Development of Simple and Rapid Procedures for Purification of Egg

Iowa State University Capstones, Theses and Retrospective Theses and Dissertations Dissertations

2007 Development of simple and rapid procedures for purification of egg yolk immuno-globulin and ovotransferrin from chicken egg and antimicrobial activity of ovo-transferrin against Escherichia coli O157:H7 and Listeria monocytogenes Kyung Yuk Ko Iowa State University

Follow this and additional works at: https://lib.dr.iastate.edu/rtd Part of the Agriculture Commons, and the Food Science Commons

Recommended Citation Ko, Kyung Yuk, "Development of simple and rapid procedures for purification of egg yolk immuno-globulin and ovotransferrin from chicken egg and antimicrobial activity of ovo-transferrin against Escherichia coli O157:H7 and Listeria monocytogenes" (2007). Retrospective Theses and Dissertations. 15913. https://lib.dr.iastate.edu/rtd/15913

This Dissertation is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Retrospective Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected].

Development of simple and rapid procedures for purification of egg yolk immunoglobulin and ovotransferrin from chicken egg and antimicrobial activity of ovotransferrin against Escherichia coli O157:H7 and Listeria monocytogenes.

Kyung Yuk Ko

A dissertation submitted to the graduate faculty

in paritial fulfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

Major: Meat Science

Program of Study Committee: Dong U. Ahn, Major Professor James S. Dickson Joseph G. Sebranek Joan E. Cunnick Elisabeth J. Huff-Lonergan

Iowa State University

Ames, Iowa

2007

UMI Microform 3274855 Copyright 2008 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.

ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, MI 48106-1346 ii

DEDICATION

Firstly, I would like to dedicate this dissertation to my loving God. During my Ph. D. study, He was my perfect supporter, refuge and fortress. Whenever I was frustrated for my weakness or a variety of works unexpected during the study in Ames, He lifted me from the mire and taught me about his deep thoughts and love. Through such difficulties and sufferings faced during ph. D study, I realized that true my pleasure and satisfaction come from only God not the circumstances or any other people. Also, I knew that there is no any shortness in my life if I have to be beside Him because He is a perfect father who knows precisely my need and joyfulness. Also He wants readily to fill them plentifully in my life. Furthermore, I am thankful to God for providing the privilege to study for me and I would like to remember forever the lessons and his love showed during ph. D study. Due to memories of his deep love rendered to me, I would like to accomplish a beautiful life to share my knowledge and privileges gladly for other people. Secondly, I want to dedicate this dissertation to my respected father, Tae-jun Ko and my mother, Aeok Hong in heaven. Even though they passed away before starting ph. D study, I believe that they have been looking at my life in heaven and want to help me readily anywhere. It was the largest suffering and sorrow in my entire life that I could not the chance to reward their love and devotion. However, I thought that only the way to reward their love is to do my best during given life including my Ph. D. study. Thirdly, I would like to dedicate this dissertation in loving memory of my niece, Min-ji Kim. Her death which happened in last year has been one of the hardest things that I have dealt with during my Ph.D. study. Although Minji’s life was cut short, she left behind a lifetime of memories. She was a special loving niece. Even though she was 21 year-old, she was my friend. During her short life, she gave many joys and pleasures to my family. The sorrow losing Min-ji made me perform more diligently and sincerely in my research. I believe that her death will not be in vain.

iii

TABLE OF CONTENTS

ABSTRACT v CHAPTER 1. GENERAL INTRODUCTION 1 Dissertation Organization 3

CHAPTER 2. LITERATURE REVIEW 4 I. Escherichia coli O157:H7 and Listeria monocytogenes 4 I-1. Escherichia coli O157:H7 4 I-2. Listeria monocytogenes 6 II. Natural antimicrobial proteins in chicken eggs 8 II-1. Egg yolk immunoglobulin (IgY) 9 II-1.1. Characteristics of IgY 9 II-1.2. Passive immunizations of animals using IgY 10 II-1.3. Separation and purification methods of IgY 12 II-2. Lysozyme 15 II-3. Ovotransferrin 17 II-3.1. Chemical and physical characteristics of ovotransferrin 18 II-3.2. Iron uptake and release of ovotransferrin 19 II-3.2.1. Iron uptake of ovotransferrin 19 II-3.2.2. Iron release of ovotransferrin 21 II-3.2.3. Effect of sodium bicarbonate on iron uptake and release 23 II-3.3. Purification methods of ovotransferrin 23 II-3.4. Activated lactoferrin 26 II-3.5. Antimicrobial activity of ovotransferrin 27 II-3.5.1. Mechanism of antibacterial action of ovotransferrin 28 II.3.5.2. Factors affecting antibacterial activity of ovotransferrin 29 II-3.6. Bactericidal peptides of ovotransferrin 31 III. Summary 32 References 33

CHAPTER 3. PREPARATION OF EGG YOLK IMMUNOGLOBULIN (IGY) FROM CHICKEN EGGS USING AMMONIUM SULFATE PRECIPITATION AND IRON EXCHANGE CHROMATOGRAPHY 60 Abstract 60 Introduction 61 Materials and Methods 63 Results and Discussion 65 Conclusion 70 References 70

CHAPTER 4. AN ECONOMIC AND SIMPLE PURIFICATION PROCEDURE FOR THE LARGE-SCALE PRODUCTION OF OVOTRANSFERRIN FROM EGG WHITE 88 Abstract 88

Introduction 89 Materials and Methods 91 Results and Discussion 95 Conclusion 99 References 99

CHAPTER 5. THE EFFECTS OF ZN2+, SODIUM BICARBONATE, CITRIC ACID ON THE ANTIBACTERIAL ACTIVITY OF OVOTRANSFERRIN AGAINST ESCHERICHIA COLI O157:H7 AND LISTERIA MONOCYTOGENES IN VITRO AND HAMS 114 Abstract 114 Introduction 115 Materials and Methods 117 Results and Discussion 121 Conclusion 129 References 130

CHAPTER 6. EDTA AND LYSOZYME IMPROVES ANTIMICROBIAL ACTIVITIES OF OVOTRANSFERRIN AGAINST ESCHERICHIA COLI O157:H7 143 Abstract 143 Introduction 144 Materials and Methods 146 Results and Discussion 151 Conclusion 157 References 158

CHAPTER 7. THE EFFECT OF EDTA AND LYSOZYME ON THE ANTIMICROBIAL ACTIVITY OF OVOTRANSFERRIN AGAINST LISTERIA MONOCYTOGENES 170 Abstract 170 Introduction 171 Materials and Methods 173 Results and Discussion 177 Conclusion 183 References 184

CHAPTER 8.GENERAL CONCLUSIONS 196

ACKNOWLEDGEMENTS 199

ABSTRACT The overall goal of this study was to develop practical, natural, and efficient antimicrobials from egg. To accomplish the goal, novel protocols for large-scale production of egg yolk immunoglobulin and ovotransferrin were developed. Also antibacterial activity of ovotransferrin against Escherichia coli O157:H7 and Listeria monocytogenes was investigated and the antimicrobial activity of ovotransferrin was improved by combinations with other antimicrobials. The developed protocol for large-scale production of egg yolk immunoglobulin (IgY) consisted of several steps such as using 0.01 % charcoal to remove lipoproteins from IgY extract, ultrafiltration to concentrate IgY solution, and 40% ammonium sulfate to precipitate IgY following dialysis. Morevoer, apo-ovotransferrin was isolated from 2 time-diluted egg white by following steps: as iron saturation of egg white, extraction of iron-bound (holo) ovotransferrin using 43% ethanol, precipitation of holo- bound ovotransferrin by 59% ethanol, dissolving the precipitate with distilled water, and ® removal iron from holo-ovotransferrin using AG 1-X2 ion exchange resin. After separating apo-ovotransferrin, the effects of citric acid, Zn- and Fe- binding, and sodium bicarbonate

(NaHCO3) on the antibacterial activity of ovotransferrin against E. coli O157:H7 and L. monocytogenes in model systems using BHI broth were investigated. Synergistic effects of EDTA and lysozyme on the antimicrobial activities of ovotransferrin were investigated toward E. coli O157 and L. monocytogenes on model systems using BHI broth and pork chops and hams. According to this study, IgY obtained by developed protocol was approximately 70 ~ 80% recovery and approximately 80% purity. The new procedure seems to be economical and practical for a large-scale purification of IgY from egg yolk since ammonium sulfate precipitation could accommodate of large volume of egg yolk. Also, the recovery rate and purity of ovotransferrin produced by the newly developed protocol was around 94% and 80% respectively. As the procedure is consisted of only a few steps, it is suitable to a large-scale ® production of ovotransferrin. Also becasuse ethanol and AG 1-X2 ion exchange resins can be regenerated, it seems to be economical and efficient. In study related to antibacterial activity of ovotransferrin, 100 mM-NaHCO3 significantly increased the antibacterial activity of ovotransferrin against both E. coliO157:H7 and L. monocytogenes. 0.5% citric acid

enhanced the antibacterial activities of ovotransferrin against E. coli O157:H7, Zn-bound ovotransferrin appeared to inhibit the growth of L. monocytogenes. So, this study suggests that the iron binding capacity of ovotransferrin was not the major mechanism of antimicrobial actions of ovotransferrin because acidic pH or Zn2+ promotes antibacterial activity of ovotransferrin toward E. coliO157:H7 or L. monocytogenes respectively. Also, 2 mg/ml EDTA improved antibacterial activity of ovotransferrin containing 100 mM-NaHCO3 against E. coli O157:H7. Also when lysozyme was combined with OS, the antibacterial activity of OS against L. monocytogenes increased slightly in vitro test. However, synergistic effect of lysozyme on antimicrobial activity of OS against L. monocytogenes did not show significantly in this study. Contrary to the model system results using BHI broth, ovotransferrin plus EDTA or/and lysozyme did not show any antimicrobial activities toward E. coli O157:H7 and L. monocytogenes in pork chops and hams. Developing preservatives such as IgY and ovotransferrin are significant not only for the food industry but also for consumers because they are derived from natural sources and generally regarded as safe. Use of these products will significantly increase the value of eggs, which is important for egg industry. Furthermore, this study claims that ovotransferrins combined with NaHCO3, EDTA, or lysozyme have high potential as natural antimicrobial agents to control E. coli O157:H7 and L. monocytogenes. Also, further studies are required to elucidate the antibacterial mechanisms of ovotransferrin and to overcome the limitations of applying ovotransferrin in food products in the future.

CHAPTER 1. GENERAL INTRODUCTION

The extension of shelf-life and control of pathogenic microorganisms in meat during refrigerated storage is a great concern to the meat industry. Meat processors strive to produce raw products that have low levels of spoilage bacteria and no pathogens on the surface of meat. Meat processing, however, is not done in a sterile environment and contamination is unavoidable, and occasionally pathogenic microorganisms may come in contact with meat. Bacterial food-borne illnesses account for an estimated 5 million cases, 60,000 hospitalizations and 1,800 deaths in the U.S. annually, and about half of the food-borne outbreaks are linked to contaminated meat (Huffman, 2002). Major pathogenic bacteria in meat or meat products include Escherichia coli O157:H7, Salmonella spp., Listeria monocytogenes, Campylobacter, Clostridium botulinum, Clostridium perfringenes, Staphylococcus aureus, Aeromonas hydrophila, and Bacillus cereus (Kotula and Kotula, 2000; Lillard, 1990; Clouser et al., 1995). Listeria monocytogenes is commonly isolated from fresh and cooked meat and 70% of the ground beef in the United States is contaminated with L. monocytogenes (Brackett, 1988). The eradication of L. monocytogenes that exist in processing lines, machines and ready-to-eat meat products is difficult because of its ability to form biofilms, which has a strong adhering ability on all kinds of surfaces. The fimbria of E. coli O157: H7 also promotes bacterial adhesion and colonization of E. coli on the mucosal surface of meat (Naidu, 2002). The reduction of the initial number of contaminated microorganisms in meat is critical for controlling propagation of pathogens. Therefore, pre-harvest decomtamination techniques of livestock as well as post-harvest decontamination of carcass and meat are considered as the major strategies to prevent or reduce the incidence of food borne diseases. Post-harvest intervention methods include chemical dehairing, hot water rinse, steam pasteurization, steam vacuum, chemical rinses, spraying lactoferrin, adding antimicrobial additives such as nicin, acids or acid salts, and irradiation (Huffman, 2002). Also, a variety of methods such as pasteurization, vacuum packaging, high pressure processing, UV radiation, irradiation, antibiotics, lactic acid, or lactoferrin addition etc. have been used as decontamination means of meat products to control pathogenic and nonpathogenic microorganisms.

Numerous methods have been introduced to prevent the growth of pathogens in meat. In recent years, however, the food industry has shown considerable interest about using natural antimicrobial agents which originate from natural sources such as plant, animal, and bacteria instead of using chemical sanitizers and additives. Due to these trends, many researchers have screened antibacterial materials from various natural sources. Natural antimicrobial materials, which are non-toxic to humans, but have inhibitory activity against undesirable microorganisms, provide great satisfaction to consumers sensitive to safety issues. Egg contains high levels of proteins such as lysozyme, egg yolk antibody (IgY) and ovotransferrin that have strong antimicrobial activities. Especially, IgY has been considered as an interesting antibacterial source that can be used to cure or prevent diseases in animals, fish, and human. IgY can easily be isolated in large scale from egg yolk without using organic solvents, and there is practically no risk or negative side effect other than egg allergy from using IgY. Ovotransferrin is a major egg white protein (12% of total egg white protein) that can be used as a natural antimicrobial agent (Valenti et al., 1983a; Banks et al., 1986; Giansanti et al., 2005; Wang and Shelef, 1991). Lactoferrin, an ovotransferrin equivalent from milk, has been given GRAS (Generally recognized as safe) status by the US FDA and is approved to use in meat processing to control pathogens, but is very expensive. Although, the antimicrobial activity of ovotransferrin against a variety of pathogenic and nonpathogenic microorganisms has been discussed, no attempt has been made to separate ovotransferrin on a commercial scale and use it in processing meat or other foods. If ovotransferrin can be separated from egg white economically on a large scale, many meat processing plants may prefer to use ovotransferrin instead of lactoferrin. Furthermore, when ovotransferrin can be separated from egg white without using hazardous chemicals, it should easily get the status of GRAS, like lactoferrin. In spite of several potential benefits of using ovotransferrin, however, little work has been done to separate ovotransferrin from egg white on a commercial scales and to use in meat processing to prevent the growth of pathogens. The first objective of this study was to develop economical, large-scale separation methods for IgY from egg yolk and ovotransferrin from egg white. The second objective was finding conditions for activating ovotransferrin separated from egg white and determining the

effects of activated ovotransferrin on the survival of food borne pathogens in model systems and ready-to-eat meat products. The central hypothesis of the study was that ovotransferrin and IgY can be purified in large quantities using simple procedures instead of expensive ion- exchange chromatography or HPLC. Also, like activated lactoferrin, the activated ovotransferrin may become a natural antimicrobial agent that can be used to control pathogens in poultry carcasses. This study developed rapid and simple protocols for large-scale production of IgY and ovotransferrin from egg. The separation protocols developed do not require sophisticated equipment or complicated steps for easy adaptation by industry. However, the results indicated that there are some weaknesses and problems to overcome for the practical application of ovotransferrin to control pathogens in meat products even though the antimicrobial efficacy of ovotransferrin in model systems was excellent.

Dissertation organization The dissertation is composed of a general introduction (Chapter 1), literature review (Chapter 2), 5 research papers (Chapter 3-7) and a general conclusion (Chapter 10). The literature review includes an overview of Escherichia coliO157:H7 and Listeria monocytogenes, chemical and physical characteristics, purification methods, and passive immunization of animals using egg yolk antibody and antibacterial properties of lysozyme and chemical and physical characteristics, iron uptake and release mechanism, purification methods, and antimicrobial activity of ovotransferrin. Chapter 3 contains preparation of IgY from egg yolk using ammonium sulfate precipitation and ion exchange chromatography. Chapter 4 describes an economic and simple purification procedure for the large-scale production of ovotransferrin from egg white. Chapter 5 elucidates the effects of Zn2+, sodium bicarbonate, and citric acid on the antibacterial activity of ovotransferrin against E. coli O157:H7 and L. monocytogenes in vitro and in hams. Chapter 6 reports enhancement of the antimicrobial activity of ovotransferrin solution containing NaHCO3 against E. coli O157:H7 by EDTA and lysozyme. Chapter 7 demonstrates the effect of EDTA and lysozyme on the antimicrobial activity of ovotransferrin against L. monocytogenes. Finally Chapter 8 is a discussion with general conclusions of this dissertation.

CHAPTER 2. LITERATURE REVIEW

I. Escherichia coli O157:H7 and Listeria monocytogenes Escherichia coli O157:H7 is one of emerging strains due to frequent disease outbreaks and meat product recalls. Also, due to risks of listeriosis caused by L. monocytogenes, the U.S. Department of Agriculture currently established a “zero tolerance” policy for L. monocytogenes in ready-to-eat (RTE) meat products. Therefore prevention or minimizing the contamination of E. coli O157:H7 and L. monocytogenes during meat processing or meat products is critical for meat industry.

I-1. Escherichia coli O157:H7 Escherichia coliO157:H7 has been used as an indicator organism to estimate if meat or meat products are contaminated by feces or digestive organs during slaughtering or processing. Illness caused by E. coli O157:H7 are frequently linked with cattle and their products, undercooked ground beef and raw milk (Firstenberg-Eden, 1997; Weeranta and Doyle, 1991). In most of E. coli O157:H7 outbreaks, the vehicle of transmission were identified as ground raw beef. E. coli O157:H7 contamination on beef carcasses were from contamination by feces or soil during the slaughtering process (Dorn, 1993). In 1982, E. coli O157:H7 was known as a human pathogen after outbreaks of hemolytic colitis caused by contaminated ground beef patties in Oregon and Michigan (Abdul-Raouf et al., 1993; Buchanan and Doyle, 1997; Foulke, 1994). However, the public was not aware of the seriousness of E. coli O157:H7 as an emerging pathogenic strain until an outbreak linked with the consumption of undercooked hamburger happened in Washington State (USDA, 1994). The outbreaks occurred in fast-food restaurants in Washington State indicated that improper cooking of patties could be responsible for the outbreak (Bell et al., 1994). Other foods associated with the outbreaks of illness caused by E. coli O157:H7 include roast beef, turkey roll, lettuce, radish sprouts, and alfalfa sprouts (Buchanan and Doyle, 1997; Semanchek and Golden, 1997). USDA-FSIS indicated that 1.8% of carcass samples taken post-processing were found to be positive for E. coli O157:H7 (Huffaman, 2002).

Generally, diarrheagenic Escherichia coli known as pathogenic strains can be classified based on virulence properties, clinical syndromes, differences in epidemiology and distinct O:H serogroups as enterotoxigenic (ETEC), enteroinvasive (EIEC), enteroaggregative (EAEC), enteropathogenic (EPEC), diffusely adherent (DAEC) and enterohemorrhagic (EHEC) (Bell, 1998b; Buchanan and Doyle, 1997; Gyles, 1994; Padhye and Doyle, 1992). E. coli O157, known as Shiga-like toxin-producing E. coli or vero cytotoxin-producing E. coli, belongs to enterohemorrhagic (EHEC) group and it sometimes produces a variety of clinical syndromes including bloody and non-bloody diarrhea, hemorrhagic colitis (HC), severe abdominal pain, cramps, hemolytic uremic syndrome (HUS) incurring severe renal failure, and thrombotic thrombocytopenic purpura (TTP) causing an impairment of the central nervous system (Bell 1998b; Fistenberg-Eden, 1997; Hathcox et al., 1995; MacDonald et al., 1998; Martin et al., 1990; Padhye and Doyle, 1992; Raghubeer et al., 1995; Doyle, 1991). EHEC strains exist naturally in cattle, other animals, and soil, and have a very low infectious dose to cause severe clinical syndromes (Bell, 2002). E. coli O157:H7 possesses specific cell-surface appendages such as fimbria or colonization factor antigens that facilitate their adhesion and anchor them to the host-tissue- matrix components such as fibronectin, collagen, elastin and mucin. Thus, they can tightly attach to beef tissues (Naidu, 2002). Furthermore, E. coli O157:H7 is quite tolerant to several harsh environments such as acid, sodium chloride, nitrite, alkali, and modified atmosphere even though the organism can be destroyed by cooking or pasteurization (Abdul-Raouf et al, 1993; Semanchek and Golden, 1997; USDA,1994; Williams and Ingham, 1997). Also, it can survive in water, beef, and chicken stored at subfreezing temperatures as well as outside the host for extended periods of time (Semantic and Golden, 1997; Wang and Doyle, 1998; Bell, 1998a). Even though E. coli O157:H7 has an optimal pH near 7.0 (Tsai and Ingham, 1997), it has higher tolerance to acidic pH than any other food borne pathogens (Brackett et al. 1994; Benjamin and Data 1995; Conner and Kostroma, 1995; Layer et al., 1995; Jordan et al., 1999). Due to high resistance to acidic pH, a variety of acidic foods such as apple cider, apple juice, fermented meat products, mayonnaise and yogurt are implicated in food-borne illness by E. coli O157:H7 (Semanchek and Golden, 1997; Tsai and Ingham, 1997).

Furthermore, Farrell et al. (1997) reported that E. coli O157:H7 survived on meat grinders, which was sanitized with peroxyacetic acid and chlorine. E. coli O157:H7 is considered as an emerging pathogenic strain due to frequent disease outbreaks and meat product recalls. So prevention or minimizing the contamination of E. coli O157:H7 during meat processing is very important for meat industry.

II-2. Listeria monocytogenes Listera monocytoenes is one of the most frequent pathogens causing food born diseases. According to the report of the Center for Disease Control and Prevention, 2,500 people suffer from listeriosis annually in the United States, indicating that infection caused by L. monocytogenes is a serious problem (CDC, 2002). Moreover, Food Safety and Inspection Service reported that one in five died from the disease (USDA-FSIS). Due to these risks, the U.S. Department of Agriculture currently established a “zero tolerance” policy for L. monocytogenes in ready-to-eat (RTE) meat products. The safety concern of L. monocytogenes is further highlighted by multistate outbreaks of listeriosis related to RTE meat products such as frankfurters and deli meats. As this organism can proliferate in the presence of curing salt at refrigerated temperature (Lou and Yousef, 1999), RTE cooked meat products have high incidence rates of contamination by L. monocytogenes due to multiple handling and processing steps performed after cooking (Uyttendaele et al., 1999). Many cases of listeriosis in humans are epidemiologically linked to the consumption of poultry products (Rorvik et al., 2003; Schwartz et al., 1988). The Nationwide Young Turkey Microbiological Baseline Data Collection Program reported that 5.9% of turkey carcass rinses and 31% of ground turkey meat in United States were contaminated by L. monocytogenes (USDA-FSIS, 1998). In Berlin, L. monocytogenes were found on 85 carcasses among 100 chickens, which are consisted of 95 broilers and 5 layers (Schonberg et al., 1989), whereas in Northern Ireland, 38 of 80 retail packs of raw chicken purchased from supermarkets were contaminated by Listeria spp. and 62% (38 of 61) of raw broiler pieces chosen from retail stores were contaminated by L. monocyotogenes (Miettinen et al., 2001).

Also, according to a report in Japan, contamination by L. monocytogenes was significantly higher in retail sliced beef (34.2%) and pork (36.4%) than on carcasses of cattle (4.9%) and swine (7.4%) (Iida et al., 1998). In Finland, six out of 50 carcasses (12%) were contaminated with L. monocytogenes (Autio et al., 2000). In the United States, ground meat products were found to be contaminated more frequently L. monocytogenes (Duffy et al., 2001). According to a survey performed at retail markets of Maryland Northern California, 577 of 31,705 samples were found to be contaminated by L. monocytogenes (Gombak et al., 2003). Generally, L. monocytogenes are found widely in plant, soil, silage, sewage, slaughterhouse waste, milk, human and animal feces, plant environments, and catering facilities (Farber and Petering, 1991; Beresford et al., 2001). L. monocytogenes is a gram- positive, highly motile, rod with rounded ends. It is a facultatively anaerobic bacterium and does not produce spores or form a capsule (Farber and Petering, 1991). L. monocytogenes can grow at temperature from 0oC to 42o C. At refrigeration temperature its growth is slow with generation times of 30 ~ 40 h at 4o C in skim milk (Malevich, 1992; Rosenow and Mirth, 1986). Usually, Listeria can grow at pH 4.5 ~ 9.6, optimally at pH 7.0 (Buchanan and Loiter, 1990; George and Lund, 1992; Parish and Higgins, 1989; Petron and Portola, 1989; Seelinger and Jones, 1986). Some Listeria require various nutrient components such as leucine, isoleucine, arginin, methionine, valine, cysteine, riboflavin, biotin, thiamine, and thioctic acid for growing (Premaratne et al., 1991; Welshimer, 1963) and they grow faster in the presence of Fe3+ and phenylalanine (Premaratne et al., 1991). Also, they commonly proliferate well in meats near or above pH 6 and poorly on meat products near or below pH 5.0 (Glass and Doyle, 1989). The organisms tolerate salt and nitrite (McClure et al., 1997) and can grow at

an water activity (aw) value below 0.93 (Farber et al., 1992; Petran and Zottola, 1989). As they can colonize, multiply and persist on processing equipment or plant substances, which may be natural habitats of the organism, they form bio-films (Rocourt and Seeliger, 1985). Low levels of L. monocytogenes in the intestines of infected animals may enter processing environments or facilities and then survive in bio-films. Therefore, these processing facilities play an important role in contamination by L. monocytogenes in some products. The contamination by L. monocytogenes in RTE meats occurs mainly during

packaging and after cooking because L. monocytogenes is sensitive to heat treatments and can be easily inactivated by cooking (Giovannacci et al., 1999; Lunden et al., 2003). Also, dust contaminated with L. monocytogenes may be a source of contamination since they can adhere to stainless steel even though the organisms show significant differences in their ability to attach to surfaces (De Ruin et al., 2003). Decayed vegetation is known as a source of many cases of listeriosis which occurred in farm animals. Due to these attributes, L. monocytogenes have been considered as one of the major food borne pathogens threatening the food industry (Rayer and Marth, 1999). For prevention or minimizing the contamination of E. coli O157:H7 and L. monocytogenes, a variety of methods including pre- and post-harvest decontamination techniques have been employed. However, due to consumers’ desire for natural antimicrobials that are environmentally, medically acceptable and safe, natural antimicrobial proteins have gained attention as a great source of antimicrobials.

II. Natural antimicrobial proteins in chicken eggs An egg is comprised of 9.5% eggshell including shell membrane, 63% egg white (albumen), and 27.5% egg yolk. An egg is composed of 75% water, 12% protein, 12% lipids, and 1% carbohydrate and minerals (Burley and Vadehra, 1989; Sugino et al., 1997). Egg whites contain a variety of chemical defenses such as lysozyme, ovotransferrin, avidin, cystic and ovomacroglobulin (called ovostatin) to protect them from microbial attack (Korpela et al., 1984; Molla et al., 1987; Miyagawa et al., 1991a, b, 1994; Blankenvoorde et al., 1998; Nakai, 2000; Ahlborn et al., 2006). Furthermore, ovomucin (Watanabe et al., 1998) and ovoinhibitor (Yolken et al., 1987) in egg albumin are reported to have antiviral properties. Egg yolk also contains antibacterial materials such as sialyloligosaccharides (Sugita-Konishi et al., 2002) and egg yolk antibodies called IgY. Eggshell membrane plays an important role as a barrier to bacterial penetration into the egg and contains active antibacterial substances (Ahlborn et al., 2006). The antibacterial activities of eggshell membrane are derived from lysozyme, ovotransferrin, and ß- N-acetylglucosaminidase, which a lysozomal enzyme associated with the hydrolytic degradation of peptidoglycan layer such as glycoproteins, mucopolysaccharides, and glycolipids in outer membranes of microorganisms (Ahlborn et al.,

2006; Vadehra et al., 1972; Winn and Ball, 1975; Schockman and Höltje, 1994). These proteins and other eggshell membrane components seem to prevent the growth of bacteria by interacting with bacterial outer membrane (Ahlborn et al., 2006). Among these egg proteins or components, egg yolk immunoglobulins that have a variety of advatages have gained attention as an antimicrobial agent for inhibition of bacterial growth in fish and human.

II-1. Egg yolk immunoglobulin (IgY) Egg yolk antibodies have great potential as a natural antibacterial source due to their binding properties to bacteria. Even though IgY is not currently recognized as a GRAS material, egg yolk antibody can be used as a secondary food additive (FSIS), which is not required to be declared in the ingredients statement nor provide any standards for identification.

II-1.1 Characteristics of IgY Chicken egg yolk is an abundant, inexpensive source of antibodies called immunoglobulin Y (IgY), which originates from serum immunoglobulins (IgG) of chicken and accumulates in yolk (Gregory, 1995). In the hen, egg yolk immunoglobulins (IgY) are actively transported from serum to yolk, and protect chicks from various infectious diseases. A large chicken egg (60 g) contains approximately 200 mg IgY, and antibodies recovered from an egg are equivalent to the amount of antibodies obtained from a rabbit every 2-3 weeks. A layer hen produces more than 300 eggs per year, and thus the production cost of chicken antibodies is much lower than that of a rabbit (Larsson et al., 1993; Schade et al., 1994). Generally, immunoglobulins protect animals or humans from bacterial, viral, and parasitic infections by inhibiting binding or colonization of bacteria, agglutinating bacteria, and neutralizing toxins (Williams and Gibbons, 1972; Kubo et al., 1973). Due to such attributes, egg yolk antibodies can inhibit the growth or propagation of some strains. Chicken antibodies have many advantages compared with the conventional mammalian antibodies. Chicken IgY reacts with more epitopes against mammalian antigens due to phylogenetic distance between birds and mammals, resulting in an amplification of

signals (Carlander et al., 1999; Larsson et al., 1993; Svensden et al., 1995; Schade et al., 1994). Poultry, having lower phylogenetic status, thus are an excellent source of varied antibody production. Mammalian IgG often binds to the rheumatoid factors that are a marker of inflammatory response in immunoassays, but chicken egg yolk antibodies do not react with rheumatoid factors. Thus, if IgY is used in immunoassays, more accurate results can be obtained (Larsson, 1988; Davalos-Pantoja et al., 2000; Carlander et al., 1999). Also, chicken egg yolk antibodies do not interfere with the immune reactions of mammalian IgG and do not activate mammalian complement (Carlander et al, 2000; Larsson et al., 1991, 1992; Akita and Nakai, 1993). Also, chicken antibodies do not bind protein A and G, which are commonly used for the isolation of IgG because of the difference in the Fc-region of chicken antibody (Akerstrom et al., 1985; Schwarzkopf and Thiele, 1996). Due to these specific properties of egg yolk antibodies, IgY has been applied in many medical fields such as xenotransplantation (Fryer et al., 1999), diagnostics (Erhard et al., 2000) and antibiotic- alternative therapy (Carlander et al., 2000). Traditionally, polyclonal antibodies have been obtained from the serum of immunized mammals such as rabbits, sheep, goat and pigs. Using chicken for the production of polyclonal antibodies seems to provide great economical benefit because of low maintenance costs for animals and the non-invasive, simple, and fast technique for IgY separation (Wu et al., 2003; Tini et al., 2002). From an animal welfare standpoint, immunoglobulin (IgY) isolated from eggs obtained from immunized chicken is also a good alternative to traditional polyclonal antibody production with mammals (Svendsen et al., 1995).

II-1.2. Passive immunizations of animal using IgY Chicken egg antibodies can prevent the growth of pathogens as IgY can block or impair the function of bacterial components related to growth. Due to many advantages of IgY in inhibition of bacteria, many researchers have reported passive immunization using IgY. Also, oral administration of egg yolk antibodies produced from immunized chickens showed promising results as prevention or therapeutic treatment for specific diseases in infants, fish, small animals, and adult humans.

Specific antibodies derived from eggs of immunized hens might be an inexpensive alternative for orally administered passive immunization against enteric infections in animals. In fact, IgY has been demonstrated to play a positive role in preventing or controlling enteric infections caused by enterotoxigenic E. coli, Salmonella typhimurium, rotavirus, and Helicobacter pylori (Sarker and Hammarstrom, 2004). Adhesion of enterotoxigenic E. coli (ETEC) to intestinal epithelium causes most of the gastrointestinal disorders in neonatal and weaned piglets (Nagy and Fekete, 1999). Egg yolk antibodies inhibited the adhesion of hemolytic E. coli K88 to piglet small intestinal mucosa (Jin et al., 1998) and showed therapeutic effects in piglet diets by inhibiting enetrotoxigenic E. coli infections which cause post weaning diarrhea (Owusu-Asiedu et al., 2002; Imberechts et al., 1997). Rabbits orally given egg yolk immunoglobulin obtained from hens immunized with enterotoxigenic E. coli were protected from diarrheal illness. In addition, specific egg yolk immunoglobulins inhibited the binding of E. coli to rabbit small bowel mucin glycoprotein in vitro (O’Farrelly, et al., 1992). Arasteh et al. (2004) reported that when IgY derived from chickens immunized with Vibrio anguillarum, an important pathogen of marine and fresh water fish, was administered orally and intraperitoneally to fish, anti V. anguilarum IgY showed protective effects against vibriosis and enhanced disease resistance. In fish, Yersinia ruckeri is a predominant bacteria isolated from rainbow trout with enteric redmouth disease. It has long periods of persistence in carrier fish and is shed in feces, leading to a continued source of infection. Feeding IgY specific for Y. ruckeri to fish decreased the mortality compared to the groups receiving normal feed. The oral feeding of specific IgY derived from immunized chickens with fish pathogens seems to be an alternative to antibiotics and chemotherapy for prevention of fish diseases (Lee et al., 2000). Recently some reports show that egg yolk antibodies can be used as therapeutic materials in human diseases. H. pylori bind several receptors on the surface of gastric epithelial cells and are often resistant to antibiotics. Egg yolk antibodies obtained from hens immunized with H. pylori whole cell lysate showed cross-reactivity with normal human flora (Shin et al., 2002), and thus the antibodies may cause difficulties by inhibiting normal flora. However, urease-specific IgY (IgY-HpU) produced by immunizing hens with urease, the

immunodominant H. pylori protein and a well-known critical virulence factor, has a significant inhibitory effect on H. pylori urease activity compared to control IgY (Shin et al., 2004). Kim (2004) reported that H. pylori-specific IgY had a property to inhibit adhesion of the bacteria to gastric epithelial cells and had powerful urease-inhibiting activity both in human and animals. Streptococcus mutans, which is the main bacterial species associated with dental caries, accumulate in the dental boil where multiple binding sites for their glucan proteins are present. Egg yolk antibodies obtained from chicken immunized with glucan-binding protein B inhibited accumulation of S. mutans within oral biofilms and protected against dental caries in a mammer similar to salivary antibodies (Smith et al., 2001). Shimizu et al. (1993) reported that encapsulation of IgY with egg lecithin/ cholesterol liposomes reduced the loss of IgY activities in gastric juices. So, these immunoglobulins could be used to prevent gastro-intestinal infections in individuals lacking an active gut immune response. Purified egg yolk antibodies can also be used for the fortification of food products, especially in infant formula since they can be purified and their activity is relatively stable at pH above 4.0, heating up to 60o C, freezing, freeze-drying, and tryptic and chymotryptic digestions (Shimizu et al., 1988). Deignan et al. (2001) reported that hen egg yolk significantly inhibited the adherence of S. typhimurium to intestinal epithelial cells in vitro and assumed that egg yolk possesses novel anti-adhesive factors. Also, Salmonella- specific IgY resulted in structural alterations of bacterial surface by binding the antigens expressed on Salmonella surface (Lee et al., 2002; Beal et al., 2004). In addition, E. coli O157:H7-specific antibodies had inhibitory effect on the growth of E. coli O157:H7 in a liquid medium (Sunwoo et al., 2002).

II-1.3. Separation and purification methods of IgY In spite of various advantages of using IgY in research and diagnostics, limitations of IgY purification protocols inhibit its practical use (Akerstrom et al., 1985; Camenisch et al., 1999; Amersham Pharmacia Biotech, 2000). The classical protocols for purifying egg yolk antibodies are normally complex and time-consuming, and are associated with a combination of precipitation and chromatographic steps because of the peculiar composition of egg yolk

and lack of specific affinity ligands (Verdoliva et al., 2000). The IgY separation methods developed so far are expensive and impractical for large scale production of antibodies. Moreover, due to the high lipid content in egg yolk and low solubility of egg yolk proteins, chromatographic techniques are considered to be less suitable for the separation of egg yolk proteins. The first step of antibody isolation from egg yolk usually is associated with the removal of lipids and lipoproteins because they are the major obstacles in isolating chicken antibodies from egg yolk (Hasen et al., 1998; Verdoliva et al., 2000). As removing lipids is a laborious and tedious task during the purification step, various methods have been tested to remove lipids or lipoproteins from egg yolk. The methods include using detergents such as cetrimide, sodium dodecyl sulfate (Sriram and Ygeeswaran, 1999), carrageenan (Chang et al., 2000; Hatta et al., 1990), sodium alginate or xantham gum (Hatta et al., 1990), solvents such as acetone (Sriram and Ygeeswaran, 1999) and chloroform (Ntakarutimana et al., 1992), and ethanol precipitation (Bade and Stegemann, 1984), and cold propane-2-ol extraction, and precipitation with polyethylene glycol (Polson et al., 1985, Akita and Nakai, 1993; Gee et al., 2003), dextran sulfate (Jensenius et al., 1981), caprylic acid (McLaren et al., 1994), or pectin (Chang et al., 2000). Also, aqueous two–phase system with phosphate and triton X-100 (Stalberg and Larsson, 2001), simple freeze and thaw method (Svendsen et al., 1995), and water dilution method under acidic conditions (Ruan et al., 2005; Akita and Nakai, 1992 ) were used for separating lipids and water soluble proteins. After removal of lipids and lipoproteins from egg yolk, egg yolk antibodies were precipitated by ammonium sulfate (Akita and Nakai, 1992, 1993; Hansen et al, 1998), sodium sulfate (Wooley and Landon, 1995) or caprylic acid-ammounium sulfate (Ruan et al 2005; McLaren et al., 1994). From the precipitated egg yolk proteins, antibodies were further purified by affinity column chromatography (Greene and Holt, 1997; Yokoyama et al., 1993), gel filtration (Devi et al., 2002), cation exchange chromatography (Fichtali et al., 1992, 1993) or DEAE-sephacel anion exchange chromatography (Akita and Nakai, 1992). Akita and Nakai (1992) suggested that a protocol using ammonium sulfate precipitation followed by ultrafiltration and gel filtration was the most efficient for IgY purification. Among various methods suggested for the removal of lipids, a water dilution

method under acidic conditions was the most efficient and economical procedure for large- scale production of egg yolk antibodies (Akita and Nakai, 1993). Svendsen et al. (1995) reported that two-times precipitation of IgY by using 25% ammonium sulfate following 40% ammonium sulfate was the most efficient and practical purification method for a large-scale production of IgY. The changes of pH values in egg yolk solution influenced the extent of interactions between polysaccharides and proteins, and the degree of precipitation of polysaccharide-lipoprotein complexes and the immunoactivity of IgY (Chang et al., 2000; Samant et al., 1993; Gurov et al., 1983). At pH higher than the pI of proteins, polysaccharides containing sulfate groups form precipitants by reacting with globulin (Samant et al., 1993). Also, 0.1% of carrageenan at pH 5.0 was excellent in removing lipoproteins from diluted egg yolk solution (Chang et al., 2000). Kim and Nakai (1996) suggested that λ- carrageenan was the most effective in precipitating lipoproteins, whereas Chang et al. (2000) reported that pectin was better than any gums. According to Akita and Nakai (1992), adjusting the pH of diluted egg yolk between 5.0-5.2 and incubating it for 6 h at 4oC reduced the amount of lipoproteins and improved the recovery of immunoglobulin in the water- soluble fraction. They reported that even though addition of salt to the diluted egg yolk increased the amount of proteins, it demonstrated an inhibitory effect on the removal of lipoprotein from the water-soluble fraction. As a lower ionic strength supports the aggregation of lipoproteins in egg yolk, increasing the dilution ratio facilitated the precipitation of lipoproteins in egg yolk. However, the increased volume made it difficult to handle the subsequent processing of IgY purification. Thus, it was necessary to reduce volume by concentrating the supernatant with ultrafiltration (Kim and Nakai, 1996, 1998) or precipitation of proteins using ammonium sulfate (Hansen et al., 1998; Cook et al., 2001). Developing a simple and large-scale production protocol for IgY production is required for the application of egg yolk antibodies in various fields. However, various methods developed for isolation and purification of egg yolk antibodies until now are not applicable for large-scale production. Most of these methods have drawbacks such as low recovery rates of IgY, complexity of procedures, or impracticality for routine separation. Among the various reviewed methods for IgY purification, a separation technique consisting

of ammonium sulfate precipitation and ultrafiltration after delipidation by water dilution method under acidic conditions have high potentials for use in large scale production of IgY. Also, pH conditions and the amount of ammonium sulfate were critical factors in antibody purification from egg yolk. As egg yolk antibodies can be lost by fouling of the membrane during the ultrafiltration process, some steps to make the water-soluble egg yolk solution free of lipoproteins or other material is absolutely required. To overcome such a problem, carrageenan or charcoal were used to remove lipid or lipoproteins completely.

II-2. Lysozyme Egg white also has several antimicrobial proteins. Lysozyme is one of the major egg white proteins involved in defense of egg from microbial invasion. It is a natural, bacteriolytic enzyme that inhibits the growth of bacteria, especially gram-positive microorganisms (Rayer and Marth, 1999). It catalyzes the hydrolysis of ß-1.4-glycosidic bond between N-acetylglucosamine (NAG) and the N-acetylmuramic acid (NAM) in the polysaccharides of certain bacterial cell wall. Also, it catalyzes the hydrolysis of 1.4- glycosidic bond in chitin, which is a polymer of N-acetyl-D-glucosamine (Jolles and Jolles, 1984; Phillips, 1966; Davis and Reeves, 2002). Lysozyme comprises 3.5% of hen egg white proteins and laying hen possesses 10-fold higher lysozyme than mammals (Osuga and Feeney, 1977). The molecular weight of egg white lysozyme is 14 kDa, and is composed of 129 amino acid residues. Lysozyme binds to ovomucin, transferrin, and ovoalbumin in egg white (Davis and Reeves, 2002). Although not yet approved as a food additive in the United States, it has a great potential to be a natural food preservative due to its properties: it is a very stable protein over a wide range of pH and heating at 100o C for 1~2 minutes, and does not cause harm to humans since lysozyme is specific for only bacterial cell walls (Davis and Reeves, 2002; Rayer and Marth, 1999). As gram-positive bacteria do not possess any outer membrane, lysozyme can hydrolyze easily the peptidoglycan layer, resulting in the lysis of bacteria (Salton and Pavlik, 1960). The antimicrobial activities of lysozyme are derived from its lytic action on certain gram-positive bacteria, but not on gram-negative bacteria possessing an outer membrane (Davis and Reeves, 2002; Davies et al., 1980). Hughey and Johnson (1987) reported that 70~

80% of L. monocytogenes in stationary phase were lysed by 10 mg/L lysozyme while cells of L. monocytogenes in the logarithmic growth phase were not lyzed by 100 mg /L of lysozyme after 12 h of incubation. Listeria can grow in nutrient-rich media such as BHI broth containing lysozyme (20 ~ 200 mg/L) unlike the listericidal effect of lysozyme in nutrient- poor media (Hughey and Johnson, 1987; Rayer and Marth, 1999). Gram-negative bacteria are less susceptible to the bacteriolytic action of lysozyme because the outer membrane prevents the access of lysozyme to the target site (Salton and Pavlik, 1960). Bacterial lipopolysaccharides induce reversible inactivation of bacteriolytic activity of lysozyme (Ohno and Morrison, 1989a, b). Among gram-negative bacteria, some genera such as Salmonella and Shigella are the most sensitive, whereas E. coli, Vibrio and Proteus are relatively resistant to lysozyme (Peterson and Hartsell, 1955). Considering these results, the antimicrobial actions of lysozyme may be attributed to the direct and indirect interaction as well as bacteriolytic actions that destroy cell wall of microorganisms. Antimicrobial activity of lysozyme against microorganisms is affected by the pH of media, lysozyme concentration, incubation temperature, growth state before treatment, and various growth modifiers or chemical treatments (Johansen et al., 1994; Rayer and Marth, 1999). Smith et al. (1991) reported that L. monocytogenes grown at low temperature was less resistant to bacteriolytic action by lysozyme than those grown at 37o C. Low pH promoted cell lysis by lysozyme, but the bacteriolytic activity of lysozyme was relatively stable at the range from pH 5.5 to 7.0 (Johansen et al., 1994). Heating (62.5o C for 15 s) L. monocytogenes in a phosphate buffer prior to treatments increased the antilisterial activities of lysozyme (Kihm et al., 1994). Carminati and Carini (1989) reported that 25 ppm and 1,000 ppm of lysozyme induced 22% and 57% reduction in L monocytogenes, respectively, and Listeria could grow in heat-treated skim milk containing lysozyme. Lysozyme at 100 mg/L did not show antibacterial activity against L. monocytogens in whole milk, but showed some antilisterial activities in mineral-free milk at 4o C. Listeria in demineralized milk containing lysozyme and heated at 55oC resulted in much larger antilisterial activity than that in mineralized milk. Combination of lysozyme and EDTA did not display antibacterial effect against L. monocytogenes in ultrahigh-temperature (UHT) milk (Payne et al. 1994). This indicated that minerals in milk protected L. monocytogenes from the lytic action of lysozyme

(Kihm et al.,1994) and the components of media influenced the antibacterial acitivity of lysozyme. The antibacterial action of lysozyme against bacteria and fungi could be improved by various chemicals such as EDTA, butylparaben, and tripolyphosphate as well as by some other naturally occurring antimicrobial agents (Johnson, 1994). Lactic acid promotes the bacteriolytic activity of lysozyme against L. monocytogenes. Also, EDTA showed a strong synergistic effect on lytic action of lysozyme against L. monocytogenes grown in BHI agar. Lysozyme (0.1 mg/ml) or 1 mM-EDTA could not inhibit the growth of some strains of L. monocytogenes, but the combination of EDTA and lysozyme inhibited their growth (Hughey and Johnson, 1987). L. monocytogenes (at 104 CFU/g) in fresh green beans, fresh corn, shredded cabbage, shredded lettuce, and carrots were inactivated by the treatment of combined lysozyme and EDTA, while Listeria in control groups grew to 106 ~ 107 levels during storage at 5o C (Hughey et al., 1989). Also, lysozyme (3 mg/ml) alone or a combination of lysozyme (3 mg/mL) and EDTA (25 mM) retarded the growth of 103 levels of L. monocytogenes in raw cod fish fillets (Wang and Shelef, 1992). The release of lipopolysaccharides from outer membrane or disruption of outer membrane by EDTA, proteinases or antibody made them more susceptible to the lytic actions of lysozyme (Salton and Pavlik, 1960; Gill and Holley, 2000; Wilson and Spitznagel, 1968; Feingold et al., 1968). Antimicrobial activity of lysozyme against gram-negative bacteria can be promoted by chemical modification such as lipophilization and/or glycosylation (Liu et al., 2000a, b) and heating, which induced partial unfolding of lysozyme (Ibrahim et al., 1996a, b) The antilisterial activity of lysozyme in foods is much lower than that in laboratory media. Also, controlling L. monocytogenes in vegetables using lysozyme is more effective than that in meat products. Even though listeriostatic property of lysozyme lasted for 2-3 weeks in fresh pork sausages, lysozyme showed no effects on the growth of L. monocytogenes during extended storage (Hughey et al., 1989).

II-3. Ovotransferrin Ovotransferrin, also known as conalbumin, is another major contributor to the egg’s defense against microbial infection and rotting, and has potential to be used as a natural

antimicrobial agent in foods. However, few studies on the antimicrobial activities of ovotrasnferrin in meat or meat products are available.

II-3.1. Chemical and physical characteristics of ovotrasnferrin Ovotransferrin known as the major iron-binding protein in egg white constitutes 12% of total egg white proteins. Ovotransferrin is a monomeric glycoprotein consisting of 686 amino acids with 78 kDa of molecular weight, has no free sulfhydryl groups or phosphorus, and binds reversibly with two Fe2+ ions per molecule and concomitantly with two CO2- ions (Abola et al., 1982; Stadelman and Cotterill, 1986; Aisen and Listowsky, 1980; Ibrahim, 1997). Ovotransferrin is composed of one N- and one C-terminal domain and one atom of transition metals such as Fe(III), Cu(III), or Al(III) can bind tightly within the interdomain cleft of each domain of ovotransferrin. Iron is bound by three phenolic hydroxyl groups of protein and one carbonate ion (Warner and Weber, 1953; Wishnia and Warner, 1961). Two lobes in Fe3+-free ovotransferrin is considered to have a conformation with an opening of the interdomain cleft (Rawas et al., 1997; Jeffrey et al., 1998; Kurokawa et al., 1999). The sequence homology between human serum transferrin and hen ovotransferrin is 51%, but a greater intimate homology is shown at C-terminal regions of these proteins (Metz-Boutigue et al., 1984). Ovotransferrin also has significant structural similarities to lactoferrin. In addition, both ovotransferrin and lactoferrin have the same iron-binding sites consisting of two phenolate oxygens from two tyrosines, a carboxylate oxygen from aspartic acid, and one imidazole nitrogen from single histidine (Metz-Bovtigue et al., 1984; Williams et al., 1982; Baker et al., 1987). Ovotransferrin has a significant similarity to lactoferrin in amino acid sequence and the global folding of the molecules, but its disulfide cross-linking patterns and the number of carbohydrate chains in the molecule is different from that of lactoferrin. Also, lactoferrin is implicated in defense function against infections, whereas ovotransferrin is mainly related to the transport of iron to target cells (Davis and Reeves, 2002; Giansanti et al., 2002). The protein moieties of transferrin in egg and chicken blood serum are identical, but the carbohydrate prosthetic groups are different (Williams, 1962a, b) A metal-free ovotransferrin is easily destroyed by physical and chemical treatments, whereas iron-bound ovotransferrin, the holo-form, is relatively stable to proteolytic

hydrolysis, chemical and thermal denaturations (Azari and Feeney, 1958). Generally, most chemical reagents and denaturing conditions reduce the affinity of ovotransferrin for iron because they change particular spatial configurations of native ovotransferrin, which is required for the formation of colored iron-bound ovotransferrin (Fraenkel-Conrat, 1950). Like most heat labile proteins in egg white, ovotransferrin forms aggregates by heating at 60o C, resulting in a milky white gel (Matsudomi et al., 1991). Preheating at 60o C induces partial unfolding and aggregation of proteins and forms soluble aggregates, which is attributed to disulfide bond and hydrophobic interactions during heating (Xu et al., 1997). Cunningham and Lineweaver (1965) reported that heating at pH 9.0 did not produce significant changes in ovotransferrin. A variety of factors such as pH, ionic strength, heating conditions, and protein concentration influenced the formation of heat-induced protein-a transparent gel or a turbid gel (Kitabatake et al., 1988). Understanding the chemical and physical properties or mechanism of iron uptake and release of ovotrasnferrin is essential for the application of ovotransferrin as a natural antimicrobiological agent to meat or meat products.

II-3.2. Iron uptake and release of ovotransferrin

II-3.2.1 Iron uptake of ovotransferrin Ovotransferrin transports and scavenges Fe (III) in eggs of poultry (Kurokawa et al., 1995). Ovotransferrin has two iron binding sites, which are located within the interdomain cleft of each lobe (Butterworth et al., 1975) and requires anions to bind specifically with Fe(III) (Bate and Schlabach, 1975). Like transferrin or lactoferrin that requires bicarbonate to overcome the effect of citrate, which may bind iron (Masson and Heremams, 1968), 2- - 3+ ovotransferrin also requires 1 molecule of CO2 as CO3 or HCO3 per atom of Fe to bind iron (Warner and Weber, 1953). Some anions such as sulfate and phosphate did not induce stable tertiary complexes with Fe(III), but they resulted in similar effects to that of bicarbonate by occupying anion-binding sites in ovotransferrin (Woodworth, 1975). Warner and Weber (1953) reported that the first and second iron atoms that bind to ovotransferrin have different association constants. In transferrin, Fe3+ binding properties are markedly different for the N- and C-lobes even though they have similar overall

conformation (Miautani et al., 2001). Abdallah and El Hage Chahine (1998) suggested that iron uptake from nitrilotriacetatoFe (III) occurred first at C-domain of ovotransferrin after interacting with bicarbonate and then the protein conformation was changed by the loss of 3 ~ 5 protons. Such a modification allows the protein to capture another Fe(III) from nitrilotriacetatoFe (III) at the N-domain. In the final step, the monoferric or diferric state of ovotransferrin lasted for about four hours to attain their equilibrium. Recently, however, Okamoto et al. (2004) suggested that iron uptake of monoferric ovotransferrin occurs mainly at N-lobe as shown at Fig. 1.

F FeN-OTF

F F Apo-OTF Differic-OTF N C

FeC-OTF

Figure 1. The scheme of iron uptake of ovotransferrin: The circle represents the N-lobe of ovotransferrin, while the square represents the C-lobe. When a lobe is loaded with iron, an “F” appears the respective circle or square. The solid and slashed arrows represent the major and minor iron uptake processes, respectively” (adapted from Okamoto et al., 2004). Apo- OTF means apo-ovotransferrin, FeC-OTF represents iron-bound monoferric ovotransferrin at C-terminal, FeN-OTF indicates iron-bound monoferric ovotransferrin at N-terminal, and Differic-OTF stands for iron-saturated ovotransferrin.

The monoferric N-terminal of ovotransferrin shows a special interdomain interaction between the two lysines on the side chain of N-domain, and is assumed to be a hydrogen- bonding interaction with the aromatic ring of a tyrosine residue. Positive charges of two lysine residues play a role to open the N-lobe domain, whereas the interaction of Fe3+ and the aspartate ligand results in the closure of the domain. Subsequently, an interdomain interaction from lysine - lysine couple plays a role to close the protein domain at pH > 6.0, indicating that a dilysine trigger sensitive to pH is considered to be responsible for the second Fe3+ uptake and subsequent domain closure, not permitting Fe3+ to release (Dewan et al., 1993). The iron uptake rate of ovotransferrin is similar to that of lactoferrin. However, the affinity of ovotransferrin to iron is the lowest among transferrin family proteins such as transferrin, lactoferrin, and ovotransferrin. (Pakdaman and El Hage Chahine, 1996; Pakdaman et al., 1998; Abdallah and EL Hage Chahine, 1998). The iron bound to transferrin is relatively stable between pH 6 and 11, but the color formed by iron binding decreases at pH < 5.5 and disappears at pH < 4 (Warner and Weber, 1953). The iron binding capacity of ovotransferrin increased as the pH of the solution increased from 6.5 to 9.5. Above pH 9.5, however, part of the ovotransferrin is irreversibly denatured (Abdallah and EL Hage Chahine, 1998). At alkaline pH, iron preferentially occupies the N-terminal binding site. However, in acidic pH, the monoferric complex is formed due to partial release of iron from N-terminal site and the Fe3+ found is mainly at C- terminal binding site (Williams et al., 1978; Butterworth et al., 1975). Also, in the mild acidic conditions, the iron-ovotransferrin complex releases Fe3+ preferentially from N-terminal binding site of ovotransferrin (Williams, 1975).

II-3.2.2. Iron release of ovotransferrin The in vitro release of Fe3+ from ferric transferrin requires the presence of a simple anion such as pyrophosphate, sulfate, or chloride. These anions in Fe3+ release are closely related to the opening of the domain in the N- or C-lobe (Baldwin and MR de Sousa, 1981; Cheuk et al., 1987; Bailey et al., 1997). The Fe3+ release of ovotranferrin is similar but not identical to that of transferrin: iron release of ovotransferrin is slower and occurs at lower pH than transferrin (Abdallah and El Hage Chahine, 1999). In apo-ovotrasnferrin, two lobes

display a conformation with an opening interdomain cleft (Jeffrey et al., 1998). In iron release, decarbonation is a prerequisite (El Hage Chahine and Pakadaman, 1995; Pakadaman and El Hage Chahine, 1996). Also, the influences of anions on the Fe3+ release are significantly different in the two lobes. The rate of anion-mediated Fe3+ release is much faster from the N-lobe than the C-lobe (Kretchmar and Raymond, 1986; Bail and Harris, 1989). The anion binding to the N-lobe of ovotransferrin follows a dual-pathway, which includes the occupation of an iron ligand residue, the opening of the domains, and the release of the synergistic carbonate anion (Muralidhara and Hirose, 2000). Unlike N-lobe, C-lobe possesses only one anion-binding site in the interdomain cleft and displays a simple monophasic pathway in Fe3+ release. Moreover, it does not facilitate Fe3+ release so that the rate of Fe3+ release from C-lobe is slower than that of the N-lobe (Kurokawa et al., 1995). The anion effect is associated with pH-dependent conformational changes in the N-lobe of ovotransferrin. Addition of chloride slows down iron release at high pH, while at low pH chloride accelerates it (Rawas et al., 1995). Iron release from ovotransferrin occurs at wide ranges of pH (pH 3.0 to 6.0). At acidic pH, the N-lobes of serum transferrin and ovotransferrin show weaker binding stability and accelerated Fe3+ release than the C-lobe (Griffiths and Humphreys, 1977; Mizutani et al., 2001). After decarbonation of the iron-loaded protein, the release of Fe3+ occurs first at the N-lobe and then at C-lobe (Abdallah and El Hage Chahine, 1999). Even though iron bound in holo-ovotransferrin or holo-transferrin is released rapidly at pH 4.0, it is released slowly in neutral media (El Hage Hanhine and Pakdaman, 1995). With transferrin, iron release is faster in the presence of citrate than formate or acetate because citrate interacts with decarbonated monoferric protein with high affinity. However, in the absence of citrate, iron release occurs at a lower pH and the intermediate complex is not formed (Abdallah and El Hage Chahine, 1999). pH itself affects the efficiency of citrate-mediated release of Fe3+ from lactoferrin in

the absence of NaHCO3 (Griffiths and Humphreys, 1977). Iron can be removed from iron- transport proteins at low pH (pH 4.5) in the presence of a large excess of citrate (Guo, 2003). Citrate-mediated Fe3+ release is slightly more efficient at pH 6.8 than at pH 7.4 (Griffiths and Humphreys, 1977).

II-3.2.3. Effect of sodium bicarbonate on iron uptake and release of ovotransferrin Bicarbonate is known as a synergistic anion (Schalabach and Bates, 1975), and is a prerequisite for the iron uptake of transferrin and lactoferrin (Aisen, 1989; Pakdaman and El Hage Chahine, 1996; Pakdaman et al., 1998) because it prevents citrate from binding iron (Griffiths and Humphereys, 1977; Peaker and Linzell, 1975). The anion serves as a bridging ligand between protein and metal ion (Aisen, 1980). With increasing bicarbonate concentrations, Fe3+-specific binding capacity of transferrin is promoted at a constant pH (Abdallah and El Hage Chahine, 1998; Workman et al., 1975 Aisen, 1989; Pakdaman and El 2- - Hage Chahine, 1997). Bicarbonate or carbonate (CO3 or HCO3 ) occupies one of Fe3+coordination sites consisting of four lateral chains of two tyrosines, one histidine, one aspartate, and a carbonate of bicarbonate (Bailey et al, 1988; Andeson et al., 1989; Zuccola, 1993; Kurokawa et al., 1995; Rawas et al., 1996). - 2- Apotransferrin interacts with two HCO3 but not CO3 . Also, each lobe of apo- - transferrin shows a different affinity constant to HCO3 . Such an affinity difference between protein and bicarbonate is not affected by the state of protonation of ovotransferrin or proton transfer, while the difference is attributed to ionic interactions associated with the side chain of arginine residues in binding sites of each lobe (Bellounis et al., 1996; Pakdaman et al, 1998). Bicarbonate may be weakly bound to apo-transferrin in the absence of a metal ion (Woodworth, 1975), but it is not clear that the binding is specific. The affinity of ovotransferrin for bicarbonate in each lobe is about three-fold weaker than that of serum transferrin (Bellounis et al., 1996; Pakdaman and El Hage Chahine, 1997).

II-3.3 Purification methods of ovotransferrin Bain and Deutsch (1948) separated essentially pure ovotransferrin from egg white using aqueous and ethanol fractionation procedures. They judged the purity and yield of the protein by electrophoretic and sedimentation behavior. Warner and Weber (1951) also fractionated ovotransferrin from egg white using NaCl and ethanol at pH 6 to 9. Ovotransferrin was also separated by fractional precipitation with ammonium sulfate or by shaking the protein solution to help precipitate ovalbumin by coagulation (Fraenkel-Conrat and Feeney, 1950; Warner, 1954; Azari and Baugh, 1967). However, the drawback of

ammonium sulfate precipitation at acidic condition is that the electrophoretic behavior of the ovotransferrin differs from native ovotransferrin (Fraenkel-Conrat and Feeney, 1950). Vachier et al. (1995) reported that using salt or ethanol precipitation, for the separation of ovotransferrin from egg white, denatures the protein and the purity of the protein was relatively low. In addition to salt or ethanol precipitation, several chromatographic methods such as gel permeation, ion-exchange, reverse-phase, hydrophobic interaction, and immobilized- ligand-affinity chromatography have been employed for the separation or purification of ovotransferrin. Ion-exchange chromatography is the most frequently used method for protein purification and isolation, and is a commonly practiced method due to its ease of use and scale-up possibilities (Larive et al., 1999; Awade, 1996). Because ovotransferrin has relatively high pI value (pH 6.1) among the egg white proteins except for lysozyme and avidin, the separation of ovotransferrin by ion exchange chromatography was mainly based on the exchange of a cation. CM-cellulose was used to separate two conalbumin fractions from egg albumin and ovotransferrin contaminated by globulins was then obtained (Rhodes et al., 1958). To obtain a purer protein, Azari and Baugh (1967) precipitated and crystallized ovotransferrin eluted from a cation column. Anion exchange chromatography using DEAE- cellulose (Mandeles, 1960) or using a Q-Sepharose Fast Flow column was also carried out to separate egg white proteins (Vachier et al., 1995). Ovotransferrin separated from anion exchange chromatography using Q-Sepharose Fast Flow column is further purified using the Q-Sepharose column a second time. The final protein obtained appeared to be homogenous on electrophoresis and its purity was around 97.5% and the amino acid composition was close to that of the nucleotide sequence of ovotransferrin (Croguennec et al., 2001). Awade et al. (1994) used a two-step chromatographic procedure involving gel permeation on a Superose 6 Prep Grade column and an anion exchange chromatography on a Q-Sepharose Fast Flow for the purification of major hen egg white proteins. The purity and yield of ovotransferrin obtained by two-step chromatographic procedure were 80% and 60%, respectively. Guerin and Brule (1992) developed an industrial-scale procedure for the separation of ovotransferrin from egg white devoid of lysozyme using a cation exchange chromatography on DUOLITE C464 and C476.

The yield and purity of the isolated ovotransferrin were 54% and 97%, respectively. The drawback of the study was, however, more than 7 hours was required for the equilibration of resins due to pH adjustment to attach ovotransferrin to resin. Ahlborn et al. (2006) separated ovotransferrin from prepared egg white solution and supernatant obtained from centrifugation after dilution of egg white solution with sodium phosphate buffer (pH 5.0) containing 2M Urea by employing Econo-Pac® High-S Cation or High-Q Anion Exchange Cartridge (5 ml). The purity and recovery of ovotransferrin separated on cation exchange chromatography was higher than that of anion exchange chromatography. However, the recovery rate of ovotransferrin obtained by cation exchange chromatography was around 37%. Owing to the metal binding property of ovotransferrin, immobilized-metal affinity chromatography has been considered as an efficient method for the isolation and purification of specific protein that has binding property with metal ion (Porath et al., 1975). Lonnerdal and Keen (1982) reported that the binding ability of ovotransferrin on metal-chelate gel is ascribed to electron-rich groups such as histidine and tryptophan, which can replace water or buffer bonded weakly to the metal complex (Lonnerdal and Keen, 1982). Immobilized metal affinity chromatography with a copper-loaded Sepharose 6B column was employed by Al- Mashikhi and Nakai (1987) who separated ovotransferrin from undiluted, blended egg white by a single chromatographic step. They obtained two fractions of ovotransferrin possessing purity of 98% and 94%. However, a part of the ovotransferrin was not bound at the loading step and was lost in the washing step, resulting in a low recovery rate. They indicated that ovotransferrin without modification of histidine strongly bound to the Cu-chelate gel, while histidine-modified ovotransferrin did not bind to Cu-chelate matrix at alkaline pH (Al- Mashikhi and Nakai, 1987). Chung et al. (1991) used a bifunctional dye-ligand chromatography using DEAE Affi-Gel Blue, as a preliminary first step, to fractionate egg white proteins. The fraction containing ovotransferrin obtained from DEAE Affi-Gel Blue was further purified by Fast Protein liquid chromatography. Based on the review of ovotransferrin purification methods reported, affinity chromatography and ion-exchange chromatography seem to be attractive choices for the isolation of ovotransferrin. However, many protocols using ion-exchange chromatography or

immobilized metal affinity chromatography are developed only on a laboratory scale, and thus are not likely to be practical for pilot-scale production of ovotransferrin. Furthermore, they require expensive resins and facilities, intensive technique for performance, and more expensive maintenance costs (Awade, 1996). Even though, a variety of protocols for a laboratory-scale separation of ovotransferrin have been developed (Azari and Baugh, 1967; Al-Mashiki and Nakai, 1987; Guerin and Brule, 1992; Shibusawa et al., 2001), no successful attempt for commercial-scale productions of ovotransferrin is available. Thus, developing an economical, simple, large-scale production method for separating ovotransferrin is very important for the commercial use of ovotransferrin as a natural antimicrobial agent in meat or meat products.

II-3.4. Activated lactoferrin As ovotransferrin has significant structural and functional similarities to lactoferrin, the knowledge about antimicrobial actions of activated lactoferrin will be helpful in understanding the mechanism of antibacterial activity of ovotransferrin. Lactoferrin, an ovotransferrin equivalent from milk, is a member of transferrin family and an iron-binding protein with 80 kDa of molecular weight. It transports iron from plasma to target cells and binds iron with high affinity (Meta-Boutigue et al., 1984). Lactoferrin is synthesized at glandular epithelial cells and secreted into mucosal fluids in response to inflammatory stimuli. It is very rich in colostrum and milk, and small amounts of lactoferrin also exist in tears, saliva, nasal fluids, pancreatic juice, gastrointestinal fluids, neutrophils, and reproductive tissue secretions (Masson et al., 1966; Levay and Viljoen, 1995). Lactoferrin has a variety of biological functions: it is involved in iron homeostasis, cellular growth and differentiation, host defense against microbial infections, anti-inflammatory activities, and protection from cancer (Levay and Viljoen, 1995; Ward et al., 2002; Brock, 2002; Harmsen et al, 1996). Activated lactoferrin is known as a microbial blocking agent and binds to the same anchor sites as bacteria on tissue surfaces with a greater affinity, and thus it can block tissue interactions with bacteria by competitive binding (Naidu, 2002). Deprivation of iron from their environment is assumed to be a central property in a variety of biological functions of

lactoferrin and results in suppression of microbial growth (Finkelstein et al., 1983). Some researchers reported that activated lactoferrin had direct bactericidal rather than bacteriostatic effect against several bacteria, including E. coli, Vibrio cholerae, S. mutans, S. pneumoniae, and Legionella pneumophila (Arnold et al., 1980, 1981, 1982; Bortner et al., 1986). Lactoferrin also possesses diverse physiological functions and interacts with various molecules including lipopolysaccharides (Elass-Rochard et al., 1995; Appelmelk et al, 1994), glycosaminoglycans (Mann et al., 1994) and cell-type-specific receptors on epithelial and immune cells (Ward et al., 2002; Brock, 2002, Suzuki et al., 2002; Baveye et al., 1999). The specific binding of activated lactoferrin against outer membrane proteins of gram-negative bacteria leads to inhibition of various cellular functions, deregulation of adhesion/ fibril synthesis, increases the permeability of bacterial outer membrane, increases the susceptibility of bacteria to hydrophobic antibiotics and lysozyme, and facilitates the release of lipopolysaccharide molecules from the cell membrane (Naidu, 2002; Ellison et al., 1988, 1991). Activated lactoferrin is recognized as GRAS material by the FDA and used in meat processing to control pathogens (Naidu et al., 2003; Kruger, 2004). Like activated lactoferrin, ovotransferrin is also expected to have similar bactericidal functions, and thus has a great potential as a natural antimicrobial compound in the future.

II-3.5. Antimicrobial activity of ovotransferrin As ovotransferrin has high affinities to di- or tri-valent metallic ions (Valenti et al., 1983b; Tranter and Board, 1982; Feeney, 1964), irons are naturally sequestered and rendered unavailable for the growth of microorganisms. Generally, lactoferrin in mammals functions mainly to defend against infections, but ovotransferrin is reported to be related mainly to iron transport like transferrin in mammals (Giansanti et al., 2002). Valenti et al. (1983b) reported that ovotransferrin possesses structurally-dependent antibacterial activity rather than iron deprivation, which is considered to be the major cause of antibacterial activity of lactoferrin in spite of remarkable structural similarities with lactoferrin. Many studies have been conducted to determine the antimicrobial effects of ovotransferrin against various bacterial species. Ovotransferrin has demonstrated strong

antibacterial activity against a wide spectrum of bacteria, including Escherichia coli (Shade and Carolin, 1944; Valenti et al., 1983a), Pseudomonas spp., Streptococcus mutans (Valenti et al., 1983a), Staphylococcus aureus and Bacillus cereus (Ibrahim, 1996; 1997), and Salmonella enteritidis (Baron et al., 1997; 2000). Valenti et al. (1985) reported that 99 among 100 species of Candida were sensitive to the antimicrobial action of ovotransferrin. Also, ovotransferrin is reported to have antiviral activity against herpes virus of Marek’s disease. Ovotransferrin is more potent against avian herpes virus than bovine lactoferrin and human lactoferrin. Even though ovotransferrin has lower activities against herpes simplex virus (HSV) than human transferrin and bovine lactoferrin, it is capable of inhibiting viral infections in the egg white of birds (Giansanti et al., 2002). Such a phenomenon may be attributed to different features of the binding sites. Lactoferrin possesses specific binding sites consisting of a cluster of positive charges at the N-terminus lobe, which are responsible for binding to glycosaminglycans (Mann et al., 1994; Wu et al., 1995) of some viruses including HSV (WuDunn and Spear 1989; Roderiquez et al., 1995), whereas ovotransferrin does not exhibit such a cluster of positive charges at N-terminus region (Williams et al., 1982, Metz-Boutigue et al., 1984, Jeltsch et al., 1987). Therefore, ovotransferrin is lacking in antiviral activity against HSV.

II-3.5.1. Mechanism of antibacterial action of ovotransferrin Although many researchers are interested in biological functions of ovotransferrin, its antimicrobial action and roles in physiological systems have not been defined clearly. Ovotransferrin plays a role in chelating iron and possibly minerals into stable complexes so that essential minerals required for microbial growth become deficient. Based on this theory, many researchers assumed that the antibacterial activity of ovotransferrin is attributed mainly to iron deprivation (Alderton et al., 1946; Fraenkel-Conrat and Feeney, 1950), which will inhibit the growth of bacteria (Arnold et al., 1980; Bullen et al., 1972; Valenti, 1983a; Mayes and Takeballi, 1983). Recently, however, several conflicting studies show the mechanisms of antibacterial action of ovotransferrin are ascribed not only to iron deprivation but also from direct interactions between ovotransferrin and microorganisms (Valenti et al., 1985). Considering that ovotransferrin has remarkable structural similarity to that of human

lactoferrin, it is postulated that the antimicrobial action could be related to direct reactions of ovotransferrin with bacterial membrane in a mamer similar to human lactoferrin (Valenti et al., 1983a; 1987). Ovotransferrin may permeate bacterial outer membranes, reach the inner membrane, and lead to permeability of other ions and dissipation of electrical potential. Thus, ovotransferrin can exert antibacterial action against gram-negative bacteria (Aguilera et al., 2003; Ibrahim, 1997). Many studies supporting direct reactions between ovotransferrin and microorganisms have been reported (Arnold et al., 1977, 1980, 1981, Valenti et al., 1982, 1985). Ovotransferrin saturated with iron displayed antifungal activity toward Candida spp. (Valenti et al., 1985), and ovotransferrin saturated with iron has more powerful bactericidal activity against E. coli than apo-ovotransferrin regardless of the amount of iron available. Ibrahim et al. (2000) suggested that the N-terminal lobe of ovotransferrin had more potent bactericidal activity against S. aureus than the C-terminal lobe regardless of the degree of iron saturation. Also, Davis and Reeves (2002) reported that the antimicrobial activity of ovotransferrin was attributed to the bactericidal domain, which is located in the N-terminal of ovotransferrin molecule rather than its iron binding capacity. In conclusion, the causes of antibacterial effect of ovotransferrin are not fully understood, and further studies related to the mechanisms of antibacterial activity of ovotransferrin are needed.

II-3.5.2. Factors affecting antibacterial activity of ovotransferrin The antimicrobial activity of ovotransferrin can be influenced by several factors such as bicarbonate and citrate concentrations, pH, composition of medium, exposure time, and bacterial species or strains (Valenti et al., 1981, 1983a; Brock, 1985). The bicarbonate in a medium can increase bacteriostatic effect, but does not affect antifungal activity of ovotransferrin (Valenti et al., 1985). The antimicrobial effect of ovotransferrin against E. coli is the highest at pH 8.0 (Valenti et al., 1983a). An excess of citrate can annul the antimicrobial activity of ovotransferrin by chelating iron, indicating that the iron binding capacity of ovotransferrin is affected by metal chelaters such as citrate (Valenti et al., 1983a; Phelpa and Antonini, 1975). However, it is also well known that the effect of citrate on

antibacterial activities of ovotransferrin depends on the species of microorganism. For example, E. coli have iron transport system mediated by citrate (Frost and Rosemberg, 1973), and thus can annul the antibacterial activity of ovotransferrin. However, addition of excess citrate does not influence the antimicrobial activity of ovotransferrin against S. aureus and Candida spp, which do not have iron citrate transport systems (Valenti et al., 1980; Valenti et al., 1985). The anitibacterial effect of ovotransferrin is variable among strains of microorganisms. Some species such as E. coli, Pseudomonas spp, and S. mutans are the most sensitive to ovotransferrin, whereas other species such as S. aureues. Proteus spp and Klebsiella spp are the most resistant species (Valenti et al., 1983a). Valenti et al. (1980) reported that some species of Staphylococci were resistant to ovotransferrin, while others were sensitive to the action of ovotransferrin. The different behavior of bacterial strains against ovotransferrin is attributed to the differences in iron transport systems in genus Staphylococcus. Some bacterial species lacking in iron transport system are much more sensitive to ovotransferrin when exposed to an iron-deficient environment. Also, the antimicrobial activity of ovotransferrin is affected by media components or matrix. Valenti et al. (1982) reported that the antibacterial activity increased as the amount of free or Sepharose-bound ovotransferrin increased. In addition, Sepharose-bound ovotransferrin showed higher antimicrobial activity than free ovotransferrin, which could be attributed to the higher stability of matrix-bound proteins. As ovotransferrin can bind metal ions such as Cu, Al, Zn, and Fe, the antimicrobial effect of ovotransferrin is decreased if it binds other metal ions besides iron. However, ovotransferrin saturated with metals as such zinc and iron still is reported to display antimicrobial activities (Ibrahim, 1997; Valenti et al., 1985; 1987). For example, Zn2+-loaded ovotransferrin was found to have more bactericidal activity than apo-ovotransferrin and other metal bound ovotransferrin. According to their explanation, the antimicrobial effect of Zn2+- ovotransferrin is attributed to the direct interaction of Zn2+-ovotransferrin molecule against the surface of bacteria rather than iron deprivation from medium (Valenti et al, 1987), and its bactericidal activity is not affected by the degree of iron saturation. In addition, the iron-

saturated N-lobe has almost two-fold greater antimicrobial activity than iron-saturated ovotransferrin (Ibrahim et al., 1998).

II-3.6. Bactericidal peptides of ovotransferrin Besides reports about antibacterial activity of intact ovotransferrin, certain peptides of ovotransferrin obtained after protein hydrolysis have been reported to have strong antimicrobial activity against some microorganisms. OTAP-92 is a cationic fragment consisting of 92 amino acid residues located within 109-200 sequence of N-lobe of ovotransferrin and contains six cysteines engaged in three intra-domain disulfide bridges. The unique structural motif of OTAP-92 plays an important role in the biological function of ovotransferrin (Ibrahim et al., 1998; Navasota et al., 1988). The bactericidal peptide is isolated by diluted-acid hydrolysis, which cleaves specifically at Asp-X sequence of ovotransferrin and produces a mixture of large disulfide-containing peptides (Ibram et al., 1998; Ingres, 1983). After proteolysis of ovotransferrin, the bactericidal domain was isolated from the hydrolysate of ovotransferrin using gel filtration and reversed-phase HPLC (Ibrahim et al., 1998). The cationic peptide corresponds to the sequence of ovotransferrin located at the lip of the N2-domain of the N-lobe of ovotransferrin. The cationic peptide of ovotransferrin exhibited potent bactericidal activity against gram-positive bacteria such as S. aureus, Bacillus subtilis, and B. cereus as well as gram-negative bacteria such as E. coli and P. aeruginosa. OTAP-92 is reported to display antibacterial activity against gram negative bacteria by crossing the bacterial outer membrane and damaging the cytoplasmic membrane (Ibrahim et al., 2000). According to analysis of computer-assisted multiple alignment, OTAP-92 has a significant structural similarity with insect defensins and bovine lactoferrin at some regions (Ibrahim et al., 1998). As the cationic antimicrobial peptide can cross the outer membrane by a self promoted uptake and dissipate the transmembrane electrochemical potential, it exhibits bactericidal properties against gram-negative bacteria (Ibrahim et al., 2000). However, when disulfide bonds of OTAP-92 were reduced, the bactericidal property was lost. Therefore, the maintenance of unique tertiary structure of OTAP-92 was essential for the accomplishment of antimicrobial activity (Ibrahim et al., 1998). Although the peptides of ovotransferrin with

antivirial effects have significant sequence homology with bovine lactoferrin peptides that show antiviral activity against MDV, intact hen ovotransferrin and bovine lactoferrin show stronger antiviral activity than the peptides. Therefore, maintaining native conformation is important for bovine lactoferrin and hen ovotransferrin to have antiviral capabilities (Giansanti et al., 2005).

III. Summary E. coli O157:H7 is considered as an emerging strain due to frequent disease outbreaks and meat product recalls. Also, the U.S. Department of Agriculture currently established a “zero tolerance” policy for L. monocytogenes in ready-to-eat (RTE) meat products due to risks of listeriosis. So prevention or minimizing the contamination of E. coli O157:H7 and L. monocytogenes during meat processing or meat product is very important for meat industry. Egg has potential as a natural antibacterial source because it contains antimicrobial proteins such as ovotransferrin, egg yolk immunoglobulins (IgY), and lysozyme. As IgY has a variety of advantages compared to traditional antibodies, many studies of passive immunization using IgY in fish and human have been reported. But some limitations in purification of IgY from egg yolk seems to inhibit its practical use. Therefore, for broad application of IgY as a antimicrobial agent, a simple and efficient protocol for IgY separation should be developed. Ovotransferrin transports and scavenges Fe (III) in poultry and has a significant similarity to lactoferrin, which has been currently recongnized as safe material (GRAS) and is permitted to use as an antimicrobial at meat processing or meat products. Also, a variety of methods including HPLC or ion exchange chromatography, gel chromatography, and affinity chromatography have been developed to separate ovotransferrin from egg white. Also, studies related to antibacterial activity of ovotransferrin have been reported, however, no application of ovotransferrin toward real products has been reported until now inspite of the attributes of ovotransferrin, which has structural and functional similiarties to lactoferrin. Ovotransferrin has two iron binding sites and requires anions, which serve as a bridging ligand between protein and metal ion, to bind specifically with Fe(III). Some anions in Fe3+ release are closely related to the opening of the domain in the N- or C-lobe.

The mechanism of antimicrobial activity of ovotransferrin is still controversial. Initially, iron binding capacity of ovotransferrin has been believed to be major mechanism of antimicrobial action of ovotransferrin. As ovotransferrin has high affinities to di- or tri-valent metallic ions, irons are naturally sequestered and rendered unavailable for the growth of microorganisms. However, subsequent studies claim that ovotransferrin reacts directly with microorganisms. Ovotransferrin has demonstrated antibacterial activity against a wide spectrum of bacteria, including E. coli, Pseudomonas spp., S. mutans, S. aureus, B. cereus, S. enteritidis, and Candida. Also, antibacterial activity of ovotransferrin has been reported to be affected by several factors such as sodium bicarbonate, citrate, pH, media, bacterial species, stability of ovotransferrin, and Zn2+ among others. In conclusion, through this review, it seemed to be necessary first to develop simple and efficient protocols for large-scale production of natural antimcriobials from eggs to use them with real products or meat processing. Also, to incresase the antimicrobial effectiveness of ovotransferrin, the understanig of precise mechanism of its antimicrobial action is considered to be essential work prior to ovotransferrin application into real products or meat processing. Furthermore, it seems to be a valuable work to apply ovotransferrin to real products.

References Abdallah, F B., and EL Hage Chahine, J. M. 1998. Hen ovo-transferrin, interaction with bicarbonate and iron uptake. Eur. J. Biochem. 258:1022-1031. Abdallah, F. B., and El Hage Chahine, J. M. 1999. Transferrins, the mechanism of iron release by ovotransferrin. Eur. J. Biochem. 263: 912-920. Abola, J. E., Wood, M. K., Chweh, A., Abraham, D., and Pulsinelli, P. D. 1982. Structure of hen ovotransferrin at 5Å resolution. In “The Biochemistry and Physiology of Iron”. Saltman, P. and Hegenauer, J. Eds. Amsterdam, The Netherlands: Elsevier Biomedical, Inc. pp. 27 – 34. Abul-Raouf, U. M., Beuchat, L. R., and Ammar, M. S. 1993. Survival and growth of Escherichia coli O157:H7 on salad vegetables. Appl. Environ. Microbiol. 59: 1999-2006.

Aguilera, O., Luis, M., Quirosb, L. M., and Fierro, J. F. 2003. Transferrins selectively cause ion efflux through bacterial and artificial membranes. FEBS Letters 548: 5-10. Ahlborn, G. J., Clare, D. A., Sheldon, B. W., and Kelly, R. W. 2006. Identification of eggshell membrane proteins and purification of ovotransferrin and ß-NAGAse from hen egg white. Prot. J. 25(1): 71-81. Aisen, P. 1980. Iron transport and storage proteins. Ann. Rev. Biochem. 49: 357-393. Aisen, P. 1989. Physical biochemistry of the transferrins-uptake 1984-1988. Phys. Bioinorg. Chem. SeR 5, Iron carriers and ion proteins (Loehr, T., ed.) pp 353-371. VCH, New York. Aisen, P. and Listowsky, I. 1980. Iron transport and storage proteins. Annu. Rev. Biochem., 49: 357-393. Akerstrom, B., Bodin, T., Reis, K., and Bjorck, L. 1985. Protein G: A powerful tool for binding and detection of monoclonal and polyclonal antibodies. J. Immunol. 135(4): 2589-2592. Akita, E. M., and Nakai, S. 1993. Comparison of four purification methods for the production of immunoglobulins from eggs laid by hens immunized with an enterotoxigenic E. coli strain, J. Immunol. Methods 160: 207-214. Akita, E. M., and Nakai, S. 1992. Immunoglobulins from egg yolk: isolation and purification. J. Food Sci. 57: 629-634. Alderton, G., Ward, W. H., and Fevold, H. L. 1946. Identification of the bacterial inhibiting iron-binding protein of egg white as conalbumin. Arch. Biochem. 11: 9-13. Al-Mashikhi, S. A., and Nakai, S. 1987. Separation of ovotransferrin from egg white by immobilized metal affinity chromatography. Agric. Biol. Chem. 51(11): 2881-2887. Amersham Pharmacia Biotech. 2000. Antibody purification handbook. No. 18: 1037-46. Andeson, B. F., Baker, H. M., Norris, G. E., Rice, D. W. and Baker, E. N. 1989. Structure of human lactoferrin-crystallographic structure analysis and refinement at 2.8 Ǻ resolution. J. Mol. Biol. 209: 711-734. Appelmelk, B. J., An, Y. Q., Greets, M., This, B.G., de Boer, H. A., Malaren, D. M., de Graff, J., and Niue’s, J. H. 1994. Lactoferrin is a lipid A-binding protein. Infect. Immunol. 62(6): 2628-32.

Arasteh, N., Aminirissehei, A. H., Yousif, A. N., Albright, L. J., and Durance, T. D. 2004. Passive immunization of rainbow trout (Oncorhynchus mykiss) with chicken egg yolk immunoglobulins (IgY). Aquaculture 231: 23-26. Arnold, R. R. Brewer, M., and Gauthier, J. J, 1980. Bactericidal activity of human lactoferrin: sensitivity of a variety of microorganisms. Infect. Immunol. 28: 893-898. Arnold, R. R., Cole, M. F., and McGheem, J. R. 1977. A bactericidal effect for human lactoferrin. Science 197: 263-265. Arnold, R. R., Russel, J. E., Chempion, W. J., and Gauthier, J. J. 1981. Bactericidal activity of human lactoferrin: influence of physical conditions and metabolic state of the target microorganism. Infect. Immunol. 32: 655-660. Arnold, R. R., Russell, J. E., Champion, W. J., Brewer, M., and Gauthier, J. J. 1982. Bactericidal activity of human lactoferrin: differentiation from the stasis of iron deprivation. Infect. Immunol. 35: 792-797. Autio, T., Sateri, T., Fredirksson-Ahomaa, M., Rahkio, M., Lunden, J., Korkeala, H. 2000. Listeria monocytogenes contamination pattern in pig slaughter houses. J. Food Prot. 63: 1438-1442. Awade, A. C. 1996. On hen egg fractionation: applications of liquid chromatography to the isolation and the purification of hen egg white and egg yolk proteins. Eur. Food Res. Technol. 202: 1-14. Awade, A.C., Moreau, S., Molle, D., Brule, G., and Maubois, J. L. 1994. Two-step chromatographic procedure for the purification of hen egg white ovomucin, lysozyme, ovotransferrin and ovalbumin and characterization of purified proteins. J. Chromatogr. A 677: 279-288. Azari, P. R., and Feeney, R. E. 1958. Resistance of metal complexes of conalbumin and transferrin to proteolysis and to thermal denaturation. J. Biol. Chem. 232: 293-302. Azari, P., and Baugh, R. F. 1967. A simple and rapid procedure for preparation of large quantities of pure ovotransferrin. Arch. Biochem. Biophys. 118: 138-144. Bade, H. and Stegemann, H. 1984. Rapid method of extraction of antibodies from hen egg yolk, J. Immunol. Methods 72: 421-426.

Bail, P. K. and Harris, W. R. 1989. Cooperativity and heterogeneity between the two binding sites of diferric transferrin during iron removal by pyrophosphate. J. Am. Chem. Soc. 111: 4457-4461. Bailey, C. T., Byrne, C., Chrispell, K., Molkenbur, C., Sackett, M., Reid, K., McCollum K., Vibbard, D., and Catelli, R. 1997. Effect of a covalently attached synergistic anion on chelator-mediated iron-release from ovotransferrin: Additional evidence for two concurrent pathways. Biochem. 36: 10105-10108. Bailey, C. T., Patch, M. G., and Carrano, C. J. 1988. Affinity labels for the anion-binding site in ovotransferrin. Biochem. 27(17): 6276-6282. Bain, J. A., and Deutsch, H. F. 1948. Separation and characterization of conalbumin. J. Biol. Chem. 172: 547-555. Baker, E. N., Rumball, S. V., and Anderson, B. F. 1987. Transferrin: insights into structure and function from studies on lactoferrin, Trends Biochem. Sci. 12: 350-353. Baldwin, D. A. and MR de Sousa, M. 1981. The effect of salts on the kinetics of iron release from N-terminal and C-terminal monoferrictransferrin. Biochem. Biophy. Res. Commun. 99: 1101-1107. Banks, J. G., Board, R. G., and Sparks, N.H.C. 1986. Natural antimicrobial systems and their potential in food preservation of the future. Biotechnol. Appl. Biochem. 8: 103-147. Baron, F., Fauvel, S., and Gautier, M. 2000. Egg Nutrition and Biotechnology, Sim, J. S., Nakai, S., and Guenter, W. Eds. CAB Int. pp 417-440. Baron, F., Gautier, M., and Brule, G. 1997. Factors involved in the inhibition of Salmonella enteritidis in liquid egg white. J. Food Protect. 60: 1318-1323. Bates, G. W. and Schlabach, M. R. 1975. The nonspecific binding of Fe 3+ to transferrin in the absence of synergistic anions. J. Biol. Chem. 250: 2177-2181. Baveye, S., Elass, E., Mazurier, J., Spik, G., and Legrand, D. 1999. Lactoferrin: a multifunctional glycoprotein involved in the modulation of the inflammatory process. Clin. Chem. Lab. Med. 37: 281-286. Beal, R. K., Wigeley, P., Powers, C., Hulme, S. D., Barrow, P. A., and Smith, A. L. 2004. Age at primary infection with Salmonella enterica serovar typhimurium in the chicken

influences persistence of infection and subsequent immunity to re-challenge. Vet. Immuno. Immunopathol. 100: 151-154. Behn, I., Hommel, U., Erhard, M., Hlinak, A., Schade, R., Schwarzkopf, C., and Staak, C. 2001. Use of polyclonal avian antibodies. In Chicken Egg Yolk Antibodies, Production and Application. Sringer Verlag pp. 255. Bell, B. P., Goldoft, M., Griffin, P. M., Davis, M. A., Gordon, D. C., Tarr, P. I., Bartleson, C. A., Lewis, J. H., Barrett, T. J. and Wells, J. G. 1994. A multistate outbreak of Escherichia coli O157:H7 associated bloody diarrhea and hemolytic uremic syndrome from hamburgers. The Washington experience. J. Am. Med. Assoc. 272: 1349-1353. Bell, C. 1998a. Accord is reached in food-poisoning case. New York Times. May 27. A16. Bell, C. 1998b. Unpasteurized apple juice. USA and Canada. pp. 25-28. In E. coli: A practical approach to the organism and its control in foods. Chapman and Hall. London, UK. Bell, C. 2002. Approach to the control of entero-haemorrhagic Escherichia coli (EHEC). Int. J. Food Microbiol. 78: 197-216. Bellounis, L., Pakadaman, R., and El Hage Chahine, J. M. 1996. Apo-transferrin proton dissociation and interactions with bicarbonate in neutral media, J. Phys. Org. Chem. 9: 111-118. Benjamin, M. M., and Datta, A. R. 1995. Acid tolerance of enterohemorrhagic Escherichia coli. Appl. Environ. Microbiol. 61: 1669-1672. Beresford, M.R., Andrew, P.W., and Shama, G. 2001. Listeria monocytogenes adheres to many materials found in food-processing environments. J. Appl. Microbiol. 90: 1000- 1005. Blankenvoorde, M. F., Can’t Hof, W., Wlgreen-Weterings, E., van Steenbergen, T. J., Brand H. S., Verman, E.V., and Nieuw Amerongen, A. V. 1998. Cystain and cystain-derived peptides have antibacterial activity against the pathogen Porphyromonas gingivalis. Biol. Chem. 379: 1371-1375. Bortner, C. A., Miller, R. D., and Arnold, R. R. 1986. Bactericidal effect of lactoferrin on Legionella pneumophila. Infect. Immunol. 51: 373-377.

Brackett, R. E. 1988. Presence and persistence of Listeria monocytogenes in food and water. Food Technol. 42(4): 162-164. Brackett, R. E., Hae, Y. Y., and Doyle, M. P. 1994. Ineffectiveness of hot acids sprays to decontaminate Escherichia coli O157:H7 on beef. J. Food Prot. 57: 198-203. Brock, J. H. 1985. The Transferrins. In: Metalloproteins II. Harrison, P. (Ed). MacMillan Press, London. pp 183-262. Brock, J. H. 2002. The physiology of lactoferrin. Biochem. Cell Biol. 80: 1-6. Buchanan, R. L. and Doyle, M. P. 1997. Foodborne disease significance of Escherichia coli O157:H7 and other enterohemorrhagic E. coli. Food Tech. 51(10): 69-76. Buchanan, R. L., and Klawitter, L. A. 1990. Effects of temperature and oxygen on the growth of Listeria monocytogenes at pH 4.5. J. Food Sci. 55: 1754-1756. Bullen, J. J., Rogers, H. J., and Leigh, L. 1972. Iron-binding proteins in milk and resistance to Escherichia coli infection in infants. Br. Med. J. 1: 69-75. Burley, R. W., and Vadehra, D. V. 1989. The Avian Egg, Chemistry and Biology. John Wiley & Sons, Inc., New York. Butterworth, R. M., Gibson, J. R. and Williams, J. 1975. Electron-paramagnetic-resonance spectroscopy of iron-binding fragments of hen ovotransferrin. Biochem. J. 149: 559-563. Camenisch, G., Tini, M, Chilow, D., Kvietikova, I., Srinivas, V., Caro, J., Spielmann, P., Wenger, R. H., and Gassmann, M. 1999. General applicability of chicken egg yolk antibodies: the performance of IgY immunoglobulins raised against the hypoxia- inducible factor 1. FASEB 13: 81-88. Carlander, D. H., Kollberg, P. E., Larsson, A., and Wejaker, P. 2000. Peroral immunotherapy with yolk antibodies for the prevention on and treatment of enteric infections. Immunol. Res. 21: 1-6. Carlander, D., Stalberg. J., and Larsson, A. 1999. Chicken antibodies : a clinical chemistry perspective. Ups J. Med. Sci. 104(3): 179-89. Carminati, D., and Carini, S. 1989. Antimicrobial acitivity of lysozyme against Listeria monocytogenes in milk. Microbiol. Aliments Nutr. 7: 49-56. CDC. 2002. Centers for Disease Control and Prevention. Multistate outbreak of listeriosis- United States. Morb. Mortal Wkly. Rep. 51: 950-951.

Chang, H. M., Lu, T. C., Chen, C. C., Tu, Y. Y., and Hwang, J. Y. 2000. Isolation of immunoglobulin from egg yolk by anionic polysaccharides. J. Agric. Food Chem. 48(4): 995-999. Cheuk, M. S., Keimg, W. M., and Loh, T. T. 1987. The effect of pH on the kinetics of iron release from diferric ovotransferrin induced by pyrophosphate J. Inorg. Biochem. 30: 121-131. Chung, C. M., Chan, S. L., and Shimizu, S. 1991.Purification of transferrins and lactoferrin using DEAE Affi-gel blue. Int. J. Biochem. 23: 609-616. Clouser, C. S., Doores, S., Mast, M. G., and Knabel, S. J., 1995. The role of defeathering in the contamination of turkey skin by Salmonella species and Listeria monocytogenes. Poultry Sci. 74 (4): 723-731. Conner, D. E., and Kotrola, J. S. 1995. Growth and survival of Escherichia coli O157:H7 under acidic conditions. Apple Environ. Microbiol. 61: 382-385. Cook, C. L., Pao, W., Frica, J. R., Anderson, B. E., and Fryer, J. P. 2001. Simple purification methods for an α-galactose-specific antibody from chicken eggs. J. Biosci. Bioeng. 91(3): 305-310. Croguennec, T., Nau, F., Pezennec, S., Piot, M., and Brulé, G. 2001. Two-step chromatographic procedure for the preparation of hen egg white ovotransferrin. Eur. Food Res. Technol. 212: 296-301. Cunningham, F. E. and Lineweaver, H. 1965. Stabilization of egg white proteins to pasteurization temperatures above 60o C. Food Technol. 19: 136-141. Davalos-Pantoja, L., Ortega-Vinuesa, J. L., Bastos-Gonzalez, D., and Hidalgo-Alvarez, R. 2000. A comparative study between the adsorption of IgY and IgG on latex particles. Biomat. Sci. Polym. Ed. 11: 657-673. Davies, B. D., Dulbecco, R., Eien, H. N., and Ginsberf, H. S. 1980. In Microbiology 3rd ed. Hagerstown, Maryland: Harper & Row. pp. 564. Davis, C. and Reeves, R. 2002. High value opportunities from the chicken egg. A report for the rural industries research and development corporation. Rural industries research and development corporation publication. pp. 1-32.

De Roin, M. A., Foong, S. C., Dixon, P. M., and Dickson, J. S. 2003. Survival and recovery of Listeria monocytogenes on ready-to-eat meats inoculated with a desiccated and nutritionally depleted dustlike vector. J. Food Prot. 66: 962-969. Deignan, I. A., Alwan, L. M., Kelly, J., and O’Farrelly, C. 2001. Hen egg yolk prevents bacterial adherences: a novel function for a familiar food. J. Food. Sci. 66: 158-161. Devi, C. M., Bai, M. V., Lal, A. V., Umashanka, P. R., and Krishnan, L. K. 2002. An improved method for isolation of anti-viper venom antibodies from chicken egg yolk. J. Biocherm. Biophys. Methods 51: 129-138. Dewan, J. C., Mikami, B., Hiros, M., and Sacchettini, C. 1993. Structural evidence for a pH- sensitive dilysine trigger in the hen ovotransferrin N-lobe: Implication for transferrin iron release. Biochem. 32: 11963-11968. Dorn, C. R. 1993. Review of foodborne outbreaks of Escherichia coli O157:H7 infection in the western United States. J. Am. Vet. Med. Assoc. 203: 1583-1587. Doyle, M. P. 1991. Escherichia coli O157:H7 and its significance in food. Int. J. Food Microbiol. 12: 289-292. Duffy, E. A., Belk, K. E., Sofos, J. N., Bellinger, G. R., Pape, A., and Smith, G. C. 2001. Extent of microbial contamination in United States pork retail products. J. Food Prot. 64: 172-178. El Hage Chahine, J. M. and Pakadaman, R. 1995. Transferrin, a mechanism for iron releases. Eur. J. Biochem. 230: 1101-1110. Elass-Rochard, E., Roseanu, A., Legrand, D., Trif, M., Salmon, V., Motas, C., Montreuil, J., and Spik, G. 1995. Lactoferrin-lipopolysaccharide interaction: involvement of the 28-34 loop region of human lactoferrin in the high-affinity binding to Escherichia coli 055B5 lipopolysaccharide. Biochem. J. 312: 839–845. Ellison, R. T. III, Giehl, T. J., and Laforce, F. M. 1988. Damage of the outer membrane of enteric gram-negative bacteria by lactoferrin and transferrin. Infect. Immuol. 56(11): 2774-2781. Ellison, R. T. III, and Giehl, T. J. 1991. Killing of gram-negative bacteria by lactoferrin and lysozyme. J. Clin. Invest. 88: 1080-1091.

Erhard, M. H., Schmidt, P., and Zinsmeister, P. 2000. Adjuvant effects of various lipopeptides and interferon-gamma on the humoral immune response of chickens. Poult. Sci. 79: 1264-1270. Farber, J. M., and Peterkin, P. I. 1991. Listeria monocyotogenes, a food-borne pathogen. Microbiol Rev. 55: 476-511. Farber, J. M., Coates, F., and Daley, E. 1992. Minimum water activity requirements for the growth of Listeria monocytogenes. Lett. Appl. Microbiol. 15: 103-105. Farrell, B. L., Ronner, A. B. and Lee Wong, A. C. 1997. Attachment of Escherichia coliO157:H7 in ground beef to meet grinders and survival after sanitation with chlorine and peroxyacetic acid. J. Food Prot. 61: 817-822. Feeney, R. E. 1964. Egg proteins. In Symposium on Foods: Proteins and their Reactions. Schults HW, Anglemier. Feingold, D. S., Goldman, J., and Kuritz, H. M. 1968. Locus of the action of serum and the role of lysozyme in the serum bactericidal reaction. J. Bacter. 96: 2118-2126. Fichtali, J., Charter, E. A., Lo, K.V., and Nakai, S. 1992. Separation of egg yolk immunoglobulins using an automated liquid chromatography system. Biotechnol. Bioeng. 40: 1388-1394. Fichtali, J., Charter, E. A., LO, K. V., and Nakai, S. 1993. Purification of antibodies form industrially separated egg yolk. J. Food Sci. 58(6): 1282-1285. Finkelstein, R. A., Skirting, C.V., and McIntosh, M. A. 1983. The role of iron in microbe- host interactions. Rev. Infect. Dis. 5: S759-s777. Firstenberg-Eden, R. 1997. E. coli rapid detection system. A rapid method for the detection of Escherichia coli O157 in meat and other foods. J. Food Prot. 60: 219-225. Fost, G. E., and Rosemberg, H. 1973. The citrate-dependent iron transport system in Escherichia coli K-12. Biochim. Biophys. Acta 330: 90-101. Foulke, J. E. 1994. How to outsmart dangerous E. coli strain. FDA Consumer. 28(1): 7-10. Fraenkel-Conrat, H. 1950. Comparison of the iron-binding activities of conalbumin and of hydroxylamidoproteins. Arch. Biochem. 28: 452-463. Fraenkel-Conrat, H., and Feeney, R. E. 1950. The metal-binding activity of conalbumin. Arch. Biochem. 29: 101-113.

Fryer, J., Firca, J., and Leventhal, J. 1999. IgY antiporcine endothelial cell antibodies effectively block human antiporcine xenoantibody binding. Xenotransplantation 6: 98- 109. FSIS. http://www.fsis.usda.gov/OPPDE/rdad/FRPubs/00-022N/ApprovalofIngredients.htm Gee, S. C., Bate, I. M., Thomas, T. M., and Rylatt, D. B. 2003. The purification of IgY by preparative electrophoresis. Protein Expression Purification 30: 151-155. George, S. M., and Lund, B. M. 1992. The effect of culture medium and aeration on growth of Listeria monocytogenes at pH 4.5. Lett. Appl. Microbiol. 15: 49-52. Giansanti, F., Massucci, M.T., Giardi, M.F., Nozza, F., Pulsinelli, E., Nicolini, C., Botti, D., and Antonini, G. 2005.Antivirial activity of ovotransferrin derived peptides. Biochem. Biophy. Res. Comm. 331: 69-73. Giansanti, F., Rossi, P., Massucci, M. T., Botti, D., Antonini, G., Valenti, P., and Seganti, L. 2002. Antiviral activity of ovotrasnferrin discloses an evolutionary strategy for the defensive activities of lactoferrin. Biochem. Cell Biol. 80: 125-130. Gill, A. O., and Holley, R. A. 2000. Surface application of lysozyme, nisin, and EDTA to inhibit spoilage and pathogenic bacterial on ham and bologna. J. Food Prot. 63: 1338- 1436. Giovannacci, I., Ragimbeau, C., Queguiner, S., Salvat, G., Vendeuvre, J. L., Carlier, V., and Ermel, G. 1999. Listeria monocytogenes in pork slaughtering and cutting plants. Use of RAPD, PFGE and PCR-REA for tracing and molecular epidemiology. Int. J. Food Microbiol. 53: 127-140. Glass, K. A., and Doyle, M. P. 1989. Fate of Listeria monocytogenes in processed meat products during refrigerated storage. Appl. Environ. Microbiol. 55: 1565-1569. Gombas, D. E., Chen, Y., Clavero, R. S., and Scott, V. N. 2003. Survey of Listeria monocytogenes in ready-to-eat foods. J. Food Prot. 66: 559-569. Greene, C. R., and Holt, P. S. 1997. An improved chromatographic method for the separation of egg yolk IgG into subpopulations utilizing immobilized metal ion (Fe3+) affinity chromatography. J. Immunol. Methods 209: 155-164. Gregory, W. 1995. IgY: Clues to the origins of modern antibodies. Immunol. Today 16: 392~398.

Griffiths, E., and Humpherys, J. 1977. Bacteriostatic effect of human milk and bovine colostrum on Escherichia coli: Importance of bicarbonate. Infect. Immunol. 15(2): 396- 401. Guerin, C., and Brule, G. 1992. Separation of three proteins from egg white. Sciences des- Aliments 112(4): 705-720. Guo, M., Harver, I., Yang, W., Coghill, L., Campopiano, D. J., Parkinson, J. A., MacGillivrary, R. R. A., Harris, W. R., and Sadler, P. J. 2003. Synergistic anion and metal binding to the ferric ion-binding protein from Neisseria gonorrhoeae. J. Biolo. Chem. 278(4): 2490-2502. Gurov, A. N., Mukhin, M. A., Larichev, N. A., Lozinskaya, N. V., and Tolstoguzoc, V. S. 1983. Emulsifying properties of proteins and polysaccharides. I. Methods of determination of emulsifying capacity and emulsion stability. Colloids Surf. 6: 35-42. Gyles, C.L. 1994. Classification of Escherichia coli. p. 31-52. In Escherichia coli in Domestic Animals and Humans. CAB international. Wllingford, UK. Hansen, P., Scoble, J. A., Hanson, B., and Hoogenraad, N. J. 1998. Isolation and purification of immunoglobulins from chicken eggs using thiophilic interaction chromatography. J. Immunol. Methods 215: 1-7. Harmsen, M. C., Swart, P. J., de Bethune, M. P., Pauwels, R., De Clercq, E., The, T. H., and Meijer, D. K. 1996. Antiviral effects of plasma and milk proteins: lactoferrin shows potent activity against both human immunodeficiency virus and human cytomegalovirus replication in vitro. J. Infect Dis. 174(2): 444-4445. Hathcox, A. K., Beuchat, L. R., and Doyle M. P. 1995. Death of enterohemorrhagic Escherichia coli O157:H7 in real mayonnaise and reduced-calorie mayonnaise dressing as influenced by initial population and storage temperature. Appl. Environ. Microbiol. 61: 4172-4177. Hatta, H., Kim, M., and Yamamoto, T. 1990. A novel isolation method for hen yolk antibody, “IgY”. Agric. Biol. Chem. 54: 2531-2535. Huffman, R. D. 2002. Current and future technologies for the decontamination of carcasses and fresh meat. Meat Sci. 62: 285-294.

Hughey, J. L., and Johnson, E. A. 1987. Antimicrobial activity of lysozyme against bacteria involved in food spoilage and food-borne disease. Apple. Environ. Microbiol. 53: 2165- 2170. Hughey, J. L., Wilger, P. A., and Johnson, E. A. 1989. Antibacterial activity of hen egg white lysozyme against Listeria monocytogenes Scott A in foods. Appl. Environ. Microbiol. 55: 631-638. Ibrahim, H. R. 1997. Insights into the structure-function relationships of ovalbumin, ovotransferrin, and lysozyme. In : Hen Eggs, Their Basic and Applied Science (Yamamoto T, Juneja, L.R., Hatta, H., and Kim, M. eds.). pp. 37-56. CRC Press, Inc. New York. Ibrahim, H. R., Higashiguchi, S., Koketsu, M., Juneja, L. R., Kim, M., Yamamoto, T., Sugnoto, Y., and Aoki, T. 1996a. Partially unfolded lysozyme at neutral pH agglutinates and kills Gram-negative and Gram-positive bacteria through membrane damage mechanism. J. Agric. Food Chem. 44: 3799-3806. Ibrahim, H. R., Higashiguchi, S., Sugimoto, Y., and Aoki, T. 1996b Antimicrobial synergism of partially-denatured lysozyme with glycine: Effect of sucrose and sodium chloride. Food Res. Int. 29(8): 771-777. Ibrahim, H. R., Iwamori, E., Sugimoto, Y., and Aoki, T. 1998. Identification of a distinct antibacterial domain within the N-lobe of ovotransferrin. Biochim. Biophys. Acta 1401(3): 289-303. Ibrahim, H. R., Sugimoto, Y., and Aoki, T. 2000. Ovotransferrin antimicrobial peptide (OATP-92) kills bacteria through a membrane damage mechanism. Biochim. Biophys. Acta 1523: 196-205. Iida T., Kanzaki, M., Nakkama, A., Kokubo, Y., Maruyama, T., and Kaneuchi, C. 1998. Detection of Listeria monocytogenes in humans, animals and foods. J. Vet. Med. Sci. 60: 1341-1343. Imberechts, H., Deprez, P., Van Driessche, E., and Pohl, P. 1997. Chicken egg yolk antibodies against F18ab fimbria of Escherichia coli inhibit shedding of F18 positive E. coli by experimentally infected pigs. Vet Microbiol. 54: 329-41. Ingris, A. S. 1983. Cleavage at aspartic acid. Methods Enzymol. 91: 324-332.

Integrin activation and cytoskeletal interaction are essential for the assembly of a fibronectin matrix. Cell 83(5): 715-24. Jeffrey, P. D., Bewley, M. C., MacGillivray, R. T. A., Mason, A. B., Woodworth, R. C., and Baker, E. N. 1998. Ligand-induced conformational change in transferrins: Crystal structure of the open form of the N-terminal half-molecule of human transferrin. Biochem. 37: 13978-13986. Jeltsch, J. M., Hen, R., Maroteaux, L., Garnier, J. M., and Chambon, P. 1987. Sequence of the chicken ovotransferrin gene. Nucleic Acids Res. 15: 7643-7645. Jensenius, J. C., Andersen, I., Hau, J., Crones, M., and Kock, C. 1981. Eggs: conveniently packaged antibodies. Methods for purification of yolk IgG. J. Immunol. Methods 46: 63- 68. Jin, L. Z., Baidoo, K., Maruquardt, R. R., and Frohlich, A. A. 1998. In vitro inhibition of adhesion of enterotoxigenic Escherichia coli K88 to piglet intestinal mucus by egg yolk antibodies. Immuno. Med. Microbiol. 21: 313-321. Johansen, C., Gram, L., and Meyer, A. S. 1994. The combined inhibitory effect of lysozyme and low pH on growth of Listeria monocytogenes. J. Food Prot. 57: 561-566. Johnson, E. A. 1994. Egg-white lysozyme as a preservative for use in foods. In: Egg Uses and Processing Technologies. New Developmemts. Sim, J. S. and Nakai, S. (Eds). International Cab, Wallingford. pp. 77-93. Jolles, P., and Jolles, J. 1984. What’s new in lysozyme research? Always a model system, today as yesterday. Mol. Cellular Biochem. 63(2): 165-189. Jordan, K. N., Oxford, L. and O’Byrne, C. P. 1999. Survival of low-pH stress by Escherichia coli O157:H7: correlation between alteration in the cell envelope and increases acid tolerance. Appl. Environ. Microbiol. 65: 3048-3055. Kihm, D. J., Leyer, G. J., An, G. H., and Johnson, E. A. 1994. Sensitization of heat-treated Listeria monocytogenes to added lysozyme in milk. Appl. Environ. Microbiol. 60: 3854- 3861. Kim, H., and Nakai, S. 1996. Immunoglobulin separation from egg yolk: a serial filtration system. J. Food Sci. 61(3): 510-512, 523.

Kim, H., and Nakai, S. 1998. Simple separation of immunoglobulin from egg yolk by ultrafiltration. J. Food Sci. 63(3): 485-490. Kim, J. W. 2004. Passive immunotherapy with egg-yolk antibodies (IgY) in human enteric infections: future prospects. In The 3rd International Symposium on Egg Nutrition for Health Promotion. Program booklet, Banff, Alberta, Canada. April. pp. 40. Kitabatake, N., Indo, K., and Doi, E. 1988. Limited proteolysis of ovalbumin by pepsin, J. Agric. Food Chem. 36: 417-420. Korpela, J., Salonen, E. M., Kuusela, P., Sarvas, M., and Vaheri, A. 1984. Binding of avidin to bacteria and to the outer membrane porin of Escherichia coli. FEMS Microbiol. Lett. 22: 3-10. Kotula, K. L., and Kotula, A. W. 2000. Microbial ecology of different types of food-fresh red meats. In The Microbiological Safety and Quality of Food (pp 359-388). B. M. Lund, T. C. Baird-Parker, & G. W. Gould (Eds), Gaithersburg, MD: Apen Publishers, Inc. Kretchmar, S. A., and Raymond, K. N. 1986. Biphasic kinetics and temperature dependence of iron removal from trasnferrin by 3, 4-LICAMS. J. Am. Chem. Soc. 108: 6216-6218. Kruger, C. L. 2004. GRAS notification: a case study. Prepared Foods 173(9): 55-56, 58. Kubo, R. T., Zimmermans, B., and Grey, H. M. 1973. Phylogeny of immunoglobulins. pp. 417- 477. In the antigens. Vol. 1. M. Sela, Ed. Academic Press, San Diego, CA. Kurokawa, H., Dewan, J. C., Mikami, B., and Sacchettini, J. C. 1999. Crystal structure of apo-hen ovotransferrin. J. Biol. Chem. 274: 28445-28452. Kurokawa, H., Mikami, B., and Hirose, M. 1995. Crystal structure of diferric hen ovotransferrin at 2.4 Ǻ resolution. J. Mol. Biol. 254: 196-207. Larive, C. K., Lunte, S. M., Zhong, M., Perkins, M. D., Wilson, S., Gokulrangan, G., Williams, T., Afroz, F., Schoneich, C., Derrick, T. S., Middaugh, C. R., and Bogdanowich-Knipp, S. 1999. Separation and analysis of peptides and proteins. Anal. Chem. 71: 389-423. Larsson, A., and Sjoquist, J. 1988. Chicken antibodies: A tool to avoid false positive results by rheumatoid factor in latex fixation tests. J. Immunol. Methods 108: 205-208. Larsson, A., Balow, R., Lindahl, T. L, and Forsberg, P. 1993. Chicken antibodies: Taking advantage of evolution –A review. Poultry Sci. 72: 1807-1812.

Larsson, A., Karlsson-Parra, A., and Sjoquist, J. 1991. Use of chicken antibodies in enzyme immunoassays to avoid interference by rheumatoid factors. Clin. Chem. 37(3): 411-414. Larsson, A., Wejaker, P., Forsberg, P., and Lindahl, T. 1992. Chicken antibodies: a tool to avoid interference by complement activation in ELISA, J. Immunol. Methods 156: 79-83. Lee, E. M., Sunwoo, H. H., Mennine, K., and Sim, J. S. 2002. In Vitro studies of chicken egg yolk antibody (IgY) against salmonella enteritidis and Salmonella typhimurium. Poultry Sci. 81: 632-641. Lee, S. B., Mine, Y., and Stevenson, R. M. W. 2000. Effects of hen egg yolk immunoglobulin in passive protection of rainbow trout against Yersinia ruckeri. J. Agri. Food Chem. 48: 110-115. Levay, P. F., and Viljoen, M. 1995. Lactoferrin: a general review. Haematologica 80: 252- 267. Leyer, G. J., Wang, L. L., and Johnson, E. A. 1995.Acid adaptation of Escherichia coli O157:H7 increases survival in acidic foods. Appl. Environ. Microbiol. 61: 3752-3755. Lillard, H. S. 1990. The impact of commercial processing procedures on the bacterial contamination and cross-contamination of broiler carcasses. J. Food Prot. 53(3): 202- 204, 207. Liu, S. T., Sugimoto, T., Azakami, H., and Kato, A. 2000a. Lipophilization of lysozyme by short and middle chain fatty acids, J. Agric. Food Chem. 48: 265-269. Liu, S. T., Azakami, H., and Kato, A. 2000b. Improvement in the yield of lipophilized lysozyme by the combination with Maillard-type glycoslyation. Nahrung 44(6): S407- s410.

Lonnerdal, B., and Keen, C. L. 1982. Metal chelate affinity chromatography of proteins. J. Appl. Biochem. 4: 203-208. Lou, Y., and Yousef, A. H. 1999. Characteristics of Listeria monocytogenes important to food processors. In: Listeria, Listeriosis and Food Safety. Ryser, E.T., Marth, E.H, Eds. New York; Marcel Dekker Inc. pp. 134-224. Lunden, J. M., Autio, T. J., Sjoberg, A. M., and Korkeala, H. J. 2003. Persistent and nonpersistent Listeria monocytogenes contamination in meat and poultry processing plants. J. Food Prot. 66: 2062-2069.

MacDonald, C. L., O’Leary, L. M., Cohen, M. L., Norris, P., Wells, G., Noll, E., Kobayashi, J. M., and Blake, P. A. 1998. Escherichia coli O157:H7, an emerging gastrointestinal pathogen. J. Am. Med. Assoc. 259: 3567-3570. McLaren, R. D., Prosser, C. G., Grieve, R. C. J., and Borissenko, M. 1994. The use of caprylic acid for the extraction of the immunoglobulin fraction from egg yolk of chicken immunized with ovine α-lactalbumin. J. Immunol. Methods 177: 175-184. Mandeles, S. 1960. Use of DEAE-cellulose in the separation of proteins from egg white and other biological materials. J. Chromatogr. 3: 256-264. Mann, D.M., Romm, E. and Migliorini, M. 1994. Delineation of the glycosaminoglycan- binding site in the human inflammatory response protein lactoferrin. J. Biol. Chem. 269: 23661-23667. Martin, D. L., MacDonald, K.L, White, K. E.. Soler, J. T., and Osterholm, M. T. 1990. The epidemiology and clinical aspects of the hemolytic uremic syndrome in Minnesota. J. Am. Med. Assoc. 323: 1161-1166. Masson, P. L., and Heremans, J. F. 1968. Metal-combining properties of human lactoferrin (red milk protein). I. The involvement of bicarbonate in the reaction. Eur. J. Biochem. 6: 579-584. Masson, P. L., Heremans, J. F., and Dive, C. 1966. An iron-binding protein common to many external secretions. Clinica Chimica Acta 14: 735-739. Matsudomi, N., Takasaki, M., and Kobayashi, K. 1991. Heat-induced aggregation of lysozyme with ovotransferrin. Agric. Biol. Chem. 155: 1651-1653. Mayes, F. J., and Takeballi, M. A. 1983. Microbial contamination of the hen's egg: A review. J. Food Protect. 46: 1092-1098. McClure, P. J., Beaumont, A. L., Sutherland, J. P., and Roberts, T. A. 1997. The effects on

growth of NaC1, pH, storage temperature and NaNO2. Int. J. Food Microbio. 34: 221- 232. Metz-Boutigue, M. H., Jolles, J., Mazurier, J., Schoentgen, F., Legrand, D., Spik, G., Montreuil, J., and Jolles, P. 1984. Human lactoferrin: amino acid sequence and structural comprisons with other transferrin. Eur. J. Biochem. 145: 659-676.

Mizutani, K., Muralidhara, B.K., Yamashita, H., Tabata, S., Mikami, B., and Hirose, M. 2001. Anion-mediated Fe3+ release mechanism in ovotransferrin C-lobe. J. Biol. Chem. 276 (38): 35940-35946. Miettinen, M.K., Palmu, L., Bjorkroth, K.J., and Korkeala, H. 2001. Prevalence of Listeria monocytogenes in broilers at the abattoir, processing plant, and retail level. J. Food Prot. 64: 994-999. Miyagawa, S., Kamata, R., Mastumoto, K., Okamura, R., and Maeda, H. 1994. Therapeutic intervention with chicken egg white ovomacroglobulin and a new quinolone on experimental Pseudomonas keratitis. Graefe’s Arch. Clin. Exper. Ophthalmol. 232: 488- 493. Miyagawa, S., Matsumoto, K., Kamata, R., Okamura, R., and Maeda, H. 1991a. Spreading of Serratia marcescens in experimental keratitis and growth suppression by chicken egg white ovomacroglobulin. Japanese J. Ophthal. 35: 402-410. Miyagawa, S., Nishno, N., Kamata, R., Okamura, R., and Maeda, H. 1991b. Effects of protease inhibitors on growth of Serratia marcescens and Pseudomonas aeruginoas. Microbiol. Patho. 11: 137-141. Molla, A., Matsumura, Y., Ymammoto, T., Okamura, R., and Maeda, H. 1987. Pathogenic capacity of proteases from Serratia marcescens and Pseudomonoas aeruginosa and their suppression by chicken egg white ovomacroglobulin. Infect. Immunol. 55: 2509-2517. Muralidhara, B. K., and Hirose, M. 2000. Anion-mediated iron release from transferrins. J. Biol. Chem. 275:12463-12469. Nagy, B., and Fekete, P. Z. 1999. Enterotoxigenic Escherichia coli (ETEC) in farm animals. Vet. Res. 30: 259-284. Naidu, A. S. 2002. Activated lactoferrin - A new approach to meat safety. Food Technol. 56(3): 40-45. Naidu, A. S., Tulpinski, J., Gustilo, K. Nimmagudda, R., and Morgan, J. B. 2003. Activated lactoferrin part 2: natural antimicrobial for food safety. Agro. Food Industry Hi-Tech. 14(3): 27-31.

Nakai, S. 2000. Molecular modifications of egg proteins for functional improvement. In : Egg Nutrition and Biotechnology Sim J. S., Nakai, S., and Guenter, W. (Eds). pp 205-217. CAB International, Oxon. Nakazato, K., Yamamura, T., and Satake, K., 1988. Different stability of N- and C-domain of diferric ovotransferrin in urea and application to the determination of iron distribution between the two domains. J. Biochem. 103: 823-828. Ntakarutimana, V., Demedts, P., Van Sande, M., and Scharpe, S. 1992. A simple and economical strategy for downstream processing of specific antibodies to human transferrin from egg yolk. J. Immunol. Methods 153: 133-140. O’ Farrelly, C., Branton, D., and Wanke, C. A. 1992. Oral ingestion of egg yolk immunoglobulin from hens immunized with an enterotoxigenic Escherichia coli strain prevents diarrhea in rabbits challenged with the same strain. Infect. Immunol. 60: 2593- 2597. Ohno, N., and Morrison, D. C. 1989a. Lipopolysaccharide interaction with lysozyme. Binding of lipopolysaccharide to lysozyme and inhibition of lysozyme enzymatic activity. J. Biolo. Chem. 264(8): 4434-4441. Ohno, N., and Morrison, D. C. 1989b. Effects of lipopolysaccharide chemotype structure on binding and inactivation of hen egg lysozyme. Europ. J. Biochem. 186(3): 621-627. Okamoto, I., Miautani, K., and Hirose, M. 2004. Iron-binding process in the amino- and carboxyl -terminal lobes of ovotransferrin; quantitative studies utilizing single Fe3+ binding mutants. Biochem. 43: 11118-11125. Osuga, D. T. and Feeney, R. E. 1977. Egg proteins. In: Food Proteins. Whitaker JR and Tannenbaum SR (Ed) AVI. Publishing Co., Westport, Connecticut pp. 209-248. Owusu-Asiedu, A., Baidoo, S. K., Nyachoti, C. M., and. Marquardt, R. R. 2002. Response of early-weaned pigs to spray-dried porcine or animal plasma-based diets supplemented with egg-yolk antibodies against enterotoxigenic Escherichia coli. J. Anim. Sci. 80: 2895-2903. Padhye, N. V., and Doyle, M. P. 1992. Escherichia coli O157:H7 epidemiology, pathogenesis, and methods for detection in food. J. Food Prot. 55: 555-565.

Pakdaman, R., and El hage Chahine, J. M. 1996. A mechanism for iron uptake by trasnferrin. Eur. J. Biochem. 236: 922-931. Pakdaman, R., and El Hage Chahine, J. M. 1997. Transferrin- the interaction of lactoferrin with hydrogen carbonate, Eur. J. Biochem. 249: 149-155. Pakdaman, R., Petitjean, M. and El Hage Chahine, J. M. 1998. Transferrins- a mechanism for iron uptake by lactoferrin. Eur. J. Biochem. 254: 144-153. Parish, M. E., and Higgins, D. P. 1989. Survival of Listeria monocytogenes in low pH model broth systems. J. Food Prot. 52: 144-147. Payne, K. D., Oliver, S. P., and Davidson, P. M. 1994. Comparison of EDTA and apolactoferrin with lysozyme on the growth of foodborne pathogenic and spoilage bacteria. J. Food Prot. 57: 62-65. Peaker, M., and Linzell, J. L. 1975. Citrate in milk: a harbinger of lactogenesis. Nature (London) 253: 464. Peterson, R. G., and Hartsell, B. 1955. The lysozyme spectrum of the Gram negative bacteria. J. Infect. Diseases 96: 75-81. Petran, R. L., and Zottola, E. A. 1989. A study of factors affecting growth and recovery of Listeria monocytogenes Scott A. J. Food Sci. 54: 458-460. Phelps, C. F., and Antonini, E. 1975. A study of kinetics of iron and copper binding to hen ovotransferrin. Biochem. J. 147: 385-391.

Phillips, D. C. 1966. The three-dimensional structure of an enzyme molecule. Sci. Am. 215: 78-90. Polson, A., Coetzer, T., Kruger, J., Maltzahn, E. V., and Merwe, K. J. V. 1985. Improvements in the isolation of IgY from the yolks of eggs laid by immunized hens. Immunol. Invest. 14: 323-327. Porath, J., Carlsson, J., Osson, I., and Belfrage, G. 1975. Metal chelate affinity chromatography, a new approach to protein fractionation. Nature 258: 598-599. Premaratne, R. J., Lin, W. J., and Johnson, E. A. 1991. Development of an improved chemically defined minimal medium for Listeria monocytogenes. Appl. Environ. Microbiol. 57: 3046-3048.

Raghubeer, E. V., Ke, J. S., Campbell, M. L., and Meyer, R. S. 1995. Fate of Escherichia coli O157:H7 and other coliforms in commercial mayonnaise and refrigerated salad dressing. J. Food Prot. 58:13-18. Ralovich, B. 1992. Data for the properties of Listeria strains (a review). Acta Microbiol. Hung. 39: 105-132. Rawas, A., Muirhead, H., and Williams, J. 1995. Structure of human diferric lactoferrin refined at 2.2 Å resolution. Acta Crystllogr. Sect. D. Biol. Crystallogr. 51: 629-646. Rawas, A., Muirhead, H., and Williams, J. 1996. Structure of diferric duct ovo-transferrin at 2.35 Ǻ resolution. Acta Crystallogr. D 52: 631-640. Rawas, A., Muirhead, H., and Williams, J. 1997. Structure of apo duck ovotransferrin: the structures of the N and C lobes are in the open form. Acta Crystallogr. Sect. D Biol. Crystallogr. 53: 464-468. Rayer, E. T., and Marth, E. H. 1999. Listeria, Listeriosis, and Food safety. 2nd ed. Marcel Dekker, Inc. New York. Rhodes, M. B., Azari, P. R., and Feeney, R. E. 1958. Analysis, fractionation, and purification of egg white proteins with celluose-cation exchanger. J. Biol. Chem. 230: 399-408. Rocourt, J., and Seeliger, H. P. R. 1985. Distribution des sepecies du genre Listeria. Zentral. Bakteriol. Mikrobiol. Hyg. A. 259: 317-330. Roderiquez, G., Oravecz, T., Yanagishita, M., Bou-Habib, D. C., Mostowski, H., and Norcross, M. A. 1995. Mediation of human immunodeficiency virus type 1 binding by interaction of cell surface heparin sulfate proteoglycans with the V3 region of envelope gp120-gp41. J. Virol. 69: 2233–2239. Rorvik, L. M., Aase, B., Alvestad, T., and Caugant, D. A. 2003. Molecular epidemiological survey of Listeria monocytogenes in broilers and poultry products. J. Appl. Microbiol. 94: 633-640. Rosenow, E. M., and Marth, E. H. 1986. Growth patterns of Listeria monocytogenes in skim, whole and chocolate milk and in whipping cream at 4, 13, 21, and 35o C. J. Food Prot. 49: 847-848. Ruan, G. P., Ma, L., He, X. W., Meng, M. J., Zhu, Y., Zhoum M. Q., Hu, Z. M., and Wang, X. N. 2005. Efficient production, purification, and application of egg yolk antibodies

against human HLA-A*0201 heavy chain and light chain (β-2m). Prot. Expre. Purif. 44: 45-51. Salton, M. R. J., and Pavlik, J. C. 1960. Studies of the bacterial cell wall. VI. Wall composition and sensitivity to lysozyme. Biochim. Biophys. Acta 39: 398-407. Samant, S. K., Singhal, R. S., Kulkarni, P. R., and Rege, D. V. 1993. Protein-polysaccharide interaction: a new approach in food formulations. Int. J. Food Sci. Technol. 28: 547-562. Sarker, S. A., and Hammarstrom, L. 2004. Passive immunotherapy with egg-yolk antibodies (IgY) in human enteric infections: future prospects. The 3rd international symposium on egg nutrition for health promotion. April. pp. 41. Schade, R., Burger, W., Schoneberg, T., Schniering, A., Schwarzkopf, C., Hlinak, A., and Kobilke, H. 1994. Avian egg yolk antibodies. The egg laying capacity of hens following immunization with antigens of different kind and origin and the efficiency of egg yolk antibodies in comparison to mammalian antibodies. Altex 11(2): 75-84. Schalabach, M. R., and Bates, G. W. 1975. The synergistic binding of anions and Fe3+ by transferrin. J. Biol. Chem. 250: 2182-2188. Schockman, G. D., and Höltje, J. V. 1994. In M, J. Guysen & R. Hakenbeck (eds.), New Comprehensive Biochemistry (pp. 131-167). Amsterdam: Elsevier Science. Schonberg, A., Teufel, P., and Weise, E. 1989. Serovars of Listeria monocytogenes and Listeria innocua from food. Acta Microbiol. Hung. 36: 249-253. Schwartz, B., Chiesielski, C. A., Broome, C. V., Gaventa, S., Brown, G. R., Gellin, B. G., Hightower, A. W., and Mascola, L. 1988. Association of sporadic listeriosis with consumption of uncooked hot dogs and undercooked chicken. Lancet 2: 779-782. Schwarzkopf, C., and Thiele, B. 1996. Effectivity of different methods for the extraction and purification of IgY. Altex 13(5): 35-39. Seelinger, H. P. R., and Jones, D. 1986. Genus Listeria. In: Sneath P.H.A., Mair N.S., Sharpe M.E., Holt, J.G. Eds. Bergey's Manual of Systemic Bacteriology. Baltimore: Williams and Wilkins. pp. 1235-1245. Semanchek, J. J., and Golden, D. A. 1997. Influence of growth temperature on inactivation and injury of Escherichia coli O157:H7 by heat, acid, and freezing. J. Food Prot. 61: 395- 401.

Shade, A. L., and Caroline, L. 1944. Raw hen egg white and the role of iron in growth inhibition of Shigella dysenteriae, Staphylococcus aureus, Escherichia coli and Saccharomyces cerevisiae. Science 100: 14-15. Shibusawa, Y., Iino, S., Shindo, H., and Ito, Y. 2001. Separation of chicken egg white proteins by high-speed countercurrent chromatography. J. Liquid Chromatogr. Related Technol. 24(13): 2007-2016. Shimizu, M., Fitzsimmons R. C., and Nakai, S. 1988. Anti-E. coli immunoglobulin Y isolated from egg yolk of immunized chickens as a potential food ingredient. J. Food Sci. 53: 1361-1366. Shimizu, M., Miwa, Y., Hashimoto, K., and Goto, A. 1993. Encapsulation of chicken egg yolk immunoglobulin G (IGY) by liposomes. Biosci.Biotech. Biochem. 57: 1445–1449. Shin J. H., Roe, I. H., and Kim, H. G. 2004, Production of anti-Helicobacter pylori urease- specific immunoglobulin in egg yolk using an antigenic epitope of H. pylori urease. J. Medical Microbiol. 53: 31-34. Shin, J. H., Yang, M., Nam, S. W., Kim, J. T., Myung, N. H., Bang, W.G., and Roe, I. H. 2002. Use of egg yolk-derived immunoglobulin as an alternative to antibiotics treatment for control of Helicobacter pylori infection. Clin. Diagn. Lab Immunol. 9: 1061-1066. Smith, J. L., McColgan, C., and Marmer, B. S. 1991. Influence of growth temperature on injury and death of Listeria monocytogenes Scott A during a mild heat treatment. J. Food Prot. 54: 166-169. Smith, D. J., King, W. F., and Godiska, R. 2001. Passive transfer of Immunoglobulin Y antibody to Streptococcus mutants glucan binding protein B can confer protection against experimental dental caries. Infect. Immunol. 69: 3135-3142. Sriram, V., and Ygeeswaran, G. 1999. Improved recovery of immunoglobulin fraction from egg yolk of chicken immunized with AsialoGM1. Russ. J. Immunol. 4(2): 131-140. Stadelman, W. J., and Cotterill, O. J. 1986. Egg Science and Technology. 3rd Ed. Westport, Cnn. Avi Pub. Co. Stalberg, J., and Larsson, A. 2001. Extraction of IgY from egg yolk using a novel aqueous two-phase system and comparison with other extraction methods. Ups J. Med. Sci. 106(2): 99-110.

Stewart, C. S., and Flint, H. J. 1999. Escherichia coli O157:H7 in farm animals. CABI Publishing, New York. pp. 169-193. Sugino, H., Nitoda, T., and Juneja, L. R. 1997. General chemical composition of hen eggs. In: Hen eggs - Their Basic and Applied Science (Yamamoto, T., Junja, L.R., Hatta, H., and Kim, M. eds.). pp 13-24. CRC Press, Inc. New York. Sugita-Konishi, Y., Sakanaka S, Sasaki, K., Juneja, L. R., Noda, T., and Amano, F. 2002. Inhibition of bacterial adhesion and Salmonella infection in BALB /c mice by sialyoligosaccharides and their derivatives from chicken egg yolk. J. Agric. Food Chem. 50: 3607-3613. Sunwoo, H. H., Lee, E. N., Mennien, K., Suresh, M. R., and Sim, J. S. 2002. Growth inhibitory effect of chicken egg yolk antibody (IgY) on Escherichia coli O151:H7. J. Food. Sci. 67: 1486-1494. Suzuki, Y. A., Shin, K., and Lonnerdal, B. 2002. Characterization of mammalian receptors for lactoferrin. Biochem. Cell Biol. 80: 75-80. Svendsen, L., Crowley, A., Ostergaard, L. H., Stodulski, G., and Hau, J. 1995. Development and comparison of purification strategies for chicken antibodies from egg yolk. Lab. Animal Sci. 45(1): 89-92. Tini, M., Jewell, U. R., Camenisch, G., Chilov, D., and Gassmann, M. 2002, Generation and application of chicken egg-yolk antibodies. Comp. Biochem. Physiol. Part A 131: 569- 574. Tranter, H. H., and Board, R. G. 1982. The inhibition of vegetative cell outgrowth and division from spores of bacillus cereus T by hen egg albumen. J. Appl. Bacteriol. 52: 67- 74. Tsai, Y., and Ingham, S. C. 1997. Survival of Escherichia coli O157:H7 and Salmonella spp. In Acidic condiments. J. Food Prot. 60: 751-755. USDA. 1994. USDA:APHIS:VS. “Escherichia coli O157:H7: Issues and Ramifications.” U.S. Dept. of Agr. and An. Plnt. Health Insp. Serv. Centers for Epidemiology and Animal Health, Fort Collins, CO. USDA-FSIS. 1998. Nationwide young turkey microbiological baseline data collection program. http://www. Fsisusda-gov/ops/baseline/yngturkpdf Accessed. May 2003.

USDA-FSIS. Listeriosis and Pregnancy: What is Your Risk?. http://www.fsis.usda.gov. Sep. 2001. Uyttendaele, M., De Troy, P., and Debevere, J. 1999. Incidence of Listeria monocytogenes in different types of meat products on the Belgian retail market. Int. J. Food Microbiol. 53: 75-80. Vachier, M. C., Piot, M., and Awade, A. C. 1995. Isolation of hen egg white lysozyme, ovotransferrin and ovalbumin, using a quaternary ammonium bound to a highly crosslinked agarose matrix. J. Chromatogr. B 664: 201-210. Vadehra, D. V., Baker, R.C., and Naylor, H. B.1972. Distribution of lysozyme activity in the exteriors of egg from Gallus gallus. Comparative Biochem. Physiol. 43: 503-508. Valenti, P., Antonin, G., Von Hunolstein, C., Visca, P., Orsi, N., and Antonini, E. 1983a. Studies of the animicrobial activity of ovotransferrin. Inter. J. Tiss. Reac. 1: 97-105. Valenti, P., Antonini, G., Rossi Fanelli, M. R., Orsi, N., and Antonini, E. 1982. Antibacterial activity of matrix-bound ovotransferrin. Antimicro. Agents Chemotherapy 21(5): 840- 841. Valenti, P., De Stasio, A., Mastromarino, P., Seganti, L., Sinibaldi, L., and Orsi, N. 1981. Influence of bicarbonate and citrate on the bacteriostatic action of ovotransferrin towards Staphylococci. FEMS Microbiol. Lett. 10: 77-79. Valenti, P., Stasio, A. D., Seganti, L., Mastromarino, P., Sinibaldi, L., and Orsi, N. 1980.

Capacity of Staphylococci to grow in the presence of ovotransferrin or CrCl3 as a character of potential pathogencity. J. Clinic. Micro. 11(5): 445-447. Valenti, P., Visca, P., Antonini, G., and Orsi, N. 1983b. Studies on the interaction between microorganisms and ovotransferrin. In Structure and Function of Iron Storage and Storage Proteins. Urishizaki, I., Aiesen, P., Listowsky, I. and Drysdale, J.W. (Ed). Elsevier, New York. Valenti, P., Visca, P., Antonini, G., and Orsi, N. 1985. Antifungal activity ovotransferrin towards genus Candida. Mycophathologia 89: 169-175. Valenti, P., Visca, P., Antonini, G., Orsi, N., and Antonini, E. 1987. The effect of saturation with Zn+2 and other metal ions on the antibacterial activity of ovotransferrin. Med. Microbiol. Immunol. 176: 123-130.

Verdoliva, A., Basile, G., and Fassina, G. 2000. Affinity purification of immunoglobulins from chicken egg yolk using a new synthetic ligand. J. Chromatogr. B. Biomed. Sci. Appl. Dec. 749(2): 233-42. Wang, C., and Shelef, L. A. 1991. Factors contribution to antilisterial effects of raw egg albumen. J. Food Sci. 56 (5): 1251-1254. Wang, C., and Shelef, L. A. 1992. Behavior of Listeria monocytogenes and the spoilage microflora in fresh cod fish treated with lysozyme and EDTA. Food Microbiol. 9: 207- 213. Wang, G., and Doyle, M. P. 1998. Survival of enterohemorrhagic Escherichia coli O157:H7 in water. J. Food Prot. 61: 662-667. Ward, P. P., Uribe-Luna, S., and Conneely, O. M. 2002. Lactoferrin and host defense. Biochem. Cell Biol. 80: 95-102. Warner, R. C. 1954. Egg proteins. In: The Proteins. Neurath, H., Bailey, K. (Ed). Academic Press, New York. Warner, R. C., and Weber, I. 1951. The preparation of crystalline conalbumin. J. Biol. Chem. 191: 173-180. Warner, R. C., and Weber, I. 1953. The metal combining properties of conalbumin. J. Am. Chem. Soc. 75: 5094-5101. Watanabe, K., Tsuge, Y., and Shimoyamada, M. 1998. Binding activities of pronase-treated fragments from egg white ovomucin with anti-ovomucin antibodies and newcastle disease virus. J. Agric. Food Chem. 46: 4501-4506. Weeranta, R. D., and Doyle, M. P. 1991. Detection and production of verotoxin 1 of Escherichia coli O157:H7 in food. Appl. Environ. Microbiol. 57: 2951-2955. Welshimer, H. J. 1963. Vitamin requirements of Listeria monocytogenes. J. Bacteriol. 85: 1156-1159. Williams, J. 1962a. Serum proteins and the livetins of hen’s egg yolk. Biochem. J. 83: 346- 355 Williams, J. 1962b. A comparison of conalbumin and transferrin in the domestic fowl. Biochem. J. 83: 355-364

Williams, J. 1975. Iron-binding fragments from the carboxyl-terminal region of hen ovotransferrin. Biochem. J. 149: 237-244. Williams, J., Elleman T.C., Kingston, I. B., Wilkins, A. G., and Kuhn, K. A. 1982. The primary structure of hen ovotransferrin. Eur. J. Biochem. 122: 297-303. Williams, J., Evans R. W., and Moreton, K. 1978. The iron-binding properties of hen ovotransferrin. Biochem. J. 173: 535-542. Williams, N. C., and Ingham, S. C. 1997. Change in heat resistance of Escherichia coli O157:H7 following heat shock. J. Food Prot. 60: 1128-1131. Williams, R. C., and Gibbons, R. J. 1972. Inhibition of bacterial antigen disposal. Sci. 177: 697- 699. Wilson, L. A., and Spitznagel, J. K. 1968. Molecular and structural damage to Escherichia coli produced by antibody, complement, and lysozyme systems. J. Bacter. 96: 1339-1348. Winn, S. E., and Ball, H. R. 1975. ß-N-acetylglucosaminidase activity of the albumen layers and membranes of the chicken’s egg. Poultry Sci. 54: 799-805. Wishnia, A., and Warner, R. C. 1961. The kinetics of denaturation of conalbumin. J. Am. Chem. Soc. 83: 2065-2071. Woodworth, R. C., Virkaitis, L. M., Woodbury, R. G., and Fava, R. A. 1975. In Pproteins of Ion Storage and Transport in Biochemistry and Medicine. Cichton, R. R. Ed. pp. 39-50. Amsterdam: North-Holland. pp 454. Wooley, J. A., and Landon, J. 1995. Comparison of antibody production to human interleukin-6 (IL-6) by sheep and chickens. J. Immunol. Methods 178: 253-265. Workman, E. F., Graham, G., and Bates, G. W. 1975. The effect of serum and experimental variables on the transferrin and reticulocyte interaction. Biochim. Biophys. Acta 399: 254-264. Wu, H. F., Monroe, D. M., and Church, F. C. 1995. Characterization of the glycosaminoglycan-binding region of lactoferrin. Arch. Biochem. Biophys. 317: 85-92. Wu, M., Chen, Q., and Wang, Y. X. 2003. Preparation of egg yolk immunoglobulin (IgY) against p53 protein expressed in E. coli. Di Yi Jun Yi Da Xue Xue Bao. 23(5): 491-493 WuDunn, D., and Spear, P. G. 1989. Initial interaction of herpes simplex virus with cells is binding to heparan sulfate. J. Virol. 63(1): 52-58.

Xu, J., Shimoyamada, M., and Watanabe, K. 1997. Gelation of egg white proteins as affected by combined heating and freezing. J. Food Sci. 62(5): 963-966. Yokoyama, H., Peralta, R.C., Horikoshi, T., Hiraoka, J., Ikemori, Y., Kuroki, M., and Kodama, Y. 1993. A two-step procedure for purification of hen egg yolk immunoglobulin G: utilization of hydroxypropylmethylcellulose phthalate and synthetic affinity ligand gel (Avid AL). Poult. Sci. 72: 275-281. Yolken, R. H., Willooughby, R., Wee, S. B., Miskuff, R., and Vonderfecht, S. 1987. Sialic acid glycoproteisn inhibit in vitro and in vivo replication of rotaviruses. J. Clinic. Invest. 79: 148-154. Zuccola, H. A. 1993. The Crystal Structure of Monoferric Human Serum Transferrin. Ph.D. thesis, Georgia Institute of Technology.

CHAPTER 3. PREPARATION OF EGG YOLK IMMUNOGLOBULIN (IgY) FROM CHICKEN EGGS USING AMMONIUM SULFATE PRECIPITATION AND ION EXCHANGE CHROMATOGRAPHY

A paper published in the Poultry Science *

K. Y. Ko1 and D. U. Ahn2

Abstract The objective of this study was developing an economical, simple, and large-scale separation method for egg yolk immunoglobulin from chicken eggs. Egg yolk diluted with 9 volumes of cold water was centrifuged after adjusting the pH to 5.0. The supernatant was added with 0.01% charcoal or 0.01% carrageenan and centrifuged at 2,800 x g for 30 min. The supernatant was filtered through a Whatman No. 1 filter paper and then the filtrate was concentrated to 20% original volume using ultrafiltration. The concentrated solution was further purified using either cation exchange chromatography or ammonium sulfate precipitation. For the cation exchange chromatography method, the concentrated sample was loaded onto a column equilibrated with 20 mM citrate-phosphate buffer at pH 4.8 and eluted with 200 mM citrate-phosphate buffer at pH 6.4. For the ammonium sulfate precipitation method, the concentrated sample was twice precipitated with 40% ammonium sulfate solution at pH 9.0. The recovery and purity of IgY were determined by ELISA and electrophoresis. The yield of IgY from the cation exchange chromatography method was 30- 40%, whereas that of the ammonium precipitation was 70 ~ 80%. The purity of IgY from ammonium sulfate method was higher than that of the cation exchange chromatography. The cation exchange chromatography could handle only a small amount of samples, whereas the ammonium sulfate precipitation could handle large volume of samples. This study suggests

*Reprinted with permission of Poultry Science, 2007, 86: 400-407 1Primary researcher and author 2Author of correspondence.

that ammonium sulfate precipitation was a more efficient and useful purification method than cation exchange chromatography for the large-scale preparation of IgY from egg yolk.

Introduction The production of antibodies from immunized chicken eggs is an excellent alternative to that from the serum of mammals (Svendsen et al., 1995). Chicken antibodies have many advantages to the traditional mammalian antibodies and have several important differences against mammalian IgG in regard to their functions. First, chicken immunoglobulin Y (IgY) can react with many epitopes of mammalian antigens due to phylogenic distance between birds and mammals, resulting in amplification of signals. Second, chicken antibodies do not react with rheumatoid factors, mammalian IgG, and bacterial Fc-receptors; they do not induce false positive results in immunoassays because they do not activate mammalian complement (Larsson et al., 1991, 1992; Davalos-Pantoja, 2000; Stalberg and Larsson, 2001), and do not bind protein A and G, which are commonly used for the isolation of IgG, due to differences in the Fc-regions (Akerstrom, 1985; Schwarzkopf and Thiele, 1996). Therefore, IgY can be applied in many medical fields such as xenotransplantation (Fryer et al., 1999), diagnostics (Erhard et al., 2000) and antibiotic-alternative therapies (Carlander et al., 1999). Furthermore, the amount of antibodies obtained from an egg is equivalent to that from 200 to 300 mL of mammalian blood and the costs for animal care per unit production of antibodies are much lower in chicken than in mammals (Larsson et al., 1993; Schade and Hlinak, 1996). However, the practical use of IgY in research and diagnostics is limited due to complex and time-consuming purification steps associated with the further purification of IgY (Akita and Nakai, 1992; Camenisch et al., 1999). The first step of IgY separation from egg yolk usually involves with the extraction of IgY from yolk. One of the major obstacles in isolating IgY from egg yolk is high concentrations of lipids and lipoproteins (Hasen et al., 1998; Verdoliva et al., 2000). Various strategies including the use of detergents such as sodium dodecylsulfate (Sriram and Ygeeswaran, 1999), carrageenan (Hatta and Kim, 1990), sodium alginate or xantham gum (Hatta et al., 1988); solvents such as acetone (Sriram and Ygeeswaran, 1999), chloroform (Ntakarutimana et al., 1992) or ethanol (Bade and Stegemann, 1984);

polyethylene glycol (Polson et al., 1985, Akita and Nakai, 1993) or dextran sulfate (Jensenius et al., 1981); aqueous two-phase system with phosphate and triton X-100 (Stalberg and Larsson, 2001); simple freeze and thaw cycling (Svendsen et al., 1995); and water dilution under acidic conditions (Akita and Nakai, 1993) have been used to remove lipids and lipoproteins from egg yolk extract. Most of these methods, however, have drawbacks such as low IgY recovery rates, complexity of procedures, or compatibility for human use. Among the methods, Akita and Nakai (1993) suggested that water dilution method under acidic conditions were suggested to be the most efficient and economical procedure for large-scale production of IgY from egg yolk. Upon removal of lipids and lipoproteins from egg yolk using acidified water, IgY can be precipitated by ammonium sulfate (Akita and Nakai, 1993), sodium sulfate (Wooley and Landon, 1995) or caprylic acid-ammonium sulfate (McLaren et al., 1994; Ruan et al., 2005). However, the water dilution (10x) method results in an extreme volume increase, which makes it difficult to use salt precipitation on a large- scale. Ultrafiltration is one of the best methods of reducing the volume of egg yolk extract, but the efficacy of ultrafiltration is greatly influenced by the presence of lipids or lipoproteins in the solution. Therefore, complete removal of lipids or lipoproteins from the water-extract of egg yolk is necessary. The changes of pH value in egg yolk solution influence the extent of interactions between polysaccharides and proteins, the precipitation of polysaccharide- lipoprotein complexes, and the recovery of immune-activity in IgY (Gurov et al., 1983; Samant et al., 1993). Chang et al. (2000) reported that addition of 0.1% of λ-carrageenan was effective in removing lipoproteins from the water-extract of egg yolk at pH 5.0. After precipitation of IgY by salts, column chromatography (Jensenius et al., 1981) is frequently used as a final step for IgY purification. However, column chromatography is expensive and impractical for the large-scale production of antibodies. Therefore, appropriate strategies for the large-scale production of antibodies with high purity and recovery are needed. The objective of this work was to develop an efficient and simple protocol for large- scale production of antibodies with high purity and recovery rates. In this study carrageenan or charcoal was added to the water-soluble fraction obtained by water dilution method to remove lipoproteins, which tend to clog membrane filter during ultrafiltration.

Materials and Methods

Preparation of egg yolk extract Fresh eggs (less than 2-day-old) were obtained from a local farm and yolk was separated from white using yolk separators. Egg yolk (4o C) was diluted with 9 volumes of cold (4o C) distilled water and homogenized for 1 min using a Waring blender at high speed. The pH of egg yolk solution was adjusted to pH 5.0 with 1 N-HCl. To determine the effect of NaCl on the extraction of IgY from egg yolk, various amounts of NaCl was added to the egg yolk before homogenization. To investigate the effect of storage temperature on the extraction of water-soluble proteins, homogenized egg yolk solutions were kept in room temperature (control), freezer, or refrigerator for 24 hrs before centrifugation at 2,800 x g for 40 min at 4oC. The turbidity and IgY yield of the supernatant were measured. After centrifugation, various concentrations of carrageen or charcoal was added to the supernatant collected and centrifuged again at 2,800 x g for 30 min at 4oC to remove residual lipoproteins in the supernatant. The supernatant was filtered through a Whitman No.1 filter paper and then concentrated to one-fifth of the original volume using a Pelican XL Biomax- 50 ultrafiltration membrane filter (cut-off size: 100 kD) installed to a Lab scale TM TFF System (Millipore, MA, USA). The concentrate was further purified for IgY either using a cation exchange chromatography or ammonium sulfate precipitation method. The turbidity and lipid content of the supernatant and filtrate were measured. The turbidity was determined by reading the absorbance of sample solutions using a Spectrophotometer (Cary 50 Bio, Varian, CA, USA) at 600 nm. Protein concentration was determined using the Bio-Rad protein assay method (Bio-Rad, Hercules, CA, USA) based on the Bradford method. Bovine Serum Albumin (1 mg protein/ml, Sigma, St. Louis, MO, USA) was used as a reference protein. The absorbance at 595 nm after 30 min of reaction with Bradford solution was measured using a spectrophotometer. Lipid content was measured using the Foch’s method (Folch et al., 1957). The concentrated sample solution by ultrafiltration was further purified using either cation exchange chromatography or ammonium sulfate precipitation method. For cation exchange chromatography, an aliquot of sample was loaded onto a column (6 mL of bed

volume) packed with pre-swollen carboxymethyl (CM) cellulose (Sigma-Aldrich St. Louis, MO, U.S.A) and equilibrated with 200 mM citrate-phosphate buffer, pH 5.0. The column was washed two times with 9 mL of 20 mM citrate- phosphate buffer, pH 5.0 and eluted with 200 mM citrate-phosphate buffer, pH 6.4. The elution profiles of samples were plotted and fractions that make a peak were pooled and analyzed for antibody activity and purity using enzyme-linked immunosorbent assay (ELISA) and sodium dodecylsulfate - polyacrylamide gel electrophoresis (SDS-PAGE), respectively. For ammonium sulfate precipitation method, the concentrated sample by ultrafiltration was first precipitated by 40% ammonium sulfate at 4o C. Then, the pellet was resuspended in 0.01M Tris-HCl (pH 8.0) to a volume equal to half of the supernatant. The sample was precipitated by 40% saturated ammonium sulfate again at 4o C, the pellet was dissolved in PBS, pH 7.4 and dialyzed against 10 mM phosphate buffer, pH 7.0 for 24 to 48 h to remove salt. The schematic diagram for the isolation of IgY from egg yolk is shown in Fig. 1. Antibody activity and purity were determined using ELISA and SDS-PAGE, respectively. All the sample preparation processes were replicated 4 times.

SDS-polyacrylamide gel electrophoresis SDS-PAGE was done under non-reducing conditions using Mini–PROTEAN II cell (Bio Rad, Richmond, CA) following instruction of the manufacturer. A 10% polyacrylamide separating gel (33% [vol/vol] acrylamide: N,N- bis-methylene acrylamide = 30% : 0.8% [wt/vol], 0.1% [wt/vol] SDS, 0.5% [vol/vol] TEMED, 0.05% [wt/vol] APS, and 0.38M Tris ·HCl, pH 8.8) was used for determination of egg yolk proteins. A 4% polyacrylamide gel (13.4% [vol/vol] acrylamide: N,N-bis-methelyene acrylamide = 30% : 0.8% [wt/vol], 0.1% [wt/vol] SDS, 0.1% [vol/vol] TEMED, 0.05% [wt/vol] APS, and 0.125M Tris ·HCl, pH 6.8) was used for the stacking gel and coomassie brilliant blue R-250 staining was used. The purity of various IgY preparations was estimated using 10% SDA-PAGE, and coomassie brilliant blue R-250 (BioRad) was used to visualize the protein bands. Broad-range SDS- PAGE molecular weight standards of 45 ~ 200 kDa (Bio-Rad) were used as markers. The purity of protein was estimated by densitometric measurements using the AlphaEaseFC software (FluorChem 8800, Alpha Innotech Corp.).

Enzyme-linked immunosorbent assay (ELISA) Immulon I micro titer 96-well plates (Dynatec Laboratories, McLean, VA) were used as the solid support. Polystyrene 96-well plates (Nalge Nunc Inc., Rochester, NY) were coated with 100 µL of IgY samples dissolved in a coating solution (0.05M carbonate buffer, pH 9.6) and incubated overnight at 4°C. After washing the wells three times with phosphate buffered saline-tween (PBS-TW: 10 mM phosphate, 0.15 M NaCl, pH 7.2, 0.05% Tween 20), 300 μL/well of blocking solution [(1% bovine serum albumin solution in PBS (10 mM phosphate, 0.15 M NaCl, pH 7.2) was added. After incubating for 1 h at room temperature, the plate was washed with PBS-TW. To each well of the plate, 100 μL primary anti-chicken IgG (1:10000 solution diluted with 1% bovine serum albumin (BSA)-conjugated alkaline phosphatase) was added and then incubated for l hr. After washing with phosphate buffered saline (PBS)-TW, each well was added with 50 μl of p-nitrophenyl phosphate solution as a substrate for color development and incubated for 30 min. The enzyme reaction was stopped by adding 50μL of 3N-NaOH and the color developed was read on an ELISA plate reader (Thermo max, Molecular Devices Corp., Sunnyvale, CA) with a 405 nm filter. For each plate, 3 kinds of controls were prepared: positive controls with reagent-grade chicken IgG (Sigma- Aldrich, St. Louis, MO, USA), non-specific antigen BSA as another control, and a negative control without antigen. All the procedures for ELISA were conducted at room temperature (about 25oC).

Statistical analysis Data were analyzed using SAS software (Release 6.11, SAS Institure Inc., Cary, NC) by the generalized linear model procedure. The Student-Newman-Keuls' multiple range test was used to compare differences among means. Mean values and standard deviation of mean (STD) were reported. Significance was defined at P < 0.05.

Results and Discussion pH adjustment on delipidation and IgY extraction Lipid content of supernatant from egg yolk solution without pH adjustment (pH 6.0) was 1.0 %, whereas that adjusted to pH 5.0 was 0.08%. Also, there was a significant

difference in the turbidity of supernatant between pH adjusted and non-adjusted egg yolk solutions (Table 1). The low turbidity of supernatant from the pH-adjusted egg yolk solution was attributed to better delipidation at pH 5.0 than at pH 6.0. However, there was no significatnt difference in the amount of obtained IgY between pH-adjustment to 5.0 and control. Similar results were obtained by Chang et al. (2000) who reported that lowering pH value of water-soluble fraction of egg yolk from pH 6.0 to 5.0 resulted in a significant decrease in lipid content (from 6.0 ~ 7.5% to 1.6 ~ 7.5%). Akita and Nakai (1992) reported that the adjustment of pH to 5.0 and 10 time dilution of yolk was crucial in removing lipids or lipoproteins from the water-soluble fraction of egg yolk. They reported that 93 ~ 96% of IgY was recovered after pH adjustment of water-soluble fraction to 5.0 ~5.2 and lowering pH value of diluted egg yolk solution resulted in a water-soluble fraction with low turbidity. Sugano and Watanabe (1961) reported that the solubility of lipoproteins at low ionic strengths decreased as the pH of water-soluble fraction was adjusted to pH 4.3. The pH of water-soluble fraction was the most important factor for IgY recovery because pH influenced the interactions between polysaccharides and proteins and the precipitation of polysaccharide-lipoprotein complexes (Gurov et al., 1983; Akita and Nakai, 1992; Samant et al., 1993). The protein contents of water-soluble fraction in both pH-adjusted and control were substantially underestimated because the Bradford’s dye-binding assay resulted in significantly lower readings for IgY than other methods (Sedmak and Grossberg 1977; Hansen et al., 1998). However, no significant difference in protein contents between pH- adjusted and pH non-adjusted treatments were observed. Akita and Nakai (1992) reported that the water-soluble fraction of egg yolk solution was free of lipids at pH 4.6 and 5.2 range, but lipid contents increased at pH below and above this range. The pH of diluted egg yolk was 5.8 ~6.3 and increased during storage. As it is difficult to separate lipids or lipoproteins at this pH, pH adjustment to 5.0 prior to centrifugation is important to eliminate lipids.

Temperature effect on extraction of water-soluble proteins Egg yolk antibodies were separated by centrifuging diluted egg yolk after pH adjustment to 5.0. The volume of supernatant obtained from frozen (24 hrs) egg yolk solution was lower than that of the control and refrigerated solution, but the turbidity of supernatants

was the lowest among the treatments (Table 2). The amounts of IgY in supernatant obtained after incubation at different temperature conditions was not significantly different. Also, there were no significant differences in total amount of IgY. Kim and Nakai (1996) and Yokoyama et al. (1993) reported that incubation of diluted egg yolk solution at freezing (-20 oC) or refrigeration temperature (4o C) was helpful in removing lipoproteins from water- soluble fraction. Larsson et al. (1993) reported that IgY was stable for 5 years of storage at cold temperature (4o C) without changes in antibody activities. However, according to our result that IgY yield obtained from freezing indicated to be lower (Table 2), some IgY seemed to be precipitated and lost during freezing, probably due to irreversible aggregation under these conditions.

Effect of NaCl in IgY precipitation As the concentration of NaCl in egg yolk solution increased, the turbidity of supernatants after centrifugation increased (Table 3). The IgY contents was not increased by the added NaCl. Kim and Nakai (1998) reported that addition of NaCl facilitated the separation of IgY from other proteins by polymerizing IgY molecules. Aggregation of IgY had no effect on Fab fragments of chicken antibody, but Fc fragments were precipitated by high salt concentration (Kubo and Benedict, 1969). Akita and Nakai (1992), however, reported that 0.16 M of NaCl resulted in an inhibitory effect in separating egg yolk granules from plasma proteins in diluted egg yolk solution, and the adverse effect of NaCl increased as the concentration of NaCl increased. In addition, if NaCl is added during extraction of IgY, dialysis step to remove the NaCl is necessary. Therefore, addition of salt to the diluted egg yolk solution would not be beneficial.

Carrageenan and charcoal effect on IgY isolation Ultrafiltration is an important tool for the large-scale preparation of IgY from egg yolk by concentrating supernatant of egg yolk solution that contains IgY. However, lipids or lipoproteins in water-soluble fraction (supernatant) should be minimized to prevent clogging of ultrafiltration membrane and improve filtration efficiency. Charcoal and carrageenan decreased the amount of lipids or lipoproteins in water-soluble fraction of egg yolk. The

optimal conditions for the highest yield of IgY from the water-soluble fraction of egg yolk was removing lipoproteins using 0.02% of λ-carrageenan at pH 4.0 (Figs. 2 and 3) or 0.01% charcoal at pH 4.0 (Figs. 4 and 5). Chang et al. (2000) reported that 0.1% λ-carrageenan at pH 5.0 was the optimal conditions for the removal of lipoproteins from 6-fold-diluted egg yolk solution. However, our result indicated that the yield of IgY at pH 5.0 was lower than that at pH 4.0. This should be caused by the differences in dilution of egg yolk. Even though some researchers suggest that addition of λ-carrageenan resulted in effective delipidation of egg yolk solution (Hatta and Kim, 1990; Kim and Nakai, 1996), some IgY should be precipitated by electrostatic forces that occur from interactions between proteins and polysaccharides (Imeson et al., 1978). Addition of 0.01% charcoal at pH 4.0 produced the highest IgY yield among the carrageenan and charcoal treatments (Fig. 6).

Cation exchange chromatography The purity of IgY solution obtained by cation exchange chromatography was not high and contained many egg yolk proteins other than IgY (Fig.7). MacCannel and Nakai (1989) used DEAE-Sephacel anion exchange chromatography to separate IgY from water-soluble fraction of egg yolk and found that the purity of IgY was 50 ~ 60%. The antibody preparation recovered by cation-exchange chromatography had 60 ~ 69% of purity (Fichtali et al., 1992; 1993). Therefore, we assumed that there are some limitations such as low purity and efficiency when ion exchange chromatography is used to purify IgY.

Ammonium sulfate precipitation The precipitation of IgY using 40% ammonium sulfate was influenced by pH of the solution, and the optimal conditions for the highest yield of IgY was at pH 9.0 (Fig. 8). The purity of IgY obtained after the second precipitation with the first precipitant using 40% ammonium sulfate at pH 9.0 appeared to be significantly high (Fig.9). This study demonstrates that 2-time precipitation of IgY using 40% ammonium sulfate at pH 9.0 produced IgY with much higher purity than cation exchange chromatography, and the efficiency of IgY purification was greater than the cation exchange chromatography methods.

Akita and Nakai (1992) reported that use of 60% ammonium sulfate produced a high IgY recovery and removed contaminating proteins, especially 38 kDa and 53 kDa proteins.

Recovery of IgY The recovery rates of IgY at each purification step are shown in Table 4. The protocol using ammonium sulfate (40%) and 0.01% charcoal was used for the recovery study. Addition of charcoal lowered the turbidity of supernatant by removing lipoprotein or lipids existing in the supernatant of egg yolk solution. This step caused no loss of IgY but speeded up the flow of ultrafiltration. About 20% of IgY was lost at ultrafilteration step, which removes proteins smaller than 50 kDa from the supernatant. Kim and Nakai (1998) also observed 15-20% loss in IgY by ultrafiltration. However, the percent loss of IgY can be reduced if larger volumes of supernatant are used for ultrafiltration. Kim and Nakai (1996) suggested that direct application of water-soluble fraction to ultrafiltration resulted in loss of IgY due to clogging of mucous lipoproteins on the membrane. The first precipitation of water-soluble fraction with 40% ammonium sulfate showed 75% IgY recovery (Table 3). After second precipitation of IgY using 40% ammonium sulfate, however, resulted in 69% yield of IgY (Table 3 and Fig. 9). Even though 6% of IgY was lost at the second precipitation step, the purity of IgY increased significantly. Akita and Nakai (1992) used 60% ammonium sulfate precipitation for purification of IgY, which resulted in 89% of IgY yield with 30% purity. Svendsen et al. (1995) used 25 ~ 45% of ammonium sulfate precipitation, which resulted in 58% yield of IgY with high purity. Most of the studies for purification or separation of specific proteins used combination methods adopting ion- exchange or gel filtration chromatography for further purification after precipitation with ammonium sulfate at first step. However the protocol developed in this study depends largely on ammonium sulfate precipitation at pH 9.0 after charcoal addition to water-soluble fraction to remove extra lipoproteins. Ammonium sulfate precipitation produced IgY with higher purity and yield than HPLC (Stec et al., 2004) or ion exchange chromatography, and was suitable for a large scale IgY preparation from egg yolk particularly with minimal steps.

Conclusion Egg yolk cooled was diluted to 10 times and adjusted to pH 5.0, and then centrifuged at 4,000 rpm for 40 min. The supernatant was filtered with filter paper, or was added with 0.01% activated charcoal or 0.02% carrageenan and then filtered. 0.01% charcoal treatment to water soluble egg yolk solution facilitates to reduce lipoprotein or lipid and then the solution became to be much clearer solution than that of carageenan. The water soluble fraction of egg yolk solution obtained after addition of charcoal following infiltration was 5 times concentrated with interfering proteins during ultrafiltration processing. The concentrated water soluble fraction was purified either using an action ion exchange column or precipitation with 40% ammonium sulfate solution at pH 9.0. The recovery of IgY from action exchange chromatography was 30 ~ 40%, whereas that of ammonium sulfate precipitation was around 70 ~ 80%. The purity of IgY from ammonium sulfate precipitation method was higher than that from anion exchange chromatography. As ammonium sulfate precipitation method could handle larger volume at once time than ion exchange chromatography, the new procedure developed at this study using precipitation of IgY using 40% ammonium sulfate and 0.01% charcoal is an economical and practical method for large- scale purification of IgY from egg yolk.

Reference Akerstrom, B., Bodin, T., Reis, K., and Bjorck, L. 1985. Protein: a powerful tool for binding and detection of monoclonal and polyclonal antibodies. J. Immunol. 135(4): 2589-2592. Akita, E. M. and Nakai, S. 1992. Immunoglobulins from egg yolk: isolation and purification. Food Sci. 57: 629-634. Akita, E. M., and Nakai, S. 1993 Comparison of four purification methods for the production of Immunoglobulins from egg laid by hens immunized with an enterotoxigenic E coli strain. J. Immunol. Methods 160: 207-214. Bade, H., and Stegemann, H. 1984. Rapid method of extraction of antibodies from hen egg yolk, J. Immunol. Methods 72: 421-426. Camenisch, G.., Tini, M., Chilow, D., Kvietikova, I., Srinivas, V., Caro, J., Spielmann, P., Wenger, R. H., and Gassmann, M. 1999. General applicability of chicken egg yolk

antibodies: the performance of IgY immunoglobulins raised against the hypoxia- inducible factor 1. FASEB 13: 81-88. Carlander, D., Stalberg, J., and Larsson, A. 1999. Chicken antibodies: a clinical chemistry perspective. Ups. J. Med Sci. 104(3): 179-89 Chang, H. M., Lu, T. C., Chen, C. C., Tu, Y. Y., and Hwang, J. Y. 2000. Isolation of immunoglobulin from egg yolk by anionic polysaccharides. J. Agric. Food Chem. 48: 995-999. Davalos-Pantoja, L., Ortega-Vinuesa, J. L., Bastos-Gonzalez, D., and Hidalgo-Alvarez, R. 2000, A Comparative study between the adsorption of IgY and IgG on latex particles, J. Biomater. Sci-Polym. E 11(6): 657-673. Erhard, M. H., Schmidt, P., Zinsmeister, P., Hoffman, A., Munster, U., Kaspers, B., Wiesmuller, K. H, and Bessler, W. G.. 2000. Adjuvant effects of various lipopeptides and interferon-gamma on the humoral immune response of chickens. Poult. Sci. 79(9): 1264- 1270 Fichtali, J., Charter, E. A., Lo, K. V., and Nakai, S. 1992. Separation of egg yolk immunoglobulin using an automated liquid chromatography system. Biotechnol. Bioeng. 40: 1388-1394. Fichtali, J., Charter, E. A., Lo, K. V., and Nakai, S. 1993. Purification of antibodies from industrially separated egg yolk. J. Food. Sci. 58: 1282-1285, 1290. Folch, J., Less, M., and Slaone-Stanley, G.. M. 1957. A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226: 497-509. Fryer, J., Firca J., Leventhal, J., Boondie, B., Malcolm, A., Lyancic, D., Gandhi, R., Shah, A., Pao, W., Abecassis, M., Kaufman, D., Stuart, F., and Anderson, B. 1999. IgY antiporcine endothelial cell antibodies effectively block human antiporcine xenoantibody binding. Xenotransplantation 6: 98-109. Gurov, A. N., Mukhin, M. A., Larichev, N. A., Lozinskaya, N. V., and Tolstoguzoc, V. S. 1983. Emulsifying properties of proteins and polysaccharides. I. Methods of determination of emulsifying capacity and emulsion stability. Colloids Surf. 6: 35-42.

Hansen, P., Scoble, J. A., Hanson, B., and Hoogenraad, N. J. 1998. Isolation and purification of immunoglobulins from chicken eggs using thiophilic interaction chromatography. J. Immunol. Methods 215: 1-7. Hatta, H., and Kim, M. 1990. A novel isolation method for hen egg yolk antibody, “IgY”. Agric. Biol. Chem. 54(10): 2531-2535. Hatta, H., Sim, J. S., and Nakai, S. 1988 Separation of phospholipids from egg yolk and recovery of water-soluble proteins. J. Food Sci. 53: 425-427. 431. Imeson, A. P., Watson, R. R., Mitchell, J. R., and Ledward, D. A. 1978. Protein recovery from blood plasma by precipitation with polyuronates. J. Food Technol. 13: 329-338. Jensenius, J. C., Andersen, I., Hau, J., Crone, M., and Koch, C. 1981. Eggs: conveniently packed antibodies. Methods for purification of yolk IgG. J. Immunol. Methods 46: 63- 68 Kim, H., and Nakai, S. 1996. Immunoglobulin separation from egg yolk: A serial filtration system. J. Food Sci. 61(3): 510-514. Kim, H., and Nakai, S. 1998. Simple separation of immunoglobulin from egg yolk by ultrafiltration. J. Food Sci. 63(3): 485-490. Kubo, R. T., and Benedict, A. A. 1969. Comparison of various avian and mammalian IgG immunoglobulins for sal-induced aggregation. J. Immunol. 103: 1022-1028. Larsson, A., Balow, R., Lindahl, T. L, and Forsberg, P. 1993. Chicken antibodies: Taking advantage of evolution –A review. Poultry Sci. 72: 1807-1812. Larsson, A., Karlsson-Parra, A., and Sjoquist, J. 1991. Use of chicken antibodies in enzyme immunoassays to avoid interference by rheumatoid factors. Clin. Chem. 37(3): 411-414. Larsson, A., Wejaker, P., Forsberg, P., and Lindahl, T. 1992. Chicken antibodies: a tool to avoid interference by complement activation in ELISA. J. Immunol. Methods 156: 79-83. MacCannel, A. A. and Nakai, S. 1989. Isolation of egg yolk immunoglobulins into subpopulations using DEAE-ion exchange chromatography, Can. Inst. Food Sci. Technol. J. 23: 42-46 McLaren, R. D., Prosser, C. G., Grieve, C. J., and Borissenko, M. 1994. The use of caprylic acid for the extraction of the immunoglobulin fraction from egg yolk of chicken immunized with ovine α-lactalbumin. J. Immunol. Methods. 177: 175-184.

Ntakarutimana, V., Demedts, P., Van Sande, M., and Scharpe, S. 1992. A simple and economical strategy for downstream processing of specific antibodies to human transferrin from egg yolk. J. Immunol. Methods. 153: 133-140. Polson, A., Coetzer, T., Kruger, J., Maltzahn, E., and van-der von Merwe, K. J. 1985. Improvements in the isolation of IgY from the yolks of eggs laid by immunized hens. Immunol. Invest. 14: 323-327. Ruan, G. P., Ma, L., He, X. W., Meng, M. J., Zhu, Y., Zhou, M. Q., Hu, Z. M., and Wang, X. N. 2005. Efficient production, purification, and application of egg yolk antibodies against human HLA- A*0201 heavy chain and light chain (β2m). Protein Expression Purification 44(1): 45 - 51 Samant, S. K., Singhal, R. S., Kulkarni, P. R., and Rege, D. V. 1993. Protein-polysaccharide interaction: a new approach in food formulations. Int. J. Food Sci. Technol. 28: 547-562. Schade, R., and Hlinak, A. 1996. Egg yolk antibodies, state of the art and future prospects. Altex 13(5): 5-9. Schwarzkopf, C., and Thiele, B. 1996. Effectivity of Different Methods for the Extraction and Purification of IgY. Altex 13(5): 35-39. Sedmak, J. J., and Grossberg, S. E. 1977. A rapid, sensitive and versatile assay for protein using Coomassie Brilliant Blue G250. Anal. Biochem. 79: 544-552 Sriram, V. and Ygeeswaran, G. 1999. Improved recovery of immunoglobulin Fraction from Egg yolk of chicken Immunized with AsialoGM1. Russ. J. Immunol. 4(2): 131-140. Stalberg, J., and Larsson, A. 2001 Extraction of IgY from egg yolk using a novel aqueous two-phase system and comparison with other extraction methods. Ups. J. Med. Sci. 106(2): 99-110. Stec, J., Bicka, L., and Kuzmak, J. 2004 Isolation and purification of polyclonal IgG antibodies from bovine serum by high performance liquid chromatography. Bull. Vet. Ins. Pulawy 48: 321-327. Sugano, H., and Watanabe, I. 1961. Isolation and some properties of native lipoproteins from egg yolk. J. Biochem. 50: 473-480

Svendsen, L., Croeley, A., Ostergaard, L. H., Stodulski, G., and Hau, J. 1995. Development and comparison of purification strategies for chicken antibodies from egg yolk. Lab. Animal Sci. 45: 89-92. Verdoliva, A., Basile, G., and Fascina, G. 2000. Affinity purification of immunoglobulins from chicken egg yolk using a new synthetic ligand. J. Chromatogr. B 749(2): 233-242. Wooley, J. A., and Landon, J. 1995. Comparison of antibody production to human interleukin-6 (IL-6) by sheep and chickens. J. Immunol. Methods. 178: 253-265. Yokoyama, H. 1993. Detection of passage and absorption of chicken egg yolk immunoglobulins in the gastrointestinal tract of pigs by use of enzyme-linked immunosorbent assay and fluorescent antibody testing. Am. J. Vet. Res. 54: 867-872.

Table 1. Changes of diluted egg yolk solution by pH adjustment to 5.0

Lipid (%) Turbidity Protein (mg/ml) IgY (mg/ml)

Control (pH 6.0) 1.00 ± 0.12 a 2.97 ± 0.15 a 0.41 ± 0.07a 0.82 ± 0.25a pH adjustment to 5.0 0.08 ± 0.02 b 0.33 ± 0.01b 0.51 ± 0.05 a 1.09 ± 0.34a

a, bMeans within a column with no common superscript differ (P < 0.05). n = 4

Table 2. Turbidity and the yields of IgY obtained from water soluble fraction after centrifugation

Yield (%)* Turbidity (600nm) IgY (mg/ml) Total IgY** (yield x IgY)

Control 77.03a 0.065a 1.11a 86.73a Refrigerator 78.23a 0.040b 1.12a 88.78a Freezing 72.03b 0.014c 1.24a 90.05a

a-cMeans within a column with no common superscript differ (P<0.05). *The volume of supernatant gained after centrifugation. n = 4. **IgY concentration x the volume of supernatant gained after centrifugation

Table 3. Changes of turbidity, protein and IgY content in adding NaCl to egg yolk water soluble fraction

Salt (%) Turbidity IgY (mg/ml)

0 0.07d ± 0.005 2.02a ± 0.18 0.1 0.06d ± 0.01 2.20a ±0.36 0.2 0.09cd ± 0.03 2.04a ± 0.20 0.3 0.14c ± 0.01 2.70a ± 0.35 0.4 0.21b ± 0.01 2.57a ± 0.11 0.5 0.51a ± 0.06 2.70a ± 0.32 a-dmeans within a column with no common superscript differ (P < 0.05). n = 4.

Table 4. The yields of IgY obtained through each processing

Purification step IgY (mg/ml) Vol. (ml) Conc. x Vol. Yield (%)

Water soluble fraction 0.33 ± 0.03 3970 ± 32.5 1310 ± 101.6 100 Charcoal and diafiltration 0.38 ± 0.01 3590 ± 28.3 1353 ± 27.4 103.5 Ultrafiltration 1.79 ± 0.10 590 ± 21.2 1056 ± 20.4 80.6 Ammonium sulfate (1st) 7.95 ± 0.06 122 ± 5.6a 970 ± 14.8 74.1 Ammonium sulfate (2nd) 15.9 ± 0.07 57 ± 3.8b 906 ± 5.5 69.2

aThe volume obtained after addition of 100ml of distilled water to the precipitate gained by first precipitation of 40% ammonium sulfate following dialysis. bThe volume gained from addition of 45ml of distilled water to the precipitate after second precipitation following dialysis. n = 4

Fig 1. Schematic diagram for the isolation of IgY from egg yolk

Egg yolk

10 times dilution with distilled water

pH adjustment to 5.0 with 1N-HCl

Centrifugation (2800 x g, 40min, 4oC) addition of 0.01% Charcoal pH adjustment to 4.0 and centrifugation

Filtration through a Whatman No. 1

Ultrafiltration

Cation exchange chromatography 40%Ammonium sulfate precipitation at pH 9.0

Dialysis with 20mM citrate phosphate buffer at pH 4.8

Fig. 2. IgY content in water soluble fraction by addition of carrageenan following centrifugation. Bars with different letters indicate significantly different values (p < 0.05, S.E = 0.17, n = 4).

a 1.2 b b b 1.0 b 0.8 0.6 0.4 IgY (m g/m l) 0.2 0.0 0 0.02 0.05 0.1 0.2

Conc. of carnageenan (%)

Fig. 3. Changes of IgY content after adding 0.02% carrageenan to water soluble fraction of egg yolk in different pH conditions. Bars with different letters indicate significantly different values (p < 0.05, S.E = 0.35, n = 4).

a 1 0.9 0.8 b 0.7 0.6 0.5 c 0.4 cd

IgY (mg/ml) 0.3 d d 0.2 0.1 0 456789 pH

Fig. 4. IgY content on each pH conditions in the addition of 0.01% charcoal addition to water soluble egg yolk protein solution. Bars with different letters indicate significantly different values (p < 0.05, S.E = 0.48, n = 4).

3.0 a

2.5

2.0 b 1.5 b b b b

IgY (mg/ml) IgY 1.0

0.5

0.0 456789 pH

Fig. 5. Comparison of IgY yields on charcoal amount added to water soluble egg yolk protein solution. Bars with different letters indicate significantly different values (p < 0.05, S.E = 0.16, n = 4).

2 1.8 a a 1.6 a a 1.4 1.2 1 0.8

IgY (mg/ml) 0.6 0.4 0.2 0 0.005 0.01 0.05 0.1

Charcoal (%)

Fig. 6. Comparison of IgY yields on the addition of different concentration of carrnageenan and 0.01% charcoal (Char*) to water soluble egg yolk protein solution. Bars with different letters indicate significantly different values (p < 0.05, S.E = 0.02, n = 4).

a 0.4 ab ab ab b b b 0.3

0.2

IgY (mg/ml) 0.1

0.0 0 0.010.020.050.10.2Char*

Carrageenan (%)

Fig. 7. SDS-PAGE patterns of egg yolk solutions obtained by cation ion exchange chromatography. 8ml of concentrated water soluble egg yolk solution by ultrafiltration were loaded on CM cellulose chromatography and the column was washed 20mM citrate phosphate buffer with pH 4.8, IgY was eluted by 200 mM citrate phosphate buffer with pH 6.4. Lane 1: marker, 2: after loading, 3 ~8: washing, 9: elution

1 2 3 4 5 6 7 8 9

IgY

Fig. 8. Changes of IgY yields on each pH condition in precipitation of 40% ammonium sulfate with concentrated water soluble egg yolk solution by ultrafiltration

6.0

b 5.0

b b 4.0

3.0 a

IgY (mg/ml) IgY a a 2.0

1.0

0.0 456789 pH

Fig. 9. SDS-PAGE patterns of sample fractions obtained from each steps including chicken centrifugation, ultrafiltration and 40% Ammonium sulfate precipitaion. Lane 1: Marker, 2: IgG, 4: after centrifugation of WSF, 5: addition of 0.01% charcoal following diafilteration, 6: after ultrafiltration, 7: after first precipitation by 40% ammonium sulfate at pH 9.0, 8: 2nd precipitation with ammounium sulfate

1 2 3 4 5 6 7 8

IgY

CHAPTER 4. AN ECONOMIC AND SIMPLE PURIFICATION PROCEDURE FOR THE LARGE-SCALE PRODUCTION OF OVOTRANSFERRIN FROM EGG WHITE

A paper which will be submitted to Poultry Science

K. Y. Ko1 and D. U. Ahn2

Abstract The objective of this study was to develop a simple and economical protocol for separating ovotransferrin from egg white. Egg white was separated from whole egg and diluted with the same volume of distilled water. To prevent denaturation during separation process, ovotransferrin in 2x-diluted egg white was converted to holo-form by adding 20 mM . FeCl3 6 H2O solution (0.25-3 ml/100 ml). The pH of egg white solution was adjusted to pH

7.0, 8.0 or 9.0, and 50 mM-NaHCO3 and 0.15M-NaCl were added to facilitate iron binding to ovotransferrin. The iron-bound ovotransferrin was separated from other egg white using different concentrations of ethanol (30-50%). Ethanol at 43% (final concentration) and pH at 9.0 was the best condition for separating iron-bound ovotransferrin from 2x-diluted egg white solution. Almost all egg white proteins including ovoalbumin were precipitated easily at 43% ethanol, but most of the iron-bound ovotransferrin remained in the supernatant. Holo- ovotransferrin in the 43% ethanol solution started to precipitate as the concentration of ethanol increased and the optimal conditions for precipitating ovotransferrin occurred when the ethanol concentration reached to 59% (final). Final holo-form ovotransferrin solution was prepared by dissolving the precipitate with distilled water. AG1-X2 ion exchange resin, at 2x theoretical content of ovotransferrin, was used to remove iron bound to ovotransferrin after pH adjustment to 4.7 using 200 mM-citric acids. The apo-ovotransferrin separated using the protocol was 80% in purity and 94% in yield. The protocol developed is simple, economical and appropriate for a large-scale production of ovotransferrin from egg white. Also, the

1 Primary researcher and author 2 Author of correspondence.

isolated ovotransferrin can be applied to human foods because the only solvent used in this

process is ethanol. Furthermore, the AG1-X2 ion exchange resin and ethanol used in this process can be regenerated and recovered.

Introduction Ovotransferrin is a major avian egg white protein that constitutes 12% of total egg white protein. It is a monomeric glycoprotein consisting of 686 amino acids with 15 disulfide bridges and has molecular weight of 76 kDa and pI of 6.1 (Abola et al., 1982; Stadelman and Cotterill, 1986). Ovotransferrin is composed of an N- and a C-terminal domain and one atom of transition metals such as Fe(III), Cu(III) and Al(III) can bind tightly within the interdomain cleft of each domain. A metal-free ovotransferrin (apo-form) is colorless and easily destroyed by physical and chemical treatments, whereas iron-bound ovotransferrin (holo-form) shows a salmon- pink color and is stable to proteolytic hydrolysis and thermal denaturation (Azeri and Feeney, 1958). Because most chemical reagents and denaturing conditions decrease the affinity of ovotransferrin for iron, particular spatial configuration of native ovotransferrin is needed for the formation of colored iron-bound ovotransferrin (Fraenkel-Conrat, 1950). The binding of 2- - 3+ iron to ovotransferrin requires 1 molecule of CO2 as CO3 or HCO3 per atom of Fe (Warner and Weber, 1953) to overcome the effect of citrate, which can bind iron (Masson and Heremams, 1968). Reiter et al. (1975), however, reported that bicarbonate rather than pH is a critical factor for bacteriostatic capability of lactoferrin. To release Fe3+ from ferric transferrin, a simple anion, such as pyrophosphate, sulfate, and chloride is required in vitro. The anion-induced Fe3+ release are closely related to the opening of a domain in either lobe (Baldwin et al., 1981; Cheuk et al., 1987; Bailey et al., 1997). At acidic pH, the N-lobe of ovotransferrin displays lower iron-binding capacity and 3+ faster release of Fe than the C-lobe. In the absence of NaHCO3, pH affects the efficiency of citrate-mediated release of Fe3+ from lactoferrin. Citrate-mediated Fe3+ release is more efficient at pH 6.8 than at pH 7.4 (Griffiths and Humphreys, 1977). Iron could be removed from iron-transport proteins at low pH (pH 4.5) in the presence of a large excess of citrate (Guo et al., 2003). Cunningham and Lineweaver (1965) reported that when holo-

ovotransferrin solution was heated at pH 9, the protein was not altered significantly. However, ovotransferrin was denatured when the protein was exposed to a pH below 4.2, but the influence of pH was reversible under certain experimental conditions (Phelps and Cann, 1956). Also, ovotransferrin can be separated from the egg white by aqueous and ethanol fractionation procedures (Bain and Deutsch, 1948; Warner and Weber, 1951), fractional precipitation with ammonium sulfate, or coagulating ovoalbumin (Warner, 1954; Azari and Baugh, 1967). However, the drawback of these techniques such as ammonium sulfate precipitation at acidic conditions and ethanol precipitation for purification of ovotransferrin denature ovotransferrin and the purity of products is relatively low (Vachier et al., 1995). In order to overcome these drawbacks from ethanol or ammonium sulfate precipitation, cation exchange chromatography such as CM-Cellulose (Rhodes et al., 1958; Azari and Baugh, 1967), DEAE-cellulose anion exchange chromatography (Mandeles, 1960), and Q-Sepharose Fast Flow column (Vachier et al., 1995) were employed for purification of ovotransferrin from egg albumin. Guerin and Brule (1992) used a cation exchange chromatography on DUOLITE C464 and C476 from egg white devoid of lysozyme for industrial-scale production of ovotransferrin. Al-Mashikhi and Nakai (1987) used a single step chromatographic method using an immobilized metal affinity chromatography (a copper- loaded Sepharose 6B column) to separate ovotransferrin from undiluted, blended egg white. Chung et al. (1991) used a bi-functional dye-ligand chromatography using DEAE Affi-Gel Blue, as the first step to fractionate egg white proteins. The fraction containing ovotransferrin obtained from DEAE Affi-Gel Blue, then was further purified by Fast Protein liquid chromatography. A two-step chromatographic procedure involving gel permeation on a Superose-6 Prep Grade column and anion-exchange chromatography on a Q-Superose Fast Flow was also employed (Awade et al., 1994). Recently, Ahlborn et al. (2006) separated ovotransferrin from egg white using Econo- Pac® High S-Cation and High Q-Anion Exchange Cartridges. Even though ion-exchange chromatography or immobilized metal affinity chromatography have been developed on a laboratory scale, the application of these techniques on pilot-scale is difficult because these methods are labor-intensive and very expensive (Awade, 1996).

The objective of this study was to develop a simple and rapid procedure for an economical, large-scale production of ovotransferrin from egg white.

Materials and Methods

Materials Chicken eggs were purchased from a local store. Egg white was separated from egg and diluted with the same volume of distilled water. Two-times diluted egg white solution was blended for 2 minutes using an electric blender (speed: level 6, Kichen Aid, St. Joseph. MI. USA) and used as a starting material for ovotransferrin separation.

Determination of FeCl3 concentration for iron saturation of egg white solution Because iron-saturated, holo-ovotransferrin is more stable to chemicals such as ethanol than the apo-form, the iron-free ovotransferrin in egg white solution was converted to iron- bound form using FeCl3 solution. The pH of 2x-diluted egg white solution was adjusted to

pH 7.0, 8.0, or 9.0 first, and then NaHCO3 and NaCl were added to make their concentrations in the egg white solution 50 mM and 0.15 M, respectively to help iron binding. . In order to determine the optimal concentration of FeCl3 6H2O for iron saturation, 0.25, 0.5,

1.0, 2.0 or 3.0 ml of 20 mM-FeCl3.6 H2O solution was added to 2x-diluted egg white solution. Sufficient mixing was conducted for 1 hr at room temperature for complete iron saturation of egg white solution. Cold ethanol (43%, final concentration) was added to the iron-saturated egg white solution to extract iron-bound ovotransferrin and then centrifugation was performed under 3,220 x g for 20 min. After filtering the supernatant including iron- bound ovotransferrin with Wathman filter papter No. 1, the iron-saturated ovotransferrin was precipitated by addition of ethanol following centrifugation. The precipitate (holo- ovotransferrin) was dissolved with 10 volumes of distilled water. The degree of iron saturation of egg white solution was estimated by measuring the optical density of the prepared solution at 468 nm since iron-bound ovotransferrin forms salmon-pink color (Azari and Feeney, 1958). That is, when there was no change of absorbance value, the solution was assumed to reach complete iron saturation.

Ethanol concentration for extraction and precipitation of Fe-ovotransferrin The egg white solution was stirred for 30 min at room temperature and then varying amounts of 100% cold ethanol (20%, 33%, 43%, and 50%, final concentrations) was added to isolate iron-bound ovotransferrin from iron-saturated egg white solution by denaturating and precipitating other egg white proteins such as ovoalbumin in the ethanol treatment. Holo- ovotransferrin was separated from the iron satured egg white solution by centrifugation at 3,220 x g for 20 min. The supernatants collected were filtered through Whatman No.1 filter paper to remove floating materials. After filtering, cold ethanol (100%), which correspons to be 53%, 56%, 59%, or 62%, was slowly added to aliquots of supernatant to precipitate iron- bound ovotransferrin. The optimal ethanol concentration for extraction and precipitation of iron-bound ovotransferrin was estimated using SDS-PAGE.

SDS-polyacrylamide gel electrophoresis A 10% polyacrylamide separating gel (33% [vol/vol] acrylamide:N,N- bis-methylene acrylamide = 30% : 0.8% [wt/vol], 0.1% [wt/vol] SDS, 0.5% [vol/vol] TEMED, 0.05% [wt/vol] APS, and 0.38M Tris ·HCl, pH 8.8) was used for determination of egg white proteins under non-reducing conditions using Mini–PROTEAN II cell (Bio Rad, Richmond, CA). A 4% polyacrylamide gel (13.4% [vol/vol] Acrylamide: N,N-bis-methelyene acrylamide = 30% : 0.8% [wt/vol], 0.1% [wt/vol] SDS, 0.1% [vol/vol] TEMED, 0.05% [wt/vol] APS, and 0.125M Tris ·HCl, pH 6.8) was used for the stacking gel and coomassie brilliant blue R-250 staining was used. Broad-range SDS-PAGE molecular weight standards of 45 ~ 200 kDa (Bio-Rad, Richmond, CA) were used as a marker. The purity of protein was estimated by densitometric measurements using the AlphaEaseFC software (FluorChem 8800, Alpha Innotech Corp.).

Iron binding capacity The quantity of ovotransferrin was determined using iron binding capacity, which was measured by the modified method of Williams (1974). The calibration curve was obtained using a commercial ovotransferrin (sigma, Sigma-Aldrich, Inc., St. Louis, USA). 50 mM-

NaHCO3 and 0.15 M-NaCl solution was added to the final apo-ovotransferrin solution to

support its iron binding capability and then its pH was adjusted to 8.0. After pH adjustment, 0.4 ml of 10 mM Fe-nitriloacetate in 50 mM Tris-HCl at pH 8.0 was added to 20 ml of prepared iron-free ovotransferrin and then the solution was kept at room temperature for 1 hour to develop color. Finally, iron binding capacity of the solution was determined by measuring absorbance at 468 nm. In order to eliminate trace amounts of metal contaminants, water used in all the procedures was treated with Chelex (Chelex-100 sodium form, Sigma- Aldrich Inc., St. Louis, USA) according to the method reported by Willard et al. (1969). To determine the effect of pH on iron binding capacity of ovotransferrin, the pH of apo-form ovotransferrin solution was adjusted to pH 4, 5, 6, 7, 8, or 9, and then the iron binding capacity of each solution was measured by the same method described above.

Removal of iron using AG1-X2 resin After separation of iron-bound ovotransferrin, iron-free ovotransferrin was produced ® by removing iron from iron saturated ovotransferrin using AG 1-X2 resin (chloride form, Bio-Rad, California, U.S.A). Firstly, to facilitate the release of iron from holo-ovotransferrin, the pH of iron saturated ovotransferrin solution was adjusted to pH 4.7 using 200 mM-citric acid based on the suggestion of Guo et al. (2003): a large excess of citrate can remove iron from iron bound transferrin family proteins forming ferric citrate. 0.1, 0.2, 0.3, 0.6, or 0.9 g ® of AG 1-X2 resin was added to 100 ml iron-bound ovotransferrin solution (6mg/ml). The mixture was gently stirred for 1 hour until the yellow color of ferric citrate completely disappears and filtered through a Whatman No.1 filter paper. After releasing iron, the residual iron in the solution was estimated by the Ferrozine test (Carter, 1971). Most of citric acid added at this step was assumed to be removed as a ferric citrate form. However, as some of citrate might remain in solution, dialysis (12kD membrane cut-off size) was conducted to remove residual citrate in presence of 20 mM-phosphate buffer, pH 6.5. In addition, to know if a dialysis step is necessary or not, ovotransferrin and protein concentration of the solution obtained after dialysis were compared to that of the solution parepared without perfoming dialysis. The ovotransferrin concentration was determined by iron binding capacity, whereas the protein contents were estimated by Lowry method (Lowry et al, 1951) using prepared reagents (DC Protein Assay, Bio-Rad, Hercules, CA)

Ferrozine test After iron removal from holo-ovotransferrin, the amount of residual iron in ovotransferrin was measured using the Ferrozine method (Carter, 1971). All reactions were performed with disposable polyethylene vessels, which are essentially free of metal contamination. Aqueous solutions were prepared with deionized distilled water. Fresh 0.1%- ascorbic acid in 0.2 N-HCl (1 ml) was added to 1 ml of iron-free ovotransferrin solution and then the mixture was stirred and allowed to stand at room temperature for 5 min. TCA (11.3% solution, 1 ml) was added to the mixture and centrifuged at 2,500 x g for 10 min. Each standard iron solution under certain concentrations was prepared by serial dilution from . 100 ppm of Fe(NH4)2SO4 6H2O solution containing 50 ppm concentrated sulfuric acid with distilled water. Ferrzine color reagent (0.8 ml), which was prepared with 75 mg ferroine, 75 mg neocuproin, and 1 drop HCl in 25 ml distilled water, was added to 2 ml sample and standard iron solution, and then the mixture was stirred gently. Finally, the optical density of the sample was measured at 562 nm after standing for 5 min at room temperature.

Yield of ovotransferrin To determine the yield of ovotransferrin, iron-bound ovotransferrin was extracted twice from iron-saturated egg white solution using 43% ethanol. The iron bound- ovotransferrin in the supernatant obtained from the first extraction by 43% ethanol was precipitated with 59% ethanol (final concentration), whereas that of the second extraction was precipitated with 64% ethanol. The precipitates were dissolved with distilled water and then the iron bound to ovotransferrin was removed using AG1-X2 resin. After iron removal, the concentration of apo-ovotransferrrin from each supernatant was determined by measuring iron binding capacity. Also, after combining the first and second supernatants, the holo-ovotransferrin was precipitated with 59% ethanol to determine the total yield of apo-ovotransferrin produced by the protocol. In calculating the yield of apo-ovotransferrin, total amount of ovotransferrin in 2 x-diluted egg white solution was assumed to be 100%. Also, therorectical value of ovotransferrin in egg white is estimated to be 14 mg/ml since ovotransferrin comprises 12%

of total egg white proteins (Abola et al., 1982), which are reported to be around 12% by Ahn and his co-workers (1997).

Statistical analysis All procedures were replicated three times and data were analyzed using the JMP software (version 5.1.1; SAS Institute., Cary, NC). Differences in the mean values were compared by Tukey HSD, and mean values and standard deviations were reported (Kuehl, 2000)

Results and Discussion

Effects of FeCl3 addition in ethanol treatment The yield of ovotransferrin was calculated by multiplying the concentration and volume of ovotransferrin. The obtained content of ovotransferrin was significantly different . depending upon the amount of added FeCl3: addition of 0.4 mM and 0.6 mM-FeCl3 6H2O for iron saturation of diluted egg white solution produced higher absorbance at 468 nm than . those of lower concentrations of FeCl3 6H2O. There was no significant difference in . absorbance values at 468 nm between 0.4 mM and 0.6 mM-FeCl3 6H2O, indicating that all the apo-ovotransferrin in egg white solution was saturated by addition of 0.4 mM . . FeCl3 6H2O (Fig. 1). The theoretical amount of FeCl3 6H2O required for saturation of apo- form ovotransferrin in 2x-diluted egg white is around 0.2 mM because an ovotransferrin molecule can bind two iron molecules (Azari and Feeney, 1958). Therefore, addition of 0.02,

0.1, 0.2, 0.4, or 0.6 mM-FeCl3.6H2O to 2 times diluted egg white solution corresponds to provide 1/4, 1/2, 1, 2, or 3 times the iron theoretically required. Fraenkel-Conrat (1950) reported that particular spatial configuration of native ovotransferrin is required for the formation of colored iron-bound ovotransferrin and most chemical reagents and denaturing conditions reduced the affinity of ovotransferrin to iron. As suggested by Azeri and Feeney (1958), the formation of holo-transferrin prevented denaturation of ovotransferrin in ethanol.

The result indicated that addition of FeCl3 was a critical step for the purification of

ovotransferrin using ethanol. Also this study confirmed the suggestion of Vachier et al. (1995) who reported that apo-ovotransferrin is denatured easily by ethanol.

Addition of ethanol to diluted egg white solution without adding FeCl3 denatured apo- ovotransferrin and co-precipitated with other proteins in egg white (Fig. l). Also, addition of

0.4mM-FeCl3.6H2O to 2x diluted egg white solution was shown to be the most appropriate for iron saturation of ovotransferrin to prevent the loss of ovotransferrin during ethanol extraction.

pH effect on iron biniding capacity and stability of holo-ovotransferrin in ETOH The pH of egg white solution had a significant effect on the recovery of the apo- ovotransferrin during ethanol extraction (Table 1). Among the different pH conditions, pH 9.0 allowed the highest recovery rate for ovotransferrin, which was around 96%. Even though iron saturation stabilized ovotransferrin by converting apo-form to holo-form, pH 9.0 conditions was found to be the most appropriate for the stabilization of holo-ovotransferrin in ethanol treatment. This indicated that pH was also a critical factor for the stability of iron bound to ovotransferrin. Also, the effect of pH conditions on iron binding capacity of apo-ovotransferrin indicated that iron binds well with ovotransferrin at pH > 6.0, but iron binding capacity of ovotransferrin appeared to be decreased at < pH 6.0 (Fig. 2). It is known that pH conditions influenced the iron binding property of ovotransferrin. Griffiths and Humphreys (1977) reported that the N-domain of ovotransferrin displays lower iron-binding stability and more accelerated release of Fe3+ than the C-domain at acidic pH. Also, they showed that pH itself affects the efficiency of citrate-mediated release of Fe3+ from lactoferrin in the absence of

NaHCO3. Based on these reports, pH is a critical factor for iron binding to native ovotransferrin and for appropriate iron binding of ovotransferrin, higher than pH 6.0 is required. Warner and Weber (1951) reported that even though ovotransferrin can be separated from egg white at any pH condition using ethanol, the intensity of salmon-pink color of iron-bound ovotransferrin solution decreased at < pH 5.5. However the color did not disappear until the pH was lower than < pH 4.0. Also Williams (1975) described that iron was bound preferentially to the N-terminal site at pH 8.0, but released at pH 5.0. At alkaline

pH conditions, iron was bound specifically at the N-terminal binding site, whereas iron binding occurs preferentially at C-terminal site in acid pH conditions (Williams et al., 1978).

Extraction and precipitation of iron-bound ovotransferrin by ethanol SDS-PAGE gel electrophoresis indicated that when ethanol was added to egg white solution at a final concentration of 33%, many proteins besides ovotransferrin were not precipitated and remained in the supernatant. However, at 43% ethanol, most other proteins were found to be precipitated and only a small amount of ovotransferrin was precipitated. Also, at 50% ethanol, slightly larger amount of ovotransferrin was found to be precipitated than that of 43% ethanol (Fig. 3). Therefore, 43% ethanol was chosen as the most suitable condition to extract holo-ovotransferrins possessing high purity and yield from iron-saturated egg white solution. Iron-bound ovotransferrin in the supernatant was precipitated by ethanol addition and the amount of ovotransferrin that remained in the supernatant decreased as the amount of ethanol increased (Fig. 4). When 2.5 ml of ethanol was added to 5 ml supernatant (corresponding to 62% ethanol), most all the ovotransferrin in the supernatant was precipitated. However, for the production of ovotransferrin both with high purity and yield, adding 2 ml of ethanol to 5 ml of the supernatant obtained from 43% ethanol extraction (59% ethanol, final) was the most appropriate conditions for precipitating iron-bound ovotransferrin.

Iron release from holo-form ovotransferrin ® When 0.6, or 0.9 g of AG 1-X2 resin was added to 100 ml iron-bound ovotransferrin solution, residual iron content was 1.8 and 1.6 ppm respectively. Free iron or iron bound to ® ovotransferrin was removed completely by first addition of 0.6 g or 0.9 g of AG 1-X2 resin, and thus a second iron releasing step by addition of 0.6 or 0.9 g of AG1-X2 resin was conducted. The content of residual iron in the solution after second iron removal was less ® than 0.5 ppm (Fig. 5). Therefore, the optimal conditions for AG 1-X2 resin to release and

remove iron from ovotransferrin was 2 times addition of 0.6 g of AG1-X2 resin to 100 ml of holo-form ovotransferrin solution. Considering the concentration of ovotrasnferrin solution

® used in this study was around 6 mg/ml, the optimal ratio of AG 1-X2 required for removing iron from holo-ovotransferrin was 2:1. Citrate was added in the holo-ovotransferrin solution because the release of Fe3+ from ferric transferrin requires presence of a simple anion, which should be closely related to opening of a domain in either C or N-lobe of ovotransferrin (Baldwin et al., 1981; Cheuk et al., 1987; Bailey et al., 1997). However, it is known that citrate binds iron (Masson and Heremans, 1968), and thus any residual citrate may result in underestimation of ovotransferrin by iron binding capacity method. In this study, citric acid was added to adjust the pH of holo-ovotransferrin to 4.7 and the added citric acid formed ferric citrate with iron. This study investigated if the dialysis as a mean to remove residual citrate is necessary because a part of citrate might remain in the solution even though most ferric citrate could be filtered. As a result, there was no significant difference in the concentrations of ovotransferrin and other proteins in the ovotransferrin solution before and after dialysis

(Table 2). Therfore, the ferric citrate was found to be easily removed by AG1-X2 resin and dialysis was not necessary.

Recovery and purity of ovotransferrin Iron-bound ovotransferrin (holo-form) in supernatant is stable at alkaline pH. When the pH of the solution was changed to 4.7 by citric acid for iron removal, however, most of the holo-form of ovotransferrin was changed to the apo-form. The apo-ovotransferrin formed in acidic conditions was denatured easily by ethanol in the solution. Therefore, it was impossible to measure the concentration of ovotransferrin in the supernatant containing ethanol, so the yield was measured only at the final stage where iron-bound ovotransferrin was precipitated with ethanol. The ethanol precipitate was dissolved with distilled water and used to determine the yield of ovotransferrin by measuring iron binding capacity. The amount of apo-ovotransferrin obtained from the first extraction by 43% ethanol was around 75.3% of total ovotransferrin in egg white solution, while the second extraction recovered 18.4% of total amount (Table 3). Also, total yield of ovotransferrin using the protocol in Fig. 7 was around 94% (Table 4), indicating that almost all ovotransferrin in egg white was recovered through the ethanol extraction and precipitation method.

Also, as shown at Fig. 6, the purity of the separated ovotransferrin appeared to be around 80%. Ovoalbumin comprising about 50% of total egg white proteins displayed the broadest band on SDS-PAGE (Lane 3). Almost all ovotransferrin was precipitated by 59% of ethanol (Lane 6), but some ovotransferrin still remained in the supernatant (Lane 5). The purity of apo-form ovotransferrin solution appeared to increase through re-addition of ethanol to final holo-ovotransferrin prepared before iron removal.

Conclusion Most of the ovotransferrin in the natural egg white exists in the apo-form. Thus, the conversion of apo-ovotransferrin to iron-bound form was the most critical step to reduce the loss or denaturation of ovotransferrin by ethanol. The amount of iron required to saturate all the ovotransferrin in egg white was about 2 times the theoretical amount to bind all of the apo-form of ovotransferrin. The holo-ovotransferrin in the presence of excess iron was more stable to ethanol treatment. At pH 9.0, the stability of iron binding in holo-ovotransferrin was significantly higher than at pH 7.0 or 8.0 during ethanol extraction and precipitation steps. The holo-ovotransferrin could be easily separated from iron-saturated egg white with 43% ® ethanol and precipitated at 59% ethanol. AG 1-X2 ion exchange resin was excellent for iron removal from holo-ovotransferrin and the citric acid added for pH adjustment seemed to play a critical role in iron releasing from holo-ovotransferrin. The protocol developed is appropriate for a large–scale production of ovotransferrin because the recovery rate is excellent (94%), the product has relatively high purity (> 80% purity), and the purification ® steps are simple and economical because ethanol can be recovered and AG 1-X2 ion exchange resin can be regenerated. The ovotransferrin produced by this method is applicable for food products because only ethanol was used to separate ovotransferrin from egg white.

Reference Abola, J. E., Wood, M. K., Chweh, A., Abraham, D., and Pulsinelli, P. D. 1982. In The Biochemistry and Physiology of Iron (Saltman, P., and Hegenaur, J. Eds) pp. 27-34. Elsevier Scientific Publishing Co., Inc., New York

100

Al-Mashikhi, S. A. and Nakai, S. 1987. Separation of ovotransferrin from egg white by immobilized metal affinity chromatography. Agric. Biol. Chem. 51(11): 2881-2887. Ahlborn, G. J., Clare, D. A., Sheldon, B. W., and Kelly, R. W. 2006. Identification of eggshell membrane proteins and purification of ovotransferrin and ß-NAGase from hen egg white. Protein J. 25(1): 71-81. Ahn, D. U., Kim, S. M., and Shu, H. 1997. Effect of egg size and strain and age of hens on the solides content of chicken eggs. Poul. Sci. 76: 914-919. Awade, A. C., Moreau, S., Molle D., Brule, G., and Maubois, J. L. 1994. Two-step Chromatographic procedure for the purification of hen egg white ovomucin, lysozyme, ovotransferrin and ovalbumin and characterization of purified proteins. J. Chromatog. A 677: 279-288. Awade, A. C. 1996. On hen egg fractionation: applications of liquid chromatography to the isolation and the purification of hen egg white and egg yolk proteins. Eur. Food Res. Technol. 202: 1-14. Azari, P., and Baugh, R. F. 1967. A simple and rapid procedure for preparation of large quantities of pure ovotransferrin. Arch. Biochem. Biophys. 118: 138-144. Azari, P. R., and Feeney, R. E. 1958. Resistance of metal complexes of ovotransferrin and transferrin to proteolysis and to thermal denaturation. J. Biol. Chem. 232: 293-302. Bailey, C. T., Byrne, C., Chrispell, K., Molkenbur, C., Sackett, M., Reid, K., McCollum, K., Vibbard, D., and Catelli, R. 1997. Effect of a covalently attached synergistic anion on cheater-mediated iron-release from ovotransferrin: Additional evidence fro two concurrent pathways. Biochem. 36: 10105-10108. Bain, J. A., and Deutsch, H. F. 1948. Separation and characterization of co albumin. J. Biol. Chem. 171: 547-555. Baldwin, D. A., and de Sousa, D. M. R. 1981. The effect of salts on the kinetics of iron release from N-terminal and C-terminal monoferrictransferrin. Biochem. Biophy. Res. Commun. 99: 1101-1107. Carter, P. 1971. Spectrophotometric determination of serum iron at the submicrogram level with a new reagent (ferrosine). Anal. Biochem. 40: 450-458.

101

Cheuk, M. S., Keimg, W. M., and Loh, T. T. 1987. The effect of pH on the kinetics of iron release from diferric ovotransferrin induced by pyrophosphate. J. Inorg. Biochem. 30: 121-131. Chung, C. M., Chan, S. L., and Shimizu, S. 1991. Purification of transferrin and lactoferrin using DEAE affi-gel blue. Int. J. Biochem. 23: 609-616. Cunningham, F. E., and Lineweaver, H. 1965. Stabilization of egg white proteins to pasteurization temperatures above 60o C. Food Technol. 19: 136-141. Fraenkel-Conrat, H. 1950. Comparison of the iron- binding activities of ovotransferrin and of hydroxylamidoproteins. Arch. Biochem. 28: 452-463. Griffiths, E., and Humpherys, J. 1977. Bacteriostatic effect of human milk and bovine colostrum on E coli: Importance of bicarbonate. Infect. Immun. 15(2): 396-401. Guerin, C., and Brule, G. 1992. Separation of three proteins from egg white. Sciences des- Aliments 112(4): 705-720. Guo, M., Harver, I., Yang, W., Coghill, L., Campopiano, D. J., Parkinson, J.A., MacGillivrary, R. R. A., Harris, W. R., and Sadler, P. J. 2003. Synergistic anion and metal binding to the ferric ion-binding protein from Neisseria gonorrhoeae. J. Bolo. Chem. 278(4): 2490-2502. Kuehl, R. O. 2000. Design of Experiments; Statistical principles of research design and analysis. 2nd Ed. New York, Duxbury Press. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: 265-275. Mandrels, S. 1960. Use of DEAE-cellulose in the separation of proteins from egg white and other biological materials. J. Chromatogram. 3: 256-264. Masson, P. L., and Heremans, J. F. 1968. Metal-combining properties of human lactoferrin (red milk protein). I. The involvement of bicarbonate in the reaction. Eur. J. Biochem. 6: 579-584. Phelps, R. A. and Cann, J. R. 1956. On the modification of co albumin by acid. II. Effect of pH and salt concentration on the sedimentation behavior, viscosity and osmotic pressure of co albumin solutions. Arch. Biochem. Biophys. 61(1): 51-71.

102

Reiter, B., Brock, J. H., and Steel, E. D. 1975. Inhibition of Escherichia coli by bovine colostrum and post-cloistral milk. II. The bacteriostatic effect of lactoferrin on a serum susceptible and serum resistant strain of E. coli. Immune. 28: 83-95. Rhodes, M. B., Azari, P. R., and Feeney, R. E. 1958. Analysis, fractionation and purification of egg white proteins with cellulose action exchanger. J. Biol. Chem. 230: 399-408. Stadelman, W. J., and Cotterill, O. J. 1986. Egg Science and Technology 3rd Ed. Westport Conn., Avi Pub. Co. Vachier, M. C., Piot, M., and Awade, A. C. 1995. Isolation of hen egg white lysozyme, ovotransferrin and ovalbumin, using a quaternary ammonium bound to a highly cross linked agars matrix. J. Chromatogram. B 664: 201-210. Warner, R. C. 1954. Egg proteins. In The Proteins. Erath H, Bailey K, (Ed). Academic Press, New York. Warner, R. C., and Weber, I. 1953. The metal combining properties of co albumin. J. Am. Chem. Soc. 75: 5094-5101. Warner, R. C., and Weber, I. 1951. The preparation of crystalline ovotransferrin J. Biol. Chem. 191: 173-180. Willard, J. M., Davis, J. J., and Wood, H. G. 1969. Phosphoenol pyruvate carboxytransphosphorylase. IV. Requirment of metal cations. Biochem. 8: 3137-3144. Williams, J. 1974. The formation of iron-binding fragments of hen ovotransferrin by limited proteolysis. Biochem. J. 141: 745-752. Williams, J. 1975. Iron-binding fragments from the carboxyl-terminal region of hen ovotransferrin. Biochem. J. 149: 237-244. Williams, J., Evans, R. W., and Moreton, K. 1978. The iron-binding properties of hen ovotransferrin. Biochem. J. 173: 535-542.

103

Table 1. Effect of pH on the yield ovotransferrin from 2x-diluted egg white solution.

First supernatant1 Second supernatant2 Vol. x Conc. pH Conc. (mg/ml)3 Vol. (ml) Conc. (mg/ml) Vol. (ml) (mg) Yield (%)*

7 9.40ab ± 0.15 61.3a ± 1.26 2.48b ± 0.62 14.0ab ± 0.25 610.7ab ± 25.9 87.2 ± 3.57 8 8.94b ± 0.82 61.7a ± 2.08 2.82b ±1.00 13.7b ± 0.14 581.4b ± 32.2 83.1 ± 4.50 9 9.69a ± 0.08 62.3a± 0.29 6.62a ± 0.62 14.3a ± 0.14 671.5a ± 0.58 95.8 ± 0.15

1The solution obtained from the first extraction with 43% ethanol. 2The solution obtained from the second extraction with 43% ethanol 3Final apo-ovotransferrin solutions free of ethanol and irons were used for measuring iron binding capacity. *The yields were calculated based on theorectical value (7.5 mg) of ovotransferrin in 2 times diluted egg white solution. a ~ bMean values with different superscript letters are significantly different (p < 0.05). n=3.

104

Table 2. The influence of citric acid on iron binding capacity of ovotransferrin.

Treatment Protein (mg/ml) Ovotransferrin (mg/ml) Residual iron (ppm)1

Without dialysis 8.59a ± 0.14 8.00a ± 0.57 0.52a ± 0.08 Dialysis 8.57a ± 0.17 8.42a ± 0.67 0.53a ± 0.06

1The amount of iron remained in the solution after iron removal was measured. a Mean values with different superscript letters are significantly different (p < 0.05, n=3).

105

Table 3. The yields of each supernatant obtained by adding 43% ethanol to egg white solution.

Extraction Conc. (mg/ml) Vol. (ml) Vol. x Conc. (mg) Yield*(%)

First1 11.99 ± 0.23 43.95 ± 0.13 527.3 ± 8.64 75.3 ± 1.23 Second2 6.65 ± 0.37 19.32 ± 0.36 128.49 ± 1.80 18.4 ± 0.26

1 Ethanol was added to 100 ml of 2x-diluted egg white to make 43% ethanol (final) and then the holo-ovotransferrin was precipitated by 59% ethanol (final). The holo- ovotransferrin in precipitate was dissolved with distilled water (a half volume of supernatant obtained from 43% ethanol extraction). 2 In second extraction, the holo-form of ovotransferrin in supernatant was precipitated by 64% ethanol. n=3. *The yields were calculated based on theorectical value (700 mg) of ovotransferrin in 100ml of 2 times diluted egg white solution.

106

Table 4. The yield of final ovotransferrin solution separated by the developed method.

Sample Conc. (mg/ml) Vol. (ml) Vol. x Conc. (mg) Yield (%)

Egg white1 7.50 100 750 100 Final solution2 6.36 ± 0.37 103 ± 2.10 655.1 ± 45.53 93.58 ± 0.6

1The theorectical concentration of ovotransferrin in 2x-diluted egg white solution used was 7.0 mg/ml. 2Apo-ovotransferrin solution produced after iron removal. n=3

107

Fig. 1. Effect of addition iron on extraction and precipitation of ovotransferrin with ethanol.

Egg white solution contained 50 mM-NaHCO3 and 0.15 M-NaCl. Bars with different letters are significantly different (p < 0.05, n = 4).

0.6 c c 0.5

b 0.4

0.3

0.2 a a

Absorbance at 468 nm 0.1

0.0 0.05 0.1 0.2 0.4 0.6

FeCl3.6H2O (mM)/ 2 x diluted egg white

108

Fig 2. Changes of iron binding capacity by different pH condition. The concentrations of ovotransferrin were determined using iron binding capacity. Bars with different letters are significantly different (p < 0.05, n = 4).

a a 12.0 a a

10.0

8.0

6.0

4.0 b

b 2.0 Ovotransferrin conc. (mg/ml) conc. Ovotransferrin

0.0 456789

109

Fig. 3. SDS-PAGE of ovotransferrin solution obtained after addition of various concentrations of ethanol. Lane 1: diluted egg white, Lanes 2~5: supernatants obtained by 20%, 33%, 43% and 50% of ethanol extraction, lanes 6~9: precipitates of 20%, 33%, 43%, and 50% of ethanol extraction

43%

Ovotransferrin

1 2 3 4 5 6 7 8 9

110

Fig. 4. SDS-PAGE of ovotransferrin solution obtained after addition of different concentrations of ethanol for precipitating holo-ovotransferrin in supernatant produced by first extraction. Lane 1: 2x-diluted egg white solution, lanes 2~5 : precipitates obtained after addition of 1 ml (53%, lane 2), 1.5 ml (56%, lane 3), 2 ml (59%, lane 4) and 2.5 ml (62%, lane 5) of ethanol to 5 ml of supernatant obtained by 43% ethanol extraction followed by dissolution of the precipitant with distilled water, lane 6~9: supernatant gained after adding 1 ml (53%, lane 6), 1.5 ml (56%, lane 7), 2 ml (59%, lane 8) and 2.5 ml (62%, lane 9) of ethanol to 5 ml of supernatant.

59%

1 2 3 4 5 6 7 8 9

111

Fig. 5. Amounts of residual iron in ovotransferrin solution after first and second treatments with AG1-X2 resin. The ovotransferrin concentration of the solution used at this study ovotransferrin was around 6.5 mg/ml. Bars with different letters indicate significant different values (p < 0.05, n = 4).

4.0 a 3.5 ab 3.0 b 2.5 c 2.0 a a c a 1.5 1.0 b R (ppm) iron esidual 0.5 b first 0.0 0.1 0.2 0.3 Second 0.6 0.9

AG1-X2 resin (g)/ 100 ml solution

112

Fig. 6. Ovotransferrin purified from 2x-diluted egg white solution. Lane 1: marker, lane 2: empty, lane 3: 2x-diluted egg white, lane 4: first supernatant obtained after ethanol extraction from iron-bound egg white solution, lane 5: supernatant obtained after precipitation of ovotransferrin with ethanol from first supernatant, lane 6: final purified ovotransferrin dissolved in the distilled water after precipitation with ethanol.

200 kD

97 kD Ovotransferrin 66 kD

45 kD

1 2 3 4 5 6

113

Fig. 7. Schematic diagram for the isolation of ovotransferrin from egg white

Dilution of egg white with 1 volume distilled water

Add NaHCO3, NaCl, and FeCl3 and adjust pH to 9.0

Stir for 30 min, add ethanol (43%, final), and centrifuge at 3,220 x g for 20 min

Supernatant (1st) Precipitant Re-extract with 43% ethanol Centrifugation at 3,220 x g, 20min

Supernatant (2nd)

Add ethanol (Supernatant: ethanol = 5: 2) and centrifuge

Dissolve precipitant with distilled water

pH adjustment to 4.7 with 200 mM citric acid

Add AG1-X2 resin and stir for 1 hr and filter

Repeat AG1-X2 resin treatment

Apo-ovotransferrin

114

CHAPTER 5. Effect of Zn2+, sodium bicarbonate, and citric acid on the antibacterial activity of ovotransferrin against Escherichia coli O157:H7 and Listeria monocytogenes in model systems and hams

A paper which will be submitted to Journal of Food Protection

K. Y. Ko1, D. U. Ahn2, H. A. Ismail, and A. Mendonca

Abstract

The influence of bicarbonate (NaHCO3) and citric acid on the antibacterial activity of apo-ovotransferrin in model systems and hams were investigated. The antibacterial activities

of 20 mg/ml ovotransferrin solution added with NaHCO3 (0, 25, 50, or 100 mM) or citric acid (0.25% or 0.5%) were estimated against five different E. coli O157:H7 and L. monocytogenes strains in BHI broth media. The antimicrobial properties of ovotransferrin saturated with Fe2+ or Zn2+ against E. coli O157:H7 and L. monocytogenes were also measured in vitro. In addition, ovotransferrin solutions containing either 100 mM-NaHCO3 or 0.5% citric acid were applied to commercial hams inoculated with E. coli O157:H7 or L. monocytogenes. The antibacterial activities of each treatment were analyzed during refrigerator storage for 4 weeks. Antimicrobial activity of ovotransferrin increased as the

concentration of added sodium bicarbonate increased. 100 mM-NaHCO3 promoted the antibacterial activity of ovotransferrin against E. coli O157:H7 and L. monocytogenes significantly. L. monocytogenes was shown to be susceptible to 0.5% citric acid alone, whereas 0.5% citric acid enhanced the antibacterial activity of ovotransferrin against E. coli

O157:H7. NaHCO3 addition annulled the strong antibacterial activity of ovotransferrin plus citric acid against E. coli O157:H7 and using sodium citrate instead of citric acid did not show any antibacterial activity against the two bacterial species. The antimicrobial activity of ovotransferrin increased significantly by acidic conditions. Unlike the model system, the

1 Primary researcher and author 2 Author of correspondence.

115

antibacterial activity of ovotransferrin added with either 100 mM-NaHCO3 or 0.5% citric acid was not detected on ham. Zn-bound ovotransferrin prevented the growth of L. monocytogenes but not E.coli O157:H7 as did apo-ovotransferrin combined with 100 mM-

NaHCO3. Also, Fe-bound ovotransferrin had little or no antibacterial activity against E. coli O157:H7 and L. monocytogenes in BHI broth culture. This study described that certain concentrations of bicarbonate and citric acid improved the antibacterial activity of ovotransferrin and Zn2+ improved the antibacterial capacity of natural ovotransferrin against certain pathogens. However, the study also suggested that there are many limitations for applying ovotransferrin directly to meat or meat products in contrast to model systems.

Introduction Ovotransferrin is a monomeric glycoprotein consisting of 686 amino acids with 78 kDa molecular weight, has no free sulfhydryl groups or phosphorus, and binds reversibly with two Fe2+ ions per molecule concomitantly with two CO2- ions (Abola et al., 1982; Stadelman and Cotterill, 1986; Aisen and Listowsky, 1980; Ibrahim, 1997). It is identified as the second major avian egg white protein constituting 12% of total egg white protein and contributes to the egg’s defense against microbial infection and rotting (Abola et al., 1982). Ovotransferrin has demonstrated antibacterial activity against a wide spectrum of bacteria, including, Escherichia coli (Shade and Carolin, 1944; Valenti et al., 1983a), Pseudomonas spp., Streptococcus mutans (Valenti et al., 1983a), Staphylococcus aureus, Bacillus cereus (Ibrahim, 1997), Salmonella enteritidis (Baron et al., 1997; 2000), and Candida (Valenti et al. 1985). The antimicrobial activity of ovotransferrin is influenced by several factors such as bicarbonate and citrate concentrations, pH conditions, composition of medium or metal irons like Zn2+ as well as bacterial species or strains (Valenti et al., 1981, 1983a, 1987; Brock, 1985). Bicarbonate known as a synergistic anion (Schalabach and Bates, 1975) is a prerequisite for the iron uptake of transferrin and lactoferrin (Aisen, 1989; Pakdaman and Elhage Chahine, 1996; Pakdaman et al., 1998) because it prevents iron binding by citrate (Griffiths and Humphereys, 1977; Peaker and Linzell, 1975). The anion may serve as a bridging ligand between proteins and metal ions (Aisen, 1980). Bicarbonate weakly binds to apo-transferrin in the absence of a metal ion (Woodworth, 1975), but it is not known if the

116

binding is specific. The affinity of ovotransferrin for bicarbonate in each lobe is about three- fold weaker than that of serum transferrin (Bellounis et al., 1996; Pakdaman and El Hage 2- - Chahine, 1997). Also, bicarbonate or carbonate (CO3 or HCO3 ) occupies one of Fe3+coordination sites consisting of four lateral chains with two tyrosines, one histidine, one aspartate, and a carbonate or bicarbonate (Bailey et al, 1988; Andeson et al., 1989; Zuccola, 1992; Kurokawa et al., 1995: Rawas et al., 1996). An excess of citrate can annul the antimicrobial activity of ovotransferrin by chelating iron, and thus the iron binding capacity of ovotransferrin is affected by metal chelates such as citrate (Valenti et al., 1983a; Phelpa and Antonini, 1975). However, it is also well known that the effect of citrate on antibacterial activity of ovotransferrin depends on the species of microorganism. For example, E. coli possessing iron transport system mediated by citrate (Frost and Rosemberg, 1973) can annul the antibacterial activity, whereas the addition of excess citrate does not influence the antimicrobial activity of ovotransferrin against S. aureus and Candida, which do not have an iron citrate transport system (Valenti et al., 1980; Valenti et al., 1985). As ovotransferrin can bind metal ions such as Cu, Al, Zn, and Fe, the antimicrobial effect of ovotransferrin is expected to decrease by binding other metal ions. Generally, iron, copper, and zinc bind the same sites in the ovotransferrin molecules (Warner and Weber, 1953). However, in some studies ovotransferrin saturated with metals such as zinc and iron still displayed antimicrobial activities (Ibrahim, 1997; Valenti et al., 1985, 1987). Valenti et al. (1987) reported that Zn2+-loaded ovotransferrin had more bactericidal activity than apo- ovotransferrin and other metal complexes, and the antimicrobial effect of Zn2+-ovotransferrin was attributed to the direct interaction of Zn2+-ovotransferrin with the surface of bacteria rather than iron deprivation from medium. However, other researchers reported that the cause of antibacterial action of ovotransferrin is due to more complex mechanisms related to direct interaction between ovotransferrin and microorganisms in addition to iron deprivation (Valenti et al., 1985). Ovotransferrin has a great potential as a natural antimicrobial agent like lactoferrin. However, no attempt has been used to study the control of pathogenic and nonpathogenic microorganisms in meat or other food processings using ovotransferrin. To identify the

117

possibility of ovotransferrin utilization as an antibacterial agent in meat, methods to activate or improve antimicrobial capability of natural apo-ovotransferrin against E. coli O157:H7 and L. monocytogenes are required. In this study, the influences of bicarbonate and citric acid as synergistic anions on the antibacterial activity of apo-ovotransferrin were investigated. Also, ovotransferrin combined with bicarbonate and citric acid was applied to ham to determine the possibility of using ovotransferrin as a natural antimicrobial agent in meat processing.

Materials and Methods

Ovotransferrin Apo-ovotransferrin used in this study was produced by the method of Ko and Ahn (2007, not published). After iron saturation of 2x-diluted egg white solution, holo- ovotransferrin (iron-free form) was separated from the egg white solution. Apo- ® ovotransferrin was obtained by removing iron bound to holo-ovotransferrin using AG 1-X2 resin (chloride form, Bio-Rad, California, U.S.A). The apo-ovotransferrin was freeze-dried and used for microbial experiments. The dried apo-ovotransferrin was dissolved in distilled water, the pH adjusted to 7.4, and NaCl added to 0.15 M. The purity of prepared ovotransferrin was around 80% and the content of residual iron in the prepared apo- ovotransferrin solution was less than 0.5 ppm. Prior to using it in microbial study, it was sterilized by 0.22 μm syringe filter (Whatman Inc. Florbam Park, NJ. USA)

Bacterial strains and growth conditions Five strains of E. coli O157:H7 (ATCC 43890, C467, FRIK 125, ATCC 43895, and 93-062) and five strains of L. monocytogenes (NADC 2045, H7962, H7969, H7762, H7596) were used. The pathogens were obtained from the microbiology laboratory of Food Science Department at Iowa State University. To activate inocula, E. coli O157: H7 and L. monocytogenes strains were individually cultured two times in brain heart infusion broths (BHI, Remel Inc., Lenexa, KS, USA) for 24 hours at 35oC. The bacteria were grown to stationary or mid-log phase and harvested. Each culture solution (5 ml) was collected in a

118

sterilized centrifuge tube and centrifuged (100,000 x g for 10 min) at 4o C. The pellets for all E. coli strains were combined, washed with saline, and re-suspended in 25 ml saline to have approximately the same number of each strain in the cocktail. The pellets for all L. monocytogenes strains were treated the same. The following method was used to determine the number of bacteria used in each experiment: after serial dilution with a 1-ml aliquot of the suspension into 0.1% peptone water (Difco Laboratories, Detroit, MI), 0.1 ml of appropriate diluted solution was distributed on the BHI agar media. After incubating at 35oC for 24 ~ 48 hr, the numbers of E. coil O157:H7 and L. monocytogenes colonies appearing on the BHI agar plates were counted.

Sodium bicarbonate effect Preliminary disc diffusion experiment (Bauer et al, 1966) was performed using ovotransferrin (5, 10, 15, or 20 mg/ml) and NaHCO3 (30, 40, or 50 mM) combinations to determine a minimum concentration of ovotransferrin showing antibacterial activity against E. coli O157:H7 and L. monocytogenes on BHI agar plates. Ovotransferrin without sodium bicarbonate did not show any antibacterial activity, but 15 or 20 mg/ml of ovotransferrin combined with 40 mM- and 50 mM-sodium bicarbonate showed some antibacterial activity against E. coli O157:H7 and L. monocytogenes on BHI agar plates. Therefore, 20 mg/ml (final) of ovotransferrin solution was used to determine the effect of sodium bicarbonate on antibacterial activity of ovotransferrin in this study. After preparing sterilized ovotransferrin

(40 mg/ml), the same volume of 2x strength BHI broth and then 2 M-sterilized NaHCO3 solution were added to make the final concentration of bicarbonate 0, 25, 50, or 100 mM in the solution. Also, 103 ~ 104 CFU/ml of E. coli O157:H7 and L. monocytogenes were inoculated into the sterilized BHI broth media containing ovotransferrin and sodium bicarbonate. After inoculation, the broth culture was incubated at 35oC for 1-8 days and the antibacterial capacity of each treatment against E. coli O157:H7 and L. monocytogenes was estimated by measuring the turbidity of BHI broth culture at 620 nm.

119

Citric acid and citrate effect As some microbial strains possess an iron transport system mediated by citrate (Frost and Rosemberg, 1973), the presence of citrate may inhibit the antibacterial activity of ovotransferrin. So, the effect of citrate or citric acid on the antibacterial activity of ovotransferrin was investigated. Apo-ovotransferrin solution (40 mg/ml) was added to the same volume of 2 x-concentrated BHI broth. Stock citric acid solution was added to the BHI broth to make the ovotransferrin solution with 0.25% or 0.5% citric acid. BHI broth culture with citric acid (but with no ovotransferrin) was inoculated with strains and used as a control. Also, 25 mM sodium citrate, equivalent to 0.5% citric acid, was added to the BHI broth containing ovotransferrin to determine if the effect of citric acid on antibacterial activity of ovotransferrin is due to acidic pH or the function of synergistic anions. Activated E. coli O157:H7 and L. monocytogenes at 104 ~ 105 CFU/ml were inoculated aseptically to each BHI broth culture. After mixing the broth culture, they were incubated at 35oC. An aliquot (1 ml) of the culture solution was serially diluted with 0.1% peptone water and 0.1 ml of each serially diluted solution was spread homogenously on the BHI agar plate. The viable cells were enumerated after incubating the BHI agar plates at 35oC for 24 ~ 48 hrs.

Antimicrobial activity of apo-, Fe-, and Zn-ovotransferrin The content of residual iron in apo-ovotransferrin used in this study was around 0.38 3+ 2+ . ppm. To saturate the apo-ovotransferrin with Fe or Zn , 0.15 ml of 0.2 M-FeCl3 6H2O or

0.16 ml of 0.2 M ZnCl2 was added to 30 ml of ovotransferrin solution (40 mg /ml) including

50 mM-NaHCO3 solution at pH 7.4 since one ovotransferrin molecule with 80 kDa binds 2

metal ions. After adding FeCl3 or ZnCl2 to ovotransferrin solution, the prepared solutions were incubated at room teperature for 2 ~ 3 hours for complete saturation and then dialysis (12kD membrane cut-off size) was conducted to remove sodium bicarbonate. After preparing the iron or zinc-saturated ovotransferrin, Zn-ovotransferrin, Fe-ovotransferrin, and apo- ovotransferrin solutions were added with the same volume of 2 x -concentrated BHI broth 4 5 culture, 50 mM- and 100 mM-NaHCO3 were added, and then 10 ~ 10 CFU/ml E. coli O15:H7 or L. monocytogenes strains were inoculated. The treatments were serially diluted in sterile 0.1% peptone water and were surface spread (0.1ml, in duplicate) on the BHI agar

120

media. After incubation at 35oC for 24 ~ 48hrs, the viable cells of E. coli O157:H7 and L. monocytogenes from each treatment were enumerated.

Antibacterial activity of ovotransferrin with citric acid or NaHCO3 on ham Commercial hams were purchased from local retail stores. Hams were sliced to 0.2 cm-thickness and vacuum packaged (-1 bar of vacuum with 10 seconds dwell time) in low o oxygen-permeable bags (nylon/polyethylene, 9.3 ml O2/m2/24 hr at 0 C; Koch, Kansas City, MO). After randomly separating the sliced hams into two groups and irradiated at 0 or 5 kGy using a Linear Accelerator (Circe IIIR; Thomason CSF Linac, Saint-Aubin France). The energy and power level were 10 Mev and 10 kW respectively, and the average dose rate was 88.3 kGy/min. To confirm the target dose, 2 alanine dosimeters per cart were attached to the top and bottom surface of a sample. The alanine dosimeter was read using a 104 Electron Paramagnetic Resonance Instrument (Bruker Instruments Inc., Billerica, MA). After irradiation, the irradiated and non-irradiated hams were transferred to microbiology lab and stored at 4oC. E. coli O157:H7 and L. monocytogenes stock suspension (0.1 ml) were inoculated aseptically on the surface of a sliced ham to have the number of cells at 106 CFU/cm2. After inoculated ham samples were manually mixed for 30 second to distribute the inocula evenly, the packaged samples were further separated into 3 groups. One ml of

ovotransferrin solutions (20 mg/ml) containing 100 mM-NaHCO3 or 0.5% of citric acid was distributed evenly on the same surface of sliced ham as in the inoculation step, vacuum- packaged (1 slice per bag, Multivac A300/16, Sepp Haggenmuller KG, Wlfertschwenden, Germany) in nylon-polyethylen bags, and stored in a refrigerator. The surviving bacteria wwa enumerated after incubating ham samples for 0, 5, 10, 15, 22, 29 and 34 days at 4oC. Each package was aseptically opened using alcohol-sterilized scissors. Sterile 0.1% peptone water (30 ~ 50 ml) was added to each bag and then processed at normal speed for 1 minute in a stomacher. Samples were serially diluted with 0.1% peptone water and 0.1 ml of the diluted E. coli O157:H7 culture solution was spread on a MacConkey sorbitol agar (Difco Laboratories, Detroit, MI), whereas those of L. monocytogenes were spread on a MOX media. The number of viable E. coli O157:H7 and L.

121

monocytogenes cells appeared on MacConkey sorbitol agar and MOX media were enumerated after incubation for 24 ~ 48 hr at 35oC.

Statistical analysis All work was replicated three times and data were analyzed using the JMP software (version 5.1.1; SAS Institute., Cary, NC). The differences in the mean values were compared by Tukey HSD, and mean values and standard error of the mean (SEM) were reported (Kuehl, 2000). If a significant F-value was obtained (p < 0.05), the Tukey-Kramer test was performed to determine whether the means of treatments were different from control. Statistical significance for all comparisons was set at p < 0.05.

Results and Discussion

Antibacterial activity of natural apo-ovotransferrin Natural apo-ovotransferrin showed a weak antibacterial property against E. coli O157:H7 (Fig. 1A), but its antimicrobial activities against L. monocytogenes were apparent after 4 days of refrigerated storage (Fig. 1B). The antibacterial activity of natural apo- ovotransferrin varied with microorganisms. Valenti et al. (1983a) reported that Pseudomonas spp., E. coli, and Streptococcus mutans were susceptible, whereas Staphylococcus aureus, Proteus spp., and Klebsiella spp. were relatively resistant to ovotransferrin. Activated lactoferrin is reported to have higher antibacterial effectiveness toward L. monocytogenes than that of E. coli O157:H7 (http://www.newswise.com/articles/view/525381). Like antibacterial activity of lactoferrin, apo-ovotransferrin was more effective in inhibiting the growth of L. monocytogenes than E. coli O157:H7.

Sodium bicarbonate

The influence of NaHCO3 against the growth of E. coli O157:H7 and L. monocytogenes was investigated before identifying its synergistic effect with ovotransferrin. The turbidity of BHI broth culture with E. coli O157:H7 decreased as the concentration of

NaHCO3 increased. However, considering the value of turbidity 100 mM-NaHCO3 alone

122

was around 0.6 from 2 days (Fig. 1A), the solution was demonstrated not to restrain the growth of E. coli O157:H7. Corral et al. (1988) reported that 120 mM (1% w/v) sodium bicarbonate reduced the number of E. coli in stationary phase by 4 log through the changes in physicochemical environment of the microorganisms. Our result, however, indicated that E.

coli O157:H7 was relatively resistant to NaHCO3 probably because we used five different

strains of E. coli O157:H7, which have different susceptibility against NaHCO3 and high adaptation capability by activation. In contrast to E. coli O157:H7, the turbidity of BHI broth culture with L. monocytogenes increased as the concentration of sodium bicarbonate increased. The 100 mM-sodium bicarbonate treatment had higer turbidity value than that of 25 mM or 50 mM sodium bicarbonate (Fig. 1B). Czuprynski and Faith (2002) reported that 50 μl of 10% sodium bicarbonate (wt/vol) enhanced the virulence of L. monocytogenes EGD or L. monocytogenes strain Scott A in mice. These results indicated that L. monocytogenes strains are less susceptible to sodium bicarbonate than and E. coli O157:H7. Especially 100 mM- sodium bicarbonate alone seemed to support the growth of L. monocytogenes after 4 day incubation compared with turbidity of control.

Sodium bicarbonate on antibacterial activity of ovotransferrin Sodium bicarbonate (25, 50 or 100 mM) was added to BHI broth containing apo- ovotransferrin to improve its antibacterial activity against E. coli O157:H7 and L.

monocytogenes. The antibacterial activity of ovotransferrin plus NaHCO3 depended on the concentration of sodium bicarbonate added (Fig. 1). As the concentration of sodium

bicarbonate increased, antimicrobial activity of ovotransferrin plus NaHCO3 increased. According to many reports, Fe3+-specific binding capacity of transferrin is promoted at constant pH as the concentration of bicarbonate increased (Abdallah and EL Hage Chahine, 1998; Workman et al., 1975 Aisen, 1989; Pakdaman and El Hage Chahine, 1997). Like transferrin, sodium bicarbonate increased the antibacterial effect of apo-ovotransferrin against E. coli O157:H7 and L. monocytogenes. Also, with E. coli O157:H7, the turbidity of

ovotransferrin solutions containing NaHCO3 was lower than that of apo-ovotransferrin or

NaHCO3 alone (Fig. 1A). This result indicated that sodium bicarbonate promoted the

123

antibacterial activity of apo-ovotransferrin against E. coli O157:H7. The combination of ovotransferrin with 100 mM sodium bicarbonate was the best condition for inhibiting the growth of E. coli O157:H7. Ovotransferrin added with 25 mM- or 50 mM- sodium bicarbonate did not completely inhibit the growth of L. monocytogenes until 4 days of incubation, whereas ovotransferrin containing 100 mM-NaHCO3 prevented the growth of L. monocytogenes. There was an apparent synergistic effect of 100 mM-NaHCO3 on the antimicrobial activity of apo-ovotransferrin against L. monocytogenes (Fig. 1B). Considering ovotransferrin plus 100 mM-NaHCO3 has the lowest turbidity value in two strains (Fig 1),

100 mM-NaHCO3 was the most appropriate concentration to improve the antibacterial activity of apo-ovotrasnferrin against E. coli O157:H7 and L. monocytogenes. Carbonate anions interact directly with the complex of transferrin- or lactoferrin-bound metals such as iron and copper (Eaton et al., 1990). Apo-transferrin is also known to interact - 2- with two HCO3 or CO3 , and the carbonic ion is considered to result in ionic interactions with arginine residues of certain side chain in binding sites of its two lobes (Bellounis et al.,

1996; Pakdaman et al, 1998). NaHCO3 at 50 mM increased the antibacterial activity of ovotransferrin against E. coli W1485, whereas ovotransferrin without NaHCO3 did not show antibacterial activity (Valenti et al., 1983a). The present study suggests that a certain concentration of sodium bicarbonate can increase the antibacterial effectiveness of ovotransferrin toward E. coli O157:H7 and L. monocytogenes.

Citric acid alone effect The effects of citric acid alone on the growth of E. coli O157:H7 and L. monocytogenes were different: L. monocytogenes was more sensitive to citric acid than E. coli O157:H7 (Fig. 2.). Also, the antibacterial activity of citric acid depended upon the concentration of citric acid. Citric acid at 0.5% showed a much greater antibacterial activity than 0.25% citric acid aginst two pathogens. Also, 0.5% citric acid alone did not completely inhibit the growth of E. coli O157:H7, even though the number of viable cells of 0.5% citric acid alone indicated to be lower than that of control or ovotransferrin itself after 2days of incubation at 35oC (Table 1 and Fig. 2A). Also in L. monocytogenes, 0.5% citric acid resulted in around 1 log reduction of viable L. monocytogenes cells from inoculated cells (104

124

CFU/ml) after 2 days of incubation (Table 2 and Fig. 2B). At the present study, citric acid showed stronger antibacterial activities against L. monocytogenes than E. coli O157:H7. Through this study, we found that L. monocytogenes are susceptible to 0.5% citric acid itself.

Citric acid and citrate effect on antibacterial activity of ovotransferrin Even though ovotransferrin plus 0.25% citric acid did not show antibacterial activity, 0.5% citric acid plus ovotransferrin appeared to be bactericidal against E. coli O157:H7 cells (Table 1 and Fig. 2A). 0.5% citric acid was found not to inhibit completely the growth of E. coli O157:H7, ovotransferrin plus 0.5% citric acid indicated to be 1 log reduction from the viable cell number inoculated initially against E. coli O157:H7. Therefore, 0.5% citric acid seems to enhanc the antibacterial activity of ovotransferrin against E. coli O157:H7. Also ovotransferrin plus 0.25% citric acid did not show significant difference from 0.25% citric acid alone in bactericidal effect against L. monocytogenes after 2 days of incubation at 35oC. Also, 0.5% citric acid alone and ovotransferrin plus 0.5% citric acid showed bactericidal effect against L. monocytogenes after 2days of incubation at 35oC (Table 2 and Fig. 2B). In contrast to E. coli O157:H7, 0.5% citric acid did not enhance antibacterial activity of ovotransferrin against L. monocytogenes since there was no significant difference in the number of viable cells between 0.5% citric acid alone and 0.5% citric acid plus ovotransferrin (Fig. 2B). In addition, the antibacterial activities of apo-ovotransferrin plus citric acid against E. coli O157:H7 and L. monocytogenes were lost in the presence of 50 mM-NaHCO3. The number of viable cells in ovotransferrin plus 0.5% citrate solution in the presence of 50 mM-

NaHCO3 was similar to that of ovotransferrin with 50 mM-NaHCO3 in both pathogens. Also the present study indicated that ovotransferrin plus 25 mM-sodium citrate solution did not inhibit the growth of the two pathogens in contrast to ovotransferrin plus 0.5% citric acid (Tables 1 and 2). Bishop et al. (1976) report that as the molar ratio of citrate to apo-lactoferrin increased, the antibacterial capacity of apo-lactoferrin toward coli form bacteria decreased. Reiter et al. (1975) reported that the presence of citrate in bovine colostrum counteracted the antibacterial effectiveness of lactoferrin against E. coli. Lee et al. (2001, 2002) reported that sodium

125

citrate has strong antimicrobial effects against gram-positive bacteria, but has a weak antibacterial activity against gram-negative microorganisms. Also, Frost and Rosemberg (1973) reported that an excess of citrate annulled the antibacterial activity of ovotransferrin against E. coli O157:H7 because it possesses an iron transport system mediated by citrate. Futhermore, an anionic chelating by citrate occurred at high pH values. As one iron molecule binds two citrate molecules, the iron binding capacity by ovotransferrin is affected by the metal chelater (Phelpa and Antonini, 1975). The effect of sodium citrate on antibacterial effectiveness of ovotransferrin is dependent upon its concentration. The growth of E. coli was inhibited less in the presence of 50 mM-sodium citrate than 10 mM-sodium citrate (Valenti et al., 1983a). The iron release of transferrin is faster in the presence of citrate than formate or acetate because citrate interacts with decarbonated monoferric proteins with high affinity. Generally, iron binding capacity of ovotransferrin is known to decrease under acidic pH (Warner and Weber, 1953, Williams et al, 1978; Butterworth et al., 1975). As 0.5% citric acid reduces the pH of ovotransferrin solution to around 4 ~ 5, the antibacterial activity of ovotransferrin should be decreased due to its low iron binding capacity. However, this study indicated that 0.5% citric acid enhanced the antibacterial activity of ovotransferrin against E. coli O157:H7. Ovotransferrin plus 25 mM sodium citrate did not show any antibacterial

activity against the two pathogens, and NaHCO3 annulled the antimicrobial activity of ovotransferrin plus 0.5% citric acid. Considering these results, the antimicrobial action of ovotransferrin seemed to be enhanced by acidic pH conditions. Activated lactoferrin is reported to interact directly with bacterial membrane materials including lipopolysaccharides (Elass-Rochard et al., 1995; Appelmelk et al, 1994), glycosaminoglycans (Mann et al., 1994), and cell-type-specific receptors on epithelial and immune cells (Ward et al., 2002; Brock, 2002, Suzuki et al., 2002; Baveye et al., 1999). Also, specific binding of activated lactoferrin against outer membrane proteins of gram-negative bacteria can lead to inhibition of various cellular functions and deregulation of adhesion/ fimbrial synthesis and increase of the permeability of bacterial outer membrane, which increase the susceptibility of cell membrane to hydrophobic antibiotics and lysozyme, and facilitate the release of lipopolysaccharides molecules from the cell membrane (Naidu, 2002;

126

Ellison III et al., 1988, 1991). Like the mechanism of antimicrobial action of activated lactoferrin, the permeability of the outer membrane of gram negative bacteria might increase when ovotransferrin reacts directly on the bacterial membrane. So increased permeability may allow the protons formed by citric acid to enter easily inside of cell. In conclusion, the synergistic effect of citric acid on antibacterial activity of ovotransferrin against E. coli O157:H7 may be due to H+ formation. Also, iron binding capability of ovotransferrin might not be the major mechanism of antimicrobial activity of ovotransferrin because the antibacterial activity of ovotransferrin against E. coli O157:H7 was improved under acidic pH, which generally occurs with a decrease of iron binding capacity of ovotransferrin.

Antibacterial activity of Apo-, Fe-, and Zn-ovotransferrin To see if the antibacterial activity of ovotransferrin is related to its metal binding property, the antibacterial activity of ovotransferrin saturated with Fe2+ or Zn2+ against E. coli O157:H7 and L. monocytogenes was determined. As expected, natural apo-ovotransferrin

(OTF) showed no antibacterial activity, but ovotransferrin plus with 50 mM-NaHCO3 appeared to have a little antibacterial activity against E. coli O157:H7 and a bacteriostatic

effect against L. monocytogenes (Fig. 3). Ovotransferrin plus 100 mM NaHCO3 showed the strongest bacteriostatic activity against both pathogens. Ovotransferrin saturated with iron had little or no bactericidal activity against E. coli O157:H7 and L. monocytogenes (Fig 3A and 3B). Ovotransferrin bound with Zn2+ also did not show antibacterial activity against E. coli O157:H7 (Fig. 3A), but it had a similar degree of antibacterial activity as apo- ovotransferrin added with 50 mM sodium bicarbonate against L. monocytogenes (Fig. 3B). These results did not agree with those of Valenti et al. (1987) who reported that ovotransferrin saturated with Zn2+ had higher antibacterial activity against E. coli than apo- ovotransferrin. Several other reports also suggested that ovotransferrin bound with metals such as zinc and iron had antibacterial activity (Ibrahim, 1997; Valenti et al., 1985, 1987). However, the present study demonstrates that Fe-bound ovotransferrin had little antimicrobial effect and the antibacterial activity of Zn-bound ovotransferrin varied with microorganisms. Zn2+ significantly enhanced antibacterial action of ovotransferrin toward L. monocytogenes, but not E. coli O157:H7. That Zn-bound ovotransferrin shows bacteristatic

127

activity against L. monocytogenes seems to conflict with a conclusion that the major mechanism of antibacterial action of ovotransferrin is ascribed to iron deprivation or metal binding capacity. Like these results, Valenti et al. (1982) suggested that the antibacterial effectiveness of ovotransferrin is not due simply to the iron deprivation from media. Many recent reports indicated that antibacterial activity of ovotransferrin may be associated with direct interactions between ovotransferrin and microorganism, regardless of its iron binding properties (Arnold et al., 1977, 1980, 1981, Valenti et al., 1982, 1985). According to these reports, ovotransferrin saturated with iron still displayed antifungal activity toward Candida (Valenti et al., 1985). Ibrahim et al. (2000) suggested that the N-terminal domain of ovotransferrin showed more potent bactericidal activity against S. aureus than the C-terminal domain regardless of the degree of iron saturation. Considering that ovotransferrin has remarkable structural similarity to that of human lactoferrin, which can react directly with bacterial surface besides the role of iron deprivation, it may not be difficult to reach an assumption that the antimicrobial action of ovotransferrin could be intimately related to direct reactions on bacterial membrane rather than iron- deprivation property of ovotransferrin alone (Valenti et al., 1983b; 1987 Davis and Reeves, 2002). Ovotransferrin may permeate bacterial outer membrane, reach the inner membrane and lead to permeation of other ions, which dissipate electrical potential so that it can exert antibacterial action against gram-negative bacteria (Aguilera et al., 2003; Ibrahim, 1997).

Application of ovotransferrin to ham Based on the results obtained from in vitro tests, the ovotransferrin solutions plus either 100 mM-sodium bicarbonate or 0.5% citric acid were applied on commercial hams to determine their antibacterial activity. During storage at 4oC for 4 weeks, the number of viable E. coli O157:H7 cells decreased slowly (Table 3), whereas that of L. monocytogenes resulted in 1.5 ~ 2 log increase in this study (Table 4) even though Amezquita and Brashears (2002) report that cooked ham can inhibit the growth of L. monocytogenes. Through this study, E. coli O157:H7 was found to be more sensitive to low temperature than L. monocytogenes. Moreover, because commercial hams contain a variety of antibacterial substances such as nitrite, salt, low water activity, etc, the growth of E.coliO157:H7 sensitive to low temperature

128

was significantly prevented on the hams, but L. monocytogenes resistant to low temperature grew slowly under such a environment. There was little significant difference in the number of viable E. coli O157:H7 and L. monocytogenes cells between non-irradiated and irradiated hams during 4o C storage in all conditions (Table 3 and Table 4). Changes in the number of L.

monocytogenes survivors in ovotransferrin plus 0.5% citric acid were low (6.2 log10 CFU/ml to 6.9 log 10 CFU/ml) in irradiated sample, whereas that of non-irradiated hams was large o (6.4 log10 CFU/ml to 8.2 log10 CFU/ml) during storage at 4 C for 4 weeks (Table 4). Zhu et al. (2000) reported that during refrigerated storage for 28 days, L. monocytogenes in non- irradiated turkey breast rolls grew, and non-irradiated samples had greater natural microflora population than irradiated samples.

The antibacterial activity of ovotransferrin combined with 100 mM-NaHCO3 or 0.5% citric acid against E. coli O157:H7 inoculated on hams was not significantly different from that of control (Table 3). In contrast to the results of model system, little or no antilisterial effect of citric acid was found in both of non-irradiated and irradiated hams. The number of viable L. monocytogenes cells in ovotransferrin plus 100 mM-NaHCO3 treatment increased more than that of any other treatments during storage at 4o C for 4 weeks. The growth of L. monocytogenes in both non-irradiated and irradiated hams was promoted by sodium bicarbonate rather than suppressed (Table 4) as predicted by our model system results (Fig. 1). This indicated that the antibacterial activity of ovotransferrin in combination with 100 mM-NaHCO3 or 0.5% citric acid was annulled completely in commercial hams (Tables 3 and 4). It might be due to media or matrix compositions, difficulties in treatment distribution, or binding of ovotransferrin to matrix. Also, as hams have a variety of factors including divalent ions that affect antibacterial activity, the antibacterial activity of ovotransferrin on hams seems to be modulated by these factors. Ellison III et al. (1988) reported that the antibacterial effectiveness of lactoferrin was not achieved in food or complex media because of divalent cations such as Ca2+ and Mg2+. Shimazaki (2000) reported that divalent cations changed the tertiary structure of lactoferrin, which prevented antibacterial action of lactoferrin. Besides these ions, the stability of ovotransferrin in matrix is a critical factor in antibacterial activity of ovotransferrin. Valenti et al. (1982) reported that Sepharose-bound ovotransferrin showed higher antimicrobial

129

activity than free ovotransferrin because matrix-bound proteins have higher stability than soluble ovotransferrin. Direct application of lactoferrin to food systems reduces antibacterial activity of lactoferrin because active substances in complex media such meat or meat products neutralize the antibacterial effectiveness of lactoferrin. Also, the activated lactoferrin can be easily diffused into tissues and the antibacterial activity of lactoferrin on meat surface is difficult to maintain. Incorporation of lactoferrin into film or finding a film matrix appropriate to maintain the antibacterial activity by preventing diffusion of lactoferrin into meat has been discussed (http://www.newswise.com/articles/view/525381) and may be applied to ovotransferrin.

Conclusion The antibacterial activity of natural apo-ovotransferrin against E. coli O157:H7 and L. monocytogenes in model systems increased as the concentration of sodium bicarbonate

increased. NaHCO3 at 100 mM markedly increased antibacterial activity of ovotransferrin against E. coli O157:H7 and L. monocytogenes. Citric acid at 0.5% enhanced antibacterial activity of apo-ovotransferrin against E. coli O157:H7, but 0.5% citric acid alone also showed a strong bactericidal activity against L. monocytogenes. Addition of NaHCO3 annulled the strong antibacterial activity of ovotransferrin plus citric acid against the two pathogens. The antimicrobial activity of ovotransferrin was greatly enhanced by acidic pH conditions. Zn-bound ovotransferrin showed bacteriostatic activity against L. monocytogenes, but Fe-bound ovotransferrin had little or no antibacterial activity against E. coli O157:H7 and L. monocytogenes. According to these results obtained in this study, the iron binding capacity of ovotransferrin does not seem to be the major cause of antibacterial action of ovotransferrin.

Ovotransferrin plus 100 mM-NaHCO3 or 0.5% citric acid did not show antibacterial activity in commercial hams, indicating that there are many limitations in using ovotransferrin to control pathogens in meat or meat products. To overcome these problems, further studies on the mechanisms of antibacterial activity of ovotransferrin and various factors that can improve antibacterial activity of ovotransferrin are needed.

130

Reference Abdallah, F B., and EL Hage Chahine, J. M. 1998. Hen ovo-transferrin, interaction with bicarbonate and iron uptake. Eur. J. Biochem. 258: 1022-1031. Abola, J. E., Wood, M. C., Chwen, A., Abraham, D., and Pulsinelli, P. D. 1982. In: The Biochemistry and Physiology of Iron. Saltman, P. and Hegenauer, J., eds. (Amsterdam, The Netherlands: Elsevier Biomedical, Inc.). pp 27-34. Aguilera, O., Luis, M., Quirosb, L. M., and Fierro, J. F. 2003. Transferrins selectively cause ion efflux through bacterial and artificial membranes. FEBS Lett. 548: 5-10. Aisen, P. 1980. Iron transport and storage proteins. Ann. Rev. Biochem. 49: 357-393. Aisen, P. 1989. Physical biochemistry of the transferrins – Uptake 1984-1988. Phys. Bioinorg. Chem. Ser. 5, Iron Carriers and ion proteins (Loehr, T., ed.) pp 353-371. VCH, New York. Aisen, P., and Listowsky, I. 1980. Iron transport and storage proteins. Ann. Rev. Biochem. 49: 357-393. Ameaquita, A., Brashears, M. M. 2002. Competitive inhibition of Listeria monocytogenes in ready- to- eat meat products by lactic acid bacteria. J. Food Prot. 65: 316-325. Andeson, B. F., Baker, H. M., Norris, G. E., Rice, D. W. and Baker, E. N. 1989. Structure of Human lactoferrin-Crystallographic structure analysis and refinement at 2.8Ǻ resolution, J. Mol. Biol. 209: 711-734. Appelmelk, B. J., An, Y. Q., Geerts, M., Thijs, B.G., de Boer, H. A., MacLaren, D. M., de Graaff, J., and Nuijens, J. H. 1994. Lactoferrin is a lipid A-binding protein. Infect. Immunol. 62(6): 2628-32. Arnold, R. R. Brewer, M., and Gauthier, J. J, 1980. Bactericidal activity of human lactoferrin: sensitivity of a variety of microorganisms. Infect. Immunol. 28: 893-898. Arnold, R. R., Cole, M. F., and McGheem J. R. 1977. A bactericidal effect for human lactoferrin. Science 197: 263-265. Arnold, R. R., Russel, J. E., Chempion, W. J., and Gauthier, J. J. 1981. Bactericidal activity of human lactoferrin: influence of physical conditions and metabolic state of the target microorganism. Infect. Immune. 32: 655-660.

131

Bailey, C. T., Patch, M. G., and Carrion, C. J. 1988. Affinity labels for the anion-binding site in ovotransferrin. Biochem. 27(17): 6276-6282. Baron, F., Fauve, S., and Gautier, M. 2000. Egg Nutrition and Biotechnology, Sims, J. S., Nikkei, S., and Günter, W. Eds. CAB Int. pp 417-440. Baron, F., Gautier, M., and Brule, G. 1997. Factors involved in the inhibition of Salmonella enteritidis in liquid egg white. J. Food Prot. 60: 1318-1323. Bauer, A. W., Kirby, W. M. M., Sherris, J. C., and Turck, M. 1966. Antibiotic susceptibility

testing by a standardized single disk method. Amer. J. Clin. Pathol. 45:493-496. Baveye, S., Leas, E., Maurer, J., Spike, G., and Lybrand, D. 1999. Lactoferrin: a multifunctional glycoprotein involved in the modulation of the inflammatory process. Clint. Chem. Lab. Med. 37: 281-286. Bellounis, L., Acadian, R., and El Hage Chahine, J. M. 1996. Apo-transferrin proton dissociation and interactions with bicarbonate in neutral media, J. Phys. Org. Chem. 9: 111-118. Bishop, J. G., Schumacher, F.L., Ferguson, L. C., and Smith, K. L. 1976. In vitro growth inhibition of Mastitis-causing coli-form bacteria by bovine apo-lactoferrin and reversal of inhibition by citrate and high concentrations of apo-lactoferrin. Infect. Immune. 14(4): 911-918. Brock, J. H. 1985. The Transferrins. In: Harrison, P.(end). Metalloproteinase II. MacMillan Press, London. pp 183-262. Brock, J. H. 2002. The physiology of lactoferrin. Biochem Cell Biol. 80: 1-6. Butterworth, R. M., Gibson, J. F., and Williams, J. 1975. Electron-paramagnetic-resonance spectroscopy of iron-binding fragments of hen ovotransferrins. Biochem. J. 149: 559-563. Corral, L. G., Post, L. S., and Montville, T. J. 1988. Antimicrobial activity of sodium bicarbonate. J. Food Sci. 53(3): 981-982. Czuprynski, C. J and Faith, N. G. 2002. Sodium bicarbonate enhances the severity of infection in neutropenic mice orally inoculated with Listeria monocytogenes EGD. Clinic Diagno Lab. Immunol. 9(2): 477-481.

132

Davis, C and Reeves, R. 2002. High value opportunities from the chicken egg. A report for the Rural industries Research and Development corporation. Rural industries Research and Development Corporation Publication. Eaton, S. S, Dubach, J., Eaton, G. R., Thurman, G., and Ambruso, D. R. 1990. Electron spin echo envelope modulation evidence for carbonate binding to iron(III) and copper(II) transferrin and lactoferrin. J. Biol. Chem. 265(13): 7138-7141. Elass-Rochard, E., Roseanu, A., Legrand, D., Trif, M., Salmon, V., Motas, C., Montreuil, J., and Spik, G. 1995. Lactoferrin-lipopolysaccharide interaction: involvement of the 28-34 loop region of human lactoferrin in the high-affinity binding to Escherichia coli 055B5 lipopolysaccharide. Biochem. J. 312: 839–845. Ellison III, R. T., Giehl, T. J., and Laforce, F. M. 1988. Damage of the outer membrane of enteric gram-negative bacteria by lactoferrin and transferrin. Infect. Immunol. 56(11): 2774-2781 Ellison III, R. T. and Giehl T. J. 1991. Killing of Gram-negative bacteria by lactoferrin and lysozyme. J. Clin. Invest. 88: 1080-1091. Frost, G. E. and Rosemberg, H. 1973. The citrate-dependent iron transport system in Escherichia coli K-12. Biochim. Biophys. Acta 330: 90-101. Griffiths, E. and Humphreys, J. 1977. Bacteriostatic effect of human milk and bovine colostrum on Escherichia coli: Importance of Bicarbonate. Infec. Immunol. 15(2): 396- 401. http://www.newswise.com/articles/view/525381. 2006. Lactoferrin deals another blow to pathogen. Source: University of Arkansas, Food Safety Consortium. Ibrahim, H. R. 1997. Insights into the structure-function relationship of ovalbumin, ovotransferrin and lysozyme. In: Hen eggs: Their Basic and Applied Science. Yamamoto, T., Juneja, I.R., Hatta, H., and Kim, M. (Eds). CRC Press. pp 37-56. Ibrahim, H. R., Sugimoto, Y., and Aoki, T. 2000. Ovtotransferrin antimicrobial peptide (OATP-92) kills bacteria through a membrane damage mechanism. Biochim. Biophys. Acta 1523: 196-205. Kuehl, R. O. 2000. Design of experiments; Statistical principles of research design and analysis. 2nd Ed. New York: Duxbury Press.

133

Kurokawa, H., Dewan, J. C., Mikami, B., and Sacchettini, J. C. 1999. Crystal structure of apo-hen ovotransferrin. J. Biol. Chem. 274: 28445-28452. Lee, Y. L., Cesario, T. Owens, J., Shanbrom, E., and Thrupp, L. D. 2002. Antibacterial activity of citrate and acetate. Nutrition 18: 665-666. Lee, Y. L., Trupp, L., Owens, J., Cesario, T., and Shanbrom, E. 2001. Bactericidal activity of citrate against gram-positive cocci. Lett. Appl. Microbiol. 33: 349-351. Mann, D. M., Romm, E., and Migliorini, M. 1994, Delineation of the glycosaminoglycan- binding site in the human inflammatory response protein lactoferrin. J. Biol. Chem. 269: 23661-23667. Naidu, A. S. 2002. Activated Lactoferrin - a new approach to meat safety. Food Technol. 56(3): 40-45. Pakdaman, R. and El hage Chahine, J. M. 1996. A mechanism for iron uptake by transferrin. Eur. J. Biochem. 236: 922-931. Pakdaman, R. and El Hage Chahine, J. M. 1997. Transferrin- the interaction of lactoferrin with hydrogen carbonate. Eur. J. Biochem. 249: 149-155. Pakdaman, R., Petitjean, M. and El Hage Chahine, J. M. 1998. Transferrins - a mechanism for iron uptake by lactoferrin. Eur. J. Biochem. 254:144-153. Peaker, M. and Linzell. J. L. 1975. Citrate in milk: a harbinger of lactogenesis. Nature 253: 464. Phelps, C. F. and Antonini, E. 1975. A study of kinetics of iron and copper binding to hen ovotransferrin. Biochem. J. 147: 385-391. Rawas, A., Muirhead, H., and Williams, J. 1996. Structure of diferric duct ovo-transferrin at 2.35 Ǻ resolution, Acta Crystallogr. D 2: 631-640. Reiter, B., Brock, J. H., and Steel, E. D. 1975. Inhibition of Escherichia coli by bovine colostrum and post colostral milk. II. The bacteriostatic effect of lactoferrin on a serum susceptible and serum resistant strain of E. coli. Immunol. 28: 83-95. Schalabach, M. R. and Bates, G. W. 1975. The synergistic binding of anions and Fe3+ by transferrin. J. Biol. Chem. 250: 2182-2188.

134

Shade, A. L. and Caroline, L. 1944. Raw hen egg white and the role of iron in growth inhibition of Shigella dysenteriae, Staphylococcus aureus, Escherichia coli and Saccharomyces cerevisiae. Sci. 100: 14-15. Shimazaki, K. 2000. Lactoferrin: A marvelous protein in milk. Anim. Sci. J. 71: 329-347. Stadelman, W. J. and Cotterill, O. J. 1986. Egg Science and Technology. 3rd Ed. Avi Pub. Co., Westport CT. Suzuki, Y. A., Shin, K., and Lonnerdal, B. 2002. Characterization of mammalian receptors for lactoferrin. Biochem. Cell Biol. 80:75-80 Valenti, P., Antonin, G., Von Hunolstein, C., Visca, P., Orsi, N., and Antonini, E. 1983a. Studies of the animicrobial activity of ovotransferrin. Inter. J. Tiss. Reac. 1: 97-105. Valenti, P., Antonini, G., Fanelli, M. R. R., and Orsi, N., and Antonini, E. 1982. Antibacterial activity of matrix-bound ovotransferrin. Antimicrobial Agents Chemother. 21(5): 840- 841. Valenti, P., De Stasio, A., Mastromarino, P., Seganti, L., Sinibaldi, L., and Orsi, N. 1981. Influence of bicarbonate and citrate on the bacteriostatic action of ovotransferrin towards Staphylococci. FEMS Microbiol. Lett. 10:77-79. Valenti, P., Stasio, A. D., Seganti, L., Mastromarino, P., Sinibaldi, L., and Orsi, N. 1980.

Capacity of Staphylococci to grow in the presence of ovotransferrin or CrCl3 as a character of potential pathogenicity. J. Clinic. Microbiol. 11(5): 445-447. Valenti, P., Visca, P., Antonini, G., and Orsi, N. 1985 Antifungal activity ovotransferrrin towards genus Candida. Mycophathologia 89: 169-175. Valenti, P., Visca, P., Antonini, G., and Orsi, N., 1983b. Studies on the interaction between microorganisms and ovotransferrin. In Structure and Function of Iron Storage and Storage Proteins. Urishizaki, I., Aiesen, P., Listowsky, I., and Drysdale, J. W. (Ed). Elsevier, New York. Valenti, P., Visca, P., Antonini, G., Orsi, N., Antonini, E. 1987. The effect of saturation with Zn+2 and other metal ions on the antibacterial activity of ovotransferrin. Med. Microbiol. Immunol. 176: 123-130. Ward, P. P., Uribe-Luna, S., and Conneely, O. M. 2002. Lactoferrin and host defense. Biochem. Cell Biol. 80: 95-102.

135

Warner, R. C. and Weber, I. 1953. The metal combining properties of conalbumin. J. Am. Chem. Soc. 75: 5094-5101. Williams, J., Evans R. W., and Moreton, K. 1978. The iron-binding properties of hen ovotransferrin. Biochem. J. 173: 535-542. Woodworth, R. C., Virkaitis, L. M., Woodbury, R. G., Fava, R. A. 1975. In: Proteins of Ion Storage and Transfort in Bicochemistry and Medicine. Ed. R. R. Cichton. pp 39-50. Amsterdam: North-Holland. pp 454. Workman, E. F., Graham, G., and Bates, G. W. 1975. The effect of serum and experimental variables on the transferrin and reticulocyte interaction. Biochim. Biophys. Acta 399: 254-264. Zhu, M. J, Mendonca, A., Ismail, H. A, and Ahn, D. U. 2000. Effects of Irradiation on Survival and Growth of Listeria monocytogenes and Natural Microflora in Vacuum Packaged Turkey Hams and Breast Rolls. Ph.D. thesis, Iowa State University, Ames, IA. Zuccola, H. A. 1993. The Crystal Structure of Monoferric Human Serum Transferrin. Ph.D. thesis, Georgia Institute of Technology.

136

Table 1. Influences of citric acid and citrate on antibacterial activity of ovtransferrin against 105 CFU/ml of E. coli O157:H7.

Treatments Storage period (days) 1 2 4 8

------Number of viable cell ( log10 CFU/ml) ------Citric acid 7.90c ± 0.65 6.88c ± 0.43 5.90b ± 1.00 2.23 c ± 0.43 OTF + Citric acid 4.71d ± 0.09 3.92d ± 0.10 2.04c ± 0.00 1.00 d ± 0.00 ab b a a OTF + Citric acid +NaHCO3 8.64 ± 0.10 8.41 ± 0.36 8.97 ± 0.19 8.74 ± 0.15 OTF + Na-Citrate 9.24a ± 0.06 9.35a ± 0.08 8.97a ± 0.05 8.22b ± 0.07 bc ab a b OTF + NaHCO3 8.40 ± 0.10 8.64 ± 0.39 8.69 ± 0.18 7.75 ± 0.03 a-c Values with different letters within a column with the same storage day are significantly different (p < 0.05, n=4)

This study used 0.5% citric acid, 25 mM sodium citrate, and 50 mM-NaHCO3. OTF: Apo-ovotransferrin (20 mg/ml).

137

Table 2. Effect of citric acid and citrate on antibacterial activity of ovtransferrin against 104 CFU/ml of L. monocytogenes.

Treatments Storage period (days) 1 2 4 8 Citric acid 3.52b ± 0.32 ** 2.78b ± 1.14 1.52b ± 0.96 1.00b ± 0.0 OTF * + Citric acid 3.52b ± 0.83 2.86b ± 0.99 2.05b ± 1.67 1.91b ± 1.78 a a a a OTF + Citric acid +NaHCO3 8.86 ± 0.14 8.60 ± 0.41 8.14 ± 0.29 7.67 ± 0.60 OTF + Na-Citrate 9.23a ± 0.11 8.62a ± 0.17 8.57a ± 0.23 8.43a ± 0.31 a a a a OTF + NaHCO3 8.60 ± 0.19 7.74 ± 0.01 8.14 ± 0.07 7.13 ± 0.04 a-b P Values with different letters within a column with the same storage day are significantly different (p < 0.05, n=4)

This study used 0.5% citric acid, 25 mM sodium citrate, and 50 mM-NaHCO3. OTF: Apo-ovotransferrin (15 mg/ml).

Table 3. Antibacterial activity of ovotransferrin solutions (20 mg/ml) plus either 100 mM-NaHCO3 or 0.5% citric acid on non-irradiated and irradiated vacuum-packaged hams inoculated with E. coli O157:H7 during storage at 4o C

Sample Treatments Storage period (days) 0 5 10 15 22 29

------Number of viable cell (log10 CFU/ml) ------

NR1 Control * 6.1a ± 0.21 5.5a ± 0.22 5.3ab ± 0.12 5.2a ± 0.10 5.0b ± 0.08 5.5a ± 0.01 a a b a b c OTF +NaHCO3 6.0 ± 0.11 5.7 ± 0.50 5.2 ± 0.01 5.2 ± 0.17 5.0 ± 0.10 5.0 ± 0.01 a a b a b c

OTF + Citric acid 5.9 ± 0.10 5.7 ± 0.04 5.2 ± 0.05 5.1 ± 0.42 5.0 ± 0.07 5.1 ± 0.11 138

IR2 Control 6.0 a ± 0.0 5.5a ± 0.06 5.5a ± 0.05 4.9a ± 0.32 5.3a ± 0.17 5.4ab ± 0.01 a a a a b ab OTF + NaHCO3 6.1 ± 0.19 5.6 ± 0.08 5.5 ± 0.06 4.9 ± 0.19 5.0 ± 0.07 5.3 ± 0.04 OTF + Citric acid 6.0a ± 0.0 5.7a ± 0.05 5.5a ± 0.08 5.0a ± 0.40 5.0b ± 0.02 5.3ab ± 0.10

a-bValues with different letters within a column with the same storage day are significantly different (p < 0.05, n = 4). *Only E. coli O157:H7 inoculation OTF: Ovotransferrin (20 mg/ml); NR1: non-irradiated ham; IR2 : irradiated ham This study used 0.5% citric acid and 100 mM sodium bicarbonate.

Table 4. Antibacterial activity of ovotransferrin solution (20 mg/ml) plus either 100mM-sodium carbonate or 0.5% citric acid on non- irradiated and irradiated vacuum-packaged hams inoculated with L. monocytogenes during storage at 4o C.

Sample Treatments Storage period (days) 0 5 10 15 22 29

------Number of viable cell (log10 CFU/ml) ------NR1 Control* 6.5a ± 0.15** 6.3ab ± 0.01 6.5 b ± 0.03 6.9b ± 0.10 7.5ab ± 0.10 8.2a ± 0.09 *** ab a a a a a OTF +NaHCO3 6.4 ± 0.11 6.4 ± 0.04 7.7 ± 0.09 7.6 ± 0.09 8.3 ± 0.21 8.4 ± 0.05 OTF + Citrate 6.4ab ± 0.14 6.5a ± 0.04 6.6b ± 0.14 7.1ab ± 0.03 7.3ab ± 0.12 8.2a ± 0.10

IR2 Control 6.4ab ± 0.04 6.2b ± 0.08 6.4b ± 0.13 6.6b ± 0.25 7.0b ± 0.08 8.0a ± 0.30 139 b a a a ab a OTF + NaHC03 6.2 ± 0.14 6.5 ± 0.01 7.5 ± 0.23 7.7 ± 0.45 7.7 ± 0.63 8.3 ± 0.47 OTF + Citrate 6.2ab ± 0.18 6.2b ± 0.03 6.4b ± 0.07 6.5b ± 0.22 6.9b ± 0.53 6.9 b ± 0.57 a-bValues with different letters within a column with the same storage day are significantly different (p< 0.05, n=4). *Only L. monocytogenes inoculation. OTF: Ovotransferrin (20 mg/ml); NR1: non-irradiated ham; IR2 : irradiated ham This study used 0.5% citric acid and 100 mM sodium bicarbonate.

140

Fig 1. The effect of sodium bicarbonate on antibacterial activity of ovotransferrin (20mg/ml) against E. coli O157:H7 (A : 103 CFU/ml inoculation) and L. monocytogenes (B: 104 CFU/ml inoculation) during 35o C incubation. Control was inoculated with only E. coli O157:H7. OTF: 20 mg/ml apo-ovotransferrin. A

1.4 1.2 1.0 0.8 0.6 0.4

Turbidity (at 620nm) 0.2 0.0 0246810 Tim e (days)

Control OTF+ 25mM NaHCO3 OTF+ 50mM-NaHCO3 OTF+100mM-NaHCO3 25mM-NaHCO3 50mM-NaHCO3 100mM-NaHCO3 OTF

0.8

0.6

0.4

Turbidity (at 620nm) (at Turbidity 0.2

0.0 0246810 Time (days)

Control OTF+ 25mM NaHCO3 OTF+ 50mM-NaHCO3 OTF+100mM-NaHCO3 25mM-NaHCO3 50mM-NaHCO3 100mM-NaHCO3 OTF

141

Fig 2. The influence of citric acid on antibacterial activity of ovotransferrin against 105 CFU/ml of E. coli O157: H7 (A) and 104 CFU/ml of L. monocytogenes (B) after 2 days of incubation at 35o C. The control was inoculated with only E. coli O157:H7 or L. monocytogenes. OTF in A: 20 mg/ml apo-ovotransferrin, OTF in B: 15 mg/ml apo- ovotransferrin. Different letters on bars represent significantly different groups (p < 0.05, n = 4). A

10.0 aab ab 9.0 b

8.0 c 7.0

6.0

5.0 CFU/ml d 10 4.0 Log Log 3.0

2.0

1.0

0.0 Control OTF 0.25% Citric acid OTF+ 0.25% Citric acid 0.5% Citric acid OTF + 0.5% Citric acid

10.0 a a 9.0 a a 8.0

7.0

6.0

CFU/ml 5.0

10 b 4.0 b Log 3.0

2.0

1.0

0.0 Control OTF 0.25% Citric acid OTF+ 0.25% Citric acid 0.5% Citric acid OTF + 0.5% Citric acid

142

Fig 3. Antibacterial activity of OTF, Fe-OTF, and Zn-OTF against E. coli O157:H7 (104 CFU/ml, A) after 1day of incubation at 35oC and L. monocytogenes (105 CFU/ml, B) after 2 days of incubation at 35oC. Control: only E. coli O157:H7 or L. monocytogenes. OTF: 20 mg/ml of apo-ovotransferrin, OTF 50: OTF + 50 mM-NaHCO3, OTF 100: OTF + 100 mM-

NaHCO3, Fe-OTF: Fe-bound ovotransferrin (20 mg/ml), Zn-OTF : Zn-bound ovotransferrin (20 mg/ml). Different letters on bars represent significantly different groups (p < 0.05, n = 4). A

12.0 a ab ab a 10.0 b

8.0 c 6.0 CFU /ml 10 4.0 Log

2.0

0.0 1 Control OTF OTF 50 OTF 100 Fe-OTF Zn-OTF

10.0 9.0 a a ab 8.0 bc bc 7.0 c 6.0 5.0 CFU /ml

10 4.0 3.0 Log 2.0 1.0 0.0 1 Control OTF OTF 50 OTF 100 Fe-OTF Zn-OTF

143

CHAPTER 6. EDTA AND LYSOZYME IMPROVES ANTIMICROBIAL ACTIVITIES OF OVOTRANSFERRIN AGAINST ESCHERICHIA COLI O157:H7

A paper which will be submitted to Journal of Food Protection

K. Y. Ko1, D. U. Ahn2 and A. Mendonca

Abstract The aim of present study was to evaluate the effect of EDTA or/and lysozyme on

antibacterial activity of ovotransferrin solution (20 mg/ml) including 100 mM-NaHCO3 (OS) against a cocktail of 5 strains of E. coli O157:H7. First, disc tests were performed to find the appropriate concentrations of EDTA and lysozyme that affect the antibacterial activity of OS. Also, the turbidity of OS solutions inoculated with E. coli O157:H7 and contained either EDTA (OSE) or lysozyme (OSL) and the viability of E. coli O157:H7 in the solutions were investigated. Furthermore, OS combined with EDTA or/and lysozyme (OSE, OSL, OSEL) were applied on irradiated pork chops and commercial hams to determine if the solutions have antibacterial activity on meat products. In addition, the effect of initial cell population on the antibacterial activity of ovotransferrin and EDTA or lysozyme combinations was determined. EDTA at 1 mg/ml plus OS (OSE) induced 3 ~ 4 log reduction in viable E. coli O157:H7 cells in BHI broth media, and 1 mg/ml lysozyme plus OS (OSL) resulted in around 0.5 ~ 1.0 log reduction during 35o C incubation for 36 hr. However, ovotransferrin plus EDTA or/and lysozyme did not show significant antibacterial effect on pork chops and hams during storage at 10o C. Also, initial cell number in media did not affect the antibacterial activity of OSE or OSEL against E. coli O157:H7. This study demonstrates that even though

combinations of ovotransferrin, NaHCO3, and EDTA (OSE) have potential to control E. coli O157:H7. Also further studies related to the mechanisms of antibacterial action of ovotransferrin in meat products are needed for food application of ovotransferrin.

1 Primary researcher and author

144

2 Author of correspondence.

Introduction Natural antimicrobials that are environmentally friendly, medically acceptable, and highly effective and economical to manufacture have gained attention due to consumers’ desire for natural food products (Payne et al., 1994). Ovotransferrin, one of the main iron- binding glycoproteins present in egg white, transports and scavenges Fe (III) in eggs of poultry (Kurokawa et al., 1995). It is a major contributor to the egg’s defense against microbial infection and rotting. Many studies have described the antibacterial properties of ovotransferrin against various microorganisms including Escherichia coli (Shade and Carolin, 1944; Valenti et al., 1983), Pseudomonas spp., Streptococcus mutans (Valenti et al., 1983), Staphylococcus aureus, Bacillus cereus (Ibrahim, 1997), Salmonella enteritidis (Baron et al., 1997; 2000), and Candida (Valenti et al., 1985). However, the antimicrobial mechanisms of ovotransferrin have not been fully elucidated yet. Initially, the iron binding capability of ovotransferrin, which limits the availability of iron required for microbial growth, was the major antibacterial mechanism of ovotransferrin. However, subsequent studies have suggested that the antimicrobial action of ovotransferrin could be intimately related to direct interactions of ovotransferrin with bacterial surface, which result in damage of outer membranes of microorganisms or destroy microbial function (Valenti et al., 1983, 1985; Arnold et al., 1981, Ibrahim, 1997; Ihbrahim et al., 2000). E. coli O157:H7 possesses specific cell-surface appendages such as fimbria or colonization factor antigens that facilitate their adhesion to host-tissue-matrix components such as fibronectin, collagens, elastin and mucin so that it can tightly attach to beef tissues (Naidu, 2002). Due to these attributes, illness caused by E. coli O157:H7 are frequently linked to cattle and their products, undercooked ground beef, and raw milk (Firstenberg-Eden, 1997; Weeranta and Doyle, 1991 Bell et al., 1994). E. coli O157:H7 is a shiga toxin- producing strain that some products contaminated with the microorganism sometimes cause bloody diarrhea in humans and cause more serious complications such as hemorrhagic colitis and hemolytic uremic syndromes (Griffin and Tauxe, 1991). The combinations of antimicrobials with different mechanisms can increase the antibacterial effectiveness. Ethylenediaminetetraacetic acid (EDTA) has been used in a

145

number of food products as a chelating agent to prevent oxidation and deteriorative reactions, which are catalyzed by metal ions. Also, it is known to enhance the effectiveness of antimicrobials and antibiotics, especially against gram-negative bacteria. For instance, EDTA promotes the antimicrobial activity of nisin, lysozyme, and monolaurin against gram- negative microorganisms (Hughey and Jonhson, 1987; Stevens et al., 1991; Razavi-Rohani and Griffiths, 1994). Even though the mechanism of antibacterial effectiveness by EDTA is not fully understood, generally chelating property to divalent cations like Ca2+ or Mg2+ has been considered as a major cause of its antibacterial activity (Shelef and Seiter, 1993). EDTA inhibits the growth of microorganisms by depriving Ca2+, Mg2+, or Fe2+, which are essential factors for microbial growth (Banin et al., 2006). Also, it destabilizes outer membrane of gram-negative bacteria by chelating Ca2+ and Mg2+ salts that function as bridges between lipopolysaccharides (LPS) of microbial outer membrane, and thus results in the release of LPS from gram-negative bacteria (Shelef and Seiter, 1993; Vaara, 1992, Banin et al., 2006). Also, chelating divalent cations facilitate the detachment of cells from biofilm and enhance the killing of biofilm-producing microorganisms by depriving Mg2+ associated with lipopolysaccharides (Banin et al., 2006). Lysozyme, a single peptide protein with 14.6 kDa, catalyzes the hydrolysis of (1-4) glycosidic linkages between N-acetylmuramic acid and N-acetylglucosamine in cell wall peptidoglycan (Proctor and Cunningham, 1988). When peptidoglycan, the major component of cell wall in both gram-positive and gram-negative bacteria, is hydrolyzed by lysozyme, the structural integrity of cell wall can be damaged (Proctor and Cunningham, 1988). Therefore, the antimicrobial activity of lysozyme is ascribed primarily to enzymatic lysis of peptidoglycan in microbial cell wall (Branen and Davidson, 2004). As the cell walls of gram- negative bacteria are protected by an outer membrane, gram-negative microorganisms are relatively resistant to antimicrobial activity of lysozyme, and thus the application of lysozyme in foods has been limited (Gill and Holley, 2003; Proctor and Cunningham, 1988). However, lysozyme gained a considerable interest for use in food systems because it is a natural enzyme produced by many animals and its activity targets a specific cellular structure of microorganism (Proctor and Cunningham, 1988).

146

Ovotransferrin alone has little antibacterial activity against E. coli O157:H7 in previous study (unpublished data). Even though 100 mM-NaHCO3 enhanced the

antibacterial activity of ovotransferrin, the ovotransferrin combining 100 mM-NaHCO3 (OS) was bacteriostatic, not bactericidal against the pathogens. Therefore, it is necessary to enhance antimicrobial activity of ovotransferrin for food application. Ovotransferrin and lysozyme are natural antimicrobials and EDTA is a chelating agent used in a wide variety of food products (Hansen et al., 2001). Therefore the use of ovotransferrin combined with EDTA or/and lysozyme on the surface of meat or meat products is an interesting approach because all those components are generally regarded as safe (GRAS). The aim of the present study was to enhance the antimicrobial activity of OS against 5 strains of E. coli O157:H7 by

combining EDTA and lysozyme. Also, the effects of NaHCO3, EDTA and lysozyme combinations on the antimicrobial activity of ovotransferrin against E. coli O157:H7 in commercial hams and irradiated pork chops were also tested.

Material and Methods

Ovotransferrin Apo-ovotransferrin (iron-free) used in this study was prepared by the method of Ko and Ahn (2007, unpublished). After iron saturation of 2x-diluted egg white solution, holo- ovotransferrin (iron-saturated) were separated using ethanol. Iron was removed from the ® holo-ovotransferrin using AG 1-X2 resin (chloride form, Bio-Rad, California, U.S.A), and then freeze-dried. The dried apo-ovotransferrin was dissolved in distilled water, its pH adjusted to 7.4, and 0.15 M NaCl was added. The purity of ovotransferrin used in this study was around 80% and the residual iron in prepared apo-ovotransferrin solution was less than 0.5 ppm. Prior to microbial study, the ovotransferrin solution was sterilized by filtering through a 0.22 μm syringe filter (Whatman Inc. Florbam Park, NJ. USA).

Bacterial strains Five different strains of E. coli O157:H7 (ATCC 43890, C467, FRIK 125, ATCC 43895, and 93-062) were used. Prior to inoculation, each strain was individually cultured in

147

brain heart infusion broth (BHI, Remel Inc., Lenexa, KS, USA) for 24 hours at 35oC and harvested twice in order to activate the strain appropriately. An aliquot (5 ml) of each culture solution was transferred to a sterilized centrifuge bottle and centrifuged at 100,000 x g for 10 min at 4oC. Pellet was collected, washed with saline, and then resuspended in 25 ml saline. An inoculation cocktail was prepared by mixing equal volumes of each cell suspension to have approximately the same population of five strains of E. coli O157:H7.

Formation of clear zone Prior to performing viability and turbidity tests to identify the antibacterial activity, disc diffusion method (Bauer et al, 1966) was used to determine the concentrations of EDTA (disodium salt, Fisher Scientific, Fairlawn, NJ, USA) and lysozyme (from egg white, Sigma- Aldrich Inc., St. Louis, USA) capable of promoting antibacterial capability of ovotransferrin. An E. coli O157:H7 cocktail containing equal number of each strain was serially diluted and then 0.1 ml of cell suspension corresponding to 105 or 106 number was spread homogenously on the BHI media agar plate. The BHI agar plates inoculated with E. coli O157:H7 were dried for 30 ~ 60 minutes in an incubator at 35oC. EDTA (1, 0, 1.5, 2.0, or 2.5 mg/ml) or lysozyme (1.0, 1.5, 2.0, or 2.5 mg/ml) was added to ovotransferrin solution or ovotransferrin

plus 100 mM-NaHCO3 (OS) solution. All solutions prepared were sterilized using 0.22 μm syringe filters (Whatman Inc. Florbam Park, NJ. USA) prior to distribution on plates. Ovotransferrin solutions (120 μl) containing different concentrations of EDTA or lysozyme were dispensed on a sterilized glass microfiber filter (2.1 cm diameter, Whatman Inc. Florbam Park, NJ. USA) placed on the dried BHI agar media. Three replications were prepared. The BHI agar plates were incubated at 35o C for 24 ~ 48 hrs and then the antibacterial activity of each ovotransferrin solution against E. coli O157:H7 was estimated by the size and clarity of clear zones.

Turbidity test The antimicrobial capacity of ovotransferrin solutions combined with EDTA or lysozyme against E. coli O157:H7 was analyzed by measuring the turbidity of solution after incubation in a BHI broth culture at 35oC. Disc test indicated that the concentrations of

148

EDTA and lysozyme showing antibacterial activity against E. coli O157:H7 when combined with ovotransferrin solutions were > 2.0 mg/ml and > 1.0 mg/ml, respectively (Table 1). After 2 ml of 40 mg/ml ovotransferrin solution was added to 2 ml of 2x-strength BHI broth, stock solutions of EDTA or lysozyme was added to the broth culture media including ovotransferrin or OS and then the final concentration of EDTA was adjusted to 2.0 or 2.5 mg/ml, and lysozyme to 1.0, 1.5, or 2.0 mg/ml. Also, 40 μl of cell suspension containing 106 ~ 107 cells of activated E. coli O157:H7 cocktail was inoculated to the prepared solution to make the initial population of E. coli O157:H7 104 ~ 105 CFU per ml. The antibacterial activity of EDTA or lysozyme solution was determined. After inoculation, the samples were incubated at 35o C for 36 hrs. Multiplication of E. coli O157:H7 in each solution was evaluated by measuring the turbidity of solution with a spectrophotometer at 620 nm every 3 - 6 hrs.

Viability test The antibacterial activity of ovotransferrin plus either EDTA or lysozyme was investigated using a viability test. First, 2 ml of ovotransferrin solution (40 mg/ml) was added to 2 ml of 2x-concentrated BHI broth media containing 100 mM-NaHCO3. Stock solution of

either EDTA or lysozyme was added to the ovotransferrin plus 100 mM NaHCO3 solution so that each solution contained 2 mg/ml of EDTA or 1 mg/ml lysozyme. Also, 40 μl of cell suspension containing 106 ~ 107 cells of activated E. coli O157:H7 was inoculated to the prepared solution to make 104 ~ 105 CFU per ml in the initial solution. EDTA or lysozyme alone was prepared as a control group. After inoculation, the prepared culture solutions were incubated at 35oC for 36 hrs. The number of viable cells was analyzed by spread plate of each culture solution (0.1 ml) after diluting (1:10) with 0.1% sterile peptone water (Difco Laboratories, Detroit, MI). The samples were incubated at 35oC for 24 ~ 36 hrs. The number of survivors on BHI agar plates was counted as colony-forming units per ml (CFU/ml) of sample.

Cell number effect

149

This study was performed to know if the population of cell affects the antibacterial activity of ovotransferrin solutions combined with various solutions. The BHI broth media containing ovotransferrin was prepared with the same method described above. OS plus either 2 mg/ml EDTA or 2 mg/ml EDTA and 1 mg/ml lysozyme were added to prepared BHI broth media. The BHI broth culture containing EDTA (2 mg/ml) and lysozyme (1 mg/ml) without added ovotransferrin was prepared to identify whether the solution itself has antibacterial activity against E. coli O157:H7. Also, 40 μl of cell suspension containing 106, 107, or 108 cells of activated E. coli O157:H7 was inoculated to 4 ml BHI broth media to make the number of E. coli O157:H7 104, 105 or 106 CFU per ml. After the prepared culture solutions were incubated at 35oC for 24 ~ 36 hrs, the viability of each treatment was measured using the same method described above.

Application to pork chop Fresh pork chops from different animals were purchased from the Meat Laboratory at Iowa State University and sliced to 1 cm-thick pieces (7 cm x 5 cm). Sliced pork chops were vacuum packaged (-1 bar of vacuum with 10 second dwell time) in low oxygen-permeable o bags (nylon/polyethylene, 9.3 ml O2/m2/24hr at 0 C; Koch, Kansas City, MO). After vacuum packaging, the sliced pork chops were irradiated at 0 or 5 kGy using a Linear Accelerator (Circe IIIR; Thomason CSF Linac, Saint-Aubin France) to kill bacteria, which may exist in the pork chop. After irradiation, the pork chops were frozen and defrosted completely prior to use for the microbial study. Each package was aseptically opened using an alcohol-sterilized scissors. E. coli O157:H7 cocktail stock suspension (0.2 ml) was inoculated on the surface of each pork chop to make the initial inoculated number 104 CFU per gram meat and then the pork chop samples were randomly divided into 5 groups. Inoculated strains were spread manually for 30 second to distribute the inoculum evenly. Lysozyme or/and EDTA were added to OS solution containing 0.15 M NaCl. The treatments are as follow: 20 mg/ml ovotransferrin containing 100 mM NaHCO3 (1OS) plus 2 mg/ml EDTA (1OSE), 1OS plus 2 mg/ml EDTA and 1 mg/ml lysozyme (1OSEL), 30 mg/ml ovotransferrin (2OS) plus 2 mg/ml EDTA (2OSE), and 2OS plus 2 mg/ml EDTA and 1 mg/ml lysozyme (2OSEL). One ml of each ovotransferrin solution was distributed on the

150 same surface of pork chops and spread manually for 30 ~ 60 sec to distribute the solution homogenously. Control group was prepared by inoculating E. coli O157:H7 to pork chops. After adding ovotransferrin to pork chops, the samples were stored in a 10oC incubator. Viable cells on the pork chops were analyzed by sprea plate method after 0, 2, 4, 8, and 12 days of storage: 30 ~ 50 ml of sterile 0.1% peptone water solution was added to each bag followed by pummeling for 1min in a stomacher at normal speed. After serial dilution of the sample with 0.1% peptone water, 0.1 ml of diluted sample was spread homogenously on the MacConkey agar (Difco Laboratories, Detroit, MI) plate in duplicate. Survivors were enumerated as CFU per ml by the same method presented above.

Application of ovotransferrin on ham Commercial hams were purchased from 3 local retail stores, and hams from each store were used as a replicate. Hams were sliced to 0.2 cm-thick pieces and vacuum-packaged (-1 bar of vacuum with 10 second dwell time) in low oxygen-permeable bags 2 o (nylon/polyethylene, 9.3 ml O2/m /24hr at 0 C; Koch, Kansas City, MO). Prior to inoculation, each package was aseptically opened using an alcohol-sterilized scissors. Activated E. coli O157:H7 cocktail stock suspension (0.1 ml) was inoculated aseptically on the surface of sliced ham to make 103 CFU/ml level. After inoculated ham samples were manually mixed for 30 second to distribute the inoculum evenly, the packaged samples were randomly divided into 5 groups. Also, 1 ml of 4 different ovotransferrin solutions prepared as pork chops studies was distributed evenly on the surface of hams. Samples inoculated with only E. coli O157:H7 without any solution added was used as a control. All hams were stored in a 10oC incubator. The number of surviving bacteria was enumerated using the same method as in the pork chop study after incubating for 0, 3, 8, and 13 days.

Statistical analysis All experiments were replicated three times and data were analyzed using the JMP software (version 5.1.1; SAS Institute., Cary, NC). The differences in the mean values were compared by Tukey HSD, and mean values and standard error of the means (SEM) were reported (Kuehl, 2000). Statistical significance for all comparisons was obtained at P < 0.05.

151

Results and Discussion Formation of clear zone EDTA alone at 2.5 mg/ml formed a weak clear zone, but less than 2.0 mg/ml did not make any clear zones. Also, all ovotransferrin solutions plus various concentrations of EDTA

(OE) did not form clear zones. Ovotransferrin solution combined with 100 mM NaHCO3 and 2.5 mg/ml EDTA (OSE) formed more apparent clear zones than 2.5 mg/ml EDTA alone. Lysozyme alone at 1.0, 1.5, 2.0 or 2.5 mg/ml did not form clear zones, but ovotransferrin plus lysozyme (OL) and OS plus lysozyme (OSL) formed clear zones (Table 1). The membrane of most bacteria is surrounded by a rigid peptidoglycan to prevent cell rupturing by high internal osmotic pressure. In gram-negative bacteria, their outer membrane lies outside the peptidoglycan but it is protected by lipopolysaccharides (LPS) molecules and is stabilized by divalent cations, which act as salt bridges on neighboring LPS molecules and proteins (Marvin et al., 1989). EDTA can destabilize outer membrane of microorganisms by chelating divalent cations such as Ca2+ and Mg2+ (Vaara, 1992; Shelef and Seiter, 1993; Pelletier et al. 1994). Payne et al. (1994) reported that more than 1 mg/ml EDTA inhibited E. coli O157:H7 in UHT milk. Jarvis et al., (2001) demonstrated that 30 mM EDTA reduced the optical density of E. coli by 50% and induced a complete loss of viability. Park (1997) reported that 0.1~ 0.4 g/l lysozyme has antibacterial activity against E. coli in nutrient broth and vegetable juice. In this study, the growth of E. coli O157:H7 appeared to be inhibited by 2.5 mg/ml EDTA. Also 1.0 ~ 2.0 mg/ml EDTA did not enhance antibacterial activity of ovotransferrin against E. coli O157:H7 significantly, whereas EDTA at 2.5 mg/ml or lysozyme at 1.0 ~ 2.5 mg/ml enhanced the antimicrobial activity of OS against 5 strains of E. coli O157:H7 synergistically. Gill and Holley (2003) reported that using cells harvested from log or stationary phase culture was a poor model for assessing antimicrobial activities in food products because the repairing ability of bacteria on log phase or stationary phase was not

152

equivalent to that of cells on lag adaptation phase. In this study, E. coli O157:H7 strains showed resistance to lysozyme and EDTA probably because the strains used were from the stationary phase.

Antibacterial activity of EDTA alone or in combination with ovotransferrin E. coli O157:H7 in BHI broth entered log phase after 6 hr incubation and the stationary phase after around 10 hr incubation at 35oC. E. coli O157:H7 in the media containing 2 mg/ml or 2.5 mg/ml EDTA alone showed a lag time for 10 hr and entered stationary phase after 18 hr of incubation (Fig. 1). OS, OS plus 2 mg/ml EDTA (OSE2), and OS plus 2.5 mg/ml EDTA (OSE2.5) inhibited E. coli O157:H7 during 35o C incubation for 36 hr (Fig. 1). However, it was difficult to assure that if there was any synergistic effect of EDTA through turbidity test alone because all turbidity values of OS, OSE2 and OSE2.5 were less than 0.1 at 620 nm. Therefore, viability test was performed to determine the synergistic effect of EDTA on antibacterial activity of OS. The number of survivors in media containing 2 mg/ml EDTA alone was not significantly different from that of control as in the turbidity test. However, a significant difference in antibacterial activity (p < 0.05) between OS and OSE2 was found in viability test (Fig. 3). Many other researchers also suggested that EDTA at 2 mg/ml synergistically enhanced the antibacterial activity of OS against E. coli O157:H7. Al-Nabulsi and Holley (2007) reported that lactoferrin plus EDTA inactivated E. coli O157:H7 and affected meat starter cultures. They concluded that EDTA enhanced the antibacterial activity of lactoferrin against E. coli O157:H7. Enhancement of lactoferrin activity by EDTA could be explained by the divalent cation-chelating attributes of EDTA, which destabilizes the outer membrane of microorganism and increases its permeability (Coughlin et al., 1983; Vaara, 1992, Boland et al., 2004). Erdei et al. (1994) identified porins, which are proteins of outer membrane and form cross-membrane channel (Stanier, et al.,1986), at the sites in which lactoferrin is bound on the surface of E. coli and demonstrated that the affinity of lactoferrin to the site is dependent upon the lipopolysaccharides (LPS) of microorganism associated with interactions of lactoferrin. Apo-lactoferrin and holo-lactoferrin induced LPS release from outer membrane of gram-negative bacteria by chelating Ca2+, and thus they are considered to

153

interact directly with LPS (Rossi et al., 2002; van Berkel et al., 1997). The bactericidal effect of lactoferrin is affected by the concentration and the degree of iron saturation in lactoferrin and is prevented by high calcium levels even though the protein does not chelate Ca2+ (Ellison and Giehl, 1991). Even though OS was bacteriostatic against E. coli O157:H7, the combination consisted of ovotransferrin, NaHCO3, and EDTA (OSE2) showed potent bactericidal effects against E. coli O157:H7 by resulting in 3 ~ 4 log reduction. Yakandawala et al. (2006) reported that a combination of ovotransferrin, protamine sulfate and EDTA reduced biofilm formation by catheter-associated bacteria including K. pneumoniae, Pseudomonoas aeruginosa, and

Staphylocococus epidermidis. Al-Nabulsi and Holley (2006) claimed that when NaHCO3 was combined with lactoferrin and EDTA, 4 log reduction in viable cells occurred due to increased lactoferrin stability by bicarbonate.

Antibacterial activity of lysozyme alone or in combination with ovotransferrin In turbidity test, lysozyme itself at 1.0, 1.5, and 2.0 mg/ml could not inhibit the growth of E. coli O157:H7. However, OS, OS plus either 1 mg/ml lysozyme (OSL1) or 2 mg/ml lysozyme (OSL2) elucidated lower than 0.1 turbidity value at 620 nm (Fig. 2). When both 1 mg/ml and 2 mg/ml lysozyme were combined OS, they showed antibacterial activity against E. coli O157:H7 in turbidity test. Therefore, l mg/ml lysozyme was chosen for viability test to determine if OSL was bactericidal or bacteriostatic after 24 hr incubation at 35oC. As shown at Fig. 3, there was no significant difference in antibacterial activity between 1 mg/ml lysozyme (L1) and control, whereas the number of survivors in OSL1 was significantly lower than that of L1 (p < 0.05). This study confirmed the result obtained from turbidity test that L1 itself did not have antibacterial activity against E. coli O157:H7. When 4.0 log CFU/ml of E. coli O157:H7 was initially inoculated to each treatment, the population of viable cells by OS was 4.5 log CFU/ml, while that of OSL1 was 3.2 log CFU/ml after 24 hr incubation at 35oC (Fig. 3). According to these results, OS is bacteriostatic, whereas OSL1 is slightly bactericidal against E. coli O157:H7. However, there was no significantly different in antibacterial activity against E. coli O157:H7 between OS and OSL.

154

Naidu et al (1993) reported that some strains of E. coli with low binding capacity to lactoferrin was recovered relatively easily from its antimicrobial actions, whereas other strains of E. coli with high binding capacity failed to be repaired. Therefore, use of 5 different E. coli O157:H7 strains may indicate increased resistance against OS or OSL treatment than that of a specific strain. Ellison and Giehl (1991) reported that combination of 2 mg/ml lactoferrin and 0.5 mg/ml lysozyme showed bactericidal capacity against E. coli. In addition, they demonstrated that lactoferrin destabilized outer membrane of gram-negative bacteria by releasing LPS. Other researchers (Yamauchi, et al., 1993 Boesman-Finkelstein and Finkelstein, 1985; Ellison and Giehl, 1991) reported that lysozyme enhanced bactericidal activity of lactoferrin and lactoferricin against E. coli even though bovine lactoferrin was bacteriostatic at most. In conclusion, this study demonstrates that L1 enhanced the antimicrobial activity of OS against E. coli O157:H7 and the combination of ovotransferrin with NaHCO3, and L1 (OSL) had a weak bactericidal property against E. coli O157:H7.

Cell number effect This study was performed to know if the population of existing cell affects the antibacterial activity of OS plus 2 mg/ml EDTA (OSE) or OS combined with 2 mg/ml EDTA and l mg/ml lysozyme (OSEL). After 24hr incubation at 35oC, OS did not restrain completely the growth of E.coli O157:H7. According to the results, in antibacteri activity of OS, there was significantly different in antibacterial activity of OS in between 104 CFU/ml and 106 CFU/ml of initial cells. As the number of initial bacteria was relatively small, the antibacterial action of OS against E. coli O157:H7 indicated to be more effective. However, antibacterial activities of OSE and OSEL, which were bactericidal against E. coli O157:H7, were not affected by the number of initial bacteria (Fig. 4). Combinations (EL) consisting of EDTA (2 mg/ml) and lysozyme (1 mg/ml) could not inhibit the growth of E. coli O157:H7 in BHI broth media when inoculated with 105 CFU/ml and 106 CFU/ml cells initially. However, when EL was combined with OS (OSEL), the solution showed a powerful bactericidal activity against E. coli O157:H7 after 35oC incubation in BHI broth for 24 hrs. Many researchers reported that the antibacterial activities of lysozyme against gram-negative and gram-positive microorganisms are related to membrane disruption (Shively and Hartsell,

155

1964a, b; Hughery and Johnson, 1987; Ellison and Giehl, 1991; Payne et al., 1994; Park, 1997; Carneiro De Melo et al., 1998; Kalashyanand et al., 1992). EDTA treatment increased the permeability of outer membrane by releasing LPS and enhanced the antibacterial action of lysozyme. When EDTA and lysozyme were combined, bactericidal activity against E. coli increased (Branen and Davidson, 2004). Contrary to these reports, this study found that 2 mg/ml EDTA plus l mg/ml lysozyme did not inhibit the growth of E. coli O157:H7. Gill and Holly (2003) reported that even though lysozyme and EDTA enhanced antibacterial activity against gram-positive organisms, the combination did not show antimicrobial activity against some gram-negative organisms. Susceptibility of E. coli O157:H7 to EDTA plus lysozyme may be dependent upon the state of strains in inoculation. Arnold et al. (1981) demonstrated that cultures on early exponential phase are more susceptible to antimicrobial action of lactoferrin than that of early stationary phase, which are more resistant. Based on such a suggestion this study used E. coli O157:H7 on stationary phase, and the susceptibility of E.coli O157:H7 to EDTA plus lysozyme was much lower than those obtained from other studies.

Ovotransferrin applications on pork chops and hams E. coli O147:H7 inoculated at 104 CFU/ml grew on the pork chop and increased to 107 CFU/ml after 10oC storage for 12 days. In contrast to the results obtained from in vitro test, 20 mg/ml OS plus 2 mg/ml EDTA (1OSE), 20 mg/ml OS combined with 2 mg/ml EDTA and 1 mg/ml lysozyme (1OSEL), 30 mg/ml OS plus 2 mg/ml EDTA (2OSE) or 30 mg/ml OS combined with 2 mg/ml EDTA and 1 mg/ml lysozyme (2OSEL) did not show any antimicrobial activities (Table 2). Unlike pork chop samples, the viable E. coli O157:H7 cells decreased slowly in commercial hams during 10oC storage for 13 days (Table 3). E. coli O157:H7 grew better in pork chops, which contain more drips by defrosting than commercial hams. Cured meat products provided several restrictive factors in the growth of some microorganisms including low storage temperature, nitrite (less than 200 ppm), NaCl, and lower water activity (Gill and Holley, 2003). Viable E. coli O157:H7 cells inoculated on hams decreased slowly due to these restrictive factors unlike in pork chops.

156

Like the pork chop study, 1OSE, 1OSEL, 2OSE or 2OSEL did not show any antibacterial activity against E. coli O157:H7 inoculated on ham (Table 3). Even though OSE and OSEL showed potent bactericidal effects against E. coli O157:H7 in vitro test, there was no significant difference in the number of survivors between control and all treatments in both pork chops and commercial hams (p < 0.05). Therefore this study demonstrates that there are some limits to overcome to apply ovotransferrin on meat or meat products. The bactericidal effect was affected by concentration and the degree of iron saturation in lactoferrin, and was prevented by high calcium levels even though the protein did not chelate Ca2+ (Ellison and Giehl, 1991). Antibacterial activities of some antimicrobials sometimes can be modulated by nutrient or laboratory media. Cutter and Siragusa (1995a, b) demonstrated that combination of nisin with chelators showed antibacterial activity against E. coli and Salmonella spp. in buffer, but it did not show any antibacterial effectiveness against the same organisms on beef. Branen and Davidson (2004) reported that in spite of antibacterial activity of EDTA and nisin combination against E. coli, the activity did not appear in 2%-fat UHT (ultra high temperature) milk stored at abuse temperatures. Also, Gill and Holly (2003) reported that a combination consisted of lysozyme and nisin had an enhanced bactericidal activity in MRS (de Man, Rogosa, Sharpe) media, but did not show such an effect in pork juice at the same conditions. Al-Nabulsi and Holley (2007) demonstrated that lactoferrin induced cell injury rather than killing E. coli O157:H7 completely, which was responsible for weaker antibacterial activities in fermented meat products. Like these reports, the results from application of antimicrobials to meat products were significantly different from those of model system. Generally, food products provide a nutrient-rich environment, so that microorganisms injured by some antimicrobials can be recovered more easily in the systems than under cell starvation conditions or limited laboratory media. In addition, many divalent cations blocked the biological activity of lactoferrin by forming tetramers in the presence of high concentration of Ca2+ and Mg2+ (Bennett et al, 1981) and modulated the membrane damage by chelators (Gill and Holley, 2003) or inhibited the direct interaction of lactoferrin to LPS (Rossi et al., 2002). Some studies suggested that Ca2+ caused the aggregation of lactoferrin (Bennett et al., 1981) and

157

hindered the LPS release by lactoferrin (Ellison et al., 1988, 1990; Ellison and Giehl, 1991; Yamauchi et al., 1993). A certain concentration of Ca2+ and Mg 2+ decreased the antibacterial activity produced by lactoferrin plus EDTA (Ellison, et al 1988) because lactoferrin was incapable of binding cell surfaces (Ellison et al., 1990). Hams and pork chops used at this study also contained a number of free ions such as Ca2+ and Mg2+. In contrast to the in vitro tests, lack of antibacterial activity of ovotransferrin in pork chops and commercial hams may be due mainly to such divalent ions. For food applications of ovotransferrin, significant amounts of chelators may be necessary to remove the divalent cations and destabilize outer membrane of E. coli present in food systems. Also, distribution or dilution effect of ovotransferrin on the products could be another explanation for the low antibacterial activity of ovotransferrin in pork chops and hams. Al-Nabulsi and Holley (2007) suggested that the low antimicrobial activity occurred by microencapsulated lactoferrin (LF) was partially ascribed to the dilution of LF when microcapsules were incorporated in the sausage batters. Even though at the present study 20 ~ 30 mg/ml of ovotransferrin was distributed on the surface of pork chops and hams, the concentration was likely to be diluted on the surface of the products. In addition, as ovotransferrin was spread manually, homogenous distribution of ovotransferrin on whole products might be impossible. Furthermore, ovotransferrin incorporated might also be penetrated or absorbed inside products so that antimicrobial action against E. coli O157:H7 might not occurr on the surface of the products. Due to significant amount of drip, which were formed during the defrosting process in pork chops, all ovotransferrin incorporated might not have remained on the surface of the pork chops. So, antimicrobial action by ovotransferrin in pork chop and hams seemed to be limited.

Conclusion EDTA at 2 mg/ml and lysozyme at 1 mg/ml enhanced antibacterial activity of

activated ovotransferrin plus NaHCO3 (OS) against E. coli O157:H7 in vitro test. EDTA at 2 mg/ml plus OS resulted in 3 ~ 4 log reduction, while 1 mg/ml lysozyme plus OS resulted in around 0.5 ~ 1 log reduction during incubation at 35oC for 36 hr. Also, initial cell number in media did not affect the antibacterial activity of OSE or OSEL against E. coli O157:H7.

158

EDTA at 2 mg/ml enhanced the antimicrobial activity of OS against E. coliO157:H7 in vitro significantly. However, OSE or OSEL did not show any antimicrobial activity on pork chops and commercial hams. Such differences were ascribed to media compositions or distribution problem. Therefore, further studies on mechanisms of antibacterial action and application methods of ovotransferrin in food are necessary.

Reference Al-Nabulsi, A. A., and Holley, R. A. 2007. Effects on Escherichia coli O157:H7 and meat starter cultures of bovine lactoferrin in broth and microencapsulated lactoferrin in dry sausage batters. Int. J. Food Microbiol. 113: 84-91. Al-Nabulsi, A. A., and Holley, R. A. 2006. Enhancing the bactericidal effects of lactoferrin against E. coli O157:H7 by chelation, temperature, and NaCl. J. Appl. Microbiol. 100: 244-255. Arnold, R. R., Russell, J. E., Champion W. J., and Gauthier, J. J. 1981. Bactericidal activity of human lactoferrin: influence of physical conditions and metabolic state of the target microorganism. Infect. Immun. 32: 655-660. Banin, E., Brady, K. M. and Greenberg, E. P. 2006. Chelator induced dispersal and killing of Pseudomonas aeruginosa cells in a biofilm. Appl. Environ. Microbiol. 72: 2064-2069. Baron, F., Fauvel, S., and Gautier, M. 2000. Egg Nutrition and Biotechnology, Sim, J. S., Nakai, S., and Guenter, W. Eds. CAB Int. pp. 417-440. Baron, F., Gautier, M., and Brule, G. 1997. Factors involved in the inhibition of Salmonella enteritidis in liquid egg white. J. Food Protect. 60: 1318-1323. Bauer, A. W., Kirby, W. M. M., Sherris, J. C., and Turck, M. 1966. Antibiotic susceptibility

testing by a standardized single disk method. Amer. J. Clin. Pathol. 45:493-496. Bell, B., Goldoft, M., Griffin, P., Dans, M., Gordon, D., Tarr, P., Bartleson, C., Lewis, J., Barret, T., Wells, J., Baron, R., and Kobayshi, J. 1994. A multi-state out-break of Escherichia coli O157:H7-associated bloody diarrhea and hemolytic uremic syndrome from hamburgers, the Washington experience. J. Am. Med. Ass. 272: 1349-1353. Bennett, R. M., Bagby, G. C., and Davis, J. 1981. Calcium-dependent polymerization of lactoferrin. Biochem. Biophys. Res. Commun. 101: 88-95.

159

Boesman-Finkelstein, M., and Finkelstein, R. A. 1985. Antimicrobial effects of human milk. FEMS Microbiol. Lett. 27: 167-174. Boland, J. S., Davidson, P. M, Bruce, B., and Weiss, J. 2004. Cations reduce antimicrobial efficacy of lysozyme-chelator combinations. J. Food Prot. 67: 285-294. Branen, J. K., and Davidson, P. M. 2004. Enhancement of nisin, lysozyme and monolaurin antimicrobial activities by ethylenediaminetetraacetic acid and lactoferrin. Int. J. Food Microbiol. 90: 63-74. Carneiro De Melo, A. M. S., Cassar, C. L., and Miles, R. J. 1998. Trisodium phosphate increases sensitivity of Gram-negative bacteria to lysozyme and nisin. J. Food Prot. 61: 839-844. Coughlin, R. T., Tonsager, S., and MaGroarty, E. J. 1983. Quantitation of metal cations bound to membranes and extracted lipopolysaccharide of Escherichia coli. Biochem. 22: 2002-2007. Cutter, C., and Siragusa, G..1995a. Population reductions of Gram negative pathogens following treatments with nisin and chelators under various conditions. J. Food Prot. 58: 977-983. Cutter, C., and Siragusa, G. 1995b. Treatments with nisin and chelators to reduce Salmonella and Escherichia coli on beef. J. Food Prot. 58:1028-1030. Ellison, R. T. III, and Giehl, J. J., 1991. Killing of Gram negative bacteria by lactoferrin and lysozyme. J. Clin. Invest. 88: 1080- 1091. Ellison, R. T. III, Giehl, R, J., and LaForce, F. M. 1988. Damage of the outer membrane of enteric gram-negative bacteria by lactoferrin and transferrin. Infect. Immun. 56: 2774- 2781. Ellison, R. T. III, LaForce, F. M., Giehl, T. J., Boose. D. S., and Dunn, B. E. 1990. Lactoferrin and transferrin damage of the gram-negative outer membrane is modulated by Ca2+ and Mg2+. J. Gen. Microbiol. 136: 1437-1446. Erdei, J., Forsgren, A., Naidu, A.S. 1994. Lactoferrin binds to porins OmpF and OmpC in Escherichia coli. Infect. Immun. 62: 1236-1240. Firstenberg-Eden, R. 1997. E. coli rapid detection system. A rapid method for the detection of Escherichia coli O157 in meat and other foods. J. Food Prot. 60: 219-225.

160

Gill, A. O., and Holley, R. A. 2003. Interactive inhibition of meat spoilage and pathogenic bacteria by lysozyme, nisin and EDTA in the presence of nitrite and sodium chloride at 24o C. Inter. J. Food Microbiol. 80: 251-259. Griffin, P., and Tauxe, R. 1991. The epidemiology of infections caused by Escherichia coli O157:H7, other enter hemorrhagic Escherichia coli and the associated hemolytic uremic syndrome. Epidemiol. Rev. 13: 60-98. Hansen , L. T., Austin, J. W., and Gill , T. A. 2001. Antibacterial effect of protamine in combination with EDTA and refrigeration. Int. J. Food Microbiol. 66: 149-161. Hughey, V. L., and Johnson, E. A. 1987. Antimicrobial activity of lysozyme against bacteria involved in food spoilage and food-borne disease. Appl. Environ. Microbiol. 53: 2165- 2170. Ibrahim, H. R. 1997. Insights into the structure - function relationship of ovalbumin, ovotransferrin and lysozyme. In: Hen eggs: Their Basic and Applied Science. Yamamoto T., Juneja I.R., Hatta H., and Kim M. (Eds). CRC Press. pp. 37-56. Ibrahim, H. R., Sugimoto, Y., and Aoki, T. 2000. Ovotransferrin antimicrobial peptide (OATP-92) kills bacteria through a membrane damage mechanism. Biochim. Biophys. Acta 1523: 196-205. Jarvis, G. N., Fields, M.W., Adamovich, D. A., Arthurs, C. E., and Russel, J. B. 2001. The mechanisms of carbonate killing of Escherichia coli. Lette. Appl. Microbiol. 33: 196- 200. Kalachyanand, N., Hanlin, M. B., and Ray, B. 1992. Sublethal injury makes Gram negative and resistant Gram positive bacteria sensitive to the bacteriocins, pediocin AcH and nisin. Lett. Appl. Microbiol. 15: 239-243. Kuehl, R. O. 2000. Design of Experiments; Statistical Principles of Rresearch Design and Analysis. 2nd Ed. New York, Duxbury Press. Kurokawa, H., Mikami, B., and Hirose, M. 1995. Crystal structure of di-ferric hen ovotransferrin at 2.4 Ǻ resolution. J. Mol. Biol. 254: 196-207. Marvin, H, J. P., Ter Beest, M. B. A., and Witholt, B. 1989. Release of outer membrane fragments from wild-type Escherichia coli and from several E. coli lipopolysaccharide mutants by EDTA and heat shock treatments. J. Bacteriol. 171: 5262-5267.

161

Naidu, A. S. 2002. Activated Lactoferrin-A New Approach to Meat Safety. Food Technol. 56(3): 40-45. Naidu, A. S., Tulpinski, J., Gustilo, K. Nimmagudda, R., and Morgan, J. B. 2003. Activated lactoferrin part 2: natural antimicrobial for food safety. Agro. Food Industry Hi-Tech. 14(3): 27-31. Park, W. B. 1997. Introduction of lysozyme susceptibility to Escherichia coli by addition of papain and polysorbate 80 in culture medium vegetable juice. Food Biotech. 6: 158-161. Payne, K. D., Oliver, S. P., and Davidson, P. M. 1994. Comparison of EDTA and apo- lactoferrin with lysozyme on the growth of food borne pathogenic and spoilage bacteria. J. Food Prot. 57: 62-65. Pelletier, C., Bourlioux, P., and van Heijenoort, J. 1994. Effects of sub-minimal inhibitory concentrations of EDTA on growth of Escherichia coli and the release of lipopolysaccharide. FEMS Microbiol. Lett. 117: 203-206. Proctor, V. A., and Cuuningham, R. E. 1988. The chemistry of lysozyme and its use as a food preservative and a pharmaceutical. Crit. Rev. Food Sci. Nutr. 26: 359-395. Razaci-Rohani, S. M., and Griffiths, M. W. 1994. The effect of mono and polyglycerol laureate on spoilage and pathogenic bacteria associated with foods. J. Food Safety 14: 131-151. Rossi, P., Giansanti, F., Boffi A., Ajello, M., Valenti, P., Chiancone, E. and Antonini, G.. 2002. Ca2+ binding to bovine lactoferrin enhances protein stability and influences the release of bacterial lipopolysaccharides. Biochem. Cell Biol. 80: 41-48. Shade, A. L. and Caroline, L. 1944. Raw hen egg white and the role of iron in growth inhibition of Shigella dysenteriae, Staphylococcus aureus, Escherichia coli and Saccharomyces cerevisiae. Science 100: 14-15. Shelef, L. A., and Seiter, J. 1993. Enhancement of the activity of novobiocin against Escherichia coli by lactoferrin. J. Dairy Sci. 82: 492-499. Shively, J. M., and Harsell, S. E. 1964b. Bacteriolysis of the Pseudomonads. II Chemical treatments affecting the lytic response. Can. J. Microbiol. 10: 911-915. Shively, J. M., and Hartsell, S. E. 1964a. Bacteriolysis of the Pseudomonads: I. Agents potentiating lysis. Can. J. Microbiol. 10: 905-909.

162

Stainer, R. Y., Ingraham, J. L., Wheelis, M. L., Painter, P. R. 1986. The Microbial World. 5th ed. Prentice-Hall a division of Simon and Schuster Englewood Cliffs, NJ, USA. Stevens , K., Sheldon, B., Klapes, H., and Klaenhammer, K. 1991. Nisin treatment for Salmonella species and other Gram negative bacteria. Appl. Environ. Microbiol. 57: 3613-3615. Vaara, M., 1992. Agents that increase the permeability of the outer membrane. Microbiol. Rev. 56: 395-411. Valenti, P., Antonin. G., von Hunolstein, C., Visca, P., Orsi, N., and Antonini, E. 1983. Studies of the animicrobial activity of ovotransferrin. Int. J. Tissue Reac. 1: 97-105. Valenti, P., Visca, P., Antonini, G., and Orsi, N. 1985. Antifungal activity ovotransferrin towards genus Candida. Mycophathologia 89: 169-175. Van Berkel, P. H. C., Geerts, M. E. J., van Veen, H. A., Mericskay, M., de Boer, H. A., and Nuijens, J. H. 1997. N-terminal stretch Arg2, Arg3, Arg4 and Arg5 of human lactoferrin is essential for binding to heparin, bacterial lipopolysaccharides, human lysozyme and DNA. Biochem. J. 328: 145-151. Weeranta, R. D. and Doyle, M. P. 1991. Detection and production of verotoxin 1 of Escherichia coli O157:H7 in food. Appl. Environ. Microbiol. 57: 2951-2955. Yakandawala, N., Gawansde, P. V., LoVetri, K., and Madhyastha, S. 2006. Effect of ovotransferrin, protamine sulfate and EDTA combination on biofilm formation by Cather-associated bacteria. J. Appl. Microbiol. 102: 722-727. Yamauchi, K., Tomita, M., Giehl, T. J. and Ellison, R. T. III. 1993. Antibacterial activity of lactoferrin and a pepsin-derived lactoferrin peptide fragment. Infect. Immun. 61: 719- 728.

163

Table 1. Effect of ovotransferrin solutions combined with NaHCO3 and EDTA or lysozyme on the formation of clear zone in BHI agar media inoculated with E. coli O157:H7 during 35oC incubation for 24 ~ 48 hrs

a Treatment Conc. (mg/ml) Control OTF OTF + NaHCO3

0 -* - - 1.0 - - - EDTA 1.5 - - - 2.0 - - +/- 2.5 +** - ++

1.0 - + ++ Lysozyme 1.5 - + ++ 2.0 - + + 2.5 - + ++

OTF: 20 mg/ml apo-ovotransferrin. a 100 mM-NaHCO3 was used at this study. * no clear zone formation on BHI agar media plate. ** formation of clear zone : the strength of clear zone was described as + (weak), ++ (middle), and +++ (strong) by its size and clarity on the BHI agar plates (n = 3).

164

Table 2. Changes of the number of E. coli O157:H7 survivors on e-beam irradiated pork

chops treated with or without ovotransferrin solutions contained 100 mM NaHCO3 plus either EDTA (2 mg/ml) or/and lysozyme (1 mg/ml) during 10o C storage

Treatments Storage period (days) 0 2 4 8 12

------Number of viable cell (log10 CFU/ml) ------Control* 4.1a ± 0.1** 4.2a ± 0.1 5.0a ± 0.1 5.2a ± 0.1 7.3a ± 0.1 1OS + EDTA 4.1a ± 0.1 4.5a ± 0.2 4.7a ± 0.6 6.0a ± 1.5 7.2a ± 0.7 1OS + EDTA + Lyso 4.2a ± 0.1 4.0a ± 0.2 5.0a ± 0.8 5.9a ± 0.7 7.2a ± 1.3 2OS + EDTA 4.2a ± 0.0 4.3a ± 0.6 5.4a ± 0.6 6.2a ± 0.3 6.9a ± 1.0 2OS + EDTA + Lyso 4.0a ± 0.2 4.2a ± 0.1 5.9a ± 0.4 5.9a ± 1.2 7.0a ± 1.3

aValues with different letters within a column are significantly different (p< 0.05). n = 3 Control: Only E. coli O157:H7 inoculation.

1OS: Ovotransferrin (20 mg/ml) + 100 mM-NaHCO3

2OS: Ovotransferrin (30 mg/ml) + 100 mM-NaHCO3 Lyso: lysozyme (1mg/ml), 2 mg/ml EDTA was used at this study

165

Table 3. Survivors of E. coli O157:H7 in commercial hams treated with or without

ovotransferrin solutions contained 100 mM-NaHCO3 plus EDTA or/and lysozyme during storage at 10o C

Treatment Storage period (days) 0 3 8 13

------Number of viable cell (log10 CFU/ml) ------Control 3.7a ± 0.0 ** 3.8 a ± 0.2 3.3 a ± 0.4 2.8 a ± 0.1 1OS + EDTA 3.5 a ± 0.0 3.5 a ± 0.1 3.6 a ± 0.1 2.9 a ± 0.2 1OS + EDTA + Lyso 3.5 a ± 0.1 3.5 a ± 0.3 3.2 a ± 0.3 3.2 a ± 0.3 2OS + EDTA 3.4 a ± 0.1 3.5 a ± 0.1 3.7 a ± 0.3 2.7 a ± 0.2 2OS + EDTA + Lyso 3.3 a ± 0.2 3.5 a ± 0.3 3.1 a ± 0.5 2.7 a ± 0.2

aValues with different letters within a column are significantly different (p< 0.05). n = 3 Control: Only E. coli O157:H7 inoculation.

1OS: Ovotransferrin (20 mg/ml) + 100 mM NaHCO3

2OS: Ovotransferrin (30 mg/ml) + 100 mM NaHCO3 Lyso: lysozyme (1 mg/ml)

166

Fig. 1. Turbidity of BHI broth cultures inoculated E. coli O157:H7 and ovotransferrin (20 o mg/ml) solution combined NaHCO3 or/and EDTA during 35 C incubation

1.3

1.1

0.9

0.7

0.5

0.3 Turbidity (at 620nm) (at Turbidity 0.1

-0.1 0 10203040 Time (hr)

E2 E2.5 OSE2 OSE2.5 C OS

C: control, only 104 CFU/ml of E. coliO157:H7 E2: 2 mg/ml EDTA E2.5: 2.5 mg/ml EDTA

OS: 20 mg/ml ovotransferrin + 100 mM-NaHCO3 OSE2: OS + 2 mg/ml EDTA OSE2.5: OS + 2.5 mg/ml EDTA.

167

Fig. 2. Turbidity of BHI broth cultures inoculated E. coli O157:H7 and ovotransferrin (20 o mg/ml) combining NaHCO3 or/and lysozyme during 35 C incubation

1.3

1.1

0.9

0.7

0.5

0.3 Turbidity (at 620nm) (at Turbidity 0.1

-0.1 0 10203040 Time (hr)

L1 L1.5 L2 OSL1 OSL1.5 OSL2 OS C

C: control, only 104 CFU/ml of E. coliO157:H7 L1: 1 mg/ml Lysozyme L1.5: 1.5 mg/ml Lysozyme L2: 2 mg/ml lysozyme

OS: 20 mg/ml ovotransferrin + 100 mM NaHCO3 OSL1: OS + 1 mg/ml lysozyme OSL1.5: OS + 1.5 mg/ml lysozyme OSL2: OS + 2 mg/ml lysozyme

168

Fig. 3. Antibacterial activity of ovotransferrin (20 mg/ml) containing 100 mM NaHCO3 and lysozyme (1 mg/ml) or EDTA (2 mg/ml) against the growth of E. coli O157:H7 in BHI broth culture during 35oC incubation for 24 ~ 36 hr.

12.0

10.0 a aa

8.0

6.0

CFU / ml CFU b 10 bc 4.0 Log

2.0 c

0.0 1 C L1 E2 OSL1 OSE2 OS

C: control, only 104 CFU/ml of E.coli O157:H7 L1: 1 mg/ml Lysozyme E2: 2 mg/ml EDTA

OS: 20 mg/ml ovotransferrin + 100 mM-NaHCO3 OSL1: OS + 1 mg/ml lysozyme OSE2: OS + 2 mg/ml EDTA a~cletters labeled on bars means there were significantly different (p < 0.05, n=3)

169

Fig. 4. Effect of initial cell population on antibacterial activity of ovotransferrin (20 mg/ml) combining 100 mM NaHCO3 and EDTA (2 mg/ml) or Lysozyme (1 mg/ml) against E. coli O157:H7 in BHI broth culture during 35o C incubation for 24 ~ 36 hr.

10.0 a a 9.0 a ab 8.0 7.0 b 6.0 5.0

4.0 c

Log 10 CFU Log 10 c 3.0 2.0 cc c c 1.0 0.0 OS (A) OSE (A) 1 OSEL (A) OS (B)

OSE (B) OSEL (B) OS (C) OSE (C) OSEL (C) EL (B) EL (C)

A: 104 CFU/ ml BHI broth B: 105 CFU/ ml BHI broth C: 106 CFU/ ml BHI broth

OS: 20 mg/ml ovotransferrin + 100 mM-NaHCO3 OSE: OS + 2 mg/ml EDTA OSEL: OS + 2 mg/ml of EDTA + 1 mg/ml lysozyme EL: 2 mg/ml EDTA + 1 mg/ml lysozyme a~cletters labeled on bars means there were significantly different (p <0.05, n=3)

170

CHAPTER 7. EFFECT OF EDTA AND LYSOZYME ON THE ANTIMICROBIAL ACTIVITY OF OVOTRANSFERRIN AGAINST LISTERIA MONOCYTOGENES

A paper which will be submitted to Journal of Food Protection

K. Y. Ko1, D. U. Ahn2 and A. Mendonca

Abstract This study evaluated the effect of EDTA and lysozyme on the antibacterial activities of activated ovotransferrin against 5 strains of L. monocytogenes. First, a disc test was performed to screen the concentrations of EDTA or lysozyme that showed antibacterial activities in ovotransferrin (O) or ovotransferrin in 100 mM NaHCO3 (OS) solution. Turbidity and viability test were conducted using O or OS solution combined with either lysozyme (OL and OSL) or EDTA (OE and OSE). Also, OS combined with 2 mg/ml lysozyme (OSL) or/and 1 mg/ml EDTA (OSLE) were applied on commercial hams to determine if the solutions show antibacterial activities on meat products. The effect of initial cell population on the antibacterial activities of ovotransferrin combined with either EDTA or lysozyme was also determined. L. monocytogenes started to grow after 1 day of incubation in the presence of > 2.0 mg/ml lysozyme. OL groups showed weak antibacterial activities against L. monocytogenes in BHI broth culture and their activities were bacteriostatic. OSL groups were bactericidal against L. monocytogenes, resulting in 1 log reduction from initial cell population. Even though OSL showed stronger antibacterial activity than OS, lysozyme did not significant effect on antibacterial activity of OS against L. monocytogenes. Also, EDTA itself at 1.0 and 2.0 mg/ml were bacteriostatic against 5 strains of L. monocytogenes. They were more susceptible to EDTA than lysozyme, and OSE1 and OSE2 had bactericidal activity against L. monocytogenes. There was a significant difference in the survivor cell populations between OS and OSE groups (p < 0.05). Therefore, EDTA enhanced the

1 Primary researcher and author 2 Author of correspondence.

171 antibacterial activity of OS against L. monocytogenes. However, ovotransferrin plus either lysozyme or/and EDTA did not show any antibacterial effect in commercial hams during storage at 10oC. In addition, the initial population of L. monocytogenes cells influenced the antibacterial activity of OSL or OSE.

Introduction Natural antimicrobial agents originat from plant, animal, and bacterial sources are expected to provide great satisfaction to consumers who are sensitive to health and safety issues. Ovotransferrin, the second major egg white protein, is an iron-binding glycoprotein and transports and scavenges Fe (III) in eggs of poultry (Kurokawa et al., 1995). Antibacterial properties of ovotransferrin against a variety of microorganisms including Escherichia coli (Shade and Carolin, 1944; Valenti et al., 1983), Pseudomonas spp., Streptococcus mutans (Valenti et al., 1983), Staphylococcus aureus, Bacillus cereus (Ibrahim, 1997), Salmonella enteritidis (Baron et al., 1997; 2000), and Candida (Valenti et al., 1985) have been reported. However, the antimicrobial mechanisms of ovotransferrin are not fully defined yet. The iron binding capability of ovotransferrin was initially believed to be the major antimicrobial action of ovotransferrin. However, recent studies demonstrated that direct interactions of ovotransferrin with bacterial surface were the major cause of antimicrobial action of ovotransferrin (Valenti et al., 1983, 1985; Arnold et al., 1981; Ibrahim, 1997; Ibrahim et al., 2000). L. monocytogenes can proliferate in the presence of curing salt at refrigerated temperature (Lou and Yousef, 1999), and colonize, multiply and persist on processing equipment or plant substances (Rocourt and Seeliger, 1985). The strains are sensitive to heat treatment and can be easily inactivated by cooking. However, ready-to-eat (RTE) meat products such as frankfurters, and deli meats have high incidence rates of listeriosis due to contamination during multiple handling and processing steps performed after cooking (Uyttendaele et al., 1999; Giovannacci et al., 1999; Lunden et al., 2003). Centers for Disease Control and Prevention (CDC, 2002) reported that 2,500 people suffer by listeriosis annually in the United States and one in five die from the disease (FSIS). Due to this risk, the U.S. Department of Agriculture currently established a “zero tolerance” policy for L.

172

monocytogenes in RTE meat products. Therefore, developing potent antimicrobials against L. monocytogenes for RTE meat is necessary. According to previous study, apo-ovotransferrin alone was not effective in controlling L. monocytogenes in BHI broth. To improve the antimicrobial activities of ovotransferrin, therefore, combinations of several antimicrobial agents are necessary (Sofos et al., 1998) The combinations of materials with different antimicrobial mechanisms are expected to increase antibacterial effectiveness since a simultaneous attack on different targets are likely to make the microorganism more difficult to overcome the environment. The use of ovotransferrin combined with EDTA or/and lysozyme as antimicrobials on the surface of meat or meat products is appealing to meat industry because all the components are generally regarded as safe (GRAS).

Lysozyme, a single peptide protein with 14.6 kDa of molecular weight catalyzes the hydrolysis of (1-4) glycosidic linkages between N-acetylmuramic acid and N- acetylglucosamine of cell wall peptidoglycan (Proctor and Cunningham, 1988). Therefore, the antimicrobial activity of lysozyme is ascribed primarily to the enzymatic lyses of peptidoglycan in the cell wall of microorganisms (Branen and Davidson, 2004). The application of lysozyme in foods has been limited because gram-negative bacteria that are protected by an outer membrane, which is relatively resistant to antimicrobial activities of lysozyme (Gill and Holley, 2003; Proctor and Cunningham, 1988). However, lysozyme gained considerable interest for use in food systems because it is a natural enzyme produced by animals and its activity targets a specific cellular structure of microorganisms (Proctor and Cunningham, 1988). Ethylenediaminetetraacetic acid (EDTA) has been used in a number of food products as a chelating agent to prevent oxidation (Hansen et al., 2001). Although the mechanism of antibacterial effectiveness by EDTA is not fully understood, its chelating property to divalent cations like Ca2+ or Mg2+ should be involved (Shelef and Seiter, 1993). EDTA inhibits the growth of microorganisms by depriving Mg2+, Ca2+, and Fe2+, which are essential factors for microbial growth, from microorganisms (Banin et al., 2006). EDTA is reported to enhance the antimicrobial activity of lactoferrin by destabilizing the outer membrane of bacteria and increasing the permeability of divalent cations (Coughlin et al., 1983; Vaara, 1992; Boland et

173

al., 2004). Also, chelating agents facilitate the detachment of cells from biofilm and enhance the killing of biofilm-producing microorganisms by depriving Mg2+ associated with lipopolysaccharides (Banin et al., 2006). The objective of this study was to enhance the antimicrobial activity of ovotransferrin solution supplemented with 100 mM-sodium bicarbonate against a cocktail of 5 strains of L. monocytogenes by combining with lysozyme or EDTA. Many studies related to antibacterial activity of ovotransferrin in vitro have been reported, but little work has been done to apply ovotransferrin as an antimicrobial agent in meat or meat products. Therefore, the present study applied OS plus lysozyme or/and EDTA on commercial hams to determine their effect on the growth of L. monocytogenes in meat products.

Material and Methods

Ovotransferrin Apo-ovotransferrin (iron-free) used in this study was prepared by the method of Ko and Ahn (2007, unpublished). After iron saturation of 2x-diluted egg white solution, holo- ovotransferrin (iron-saturated) were separated using ethanol. Iron was removed from the ® holo-ovotransferrin using AG 1-X2 resin (chloride form, Bio-Rad, California, U.S.A), and then freeze-dried. The dried apo-ovotransferrin was dissolved in 0.15 M NaCl solution and pH adjusted to 7.4. The purity of ovotransferrin used in this study was around 80% and the residual iron in prepared apo-ovotransferrin solution was less than 0.5 ppm. Prior to microbial study, the ovotransferrin solution was sterilized by passing through a 0.22 μm syringe filter (Whatman Inc. Florbam Park, NJ. USA).

Bacterial strains Five different strains of Listera monocytogenes (NADC 2045, H7962, H7969, H7762, and H7596) were used. Prior to inoculation, each strain was individually cultured on brain heart infusion broth (BHI, Remel Inc., Lenexa, KS, USA) for 24 hours at 35oC and harvested twice in order to activate the strain appropriately. Each strain culture solution (5 ml) was transferred to a sterilized centrifuge bottle and centrifuged at 100,000 x g for 10 min at 4oC.

174

The pellet was collected, washed with saline, and then resuspended in 25 ml saline. Inoculation cocktail was prepared by mixing equal volumes of each cell suspension to have approximately the same population of five strains of L. monocytogenes. The number of inoculated L. monocytogenes strain was analyzed on BHI agar plates after serial dilution with 0.1% peptone (Difco Laboratories, Detroit, MI) water and incubation at 35oC.

Formation of Clear Zone Prior to analyzing the antibacterial activity through viability and turbidity tests, disc diffusion method (Bauer et al, 1966) was used to determine the concentration of EDTA (disodium salt, Fisher Scientific, Fairlawn, NJ, USA) or lysozyme (from egg white, Sigma- Aldrich Inc., St. Louis, USA) capable of promoting antibacterial capability of ovotransferrin. L. monocytogenes cocktail containing equal amount of each strain was diluted in series and then 0.1 ml of cell suspension corresponding to 105 or 106 CFU/ml was spread homogenously on the BHI agar media plates. The BHI agar plates inoculated with L. monocytogenes were dried for 30 ~ 60 minutes in an incubator at 35o C. EDTA (1, 0, 1.5, 2.0, or 2.5 mg/ml) or lysozyme (1.0, 1.5, 2.0, or 2.5 mg/ml) was added to ovotransferrin solution

or ovotransferrin solutions in 100 mM-NaHCO3 (OS). All solutions prepared were sterilized using 0.22 μm syringe filters (Whatman Inc. Florbam Park, NJ. USA) prior to performing this experiment. Ovotransferrin solutions (120 μl) containing different concentrations of EDTA or lysozyme were dispensed to sterilized glass microfiber filter (2.1cm diameter, Whatman Inc. Florbam Park, NJ. USA) on the dried BHI agar media. Three replications were prepared. The BHI agar plates were incubated at 35oC for 24 ~ 48 hrs and then antibacterial activity of each ovotransferrin solution against L. monocytogenes was estimated through the existence of clear zones.

Turbidity test The antimicrobial capacity of ovotransferrin solutions combined with EDTA or lysozyme against the growth of L. monocytogenes was analyzed by measuring the turbidity of solution after incubation in a BHI broth culture at 35o C. Disc diffusion test indicated that when ≥ 2.0 mg/ml EDTA and ≥ 1.5 mg/ml lysozyme was combined with ovotransferrin, the

175

solutions had antibacterial activity against L. monocytogenes (Table 1). After 2 ml of 40 mg/ml ovotransferrin solution was added to 2 ml of 2x-strength BHI broth, stock solutions of EDTA or lysozyme was added to the media including ovotransferrin (O) or OS. The final concentration of lysozyme was adjusted to be 2.0 mg/ml (OL2 and OSL2), 2.5 mg/ml (OL2.5 and OSL 2.5), or 3.0 (OL3 and OSL3) mg/ml, and that of EDTA was adjusted to 1.0 mg/ml (OE1 and OSE1), 1.5 mg/ml (OE1.5 and OSE1.5) or 2.0 mg/ml (OE2 and OSE2.5). Also, 40 μl of cell suspension containing 106 ~107 level of activated L. monocytogenes was inoculated to the prepared solution to make the initial population of L. monocytogenes 104 ~ 105 CFU per ml. EDTA and lysozyme solution without ovotransferrin was prepared to determine the antibacterial activity of EDTA or lysozyme alone. After inoculation, the samples were incubated at 35o C for 36 hrs. Proliferation degree of L. monocytogenes in each treatment was evaluated by measuring its turbidity with a spectrophotometer at 620 nm every 3 ~ 6 hrs.

Viability analysis The antibacterial activities of ovotransferrin plus lysozyme (OSL) and ovotransferrin plus EDTA (OSE) were investigated using a viability test. First, 2 ml of ovotransferrin solution (40 mg/ml) was added to 2 ml of 2x-concentrated BHI broth media, and then 100 mM sodium bicarbonate was added. Stock solution of either EDTA or lysozyme was added to the BHI broth culture contained with ovotransferrin alone or OS. The final concentration of lysozyme and EDTA in solution was identical to that of turbidity test. Also, 40 μl of cell suspension containing 106 ~ 107 cells of activated L. monocytogenes was inoculated to make initially 104 ~ 105 CFU per ml in BHI broth culture. EDTA or lysozyme alone was prepared as a control group. After inoculation, the prepared culture solutions were incubated at 35oC for 36 hrs. The number of viable cells was analyzed by spreading each culture solution after diluting (1:10) with 0.1% sterile peptone water (Difco Laboratories, Detroit, MI), spreading 0.1 ml of diluted samples homogenously on a BHI agar plate, and incubating at 35oC for 24 ~ 36 hrs. The number of survivors on BHI agar plates was counted as colony-forming units per ml (CFU/ml) of sample.

176

Effect of initial cell number This study was performed to know if the populations of initial cells affect the antibacterial activity of ovotransferrin combined with various additives. The BHI broth media containing ovotransferrin were prepared with the same method described above. Stock solution of either EDTA or lysozyme was added to OS so that the final concentration of lysozyme and EDTA in the ovotransferrin solution became 2 mg/ml lysozyme (OSL2) or 1 mg/ml EDTA (OSE1), respectively. EDTA (2 mg/ml) or lysozyme (1 mg/ml) alone was prepared to identify antibacterial activity of the solution itself against L. monocytogenes. Also, 40 μl of cell suspension containing 106, 107, or 108 cells of activated E. coli O157:H7 was inoculated to 4 ml BHI broth media to make the number of L. monocytogenes 104, 105 or 106 CFU per ml. After the prepared culture solutions were incubated at 35oC for 24 ~ 36 hrs, and the viability of each treatment was measured using the same method as above.

Application of ovotransferrin to hams and pork chops Commercial hams were purchased from local retail stores. Hams were sliced to 0.2 cm-thick pieces and vacuum packaged (-1 bar of vacuum with 10second dwell time) in low 2 o oxygen-permeable bags (nylon/polyethylene, 9.3ml O2/m /24hr at 0 C; Koch, Kansas City, MO). Prior to inoculation, each package was aseptically opened using an alcohol-sterilized scissors. Activated L. monocytogenes cocktail stock suspension (0.1 ml) was inoculated aseptically on the surface of sliced ham to make the initial cell number around 103 CFU /ml. After manually mixing for 30 second to distribute the inocula evenly, the packaged samples were randomly divided into 7 groups. Lysozyme or/and EDTA were added to ovotransferrin solution included 100 mM NaHCO3 and 0.15 M NaCl. Therefore, the finally treatments comprised of 2 mg/ml lysozyme (L), 2 mg/ml lysozyme plus 1 mg/ml EDTA (LE), 20 mg/ml of ovotransferrin plus 2 mg/ml lysozyme (1OSL), 20 mg/ml of ovotransferrin plus 2 mg/ml lysozyme and 1 mg/ml EDTA (1OSLE), 30 mg/ml of ovotransferrin plus 2 mg/ml lysozyme (2OSL), and 30 mg/ml of ovotransferrin plus 2 mg/ml lysozyme and 1 mg/ml EDTA (2OSLE). Each prepared ovotransferrin solution (1 ml) was distributed evenly on the surface of a ham. Control group was prepared by inoculating L. monocytogenes to hams without adding any solution, and all hams were stored in a 10oC incubator.

177

Viable L. monocytogenes cells on the pork chops were analyzed after 0, 2, 5, 10, 15, 22, and 29 days of storage: 30 ~ 50 ml of sterile 0.1% peptone water solution was added to each vacuum package bag followed by pummeling for 1 minute in a stomacher at normal speed. After serial dilution of the sample with 0.1% peptone water, 0.1 ml of diluted sample was spread homogenously on MOX media (Difco Laboratories, Detroit, MI) plates in duplicate. Survivors were enumerated as CFU per ml by the same method as above.

Statistical analysis All experiments were replicated three times and the data were analyzed using the JMP software (version 5.1.1; SAS Institute., Cary, NC). The differences in the mean values were compared by Tukey HSD, and means values and standard error (SEM) were reported (Kuehl, 2000). Statistical significance for all comparisons was obtained at p < 0.05.

Results and Discussion

Formation of clear zone Lysozyme at 1.0 and 1.5 mg/ml did not form clear zones, but 2.0, 2.5 and 3.0 mg/ml lysozyme did. The strength of clear zone formed by lysozyme combined with ovotransferrin or OS did not show significant difference compared with that of lysozyme (Table 1). L. monocytogenes was susceptible to ≥ 2.0 mg/ml lysozyme. The antimicrobial activity of lysozyme has long been believed to be due to its bacteriolytic actions, which hydrolyze ß- 1.4-glycosidic bond in peptidoglycan of gram-positive bacteria that do not possess any outer membrane, but gram-negative bacteria with outer membrane is reported to be relatively resistant to lysozyme (Davis and Reeves, 2002; Davies et al., 1980; Spitznagel, 1984). The antibacterial action of lysozyme against L. monocytogenes in vegetables were strong, but that in animal-derived foods such as pork sausage and cheese were weak (Hughey et al., 1989). Carminati and Carni (1989) reported that no antimicrobial activities by lysozyme against L. monocytogenes were shown in whole milk. EDTA at 1.5 and 2.0 mg/ml made clear zone, but no clear zone was formed at 1.0 mg/ml EDTA. When EDTA was combined with ovotransferrin or OS, the strength of clear

178

zone decreased (Table 1). Zhang and Mustapha (1999) demonstrated that EDTA itself was not lethal to L. monocytogenes under the concentrations permitted in foods. EDTA is usually used in combination with other preservatives and promotes the suppression of L. monocytogenes in meat (Monk et al., 1996; Parente et al., 1998), but sometimes EDTA seems to modulate the antibacterial activity of antimicrobials such as nisin (Zhang and Mustapha, 1999). This study indicated that more than 1.5 mg/ml of EDTA inhibited the growth of L. monocytogenes. However, when EDTA was combined with ovotransferrin or OS, the strength of clear zone decreased. Based on the disc test, more than 2 mg/ml lysozyme or 1 mg/ml of EDTA were chosen in turbidity and viability tests, and tests were performed to identify EDTA and lysozyme effects on antibacterial activities of ovotransferrin.

Lysozyme on antibacterial activity of ovotransferrin As shown in Fig. 1, L. monocytogenes entered log phase after 7 ~ 8 hr incubation at 35oC in BHI broth culture. They seemed to face stationary phase after 24 hr incubation and then death. Lysozyme at 2 mg/ml delayed the growth of L. monocytogenes for 24 hrs, whereas lysozyme 2.5 mg/ml and 3 mg/ml delayed it for 30 hrs and 36 hr, relatively. Therefore, L. monocytogenes seemed to have a lag time to adapt new environment when lysozyme was added (Fig. 1). Even though lysozyme alone delayed the growth of L. monocytogenes during initial period, L. monocytogenes started to grow after 24 ~ 36 hr incubation at 35oC. The turbidity values obtained from all OL or OSL groups were lower than 0.1 as in OS treatment (Fig 1). According to turbidity test, L. monocytogenes were found to be controlled by OL or OSL groups even though they were resistant to lysozyme alone under the concentrations used in this study. In viability test, which measures the number of viable L. monocytogenes cells after 48 hr incubation at 35o C, with 2.0, 2.5 or 3.0 mg/ml lysozyme was not different from that of the control (p < 0.05) (Fig. 2). The activities of lysozyme against L. monocytogenes is reported to vary depending upon food products or media: L. monocytogenes are susceptible to lysozyme in media and phosphate buffer, whereas they are highly resistant to lysozyme in whole milk containing diavalent cations such as Ca2+ and Mg2+ (Kihm et al., 1994). As BHI media used in this study is relatively nutrient-rich media, injured cells of L. monocytogenes by lysozyme

179

treatment should have been repaired easily in the media. Thus they showed resistance to 2.0 ~ 3.0 mg/ml lysozyme. Another possible explanation for resistance of L. monocytogenes to lysozyme may be attributed to using a cocktail of 5 different strains of L. monocytogenes instead of using a specific strain. The number of viable cells obtained when using OS slightly increased from 4.6 log CFU/ml, the number of the initial inoculation, to 4.9 log CFU/ml during 48 hrs incubation at 35oC. Therefore, OS seemed to be bacteriostatic against L. monocytogenes. Also the number of viable L. monocytogenes cells in OL2, OL2.5, and OL3 increased from 4.6 log CFU/ml, the number of initial inoculated cell, to 5.1, 7.2, and 6.0 log CFU/ ml, respectively (Fig. 2). OL groups showed a weak antibacterial activity against L. monocytogenes compared with control. However, OSL groups showed a stronger antilisterial activity against L. monocytogenes than that of OL groups. In addition, OSL2, OSL2.5, and OSL3.0 were listericidal considering their changes from 4.6 log CFU/ml to 3.3, 3.0 or 2.9 log CFU/ml, respectively. Also there was significant difference in antibacterial activity against L. monocytogenes between OL and OSL under lysozyme at 2.5 mg/ml and 3.0 mg/ml (Fig. 2). Therefore, this study demonstrates that 100 mM sodium bicarbonate enhanced the antimicrobial activity of OL against L. monocytogenes, while lysozyme at 2.0, 2.5, or 3.0 mg/ml did not show clearly effect on antibacterial activity of OS toward L. monocytogenes. Payne et al. (1994) demonstrate that a combination of 15 mg/ml lactoferrin and 0.15 mg/ml lysozyme delayed the growth of L. monocytogenes in UHT (ultra high temperature) milk, whereas Boesman-Finkelstein and Finkelstein (1985) report that synergistic activity by lactoferrin plus lysozyme was not clearly found. Cultures inoculated with the strains at early exponential phase are relatively susceptible to lactoferrin, while those at the early stationary phase are comparatively resistant to the antimicrobial (Arnold et al., 1981). However, using multiple L. monocytogenes strains at stationary phase, and nutrient rich media resulted in a weak synergistic effect of lysozyme on antibacterial activity of ovotransferrin.

EDTA on antibacterial activity of ovotransferrin EDTA alone at 1.0, 1.5 or 2.0 mg/ml, OS, OE group, and OSE group showed < 0.1 optical density at 620 nm (Fig. 3). This confirmed that L. monocytogenes were susceptible to

180

EDTA as shown in disc test. As EDTA alone showed a lower turbidity than OE or OSE at initial point, it was impossible to compare the antibacterial activity of OE or OSE with EDTA itself by turbidity analysis (Fig. 3). As almost all treatments showed similar turbidity values of less than 0.1, it was difficult to determine if EDTA had a synergistic effect with ovotransferrin using turbidity test. However, according to turbidity test, EDTA increased the antibacterial activity of OS (Fig.3). In viability test, E1 and E2 resulted in a decrease in viable cell numbers (from 4.6 log CFU/ml to 4.1 and 4.0 log CFU/ml, respectively), whereas OSE1 and OSE2 occurred 1 log reduction in cell numbers (from 4.6 log CFU/ml to 3.1 and 3.1 log CFU/ml, respectively) (Fig. 4). Based on these results, L. monocytogenes was found to be more susceptible to EDTA than lysozyme under the concentrations used in this study. This result is consistent with that of Parente et al. (1998) who reported that the survivors of L. monocytogenes decreased when the concentration of EDTA increased. Considering that the survivors of OS treatment were 4.9 log CFU/ml and those of OSE were 3.1 log CFU/ml, EDTA addition to OS was found to result in 1.8 log reduction (Fig. 4). As the bactericidal action of OSE against L .monocytogenes was not large and EDTA itself had bacteriostatic activity against L. monocytogenes, EDTA seemed to have an additive effect on antibacterial activity of OS rather than a synergistic effect. Like the result shown at

this study, Al-Nabulsi and Holley (2006) reported that the combination consisted of NaHCO3, lactoferrin, and EDTA accomplished 4 log reduction. EDTA chelates divalent cations of outer membrane (Coughlin et al., 1983; Vaara, 1992; Boland et al., 2004), which enhanced the activity of lactoferrin through the destabilization of outer membrane of gram negative bacteria and increased membrane permeability. Negatively charged head groups of phosphatidylglycerol and caridiolipin, which are predominant phospholipids existing in cytoplasmic membrane of L. monocytogenes (O’Leary and Wilkinson), can be neutralized by divalent cations such as Mg2+ and Ca2+. As a result, more rigid membrane is formed by condensation of membrane lipids (Harwood and Russel, 1984) and became less efficient for nisin insertion and pore formation (Abee et al., 1994). As EDTA is combined with nisin and leucosin, their bactericidal effect against L. monocytogenes in TSBYE is enhanced (Parente et al., 1998)

181

probably due to the chelating property of EDTA to divalent cations that play a role to protect L. monocytogenes (Abee et al., 1994). Aguilera et al (2003) reported that transferrin family proteins including ovotransferrin interacted directly with cell membrane and resulted in dissipation of electrical potential inside cells. In conclusion, EDTA increased the antibacterial activity of OS against L. monocytogenes. However, antimicrobial effects by EDTA itself seemed to be the major contributor to the listericidal activity of OSE. Antibacterial activity of the combinations consisted of EDTA and OS, which were shown to be bacteriostatic, was likely to be ascribed to the additive interactions rather than synergistic effect.

Cell number effect OS treatment to BHI broth media inoculated with 104 and 105 CFU/ml occurred 1 log increase from the initial cell number during 35oC incubation for 48 hrs. However, the viable cells in BHI broth culture inoculated with 106 cells and treated with OS did not indicate any increase (Fig. 5). BHI broth culture added with OSL2 and inoculated with 104, 105 or 106 CFU/ml of L. monocytogenes resulted in 3.3, 3.7, or 4.5 log CFU/ml of viable cells, respectively, after 48 hrs incubation at 35oC. BHI broth inoculated with 104, 105 or 106 CFU/ml of L. monocytogenes cells and OSE1 showed 3.1, 4.0, or 4.4 log CFU/ml of viable cells, respectively (Fig 5). These results demonstrate that OSL2 or OSE1 had listericidal activity against L. monocytogenes and the number of initial cells existing in the media affected the antibacterial activity of OSL2 and OSE1 (p < 0.05). However, the influence by initial cell population on antimicrobial actions of ovotransferrin against L. monocytogenes was not overwhelming.

Ovotransferrin application on commercial hams L. monocytogenes with 104 initial cells grew slowly in commercial hams to 105 CFU/ml during storage at 10oC for 29 days. In contrast to in vitro test, 1OSL, 1OSLE, 2OSL, and 2OSLE did not show any antimicrobial activity against L. monocytogenes. No significant difference between control and all treatments were detected (Table 2). Therefore, it was suggested that even though OSL2 and OSE1 showed clear bactericidal effect against L.

182

monocytogenes in model systems, they did not have antibacterial activities against L. monocytogenes in commercial hams during storage. Antibacterial activities of some antimicrobials shown in buffer or laboratory media sometimes can be modulated in nutrient media or in presence of cations and salts, which are capable of supporting the growth of microorganisms (Branen and Davidson, 2000; Bellamy et al., 1992; Jones et al., 1994). Murdock et al. (2007) reported that the inhibitory activities of lactoferrin against L. monocytogenes could be eliminated or reduced in some media like trypticase soy broth since they provided many divalent cations that might affect the antibacterial activity of lactoferrin. Exogenous Ca2+ and Mg2+ may block the biological activity of lactoferrin by forming lactoferrin tetramers, increasing iron strength in the presence of high concentration of Ca2+ and Mg2+ or inhibiting binding to cell surface (Bennett et al, 1981; Viejo-Díaz et al, 2004; Ellison et al., 1988, 1990) Low level of EDTA synergistically enhanced the antibacterial activity of lysozyme against L. monocytogenes (Branen and Davidson, 2004). Payne et al. (1994) reported that more than 1 mg/ml EDTA synergistically enhanced the effect of 0.1 ~ 0.2 mg/ml lysozyme against L. monocytogenes in UHT milk. EDTA enhanced the antibacterial activity by lysozyme against L. monocytogenes in microbiological media as well as food products (Hughey and Johnson, 1987, Hughey et al., 1989). Johnson (1994) reported that antibacterial action of lysozyme is limited to a part of bacteria and fungi and could be improved by certain substances such as EDTA, butyl paraben, tripolyphosphate, and some other naturally occurring antimicrobial agents. However, EDTA plus lysozyme did not show any antimicrobial activity against L. monocytogenes in hams in this study. Such a difference may be due partially to media compositions. Antimicrobial combinations consisting of lysozyme and nisin had an enhanced bactericidal activity in MRS media, but they did not show such an effect in pork juice at the same conditions (Gill and Holly, 2003). As food products provide a nutrient-rich environment, injured microorganisms by antimicrobials may be recovered more easily in food systems than those under cell starvation or limited media conditions. Another possible explanation about no antibacterial activities by OSL and OSE in commercial hams could be distribution problem or dilution effect of ovotransferrin to the products, state of strains in inocula, or stability of ovotransferrin on the surface of meat

183

products. Generally, antimicrobial activity of ovotransferrin is dependent upon dose of ovotransferrin. In the present study, 1 ml of 20~ 30 mg/ml ovotransferrin was distributed on the surface of hams, and the concentration was likely to be diluted in certain part of the product so that low antibacterial effect against L. monocytogenes might be shown in the area. Also, ovotransferrin was spread manually, and thus it is difficult to distribute ovotransferrin uniformally or homogenously on whole products. Also, ovotransferrin incorporated could have been penetrated or absorbed inside the product so that it could not be involved in antimicrobial activities. Stiles (1996) demonstrated that even though antimicrobials such as bacteriocins are effective in inhibiting pathogens including L. monocytogenes and other spoilage microorganisms in model systems, high population of survivors or recovering features is often observed in food systems. This study showed that there are many limitations in applying ovotransferrin as antimicrobial agent in food systems at this point.

Conclusion Lysozyme at 2 mg/ml in BHI media delayed the growth of L. monocytogenes for 24 hrs at 35oC, but it started to adapt new environment. OL2 appeared to be bacteriostatic activity and OSL2 indicated to have bactericidal effect against L. monocytogenes in vitro test. Also, 100 mM sodium bicarbonate was found to enhanc the antimicrobial activity of OL against L. monocytogenes, while lysozyme at 2.0, 2.5, or 3.0 mg/ml did not show significant effect on antibacterial activity of OS toward L. monocytogenes. EDTA alone at 1 mg/ml was bacteriostatic against L. monocytogenes and OSE1 was bactericidal. However, it was difficult to determine whether EDTA has synergistic effect on ovotransferrin since EDTA itself had a significant antibacterial activity against L. monocytogenes in BHI broth culture. The initial cell number of cells in media affected the antibacterial activity of OSE or OSL toward L. monocytogenes, but the effect of cell population on antimicrobial action of ovotransferrin was not significant. Contrary to in vitro test, OSL, OSEL, L, and LE did not show any antimicrobial activities in commercial hams during incubation at 10oC for 29 days. There was apparent difference in antibacterial activities of ovotransferrin between model systems and real products. This study suggested that OSL2 had a great potential as a natural antimicrobial

184

agents to control L. monocytogenes in vitro, but further studies are needed to improve the antimicrobial activity of ovotransferrin in food products.

Reference Abee, T. Rombouts, F. M., Hugenholtz, J., Guihard, G., and Letellier, L. 1994. Mode of action of nisin Z against Listeria monocytogenes Scott A grown at high and low temperatures. Appl. Environ. Microbiol. 60 (6): 1962-1968. Aguilera, O., Quiros, L. M., and Fierro, J. F. 2003. Transferrins selectively cause ion efflux through bacterial and artificial membranes. FEBS Lett. 548: 5-10. Al-Nabulsi, A. A., and Holley, R. A. 2006. Enhancing the bactericidal effects of lactoferrin against E. coli O157:H7 by chelation, temperature, and NaCl. J. Appl. Microbiol. 100: 244-255. Arnold, R. R., Russell, J. E., Champion W. J., and Gauthier, J. J. 1981. Bactericidal activity of human lactoferrin: influence of physical conditions and metabolic state of the target microorganism. Infect. Immunol. 32: 655-660. Banin, E., Brady, K. M. and Greenberg, E. P. 2006. Chelator induced dispersal and killing Pseudomonas aeruginosa cells in a biofilm. Appl Environ Microbiol. 72: 2064-2069. Baron, F., Fauvel, S., and Gautier, M. 2000. Egg Nutrition and Biotechnology, Sim, J. S., Nakai, S., and Guenter, W. eds. CAB International. pp. 417-440. Baron, F., Gautier, M., and Brule, G. 1997. Factors involved in the inhibition of Salmonella enteritidis in liquid egg white. J. Food Protect. 60: 1318-1323. Bauer, A. W., Kirby, W. M. M., Sherris, J. C., and Turck, M. 1966. Antibiotic susceptibility

testing by a standardized single disk method. Amer. J. Clin. Pathol. 45:493-496. Bellamy, W., Takase, M., Wkabayashi, H., Kawase, K., and Tomita, M. 1992. Antibacterial spectrum of lactoferricin B, a potent bactericidal peptide derived from the N-terminal region of bovine lactoferrin. J. Appl. Bacteriol. 73: 472-479. Bennett, R. M., Bagby, G. C., and Davis, J. 1981. Calcium-dependent polymerization of Lactoferrin. Biochem. Biophys. Res. Commun. 101: 88-95. Boesman-Finkelstein, M., and Finkelstein, R. A. 1985. Antimicrobial effects of human milk. FEMS Microbiol. Lett. 27: 167-174.

185

Boland, J. S., Davidson, P. M, Bruce, B., and Weiss, J. 2004. Cations reduce antimicrobial efficacy of lysozyme-chelator combinations. J. Food Prot. 67: 285-294. Branen J. K., and Davidson, P. M. 2004. Enhancement of nisin, lysozyme and monolaurin antimicrobial activities by ethylenediaminetetraacetic acid and lactoferrin. Int. J. Food Microbiol. 90: 63-74. Branen, J., and Davidson, P. M. 2000. Activity of hydrolyzed lactoferrin against foodborne pathogenic bacteria in growth media: the effect of EDTA. 30: 233-237. Carminati, D., and Carini, S. 1989. Antimicrobial activity of lysozyme against Listeria monocytogenes in milk. Microbiol. Aliments Nutr. 7: 49-56. CDC: Centers for Disease Control and Prevention. 2002. Multistate outbreak of listeriosis- United States. Morb. Mortal. Wkly. Rep. 51: 950-951. Coughlin, R. T., Tonsager, S., and MaGroarty, E. J. 1983. Quantitation of metal cations bound to membranes and extracted lipopolysaccharide of Escherichia coli. Biochem. 22: 2002-2007 Davies, B. D., Dulbecco, R., Eien, H. N., and Ginsberf, H. S. 1980. In Microbiology 3rd ed. Hagerstown, Mariland: Harper & Row. pp. 564. Davis, C., and Reeves, R. 2002. High value opportunities from the chicken egg. A report for the rural industries research and development corporation. Rural industries Research and Development Corporation Publication. Ellison, R. T. III, Giehl, R, J., and LaForce, F. M. 1988. Damage of the outer membrane of enteric gram-negative bacteria by lactoferrin and transferrin. Infect. Immunol. 56: 2774- 2781. Ellison, R. T. III, LaForce, F. M., Giehl, T. J., Boose. D. S., and Dunn, B. E., 1990. Lactoferrin and transferrin damage of the gram-negative outer membrane is modulated by Ca2+ and Mg2+. J. Gen. Microbiol. 136: 1437-1446. FSIS: www. FSIS. USDA Gov/ oppde/larc. Gill A. O., and Holley R. A. 2003. Interactive inhibition of meat spoilage and pathogenic bacteria by lyozyme, nisin and EDTA in the presence of nitrite and sodium chloride at 24oC. Inter. J. Food Microbiol. 80: 251-259.

186

Giovannacci, I., Ragimbeau, C., Queguiner, S., Salvat, G., Vendeuvre, J. L., Carlier, V., and Ermel, G. 1999. Listeria monocytogenes in pork slaughtering and cutting plants. Use of RAPD, PFGE and PCR-REA for tracing and molecular epidemiology. Int. J. Food Microbiol. 53: 127-140. Hansen , L. T. Austin, J. W., and Gill , T. A. 2001. Antibacterial effect of protamine in combination with EDTA and refrigeration. Int. J. Food Microbiol. 66: 149-161. Harwood, J. L., and Russel, J. N. 1984. Lipids in plants and microbes. George Allen and Unwin, Ltd, London. Hughery, V. L., Wilger, P. A., and Johnson, E. A. 1989. Antibacterial activity of hen egg white lysozyme against Listeria monocytogenes Scott A in foods. Appl. Environ. Microbiol. 55: 631-638. Hughey, V. L., and Jonhson, E. A. 1987 Antimicrobial activity of lysozyme against bacteria involved in food spoilage and food-borne disease. Appl. Environ. Microbiol. 53: 2165- 2170. Ibrahim, H. R. 1997. Insights into the structure function relationship of ovalbumin, ovotransferrin and lysozyme. In: Hen eggs: Their Basic and Applied Science. Yamamoto T, Juneja IR, Hatta H, and Kim M. (Eds). CRC Press Inc., New York, pp. 37-56. Ibrahim, H. R., Sugimoto, Y., and Aoki, T. 2000. Ovotransferrin antimicrobial peptide (OATP-92) kills bacteria through a membrane damage mechanism. Biochim. Biophys. Acta 1523: 196-205. Jones, E. M., Smart, A., Bloomberg, G., Burgess, L., and Millar, M. R. 1994. Lactoferricin, a new antimicrobial peptide. J. Appl. Bacteriol. 77: 208-214. Johnson, E. A. 1994. Egg-white lysozyme as a preservative for use in foods. In: Egg Uses and Processing Technologies. New Developmemts. Sim, J. S. and Nakai, S. (Eds). International Cab, Wallingford. pp. 77-93. Kihm, D. J., Leyer, G. J., AN, G. H., and Johnson, E. A. 1994. Sensitization of heat-treated Listeria monocytogenes to added lysozyme in milk. Appl. Environ. Microbiol. 60 (10): 3854-3861. Kuehl, R. O. 2000. Design of Experiments; Statistical Principles of Research Design and Analysis. 2nd Ed. New York: Duxbury Press.

187

Kurokawa, H., Mikami, B., and Hirose, M. 1995. Crystal structure of diferric hen ovotransferrin at 2.4 Ǻ resolution. J. Mol. Biol. 254: 196-207. Lou, Y., and Yousef, A. H. 1999. Characteristics of Listeria monocytogenes important to food processors. In: Ryser, E.T., Marth, E.H, eds. Listeria, listeriosis and food safety. New York; Marcel Dekker Inc. pp. 134-224. Lunden, J. M., Autio, T. J., Sjoberg, A. M., and Korkeala, H. J. 2003. Persistent and nonpersistent Listeria monocytogenes contamination in meat and poultry processing plants. J. Food Prot. 66: 2062-2069. Monk, J. D., Beuchat, L. R., and Hathcox, A. K. 1996. Inhibitory effects of sucrose monolaurate, alone and in combination with organic acids, on Listeria monocytogenes and Staphylococcus aureus. J. Appl. Bacteriol. 81(1): 7-18. Murdock, C. A., Cleveland, J., Mattews, K. R., and Chikindas, M. L. 2007. The synergistic effect of nisin and lactoferrin on the inhibition of Listeria monocytogenes and Escherichia coliO157:H7. Lett. Appl. Microbiol. 44: 255-261.

O’Leary, W. M., and Wilkinson, S. G. 1988. Gram-positive bacteria. pp. 117-201. In In C. Ratledge and S. G. Wilkinson (ed.), Microbial lipids, vol. 1. Academic Press, London, Great Britain. Parente, E., Giglio, M. A., Ricciardi, A., and Clementi, F. 1998. The combined effect of nisin, leucocin F10, pH, NaCl and EDTA on the survival of Listeria monocytogenes in broth. Inter. J. Food Microbiol. 40: 65-75. Parente, E., Moles, M., and Ricciardi, A. 1996. Leucocin F10, a bacteriocin from Leuconostoc carnosum. Int. J. Food Microbiol. 33: 231-243. Payne, K. D., Oliver, S. P., and Davidson, P. M. 1994. Comparison of EDTA and apo- lactoferrin with lysozyme on the growth of food borne pathogenic and spoilage bacteria. J. Food Prot. 57: 62-65. Proctor, V. A., and Cuuningham, R. E. 1988. The chemistry of lysozyme and its use as a food preservative and a pharmaceutical. Crit. Rev. Food Sci. Nutr. 26: 359-395 Rocourt, J., and Seeliger, H. P. R. 1985. Distribution des sepecies du genre Listeria. Zentral. Bakteriol. Mikrobiol. Hyg. A. 259: 317-330.

188

Shade, A. L., and Caroline, L. 1944. Raw Hen Egg White and the role of iron in growth inhibition of Shigella dysenteriae, Staphylococcus aureus, Escherichia coli and Saccharomyces cerevisiae. Sci. 100: 14-15. Shelef, L. A., and Seiter, J. 1993. Enhancement of the activity of novobiocin against Escherichia coli by lactoferrin. J. Dairy Sci. 82: 492-499. Sofos, J. N., Beuchat, L. R., Davidson, P. M., and Johnson, E. A., 1998. Naturally occurring antimicrobials in food. Task force report No. 132. Council for Agricultural Science and Technology, Ames, IA. pp. 103. Spitznagel, J. K. 1984. Non oxidative antimicrobial reactions of leucocytes. In Regulation of leucocytes Function. Contemporary Topics in Immunol. 14: 283-343. Stiles, M.E., 1996. Biopreservation by lactic acid bacteria. Antonie van Leeuwenhoek 70: 331-345. Uyttendaele, M. De Troy, P., and Debevere, J. 1999. Incidence of Listeria monocytogenes in different types of meat products on the Belgian retail market. Int. J. Food Microbiol. 53: 75-80. Vaara, M. 1992. Agents that increase the permeability of the outer membrane. Microbiol. Rev. 56: 395-411. Valenti, P., Antonin. G., Von Hunolstein, C., Visca, P., Orsi, N., and Antonini, E. 1983. Studies of the antimicriobial activity of ovotransferrin. Inter. J. Tissue Reac. 1: 97-105. Valenti, P., Visca, P., Antonini, G., and Orsi, N. 1985 Antifungal activity ovotransferrin towards genus Candida. Mycophathologia 89: 169-175. Viejo-Díaz, M., Andrés, M. T., and Fierro, J. F. 2004. Modulation of in vitro fungicidal activity of human lactoferrin against Candida albicans by extracellular cation concentration and target cell metabolic activity. Antimicrobial agents and chemotherapy 48: 1242-1248. Zhang, S. S., and Mustapha, A. 1999. Reduction of Listeria monocytogenes and Escherichia coli O157:H7 numbers on vacuum-packaged fresh beef treated with nisin or nisin combined with EDTA. J. Food Protect. 62(10): 1123-1127.

189

Table 1. Effect of ovotransferrin combined with NaHCO3 and EDTA or NaHCO3 and lysozyme on clear zone formation on BHI agar media plates inoculated with L. monocytogenes during 35o C incubation for 24 ~ 48 hrs

a b Treatment Conc. (mg/ml) Control OTF OTF + NaHCO3

0 -* - -/+ 1.0 - - - 1.5 - +/- - Lysozyme 2.0 ++ ++ +/- 2.5 ++ ++ ++ 3.0 ++ +++ ++

1.0 - - - EDTA 1.5 ++ +** + 2.0 ++ + ++

a P 20 mg/ml ovotransferrin b P 100 mM NaHCO3 was used at this study. * P No clear zone formed on BHI agar media plate. ** P Formation of clear zone: the strength of clear zone was described as + (weak), ++ (middle), and +++ (strong) by its size or clarity on the BHI agar. (n = 3)

Table 2. Viable cells of L. monocytogenes on commercial hams treated with or without ovotransferrin in 100 mM o NaHCO3 (OS) plus or/and lysozyme (2 mg/ml) and/or EDTA (1 mg/ml) during 10 C storage

Treatments Storage period (days) 0 2 5 10 15 22 29 ------Number of viable cells (log 10 CFU/ml) ------Control 4.5a ± 0.0* 4.6 a ± 0.2 5.0 a ± 0.0 5.3 a ± 0.4 5.8 a ± 0.0 5.4 a ± 0.2 5.3 a ± 0.4 1OSL 4.3 a ± 0.1 4.7 a ± 0.1 5.1 a ± 0.1 5.5 a ± 0.1 5.6 a ± 0.4 5.5 a ± 0.9 5.2 a ± 0.3 1OSLE 4.4 a ± 0.1 4.6 a ± 0.0 5.0 a ± 0.1 5.1 a ± 0.2 5.9 a ± 0.3 5.6 a ± 0.2 6.0 a ± 0.4 2OSL 4.4 a ± 0.0 4.6 a ± 0.0 5.0 a ± 0.1 5.2 a ± 0.2 5.2 a ± 0.1 5.5 a ± 0.0 5.4 a ± 0.6 2OSLE t 4.3 a ± 0.1 4.6 a ± 0.1 4.9 a ± 0.2 5.1 a ± 0.1 5.6 a ± 0.2 5.7 a ± 0.3 5.4 a ± 0.2 190 L 4.3 a ± 0.0 4.5 a ± 0.1 4.9 a ± 0.1 4.9 a ± 0.3 5.4 a ± 0.3 5.0 a ± 1.0 5.4 a ± 0.5 LE 4.4 a ± 0.2 4.5 a ± 0.1 5.0 a ± 0.1 5.1 a ± 0.1 5.5 a ± 0.5 5.6 a ± 0.3 6.2 a ± 0.6 aMeans within a column with no common superscript differ (P < 0.05, n = 4). Control: only L. monocytogenes inoculated. *Mean ± standard deviation.

1OSL: 20 mg/ml ovotransferrin + 100 mM NaHCO3 + 2 mg/ml lysozyme

1OSLE: 1OSL + 1 mg/ml EDTA , 2OSL: 30 mg/ml ovotransferrin + 100 mM NaHCO3 + 2 mg/ml lysozyme 1OSLE: 2OSL + 1 mg/ml EDTA, L: 2 mg/ml lysozyme LE: 2 mg/ml lysozyme + 1 mg/ml EDTA

191

Fig.1. Turbidity of BHI broth cultures inoculated L. monocytogenes and ovotransferrin (20 o mg/ml) combined with NaHCO3 or/and EDTA during 35 C incubation

1.2

0.8

0.6

0.4

0.2 Turbidity (at 620nm) 0

-0.2 0 1020304050

Time (hrs)

Control L2 L2.5 L3 OS OL2 OL2.5 OL3 OSL2 OSL3 OSL2.5

C: control, 104 CFU/ml of L. monocytogenes L2: 2 mg/ml Lysozyme L2.5: 2.5 mg/ml Lysozyme L3: 3.0 mg/ml Lysozyme

OS: 20 mg/ml ovotransferrin + 100 mM-NaHCO3 OL2: 20 mg/ml ovotransferrin + L2 OL2.5: 20 mg/ml ovotransferrin + L 2.5 OL3: 20 mg/ml ovotransferrin + L 3.0 OSL2: OS + 2 mg/ml lysozyme OSL2.5: OS + 2.5 mg/ml lysozyme OSL3: OS+ 3.0 mg/ml lysozyme

192

Fig 2. Antibacterial activity of ovotransferrin (20 mg/ml) combined with 100 mM NaHCO3 and/or lysozyme against the growth of L. monocytogenes on BHI broth culture during 35oC incubation for 24 ~ 36 hr

12.0 11.0 a a a 10.0 a 9.0 ab 8.0 b 7.0 bc bcd 6.0 CFU ml / cd d 10 5.0 d 4.0

Log 3.0 2.0 1.0 0.0 1 C L2 L2.5 L3 OL2 OL2.5 OL3 OS OSL2 OSL2.5 OSL3

a-d Bars with different letters indicate significantly different values (P < 0.05. n=4). C: control, only 104 CFU of L. monocytogenes L2: 2 mg/ml Lysozyme L2.5: 2.5 mg/ml Lysozyme L3: 3.0 mg/ml Lysozyme OL2: 20mg/ml ovotransferrin + L2 OL2.5: 20mg/ml ovotransferrin + L2.5 OL3: 20mg/ml ovotransferrin + L 3.0

OS: 20 mg/ml ovotransferrin + 100 mM-NaHCO3 OSL2: OS + 2 mg/ml lysozyme OSL2.5: OS + 2.5 mg/ml lysozyme OSL3: OS+ 3.0 mg/ml lysozyme

193

Fig 3. Turbidity of BHI broth cultures inoculated L. monocytogenes and ovotransferrin (20 o mg/ml) combined with NaHCO3 or/and EDTA during 35 C incubation

0.10

0.08

0.06

0.04

0.02

0.00

-0.02 Optical density at 620nm at density Optical -0.04 0 102030405060 Time (hrs)

E1 E1.5 E2 OE1 OE1.5 OE2 OSE1 OSE1.5 OSE2 OS

C: control, only 104 CFU / ml of L. monocytogenes E1: 1 mg/ml EDTA E1.5: 1.5 mg/ml EDTA E2: 2.0 mg/ml EDTA

OS: 20 mg/ml ovotransferrin + 100 mM NaHCO3 OE1: 20 mg/ml ovotransferrin + E1 OE1.5: 20 mg/ml ovotransferrin + E2 OE2: 20 mg/ml ovotransferrin + E3 OSE1: OS + E1 OSE1.5: OS + E1.5 OSE2: OS+ E2

194

Fig 4. Antibacterial activity of ovotransferrin (20 mg/ml) combined with 100 mM NaHCO3 and/or EDTA (1 mg/ml) against the growth of L. monocytogenes in BHI broth culture during 35oC incubation for 48 hr

10.0 a 9.0 8.0 7.0 6.0 b 5.0

CFU / ml CFU / cc 10 4.0 dd

Log 3.0 2.0 1.0 0.0 1 C E1 E2 OS OSE1 OSE2

a-d Bars with different letters indicate significantly different values (P < 0.05. n=4). C: control, only 104 cells of L. monocytogenes E1: 1 mg/ml EDTA E2: 2.0 mg/ml EDTA

OS: 20 mg/ml ovotransferrin + 100 mM NaHCO3 OSE1: OS + E1 OSE2: OS + E2

195

Fig 5. Effect of initial cell population on the antibacterial activity of ovotransferrin (20 mg/ml) solution combined with 100 mM NaHCO3 and EDTA (2 mg/ml) or lysozyme (1 mg/ml) against L. monocytogenes in BHI broth culture during 35oC incubation for 48 hr

8.0 7.0 a a 6.0 b b 5.0 bcd bc de cde 4.0 e CFU / ml / CFU

10 3.0 2.0 Log 1.0 0.0 OS (A) OSL2 (A) OSE11 (A) OS (B) OSL2 (B)

OSE1 (B) OS (C) OSL2 (C) OSE1 (C)

a-d Bars with different letters indicate significantly different values (P < 0.05. n=4). A: 104 CFU/ ml BHI broth B: 105 CFU/ ml BHI broth C: 106 CFU/ ml BHI broth

OS: 20 mg/ml ovotransferrin + 100 mM NaHCO3 OSL2: OS + 2 mg/ml lysozyme OSE1: OS + 2 mg/ml lysozyme + 1 mg/ml of EDTA

196

CHAPTER 8. GENERAL CONCLUSIONS

Foodborne illnesses are a significant problem and a major public health concern. A number of methods including the use of synthetic and natural antimicrobial agents have been developed to control the growth of pathogens in foods, but there are growing consumer demands for ‘natural’ antimicrobials or methods, which are environmentally friendly, medically acceptable, highly effective, and economical to manufacture. Egg contains various antimicrobial proteins including lysozyme, egg yolk antibody (IgY), and ovotransferrin. As IgY has a great potential for use as a natural antimicrobial agent, this study developed a novel protocol for a large-scale production of IgY from egg yolk. The new protocol used 0.01 % charcoal to remove lipoproteins, ultrafiltration to concentrate IgY solution, and 40% ammonium sulfate to precipitate IgY. The recovery of IgY produced by the protocol was around 70 ~ 80% and its purity were much higher than that of IgY obtained by anion exchange chromatography. As ammonium sulfate precipitation method could handle a large volume of IgY solution at a time, the new procedure was suitable for a large-scale purification of IgY from egg yolk. Ovotransferrin that has a high potential as a natural antimicrobial agent was separated from egg white using developed protocol. That is, apo-ovotransferrin in egg white was saturated with iron to prevent its denaturation during ethanol treatment. Holo-ovotransferrin was extracted by 43% ethanol and precipitated by 59% ethanol from iron-saturated egg white solution. After dissolving precipitated holo-ovotransferrin with distilled water, the iron bound ® to holo-ovotransferrin was removed using AG 1-X2 ion exchange resins so that apo- ovotransferrin was obtained. The recovery rate of apo-ovotransferrin was around 94%, whereas the purity was greater than 80%. The developed method was simple and economical for a large-scale production of ovotransferrin because the protocol used only ethanol and AG ® 1-X2 ion exchange resins and they can be recovered and reused. In addition, the ovotransferrin obtained from this protocol can be easily applied of food products because only ethanol was used as a solvent for separation of apo-ovotransferrin. Ovotransferrin was originally thought to inhibit microbial growth by limiting iron from the environment. However, the precise mechanism of antibacterial action of ovotransferrin

197

was not defined clearly. This study investigated the effect of citric acid, NaHCO3 on the antibacterial activity of ovotransferrin against E. coli O157:H7 and L. monocytogenes to understand the antimicrobial action of ovotransferrin. The antibacterial activities of Apo-, Zn-, and Fe-bound ovotransferrin against E. coli O157:H7 and L. monocytogenes were also investigated. The antibacterial activity of natural apo-ovotransferrin against E. coli O157:H7 and L. monocytogenes in model systems increased as the concentration of sodium

bicarbonate increased. Ovotransferrin combined with 100 mM-NaHCO3 (OS) resulted in marked increase in antibacterial activity against the two pathogens. Iron binding capacity of ovotransferrin was generally decreased under acidic pH. Also, the antibacterial activity of ovotransferrin against E. coli O157:H7 increased when ovotransferrin was combined with 0.5% citric acid even though the pH of media decreased to pH 4 ~ 5 by citric acid. The growth of L. monocytogenes was prevented by Zn-saturated ovotransferrin, but E. coli O157:H7 was not. These results suggested that iron binding capacity was not the major antimicrobial mechanism of ovotransferrin because acidic condition formed by addition of citric acid improved antibacterial acticity of ovotransferrin against E. coliO157:H7. Instead, ovotransferrin was likely to interact directly with bacterial membrane and induced a variety of physicochemical changes, which affected the survival of microorganisms. Ovotransferrin alone and OS showed limited antibacterial effect on the growth of E. coli O157:H7 and L. monocytogenes, but addition of EDTA improved the antibacterial activity of ovotransferrin or OS against two pathogens. EDTA (2 mg/ml) plus OS (OSE) caused 3 ~ 4 log reduction in E. coli O157:H7, while EDTA (1mg/ml) plus OS resulted in around 1 log reduction of L. monoyctogenes during after 36 hr incubation at 35oC. But it was difficult to determine whether EDTA has a synergistic effect with ovotransferrin since EDTA itself showed bacteriostatic property against L. monocytogenes. Through this study, use of OSE seemed to be an effective approach to control E. coli O157:H7. Also, the effect of lysozyme on antibacterial activity of OS against E. coliO157:H7 and L. monocytogenes was not clearly found in the present study. However when lysozyme at 2.5 or 3.0 mg/ml lysozyme was combined with OS, the combinations showed to have bactericidal activity against L. monocytogenes, resulting in 1 log reduction of initial cell number in vitro test.

198

Contrary to the results obtained in model systems using BHI media, ovotransferrin

plus 100 mM NaHCO3 or 0.5% citric acid and ovotransferrin plus EDTA or/and lysozyme (OSE or OSL) did not show significant antibacterial activity against E. coli O157:H7 and L. monocytogenes in pork chops and hams. Such differences in antibacterial activity of ovotransferrin combined with other antimicrobials between model systems and real products seemed to be ascribed to different media compositions such as the presence of many divalent cations, distribution problems, and low stability of unbound ovotransferrin on the surface of meat products. In conclusion, this study suggested that there is high possibility of using ovotransferrin and its combinations with NaHCO3, EDTA, or lysozyme as natural antimicrobial agents to control E. coli O157:H7 and L. monocytogenes in foods. However, elucidation of the precise antibacterial mechanisms of ovotransferrin and improvements in antibacterial property of ovotransferrin and its combinations with other agents is needed for their use in food systems.

199

AKNOWLEDGEMENTS

I deeply appreciate my major professor, Dr. Dong Uk Ahn. I am thankful to him for giving the opportunity to study in Meat Science Program. His guidance, patience, encouragement, and warmth made me greatly accomplish the Ph. D. study. Also, I thank to learn his faithful attitude about the research. One of thanks toward my advisor is that I have gained fast response and useful idea due to his diligence whenever I need his advice and comments. Wherever I go after graduation, I will pray to God for him and his lab as or like always. Also, I would like to thank my Ph. D committee members, Dr, Joseph G. Sebranek, Dr. James S. Dickson, Dr. Joan E. Cunnick, and Dr. Elisabeth J. Huff-Lonergan. They gave invaluable comments and guidance to my research and helped me sincerely to accomplish Ph. D program. I am grateful to my laboratory colleagues: Dr. Kichang Nam, Dr. Eun Joo Lee, Dr. Byungrok Min, Hesham, Dr. Soo Min Kim, Dr. Hojin Kang, and others. Due to their help and support, I could finish with satisfaction my research. Also, I want to thank my sisters and brothers, who have given me endless love and financial support until now. Their care and love gave me many tips in realizing a sound and plentiful life. Especially, my oldest brother, Sang Young Ko, is such a wonderful and warm- heart person and has taken care of me like his child. Due to his sincere love, I could endure and be patient in the difficulties that I faced during the Ph. D. degree. Moreover, I appreciate deeply my forever spiritual friends, Ji-Eun Hong and Ki-In Rye. They have helped me readily whenever I need and always encouraged me sincerely. No matter what I was facing, they always have stood by me and prayed for me. Their financial support and confidence toward me became great power in achieving the Ph.D. study. Also, I am truly grateful to my sincere mentors, minister, Ji-Yeon Choi and Bessie Fick. They are sincerely mentors who God sent to me and have taught and induced me to go to a right way with praying for me. Also, I am thankful to Jerry Mattew, who has taken care me whenever I had some trouble and showed me his dedicated life toward other people. Finally, I would like to thank a number of friends including Ames Korean Christian Reformed Church members who have been praying and helping for me during Ph. D degree.