Content uploaded by Brian W Sheldon
Author content
All content in this area was uploaded by Brian W Sheldon
Content may be subject to copyright.
Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
ABSTRACT
AHLBORN, EUGENE JOSEPH. Identifying the Components in Eggshell Membranes
Responsible for Reducing the Heat Resistance of Bacterial Pathogens. (Under the
direction of Dr. Brian W. Sheldon.)
In 2001 it was discovered that exposure to egg shell membranes (ESM)
significantly reduced the thermal resistance and/or inhibited selected Gram-positive and
Gram-negative foodborne bacterial pathogens. Methods to extract these enzyme-rich
shell membranes are readily available and offer egg processors potential economic value
as a value-added product and provide processors with a ‘natural’ antibacterial adjuvant to
sensitize bacterial pathogens and spoilage organisms to other stresses for food or
pharmaceutical applications. However, a greater understanding of the egg shell
membrane components that deliver antimicrobial activity is critical with respect to
developing a better understanding of how it may be used in practical applications.
The first research objective focused on extracting enzymatically and biologically
active ESM protein fractions in order to evaluate their specific biological role in
providing defense against bacterial pathogens. Numerous attempts to purify active
fractions of β-N-acetylglucosaminidase (β-NAGase), lysozyme and ovotransferrin from
the ESM proved somewhat limited; however, isoelectric focusing and ion exchange
chromatography provided a technical means whereby relatively pure protein/enzyme
samples were obtained. SDS-PAGE electrophoresis and Matrix-Assisted Laser
Desorption Time-of-Flight Mass Spectrometry were used to partially characterize the
individual proteins. The results showed that hen egg white (HEW) β-N-
acetylglucosaminidase was isolated using a two-step chromatographic procedure
(isoelectric focusing followed by cation exchange chromatography), during which
relatively pure fractions of ovotransferrin were also obtained. Purified HEW and ESM
fractions, such as reported in this study, will be required in order to fully elucidate the
mechanism responsible for the antimicrobial properties of the ESM.
The second research objective addressed the identification of ESM components
that reduce the heat resistance of bacterial pathogens. Also, we attempted to define the
possible mechanism of action. Decimal reduction times (D54°C-values) for Salmonella
enterica serovar Typhimurium were determined using various preparations of ESM and
purified ESM and HEW proteins. Ovotransferrin, lysozyme and β-NAGase were
identified as primary contributors to ESM antimicrobial activity. Additionally, cellular
loss of Ca2+, Mg2+, Na+ and K+ from S. Typhimurium cells, after treatment with ESM
components was observed. Transmission electron microscopy indicated that the integrity
of the outer membrane was compromised when a combination of ovotransferrin,
lysozyme and β-NAGase was applied to the cells. Thus, we hypothesized that ESM-
bound ovotransferrin functions as an ion chelator and disrupts lipopolysaccharide
stability, allowing lysozyme and β-NAGase greater access to the peptidoglycan, resulting
in a compromised outer membrane and increasing susceptibility to heat and possibly
other stressors (i.e. pressure, osmotic changes).
The third objective of this research was to determine if there are differences in
enzymatic and biological activity of ESM as a function of layer breeds, age and ESM
stabilization methods. The D54°C-values measured for ESM-treated S. Typhimurium were
not adversely impacted by layer breed or age, indicating that ESM biological activity
remains fairly constant across different layer and breeds and throughout the laying cycle.
Significant reductions in biological activity was not observed in refrigerated (4°C), frozen
(-20°C) or freeze-dried ESM, however significant reductions in the enzymatic (16 to
31%) and biological (80 to 90%) activity were observed for air and heat dried ESM,
respectively. Refrigerated and frozen ESM also exhibited gradual loss of enzymatic
activity during a 6 month storage study, where as air, heat and freeze-dried ESM showed
no significant loss of enzyme activity after processing treatments.
The enzymatic and biological activities reported in this dissertation should
warrant further research with respect to evaluating the potential commercial value of
eggshell membranes as a processing adjuvant in food and pharmaceutical applications.
IDENTIFYING THE COMPONENTS IN EGGSHELL MEMBRANES
RESPONSIBLE FOR REDUCING
THE HEAT RESISTANCE OF BACTERIAL PATHOGENS
By
EUGENE JOSEPH AHLBORN
A dissertation submitted to the Graduate Faculty of
North Carolina State University
in partial fulfillment of the
requirements for the Degree of
Doctor of Philosophy
DEPARTMENT OF FOOD SCIENCE
Raleigh, 2005
APPROVED BY:
_________________________ _________________________
Dr. Sophia Kathariou Dr. Todd R. Klaenhammer
_________________________ _________________________
Dr. Harold E. Swaisgood Dr. Brian W. Sheldon
Chair of Advisory Committee
ii
BIOGRAPHY
Eugene Joseph Ahlborn, the son of Dr. Ernest S. and Dorothy N. Ahlborn was
born on October 25, 1969 in Ceder City, Utah. After a family move in 1974, the author
was raised (along with his seven siblings) in Twin Falls, Idaho. He graduated from Twin
Falls High School in 1988 and enrolled at Brigham Young University (Provo, UT). He
married Sheri M. Broadhurst March 1, 1996 and shortly after began his family. He
graduated with a Bachelor of Science degree in Nutritional Science in 1998 and later
earned a Master of Science degree in Food Science under the direction of Dr. Clayton
Huber. After graduation in December, 2001, he accepted a research assistantship at
North Carolina State University and began pursuit of a Doctor of Philosophy degree in
Food Science under the direction of Dr. Brian W. Sheldon.
iii
ACKNOWLDEGEMENTS
I would like to thank my advisor, Dr. Brian Sheldon, for all his assistance and
guidance during the course of my degree. I am especially grateful for the ‘open-door’
policy provided to me. I express my gratitude to Dr. Sophia Kathariou, Dr. Robert Kelly,
Dr. Todd Klaenhammer and Dr. Harold Swaisgood for their input and assistance as part
of my Ph.D. graduate committee. I would also like to thank Dr. Debbie Clare for her
continual assistance throughout my work here at NCSU. Although not an actual member
of my graduate committee, she will be considered a member in my eyes.
Appreciation is extended to Dr. Allen Foegeding and Paige Luck for allowing me
to invade their laboratory for much of my work. I am also grateful to the entire Food
Science Faculty and Staff for all their assistance and encouragement. The camaraderie
and friendships will be missed, but never forgotten. I would also like to acknowledge the
financial support I received during my degree through the North Carolina Agriculture
Foundation and the American Egg Board through their Research Fellowship Program.
Lastly, I would like to express my sincere gratitude and love to my wife, Sheri.
Thank you for your love and support during this time. Both Sheri and I thank our
parents, Ernest and Dorothy Ahlborn and Winfield and the late Carolyn Broadhurst for
their encouragement, love and support. Thank you all.
iv
TABLE OF CONTENTS
Page
LIST OF TABLES…………………………………...……………………………..……vii
LIST OF FIGURES………………………………………...……….….…..…...…..…..viii
INTRODUCTION……………………………………………………..……...………..…1
LITERATURE
REVIEW…………………………………………………………………..…..……….….4
I. Eggs and Egg Shell Waste……………………………….……….……..………..…4
II. Eggshell Function and Composition…………...…………………….…………….5
A. Natural Defenses of the Egg Against Microbial Invasion…...…...……………5
B. Components of the Eggshell Membrane……………….….…..…………..…...6
C. Chemical Defenses of the Hen’s Egg…….…………….…..….………….…..10
III. The Bacterial Cell Wall……………………………………………………..........11
A. Peptidoglycan……………………………………………..….….…….…...…13
B. Outer Membrane and Lipopolysaccharide Layer…………..….…...…………15
C. Outer Membrane Integrity.……………………...…...…….………………….17
D. Lethal Effects of Heat on Bacteria…….…………………………..….…….18
E. High Pressure Processing………………...…………………………………..…….…20
IV. Lysozyme………………………………………………………………..…..……21
A. Lysozyme………………………………………………………………..….….21
B. Properties and Activity of Lysozyme……….………………………..….….….24
C. Bacterial Sensitivity to Lysozyme…………….……………………….…...…..25
D. Lysozyme and Food…………………......……………………………....……..27
v
V. β-N-acetylglucosaminidase……………………………………………………..…28
A. β-N-acetylglucosaminidase………………………….……………………..…28
B. β-N-acetylglucosaminidase from Hen Eggs………………….…………….....29
C. β-N-acetylglucosaminidase Activity…………………………………….……31
VI. Ovotransferrin……………………………………………………………..….…..32
A. Ovotransferrin……………………………………………………………...…32
B. Action of Ovotransferrin on Bacteria…….………………………………..….33
VII. Food Preservation and the Hurdle Concept…………………………………..…34
VIII. Conclusion……………………………………………………………………...35
REFERENCES………………………………………………………………………….37
MANUSCRIPT 1………………………………………………………………….…......47
Title Page: Identification of Eggshell Membrane Proteins and Purification
of Ovotransferrin and B-NAGase from Hen Egg White……….…....48
Abstract………………………………………………………………………..…49
Introduction…………………………………………………………………..…..50
Materials and Methods………………………………………………………..….53
Results and Discussion….…………………………………………………….….60
References………………………………………………………………………..68
MANUSCRIPT 2………………………………………………………………………...80
Title Page: Identifying the Components in the Eggshell Membrane Responsible
for Reducing the Heat Resistance of Bacterial Pathogens……….…....81
Abstract………………………………………………………………………..…82
Introduction…………………………………………………………………..…..84
Materials and Methods………………………………………………………..….87
Results and Discussion……………………………………………………….….98
Conclusions………………………………………………….…………….……106
References…………………………………………………...………………….108
vi
MANUSCRIPT 3………………………………………………………………….…....121
Title: Enzymatic and Biological Activity in Egg Shell Membranes
as Influenced by Layer and Storage Variables…………………………..122
Abstract………………………………………………...……………………….123
Introduction…………………………………………………………………..…125
Materials and Methods……………………………………………………….…126
Results and Discussion…………………………………………………………133
Conclusions……………………………………………………………………..137
References………………………………………………………………………139
vii
LIST OF TABLES
MANUSCRIPT 1:
Table I. Protein extraction data of ESM with various chemicals……..……..……71
Table II. Purification table of β-NAGase from HEW….………….….…..……….72
MANUSCRIPT 2:
Table 1. Enzymatic and biological activities of ESM components………..…..112
Table 2. D-values (min) for purified enzyme fractions…………………..……113
Table 3. Extracellular ion supernatant concentrations of Salmonella
Typhimurium treated (37°C, 30 min) with eggshell membranes
(ESM), ovotransferrin, or a combination of ovotransferrin, lysozyme
and β-N-acetylglucosaminidase (β-NAGase)………………………………………….114
viii
LIST OF FIGURES
LITERATURE REVIEW: Page
Figure 1. Diagram of the egg shell and egg shell membrane structure………...…7
Figure 1. Scanning electron micrograph of the egg shell and egg shell
membrane……………………………………………………………………...….7
Figure 2. Scanning electron micrograph of the inner eggshell membrane….....…9
Figure 3. Profiles of the cell envelope for Gram-positive and
Gram-negative bacteria ......……………………………………………..….……12
Figure 5. Caricature depictions of the structural makeup and components
of Gram-positive and Gram-negative cell walls……………………………...….12
Figure 6. Schematic diagram of a peptidoglycan sheet……………………...….14
Figure 7. Mechanism of action for HEW lysozyme……….………………..…..23
MANUSCRIPT 1:
Figure 1. SDS-PAGE of ESM bound proteins extracted under various
conditions…………………………………………………………………..…….73
Figure 2. (a and b) MALDI-TOF analysis of ESM proteins.....…………….…...74
Figure 3. pH gradient and absorption (400nm) of β-NAGase activity
obtained through isoelectric focusing of hen egg white…………………………75
Figure 4. Cation-exchange chromatograph for purification of
ovotransferrin and β-N-acetylglucosaminidase………………………………….76
Figure 5. SDS-PAGE of hen egg white protein purification………………….…77
ix
Figure 6. (a) MALDI-TOF analysis of purified HEW ovotransferrin at a
full scale and (b) a partial scale ….……………………………………………...78
Figure 7. (a) MALDI-TOF of β-N-acetylglucosaminidase fraction at
full scale and (b) a partial scale…………..………………………………...…….79
MANUSCRIPT 2:
Figure 1. Mean D54°C-values (minutes) (n=3) for Salmonella Typhimurium
following treatment (incubation at 37°C for 30 min) with various
concentrations of egg shell membranes (grams) to bacterial
suspensions (ml)……………………………………………………………...…115
Figure 2. Mean D54°C-values (minutes) (n=3) for Salmonella Typhimurium
following treatment (incubation at 37°C for 30 min) with treated egg shell
membranes (1:20 ratio of g ESM to mls of bacterial suspension)……..……….116
Figure 3. Mean D54°C-values (minutes) (n=3) for purified egg white
ovotransferrin (Ovotrans), lysozyme, ovotransferrin and lysozyme, and an
ovotransferrin, lysozyme, and β-N-acetylglucosaminidase (B-NAGase) treatments
compared to no treatment (control) and eggshell membrane
(ESM)-treated Salmonella Typhimurium………………………………………117
Figure 4. Transmission electron micrographs of treated (37°C, 30 min)
Salmonella Typhimurium at 32,000 x magnification ……………………….....118
Figure 5. Transmission electron micrographs of treated (37°C, 30 min)
Salmonella Typhimurium at 6,000 x magnification ……………….………..…119
Figure 6. SDS-PAGE of Salmonella Typhimurium cell supernatant after
treatment with ESM or purified enzyme components………………………….120
MANUSCRIPT 3:
Figure 1. Comparison of the enzymatic activity of lysozyme and
β-N-acetylglucosaminidase in eggshell membranes from White
Leghorn (WL) and Rhode Island Red (RIR) layers at 25-27 and
78-80 weeks of age……………………………………………………………..141
Figure 2. Comparison of the enzymatic activity of lysozyme and
x
β-N-acetylglucosaminidase versus biological activity [D-values (min)
of ESM-treated Salmonella Typhimurium (37°C, 30 min) followed
by heat inactivation (54°C)] of eggshell membranes from White Leghorn
layers at 33, 50 and 81 weeks of age……………………………………..…….142
Figure 3. β-N-acetylglucosaminidase activity as influenced by
membrane stabilization method and storage time………………………………143
Figure 4. Lysozyme activity as influenced by membrane stabilization
method and storage time………………………………….…………………….144
Figure 5. D-values (min) for ESM-treated Salmonella Typhimurium
(37°C, 30 min) followed by heat inactivation (54°C) as influenced by
membrane stabilization method and storage time……………………………....145
Introduction
In 1999 Mead and colleagues (1999), in conjunction with the Centers for Disease
Control and Prevention, presented their data on the incidences of foodborne illnesses in
the United States. They estimated that over 76 million cases of foodborne illness occur
annually resulting in over 325,000 hospitalizations and over 5,000 deaths. These
staggering figures come from a country that boasts itself as a leading producer of the
world’s safest food. Even more alarming is that some consider these numbers to be an
underestimation.
Although more than 250 foodborne diseases have been described, known bacteria
(e.g., Campylobacter, Salmonella, E. coli O157:H7), viruses (e.g., Norwalk-like,
caliciviruses), and parasites (e.g., Giardia, Cyclospora) account for an estimated 14
million illnesses. It is difficult to calculate the exact cost as a result of such illnesses;
however, medical costs and lost wages due to foodborne salmonellosis, only one of many
foodborne infections, have been estimated to be more than $1 billion per year (CDC,
2003).
Implementing preventative means such as Good Manufacturing Practices
(GMPs), Good Agricultural Practices (GAPs), and Hazard Analysis Critical Control
Points (HACCP) have certainly made a positive impact at reducing the risks of foodborne
illnesses. In 2002, a report from the CDC’s Emerging Infections Program Foodborne
Diseases Active Surveillance Network (FoodNet) showed a decrease in the major
bacterial foodborne illnesses. However, some infections failed to show a steady decline
in their occurrence, indicating that increased efforts are needed to further reduce the
incidence of foodborne illnesses, especially considering the emergence of new, more
1
resilient pathogens (e.g., multidrug-resistant Salmonella). This information, along with
demands and concerns from consumers, has opened the door for food and pharmaceutical
producers to explore new options in food and drug safety.
In the last decade, consumers have shown an increasing demand for minimally
processed foods and sustained functionality of naturally occurring bioactive ingredients.
Several factors have fueled the public’s interest in exploring options for minimally
processed foods and bioactive compounds to enhance nutrition including concerns about
the safety and tolerance of synthetic preservatives; the suspected link between the
overuse of subtherapeutic antibiotics as animal growth promotants and the development
of multi-drug resistance in microbes, as well the increased media attention given to diet
and health (Naidu, 2000).
A number of products have been approved by regulatory agencies for use as direct
food antimicrobials. However, the limited spectrum of antimicrobial activity of some of
these substances has led to the continued search for more effective antimicrobials among
naturally occurring compounds (Davidson and Zivanovic, 2003). Although not often
associated with food safety, great potential lies in the area of egg shell waste.
In 2001, Poland and Sheldon discovered that exposure to egg shell membranes
significantly reduced the thermal resistance and/or inhibited selected Gram-positive and
Gram-negative foodborne bacterial pathogens [i.e., up to 3 log reduction in L.
monocytogenes populations and 83 - 87% reduction in thermal decimal reduction times
(D-values) for Salmonella enterica serovar Typhimurium, Salmonella enterica serovar
Enteritidis, and Escherichia coli O157:H7; Poland and Sheldon, 2001]. Methods to
easily extract these enzyme-rich shell membranes are readily available (Winn and Ball,
2
1996; MacNeil, 1998, 2001) and offer egg processors potential economic value as a
value-added product and provide processors a ‘natural’ processing adjuvant to sensitize
bacterial pathogens and spoilage organisms for food or pharmaceutical applications.
However, a greater understanding of the egg shell membrane (i.e. the components
responsible for the antimicrobial activity and their activity and stability) is essential to
better understand how it may be used in practical applications. This is the objective of
our studies.
3
Literature Review
I. Eggs and Egg Shell Waste
The annual per capita consumption of eggs in the U.S. is currently 240. As a
major contributor to the nation’s food supply, 28% of consumed eggs come from
processing plants manufacturing products such as pasteurized liquid whole egg, cakes,
pies, pasta, etc. (Buddington, 1999). The total annual eggshell waste is derived from
50,000,000 cases of eggs at 360 eggs per case, generating 18 billion eggshells or about
250,000,000 pounds of eggshell waste per year (Hemple, 1999). Approximately 90% of
this waste volume is calcium carbonate, while the other 10% is comprised of
proteinaceous eggshell membranes (ESM). Eggshell waste from egg breaking operations
has become such a waste problem that the U. S. Environmental Protection Agency ranked
it among the top ten most serious waste products from food processing and
manufacturing industries (Hemple, 1999).
Eggshell waste is a serious matter for the "egg-breaking" industry. Companies
are spending up to $100,000 a year to dispose of eggshells in landfills, many of which are
reaching capacity. Additionally, landfills do not want eggshells because the protein-rich
membrane which adheres to the shell attracts rats and other vermin (Walton et al., 1973).
Eggshells have the potential to be a valuable commodity, selling for about $100 a ton.
Rich in calcium, eggshells can be used to fortify foods from animal feed to orange juice.
When finely ground the shells can also be used as a substitute for pulp in paper, used in
the cosmetic industry as an ingredient in facial scrubs, or as a soil amendment by farmers
to change soil pH levels on their fields. However, the vast quantities produced annually
in the U. S. dictate that other feasible uses be identified (Hemple, 1999). Attempts by
4
some companies to dry the shells for further use have failed due to the adhering
membranes which led to spoilage problems in the final product. This thin membrane,
designated as a nuisance, has potentially more value than shell-derived products.
II. Eggshell Function and Composition
A. Natural Defenses of the Egg Against Microbial Invasion
The intact egg is naturally equipped with several physical and chemical defenses
aimed at protecting the developing embryo from physical harm and microbial invasion.
Primary physical barriers to microbial penetration of the egg are the cuticle, shell and the
inner and outer shell membranes. The cuticle is a 0.01 mm-thick external protein layer
deposited over the surface of the shell. The cuticle interferes with bacterial penetration
and limits water loss from the egg interior by sealing the pores located on the shell
surface. However, Sparks and Board (1985) demonstrated that bacterial penetration of
the shell pores occurred at a high incidence when an egg with an amorphous cuticle was
placed on cotton wool moistened with a bacterial culture. However, there was a
decreased incidence of penetration when a shell with a vesicular cuticle was challenged
with bacteria in an identical manner.
The shell comprises approximately 9-12% of the total egg weight. It consists
primarily of calcium carbonate (94%) with small amounts of magnesium carbonate (1%),
calcium phosphate (1%) and organic matter, chiefly protein (4%) (Stadelman, 1973).
The shell not only provides structural integrity to the egg, but contains numerous pores
(ca. 7,000 to 17,000 per egg) that permit the diffusion of respiratory gasses. It is
important to note that the pores are often many times the diameter of bacterial cells, and
even in eggs with undamaged cuticles, as many as 10-20 pores lack either an adequate
5
cover or plug of cuticle which provide the portals for bacteria to infect the internal
contents of the egg (North, 1978; Board and Tranter, 1994). The thickness of the
eggshell, as influenced by age, hereditary, nutritional and environmental factors, also
may play a role in preventing bacterial penetration (Taylor and Martin, 1929).
B. Components of the Eggshell Membrane
The eggshell is lined with two light-pink colored membranes, each of which is
composed of highly cross-linked proteins similar to keratin, collagen and elastin (Baker
and Balch, 1962). A diagram and scanning electron micrograph of the eggshell and
membrane structure is presented in Figure 1 and 2. The eggshell membrane (ESM)
consists of an outer and inner membrane with a thickness of approximately 50 µm and 15
µm respectively. The outer membrane is located just inside the shell and the inner
membrane is located between the outer membrane and albumin. The outer shell
membrane is attached firmly to the shell by numerous cones on the shell surface
extending into the membrane through fibril associations, while the inner membrane lies
immediately over the albumen. The membranes are held firmly together, except at the
blunt end where they separate to enclose the air space. Their structure is similar to a
meshwork of entangled threads which aids in obstructing invading microorganisms. It
has been determined that the membranes are largely composed of protein which, because
of its high content of cystine and its insolubility, is described as keratin-like (Cooke and
Balch, 1970).
Early electron microscopy by Masshoff and Stolpmann (1961) showed that the
membrane fibers consisted of a central core having a fine, fibrillar structure surrounded
by a fine-granular sheath consisting of protein and mucopolysaccharide. Although fibers
6
from the membranes were similar in structure, fibers from the inner membrane were
thinner. Figure 3 is an electron micrograph of the fiberous inner eggshell membrane.
Gaps, spanned by delicate strands, were present between the core and sheath. Within
both membranes, some fibers were fused together in places by their sheaths. Simons and
Wiertz (1963) confirmed these results, but they found that the gaps between core and
sheath were not visible in all places. Later electron micrographs of the shell membranes
have provided evidence that each fiber of a membrane has an electron-dense core and the
higher mantle layer has a high content of polysaccharides (Powrie, 1973).
Eggshell membrane histochemical studies have shown the presence of protein and
firmly-bound sugars (Simkiss, 1958; Masshoff and Stolpmann, 1961; Robinson and
King, 1968). Starcher and King (1980) identified desmosine and isodesmosine as the
major cross-links in eggshell membranes, which were later recognized as giving eggshell
membrane its high insolubility and flexibility to support the egg white before the
formation of the shell (Takahashi et al., 1996). In addition, allysine (α-aminoapipic-δ-
semialdehyde), the reactive precursor of cross-links, and its aldol condensation product
are present in significant quantities in eggshell membrane (Crombie et al., 1981). Harris
et al. (1980) showed that lysyl oxidase, located in the isthmus of the hen oviduct in a
copper-rich region, is required for the biosynthesis of these cross-links in the eggshell
membrane. Lysyl oxidase activity was found in shell membranes and was shown to be
coupled with catalase. This coupling enzyme system is considered to be involved in the
biosynthesis of eggshell membranes and protects the embryo against hydrogen peroxide.
Lifshitz et al. (1963) concluded that the inner shell membrane is the most
important single barrier to bacterial penetration into the egg, with shell membranes acting
8
as bacterial filters, and containing antibacterial substances. Elliott and Brant (1957) were
the first to report the presence of lysozyme in eggshell membranes and Winn and Ball
(1975) identified measurable amounts of β-N-acetylglucosaminidase activity within the
membranes. Ovotransferrin was later identified by immunofluorescence as being another
component of the ESM (Gautron et al., 2001). These proteins as well as other
unidentified compounds may contribute to the antimicrobial properties exhibited in the
ESM.
C. Chemical Defenses of the Hen’s Egg
Apart from the physical barriers, the avian egg is equipped with a number of
chemical defenses which inhibit or prevent microbial growth in the albumen (Board and
Tranter, 1994). The alkalinity of the albumen, which increases from an initial pH of
approximately 7.6 to greater than 9.0 after one week at 25°C, is a major naturally-
occurring antimicrobial factor. Egg albumen (composed of approximately 88% water,
10% protein, 1% carbohydrate, 0.5% lipid and 0.5% minerals) contains lysozyme, avidin,
conalbumin (ovotransferrin), ovomucoid and other components that possess some degree
of antimicrobial activities. Ovomucoid has antibacterial activity associated with its
ability to inhibit proteolytic enzymes (i.e., trypsin inhibitor) important to microbial
growth (Garibaldi, 1960). Lysozyme is primarily known for its bacteriolytic properties
towards lysing the cell walls of Gram-positive bacteria. Avidin is believed to inhibit
microbial growth by binding biotin which is required by bacteria as an essential enzyme
cofactor. Conalbumin (ovotransferrin), which constitutes 10-12% of the total egg white
solds, acts as a chelating protein that binds iron and other minerals into a stable complex
and potentially produces a deficiency of minerals for invading microorganisms. The
10
hen’s egg also contains a β-N-acetylglucosaminidase (Lush and Conchie, 1966; Winn and
Ball, 1975) which is classified as a lysosomal enzyme that contributes to the hydrolytic
degradation of glycoproteins, mucopolysaccharides and glycolipids (Robinson and
Stirling, 1968). A more thorough discussion of lysozyme, conalbumin (ovotransferrin)
and β-N-acetylglucosaminidase will be presented later.
III. The Bacterial Cell Wall
The mechanism of antimicrobials in food preservation is either through
controlling the growth of a contaminating microorganism (bacteriostatic) or killing the
microorganism directly (bactericidal) (Tortora et. al., 1986a). To better understand the
impact of antimicrobial agents requires an understanding of the structure and composition
of the bacterial cell wall.
The bacterial cell wall is a semirigid structure that is responsible for the
characteristic shape of the cell. The cell wall surrounds the underlying fragile plasma
(cytoplasmic/inner) membrane and protects it, the internal structures, and the cell
medium from potentially adverse changes in its surrounding environment. The primary
function of the cell wall is to prevent bacterial cells from rupturing when the osmotic
pressure inside the cell is greater than that outside the cell (Tortora et al., 1986). The cell
wall also serves as a medium for nutrient transport and release of waste products, as well
as providing resistance to phagocytes and cellular exchange of DNA (Nikaido and Vaara,
1987). Figure 4 and 5 are representations of the components which make up Gram-
positive and Gram-negative bacteria cell wall.
11
Figure 4. Profiles of the cell envelope for Gram-positive and Gram-negative bacteria. The Gram-
positive wall is a uniformly thick layer external to the plasma membrane composed mainly of
peptidoglycan (murein). The Gram-negative wall appears thin and multilayered consisting of a
phospholipid-lipopolysaccharide (LPS) outermembrane, a relatively thin peptidoglycan sheet and
the plasma membrane. The space between the inner (plasma) and outer membranes (wherein the
peptidoglycan resides) is called the periplasm.
Gram-positive cell wall Gram-negative cell wall
Figure 5. Cartoon depictions of the structural makeup and components of Gram-positive and Gram-
negative cell walls. (from http://student.ccbcmd.edu/courses/bio141/lecguide, Copyright © Gary E.
Kaiser used with permission) .
12
A. Peptidoglycan
Peptidoglycan (mucopeptide/murein) is a complex macromolecular polymer of
amino sugars cross-linked by short peptides. It is the major component of Gram-positive
bacterial cell walls and is responsible for cell wall integrity and rigidity. The
peptidoglycan polymer is composed of an alternating sequence of two amino sugars
related to glucose termed N-acetylglucosamine (NAG) and N-acetylmuramic acid
(NAMA). The N-acetylglucosamine and N-acetylmuramic acid molecules are primarily
linked together by β-1,4 glycosidic bonds and alternate in rows, each row forming a
carbohydrate “backbone.” The NAG and NAMA molecules may also be linked together
by β-1,3 and β-1,6 glycosidic bonds. Each row is composed of 10 to 65 amino sugars.
Each NAMA molecule is attached to a short tetrapeptide side chain of the amino acids
and amino acid derivatives L-alanine, D-glutamic acid, diaminopimelate (DAP) and D-
alanine. Cross linking peptide bonds between DAP and D-alanine in different linear
sugar chains form a two-dimensional grid that is strong and rigid, although the exact
amino acid sequence and overall structure varies within each bacterial species (Tortora et
al., 1986; Proctor and Cunningham, 1988). The structural formulas for NAG and NAMA
with its amino acid side chains and linking is shown in Figure 6.
The cell wall in most Gram-positive bacteria is composed of multiple layers of
peptidoglycan and is significantly thicker than that of Gram-negative bacteria.
Peptidoglycan of Gram-positive bacteria may comprise as much as 90% of the cell wall
compared to as little as 5-10% in some Gram-negative bacteria (Withholt et al., 1976;
Prescott et al., 1999). The degree of cross-linking between adjacent peptides also varies
between organisms. Some Gram-positive organisms may have close to 100% cross-
13
Figure 6. Schematic diagram of a peptidoglycan sheet. The glycan backbone (a) is a repeat polymer
of two amino sugars, N-acetylglucosamine (NAG) and N-acetylmuramic acid (NAMA). Attached to
the N-acetylmuramic acid is a tetrapeptide consisting of L-alamine (L-ala)- D-glutamic acid (D-glu) –
meso-diaminopimelate (DAP) – and D-alamine (D-ala). Abbreviated structure (b) of the muramic
acid subunit. Nearby tetrapeptide side chains may be linked to one another by an interpeptide bond
between DAP on one chain and D-ala on the other (c). The polymeric form of the molecule (d). The
free amino group of DAP is substituted with a glycine pentapeptide (gly-gly-gly-gly-gly-) which then
becomes an interpeptide bridge forming a link with a carboxy group from D-ala in an adjacent
tetrapeptide side chain. Gram-positive peptidoglycans differ from species to species, mainly in
regards to the amino acids in the third position of the tetrapeptide side chain and in the amino acid
composition of the interpeptide bridge.
14
bridging between peptides. In contrast, the frequency of cross-linking in E. coli (a Gram-
negative bacterium) may be as low as 30% resulting in more of a gel-like structure, rather
than a more rigid, compact layer. Gram-positive cell walls usually contain large amounts
of teichoic acids, polymers of glycerol or ribitol joined by phosphate groups. Teichoic
acids appear to extend to the surface of the peptidoglycan and help give the Gram-
positive cell wall its negative charge. These molecules may be important in maintaining
the structure of the wall, however the exact function is still unclear (Prescott et al., 1999).
B. Outer Membrane and Lipopolysaccharide Layer
Apart from differences in the peptidoglycan, Gram-negative bacteria have two
unique regions that surround the outer plasma membrane: the periplasmic space and the
lipopolysaccharide (LPS) layer. The periplasmic space separates the outer plasma
membrane from the peptidoglycan layer and contains periplasmic enzymes and other
proteins that participate in nutrient acquisition (i.e. hydrolytic enzymes attacking nucleic
acids and phosphorylated molecules, and binding proteins involved in transport of
material into the cell). The periplasmic space also contains enzymes involved in
peptidoglycan synthesis and modification of toxic compounds that could harm the cell.
Electron micrographs have identified a similar but small periplasmic space in some
Gram-positive bacteria, but do not appear to have as many periplasmic proteins (Prescott
et al., 1999).
The outer membrane is a protective barrier of Gram-negative bacteria and lies
outside the thin peptidoglycan layer. It serves in preventing/slowing the entry of toxic
substances (i.e. bile salts, antibiotics) that might kill or injure the bacterium. The LPS
layer constitutes the majority of the outer membrane and is located adjacent to the
15
exterior peptidoglycan layer. It is a complex phospholipid bilayer containing both lipid
and carbohydrate consisting of three parts: (1) lipid A, (2) the core polysaccharide and (3)
the O side chain (antigen). A small lipoprotein, Braun’s lipoprotein, is covalently bound
to the underlying peptidoglycan and embedded in the outer membrane by its hydrophobic
end. Although complex, the outer membrane is not impermeable and can permit the
passage of small molecules (ca. 600-700 daltons) through porin proteins which cluster
together and span the outer membrane.
Adjacent LPS molecules of Gram-negative bacteria are also stabilized by the
presence of divalent cations (i.e. Mg2+, Ca2+) in the outer membrane, which decrease
electrostatic repulsions and increase LPS-LPS associations. These strong associations are
believed to be the primary reason large hydrophilic molecules and most hydrophobic and
amphiphilic molecules are prevented from gaining access to the cell (Nikaido and Nakae,
1979; Nikaido and Vaara, 1987). Alterations to the LPS, including removal of the
divalent cations that stabilize the outer membrane, result in compromised functioning of
this lipid bilayer barrier (Vaara, 1992).
The LPS is important for several reasons. The core polysaccharide usually
contains charged sugars and phosphates which contribute to the negative charge on the
bacterial surface, which along with the O antigen, aids the bacteria in avoiding various
defenses of the host organism. Lipid A is a major constituent of the outer membrane and
helps stabilize membrane structure. Additionally, lipid A often is toxic. As a result, the
LPS can act as an endotoxin and cause some of the symptoms that arise in Gram-negative
bacterial infections (Prescott et al., 1999).
16
C. Outer membrane integrity
Access to intracellular organelles or components of Gram-negative bacterial cells
by many antimicrobials is limited due to the outer membrane (OM) which serves as an
efficient outer permeability barrier against macromolecules. However, there is a
propensity for strong interactions between the OM with highly cationic molecules
because of the predominantly anionic nature of lipopolysaccharides and membrane
proteins on the OM of Gram-negative microorganisms (Nikaido, 1996). Several cationic
molecules have been shown to disrupt the integrity of the OM, resulting in loss of the
barrier function without exhibiting direct bactericidal activity. The compounds such as
polyethyleneimine (PEI) and polymyxin B nonapeptide (PMBN) bind to and functionally
weaken the OM of gram-negative bacteria, sensitizing the organism to induced lysis or
penetration by detergents, hydrophobic antibiotics or probes (Helander et al., 1997; Vaara
and Vaara, 1983).
Electron micrographs have shown considerable structural alterations caused by
the binding of these agents to the OM (Vaara and Vaara, 1983; Helander et al., 1998a;
Helander et al., 2001). Vaara and Vaara (1983) also showed that other polycationic
substances (protamine and certain polylysines) are able to disrupt the OM and
simultaneously release major portions of LPS from the cells. EDTA has also been shown
to release portions of LPS. This action is attributed to its metal-chelating action and
removal of divalent cations that are essential for OM stabilization (Hukari et al., 1986).
These substances, which lack inherent toxicity while increasing the OM
permeability by various mechanisms, are termed permeabilizers (Vaara and Vaara, 1983).
There lies significant potential for food-grade permeabilizers with application to food
17
protection in that they would sensitize pathogenic Gram-negative bacteria to other
inhibitory hurdles (e.g., heat, pressure, or other substances) by disrupting the OM barrier.
For example, Hughey and Johnson (1987) demonstrated that the resistance of Gram-
negative bacteria to lysozyme was diminished when the OM had been disrupted by
EDTA.
Located beneath the cell wall (OM in Gram-negative bacteria), the cytoplasmic or
inner membrane (IM) lacks the integrity of the peptidoglycan and OM. A delicate and
semipermeable lipoprotein, it is responsible for controlling the entry and exit of solutes
from the interior of the cell. Damage to this membrane (through membrane-active
chemicals or physical processes such as high temperature and freezing) greatly affects the
bacteria (Russell, 1998). Damage to this membrane can be detected very easily by
measuring the extent of intracellular leakage (K+ ions, 260 nm-absorbing materials,
nucleotides, denatured proteins, amino acids) from heated cells (Russell and Harries,
1968; Allwood and Russell, 1970; Tomlins and Ordal, 1970, Beuchat, 1978).
D. Lethal Effects of Heat on Bacteria
Temperature response varies between bacteria, including the lethal effects of high
temperatures. Non-sporulating bacteria are readily inactivated at temperatures of about
50° C and above. Generally, the rate of inactivation increases as the temperature is raised
(Russell, 1998). “Several factors, such as composition and pH of the menstruum, the
type of organism (there could be a strain-dependent response), the growth conditions,
heating method and recovery conditions influence the rate of bacterial inactivation
(Russell, 2003).” Most cellular component are likely affected to some degree by high
temperatures (e.g., the denaturation or coagulation of proteins, breaks in DNA, lesions in
18
RNA and damage to the outer layers/membranes) and have been suggested as
possibilities as to the lethal affect in thermal inactivation of microorganisms.
Early thought prescribed that the denaturation or coagulation of proteins involved
in cell respiration or cell multiplication was the cause of death (Banwart, 1979). At
temperatures ranging from 50° to 60° C, leakage of cellular components into the
suspending medium indicates damage to the permeability of the cell. However, at higher
temperatures, death can precede leakage. Scheie and Ehrenspeck (1973) suggested that
heat caused a denaturation of proteins in the cell envelope of E. coli, thus weakening the
peptidoglycan layer sufficiently to prevent multiplication. Further injury results from
internal osmotic pressure which ruptures the cell membrane at weakened areas. Banwart
(1973) assumed that destruction of cells at temperatures that cause sublethal injury may
involve the cell membrane. However, if death precedes leakage at higher temperatures,
other mechanisms must be involved. Russell and Harries (1968) reported “that for non-
sporeforming cells, such as E. coli, RNA degradation is closely related to heat induced
death. Protein coagulation, if not the primary lethal event, is important in the thermal
destruction of microorganisms, as denaturation of proteins can occur in several areas of
the cell.”
“To assign a lethal effect to a single alteration within an organism is difficult, if
not impossible. Because some changes are more pronounced than others, some of which
may be repairable and depend upon the intensity of the temperature applied, it is
advisable to determine the effects of a range of temperatures on the components of
bacterial cells (Russell, 2003).”
19
E. High Pressure Processing
High hydrostatic pressure (HHP) processed foods have been commercially
available since 1990 in Japan. In 1996, the availability of these foods spread to Europe
and the United States (Knorr et al., 1998). Compared with heat treatments, high pressure
processing offers major benefits to the food and pharmaceutical industry. Because of the
pressure range used for biotechnological interest, there is virtually no effect on covalent
bonds. As a result, natural compounds such as flavors, aromas, dyes and
pharmacologically active molecules are not adversely impacted (Smelt, 1998).
High pressure causes inactivation of vegetative microorganisms through
membrane modification, inactivation of key enzymes, and inhibition of protein
biosyntheses (Abe et al., 1999). Microbial growth is generally inhibited at pressures of
20–130 MPa and cell death occurs in the range of 130–800 MPa. Tolerance to high
pressures varies according to the species, strain and suspending mediums used (Abee and
Wouters, 1999).
Although the exact mechanisms of cellular damage by HHP have not been
elucidated, moderate pressure may result in sublethal injuries of bacteria. As pressure
increases, cellular membrane damage becomes more severe and coincides with a rapid
increase in the death rate (Kalchayanand et al., 1998). More comprehensive reviews
regarding the effects of high pressure on microorganisms can be found in the following
articles (Abee and Wounters, 1999; Linton and Patterson, 2000; and De Angelis and
Gobbetti, 2004).
20
IV. Lysozyme
A. Lysozyme
The Commission on Enzymes has assigned the classification numbers 3.2.1.17 to
lysozyme. C-type (chick type) lysozymes are a homologous family of bacteriolytic
enzymes which catalyze the hydrolysis of β-1,4-glycosidic bonds of polysaccharides,
thereby compromising the peptidoglycan layer of bacterial cell walls and polymers of N-
acetyl-D-glucosamine (Jolles and Jolles, 1984). They are present in a variety of species
ranging from microorganisms to invertebrates, to mammals.
Lysozyme was first discovered in 1922 by Alexander Fleming. While suffering
from a cold, he found that his nasal mucous dissolved bacteria on agar plates. He soon
discovered that the antibacterial action was due to an enzyme, but that it was only
effective against certain bacteria and not those most infectious to man (Phillips, 1966). In
1963, Jolles and colleagues at the University of Paris and Canfield at the Columbia
University College of Physicians and Surgeons discovered the chemical make-up of hen
egg white lysozyme and mapped the single polypeptide chain sequence comprised of 129
amino acid residues. Lysozyme is linked in four places by disulfide bridges between
cystine residues 64 and 80, 76 and 94, 6 and 127, and 30 and 115. Lysozyme hydrolyzes
a number of substrates, but particularly the alternating polysaccharide copolymers of N-
acetyl glucosamine (NAG) and N-acetyl muramic acid (NAM) which represent the basic
unit polysaccharide structure (peptidoglycan) of many bacterial cell walls. Lysozyme
cleaves the β(1-4) glycosidic linkage connecting the C1 carbon of NAM to the C4 carbon
of NAG. Tranter (1994) discovered that at least two of the disulfide bonds must be intact
21
for lysozyme to maintain its biological activity, and reduction of all four bonds results in
a loss of all activity.
Jolles and Jolles (1984) sequenced hen egg white lysozyme and analyzed the
three-dimensional structure by X-ray crystallography. It is an ellipsoidal molecule (45 x
30 x 30 Å) with two globular lobes (residue 5-36 and 98-129 comprising the first lobe
and residues 40-94 comprising the second). A deep cleft lies between the two lobes
which is able to host six hexose units (subsites one to six) held in place by hydrogen
bonds and hydrophobic interactions. Amino acid residues are thought to contribute
significantly to the specific binding of substrates. The optimal substrate is a (NAG-
NAMA)3 hexasaccharide, with the active site of lysozyme including binding sites for
each sugar ring of the hexasaccharide, these being designated sites A through F. When
the fourth hexose (D) in the chain becomes twisted out of its normal position (Figure 7a),
a strain is imposed on the C-O bond on the ring-4 side of the oxygen bridge between
rings D (NAMA) and E (NAG). Lysozyme’s mechanism of action involves residue 35,
glutamic acid (Glu-35), which is about 3Å from the -O- bridge that is to be broken. The
free carboxyl group of glutamic acid is a hydrogen ion donor and transfers a hydrogen
ion to the oxygen atom, breaking the already-strained bond between the oxygen atom and
the carbon atom of the forth hexose (Figure 7b). This carbon atom acquires a positive
charge that in turn temporarily attracts the negatively-charged carboxyl ion of residue 52,
aspartic acid (Asp-52) which stabilizes the structure long enough for an -OH ion (from a
spontaneously dissociated water molecule) to unite with the carbon (Figure 7c). The
hydrogen ion (H+) remaining from the dissociated water can replace that lost by Glu-35 at
22
Figure 7. Mechanism of action for HEW lysozyme. (a.) When the fourth hexose (D) in the chain becomes
twisted out of its normal position, a strain is imposed on the C-O bond on the oxygen bridge between rings
D (NAMA) and E (NAG). The free carboxyl group from lysozyme residue 35, glutamic acid (Glu-35) is a
hydrogen ion donor and transfers a hydrogen ion to the oxygen atom, breaking the already-strained bond
between the oxygen atom and the carbon atom of the forth hexose (b). This carbon atom acquires a
positive charge that in turn temporarily attracts the negatively-charged carboxyl ion of residue 52, aspartic
acid (Asp-52) which stabilizes the structure long enough for an -OH ion (from a spontaneously dissociated
water molecule) to unite with the carbon (Figure 7c). The hydrogen ion (H+) remaining from the
dissociated water can replace that lost by Glu-35
23
which point the polysaccharide is broken and the enzyme is free to attach to a new
location on the bacterial cell wall and continue its digestion activity.
Generally speaking, endo-N-acetylmuramidases, such as lysozyme, hydrolyze the
glycosidic linkages between N-acetylmuramic acid and N-acetylglucosamine, resulting in
the release of N-acetylmuramic acid residues at the reducing end. Endo-N-
acetylglucosaminidases, such as β-N-acetylglucosaminidase, hydrolyze the glycosidic
bonds between N-acetylglucosamine and N-acetylmuramic acid, releasing fragments with
N-acetylglucosamine at the reducing end (Ghuysen, 1968).
B. Properties and Activity of Lysozyme
The molecular weight of lysozyme is approximately 14,400 daltons with an
isoelectric point of pH 10.7. Reports regarding its maximum activity rate related to pH
vary and have been reported to between pH 3.5 to 7.0 (Powrie, 1977) or as high as pH 9.2
(Davies et al., 1969). Ionic strength of buffers may be responsible for the extreme
variance in the reported data. Hen egg white lysozyme is remarkably heat stable,
particularly at acidic pH. At pH 4 to 5 lysozyme can retain some activity when boiled
(100°C) for 1 to 2 min. However, at higher pHs, it is readily denatured (Meyer et al.,
1936). Matsuoka and co-workers (1966) also reported that lysozyme had increased heat
stability in acidic solutions (pH 4.5, 100°C, 3 min; pH 5.29, 100°C, 30 min).
Other variables that have been shown to impact the heat stability of lysozyme
including medium pH and degree of purity. Sandow (1926) reported unpurified egg
white lysozme was inactivated in 15 minutes at 65°C, pH 8.0; where as heating at pH 5.0
for over 60 minutes at 65°C did not result in any loss of activity. Sandow also reported
lysozyme to be over 50 times more stable in phosphate buffer (pH 6.2) than in egg white
24
(pH 9.2) at 62.5°C. Epstein and Chain (1940) reported that pure lysozme preparations
showed less stability to heat than impure ones. However, Smolelis and Hartsell (1952)
reported that in phosphate buffer (pH 6.2), highly purified lysozyme can be exposed to
80°C for 30 minutes with only 5% loss of activity. When exposed to 100°C for 20
minutes, only a 25% loss of lytic activity was detected. Cunningham and Lineweaver
(1965) also showed egg white lysozyme to be more heat-stable (62.5°C) at pH 7 than pH
9. They reported a loss of approximately 10% at pH 7 compared to over a 95% loss at
pH 9 after 10 minutes of heating. They reported no inactivation of lysozyme in
phosphate buffer under identical heating conditions, even at pH 9. However, at 65°C, pH
9.0, inactivation in phosphate buffer occurred in 10 min.
C. Bacterial Sensitivity to Lysozyme
As previously mentioned, lysozyme hydrolyzes the alternating polysaccharide
copolymers of N-acetyl glucosamine (NAG) and N-acetyl muramic acid (NAM) of the
peptidoglycan of many bacterial cell walls. Due to the absence of a lipopolysaccharide
layer (LPS), which inhibits lysozyme accessibility, Gram-positive bacteria are more
susceptible to the lytic activity of lysozyme. Salton and Pavlik (1960) tested various
Gram-positive bacteria to determine their degree of susceptibility to lysozyme. Strains of
Bacillus, Corynebacterium, Lactobacillus, Micrococcus, Sarcina, Sporosarcina,
Staphylococcus, and Streptococcus were incubated for several hours with 50 µg of egg-
white lysozyme per ml of solution. Although all showed some degree of sensitivity to
digestion with lysozyme, the percent reductions in populations varied from 9% for the
most resistant (Staphylococcus aureus) cell walls to 98% for the most sensitive
(Micrococcus lysodeikticus). The difference in lysozyme sensitivity may be attributed to
25
a variance in β-glycosidic bonds present in the peptidoglycan of the various
microorganisms. Lysozyme primarily affects the β-1,4-N-acetylhexosaminidase bond,
however cell walls may be composed of not only β-1,4 glycosidic bonds but also 1,3 and
1,6 bonds as well. The degree of sensitivity would relate directly to the proportion of β-
1,4 glycosidic bonds present in the cell walls. Therefore, Gram-positive bacteria with
higher β-1,3 and 1,6 glycosidic bonds would exhibit greater resistance to the action of
lysozyme (Salton and Pavlik, 1960; Vakil et al., 1969).
Gram-negative bacteria are generally more resistant to lysozyme unless the LPS
layer is disrupted by physical or chemical means. Treatments such as 2-amino-2-
hydroxymethylpropane-1,3-diol (Tris) and ethylenediaminetetraacetate (EDTA) (Leive,
1968) bind ions (i.e. Ca2+, Mg2+) that are essential for maintaining the integrity of the
LPS layer. Antibiotics such as polymyxin B and aminoglycosides can also bind to the
LPS and result in disruption of the outer layers of the cell. Additionally, shifts in the pH,
osmotic shock (Birdsel and Cota-Robles, 1967), freeze-thaw cycling (Ray et al., 1984)
and drying (Webb, 1969) can predispose the bacterial cell to the lytic action of lysozyme.
Peterson and Hartsell (1955) evaluated 135 Gram-negative bacterial species for
sensitivity to lysozyme and concluded that only under extreme conditions [exposure to
dilutions of hydrochloric acid (pH 3.5, 45°C, 1 hr) followed by adjustment to pH 9.8 with
0.05 N sodium hydroxide] does lysozyme act on Gram-negative cells. The degree of
lysis towards each species varied from sensitive (Salmonella, Brucella), moderately
sensitive (Klebsiella, Shigella, Neisseria, Pseudomonas, Pasteurella, Erwinia,
Escherichia) to insensitive (Vibrio, Proteus). Hughey and Johnson (1987) evaluated
lysozyme against food pathogens and spoilage bacteria and showed certain strains of
26
Clostridium botulinium, C. thermosaccharolyticum, C. tyrobutyricum, Bacillus
stearothermophilus and Listeria monocytogenes were effectively inhibited or lysed by
lysozyme. Lysozyme activity was also influenced by the presence of chelating agents
(EDTA), as well as environmental conditions such as temperature, growth phase and
medium. However, regardless of experimental conditions, some bacteria evaluated were
resistant.
D. Lysozyme and Food
Because of its specificity of action towards bacterial cell wall components and its
lack of toxicity to humans, lysozyme is attractive as a food preservative. Bacterial cell
walls are comprised of structural components (i.e. peptidoglycan) unlike those found in
eukaryotic cells (plant and animal cells). The cell walls present a prime target for
lysozyme and other antimicrobials in that these chemicals will often damage bacterial cell
walls or interfere with their synthesis without harming the animal host cells.
Use of lysozyme in foods was introduced in Europe and Asia in the 1960s after an
efficient extraction method was developed from egg albumen. Lysozyme has found use
as an antimicrobial agent in various foods. For example, the Japanese have become the
largest users of lysozyme in practical applications, using it as a preservative on fresh
fruits and vegetables, seafoods and meats, tofu bean curd, and wine and sake. A process
has also been patented in the United Kingdom whereby lysozyme is added to butter or
milk for cheese making to prevent growth of undesirable microorganisms (Proctor and
Cunningham, 1988).
Despite being declared safe for the use in foods by the Joint FAO/WHO Expert
Committee on Food Additives in 1992, the Food and Drug Administration (FDA) did not
27
grant GRAS (generally recommended as safe) status for lysozyme until 1998 when it was
approved for usage in cheese. More recently, the FDA has also granted GRAS status for
limited use in other specific foods (i.e. wine, beer, frankfurters). Current use of lysozyme
as an antimicrobial agent in the U.S. food industry has been limited inpart due to the
restricted or limited spectrum of lytic activity against some Gram-positive bacteria and
loss of some lytic activity in the presence of food matrix components such as
carbohydrates and lipids (Hayashi et al., 1989).
V. β-N-acetylglucosaminidase
A. β-N-acetylglucosaminidase
β-N-acetylglucosaminidase is a ubiquitous glycocosidase exhibiting hydrolase
activity against mucopolysaccharides, mucoproteins, glycoproteins and other
mucosubstances (Levvy and Conchie, 1966). It is generally considered to be a lysosomal
enzyme and attacks oligosaccharides derived from hyaluronic acid effectively removing
terminal nonreducing N-acetylglucosamine residues (Weissmann et al., 1964). β-N-
acetylglucosaminidase is ubiquitous and found in animals (i.e., boar epididymis, hen
ovoduct), plant tissues (i.e., jack bean), produced in prokaryotic organisms (i.e.,
Lactococcus lactis, Clostridium perfringens) and found in humans.
Bacteria synthesize peptidoglycan hydrolases capable of hydrolyzing their own
peptidoglycan (Schockman and Höltje, 1994). These hydrolases are synthesized during
cellular growth and are involved in a number of cellular functions that occur in cell wall
remodeling such as cell separation after division, cell wall turnover and cell wall
expansion (Smith et al., 2000). N-acetylglucosaminidase is one of four different genera
involved in cleaving the chemical bonds inside the peptidoglycan molecule. It is
28
assumed to contribute to the hydrolytic degradation of glycoproteins,
mucopolysaccharides and glycolipids (Robinson and Stirling, 1968).
Very few studies exploring its antimicrobial activity have been initiated, however
Martin and Kemper (1970) reported N-acetylglucosaminidase from Clostridium
perfringens to be lytic against some Gram-negative bacteria, Kaplan et al. (2003)
identified a soluble β-N-acetylglucosaminidase that causes detachment and dispersion of
Actinobacillus actinomycetemcomitans biofilm cells, and Huard et al. (2003) identified an
N-acetylglucosaminidase from Lactococcus lactis that exhibited hydrolyzing activity on
the peptidoglycan layer of several Gram-positive bacteria.
B. β-N-acetylglucosaminidase from Hen Eggs
Lush and Conchie (1966) first identified the presence of β-N-
acetylglucosaminidase (EC 3.2.1.30) activity in hen’s egg white. They reported two
know carbohydrases which occur in hen egg albumen, β-N-acetylglucosaminidase and
lysozyme. They found β-N-acetylglucosaminidase to be present in all hen egg albumen
samples (n=39) from six strains of hens ranging in concentraitons from 2320 to 8880 µg
(expressed as µg p-nitrophenol released from phenyl N-acetyl-β-glucosaminide by 1-ml
albumen in 1-hr at 37°C). They also observed variations in the level of activity across the
three main layers of egg albumen.
Several studies have reported varying concentrations of β-N-
acetylglucosaminidase found in pooled fresh chicken eggs. Lush and Conchie (1966)
first reported a 2-fold variation in the average concentration of β-N-
acetylglucosaminidase isolated from 6 to 9 hens from 6 layer strains including
Thornbergs 606, Sterling White Link, Brown Leghorn, White Leghorn (strains A and B)
29
and Rhode Island Red. Henderson and Robinson (1969) later reported up to a 6-fold
difference in the β-N-acetylglucosaminidase activity detected in samples of fresh egg
white, whereas Donovan and Hansen (1971a) observed a 3-fold variation in enzymatic
activity across individual fresh eggs. Their study followed the release of p-nitrophenol
from p-nitrophenyl-N-acetyl-β-D-glucosaminide using a modified procedure of Lush and
Conchie (1966). The β-N-acetylglucosaminidase activity in pooled egg white from 15
different lots of egg albumen had a mean value of 4.12 mµmoles/min (expressed as p-
nitrophenol released from p-nitrophenyl-N-acetyl-β-D-glucosaminide by 0.01-ml
albumen in 25-min at 37°C) with a standard deviation of 0.50 mµmoles/min. No
seasonal variations in activity were observed.
Tarentino and Maley (1971) purified two isozymes of β-N-acetylglucosaminidase
from hen oviduct with reported molecular weights of 118,000 (Type I) and 158,000
daltons (Type II) and pI values of 6.45 and 6.86, respectively. Kinetic constants and
activity towards various substrates exhibited virtually no differences between the two.
However, the Type I enzyme is the predominant species. Lucas (1979) later showed that
only Type I isozyme is found in egg white and suggested that it may play a role in the
catabolism of glycoproteins in both egg white and the oviduct.
Contrary to the findings of Tarentino and Maley reported above, Ogawa and
Nakamura (1983) and Ogawa et al., (1983) also purified a β-N-acetylhexosaminidase
(NAHase) from both hen oviduct and egg white, and then chicken liver and chicken
serum. Oviduct and liver NAHase was reported to have a similar MW of 53,000, while
serum and egg white NAHase had a MW of 68,000. They hypothesized that these
30
variations could be because lysosomal enzymes may be synthesized as precursors with a
subunit molecular weight higher than those of mature forms.
C. β-N-acetylglucosaminidase Activity
β-N-acetylglucosaminidase activity in egg white responds negatively to increases
in both temperature and pH. There is gradual loss in its activity as the pH exceeds 9,
especially at ambient temperatures. Lush and Conchie (1966) evaluated albumin β-N-
acetylglucosaminidase activity in eggs stored at room temperature. They observed a
rapid decline in enzymatic activity from a mean of 3283 units to 250 units after storage
for 10 days and then decreased to 30 units after 19 days (expressed as µg p-nitrophenol
released by 1-ml albumen in one hour at 37°C). They also found that the enzyme activity
was not affected by freezing/thawing or homogenization of the albumen. Donovan and
Hansen (1971a) showed that when refrigerated, the enzyme activity in fresh eggs
remained high for a considerable length of time. Winn and Ball (1975) reported that the
loss of β-N-acetylglucosaminidase activity is closely associated to the increase in
albumen pH. They recorded a rapid decline in enzyme activity to near extinction
between 3 to 6 days of age at pH 9.2 to 9.4 when eggs were stored in open flats at 25°C.
In contrast, β-N-acetylglucosaminidase in shell membrane was found to be very
stable. Winn and Ball (1975) reported that, for eggs stored for nine days at room
temperature, activity in the membrane was not affected by pH changes in the surrounding
albumen (which reached pH values of over 9.0). They proposed that the enzyme might
be protected as a component of the membrane. They further suggested that if β-N-
acetylglucosaminidase is an integral part of the shell membrane, it might play a
bacteriostatic role or assist in slowing bacterial penetration into the egg.
31
The optimum pH for β-N-acetylglucosaminidase activity varies depending upon
the source of the enzyme. Donovan and Hansen (1971a) determined the optimum pH of
hen albumen β-N-acetylglucosaminidase activity to be 3.0 with at least 20% of maximum
activity observed over a pH range of 1.5 to 5.5. The enzyme activity in egg albumen
(from pH 6.8 to 8.8) was stable for several hours at ambient temperature (24°C). When
the egg albumen to pH was increased to 9.6 with 1 M NaOH, the enzyme activity was
rapidly lost.
Henderson and Robinson (1969) were the first to report that β-N-
acetylglucosaminidase from hen albumen is heat inactivated when exposed to
temperatures near 60°C. They determined the activation energy for heat denaturation in
egg albumen (no pH specified) to be 62.9 kcal/mole. Later, Donovan and Hansen
(1971b) demonstrated that the kinetics of β-N-acetylglucosaminidase heat inactivation in
egg white and whole egg (pH 7.0) followed a first order rate when heated between 58-
62°C and an activation energy of 91 kcal/mole in egg albumen and 73 kcal/mole in whole
egg. At 60°C (pH 7.0), approximately 66% of the enzyme activity was destroyed by
heating for 3.5 min. To date, there has been no published literature pertaining to the
antimicrobial activity of hen β-NAGase.
VI. Ovotransferrin
A. Ovotransferrin
Ovotransferrin, also referred to as conalbumin, constitutes about 13% of egg
white proteins. It belongs to the transferrin family and shows about 50% sequence
homology with mammalian serum transferrin and lactoferrin. It is a glycoprotein
consisting of 686 amino acid residues with 15 disulfide bridges, a MW of approximately
32
77,700 and an isoelectric point (pI) of 6.1-6.5 (Abola et al., 1982). It is folded into two
lobes referred to as the N- and C-lobe. Although similar in structure, they differ
functionally. The two lobes are linked together with a short connecting peptide and each
lobe can be divided into two domains enclosing a hydrophilic cleft which acts as a
transitional metal (i.e. (Fe[III], Cu[III], Al[III]) binding site.
B. Action of Ovotransferrin on Bacteria
It was believed the antimicrobial activity exhibited in ovotransferrin was only a
result of iron deprivation (iron being an essential growth factor for most microorganisms)
due to the sequestered iron in the binding site (Alderton, Ward and Fevold, 1946). The
antimicrobial activity of ovotransferrin (Valenti et al., 1983) against a variety of
microorganisms, including pathogenic Escherichia coli, Pseudomonas aeruginosa and
Vibrio cholera, has been demonstrated (Boesman-Finkelstein and Finkelstein, 1985). In
vivo, ovotransferrin has been shown to have therapeutic properties against acute enteritis
in infants (Corda et al., 1983). Several reports of ovotransferrin retaining antimicrobial
activity, even when complexed with metals such as zinc and iron, have been reported
(Valenti et al., 1987; Valenti et al., 1985; Ibrahim, 1996).
Structural similarity between lactoferrin and ovotransferrin exist, however
evidence suggests that ovotransferrin possesses structurally-dependent bactericidal
activity other than the iron deprivation effect reported for lactoferrin. Regardless of the
degree of iron saturation, ovotransferrin exhibited strong antimicrobial action against
Gram-positive Staphylococcus aureus. For the Gram-negative organism Escherichia
coli, the iron-bound ovotransferrin was more bactericidal than the iron-free ovotransferrin
suggesting that the mechanism of action is attributed to factors other than the iron
33
deprivation effect (Ibrahim et al., 1998). Aguilera and collegues (2003) demonstrated the
ability of ovotransferrin to permeate the outer membrane of Escherichia coli. This
interaction was attributed to the anionic nature of the outer membrane and the cationic
nature of ovotransferrin. Although ovotransferrin was not able to cause extensive
damage to the bacteria, there was a marked increase in extracellular [K+] concentrations,
without detectable changes in [Na+]. A dissipation of the electrical potential was
observed along with increased sensitivity of bacterial cells to the hydrophobic antibiotic,
actinomycin D. Actinomycin D, which normally possesses limited action against Gram-
negative bacteria, inhibited RNA and protein synthesis upon disruption of the outer
membrane.
VII. Food Preservation and the Hurdle Concept
The issue of food safety is a priority among consumers, and this desire for safer
food has transferred to processors. To increase the safety of foods, processors have
introduced “safety hurdles” to help ensure/increase the level of safety. Refrigeration and
freezing, thermal processing, and the addition of chemical preservatives are all
preservation methods, some of which are used simultaneously or sequentially. However,
during the last decade consumers have become more interested in alternative
preservatives to reduce the use of chemical preservatives (Wang and Shelef, 1992).
Additionally, Ray (1992) reported consumer concerns about the possible health risks of
foods preserved with “unnatural” chemicals as well as the loss of nutritional and
functional value of harshly processed foods.
Consumer’s preference for refrigerated foods devoid of preservatives is based on
their perception that these foods are more nutritious, healthy and “natural” as opposed to
34
“harshly” processed and “chemically” preserved foods (Ray, 1992). The food industry
has responded to consumer demands for healthier foods by introducing products to the
market place that receive minimal processing and contain reduced levels of traditional
preservations such as salt, sugar, acids, and chemical preservatives. These “minimally
processed foods” are products that receive a heat process or other preservation treatment
for reducing the microbiological load in foods, but does not produce “commercial
sterility” (Anon., 1998).
The hurdle concept, or hurdle technology, is an approach that combines several
inhibitory hurdles that together can effectively reduce microbial pathogens in foods
(Leistner and Gorris, 1995). Common hurdles include physical and chemical
components such as refrigeration, modified atmosphere packaging (MAP), thermal
treatments, water activity (aw), pH and preservatives. When used together hurdles may
act synergistically, thus enabling the use of lower levels of each hurdle than would
normally be necessary if used alone. Incorporation of multiple barriers or hurdles are
recommended in food products to inhibit or minimize microbial growth. Chemical
preservatives are a common hurdle employed as an adjunct to product formulations.
IX. Conclusion
Eggshells and their membranes from egg breaking operations are a significant
waste product for processors. As previously mentioned, ESM has demonstrated
significant antimicrobial activity (Poland and Sheldon, 2001). These enzyme-rich shell
membranes are readily extractable and offer egg processors potential economic value as a
food or pharmaceutical processing adjuvant or “natural” biocide to sensitize bacterial
pathogens and spoilage organisms to other hurdles. Use of egg shell membranes, or their
35
components as an antimicrobial processing aid for heat-sensitive foods and
pharmaceutical products could become a reality. This may lead to reduced thermal
process requirements (lower process temperatures and times) while still attaining a
pathogen-free product with extended shelf-life. Consumers will perceive these natural
ingredients as more acceptable while processors will gain a new value-added product
having potentially significant market value. Reduced thermal processing requirements
may also result in products that maintain higher nutrient levels, have improved
functionality and potentially lower processing costs.
Our work involves the identification of components of the ESM responsible for
the observed antimicrobial activity including their enzyme activity and stability in the
membrane. Insight gained into the responsible components and their mechanism of
action in lowering the heat resistance of bacteria will provide additional information on
optimizing this activity and identify other potential applications for egg shell membranes
and albumen. The end result may be the identification of new functional proteins or
components derived from eggshell membranes and egg white that have significant
economic value.
36
References:
Abe, F., Kato, C. and K. Horikoshi. 1999. Pressure-regulated metabolism in
microorganisms. Trends Microbiol. 7:447-453.
Abee, T., and J. A. Wouters. 1999. Microbial stress response in minimally processing.
Int. J. Food Microbiol. 50, 65-91.
Abola, J. E., Wood, M. C., Chwen, A., Abraham, D. and P. D. Pulsinelli. 1982. In: The
Biochemistry and Physiology of Iron. Saltman P and Hegenarer J (Eds). Elsevier.
Aguilera, O., Quiros, L. M. and J. F. Fierro. 2003. Transferrins selectively cause ion
efflux through bacterial and artificial membranes. FEBS LETTERS. 548:5-10.
Alderton, G., Ward, W. H., and H. L. Fevold. 1946. Identification of the
bacteria-inhibiting, iron-binding protein of egg white as conalbumin. Arch Biochem
Biophys. 11:9-15.
Allwood, M. C. and A. D. Russell. 1970. Mechanisms of thermal injury in non-
sporulating bacteria. Adv Appl Micro. 12:89-119.
Anonymous. 1988. Safety consideration of the new generation refrigerated foods.
Refrigerated food and microbiological criteria committee of the National Food Processors
Association. Dairy Food Sanit. 8:5-7.
Arora, K. L., D. V. Arneja, and D. Chand. 1974. A report on the lysozyme concentration
of egg white of some breeds of fowl. Indian J. Poult. Sci. 9:162-165.
Baker, J. R., and D. A. Balch. 1962. A study of the organic material of hen’s-egg shell.
Biochem. J. 82:352-361.
Baker, R. C. 1990. Survival of Salmonella enteritidis on and in shelled eggs, liquid eggs,
and cooked egg products. Dairy, Food Environ. Sanit. 10:273-275.
Banwart, G. J. 1979. Basic Food Microbiology. AVI Publishing Co. Westport,
Connecticut. 661-662
Beckers, H. J., P. S. Soentoro, and E. H. M. Delfgou-van Asch. 1987. The occurrence of
Listeria monocytogenes in soft cheeses and raw milk and its resistance to heat. Int. J.
Food Microbiol. 4:249-256.
Beddell, C. R., C. C. F. Blake, and S. J. Oatley. 1975. An X-ray study of the structure
and binding properties of iodine-inactivated lysozyme. J. Mol. Biol. 97:642-654.
37
Beuchat, L. R. 1978. Injury and repair of Gram-negative bacteria, with special
consideration of the involvement of the cytoplasmic membrane. Adv Appl Micro.
23:219-243.
Birdsel, D. C. and E. H. Cota-Robles. 1967. Production and ultrastructure of lysozyme
and ethylenediaminetetra-acetate-lysozyme spheroplasts of Escherichia coli. J Bacteriol.
93:427-437.
Board, R. G. 1968. Microbiology of the egg: A review. pp. 133-162. In T. C. Carter (ed.).
Egg Quality. A Study of the Hen’s Egg. Oliver and Boyd, Edinburgh, Scotland.
Board, R. G. and H. S. Tranter. 1994. The Microbiology of Eggs. pp. 81-104. In W. J.
Stadelman and O. J. Cotterill (eds). Egg Science and Technology, 4th ed. The Haworth
Press, Inc. Binghamton, NY.
Boesman-Finkelstein, M. and R. Finkelstein. 1985. Antimicrobial effects of human milk:
Inhibitory activity on enteric pathogen. FEMS Microbiology and Immunology 27:167.
Bradshaw, J. G., Peeler, J. T. and R. M. Twedt. 1977. Thermal inactivation of ileal loop-
reactive Clostridium perfringens type A strains in phosphate buffer and beef gravy. Appl.
Environ. Microbiol. 34:280-284.
Buddington, L. R. 1999. Personal communication. United Egg. Atlanta, GA. given to
Poland, A. L. and B.W. Sheldon, 2001. Altering the thermal resistance of foodborne
bacterial pathogens with an eggshell membrane waste by-product. J. Food Prot. 64:486-
492.
Carminati, D. and S. Carini. 1989. Antimicrobial activity of lysozyme against Listeria
monocytogenes in milk. Microbiol. Aliment. Nutr. 7:49-56.
CDC. 2003. Public Health Laboratory Information System surveillance data, Salmonella
annual summaries. at http://www.cdc.gov/ncidod/dbmd/phlisdata/salmonella.htm
Cooke, A.S. and D.A. Balch. 1970. Studies of membrane, mammillary cores and cuticle
of the hen eggshell. Brit. Poult. Sci. 11:345-352.
Corda, R., Biddau, P., Corrias, A. and E. Puxeddu. 1983. Conalbumin in the treatment of
acute enteritis in the infant. Int. J. Tissue React.1:117-123.
Crombie, G., Snider R., Faris, B. and C. Franzbrau. 1981. Lysine-derived cross-links in
the egg shell membrane. Biochim. Biophys. Acta. 640:365-367.
Cunningham F. E., and H. Lineweaver. 1965. Stabilization of egg-white proteins to
pasteurizing temperatures above 60ºC. Food Technol. 19:1442-1447.
38
Davidson, P. M. 2001. ‘Chemical preservatives and natural antimicrobial compounds’,
In: Doyle, M.P., Beuchat, L.R. and T.J. Montville. Food Microbiology, Fundamentals
and Frontiers, 2nd ed, American Society for Microbiology, Washington, DC.
Davidson, P. M. and S. Zivanovic. 2003. ‘The use of natural antimicrobials’, In: Zeuthen,
P. and L. Bogh-Sorensen. 2003. Food preservation techniques. CRC Press, Boca Raton,
FL.
Davies, A., Neuberger, A., and B.Wilson.1969. The dependence of lysozyme activity on
pH and ionic strength. Biochim. Biophys. Acta. 178: 294.
Donnelly, C. W., Briggs, E. H. and L. S. Donnelly. 1987. Comparison of heat resistance
of Listeria monocytogenes in milk as determined by two methods. J. Food Prot. 50:14-17.
Donovan, J. W. and L. U. Hansen. 1971a. The β-N-acetylglucosaminidase activity of egg
white. 1. Kinetics of the reaction and determination of the factors affecting the stability of
the enzyme in egg white. J. Food Sci. 36:174-177.
Donovan, J. W. and L. U. Hansen. 1971b. The β-N-acetylglucosaminidase activity of egg
white. 2. Heat inactivation of the enzyme in egg white and whole egg. J. Food Sci.
36:178-181.
Elliott, L. E. and A. W. Brant. 1957 Effect of saline and eggshell membrane on bacterial
growth. Food Research. 22:241-250.
Epstein, L. A. and E. Chain. 1940. Some observations on the preparation and properties
of the substrate of lysozyme. Brit. J. Exptl. Path. 21:339-355.
Foegeding, P. M. and S. B. Leasor. 1990. Heat resistance and growth of Listeria
monocytogenes in liquid whole egg. J. Food Prot. 53(1):9-14.
Gallo, A. A., Swift, T. J., and H. Z. Sable. 1971. Magnetic resonance study of the Mn+2-
lysozyme complex. Biochem. Biophys. Res. Commun. 43:1232-1238
Garibaldi, J. A. 1960. Factors in egg white which control growth of bacteria. Food Res.
25:337-344.
Gautron, J., Hincke, M. T., Panheleux, M., Garcia-Riuz, J. M., Boldicke, T., and Y. Nys.
2001. Ovotransferrin is a matrix protein of the hen eggshell membranes and basal
calcified layer. Connect Tissue Res. 42(4):255-67.
Ghuysen, J.M. 1968. Use of bacteriolytic enzymes in determination of wall structure and
their role in cell metabolism. Bacteriol. Rev. 32:425-464.
Gill, A. O. and R. A. Holley. 2000. Inhibition of bacterial growth on ham and bologna by
lysozyme, nisin and EDTA. Food Res. Int. 33:83-90.
39
Gorin, G., S. F. Wang, and L. Papapavlou. 1971. Assay of lysozyme by its lytic action on
M. lysodeikticus cells. Analyt. Biochem. 39:113-127.
Grasselli, M., S. A. Camperi, A. A. Navarro del Canizo and O. Cascone. 1999. Direct
lysozyme separation from egg white by dye membrane affinity chromatography. J. Sci.
Food Agric. 79:333-339.
Harris, E. D. and J. E. Blount. 1980. Localization of lysyl oxidase in hen oviduct:
Implications in egg shell membrane formation and composition. Science. 208:55-56.
Haurd, C., Miranda, G., Wessner, F., Bolotin, A., Hansen, J., Foster, S. J. and M. P.
Chapot-Chartier. 2003. Characterization of AcmB, an N-acetylglucosaminidase autolysin
from Lactococcus lactis. 149:695-705.
Hayashi, K., T. Kasumi, N. Kubo, K. Haraguchi, and N. Tsumura. 1989. Effects of N-
acetylmuramidase from Streptomyces rutgersensis H-46 as a food preservative. Agri
Biol. Chem. 53:3173-3177.
Hemple, H. 1999. Personal communication to Poland, A. L. and B.W. Sheldon, 2001.
Altering the thermal resistance of foodborne bacterial pathogens with an eggshell
membrane waste by-product. J. Food Prot. 64:486-492.
Henderson, A. E. and D. S. Robinson. 1969. Effect of heat pasteurization on some egg
white enzymes. J. Sci. Food Agr. (Great Britain). 20:755-760.
Hughey, V. L. and E. A. Johnson. 1987. Antimicrobial activity of lysozyme against
bacteria involved in food spoilage and food-borne disease. Appl. Environ. Microbiol.
53:2165-2170.
Hughey, V. L., Wilger, R. A. and E. A. Johnson. 1989. Antibacterial activity of hen egg
white lysozyme against Listeria monocytogenes Scott A in foods. Appl. Environ.
Microbiol. 55:631-638.
Hukari, R., Helander, I. M. and M. Vaara. 1986. Chain length heterogeneity of
lipopolysaccharide released from Salmonella Typhimurium by
ethylenediaminetetraacetic acid or polycations. Eur. J. Biochem. 154:673-676.
Ibrahim, H. R. in: Hen Eggs: Their Basic and Applied Science, CRC Press, Inc., New
York, 1996. pp 37-57.
Ibrahim, H. R. 1997. Insights into the structure function relationship of ovalbumin,
ovotransferrin and lysozyme. In: Hen eggs: Their Basic and Applied Science.
Yammamoti T, Juneja LR, Hatta H and Kim M (Eds). CRC Press Inc., New York
40
Ibrahim, H. R., Iwamori, E., Sugimoto, Y. and T. Aoki. 1998. Identification of a distinct
antibacterial domain within the N-lobe of ovotransferrin. Biochim. Biophys. Acta.
1401:289-303.
Janolino, V. G. and H. E. Swaisgood 1982. Analysis and optimization of methods using
water-soluble carbodiimide for immobilization of biochemicals to porous glass.
Biotechnol. Bioengr. 24:1069-1080.
Jolles, J., Jaurequi-Adell, J., Bernier, I. and P. Jolles. 1963. The chemical structure of
hen’s egg white lysozyme: the detailed study. Biochem. Biophys. Acta. 78:668-689.
Jolles, P. and J. Jolles. 1984. What’s new in lysozyme research? Mol and Cell. Biochem.
63:165-189.
Kalchayanand, N., Sikes, A., Dunne, C. P. and B. Ray. 1998. Factors influencing death
and injury of foodborne pathogens by hydrostatic pressure-pasteurization. Food Micro.
15:207-214.
Kaplan, J. B., Ragunath, C., Rasmasubbu, N. and D. H. Fine. 2003. Detachment of
Actinobacillus actinomycetemcomitans biofilm cell by an endogenous β-hexosaminidase
activity. J. Bacter. 185:4693-4698.
Karatzas, A. K., Kets, E. P. W., Smid, E. J. and M. H. J. Bennik. 2001. The combined
action of carvacrol and high hydrostatic pressure on Listeria monocytogenes Scott A. J.
Appl. Micro. 90:463-469.
Kihm, D. J., Leyer, G. J., An, G. H. and E. A. Johnson. 1994. Sensitization of heat treated
Listeria monocytogenes to added lysozyme in milk. Appl. Environ. Microbiol. 60:3854-
3861.
Knorr, D., Heinz, V., Lee, D. U., Schluter, O. and M. Zenker. 1998. High pressure
processing of foods: introduction. In: Proceedings of VTT Symposium Fresh Novel
Foods by High Pressure. K. Autio (ed). Pp. 9-20. Espoo, Finland.
Leister, L. 2000. Basic aspects of food preservation by hurdle technology. Int. J. Food
Micro. 55:181-186.
Leive, L. 1968. Studies on the permeability change produced in coliform bacteria by
ethylenediaminetetraacetate. J. Biol. Chem. 243:2373-2380.
Levvy, G. A. and J. Conchie. 1966. Mammalian glycosidases and their inhibition by
aldonolactones. Pp. 571-580. In: S. P. Colwick and N. O. Kaplan (eds.). Methods in
Enzymology, Complex Carbohydrates, vol. 8. Academic Press, New York.
Lifshitz, A., Baker, R. C. and H. B. Naylor. 1963. The relative importance of chicken egg
exterior structures in resisting bacterial penetration. J. Food Sci. 28:94-99.
41
Linton, M. and M. F. Patterson. 2000. High pressure processing of foods for
microbiological safety and quality. Acta Micro. Immunol. Hung. 47:175-182.
Lucas, J. J. 1979. Effect of hormone treatment on chick oviduct β-N-
acetylglucosaminidase isozymes and other acid hydrolases. Arch. Biochem. Biophys.
197:96-103.
Lush, I. E. and J. Conchie. 1966. Glycosidases in the egg albumen of the hen, the turkey
and the Japanese quail. Biochemica et Biophysica Acta. 130:81-86.
MacNeil, J. H. 1998. Waste egg shell components recovery. World Patent # WO
9841326.
MacNeil, J. H. 2001. Method and apparatus for separating a protein membrane and shell
material in waste egg shell. US Patent # US 6176376.
Martin, H. H. and S. Kemper. 1970. Endo-N-acetyl-glucosaminidase from Clostridium
perfringens, lytic for cell wall murein of Gram-negative bacteria. J. Bact. 102:347-350.
Masshoff, W. and H. J. Stolpmann. 1961. Licht- und electronmikroscopishe
Untersuchungen an der Schalenhaut und Kalkschale des Huhnereies. Z. Zellforsch.
55:818-832.
Matsuoka, Y., Hidaka, Y. and M. Yashima. 1966. Japanese Patent 41-150.
Meyer, K., Thompson, R., Palmer, J. W. and D. Khorazo. 1936. The purification and
properties of lysoyzme. J. Biol. Chem. 113:303-309.
Mead, P. S., Slutsker, L., Dietz, V., McCaig, L. F., Bresee, J. S., Shapiro, C., Griffin, P.
M. and R. V. Tauxe. 1999. Food-related illness and death in the United States. Emerg.
Infect. Dis. 1999. 5:607-625.
Moore, L. V. H., Bourne, D. M. and W. E. C. Moore. 1994. Comparative distribution and
taxonomical value of cellular fatty acids in thirty-three genera of anaerobic Gram-
negative bacilli. Int. J. Syst. Bacteriol. 44:338-347.
Nagao, E. and J. A. Dvorak. 1997. An integrated approach to the study of living cells by
atomic force microscopy. J. Microscopy. 191(1):8-19.
Naidu, A. S. 2000. Natural food antimicrobial systems. pp. 1-16. CRC Press LLC., Boca
Raton, FL.
Nikaido, H. 1996. Outer membrane. 2nd edn. In: Neidhardt, F. C. (Ed.). 1996. Escherichia
coli and Salmonella: Cellular and Molecular Biology, vol. 1. American Society for
Microbiology. Washington, DC. pp. 29-47.
42
Nikaido, H. and T. Nakae. 1979. The outer membrane of gram-negative bacteria. Adv.
Microb. Physiol. 20:163-250.
Nikaido, H., and M. Vaara. 1987. Outer membrane. pp.7-22. In F. C. Neidhardt (ed.).
Escherichia coli and Salmonella Typhimurium, vol. 1. Amer. Soc. for Microbiol.
Washington, DC.
North, M. O. 1978. Commercial chicken production manual. AVI Publishing Co.,
Westport, CT.
Ogawa, Y., Nakamura. R, and Y. Sato. 1983. Purificaiton of β-N-Acetylhexosaminidase
from egg white and the microsomal and lysosomal fractions of hen oviduct. Agric. Biol.
Chem. 47:2085.
Ogawa, Y., and R. Nakamura. 1983. β-N-Acetylhexosaminidase obtained from chicken
liver and chicken serum. Agric. Biol. Chem. 48(1):229-230.
Peterson, R. G. and B. Hartsell. 1955. The lysozyme spectrum of the Gram-negative
bacteria. J. Infect. Diseases. 96:75-78.
Poland, A. L. and B.W. Sheldon, 2001. Altering the thermal resistance of foodborne
bacterial pathogens with an eggshell membrane waste by-product. J. Food Prot. 64:486-
492.
Powrie, W. D. 1973. Chemistry of eggs and egg products. pp. 61-90. In W. J. Stadelman
and O. J. Cotterill (eds). Egg Science and Technology, AVI Publishing Co., Westport,
CT.
Powrie, W. D. 1977. Chemistry of eggs and egg products. pp. 65-91. In W. J. Stadelman
and O. J. Cotterill (eds). Egg Science and Technology, 2nd ed. AVI Publishing Co.,
Westport, CT.
Prescott, L. M., Harley, J. P., and D. A. Klein. 1999. Procaryotic cell structure and
function. pp. 37-71. In Microbiology, 4th ed. WCB/McGraw-Hill., Boston, MA.
Proctor, V. A. and F. E. Cunningham. 1988. The chemistry of lysozyme and its use as a
food preservative and a pharmaceutical. Crit. Rev. Food Sci. Nutr. 26:359-395.
Puvanakrishnan, R. and S. M. Bose. 1980. Studies on the immobilization of trypsin on
sand. Biotechnol. Bioengr. 22:919-928.
Ray, B., Johnson, C. and B. Wanismail. 1984. Factors influencing lysis of frozen
Escherichia coli cells by lysozyme. Cryo-Letters. 5:183-190.
43
Ray, B. 1992. The need for biopreservation. pp.1-23. In: B. Ray and M. Daeschel (eds.).
Food Biopreservatives of Microbial Origin. CRC Press, Boca Raton, FL.
Robinson, D. S. and N. R. King. 1968. Mucopolysaccharides of an avian egg-shell
membrane. J. Roy. Micr. Soc. 88:13-22.
Robinson, D. and J. L. Stirling. 1968. N-Acetyl-β-Glucosaminidases in Human Spleen.
Biochem. J. 107:321-327.
Russell, A. D. and D.Harries. 1968. Damage to Escherichia coli on exposure to moist
heat. Appl. Microbiol. 16:1394-1399.
Russell, A. D. 1998. Microbial susceptibility and resistance to chemical and physical
agents. In: Balows, A. & Duerden, B. I. (eds) Topley & Wilson’s Microbiology and
Microbial Infections Vol. 2, Systematic Bacteriology, pp. 149-184. Arnold, London.
Russell, A. D. 2003. Lethal effects of heat on bacterial physiology and structure. Sci
Progress. 86:115-137.
Salton M. R. J. and J. C. Pavlik, 1960. Studies of the bacterial cell wall. VI. Wall
composition and sensitivity to lysozyme. Biochimica et Biophysica Acta. 39:398.
Sandow, A. 1926. The anti-bacterial activity of egg white. Proc. Soc. Exp. Biol. Med.
24:172-175.
Scheie, P., and S. Ehrenspeck. 1973. Large surface blebs on Escherichia coli heated to
inactivating temperatures. J. Bacteriol. 114:814-818.
Schuman, J. D., Sheldon, B. W. and P. M. Foegeding. 1998. Thermal resistance of
Aeromonas hydrophila in liquid whole egg. J. Food Prot. 60:231-236.
Shockman, G. D., and J. V. Höltje. 1994. Microbial peptidoglycan (murein) hydrolases.
In New Comprehensive Biochemistry, vol. 27, Bacterial Cell Wall, pp131-167. Edited by
J.M. Guysen & R. Hakenbeck, Amsterdam: Elsevier Science.
Simkiss, K. 1958. The structure of the egg-shell with particular reference to the hen.
Ph.D. Thesis, University of Reading.
Simons, P. C. M. and G. Wiertz. 1963. Notes on the structure of membranes and shell in
the hen’s egg: an electron microscopical study. Z. Zellforsch. 59:555-567.
Smelt, J. P. P. M. 1998. Recent advances in the microbiology of high pressure
processing. Trends Food Sci. Tech. 9:152-158.
Smith, T. J., Blackman, S. A. and S. J. Foster. 2000. Autolysins of Bacillus subtilis:
multiple enzymes with multiple functions. Microbiology. 146:249-262.
44
Smolelis, A. N. and S. E. Hartsell. 1952. Factors affecting the lytic activity of lysozyme.
J. Bacteriol. 63:665.674.
Sparks, N. H. C. and R. G. Board. 1985. Bacterial penetraton of the recently oviposited
shell of hens’ eggs. Aust. Vet. J. 62:169-170.
Stadelman, W. J. 1973 Quality identification of shell eggs. Pp. 26-37. In W. J. Stadelman
and O. J. Cottrill (eds.). Egg Science and Technology. AVI Publishing Co., Westport,
CT.
Starcher, B. C. and S. G. King. 1980. The presence of desmosine and isodesmosine in
eggshell membrane protein. Connect. Tissue Res. 8:53-55.
Sun, H., Li, H., and P. J. Sadler. 1999. Transferrin as a metal ion mediator. Chem Rev.
99:2817-2842.
Takahashi, K., K. Shirai, M. Kitamura and M Hattori. 1996. Soluble egg shell membrane
protein as a regulating material for collagen matrix reconstruction. Biosci. Biotechnol.
Biochem. 60:1299-1302.
Tarentino, A. L. and F. Maley. 1971. Multiple forms of a highly purified β-N-
acetylhexosaminidase from hen oviduct. Arch. Biochem. Biophys. 147:446-456.
Taylor, L. W. and J. H. Martin. 1929. Factors influencing thickness of the eggshell. Poult.
Sci. 8:39-44.
Tomlins, R. I. and Z. J. Ordal. 1976. Thermal injury and inactivation in vegetative
bacteria. In: Skinner, F. A. & Hugo, W. B. (eds) Inhibition and Inactivation in Vegetative
Microbes. Society for Applied Bacteriology Symposium Series No. 5, pp. 153-190.
Academic Press, London.
Tortora, G. J., Funke, B. R., and C. L. Case (eds.). 1986a. Antimicrobial Drugs. pp. 493-
517. In Microbiology, An Introduction, 2nd ed. Ch.18. The Benjamin/Cummings
Publishing Company, Menlo Park, CA.
Tortora, G. J., Funke, B. R., and C. L. Case (eds.). 1986b. Functional anatomy of
prokaryotic and eucaryotic cells. pp. 79-112. In Microbiology, An Introduction, 2nd ed.
Ch.4. The Benjamin/Cummings Publishing Company, Menlo Park, CA.
Tranter, H. S. 1994. ‘Lysozyme, ovotransferrin and avidin’, in Dillion, V. M. and R. G.
Board. Natural Antimicrobial Systems and Food Preservation, CAB Intl, Wallingford,
UK.
Vakil. J. R., Chandan, R. C., Parry, R. M. and K. M. Shahani. 1969. Susceptibility of
several microorganisms to milk lysozymes. J. Dairy Sci. 52:1192-1196.
45
Valenti, P., Antonin, G., Von Hunolstein, C., Visca, P., Orsi, N. and E. Antonini. 1983.
Studies of the antimicrobial activity of ovotransferrin. Int. J. Tissue React. 1:97-105.
Valenti, P., Visca, P., Antonini, G., Orsi, N., and E. Antonini. 1987. The effect of
saturation with Zn2+ and other metal ions on the antibacterial activity of ovotransferrin.
Med. Microbiol. Immunol. Berl. 176:123-130.
Valenti, P., Visca, P., Antonini, G. and N. Orsi. 1985. Antifungal activity of
ovotransferrin towards genus Candida. Mycopathologia. 89:169-175.
Vaara, M. 1992. Agents that increase the permeability of the outer membrane. Microbiol.
Rev. 56:395-411.
Walton, H. V., Cotterill, O. J. and J. M. Vandepopuliere. 1973. Composition of shell
waste from egg breaking plants. Poult. Sci. 52:1192-1197.
Wang, C. and L. A. Shelef. 1992. Behavior of Listeria monocytogenes and the spoilage
microflora in fresh cod fish treated with lysozyme and EDTA. Food Microbiol. 9:207-
213.
Webb, S. J. 1969. The effects of oxygen on the possible repair of dehydration damage by
Escherichia coli. J. Gen. Microbiol. 58:317-326.
Weissmann, B., Hadjiioannou, S. and J. Tornheim. 1964. Oligosaccharase activity of β-
N-acetyl-D-glucosaminidase of beef liver. J. Biol. Chem. 239:59-63.
Wilkins, D. M. and R. G. Board. 1989. ‘Natural antimicrobial systems’, in Gould, G. W.
Mechanisms of Action of Food Preservation Procedures, Elsevier Appl. Sci. London.
Winn, S. E. and H. R. Ball, Jr. 1975. β-N-acetylglucosaminidase activity of the albumen
layers and membranes of the chicken’s egg. Poult. Sci. 54:799-805.
Winn, S. E. and H. R. Ball, Jr. 1996. Personal communication. North Carolina State
University. Raleigh, N. C. to Poland, A. L. and B.W. Sheldon, 2001. Altering the thermal
resistance of foodborne bacterial pathogens with an eggshell membrane waste by-
product. J. Food Prot. 64:486-492.
Witholt, B., Van Heerikhuizen, H., and L. De Leij. 1976. How does lysozyme penetrate
through the outer bacterial membrane? Biochemica et Biophysica Acta. 443:534-544.
Zahringer, U., Lindner, B. and E. Rietschel. Molecular structure of lipid A, the endotoxic
center of bacterial lipopolysaccharides. Adv. Carbohydr. Chem. Biochem. 50:211-276.
Zeuthen, P. and L. Bogh-Sorensen. 2003. Food preservation techniques. CRC Press,
Boca Raton, FL.
46
47
MANUSCRIPT 1
[to be submitted to the Journal of Protein Chemistry]
48
Identification of Eggshell Membrane Proteins and Purification
of Ovotransferrin and B-NAGase from Hen Egg White
G. J. Ahlborn1, D. A. Clare1, B. W. Sheldon1,2 and R. W. Kelly3
Department of Food Science1, Department of Poultry Science2, Department of
Chemical Engineering and Biotechnology3
North Carolina State University
Raleigh, NC 27695
49
Abstract
Exposure of selected Gram-positive and Gram-negative bacterial pathogens to egg shell
membranes (ESM) significantly reduced their thermal resistance and/or inactivated cells.
Although the components responsible for this antibacterial activity have not been
conclusively identified, several proteins associated with the ESM have been identified
including β-N-acetylglucosaminidase, lysozyme and ovotransferrin (each displaying
varying degrees of antibacterial activity). Numerous attempts to purify active fractions of
β-N-acetylglucosaminidase, lysozyme and ovotransferrin from the ESM proved
somewhat limited; however, hen egg white (HEW) β-N-acetylglucosaminidase was
purified using a two-step chromatographic procedure (isoelectric focusing followed by
cation exchange chromatography), from which pure fractions of ovotransferrin were also
obtained. SDS-PAGE electrophoresis and Matrix-Assisted Laser Desorption Time-of-
Flight Mass Spectrometry were then used to partially characterize the individual protein
components. Purified protein/enzyme fractions, such as these will be required in order
to fully elucidate the mechanism responsible for the antimicrobial properties associated
with the ESM.
Keywords: β-N-acetylglucosaminidase, ovotransferrin, isoelectric focusing, SDS-PAGE,
eggshell membrane, hen egg white
50
1. INTRODUCTION
The intact egg is naturally equipped with several physical and chemical defenses
aimed at protecting the developing embryo from physical harm and microbial invasion.
Primary physical barriers to microbial penetration of the egg include the cuticle, shell and
the inner and outer shell membranes. The light-pink colored membranes, composed of
highly cross-linked proteins similar to keratin, collagen and elastin, are structurally
similar to a meshwork of entangled threads which aid in obstructing invading
microorganisms (Baker and Balch, 1962).
Apart from the physical barriers, the avian egg is also equipped with a number of
chemical defenses which inhibit or prevent microbial growth in the albumen. Egg
albumen (composed of approximately 88% water, 10% protein, 1% carbohydrate, 0.5%
lipid and 0.5% minerals) contains lysozyme, avidin, conalbumin (ovotransferrin),
ovomucoid and other components that possess varying degrees of antimicrobial activity.
The alkalinity of the albumen, which increases from an initial pH of approximately 7.6 to
greater than 9.0 after one week at 25°C, is a major naturally-occurring antimicrobial
factor (Board and Tranter, 1994).
Ovomucoid inhibits proteolytic enzymes (i.e., trypsin inhibitor) important for
microbial growth (Garibaldi, 1960). Lysozyme is a bacteriolytic enzyme which catalyzes
the hydrolysis of β-1,4-glycosidic bonds of polysaccharides, thereby compromising the
peptidoglycan layer of bacterial cell walls and polymers of N-acetyl-D-glucosamine
(Jolles and Jolles, 1984). Avidin is believed to inhibit microbial growth by binding biotin
required by most all bacteria as an essential enzyme cofactor. Conalbumin (or
ovotransferrin) acts as a chelator, binding iron and possibly other minerals into a stable
51
complex and potentially producing a deficiency of essential minerals required for
invading microorganisms. Initially, it was believed that the antimicrobial properties of
ovotransferrin were solely due to iron deprivation, iron being an essential growth factor
for most bacteria (Alderton et al., 1946). However, several more recent reports have
described the retention of ovotransferrin antimicrobial activity, even when the protein is
complexed with metal ions such as zinc and iron (Valenti et al., 1987; Valenti et al.,
1985; Ibrahim, 1996).
The hen’s egg also contains a β-N-acetylglucosaminidase classified as a
lysosomal enzyme that contributes to the hydrolytic degradation of glycoproteins,
mucopolysaccharides and glycolipids (Weissmann et al., 1964; Lush and Conchie, 1966;
Robinson and Stirling, 1968; Winn and Ball, 1975). N-acetylglucosaminidase is one of
four different genera involved in cleaving chemical bonds inside the peptidoglycan layer
(Schockman and Höltje, 1994). Thus, the complex interactions among all of these
physical and chemical components within the intact egg provides the greatest protection
to the developing embryo, although Lifshitz et al. (1963) proposed that the inner shell
membrane may be the single most important barrier to bacterial penetration into the egg,
with shell membranes acting as bacterial filters, and containing active antibacterial
substances.
Histochemical studies using eggshell membranes confirmed the presence of
protein and firmly-bound sugars (Simkiss, 1958; Robinson and King, 1968). Starcher
and King (1980) identified desmosine and isodesmosine as the major cross-links within
these membranes, a finding later recognized as providing the eggshell membrane, itself,
with a high degree of insolubility and flexibility, an attribute that supports the egg white
52
before the formation of the shell (Takahashi et al., 1996). In addition, allysine (α-
aminoapipic-δ-semialdehyde), the reactive precursor of cross-links, and its aldol
condensation product, are present in significant quantities within shell membranes
(Crombie et al., 1981). Harris et al. (1980) showed that lysyl oxidase, located in a
copper-rich region of the isthmus of the hen oviduct, is required for the biosynthesis of
these cross-links. These investigators also found lysyl oxidase activity in shell
membranes, found to be coupled with catalase. This coupled enzyme complex was
considered to be involved in the biosynthesis of eggshell membranes, protecting the
embryo against hydrogen peroxide (Harris et al., 1980). Although these components
have not been specifically identified as ‘antibacterial’, several other proteins identified in
the ESM have been documented as exhibiting antimicrobial behavior.
Elliott and Brant (1957) were the first to report the presence of lysozyme in ESM,
while Hincke and colleagues (2000) recently determined that lysozyme was
heterogeneously distributed throughout the ESM using a colloidal-gold
immunocytochemical localization detection method. Winn and Ball (1975) initially
identified measurable amounts of β-N-acetylglucosaminidase activity within shell
membranes. Ovotransferrin was conclusively localized within the membrane using
immunofluorescence (Gautron et al., 2001).
To date, there are a limited number of studies that describe protein purification
from eggshell membranes. In the early work of Vadehra et al. (1972), lysozyme and
other minor protein components were extracted from the ESM using normal saline (0.9%
NaCl), phosphate buffers and distilled water solutions. Later, Wong et al. (1984) and
Takahashi et al. (1996) solubilized collagen-like egg shell membrane proteins through a
53
combination of performic acid oxidation and pepsin digestion. Most recently, Yi et al.
(2004) examined these collagen-like proteins in more detail.
Previous results from our own laboratory demonstrated that eggshell membrane-
bound components were capable of reducing the heat resistance and/or inhibiting the
growth of selected Gram-positive and Gram-negative foodborne bacterial pathogens
suspended in 0.1% peptone water (Poland and Sheldon, 2001). More specifically, 83 -
87% reductions in thermal decimal reduction times (D-values) for Salmonella enterica
serovars Typhimurium and Enteritidis (D54ºC), and Escherichia coli O157:H7 (D52ºC)
were noted. Also, a 3 log reduction in the population of L. monocytogenes was observed
following incubation for 45 min at 37ºC. The exact nature of the protein and non-protein
constituents responsible for these antibacterial properties was not identified. Therefore,
the focus of this study was designed to extract enzymatically and biologically active
fractions of β-N-acetylglucosaminidase, lysozyme, and ovotransferrin in order to evaluate
their function in providing ESM antimicrobial protection.
2. MATERIALS AND METHODS
2.1 Chemicals.
Clostridium histolyticum collagenase, porcine pancreatic elastase, and 4-
nitrophenyl N-acetyl-β-D-glucosaminide were obtained from Sigma Chemicals (St.
Louis, MO). Triton X-100, Chaps, Tween-20, polyethylene 9-lauryl sulfate (P-9-L),
guanadine HCl (GauHCl), urea, ammonium sulfate, sodium dodecyl sulfate (SDS) and
lithium dodecyl sulfate (LDS) were purchased from Fisher Scientific (Pittsburg, PA ).
54
Bacillus licheniformis PWD-1 keratinase was provided by Dr. Jason Shih (NCSU,
Raleigh, NC). All other chemicals used for these experiments were A.C.S. certified.
2.2 Eggshell membrane extraction.
One to four day-old eggs from 24-33 week old White Leghorn (WL) layers were
washed with a nylon brush in cool (15-18°C) water with 100 ppm sodium hypochlorite
and rinsed in sterilized deionized water. Eggs were broken, their contents emptied and
residual albumen removed with distilled, deionized water. Membranes were either
carefully extracted by hand ensuring that both the inner and outer membrane remained
intact or extracted according to a modified procedure (Winn and Ball, 1996; and Poland
and Sheldon, 2001) which mimics commercial processing techniques. Briefly, emptied
egg shells were placed in a KitchenAid® Food Processor (Model # KFP600WH)
containing 400 ml of sterile water, homogenized for 10 minutes, poured into a sterile
container, and left undisturbed for 5 minutes to allow shell fragment settling. The top
layer of the aqueous suspension, containing the membrane fragments, was decanted into a
buchner funnel containing Whatman #1 filter paper and vacuum-dried for 10 minutes.
An additional 400 ml of sterile water was added to the shell fragments to suspend any
residual ESM remaining from the extraction process, and then removed as described.
Filter cakes of the compacted membrane fragments were removed from the filter paper
and stored under refrigeration until further use (less than weeks) in a sterile petri dish
wrapped in aluminum foil.
2.3 Protein extraction from ESM
55
Numerous buffer and detergent solutions were evaluated for their capacity to
extract and solubilize ESM proteins (Table 1). For these trials, all buffers were prepared
using HPLC grade water and sterile filtered (0.22µm, Millipore) prior to use. Fifty ml of
each extraction solution was combined with 2 g fresh ESM in a 100-ml beaker with a
nylon stir bar. Extraction conditions included: 1) continuous stirring (200 rpm) for 2
hours at 25°C; 2) continuous stirring (200 rpm) for 2 hours at 37°C; 3) continuous stirring
(200 rpm) for 24 hours at 4°C; or a combination of the described conditions. Those
solutions resulting in the highest amount of active proteins/enzymes were further utilized
for purification. Following extraction, solubilized proteins were either concentrated to 1-
mL volumes in an Amicon Ultra-15 centrifuge concentrator tube (4,000 x g, 30-90 min)
(Millipore Corp, Bedford, MA) or salted out with a 65% solution of ammonium sulfate
(24 hr, 4°C). Precipitated proteins were resuspended in 1-mL volumes using 25 mM
sodium phosphate buffer (pH 6.0 – 7.2) diluent and dialyzed against the buffer prior to
use.
2.4 Preparation of egg white solution
Ten, four day-old eggs from WL layers (24-33 weeks of age) were cracked open,
and the albumen portion (~ 35 mL/egg) pooled, and then diluted with 350 mL of a 5 mM
sodium phosphate buffer solution (pH 5.0) containing 2 M urea. This mixture developed
a muciod precipitate which was removed by centrifugation (5,000 x g, 10 min, 4°C).
Following centrifugation, the egg white solution was stored at 4°C.
2.5 Protein Content
56
Protein concentration was measured using the BCA (bicinchoninic acid) assay kit
(Pierce Inc., Rockford, IL), and a standard curved developed with bovine serum albumin.
Assays were blanked using each extraction buffer and standard protein curves defined for
each buffering condition. All samples undergoing further purification steps were
dialyzed against 25 mM sodium phosphate (pH 6.0 or 7.2).
2.6 β-NAGase Assay
The release of p-nitrophenol from 4-nitrophenyl N-acetyl-β-D-glucosaminide was
followed using a modified procedure of Lush and Conchie (1966), Donovan and Hansen
(1971) and Winn and Ball (1975). For liquid assays, the incubation mixture contained
10-µl of the enzyme sample, 0.3 ml substrate (0.76 µmole 4-nitrophenyl N-acetyl-β-D-
glucosaminide in 0.1 M citrate buffer, pH 3.0) and 0.2 ml distilled, deionized water.
Samples were incubated at 37ºC for 25 min and the reaction stopped upon addition of
0.66 ml of 0.2 M Na2CO3 to the incubation mixture. The mixture was briefly vortexed
and the absorbency read at 400 nm using a Shimadzu UV 160U UV-Visible Recording
Spectrophotometer (Shimadzu Corp., Kyoto, Japan). Protocols for evaluating ESM
activity after extraction differed slightly. Under these experimental conditions, the
incubation mixture consisted of 0.01 g (± 0.001 g) of membrane fragments, 0.9 ml
substrate and 0.6 ml distilled, deionized water. This mixture was then incubated at 37ºC
for 25 min and the reaction stopped by adding 2.0 ml of 0.2 M Na2CO3. The contents
were briefly vortexed, the ESM removed, and the absorbency read at 400 nm. The
spectrophotometer was zeroed using the reaction solution alone and the enzymatic
activity was calculated according to the following equation:
57
Equation 1 = Units/mg = ∆ A400 / total volume___________
minutes * 10.8 * mg protein
where 10.8 = extinction coefficient of the substrate at the given absorbance
2.7 Isoelectric Focusing.
Isoelectric focusing was conducted with the RotoforTM Preparative IEF Cell
(BioRad Corp., Hercules, CA) powered by a LKB Bromma 2297 Macrodrive 5 constant
power supply (Bromma, Sweden) and cooled with the Isotemp Refrigerated Circulator
Model 901 (Fisher Scientific, Pittsburg, PA). A pH gradient was first established in the
focusing chamber by adding 38 ml Milli-Q water and 1.5 ml Bio-Lyte 6/8 (#163-1163) or
15µl Bio-Lyte 3/10 (163-2094) ampholytes. A voltage current was supplied by running
15 watts of power for approximately 40 minutes. Afterwards, the current was
temporarily turned off, 20 ml of the egg white or ESM solution added to the chamber,
and the current reapplied. The focusing chamber ran for approximately 2 hours, at which
point there no voltage change, indicating that isoelectric focusing was complete. Samples
(20) were removed from the chamber with the harvesting needle array and evaluated for
pH (Orion Expandable ionAnalyzer EA 940, Orion Research, Boston, MA).
Ampholytes were removed by adding 1M NaCl to the pooled sample
disassociating bound ampholytes from the proteins, and concentrated through
centrifugation at 4,000 x g for 30 min at 4°C using an Amicon Ultra-15 centrifuge tube
(Millipore Corp, Bedford, MA). Afterwards, buffer exchange (50 mM sodium
phosphate, pH 6.0) was accomplished through dialysis with a Pierce Slide-A-Lyzer®
Cassette (0.5-3.0 ml, 3,500 MWCO), which also functioned to remove residual
ampholytes and/or salt. All fractions were assayed for total protein, β-N-
58
acetylglucosaminidase activity, and/or evaluated for protein banding patterns using SDS-
PAGE.
2.8 Ion Exchange Column Chromatography
Ion exchange liquid chromatography was performed on the BioLogic LP low-
pressure chromatography system (BioRad) linked to an in-line UV detector (254 nm),
conductivity meter, and a fraction collector operated by the LP Data View software
system. An Econo-Pac® High S Cation or High Q Anion Exchange Cartridge (5 ml) was
employed for protein separation following isoelectric focusing. The liquid phase (Buffer
A) consisted of a 25 mM sodium phosphate buffer (pH 6.0 or 7.2, respectively), and
proteins were eluted (Buffer B) with the same buffer containing 1 M NaCl. HEW
proteins were removed from the column using step-wise gradient conditions (0-30 min
100% Buffer A; 30-40 min 60% Buffer A, 40% Buffer B; 40-50 min 50% Buffer A, 50%
Buffer B; 50-60 min 40% Buffer A, 60% Buffer B; 60-90 min 100% Buffer B) at a flow
rate of 1.5 mL/min. ESM proteins were removed from the column using linear gradient
conditions (0-30 min 100% Buffer A, 30-90 min 0-100% Buffer B) at a flow rate of 1.5
mL/min. Fractions (3 mL) were collected every 2 minutes and peaks corresponding to
protein elution were assayed for enzyme activity and analyzed for protein banding
profiles using SDS-PAGE. Eluted proteins were also analyzed by MALDI-TOF to
further verify protein identity based on mass.
2.9 SDS-PAGE
59
The purity of the protein/enzyme SDS-PAGE using the Invitrogen NuPAGETM 4-
12% Bis-Tris Gels (1.0 mm x 10 wells, #NP0321BOX) according to the procedures
adapted from the methods of Laemmli (1970). Lithium dodecyl sulfate was added to all
protein samples and heated at 75°C for 10 min prior to gel loading. Electrophoresis was
performed at a constant current (155 V) for 35 minutes using NuPAGETM MES SDS
Running Buffer (#NP0002). Protein staining was achieved with Coomassie Brilliant
Blue R-250, while destaining was accomplished using a clearing solution prepared with
25% methanol, 7.5% acetic acid and 67.5% water.
2.10 MALDI-TOF
Molecular weights for heterogeneous and homogeneous fractions of ESM and HEW
proteins were determined through Matrix Assisted Laser Desorption Ionization - Time
of Flight (MALDI-TOF) analysis. All samples were dialyzed against Milli-Q water or
10 mM sodium carbonate buffer (pH 7.0) after detergent removal. One µl of each
protein fraction (100 - 400 pmol/ml) was added to 20 µl of the matrix solution (30%
acetonitrile, 70% Milli-Q water and 0.1% trifluoroacetic acid), saturated with sinapinic
acid, and 0.5 – 1.0µl dispersed onto the target and allowed to dry (~ 15 min). The
analysis carried out using a Bruker Daltonics OmniFLEX™ bench-top MALDI-TOF
mass spectrometer (Billarica, MA. 01821) in reflection mode at an operation voltage of
19 kV according to standard protocols. The MALDI-TOF analyzer was calibrated with
pure fractions of lysozyme and bovine serum albumin and data analysis conducted with
the XMASS 5 (1.1) and Flex Control (1.1) acquisition programs.
60
3. RESULTS AND DISCUSSION
3.1 Extraction of ESM proteins
Extraction of ESM-bound proteins proved to be very difficult. Table 1 summarizes
numerous attempts to remove active proteins/enzymes according to pH, total protein
extracted, and β-NAGase activity. Milli-Q water or individual aqueous buffers had little
effect on solubilizing ESM proteins. With the addition of higher molar salt
concentrations, extraction improved only slightly. As evidenced by SDS-PAGE
analysis (Figure 1), most extraction solutions containing detergents and/or chaotropic
agents improved solubility of lysozyme, ovoalbumin and ovotransferrin. Assays prior to
protein extraction showed β-NAGase activity to be most abundant in the ESM; however,
solubilization of the enzyme was limited. Therefore, increasing total protein recovery
while improving β-NAGase yield became the focal point for evaluating all extraction
methodologies.
3.2 SDS-PAGE and MALDI-TOF Analysis
As depicted in Figure 1, SDS and LDS proved optimal, yielding the highest amount
of total protein (2.45 and 2.72 mg/ml, respectively). Only keratinase digestion proved
more effective with respect to total protein yield, although protein denaturation was
extensive and unsuitable for further analytical procedures. SDS and LDS fractions
exhibited the highest number of individual protein bands, specifically 7 identifiable SDS
bands as well as several undefined regions which may correlate with proteins (Figure 1,
Lane 2). Acidic versus basic conditions for protein extractions and detergent removal
protocols produced different peaks corresponding to solubilized proteins when samples
61
were analyzed with MALDI-TOF (Figure 2). Extractions performed at acidic pH in SDS
is presented in Figure 2a; while Figure 2b shows the proteins solubilized with SDS at a
basic pH. Calculating MALDI-TOF molecular weights in a heterogeneous protein
solution can result in slight variations compared to actual molecular weight
determinations. Therefore both purified and heterogeneous ESM and HEW fractions
were analyzed for comparison.
Although several of the peaks were barely distinguishable, the molecular weight for
approximately 11 of them was determined and correlated with known ESM protein bands
(Figure 1). A faint band (labeled A) corresponds to peaks at approximately 4-5,000 kDa,
while band B represents one or more peaks between 9 and 11 kDa. Band C was
identified as lysozyme at 14.4 kDa, while a double protonated molecule of lysozyme was
observed at 7.2 kDa and a lysozyme dimer was likely shown in the ‘expanded’ 28 kDa
peak. A faint band, (D), (obtained using basic pH protocols) was observed at 22.3 kDa,
and may represent ovoglycoprotein or a collagen-like protein described by Takahashi et
al. (1996). The stained region labeled (E) possibly corresponds to ovomucoid with a
MW of 28.7 kDa and as previously mentioned might represent a lysozyme dimer. (F)
represents a mixture of several proteins, including ovoalbumin at 45 kDa. Peak G at 53.3
kDa, was identified as β-NAGase and band (H) as ovotransferrin at (77.8 kDa). Band I
(at 90 kDa) was not identified. Ovomucin, (band J) at 110 kDa, was not detected through
MALDI analysis. Band K, a minor peak at 68.3 kDa, might represent a β-NAGase
isozyme based on the previous observations of Ogawa et al. (1983).
It appears that the majority of proteins from the hen egg white may also be present in
the shell membrane. This hypothesis is supported by the fact that the membrane and egg
62
white are both formed in the hen oviduct and components of each may easily become
integrated with each other. However, further analysis of purified proteins from both
ESM and HEW must be compared more closely (i.e. comparing the amino acid
sequences of each).
Ionic detergents, in general, serve as efficient solubilizing agents but have a greater
tendency to denature proteins by destroying native three dimensional structures which
can be crucial for functional activity. This was apparent in protein samples that exhibited
an SDS-banding pattern identical to active HEW β-NAGase as reported in early literature
(Ogawa and Nakamura, 1983; Ogawa et al., 1983). However, these solubilized protein
samples were devoid of β-NAGase enzymatic activity as a result of detergent extraction.
Nonionic detergents are less denaturing and can often be used to improve solubility
of membrane proteins and retain biological functionality. Triton-X 100 and Tween 20,
two nonionic detergents prepared in phosphate buffer, proved effective by increasing the
total amount of recovered protein (0.84 and 0.81 mg/ml respectively). With the addition
of 2 M GauHCl (a chaotropic agent) and DTT (disulfide reductant) to the detergent
solutions, the total protein extracted increased almost 3-fold (2.27 and 1.98 mg/ml).
However, despite SDS-banding patterns and MALDI-TOF peaks corresponding to
established molecular weights for active β-N-acetylhexosaminidase activity, β-NAGase
activity was absent from both the protein precipitate and membrane fragments after
extraction. These results indicate that both Triton-X and Tween irreversible disrupted the
native structure of β-NAGase. Although somewhat limited, 2 M GauHCl containing 25
mM DTT in 50 mM sodium phosphate buffer (pH 5.8) was found to be the only
experimental condition for extracting protein fractions that show measurable β-NAGase
63
activity. Further increases in either GauHCl or DTT concentrations did not improve
recovered β-NAGase activity in the extracts.
Although the ESM has been characterized as a network comprised of cross-linked
proteins similar to keratin, collagen and elastin (Baker and Balch, 1962), only PWD-1
keratinase demonstrated any hydrolytic activity towards the ESM. After 1 hour,
keratinase digestion released over 2 mg/ml of total protein and after a 6 hour treatment
over 4.4 mg/ml was recovered. As anticipated, keratinase hydrolysis (which randomly
cuts all proteins), produced a large, unidentifiable smeared band, between 1 and 6 kDa as
evidenced by SDS-PAGE (data not shown). β-NAGase activity was also absent in the
digest, and the loss of enzymatic activity on partially digested fragments was inversely
related to keratinase exposure time. Digestion with elastinase, collagenase or a
combination of the two did not improve protein recovery, nor did it decrease β-NAGase
activity in the treated ESM. Exposure of ESM to either immobilized or soluble trypsin
slightly improved the quantity of protein extracted, however, these treatments did not
solubilize active fractions of β-NAGase.
Not only can detergents denature proteins, leading to a loss of functional activity, but
many interfere with subsequent analyses, such as mass spectroscopy, chromatography,
and other analytical procedures. Detergents that are not bound to proteins are often
dialyzable under certain conditions; whereas bound detergents (covalent or non-covalent)
often present more of a challenge. The ProteoSpinTM Detergent Clean-up Micro Kit
(Norgen Biotek, Canada) was used to remove up to 95% of either SDS or Triton-X with a
minimal loss of protein. Although most of the detergent was successfully removed, a loss
of total protein occurred, ranging from 10 to 35% of the total.
64
3.3 Purification of HEW β-NAGase and Ovotransferrin
A purification table for β-NAGase from HEW is presented in Table 2. The specific
activity of the starting material was relatively low (2.93 x 10-3 U/mg). After isoelectric
focusing, the majority of β-NAGase activity was typically isolated in 3 or 4 sample
chambers, which exhibited a pH of approximately 5.5-6.5 (Figure 3). The isoelectric
focusing protocol was repeated with the remaining samples and fractions exhibiting
enzyme activity greater than the starting activity (i.e. A400nm > 0.2) were pooled.
Afterwards, the specific activity had increased (1.37 x 10-2 u/mg) although only 63 % of
β-NAGase was recovered.
Cation exchange chromatography is based on electrostatic interactions between
the protein and the surface matrix of the ion exchange resin. Ideally, there are three to six
interaction points involved per protein molecule. Positive counterions from the resin are
displaced by the protein, itself, as the positive surface charges on the protein adhere.
Proteins bind to the cation exchange resin if the pH of mobile phase (buffer) is lower than
the isoelectric point of the protein. The protein is then eluted with an increasing ionic
strength buffer (1 M NaCl in this case). Figure 4 depicts the chromatogram obtained
upon elution of the HEW proteins from the column while Figure 5 illustrates the SDS-
PAGE banding profile characteristics of various samples collected throughout the
purification process. The primary protein peak was visualized between 30 and 40
minutes of the run, and this sample (Figure 5, Lane 4) was later identified to be a very
pure fraction of ovotransferrin while the smaller peak (Figure 5, Lane 5) was determined
to be β-NAGase. Following chromatography, only 37% of the initial activity was
65
recovered, although the specific activity increased to 0.464 U/mg reflecting a relative
purification of over 150 fold (Table 2).
Previously, Tarentino and Maley (1971) purified two isozymes of β-N-
acetylglucosaminidase from hen oviduct with reported molecular weights of 118,000
(Type I) and 158,000 daltons (Type II) and pI values of 6.45 and 6.86, respectively.
Kinetic constants and enzymatic activity towards various substrates showed virtually no
differences between these two isozymic forms, although it was established that the Type I
enzyme was the predominant species. Lucas (1979) later showed that only Type I
isozyme was found in egg white and suggested that it may play a role in the catabolism of
glycoproteins in both egg white and the oviduct.
Contrary to these findings, Ogawa and Nakamura (1983) and Ogawa et al., (1983)
purified a β-N-acetylhexosaminidase (NAHase) from hen oviduct, egg white, chicken
liver and chicken serum. Oviduct and liver NAHase was reported to have a similar MW
of 53,000 Daltons while the size of serum and egg white enzymes were reported as
68,000 Daltons. They hypothesized that these variations may be due to the synthesis of
lysosomal precursors that exhibited a higher subunit molecular weight than those of
mature forms. Kato and colleagues (1997) also purified an endo- β-N-
acetylglucosaminidase from the hen oviduct with a molecular weight of 52,000 Daltons
by SDS-PAGE analysis and 54,000 Daltons using gel filtration methods. More recently,
the complete genome for the chicken (gallus gallus) has been published, including the
sequence for β-N-acetylglucosaminidase, estimated to be approximately 98,000 Daltons.
This finding suggests a post-translational modification of β-NAGase. Our own data
evidenced the molecular weight of ESM and HEW β-NAGase as 53,200 and 52,500
66
Daltons respectively, although minor amounts of a 68,300 dalton ESM protein was
identified through MALDI-TOF analysis. The later may represent a precursor form of
the enzyme. Future work will be needed to fully characterize and complete the amino
acid sequences for both the ESM and HEW enzymes, which may then be compared to the
genomic sequence for further verification.
Fraenkel-Conrat and Feeney (1950) were the first to separate ovotransferrin from
other egg white proteins using ammonium sulfate precipitation at pH 3. However, this
method altered the electrophoretic behavior of the protein. Since that study, ion
exchange (Rhodes et al., 1958; Azari and Baugh, 1967; Jeltsch and Chambon, 1982;
Guerin and Brule, 1992; Croguennec et al., 2001) and affinity chromatography (Al-
Mashikhi and Nakaï, 1987; Chung et al., 1991) techniques have been used successfully
for purifying ovotransferrin. Herein, the separation of ovotransferrin, from β-NAGase
purification, resulted in a protein fraction of a greater purity than several commercially
prepared samples. MALTI-TOF analyses identified the molecular weights of HEW and
ESM ovotransferrin as approximately 77,400 and 77,800 Daltons, respectively.
MALDI-TOF analysis established the molecular weight of ESM lysozyme as very
similar to that of the HEW lysozyme standard (14,423 and 14,407 Daltons, respectively).
Hen egg white lysozyme purification methodology has been refined, hence the
commercial availability of the enzyme. As we were only able to purify small amounts of
lysozyme from ESM, our preliminary findings suggest that purification of lysozyme from
ESM using these methods is not cost effective as compared to purchasing these materials
from commercial sources.
67
β-N-acetylglucosaminidase, lysozyme and ovotransferrin were purified from ESM
and HEW. Although they appear to be very similar in molecular weight, further studies
are required to determine whether differences exist in both conformation and activity.
The ESM, itself, provides a very stable matrix for enzymes/protein components (Ahlborn
et al., 2005) and has the potential for use a ‘natural’ processing adjuvant. However,
further research is required to identify more effective methods for solubilizing and
extracting ESM proteins. Eventually, these types of studies will lead to the identification
of specific membrane components that provide antibacterial activity as well as deliver
protection to the developing chick embryo.
68
References
Al-Mashikhi, S. A., and Nakaï, S. (1987). Agric Biol Chem. 51,2881–2887
Alderton, G., Ward, W. H., and Fevold, H. L. (1946). Arch Biochem Biophys. 11,9-15.
Ahlborn, G. J., and Sheldon, B. W. (2005). Poult. Sci (in review).
Azari, P., Baugh, R. F. (1967). Arch. Biochem. 118,138–144.
Baker, J. R., and Balch, D. A. (1962). J. Biochem. 82,352-361.
Board, R. G., and Tranter, H. S. (1994). In Egg Science and Technology 4th ed.
(Stadelman, W. J., and Cotterill, O. J. eds). The Haworth Press, Inc. Binghamton, NY,
pp. 81-104.
Chung, C. M., Chan, S. L., and Shimizu, S. (1991). Int. J. Biochem. 23,609–616
Croguennec, T., Nau, F., Pezennec, S., Piot, M., and Brulé, G. (2001). Eur. Food Res.
Technol. 212,296-301.
Crombie, G., Snider R., Faris, B., and Franzbrau C. (1981). Biochim. Biophys. Acta.
640,365-367.
Donovan, J. W., and Hansen, L. U. (1971). J. Food Sci. 36,174-177.
Elliott, L. E., and Brant, A. W. (1957). Food Research. 22,241-250.
Fraenkel-Conrat, H., and Feeney, R. E. (1950). Arch Biochem. 29,101-113.
Garibaldi, J. A. (1960). Food Res. 25,337-344.
Gautron, J., Hincke, M. T., Panheleux, M., Garcia-Riuz, J. M., Boldicke, T., and Y. Nys.
(2001). Connect Tissue Res. 42,255-67.
Guerin, C., and Brule, G. (1992). Sci. Aliment. 12,705–720.
Harris, E. D., and Blount. J. E. (1980). Science. 208,55-56.
Hincke, M. T., Gautron, J., Panheleux, M., Garcia-Ruiz, J., McKee, M. D., and Nys, Y.
(2000). Matrix Biology. 19,443-453.
Ibrahim, H. R. (1996). In Hen Eggs: Their Basic and Applied Science, CRC Press, Inc.,
New York, pp 37-57.
Jeltsch, J. M., and Chambon, P. (1982). Eur. J. Biochem. 122,291–295.
69
Jolles, P., and Jolles, J. (1984). Mol. and Cell. Biochem. 63,165-189.
Kato, T., Hatanaka, K., Mega, T., and Hase, S. (1997). J. Biochem. 122,1167-1173.
Laemmli, U. K. (1970). Nature. 227,680-685.
Levvy, G. A., and Conchie, J. (1966). In Methods in Enzymology, Complex
Carbohydrates (Colwick, S. P., and Kaplan, N. O. eds.). Academic Press, New York, pp.
571-580.
Lifshitz, A., Baker, R. C., and Naylor, H. B. (1963). J. Food Sci. 28,94-99.
Lucas, J. J. (1979). Arch. Biochem. Biophys. 197,96-103.
Lush, I. E., and Conchie, J. (1966). Biochem. et Biophys. Acta. 130,81-86.
Ogawa, Y., and Nakamura, R. (1983). Agric. Biol. Chem. 48,229-230.
Ogawa, Y., Nakamura, R., and Sato, Y. (1983). Agric. Biol. Chem. 47,2085-2089.
Poland, A. L., and Sheldon, B. W. (2001). J. Food Prot. 64,486-492.
Rhodes, M. B., Azari, P. R., and Feeney, R. E. (1958). J. Biol Chem. 230,399–408.
Robinson, D. S., and King, N. R. (1968). J. R. Microsc. Soc. 88,13-22.
Robinson, D., and Stirling, J. L. (1968). Biochem. J. 107,321-327.
Simkiss, K. (1958). Ph.D. Thesis, University of Reading.
Starcher, B. C., and King, S. G. (1980). Connect. Tissue Res. 8,53-55.
Shockman, G. D., and Höltje, J. V. (1994). In New Comprehensive Biochemistry
(Guysen, J. M., and Hakenbeck, R. eds), Elsevier Science, Amsterdam, pp131-167.
Takahashi, K., Shirai, K., Kitamura, M., and Hattori, M. (1996). Biosci Biotechnol
Biochem. 60,1299-1302.
Tarentino, A. L., and Maley, F. (1971). Arch. Biochem. Biophys. 147,446-456.
Vadehra, D. V., Baker, R. C., and Naylor, H. B. (1972). Comp. Biochem. Physiol.
43,503-508.
Valenti, P., Antonin, G., Von Hunolstein, C., Visca, P., Orsi, N., and Antonini, E. (1983).
Int. J. Tissue React. 1,97-105.
70
Valenti, P., Visca, P., Antonini, G., and Orsi, N. (1985). Mycopathologia. 89,169-175.
Valenti, P., Visca, P., Antonini, G., Orsi, N., and Antonini, E. (1987). Med. Microbiol.
Immunol. Berl. 176,123-130.
Weissmann, B., Hadjiioannou, S., and Tornheim, J. (1964). J. Biol. Chem. 239,59-63.
Winn, S. E., and Ball, H. R. Jr. (1975). Poult. Sci. 54,799-805.
Winn, S. E., and Ball, H. R. Jr. (1996). Personal communication. North Carolina State
University. Raleigh, N. C.
Wong, M., Hendrix, M. J., von der Mark, K., Little, C., and Stern, R. (1984. Dev Biol.
104,28-36.
Yi, F., Guo, Z. X., Zhang, L. X., Yu, J., and Li, Q. (2004). Biomaterials. 25,4591-4599.
71
Table I. Protein extraction data of ESM with various chemicals.
Extraction agents
pH
Total Protein
Extracted
(mg/ml)
Supernatant
β-NAGase
activity (U/ml)
ESM
β-NAGase
activity (U/mg)
Milli-Q water
6.8 <0.01 ND 12.3
50 mM sodium phosphate buffer
0.1 M NaCl
1 M NaCl
1 M NaCl, 5 mM EDTA
7.0 0.02
0.05
0.08
0.09
ND
ND
ND
13.1
12.2
10.8
11.1
50 mM sodium phosphate buffer (RT/Cold)
2 M GuaHCl ,
2 M GuaHCl , 25 mM DTT
2 M Urea
2 M Urea, 25 mM DTT
5.8
6.4
0.42
0.84
0.10
0.20
ND
ND
3.02 x 10-3
ND
2.2 x 10-4
10.1
9.7
12.3
11.8
50 mM potassium phosphate buffer
PWD-1 Keratinase 1 hour
Keratinase 6 hours
8.0
2.1
4.4
ND
ND
10.1
5.4
50 mM Tris buffer
Elastase
Collagenase (0.2 M Ca2Cl)
Elastase and Collagenase (0.2 M Ca2Cl)
Trypsin
Pepsin
7.0
8.1
<0.01
<0.01
0.03
0.03
0.25
0.18
ND
ND
ND
ND
ND
ND
10.7
12.8
11.6
10.1
15.4
11.1
50 mM sodium phosphate buffer
1% Triton X-100
1% P-9-L
1% Chaps
1% Tween-20
2 M GauHCl, 25 mM DTT with:
1% Triton X-100
1% P-9-L
1% Chaps
1% Tween-20
6.0
5.8
0.02
0.84
0.52
0.66
0.81
2.27
1.43
1.22
1.98
ND
ND
ND
ND
ND
ND
ND
ND
ND
12.5
0
3.8
11.1
2.7
ND
3.1
3.4
ND
50 mM sodium phosphate buffer
SDS
LDS
6.0
2.45
2.72
ND
ND
ND
ND
ND = activity not detected
72
Table II. Purification table of β-NAGase from HEW
Purification Step
Activity
(U)
Protein
(mg)
Sp act
(U/mg)
Recovery
(U/ml)
Relative
Purification
Initial content
Diluted HEW
0.088
30.0
2.93 x10-3
100%
1
Isoelectric Focusing
Rotofor®
0.213
15.6
1.37 x10-2
63%
4.66
Ion exchange
chromatography:
High S cation column
0.0153
0.033
0.464
37%
158.4
73
MW- kDa
98 ← K
← I
←H
J→
52 ← G L→
← F
← E
31
←D
←C
14
←B
←A
Figure 1. SDS-PAGE of ESM bound proteins extracted under various conditions. 1: MW Marker,
2: LDS/β-mercap, 3: 50 mM NaP/2 M GauHCl/25 mM DTT, 4: 1% P-9-L, 5: 1% Chaps, 6: 1%
Triton X-100, 7: 1% Tween-20, 8: 50 mM NaP/2 M Urea/25mM DTT, 9: 1% Chaps with 2 M
Urea/25mM DTT, 10: 1% Tween 20 with 2 M Urea/25mM DTT. The faint band labeled (A) may
correspond to peaks at approximately 4-5,000 kDa, while band (B) might represent one or more
peaks between 9 and 11 kDa. Band (C) is lysozyme at 14.4 kDa. A faint band (D) represents the
peak from Figure 2b (basic pH protocols) at 22.3 kDa, which may be ovoglycoprotein or a collagen-
like protein. The stained region labeled (E) could correspond to ovomucoid with a MW of 28.7 kDa.
(F) may represent a mixture of ovoalbumin at 44.6 kDa with 48 kDa lysyl oxidase. (G) represents a
combination of three proteins between 51 and 55 kDa, with the peak at 53.2kDa, being β-NAGase.
Band (H) is ovotransferrin at 77.8 kDa, while (I) may represent peaks at 90 kDa (an unidentified
protein). Although there were no corresponding MALDI peaks, (J) or (K) could be ovomucin at 110
kDa. (L) potentially represents the peak at 68.3 kDa, also a β-NAGase protein.
74
a.i. a
b
Figure 2. (a) MALDI-TOF analysis of ESM protein mixture. The system calibrated to lysozyme and
bovine serum albumin. Peaks at 4,289 and 5,889 Daltons are unidentified. The peak at 14,423 is
lysozyme, the peak at 7,232 is a double protonated lysozyme molecule. The peak at 28,769 is thought
to correspond to ovomucoid but may also contain a lysozyme dimer. The peak at 44,659 is believed
to be ovoalbumin. The peak at 53,200 was identified as β-N-acetylglucosaminidase, while the
surrounding peaks are unknown. Ovotransferrin is represented by the peak at 77,845, but may also
correspond to a protonated dimer molecule as well. (b) is the analysis of a more homogeneous
fraction of protonated molecule at 22,396, which is thought to be a collagen-like protein. The peaks
at 11,237 and 44.822 correspond to a double protonated molecule and a dimer of the 22,396 protein.
Peaks at 67,358 and 90,436 are unidentified but may also correspond to dimers/trimers of other
proteins. An N-acetylglucosaminidase could correspond to the peak at 68,903.
b
75
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Rotofor Chamber
pH
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Abs
400
Abs 400 pH
Figure 3. pH gradient and A400nm values, reflective of β-NAGase activity in samples collected during
isoelectric focusing of hen egg white.
76
0
10
20
30
40
50
60
70
80
90
100
010203040506070
Time (min)
- - - - - Conductivity (mS/cm)
-0.001
0.004
0.009
0.014
0.019
0.024
0.029
_______
UV [A.U. 254]
Peak 1→
← Peak 2
Figure 4. Cation-exchange chromatograph during purification of ovotransferrin and β-N-
acetylglucosaminidase.
77
1 2 3 4 5 6
Figure 5. SDS-PAGE of hen egg white protein purification. Lanes 1: MW marker, 2: Crude HEW, 3:
Rotofor IEF, 4: Ovotransferrin from cation exchange (77.4 kDa), 5: β-NAGase - Cation exchange
(52.5 kDa) and 6: Ovotransferrin - Anion exchange
78
a.i. a.
b
Figure 6. MALDI-TOF analysis of purified HEW ovotransferrin at full scale (a) and partial scale
(b).
79
a.i. a.
b.
Figure 7. MALDI-TOF of ‘pure’ B-NAGase fraction at full scale (a) and partial scale used to
compare molecular weights of purified protein samples.
80
MANUSCRIPT 2
[to be submitted to the Journal of Food Protection]
81
Identifying the Components in the Eggshell Membrane
Responsible for Reducing the Heat Resistance of
Bacterial Pathogens
Gene Ahlborn1 and Brian W. Sheldon2
Department of Food Science1 and Poultry Science2
North Carolina State University
Raleigh, NC 27695
82
Abstract
The biological activity (D-value determination) of eggshell membranes (ESM) was
examined to determine the membrane components responsible for their observed
antibacterial activity and their mechanisms of action. Salmonella enterica serovar
Typhimurium suspended in peptone water was exposed to ESM at ratios of 1:10 to 1:60
(grams of ESM to ml of bacterial suspension, ca 107-8 CFU/ml). D54°C-values at the 1:10
ratio were significantly reduced compared to the control (0.69 min and 5.34 min
respectively). As ESM ratios decreased, D-values followed a linear increase until
reaching the 1:60 ratios at which point there was no longer a significant difference in
D54°C-values compared to the control (5.02 min versus 5.34 min). Biological and
enzymatic activities (i.e., -N-acetylglucosaminidase, lysozyme and ovotransferrin) of
ESM treated with trypsin, lipases or heat denatured were compared to those of untreated
ESM. Trypsin treated ESM lost all biological activity (D54°C-values of 5.12 and 5.38 min
for immobilized and solubilized trypsin, respectively), but showed no significant loss of
enzymatic activities. Treatments with porcine lipase and a lipase cocktail did not impact
biological or enzymatic activities. Heat denaturation (80 and 100°C for 15 min) of ESM
resulted in significant decreases in biological activity (3.99 and 4.43 min D-values,
respectively) and complete loss of -N-acetylglucosaminidase (-NAGase) activity.
Lysozyme and ovotransferrin activities remained, but were significantly reduced.
Purified ESM and hen egg white (HEW) components (i.e., -NAGase, lysozyme and
ovotransferrin) were added to S. Typhimurium suspensions (in 0.1% peptone water) at
varying concentrations to evaluate their biological activity. Lysozyme or -NAGase
alone resulted in D54°C-values of 4.50 and 3.68 min, respectively. Ovotransferrin was
83
more effective yielding a 2.44 min D54°C-values, while combinations of ovotransferrin
with lysozyme or -NAGase yielded similar D-values (2.08 and 2.23 min). However, a
combination of all three components produced a D-value of 1.47 min, which was similar
to the ESM treatment. Exposing S. Typhimurium cells to a mixture of ovotransferrin,
lysozyme and -NAGase or ESM resulted in significant increases in extracellular
concentrations of Ca+2, Mg+2, and K+. Sodium losses were only detected with purified
components. Transmission electron microscopy of S. Typhimurium cells treated with
ESM and purified components indicated structural damage to the outer membrane,
although cell lysis was only visible with a combination of ovotransferrin, lysozyme and
-NAGase. It was hypothesized that these components (which are also found in the
ESM) removed cations (i.e., Ca+2 and Mg+2) required for lipopolysaccharide (LPS)
stability, effectively sensitizing bacteria to heat and possibly pressure, osmotic changes or
other stressors.
Keywords: eggshell membrane, β-N-acetylglucosaminidase, lysozyme, ovotransferrin,
antimicrobial, mechanism
84
Introduction
In 1999 Mead and colleagues (1999), in conjunction with the Centers for Disease
Control and Prevention, presented their data on the incidences of foodborne illnesses in
the United States. They estimated that over 76 million cases of foodborne illness occur
annually resulting in more than 325,000 hospitalizations and over 5,000 deaths.
Although more than 250 foodborne diseases have been described, known bacteria (e.g.,
Campylobacter, Salmonella, E. coli O157:H7), viruses (e.g., Norwalk-like, caliciviruses),
and parasites (e.g., Giardia, Cyclospora) account for an estimated 14 million illnesses. It
is difficult to calculate the exact cost as a result of such illnesses; however, medical costs
and lost wages due to foodborne salmonellosis, only one of many foodborne infections,
have been estimated to be more than $1 billion per year (Anonymous, 2003).
Implementing preventative means such as Good Manufacturing Practices
(GMPs), Good Agricultural Practices (GAPs), and Hazard Analysis Critical Control Point
programs (HACCP) have certainly made a positive impact at reducing the risks of
foodborne illnesses. In 2002, a report from the CDC’s Emerging Infections Program
Foodborne Diseases Active Surveillance Network (FoodNet) showed a decrease in the
major bacterial foodborne illnesses. However, some infections failed to show a steady
decline in their occurrence, indicating that increased efforts are needed to further reduce
the incidence of foodborne illnesses, especially considering the emergence of new, more
resilient pathogens (e.g., multidrug-resistant Salmonella).
In the last decade, consumers have shown an increasing demand for minimally
processed foods and sustained functionality of naturally occurring bioactive ingredients.
Several factors have fueled the public’s interest in exploring options for minimally
85
processed foods and bioactive compounds to enhance nutrition including concerns about
the safety and tolerance of synthetic preservatives; the suspected link between the
overuse of subtherapeutic antibiotics as animal growth promotants, the development of
multi-drug resistance in microbes, as well as the increased media attention given to diet
and health (Naidu, 2000).
A number of products have been approved by regulatory agencies for use as direct
food antimicrobials. However, the limited spectrum of antimicrobial activity of some of
these substances has led to the continued search for more effective antimicrobials among
naturally occurring compounds (Davidson and Zivanovic, 2003). Although not often
associated with food safety, great potential lies in the area of egg shell waste. In
preliminary studies, Poland and Sheldon (2001) demonstrated that eggshell membrane-
bound components were capable of reducing the heat resistance and/or inhibiting the
growth of selected Gram-positive and Gram-negative foodborne bacterial pathogens
suspended in 0.1% peptone water. Reductions in thermal decimal reduction times (D-
values) of 83 - 87% were observed for Salmonella enterica serovars Typhimurium and
Enteritidis (D54°C) and Escherichia coli O157:H7 (D52°C) and up to a 3 log reduction in L.
monocytogenes populations following incubation with eggshell membranes (ESM) for 30
min at 37ºC.
The pink-colored ESM consists of an outer and inner membrane with a thickness
of approximately 50 µm and 15 µm respectively. The outer membrane is located just
inside the shell and the inner membrane is located between the outer membrane and
albumen. Their structure is similar to a meshwork of entangled threads which aids in
obstructing invading microorganisms. Lifshitz et al. (1963) concluded that the inner shell
86
membrane may be the most important single barrier to bacterial penetration into the egg,
with shell membranes acting as a bacterial filter and also containing antibacterial
substances.
Several compounds possessing known antimicrobial characteristics have been
identified in the ESM. Elliott and Brant (1957) were the first to report the presence of
lysozyme in eggshell membranes and Winn and Ball (1975) identified measurable
amounts of -N-acetylglucosaminidase activity within the membranes. Ovotransferrin
was later identified by immunofluorescence as being another component of the ESM
(Gautron et al., 2001). These proteins as well as other unidentified compounds may
contribute to the antimicrobial properties exhibited in the ESM.
Methods to extract these enzyme-rich shell membranes are readily available
(MacNeil, 2001; Winn and Ball, 1996) and offer egg processors potential economic value
as a ‘natural’ processing adjuvant in food or pharmaceutical products to sensitize
bacterial pathogens and spoilage organisms to heat or other treatments. However, a
greater understanding of the egg shell membrane (i.e. the components responsible for the
antimicrobial activity and possible mechanisms of action), the focus of our studies, is
essential to better determine how it may be used in practical applications.
87
Materials and Methods
Chemicals
Chicken egg white lysozyme (EC 3.2.1.17), chicken egg white ovotransferrin
(#C-0755), porcine pancreatic lipase (EC 3.1.1.3, #L-3126), wheat germ lipase (EC
3.1.1.3, #L -3001), Thermomyces lanuginosus lipase (EC 3.1.1.3, #L-0902), Candida
albicans lipase (EC 3.1.1.3, #L-4777), Micrococcus lysodeikticus ATCC 4689 (#M-3770)
and 4-nitrophenyl N-acetyl--D-glucosaminide (#N-9376) were all obtained from Sigma
Chemicals (St. Louis, MO). TPCK-treated bovine pancreas trypsin (#T-1426) was also
obtained from Sigma Chemicals and immobilized on controlled pore glass beads
following the procedures described by Janolino and Swaisgood (1982). BHI broth and
agar were obtained from Difco Laboratories (Detroit, MI). Hen egg white ovotransferrin
and -N-acetylglucosaminidase were obtained using the procedures described by Ahlborn
and colleagues (2005). All other chemicals and buffers used were certified A.C.S grade.
Bacteria and culture conditions
Salmonella enterica serovar Typhimurium ATCC 14028 was obtained from the
American Type Culture Collection (Rockville, MD) and maintained in double strength
brain heart infusion (BHI) broth (Difco Laboratories, Detroit, MI) plus 20% glycerol and
stored at –20ºC. Working stock cultures were propagated using two successive transfers
in BHI broth (37ºC, 18-24 hr) prior to use.
Individual 10-ml cultures were prepared by transferring a 0.1-ml subsample of
Salmonella Typhimurium (ST) to 10-ml of fresh BHI broth (2 h, 37ºC). The cells were
pelleted by centrifugation (15 min at 5ºC at 8,000 x g), resuspended in sterile 0.1%
88
peptone water (PW), and centrifuged again. Following resuspension in 10-ml PW, optical
density at 600 nm was determined and adjusted 0.5 (ca 107-8 CFU/ml).
Eggshell membrane extraction
Egg shell membranes were extracted using a ‘commercial-like’ modified
procedure described by Poland and Sheldon (2001). One hundred fresh (within 2 days of
lay), non-fertile, White Leghorn (Hyline) eggs were washed with a nylon brush in cool
(15-18°C) water containing 200 ppm sodium hypochlorite and rinsed in sterilized
deionized water. Eggs were broken, their contents emptied and residual albumen
removed by rinsing with distilled, deionized water. Approximately one-third of the
emptied egg shells were ground for 10 minutes in a KitchenAid® Food Processor (Model
# KFP600WH) containing 400 ml of sterile water before being poured into a sterile
container and left undisturbed for 5 minutes to allow shell fragments to settle. The top
layer of the aqueous suspension containing the membrane fragments was decanted into a
buchner funnel containing Whatman #1 filter paper and vacuum-dried for 10 minutes.
An additional 400 ml of sterile water were added to the shell fragments to remove any
residual membrane fragments remaining from the first process. This process was
repeated with the remaining shells. Filter cakes of the compacted membrane fragments
were removed from the filter paper, pooled together and stored in a sterile petri dish
wrapped in aluminum foil and refrigerated until used (less than 1 week).
β-NAGase Assay
The release of p-nitrophenol from 4-nitrophenyl N-acetyl--D-glucosaminide was
followed using a modified procedure of Lush and Conchie (1966), Donovan and Hansen
(1971), and Winn and Ball (1975). The incubation mixture containing 0.01 g (± 0.001 g)
89
of membrane fragments, 0.9 ml substrate (0.76 µmole 4-nitrophenyl N-acetyl--D-
glucosaminide in 0.1 M citrate buffer, pH 3.0) and 0.6 ml distilled, deionized water was
incubated at 37ºC for 25 min, and the reaction stopped by adding 2.0 ml of 0.2 M
Na2CO3. The mixture was briefly vortexed, the ESM removed, and the absorbency (at
400nm) read in a narrow absorption cell (1-cm light path) using a Shimadzu UV 160U
UV-Visible Recording Spectrophotometer (Shimadzu Corp., Kyoto, Japan) blanked
against the reaction solution minus the ESM. Enzyme activity was calculated according
to Equation 1. ESM samples (treated and non-treated) were evaluated in triplicate, and
the values presented are the average of the three.
Equation (1) = Units/mg = A400 / total volume___________
minutes * 10.8 * mg protein
where 10.8 = extinction coefficient of the substrate at the given absorbance
Lysozyme Assay
Lysozyme activity was determined using an adaptation of the assay described by
Shugar (1952) as measured by the change in optical density (450 nm) following exposure
of Micrococcus lysodeikticus to the ESM. A cell suspension was produced by adding
0.015% (w/v) lyophilized Micrococcus lysodeikticus cells (Sigma, # M-3770) to 66 mM
potassium phosphate buffer (pH 6.24) at 25°C. The resulting suspension absorbance
(450nm) was between 0.6 and 0.7. Random samples (0.01 g ± 0.001 g) from various
ESM treatments were added to 5.0 ml of the M. lysodeikticus suspension and mixed by
inversion. After two minutes (designated as the optimal time to observe enzyme activity
as indicated by a linear decrease in OD), ESM were removed, the decrease in absorbance
recorded, and enzyme activity calculated according to Equation 2. ESM samples (treated
90
and non-treated) were evaluated in triplicate, and the values presented are the average of
the three.
Equation (2) units/mg solid = A450 / time (in minutes)__
0.001 x mg/solid/reaction mix
where 0.001 = change in absorbance at A450nm as per the Unit definition
Ovotransferrin activity
The iron-binding properties of the ESM were determined using a modified
protocol described by Tranter et al. (1983) and Crogennec et al. (2001). Briefly, 5 ml of
a solution (6 mg/ml) of filtered, (0.45 µm, Millipore) ferric ammonium sulphate
[Fe(NH4)2(SO4)2 · 6H2O] was added to 0.030 g (± 0.002 g) ESM. The solution was
incubated with agitation (37°C for 30 min), after which shell membranes containing
bound iron were washed 3 times is sterile deionized water and ashed in a muffle furnace
(600°C, 24 hours). The ash was acid digested (6 N HCl) and iron content determined
through Inductively Coupled Plasma (ICP) emission spectroscopy. Ovotransferrin iron-
binding activity was determined in triplicate (n=3) and 6.5 µg/g (for iron occurring
naturally in the ESM) subtracted from iron values obtained.
D-value determinations
Decimal reduction times (D-values) were determined using the combined
methods of Poland and Sheldon (2001) and the immersed sealed capillary tube procedure
described by Schuman and colleagues (1998) and Foegeding and Leasor (1990).
Salmonella enterica serovar Typhimurium ATCC 14028 (ST) was selected on the
91
grounds that in preliminary trials cells from ST showed the greatest sensitivity to ESM
(Poland and Sheldon, 2001; Ahlborn and Sheldon, 2005). Mid-log phase cell populations
of ST were produced by culturing cells for 24 hours in 10 ml BHI broth at 37ºC in a
rotating water bath shaker (Model G76D, New Brunswick Scientific Co., Inc., Edison,
NJ), followed by transfer of 100 µl of the cell suspension into 10 ml of BHI broth and
incubation at 37ºC for 2 hours. Cells were harvested by centrifugation (15 min at 4ºC at
8,000 x g), washed in sterile 0.1% PW, centrifuged again as described and suspended in
10 ml of 0.1% PW (ca. 107-8 CFU/ml). Zero to one gram of the pooled and treated ESM
fragment extracts were added to 20 ml of the peptone bacterial suspension and incubated
with agitation at 37ºC for 30 minutes.
Following incubation, ESM were removed from the peptone bacterial suspension
[using a Stomacher bag containing a nylon filter (Stomafilter P-type, Gunze Sangyo, Inc.,
Tokyo Japan)] and the inoculated PW (0.05 ml) was dispensed into individual glass
capillary tubes (0.8 to 1.1 mm i.d. by 90 mm long; no. 9530-4, Fisher Scientific,
Pittsburg, PA) using a syringe fitted with a 100-mm blunt needle (head space of
approximately 4mm). Filled capillary tubes were then heat-sealed using a propane torch
and kept on ice until heated (less than 30 minutes). Tubes were placed upright in a mesh-
screen-covered test tube rack and rapidly submerged into a preheated (54ºC) circulating
water bath (Model W19, Haake, Inc., Karlsruhe, Germany) with a temperature control
module accurate to ±0.05ºC (Model DC1, Haake, Inc., Karlsruhe, Germany). At six to
eight evenly spaced intervals, duplicate tubes were removed from the water bath and
immersed in an ice-water slurry for 5-10 minutes. Capillary tubes were cleansed of any
residual contamination by submerging the tubes in sodium hypochlorite (500 ppm, pH
92
6.5) for 5 seconds followed by three successive rinses in sterile, deionized water. Tubes
were aseptically transferred to individual test tubes containing 5 ml of sterile PW and
finely crushed using a sterile glass rod. This initial 10-2 dilution was then agitated on a
vortex mixer, serially diluted in PW and spiral plated on BHI agar using a Spiral Biotech
Autoplater 4000 (Spiral Biotech Inc, Norwood, MA). Plates were incubated at 37ºC for
18-24 hours and colonies enumerated using a Microbiology International ProtoCOL
automatic colony counting system (Model 60000, Synoptics Ltd., UK).
Triplicate thermal inactivation trials were conducted and survivor curves (log S.
Typhimurium/ml versus heating time) were plotted for each trial from the various
treatments described below. Best-fit linear regression lines were drawn through each
data set and D-values were calculated as the negative reciprocal of the survivor slope
obtained by regression analysis. D-values represent the average of the three thermal
inactivation trials.
Statistical Analysis
Mean decimal reduction times obtained via the ISCT procedure were calculated
comparing the control (no membrane) treatment with the experimental (with membrane
variables) treatment. Statistical analysis of the D-values was determined by analysis of
variance (ANOVA, P 0.05) and the means separated by comparison of each mean pair
using the student’s t-test ( LSD, P 0.05). The residual replicate by treatment mean
square was used for testing the main effects (treatment, replicates). Statistical analysis of
enzyme activity studies were determined using the General Linear Model (GLM) and
Least Squares Means (LSM) with P 0.05 (SAS Institute, Inc., 1990).
93
Treatments
Minimum Inhibitory Concentration.
Ten ml of the PW/bacterial suspensions (ca log 7-8 CFU/ml) were placed in seven
sterile Erlenmeyer flasks. One gram (representing a 1:10 ratio), 0.5 g (1:20 ratio), 0.33 g
(1:30 ratio), 0.25 g (1:40 ratio), 0.2 g (1:50 ratio) and 0.17 g (1:60 ration) of ESM were
respectively added to one of the flasks. The seventh flask (without membrane) served as
the control. Bacterial suspensions were incubated with agitation (150 rpm) at 37º for 30
minutes. Following incubation, all suspensions were poured into a sterile filtering-
stomacher bag, and bacterial suspensions were removed with a sterile pipette and
aseptically transferred to a sterile test tube (effectively removing all membrane fragments
from the suspension). Samples were placed on ice (< 10 min) until transferred to
capillary tubes for D-value determination as described above.
Trypsin and immobilized trypsin treated ESM.
Nine samples of 1.5 grams of ESM were placed in sterile 50-ml polypropylene
graduated tubes (30 x 115 mm). Three tubes each received one of the following
treatments in 46 mM Tris buffer (pH 8.1) with 12 mM CaCl2: (1) 20 ml of trypsin (200
units/ml) in the Tris buffer; (2) 20 ml of Tris buffer with 3 ml immobilized trypsin (97.8
units/ml); (3) 20 ml of Tris buffer as a control. Tubes were laid horizontally in a
controlled environment incubator/shaker (New Brunswick Scientific-Series 25, Edison,
N. J.) and incubated (37ºC, 100 rmp) for 3 hours. Samples were removed and
membranes were rinsed 5 times in sterile, ddH2O to remove any residual trypsin from the
membranes. Excess water was removed from the membranes by pat-drying them with
94
sterile Kimwipes (Kimberly-Clark Corp, Roswell, GA). One gram was removed from
each sample, added to 20 ml of the previously described bacterial suspension. Bacterial
suspensions were incubated with agitation (150 rpm) at 37º for 30 minutes. Following
incubation, all suspensions were poured into a sterile filtering-stomacher bag and
bacterial suspensions removed with a sterile pipette and aseptically transferred to a sterile
test tube (removing all membrane fragments from the suspension). Samples were placed
on ice (< 10 min) until transferred to capillary tubes as described above. Remaining ESM
fragments were evaluated for enzymatic activity as previously described.
Heat inactivation of ESM proteins.
Shell membranes were placed in sterile, deionized water and heated to either 80 or
100°C for 15 minutes after which membranes were rinsed with sterile water and excess
water removed as described above. One gram from each sample was added to 20 ml of
the bacterial suspension, incubated, and treated as previously described for D-value
determination and enzymatic activity.
Treatment of ESM with lipase.
Three treatments consisting of either (1) a buffer control, (2) porcine lipase (4,000
units), or (3) a combination of porcine lipase (2,000 units), wheat germ lipase (300 units),
Thermomyces lanuginosus lipase (5,000 units) and Candida albicans lipase (300 units),
were added to 100 ml of 50 mM sodium phosphate buffer (pH 7.0) and placed in a sterile
120 ml beaker containing five grams of ESM from a common pool and a Teflon stirbar.
Membrane treatments were incubated (4 hours, 37°C) with mild stirring (100 rpm) and
95
stored (4°C) overnight. The following morning, treatments were removed from the
refrigerator and incubated for an additional 2 hours (37°C, 100 rpm). Following
incubation, membranes were removed from the treatments and rinsed with five
successive rinses of 300 ml sterile ddH2O (at which point lipase activity was no longer
present in the ESM or rinsate samples). One gram ESM samples were removed, added to
20 ml of the bacteria suspension, and D-values and enzyme activities were determined as
described. Lipase activity was determined according to the procedures of Rajeshwara
and Prakash (1994).
D-value determination with purified components of ESM and hen egg white.
Due to the difficulty in obtaining sufficient amounts of purified ESM proteins
(i.e., -N-acetylglucosaminidase and ovotransferrin), the ISCT procedure previously
described was modified using 1 to 2 ml bacterial suspensions (ca 107-8 CFU/ml). D-
values were determined by incubating (30 min, 37°C) various concentrations of the
purified fractions of ovotransferrin (0.2µM – 6.0µM), lysozyme (0.02 µM - 0.2 µM) and
-N-acetylglucosaminidase (0.01 µM – 0.02 µM) with the bacterial suspension D-values
calculated as described.
Transmission Electron Microscopy
Individual cultures samples were prepared by transferring a 0.2-ml subsample of
Salmonella Typhimurium to 20-ml of fresh BHI broth (2.5 h, 37ºC). The cells were
pelleted by centrifugation (15 min at 5ºC at 3,000 x g), resuspended in sterile 0.1% PW,
and centrifuged again. Following resuspension in 10-ml PW, optical density at 600 nm
96
was recorded (approximately 0.5 or 107-8 CFU/ml). Cells then received one of seven
treatments: (1) control- no additional treatment, (2) 1 g fresh ESM, (3) 0.2 µM
ovotransferrin, (4) 0.04 µM lysozyme (5) 0.2 µM ovotransferrin and 0.04 µM lysozyme,
(6) 0.2 µM ovotransferrin and 0.01 µM -NAGase, or (7) 0.2 µM ovotransferrin, 0.04
µM lysozyme, and 0.01 µM -NAGase. All samples were incubated with agitation at
37ºC for 30 minutes and the ESM removed as described above. Samples were
centrifuged (as described), the supernatant discarded and the pellets resuspended in 4 ml
of 3% GTA in 0.1 M Na cacodylate buffer (pH 7.2) and fixed at 4°C for 1 week.
Tubes were then centrifuged (3,000 x g, 7 min) and the supernatant removed.
Pellets were transferred to microfuge tubes with 1-ml 0.1 M sodium cacodylate buffer,
held for 10 min at 4°C and centrifuged for 2 min. All samples were rinsed with the same
buffer for two additional 15-min changes. Following removal of the last supernatant, 1.0
ml of 2% osmium tetroxide in 0.1 M sodium cacodylate buffer was added to resuspend
the cells and then the cells were fixed for two hours on ice in the dark. Samples were
centrifuged as described above, osmium tetroxide removed and washed an additional two
times in clean buffer before being held overnight at 4°C. The following day, samples
were centrifuged (2 min, 3,000 x g), rinsed and centrifuged as described. The pellet was
resuspended in 2% agarose (in 0.1 M sodium cacodylate buffer) and centrifuged (5 min,
3,000 x g). The samples were chilled on ice for 30 minutes and the gelled agarose
removed from the microfuge tubes. The tip containing cell samples were cut off with a
clean razor blade, cut into blocks and placed in cold buffer. The blocks were rinsed with
a brief distilled water wash and En-Bloc stained with 1% aqueous uranyl acetate in the
dark (4°C, 12 hours).
97
Following staining, samples were rinsed with a change of cold distilled water and
then dehydrated with 1-hour changes of cold 30%, 50%, 70%, 95% and 100% ethyl
alcohol (EtOH) warming to room temperature and followed by two changes in 100%
EtOH at room temperature (23°C). Infiltration proceeded 24-hours later by subsequent
changes of 1:2 Spurr’s reagent: EtOH, 1:1 Spurr’s reagent: EtOH, and 3:1 Spurr’s
reagent: EtOH, and concluded with an 8-hour change of 100% Spurr’s reagent (23°C),
and a 5-hour change of 100% Spurr’s under a vacuum. Samples were embedded
overnight with Epon 812 - Araldite M epoxy resin mixture in BEEM® capsules.
Ultrathin sections were cut using diamond knives on the ultramicrotome Reichert–Um
03, placed on 200-mesh nickel supporting grids for transmission electron microscopy,
stained with 4% uranyl acetate for 1 hour, and then followed by a 0.2% aqueous solution
of Reynolds lead citrate (Reynolds, 1963) at room temperature. Each stain was followed
by triple distilled water washes (1 min, 23°C). Examination took place using a JOEL
100S Transmission Electron Microscope JEM 100C operated at 80 kV.
Atomic emission spectrometry
Loss of intracellular cations were determined using an adapted protocol from
Aguilera et al., (2003). Salmonella Typhimurium cells were obtained by transferring 3
ml of S. Typhimurium cells from a 24 hour stock culture to 200 ml of BHI broth and
incubating in a circulating water bath (2 hr, 37°C). Cells were centrifuged (8,000 x g, 15
min), washed three times (in either Milli-Q or peptone water) and resuspended (A600=2.0)
in either deionized Milli-Q water or 0.1% PW (Difco). Suspended cells (10 ml) were
incubated at 37°C for 30 min in the presence or absence of ESM (2.0 g), ovotransferrin
98
(0.5 µM), and a combination of ovotransferrin (0.5 µM), lysozyme (0.08 µM) and -
NAGase (0.01 µM). Aliquots (10 ml) were filtered through a stomacher bag containing a
nylon filter (to remove ESM; all samples however received the same treatment),
centrifuged at 8000 × g for 10 min, the supernatants collected, filtered through a 0.45µm
syringe filter, and Ca+2, Mg+2, K+ and Na+ quantified using an atomic emission
spectrometry detector (UNICAM 929, Unicam Limited, Cambridge, UK). Controls
(bacteria-free) containing only Milli-Q water or peptone water alone and with either
ESM, ovotransferrin or a combination of enzymes were evaluated independently. The
trial was duplicated, with samples run in duplicate.
Results and Discussion
Minimum Inhibitory Concentrations.
Figure 1 shows the effects of various concentrations of ESM on the D54-values for
Salmonella Typhimurium. At 54°C, the control treatment (no added ESM) required over
5 minutes of heating time (5.34) to yield a 1 log reduction in bacterial population. When
eggshell membranes were added at a ratio of 1 g of membrane to 10 ml of the S.
Typhimurium suspension, the D54°C-value was significantly reduced over 7-fold (to 0.69
min). Significant increases in the D54°C-values were detected as the concentration of
ESM to cell population decreased. At a ratio of 1 g to 60 ml of the bacterial suspension,
there was no significant difference in the heat resistance of S. Typhimurium compared to
the ESM-free control treatment (5.02 min vs 5.34 min).
Trypsin Treated ESM
99
Due to the difficulty in extracting and purifying the components comprising ESM,
an indirect approach was first taken to access the impact of individual components on the
observed antimicrobial activity of ESM. The biological activities (reduction in D54°C-
values) of ESM following exposure to various treatments are shown in Figure 2.
Subjecting the ESM to suspended or immobilized trypsin resulted in all loss of
antimicrobial activity as depicted by significant increases in D54°C-values(5.38 min and
5.12 min respectively) compared to the 1:20 ESM treatment (1.65 min). Trypsin is a
protease which cleaves proteins between the lysine and arginine bond. Although loss of
the biological activity was not surprising, it was interesting to note that when ESM was
exposed to either solubilized or immobilized trypsin, there were no significant reductions
in -NAGase activity and a slight decrease in lysozyme activities as measured by the
release of p-nitrophenol from 4-nitrophenyl N-acetyl--D-glucosaminide or change in
optical density (450 nm) following exposure to Micrococcus lysodeikticus, respectively
(Table 1). This finding may be related to several possibilities: 1) the active sites
responsible for enzymatic activity lack lysine-arginine bonds and are therefore unaffected
by trypsin; 2) the active sites are partially protected by membrane components; or 3)
trypsin is able to cleave specific sites in the enzyme yielding peptide fragments adhered
to the ESM that work cooperatively to produce the specific enzymatic activity. Unlike
the enzyme activities, the iron-binding capabilities of membrane-bound ovotransferrin
were significantly reduced by exposure to trypsin (Table 1).
Heat Inactivation of ESM-bound components
Membranes subjected to heat treatments (80 and 100°C for 15 min) prior to
exposure to S. Typhimurium cells lost significant biological activity (i.e., from 1.65 min
100
to 3.99 and 4.43 min respectively). Although the heat ‘inactivated’ membranes retained
some biological activity, it was greatly diminished. If protein components of ESM were
primarily responsible for the reduction in D-values, heat denaturation of these proteins
would occur to varying degrees with exposure to heat. Membrane-bound -NAGase was
more susceptible to heat degradation, with no activity detected following the 15-min heat
treatments. Membrane-bound ovotransferrin retained some activity while lysozyme
retained up to a third of its activity (14.0 ± 4.8 U/mg at 80°C; 4.3 ± 2.3 U/mg at 100°C)
(Table 1). The retention of lysozyme activity is supported by the work of Masschalck
and colleagues (2001) in which partially heat-denatured lysozyme (80°C, 20 min)
retained 11% of its activity.
Treatment of ESM with lipase
Lipid components, particularly fatty acids, have been observed to have
antimicrobial characteristics in some instances (Isaacs and Thormar, 1991; Coonrod,
1987; Thormar, 1987). Although the lipid content (dry weight) of ESM is only 1.35%
(Suyama et al., 1977), we elected to treat ESM with porcine lipase and a lipase cocktail to
evaluate whether lipid components may contribute to the observed antimicrobial activity.
After treatment with the lipases, ESM were thoroughly rinsed to remove the lipase and
D54°-values determined. ESM biological activity was not affected by treatment with the
lipases (D54-values of 1.54 and 1.18 min respectively) demonstrating that lipid
components are not probable contributors to the ESM biological activity. Furthermore,
these two lipase treatments also did not affect ovotransferrin, lysozyme and -NAGase
activities (Table 1).
101
D-value determination of purified ESM and HEW components
As previously mentioned, the insolubility of the ESM and the difficulty in
obtaining purified and active components necessitated that hen egg white (HEW) be used
as a source of identified egg membrane components (i.e., ovotransferrin, lysozyme, -
NAGase). However, it should be remembered that purified components from HEW and
their membrane-bound counterparts may exhibit different characteristics both in structure
as well as activity (biological and enzymatic). Inhibitory concentrations equivalent to
that exhibited by ESM (i.e. D-value reductions) were first determined for the “purified”
HEW components using enzymatic values for lysozyme and -NAGase activities as the
reference (Ahlborn and Sheldon, 2005).
Using the 1.65 min D54°C-value as a benchmark (Figure 2), 0.04µM of lysozyme
and 0.02µM of purified -NAGase represented the equivalent enzyme activity to
membrane-bound enzymes. As ovotransferrin concentrations (as determined by iron-
binding capabilities) varied greatly between membrane samples, D54°-values for a range
of ovotransferrin (0.1 to 4.0 µM) were determined (Table 2). Using D-value
determination protocol described previously, 0.2 µM of purified ovotransferrin resulted
in significant reductions in D54°-values. Increasing ovotransferrin concentrations above
0.2 µM did not significantly decrease D54°-values, while decreasing amounts of
ovotransferrin resulted in reduced biological activity (increased D54°C-values) (Table 2).
Ovotransferrin alone at 0.2 µM significantly reduced D54°-values (2.44 min versus
5.34 min) compared to the ST control (Figure 3) while lysozyme at 0.04 µM also resulted
in a significant reduction in D54°-values (3.68 min) although the values were not as low as
those produced by ovotransferrin. Incubation of ST cells with 0.02 µM -NAGase
102
showed no significant reduction in D54°-values (4.92 min) and reductions in population
were not observed for any treatment alone (data not shown). D54°-values for treatment
with lysozyme and ovotransferrin in combination were 2.08 min, while -NAGase and
ovotransferrin together resulted in a mean D54°-value of 2.30, neither of which were
different from ovotransferrin alone. A combination of all three components was required
to yield similar D54°-values as detected following ESM treatment (1.47 min compared to
1.65 min), indicating that these three components are in fact responsible for the ESM
antibacterial activity.
Atomic Emission Spectrometry
Extracellular concentrations of Ca+2, Mg+2, K+ and Na+ were determined
following treatment of Salmonella Typhimurium cells with ESM, purified ovotransferrin,
or a combination of ovotransferrin, lysozyme and -NAGase in Milli-Q water.
Moreover, the release of these ions following treatment of S. Typhimurium cells with
ESM in peptone water was also examined (Table 3). Aguiler and colleagues (2003)
demonstrated that ovotransferrin as well as other transferrins are able to cause selective
ion efflux through bacterial membranes. They reported that extracellular concentrations
of K+ increased when E. coli cells were treated with transferrins while Na+ concentrations
were not affected. Although the electrical potential of the cell () was significantly
reduced and even lost in some instances, they observed no loss of E. coli viability.
Similar to the observations by Aguiler and colleagues, treatment with ovotransferrin
alone resulted in significant extracellular concentrations of K+ while sodium
concentrations were unaffected. When lysozyme and -NAGase were used in
combination with ovotransferrin significant concentrations of extracellular K+, Na+, Mg+2
103
and Ca+2 were observed. Compared to the controls (ESM in Milli-Q water and S.
Typhimurium suspended in Milli-Q water), an increase in extracellular concentrations of
Mg+2, Ca+2, and Na+ were also observed when S. Typhimurium cells were treated with
ESM. Eggshell membrane treatments in peptone water yielded significant cellular losses
of K+, Na+, Mg+2 and Ca+2 concentrations compared to ESM treatments in Milli-Q water.
Transmission Electron Microscopy
The outer membrane normally functions as a protective barrier for Gram-negative
bacteria and lies outside the thin peptidoglycan layer. It serves in preventing/slowing the
entry of toxic substances (i.e. bile salts, antibiotics) that might kill or injure the
bacterium. The LPS layer constitutes the majority of the outer membrane and is located
adjacent to the exterior peptidoglycan layer. Although complex, the outer membrane is
not impermeable and can permit the passage of small molecules (ca. 600-700 Daltons)
through porin proteins which cluster together and span the outer membrane (Prescott et
al., 1999).
Lipopolysaccharide associations in Gram-negative bacteria are also stabilized by
the presence of divalent cations (i.e. Mg2+, Ca2+) in the outer membrane, which serve to
decrease electrostatic repulsions and increase LPS-LPS relations. These strong
associations are believed to be the primary reason why large hydrophilic molecules and
most hydrophobic and amphiphilic molecules are prevented from gaining access to the
cell (Nikaido and Nakae, 1979; Nikaido and Vaara, 1987). Alterations to the LPS,
including removal of the divalent cations that stabilize the outer membrane, result in
compromised functioning of this lipid bilayer barrier (Vaara, 1992).
104
Transferrins, including ovotransferrin, have been shown to bind aluminum,
copper, and zinc in the identical manner of iron uptake. (Li and Yang, 2004; Garrett et
al., 1991). Bovine lactoferrin has also been recently shown to bind calcium (Rossi et al.,
2002). Ovotransferrin alone appears to be unable to bind/effect the Mg2+and Ca2+ cations
through similar actions, however, ovotransferrin with the addition of lysozyme and -
NAGase (or ESM-bound ovotransferrin, lysozyme and -NAGase) is able to affect
cellular losses of these and possibly other cations. Whether or not the Mg2+and Ca2+ in
the resulting extracellular fluid is from the LPS or from inside the cell itself requires
more analysis.
Through transmission electron micrographs (Figures 4 and 5) ovotransferrin and
the ESM treated S. Typhimurium cells show some indication of cellular disruption to the
outer membrane/LPS integrity. Compared to a more compact or tightly joined outer
membrane/LPS in the control micrographs (Figure 4a), ovotransferrin-treated cells
(Figure 4b) appear to be less ‘compact’, while ESM treated cells (Figure 4c) appear to
have an outer membrane with an increased loss in structural integrity. Salmonella cells
treated with a combination of ovotransferrin, lysozyme and -NAGase exhibit a dramatic
difference in cellular integrity (Figure 4d), where outer membranes are ‘jagged’ (or
misshaped) and cellular material is scattered throughout the micrograph (Figure 5d),
indicating cell lysis.
As previously noted, the outer membrane and LPS-LPS associations help to
prevent larger molecules from gaining access into the cell. Loss of LPS Mg2+and Ca2+
could explain the observed disruption to the outer membrane and LPS structure. If this is
the case, hydrolases such as lysozyme and -NAGase may gain access to the
105
peptidoglycan and further disrupt cellular integrity. This hypothesis is supported by the
work of Handcock and Wong (1984) where they assessed the impact of nitorcefin on the
permeability of Pseudomonas aeruginosa cells in the presence of the cation chelator
EDTA. They theorized that EDTA sequestered Mg+2 ions essential for LPS-LPS cross-
bridges. They also observed that the addition of Mg+2 ions countered the permabilizing
effect by perhaps saturating EDTA with the supplemented Mg+2 ions, thereby preventing
any further action on cellular Mg+2. Stan-Lotter and colleagues (1979) also demonstrated
that the addition of cations (Mg2+and Na+) to the growth medium substantially reduced
the inhibitory effect of several antibiotics and the dye gentian violet on S. Typhimurium
mutants. More recently, Branen and Davidson (2004) demonstrated that the addition of
either EDTA or lactoferrin enhanced the activity of nisin, monolaurin and lysozyme
against Listeria monocytogenes, Escherichia coli, and Salmonella Enteritidis. These
findings suggest that ovotransferrin combined with lysozyme and -NAGase may remove
cations essential for proper LPS stability.
Differences in D-values resulting in treatment of bacteria cells with ESM versus
the purified components might be attributed to “inaccessibility” issues. The purified
ovotransferrin and enzymes have greater accessibility to the cell structures, whereas ESM
lysozyme and -NAGase are bound in the matrix of the membrane and are probably not
able to associate as readily with the cells. Although damage to the outer membrane by
ESM (i.e. loss of divalent cations) may not be bactericidal, it may result in a weakened
cell that exhibits greater susceptibility to damage through heat, pressure and osmotic
stressors. Hitchener and Egan (1977) demonstrated this effect in exposing E. coli K-12
cells to EDTA and sublethal heat (48°C). While sublethal heat treatments alone resulted
106
in the release of less than 20% of LPS components and EDTA produced some structural
damage to the LPS, these individual treatments did not impact death rates. However, the
combination of sublethal heat and EDTA resulted in a 50% loss of the LPS components
from the cells and a significant decrease in D48°C -values. Moreover, the addition of
0.5mM Mg+2 to the heating medium again provided a protection to the cells from death
and structural injury. Another possible reason for the decreased activity of ESM-bound
components may be attributed to the high content of cations present in the membrane
itself, including high levels of Na+, Ca2+, Cu2+, and to a lesser extent Mg2+ which may
interfere or limit sequestering or binding of extraneous cations.
It is also important to note that S. Typhimurium cells treated with ESM showed a
0.4 log (±0.1 log) reduction in population after incubation for 30 min at 37°C versus a 0.8
log reduction (±0.1 log) for cells treated with a combination of the purified proteins.
Intracellular protein loss of treated versus non-treated S. Typhimurium cells was also
determined and then subsequently examined by SDS-PAGE. There was indication of
intracellular protein loss from treatment with both ESM and purified membrane
components (0.03 µg/ml and 0.08 µg/ml respectively). SDS-PAGE revealed only two
additional bands (which were also evident in the supernatant of sonicated S.
Typhimurium cells) in the treated cell supernatant that were not found in the gel lanes
containing ESM or enzymes alone, indicating that the bands are S. Typhimurium
intracellular proteins (Figure 6).
CONCLUSIONS
From the above work, ovotransferrin, lysozyme and -N-acetylglucosaminidase
from the ESM are the primary components responsible for the antibacterial activity
107
exhibited in the ESM. The combination of the proteins are able to interfere is LPS-LPS
interactions, sensitizing the outer membrane to the lethal affects of heat and possibly
pressure and osmotic stressors as well. Purified proteins from the HEW were also
demonstrated to have an increased bactericidal affect and disrupt the integrity of the outer
membrane to a greater degree than the ESM-bound components.
Significant work is still required to determine the true potential that egg shell
membranes (and its components) may have as a natural antimicrobial or processing aid in
heat-sensitive food and pharmaceutical products. The impact of cations (i.e., Ca2+and
Mg2+) and other food components (i.e., proteins, carbohydrates and lipids) present in the
matrix of food and pharmaceutical products on ESM-mediated antimicrobial activity is
yet to be fully explored. The economic feasibility as well as an understanding of how
physical barriers and exposure conditions (temperature and time parameters) affect the
efficacy of ESM must also be explored. However, if successful, the application of ESM
or its components may lead to the reduction of thermal process requirements for foods
(lower process temperatures and times) and yield products with an extended shelf-life and
reduced populations of microorganisms. Reduced thermal processing requirements may
also result in food products that have higher nutrient levels, have improved functionality
and potentially lower processing costs.
108
References:
Aguilera, O., Quiros, L. M., and J. F. Fierro. 2003. Transferrins selectively cause ion
efflux through bacterial and artificial membranes. FEBS. 548:5-10.
Ahlborn, G. J., Clare, D., Sheldon, B. W., and R. Kelly. 2005. Identification of eggshell
membrane proteins and purification of ovotransferrin and -N-acetylglucosaminidase
from hen egg white. J. Protein Chem. (in review).
Ahlborn, G. J., and B. W. Sheldon. 2005. Enzymatic and biological activity in egg shell
membranes as influenced by layer and storage variables. J. Poult. Sci. (in review).
Anonymous. 2003. CDC (Centers for Disease Control and Prevention). Public Health
Laboratory Information System surveillance data, Salmonella annual summaries at
http://www.cdc.gov/ncidod/dbmd/phlisdata/salmonella.htm
Branen, J. K., and P. M. Davidson. 2004. Enhancement of nisin, lysozyme, and
monolaurin antimicrobial activities by ethylenediaminetetraacetic acid and lactoferrin. Int
J. Food Microbiol. 90(1):63-74.
Coonrod, J. D. 1987. Role of surfactant free fatty acids in antimicrobial defenses.
Eur. J. Respir. Dis. Suppl. 153:209-14.
Crogennec, T., Nau, F., Pezennec, S., Piot, M., and B. Gerard. 2001. Two-step
chromatographic procedure for the preparation of hen egg white ovotransferrin. Eur.
Food Res. Technol. 212:296-301.
Crombie, G., Snider R., Faris, B., and C. Franzbrau. 1981. Lysine-derived cross-links in
the egg shell membrane. Biochim. Biophys. Acta. 640:365-367.
Davidson, P. M., and S. Zivanovic. 2003. The use of natural antimicrobials, p22-45. In
Zeuthen, P., and L. Bogh-Sorensen. (eds). Food preservation techniques. CRC Press,
Boca Raton, FL.
Donovan, J. W., and L. U. Hansen. 1971. The -N-acetylglucosaminidase activity of egg
white. 1. Kinetics of the reaction and determination of the factors affecting the stability of
the enzyme in egg white. J. Food Sci. 36:174-177.
Elliott, L. E., and A. W. Brant. 1957 Effect of saline and eggshell membrane on bacterial
growth. Food Research. 22:241-250.
Foegeding, P. M., and S. B. Leasor. 1990. Heat resistance and growth of Listeria
monocytogenes in liquid whole egg. J. Food Prot. 53:9-14.
109
Gautron, J., M. T. Hincke, M. Panheleux, J. M. Garcia-Riuz, T. Boldicke, and Y. Nys.
2001. Ovotransferrin is a matrix protein of the hen eggshell membranes and basal
calcified layer. Connect. Tissue Res. 42:255-67.
Garratt, R. C., R. W. Evans, S. S. Hasnain, P. F. Lindley, and R. Sarra. 1991. X.a.f.s.
studies of chicken dicupric ovotransferrin. Biochem. J. 280:151-155.
Hancock, R. E., and P. G. Wong. 1984. Compounds which increase the permeability of
the Pseudomonas aeruginosa outer membrane. Antimicrob. Agents Chemother. 26:48-52.
Hitchener, B. J., and A. F. Egan. 1977. Outer-membrane damage in sublethally heated
Escherichia coli K-12. Can. J. Microbiol. 23:311-8.
Isaacs, C. F., and H. Thormar. 1991. The role of milk-derived antimicrobial lipids as
antiviral and antibacterial agents. Adv. Exp. Med. Biol. 310:159-65.
Janolino, V. G., and H. E. Swaisgood. 1982. Analysis and optimization of methods using
water-soluble carbodiimide for immobilization of biochemicals to porous glass.
Biotechnol. Bioengr. 24:1069-1080.
Li, Y. Q., and B. S. Yang. 2004. Aluminum ion removal from monoaluminum
ovotransferrin by pyrophosphate. Chinese J. Chem. 22,1153-1157.
Lifshitz, A., Baker, R. C., and H. B. Naylor. 1963. The relative importance of chicken
egg exterior structures in resisting bacterial penetration. J. Food Sci. 28:94-99.
Lush, I. E., and J. Conchie. 1966. Glycosidases in the egg albumen of the hen, the turkey
and the Japanese quail. Biochemica et Biophysica Acta. 130:81-86.
MacNeil, J. H. 2001. Method and apparatus for separating a protein membrane and shell
material in waste egg shell. US Patent # US 6176376.
Masschalck, B., R. Van Houdt, E. G. Van Haver, and C. W. Michaels. 2001. Inactivation
of gram-negative bacteria by lysozyme, denatured lysozyme, and lysozyme-derived
peptides under high hydrostatic pressure. Appl. Environ. Microbiol. 67:339-44.
Mead, P. S., L. Slutsker, V. Dietz, L. F. McCaig, J. S. Bresee, C. Shapiro, P. M. Griffin,
and R. V. Tauxe. 1999. Food-related illness and death in the United States. Emerg. Infect.
Dis. 1999. 5:607-625.
Naidu, A. S. (ed). 2000. Natural food antimicrobial systems. pp. 1-16. CRC Press LLC.,
Boca Raton, FL.
Nikaido, H., and T. Nakae. 1979. The outer membrane of gram-negative bacteria. Adv.
Microb. Physiol. 20:163-250.
110
Nikaido, H., and M. Vaara. 1987. Outer membrane. pp.7-22. In F. C. Neidhardt (ed.).
Escherichia coli and Salmonella Typhimurium, vol. 1. Amer. Soc. for Microbiol.
Washington, DC.
Poland, A. L., and B.W. Sheldon. 2001. Altering the thermal resistance of foodborne
bacterial pathogens with an eggshell membrane waste by-product. J. Food Prot. 64:486-
492.
Prescott, L. M., Harley, J. P., and D. A. Klein. 1999. Prokaryotic cell structure and
function. pp. 37-71. In Microbiology, 4th ed. WCB/McGraw-Hill., Boston, MA.
Rajeshwara, A. N., and V. Prakash. 1994. Structural stability of lipase from wheat germ.
Int. J. Pept. Protein Res. 44:435-40.
Reynolds, E. S. 1963. The use of lead citrate at high pH as an electronopaque stain in
electron microscopy. J. Cell Biol. 17:208–212.
Rossi, P., G. Francesco, A. Boffi, M. Ajello, P. Valenti, E. Chiancone, and G. Antonini.
2002. Ca2+ binding to bovine lactoferrin enhances protein stability and influences the
release of bacterial lipopolysaccharide. Biochem. Cell Biol. 80:41–48.
SAS Institute Inc. 1990. SAS/STAT User’s Guide, version 6, 4th edition. SAS Institute,
Inc. Cary, NC.
Schuman, J. D., Sheldon, B. W., and P. M. Foegeding. 1998. Thermal resistance of
Aeromonas hydrophila in liquid whole egg. J. Food Prot. 60:231-236.
Shugar, D. 1952. The measurement of lysozyme activity and the ultra-violet inactivation
of lysozyme. Biochim. Biophys. Acta 8, 302-9.
Stan-Lotter, H., M. Gupta, and K. E. Sanderson. 1979. The influence of cations on the
permeability of the outer membrane of Salmonella typhimurium and other gram-negative
bacteria. Can. J. Microbiol. 25:475-85.
Suyama, T., H. Nakamura, M. Ishida, and S. Adachi. 1977. Lipids in the exterior
structures of the hen egg. J. Ag Food Chem. 25:799-803.
Takahashi, K., K. Shirai, M. Kitamura, and M Hattori. 1996. Soluble egg shell membrane
protein as a regulating material for collagen matrix reconstruction. Biosci. Biotechnol.
Biochem. 60:1299-1302.
Thormar, H., C. E. Isaacs, H. R. Brown, M. R. Barshatzky, and T. Pessolano. 1987.
Inactivation of enveloped viruses and killing of cells by fatty acids and monoglycerides.
Antimicrob. Agents Chemother. 31:27-31.
111
Tranter, H. S., N. H. C. Sparks, and R. G. Board. 1983. A note on the structure and iron-
binding properties of egg-shell membranes. Brit. Poult. Sci. 24:123-130.
Vaara, M. 1992. Agents that increase the permeability of the outer membrane. Microbiol.
Rev. 56:395-411.
Winn, S. E., and H. R. Ball, Jr. 1975. -N-acetylglucosaminidase activity of the albumen
layers and membranes of the chicken’s egg. Poult. Sci. 54:799-805.
Winn, S. E., and H. R. Ball, Jr. 1996. Personal communication. North Carolina State
University. Raleigh, N. C.
112
Table 1. Enzymatic and biological activities of ESM components
ESM Treatment
Biological (D54-value)
Activity
(minutes)
Lysozyme
Activity
(U/mg)
-NAGase
Activity
(U/mg)
Ovotransferrin
Activity
(µg iron/g ESM)
Control
1.65 c
41.4 ± 3.9 a
12.2 ± 1.6 a
110.2 ± 10.6a
Immobilized Trypsin
5.12 a
36.7 ± 4.2 a
11.3 ± 1.1 a
45.7 ± 8.4 b
Trypsin (suspended)
5.38 a
37.8 ± 2.2 a
9.7 ± 1.8 a
39.7 ± 11.7 b
80°C for 15 min
3.99 b
14.0 ± 4.8 b
0 b
31.7 ± 12.0 bc
100°C for 15 min
4.43 b
4.3 ± 2.3 c
0 b
17.1 ± 4.3 c
Porcine Lipase
1.54 c
42.8 ± 4.5 a
11.8 ± 1.4 a
100.1 ± 9.5 a
Lipase Cocktail
1.18 c
40.6 ± 5.2 a
10.9 ± 1.7 a
104.7 ± 16.5 a
a-c Mean values (n=3) within D-values and protein/enzyme activity values with different letter
superscripts are significantly different (α ≤ 0.05).
113
Table 2. D54°C-values (min) for Salmonella Typhimurium
pre-exposed to purified enzyme fractions (37°C, 30 min).
µM added
Sample Ovotransferrin Lysozyme -NAGase
D-value
(min)
1 4.00 - - 2.38 c
2 2.00 - - 2.31ND
3 0.50 - - 2.68 ND
4 0.20 - -- 2.44 c
5 0.10 - - 3.15 ND
6 - 0.02 - 4.11 d
7 - 0.04 - 3.68 ND
8 - 0.10 - 3.58 d
9 - - 0.01 4.89 f
10 - - 0.02 4.78 f
11 0.20 0.04 - 2.08 b
12 0.20 - 0.01 2.40 c
13 0.02 0.10 - 2.00 b
14 0.02 0.04 0.01 1.47 a
Control (no treatment) 5.35 f
a-f Mean values (n=2) within ion types with different letter superscripts
are significantly different (α ≤ 0.05). ND Indicates that the values were not
analyzed statistically (n=1).
114
Table 3. Extracellular ion supernatant concentrations of Salmonella Typhimurium treated
(37°C, 30 min) with eggshell membranes (ESM), ovotransferrin, or a combination of
ovotransferrin, lysozyme and -N-acetylglucosaminidase (-NAGase).
Ca K Mg Na
Sample ID mg/L mg/L mg/L mg/L
Milli-Q water <0.05f0.09h<0.05f<0.05g
Milli-Q water with S. Typhimurium <0.05f1.30e<0.05f0.26g
Milli-Q water and ESM 2.20d0.78g1.85d2.26f
Milli-Q water, ESM, S. Typhimurium 2.95c1.20e2.68c3.41e
Milli-Q, ovotransferrin <0.05f0.06h<0.05f17.14d
Milli-Q, ovotransferrin, S. Typhimurium <0.05f0.97f<0.05f16.34d
Milli-Q, ovotransferrin, lysozyme,
-NAGase <0.05f<0.05h<0.05f19.16c
Milli-Q, ovotransferrin, lysozyme,
-NAGase, S. Typhimurium 0.45e2.15d0.46e43.01a
peptone <0.05f2.63c<0.05f20.61b
peptone, S. Typhimurium 0.16f2.69bc 0.11f18.90c
peptone, ESM 5.27b2.80b4.00b20.28b
peptone, ESM, S. Typhimurium 7.79a3.29a6.41a20.69b
a-g Mean values (n=2) within ion types with different letter superscripts are significantly different (α
≤ 0.05).
115
4.19
f
3.14
e
2.35
d
1.36
c
0.69
b
5.34
a
5.02
af
0
1
2
3
4
5
6
control 1 to 10 1 to 20 1 to 30 1 to 40 1 to 50 1 to 60
Ratio ESM (g) to Bacterial suspensions (10
7
-
8
CFU/ml)
Time (min)
Figure 1. Mean D54°C-values (minutes) (n=3) for Salmonella Typhimurium following treatment
(incubation at 37°C for 30 min) with various concentrations of egg shell membranes (grams) to
bacterial suspensions (ml). a-f Mean D-values with different letter superscripts are significantly
different (α ≤ 0.05).
116
5.34
a
5.12
a
1.65
b
4.43
c
1.54
b
5.38
a
3.99
c
1.18
b
0
1
2
3
4
5
6
Control ESM (1:20
wt/vol) Immobil
Tryp ESM Trypsin Tx
ESM Heated
ESM 80°C Heated
ESM 100°C Porc ine
Lipase Tx
ESM
Lipase
Cocktail Tx
ESM
Time (min)
Figure 2. . Mean D54°C-values (minutes) (n=3) for Salmonella Typhimurium following treatment
(incubation at 37°C for 30 min) with treated egg shell membranes (1:20 ratio of g ESM to mls of
bacterial suspension). Treatments are: ESM (no alteration to shell membranes), Immobil Tryp ESM
and Trypsin Tx ESM (shell membranes exposed immobilized trypsin and suspended trypsin), ESM
heated at 80°C for 15 min, ESM heated at 100°C for 15 min, ESM treated with porcine lipase, and
ESM treated with a lipase cocktail. a-f Mean D-values with different letter superscripts are
significantly different (α ≤ 0.05).
117
5.34
a
1.65
b
2.44
c
3.68
d
4.92
a
2.08
c
2.23
c
1.47
b
0
1
2
3
4
5
6
Control ESM 1:20 Ovotrans Lysozyme B-NAGase Ovotrans
Lysozyme Ovotrans
B-NAGase Ovotrans
Lysozyme
B-NAGase
Time (min)
Figure 3. Mean D54°C-values (minutes) (n=3) for purified egg white ovotransferrin (Ovotrans),
lysozyme, ovotransferrin and lysozyme, and a ovotransferrin, lysozyme, and β-N-
acetylglucosaminidase (B-NAGase) treatments compared to no treatment (control) and eggshell
membrane (ESM)-treated Salmonella Typhimurium. a-f Mean D-values with different letter
superscripts are significantly different (α ≤ 0.05).
118
a b
c d
Figure 4. Transmission electron micrographs of treated (37°C, 30 min) Salmonella Typhimurium at
32,000 x magnification (magnification bar represents 300 nm). (a) control cell receiving no treatment
(b) with the addition of ovotransferrin, (c) in the presence of eggshell membranes and (d) with the
addition of ovotransferrin, lysozyme and β-N-acetylglucosaminidase. (picture C will be replaced)
119
a b
c d
Figure 5. Transmission electron micrographs of treated (37°C, 30 min) Salmonella Typhimurium at
6,000 x magnification (magnification bar represents 1 µm). (a) control cell receiving no treatment (b)
with the addition of ovotransferrin, (c) in the presence of eggshell membranes and (d) with the
addition of ovotransferrin, lysozyme and β-N-acetylglucosaminidase.
120
B
A
1 2 3 4 5 6 7 8 9 10
Figure 6. SDS-PAGE of Salmonella Typhimurium cell supernatant after treatment with ESM or
purified enzyme components. Lane 1- molecular weight marker, Lane 2….. Bands A and B
indicating cellular proteins lost from Typhimurium cells.
(CAN”T see well- need to re-run and will explain gel in more
detail after running.)
121
MANUSCRIPT 3
[to be submitted to the Journal of Poultry Science]
122
Enzymatic and biological activity in egg shell membranes as
influenced by layer and storage variables
Gene Ahlborn1 and Brian W. Sheldon2
Department of Food Science1 and Poultry Science2
North Carolina State University
Raleigh, NC 27695
123
Abstract
Eggshell membrane-bound components are capable of reducing the heat
resistance and inhibiting the growth of selected foodborne bacterial pathogens suspended
in 0.1% peptone water. Lysozyme and -N-acetylglucosaminidase (-NAGase) are two
of the membrane components identified as contributing to the observed antimicrobial
activity. The enzymatic and biological (reduced heat resistance as measured by decreases
in D-values) activities of the eggshell membrane (ESM) enzymes lysozyme and -
NAGase were examined. ESM enzyme activities were compared between White
Leghorn (WL) and Rhode Island Red (RIR) layers as a function of age (25-27, 78-80
weeks). Twenty-four week-old WL hens produced ESM with 28% higher lysozyme
activity than RIR layers. Moreover, older WL layers produced ESM with 17% less
lysozyme activity than ESM from RIR layers. Similarly, -NAGase enzyme activities
differed by hen age within breeds with younger hens yielding ESM with 14-16% more
enzyme activity. D54°C-values (range of 1.9 – 2.0 minutes) of White Leghorn ESM-
treated Salmonella enterica serovar Typhimurium test strain did not differ as a function of
bird age (33, 50, 81 weeks). Moreover, there was a significant inverse correlation
between lysozyme and -NAGase activities and D54C-values. ESM Lysozyme and -
NAGase activities varied somewhat over a 6-month storage study following treatment
with one of five different processing methods [i.e., storage at 4oC, -20oC, or ambient air
storage after freeze drying or air drying (23oC for 73 hours or forced air drying at 50oC
for 36 hours)]. Both air and forced air convection oven drying yielded significant
reductions in -NAGase ESM activity (18% and 31%) and lysozyme activity (16% and
31%) after the initial 24 hours and then remained fairly stable during the extended storage
124
period (1-6 months). Freeze-dried samples retained the most enzymatic activity (95% of
the original activity) throughout the 6-month trial while refrigerated ESM loss 20% and
18% of the -NAGase and lysozyme activity, respectively. Frozen ESM lost 22% of the
-NAGase activity yet lysozyme was nearly unaffected (4% loss) after 6-months. ESM
biological activities against S. Typhimurium were not adversely impacted by layer breed
or age indicating that ESM lysozyme and -NAGase activity remain active for extended
storage periods. Twenty-four hours after processing treatments significant reductions in
the biological activity were observed for air and heat dried ESM, with D54°C-values of 4.2
and 4.8 min, respectively. Significant reductions were not observed in the biological
activity of ESM 24 hours and 6 months after processing treatments. These findings
suggest that ESM may have potential commercial value as a processing adjuvant in food
and pharmaceutical product applications.
Keywords: eggshell membrane, β-N-acetylglucosaminidase, lysozyme, stability, activity
125
INTRODUCTION
Numerous bacterial pathogens are transmitted via the food chain. In 1999 the
Centers for Disease Control and Prevention (CDC) reported that an estimated 76 million
persons contract foodborne illnesses each year in the United States (Mead et al., 1999).
Although more than 250 foodborne diseases have been described, identified bacteria
(e.g., Campylobacter, Salmonella, E. coli O157:H7), viruses (e.g., Norwalk-like,
caliciviruses), and parasites (e.g., Giardia, Cyclospora) account for an estimated 14
million illnesses. It is difficult to calculate the exact cost as a result of such illnesses;
however medical costs and lost wages due to foodborne salmonellosis have been
estimated to be more than $1 billion/year (Anonymous, 2003). This information along
with increasing consumer demands for naturally/minimally processed foods (Ray, 1992;
Zeuthen and Sorensen, 2003) has opened the doors for food processors to explore new
options in food safety.
In preliminary studies, Poland and Sheldon (2001) demonstrated that eggshell
membrane-bound components were capable of reducing the heat resistance or inhibiting
the growth of selected Gram-positive and Gram-negative foodborne bacterial pathogens
suspended in 0.1% peptone water [i.e., 83 - 87% reduction in thermal decimal reduction
times (D-values) for Salmonella enterica serovars Typhimurium (D54°C) and Enteritidis
(D54°C), Escherichia coli O157:H7 (D52°C) and up to a 3 log reduction in L.
monocytogenes populations following incubation for 45 min at 37ºC. Ahlborn and
Sheldon (2005) identified that lysozyme, -N-acetylglucosaminidase (-NAGase) and
ovotransferrin were the primary components in ESM responsible for decreasing the heat
resistance of these selected foodborne bacterial pathogens. Methods to extract these
126
enzyme-rich shell membranes are readily available (MacNiel, 2001; Winn and Ball,
1996) and offer egg processors potential economic value as a ‘natural’ processing
adjuvant in food or pharmaceutical products to sensitize bacterial pathogens and spoilage
organisms to heat or other treatments.
If ESM are to be used as a food safety hurdle, it would be beneficial to correlate
measurable enzymatic activity with the antimicrobial (or biological) activity as well as to
determine if there are differences in enzymatic and biological activity as a function of
layer breeds, age and methods of stabilizing (or preparing) ESM for subsequent use. The
purpose of this study was to examine the impact of layer breed, bird age, membrane
stabilization treatment, and storage time on the enzymatic and biological activity of
lysozyme and -NAGase in eggshell membranes.
MATERIALS AND METHODS
Materials
Salmonella enterica serovar Typhimurium ATCC 14028 was obtained from the
American Type Culture Collection (Rockville, MD) and stock cultures were maintained
in sterile double-strength Brain Heart Infusion (BHI) broth plus 20% (vol/vol) glycerol
and stored at -20°C. Lyophilized Micrococcus lysodeikticus ATCC 4689 cells (M-3770)
and 4-nitrophenyl N-acetyl--D-glucosaminide (N-9376) were obtained from Sigma
Chemicals (St. Louis, MO). BHI broth and agar were obtained from Difco Laboratories
(Detroit, MI). All other chemicals and buffers used were reagent grade.
127
Experimental Design
Layer Breed and Age Trials. Fifty eggs per group were collected from White
Leghorn (WL) and Rhode Island Red (RIR) layers at 25-27 weeks and 78-80 weeks of
age. Birds were housed at Carolina Eggs (Nashville, N. C.) and the North Carolina
Department of Agriculture and Consumer Services Peidmont Research Station and feed
standard corn and soybean layer rations under a 16 hour light, 8 hour dark cycle.
Membranes were extracted and prepared as described below with subsequent analysis of
the enzymatic activities of lysozyme and -NAGase.
Membrane Stabilization Trials. Membranes from 100 eggs produced by WL
layers (25-27 weeks) were evaluated. Following extraction (described below), the pooled
ESM were randomly divided into five subgroups and lysozyme and -NAGase activity
determined. Each subgroup was randomly assigned to one of five membrane
stabilization treatments: (1) refrigerated and stored at 4°C, (2) frozen and stored at -20°C,
(3) lyophilized (freeze-dried under identical conditions as described below for
determining solids content), (4) dried at ambient temp (ca 23°C) for 72 h, and (5) dried at
50°C for 36 h in a forced air convection oven. The later three subgroups were maintained
at room temperature following the stabilization treatment. ESM subgroups were then
assayed for lysozyme and -NAGase activity 24 hours following the stabilization
treatments and then at 1, 2, 4 and 6 months under the described storage conditions. D-
54°C-values against S. Typhimurium were determined for treated ESM 24 hours and 6
months after stabilization treatments.
Enzymatic versus Biological Activity. Fifty fresh eggs from WL layers at 33
weeks, 50 weeks and 81 weeks of age were collected and their membranes were extracted
128
using the more commercial-like methods described below. Lysozyme and -NAGase
activity were measured and heat inactivation D54°C-values determined for each group of
membranes.
Methods
Membrane Extraction. One to four day old eggs were washed with a nylon brush
in cool (15-18°C) water containing 100 ppm sodium hypochlorite and rinsed in sterilized
deionized water. Eggs were broken, their contents emptied and residual albumin
removed by rinsing with distilled, deionized water. Membranes were carefully extracted
by hand (ensuring that both the inner and outer membrane remained intact) and up to 30
samples (4.7 mm dia.) excised from each membrane using a die cutter. Excised disks
from identical layer variables were pooled, randomly divided in sub-groups of 10 disks
each (totaling 0.01 g ± 0.002 g), and treated as described in the experimental design.
For comparison of enzymatic versus biological activity, a ‘commercial-like’
process was used to extract ESM using a modified procedure described by Winn and Ball
(1996) and Poland and Sheldon (2001). Briefly, eggs were washed and cleaned of
surface contamination and rinsed as described above. The emptied egg shells were
ground for 10 minutes in a KitchenAid® Food Processor (Model # KFP600WH)
containing 400 ml of sterile water before being poured into a sterile container and left
undisturbed for 5 minutes to allow shell fragments to settle. The top layer of the aqueous
suspension containing the membrane fragments were decanted into a buchner funnel
containing Whatman #1 filter paper and vacuum-dried for 10 minutes. An additional 400
ml of sterile water was added to the shell fragments to remove any residual membrane
fragments remaining from the first process. Filter cakes (from the same egg variables) of
129
the compacted membrane fragments were removed from the filter paper, pooled together
and stored in a sterile petri dish wrapped in aluminum foil and refrigerated until used
(less than 24 hours).
Solids Content. Solids content was determined by weighing extracted membranes
before and after freeze-drying. Briefly, ESM were submerged in liquid nitrogen for 10
minutes. To prevent the ESM from thawing, membranes were quickly removed from the
liquid nitrogen and placed on the pre-cooled plate of an LabConco Series 5 Freeze Dryer
(LabConco Corporation, Kansas City, MO) and dried for 18 hours at -20°C under 10-2
Torr pressure.
Protein Content. The Kjeldahl Method was used to determine total protein
content required in the enzyme assay equations. Percent protein was obtained using the
conversion factor 6.25 from the average of three determinations from each sample.
Lysozyme Assay
Lysozyme activity was determined using an adaptation of the assay described by
Shugar (1952) as measured by the change in optical density (450 nm) following exposure
of Micrococcus lysodeikticus to the ESM. A cell suspension was produced by adding
0.015% (w/v) lyophilized Micrococcus lysodeikticus cells (Sigma, # M-3770) to 66 mM
potassium phosphate buffer (pH 6.24) at 25°C. The resulting suspension absorbance
(450nm) was between 0.6 and 0.7. ESM samples [each sample comprised of 10 disks
(4.7 mm dia, totaling 0.01 g ± 0.002 g) or 0.01 g ± 0.001 g of membrane fragments] from
the treated groups were added to 5.0 ml of the M. lysodeikticus suspension and mixed by
inversion. After two minutes [designated as the optimal time to observe enzyme activity
as indicated by a linear decrease in OD (unpublished data)], the disks were removed, the
130
decrease in absorbance recorded, and enzyme activity calculated according to Equation 1.
Twenty samples (n = 20) were evaluated for each of the five treatment groups (three
treatment groups for biological vs enzymatic trial).
Equation (1) units/mg solid = A450 / time (in minutes)__
0.001 x mg/solid/reaction mix
where 0.001 = change in absorbance at A450nm as per the Unit definition
β-NAGase Assay
The release of p-nitrophenol from 4-nitrophenyl N-acetyl--D-glucosaminide was
followed using a modified procedure of Lush and Conchie (1966), Donovan and Hansen
(1971), and Winn and Ball (1975). The incubation mixture containing either 10 ESM
disks (4.7 mm dia. totaling 0.01 g ± 0.002 g) or 0.01 g (± 0.001 g) of membrane
fragments, 0.9 ml substrate (0.76 µmole 4-nitrophenyl N-acetyl--D-glucosaminide in 0.1
M pH 3.0 citrate buffer) and 0.6 ml distilled, deionized water was incubated at 37ºC for
25 min. The reaction was stopped by adding 2.0 ml of 0.2 M Na2CO3 to the incubation
mixture. The mixture was then briefly vortexed, the ESM disks removed, and the
absorbency (at 420nm) of the solution read in a narrow path absorption cell (1-cm light
path) using a Shimadzu UV 160U UV-Visible Recording Spectrophotometer (Shimadzu
Corp., Kyoto, Japan) blanked against the reaction solution minus the ESM disks.
Enzyme activity was calculated according to Equation 2. Sample sizes were identical to
the lysozyme assay.
Equation (2) = Units/mg = A400 / total volume___________
minutes * 10.8 * mg protein
where 10.8 = extinction coefficient of the substrate at the given absorbance
131
Bioassay (D-value determination)
Decimal reduction times (D-values) were determined using the combined
methods of Poland and Sheldon (2001) and the immersed sealed capillary tube procedure
described by Schuman and colleagues (1998). Salmonella enterica serovar Typhimurium
ATCC 14028 (ST) was selected on the grounds that in preliminary trials cells from ST
showed the greatest sensitivity to ESM (Poland and Sheldon, 2001; Ahlborn and Sheldon,
2005). Mid-log phase cell populations of ST were produced by culturing cells for 24
hours in 10 ml BHI broth at 37ºC in a rotating water bath shaker (Model G76D, New
Brunswick Scientific Co., Inc., Edison, NJ), followed by transfer of 100 µl of the cell
suspension into 10 ml of BHI broth and incubation at 37ºC for 2 hours. Cells were
harvested by centrifugation (15 min at 4ºC at 8,000 x g), washed in sterile 0.1% peptone
water (PW) and suspended in 10 ml of 0.1% PW (ca 108 CFU/ml). Zero or one gram of
the pooled ESM fragment extract was added to 20 ml of the peptone bacterial suspension
and incubated with agitation at 37ºC for 30 minutes.
Following incubation, ESM were removed from the peptone bacterial suspension
[in a Stomacher bag containing a nylon filter (Stomafilter P-type, Gunze Sangyo, Inc.,
Tokyo Japan)] and the inoculated PW (0.05 ml) was dispensed into individual glass
capillary tubes (0.8 to 1.1 mm i.d. by 90 mm long; no. 9530-4, Fisher Scientific,
Pittsburg, PA) using a syringe fitted with a 100-mm blunt needle (head space of
approximately 4mm). Filled capillary tubes were then heat-sealed using a propane torch
and kept on ice until heated (less than 30 minutes). Tubes were placed upright in a mesh-
screen-covered test tube rack and rapidly submerged into a preheated (54ºC) circulating
water bath (Model W19, Haake, Inc., Karlsruhe, Germany) with a temperature control
132
module accurate to ±0.05ºC (Model DC1, Haake, Inc., Karlsruhe, Germany). At six to
eight evenly spaced intervals, duplicate tubes were removed from the water bath and
immersed in an ice-water slurry for 5-10 minutes. Capillary tubes were immersed in
sodium hypochlorite (500 ppm, pH 6.5) for 5 seconds and then rinsed three times in
sterile, distilled deionized water. Tubes were aseptically transferred to individual test
tubes containing 5 ml of sterile PW and finely crushed using a sterile glass rod. This
initial 10-2 dilution was then agitated on a vortex mixer, serially diluted in PW and spiral
plated on BHI agar using the Spiral Biotech Autoplater 4000 (Spiral Biotech Inc,
Norwood, MA). Plates were incubated at 37ºC for 18-24 hours and enumerated using a
Microbiology International ProtoCOL automatic colony counting system (Model 60000,
Synoptics Ltd., UK).
Triplicate thermal inactivation trials were conducted and survivor curves (log
viable S. Typhimurium/ml versus heating time) were plotted for each trial. Best-fit linear
regression lines were drawn through each data set and D-values were calculated as the
negative reciprocal of the survivor slope obtained by regression analysis. D-values
represent the average of the three thermal inactivation trials.
Statistical Analysis
Statistical analysis of enzyme activity studies were determined using the General
Linear Model (GLM) and Least Squares Means (LSM) with P 0.05. Statistical
analysis of the D-values was determined by analysis of variance (ANOVA, P 0.05) and
the means separated by comparison of each mean pair using the student’s t-test (LSD, P
0.05). The residual replicate by treatment mean square was used for testing the main
effects (treatment, replicates) (SAS Institute, Inc., 1990).
133
RESULTS AND DISCUSSION
Breed and Age Trial
Figure 1 shows the enzymatic activity of lysozyme and -NAGase as influenced
by layer breed and age. Shell membrane lysozyme activity (43.4 U/mg) was greatest in
the WL layers at 25-27 weeks of age. ESM from 78-80 week old WL layers had
significantly lower lysozyme activities (17.1%) than ESM from the younger birds.
Contrary to the differences observed in WL layers, no difference in ESM lysozyme
activity was detected between young and old RIR layers. ESM lysozyme activity from
25-27 week old WL layers was 28% greater than the RIR counterparts, however no
significant breed difference was observed for membranes extracted from the 78-80 week
old layers. -N-acetylglucosaminidase activity was highest in the ESM extracted from
25-27 week old WL and RIR layers (13.2 and 12.6 U/mg respectively). Significant
reductions in -NAGase activity within breeds were observed in ESM extracted from the
older birds (14.4% in WL layers and 15.9% in RIR layers).
Comparison of Enzymatic and Biological Activity
The time and resources required to determine biological activity (through heat
inactivation trials) can be costly and time consuming. Thus, the relationship between
biological and enzymatic activities was determined for lysozyme and -NAGase. Figure
2 depicts the enzymatic and biological (D54°C-values) activity of lysozyme and -NAGase
detected in ESM from WL layers at three different ages (33, 50, and 81 weeks.)
Lysozyme and -NAGase activity was greatest in the ESM from 33 week old
layers. A significant reduction in both ESM enzyme activities was found in the 50 week
old layers. At 81 weeks, a slight numerical increase in both lysozyme and -NAGase
134
activity was observed in the ESM, although not significantly different than membranes
from 50 week old layers. This reduction in enzyme activity may correspond to the hen’s
laying cycle. Egg production begins at approximately 19 weeks of age, reaches a
maximum between 24-28 weeks of age and then gradually declines as the layers reach the
forced molting period (through feed restriction or reduced caloric intake) where egg
production declines significantly. Following the forced molting period (around 60-64
weeks of age), egg production increases again, reaching a peak output at approximately
72-78 weeks of age and thereafter followed by a decline in egg production. It is well
established that hormone levels such as estrogen (Beck and Hansen, 2004), corticosterone
and thyroid hormones (Davis et al., 2000) change throughout the egg-laying cycle. We
hypothesize that perhaps as egg production decreases, other hormonal and physiological
changes occur in the hen such as a reduction in lysozyme and -NAGase levels in the
oviduct, albumen and shell membrane. After molting, an increase in the expression of
these enzymes should coincide with the increase in egg production after restoration of
normal hormonal levels.
In comparing the enzymatic and biological activities, a significant inverse
correlation was observed for lysozyme (r = 0.998) and -NAGase (r = 0.992). Shell
membranes from 33 week old layers (having the highest lysozyme and -NAGase
activity) produced the lowest D-value (1.9 min). However, no significant differences in
D-values were observed across hen age groups. After molting, the 81 week old layers
produced similar lysozyme and -NAGase ESM activities as the 33 (lysozyme) and 50
week old layers (lysozyme and -NAGase). Layer breeds within age groups did not
135
influence D-values (unpublished data). Thus, age and breed apparently do not adversely
affect the biological activity of ESM.
Membrane Stability Trial
After extraction, representative membrane samples were taken and enzymatic
activities determined from pooled samples prior to subjecting subsamples to one of five
“stabilization treatments”. No significant difference was observed in the samples prior to
the processing treatments with lysozyme and -NAGase activity averaging 42.4 (± 3.3)
and 13.5 (± 0.75) units/mg respectively. Figure 3 depicts the activity of -NAGase
subjected to the five treatments over a six month period. Twenty-hours following
treatments, no significant change in enzyme activity was observed in the refrigerated,
frozen and freeze-dried samples. In contrast, air and heat dried samples lost 18% and
31% of their respective activity. The three drying methods (freeze, air and heat dried)
produced fairly consistent and stable enzyme activity over the next six months. However
the -NAGase activity of refrigerated and frozen ESM gradually decreased over time,
with samples loosing approximately 20 and 22% of their enzyme activity, respectively, at
the end of six months.
Similar trends were observed with lysozyme activity as illustrated in Figure 4.
Following treatment, the greatest loss of activity after 24 hours was detected in the air
and heat dried samples (16% and 31% respectively), after which no significant further
loss was observed in the succeeding six months. After 6 months, lysozyme activity had
declined by 18% in refrigerated samples although there were no significant decreases in
frozen membranes (4%) and freeze-dried membranes (< 1%) throughout the extended
storage.
136
With the exception of the initial 24 hour period for air and heat dried ESM,
enzyme activity remained stable over the six month storage period. Lysozyme activities
were also generally more variable than -NAGase (larger standard deviations). These
larger variances may be explained by data presented by Hincke and colleagues (2000)
who observed using a colloidal-gold immunocytochemical localization detection
methodology that lysozyme was distributed heterogeneously throughout the ESM
membrane. The apparent random distribution of lysozyme could explain the larger
deviations observed within treatment subgroups and samples despite a representative
sampling and pooling protocol. This hypothesis was supported by preliminary trials in
which we examined the lysozyme and -NAGase activity of 20 (4.7 mm dia., 0.4-0.5 µg
each) excised disks taken from various sections of an egg. Lysozyme activity varied by
as much as 270% (20.4 – 55.1 units/mg) while -NAGase activity only varied by as much
as 64% (9.1 – 14.2 units/mg) (unpublished data).
Figure 5 presents the D54°C-values (min) for S. Typhimurium pre-exposed (37°C,
30 min) to ESM following the membrane stabilization treatments. No difference was
observed in D-values from bacteria exposed to fresh, frozen and freeze-dried ESM
samples taken 24 hours after their respective treatments. There was a significant loss of
biological activity in membranes that were air dried, with D-values being only slightly
lower than the control treatment (4.2 compared to 5.3 min respectively). Heat ‘stabilized’
membranes also lost significant biological activity (D-value = 4.8 min) and were not
different than air dried membranes or the control (no ESM treatment) No significant
differences in D-values within processing treatments were detected between the 24-hour
137
and 6-month samples indicating that the components responsible for increasing the heat
sensitivity of S. Typhimurium were stable.
CONCLUSION
Although the focus of this study was directed only at lysozyme and -NAGase,
the properties and activity of ESM ovotransferrin, the other primary component presumed
to contribute to the antibacterial activity of the ESM, have not yet been fully evaluated.
Preliminary data indicates that the chelating (iron-binding) properties of membrane
bound ovotransferrin follow similar trends observed with lysozyme and -NAGase
(unpublished data). Further evaluation is required to make a complete assessment.
The application of the egg shell membrane (and its components) as a natural
antimicrobial may lead to their use as processing aids in heat-sensitive food and
pharmaceutical products resulting in a reduction of thermal process requirements (lower
process temperatures and times) yet still attaining a product with extended shelf-life and
reduced levels of microorganisms. Reduced thermal processing requirements may also
result in food products that have higher nutrient levels, have improved functionality and
potentially lower processing costs. Consumers will perhaps perceive these “natural”
ingredients as more acceptable while egg processors will gain a new value-added product
having potentially significant market value.
If food and pharmaceutical processors are able to utilize ESM as a processing
adjuvant, the economic feasibility of the ESM extraction and stabilization methods will
need to be thoroughly evaluated. Although freeze-drying maintained the greatest activity
over time, any benefit captured by using freeze-dried ESM having higher enzymatic and
138
biological activity may be offset by the expensive freeze-drying process. Thus, other
stabilization methods may be more practical. More investigations are needed to evaluate
the effectiveness of ESM in actual food or pharmaceutical applications to realize its real
market potential. The impact of food components (i.e., proteins, fats, carbohydrates),
physical barriers and exposure temperatures and time on ESM biological activity are a
few challenges that must be explored.
139
References:
Ahlborn, G. J., and B. W. Sheldon. 2005. Identifying the components in the eggshell
membrane responsible for reducing the heat resistance of bacterial pathogens.
(J. Food Prot. in review)
Anonymous. 2003. Centers for Disease Control and Prevention Subject: Public Health
Laboratory Information System surveillance data, Salmonella annual summaries.
http://www.cdc.gov/ncidod/dbmd/phlisdata/salmonella.htm
Beck, M. M., and K. Hansen. 2004. Role of estrogen in avian osteoporosis. Poult. Sci.
83:200-206.
Davis, G. S., K. E. Anderson, and A. S. Carroll. 2000. The effects of long-term caging
and molt of single comb White Leghorn hens on herterophil to lymphocyte ratios,
corticosterone and thyroid hormones. Poultry Sci. 79:514-518.
Donovan, J. W., and L. U. Hansen. 1971. The -N-acetylglucosaminidase activity of egg
white. 1. Kinetics of the reaction and determination of the factors affecting the stability of
the enzyme in egg white. J. Food Sci. 36:174-177.
Hincke, M. T., J. Gautron, M. Panheleux, J. Garcia-Ruiz, M. D. McKee, and Y. Nys.
2000. Identification and localization of lysozyme as a component of eggshell membranes
and eggshell matrix. Matrix Biology. 19:443-453.
Lush, I. E., and J. Conchie. 1966. Glycosidases in the egg albumen of the hen, the turkey
and the Japanese quail. Biochemica et Biophysica Acta. 130:81-86.
MacNeil, J. H. 2001. Method and apparatus for separating a protein membrane and shell
material in waste egg shell. US Patent # US 6176376.
Mead, P. S., L. Slutsker, V. Dietz, L. F. McCaig, J. S. Bresee, C. Shapiro, P. M. Griffin,
and R. V. Tauxe. 1999. Food-related illness and death in the United States. Emerg.
Infect. Dis. 1999. 5:607-625.
Poland, A. L., and B.W. Sheldon. 2001. Altering the thermal resistance of foodborne
bacterial pathogens with an eggshell membrane waste by-product. J. Food Prot. 64:486-
492.
Ray, B. 1992. The need for biopreservation. pp.1-23. In: B. Ray and M. Daeschel (eds.).
Food Biopreservatives of Microbial Origin. CRC Press, Boca Raton, FL.
SAS Institute Inc. 1990. SAS/STAT User’s Guide, version 6, 4th edition. SAS Institute,
Inc. Cary, NC.
140
Shugar, D. 1952. The measurement of lysozyme activity and the ultra-violet inactivation
of lysozyme. Biochemical et Biophysica Acta. 8:302-309.
Schuman, J. D., B. W. Sheldon, P. M. Foegeding. 1998. Thermal resistance of
Aeromonas hydrophila in liquid whole egg. J. Food Prot. 60:231-236.
Winn, S. E., and H. R. Ball, Jr. 1975. -N-acetylglucosaminidase activity of the albumen
layers and membranes of the chicken’s egg. Poult. Sci. 54:799-805.
Winn, S. E., and H. R. Ball, Jr. 1996. Personal communication. North Carolina State
University. Raleigh, N. C. to Poland, A. L. and B.W. Sheldon, 2001. Altering the thermal
resistance of foodborne bacterial pathogens with an eggshell membrane waste by-
product. J. Food Prot. 64:486-492.
Zeuthen, P., and L. Bogh-Sorensen. 2003. Food preservation techniques. CRC Press,
Boca Raton, FL.
141
43.4
A
36.0
B
33.8
B
31.8
B
13.2
a
11.3
b
12.6
ab
10.6
bc
0
5
10
15
20
25
30
35
40
45
50
WL (25) WL (78) RIR (25) RIR (78)
Enzyme activity
Units/mg
Lysozyme β-NAGase
Figure 1. Comparison of the enzymatic activity of lysozyme and β-N-acetylglucosaminidase in
eggshell membranes from White Leghorn (WL) and Rhode Island Red (RIR) layers at 25-27 and 78-
80 weeks of age. ABab Means with different letter superscripts within enzyme types differ
significantly (α ≤ 0.05) (n=3).
142
1.9 2.1 2.0
42.6
A
12.8
a
33.6
B
9.7
b
37.6
AB
10.9
b
0.0
10.0
20.0
30.0
40.0
50.0
Lysozyme (U/mg) β-NAGase (U/mg)
Units/mg or minutes
33 w eeks 45-50 w eeks 81 w eeks
D
54°C
-value (min)
Figure 2. Comparison of the enzymatic activity of lysozyme and β-N-acetylglucosaminidase versus
biological activity [D-values (min) of ESM treated Salmonella Typhimurium (37°C, 30 min) followed
by heat inactivation (54°C)] of eggshell membranes from White Leghorn layers at 33, 50 and 81
weeks of age. ABab Means with different letter superscripts within enzyme types differ significantly
(α ≤ 0.05) (n=3).
143
b
b
a
b
c
50
60
70
80
90
100
0123456
Time (months)
% Enzyme Activit
y
Refriger ated Frozen Freeze-dried Air dried Heat dried
Figure 3. β-N-acetylglucosaminidase activity as influenced by membrane stabilization method and
storage time. (♦) refrigerated and stored at 4°C, (□) frozen and stored at -20°C, (▲) lyophilized
(freeze-dried), (○) dried at ambient temp (ca 23°C) for 72 h, and (*) dried at 50°C for 36 h in a forced
air convection oven. abc Means with different letter superscripts within treatments differ significantly
(α ≤ 0.05) (n=20).
144
bc
ab
a
cd
d
50
60
70
80
90
100
0123456
Time (months)
% Enzyme Activit
y
Refrigerated Froz en Freeze-dried Air dried Heat d ri ed
Figure 4. Lysozyme activity as influenced by membrane stabilization method and storage time.
Treatments are: (♦) refrigerated and stored at 4°C, (□) frozen and stored at -20°C, (▲) lyophilized
(freeze-dried), (○) dried at ambient temp (ca 23°C) for 72 h, and (*) dried at 50°C for 36 h in a forced
air convection oven. abcd Means with different letter superscripts within treatments differ
significantly (α ≤ 0.05) (n=20).
145
1.7 1.7
2.1
1.6
4.2
4.8 5.3
1.7
4.0
4.9
5.4
1.8
0
1
2
3
4
5
6
7
Fresh Frozen Freeze-dried Air Dried Heat Dried Control
ESM stabilization treatment
D54C-value (min)
D-value 24 hours D-value at 6 months
a a a a a a b b bc bc c c
Figure 5. D-values (min) for ESM treated Salmonella Typhimurium (37°C, 30 min) followed by heat
inactivation (54°C) as influenced by membrane stabilization method and storage time. Treatment
definitions are: (Fresh) refrigerated and stored at 4°C, (Frozen) frozen and stored at -20°C, (Freeze-
dried) lyophilized and stored at room temperature, (Air Dried) dried at ambient temp (ca 23°C) for
72 h, and (Heat Dried) dried at 50°C for 36 h in a forced air convection oven. Control is the D-value
of S. Typhimurium without an ESM treatment. abc Means with different letter superscripts within
treatment varibles differ significantly (α ≤ 0.05) (n=3).
