. 8 at 3.7C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 5 16. Carbonyl values in washed tilapia muscle containing different forms of Hb at pH 6. 8 at 25C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 0 2 4 6 8nM carbonyls/mgDay (c) CO Hb (b) Met Hb (a) Oxy Hb 0 10 20 30 40 50 60 70 80 90 0 4 8 12 16 20 24nM carbonyls/mg muscleWeek (b) CO -Hb (a) Met -Hb (a) Oxy-Hb

PAGE 147

147 Figure 517. Carbonyl values in washed tilapia muscle containing different forms of Hb at pH 7 .3 at 3.7C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 518. Carbonyl values in washed tilapia muscle containing different forms of Hb at pH 7 .3 at 25C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistical ly (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 0 2 4 6 8nM carbonyls/mgDay (b) CO Hb (c) Met Hb (a) Oxy Hb 0 10 20 30 40 50 60 70 80 90 0 4 8 12 16 20 24nM carbonyls/mg muscleWeek (a) CO Hb (a) Met Hb (a) Oxy Hb

PAGE 148

148 Figure 519. %CO released during 3.7C storage at different pH in washed tilapia muscle containing CO Hb, averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 5 20. %CO released during 25C storage at different pH in washed tilapia muscle containing CO Hb, averaged across Hb concentration and NaCl added. Letters wit hin the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. -5 0 5 10 15 20 25 30 0 2 4 6 8 % CODay (a) pH6.3 (b) pH6.8 (c) pH7.3 0 10 20 30 40 50 60 0 4 8 12 16 20 24% COWeek (a) pH6.3 (b) pH6.8 (c) pH7.3

PAGE 149

149 Figure 521. %Oxy Hb in washed tilapia muscle containing different forms of Hb at pH a) 6.3, b) 6.8, and c) 7.3 at 3.7C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Oxy -HbDay (c) Met Hb (a) CO Hb (b) Oxy Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Oxy -HbDay (c) Met Hb (a) CO Hb (b) Oxy Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Oxy -HbDay (c) Met Hb (a) CO Hb (b) Oxy Hb (c)

PAGE 150

150 Figure 522. %Oxy Hb in washed tilapia muscle containing different forms of Hb at pH a) 6.3, b) 6.8, and c) 7.3 at 25C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Oxy -HbWeek (c) Met -Hb (a) CO -Hb (b) Oxy-Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Oxy -HbWeek (c) Met Hb) (a) CO Hb (b) Oxy Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Oxy -HbWeek (c) Met Hb (a) CO Hb (b) Oxy Hb (c)

PAGE 151

151 Figure 523. %Met Hb in washed tilapia m uscle containing different forms of Hb at pH a) 6.3, b) 6.8, and c) 7.3 at 3.7C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Met -HbDay (a) Met Hb (c) CO Hb (b) Oxy Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8% Met -HbDay (a) Met Hb (c) CO Hb (b) Oxy Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Met -HbDay (a) Met -Hb (c) CO -Hb (b) Oxy-Hb (c)

PAGE 152

152 Figure 524. %Met Hb in washed tilapia muscle containing different forms of Hb at pH a) 6.3, b) 6.8, and c) 7.3 at 25C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Met -HbWeek (a) Met Hb (c) CO Hb (b) Oxy Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Met -HbWeek (a) Met Hb (c) CO Hb (b) Oxy Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Met -HbWeek (a) Met Hb (c) CO Hb (b) Oxy Hb (c)

PAGE 153

153 Figure 525. %Deoxy Hb in washed tilapia muscle containing different forms of Hb at pH a) 6.3, b) 6.8, and c) 7.3 at 3.7C averaged across Hb concentration and NaCl added. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Deoxy -HbDay (a) Met Hb (a) CO Hb (a) Oxy Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8% Deoxy -HbDay (a) Met Hb (c) CO Hb (b) Oxy Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Deoxy -HbDay (a) Met -Hb (c) CO -Hb (b) Oxy-Hb (c)

PAGE 154

154 Figure 526. %Deoxy Hb in washed tilapia muscle containing different forms of Hb at pH a) 6.3, b) 6.8, and c) 7.3 at 25C averaged across Hb concentration and NaCl added Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Deoxy -HbWeek (b) Met -Hb (b) CO -Hb (a) Oxy-Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Deoxy -HbWeek (a) Met -Hb (b) CO -Hb (b) Oxy-Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Deoxy -HbWeek (a) Met Hb (b) CO Hb (b) Oxy Hb (c)

PAGE 155

155 Figure 527. Images of wa shed tilapia muscle containing oxy CO and m et Hb at a concent ration of 12mol Hb/kg muscle. a) pH 6.3, Day 0, 3, and 6 b) pH 6.8, Day 0, 3, and 6, and c) pH 7.3, Day 0, 3, and 6, Obtained during storage for six days. pH 6.3 a*Value Standard CO Hb Oxy Hb Met Hb pH 6.8 a*Value Standard CO Hb Oxy Hb Met Hb pH 7.3 a*Value Standard CO Hb Oxy Hb Met Hb

PAGE 156

156 CHAPTER 6 THE ROL E OF SODIUM CLORIDE ON THE PRO OXIDATIVE ACTIVITY O F OXY, CO AND MET HEMOGLOBIN ON IN A W ASHED MINCED TILAPIA WASHED SYSTEM Introduction Sodium chloride ions are found in all living cells and can act as a prooxidant of muscle lipid peroxidation ( 82 ) Kanner and others ( 83 ) found that presence of NaCl initiates reactions producing superoxide anion radical (O2 -) which results in formation of hydroxyl radical. The prooxidative effect and the increase of lipid peroxidation by NaCl in model systems were studied and the elevation of free iron in tissues was ascribed to NaCl ( 82) Wallace and others ( 84 ) found that the stability of oxy Hb and its oxidation to m et Hb may have been affected by the presence of NaCl by shifting ferrous ions from interaction with oxygen to reaction with hydroperoxides and dec omposing these compounds to free radicals, accelerating the peroxidation process. Harel ( 85 ) investigated the effect of NaCl on autoxidation of ferrous and cuprous ions in the presence of ascorbic acid and iron chelators. The generation of hydroxy radicals by ascorbic acid and metal ion's was inhibited by NaCl. NaCl also inhibited the oxidation of ascorbic acid by preventing the interaction of Fe or Cu with oxygen. The chloride anion interacts with the iron ion inhibiting the ferrous ion oxidation. Calcium chloride, magnesium chloride, and Lithium chloride showed similar results. Wallace and others ( 84) demonstrated that NaCl accelerates the decomposition of oxy Hb to m et Hb Harel ( 85 ) postulating that NaCl prevents or disturbs the interaction between heme iron and oxygen in the same way.

PAGE 157

157 Materials and Metho ds The following methods have been described in detail in chapter 3: Preparat ion of Washed Minced Tilapia Muscle (MTWM), Collection of Fish Blood, Preparation of Hemolysa te, Quantification of Hb Levels in Hemolysate, Oxy CO and Met Hb Preparation, Sample Preparation: Addition of Hb and NaCl, Determination of Peroxide Value (PV) Determination of Thiobarbituric Acid Reactive Substances (TBARS) Determination of Carbonyl Groups Heme Group Autoxidation, and Color Analysis. The Gas Chromatography (GC) Method was described in Chapter 4. Results Lipid Oxidation Analysis Lipid hydroperoxide results obtained at 3.7C with no added salt demonstrated, as shown previously, that oxy Hb was significantly more proox idative than CO Hb and met Hb. Oxy Hb did not only lead to significant oxidation two days before the other forms but also resulted in higher maximum hydr operoxide values (Figure 61). CO Hb and m et Hb led to a similar development of oxidation, with CO Hb having significantly lower TB ARS than m et H b on day 8 only. At 25C storage over 24 weeks (Figure 62) there was little difference between the three forms, although CO Hb at week 2024 was significantly less pro oxidative than oxy Hb. Lipid hydroperoxides formed faster in the presence of 150 mM NaCl at 3.7C than in the absence of salt (Figure 6 3). Oxy Hb samples were already significantly oxidized on day 2, and hydroperoxide values were significantly (p 0.05) higher at day 2 and 4 compared to CO Hb and met Hb. Samples with CO and m et Hb did not differ significantly (p in lipid hydroperoxides throughout storage at 3.7C. At 25C, no

PAGE 158

158 significant differences were found between the three forms of Hb, except CO Hb was significantly (p less pro oxidative than met and oxy Hb at week 24 (Figure 64). At 3.7 C storage in the presence of 450mM NaCl a similar development of lipid hydroperoxides was seen for the three forms of Hb as in the presence of 150 mM NaCl (Figure 65). CO Hb, however, gave significantly (p l ower hyd roperoxide values than oxy and m et Hb on day 4 and 6. At 25C storage overall no significant differences were noted between the three forms of Hb, except at week 4 where oxy Hb was significantly more prooxidative than met Hb and week 24 were oxy Hb had significantly higher values than CO and m et Hb (Figure 66). More lipid hydroperoxide formation was seen at 25C in the presence of 450 mM NaCl, compared to the lower salt levels tested. At 3.7C storage with no added salt, washed tilapia muscle containi ng m et Hb developed higher TB ARS values sooner than CO and oxy Hb, which did not differ from each other except on day 8 when CO Hb had s ignificantly higher TBARS than oxy Hb (Figure 67). Met Hb differed significantly from CO Hb only on day 6, being the m ore prooxidative. Met Hb differed significantly from oxy Hb day 6 and 8. At 25C, oxy Hb led to the most oxidation, having significantly higher TB ARS than CO Hb from week 1624 m et Hb from week 1224 (Figure 6 8). CO and m et Hb did not differ significantly during frozen storage. A sooner onset of lipid oxidation (TBARS) was seen for all forms in the presence of 150 mM NaCl 3.7C compared to samples with no added salt (Figure 69), which is in agreement with the lipid hydroperoxide data. Oxy Hb samples peaked in TBARS values at day 0, having significantly high er values than CO and m et Hb. However, by day 6

PAGE 159

159 o xy Hb had significantly signific antly lower TBARS than CO and m et Hb. Although CO Hb lagged in oxidation behind m et Hb, the two forms did not differ significantly from each other. Met Hb did however lead to the highest level of TBARS f ormed of all three forms tested Very similar trends in the formation of TBARS was seen at 25C in the presence of 150 mM NaCl compared to no added salt, although oxidation values were a bit higher in the system with salt (Figure 610). CO and m et Hb did not differ significantly in pro oxidative activity, while oxy Hb was significantly more pro oxidative than CO Hb at week 12 and both CO and m et Hb from weeks 2024. TBARS development in washed tilapia muscle at 3.7C with 450 mM added NaCl (Figure 611) was similar to samples with 150mM NaCl. Although oxy Hb had the highest level of TBARS at day 2, it did not differ significantly from CO Hb, but both differed from m et Hb which was significantly more prooxidative on day 6 only. Higher TBARS values were formed in samples with 450 mM NaCl at 25C compared to the other salt levels tested (Figure 612) Samples with CO and m et Hb did not differ significantly in TBARS v a lues except for week 20 where m et Hb was significantly more prooxidative than CO Hb. Oxy Hb was signifi cantly more prooxidative than met and o xy Hb from weeks 2024. Protein Oxidation Analysis Formation of carbonyls as a function of Hb type was similar at all added salt levels tested at 3.7C, where oxy Hb led to an increase in carbonyl groups formed and also gave the highest values (Figures 613, 6 15 a nd 617). In the absence of added salt o xy Hb had significant ly higher carbonyl values than met and oxy Hb on days 26 (Figure 613). CO Hb had higher carbonyl values than m et Hb only on day 0. In the presence of 150mM NaCl found no significant differences in carbonyl values were found

PAGE 160

160 between CO and m et Hb. Oxy Hb had however significantly higher carbonyl values than CO and m et Hb from days 26 and significantly higher values than CO Hb on day 8. Oxy Hb was significantly more prooxidative with res pect to protein oxidation than met and oxy Hb in the presence of 450mM NaCl (Figure 617). CO Hb and m et H b differed significantly on day 2 only, CO Hb giving higher carbonyl values. Similar trends in formation of carbonyl values were observed at 25C regardless of salt concentration of the washed tilapia muscle system with added Hb (Figures 6 14, 6 16 and 618). Oxy Hb was significantly m ore prooxidative than CO and m et Hb from week 412 in the presence of no added salt (Figure 614). However, carbonyls declined after 16 weeks and from week 2024, ox yHb was less pro oxidative than the other Hbs. CO Hb had significantly (p lower carbonyl values than m et Hb from week 812, but significantly h igher values at week 24, where m et Hb had leve led off. In the presence of 150mM NaCl at 25C (Figure 6 16) o xy and m et Hb led to significantly (p 0.05) higher level of carbonyls than CO Hb However, it declined rapidly after week 12 and at week 24 had significantly lower values than both CO and m et Hb forms (Figure 616). Met Hb was more prooxidative than CO Hb at weeks 1220. In the precence of 450mM NaCl at 25C, samples with CO Hb were signifi cantly less pro oxidative than met and oxy Hb throughout storage, except for week 24 where it had significantly higher carbonyl values (Figure 6 18) Met Hb had sign ificantly lower carbonyls than oxy Hb at week 12, but significantly higher values week 24. CO Release The % of CO released during storage at 3.7C of washed tilapia muscle containing CO Hb with no added 150, and 450mM of NaCl (Figure 6 19) showed that a greater % of CO was released from the samples that contained 150mM NaCl. A greater

PAGE 161

161 % of CO was released on day 0 by samples containing all three forms of Hb, but the following two days the % of CO released was minimal but began to increase on day 6 and 8. At 25C storage (Figure 620), the greatest % of CO was released by the samples containing 0.0mM NaCl. The % of CO released increased with increasing weeks of storage. Color Analysis Changes in a*values (redness) during 3.7C storage of washed tilapia m uscle containing oxy CO and m et Hb and no added NaCl demonstrated significant differences for the three forms of Hb (Table 6 1). Oxy Hb had significantly higher a*values than m et Hb throughout fresh storage, and significantly higher a*values than CO Hb on days 68. CO Hb had significantly higher a*values than m et Hb on all days except day 8, where they were equal. At 25C, samples with CO Hb had significantly higher a*values than m et Hb throughout each week of stor age and significantly higher a*value t han oxy Hb at week 12 (Table 62). Oxy Hb had significantly higher a*values than m et H b at all weeks, except week 12. In the presence of 150mM NaCl at 3.7C, initial a*values were higher than in the absence of added salt (Tabl e 61). Samples with CO Hb and o xy Hb had both sign ificantly higher a*values than m et Hb throughout the storage period, while CO Hb had significantly higher a*values than oxy Hb on day 2 only. CO Hb had significantly higher a*values than m et Hb throughout the 24 weeks of storage at 2 5C and si gnificantly higher values than oxy Hb, except for week 0. Oxy Hb had significantly higher a*values than m et Hb at weeks 0, 4, 8, 16, and 20 (Table 62). Initial a*values of all samples increased even more in the precence of 450 mM NaCl than the other levels studied (Table 6 1). CO Hb had significantly h igher a* values

PAGE 162

162 than m et Hb throughout the 8 day storage period and sign ificantly higher a*values than o xy Hb on all days of storage except day 6. Oxy Hb had significantly higher a*values than m et H b on all days of storage. CO Hb had significantly higher a*values than both met Hb and oxy Hb throughout the 24 weeks of storage at 25C (Table 62). Oxy Hb had significantly higher a*values than m et Hb at weeks 0, 4, 12, and 20. Heme Group Autoxidation % o xy Hb declined over the 8 days at 3.7C for all Hb forms tested regardless of salt level in the model system. Oxy and CO Hb had significantly greater % oxy Hb than m et Hb throughout each day of storage at 3.7C, regardless of salt concentration tested (Figure 621). Both in the absence of added salt and in the presence of 150 mM NaCl, CO Hb had greater % o xy Hb on days 04 than oxy Hb (Figure 621a,b). However, in the presence of 450 mM NaCl, CO Hb had greater % oxy Hb than oxy Hb on days 0, 4 and 6. % o xy Hb values also declined over time at 25C, regardless of Hbform tested and salt concentration. Results at 25C were similar to the results at 3.7C where CO and oxy Hb had significantly greater % o xy Hb throughout the 24 weeks of storage than m et H b regardless of salt concentration (Figure 622). In the absence of added salt, CO Hb had significantly more % oxy Hb than oxy Hb from week 016, which was extended to week 20 in the presence of 150 mM and 450 mM NaCl. % m et Hb increased during storage at 3.7C for all samples, irrespective of salt levels tested (F igure 6 22). As expected, m et Hb had significantly greater % m et Hb than CO and oxy Hb at all salt levels tested. CO Hb had significantly less % m et Hb than oxy Hb day 0, 2, and 4 in the absence of salt and in the presence of 450 m M NaCl, but was only lower in % m et Hb at days 0 and 2 i n the presence of 150 mM NaCl

PAGE 163

163 During 25C storage there were minor changes in % met Hb for samples with m et Hb, while the two other forms increased in % me t H b over time (Figure 624). Met Hb had significantly higher % m et H b than CO and oxy Hb throughout the 24 weeks of storage, regardless of salt levels tested. CO Hb had significantly less % met Hb formed than oxy Hb from weeks 016, at no added salt and in the presence of 150 mM NaCl, while in the presence of 450 mM NaCl the time was reduced to 12 weeks. No significant differences were seen between samples with oxy and met Hb in terms of % d eoxy Hb at 3.7C, regardless of salt concentration (Figure 6 25). In the absence of salt and in the presence of 150 mM NaCl, CO Hb ha d less % deoxy Hb than m et Hb from days 04, and l ess than oxy Hb from days 24. In the presence of 450 mM NaCl the values were lower for CO Hb compared to m et Hb on days 0, 4 and 6, and d ays 0 and 4 compared to oxy Hb. During the entire 25C storage period % d eoxy Hb di d not differ significantly for oxy and m et Hb, regardless of salt concentration tested with the exception for samples with 450 mM NaCl where m et Hb had significantly great er % d eoxy Hb from weeks 1624 (Figure 26a). Samples with CO Hb were however affected somewhat by salt concentration. In the absence of salt, CO Hb had significantly less % deoxy Hb than m et Hb at weeks 0 and 8, and significantly less than o xy Hb a t week 8. In the presence of 150 mM NaCl, CO Hb had significantly less % d eoxy H b than oxy Hb from weeks 01 2, and significantly less than m et Hb at week 0 only. In the presence of 450 mM NaCl, CO Hb had significantly less % deoxy Hb than o xy Hb from weeks 012 and at week 24, while it was significantly less than m et Hb week 0.

PAGE 164

164 Discussion NaCl is a common additive used in the food industry for a variety of purposes, including inhibition of microbial growth and flavor enhancement ( 98) The extensive use of NaCl in food preparation and preservation dictates the necessity for thoroughly e xamining its effects on food quality. Conflicting data exists among previous research scientists investigating the influence of addition of salt on the oxidative properties of muscle food systems. While results of some observations suggest that NaCl may ex hibit prooxidative activities, accelerating lipid and protein oxidation in muscle food systems, including fish ( 82, 108) others report opposite effects ( 10, 109 ) which makes further exploration of this topic important. Ability of various s alt concentrations to induce changes in protein solubility and gelation was previously described ( 98) and may affect protein oxidation. In this experiment the effects of NaCl (150 and 450mM) concentrations on the propert ies of different forms of Hbs (oxy CO and m et Hbs) in a washed tilapia muscle system were investigated under refrigeration (3.7C) and frozen ( 25C) storage conditions. The results of the present investigation strongly support previous suggestions that higher concentration of NaCl may enhance prooxidative activities of Hb that is confirmed by TBARS scores and lipid peroxidation analysis ( 110 ) It appears that both rate and magnitude of Hb oxidation increases with increase in NaCl concentration regardless of storage conditions, although under frozen storage conditions the rate of oxidation was somewhat retarded (Figure 621, 622). The significant finding of this study was that the oxidative activities of CO Hb were overall less affected by increased concent rations of NaCl as compared to oxy and m et Hb (F igure 61 6 6). It was previously reported that NaC l may disrupt the stability of oxy -

PAGE 165

165 Hb, accelerating its deoxygenation and formation of m et Hb ( 84) From accumulated data it appears that the pro oxidative activit ies of NaCl may be due, in part, to its ability to hasten release of iron from heme ( 82) In particular, the chloride anion was implicated in Hb autoxidation, serving as a substrate for a chloroperoxidase in the pr esence of preformed hydrogen peroxide ( 84, 111) In addition, it was proposed that NaCl may have the ability to accelerate lipid peroxidation by shifting ironoxygen to ironhydroperoxide interactions. A significantly higher rate of lipid peroxidation and TB ARS formation was observed for oxy Hb at 3.7C with NaCl present compared to the other forms (Figure 67 6 12). It may be hypothesiz ed that rapid deoxygenation of oxy Hb under these conditions may be further augmented by the presence of NaCl, leading to increased oxidation. % oxy Hb decreased significantly for all samples during 8 days of storage (Figure 6 21a, b, and c). C O Hb autoxiation was least affect ed by NaCl additi on (Figure 623a, b, and c ) Moreover, it was also suggested that addition of NaCl may decrease the pH level and Hb affinity to oxygen which would contribute to increased Hb autoxidation and subsequently lipid peroxidation ( 112 ) In addition, other factors besides Hb autoxidation may influence lipid peroxidation under described conditions. Previous studies indicate that the addition of salt may suppress glutathione peroxidase activi ties in refrigerated ground pork muscle that may result in increase of lipid peroxidation ( 113 ) Another study on oxidative changes in salted herring found that the decreased levels of alphatocopherol in the ripening salted herring may further enhance oxidative processes ( 114) At 2 5 C however, the same rate of formation of TBARS and lipid hydroperoxides was observed for all forms of Hb, with oxy Hb producing signi ficantly higher levels of

PAGE 166

166 TBARS by the end of storage at any NaCl concentration (Figure 68, 6 10, and 6 12). CO Hb was least susceptible to oxidation without added NaCl and with the 150 mM of NaCl under frozen storage conditions (Figure 62, 6 4, 6 8 and 6 1 0 ) This implies that lower temperatures may slow the oxidation rate of muscle foods containing Hb independently of the presence or absence of NaCl and type of Hb. Protein oxidation (assessed by carbonyl formation) catalyzed by all forms of Hb was not significantly affected by addition of NaCl at either storage temperature (Figure 6 136 18). However, at 3.7C and 25C, the pattern of carbonyl production was different for each form of Hb and the days/weeks on which oxi dations occurred. Samples with oxy Hb did however result in the highes t levels of carbonyls formed. It is speculated that oxy Hb is a main promoter of lipid oxidation. Richards and Hultin ( 4 ) reported that oxy and deoxy Hb produced high levels of lipid peroxide f ormation, whereas m et Hb caused little peroxidation. These findings might indicate that oxy Hb could also play an important role as a catalyst in promoting protein oxidation. Eymard and others ( 35 ) investigated the link between lipid oxidation and protein oxidation during processing and storage of horse mackerel using fish minces with differences in lipid and protein fractions and different oxidative levels. The authors concluded that lipid and protein oxidation developed simultaneously but it was difficult to determine how they are linked. Lipid and protein oxidation share the same catalysts, and they can develop independently of one another, or in parallel, or they can interact with each other ( 35 ) Our results are consi stent with a recent report on the effects of NaCl on protein oxidation in frozen yellowtail meat which found that the addition of NaCl did not have any significant effect on the protein carbonyls content in the yellowtail meats ( 110)

PAGE 167

167 Further exploration of effects of NaCl on protein oxidation with emphasis on the detailed mechanism(s) involved in this process is warranted. Changes in a*values observed during 3.7C storage were unexpected. Without addition of salt and with interm ediate (150 mM) concentration, oxy Hb demonstrated overall higher color stability as compared to CO Hb and, as expected, m et Hb (Table 61). Th e above results are consistent with the loss of oxy Hb and the rate of deoxygenation and Hb autoxidation presented in Figures 621, 6 23, 6 25. Although, at the highest concentration of NaCl (450 mM) Red color at 3.7 C significantly correlated positively with % oxy Hb and negatively with % deoxy and met Hb (Appendix C, Table C 1). However, at the highest concentration of NaCl (450 mM) red color was more stable for CO Hb at day 1 (Table 61). During 25C storage, a*value was significantly higher for CO Hb with and without salt (Table 62). These results suggest that treatment of fish muscle with CO may delay color deterioration under specific NaCl concentrations and temperature conditions. These findings support the results of earlier studies that evaluated color stability of different kinds of meat exposed to CO treatment ( 115) This may be explained by CO replacing oxygen; it slowes the oxidation and thus deterioration of color in muscle meat systems including fish. It was also interesting to note that as salt levels was increased, the initial a*value of all samples was increased.

PAGE 168

168 Conclusion This work suggests that NaCl is highly pro oxidative in a system containing oxy Hb and membrane lipids, thus providing insights into how oxidation in seafood based products containing salt could be potentially controlled. Oxy Hb maintained its catalytic effect and is believed to catalyze oxidation by the breakdown of preformed lipid hydroperoxides The low pro oxidative activity of CO Hb is due in part to CO increasing the stability of heme pr otein structure and slowing its autoxidation. CO treatment enhanced the red color of fish muscle stored at freezer temperatures with and without added NaCl.

PAGE 169

169 Table 6 1 Changes in a*value in washed tilapia muscle cont aining different forms of Hb at concentrations of added NaCl (0, 150, and 450 mM) at 3.7C averaged across pH level and Hb concentration. a* value 0 added NaCl Storage time (day) Oxy Hb a CO Hb b Met Hb c 0 18.54.8 18.32.0 7.41.3 2 17.65.9 17.52.1 9.92.0 4 15.39.5 16.02.7 10.01.8 6 15.010.5 9.24.3 7.92.5 8 14.910.5 6.93.0 6.93.0 a* value 150.0mM NaCl Storage time (day) Oxy Hb a CO Hb a Met Hb b 0 19.74.3 22.03.0 8.11.8 2 16.08.2 21.23.7 10.32.3 4 15.59.6 16.77.3 7.72.0 6 15.110.3 11.98.6 5.72.6 8 13.49.2 10.87.3 4.52.2 a* value 450.0mM NaCl Storage time (day) Oxy Hb b CO Hb a Met Hb c 0 21.54.6 25.43.6 8.71.9 2 16.29.6 23.56.5 10.52.3 4 14.79.3 18.910.8 6.51.6 6 10.49.9 12.910.6 4.82.0 8 8.08.8 11.89.4 2.91.0 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n=18). Letters within the same rows indicate statistically (<0.05) significant differences separated by Tukeys HSD.

PAGE 170

170 Table 6 2 Changes in a* value in washed tilapia muscle conta ining di fferent forms of Hb at concentrations of added NaCl (0, 150, and 450 mM) at 25C averaged across pH level and Hb concentration. a* value 0 added NaCl Storage time (day) Oxy Hb b CO Hb a Met Hb c 0 18.54.8 18.32.0 7.41.3 4 19.84.4 22.93.2 13.42.4 8 18.04.4 21.14.3 11.92.5 12 13.26.4 19.63.5 11.42.3 16 17.25.9 18.63.7 11.22.4 20 17.37.2 17.33.5 10.82.2 24 15.57.7 16.93.9 10.12.7 a* value 150.0mM NaCl Storage time (day) Oxy Hb b CO Hb a Met Hb c 0 19.74.3 22.03.0 8.11.8 4 16.66.7 24.83.2 12.33.0 8 15.96.5 22.84.4 11.52.2 12 12.87.3 21.44.7 10.21.9 16 14.47.0 20.35.1 10.22.4 20 14.29.1 19.54.9 9.92.9 24 12.69.6 18.55.7 9.73.2 a* value 450.0mM NaCl Storage time (day) Oxy Hb b CO Hb a Met Hb c 0 21.54.6 25.43.6 8.71.9 4 15.27.1 27.16.8 11.12.5 8 13.57.0 25.07.8 9.72.1 12 14.68.4 23.37.8 9.32.6 16 12.09.0 23.18.4 9.22.9 20 14.010.3 21.08.2 8.73.0 24 12.211.1 19.79.4 8.63.6 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n=18). Letters within the same rows indicate statistically (<0.05) significant differences separated by Tukeys HSD.

PAGE 171

171 Figure 61 Lipid hydroperoxide values in washed tilapia muscle at 3.7C with no added NaCl and containing different forms of Hb, averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 62 Lipid hydroperoxide values in washed tilapia muscle at 25C with no added NaCl and containing different forms of Hb, averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys H SD. -100 0 100 200 300 400 500 600 0 2 4 6 8Lipid hydroperoxide mol/kg muscleDay (c) CO Hb (b) Met Hb (a) Oxy Hb -100 0 100 200 300 400 500 600 0 4 8 12 16 20 24Lipid hydroperoxide mol/kg muscleWeek (b) CO -Hb (a) Oxy-Hb (ab) Met -Hb

PAGE 172

172 Figure 63 Lipid hydroperoxide values in washed tilapia muscle at 3.7C with 150 mM NaCl added and containing different forms of Hb, averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05 ) significant differences separated by Tukeys HSD. Figure 64 Lipid hydroperoxide values in washed tilapia muscle at 25C with 150 mM NaCl added and containing different forms of Hb, averaged across pH level and Hb concentration. Letters within the l egend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. -100 0 100 200 300 400 500 600 0 2 4 6 8Lipid hydroperoxide mol/kg muscleDay (b) CO Hb (b) Met Hb (a) Oxy Hb -100 0 100 200 300 400 500 600 0 4 8 12 16 20 24Lipid hydroperoxide mol/kg muscleWeek (b) CO -Hb (a) Met -Hb (a) Oxy-Hb

PAGE 173

173 Figure 65 Lipid hydroperoxide values in washed tilapia muscle at 3.7C with 450 mM NaCl added and containing different forms of Hb, averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 66 Lipid hydroperoxide values in washed tilapia muscle at 25C with 450 mM NaCl added and cont aining different forms of Hb, averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. -100 0 100 200 300 400 500 600 0 2 4 6 8Lipid hydroperoxide mol/kg muscleDay (c) CO Hb (b) Met Hb (a) Oxy Hb -100 0 100 200 300 400 500 600 0 4 8 12 16 20 24Lipid hydroperoxide mol/kg muscleWeek (a) CO Hb (a) Met Hb (a) Oxy Hb

PAGE 174

174 Figure 67 TBARS values in washed tilapia m uscle at 3.7C with 150 mM NaCl added and containing different forms of Hb, averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 68 TBARS values in washed tilapia muscle at 25C with no added NaCl and containing different forms of Hb, averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. -5 0 5 10 15 20 25 30 35 0 2 4 6 8mol TBARS/kg muscleDay (b) CO -Hb (a) Met -Hb (b)Oxy-Hb -5 0 5 10 15 20 25 30 35 0 4 8 12 16 20 24mol TBARS/kg muscleWeeks (b) CO Hb (b) Met Hb (a) Oxy Hb

PAGE 175

175 Figure 69 TBARS values in washed tilapia muscle at 3.7C with 150 mM NaCl added and containing different forms of Hb, averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (< 0.05) significant differences separated by Tukeys HSD. Figure 610. TBARS values in washed tilapia muscle at 25C with 150 mM NaCl added and containing different forms of Hb, averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. -5 0 5 10 15 20 25 30 35 0 2 4 6 8mol TBARS/kg muscleDay (b) CO Hb (a) Met Hb (b) Oxy Hb -5 0 5 10 15 20 25 30 35 0 4 8 12 16 20 24mol TBARS/kg muscleWeeks (b) CO Hb (b) Met Hb (a) Oxy Hb

PAGE 176

176 Figure 611. TBARS values in washed tilapia muscle at 3.7C with 450 mM NaCl added and containing different forms of Hb, averaged across pH level and Hb concentr ation. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 612. TBARS values in washed tilapia muscle at 25C with 450 mM NaCl added and containing different forms of Hb, averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. -5 0 5 10 15 20 25 30 35 0 2 4 6 8mol TBARS/kg muscleDay (b) CO Hb (a) Met Hb (b) Oxy Hb -5 0 5 10 15 20 25 30 35 0 4 8 12 16 20 24mol TBARS/kg muscleWeeks (b) CO Hb (b) Met Hb (a) Oxy Hb

PAGE 177

177 Figure 613. Carbonyl va lues in washed tilapia muscle at 3.7C with 150 mM NaCl added and containing different forms of Hb, averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by T ukeys HSD. Figure 614. Carbonyl values in washed tilapia muscle at 25C with no added NaCl and containing different forms of Hb, averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05) sig nificant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 0 2 4 6 8nM carbonyls/mgDay (b) CO Hb (c) Met Hb (a) Oxy Hb 0 10 20 30 40 50 60 70 0 4 8 12 16 20 24nM carbonyls/mg muscleWeek (b) CO Hb (ab) Met Hb (a) Oxy Hb

PAGE 178

178 Figure 615. Carbonyl values in washed tilapia muscle at 3.7C with 150 mM NaCl added and containing different forms of Hb, averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 616. Carbonyl values in washed tilapia muscle at 25C with 150 mM NaCl added and containing different forms of Hb, averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 0 2 4 6 8nM carbonyls/mgDay (b) CO Hb (b) Met Hb (a) Oxy Hb 0 10 20 30 40 50 60 70 0 4 8 12 16 20 24nM carbonyls/mg muscleWeek (b) CO Hb (a) Met Hb (a) Oxy Hb

PAGE 179

179 Figure 617. Carbonyl va lues in washed tilapia muscle at 3.7C with 450 mM NaCl added and containing different forms of Hb, averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by T ukeys HSD. Figure 618. Carbonyl values in washed tilapia muscle at 25C with 450 mM NaCl added and containing different forms of Hb, averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (< 0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 0 2 4 6 8nM carbonyls/mgDay (b) CO Hb (b) Met Hb (a) Oxy Hb 0 10 20 30 40 50 60 70 0 4 8 12 16 20 24nM carbonyls/mg muscleWeek (b) CO Hb (a) Met Hb (a) Oxy Hb

PAGE 180

180 Figure 619. %CO released during 3.7C storage of washed tilapia muscle containing CO Hb with 0, 150, and 450 mM NaCl added to the system. The effect of forms of Hb (Oxy CO and Met Hb) by storag e interaction averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. Figure 620. % CO released during 25C storage of washed tilapia muscle containing CO Hb with 0, 150, and 450 mM NaCl added to the system. The effect of forms of Hb (Oxy CO and Met Hb) by storage interaction averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05 ) significant differences separated by Tukeys HSD. 0 5 10 15 20 25 30 35 40 45 50 0 2 4 6 8% CODay (b) 0.0mM (a) 150mM (b) 450mM 0 5 10 15 20 25 30 35 40 45 50 0 4 8 12 16 20 24 % COWeek (a) 0.0mM (c) 150mM (b) 450mM

PAGE 181

181 Figure 621. %Oxy Hb formed in washed tilapia muscle containing di fferent forms of Hb at conc entrations of added NaCl a) 0, b) 150, and c) 450 mM at 3.7C averaged across pH level and Hb concentrat ion. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % OXy -HbDay (c) Met Hb (a) CO Hb (b) Oxy Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Oxy -HbDay (c) Met Hb (a) CO Hb (b) Oxy Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Oxy -HbDay (c) Met -Hb (a) CO -Hb (b) Oxy-Hb (c)

PAGE 182

182 Figure 62 2 %Oxy Hb formed in washed tilapia muscle contain ing different forms of Hb at concentrations of added NaCl a) 0, b) 150, and c) 450 mM at 25C averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Oxy -HbWeek (c) Met Hb (a) CO Hb (b) Oxy Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Oxy -HbWeek (c) Met Hb (a) CO Hb (b) Oxy Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Oxy -HbWeek (c) Met Hb (a) CO Hb (b) Oxy Hb (c)

PAGE 183

183 Figure 623. %Met Hb formed in washed tilapia muscle contain ing different forms of Hb at conc entrations of added NaCl a) 0, b) 150, and c) 450 mM at 3.7C averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8% Met -HbDay (a) Met -Hb (c) CO -Hb (b) Oxy-Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Met -HbDay (a) Met Hb (c) CO Hb (b) Oxy Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Met -HbDay (a) Met -Hb (c) CO -Hb (b) Oxy-Hb (c)

PAGE 184

184 Figure 624. %Met Hb formed in washed tilapia muscle contain ing different forms of Hb at conc entrations of added NaCl a) 0, b) 150, and c) 450 mM at 25C averaged across pH level and Hb concentration. Letters within the legend of the f igure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Met -HbWeek (a) Met Hb (c) CO Hb (b) Oxy Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Met -HbWeek (a) Met Hb (c) CO Hb (b) Oxy Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Met -HbWeek (a) Met Hb (c) CO Hb (b) Oxy Hb (c)

PAGE 185

185 Figure 625. %Deoxy Hb formed in washed tilapia muscle contain ing different forms of Hb at conc entrations of added NaCl a) 0, b) 150, and c) 450 mM at 3.7C averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8% Deoxy -HbDay (a) Met Hb (b) CO Hb (a) Oxy Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 % Deoxy -HbDay (a) Oxy Hb (b) CO Hb (a) Met Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8% Deoxy -HbDay (a) Met -Hb (b) CO -Hb (a) Oxy-Hb (c)

PAGE 186

186 Figure 626. %Deoxy Hb formed in washed tilapia muscle containing different forms of Hb at conc entrations of added NaCl a) 0, b) 150, and c) 450 mM at 25C averaged across pH level and Hb concentration. Letters within the legend of the figure indicate statistically (<0.05) significant differences separated by Tukeys HSD. 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24% Deoxy -HbWeek (a) Met Hb (b) CO Hb (a) Oxy Hb (a) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24% Deoxy -HbWeek (a) Met Hb (b) CO Hb (a) Oxy Hb (b) 0 10 20 30 40 50 60 70 80 90 100 0 4 8 12 16 20 24 % Deoxy -HbWeek (a) Met Hb (b) CO Hb (b) Oxy Hb (c)

PAGE 187

187 CHAPTER 7 EFFECT OF THE OXIDATION OF OXY CO AND MET HEMOGLOBIN ON THE QUALITY OF REFRIGERA TED WASHED MINCED TILAPIA MUSCLE Introduction World consumption of fish has increased dramatically over the last 50 years and this increase is now only met with increased capacity of aquaculture ( 116) The stability of seafoods and the extension of the shelf life of seafood products has become crucial for the seafood industry. Of prime importance is the red color found in the dark muscle of fish like mahi m ahi, tuna, and tilapia. The presence of this red color influences the consumer and is interpreted as freshness, thereby increasing its market value. The brown or dark color of fish suggests lack of freshness and poor quality ( 79) Maintaining this red color has been achieved in the past through refrigeration, fr eezing and modified packaging. More recently, the use of carbon monoxide and filtered wood smoke has been used to enhance this red color in fish muscle ( 69) Carbon monoxide (CO) has been used to stabilize the red color in fish muscle. However, few studies have been published on the effect of CO on the oxidative stability of seafood products by stabilizing the heme protein, a key pro oxidant, and preventing oxidation of oxygenated hemoglobin and myoglobin, (red color) to methemoglobin and myoglobin (brown color). Quality of fish may have different meanings for different people. Furthermore, quality can be assessed by instrumental and sensory evaluations. Since the consumer is the ultimate judge of quality correlation of instrumental and sensory evaluations will provide the best evidence of quality. In the present study we examined effects of different degrees of oxidation on the development of painty/off odor and color changes in a washed tilapia muscle model

PAGE 188

188 system at 3.7C assessed by a comparison between subjective (sensory evaluation), chemical (TBARS), an d physical (a*value) analysis. Sensory evaluation is commonly utilized to assess the quality and freshness of fish and fish products during processing and storage. However, it is believed that various processing methods may mask (alter, modify to a certain degree) the subjective perception of freshness of seafood ( 117 ) For instance, some manipulations aimed to stabilize and prevent deterioration of the color may give false impression on quality and freshness of muscle foods to a consumer. I n this part of the experiment, we aimed to determine the correlations between subjective (sensory evaluation) and objective methods (chemical presented by TBARS scores, and physical presented by a*values), and therefore verify the validity of descriptive analysis in rating of freshness of seafood (particularly tilapia) under different processing and storage conditions. Materials and Methods The following methods have been described in detail in chapter 3: Preparati on of Washed Minced Tilapia Muscle (MTWM), Collection of Fish Blood, Preparation of Hemolysate, Quantification of Hb Levels in Hemolysate, Oxy CO and Met Hb and Determination of Thiobarbituric Acid Reactive Substances (TBARS) Sample Preparation : Addition of Hb and NaCl Samples of tilapia was hed model system previously prepared and stored at 80C were thawed rapidly under running water (20C) and kept on ice. The pH was adjusted to the desired pH of (6.3, 6.8, or 7.3) using 2N NaOH or 2N HCl. After the desired pH was established, moisture content was determined using a moisture balance (CSI Scientific Company, Inc., Fairfax, VA).

PAGE 189

189 Oxy hemoglobin previously prepared was thawed under running water and added to the muscle system to give a final concentration of 12 mol/kg washed muscle. Hb was mix ed manually into the WMTM system using a plastic spatula. The homogenous color of the minced mus cle indicated adequate mixing. Hb was added to the system at three pH levels (6.3, 6.8, and 7.3). Concentrations of added NaCl to the samples were 0 and 450 mM NaCl was mixed manualy Samples were plated in Petri dishes (~25 g) covered with a lid, and stored at 3.7C for 6 days All samples were stored i n duplicate. The various combinations were conducted as presented in Table 7 1. CO Hb samples were prepared by gassing the washed system with 100% CO for two hours on ice. The WMTM was placed in a gastight vacuum bags equipped with a silicon septum valve obtained from LabPure Instruments. CO Hb was added to the muscle after removal from the bags to give a final concentration of ~ 12mol/kg washed muscle. CO Hb was mixed manually using a plastic spatula. CO Hb was added to the system at three pH levels (6.3, 6.8, and 7.3). Samples were placed in Petri dishes (~25 g) covered with a lid on, and stored at 3.7C for 6 days. All samples were stored in duplicate (Table 7 1). Met Hb, previously prepared, was mixed manually with the washed system using a plastic spatula to a final concentration of ~ 12 mol/ k g washed muscle. Met Hb was added to the system at three pH level s (6.3, 6.8, and 7.3). Samples were plated in Petri dishes (~25 g) covered with a lid on, and stored at 3.7C for 6 days All samples were stored in duplicate (Table 71). One g of sample was taken at day 0, 3, and 6 for chemical analysis. The remainder of the samples were returned to 3.7C. For color analysis, 12 g of samples

PAGE 190

190 were stored with a lids on. Due to the difficulty of mixing samples containing 450 mM NaCl, samples were taken from the core and side of the dishes. These samples were then stored in aluminum foil at 80C until analysis was performed. Descriptive Sensory Analysis Sensory analysis was conducted with modification of the procedure of Richards and others ( 118 ) to assess the qualitative and quantitative off odor of oxidation. A descriptive analysis sensory test was used. Thirteen panelists were chosen based on their abil ity to detect the rancid odor. Panelists were screened using a simple triangle test. Panelists were asked to sniff the samples in dark cups and identify the odd sample. The design of this Triangle Test is presented in Appendix D Based upon the results of the triangle test, 13 panelists were then selected to be trained to rate the off odor (painty/rancid odor) using a 15 point scale, 1 being no off odor and 15 the strongest ( Appendix E ). Three sessions, 60 min each, were used to train the panelists to correctly rate the intensity of the off odor (painty/rancid odor). A 1015 min break was taken between each set of samples. Panelist were trained using a set of samples previously prepared as a standard at pH 6.3 oxy Hb day 0, 2, 4, and 6 with and without NaCl. Table 7 2 represents treatments used to train panelis ts. Six evaluation sessions followed the training sessions. Table 73 represents the combinations of treatments evaluated by panelists. The samples within a session were randomly presented. For each sensory session, samples stored at 3.7C for 0, 3, and 6 days were used. Oxy, CO and Met Hb at a Hb concentration of 12 L/kg muscle were evaluated in duplicate in each session. Hemolysate was thawed under running water and mixed with the tilapia minced wash muscle in the form of Hb (oxy CO and met Hb) and pH (6.3, 6.8, and 7.3) desired ( Table 74 ). Sample were placed in capp ed dark

PAGE 191

191 cups (11 g) and stored at 3.7C for six days. Day zero samples were vacuum packed and stored at 80C immediately followed by day 3 and day 6. The sensory testing was conducted in the FSHN sensory lab where panelists were seated in separate booths. Color of the samples was masked using red light to avoid the influence of color on the rating of the odor intensity. Samples were thawed to 3.7C 2 h prior to the test and kept on ice until presented to the panelist. Samples were assigned three random digit numbers. Panelists were asked to sniff the samples in the order presented from left to right and rate the intensity of the off odor. Test ballots are presented in Appendix E Total n umber of samples evaluated in each session was 9 and each sample was evaluated in duplicate. Panelists wer e asked to take a 10 15 min break between each replication to avoid fatigue. Color Analysis Deterioration of red color (a*value) during oxidation in the washed muscle samples was measured using a digital Color Machine Vision System (CMVS) ( 97) The CMVS measures the average (a*value) for each sample. Pictures of samples were taken at the same intervals that samples were taken for chemical and sensory analysis. Change in color was correlated with TBARS and sensory results. In a closed chamber (impermeable to stray light), a 1 2 g sample was placed. A digital camera (Nikon D200 Digital Camera, Nikon Corp., Japan) facing the bottom of the chamber was used to capture pictures of the samples. Two fluorescent lights (top of the chamber), each to simulate illumination by noonday summer sun (D65 illumination), were used to obtain uniformity of light. The Nikon D200 Settings used are shown in Table 7 4 A red reference tile was placed with each picture and used as a standard for the redness (a*value).

PAGE 192

192 S tatistical Analysis The mean and standard deviation of a*value for oxy, CO and m et Hb ; pH values 6.3, 6.8, and 7.3; and NaCl concentrations added of 0 and 450 during storage at 3.7C for 0, 3, and 6 days are presented. Statistical comparisons were made between treatments using the General Linear Model Procedure and the mean separations were performed using Tukeys Least Square Means (SAS software 9.2). A statistical significance was reported as p Hb concentrations, pH values, and NaCl concentrations on lipid/protein oxidation is reported, Fishers Z transformation correlation comparison and Correlated Correlation Comparison method were conducted ( 119, 120) comparing sensory scores and TBARS, sensory scores and a* value, and TBARS and a*value. Results Sensory Evaluation A total of 13 trained panelists were used to evaluate the development of the painty odor in the tilapia muscle during storage at 3.7C for 6 days as a measurement of oxidation. Using a 15 point scale, 1 being no off odor and 15 the strongest ( Appendix E ), the outcomes showed that the panelists detected painty/rancid odor under various treatments of the samples. When panelists were asked to evaluate samples containing the three different forms of Hb (oxy CO and m et Hb ), no odor was detected on day 0 for any of the three Hb forms ( Figure 71 ). However, for oxy Hb samples panelists were not able to detect (p<0. 05) the painty odor by day 3. Similarly, no odor was det ected for CO Hb samples on day 3. M et Hb samples on the other hand, showed a significant development of the painty/rancid odor detect ed by the panelists on day 3. On day 6, for oxy and CO Hb samples panelists scores showed a significant difference in the odor

PAGE 193

193 compared to day 3 but metHb samples displayed no sign ificant change between day 3 and 6 in the panelists rat ing of the painty/rancid odor. The results of the interaction of Hb form by day ( Figure 71 ) indicated that day 3 and day 6 for met Hb samples were significantly different compared to oxy and CO Hb evaluated on the same days. Figure 72 s hows the results for the effect of the pH on the development on the oxidation assessed by the increase of the painty/rancid odor. It was observed that at pH 6.3, there was a significant differenc e between day 0 and day 3, but no difference in day 6 compared to day 3. The panelists were unable to detect the difference of the sample at pH 6.8 on day 0 or day 3. However, panelists score showed a significant increase in the odor by day 6 for pH 6.8 samples. For pH 7.3, there was no significant difference in th e panelists rating between day 0, 3, or 6. The results of the interaction of panelists by day ( Figure 7 2 ) showed that day 3 was rated significantly higher for pH 6.3 samples only. While day 6 for pH 6.8 was not different (p<0.05) from day 6 for pH 6.3 an d 7.3, day 6 for pH 6.3 was sign ificantly different from day 6 for pH 7.3. Regardless of the concentration of NaCl used, panelists scores indicated a significant development of the painty/rancid odor after storage for 3 days at 3.7C ( Figure 73 ). However, no additional odor was detected by day 6 for either samples with and without added NaCl The results of the interaction of NaCl by day ( Figure 7 3 ) showed there was no NaCl effect. Lipid Oxidation Analysis TBARS was used to measure lipid oxidation in the muscle system. The effect of different forms of Hb (oxy CO and met Hb) is presented in Figure 74 Met Hb had significantly (p 0.05) less oxidation lag time compared to oxy and CO Hb samples However, t he amount of TBARS developed on day 6 did not dif fer significantly from day

PAGE 194

194 3 for met Hb samples Oxy Hb was not significantly different fr om CO Hb on day 0 or day 3, however, oxy Hb showed more stability on day 6 of storage compared to CO Hb. The interact ion of day by TBARS showed that for all three forms of Hb, day 3 was signif icantly higher in TBARS for met Hb samples compared to oxy and CO Hb ( Figure 7 4 ). The oxidation lag phase was found to decrease significantly as pH level decreased ( Figure 75 ). For pH 6.3, there was a signifi cant difference between day 0 and day 3 in the amount of TBARS formation, but not between day 3 and day 6 The same pattern of oxidation was observed for pH 6.8. After 6 days of storage at pH 7.3, no significant change in the amount of TBARS was observed. The results of the interaction of day by TBARS indicated that all three pH levels for day 0 and day 6 had no significant difference in the amount of TBARS detected. However, at day 3, pH 7.3 was significantly different from pH 6.3 in the amount of TBARS, but not different from pH 6.8 which also did not also differ from pH 6.3. Thus, the interaction of day by TBA RS indicates that only day 3 had significant differences for the three levels of pH ( Figure 75 ). The effect of NaC l on the development of ox idation during storage for 6 days at 3.7C is shown in Figure 76 Regardless of the concentration of added NaCl (0, and 450 mM), the oxidation lag time did not differ significantly between different concentrations. However, there was a significant difference between the amount s of TBARS detected on day 3 compared to day 0 for both concentrations ( Figure 76 ). Day 6 was not signif icantly different from day 3 for both concentrations. The interaction of

PAGE 195

195 day by TBARS indicated no significant difference between both concentrations of NaCl on each day of analysis ( Figure 76 ). Correlation between Sensory Scores and TBARS The amount of TBA RS formed during storage for 6 days was highly correlated ( Figure 77 ) with panelists ratings of painty/rancid odor (R= 0.95) at p<0.0001. As oxidation, measured using TBARS, increased in the muscle system, the panelists rating of painty/rancid odor increased. Hb forms, pH levels, and NaCl concentrations had no significant effect on the correlation between TBARS and sensory scores. Color Analysis Effect of different forms of Hb (oxy CO and met Hb), different pH levels (6.3, 6.8, and 7.3), and effect of different concentrations of added NaCl (0, and 450 mM) on the stability of Hb color in tilapia washed system are presented in Table 75 Color analysis was conducted using a Color Machine Vision Systems where average a* (redness) value of the whole surface of the tilapia muscle was measured during 6 days of storage at 3.7C ( Table 7 5 ). Redness of samples containing oxy Hb decre ased significantly after 3 days of storage. Further decrease in redness for oxy Hb on day 6 was not signif icantly different from day 3. CO Hb maintained its redness through storage at day 3, but had a signific ant decrease in color by day 6. Color of m et Hb samples did not change sign ificantly during storage for 6 days, however, its redness was significantly less than oxy a nd CO Hb forms for each day of analysis as indicated by the interaction of day by a*value ( Table 75 ). Redness of samples was affected by the level of pH ( Table 7 5 ). As pH increased, the a*value (redness) for samples became more stable during storage. Low pH (6.3)

PAGE 196

196 samples had significantly lower a*values for each day of storage. After 6 days of storage, pH 6.8 samples showed a significant loss of redness. High pH (7.3) maintained a stable a*value during storage ( Table 7 5 ). The interaction of day by a*val ue indicated a significant difference in the redn ess for all pH levels on day 6 with pH 6.3 having significantly lower a*value and pH 7.3 having higher a*value. NaCl had an effect on red color (Table 7 5) In samples with no added NaCl a significant decrease in a*value was detected on day 6 of storage. However, addition of 450 mM NaCl/kg muscle resulted in significant loss of redness by day three and a further significant decrease was detected on day 6. The interaction of NaCl by day indicated a signi ficant loss of redness for the c oncentration of 450mM on day 6 ( Table 7 5 ). Correlation between Sensory Scores and a*Value As a*value decreased, the panelists scores of painty/rancid odor increased and were highly negatively correlated (R= 0.81, p<0.0001) ( Figure 78 ). For correlation between panelists scores and color, pH levels and NaCl concentrations had no significant effect. However, different forms of Hb did (Table 7 6) Met Hb samples did not have as strong a correlation (R= 0.66 ) as the other form s (oxy and CO had correlations of R= 0.96). The p value for comparing met and oxy is 0.0013 and p value for comparing met and CO was 0.0017 ( Table 7 6 ). Correlation between TBARS and a*Value It was observed that there was a correlation (R= 0.71) between TBARS and a*value at p<0.0001 ( Figure 79 ). The a*value (redness) decreased as oxidation proceeded over time during storage for six days at 3.7C and TBARS value increased. For correlation between TBARS and a*value, p H levels and NaCl addition had no

PAGE 197

197 signi ficant effects. However, different forms of H b had an effect. The met Hb samples (R= 0.50) did not have as strong a correlation as oxy Hb (R= 0.84, p =0.0663) and the CO Hb samples (R= 0.88, p =0.026) as presented in Table 7 6 The correlation between TBAR S and a*value was significantly different from the correlation between panelist and a*value (p =0.0001) ( Figure 79 ). Discussion The degree of deterioration in sensory quality and freshness (odor, color, texture, and flavor) of seafood is thought to be highly dependent on the degree of lipid peroxidation, which in turn may be affected by a number of conditions, including pH, storage temperature, prevalence of certain types of Hb, extent of storage, etc ( 80, 81) Specifically, the level of rancidity of the lipids marked by painty/off odor development in var ious fish samples was correlated to a degree of lipid peroxidation ( 23 ) The findings of this study demonstrated a statistically significant correlations between trained panelists scores, the development of thiobarbituric acid reactive substances (TBARS), and deterioration in red color (a*value). Furthermore, the observed correlation was not significantly affected by the type of treatment, Hb form, pH levels, NaCl added concentrations, or extent of storage, verifying the usefulness of descriptive analysis to assess oxidation in the washed tilapia muscle system. The accuracy of the sensory evaluation as well as the strong relationship between painty odor and the degree of lipid peroxidation in ti lapia muscle meat in this experiment certifies the use of painty/off odor as a valid marker of lipid oxidation. Clear association was observed between panelists scores and results of lipid oxidation analysis determined by the TBARS method ( Figure 7 7 ). Th is is in agreement with previous reports investigating different markers of lipid oxidation in fish products.

PAGE 198

198 The TBARS method has been shown to be a useful measure of rancidity development, and also correlates well with the results of sensory evaluation i n many seafood products ( 11, 55 ) Specifically, when asked to evaluate the odor in fish samples treated with different forms of Hb, it appears that panelists ratings were accurately corresponding to the presence and quantity of products of oxidation expressed as TBARS. The addition of m et Hb to tilapia produced a significant rancid odor by day three which did not ch ange significantly by day 6 ( Figure 71 ). In contrast, in the samples with Oxy and CO Hb, off o dor was not detectible on day 3, and appeared only by day 6. TBARS analysis demonstrated shorter oxidation lag time and time needed to reach a plateau in samples treated with m et Hb ( Figure 74 ). The m et form of Hb has been shown in numerous experiments to induce lipid oxidation in different muscle systems ( 80) It was proposed that the prooxidant activity of m et Hb is due to its structural characteristics promoting autoxidation, and ultimately, li pid peroxidation. Recent research has found that met Hb is a potential pro oxidant at the pH found in fresh meat ( between 5.3 and 6.2) which emphasizes the importance of hydroperoxides in met Hb initiated lipid oxidation ( 53) Electrostatic and hydrophobic interactions are involved when Hb binds to phospholipids. Met Hb affects the structural and physiochemical parameters of the lipid water interface ( 54) This results in the formation of hemichrome, a poor initiator of lipid oxidation ( 41) at physiological pH in model systems containing long chain free fatt y acids, resulting in non catalytic activity. At lower pH values, the electrostatic and hydrophobic interactions are not involved, most likely due to the different charge distribution on both the fatty acid and the heme protein ( 5 ) Thus, in the presence of lipids, met Hb at physiological pH, due to the formation of the non catalytic hemichrome,

PAGE 199

199 can undergo rapid neutralization. However, in a high lipophilic environment, denaturation of the heme protein may result in exposure of the heme group to the surrounding lipids and induce lipid peroxidation. However, at lower pH values, met Hb is able to initiate lipid oxidation in a lipid hydroperoxidedependent mechanism. In contrast, o xy Hb and CO Hb have lesser proox id ant activity which was reflected by slower TBARS production as well as retarded devel opment of rancid odor by day 6 (Figures 54 and 57) This supports results of previous investigations ( 57 ) Replacing oxygenated Hb and Mb with CO Hb and CO Mb in the dark muscle of fish was shown to offer some protection against lipid oxidation ( 4, 69, 70 ) High concentrations of unsaturated fatty acids (e.g. linoleic acid) in dark muscle of fish, in the presence of oxy Hb and deoxy Hb may result in increased levels of lipid peroxides. This is due mainly to the autoxidation and deoxygenation of Hb, especially at reduced pH levels. CO treatment re tards autoxidation of Hb and myoglobin to the ferric form, and thus plays a role in decreasing lipid oxidation, and possibly extending the shelf life of fish fillets treated with CO. Danyali ( 71 ) found that CO and filtered smoke (FS) treatment can be effective in retarding lipid oxidation. Filtered smoke, which contains phenolics that have the potential to serve as antioxidants, was not more effective than the other gas treatments in retarding oxidation development. After tuna steaks were treated with FS, 4%, 18% and 100% CO, Kristinsson and others ( 72) measured lipid oxidation using TBARS (thiobarbituric acid reactive substances). They found that the 4% CO treated tuna had higher TBARS than untreated tuna. The authors theorized that the higher levels of oxygen and CO2 (which results in a drop in muscle pH) in the 4% CO may have promoted oxidation of heme prote in. The 18% CO and FS (containing

PAGE 200

200 approximately 18% CO) may have protected the heme molecule against the action of CO2. Garner and others ( 121) reported that 100% CO was more effective than 100% nitrogen in retarding lipid oxidation, which suggests that the effect is not due solely to the absence of oxygen. Richards and Hultin ( 4 ) reported the presence of pro oxidants of fish muscle, which may be reduced through the stabilization of the heme protein by CO ( 72) Kristinsson and others ( 72 ) conclude that treating yellowfish tuna with medium to high levels of CO may stabilize the heme protein molecule against lipid oxidation. Tested pH levels (6.3, 6.8, and 7.3) were chosen to reflect the range of post mortem pH of fish products. Our findings revealed accelerated formation of TBARS at an acidic pH when compared to more basic pH levels. These findings are in agreement with previous investigations describing more rapid lipid oxidation at lower pH values. Richards and others ( 80 ) suggest t hat at low pH val ues (pH 6.3), tilapia Hb subunits dissociate faster and increase release of hemin. This may be due to the weakened hemeglobin linkage caused by the protonation of proximal histidine. The structural change in Hb increases the proton transf er mechanism. Kristinsson and others ( 69 ) and Kristinsson and Hultin ( 10 ) reported that at lower pH, the protein is partly unfolded, giving the heme portion greater ability to participate in oxidation. At higher pH levels, oxy and CO Hb may have greater stability of the protein structure, resulting in lower lipid peroxide formation. In contrast, m et Hb, which is readily formed at acidic pH, is believed to catalyze oxidation by the breakdown of preformed lipid peroxides ( 11 ) A greater degree of rancid odor at lower pH values (6.3) compared to intermediate (6.8) versus highest (7.3) was also observed. This trend was not affected by the extent of storage (0, 3, or 6 days).

PAGE 201

201 The degree of redness of fish muscle is known to be affected by a multitude of factors, most importantly: pH, type of Hb present, and extent and temperature of storage ( 122 ) In this study, samples with CO Hb displayed the greatest stability of a*val ues, with significant decrease in color only by day 6 (Figure 7 1 ). It was previously described that CO treated samples have the ability to retard lipid oxidation, and stabilizing redness of fish muscle samples ( 122 ) The data fr om the present experiment establishes a strong link between lipid oxidation and degree of color stability/changes. Met Hb, recognized as a strong promoter of lipid oxidation in muscle food systems, produced a considerable decrease in redness on day 0, and did not significantly change by day 6 (Table 75). Although the rancid odor of samples contained m et Hb was detected by panelists only by day three, no further changes in odor were detected by day 6 (Figure 72). Results of color analysis for o xy Hb were s omewhat intermediate between CO and met H b, with decrease in redness detected by day 3 without significant further changes by day 6. Sensory evaluation of corresponding samples showed off odor by day 3 in creasing in intensity by day 6. Lastly, in the present study we observed no significant effect of addition of different concentrations of added NaCl (0, and 450 mM) on TBARS, a*value, or sensory scores. It was previously reported that NaCl may disrupt the stability of oxy Hb, accelerating its deoxygenation and formation of met Hb ( 84) From accumulated data it appears that the prooxidative activities of NaCl may be due, in part, to its ability to hasten release of iron from heme ( 82) In particular, the chloride anion was implicated in Hb autoxidation, se rving as a substrate for a chloroperoxidase in the presence of preformed hydrogen

PAGE 202

202 peroxide ( 84, 111) In addition, it was proposed that NaCl may have the ability to accelerate lipid peroxidation by shifting ironoxygen to ironhydroperoxide interactions. Osinchak and others ( 123 ) used a press juice of atlantic mackerel ( model sys tem ) to examine the effect of NaCl on lipid oxidation. The authors found that NaCl was clearly a potent pro oxidant of lipid oxidation which they moniterd by the development of TBARS. Harel ( 85) found NaCl t o inhibit the hydroxyl radical. It was suggested that NaCl inhibits ferrous i r on oxidation. Wallace and others ( 84 ) found that NaCl affect e d the stability of oxy Hb and accelerated the oxidation of oxy Hb to met Hb, thus enhancing Hb autoxidation. This appeared not to be consistent with the findings of the washed tilapia model system. Conclusion Sensory analysis of washed mince muscle containing 12 mol Hb/kg of CO oxy and met Hb at pH 6.3, 6.8, and 7.3 with and without added salt revealed that panelists detection of rancidity was highly correlated with TBARS analysis. Sensory scores correlated negatively with a* value, giving further evidence of the accuracy of sensory analysis to detect rancidity in a washed mince model system. The significant negative correlation between a*value and TBARS gives a measure of the consistency of the results presented. Panelists found no significant differences in samples with and without addition of NaCl.

PAGE 203

203 Table 7 1. T reatments used in the study. Hemoglobin was added to washed tilapia muscle at a concentration of 12 mol Hb/kg muscle and stored at 3.7C. Hb form pH NaCl added (mM/Kg ) Oxy Hb 6.3 0 450 6.8 0 450 7.3 0 450 CO Hb 6.3 0 450 6.8 0 450 7.3 0 450 Met Hb 6.3 0 450 6.8 0 450 7.3 0 450

PAGE 204

204 Table 7 2. Reference s amples used to train panelists for the descriptive analysis testing Treatment pH Hb [mol/kg] Days Oxy Hb with out added NaCl 6.3 12 0, 2, 3, 4, 6 Oxy Hb with 450 mM NaCl 6.3 12 0, 2, 3, 4, 6

PAGE 205

205 Table 7 3. C ombination of treatments used by panelist s to rate the formation of painty /rancid odor as an indicator of lipid oxidation Session Treatment pH NaCl added [mM/kg muscle] Storage Days 1 Oxy CO Met Hb 6.3 0 0, 3, 6 2 Oxy CO Met Hb 6.3 450 0, 3, 6 3 Oxy CO Met Hb 6.8 0 0, 3, 6 4 Oxy CO Met Hb 6.8 450 0, 3, 6 5 Oxy CO Met Hb 7.3 0 0, 3, 6 6 Oxy CO Met Hb 7.3 450 0, 3, 6

PAGE 206

206 Table 7 4. Nikon D200 Settings used for measurement of change in color during storage period. Setting Specification Lens Focal length Sensitivity Optimize image High ISO NR Exposure mode Metering mode Shutter speed and aperture Exposure compensation (in camera) Focus mode Long exposure NR Exposure compensation (by capture NX) Sharpening Tone compensation Color mode Saturation Hue adjustment White balance Zoom VR 18 200 mm F 3.5 5.6 G 36 mm ISO 100 Custom Off Manual Multi pattern 1/3s F/11 1.0 EV Manual Off 0 EV Normal Normal Mode I Normal 0 Direct sunlight Manual

PAGE 207

207 Table 7 5. Changes in a* value of tilapia muscle after the addition of the Hb and NaCl at three different pH levels followed by storage for 6 days at 3.7C Storage time (day) Oxy Hb CO Hb Met Hb 0 26.32.6 a* 23.33.1 a* 8.60.6 a** 3 20.010.8 b* 20.28.5 a* 6.93.5 a** 6 16.613.1 b* 14.811.4 b* 3.61.1 a** Storage time (day) pH 6.3 pH 6.8 pH 7.3 0 17.87.1 a* 19.59.0 a* 20.99.8 a* 3 7.36.5 bc** 18.89.7 a* 21.18.5 a* 6 3.11.3 c* 12.611.1 b** 19.311.7 a*** Storage time (day) 0 added NaCl 450 mM NaCl 0 18.48.0 a* 20.48.9 a* 3 17.39.2 ab* 14.111.0 b* 6 14.112.1 b* 9.210.2 c** Values are means Standard Deviations for all pixels of the surface of the minced muscle samples. Letters within the same column indicate statistically (<0.05) significant differences for storage times. Symbols within the same rows indicate statistically (<0.05) significant differen ces between oxy, CO and Met Hb.

PAGE 208

208 Table 7 6. Fishers z transformation correlation comparison (R, p<0.05) between sensory scores, TBARS, and a*value in tilapia washed muscle during storage for 6 days at 3.7C Sensory Scores and TBARS Oxy Hb CO Hb Met Hb pH6.3 pH6.8 pH7.3 0.0NaCl 450NaCl 0.94 0.97 0.94 0.93 0.97 0.98 0.96 0.95 Sensory Scores and a*value Oxy Hb CO Hb Met Hb pH6.3 pH6.8 pH7.3 0.0NaCl 450NaCl 0.96 0.96 0.66 0.83 0.80 0.76 0.80 0.84 TBARS and a*value Oxy Hb CO Hb Met Hb pH6.3 pH6.8 pH7.3 0.0NaCl 450NaCl 0.84 0.88 0.50 0.69 0.73 0.69 0.69 .073

PAGE 209

209 Figure 71 Hb form by storage interaction. Change in painty/rancid off odor of tilapia washed muscle samples with different Hb forms added. Samples were stored for six days at 3.7C. Different letters for each Hb form (Oxy, CO, Met Hb) indicate statistically significant differences <0.05. Different symbols within each day indicate st atistically significant differences <0.05. Met Hb differed from CO and Oxy Hb on day 3 and 6. Figure 72 pH by storage interaction. Change in painty/rancid off odor of tilapia washed muscle samples with different Hb forms added. Samples were stored for si x days at 3.7C. Different letters for each pH level indicate statistically significant differences <0.05. Different symbols within each day indicate statistically significant differences <0.05. 1 3 5 7 9 11 13 15 0 3 6 Sensory ScoresDay Oxy Hb CO Hb Met Hb a* a* b ab ab b b* b* a* 1 3 5 7 9 11 13 15 0 3 6 Sensory ScoresDay pH 6.3 pH 6.8 pH 7.3 a* a a a* a* b* ab b* b*

PAGE 210

210 Figure 73 NaCl concentration by storage interaction. Ch ange in painty/rancid off odor at d ifferent NaCl concentrations during storage of tilapia washed muscle for six days at 3.7C. Different letters for each concentration indicate statistically significant differences <0.05. Different s ymbols within each day indicate statistically significant differences <0.05. Figure 74 TBARS by storage interaction. Development of TBARS at different Hb forms during storage of tilapia washed muscle for six days at 3.7C. Different letters for each Hb form indicate statistically significant differences <0.05. Different symbols within each day indicate statistically significant differences <0.05. Only day 3 showed a significant difference between Met Hb versus CO and Oxy Hb. 1 3 5 7 9 11 13 15 0 3 6 Sensory ScoresDay 0.0 mM NaCl 450 mM NaCl a* a* b* b* b* b* 0 5 10 15 20 25 30 35 40 0 3 6 mol TBARS/kg muscleDay Oxy -Hb CO -Hb Met -Hba b ab a a a b b* a

PAGE 211

211 Figure 75 pH by storage interaction. Development of TBARS at different pH levels during storage of tilapia washed muscle for six days at 3.7C. Different letters for each pH level indicate statistically significant differences <0.05. Different symbols within each day indicate statistically significant differences <0.05. Only day 3 showed a significant difference between pH 6.3 versus pH 6.8 and 7.3. Figure 76 NaCl by storage interaction. Development of TBARS at different NaCl concentrations during storage of tilapia washed muscle for six days at 3.7C. Different letters for each concentration indicate statistically significant differences <0.05. S ymbols within each day indicate there was no statistically significant difference <0.05 on each day for either concentration of NaCl. 0 5 10 15 20 25 30 35 40 0 3 6 mol TBARS/kg muscleDay pH 6.3 pH 6.8 pH 7.3 a a b* b a b* b a a 0 5 10 15 20 25 30 35 40 0 3 6 mol TBARS/kg muscleDay 0.0mM NaCl 450mM NaCl a* b* b* a* b* b*

PAGE 212

212 Figure 77 Correlation (p<0.0001) between the development of TBARS and sensory scores during storage of tilapia washed muscle for six days at 3.7C. Figure 78 C orrelation (p< 0.0001) between change in a*value and sensory scores during storage of tilapia washed muscle for six days at 3.7C. y = 0.2263x + 1.3726 R = 0.9502 0 2 4 6 8 10 12 14 0 10 20 30 40 50 60Sensory Scoresmol TBARS/kg muscle y = -0.299x + 8.8825 R= -0.81 0 2 4 6 8 10 12 14 0 5 10 15 20 25 30 35Sensory Scoresa*-Value

PAGE 213

213 Figure 79 Correlation (p< 0.0001) betw een the development of TBARS and change in a* value during storage of tilapia washed muscle for six days at 3.7C. y = -0.4586x + 21.363 R = -0.71 -10 -5 0 5 10 15 20 25 30 35 0 10 20 30 40 50 60a*-Valuemol TBARS/kg muscle

PAGE 214

214 CHAPTER 8 SUMMA RY AND CONCLUSION Lipid and protein oxidation was measured using washed minced tilapia muscle (WMTM) adjusted to pH 6.3, 6.8 and 7.3. Oxy Hb, CO Hb, and m et Hb at concentrations of 6, 9, and 12 mol/ kg muscle, and NaCl at added concentrations of 0, 150 and 450 mM were added to the WMTM. Samples were stored at 3.7C for 8 days and 25C for 24 weeks and assessed at intervals for oxidation. Lipid oxidation was monitored by following thiobarbituric acid reactive substances (TBARS) and lipid hydroperoxides (LOOH). Protein oxidation was measured by the DNPH method. Hb oxidation state was monitored and change in color was measured using Color Vision Machine System. Formation of m et Hb was monitored spectrophotometrically. Rancidity was evaluated with a sens ory panel (Appendix E ). CO Hb was overall significantly less pro oxidative compared to o xy and m et Hb r egardless of concentration used. The low pro oxidative activity of CO Hb is likely due in part to CO increasing the stability of heme protein structure, thereby reducing oxidative processes and slowing down autoxidation. The reduced reactivity of CO Hb with hydrogen peroxides may be due t o the strong affinity of CO to Hb. T he effect of pH on oxidation superseded NaCl concentration, Hb form and concentra tion, or temper ature in influencing oxidation and controlling f or deterioration of fish muscle. Low pH increased the susceptibility of the washed minced tilapia muscle to oxidation, the lower the pH, the greater oxidation for all three forms of Hb. The color and sensory quality of the washed tilapia muscle also decreased with decreasing pH levels Temperature significantly affected the auto oxi dation of CO o xy and m et Hb with greater % of m et Hb formed at 25 C.

PAGE 215

215 NaCl significantly affected the stability of o xy Hb and its oxidation to m et Hb CO Hb, on the other hand, was significantly less prooxidative compared to other forms of Hb, regardless of concentration used. The low prooxidative activity of the oxidized form ( Met Hb ) could be a result of the absence of oxygen to form hydrogen peroxide required for m et Hb oxidation. The higher the concentration of NaCl, the greater the oxidation to m et Hb r egardless of the concentration of Hb. Higher concentrations of NaCl also decreased the stability of color, and the sensory quality indicating NaCl has a high prooxidant act ivity in a system containing oxy Hb and membrane lipids, thus providing insights how oxidation in seafood based products containing salt could be potentially controlled. Sensory results are highly correlated with TBARS, indicating that Hb effectively catalyzed oxidation of the washed minced tilapia muscle system. In addition, the high co rrela tion of sensory results with a* values further verifies the use of sensory analysis for detecting oxidation of seafood products. In conclusion, Hb form and concentration, pH temperature, NaCl concentration, all affec ted the oxidative activity of o xy CO and m et Hb The results of this research suggest that tilapia minced muscle treated with C O could have greater stability to oxidation, thus resulting in increased color stability and sensory quality.

PAGE 216

216 APPENDIX A RESEARCH SCHEME Hb Oxy Hb CO Hb Met Hb NaCl 150 mM 450 mM pH 6.8 6.3 7.3 Temperature 25C 3.7 C PV TBARS Protein oxidation Loss of CO Extract and analysis (Hb oxidation state) Hemin loss Sensory analysis Color analysis Mechanism of Oxidation 0.0 mM Hb Conc. 6, 9, and 12 mol/kg Tilapia Washed Muscle Model System

PAGE 217

217 A PPENDIX B CHANGES IN L* VALUE AND B* VALUE Table B 1. Changes in L*value in washed tilapia muscle containing different forms of Hb at a concentration of 6, 9, and 12mol/kg muscle at 3.7C averaged across pH level and NaCl concentration. L* value 6mol Hb/kg muscle Storage time (day) Oxy Hb b CO Hb c Met Hb a 0 74.81.5 74.43.3 78.11.8 2 75.01.4 74.43.6 77.31.8 4 75.91.9 74.33.6 76.81.7 6 76.02.6 74.04.4 77.11.7 8 76.02.9 73.05.1 78.01.9 L* value 9mol Hb/kg muscle Storage time (day) Oxy Hb c CO Hb b Met Hb a 0 70.82.0 73.72.4 74.62.0 2 71.11.8 73.22.3 74.02.3 4 71.92.3 73.12.6 73.51.6 6 72.33.0 73.63.0 73.91.8 8 72.43.6 73.53.1 75.12.6 L* value 12mol Hb/kg muscle Storage time (day) Oxy Hb b CO Hb a Met Hb a 0 68.02.0 71.72.8 71.91.9 2 68.21.9 71.13.1 70.61.6 4 69.22.4 71.12.9 70.31.3 6 69.43.1 72.13.0 70.71.6 8 69.63.8 71.93.7 72.22.3 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n=18). Letters within the same rows indicate statistically (<0.05) significant differences separated by Tukeys HSD.

PAGE 218

218 Table B 2. Changes in L*value in w ashed tilapia muscle containing different forms of Hb at a concentration of 6, 9, and 12mol/kg muscle at 25C averaged across pH level and NaCl concentration. L* value 6mol Hb/kg muscle Storage time (weeks) Oxy Hb b CO Hb c Met Hb a 0 74.81.5 74.43.3 78.11.8 4 73.20.8 71.33.8 75.61.0 8 73.41.6 72.03.8 75.00.7 12 71.52.9 72.43.3 75.40.9 16 73.71.3 71.74.0 75.90.9 20 75.10.9 72.03.9 75.91.0 24 76.61.8 70.64.8 76.11.2 L* value 9mol Hb/kg muscle Storage time (day) Oxy Hb b CO Hb b Met Hb a 0 70.82.0 73.72.4 74.62.0 4 69.51.3 69.72.8 71.81.3 8 69.91.8 69.82.8 71.41.2 12 68.91.8 70.72.4 71.71.2 16 69.71.1 70.32.6 72.11.2 20 71.82.0 69.92.9 72.21.3 24 73.13.2 69.13.0 71.81.6 L* value 12mol Hb/kg muscle Storage time (day) Oxy Hb b CO Hb a Met Hb a 0 68.02.0 71.72.8 71.91.9 4 65.51.0 68.12.8 68.20.9 8 66.01.2 68.13.3 67.61.0 12 66.53.5 69.02.7 67.61.4 16 65.82.0 68.32.7 68.31.3 20 67.63.3 68.12.6 68.11.5 24 68.83.5 68.62.6 67.81.5 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n=18). Letters within the same rows indicate statistically (<0.05) significant differences separated by Tukeys HSD.

PAGE 219

219 Table B 3. Changes in b*value in washed tilapia muscle containing different forms of Hb at a concentration of 6, 9, and 12mol/kg muscle at 3.7C averaged across pH level and NaCl concentration. b* value 6mol Hb/kg muscle Storage time (day) Oxy Hb b CO Hb b Met Hb a 0 1.81.4 1.81.5 2.61.0 2 2.71.7 3.72.0 4.00.9 4 4.41.8 4.22.2 4.41.0 6 4.71.6 4.72.6 5.21.7 8 4.42.6 5.12.2 6.81.8 b* value 9mol Hb/kg muscle Storage time (day) Oxy Hb b CO Hb b Met Hb a 0 3.41.8 3.11.5 3.81.2 2 4.52.0 4.61.7 5.21.0 4 6.52.2 5.22.6 6.21.6 6 7.12.5 6.32.5 7.52.0 8 6.43.2 7.23.2 9.31.6 b* value 12mol Hb/kg muscle Storage time (day) Oxy Hb b CO Hb c Met Hb a 0 4.72.3 3.91.8 5.01.6 2 5.82.1 5.01.7 6.61.4 4 7.72.1 5.52.5 7.61.8 6 8.52.4 7.42.7 9.02.2 8 7.83.5 8.62.8 10.52.1 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n=18). Letters within the same rows indicate statistically (<0.05) significant differences separated by Tukeys HSD.

PAGE 220

220 Table B 4. Changes in b*value in washed tilapia muscle containing different forms of Hb at a concentration of 6, 9, and 12mol/kg muscle at 3.7C averaged across pH level and NaCl concentration. b* value 6mol Hb/kg muscle Storage time (day) Oxy Hba CO Hbc Met Hbb 0 1.81.4 1.81.5 2.61.0 4 8.71.6 5.11.7 5.60.5 8 9.92.0 5.21.7 6.31.2 12 10.13.3 4.31.4 6.11.4 16 7.44.4 3.02.0 6.71.8 20 6.02.4 2.21.4 7.01.5 24 6.23.2 3.02.0 6.91.8 b* value 9mol Hb/kg muscle Storage time (day) Oxy Hba CO Hbc Met Hbb 0 3.41.8 3.11.5 3.81.2 4 11.01.5 7.12.2 7.80.8 8 12.11.9 7.12.2 8.41.7 12 12.22.5 5.71.7 8.62.0 16 9.94.0 4.61.9 9.42.0 20 8.42.2 3.62.1 9.12.1 24 8.22.5 4.62.0 9.92.6 b* value 12mol Hb/kg muscle Storage time (day) Oxy Hba CO Hbc Met Hbb 0 4.72.3 3.91.8 5.01.6 4 13.62.1 8.01.7 10.11.4 8 14.42.2 7.81.7 11.01.7 12 16.24.5 7.32.4 11.82.2 16 12.44.1 6.32.0 12.02.6 20 11.02.1 5.32.5 11.92.8 24 10.92.6 4.91.3 12.82.4 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n=18). Letters within the same rows indicate statistically (<0.05) significant differences separated by Tukeys HSD.

PAGE 221

221 Table B 5 Changes in L*value in washed tilapia muscle containing different forms of H b at pH 6.3, 6.8, and 7.3 at 3.7C averaged across Hb and NaCl concentration. L* value pH 6.3 Storage time (day) Oxy Hb a CO Hb a Met Hb b 0 71.93.8 74.33.2 73.93.0 2 72.53.2 73.53.9 73.02.9 4 74.92.9 74.63.0 73.42.3 6 76.32.9 76.32.3 74.52.1 8 76.82.9 75.92.8 76.22.3 L* value pH 6.8 Storage time (day) Oxy Hb c CO Hb c Met Hb a 0 71.03.1 72.53.0 73.92.8 2 70.93.3 72.83.2 72.73.0 4 71.23.1 71.43.2 72.02.8 6 71.32.8 71.73.0 72.03.0 8 71.53.1 71.63.7 72.52.9 L* value pH 7.3 Storage time (day) Oxy Hb c CO Hb b Met Hb a 0 70.73.2 73.02.8 76.92.9 2 70.83.2 72.32.9 76.13.2 4 70.83.1 72.53.0 75.23.3 6 70.23.1 71.63.2 75.23.3 8 69.63.3 70.93.6 76.62.9 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n=18). Letters within the same rows indicate statistically (<0.05) significant differences separated by Tukeys HSD.

PAGE 222

222 Table B 6 Changes in L*value in washed tilapia muscle containing different forms of H b at pH 6.3, 6.8, and 7.3 at 25C averaged across Hb and NaCl concentration. b* value pH 6.3 Storage time (day) Oxy Hb b CO Hb b Met Hb a 0 1.31.7 1.81.5 3.61.7 2 4.13.4 3.51.7 4.71.6 4 7.43.5 5.43.5 6.02.5 6 7.53.0 7.53.6 7.72.6 8 7.43.1 8.43.3 8.12.5 b* value pH 6.8 Storage time (day) Oxy Hb b CO Hb c Met Hb a 0 4.01.6 2.61.3 3.71.5 2 4.31.5 3.91.4 5.41.3 4 5.41.3 3.61.1 5.91.7 6 6.93.1 5.52.5 7.72.8 8 6.84.3 7.82.9 9.52.3 b* value pH 7.3 Storage time (day) Oxy Hb b CO Hb b Met Hb a 0 4.61.6 4.41.5 4.21.6 2 4.71.6 5.91.5 5.71.6 4 5.71.4 6.01.4 6.21.8 6 6.01.5 5.41.2 6.21.8 8 4.41.6 4.71.3 9.12.3 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n=18). Letters within the same rows indicate statistically (<0.05) significant differences separated by Tukeys HSD.

PAGE 223

223 Table B 7 Changes in b*value in washed tilapia muscle containing different forms of H b at pH 6.3, 6.8, and 7.3 at 3.7C averaged across Hb and NaCl concentration. b* value pH 6.3 Storage time (day) Oxy Hb b CO Hb b Met Hb a 0 1.31.7 1.81.5 3.61.7 2 4.13.4 3.51.7 4.71.6 4 7.43.5 5.43.5 6.02.5 6 7.53.0 7.53.6 7.72.6 8 7.43.1 8.43.3 8.12.5 b* value pH 6.8 Storage time (day) Oxy Hb b CO Hb c Met Hb a 0 4.01.6 2.61.3 3.71.5 2 4.31.5 3.91.4 5.41.3 4 5.41.3 3.61.1 5.91.7 6 6.93.1 5.52.5 7.72.8 8 6.84.3 7.82.9 9.52.3 b* value pH 7.3 Storage time (day) Oxy Hb b CO Hb b Met Hb a 0 4.61.6 4.41.5 4.21.6 2 4.71.6 5.91.5 5.71.6 4 5.71.4 6.01.4 6.21.8 6 6.01.5 5.41.2 6.21.8 8 4.41.6 4.71.3 9.12.3 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n=18). Letters within the same rows indicate statistically (<0.05) significant differences separated by Tukeys HSD.

PAGE 224

224 Table B 8 Changes in b*value in washed tilapia muscle containing different forms of H b at pH 6.3, 6.8, and 7.3 at 25C averaged across Hb and NaCl concentration. b* value pH 6.3 Storage time (day) Oxy Hb a CO Hb c Met Hb b 0 1.31.7 1.81.5 3.61.7 4 9.12.2 5.81.8 7.11.6 8 10.02.6 6.02.1 8.42.4 12 9.02.7 5.22.2 7.62.9 16 8.62.0 3.32.2 8.22.9 20 9.82.1 2.62.4 8.02.9 24 11.01.8 2.81.6 9.22.8 b* value pH 6.8 Storage time (day) Oxy Hb a CO Hb c Met Hb b 0 4.01.6 2.61.3 3.71.5 4 12.22.0 6.52.0 8.12.1 8 13.41.6 6.42.0 9.72.7 12 16.04.0 5.42.1 10.23.0 16 14.93.2 5.22.1 11.22.9 20 8.93.4 4.12.1 10.53.1 24 8.43.4 4.61.6 10.64.0 b* value pH 7.3 Storage time (day) Oxy Hb a CO Hb c Met Hb a 0 4.61.6 4.41.5 4.21.6 4 12.12.5 7.92.4 8.32.4 8 13.02.6 7.62.2 7.62.0 12 13.62.9 6.62.2 8.62.6 16 6.22.9 5.42.3 8.72.5 20 6.62.6 4.52.2 9.52.5 24 5.92.5 5.11.9 9.83.0 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n=18). Letters within the same rows indicate statistically (<0.05) significant differences separated by Tukeys HSD.

PAGE 225

225 Table B 9 Changes in L*value in washed tilapia muscle containing different forms of Hb at concentrations of added NaCl ( 0, 150, and 450mM ) at 3.7C averaged across pH level and Hb concentration. L* value 0mM NaCl Storage time (day) Oxy Hb c CO Hb a Met Hb b 0 72.83.0 75.91.7 75.82.6 2 72.73.0 75.72.1 74.63.0 4 73.23.4 75.32.2 73.93.0 6 73.44.3 75.31.5 73.23.3 8 73.14.9 75.22.1 74.23.5 L* value 150.0mM NaCl Storage time (day) Oxy Hb c CO Hb b Met Hb a 0 71.82.8 73.42.2 75.53.0 2 71.92.8 72.92.5 74.53.4 4 72.73.3 72.92.4 74.03.1 6 72.83.8 73.52.7 74.63.1 8 72.64.2 72.83.3 75.62.9 L* value 450.0mM NaCl Storage time (day) Oxy Hb b CO Hb b Met Hb a 0 69.03.2 70.52.3 73.43.4 2 69.63.3 70.02.5 72.83.6 4 71.03.7 70.43.3 72.83.2 6 71.63.8 70.74.3 73.93.0 8 72.14.0 70.44.8 75.53.4 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n=18). Letters within the same rows indicate statistically (<0.05) significant differences separated by Tukeys HSD.

PAGE 226

226 Table B 10 Changes in L*value in washed til apia muscle containing different forms of Hb at concentrations of added NaCl (0, 150, and 450mM) at 25 C averaged across pH level and Hb concentration. L* value 0mM NaCl Storage time (weeks) Oxy Hb b CO Hb a Met Hb a 0 72.83.0 75.9 1.7 75.82.6 4 68.53.5 71.92.9 71.83.2 8 69.03.3 72.43.3 71.73.3 12 69.83.4 73.32.0 72.13.4 16 69.23.8 72.52.6 72.73.1 20 70.44.0 72.72.2 72.63.1 24 72.34.8 71.92.4 72.93.4 L* value 150.0mM NaCl Storage time (day) Oxy Hb b CO Hb b Met Hb a 0 71.82.8 73.42.2 75.53.0 4 69.93.2 70.22.2 72.13.3 8 70.23.2 70.52.4 71.23.3 12 69.93.1 70.92.2 71.63.4 16 70.03.7 70.22.7 71.93.5 20 72.13.4 70.12.7 71.83.5 24 73.34.0 69.42.6 71.93.5 L* value 450.0mM NaCl Storage time (day) Oxy Hb b CO Hb c Met Hb a 0 69.03.2 70.52.3 73.43.4 4 69.83.3 67.13.1 71.73.3 8 70.23.8 67.03.1 71.23.1 12 67.13.2 68.02.6 70.93.5 16 70.03.4 67.53.0 71.63.6 20 72.04.0 67.13.2 71.74.0 24 72.94.4 67.14.1 70.83.9 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n=18). Letters within the same rows indicate statistically (<0.05) significant differences separated by Tukeys HSD.

PAGE 227

227 Table B 11 Changes in b*value in washed tilapia muscle containing different forms of Hb at concentrations of added NaCl ( 0, 150, and 450mM ) at 3.7C averaged across pH level and Hb concentration. b* value 0mM NaCl Storage time (day) Oxy Hb b CO Hb c Met Hb a 0 2.32.4 1.71.3 2.51.3 2 2.82.3 2.71.2 4.21.5 4 4.71.7 2.81.3 4.71.6 6 5.11.5 3.91.5 5.31.5 8 4.11.1 5.52.1 6.92.1 b* value 150.0mM NaCl Storage time (day) Oxy Hb b CO Hb b Met Hb a 0 3.11.9 2.91.6 3.71.0 2 4.31.4 4.51.3 5.31.1 4 6.31.9 5.01.5 6.01.7 6 6.41.8 6.52.2 7.52.4 8 5.12.4 7.03.1 9.21.8 b* value 450.0mM NaCl Storage time (day) Oxy Hb a CO Hb b Met Hb a 0 4.51.7 4.11.8 5.11.3 2 6.02.0 6.01.3 6.21.3 4 7.52.8 7.22.1 7.41.7 6 8.83.0 8.12.7 8.82.2 8 9.43.3 8.53.3 10.61.7 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n=18). Letters within the same rows indicate statistically (<0.05) significant differences separated by Tukeys HSD.

PAGE 228

228 Table B 12 Changes in b*value in washed tilapia muscle contai ning different forms of Hb at concentrations of added NaCl ( 0, 150, and 450mM ) at 25C averaged across pH level and Hb concentration. b* value 0mM NaCl Storage time (day) Oxy Hb a CO Hb c Met Hb b 0 2.32.4 1.71.3 2.51.3 4 10.83.0 5.41.4 7.32.2 8 11.23.1 5.21.3 7.12.2 12 11.23.8 4.21.3 7.22.5 16 8.63.9 3.21.6 7.62.2 20 6.72.1 2.01.5 7.42.4 24 6.72.5 3.01.3 7.52.5 b* value 150.0mM NaCl Storage time (day) Oxy Hb a CO Hb c Met Hb b 0 3.11.9 2.91.6 3.71.0 4 11.22.7 6.41.6 8.32.1 8 12.12.6 6.11.5 9.32.4 12 12.74.0 5.31.8 9.42.7 16 9.63.8 4.01.7 10.53.1 20 8.62.7 3.41.9 10.42.8 24 8.93.4 4.21.3 10.52.6 b* value 450.0mM NaCl Storage time (day) Oxy Hb a CO Hb c Met Hb a 0 4.51.7 4.11.8 5.11.3 4 11.42.3 8.42.5 7.91.9 8 13.02.2 8.72.0 9.32.3 12 14.74.7 7.71.9 9.93.1 16 11.55.7 6.72.2 10.12.9 20 10.03.3 5.72.1 10.22.9 24 9.63.5 5.32.4 11.63.3 Values are means Standard Deviations for all pixels of the surface of the minced muscle samples (n=18). Letters within the same rows indicate statistically (<0.05) significant differences separated by Tukeys HSD.

PAGE 229

229 APPENDIX C PEARSON CORRELATION COEFFICIENT Table C 1 Pearson correlation c oefficient for s amples stored at 3.7 C for 8 day s* TBARS PV Carbon yls Oxy Hb Met Hb Deoxy Hb L* value a* value b* value %CO TBARS 1 0.84* 0.02 0.64* 0.65* 0.48* 0.36* 0.25* 0.22* 0.33* PV 0.84* 1 0.06 0.73* 0.73* 0.60* 0.22* 0.37* 0.06 0.46* Carbonyls 0.02 0.06 1 0.43* 0.41* 0.37* 0.08 0.22* 0.20* 0.19* Oxy Hb 0.64* 0.73* 0.43* 1 0.98* 0.85* 0.08 0.50* 0.27* 0.60* Met Hb 0.65* 0.73* 0.41* 0.98* 1 0.75* 0.09 0.48* 0.26* 0.58* Deoxy Hb 0.48* 0.60* 0.37* 0.85* 0.75* 1 0.01 0.45* 0.25* 0.54* L* value 0.36* 0.22* 0.08 0.08 0.09 0.01 1 0.59* 0.69* 0.10 a* value 0.25* 0.37* 0.22* 0.50* 0.48* 0.45* 0.59* 1 0.50* 0.43* b* value 0.22* 0.06 0.20* 0.27* 0.26* 0.25* 0.69* 0.50* 1 0.35* %CO 0.33* 0.46* 0.19* 0.60* 0.58* 0.54* 0.10* 0.43* 0.35* 1 Number of observations (N=810), P

PAGE 230

230 Table C 2 Pearson correlation coefficient for samples stored at 25C for 24 weeks TBARS PV Carbony ls Oxy Hb Met Hb Deoxy Hb L* value a* value b* value %CO TBARS 1 0.73* 0.07 0.70* 0.71* 0.59* 0.12 0.62* 0.46* 0.12* PV 0.73* 1 0.08 0.70* 0.70* 0.63* 0.11 0.71* 0.62* 0.15* Carbonyls 0.07 0.08 1 0.22* 0.19* 0.24* 0.17* 0.12 0.04 0.15* Oxy Hb 0.70* 0.70* 0.22* 1 0.99* 0.90* 0.16* 0.75* 0.36* 0.13* Met Hb 0.71* 0.70* 0.19* 0.99* 1 0.82* 0.14* 0.74* 0.36* 0.10 Deoxy Hb 0.59* 0.63* 0.24* 0.90* 0.82* 1 0.17* 0.67* 0.29* 0.21* L* value 0.12 0.11 0.17* 0.16* 0.14* 0.17* 1 0.52* 0.19* 0.08 a* value 0.62* 0.71* 0.12 0.75* 0.74* 0.67* 0.52* 1 0.37* 0.19* b* value 0.46* 0.62* 0.04 0.36* 0.36* 0.29* 0.19* 0.37* 1 0.06 %CO 0.12* 0.15* 0.15* 0.13* 0.10 0.21* 0.08 0.19* 0.06 1 Number of observations (N=1134), P

PAGE 231

231 APPENDIX D TRIANGLE TEST BALLOT Name: Date: Product: Tilapia washed muscle with 12mol/kg hemoglobin Please sniff the samples from left to right, in the order presented. Two samples are the same and one is different; please determine which is the odd sample and circle the code that represents it. If you cannot differentiate between the samples, please tak e a guess. Instruction: Which is the odd sample? 346 576 021

PAGE 232

232 APPENDIX E DESCRIPTIVE ANALYSIS BALLOT Sample: Fish muscle Date: Name: Panelist #: Attribute: painty/rancid odor Please sniff/smell samples in front of you in the order presented and r ate the intensity of the painty/rancid odor using the 15 point scale presented below (none to strong) Strong None Strong None Strong None Strong None Strong None Strong None Strong None Strong None Strong None

PAGE 233

233 LIST OF REFERENCES (1) Richards, M.; Dettmann, M. Comparative analysis of different hemoglobins: autoxidation, reaction with peroxide, and lipid oxidation. J. Agric. Food Chem 2003, 51 3886 3891. (2) German, J. Food processing and lipid oxidation. Advances in experimental Med. Biol 1999, 459, 2350. (3) Miller, D. Food chemistry John Wiley & Sons : New York 1998; p 168. (4) Richards, M.; Hultin, H. Effect of pH on lipid oxidation using trout hemolysate as a catalyst: a possible role for deoxyhemoglobin. J. Agric. Food Chem 2000, 48, 31413147. (5) Baron, C.; Andersen, H. Myoglobin induced lipid oxidation. A review. J. Agric. Food Chem 2002, 50 3887 3897. (6) Simic, M. Oxygen radicals in biology and medicine. Plenum Press New York: 1988. (7) Kanner, J. Oxidative processes in meat and meat products: quality implications. Meat Sci 1994, 36, 169189. (8) Undeland, I.; Hultin, H.; Richardss, M. A dded triacylglycerols do not hasten hemoglobinmediated lipid oxidation in washed minced cod muscle. J. Agric. Food Chem 2002, 50 6847 6853. (9) Dugan, L. R., Jr. The science of meat and meat products Freeman CO., : San Francisco, California, 1971; p p 1 33. ( 10) Kristinsson, H.; Hultin, H. The effect of acid and alkali unfolding and subsequent refolding on the pro oxidative activity of trout hemoglobin. J. Agric. Food Chem 2004, 52 5482 5490. (11) Rich ards, M.; Hultin, H. Contributions of blood and blood components to lipid oxidation in fish muscle. J. Agric. Food Chem 2002, 50 555 564. (12) Kranen, R.; Van Kuppevelt, T.; Goedhart, H.; Veerkamp, C.; Lambooy, E.; Veerkamp, J. Hemoglobin and myoglobin content in muscles of broiler chickens. Poultry Sci 1999, 78, 467 472 (13) Port er, P.; Kennish, J.; Kramer, D. The effects of exsanguination of sockeye salmon on the changes n the lipid composition during frozen storage. EG Bligh eds Seafood and Tech. 1992, 76 84.

PAGE 234

234 (14) Hultin, H. Oxidation of lipids in seafoods In Seafood: chemistry, processing, technology and quality ; Shahidi, F., Botta, J. R., Eds.; Blacki Academic and Professional: London, 1994; p p 49 74. (15) Bakker, G.; Boyer, R. Iron incorporation into apoferritin. The role of apoferritin as a ferroxidase. J. Biol Chem 1986, 261 13182 13185. (16) Frankel, E. Methods to determine extent of oxidation. In Lipid Oxidation. 2nd edition ed.; The Oily Press. PJ Barnes & A ssociates: B ridgewater, England, 2005; p p 99 127. (17) Silva, J.; Chamul, R. Composition of marine and freshwater finfish and shellfish species and their products. 2000 31 45. (18) Exler, J.; Weihrauch, J. Comprehensive evaluation of fatty acids in foods. J. the American Dietetic Assn. 1977, 71, 518 525 (19) Kiessling, A.; Storebakken, T.; Aasgaard, T.; Kiessling, K. Changes in the structure and function of the epaxial muscle of rainbow trout (Oncorhynchus mykiss) in relation to ration and age. 1. Growth dynamics. Aquaculture (Netherlands) 1991, 93, 335356. (20) Hsieh, R.; Kinsella, J. Oxidation of polyunsaturated fatty acids: mechanisms, products, and inhibition with emphasis on fish. Advances in food and nutrition research (USA) 1989. (21) Hultin, H. TrimethylamineN oxide (TMAO) demethylation and protein denaturation in fish muscle. In Advances in seafood biochemistry: composition and quality; Technomic Publishing Co. : Lancaster, 1995; p p 2542. (22) Undeland, I.; Kristinsson, H.; Hultin, H. Hemoglobinmediated oxidation of washed minced cod muscle phospholipids: Effect of pH and hemoglobin source. J. Agric. Food Chem 2004, 52, 44444451. (23) Richards, M.; Hultin, H. Rancidity development in a fish model system as affected by phospholipi ds. J. Food Lipids 2001, 8 215 230. (24) Decker, E.; Warner K.; Richards, M.; Shahidi, F. Measuring antioxidant effectiveness in food. USDA, ARS. 2005. (25) Decker, E.; Hultin, H. Factors influencing catalysis of lipid oxidation by the soluble fraction of mackerel muscle. J. Food Sci 1990, 55, 947 950. (26) Phillippy, B.; Hultin, H. Some factors involved in trimethylamine N oxide (TMAO) demethylation in post mortem red hake muscle. J. Food Biochem 1993, 17, 251 266.

PAGE 235

235 (27) Jia, T.; Kelleher, S.; Hultin, H.; Petill o, 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, 1195 1201. (28) Hwang, K.; Regenstein, J. Protection of m enhaden mince lipids from rancidity during frozen storage. J. Food Sci 1989, 54, 11201124. (29) Fujii, T.; Hirayama, M.; Okuzumi, M.; Yasuda M.; Nishino, H.; Yokoyama, M. Shelf life studies on fresh sardine [Sardinops melanostictus] packaged with carbon dioxidenitrogen gas mixture. Bulletin of the Japanese Society of Scientific Fisheries (Japan) 1989. (30) Kjaersgrd, I.; Nrrel ykke, M.; Baron, C.; Jessen, F. Identification of carbonylated protein in frozen rainbow trout (Oncorhynchus mykiss) fillets and development of protein oxidation during frozen storage. J. Agric. Food Chem 2006, 54 94379443. (31) Baron, C.; Kjaersgr d, I.; Jessen, F.; Jacobsen, C. Protein and lipid oxidation during frozen storage of rainbow trout (Oncorhynchus mykiss). J. Agric. Food Chem 2007, 55, 8118 8127 (32) Tironi, V.; Tomas, M.; Anon, M. Structural and functional changes in myofibrillar proteins of sea salmon (Pseudopercis semifasciata) by interaction with malonaldehyde (RI). J. Food Sci 2002, 67 929 935. (33) Dalle Donne, I.; Rossi, R.; Giustarin i, D.; Milzani, A.; Colombo, R. Protein carbonyl groups as biomarkers of oxidative stress. Clinica Chimica Acta 2003, 329, 23 38. (34) stdal, H.; Davies, M.; Anderse n, H. Reaction between protein radicals and other b iomolecules. Free Radical Biol. Med. 2002, 33 201209. (35) Eyma rd, S.; Baron, C.; Jacobsen, C. Oxidation of lipid and protein in horse mackerel (Trachurus trachurus) mince and washed minces during processing and storage. Food Chem 2009, 114 57 65. (3 6) Levine, R.; Williams J.; Stadtman, E.; Shacter, E. Carbonyl assays for determination of oxidatively modified proteins. Methods in Enzymol 1994, 233, 346357. (37) Anderson, C.; Wu, W. Analysis of carbon monoxide in commercially treated Tuna (Thunnus spp.) and Mahi Mahi (Coryphaena hippurus) by gas chromatography/mass spectrometry. J. Agric. Food Chem 2005, 53 7019 7023.

PAGE 236

236 (38) Grunwald, E.; Richards, M. Mechanisms of heme proteinmedi ated lipid oxidation using hemoglobin and myoglobin variants in raw and heated washed muscle. J. Agric. Food Chem 2006, 54, 82718280. (39) Kanner, J.; German, J.; Kinsella, J. Initiation of lipid peroxidation in biological systems. CRC Critical Rev. in Food Sci. Nutr 1987, 25 317 364. (40) Puppo, A.; Halliwell, B. Formation of hydroxyl radicals from hydrogen peroxide in the presence of iron. Biochem. J 1988, 249, 185190. (41) Baron, C.; Skibsted, L.; Andersen, H. Peroxidation of linoleate at physio logical pH: hemichrome formation by substrate binding protects against metmyoglobin activation by hydrogen peroxide. Free Radical Biol. Med. 2000, 28, 549 558. (42) Anders en, H.; Pellett, L.; Tappel, A. Hemichrome formation, lipid peroxidation, enzyme inactivation and protein degradation as indexes of oxidative damage in homogenates of chicken kidney and liver. Chemico biological Int 1994, 93 155 164. (43) Stryer, L. Oxygentransporting proteins: myoglobin and haemoglobin. In Biochemistry ; 3rd ed.; WH Freeman and Company: New York 1988; p p 143 176. (44) Levy, A.; Rifkind, J. Low temperature formation of a distal histidine complex in hemoglobin: a probe for heme pocket flexibility. Biochem 1985, 24, 6050 6054. (45) Liong, E.; Dou, Y.; Scott, E.; Olson, J.; Phillips, G. Waterproofing the heme pocket role Of proximal amino acid side chains in preventing hemin loss from myoglobin. J. Biol. Chem 2001, 276, 90939100. (46) Kanner, J.; Harel, S. Initiati on of membranal lipid peroxidation by activated metmyoglobin and methemoglobin. Arch. Biochem. Biophys 1985, 237, 314 326 (47) Kanner, J.; Harel, S. Lipid peroxidation and oxidation of several compounds by H2O2 activated metmyoglobin. Lipids 1985, 20 625634. (48) Uno, T.; Takeda, A.; Sh imabayashi, S. Effects of imidazoles and pH on the peroxidase activity of the heminhydrogen peroxide system. Inorganic Chem 1995, 34 1599 1607. (49) S ato K.; H egarty G., Warmed over flavor in cooked meats. J. Foo d Sci 1971, 36, 1098 1102. (50) Monahan, F.; Crackel, R.; Gray, J.; Buckley, D.; Morrissey, P. Catalysis of lipid oxidation in muscle model systems by haem and inorganic iron. Meat Sci 1993, 34, 95 106.

PAGE 237

237 (51) Love, J.; Pearson, A. Metmyoglobin and nonheme iron as prooxidants in cooked meat. J. Agric. Food Chem 1974 22, 10321034. (52) Tichivangana, J. Z.; Morrissey, P. A. Metmyoglobin and inorganic metals as pro oxidants in raw and cooked muscle systems. Meat Sci 1985, 15, 1071 18. (53) Yos hida, Y.; Kashiba, K.; Niki, E. Free radical mediated oxidation of lipids induced by hemoglobin in aqueous dispersions. Biochimica et Biophysica Acta. 1994, 1201, 165 181 (54) Gorbenko, G. Bromothymol blue as a probe for structural changes of model membranes induc ed by hemoglobin. BBABiomembranes 1998, 1370, 107 118. (55) Richards, M.; Modra, A.; 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. (56) Richards, M.; Dettmann, M.; Gr unwald, E. Pro oxidative characteristics of trout hemoglobin and myoglobin: a role for released heme in oxidation of lipids. J. Agric. Food Chem 2005, 53, 1023110238. (57) Krist insson, H.; Mony, S.; Petty, H. Properties of tilapia carboxy and oxyhemoglobin at postmortem pH. J. Agric. Food Chem 2005, 53, 36433649. (58) Grunwald, E.; Richards, M. Studies with myoglobin variants indicate that released hemin is the primary promoter of lipid oxidation in washed fish muscle. J. Agric. Food Chem 2006, 54, 44524460. (59) Hargrove, M.; Krzywda, S.; Wilkinson, A.; Dou, Y.; Ikeda Saito, M.; Olson, J. St ability of myoglobin: a model for the folding of heme proteins. Biochem 1994, 33, 1176711775. (60) Aranda IV, R.; Cai, H.; Worley, C.; Levin, E.; Li, R.; Olson, J.; Phillips Jr, G.; Richards, M. Structural analysis of fish versus mammalian hemoglobins: Effect of the heme pocket environment on autooxidation and hemin loss. Proteins: Structure, Function, and Bioinformatics 2009, 75, 217 230. (61) Hargrov e, M.; Wilkinson, A.; Olson, J. Structural factors governing hemin dissociation from metmyoglobin. Bio chem 1996, 35, 11300 11309. (62) Hargrove, M.; Whitaker, T.; O lson, J.; Vali, R.; Mathews, A. Quaternary structure regulates hemin dissociation from human hemoglobin. J. Biol. Chem 1997, 272 1738517389.

PAGE 238

238 (63) Aranda, R.; Levin, E.; Schotte, F.; Anfinrud, P.; Phillips, G. Time dependent atomic coordinates for the dissociation of carbon monoxide from myoglobin. Acta Crystallographica Section D: Biol Crystallography 2006 62, 776 783. (64) Arihara, K.; Cassens, R.; Greaser, M.; Luchansky, J.; Mozdziak, P. Localization of metmyoglobin reducing enzyme (NADH cytochrome b5 reductase) system components in bovine skeletal muscle. Meat Sci 1995, 39, 205213. (65) Bekhit, A.; Faustman, C. Metmyoglobin reducing activity. Meat Sci 2005, 71, 407439. (66) Reddy, L.; Carpenter, C. Determination of metmyoglobin reductase activity in bovine skeletal muscles. J. Food Sci 1991, 56 1161 1164. (67) Taylor, D.; Hochstein, P. Inhibition by adriamycin of a metmyoglobin reductase from beef heart. Biochem. pharmacology 1978 27, 20792092. (68) Hag ler, L.; Coppes, R.; Herman, R. Metmyoglobin reductase. Identification and purification of a reduced nicotinamide adenine dinucleotide dependent enzyme from bovine heart which reduces metmyoglobin. J. Biol. Chem 1979, 254, 65056514. (69) Kristinsson, H.; Mony, S.; Demir N.; Balaban, M.; Otwell, W. The effect of carbon monoxide and filtered smoke on the properties of aquatic muscle and selected muscle components 1st Joint Trans Atlantic Fisheries Conf., TAFT : R eyjavik, Iceland, 2003; Available at: http://www.rfisk.is/taft2003/First.htm: (70) Garner, K. The effects of carbon monoxide on the muscle quality of Spanish mackerel MS thesis Department of Food Sc ience and Human Nutrition, University of Florida, Gainesville, FL, 2004. (71) Danyali, N. The effect of carbon monoxide and filtered smoke on quality and safety of yellowfin tuna. MS t hesis Department of Food Science and Human Nutrition, University of Florida, Gainesville, FL, 2004. (72) Kristinsson, H.; Ludlow, N.; Ba laban, M.; Otwell, W.; Welt, B. Muscle quality of yellowfin tuna (Thunnus albacares) steaks after treatment with carbon monoxide gases and filtered wood smoke. J. Aquatic Food Prod. Technol 2006, 15, 49 67. (73) Garner, K.; Kristinsson, H. Quality of Spanish mackerel (Scomberomorous maculatus) muscle as affected by carbon monoxide and filtered smoke gas treatment IFT Annual Meeting; Institute of Food Technologists, July 1216, 2004. (74) Srheim O.; Nissen, H.; Nesbakken, T. The storage life of beef and pork packaged in an atmosphere with low carbon monoxide and high carbon dioxide. Meat Sci 1999, 52, 157164.

PAGE 239

239 (75) Kropf, D. Effects of retail display conditions on meat color Reciprocal Meat Conference Proceedings 1980; pp 1532. (76) Phillips, C. Review: modified atmosphere packaging and its effects on the microbiological quality and safety of produce. Intl. J. Food Sci. Technol 1996, 31 463479. (77) FDA. Agency response letter GRAS Notice NO. GRN. 000015. 2002, Available at: http://www.cfsan.fda.gov/Rdb/opag143.html (78) FDA. Agency response letter GRAS Notice NO. GRN. 000143. 2004, Available at: http://www.cfsan.fda.gov/Rdb/opag143.html (79) Otwell, W Use of carbon monoxide and filtered smokes in fish processing. In Modified atmospheric processing and packaging of fish: filtered smokes, c arbon monoxide, and reduced oxygen packaging. Otwell, W.; Kristinsson, H.; Balaban, M. Eds.; Blackwell Publishing: Ames, Iowa, 2006; pp 3 13. (80) Richards, M.; Nelson, N.; Kristinsson, H.; Mon y, S.; Petty, H.; Oliveiras, A. Effects of fish heme protein s tructure and lipid substrate composition on hemoglobinmediated lipid oxidation. J. Agric. Food Chem 2007, 55 3643 3654. (81) Pa zos, M.; Medina, I.; Hultin, H. Effect of pH on hemoglobincatalyzed lipid oxidation in cod muscle membranes in vitro and in situ. J. Agric. Food Chem 2005, 53 3605 3612. (82) Kanner, J.; Salan, M.; Harel, S.; Shegalovich, I. Lipid peroxidation of muscle food: the role of the cytosol ic fraction. J. Agric. Food Chem 1991, 39 242 246. (83) Kanner, J.; Harel S.; Hazan, B. Muscle membranal lipid peroxidation by an" iron redox cycle" system: initiation by oxy radicals and sitespecific mechanism. J. Agric. Food Chem 1986, 34, 506 510. (84) Wallac e, W.; Maxwell, J.; Caughey, W. The mechanisms of hemoglobin autoxidation. Evidence for protonassisted nucleophilic displacement of superoxide by anions. Biochem. Biophysical Res. Communications 1974, 57, 11041116. (85) Harel, S. Oxidation of ascorbic acid and metal ions as affected by NaCl. J. Agric. Food Chem 1994, 42 2402 2406. (86) Fyhn, U.; Fyhn, H.; Davis, B.; Po wers, D.; Fink, W.; Garlick, R. Hemoglobin heterogeneity in Amazonian fishes. Comp. Biochem. Physiol 1979, 62 39 66.

PAGE 240

240 (87) Bradford, M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of proteindye binding. Analytical biochem 1976, 72 248 254. (88) De Young, A.; Kwiatkowski, L.; Noble, R. Fish hemoglobins. Methods in enzymology 1994, 231, 124. (89) Robinson, H.; Hogden, C. The biuret reaction in the determination of serum proteins I. A study of the conditions necessary for the production of a stable color which bears a quantitative relationship to the protein concentration. J. Biol. Chem 1940, 135 707 725. (90) Lee, C.; Trevino, B.; Chaiyawat, M. A simple and rapid solvent extraction method for determining total lipids in fish tissue. J. AOAC Intl 1996, 79 487 492. (91) Anderson, R.; Davis, S. An organic phosphorus assay which avoids the use of hazardous perchloric acid. Clinica Ahimica Acta; Intl. j. Clinical Chem 1982, 121 111121 (92) Decker, E.; Welch, B. Role of ferritin as a lipid oxidation catalyst in muscle food. J. Agric. Food Chem 1990, 38, 674 677. (93) Lemon, D. An improved TBA test for rancidity In: Woyewoda, A.D., KE, P.J. and Burns, B.G., Eds ed.; Halifax Laboratory, Halifax: Nova Scotia, 1975; 51 6 5 72. (94) Kj, I.; Norrel ykke, M.; Baron, C.; Jessen, F. Identification of carbonylated protein in frozen rainbow trout (Oncorhynchus mykiss) fillets and development of protein oxidation during frozen storage. J. Agric. Food Chem 2006, 54, 9437 9446. (95) Krzywicki, K. The determination of haem pigments in meat. Meat Sci 1982, 7 2936. (96) Tang, J.; Faustman, C.; Hoagland, T. Krzywicki revisited: Equations for spectrophotometric determination of myoglobin redox forms in aqueous meat extracts. J. Food Sci 2004, 69, 717 720. (97) Balaban, O.; Luzuriaga, D. Measuring color in foods. Machine vision system provides program for analysis. Resource: Engineering & Technol. for a Sustainable World. 2001, 8 10 11. (98) Foeged ing, E.; Lanier, T.; Hultin, H. Characteristics of edible muscle tissues In Food Chemistr y; Fennema, O., Ed.; Marcel Dekker: New York, 1996; p p 879942. (99) Carlse n, C.; Mller, J.; Skibsted, L. Heme iron in lipid oxidation. Coordination Chem. Rev 2005, 249 485 498.

PAGE 241

241 (100) Baron, C.; Skibsted, L.; Andersen, H. Concentration effects in myoglobin catalyzed peroxidation of linoleate. J. Agric. Food Chem 2002, 50, 883 888. (101) Adams, S.; Green, P.; Claxton, R.; Simcox, S.; Williams, M.; Walsh, K.; Leeuwenburgh, C. Reactive carbonyl formation by oxidative and non oxidative pathways. Front Biosci 2001, 6, A1724. (102) Mantilla, T. Euthanasia of tilapia using carbon monoxide for color fixation and color stabilization. MS thesis, Department of Food Science and Human Nutrition, University of Florida, Gainesville, FL, 2005. (103) Huo, L. Carbon monoxide determination and penetration in treated fish products. PhD Dissertation, Department of Food Science and Human Nutrition, University of Florida, Gainesville, FL, 2008. (104) Riggs, A. Properties of hemoglobins In Fish physiology. The NerVous system, circulation, and respiration; Hoar, W. S., Randall, D. J., Eds.; Academic Press: New York, 1970; Vol. IV, p p 209252. (105) Richard s, M.; Ostdal, H.; Andersen, H. Deoxyhemoglobinmediated lipid oxidation in washed fish muscle. J. Agric. Food Chem 2002, 50 12781283. (106) Hargrove M S ; Singleton, E W ; Quilling M L ; Ortizfl L A ; P hillips, G. N.; Jr.11; S, a. O. J. His64(E7) --Tyr apomyoglobin as a reagent for measuring rates of hemin dissociation*. The J. Biol. Chem 1994, 269, 42074214. (107) Buaneow, C.; Usawakesmanee, W.; Siri pongvutikorn, S.; Tongraung, C. Effect of pH and ATP on lipid oxidation in unwashed and washed seabass mince (Lates calcari fer) mediated by hemoglobin. Songklanakarin J. Sci Technol 2008, 30, 127134 (108) Osinchak, J. ; Hultin, H.; Zajic ek, O.; Kelleher, S.; Huang, C. Effect of NaCl on catalysis of lipid oxidation by the soluble fraction of fish muscle. Free Radical Biol. Med. 1992, 12, 35 47 (109) Nambudiry, D. Lipid oxidation in fatty fish: the effect of salt content in the meat. J. Food Sci. Technol 1980, 17, 176178. (110) Shimizu, Y.; Kir iake, S.; Ohtubo, S.; Sakai, T. Effect of NaCl on protein and lipid oxidation in frozen yellowtail meat. Biosci. Biotechnol. Biochem 2009 73, 923 925. (1 11) Morrison, M.; Schonbaum, G. Peroxidasecatalyzed halogenation. Annu. Rev. Biochem 1976, 45, 861888.

PAGE 242

242 (112) Irzhak, L.; Tiurnin, A. Effect of sodium chloride on the properties of hemoglobin in the erythrocyte. Fiziologicheski zhurnal SSSR imeni IM Sechenova. 1985, 71, 867872 (113) Hernndez, P.; Park, D.; Soon Rhee, K. Chloride salt type/ionic strength, muscle site and refrigeration effects on antioxidant enzymes and lipid oxidation in pork. Meat Sci. 2002, 61, 405410. (114) Anders en, E.; Andersen, M.; Baron, C. Characterization of Oxidative Changes in Salted Herring (Clupea harengus) During Ripening. J. Agric. Food Chem. 2007, 55, 9545 9553. (115) Jayasingh, P.; Cornforth, D.; Carpenter, C.; Whittier, D. Evaluation of carbon monoxide treatment in modified atmosphere packaging or vacuum packaging to increase color stability of fresh beef. Meat Sci. 2001, 59 317 324. (116) Csirke, J. Global production and state of marine fishery resources. Review of the state of world marine fishery resources. FAO Fisheries Technical Paper 2005, 457, 1 9. (117) Botta, J. R. Freshness quality of seafood: a review. In Seafood: chemistry, processing, technology and quality ; Shahidi, F., Botta, J. R., Eds.; Blacki Academic and Professional: London, 1994; pp 140167. (118) Richard s, M.; Kelleher, S.; Hultin, H. Effect of washing with or without antioxidants on quality retention of mackerel fill ets during refrigerated and frozen storage. J. Agric. Food Chem 1998, 46 4363 4371. (119) Fisher, R. Statistical methods for research workers Edinburgh, UK: Oliver and Boyd 1970. (120) Meng, R.; Rubin, X.; Rubin, D. Comparing correlated correlation coefficients. Psychological Bulletin. 1992, 111, 172 175. (121) Garner, K.; Mony, S.; Kristinsson, H. The effect of carbon monoxide on autoxidation, peroxidase and prooxidative activity of hemoglobin from tilapia. IFT An nual Meeting; Institute of Food Technologists Abstract # 425, Chicago, I L, 2003. (122) Suman S.; Mancini, R.; Faustman, C. Lipid oxidation induced carboxymyoglobin oxidation. J. Agric. Food Chem 2006 54, 92489253. (123) Osinchak, J.; Hultin, H.; Zajic ek, O.; Kelleher, S.; Huang, C. Effect of NaCl on catalysis of lipid oxidation by the soluble fraction of fish muscle. Free radical biology & medicine. 1992, 12, 35 47.

PAGE 243

243 BIOGRAPHICAL SKETCH Sara Abdulmajeed Aldaous was born in Riyadh, Saudi Arabia, on October 31, and grew up in Jeddah, on the west coast. She graduated in 1995 from King Abdulaziz University, Jeddah, Saudi Arabia, College of Home Economics, Department of Food and Nutrition. Sara ranked in the top 5 percent of this departments gr adua tes over the past 15 years. From 19952001, she joined the team of Food and Nutrition at King Abdulaziz University, teaching lab cou r ses in Food Science and Nutrition. In 2001 she received a scholarship to study abroad to pursue her m aster s and PhD degrees in food science. In Fall 2003 she joined the Department of Food Science and Human Nutrition at the Univ ersity of Maine, Orono, Maine. After she received her m aster s degree, she joined the Department of Food Science and Human Nutrition at the University of Florida, Gainesville, Florida, in Fall 2005 to pursue her PhD in Food Science. During her graduate career her outs tanding performance in academics was recognized by The Un iversity of Florida with the presentation of the Marilyn Little Altrusa Scholarship for outstanding achievement by an international female. She also presented seven posters specifying findings of her researc h at four different conferences, one of which rec eived 2nd place at the Institute of Food Technologists ( IFT ) Aquatic Food Products Division Paper Competition in 2008.