Citation
Factors affecting growth and culturing of Campylobacter jejuni

Material Information

Title:
Factors affecting growth and culturing of Campylobacter jejuni
Creator:
Mello, Indauê Ieda Giriboni de
Place of Publication:
[Gainesville, Fla.]
Publisher:
University of Florida
Publication Date:
Language:
English

Subjects

Subjects / Keywords:
Alluvial islands ( jstor )
Bacteria ( jstor )
Breasts ( jstor )
Campylobacter ( jstor )
Campylobacter jejuni ( jstor )
Food ( jstor )
Isothiocyanates ( jstor )
Microbiology ( jstor )
Poultry ( jstor )
Vapors ( jstor )
ACID, ALLYL, BILE, CAMPYLOBACTER, ELECTRON, ISOTHIOCYANATE, JEJUNI, MICROSCOPY, MUCIN, SHOCK
Animal Sciences thesis, Ph.D ( lcsh )
Dissertations, Academic -- Animal Sciences -- UF ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent) ( marcgt )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Summary:
ABSTRACT: Two studies were conducted to examine the effect of five factors on the growth and culturability of Campylobacter jejuni: exposure to allyl isothiocyanate (AIT), acid shock, prolonged storage, bile juice and gastric mucin. Microbiological analyses revealed the antimicrobial properties of AIT against Campylobacter and psychrotrophic organisms. Chicken breasts were packaged in a Fresh-R-Pak tray and stored in a retail case for 7 days at 4 degrees C with and without vaporized AIT. Exposure to 100, 300, and 600 microliters of AIT restricted the growth of most spoilage organisms (P<0.05) and strongly inhibited the growth of Campylobacter. Muscle color was also affected by exposure to AIT (P<0.05). Exposure to AIT caused an increase in the degree of lightness (L*) and yellowness (b*) of the muscle whereas it reduced the degree of redness (a*). The safety and shelf life of packaged, fresh, chicken breast may be enhanced by storage in an atmosphere containing vapors of AIT. The combined effects of acid shock, bile juice and gastric mucin were investigated to test their impact on the recovery and growth of aged, nutrient-deprived C. jejuni. Campylobacter cells were deprived of nutrients by storing them in phosphate buffer and acid shocked after 3 months of storage at 4 degrees C. There was no individual, synergistic or additive effect with regard to the recovery of C. jejuni with any of the treatments tested.
Summary:
ABSTRACT (cont.): The morphology of C. jejuni was studied by transmission electron microscopy for a period of 29 days. Growth curves were quantified on the basis of plate counts. Unusually shaped cells were observed, possibly intermediate stages between spiral and coccoid forms. It was deduced that during prolonged storage and lack of nutrients in the medium, the process of morphological transformation to coccoid forms involves a loss of helical morphology, shortening and thickening of the cell and the formation of club shaped intermediate forms.
Thesis:
Thesis (Ph. D.)--University of Florida, 2002.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
General Note:
Title from title page of source document.
General Note:
Includes vita.
Statement of Responsibility:
by Indauê Ieda Giriboni de Mello.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Mello, Indauê Ieda Giriboni de. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
12/1/2003
Resource Identifier:
029834840 ( ALEPH )
82570406 ( OCLC )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

FACTORS AFFECTING GROWTH AND CULTURING OF Campylobacter jejuni By INDAU IEDA GIRIBONI DE MELLO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2002

PAGE 2

COPYRIGHT 2002 by Indau Ieda Giriboni de Mello

PAGE 3

I dedicate this dissertation to my mother, Vildes Claudio Giriboni de Camargo Mello, who has always believed in me and has provided me with unconditional love, guidance, encouragement, and financial support throughout my long and never ending education. The journey that culminated with this work may have never begun without my mother.

PAGE 4

ACKNOWLEDGMENTS First and foremost, I would like to express my most sincere and humble gratitude to God for giving me so many opportunities for personal and spiritual advancement, as well as the courage, perseverance, and determination to follow my dreams and goals. A dissertation such as this is the result of many individuals who have spent their time and efforts to assure that this work would be completed. I have many people to thank. I wish to thank my major professor, Dr. Sally Williams, for her assistance and support during my doctoral program. Thanks are also due to my supervisory committee members, Dr. Henry Aldrich, Dr. Gary Butcher, and Dr. Richard Miles, for their advice and suggestions during my doctoral program. Thank must be also granted to Dr. Ramon Littell for his help with the statistical analyses and to Dr. James Lindsay, who unknowingly sparked my interest and love for food microbiology. Finally, I thank the Department of Animal Science at the University of Florida for the financial support they have provided me with. My deepest appreciation goes to my greatest blessing, my fianc, Luis Carlos Nogueira, the most decent, kind, loving and righteous person I have ever known. I am most thankful for his so–much–needed emotional support, iv

PAGE 5

encouragement, companionship, dedication and love during the most difficult times of my doctoral studies and life. I am grateful for his understanding and patience when I was mentally exhausted, for making me feel that everything would be fine, and above all else for reminding me to always have faith in God. Various others are appreciated for their indispensable and crucial help. A hearty and special thank you goes to Jan Hunt and Dr. Don Bark, from the U.S. Food and Drug Administration, for teaching me how to work with Campylobacter and for inspiring my career in food safety. A thank you is also dedicated to Donna Williams for her uncountable assistance with the electron microscope. Furthermore, I could not have hand–poured all the microbiological plates needed for my research without the unselfish aid of Sonia Nang, Jennifer Gollwitzer and Dr. Clay Walker (and their automatic pour–plate machine), from the Oral Biology Department. Thanks are also extended to Anne Benner and Sergei Sennikov for helping me with the microbial and colorimetric analyses of the chicken breasts used in my study. Many other friends had a positive impact on my life and on this dissertation without even knowing. My dear friends Patrick Lannon, Andreas Kavazis, Justin Sobota, Bernhard Bodmann, and Dr. Ronald Schmidt believed in me when they decided to help me create the International Gourmet Association (IGA). The IGA meetings became my safe harbor where I was able to be myself, learn about leadership, patience, tolerance, sagacity, and where I had the opportunity to meet several individuals who became important in some aspects v

PAGE 6

of my life. Many thanks go also to all my friends who attended to the exit seminar of my defense and gave me so much support, cheering me up and keeping me confident. A special thank you is extended to a few very dear friends, Dr. Maria Luiza Braga Fernandes, Dr. Rogrio Lacaz Ruiz, and Gillian Folkes, for always being there for me when I needed something, no matter how far or close I might be. Their friendship will be forever cherished. A gesture of appreciation goes to Dr. Anthony Pescatore and Mike Ford, who have always kept an interest in my academic progress even after I left the University of Kentucky and for taking me under their wings in every scientific meeting we attended. Their warm thoughts and inquiries were always received with happiness. And above all else, I would like to express my most sincere gratitude to a special friend, Luis Maia de Mello Massa, for assisting me through the admission process of my doctoral program and for his encouragement, especially during the first two years of my PhD work. I have been blessed in my life with a remarkably supportive family. This dissertation also belongs to them for encouraging my continuous pursuit of academic excellence and for supporting me in every step along the way. Many thanks go to my beloved sister, Amait Iara Giriboni de Mello, and my dearest brother, Rubner Vilens Giriboni de Mello, for the love they have always given me. My sister has been an inspiration and an example of dedication to professional life; her vehement sense of justice has constantly reminded me that vi

PAGE 7

fairness and moral rightness must prevail, at any cost. And foremost, I thank her for providing me with an enormous demonstration of her love during my doctoral studies by naming her daughter after me. In spite of my brother’s desire to pass unnoticed through my life, I know that hidden inside him, within his big heart, is a genuine wish to see me succeed and thrive; and for that I am most grateful. In all the ups and downs that I have encountered, he has always kept the doors open for me. I admire his search for self–identity and truth. A profound gratefulness, respect and admiration are devoted to my mother, Vildes Claudio Giriboni de Camargo Mello, to whom this work is specially dedicated. And last but not least, I deeply thank my late father, Rubens Camargo Mello, who always taught me the value and importance of education, perseverance, and endless quest for knowledge. Of humble background, he succeeded and worked hard throughout his life to provide for his family even after his death. For this and so many other reasons I am honored to be his daughter. vii

PAGE 8

TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................iv LIST OF TABLES.............................................................................................................xi LIST OF FIGURES.........................................................................................................xiv ABSTRACT...................................................................................................................xvii CHAPTER 1 INTRODUCTION.......................................................................................................1 2 LITERATURE REVIEW..............................................................................................4 General Microbiology of Campylobacter spp...........................................................4 Historical Background..........................................................................................4 Morphological and Biochemical Properties....................................................11 Antibiotic Resistance and Sensitivity...............................................................16 Pathogenesis and Epizootiology.......................................................................18 Virulence and Toxin Production.......................................................................21 Poultry as Reservoir and Vehicle of Campylobacter.............................................29 General Properties of Fresh Poultry.................................................................32 Poultry microorganisms...............................................................................33 Early studies on Campylobacter and poultry..............................................37 Prevention and Control of Campylobacter........................................................45 Factors that Have a Potential Effect on Growth and Survival of Campylobacter............................................................................................................50 Allyl Isothiocyanates..........................................................................................50 Acid–Shock...........................................................................................................59 Importance of Acid–shock in Foodborne Pathogens.....................................62 Methods for Culturing and Analyzing Campylobacter........................................68 Selective Culture Media.....................................................................................68 Culturing Methodology.....................................................................................76 Porcine bile juice in culture media.............................................................78 Porcine gastric mucin in culture media.....................................................81 Microbiological Electron Microscopy...............................................................85 viii

PAGE 9

3 ANTIMICROBIAL PROPERTIES OF ALLYL ISOTHIOCYANATE.................93 Material and Methods.............................................................................................94 Selective Culture Media and Agar....................................................................94 Bacterial Culture..................................................................................................95 Allyl Isothiocyanate............................................................................................96 Assays for Dose–Related Bacteriostatic Activity of Allyl Isothiocyanate on Fresh Poultry........................................................................................................96 Antimicrobial sensitivity test assay............................................................97 Chicken drumstick assay.............................................................................98 Chicken Breast Study........................................................................................100 Treatment of chicken breasts.....................................................................101 Sample storage............................................................................................103 Statistical analyses.......................................................................................104 Sample Analyses...............................................................................................106 Campylobacter jejuni analysis..................................................................107 Spoilage microbial analysis.......................................................................108 pH analysis...................................................................................................108 Colorimetric analysis..................................................................................109 Results and Discussion..........................................................................................110 Microbial Analyses............................................................................................111 Psychrotrophic organism analysis............................................................112 Campylobacter spp. analysis........................................................................128 pH Analysis........................................................................................................131 Colorimetric Analysis.......................................................................................135 L* value.........................................................................................................137 a* value.........................................................................................................151 b* value.........................................................................................................161 Conclusion..............................................................................................................172 4 EFFECT OF ACID–SHOCK ALONE AND IN COMBINATION WITH PORCINE BILE AND MUCIN ON CAMPYLOBACTER JEJUNI STORED AT 4C FOR THREE MONTHS...................................................................................176 Introduction............................................................................................................176 Material and Methods...........................................................................................177 Selective Culture Media and Agar..................................................................177 Bacterial Culture................................................................................................178 Porcine Bile Juice Collection and Preparation..............................................179 Treatment Preparation......................................................................................180 Sample Storage..................................................................................................183 ix

PAGE 10

Microbial Analysis............................................................................................183 Transmission Electron Microscopy Analyses...............................................186 Negative staining........................................................................................186 Specimen preparation for sectioning.......................................................187 Sectioning methodology............................................................................189 Section staining procedure........................................................................193 Acid–Shock Treatment.....................................................................................194 Results and Discussion..........................................................................................195 Microbial Culturability.....................................................................................195 Acid–Shock Treatment.....................................................................................203 Experimental Media..........................................................................................204 Electron Micrographs.......................................................................................207 Conclusion..............................................................................................................230 5 SUMMARY..............................................................................................................234 APPENDIX P–VALUES ASSOCIATED WITH THE EFFECTS OF ALLYL ISOTHIOCYANATE AND DAYS OF STORAGE ON NUMBER OF PHYSCHROTROPHIC ORGANISMS AND COLOR OF CHICKEN BREASTS..................................................................................................................240 REFERENCES................................................................................................................245 BIOGRAPHICAL SKETCH.........................................................................................271 x

PAGE 11

LIST OF TABLES Table page 1 Combinations of AIT concentration and storage day........................................105 2 Sources of variation and degrees of freedom used in the analysis of variance (ANOVA)...........................................................................................................106 3 Analysis of variance for total psychrotrophic counts observed in skinless, boneless chicken breasts stored at 4C and exposed to allyl isothiocyanate vapors (AIT) for 7 days....................................................................................114 4 Treatment means for total psychrotrophic counts in skinless, boneless chicken breasts stored at 4C and exposed to allyl isothiocyanate vapors (AIT) for 7 days..................................................................................................................115 5 P–values associated with linear, quadratic and cubic effects of allyl isothiocyanate vapors (AIT) for total psychrotrophic counts observed in skinless, boneless chicken breasts stored at 4C for 7 days........................122 6 Prediction regression equations for the linear and quadratic curves for total psychrotrophic counts observed in skinless, boneless chicken breasts exposed to allyl isothiocyanate vapors (AIT) for 7 days.............................122 7 P–values associated with linear, quadratic and cubic effects of days of storage on total psychrotrophic counts observed in skinless, boneless chicken breasts exposed to allyl isothiocyanate vapors (AIT)..................................126 8 Prediction regression equations for the linear and quadratic curves for total psychrotrophic counts observed in skinless, boneless chicken breasts stored for 7 days while exposed to allyl isothiocyanate vapors (AIT)......126 9 Analysis of variance for pH values of skinless, boneless chicken breast homogenate stored at 4C and exposed to allyl isothiocyanate vapors (AIT) for 7 days.................................................................................................132 xi

PAGE 12

10 Treatment means for pH values of skinless, boneless chicken breast homogenate stored at 4C and exposed to allyl isothiocyanate vapors (AIT) for 7 days.................................................................................................132 11 Analysis of variance for L* values of skinless, boneless chicken breasts stored at 4C and exposed to allyl isothiocyanate vapors (AIT) for 7 days.........139 12 Treatment means for L* values of skinless, boneless chicken breasts stored at 4C and exposed to allyl isothiocyanate vapors (AIT) for 7 days..............141 13 P–values associated with linear, quadratic and cubic effects of allyl isothiocyanate vapors (AIT) on L* values of skinless, boneless chicken breasts stored at 4C for 7 days.......................................................................147 14 Prediction regression equations for the linear and quadratic curves for L* values of skinless, boneless chicken breasts exposed to allyl isothiocyanate vapors (AIT) for 7 days....................................................................................147 15 P–values associated with linear, quadratic and cubic effects of days of storage on L* values of skinless, boneless chicken breasts exposed to allyl isothiocyanate vapors (AIT)............................................................................149 16 Prediction regression equations for the linear and quadratic curves for L* values of skinless, boneless chicken breasts stored for 7 days while exposed to allyl isothiocyanate vapors (AIT)...............................................149 17 Analysis of variance for a* values of skinless, boneless chicken breasts stored at 4C and exposed to allyl isothiocyanate vapors (AIT) for 7 days.........151 18 Treatment means for a* values of skinless, boneless chicken breasts stored at 4C and exposed to allyl isothiocyanate vapors (AIT) for 7 days..............152 19 P–values associated with linear, quadratic and cubic effects of allyl isothiocyanate vapors (AIT) on a* values of skinless, boneless chicken breasts stored at 4C for 7 days.......................................................................157 20 Prediction regression equations for the linear and quadratic curves for a* values of skinless, boneless chicken breasts exposed to allyl isothiocyanate vapors (AIT) for 7 days....................................................................................157 21 P–values associated with linear, quadratic and cubic effects of days of storage on a* values of skinless, boneless chicken breasts exposed to allyl isothiocyanate vapors (AIT)............................................................................160 xii

PAGE 13

22 Prediction regression equations for the linear and quadratic curves for a* values of skinless, boneless chicken breasts stored for 7 days while exposed to allyl isothiocyanate vapors (AIT)...............................................160 23 Analysis of variance for b* values of skinless, boneless chicken breasts stored at 4C and exposed to allyl isothiocyanate vapors (AIT) for 7 days.........162 24 Treatment means for b* values of skinless, boneless chicken breasts stored at 4C and exposed to allyl isothiocyanate vapors (AIT) for 7 days..............163 25 P–values associated with linear, quadratic and cubic effects of allyl isothiocyanate vapors (AIT) on b* values of skinless, boneless chicken breasts stored at 4C for 7 days.......................................................................168 26 Prediction regression equations for the linear and quadratic curves for b* values of skinless, boneless chicken breasts exposed to allyl isothiocyanate vapors (AIT) for 7 days....................................................................................168 27 P–values associated with linear, quadratic and cubic effects of days of storage on b* values of skinless, boneless chicken breasts exposed to allyl isothiocyanate vapors (AIT)............................................................................170 28 Prediction regression equations for the linear and quadratic curves for b* values of skinless, boneless chicken breasts stored for 7 days while exposed to allyl isothiocyanate vapors (AIT)...............................................170 29 Composition of experimental media....................................................................181 30 P–values associated with the combination of effects of allyl isothiocyanate vapors (AIT) and storage on total psychrotrophic counts of skinless, boneless chicken breasts stored at 4C...........................................................241 31 P–values associated with the combination of effects of allyl isothiocyanate vapors (AIT) and storage on L* values of skinless, boneless chicken breasts stored at 4C.......................................................................................................242 32 P–values associated with the combination of effects of allyl isothiocyanate vapors (AIT) and storage on a* values of skinless, boneless chicken breasts stored at 4C.......................................................................................................243 33 P–values associated with the combination of effects of allyl isothiocyanate vapors (AIT) and storage on b* values of skinless, boneless chicken breasts stored at 4C.......................................................................................................244 xiii

PAGE 14

LIST OF FIGURES Figure page 1 Campylobacteraceae taxonomy from 1991 to 1995 with historical maximum number of species.................................................................................................8 2 Campylobacteraceae taxonomy with current number of species......................10 3 "Preliminary FoodNet Data on the Incidence of Foodborne Illnesses – Selected Sites, United States, 2000" and "Preliminary FoodNet Data on the Incidence of Foodborne Illnesses – Selected Sites, United States, 2001.".....................................................................................................................31 4 Glucosinolates and main products of hydrolysis by the enzyme myrosinase...........................................................................................................53 5 Fresh–R–Pax tray.................................................................................................102 6 Proprietary absorbent gel material.......................................................................102 7 Trays containing three chicken breast halves prior to receiving AIT ............102 8 Sealed tray containing measured amount of AIT and three chicken breast halves..................................................................................................................102 9 Display retail case...................................................................................................103 10 AIT treated chicken halves stored in the retail display case............................103 11 Psychrotrophic counts of halved chicken breasts exposed to allyl isothiocyanate vapors (AIT) and stored for 7 days at 4C..........................117 12 Psychrotrophic counts of halved chicken breasts stored at 4C for 7 days while being exposed to allyl isothiocyanate vapors (AIT)..........................127 13 Mean pH of homogenate from halved chicken breasts stored at 4C for 7 days while being exposed to allyl isothiocyanate vapors (AIT)..........................133 xiv

PAGE 15

14 Mean pH of homogenate from halved chicken breasts exposed to allyl isothiocyanate vapors (AIT) for 7 days at 4C..............................................134 15 Framework of a Color Model System, showing the L*, a*, and b* values in the form of three axes..............................................................................................138 16 Mean L* values of halved chicken breasts stored at 4C for 7 days while being exposed to allyl isothiocyanate vapors (AIT)...............................................143 17 Mean L* values of halved chicken breasts exposed to allyl isothiocyanate vapors (AIT) for 7 days at 4C.........................................................................145 18 Mean a* values of halved chicken breasts stored at 4C for 7 days while being exposed to allyl isothiocyanate vapors (AIT)...............................................153 19 Mean a* values of halved chicken breasts exposed to allyl isothiocyanate vapors (AIT) for 7 days at 4C.........................................................................156 20 Mean b* values of halved chicken breasts stored at 4C for 7 days while being exposed to allyl isothiocyanate vapors (AIT)...............................................164 21 Mean b* values of halved chicken breasts exposed to allyl isothiocyanate vapors (AIT) for 7 days at 4C.........................................................................166 22 Diagram of the experimental design....................................................................184 23 Number of Campylobacter jejuni cells in stationary phosphate buffer stored aerobically at 4 C.............................................................................................196 24 Helically shaped C. jejuni on Day 1......................................................................208 25 Campylobacter jejuni on Day 1 presenting different morphological shapes....208 26 Spiral C. jejuni. Showing flagellum, flagellum insertion point, polar membrane, cell membrane, outer membrane, and periplasmic space......209 27 Coccoid and spiral shaped C. jejuni on Day 22. Showing flagellum...............209 28 Different sizes of negatively stained flagellated C. jejuni..................................210 29 Spiral C. jejuni..........................................................................................................212 30 Polar end of C. jejuni. Showing flagellum, flagellum insertion point, polar membrane, outer membrane, and cell membrane.......................................212 31 Polar membrane of helical C. jejuni......................................................................213 xv

PAGE 16

32 Morphological transformation of helical C. jejuni to coccoid...........................219 33 Duplicated cell membrane of C. jejuni during transition from spiral to coccoid shape...................................................................................................................220 34 Campylobacter jejuni on Day 22 of storage............................................................229 35 Campylobacter jejuni spheroplasts..........................................................................229 xvi

PAGE 17

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy FACTORS AFECTING GROWTH AND CULTURING OF Campylobacter jejuni By Indau Ieda Giriboni de Mello December 2002 Chair: Sally K. Williams Major Department: Animal Sciences Two studies were conducted to examine the effect of five factors on the growth and culturability of Campylobacter jejuni: exposure to allyl isothiocyanate (AIT), acid shock, prolonged storage, bile juice and gastric mucin. Microbiological analyses revealed the antimicrobial properties of AIT against Campylobacter and psychrotrophic organisms. Chicken breasts were packaged in a Fresh–R–Pak tray and stored in a retail case for 7 days at 4 C with and without vaporized AIT. Exposure to 100, 300, and 600 L of AIT restricted the growth of most spoilage organisms (P<0.05) and strongly inhibited the growth of Campylobacter. Muscle color was also affected by exposure to AIT (P<0.05). Exposure to AIT caused an increase in the degree of lightness (L*) and yellowness (b*) of the muscle whereas it reduced the degree of redness (a*). The safety and shelf life of packaged, fresh, chicken breast may be enhanced by xvii

PAGE 18

storage in an atmosphere containing vapors of AIT. The combined effects of acid shock, bile juice and gastric mucin were investigated to test their impact on the recovery and growth of aged, nutrient–deprived C. jejuni. Campylobacter cells were deprived of nutrients by storing them in phosphate buffer and acid–shocked after 3 months of storage at 4 C. There was no individual, synergistic or additive effect with regard to the recovery of C. jejuni with any of the treatments tested. The morphology of C. jejuni was studied by transmission electron microscopy for a period of 29 days. Growth curves were quantified on the basis of plate counts. Unusually shaped cells were observed, possibly intermediate stages between spiral and coccoid forms. It was deduced that during prolonged storage and lack of nutrients in the medium, the process of morphological transformation to coccoid forms involves a loss of helical morphology, shortening and thickening of the cell and the formation of club–shaped intermediate forms. xviii

PAGE 19

CHAPTER 1 INTRODUCTION The concern with pathogens in our food system has recently increased, especially after the large number of bacterial foodborne outbreaks involving meat and ready–to–eat products. Continuous advances worldwide in food manufacturing, production and distribution have enabled the food industry to expand and supply high quality products to consumers on a year–round basis. According to the Centers for Disease Control and Prevention, in 2001 Campylobacter was the second leading cause of gastroenteritis in the United States, being responsible for 13.8 cases of illnesses per 100,000 persons (Centers for Disease Control and Prevention, 2002). Campylobacter is commonly found in the intestinal tracts of humans and animals. It is believed that about 80% of all poultry sold for human consumption in the United States is contaminated with Campylobacter (Centers for Disease Control and Prevention, 2000). The food industry relies largely on a combination of hurdles to achieve and maintain the safety of food products by inhibiting, killing or inactivating contaminating microorganisms in food products. The use of hurdles, usually applied simultaneously or in a sequential way, assures that in specific cases, pathogenic and spoilage organisms are safely killed before reaching the consumer. 1

PAGE 20

2 As the food industry works hard on new interventions to assure safe food, consumers are increasingly demanding safer, cheaper and healthier food, without large amounts of preservatives, sodium and other shelf extending ingredients (Knchel and Gould, 1995). Antimicrobial agents derived from plants have been long investigated and they can be used in food products to meet the consumer’s demands and expectations (Walker et al., 1925; Delaquis and Mazza, 1995; Kyung and Fleming, 1996; Shofran et al., 1998; Ward et al., 1998; Cherry, 1999; Delaquis et al., 1999; Kim et al., 2002). Isothiocyanates, including allyl isothiocyanate (AIT), are natural plant–origin compounds derived from a group of glycosides stored within cell vacuoles of all plants belonging to the family Cruciferae. Allyl isothiocyanate has been widely studied against both spoilage and pathogenic organisms and has proven to inhibit their growth (Walker et al., 1937; Carter et al., 1963; Isshiki et al., 1992; Tsunoda, 1994; Brabban and Edwards, 1995; Delaquis and Mazza, 1995; Shofran et al., 1998; Ward et al., 1998; Cherry, 1999; Delaquis et al., 1999; Kim et al., 2002). To the present date, no data could be found related to the effect of AIT vapors on the shelf life and presence of Campylobacter jejuni on poultry products available to the consumer. Moreover, no data are available on how poultry products are affected by storage when being constantly exposed to the AIT vapors trapped inside a sealed food package. Low storage temperatures (4 C) allow Campylobacter to remain viable in a food product during storage (Hazeleger et al., 1998). Given the importance of

PAGE 21

3 water and poultry as vehicles of C. jejuni, studies examining the survival of C. jejuni stored at 4 C for prolonged periods of time and its capability to withstand the low acidity of the stomach when inside the human’s gastrointestinal system are extreme important. The low pH of the stomach is considered to be the body’s first defense against microorganisms and if the bacteria survive the acidic environment of the stomach, they will have a better change to thrive in the small intestine, multiplying and causing damage. In addition, it is important to investigate the impact of bile salts and mucin on the recovery and growth of acid–shocked Campylobacter. Combined, these studies have the potential for helping to understand how Campylobacter interacts with humans and how to better culture the organism in the laboratory.

PAGE 22

CHAPTER 2 LITERATURE REVIEW General Microbiology of Campylobacter spp. Historical Background The microaerophilic vibrios, today considered to be Campylobacter, were first recognized about 90 years ago. In 1909 two veterinarians, McFadyean and Stockman, discovered a microorganism associated with epizootic abortion in ewes (cited by Franco, 1988). Four years later, these same researchers demonstrated that this “vibrio” microorganism could be observed in infectious abortions in sheep and pregnant cows (Fox, 1982; Franco, 1988). In 1919, Smith isolated a microaerophilic spirillum bacterium from aborted calf tissues which was the same microorganism isolated by McFadyean and Stockman. Because this bacterium had a comma–shaped morphology, Smith and Taylor proposed the name “Vibrio fetus” and the disease became known as vibrionic abortion (Fox, 1982; Butzler, 1984; Franco, 1988, Skirrow, 1977). In 1931, a new “vibrio” that caused dysentery in calves during the winter was discovered and named Vibrio jejuni by Jones et al. (cited by Butzler, 1984; Franco, 1988). Similar microorganisms continued to be described as being the cause of dysentery in swine (V. coli) and humans (V. fetus) up to 1947 and up to 1959 as a pathogen (V. fetus veneralis) in cases of enzootic sterility in cows (Butzler, 1984). 4

PAGE 23

5 The three identified vibrio microorganisms—V. jejuni, V. coli, and V. fetus—were named in association with specific diseases in animals rather than to any observed taxonomic differences among them (Butzler, 1984). The possibility of taxonomic differentiation started when King (1957) showed that catalase–positive microaerophilic vibrios could be distinguished by their ability to grow at different temperatures. One group corresponded closely to the already existing description of V. fetus while the other group was simply called “related vibrios.” She later suggested that her “related vibrios” were identical to the V. jejuni and to the V. coli observed in the dysenteric swine of Doyle (Butzler, 1984). During that time King (1957) was grouping the vibrio bacteria into groups, and the clinical entities of bovine vibrionic abortions started to be better understood. One type of sporadic abortion often occurred among pregnant cows that were members of a fertile herd. The infection that resulted in abortion probably resulted from the establishment of V. fetus in the intestinal tract of the cow. The other type of sporadic abortion occurred in a herd with reduced conception rates, suggesting a case of infectious infertility. Florent demonstrated, in 1959, that this type of abortion was transmitted venereally from symptomless bulls to cows and that infertility was the major consequence of the infection (Butzler, 1984). Thus, he called this vibrio V. fetus var. venerealis. In 1963, the microaerophilic vibrios were assigned by Sebald and Vron to a new genus called Campylobacter (Hbert et al., 1982). Ten years later, Vron and Chatelain (1973) published the first comprehensive taxonomy of Campylobacter

PAGE 24

6 and recognized the microorganisms Vibrio jejuni and Vibrio coli in their classification of the genus. From that point on, V. jejuni and V. coli became Campylobacter jejuni and Campylobacter coli, names that were later accepted by an International Committee of Systematic Bacteriology (Skerman et al., 1980; Skirrow and Benjamin, 1980). Because it was not yet possible to distinguish C. jejuni from C. coli, both bacteria were grouped together as C. fetus subsp. jejuni in the 8th edition of Bergey’s Manual of Determinative Bacteriology (Smibert, 1974). For a long time nobody really knew if C. jejuni and C. coli were different species or merely biotypes or variants within the same species. In 1980, Harvey (1980) showed by the use of two tests that it was possible to distinguish not only C. jejuni from C. coli, but also two biotypes of C. jejuni on the basis of their ability to hydrolyze hippurate. In the 8th edition of Bergey’s Manual of Determinative Bacteriology (Buchanan and Gibbons, 1974), the genera Spirillum and Campylobacter constituted the family Spirillaceae. Grouping of these taxa in one family was based mainly on a number of morphological features. Because the phylogenetic relationship of these organisms were still unknown, the use of the family name Spirillaceae was discontinued and the genus Campylobacter was considered to belong to the Proteobacteria class (Vandamme and Ley, 1991). In 1991, Vandamme and Ley proposed that the genera Campylobacter and Arcobacter should be included in a separate family, named Campylobacteraceae. Their rationale was based on the enormous genotypic and phenotypic differences

PAGE 25

7 observed among these two genera and the other members of the class Proteobacteria. During the time the “vibrio” microaerophilic microorganisms were being discovered, Tunnicliff (1914) isolated a strictly anaerobic vibrio in the sputum of patients suffering from acute bronchitis. Prvot, in 1940, named the Tunnicliff anaerobic vibrio V. sputorum (Butzler, 1984). Later in 1953, Florent (cited by Butzler, 1984) isolated and named Vibrio bubulus, a saprophytic vibrio found in bovine semen and in the vagina of cows. Loesche et al. (1965) compared the characters of both V. sputorum and V. bubulus and suggested that the differences existing between the two were not sufficient to separate them into two species. The two microorganisms were then named V. sputorum subsp. sputorum and V. sputorum subsp. bubulus. A new subspecies of V. sputorum, V. sputorum subsp. mucosalis was later isolated from the lesions of porcine intestinal adenomatosis by Rowland et al. (1973). All these vibrio bacteria were finally renamed Campylobacter when they where compared with the members of the genus Campylobacter. Proper characterization of microorganisms is very important for taxonomists when they are setting up and surveying culture collections (Stonnet et al., 1995). The taxonomy (Figure 1) of the family Campylobacteraceae has been significantly revised since it was proposed in 1991. For this reason,

PAGE 26

8 Campylobacteraceae Campylobacter butzleri cinaedi coli concisus curvus cryaerophilus fennelliae fetus fetus venerealis gracilis helveticus hyoilei hyointestinalis hyointestinalis lawsonii jejuni jejuni doylei mucosalis mustelae nitrofigilis pylori pylori mustelae rectus showae sputorum sputorum bubulus mucosalis upsaliensis Arcobacter Figure 1 Campylobacteraceae taxonomy from 1991 to 1995 with historical maximum number of species. Figure elaborated based on the text published by Euzby (1997)

PAGE 27

9 Euzby (1997) compiled and published a comprehensive list of bacterial names, which includes several new Campylobacter that have been described over the last decade. Some of the new discoveries include C. consisus (Tanner et al., 1981); C. lari (Benjamin et al., 1983); C. nitrofigilis (McClung et al., 1983); C. pylori (Marshall et al., 1984); C. curvus (Tanner et al., 1884, cited by Euzby, 1997); C. hyointestinalis (Gebhart et al., 1985); C. cryaerophila (Neill et al., 1985); C. cinaedi (Totten et al., 1985); C. fennelliae (Totten et al., 1985); C. mustaleae (Fox et al., 1989); C. butzleri (Kiehlbauch et al., 1991); C. rectus (Vandamme et al., 1991); C. upsaliensis (Sandstedt and Ursing, 1991); C. helveticus (Stanley et al., 1992); C. showae (Etoh et al., 1993); C. gracilis (Vandamme et al., 1995); and C. hyoilei (Alderton et al., 1995). Campylobacter lari was originally called C. laridis (named after its occurrence in the intestinal tracts of gulls) by Benjamin et al. (1983) but it was revised to C. lari by Graevenitz (1990) in accordance to the International Code of Nomenclature of Bacteria and C. pyloridis was revised to Campylobacter pylori (Marshall and Goodwin, 1987). Several taxa previously described as Campylobacter have now been reassigned to the genus Arcobacter or Helicobacter. Vandamme et al. (1991) proposed several changes in the taxonomy, which were later accepted by The International Committee on Systematic Bacteriology. After approval, the genus Arcobacter included A. cryaerophilus, A. nitrofigilis, A. butzleri, and A. skirrowii. The genus Helicobacter contained H. pylori, H. mustelae, H. cinaedi and H. fennelliae,

PAGE 28

10 Campylobacteraceae Campylobacter coli concisus curvus fetus fetus venerealis gracilis helveticus hyointestinalis hyointestinalis lawsonii jejuni jejuni doylei mucosalis rectus showae sputorum sputorum bubulus upsaliensis Arcobacter Figure 2 Campylobacteraceae taxonomy with current number of species. Figure elaborated based on the text published by Euzby (1997)

PAGE 29

11 which were originally described as Campylobacter (Ursing et al., 1994). In addition, Campylobacter hyolei was found to be a subjective synonym of Campylobacter coli (Vandamme et al., 1997). As of 1997 (Figure 2), the genus Campylobacter consisted of 14 species and five subspecies (Euzby, 1997). Morphological and Biochemical Properties Campylobacter are curved or occasionally straight rods, with size ranging from 0.2 to 0.9 m wide and 0.5 to 5 m long. The rods may appear as spiral, S—, V—, or comma–shaped forms and may also occur in short or occasionally long chains. The cells may become spherical or coccoid, especially in old cultures. They are non–spore–forming and gram–negative. The cells are highly motile by means of single or occasionally multiple unsheathed flagella at one or both ends (Ursing et al., 1994). According to On et al. (1995), the differences observed in the flagellar arrangement of certain species strains is high, explaining why the number of flagella should not be considered as an important taxonomic criterion for the Campylobacter genus. When working with C. hyointestinalis, On et al. (1995) observed both mono and biflagellated cells (of the same general shape and size) within the same culture of strain. They proposed that genetic changes in Campylobacter might occur as a result of spontaneous mutation, as well as mechanisms such as natural or plasmid–borne transformation, which could explain the considerable phylogenetic diversity observed within the genus. Optimum temperature for growth ranges from 30 to 42C. They are usually microaerophilic; i.e., they grow best in an atmosphere containing 5 to

PAGE 30

12 10% oxygen and have a respiratory type of metabolism. Some species may be aerotolerant. The G+C content of the DNA ranges from 28 to 46 mol%. Campylobacter do not oxidize or ferment carbohydrates; therefore neutral or acidic end–products are not produced, which makes development of differential media difficult. These organisms exhibit a respiratory, chemoorganotrophic metabolism. As for energy sources and carbon incorporation, Campylobacter prefers amino acids and intermediates of the tricarboxylic acid cycle (Ursing et al., 1994; Fox, 1982). Colony morphology of Campylobacter on agar plates varies considerably, according to the basal medium, bacterial strain, level of moisture on the surface of the agar, incubation time and incubation temperature. Thus, colony morphology should not be used as the main distinguishing factor when selecting for Campylobacter. Nonetheless, the most typical morphologies vary from round colonies to irregular ones with smooth edges. Colonies can also be non–distinctive, with morphology varying from a thick translucent white growth to a spreading, film–like transparent growth. On blood agar, some colonies can produce a rainbow sheen (Hunt and Abeyta, 1995). Visible colonies usually appear on the plating media within 24 to 48 hours of incubation with some growth occurring, occasionally, after 72 to 96 hours (Franco, 1988). Campylobacter tend to multiply slower than the usual enteric flora; thus it is very difficult to isolate this organism from fecal specimens without the use of selective techniques (Franco, 1988).

PAGE 31

13 Campylobacter species will survive in some foods (but not multiply) at refrigeration temperature for prolonged periods of time, especially if food is kept airtight. The bacteria will survive for only a few days at room temperature. The total Campylobacter count in foods such as poultry and milk usually decreases by 2 logs at C. However the organisms can often survive beyond the shelf life of refrigerated products (Hunt and Abeyta, 1995). Exposing Campylobacter to experimental stresses such as air, drying, low pH, heating, freezing, and prolonged storage will damage cells and hinder recovery to a greater degree than for most bacterial strains. Campylobacter possess limited ability to produce oxygen–neutralizing enzymes; thus, the simple act of opening food packages and entrance of fresh oxygen into the product weakens the microorganism (Hunt and Abeyta, 1995). Although Arcobacter and Campylobacter share many similar characteristics, key features of the genus Campylobacter can be used to distinguish the two genera. These features include the fact that Arcobacter grows at 15C but not at 42C, its optimal temperature for aerobic growth is 30C, and its G+C content of the DNA ranges from 27 to 30 mol% (Ursing et al., 1994). Other Campylobacter species are not considered relevant in terms of foodborne illness. Nonetheless, differentiation of the different species is important when surveilling gastroenteritis cases. Campylobacter showae is one species of Campylobacter, which differs largely from the main species, C. jejuni. One of the main differences relates to its morphology; C. showae are gram–

PAGE 32

14 negatives, straight (not curved) rods. Their size ranges from 0.5 to 0.8 m wide and 2 to 5 m long with round ends and two to five unipolar unsheathed flagella. They prefer to grow under anaerobic conditions but they can thrive in a microaerophilic atmosphere if fumarate and formate or hydrogen (H 2 ) is present. C. showae are oxidase positive, as well as arylsulfatase positive, and most strains decompose hydrogen peroxidase. They are asacchrolytic and reduce nitrate to nitrite. C. showae do not produce lysine and ornithine decarboxylase, alkaline phosphatase, urease, gamma–glutamyl transferase and do not hydrolyze hippurate; however they produces hydrogen sulfite (H 2 S). The pathogenicity and importance of this species are still mainly unknown. The G+C content of the DNA is high varying from 44 to 46 mol% (Etoh et al., 1993). Another species that differs from C. jejuni is Campylobacter hyointestinalis. C. hyointestinali are gram–negative curved or loosely helical rods with size ranging from 0.2 to 0.5 m wide and 1.2 to 2.5 m long. The cells are highly motile by means of a single or occasionally two unsheathed polar flagella. Optimal growth occurs under microaerophilic conditions with some strains being capable of growing anaerobically. Inclusion of blood in the culture medium enhances culturability but is not essential for growth. After 48 hours of growth on 5% blood agar at 37 C, colonies are mostly circular and convex, and colony size varies from pinpoints up to 2.0 mm in diameter. A greenish color, accompanied by alpha–hemolytic activity, may be apparent in cultures of some strains. All strains grow at 42 C. As with C. jejuni, C. hyointestinalis are oxidase

PAGE 33

15 and catalase positive, and are able to reduce selenite and nitrate. Nalidixic acid can be used as a distinguishing tool because C. hyointestinalis, as opposed to C. jejuni, are nalidixic acid resistant. C. hyointestinalis are associated with enteric disorders, but their true pathogenicity is still unknown (On et al., 1995). Campylobacter hyointestinalis subsp. hyointestinalis has the same characteristics given for the C. hyointestinalis. The G+C content of 34 to 36 mol% is lower than that found in other species of Campylobacter, for example, C. showae. This organism has been previously isolated from intestines of several animals such as pigs, hamsters, cattle, deer and even from human feces (On et al., 1995). The other subspecies of C. hyointestinalis, Campylobacter hyointestinalis subsp. lawsonii, are spirals or slightly curved rods, measuring about 0.2 m in diameter and 1.4 to 2.0 m in length. When observed by wet mount, all cells are motile by means of one or two polar flagella and move in a corkscrew type motion. Colonies incubated for 72 hours on 5% calf blood agar are quite small (ranging in size from pinpoint up to 1 mm in diameter), have a slightly raised round shape morphology and smooth edges (On et al., 1995). Due to the early confusion observed in the taxonomy for Campylobacter, Campylobacter hyoilei was described as Campylobacter coli (Vandamme et al., 1997). Thus, the description of C. hyoilei published by Alderton et al. (1995) can be extrapolated to C. coli. Campylobacter hyoilei are gram–negative, comma–shaped, non–encapsulated, non–spore–forming bacteria. Most strains grow under microaerophilic or anaerobic conditions at both 37 and 42C, and cells are motile

PAGE 34

16 by means of bipolar flagella. Oxidase and catalase are produced, meaning that this species are oxidase and catalase positive, and urease is not produced. Similar to C. jejuni, C. hyoilei is susceptible to nalidixic acid and the G+C content of this species is around 33.5 to 36.5 mol% (Alderton et al., 1995). Antibiotic Resistance and Sensitivity There are basically two types of resistance to antibiotics: intrinsic and acquired. Intrinsic resistance to an antibiotic means that every strain in the species is resistant to a certain compound while acquired resistance is that resulting from a chromosomal mutation or acquisition of a foreign DNA (plasmid or transposon) which carries the resistance to the specific substance. The mechanism of how intrinsic resistance occurs is not yet completely understood. The intrinsic resistance, most likely, involves blocking certain antibiotics from penetrating the cell or pumping antibiotics out of the cell to the drug faster than it can enter. All C. jejuni and C. coli isolates are intrinsically resistant to a number of antibiotics, including bacitracin, novobiocin, rifampin, streptogramin B, trimethoprim, vancomycin, and usually cephalothin. Campylobacter jejuni and C. coli are also susceptible to nalidixic acid, cirpofloxacin, norfloxacin, and ofloxacin (Taylor and Courvalin, 1988). The whole issue about susceptibility or resistance to specific antibiotics is extreme important when cure of infected hosts is needed and when subtherapeutic levels of antibiotics belonging to certain families are incorporated into animal feed to function as growth promoters.

PAGE 35

17 According to Taylor and Courvalin (1988), the Campylobacter genus, with its gram–negative cell wall but very low G+C content, has apparently been able to acquire resistance determinants from both gram–positive and gram–negative organisms, although the former seem to be the more common source. Campylobacter and Enterococcus occupy a common niche (the human and animal gastrointestinal tracts) and DNA exchange between these two species occurs within this environment is very likely to occur. The transfer of genetic material is believed to occur because direct transfer of plasmid DNA from gram–positive cocci (Enterococcus species) to gram–negative bacteria (E. coli) has been previously accomplished in vitro. Acquired antibiotic resistance in Campylobacter species includes, but are not limited to, tetracyline, minocycline, kanamycin, chloramphenicol, streptomycin, spectinomycin, erythromycin, ampicillin and nalidixic acid. Erythromycin is accompanied by cross–resistance to spiramycin, tylosin, and clindamycin (Blaser et al., 1982; Taylor and Courvalin, 1988; Shane, 1997). The biochemical mechanisms involved in resistance to macrolides is extremely important, especially in the case of erythromycin, since this is the drug of choice for treatment of serious Campylobacter infections (Blaser et al., 1982; Taylor and Courvalin, 1988). Furazolidone is another drug that has been shown to be effective against Campylobacter species (Shane, 1997).

PAGE 36

18 Pathogenesis and Epizootiology The reservoir for Campylobacter species in nature is enormous since this organism has been revealed to be quite ubiquitous. Campylobacter is a pathogenic or commensal organism in cattle, sheep, swine, fowls, dogs, cats and monkeys, and it has been isolated from both fresh and sea water. Investigation of outbreaks of Campylobacter has been linked to the consumption of contaminated water, poultry, and raw milk from infected cows (Blaser, 1980). Two types of diarrhea are observed with campylobacteriosis: an inflammatory diarrhea, with fever and slimy, often bloody stools containing leukocytes; and noninflammatory diarrhea, with watery stools and the absence of leukocytes and blood (Wassenaar, 1997). Although campylobacteriosis does not usually lead to death and it is self–limiting, it has been estimated that 500 people with Campylobacter infections die each year, possibly due to other complications associated with the disease such as Guillain–Barr syndrome and Miller Fisher Syndrome (Dreesen, 1997). The pathogenic mechanisms of how campylobacteriosis occur are not totally understood, but important virulence factors include motility, chemotaxis, ability to translocate and the production of toxins (Stern et al., 1992a; Dreesen, 1997). Apparently, the virulence factors that are involved in the infection greatly influence the symptoms of the disease. The different clinical signs and epidemiology seen in the expression of pathogenicity of Campylobacter species

PAGE 37

19 suggests that pathogenesis results from a combination of different bacterial properties and host defenses. Motility, which is achieved by means of a single flagellum at one or both ends of the bacteria, has an important role in virulence because it is required for the bacteria to reach the attachment sites and penetrate into the intestinal cells. Loss of motility is detrimental to Campylobacter because the bacterium looses its ability to colonize the gastrointestinal tract and cause infection. The importance of motility as a virulence factor is best demonstrated by true isogenic nonflagellated mutants, which are unable to colonize the intestine of experimental animals (Guerry et al., 1992). Campylobacter jejuni contains two flagellin genes, flaA and flaB; the wild–type bacteria express flaA only, but flaB can be expressed under certain conditions. According to Wassenaar et al. (1993), the presence of flagella composed of flagellin A, rather than motility is essential for optimal bacterial colonization and that inactivation of the gene that encodes flagellin B (flaB) had no effect on bacterial motility. The flagellum of Campylobacter has a much broader role than to solely provide motility to the bacterium. It is believed that the flagella may also play a role in the internalization and translocation of C. jejuni. The flagellin has been proposed as an adhesin in the binding of Campylobacter to culture cells. According to Grant et al. (1993), C. jejuni is able to become internalized within human intestinal epithelial cells and traverse monolayers of polarized human colonic carcinoma (Caco–2) cells. This translocation across the epithelial cell

PAGE 38

20 barrier may reflect a pathogenic mechanism by which this organism can gain access to submucosal tissue, leading to the tissue damage and inflammation. Campylobacteriosis is a self–limiting gastroenteritis and recovery is completed in approximately 7 days even without medical interference. The problem with campylobacteriosis however, is the possibility that exists that individuals inflicted by the disease will suffer from neurological sequelae months or years afterwards (Lindsay, 1997; Ang et al., 2001). Two neuropathies are associated with C. jejuni infections: Guillain–Barr syndrome (GBS) and Miller Fisher syndrome (MFS). Both of those syndromes are characterized by being acute or sub–acute, acquired, immune–mediated neuropathies frequently proceeded by an infection illness, more commonly, campylobacteriosis. The GBS is characterized by alexia, motor paralysis, acellular increase in the total protein content in the cerebrospinal fluid and inflammatory demyelinating polyradiculoneuropathy (Lindsay, 1997). All GBS cases have propensity to the nerve roots, cause mononuclear infiltration of peripheral nerves, which eventually leads to primary axonal degeneration (demyelination) (Lindsay, 1997). Ang et al. (2001) demonstrated the role of molecular mimicry in the development of GBS and MFS in patients with post Campylobacter infections neuropathies. Molecular mimicry in believed to be the cause of GBS and C. jejuni because a few peripheral nerves of the human neurological system share similar molecules with antigens on the surface of C. jejuni cells (Lindsay, 1997).

PAGE 39

21 Campylobacter jejuni is a gram–negative bacterium and as so, it contains a lipopolysaccharide structure (LPS) attached to its outer membrane. The core oligosaccharides of its LPS contain ganglioside–like structures, which are similar to human’s GM 1 and GQ 1b gangliosides (Ang et al., 2001). The LPS structure is very antigenic, i.e., it strongly stimulates the immune system. Upon exposure to C. jejuni, the immune system produces antibodies against the LPS structure as an attempt to fight the infection. Due to the similarity of the core oligosaccharides of the LPS and the gangliosides, after the infection, the anti–GM 1 and the anti–GQ 1b antibodies attack the GM 1 and GQ 1b gangliosides on the neuromuscular junction contributing to the appearance of neurological symptoms (Lindsay, 1997; Ang et al., 2001). Anti–GM 1 antibodies are associated with motor neuropathy and anti–GQ 1b antibodies are associated with oculomotor symptoms (Ang et al., 2001). Virulence and Toxin Production Studies on the pathogenesis of C. jejuni demonstrated that for this organism to cause disease, the susceptibility of the host and the relative virulence of the infecting strain are important. Infection with C. jejuni results from the ingestion of contaminated water or food, and the infective dose can be as low as 800 organisms. It is generally accepted that for a microorganism to cause a foodborne illness and to initiate an infection, it must penetrate the gastrointestinal mucus, adhere to the gut enterocytes and once adhered, it can induce diarrhea. In the case of Campylobacter spp. however, it has been

PAGE 40

22 suggested that adherence does not play a role in the infection process. Beery et al. (1988) observed that Campylobacter is able to colonize the crypt mucus without attaching to the crypt microvilli. Lee et al. (1986) noted that Campylobacter is highly motile and rapidly transverse along the intestinal mucus without adhering to the intestinal surface of inoculated mice. Meinersmann et al. (1991) demonstrated that the colonizing Campylobacter stays in the middle of the crypts, without adhering to the epithelial cells. The crypt epithelium is not even damaged during colonization of chicken ceca. Apparently, mucin is much more important in the colonization process than adherence to epithelium cells. Among the factors that influence pathogenicity of Campylobacter spp., toxin production is perhaps the most widely accepted. Campylobacter jejuni synthesizes several toxins, mainly classified into enterotoxin or cytotoxin and that vary from strain to strain. During infection with Campylobacter, the levels of all immunoglobulin classes rise in humans. Of these, IgA is the most important as it can cross the gut wall in an attempt to combat the invading organism. IgA immobilizes organisms, causing them to aggregate, and activates complement. In addition, IgA provides short–term immunity to the host against a recurrent infection with the same organism. Other immunoglobulins do not cross the intestinal wall; instead they act on bacteria entering the blood stream, thus preventing bacteraemia (Wallis, 1994). Recurrent Campylobacter infections may induce partial immunity thus preventing the onset of campylobacteriosis. This partial immunity however,

PAGE 41

23 does not seem to prevent bacteria from colonizing the gastrointestinal tract of the infected host. Such partial immunity is exemplified in studies with young children in developing countries. In those countries, children generally only fall ill with campylobacteriosis up to the age of two; thereafter infections are usually asymptomatic. This partial immune protection is concomitant with raised specific antibodies. In addition to children, workers of poultry slaughterhouses who are repetitively exposed to Campylobacter exhibit rapid acquisition of immunity and protection from subsequent disease (Kaijser and Meinersmann, 1996). Although synthesis of several toxins has been reported, their mechanism of action and their importance in disease remains unclear. Even the very existence of some of the toxins has been questioned. The problem is that some of the researchers who have worked with Campylobacter for years have failed to detect any toxin produced by Campylobacter species. Wassenaar (1997) wrote an excellent review on toxins produced by Campylobacter. In his review, toxins were classified in two classes, depending on their primary mode of action, i.e., enterotoxins and cytotoxins. Enterotoxins (also called cytotonic toxins) are secreted proteins, which have the ability to bind to a cellular receptor, enter the cell, and elevate intracellular cyclic AMP (cAMP) levels. Two well–known enterotoxins include Vibrio cholerae toxin (CT) and Escherichia coli heat–labile toxin (LT). Both toxins have an A/B motif: a larger A subunit, which has enzymatic activity, and the smaller B subunit, which is characterized as a

PAGE 42

24 pentamer and is responsible for binding to a receptor in a cell. When the toxin attaches to the cell, the B subunit binds to the receptor ganglioside GM1 and the A subunit is transported into the cell via receptor–mediated endocytosis. The A subunit is activated inside the cell by proteolytic cleavage and reduction of disulfide bonds. Upon activation, the A subunit “ADP–ribosylates” a GTPase that functions in the production of cAMP, deregulating the cellular adenylate cyclase regulatory system. As a result, the intracellular levels of cAMP rise, and ion flux changes cause excess secretion of fluid resulting in watery diarrhea. In vitro activity of enterotoxins can be demonstrated by the elongation of exposed cultured Chinese hamster ovary (CHO) cells, rounding of mouse adrenal tumor cells (Y–1), and by measuring intracellular levels of cAMP in exposed cells. On the contrary, in vivo activity can be demonstrated using the rabbit or rat ileal loop test (RILT) or the removable intestinal tie adult rabbit diarrhea (RITARD) model. Ruiz–Palacios et al. (1983) were the first authors to describe the production of a cholera–like enterotoxin by C. jejuni. This enterotoxin caused intraluminal fluid secretion in rat RILT model and elongation of CHO cells with an increase in intracellular levels of cAMP. Other researchers later observed similar results when working with some strains of C. jejuni. Because the enterotoxin was detected in cultures of C. jejuni, the toxin was named CJT (C. jejuni toxin), CYTON (cytotonic toxin) or CCT (Campylobacter cytotonic toxin). Neutralization of CJT with anti–LT and anti–CT antibodies suggested immunological

PAGE 43

25 similarities between the three toxins. Fernandez and Toro (1998) noticed that alterations produced in small intestine by the enterotoxigenic capacity of C. jejuni were similar to the hydric, electrolytic and pathological changes caused by CT and LT toxins. The C. jejuni enterotoxin altered the electrolyte exchange mechanism by causing an increase in sodium and chloride ions secretion into the intestinal lumen and increase in tissue cAMP levels. Daikoku et al. (1989) studied toxin production in Campylobacter species and found that both types of toxins—enterotoxin and cytotoxin—were produced by Campylobacter spp. and that they could be isolated from patients suffering from diarrhea. The maximum enterotoxin activity occurred 24 hours after incubation, a time at which cell growth reached the stationary phase. Dillon et al. (1995) reported that maximal enterotoxin production occurred when cultures approached stationary phase and that production yields were influenced by culture conditions, such as medium supplements and growth temperature. For instance, IsoVitaleX (1%) is often added to the culture medium to increase enterotoxin production. The component responsible for this effect is probably ferric iron. Culture filtrates become more strongly positive after polymyxin B treatment, perhaps because the enterotoxin is located in the periplasm (Wassenaar, 1997). Similar results concerning simultaneous production of enterotoxin and cytotoxin by Campylobacter spp. were observed by Florin and Antillon (1992). Interestingly, they observed that bacteria strains that produced only enterotoxin

PAGE 44

26 were recovered from patients with watery diarrhea while strains that produced only cytotoxin were recovered from patients with bloody diarrhea. This, however, was not seen in other studies where patients infected with toxigenic strains more often suffered from watery diarrhea instead of bloody diarrhea (Ruiz–Palacios et al., 1983). Skirrow and Blaser (1992) speculated that because individuals who travel to developing countries usually suffer from inflammatory diarrhea after Campylobacter infection, the difference between watery and inflammatory diarrhea is more likely to be determined by the host susceptibility than by bacteria strain. The controversy about toxin production among Campylobacter strains remains because there are some evidences that some C. jejuni and C. coli are nonenterotoxic toward CHO and Y–1 and do not cause fluid secretion in rabbit RILT model (Wadstrom et al., 1983). Also, there is no homology at the DNA level between the genes coding for CJT and CT, despite the immunological cross–reactivity in vitro and in vivo between CJT and CT or LT (Perez–Perez et al., 1992). Another concern is that the reported frequency of enterotoxin production in clinical samples varies widely among isolates, with high frequencies reaching 100% and at the other extreme, none. Apparently, these conflicting results reflect differences in methodology such as the number of bacterial passages, strain storage and culture conditions (Wassenaar, 1997). The literature available on cytotoxin is confusing because different cell lines, bacterial culture conditions, and strains make it very difficult to compare

PAGE 45

27 the data available. On reading the literature, it is hard to conclude if the damage observed in the cells is the result of the toxin or due to the cell line or bacterial strain used. Actually, several cytotoxins are described in the literature including cytolethal distending toxin (and cytolethal rounding toxin), hepatotoxin, hemolysin, and toxic factor. Cytotoxins are proteins that kill target cells by either acting intracellularly or by forming pores in the cells. The cytotoxins that act intracellularly generally bind to the cells and are processed before they reach the cell cytoplasm. Different mechanisms of toxicity exist, but two predominate: inhibition of cellular protein synthesis and inhibition of actin filament formation. Two well–known cytotoxins that act by inhibiting protein synthesis include Shigella dysenteriae toxin and E. coli Shiga–like toxin (Stx), also known as verotoxin. These two toxins resemble CT and LT in that they contain two subunits, the A subunit with enzymatic activity and a pentamer of B subunits. In this case, however, the B subunits bind to the sphingoglycolipid receptor Gb 3 instead of the ganglioside GM 1 . Upon uptake of the holotoxin, the subunit A is proteolytically cleaved and the activated fragment inactivates ribosomes by depurination of a nucleotide in the 28S rRNA. The resulting inhibition of protein synthesis leads to cell death. In vitro activity of both toxins can be measured in sensitive cells like African green monkey kidney (Vero) cells and human tumor epithelial (HeLa) cells (Wassenaar, 1997).

PAGE 46

28 In addition to acting intracellularly, cytotoxins can form pores in target membrane cells. Hemolysin is an example of such cytolysin, in which the lytic activity targets erythrocytes, but the real importance of cytotoxin lies in their action toward nucleated cells. Pore–forming toxins elicit a broad spectrum of secondary reactions in nucleated cells, including cytokine release, cytoskeleton dysfunction, secretion of granule constituents, and generation of lipid mediators (Wassenaar, 1997). These immune reactions usually cause profound local and distant effects in host tissues. The main problem with a toxin that acts in this manner is that the killing of cells from the host’s immune system such as leukocytes, granulocytes and macrophages may allow the organism to invade host defenses and cause infection. Schulze et al. (1996) investigated C. jejuni isolates from an outbreak and also from poultry sources using a CHO–K1 cell culture to analyze for cytotoxin production. The cytotoxin obtained in their study caused the formation of large, rounded polymorphic and elongated cells that were very slow growers. The cytotoxin induced a slowly developing distention (inflating and stretching) of HeLa, Vero and CHO but had no effect in Y cells. In addition to causing elongation of cells, Whitehouse et al. (1998) observed that the cytolethal distending toxin from C. jejuni causes a rapid and specific cell cycle arrest at the G2 phase in HeLa and Caco–2 cells. According to them, this provides a model for the generation of diarrheal disease by Campylobacter and other bacteria that produces such toxin.

PAGE 47

29 When Schulze et al. (1998) was studying the formation of cytotoxins by Campylobacter, they observed three morphologically different toxins. The first cytotoxin was the same as the one discovered by Schulze et al. (1996). The second cytotoxin produced a rounding of cells without any changes in their size of the cell and a reduction in growth, which was named cytolethal rounding toxin (CLRT). The third cytotoxin caused cells to enlarge in a polymorphic way as well as to acquire a small rounded shape. It is now recognized that some Campylobacter strains can destroy red blood cells, i.e., have hemolytic activity when incubated for 4 days at 42 C. The production of a hemolysin may be related to virulence of those strains. However, different studies have produced conflicting results regarding the hemolytic activity, and it has not been determined yet whether the hemolysin is secreted or cell associated. Interestingly, the hemolytic activity has been observed only in old cultures, suggesting that the hemolytic factor could be an intracellular component that is released when the bacteria die and lyse. Most of the hemolytic activity seen in those strains of Campylobacter is beta–hemolysis, which produces a clear zone around the colonies on blood plates (Wassenaar, 1997). Alpha–hemolytic–like activity was also reported in some strains when the pH of the basal medium was around 6.0 to 6.5 (Misawa et al., 1995). Poultry as Reservoir and Vehicle of Campylobacter Several authors have shown that Campylobacter species, especially C. jejuni, is one of the leading causes of bacterial diarrheal illness in humans in the United

PAGE 48

30 States and other industrialized countries. Figure 3 shows the CDC’s 2001 preliminary data on the incidence of foodborne illnesses in the United States. According to CDC, in 2001, Campylobacter spp. was the second leading cause of gastroenteritis in the US being responsible for 13.8 cases of illnesses per 100,000 persons (Centers for Disease Control and Prevention, 2002). An estimated 2.4 million persons are affected with Campylobacter alone each year (Mead et al., 1999). Campylobacter is commonly found in the intestinal tracts of humans and animals, poultry being the main reservoir for some species such as C. jejuni. It is believed that a stunning 80% of all poultry sold for human consumption in the United States is contaminated with Campylobacter (Centers for Disease Control and Prevention, 2000). Based on this number, it is not a surprise that poultry and poultry products are the most often incriminated food as the cause of campylobacteriosis (Blaser, 1980; Grant et al., 1980; Blaser et al., 1982; Blaser, 1982; Oosterom et al., 1983; Genigeorgis et al., 1986; Harris et al., 1986; Tauxe et al., 1987; Annan–Prah and Janc, 1988; Clark and Bueschkens, 1988; Franco, 1988; Boer and Hahn, 1990; Stern et al., 1992a; Bryan and Doyle, 1995; Dreesen, 1997). Just to give an idea of the impact poultry has on the dissemination of foodborne diseases, during 1968 to 1977, meat and poultry were responsible for 54% of reported outbreaks. Poultry accounted for 19.8% of the incidents. In

PAGE 49

31 Figure 3 "Preliminary FoodNet Data on the Incidence of Foodborne Illnesses – Selected Sites, United States, 2000" and "Preliminary FoodNet Data on the Incidence of Foodborne Illnesses – Selected Sites, United States, 2001." Figure obtained from the CDC's Foodborne Diseases Active Surveillance Network (FoodNet), 2001

PAGE 50

32 another surveillance, from 1966 to 1974, poultry accounted for 15% of all foodborne disease outbreaks for which vehicles were identified. From 1977 to 1984, poultry was identified as a vehicle in approximately 10% of the outbreaks. During this interval, turkeys were responsible for 56% and chickens for 44% of the total poultry outbreaks (Bryan and Doyle, 1995). Similar reports have been published since then. According to Shane (1997), recovery of Campylobacter from chicken carcasses is approximately six times higher than from pork or beef, and ranges from 30 to 100% of specimens surveyed. According to the Centers for Disease Control and Prevention 2000 data, about 80% of all poultry sold for human consumption on that year in the United States was contaminated with Campylobacter (Centers for Disease Control and Prevention, 2000). General Properties of Fresh Poultry Microbial growth in fresh meat is the primary factor associated with spoilage, and loss of meat quality and money (Brewer et al., 1995). The inherent characteristics of poultry and meat such as high protein and moisture content make them an ideal media for bacterial growth and survival. This explains why these products are highly perishable and often contaminated with pathogens. It is well known that there is a relationship between the water content of a product and its perishability—the higher the water content in a food, the more likely is the food to support bacterial growth. However, because water usually associates to molecules with different intensities, i.e., water binds strongly to some molecules and weakly to others, water content alone is not a good indicator

PAGE 51

33 of perishability. Thus, rather than using water content, water activity (a w ) should be used as an indicator of susceptibility to spoilage. Water activity represents the ratio of the vapor pressure of a solution to that of pure water (Troller and Scott, 1992). In other words, a w estimates how much water is engaged in weak associations with molecules, thus being available to support degradative activites, such as growth of microorganisms. The a w of fresh poultry is usually around 0.99 to 0.98, depending on storage time and conditions (Silliker et al., 1980). Bacterial growth depends on a minimal a w , which varies according to the microorganism in question. Campylobacter jejuni requires a minimum a w of 0.97 for growth, Pseudomonas fluorescens needs 0.97, and Salmonella spp. and E. coli require 0.95 (Troller and Scott, 1992). Another important property observed in fresh poultry with regards to microbial growth is pH level. The pH has a substantial effect on the activity of most bacterial enzymes. The pH of fresh poultry ranges from 5.7 to 5.9, which is around the optimal levels for many pathogenic and spoilage bacteria. Poultry microorganisms Islam et al. (1978) observed that the shelf life of poultry meat ended when the total psychrotrophic counts reached 10 6 to 10 7 colonys forming unity (cfu)/g. However, Barnes and Thornley (1976) observed that psychrotrophic microflora produced unacceptable off–odor when their population reached 10 7 to 10 8 cfu/g. Other researchers considered that shelf life of meat products ended when the

PAGE 52

34 total bacterial counts reached 10 7 cfu/g where the meat has unacceptable off–odor, and is considered spoiled (Cunningham, 1979; Brewer et al., 1995). Most spoilage organisms in poultry are psychrotrophs, i.e., they grow in foods at refrigerated temperatures but have temperature optima above 20 C. According to Cousin et al. (1992), psychrotrophs can be defined as those microorganisms that produce visible growth at 7 C 1 within 7 to 10 days, regardless of their optimum growth temperatures. Interestingly, psychrotrophs are considered a subgroup of the mesophiles rather than being psychrophiles. Psychrotrophs are ubiquitous in nature and include yeasts, molds, and a broad range of bacteria. The major psychrotrophic bacteria found in poultry include species of Acinetobacter, Aeromonas, Alcaligenes, Cytophaga, Flavobacterium, Moraxella, Micrococcus, and the most predominant of all, Pseudomonas (McMeekin, 1975). It is important to compare the differences observed in the microflora of poultry and humans because microorganisms such as Campylobacter and Salmonella can survive, grow and multiply in the gastrointestinal tract of both hosts, yet they are commensal organisms in one, i.e., poultry, and pathogens in the other, i.e., humans. Maybe, when researchers have a full understanding of the dynamics and interactions that occur in the gastrointestinal tracts of humans and animals, zoonosis will be better controlled and possibly eliminated from our food system. In contrast to poultry, humans have a much more heterogeneous microflora. According to Simon and Gorbach (1986) the microflora of the

PAGE 53

35 stomach is composed of mainly gram–positive and aerobic bacteria with the most commonly isolated species being streptococci, staphylococci, lactobacilli, and various fungi. The small intestine acts as a zone of transition between the flora of the stomach, usually sparse and meager, and the plentiful flora of the colon. At the proximal small bowel, the microflora is similar to that of the stomach and it is present in low concentrations (10 3 to 10 4 cfu/ml); some Veillonellae, Actinomycos spp., coliforms and anaerobic bacteria are also found in even lower concentrations. As bacteria approach the distal ileum, gram–negative bacteria start to grow more efficiently than gram–positive, outgrowing them. As expected, coliforms predominate and anaerobic bacteria such as Bacteroides, Bifidobacterium, Fusobacterium, and Clostridium are found in significant numbers. Bacterial concentration continues to increase from the distal to the ileocecal sphincter reaching its maximum level at the colon, which can have from 10 11 to 10 12 cfu/ml of bacteria. In the colon however, anaerobic bacteria predominate and the most commonly bacteria found are the Bacterioides, Bifidubacterium, Eubacterium, anaerobic gram–positive cocci, Clostridia, Enterococci, and various species of Enterobacteriaceae. As previously discussed, Campylobacter spp. is widely found in poultry products. Among the Campylobacter spp. present in poultry carcasses, C. jejuni is by far the most prevalent. Dickins et al. (2002) observed that 80% of 448 Campylobacter–positive poultry carcasses were C. jejuni whereas only 14% were C. coli and the remaining 6% were unidentified Campylobacter spp. Other

PAGE 54

36 pathogenic organisms of importance include Salmonella spp., and E. coli (Rose et al., 2002; Snchez et al., 2002). Rose et al. (2002) observed that 10.2% of 22,484 broiler carcasses obtained from federally inspected broiler establishments for a period of three years where contaminated with Salmonella. Ground poultry and turkey had increased levels of contamination; 14.4% of 735 samples of ground poultry and 29.7% of 3,192 samples of ground turkey were contaminated with Salmonella (Rose et al., 2002). One of the reasons poultry is usually contaminated with pathogens is the way the birds are raised and processed. In the United States, during processing of the birds, poultry carcasses are chilled by immersing them in chlorinated cold water. While fast chilling helps to reduce pathogen growth, the practice of immersing several carcasses in the same chill water increases cross–contamination, thus spreading any organism found in one carcass to several others. Snchez et al. (2002) observed that the levels of Campylobacter, Salmonella, and psychrotrophic organisms were higher in immersion–chilled broilers as compared to air–chilled counterparts. Although bacteria are the main organisms responsible for spoilage of poultry products, yeasts are often involved in the spoilage process. Yeasts growing on refrigerated poultry exert proteolytic and lipolytic activities, which liberates substrates that enhance the growth of bacteria in the poultry product (Hinton et al., 2002). Yeasts of the genera Candida are the most prevalent in poultry carcasses, followed by Cryptococcus spp., Yarrowia lipolytica,

PAGE 55

37 Zygosaccharomyces rouxii, Trichosporum beigelii, Kluyveromyces marxianus, and Hansenula anomala (Hinton et al., 2002). Early studies on Campylobacter and poultry Boer and Hahn (1990) isolated C. jejuni from 170 of 279 samples of chicken’s products, which represents a contamination rate of 61%. Harris et al. (1986) analyzed a wide variety of potential risk factors, i.e., food, travel, water, animal and human contact, implicated as the cause of C. jejuni and C. coli infections to see at which degree consumption of specific meats played an etiologic role in Campylobacter outbreaks. As expected, they found that consumption of contaminated poultry was an important cause of campylobacteriosis. But, more important than that, they found that consumption of fresh chicken and Cornish game hen was twice as often by cases than by controls and that the association of raw chicken with campylobacteriosis was even stronger then that with chicken in general. This reveals that the incidence of poultry contamination in the field or during processing and handling of the product is higher than cross–contamination during cooking. In addition, they found that of the retail meats sampled in the survey, chicken and game hen were the only ones commonly found to harbor C. jejuni and C. coli. To better understand the epidemiology of Campylobacter infections, the Centers for Disease Control and Prevention did a national surveillance in 1983 (Tauxe et al., 1987). Interesting results were obtained with the surveillance. They found that person–to–person transmission and point source outbreaks are

PAGE 56

38 unusual to occur. In addition they confirm that poultry is the single dominant nationwide source of campylobacteriosis and that there is a seasonal and an age–specific isolation rate. The seasonal distribution of Campylobacter isolation is pretty constant along the years, with summer being the peak of contamination in retail poultry. According to Tauxe et al. (1987), the similarity of the seasonal distribution in all age groups and geographical regions suggests that a single route of transmission could be the major source of campylobacteriosis. Infants have higher incidence of campylobacteriosis followed by young adults (20 to 30 years of age); within infants, the second month of life is at highest risk (Tauxe et al., 1987; Bryan and Doyle, 1995). These peaks relate to times of weaning and when young adults move out of their parent’s house and set up housekeeping and food preparation on their own (Beery et al., 1988). Smitherman (cited by Genigeorgis et al., 1986) observed that, differently from humans, the probability of infection in chickens increase with age and that the average age of infection manifested by fecal shedding was 40 to 45 days. Campylobacter jejuni is a commensal bacterium in poultry, colonizing the ceca without causing discernible symptoms of the disease (Franco, 1988; Dreesen, 1997). The principal change associated with C. jejuni infection in chickens comprises distention of the intestinal tract extending from the distal duodenal loop to the bifurcation of the ceca. Accumulation of watery fluid and mucus

PAGE 57

39 occurs and depending of the cytotoxic properties of the Campylobacter involved, hemorrhages may be present (Shane, 1997). The intestinal tract of poultry is colonized with Enterobacteriaceae (including coliforms), clostridia and fecal streptococci within a day of hatching. As the bird ages, colonization by these microorganisms diminish and are replaced by a characteristic microflora in different regions of the intestines. C. jejuni colonizes primary the lower gastrointestinal tract of chicks and may be present in that location in populations as high as 10 7 cfu/g (Bryan and Doyle, 1995; Beery et al., 1988). The colonization is higher in the ceca, large intestine and cloaca where densely packed cells are localized in the mucus within crypts (Bryan and Doyle, 1995; Beery et al., 1988; Meinersmann et al., 1991). In the chickens, the bacteria are found mainly in the ceca, large intestine, and cloaca (Soerjadi et al., 1982; Beery et al., 1988). Campylobacter appear to pervade the lumina of crypts without attachment to crypt microvilli (Bryan and Doyle, 1995; Beery et al., 1988; Meinersmann et al., 1991). Campylobacter jejuni is chemoattracted to mucin, which can be utilized as a sole substrate for growth, and to L–fucose, an important constituent of mucin. Apparently, chemoattraction to mucin may also attract the bacteria to the mucus, in which they move by highly active flagella to mucus–filled crypts where the organisms then establish themselves (Bryan and Doyle, 1995). In contrast to poultry, eggs are believed to be free of Campylobacter species. Contamination of freshly laid eggs occurs with other bacteria, but contamination

PAGE 58

40 of the egg in the field has not been established for C. jejuni (Clark and Bueschkens, 1988; Jay, 1992). Izat and Gardner (1988) observed that the inherent characteristics of the eggshell make very unlikely the possibility of contamination under proper sanitation and normal plant operating conditions. For Bryan and Doyle (1995), however, proper sanitation was a not sine qua non condition for obtaining Campylobacter–free eggs because they observed that eggs that laid by Campylobacter–positive hens did not contained Campylobacter. Also, samples obtained from the yolks, albumin and shell surfaces of such eggs were not positive for Campylobacter indicating that this organism is not likely to be transmitted by the eggs. Even if feces of contaminated hens come in contact with the eggs during ovoposition, under commercial conditions Campylobacter would not survive the desiccation on the egg surface. According to Shane (1997) vertical transmission of Campylobacter by either transovarian infection or by penetration of the egg shell after ovoposition has not been shown, thus explaining why eggs are not considered one of the sources of campylobacteriosis for humans. The mode of action and vehicle of transmission of Campylobacter are not yet completely understood, however some suggestions have been made about how the pathogenicity is carried. It has been proposed that C. jejuni colonizes the human intestinal mucus in a manner similar to that of the spiral bacteria of the normal microflora, which are spiral, microaerophilic and can penetrate deeply into the intestinal crypts (Stern et al., 1992a). Apparently, the adaptation to the

PAGE 59

41 intestinal mucus niche may be an important determinant of the virulence of C. jejuni. Campylobacteriosis that cause profuse watery diarrhea might indicate that these strains could be enterotoxigenic, while dysentery–like disease suggests that some strains of the species might be invasive (Jay, 1992; Stern et al., 1992a). Evidences that Campylobacter may be invasive comes from the nature of the symptoms, the rapid development of high agglutinin titers after infection, recovery of the organism from the peripheral blood during acute phase of the disease, and the findings that C. jejuni can penetrate HeLa cells (Jay, 1992). Even though the scientific community recognizes the fact that retail poultry is highly contaminated by Campylobacter, how the flocks get contaminated with the organism and how the spreading of the pathogen is still without consensus among several researchers since young chicks are usually free of Campylobacter. Several factors are believed to help the introduction and spread of Campylobacter in the poultry houses including, but not being limited to, the environment in and near rearing houses, wild birds, human transit, footwear and clothing of farm staff, and contaminated litter (Bryan and Doyle, 1995). Dreesen (1997) pointed that horizontal transmission from some environmental sources is indeed the most likely way young chicks are infected with Campylobacter. Sources vary from close contact of other farm animals (such as pigs, cattle, sheep and other fowl) to providing non–chlorinated or otherwise non–treated water to the broilers. Houseflies (Musca domestica) and litter are significant sources of contamination, which play an important role in the

PAGE 60

42 dissemination of avian campylobacteriosis (Montrose et al., 1985; Shane et al., 1985). The role of litter in horizontal transmission is of extremely importance because chickens carrying intestinal C. jejuni excrete the organism in the feces for up to 63 days when transferred from litter to wire, regardless of any coprophagy among birds (Montrose et al., 1985). Some researchers can consider poultry incubators a potential source of Campylobacter contamination. Clark and Bueschkens (1988) showed that one infected chick in an incubator can potentially infect other chicks before the birds reach the farm in transporting boxes. However, it has been already demonstrated that vertical transmission of Campylobacter is very unlikely to occur; therefore the possibility of having an infected chicken in the incubator is quite low. Furthermore, incubators are generally disinfected with formaldehyde, which reduces bacterial numbers on the surface of the eggs. According to Blaser (1980) and Blaser et al. (1980), although the venereal route is important for transmission of Campylobacter species in some animals, transmission by this route in humans has not been documented, neither has person–to–person transmission. Vertical transmission from mother to neonate, however, has been shown. The infection is probably related to the rupture of the membranes or during the passage of the baby through the birth canal. Bryan and Doyle (1995) analyzed several factors that contribute to outbreaks of poultry–borne disease and they found that improper cooling of food is responsible for 48% of the cases, while food prepared a day or more before

PAGE 61

43 serving accounts for 34%; inadequate cooking or thermal processing, 27%; an infected/colonized person touching cooked foods, 23%; inadequate reheating of cooked foods, 20%; improper hot storage of cooked foods, 19%; cross–contamination of cooked food from raw foods, 15%; inadequate cleaning of equipment, 11%; and ingestion of raw products, 8%. The fact that intestinal tracts of some turkeys and broilers are loaded with Campylobacter contributes to contamination of the carcasses during processing reflecting the high contamination levels found on poultry meat (Shane, 1997). The levels of C. jejuni in the intestinal tract of turkeys and poultry can vary between 10 4 to 10 8 cfu/g and often the majority of birds (about 75%) reach the slaughterhouse harboring the bacteria (Beery et al., 1988). Bacterial population on surfaces of raw poultry carcasses at the end of processing ranges, usually, from 10 3 to 10 5 aerobic mesophilic organisms per cm 2 . Evisceration and giblet operations are other major points of spread of bacteria (Oosterom et al., 1983; Franco, 1988; Bryan and Doyle, 1995). The problem is that many bacteria found in poultry adhere firmly to the carcasses during processing, making decontamination of the carcass hard to be accomplished (Bryan and Doyle, 1995). Fecal contamination of feathers, skin, and hides occurs during poultry and livestock production and slaughter, with the potential for carcass contamination during processing. The transport of chickens from the farm to the slaughterhouse helps to spread the disease potentiating the distribution of pathogens on the carcasses. In average, there is an increase of 10 4 cfu of

PAGE 62

44 Campylobacter spp. per poultry carcass if the chicken is transported to a processing facility to be slaughtered (Stern et al., 1995). The mechanism by which the carcass becomes contaminated during processing initiates with the retention of bacteria within a liquid film on the skin from which they migrate to the skin becoming entrapped in the ridges and crevices (Bryan and Doyle, 1995). The retention of the bacteria does not necessarily begin on the carcass; instead it starts on the live birds. After that, bacterial retention is enhanced with the scalding process and it continues throughout the whole processing of the bird (Bryan and Doyle, 1995). Not surprisingly, the level of carcass contamination is somehow proportional to the microbial concentration of the processing water. The hot water used during scalding opens the feather follicles so the feathers can be easily plucked (Bryan and Doyle, 1995). Once the follicles are opened, they remain open throughout processing until chilling when they close thereby retaining microorganisms that have attached inside the follicle contaminating the subcutaneous layer (Bryan and Doyle, 1995). In addition to the retention of the microorganisms, during chilling, certain bacteria (e.g. Salmonella) quickly adhere and attach to polysaccharides and to the material surrounding collagen fibers. Since attachment of bacteria is time–dependent, more bacteria are transferred from the water film on the surface of the carcass to the skin as chilling time increases. Furthermore, entrapment of water may protect the bacteria attached to the skin against bactericides (Bryan and Doyle, 1995).

PAGE 63

45 Even when processing has finished, contamination with Campylobacter does not cease. Cross–contamination of cooked foods from raw chicken is the primary pathway of spreading of pathogens such as Campylobacter. During food preparation, contamination of surfaces such as cutting–boards and plates, ready–to–eat foods containing raw and cooked ingredients, and poorly washed hands represent the significant way to increase and disseminate Campylobacter (Boer and Hahn, 1990). Prevention and Control of Campylobacter Although most of the pathogens present in contaminated poultry can be safely eliminated during the process of cooking and the spread of the same pathogens can be avoided by basic hygiene practices, the blame and the responsibilities for food outbreaks falls always on the poultry and meat industry. Between 1988 and 1992, 2,423 outbreaks of foodborne disease were reported to the Centers for Disease Control and Prevention (CDC). Of these outbreaks, 1,435 had information reported concerning contributing factors. During this 5–year period, 59% of all outbreaks were due to improper holding temperatures, and 36% were caused by poor personal hygiene of food workers (Bean et al., 1996). Unfortunately, consumers like to assume that food is sterile, when in reality it is not and most likely it will never be. Consumers want food that it is unrealistically safe regardless of how the product is handled after it is out of the company’s control. Altekruse et al. (1996) reported the results of a national telephone survey on consumer knowledge of

PAGE 64

46 foodborne microbial hazards and food–handling practices. About one–third of the respondents reported that they do not wash hands or take any precautions to prevent cross–contamination from raw meat. On the other hand, the food industry is also guilty of unsafe hygiene and deficient food–handling practices. Several major outbreaks in the history of the United States’ meat industry can be cited as examples of poor control exerted by the food manufacturers and/or distributors. In 1997, Hudson Foods Inc. recalled 25 million pounds of frozen ground beef suspected of being contaminated with E. coli O157:H7. That was the largest recall of meat products in U.S. history (Knight and Pierce, 1997; Henkel, 1998). A year later, a deadly outbreak involving Listeria monocytogenes found in various brands of hot dogs forced Sara Lee to recall 15 million pounds of product. Not long ago, in July of 2002, ConAngra Beef Company was responsible for the second largest recall of beef products, recalling 19 million pounds of beef trimmings, and fresh and frozen ground beef that were suspected to be contaminated with E. coli O157:H7 (Veneman and Murano, 2002). Although the United States Food and Drug Administration (FDA) and the United States Department of Agriculture (USDA) have no authority under the Federal Food, Drug, and Cosmetic Act (Sec. 412 and Sec. 518 Food Drug and Cosmetic Act; Sec. 351 Public Health Service Act) to order recalls, companies try to remove products from the marketing as expeditiously as possible, in order to avoid irremediable damages to company. Listeria spp., E. coli 0157:H7, Salmonella

PAGE 65

47 spp., Campylobacter spp. and other pathogens can be safely killed by proper cooking, but the final responsibility for a safe food at the consumer’s table lies, undoubtedly, on the food industry. Prevention and control of Campylobacter infection at the animal level is hard and sometimes impractical. Under commercial conditions, unrestricted movement of personnel, recycling of litter, and the use of earth–floor convection–ventilated housing subject to ingress of flies, vermin, and possibly wild birds and mice, all contribute to colonization of the intestinal tract of poultry with Campylobacter (Shane, 1997). Franco (1988) suggested a complete decontamination of the chicken house as a way of preventing Campylobacter from spreading between flocks. Decontamination includes removal of the litter after raising each flock, i.e., no recycling of the litter, disinfection of the equipment and buildings and a rest period between flocks of at least seven days. The efficacy of these practices on the elimination of Campylobacter however is questionable. Removal and elimination of litter, thorough cleaning and disinfection of building and equipment with quaternary ammonium, iodine or chlorine, rest periods between flocks of 10 to 15 days and pulverization of new litters with quaternary ammonium are common husbandry measures practiced in the poultry industry in Brazil (Englert, 1991). Still, Brazilian broilers are not Campylobacter–free. Thus, measures for eliminating and preventing Campylobacter infections in poultry need to go beyond sanitary practices.

PAGE 66

48 Another food that is usually implicated in foodborne illnesses due to Campylobacter infections is unpasteurized milk. Campylobacter species are heat–sensitive and are easily destroyed by milk pasteurization temperatures. Prevention of campylobacteriosis can be accomplished by avoiding unpasteurized milk, raw or undercooked poultry products, and cross–contamination of read–to–eat products (Jay, 1992). Nurmi and Rantala (1973) developed the concept of competitive exclusion as a mean of reducing Salmonella infections in broilers. Competitive exclusion consists of administering intestinal contents or even bacterial cultures from older chickens to young chicks (Nurmi and Rantala, 1973; Wierup et al., 1988). Their hypothesis was that as the development of the intestinal flora is delayed when broilers are produced in a milieu low in microbes, the newly hatched birds are in a transitional state between germ–free and normal animals for a few days. During the first days after hatching, the defense against bacteria entering the body through the mucosa or colonizing on it is probably weaker than normal, making the young chicks susceptible to colonization by several bacteria. Because the chick’s resistance against some bacteria increases with age, Nurmi and Rantala suggested that it is important to protect the very young birds against colonization with pathogenic microorganisms. Competitive exclusion can be achieved with defined or undefined cultures. Undefined culture consists of introducing suspensions of fecal droppings or cecal contents from adult chickens into day–old birds. Problems

PAGE 67

49 associated with undefined culture include large variability in the efficacy of colonization in young birds and a possible spread of avian or human pathogens to the birds. These concerns can be addressed by defining the bacteria used in the competitive exclusion treatment, i.e., by using the defined culture instead of undefined. An undefined culture can be transformed into a defined culture by manipulating bacteria composition, medium selection, use of antibiotics, and growth conditions. Schoeni and Doyle (1992) and Schoeni and Wong (1994) demonstrated that colonization of the chick cecal crypts with C. jejuni could be inhibited by using bacterial isolates that occupied the same niche of Campylobacter and that those isolates produced metabolites antagonistic to the pathogen in question. Similar inhibition of C. fetus subsp. jejuni colonization in poultry was observed by Soerjadi et al. (1982). The benefits of competitive exclusion go beyond the colonization aspect. By–products of bacterial metabolism such as short chain fatty acids (acetic, propionic, and butyric acid) may help to create an intraluminal environment, which restricts microbial growth by inhibiting bacterial proliferation (Simon and Gorbach, 1986). At sufficient low pH, these three short chain fatty acids are undissociated and capable of entering the bacterial cell, inhibiting its metabolism. Furthermore, some species of bacteria have the ability to produce bacteriocins, which are specific antibiotic–like substances that inhibit the growth of other bacteria on an autoregulatory way (Simon and Gorbach, 1986).

PAGE 68

50 Some researchers have noticed that the administration of dietary carbohydrates helps to reduce and inhibit bacterial colonization. Schoeni and Wong, 1994 reported that carbohydrates might enhance the efficacy of competitive exclusion by inhibiting adherence of pathogens to the crypts, decreasing cecal pH, and/or influencing bacterial populations in the intestine. However, Simon and Gorbach (1986) noticed little effect on the composition of fecal microflora when poultry diet was modified. According to the later authors, diet may influence metabolic activities of the microflora, but certainly not on the taxonomic bacterial composition. Factors that Have a Potential Effect on Growth and Survival of Campylobacter Allyl Isothiocyanates The food industry relies largely on a combination of deleterious stresses or hurdles to keep their products safe by inhibiting, killing or inactivating contaminating microorganisms in food systems. In order to achieve a satisfactory level of safety and preservation, a combination of intrinsic and extrinsic hurdles (implicit inhibitory and lethal factors) are used. These hurdles are usually applied simultaneously or in a sequential way as a means to abbreviate the life span or inactivate the development of unwanted organisms in the food product (Rowan, 1999). As the food industry works hard on new interventions to assure safe food, consumers are increasingly demanding a safer and cheaper but healthier and more natural foods that lack preservatives, sodium and other shelf extending ingredients. And as expected, changes in

PAGE 69

51 processing and addition or removal of food additives largely influence the physiology of food spoilage as well as pathogenic organisms (Knchel and Gould, 1995). Although death rates from food–related illness are considerably lower in developed countries, it has been pointed out that up to 5% of episodes of acute gastroenteritis lead to serious, and often chronic sequelae (including rheumatic conditions such as Reiter’s syndrome and ankylosing spondylitis), nutritional and malabsorption problems, haemolytic–uraemic syndrome (caused by verotoxin–producing stains of Escherichia coli), and other illness such as Guillain–Barr syndrome following infection by Campylobacter species (Lindsay, 1997; Rowan, 1999; Ang et al., 2001). It is important to note however that the reasons for such increased numbers are not necessarily due to an increase in the infection rate, but most likely due to a combination of factors such as better reporting of the illness, changes in husbandry practices, changes in food marketing and consumption habits, expanded international and domestic travel, increased number of sensitive populations, improved and more sensitive detection methods of pathogens in samples and identification of new and emerging pathogens. The problem with Campylobacter and other pathogenic bacteria is that contamination can occur after processing even if any hurdle applied to the product during processing is efficacious. Cross–contamination of cooked foods from raw chicken is the primary pathway of spreading the organism after

PAGE 70

52 processing. During food preparation, contamination of surfaces, such as cutting–boards and plates, ready–to–eat foods containing raw and cooked ingredients and poorly washed hands represent the most important areas for cross–contamination (Boer and Hahn, 1990). In order to reduce cases of cross–contamination, poultry should be free of Campylobacter to begin with. One possible way of reducing the number of organisms present in poultry is to treat carcasses with allyl isothiocyanate (AIT) during processing. Isothiocyanates are natural plant–origin compounds derived from a group of glycosides stored within cell vacuoles of all plants belonging to the family Cruciferae, also known as Brassicaceae (Mazza, 1984; Delaquis and Mazza, 1995; Delaquis et al., 1999; Kim et al., 2002). Other plants such as papaya are also known to contain these compounds (Tang, 1973). Upon injury to the plant tissues, glucosinolates are quickly hydrolyzed by the cell wall enzyme myrosinase (Delaquis and Mazza, 1995; Shofran et al., 1998). This reaction leads to the release of isothiocyanates, nitriles and thiocyanates (Figure 4). Antimicrobial activity in plants or plant extracts has been long investigated (Delaquis and Mazza, 1995; Kyung and Fleming, 1996; Shofran et al., 1998; Ward et al., 1998; Cherry, 1999; Delaquis et al., 1999; Kim et al., 2002). Early research has shown the preservative effect of the essential oils of the mustard seed (Walker et al., 1925). The compounds responsible for the antimicrobial esters of isothiocyanic acid, or isothiocyanates (Hoffman and Evans, 1911; Mazza,

PAGE 71

53 Glucosinolate Nitriles R–CN + S + Isothiocyanates R–N=C=S + H2SO4 Myrosinase + H2O D–glucose + R–S–CN Thiocyanates Figure 4 Glucosinolates and main products of hydrolysis by the enzyme myrosinase (adapted from Delaquis and Mazza, 1995)

PAGE 72

54 esters of isothiocyanic acid, or isothiocyanates (Hoffman and Evans, 1911; Mazza, 1984; Delaquis and Mazza, 1995; Kyung and Fleming, 1996; Shofran et al., 1998; Ward et al., 1998; Delaquis et al., 1999; Kim et al., 2002). Mazza (1984) reported that steam distillation of horseradish root yields an essential oil, which comprises up to 90% AIT, 4 to 10% 2–phenethyl isothiocyanate (PEIT) and a mixture of several compounds. Sinigrin is a glucoside (molecule of glucose linked to a molecule of AIT) found in cabagge, Brassica oleracea var. capitata that upon hydrolysis, as in the case of other Cruciferae, produces AIT upon injury (Walker et al., 1937). Sinigrin is hydrolyzed by myrosinase yielding four distinct coumpounds: allyl isothiocyanate (AIT), allyl thiocyanates (ATC), allyl cyanadie (AC), and 1–cyano–2,3–epithiopropane (CETP) (Shofran et al., 1998). Yields of each compound are not fixed and will vary according to the pH of the reaction and the presence of ferrous ion and an epithiospecifier protein on myrosinase. For instance, AITs are produced at neutral pH while ACs are produced at pH 4; CETPs are formed when ferrous ion and the myrosinase protein are combined (Shofran et al., 1998). To avoid unwanted breakdown of sinigrin inside the plant, the glucoside is present in certain cells while the myrosinase enzyme is present in other cells (Walker et al., 1937). Sinigrin, by itself, has no antimicrobial importance; its hydrolysis products however are widely used as antimicrobial agents inhibiting the growth of several microorganisms (Brabban and Edwards, 1995). Allyl isothiocyanate

PAGE 73

55 isolated from mustard seed and horseradish is well recognized as being a potent antimicrobial compound (Delaquis and Mazza, 1995; Shofran et al., 1998; Cherry, 1999). Depending on the bacterium, a minimum inhibitory concentration of 50 to 500 ppm is sufficient to curb growth (Kyung and Fleming, 1996). Shofran et al. (1998) reported that when compared to all sinigrin’s hydrolysis products, AIT proved to be the most inhibitory against a large selection of bacteria and yeasts, even being considered more inhibitory than sodium benzoate, a common food preservative. Furthermore, contrary to sodium benzoate, the antimicrobial activity of AIT varied very little over pH 4 to 7. According to Shofran et al. (1998), the main difference in the antimicrobial potencies over a wide range of pH between those two food preservatives is related to their protonic state; the fully protonated form of benzoic acid penetrates the cell membrane, dissociates to lower internal cell pH stopping growth. Allyl isothiocyanate, on the other hand, does not have an acidic proton with which it can dissociate, thus diffusing more readily through the cell at any pH (Shofran et al., 1998). Isothiocyanates have been determined to be an effective antimicrobial with broad spectrum bacteriostatic, bactericidal, and antifungal activity toward Aspergillus, Gibberella, Colletrotrichum, Penicillium, Cladosporium, Escherichia coli O157:H7, generic E. coli, Staphylococcus aureus, Listeria monocytogenes, Salmonella typhimurium, and Saccharomyces cerevisiae (Walker et al., 1937; Carter et al., 1963; Tsunoda, 1994; Shofran et al., 1998, Ward et al., 1998; Delaquis et al., 1999). The isothiocyanates have antibacterial activity against both gram–negative and

PAGE 74

56 gram–positive bacteria (Isshiki et al., 1992, Shofran et al., 1998, Cherry, 1999) and they are believed to be effective against Campylobacter organisms. So far, no work on the effects of isothiocyanates on Campylobacter jejuni of poultry source has been published in the literature. Although isothiocyanates are known to have antimicrobial properties, few studies were published on the effect of isothiocyanates on shelf stability of foods, especially poultry. The reason for this lack of research is most likely due to the fact that synthetic preservatives (as opposed to isothiocyanates) are relatively inexpensive and highly effective. But, restrictions imposed by the food industry, consumers and regulatory agencies on the use of some common synthetic food preservatives have drawn attention back to natural compounds, particularly those from plants. Papadopoulos et al. (1991a) stated that compounds that can extend the shelf life of products highly susceptible to spoilage, without significantly affecting the palatability or visual characteristics of cooked products, are indeed of great value to the food industry. Delaquis et al. (1999) demonstrated that exposure of roast beef to 20 ml of vaporized horseradish essential oil restricted the growth of most spoilage bacteria, including Pseudomonas spp. and Enterobacteriaceae, and preserved the color of the cooked meat. Kim et al. (2002) reported that AIT combined with acetic acid extended the shelf life of cooked rice, proving that the combination of both compounds works as a potent antimicrobial. This extension of shelf life was acquired without any detrimental effects on the sensory qualities of the cooked rice. The growth of several

PAGE 75

57 foodborne microorganisms was completely inhibited for seven days in agar plates kept at 12 C when volatile distillate of horseradish root was added to the plates (Ward et al., 1998). The use of AIT in the food industry is not new. The compound is added to mayonnaise and horseradish to serve as flavor enhancer. Allyl isothiocyanate is legally added to food products in Japan as food preservatives provided that the compound is extracted from natural sources and it is not synthetically made. The AIT is required to be acquired from natural sources because during production of the synthetic counterpart, the compound can be contaminated with traces of allyl chloride (Clark, 1992). Despite its potential as an antimicrobial, addition of AIT to food products is limited because of its strong pungent flavor, especially if purified isothiocyanates or crude essential oil preparations are used. The mode of action of AIT is still not totally known. It is believed that AIT inhibits survival and growth of microorganisms by binding to sulfhydryl groups on the active sites of enzymes responsible for the bacterium’s metabolism (Tang, 1974; Kawakishi and Kaneko, 1985; Kolm et al., 1995; Shofran et al., 1998). Isothiocyanates react easily with free amino groups of amino acids and proteins to give their thiourea derivatives (Kawakishi and Kaneko, 1985; Kawakishi and Kaneko, 1987). Thus, the isothiocyanates derived from Brassica plants may decrease the function of bacterial proteins by interacting with them (Kawakishi and Kaneko, 1987). Tang (1974) reported similar explanation for the inhibitory

PAGE 76

58 effects of benzyl isothiocyanate on papain. The enzyme glutathione transferase seems to important on the inhibitory function of AIT because it promotes transfer of a thiol group from a molecule called g–glutamylcysteinylglycine to the electrophilic central carbon of the isothiocyanate group forming a dithiocarbamate (Kawakishi and Kaneko, 1985; Kolm et al., 1995). In addition to being a potent antimicrobial, isothiocyanates can also be beneficial to humans. Recent research on isothiocyanate has shown that it can inhibit chemicals responsible for tumor formation and that AIT can prevent cancer (Caragay, 1992). Interestingly, while there are researchers showing that isothiocyanates can inhibit cancer formation, others have shown that they can cause cancer. Musk and Johnson (1993) studied the mutagenic and carcinogenic properties of allyl isothiocyanate and other isothiocyanates and found them to be mutagenic. For the immediate future, there is an urgent need for studies on food preservation and safety that focus on the multiplicity of sub–lethal stresses that are commonly experienced by foodborne bacterial pathogens during food manufacture, distribution and storage. Natural compounds such as allyl isothiocyanate have the potential to intervene on these processes without many deleterious effects on the product (Delaquis and Mazza, 1995; Shofran et al., 1998; Cherry, 1999; Delaquis et al., 1999; Kim et al., 2002). The use of AIT can be advantageous to the food industry because pathogenic bacteria have not developed resistance to the compound. This is an important factor when

PAGE 77

59 choosing an antimicrobial agent because foodborne bacterial pathogens are quickly adapting to the deleterious stresses to which they are being submitted during manufacturing and distribution of food products. Acid–Shock Campylobacter prefer to grow in neutral pH, but must nevertheless contend with acid encounters when passing through the gastrointestinal tract. During passage through the gastrointestinal tract, Campylobacter cells must travel through the extreme acid environment of the stomach, followed by a slightly higher pH in the duodenum and survive amidst high concentrations of volatile fatty acids in the intestine. According to Foster (1995), moderate acid conditions down to pH 4.5 in some bacteria, such as Salmonella typhimurium, are handled well by constitutive pH homeostasis systems that serve to maintain a more neutral internal pH relative to an acid external environment. There is little information on the effect of pH on Campylobacter species. Most of the studies available focus on how different pHs negatively affect growth of Campylobacter. There is no evidence in the literature that an acid shock treatment, i.e., exposure of Campylobacter to very low pH for a specific amount of time, was attempted as a way of recovering starved Campylobacter cells. The idea of looking for an inducible acid shock system on Campylobacter cells stored over prolonged periods of time originated from the differences in pH seen in the gastrointestinal tract of any warm–blooded animal. The normal passage of Campylobacter cells through the acid environment of the stomach is by

PAGE 78

60 definition an acid shock. It has been demonstrated several times that starved cells have the ability to resuscitate, i.e., become culturable, after passage through the gastrointestinal tract. Therefore, it seemed reasonable to predict that the starved and/or severely injured cells might be programmed to view exposure to low or mild acid as a predictor of future exposure to optimal conditions for colonization and pathogenesis inside a host. If this prediction holds true, exposure to low pH will induce the expression of genes responsible for recovery of starved cells, which acts similarly to other survival systems present in bacteria, such as heat shock, oxidative stress, DNA damage, and osmotic shock system. Christopher et al. (1982) demonstrated that Brucella broth adjusted to a pH of 5 to 9 with 10% lactic acid caused either a decrease or increase in the number of viable Campylobacter fetus cells. At pH 5 and 9, no survivors were detected while the number of cells recovered at pH 6 thru 8 increased rapidly. When studying low pH adaptation and acid tolerance response of Salmonella typhimurium, Foster (1995) observed that survival at pH 3 requires the synthesis of 50 proteins when cells undergo an acid shock. These proteins are called acid–shock proteins and they apparently serve as a signal for the induction of genes related to pathogenesis. In addition to the acid–shock system, other stress–response systems are found in several bacteria. For instance, two very important stress–response systems are the osmotic– and the heat–shock response. The importance of these two systems relates also to cross–protection that they give the bacteria to other

PAGE 79

61 environmental stresses such as heat, salt, and irradiation of foods (Rowan, 1999; Rowe and Kirk, 1999; Gahan et al., 1996; Berry and Cutter, 2000). The heat stress response system causes an increase in the synthesis of stress chaperone proteins (Hsp). The Hsp act by altering the kinetics of protein folding favoring correctly assembled molecules and protein complexes (Heredia et al., 1997; Heredia et al., 1998; Rowan, 1999). According to Heredia et al. (1998) and Rowan (1999) bacterial cells containing these stress proteins appear to have enhanced potential to repair damages, resist further stresses and to prevent cell death. In some enteric bacteria, such as E. coli and S. typhimurium, the induced synthesis of stress proteins involved in general stress response including the heat shock proteins and certain outer membrane proteins, is believed to help the cell to maintain intracellular homeostasis in adverse environments (Heredia et al., 1997; Heredia et al., 1998; Rowan, 1999). Another point to consider when studying acid–shock is the fact that bacterial cells in stationary phase are more resistant to environmental stresses (Booth, 1998). Apparently, the cornerstone of the tolerance some microorganisms exhibit when exposed to sub–lethal stresses is due to the activation of existing enzymes and transport systems regulated by changes in gene expression. Both specific and general regulons, i.e., a group of coordinately–regulated genes that are activated in response to an environmental parameter, are involved in reducing the susceptibility of an organism to stress (Rowan, 1999). When working with E. coli, Rowe and Kirk (1999) observed that

PAGE 80

62 entry into the stationary phase growth or a period of starvation had the ability to induce synthesis of protective proteins encoded by the rpoS gene; this gene gives the bacterium resistance to a range of chemical and physical stresses. Rowe and Kirk (1999) demonstrated that cross–protection of E. coli O157:H7 to lethal levels of acid and salt is mediated, at least in part, by the same rpoS gene. The importance of resistance to sub–lethal environmental stress such as starvation goes beyond providing cross–protection to bacteria. Some environmental signals that have been found to have a modulating influence on the expression of virulence determinants in bacteria include carbon dioxide (CO 2 ), temperature, iron, pH, starvation, carbon source, osmolarity, growth phase, calcium ions (Ca +2 ), oxidative stress, phenolic compounds, monosaccharides, phosphate, sulfate (SO 4 ), nicotinic acid and amino acids (Rowan, 1999). Several foodborne bacteria, including S. typhimurium, L. monocytogenes, Vibrio cholera, Shigella spp., Yersinia spp., E. coli, Staphylococcus aureus, and Pseudomonas aeruginosa, have expressed their virulent determinants under environmental stress (Rowan, 1999; Rowe and Kirk, 1999). Importance of Acid–shock in Foodborne Pathogens The major host defense mechanisms against bacterial overgrowth in the normal bowel and later in the small intestine are the normal peristaltic activity of the gastrointestinal tract itself and the gastric acid (Simon and Gorbach, 1986; Villarreal et al., 2000). Gastric acid has also been implicated as a protective mechanism in maintaining the bacterial population of the upper small bowel to a

PAGE 81

63 low number. As expected, the gastric juice acts also as the main barrier against pathogens. The importance of such barrier can be easily observed in the reduction of infective dose when pathogens are administered to patients in conjunction with acid neutralizers, like food or bicarbonates (Simon and Gorbach, 1986). The infective dose of Vibrio cholera decreases from 10 11 to 10 6 total organisms when volunteers are pretreated with sodium bicarbonate prior to inoculation. Due to an adaptation process, many enteric pathogens are capable of surviving diverse environmental conditions, including the diversified conditions found in the human’s gastrointestinal tract. The enteric bacteria may persist outside of the host in an aqueous, low–nutrient environment by entering into a state of dormancy. Upon ingestion by a host, or transfer to a suitable environment, the dormant bacterium may become active causing infection, leading to disease (Pace et al., 1997). In order for the bacterium to induce the disease in the host, it must be able to adhere to the gastrointestinal system, replicate, produce virulent factors and depending on the organism, invade the tissue before it is flushed out of the intestine. Thus, the ability to respond rapidly and efficiently is essential for the success of the pathogen; it requires that the bacterium be able to sense its surroundings. The idea that bacteria can sense their surroundings is not new. Bacteria can sense the environment and thus communicate with each other by small molecules called autoinducers. Often, each bacterium in a population

PAGE 82

64 synthesizes and releases into the environment a constant amount of a low–molecular–weight signaling molecule, the autoinducer, whose concentration reflects the total number of bacteria in the population (Aizawa et al., 2000). The theory is that at very low population density, the concentration of autoinducers is so low that the bacterial population does not notice their presence, but as the population increases, so does the concentration of autoinducers. When a certain threshold density is reached, it triggers some activity in bacteria capable of detecting the autoinducer. The autoinducers accumulate to sufficient concentrations only when there is a critical mass of cells in a confined environment (Greenberg, 1997). The terminology for such event is described as quorum sensing, which is described as the mechanism that alters gene expression in response to bacterial population density (Aizawa et al., 2000). Quorum sensing was first described in a symbiotic association of a population of Vibrio fischeri that colonized the light organ of the Hawaiian squid Euprymma scolopes, providing bioluminescence to the squid in exchange for a supply of nutrients (Greenberg, 1997; Aizawa et al., 2000). Other quorum sensing systems were later observed in several bacteria, both gram–positive and gram–negative. In bacteria, autoinducers include peptides in gram–positive and acylated homoserine lactones in gram–negative bacteria (Greenberg, 1997; Aizawa et al., 2000). Quorum sensing has been demonstrated in foodborne pathogens such as E. coli O157–H7, Yersinia enterocolitica, and Salmonella enterica serovar Typhimurium 14028 (Greenberg, 1997; Aizawa et al., 2000; Institute of

PAGE 83

65 Food Technologists, 2002). Quorum sensing has been linked to virulence in enterohemorrhagic (EHEC) and enteropathogenic (EPEC) E. coli (Sperandio et al., 1999). It is believed that the low infectious dose required by E. coli O157:H7 is partially because of quorum sensing. E. coli can be induced to colonize the intestine by quorum sensing of signals released by nonpathogenic E. coli that is part of the normal flora of the host because bacteria can recognize generic autoinducers (Aizawa et al., 2000; Institute of Food Technologists, 2002). Quorum sensing of autoinducers from generic E. coli is likely to occur based on studies on exotoxin production in Pseudomonas aeruginosa. Pseudomonas does not produce exotoxins when present at low densities; it waits until cell density has reached such a high level that the exotoxins produced could overwhelm the host’s defense, depriving it of the chance to respond immunologically (Greenberg, 1997). Of great importance is the fact that in both E. coli and Salmonella one type of autoinducer called autoinducer 2 (AI–2) is synthesized during the exponential phase and is degraded at the onset of the stationary phase (Aizawa et al., 2000). So far, it is still unknown which autoinducer Campylobacter produces and if it is linked to virulence or to resistance to environmental stresses. Is the morphological transformation of spiral to coccoid cells a result of quorum sensing when Campylobacter is under stress? Is there a quorum sensing system that induces morphological transformation in bacteria? At present, the answers to these questions are unknown. However, the possible relationship between

PAGE 84

66 quorum sensing and survival to environmental stress should be considered and evaluated in the future. A paradox associated with C. jejuni is that this pathogen is sensitive to several environmental factors such as exposure to air, drying, low pH, heating, freezing and prolonged storage (Hunt and Abeyta, 1995) yet Campylobacter is the leading cause of gastrointestinal illnesses in the United States. One possible explanation for this phenomenon may be that while Campylobacter species are sensitive to the environmental factors listed above when present in the food system, the microorganism may regain its strength and proliferate more readily inside the host. Furthermore, the lack of an outstanding culture medium and easy–to–follow methodology for culturing Campylobacter may contribute to low recoveries of the organism when injured or present in very low numbers in the food system. Campylobacter species are catalase positive (Hunt and Abeyta, 1995). Catalase catalyzes the formation of water and oxygen from hydrogen peroxide when oxygen is used by the microorganism as a terminal electron acceptor during respiration. Hydrogen peroxide must be converted to water and oxygen because of its toxicity to the cell. Catalase also enables the organism to overcome the peroxides generated from the oxidative burst caused by host macrophages and neutrophils that are recruited to the submucosa when the bacteria are invading the small intestine mucosa (Lee et al., 1997).

PAGE 85

67 The idea of studying acid–shock, acid–tolerance and cross–protection in foodborne pathogens is not limited to Gram–negative bacteria. Marron et al. (1997) reported that L. monocytogenes exhibited a significant acid–tolerance response following a 1–hour exposure to milk acid (pH 5.5), which was capable of subsequently protecting cells from severe acid stress (pH 3.5). This same protection against acidic pH provided the bacterium protection against lethal levels of other stresses such as heat, ethanol, salt and crystal violet. The acid–adaptation also enhanced the survival of L. monocytogenes in foods containing lactic acid, citric acid, or acetic acid. Gahan et al. (1996) reported similar results to that published by Marron et al. (1997). In his study, Gahan et al. (1996) demonstrated that adaptation to sub–optimal growth environments gave L. monocytogenes the ability to alter its cellular physiology in a way that it made the bacterium more resistant to further stresses. The researchers showed that L. monocytogenes that had been previously exposed to low pH were able to survive in acidic foods such as orange juice, salad dressing, yogurt, and cottage cheese, otherwise not suitable for their survival. The importance of the acid tolerance and cross–protection has been widely studied in S. typhimurium and E. coli (Gahan et al., 1996; Berry and Cutter, 2000). Both S. typhimurium and E. coli have shown to alter synthesis of several outer membrane and heat shock proteins to maintain intracellular homeostasis in a pernicious environment (Gahan et al., 1996). Berry and Cutter (2000) reported

PAGE 86

68 that the acid tolerance of E. coli O157:H7 is enhanced or sustained longer upon storage at refrigeration temperature. Few papers have been published on the production of cold–shock proteins in Campylobacter. Hazeleger et al. (1998) and Chan et al. (2001) observed that Campylobacter have a very narrow growth temperature with the minimum being around 31 to 32 C. In order for bacteria to grow at temperatures far below their optimum, synthesis of cold–shock proteins is often needed. For instance, E. coli is known to synthesize several cold–shock proteins including CspA, which is believed to act as an RNA chaperone to block the formation of secondary structures in the mRNA (Yamanaka et al., 1998). Synthesis of similar protein however was not found to occur in C. jejuni, thus explaining why Campylobacter survive at low temperatures but are less likely to grow (Hazeleger et al., 1998). Methods for Culturing and Analyzing Campylobacter Selective Culture Media Several growth media for Campylobacter spp. have been developed over the years. Nonetheless, no consensus has been achieved yet on which medium should be selected as the standard medium for growing and culturing Campylobacter in the laboratory. One of the difficulties in developing an optimum selective and differential medium for Campylobacter is the fact that this pathogen is a fastidious organism, which requires special growth temperature, gaseous environment and nutrient–rich basal medium. Another problem is that cultures of Campylobacter are usually overgrown by other cultures such as

PAGE 87

69 coliform bacteria, Proteus spp. and yeasts and molds (Goossens and Butzler, 1992; Stern et al., 1992b; Jeffrey et al., 2000). Media that have gained the attention of several researchers and laboratories for the isolation of Campylobacter include Skirrow formulation, Butzler’s Agar, Campylobacter Blaser Agar, Preston, Semisolid Blood–free Selective Medium (SSM), Campy–Brucella Agar Plate (Campy–BAP), Campylobacter Cefoperazone Desoxycholate Agar (CCDA), Abeyta–Hunt–Bark Agar (Campy–FDA), Campylobacter Thioglycollate medium, Brucella Broth with 0.16% Agar, Semi–Solid Campylobacter Medium, and Campy–Cefex. These are the media that have been widely used, tried and modified (Wang et al., 1980; Bolton and Coates, 1983; Goossens and Butzler, 1992; Stern et al., 1992b; Hunt and Abeyta, 1995; Hunt et al., 1998; Jeffrey et al., 2000). All of the media listed differ in the source and amount of basal medium, presence or absence of growth enhancers such as horse or sheep blood, source and concentration of antibiotics, and presence of special components like charcoal. Nevertheless, they are all nutrient–rich media capable of supporting growth of a large number of fastidious organisms without much selectivity towards Campylobacter spp. When looking at a medium formulation, it is important to note that the rate of recovery of the desired organism is as crucial as other features. An ideal Campylobacter medium must provide excellent recovery of organisms while providing substantial selectivity against background flora and allowing for easy and quick differentiation of bacterial species. Furthermore, the medium has to be

PAGE 88

70 easy to prepare, fairly inexpensive and it must have good shelf life. If possible, additions of supplements following heat–steam sterilization should be minimized to reduce the risk of post–sterilization contamination of the medium. In addition, the medium should allow the organism to grow into isolated and distinct colonies without inducing swarming, which restricts quantification of the organism. The first selective medium developed specifically for Campylobacter spp. was described by Skirrow and became known as the Skirrow formulation. The medium contained a nutrient base, 7% lysed horse blood and a combination of antimicrobials—trimethoprim, polymyxin B and vancomycin. Campylobacter was able to grow well on the medium, but so did fecal background flora and Proteus spp., which often predominated (Goossens and Butzler, 1992). Campy–Bap was developed shortly the development of the Skirrow’s medium. It contained a Brucella agar base, instead of the blood agar base found in the Skirrow medium, and 10% sheep blood. Both media included nutritionally rich ingredients, which contained high levels of thymidine to reverse the inhibitory effect of the antimicrobial trimethoprim (Goossens and Butzler, 1992). Trimethoprim is added to the medium to suppress growth of some gram–negative organisms present in fecal flora, especially Pseudomonas spp. The same is true for other antimicrobials such as cefoperazone, and polymyxin B. Vancomycin and rifampin, on the other hand, inhibit growth of gram–positive bacteria. Rifampin was also toxic to Campylobacter probably due to the accumulation of hydrogen

PAGE 89

71 peroxide inside the cell (Goossens and Butzler, 1992). And finally, cycloheximide and amphotericin B are added to media to suppress the growth of yeasts. Another Campylobacter medium that was widely used was the Butzler’s Agar, which contained Columbia agar base instead of the Brucella agar or blood–base agar. It was necessary to add 5% defibrinated sheep blood, cefoperazone, rifampicin, amphotericin B and colistin to the Butzler’s agar. As an alternative to the current commercial media available for culturing Campylobacter, the United States Department of Agriculture (USDA) developed the Campy–Cefex medium. The main differences in the medium as compared to CCDA and Campy–Bap was the lack of blood, the incorporation of FBP—a mix of Ferrous Sulfate (FeSO4.7H2O), Sodium Metabisulfite and Sodium Pyruvate—and the addition of lysed horse blood. The list of antimicrobials was reduced to only sodium cefoperazone and sodium cycloheximide as opposed to vancomycin, polymyxin B, trimethoprim lactate, amphotericin B and cephalothin, which are added to the Campy–Bap medium. The medium was effective in recovering C. jejuni and slightly superior to the other two media listed above (Stern et al., 1992b). George et al. (1978) reported that the incorporation of FBP to solid media enhanced the growth of Campylobacter. As a matter of fact, the culture medium Abeyta–Hunt–Bark developed by the United States Food and Drug Administration (FDA) relies on FBP to augment the growth of C. jejuni (Hunt and Abeyta, 1995; Hunt et al., 1998). In addition to FBP, heart infusion agar,

PAGE 90

72 sodium cefoperazone, rifampicin and amphotericin B are included in the medium. Bolton et al. (1983) compared the efficacy of 5 different selective media for isolation of Campylobacter: Skirrow’s, Butzler’s, Blaser’s, Campy–Bap and Preston’s media. According to the authors, Preston medium preceded by enrichment on modified Preston Enrichment Broth proved to be the most selective medium for Campylobacter while Butzler was the least effective. Since all media were incubated under the same conditions of time, temperature and atmosphere, the difference observed in isolation rates were due to the basal medium, growth promoting additives such as amount and source of blood, and inhibitory supplements such as antibiotics present in each media formulation. How FBP works in the media is still not completely known, but it is believed that the chemical mixture quenches superoxide anions generated in the medium. Campylobacter are very sensitive to superoxide anions (Goossens and Butzler, 1992; Corry et al., 1995; Blais and Phillippe, 1999). It is also considered that FBP can improve the aerotolerance of Campylobacter (Smibert, 1978; Goossens and Butzler, 1992). According to Smibert (1978), iron and bisulfite together act non–enzymatically to destroy superoxide radicals, whereas pyruvate can destroy hydrogen peroxide. The incorporation of FBP mix into media is especially important when medium is exposed to light and air because high–energy radicals such as superoxide and peroxides are generated by photochemical reduction of medium components (Smibert, 1978). If present in

PAGE 91

73 the medium, these high–energy radicals inhibit the growth of Campylobacter. The addition of FBP to the medium reduces the hydrogen peroxide toxicity by a factor of 10 (Smibert, 1978). Often other oxygen scavengers such as glutathione are added to the Campylobacter media to assist in the recovery of stress–damaged bacteria. Glutathione, as well as pyruvate, are believed to protect the cells against self–destruction by reducing the oxygen tension of the media and by stimulating the expression of genes associated with the stationary phase, which allows cells to adapt to stress responses protecting them against denaturation (Dodd et al., 1997). There is a controversy on which supplement works best for Campylobacter recovery: blood or FBP. Both supplements are added to the medium to provide some protection against toxic oxygen products, thus promoting growth. In addition, blood contains enzymes such as catalase, peroxidase, and peroxide dismutase, which further decompose toxic oxygen derivatives (Goossens and Butzler, 1992). Another alternative to the blood or FBP is the incorporation of charcoal to the medium. Bolton and Coates (1983) observed that charcoal could effectively replace blood in media formulations. The CCDA medium contains 4% charcoal instead of the typical 5% blood observed in other Campylobacter media. This charcoal–containing medium provides excellent selectivity and good recovery of Campylobacter (Stern et al., 1992b). Nonetheless, CCDA has its drawbacks such as the need for constant shaking of the medium during plate

PAGE 92

74 pouring to avoid precipitation of the charcoal, and the fact that the dark opaque color of the medium makes identification of the translucent colonies of Campylobacter a difficult task. Another problem in dealing with fastidious organisms arises when samples need to be taken in the field and transferred to the laboratory for further analysis. With this in mind, Jeffrey et al. (2000) developed a field–suitable, semisolid aerobic enrichment medium of isolation of C. jejuni. The medium contains, among other components, Thioglycollate basal medium, beef heart infusion, and a long list of antimicrobials similar to Campy–Bap. The addition of Thioglycollate helped to allow for aerobic incubation while the other components gave this medium a triple function, i.e., enrichment, holding, and transport. This field–suitable semi–solid medium resembles the Semisolid Blood–free Selective Medium (SSM medium) used to isolate and differentiate Campylobacter organisms on the basis of swarming. Goossens et al. (1989) developed the SSM medium to give researchers and clinical microbiologists an easy and inexpensive way to isolate Campylobacter from stool samples. The SSM medium is composed of Mueller–Hinton broth supplemented with cefoperazone and trimethoprim and a smaller amount of agar than the Jeffrey’s medium (0.28% to 0.4% as compared to 0.6% in the field–suitable semisolid medium). The only problem with the SSM medium was the inability to identify nonmotile Campylobacter organisms.

PAGE 93

75 The osmolarity of the culture medium has proven to be as important to bacterium growth as its components. Growth medium with elevated osmolarity has been linked to changes in cell morphology of E. coli and Bacillus subtilis (Ruzal and Sanchez–Rivas, 1994; Schleyer et al., 1993). On the other hand, low–osmolarity has been shown to alter the cell morphology of Campylobacter jejuni to an atypical coccoid–form (Reezal et al., 1998). This controversy regarding which solid media should be used when culturing Campylobacter can be also applied to enrichment broths. As with solid media, broths differ in nutrient composition, oxygen–quenching agent, optimum temperature and incubation time, and required atmospheric condition. Among five widely used Campylobacter Enrichment Broths (Rosef, Hunt, Yamazaki, Bolton and Tran), all of them contain yeast extract while the majority contains sodium chloride, FBP and peptone. Rosef broth differs from the others by incorporating Resazurin; Hunt broth combine both FBP and lysed horse blood in the same formulation; Yamazaki broth relies on erythrocyte digest, beef extract and potassium bicarbonate as a substitute for FBP and blood; Bolton broth is the most different one and boosts its formula by adding peptic digest of animal tissue, lactalbumin hydrolysate, two FBP components, sodium carbonate, haemin, saponin lysed horse blood and n–ketoglutamic acid; and finally, Tran broth contains FBP, casein hydrolysate, a–ketoglutamic acid, sodium carbonate and charcoal, instead of blood (Blais and Phillippe, 1999).

PAGE 94

76 In an attempt to test many of the claims that have been made concerning the feasibility of the broths being used for enriching Campylobacter, Blais and Phillippe (1999) compared the growth of Campylobacter when incubated in Rosef’s, Hunt’s, Yamazaki’s, Bolton’s and Tran’s broths. The most effective method for culturing the organism was to grow it on Hunt broth with lysed horse blood, with 5% CO 2 and moist air at 40 C. This methodology however has been tested only in pure cultures and not with poultry products, which are rich in background flora. Culturing Methodology Another problem widely seem with Campylobacter spp. is the lack of a suitable culturing methodology. This is easily noticed by reviewing the literature and data available on the incidence of Campylobacter–contaminated poultry carcasses. It is estimated that incidences in retail ranges from 2% to 98% of the carcasses tested (Denton, 2002). According to CDC, about 80% of all retail poultry carcasses are contaminated with this pathogen (Centers for Disease Control and Prevention, 2000). How can it be that such wide variation in levels of contamination is reported in the literature? The explanation for such question reflects the absence of a generally accepted method for its isolation and recovery on a consistent basis. The idea that a standardized, reliable, fast and accurate methodology for testing foods for the presence of Campylobacter is important to the food industry and to consumers is neither new nor surprising. The fact that in 2001,

PAGE 95

77 Campylobacter was the second leading cause of gastroenteritis in the US (Centers for Disease Control and Prevention, 2002) warrants the need for the development of such methodology. Nonetheless, it is important to note that good recovery and isolation methodologies are the result of a thorough understanding of the biochemistry and dynamics of the organism. The environmental sources of Campylobacter have not been clearly established. Due to this gap in the understanding of this pathogen, no proven intervention strategies are currently available for use by poultry producers, farmers and poultry processors for reduction and control of the organism. Until the ecology of Campylobacter is fully understood, attempts in achieving the best procedures for culturing the pathogen will be based on assumptions and experimentations. Thus, it is pivotal for the food industry that the scientific community focuses on basic research that will help to elucidate the life cycle of this intriguing organism. Some authors have even considered that this shortage of knowledge on how to control Campylobacter combined with a deep understanding of the ecology of other pathogens such as Salmonella are contributing for an increase in levels of Campylobacter in poultry (Denton, 2002). Denton (2002) reported that traditional reduction strategies employed for Salmonella reduction and control have repeatedly shown to be ineffective against Campylobacter. Furthermore, it is believed that the massive Salmonella control in the poultry industry is creating a suitable environment conducive to the emergence of Campylobacter.

PAGE 96

78 Denton (2002) has pointed out that several laboratories have invested time and resources to develop reliable culturing methodologies without much success. According to the author, Campylobacter is by nature very difficult to isolate and culture with accuracy and precision from live animals and foods, and United Stated Department of Agriculture Food Safety and Inspection Service (FSIS) has recently asked the National Advisory Committee on Microbiological Criteria for Food (NACMCF) to review the methodologies used for Campylobacter detection in the 1994 and 1999–2000 baseline studies in young chickens and compare them to recent advances in methodologies. Porcine bile juice in culture media Campylobacter is a gram–negative bacterium, i.e., when appropriately stained using the Gram stain procedure Campylobacter is visualized with a pink counterstain called safranin. What makes the differentiation possible between the two types of bacteria – gram–negative and gram–positive – is the composition of their cell wall. The ability of a bacterium to retain or not a crystal violet–iodine stain is related to the thickness and components of its cell wall. The cell wall of the gram–negative bacteria has an additional layer—the outer membrane—as compared to the cell wall on gram–positive bacteria, which is made of proteins and phospholipids structured as a fluid lipid bilayer. A structure called lipopolysaccharide (LPS) is attached to this outer membrane and sticks out of the cell. The LPS contain a lipid A, a core and a repeating oligosaccharide called O–antigen. This LPS layer is very important in gram–

PAGE 97

79 negative bacteria as it confers resistance against hydrophobic compounds, such as bile salts. Bile salts act as detergents, solubilizing the membrane proteins and leading to the destruction of the cell. Therefore, cells that have the ability to resist the harmful effects caused by bile salts seem to be better suited for passage through the small intestine of warm–blooded animals. But it is important to note that even cells that contain LPS can be destroyed by bile if its concentration is high. The human liver secretes 600 to 1,200 ml of bile per day into the duodenum (Rhoades and Tanner, 1995). Bile contains bile salts, bile pigments, cholesterol, phospholipids and proteins and it is formed in the liver from cholesterol and bile acids. The bile acids are then converted into bile salts as conjugates of taurine or glycine and secreted in the duodenum (Rhoades and Tanner, 1995). Bile acids and their conjugates are essential components of bile. The physiological effects of bile acids include: induction of bile flow; feedback inhibition of bile acids synthesis, modulation of cholesterol synthesis; elimination of cholesterol; and facilitation of dispersion and absorption of lipids and fat–soluble vitamins (Brunton, 1996). Normal functioning of the gastrointestinal tract depends on appropriate synthesis and enterohepatic circulation of bile salts (Brunton, 1996). Several studies have been published on bacterial tolerance to bile salts. A common finding that has been reported is that adaptation to lethal concentrations of bile salts leads to a significant cross–protection towards heat

PAGE 98

80 shock (Christopher et al., 1982; Flahaut et al., 1996). According to Flahaut et al. (1996) pretreatment with bile salts resulted in the induction of a subset of the heat shock proteins (18 out of 21 bile salt stress polypeptides are also induced by heat), which may be responsible for the observed cross–protection effect. Again, similar to the effects caused by acid shock, heat and bile salt shock induces the formation of proteins that help to protect the cell against external stress factors. Bile is usually added to selective bacterial culture medium as a bacterial growth inhibitor. High levels of bile salts can indeed be toxic to bacteria, but low levels of bile might be important in regulating bacterial physiology and it may play an important role in the host–pathogen interaction (Pace et al., 1997). Pace and colleagues showed that bile can enhance growth of nutrient–deprived Vibrio parahaemolyticus and enhance its virulence, bacterial capsule size, and adherence to epithelial cells suggesting that the bile acid–containing environment found in the human host favors the growth of that particular bacterium and that bile acids enhance the expression of such virulent factors. According to the authors, bile might act on the bacteria through a calcium–dependent mechanism, which might be analogous to the low–calcium response exhibited by other pathogenic bacteria. Pace et al. (1997) also reported that in one of their unpublished studies, Campylobacter jejuni cultured with bile was able to adhere and invade epithelial cells at greater level as compared to that cultured without bile. Carvalho et al. (1997) studied the incidence of C. jejuni in the viscera, gallbladder and bile of

PAGE 99

81 broiler chickens. According to the authors, liver was the organ of choice for Campylobacter infection. The frequency of C. jejuni in the bile was low (6.9%) nevertheless it showed that Campylobacter could survive and maybe multiply in high concentrations of bile. In another study, Bertchinger (cited by Carvalho et al., 1997) observed that as much as 21% of the bile samples from chickens contained Campylobacter spp. Porcine gastric mucin in culture media In addition to bile juice and bicarbonate, bacteria encounter mucus in the small intestine. The human small intestine secretes about 2 to 3 liters/day of isotonic alkaline fluid. This secretion is derived mainly from cells in the crypts of Lieberkhn, which are tubular glands located at the base of intestinal villi. The intestinal secretion most likely helps to maintain the fluidity of the chyme and may also play a role in washing away pathogens. Present in this intestinal secretion are various mucins secreted by specialized epithelial cells in the small intestine and colon called globet cells (Ktyi, 1990; Corfield et al., 1992; Mantle and Rombough, 1993; Bradshaw et al., 1994; Rhoades and Tanner, 1995; Sharma et al., 1997). The globet cells secrete mucus into the intestinal secretions and this mucus lubricates the mucosal surface and protects it from the mechanical damage by solid food particles. In the small intestine, it may also provide a physical barrier to the entry of microorganisms into the mucosa. Interestingly, some enteric

PAGE 100

82 pathogens interact with this mucous material in the small intestine and depend on them to invade and survive inside the host (Ktyi, 1990). Mucins are glycoproteins that are high in carbohydrates (70 to 80%) and which form a gel when in solution. They are extremely diverse in structure and are usually very large molecules (>2 x 106–kDa). These glycoproteins consist of numerous oligosaccharide side chains (O–linked glycoproteins) are primarily sialic acid, galactose, N–acetylglucosamine and fucose linked via either serine or threonine residues to the polypeptide backbone (Mantle and Rombough, 1993; Bradshaw et al., 1994). The oligosaccharides are present in different combinations though each completed chain terminates in an a–anomerically linked saccharide: sialic acid and fucose are always terminal and the galactose and N–acetylglucosamine are b–linked (Corfield et al., 1992; Bradshaw et al., 1994). These chains are the ones that confer the characteristic physical–chemical properties on the glycoproteins that make up the mucus gel and the polymeric structure observed in mucin is critical for the gel formation. Mantle and Rombough (1993) observed that the reduction of the disulfide bridges that hold together the glycoprotein monomers or the proteolytic digestion of nonglycosylated mucin peptide releases glycoprotein monomers that do not gel under normal physiological conditions, confirming that the polymeric structure of mucin is indeed essential for gel formation. The diversity of the oligosaccharide side chains and of the polypeptide backbone creates a complex molecule that requires the action of several bacterial

PAGE 101

83 enzymes to be degraded (Bradshaw et al., 1994). The high number of the chains grouped in repetitive pattern protects the underlying polypeptide backbone against proteolytic cleavage by bacterial enzymes, except in regions lacking oligosaccharide chains or after the side chains have been degraded (Corfield et al., 1992). Fecal extracts and anaerobic fecal cultures contain enteric bacteria called mucin oligosaccharide–degrading (MOD) strains, which synthesize linkage–specific enzymes capable of degrading the oligosaccharide chains that make up the mucus (Corfield et al., 1992). Most of these enzymes are called glycosidases and they are exoglycosidases, which act sequentially to catalyze hydrolytic cleavage of the glycosidic linkages beginning at the outer non–reducing end of each chain (Corfield et al., 1992). Ktyi (1990), and Mantle and Rombough (1993) observed that several virulent and non–virulent enteric bacteria utilize mucin as a source of energy and depend on it for growth. Some of these organisms are of major concern in the food industry such as E. coli, Campylobacter jejuni, S. typhimurium, Clostridrium perfringens and Shigella flexneri (Ktyi, 1990; Mantle and Rombough, 1993; Sharma et al., 1997). The authors observed that purified rabbit small intestinal mucin was shown to enhance growth of Yersinia enterocolitica and it served as the sole source of nutrient (nitrogen and carbon source) for the bacteria. The same effect was seem in both virulent and nonvirulent strains, even if the organisms were being cultured in the presence of a complex background rich in nutrients.

PAGE 102

84 Oligosaccharide chains of human colonic mucin glycoproteins appear to have a high content of sialic acids whose hydroxyl groups on carbons 7, 8, and 9 are highly substituted by acetyl esters (Corfield et al., 1992). Although O–acetyl substitution of the C hydroxyl group strongly inhibits cleavage of sialic acid from glycoproteins by bacterial sialidases, O–acetyl substitution on C–7, C–8, and C–9 hydroxyl groups also inhibits cleavage, but less strongly (Corfield et al., 1992). Therefore, O–acetyl groups on sialic acids have been regarded as a protective adaptation that prevents or retards mucin oligosaccharide chain degradation by inhibiting cleavage of the terminal sialic acid at the outer non–reducing end of chain (Corfield et al., 1992). As previously stated, interaction of the enteric organisms with small intestinal mucous is very important. Sylvester et al. (1996) observed that Campylobacter upsaliensis adheres to mucin and that mucin fucose seems to be a chemotactic factor to Campylobacter jejuni. According to the author, specific adherence to mucin receptors by organisms having the peculiar corkscrew motility typical of Campylobacter species may actually help penetration of the mucus barrier by allowing the bacteria to dock in place before they bore through the mucus to reach the underlying epithelial cells. Lee et al. (1997) observed that an uncharacterized coccoid organism responsible for some types of gastric inflammation in humans expresses adhesin that binds to gastric mucin. The adhesin activity was detected only after the isolate was exposed to an acidic pH, suggesting that in the natural process of

PAGE 103

85 infection, the low pH of the stomach triggers a physiological modification on the cell surface component, which provides the cell with adhesin activity. Preliminary taxonomic classification of the coccoid organism indicated similarities to the Staphylococcus DNA homology groups containing S. cohnii and S. xylosus. Sharma et al. (1997) were able to demonstrate that the intestinal mucosa of rats is capable of adapting itself to different diets and microbial populations by modifying mucin composition, crypt–villus architecture and goblet cell glycoconjungates. Apparently this type of adaptation is not restricted only to rats; the same authors showed that the composition of the poultry feed, in particular the consumption of diets supplemented with the enzyme xylanase, alters the viscosity and composition of the mucin by changing the goblet cell glycoconjugates containing sialic acid and N–acetylglucosamine residues of the chicken intestinal tract. Microbiological Electron Microscopy Several morphological forms of Campylobacter spp. have been described in the literature (Rhoades, 1954; Ogg, 1962; Ng et al., 1985; Rollins and Colwell, 1986; Moran and Upton, 1987; Beery et al., 1988; Jones et al., 1991; Meinersmann et al., 1991; Buswell et al., 1998; Federighi et al., 1998; Reezal et al., 1998; Lzaro et al., 1999; Karlyshev et al., 2001). These morphological forms include spirals, S–shapes, commas, gull shapes, dimpled shapes, club–shape, donut, and coccoid shapes. The spiral shape however is the most predominant and it is considered

PAGE 104

86 one of the distinguishing characteristics of Campylobacter species. Both the spiral morphology and the darting motility of Campylobacter are used to differentiate this organism from other enteric bacteria. All of the above researchers have also reported in their publication the appearance of coccoid forms of Campylobacter in aging and stressed cultures. The spiral and S–shape forms of Campylobacter predominates in young cultures and on the actively growing cells such as the cells located at the periphery of a colony (Rhoades, 1954; Ng et al., 1985). The coccoid form on the other hand is mostly found in older cultures and on the center of a colony where mostly inactive cells are present (Ogg, 1962; Ng et al., 1985). Some forms described in the literature as donut or club–shape is believed to be, by some authors, an intermediate stage between the active spiral form and the inactive coccoid form. Apparently, all Campylobacter species start having a spiral shape and as time progresses and the cell ages, the transformation from spiral shape to coccoid takes place. The process by which the cells change their morphology is not well documented in the literature and no work that demonstrates the sequential changes in cell morphology has been published. Therefore, the fact that one spiral Campylobacter eventually becomes coccoid is based on the inferences and on the research done with other spiral–shaped bacteria such as Spirillum, Vibrio and Oceanospirillum. The morphological changes observed in Campylobacter are

PAGE 105

87 also believed to be concomitant with a rapid decline in culturability (Rhoades, 1954; Ogg, 1962; Ng et al., 1985; Rollins and Colwell, 1986; Jones et al., 1991). Despite the fact that the sequential steps observed with Campylobacter during the morphological changes from spiral to coccoid have not been described in the literature, it is believed that the cell starts to enlarge on one end forming a club shape. The cell wall starts to separate from the cytoplasm and it begins to stretch out. As the cell wall stretches, it loses its integrity and cytoplasmic material leaks out causing the cell to quickly degenerate. This degenerative state can be accompanied by bleb formation (Ng et al., 1985) or not. By the end of the morphological and physiological transformations, the cell becomes a round, empty cytoplasm with a stretched cell membrane. Some electron micrographs published in the literature depict such empty round cells and describe them as dead or severely damaged cells (Ng et al., 1985; Rollins and Colwell, 1986). But not all coccoid cells are regarded as a degenerative state of Campylobacter spp. As a matter of fact, some researchers consider the coccoid form of Campylobacter a survival state, which allows the bacteria to survive and remain viable during stressful situations such as lack of nutrients, low molality, low temperatures, etc. (Rollins and Colwell, 1986; Reezal et al., 1998). As a fastidious organism, Campylobacter requires nutrient–rich environments and normally grows well only in vivo. Transmission to a new host involves a period of exposure to hostile environments and stressful conditions. The coccoid cells that

PAGE 106

88 appear to remain viable have condensed cytoplasm and intact cell wall, which differ from the round, degenerative cells. Nevertheless, even if these dense coccoid cells are still viable their recovery and culture in the laboratory is difficult when conventional culture techniques are employed (Reezal et al., 1998). At high temperatures, Campylobacter spp. quickly changes from a culturable spiral–shaped cell to nonculturable coccoid–shaped cell. When maintained under low–temperatures in nutrient–poor environment however, this morphological transformation can take as long as 1 to 4 months to take place (Hazeleger et al., 1998). Similar transformation from a motile, spiral–shaped cell to a coccoid form is observed with Helicobacter pylori. H. pylori is associated with large numbers of gastric disorders in the industrialized world, and cultures from individuals suffering from such disease bear the bacterium in both morphological states. Several authors have observed that, as seen in Campylobacter species, old cultures or under aerobic conditions, the motile, spiral–shaped H. pylori changes into a nonmotile coccoid form and became nonculturable and noninfectious. Thus, it has been speculated that these coccoid forms appear under unfavorable conditions as an infectious agent that can survive outside the human host (Lee et al., 1997). Campylobacter and Helicobacter species share several similar characteristics. Both species are gram–negative spiral–shaped cells at young age, which will eventually change to a coccoid shape as the colony ages. The coccoid shape also

PAGE 107

89 appears to be associated with a decrease in ability to subculture the bacterium. They are catalase and oxidase positive. Plates must be incubated microaerophilically and colony growth is slow. Colonies from both species appear, depending on the culture medium, as small, pinpoint, translucent, nonhemolytic colonies. As with Campylobacter, Helicobacter cells are motile by means of a single, polar or bipolar flagellum (Murray et al., 1995). Because of the similarities these two species share, until 1989 Helicobacter pylori were classified as Campylobacter pylori. To date, knowledge acquired for one species is often extended to the other. When working with dogs, Fox et al. (1996) observed that several species of Helicobacter (H. hepaticus, H. bilis, H. pullorum, and H. canis) had a tremendous resistance to bile (1.5% desiccated ox bile), which probably contributed to the ability of the organisms to colonize the liver. H. canis was found to grow well in Campylobacter selective media at 42 C under microaerophilic conditions. An interesting point about these coccoid cells is that they are very difficult to culture in the laboratory. In addition to H. pylori, uncharacterized coccoid cells were also found in human gastric inflammations. Scanning electron micrographs of such cells revealed a large variety of cell morphologies, which range from perfectly spherical forms to dented ones that ended up giving a donut–shaped form to the organism. They also observed that the coccoid–form bacteria bound to porcine gastric mucin only if they were previously pre-exposed to a pH of 3.5

PAGE 108

90 or lower, which approaches the prevailing pH of the human and porcine stomach. One of the puzzles that still remain to be solved with Campylobacter relates to its helical morphology. It is not known what causes the bacterium to have a spiral shape and what mechanisms are involved during the transition from spiral to coccoid cells. Much of the work on the dynamics and composition of helically shaped bacteria has been done on Spirochetes, Vibrio and Aquaspirillum (Williams and Rittenberg, 1956; Williams and Rittenberg, 1957; Joseph and Canale–Parola, 1972; Greenberg and Canale–Parola, 1977; Canale–Parola, 1984; Krieg, 1984; Johnson and Faine, 1984). Not surprisingly, most of the helically shaped bacteria transform into “spherical bodies” under unfavorable conditions and as culture ages. However, the spherical forms of the bacteria are often regarded as nonviable (Canale–Parola, 1984; Krieg, 1984; Johnson and Faine, 1984). The ability to undergo morphological transformation appears to be genetic because old cultures of some species of Aquaspirillum, such as A. fasciculus, show massive amounts of coccoid forms as opposed to others (Krieg, 1984). Joseph and Canale–Parola (1972) observed that the helical shape of spiral bacteria is maintained by the peptidoglycan layer because the purified peptidoglycan layer of Spirochaeta species retained its coiled configuration. Also, treatment of the spiral cells with lyzosyme and penicillin, which interfere in the synthesis of peptidoglycan layer, resulted in loss of the helical shape (Joseph et al., 1973; Canale–Parola, 1978).

PAGE 109

91 Later, Greenberg and Canale–Parola (1977) reported that the helical shape of Spirochetes could be lost through mutation, which results in a rod shaped bacterium, frequently with bent ends. Based on their findings, the researchers concluded that the helical configuration of the spirochetes was maintained by the peptidoglycan layer. It is not known if the peptidoglycan of helically shaped Campylobacter differs from the coccoid shaped cell. Baker and Park (1975) reported that formation of coccoid cells was correlated with a reduction in glycan chain length and a decrease in the frequency of peptide crosslinking. The general chemical composition of peptidoglycan layer from a spiral shaped Vibrio sp. NCTC 4716 was similar to that from the coccoid shaped cell. However, the peptidoglycan from the coccoid cells was less cross–linked than that from the spiral shaped ones. Also, the peptidoglycan of the coccoid cell was more resistant to digestion by agents such as EDTA and lysozyme, which indicated that portions of the peptidoglycan that are sensitive to an N–acetylmuramidase had been lost during morphological transformation. The helical shape of bacterial cells can appear in cells that are usually not helical. Bacillus subtilis, a rod–shaped bacterium can grow in the form of a helix when cell division is suppressed, cells are cultivated in rich media at low densities, and when there is a deficiency in some autolytic enzymes. The helical growth is believed to be caused by the insertion of wall polymers in the cell wall, which introduces stress into the cell surface leading to a shape deformation

PAGE 110

92 (Mendelson, 1982). The formation of these microfibers were later found to be inhibited by penicillin G, indicating that the peptidoglycan was involved in the helical shape of B. subtilis (Zaritsky and Mendelson, 1984). Cooper (2001) proposed that the spiral shape of helical bacteria, such as Vibrio cholerae, is not the result of an asymmetrical positioning of the peptidoglycan; rather, the bacterium originally grows in a helical shape. According to the researcher, the peptidoglycan is synthesized in a continuous and permanent helical shape similar to a telephone cord. If the telephone cord is cut in small pieces, each resulting part will have a short comma or spiral shape as observed with Vibrio and Campylobacter cells. Therefore, synthesis of the peptidoglycan in helical bacteria is different than in rod–shaped bacteria. Koch (2000) published an excellent review on bacterial cell wall and peptidoglycan synthesis. The review explains what is known about cell wall growth and division, including the role of physics in cell wall formation, and growth patterns of the peptidoglycan for rod– and coccus–shaped bacteria. Unfortunately, there are no explanations for the spiral growth of helical bacteria and how the geometry of the peptidoglycan differs from the rods and cocci organisms.

PAGE 111

CHAPTER 3 ANTIMICROBIAL PROPERTIES OF ALLYL ISOTHIOCYANATE The concern about pathogens in our food system has increased over the past years, especially after an increased number of outbreaks involving meat and ready–to–eat products. Continuous advances worldwide in food manufacturing, production and distribution have enabled the food industry to expand and supply high quality products to consumers on a year–round basis. This luxury however, comes with a price: a higher incidence of opportunistic pathogens seemed to have found their way through the cracks of the system, finding optimal niches for their survival and growth, resulting in increased numbers of foodborne illnesses in humans. There are several ways these opportunistic pathogens can be eliminated from our food supply. The use of hurdles assures that in specific cases, pathogenic and spoilage organisms are safely destroyed or removed from the food before reaching the customer. The hurdles in this case, can be anything from physical interference to the use of chemical disinfectants and naturally derived food additives with antimicrobial properties. Allyl isothiocyanate (AIT) is one naturally derived antimicrobial that has been widely studied against both spoilage and pathogenic organisms and has proven to be successful in inhibiting their growth (Walker et al., 1937; Carter et al., 1963; Isshiki et al., 1992; Tsunoda, 93

PAGE 112

94 1994; Brabban and Edwards, 1995; Delaquis and Mazza, 1995; Shofran et al., 1998; Ward et al., 1998; Cherry, 1999; Delaquis et al., 1999; Kim et al., 2002). There is considerable interest in the application of gaseous AIT in packaged foods due to its potent antimicrobial activity (Delaquis and Mazza, 1995; Shofran et al., 1998; Kim et al., 2002). The antimicrobial properties of gaseous AIT seems to be much higher than the aqueous AIT due to the poor solubility and decomposition of the compound (Delaquis and Mazza, 1995; Kim et al., 2002). To this date, there is no literature available on the effect of AIT vapors on shelf life and numbers of Campylobacter spp. in poultry products. The objective of this study was to determine the potential antimicrobial activity of allyl isothiocyanate against psychrotrophic organisms and Campylobacter spp. in poultry products. The potential importance of AIT to the poultry industry and its applicability and feasibility was considered. Material and Methods Selective Culture Media and Agar Unless otherwise specified, the methodology used in this study for storing and culturing C. jejuni, was similar to that described in the Campylobacter chapter of the Bacteriological Analytical Manual (Food and Drug Administration, 1998). Plates used to culture Campylobacter spp. consisted of Jan–Abeyta–Hunt medium supplemented with Bolton’s Supplement (Oxoid, Inc., Nepean, Ontario, Canada, K2G 1E8, cat. No. SR0183E). Plates were manually pre–poured at the beginning of each trial and stored in the dark at 4 C until needed. Psychrotrophic

PAGE 113

95 organisms were plated onto Petrifilm Aerobic count Plate (3M, St. Paul, MN, 55144). Campylobacter Enrichment Broth (Med–Ox Diagnostics Inc., Ogdensburg, NY, 13669, cat. No. AM7526) test tubes and Jan–Abeyta–Hunt plates were incubated in a microaerophilic environment, i.e., 5% oxygen (O 2 ), 10% carbon dioxide (CO 2 ) and 85% nitrogen (N 2 ), using a gassed jar system. Two 3.5 L Campy Gas Generating Gas Pak envelops (Oxoid, Inc., Nepean, Ontario, Canada, K2G 1E8, cat. No. CN0035A) were added to each 7 L gas jar and one 2.5 L Campy Gas Generating Gas Pak (Oxoid, Inc., Nepean, Ontario, Canada, K2G 1E8, cat. No. CN0025A) was added to each small 2.5 L gas jar. Unused jars were cleaned and stored opened with the screw clamp lids stored separately. Bacterial Culture A freeze–dried culture of Campylobacter jejuni subsp. jejuni (ATCC 33560) was purchased from ATCC (American Tissue Culture Collection, Manassas, Virginia, 20110). Upon receiving, culture was propagated according to ATCC’s instructions. The freeze–dried pellet was hydrated in 6 mL of Fluid Thioglycollate (Oxoid, Inc., Nepean, Ontario, Canada, K2G 1E8, cat. No. CM0391B). Several drops of the suspension were used to inoculate slants and plates of Trypticase Soy Agar (Oxoid, Inc., Nepean, Ontario, Canada, K2G 1E8, cat. No. CM0129B) supplemented with 5% lysed horse blood. Test tube, slants and plates were incubated at 37 C, in anaerobe jars fitted with Campy–Paks, for 48–72 hours. Identification of Campylobacter colonies was based on colony

PAGE 114

96 morphology, wet mount, and catalase and oxidase tests. Typical colonies were stored on ceramic beads (Protect Beads, Med–Ox Diagnostics Inc., Ogdensburg, NY, 13669, cat. No. STCTS70A) at C until needed. A few days before needed, several beads were thawed and plated on pre–poured antibiotic–free Abeyta–Hunt–Bark Agar plates. Plates were incubated at 42 C for 48 hours under microaerophilic conditions, i.e., 5% O 2 , 10% CO 2 and 85% N 2 . At the first day of the study, a loopfull of the C. jejuni culture was suspended in 9 ml of sterile Phosphate Buffer (pH 7.2) and let set for 5 minutes. Allyl Isothiocyanate The allyl isothiocyanate (95% purity) used in this study was purchased from Sigma–Aldrich Chemical Company, Inc. (St. Louis, MO, 63118, cat. No. 377430–100g). The chemical was stored refrigerated, in a closed container protected from direct light throughout the length of the study. Pure AIT (95% purity) used throughout the study, unless otherwise stated. When lower concentrations of AIT were needed, the isothiocyanate was diluted with commercial grade corn oil. Allyl isothiocyanate concentrations of 0.01%, 0.025%, 0.05%, 0.075%, 0.1%, 0.15%, and 0.2% and 95% were obtained through dilutions when necessary. Assays for Dose–Related Bacteriostatic Activity of Allyl Isothiocyanate on Fresh Poultry Different approaches were investigated to determine the best assay as well as the most efficient dose–related bacteriostatic activity of allyl isothiocyanate on

PAGE 115

97 fresh poultry. The assays tested the dose of AIT needed to achieve the desired antimicrobial effect in poultry products as well as the best mode of application of the compound. Antimicrobial sensitivity tests were also performed. Antimicrobial sensitivity test assay Several 0.635 cm blank paper discs used for antimicrobial sensitivity test procedures (Beckton–Dickinson Microbiology Systems, Sparks, MD, 21152) were impregnated with pure AIT (95% purity) or with a mixture of 0.01%, 0.025%, 0.05%, 0.075%, 0.1%, 0.15%, and 0.2% AIT–Corn oil. A sterile cotton swab that had been dipped into the solution containing a fresh culture of C. jejuni subsp. jejuni was used to spread the entire surface of Abeyta–Hunt–Bark Agar plates. The plates were allowed to dry for 5 minutes. Using sterile forceps, the AIT–, and AIT–Corn oil–impregnated discs were randomly placed over the agar surface, four discs per plate, at equal distance from each. Plates were incubated at 42 C for 48 hours under microaerophilic conditions. Growth and any zone of inhibition were observed after incubation. After 48 hours of incubation, growth was observed in all plates except those that had discs impregnated with pure AIT (95% purity). On these plates, growth inhibition of C. jejuni was complete, i.e., no colonies were seen in any part of the plate. Complete inhibition was due to volatilization of AIT, which was confined inside the Petri dish and could not dissipate. All plates that contained the 95% AIT impregnated discs were kept open for a minute under the laminar flow hood until all the AIT vapors dissipated. Plates were then incubated for an

PAGE 116

98 additional 48 hours under microaerophilic conditions. Interestingly, growth of C. jejuni was observed in all plates that were incubated for a second time without the presence of AIT vapors, which had been released from the Petri dish. The antimicrobial sensitivity test confirmed the hypothesis that AIT vapor can be used as an effective bacteriostat against C. jejuni. Chicken drumstick assay Chicken drumsticks (bone in and skin on) were purchased from a local retail store. Each drumstick was placed in a sterile stomacher bag and inoculated with 1 mL of C. jejuni bacterial suspension. Each inoculated drumstick was placed in an aluminum foil lined plastic container along with a disposable Petri dish containing 10 mL of pure AIT (95% purity). Negative controls consisted of a chicken drumstick inoculated with C. jejuni and incubated with a Petri dish containing corn oil. Each plastic container was placed inside a food–grade, vapor–barrier, metalized poly pouche (12” x 18”, 2 MIL, cat. No 183, Associated Bag Company, Milwaukee, WI, 53207) and heat–sealed to insure that the AIT vapors would not dissipate. A positive control was obtained by inoculating chicken drumsticks with 0.1 mL of the C. jejuni bacterial suspension and incubated for 48 hours to check for bacterial viability. Each metalized pouche was opened at a different time after sealing: 0 (control, i.e., no exposure to AIT vapors), 1, 5, 10, 15, and 30 minutes. Exposure time constituted different treatments. Once the drumsticks had been exposed to the AIT vapors for the pre–determined length of time, pouches were opened and

PAGE 117

99 each drumstick was placed inside a sterile stomacher bag along with 18 ml of sterile 0.1% peptone and lightly homogenized by hand. A sample of each solution was used to streak Abeyta–Hunt–Bark Agar plates. Plates were allowed to dry for 5 minutes, and then incubated at 42 C for 48 hours under microaerophilic conditions. Plates were checked for growth at 24 and 48 hours of incubation. All treatments and plating were performed in duplicate. No growth was observed in any plate after 48 hours of incubation indicating that the amount of AIT used should be lowered. Other modifications that needed to be made in the protocol were observed: pure AIT should not be placed in plastic containers because it might melt them—the disposable Petri dishes melted upon prolonged contact with AIT; and pure AIT should not enter in contact with the chicken meat because it may denature the muscle proteins resulting in a product with a mushy consistency. Three additional tests were performed using a similar protocol as described above, but with the exposure of chicken drumsticks to different amounts of AIT vapors. First, chicken drumsticks were exposed to 5 ml of AIT vapors for 0, 1, 5, 10, 15, and 30 minutes. Because total inhibition of C. jejuni was achieved with 5 ml of AIT vapors, the amount of vaporized AIT was reduced to 1 ml and later only 1 drop while keeping the same exposure times. Recovery of C. jejuni was possible when chicken drumsticks were exposed for 1 minute to vapors resulting from 1 drop of AIT. Based on these results, a last trial was conducted using different exposure times and amounts of AIT. Chicken

PAGE 118

100 drumsticks heavily inoculated with C. jejuni were exposed for 0, 1, 2, and 3 minutes to vapors of AIT resulting from 10, 50, 250, 500, and 1000 L of AIT. Bacterial growth was observed for all treatments after 48 hours of incubation, probably due to the heavy inoculation with C. jejuni and the limited amount of AIT vapors. The levels of contamination in retail chickens vary according to month of the year and to sanitary conditions of the processing plant that processed the chickens but on average carcasses are contaminated with low counts of C. jejuni. Cason et al. (1997) reported that levels of contamination vary from 10 2 to 10 4 cfu of Campylobacter spp. per carcass. A final antimicrobial test was performed with chicken drumsticks inoculated with 10 3 cfu/ml of C. jejuni and exposed for 0, 1, 2, and 3 minutes to vapors derived from 0, 100, 300 and 600 L of 100 g of poultry product. The results obtained from this test were used to define the level of AIT needed to inhibit C. jejuni from poultry products without exposing the food to excessive amounts of isothiocyanates. Chicken Breast Study Three hundred and fifty pounds of fresh skinless, boneless chicken breasts were purchased from Gold Kist Poultry Processing Plant (Live Oak, Fl, P.O. Drawer 1000). Chicken breasts were randomly selected immediately after processing and prior to packaging for distribution. Breasts were placed in clean plastic bags (Cryovac Sealed Air Corporation, South Carolina) and transported in coolers containing crushed ice to the University of Florida’s Meat Science

PAGE 119

101 Microbiology Laboratory. Upon arrival, chicken breasts were placed in a walk–in refrigerator, and subjected to antimicrobial treatments on the same day of purchase. Treatment of chicken breasts Upon arrival at the University of Florida Meat Science Microbiology Laboratory, all chicken breasts were carefully examined for remaining skin or bone fragments, separated into two breast halves and individually weighed. One hundred and sixty eight Fresh–R–Pax trays (Maxwell Chase Technologies – LLC, P.O. Box 1676, Douglasville, GA, 30133) were randomly labeled with a treatment combination (Figure 5). The bottom of the Fresh–R–Pax tray is divided into 12 chambers, each containing a proprietary absorbent gel material manufactured with GRAS (generally regarded as safe) components that gelatinizes when in contact with any purge released by the product (Figure 6). A group of three skinless, boneless chicken breast halves of known weight was transferred to a labeled Fresh–R–Pax tray (Figure 7). Chicken breasts were not inoculated with any bacterial culture and the initial bacterial load in the poultry was assumed to be representative of the contamination level found in retail products. According to each treatment assignment, 0, 100, 300 or 600 L pure AIT (95% purity)/100 g chicken breast was pipetted into a corner chamber at the bottom of the tray. Trays were placed inside a 50 CL polyester high gas barrier clear pouche (Serv–Pack Corp., Hollywood, FL, 33023) and heat–sealed to insure complete exposure of chicken breasts to vaporized AIT (Figure 8).

PAGE 120

102 Figure 5 .Fresh–R–Pax tray Figure 6 Proprietary absorbent gel material in one of the chambers at the bottom of tray Figure 7 Tray containing three chicken breast halves Figure 8 Trays containing three chicken breast halves prior to receiving AIT treatment

PAGE 121

103 Figure 9 Display retail case Figure 10 AIT treated chicken halves stored in the retail display case Sample storage All 168 sealed Fresh–R–Pax trays containing three chicken breast halves and 0, 100, 300 or 600 L AIT/100 g chicken breast were transferred to two retail display cases at the University of Florida Meat Science Store (Figure 9). Samples were labeled as part of a research study and kept at 4 C in the retail display cases for a maximum period of 7 days (Figure 10). Due to the large number of samples and small size of the display cases, trays were stacked in two layers. To insure that all trays were exposed to the same amount of light and cold air, all trays were randomly rearranged in the display cases once a day throughout the length of the study. Immediately after being placed in the retail cases, all trays assigned to the 0–day storage time treatment were transferred back to the microbiology laboratory for analysis. At each subsequent storage day (3, 5 and 7 days), six trays from each level of AIT were removed from the display cases and

PAGE 122

104 transferred to the laboratory. Samples were tested for presence of Campylobacter spp., psychrotrophic organisms, pH of the poultry homogenate, and color of the surface muscle. A subjective odor analysis of the retail cases during storage of samples was conducted daily to assure that the vapors were contained inside the packaging and not being released into the environment. Despite the fact that there were enough treated samples for the study to continue for up to 14 days of storage, all analyses were terminated after 7 days. The reason for that was the advanced state of putridness observed in samples that were not exposed to any level of AIT and the unsatisfactory general appearance and color of the chicken breasts that were stored while being constantly exposed to the AIT vapors. Continuation of the study would not have contributed any relevant information to the scientific community. Statistical analyses One hundred sixty eight trays containing skinless, boneless chicken breast halves were subjected to 2 experimental factors, each containing four levels: Factor A, AIT concentration (0, 100, 300 and 600 L of AIT/100 g of chicken breast); and Factor B, storage time of chicken breasts prior to analysis (0, 3, 5 and 7 days of storage). The combination of 4 levels of Factor A and 4 levels of Factor B allowed for 16 different combinations of the factors (Table 1). Exposure to 0 L of AIT was used as control treatment. Each treatment combination was replicated 6 times.

PAGE 123

105 Table 1 Combinations of AIT concentration and storage day Level of Pure AIT (L/100 g Chicken Breast Halves) Factor 0 100 300 600 0 0 AIT, 0 day 100 AIT, 0 day 300 AIT, 0 day 600 AIT, 0 day 3 0 AIT, 3 days 100 AIT, 3 days 300 AIT, 3 days 600 AIT, 3 days 5 0 AIT, 5 days 100 AIT, 5 days 300 AIT, 5 days 600 AIT, 5 days Storage Time (days) 7 0 AIT, 7 days 100 AIT, 7 days 300 AIT, 7 days 600 AIT, 7 days Data were subjected to analysis of variance for a completely randomized design (Snedecor and Cochran, 1989) with a 4x4 factorial arrangement of treatments (level of AIT and storage day) (Table 2). The SAS software program (SAS Institute, 1990) was used to determine significant differences among treatments. The procedure PROC GLM from SAS was used to perform the statistical analysis. When significant differences were observed, treatments were analyzed using linear, quadratic and cubic regressions. A tray containing 3 skinless, boneless, chicken breast halves was considered an experimental unity. The linear model for the study was: Y ijk = + i + j + () ij + E (ij)k Where: Y ijk = variable measured on a tray of three chicken breast halves exposed to i amount of AIT stored for j days = overall mean

PAGE 124

106 i = fixed effect of the i th level of AIT on microbial count, pH and color of chicken breast i = 0, 100, 300, 600 j = fixed effect of the j th storage day on microbial count, pH and color of chicken breast j = 0, 3, 5, 7 () ij = interaction of i th level of AIT and j th storage day on microbial count, pH and color of chicken breast. E (ij)k = random effect of all unspecified variables peculiar to the k th experimental unit assigned to the treatment combination involving the i th level of AIT and the j th storage day ij NID (0, b 2 ) Table 2 Sources of variation and degrees of freedom used in the analysis of variance (ANOVA) Source Degrees of freedom Model 15 Level of AIT 3 Storage time 3 Level of AIT * Storage time 9 Error 80 Total 95 Sample Analyses Treated samples were analyzed for Campylobacter jejuni, psychrotrophic organisms, pH, and objective color.

PAGE 125

107 Campylobacter jejuni analysis Presence or absence of C. jejuni on samples was conducted using the methodology for isolating Campylobacter species from food and water described in the Campylobacter chapter of the United States Food and Drug Administration Bacteriological Analytical Manual (Hunt and Abeyta, 1995). At days 0, 3, 5, and 7, 24 trays containing three skinless, boneless chicken breast halves that had been assigned to each level of AIT treatment were removed from the display retail case and transferred to the Microbiology Laboratory. The 50 CL polyester high gas barrier clear pouche that surrounded each tray was carefully opened and all three chicken halves were aseptically transferred with sterile metal tongs to a sterile stomacher bag along with 50 mL of 0.1% peptone water. Bags were gently massaged by hand for 2 minutes to loosen any attached microorganisms. One milliliter of bacterial rinse from the stomacher bag was transferred to a 9–mL test tube containing 0.1% peptone water and serial diluted to 10 . The extent of the serial dilution was adjusted throughout the trial as needed; i.e., the number of dilutions decreased depending on the bacterial growth obtained at the previous sampling day. From the serial dilution test tubes, 0.1 mL was plated out, in duplicates, onto Abeyta–Hunt–Bark plates. Plates were closed with laboratory masking tape and placed inside a metalized poly pouche. The metalized poly pouche was flushed three times with a microaerophilic gas mixture containing 5% O 2 , 10% CO 2 and 85% N 2 and heat–sealed. Pouche was

PAGE 126

108 incubated at 42 C for 48 hours. After 48 hours of incubation, plates containing typical Campylobacter colonies were recorded as positive samples. Due to a large number of colonies that spread over the agar instead of forming isolated colonies, samples were considered either positive or negative for the presence of Campylobacter. Colonies were confirmed as C. jejuni as previously described. Spoilage microbial analysis Psychrotrophic counts were conducted using Petrifilm Aerobic Count Plates (3M Microbiology Products, St. Paul, MN, 55144) method for total plate count. Each Petrifilm was labeled with the appropriated sample number and dilution. One milliliter of each serial dilution prepared for Campylobacter was plated, in duplicate, onto Aerobic Count Plate Petrifilm. Petrifilms were incubated aerobically at 25 C for 5 days. The actual serial dilutions plated were adjusted throughout the trial as needed. Counts were reported as colony forming units of psychrotrophic organisms per gram (cfu/g) of sample. pH analysis The pH of each sample was obtained from the initial bacterial rinse prepared for the microbiological assay (Kempton and Bobier, 1970). After the microbial analysis had been performed, a small aliquot of the bacterial rinse from the stomacher bag containing the chicken halves was poured into a clean beaker and the pH of the solution was measured with a portable pH meter (Oyster, Extech Instruments Corporation, Waltham, MA, 02451). Triplicate measurements were recorded per sample.

PAGE 127

109 Colorimetric analysis Color measurements of experimental chicken breasts were obtained with a portable colorimeter (Minolta Chroma Meter CR100, Minolta, Osaka 541 Japan). The color of the chicken breast was measured at three different random positions at days 0, 3, 5, and 7. Measurements were obtained by placing the optical reader over the 50 CL polyester high gas barrier clear pouche and pressing it over each chicken breast. The colorimeter was calibrated as recommended by the manufacturer prior to each sampling period. Calibration was performed with a piece the 50 CL pouche material covering the optical reader. The optical reader was covered during calibration because color was measured on packaged chicken breasts. Samples were evaluated for L* (degree of luminescence or lightness), a* (degree of redness), and b* (degree of yellowness) values. CIELab (L* a* b*) was developed by the Commission Internationale d’Eclairage (CIE) in 1976 as an international standard for color measurement (Papadakis et al., 2000). The degree of lightness (L* value) varies from zero, which indicates black, to 100, which indicates absolute white. The higher the L* value, the lighter the sample. The degree of redness (a*) and of yellowness (b*) vary from –120 to +120. Positive a* values indicate higher degree of red while negative values represent higher degree of green. Positive b* values indicate high degree of yellow while negative values indicate high degree of blue.

PAGE 128

110 Results and Discussion Initial experiments were conducted to determine the methodology and levels of AIT required to inhibit Campylobacter jejuni in poultry products, especially premium parts such as skinless and boneless chicken breasts and drumsticks. Chicken drumsticks, instead of whole carcasses were selected for the AIT dose–related bacteriostatic assays due to practical reasons. Based on prior experimentation in this study and work published in the literature review (Delaquis and Mazza, 1995), it was determined that the antimicrobial properties of AIT were superior when the isothiocyanate was vaporized instead of in a liquid form. Thus, poultry products were exposed to vaporized AIT instead of being sprayed, misted or vacuum tumbled with the isothiocyanate. Furthermore, AIT is a very viscous and toxic oil. The handling of AIT requires extensive safety training and precautions, which limits its use. The spraying/misting and vacuum tumbling methods requires that large amounts of AIT be handled and applied to the chicken. Previous experience with AIT showed that prolonged contact of the chemical with poultry flesh affected the physical characteristics of the poultry product; the flesh loses its original consistency, becoming mushy and disintegrating easily. Moreover, pure AIT melts plastic upon prolonged contact, further preventing any attempts of treating products through spraying/misting and vacuum tumbling. This is a major concern when trying to apply the

PAGE 129

111 knowledge obtained in a laboratory setting into a commercial facility where several machines have parts that are made of plastic. In addition to the practical implications of directly exposing or submerging poultry to a liquid form of AIT, the presence of chemical residues in food products when in such close contact with each other must be accounted for. Allyl isothiocyanate is a fat–soluble compound and there is a possibility that poultry products, especially poultry skin, could retain large amounts of residue if the chemical is in direct contact with the product. The use of vaporized AIT is less likely to leave residues on the product, does not change the consistency of the muscle and does not require handling of large amounts of AIT during the antimicrobial treatment application. Microbial Analyses In this study, the samples used for the microbial analysis were not challenged with C. jejuni nor psychrotrophic organisms. This study focused on the actual microflora found on skinless, boneless chicken breasts sold to the consumer at retail level. Although unusual in the food microbiology field, unchallenged studies provide information that is not obtainable when samples are purposely inoculated with the desired organism. One important aspect when dealing with Campylobacter spp. is the fact that they are usually outgrown by the background microflora (Hunt and Abeyta, 1995).

PAGE 130

112 Psychrotrophic organism analysis Microbial analyses of psychrotrophic organisms were performed to determine if exposure of chicken breasts to vapors of allyl isothiocyanate (AIT) would significantly reduce the number of organisms present, thus increasing the shelf life of the product. Psychrotrophic counts are widely used to assess the microbial load of fresh poultry and meat products (Jay, 2002). The initial microbial load helps to establish overall product quality, handling and storage history of the food. It also gives information on the product’s shelf life. Bacterial growth follows a very characteristic pattern, which consists of four distinct phases: lag phase, logarithmic or exponential phase, stationary phase and death phase (Forrest et al., 1975). The lag phase is very important for the food industry because it represents the bacterial count present in a certain environment that allows for bacterial survival but not for bacterial growth (Forrest et al., 1975). As soon as the environmental conditions change and become more suitable for the bacteria, multiplication and growth take place marking the beginning of the exponential phase. Any measure taken by the food industry to extend the lag phase while retarding the beginning of the exponential phase is relevant in terms of food safety (control of pathogens) and shelf life (control of spoilage organisms) of the product. In the preliminary studies, it was observed that AIT exerted a bacteriostatic effect, i.e., inhibited bacterial growth, in both Campylobacter and spoilage organisms instead of having a bactericidal effect, i.e., killed bacteria.

PAGE 131

113 While bactericides shift the bacterial growth to the right, straight to the death phase, bacteriostats extend the lag phase. Similar results were observed by Delaquis and Sholberg (1997), in which they reported that S. typhimurium, L. monocytogenes Scott A, and E. coli O157:H7 were inhibited when the agar surface on which they were being cultured was exposed to 1 ppm of AIT/L of gas over the surface. Shofran et al. (1998) observed that 50 to 1,000 ppm of AIT was the minimum inhibitory concentration for bacteria species cultured in medium broth. The design of the experiment cited herein was a 4 x 4 factorial arrangement (effect of AIT levels and effect of storage time on chicken breasts), which gives 16 treatment combinations. Thus, the same treatment means can be analyzed in two ways: based on storage time and on level of AIT. One comparison tests the effect of storage time on the bacterial population while the other tests the effect of different levels of AIT. Table 3 shows the analysis of variance for total psychrotrophic counts recorded in chicken breasts exposed to vapors from 0, 100, 300 and 600 L of AIT per 100 g of meat for up to 7 days of storage. The results indicated that the level of AIT, the number of days that meat was exposed to AIT vapors (days of storage) and the interaction of AIT levels and days of storage had a significant impact on the spoilage microflora of the chicken breast.

PAGE 132

114 Table 3 Analysis of variance for total psychrotrophic counts observed in skinless, boneless chicken breasts stored at 4C and exposed to allyl isothiocyanate vapors (AIT) for 7 days Source df SS Pr > F Model Level of AIT 3 16.8917 <0.0001 Days of storage 3 133.9910 <0.0001 Level of AIT * Days of storage 9 52.7078 <0.0001 Error 64 42.2401 Total 79 245.8308 Despite the fact that an initial 24 samples of chicken breasts were analyzed for psychrotrophic organisms shortly after the experimental setup, differences in initial average bacterial count was observed. Table 4 shows the treatment means for total psychrotrophic organism counts in chicken breast meat exposed to AIT vapors and stored for up to 7 days at 4 C. The initial psychrotrophic count for the control chicken breast meat was 3.53 log 10 cfu/g, which was significantly higher (P<0.05) than that observed in the chicken breasts exposed to 600 L of AIT (2.12 log 10 cfu/g) but not different than those exposed to 100 L or 300 L of AIT (2.93 and 2.85 log 10 cfu/g, respectively). Furthermore, there were no differences in psychrotrophic count in chicken breasts exposed to 100 L, 300 L, and 600 L of AIT (2.93, 2.85, and 2.12 log 10 cfu/g, respectively). The initial microbial count for psychrotrophic organisms observed was slightly higher than those published in the literature. Although variable, contamination of chilled poultry carcasses includes 10 2 0 5 cfu/g of mesophilic

PAGE 133

115 aerobes, 10 1 0 3 cfu/g of psychrotrophs, 10 3 4 cfu/g of Enterobaceteriaceae, 10 1 –10 5 cfu/g of E. coli, 10 3 cfu/g of S. aureus and <10 2 cfu/g of C. perfringens (Campbell et al., 1983; Grau, 1986). Two main reasons may explain the slightly higher initial contamination levels observed in this study. Table 4 Treatment means for total psychrotrophic counts in skinless, boneless chicken breasts stored at 4C and exposed to allyl isothiocyanate vapors (AIT) for 7 days * Data shown are average means of six replicate trays containing 3 halved chicken breasts Days of Storage (Log 10 cfu/g) * Treatments 0 3 5 7 0 L AIT/100 g sample 3.53 a w 4.51 b w 5.18 b w 7.94 c w 100 L AIT/100 g sample 2.93 a wx 2.92 a x 2.94 a x 3.70 a x 300 L AIT/100 g sample 2.85 a wx 2.87 ab x 1.93 ac xy 1.64 c y 600 L AIT/100 g sample 2.12 ab x 1.18 ab y 0.66 a y 2.54 b xy a–c Values within the same row with the same superscript are not significantly different (p>0.05) w–y Values within the same column with the same subscript are not significantly different (p>0.05) First, chicken breasts were used instead of chicken carcasses. The extra handling of the chicken product during processing, cutting and deboning increases the potential for contamination. Second, in spite of the fact that chicken breasts were transported to the microbiology laboratory on ice, the chicken was kept at room temperature for most of the duration of experimental setup and treatment application, which took approximately six hours. According to the 1999 Food Code, potentially hazardous food must be kept at a temperature below 5 C or above 60 C, at all times in order to avoid microbial growth (U. S. Department of Health and Human Services, 1999).

PAGE 134

116 The difference observed in the bacterial count on Day 0 of storage between control samples and those exposed to 600 L of AIT is due to the strong effect that AIT exerted on the microbial flora. One of the requirements for a well–designed experiment is to assure that all samples are treated equally. For this study alone, a total of 168 trays (samples) containing three halved chicken breasts were prepared. Samples that were removed from the display case on the first day of the study (storage day 0) were exposed to the vapors of AIT for about 6 hours before they were analyzed for microbial counts. Allyl isothiocyanate is a very strong antimicrobial and this 6 hour–exposure time was sufficient to significantly reduce the population of psychrotrophic organisms present in the chicken breast. Based on the results observed in the preliminary chicken drumstick study, in which poultry was exposed to the same amount of AIT for a period of 0, 1, 2 or 3 minutes, a 1.4 log reduction in the total bacterial count with the maximum level of AIT was not surprising even on Day 0 of storage. As storage time increased from zero to seven days, the difference observed in total psychrotrophic counts among levels of AIT increased slightly. After three days of storage, the psychrotrophic count for the control samples was significantly higher (P<0.05) than those seem in samples exposed to 100, 300 and 600 L of AIT (4.51 vs. 2.92, 2.87, and 1.18 log 10 cfu/g, respectively). Exposure to 100 or 300 L of AIT

PAGE 135

117 012345678901234567Storage DayPsychrotrophic counts (log10 CFU/g) 0 AIT 100 AIT 300 AIT 600 AIT Figure 11 Psychrotrophic counts of halved chicken breasts exposed to allyl isothiocyanate vapors (AIT) and stored for 7 days at 4C

PAGE 136

118 reduced the bacterial count by approximately 1.6 logs while exposure to 600 L of AIT further reduced the bacterial numbers by 3.3 logs. As storage time progressed, the differences in total bacterial numbers continued to increase. Figure 11 illustrates the differences observed in psychrotrophic counts over time. At day 5 of storage, control samples had a higher bacterial population as compared to samples that were exposed to any level of AIT. Exposure to 600 L of AIT reduced total bacterial numbers by 4.5 logs as compared to no exposure (0.66 vs. 5.18 log 10 cfu/g). Levels of AIT as low as 100L of AIT reduced bacterial population by 2.2 logs (2.94 vs. 5.18 log 10 cfu/g) while 300 L caused a 3.2 log reduction in bacterial counts (1.93 vs. 5.18 log 10 cfu/g). Even though there was a significant reduction in bacterial population when chicken breasts were exposed to vapors of AIT, there was no significant difference in bacterial numbers if 100 L or 300 L were used; both levels significantly reduced bacterial population by the same amount. The differences in the number of psychrotrophic organisms present in chicken breasts that were exposed to AIT increased even more after seven days of storage. No exposure to AIT allowed for maximum bacterial growth, which accounted for 7.94 log 10 cfu of psychrotrophic organisms per gram of chickenbreast. Such population was 4.2 logs higher than that found in chicken breasts that were exposed to 100 L of AIT (7.94 vs. 3.70 log 10 cfu/g) and 6.3 logs higher than those exposed to 300 L of AIT (7.94 vs. 1.64 log 10 cfu/g). Exposure to 600 L significantly reduced bacterial numbers in chicken breast after 7 days

PAGE 137

119 of storage by 5.4 logs (7.94 vs. 2.54 log 10 cfu/g). Surprisingly, a 1–fold increase in the level of AIT (from 300 L to 600 L) had no significant effect on bacterial population; actually there was a difference in the numerical value of bacterial numbers but not a difference in statistical value. Similar reduction in bacterial population was observed in inoculated roast beef slices stored with horseradish distillate for 7 days (Ward et al., 1998). According to the authors, horseradish distillate is made of 90% allyl isothiocyanate, 9% 2–phenethyl isothiocyanate and traces amounts allyl thiocyanates. Increasing levels of horseradish distillate concentrations caused a steady decline in the number of Listeria monocytogenes, Staphylococcus aureus, Escherichia coli, Salmonella typhimurium, and Serratia grimesii. Figure 11 clearly illustrates the differences observed in total bacterial counts observed over time according to the exposure level of AIT. The changes seen in bacterial numbers as the storage time increased was used to predict regression models that best describe the behavior of the bacterial population (Table 5). Based on the regression models presented in Table 5, a series of prediction regression equations were calculated (Table 6). The regression models presented in Table 5 must be analyzed in conjunction with Figure 11 because, for this experimental condition (set of data), the model that best fits the treatment means does not necessarily best represent the behavior of the bacterial population. Despite the fact that the experimental conditions observed represents as best as it can the behavior of the population of

PAGE 138

120 inference, one must not forget that results obtained with any study are laboratorial results, which may or may not correlate well with reality. This is due to the fact that the results obtained did not account for the various factors that are usually found in–situ, such as the large variety of microorganisms found in foods (and its interactions); presence of bacteriophages; spatially heterogeneous growth of organisms throughout the food; degradation products that end up being part of the food ecosystem; presence of biofilms; excessive handling of the product at the retail store; packaging at the place of sale; and so on. As stated by Fleet (1999), much of our knowledge in food microbiology is derived from studies in the laboratory, which lack the diversity of physiological states of the microbial cells found in in–situ environments. The in–situ environment in this case refers to the attachment or entrapment of microbial cells with a solid substrate, in which the cells are immobilized and localized in high densities (Fleet, 1999). Also, cell signaling (quorum sensing) usually plays an important role in in–situ environments. Furthermore, laboratorial studies focus on a limited number of stress factors—usually only one or two stressors are tested. In the case of in–situ food environments, very rarely there is only one single factor involved. Microbial cells may be simultaneously exposed to a combination of factors, such as low pH, low temperature, high NaCl concentration, acetic acid, background bacteria, high levels of by–products, etc (Fleet, 1999). The outcome of the interactions might be

PAGE 139

121 different than those found in the laboratory. In addition, the fact that every ecosystem is dynamic, therefore it is in a constant changing state, plays a role on the results obtained. The severity of the stress inflicted on the bacteria can change with time as will the physiological state of the microbial cells (Fleet, 1999). For instance, cells in the stationary phase of growth are usually more resistant to stresses than cells that are on the exponential phase, and cells in the stationary phase generally dominate in many foods (Fleet, 1999). One of the factors that highly influences how the results obtained with a sample population correlate with the population of inference is the number of replications analyzed. The number of replications used in this study was statistically adequate, however it was not sufficient to minimize the impact of off–values observed during the study. For example, the slight increase in bacterial numbers seen in chicken breasts exposed to 600 L of AIT after seven days of storage is more likely an artifact of the data than it is an accurate representation of reality. If an infinite number of samples were analyzed, probably bacterial numbers would have plateaued with time, instead of increasing. This rationale was used to select the significant quadratic or linear effects of a particular variable in this study instead of a significant cubic effect. Table 4 shows the effect of storage day on AIT treated and untreated chicken breasts. Chicken breasts that were not exposed to vapors of AIT had the highest increase in bacterial numbers over time. As expected, the increase in

PAGE 140

122 Table 5 P–values associated with linear, quadratic and cubic effects of allyl isothiocyanate vapors (AIT) for total psychrotrophic counts observed in skinless, boneless chicken breasts stored at 4 C for 7 days Days of Storage P–values 0 3 5 7 Model 1 Linear effect of AIT <0.0001 0.0001 <0.0001 <0.0001 Quadratic effect of AIT 0.6344 0.5695 0.0688 <0.0001 Cubic effect of AIT 0.0215* 0.0227* 0.1368 0.0129* Model 2 Linear effect of AIT <0.0001 0.0005 <0.0001 <0.0001 Quadratic effect of AIT 0.6787 0.6190 0.0785 <0.0001 Model 3 Linear effect of AIT <0.0001 0.0003 0.0001 — *Cubic effect of AIT does not reflect data properly; therefore the cubic effect was removed in Model 2 Table 6 Prediction regression equations for the linear and quadratic curves for total psychrotrophic counts observed in skinless, boneless chicken breasts exposed to allyl isothiocyanate vapors (AIT) for 7 days Days of Storage R 2 Regression Equation 0 0.61 Log 10 Psychrotrophic = 3.3819 – 0.0020 (AIT) 3 0.51 Log 10 Psychrotrophic = 4.0759 – 0.0047 (AIT) 5 0.57 Log 10 Psychrotrophic = 4.0759 – 0.0047 (AIT) 7 0.88 Log 10 Psychrotrophic = 7.2510 – 0.0325 (AIT) + 0.00004 (AIT) 2

PAGE 141

123 bacterial population was constant throughout the 7–day period, reaching a maximum 7.94 cfu/g. Such high microbial population after seven days of storage was expected because the chicken breasts were stored at 4 C. The average shelf life of muscle foods stored at 0 C is 2 weeks. However, if the storage temperature is above 0 C, shelf life is drastically reduced—for every 5 C increase in storage temperature the shelf life of the product is reduced by half (Sofos, 1994). After three days of storage, bacterial counts showed almost a 1–log increase as compared to the initial count and at day five counts increased by 1.6 logs (4.51, 3.53, and 5.18 log 10 cfu/g, respectively). Although the bacterial numbers were not significantly different after three and five days of storage, they were significantly higher than the initial total bacterial count. Bacterial population continued to increase as time progressed, as seen with the 2.7 logs increase in the number of psychrotrophic organisms in day 7 as compared to day 5 (7.94 vs. 5.18 log 10 cfu/g, respectively) and a 3.4 logs increase when compared to day 3 (7.94 vs. 4.51 log 10 cfu/g, respectively). The most dramatic difference was observed when the total number of psychrotrophic organisms after seven days of storage was compared to the initial count (7.94 vs. 3.53 log 10 cfu/g). Bacterial population increased by 4.4 logs throughout the duration of the study, reaching a population of 7.94 log 10 cfu/g, which generally means spoiled product (Brewer et al., 1995). A muscle food is considered spoiled when certain products of enzymatic metabolism, both of

PAGE 142

124 natural and microbial origin, make it unacceptable, offensive and unpalatable to the human senses (Sofos, 1994). According to Sofos (1994), Pseudomonas spp, Enterobacteriaceae, Brochothrix thermosphacta, lactic acid bacteria of the genera Lactobacillus and Leuconostoc, Aeromonas spp., and Shewanella putrefaciens predominate in spoiled muscle foods. In the case of poultry, the majority of spoilage organisms present in the product are Gram–negative, rod–shaped bacteria such as Pseudomonas and Shewanella spp. These two species of bacteria are implicated in the formation of off–odors and slime on the surface of the poultry as well as in the process of discoloration of the muscle. An increase in bacterial numbers as storage time increased was not seen in samples treated with AIT. When chicken breasts were exposed to 100 L of AIT no significant difference in numbers of psychrotrophic organisms were observed at days 3, 5 and 7 of storage as compared to the initial bacterial load (2.92, 2.94, and 3.70 vs. 2.93 log 10 cfu/g). The final bacterial load of chicken breasts was reduced by 1.2 logs as compared to the initial count (1.64 vs. 2.85 log 10 cfu/g) when the level of AIT increased to 300 L. Surprisingly, 600 L of AIT did not significantly reduce the numbers of psychrotrophic organisms at the end of seven days of storage. However, a significant reduction in the number of psychrotrophic organisms was observed after three and five days of storage as compared to the initial count (1.18, and 0.66, vs. 2.12 log 10 cfu/g).

PAGE 143

125 The fact that such effect occurred in the first five days of storage but not after seven days, leads to the conclusion that the samples were contaminated during microbial analysis. This is especially true if we take into consideration the reduction pattern seen in the bacterial population as AIT level was increased from 0 to 100 L, 100 L to 300 L throughout the 7–day period, and from 300 L to 600 L in the first 5 days of storage. Table 7 shows the linear, quadratic and cubic effects of storage day on the psychrotrophic organism population found on untreated or AIT treated chicken breasts. Table 8 presents the prediction regression equations, as well as the coefficient of correlation (R 2 ) for the best fitted model. The regression models described on Table 7 must be analyzed in conjunction with Figure 12. The effect of storage day on the microbial flora of chicken breasts was different according to the level of AIT used. When chicken breasts were not exposed to AIT (control samples), storage day had a quadratic effect in the microbial population. With the addition of 100 L of AIT to the trays containing chicken breasts, storage day had a linear effect on bacterial population. Similar linear effect of storage day was observed when 300 L of AIT was used. At maximum level of AIT (600 L0, storage day had an quadratic effect on the number of psychrotrophic organisms present in the chicken breast stored for up to seven days at 4 C.

PAGE 144

126 Table 7 P–values associated with linear, quadratic and cubic effects of days of storage on total psychrotrophic counts observed in skinless, boneless chicken breasts exposed to allyl isothiocyanate vapors (AIT) Levels of AIT (L AIT/100 g sample) P–values 0 100 300 600 Model 1 Linear effect of Day <0.0001 0.1507 0.0036 0.9200 Quadratic effect of Day 0.0196 0.2186 0.3307 0.0283 Cubic effect of Day 0.2362 0.5745 0.2316 0.2075 Model 2 Linear effect of Day <0.0001 0.1437 0.0038 0.9275 Quadratic effect of Day 0.0204 0.2107 0.3360 0.0300 Model 3 Linear effect of Day — 0.1488 0.0036 — Table 8 Prediction regression equations for the linear and quadratic curves for total psychrotrophic counts observed in skinless, boneless chicken breasts stored for 7 days while exposed to allyl isothiocyanate vapors (AIT) AIT (L AIT/100 g sample) R 2 Regression Equation 0 0.74 Log 10 Psychrotrophic = 3.6001 – 0.05120 (Day) + 0.08738 (Day) 2 100 — Log 10 Psychrotrophic = 2.32728 300 0.32 Log 10 Psychrotrophic = 3.04359 – 0.1910 (Day) 600 0.64 Log 10 Psychrotrophic = 2.2133 – 0.8354 (Day) + 0.1228 (Day) 2

PAGE 145

127 01234567890100200300400500600AIT LevelPsychrotrophic counts (log10 CFU/g) Day 0 Day 3 Day 5 Day 7 Figure 12 Psychrotrophic counts of halved chicken breasts stored at 4C for 7 days while being exposed to allyl isothiocyanate vapors (AIT)

PAGE 146

128 Overall, exposure to vapors of AIT significantly inhibited the growth of psychrotrophic organisms reducing its population on chicken breasts stored at 4 C for 7 days. The reduction in the number of psychrotrophs as compared to the control samples was higher as the storage day increased. Similar reduction in the number of total aerobes was observed by Delaquis et al. (1999). Addition of 20 L of essential horseradish oil/L inhibited the growth of most spoilage bacteria on pre–cooked roast beef. Kim et al. (2002) reported that AIT in combination with acetic acid extended the shelf life of cooked rice stored at various temperatures. Campylobacter spp. analysis When plated on solid agar media, some strains of Campylobacter have the tendency to spread on the surface of the agar, which can sometimes be characterized by a rainbow sheen (Hunt and Abeyta, 1995). When these strains are plated on solid media, colony quantification is impossible to be performed because there are no ways to distinguish isolated colonies. In those cases, quantification is based only on a positive–and–negative basis. Plates that present any bacterial growth are considered positives while plates containing no growth are counted as negative. Statistical analysis of the effects of AIT and days of storage on the incidence of Campylobacter over time was not needed, since all samples that had been exposed to any level of AIT were Campylobacter negative. The initial incidence of naturally–occurring Campylobacter on the control chicken breasts were around 66%, i.e., four out of six samples tested positive for Campylobacter.

PAGE 147

129 When chicken breasts were exposed to 100, 300 and 600 L, all samples tested negative for Campylobacter indicating that exposure to 100 thru 600 L of AIT per 100 grams of sample completely inhibits bacterial growth. Similar results were observed throughout the study, regardless of days of storage or level of AIT. Aproximately 60% of the control samples continued to test positive for Campylobacter while all AIT treated samples tested negative. Given the extent of antimicrobial activity of AIT against C. jejuni in the preliminary studies, exposure of the chicken breasts to the vapors of AIT for a period of up to seven days of storage was expected to fully inhibit bacterial growth. Allyl isothiocyanate was extremely efficient in inhibiting the growth of C. jejuni. Addition of 100, 300 and 600 L of AIT per 100 grams of chicken breast in trays containing the product completely curbed the growth of the organism. Surprisingly, AIT was more efficient in inhibiting the growth of naturally occurring Campylobacter than spoilage organisms. When chicken breasts were exposed to 600 L of AIT for a period of seven days, a maximum of 5.4 log reduction in the numbers of psychrotrophic organisms was obtained when compared to the control samples (Table 4). Differences in sensitivity of strains to AIT, surface microenvironments, decomposition of the substrates, and the decline of residual AIT could all account for the variability seen among psychrotrophic organisms and Campylobacter. So far, there are no studies published in the literature on how different levels alone or in combination with prolonged storage affect the growth

PAGE 148

130 and survival of Campylobacter, specially using a food product. But research has been conducted on the effect of plant oils, other than AIT, on the in vitro survival of Campylobacter jejuni. Friedman et al. (2002) analyzed the effect of 93 essential oils and 23 oil compounds on the growth of Campylobacter jejuni. Based on their study, they reported that growth of Campylobacter was reduced by oils extracted from marigold, ginger root, jasmine, patchouli, gardenia, cedarwood, carrot seed, mugwort, spikenard and orange bitter, and by the oil compounds cinnamaldehyde, estragole, carvacrol, benzaldehyde, citral, thymol, eugenol, perillaldehyde, carvone R, and geranyl acetate. Ward et al. (1998) reported that the amount of vaporized AIT in the headspace of samples that were treated with AIT prior to being packaged declined over time. According to the authors, this may be due to spontaneous decomposition and nucleophilic attack of water and hydroxide ions in atmospheres containing water vapor. In addition plastics can absorb AIT, which further reduces the amount of residual AIT in the headspace. Despite the fact that headspace measurements were not conducted herein, it is the author’s belief that the proprietary gel material present in the bottom chambers of the Fresh–R–Paxt trays may reduce AIT absorption by the plastic. When AIT is applied to the gel material, the liquid compound gelatinizes, which may inhibit plastic absorption of AIT. Another fact that must be considered when comparing the inhibitory effect of AIT against Campylobacter and spoilage organisms is the ability

PAGE 149

131 psychrotrophs have to outcompete Campylobacter. The synergistic effect of AIT combined with the slow growing characteristic of Campylobacter can explain why a complete inhibition of Campylobacter and a partial inhibition of psychrotrophic organisms were observed over time. Based on the results obtained in this study, it was concluded that the extent of inhibition caused by exposure to vaporized AIT is species specific and concentration dependent. Campylobacter was completely inhibited by 100 l to 600 L of AIT after only six hours of exposure while a partial inhibition of spoilage organisms was obtained with the same levels. pH Analysis Table 9 shows the analysis of variance for pH values of chicken breasts homogenate exposed to vapors from 0, 100, 300 and 600 L of AIT per 100 g of meat and stored for up to 7 days at 4 C. The results presented indicate that the level of AIT, the number of days that meat was exposed to AIT vapors (days of storage) and the interaction of AIT levels and days of storage had no significant effect on the pH of the chicken breast homogenate. Treatment means for pH values of chicken breast homogenate exposed to AIT vapors and stored for up to 7 days at 4 C are presented in Table 10. The initial pH of the control chicken breast was 5.86, which was not significantly different than the pH of chicken breasts exposed to 100 L, 300 L, and 600 L of AIT (5.89, 5.88, and 5.89, respectively). Furthermore, no significant differences in

PAGE 150

132 Table 9 Analysis of variance for pH values of skinless, boneless chicken breast homogenate stored at 4C and exposed to allyl isothiocyanate vapors (AIT) for 7 days Source df SS Pr > F Model Level of AIT 3 0.0685 0.2721 Days of storage 3 0.0254 0.6886 Level of AIT * Days of storage 9 0.0187 0.9991 Error 80 1.3789 Total 95 1.4916 Table 10 Treatment means for pH values of skinless, boneless chicken breast homogenate stored at 4C and exposed to allyl isothiocyanate vapors (AIT) for 7 days Days of Storage (pH value) * Treatments 0 3 5 7 0 L AIT/100 g sample 5.86 a 5.92 a 5.85 a 5.93 a 100 L AIT/100 g sample 5.89 a 5.93 a 5.93 a 5.98 a 300 L AIT/100 g sample 5.88 a 5.93 a 5.86 a 5.93 a 600 L AIT/100 g sample 5.89 a 5.90 a 5.91 a 5.95 a * Data shown are average means of six replicate trays containing 3 halved chicken breasts aValues within the same row and same column with the same superscript are not significantly different (p>0.05)

PAGE 151

133 5.505.705.906.106.306.500100200300400500600AIT LevelpH Day 0 Day 3 Day 5 Day 7 Figure 13 Mean pH of homogenate from halved chicken breasts stored at 4C for 7 days while being exposed to allyl isothiocyanate vapors (AIT)

PAGE 152

134 5.505.705.906.106.306.5001234567Storage DaypH 0 AIT 100 AIT 300 AIT 600 AIT Figure 14 Mean pH of homogenate from halved chicken breasts exposed to allyl isothiocyanate vapors (AIT) for 7 days at 4C

PAGE 153

135 the pH of the chicken breast homogenate were observed after seven days of storage, regardless of the level of AIT used. Figure 13 and Figure 14 show the effect of AIT and storage days on the pH of the chicken homogenate. Colorimetric Analysis Consumer purchases are highly influenced by the appearance of product they are ready to buy. Color is the primary criterion by which consumers evaluate meat quality and acceptability (Cornforth, 1994) and it is usually considered the most important sensory characteristic in the aspect of a food, especially of animal origin (MacDougall, 1994). Color can be considered both a subjective and an objective measurement, depending on how it is perceived. When color is perceived by the eye, it is considered a subjective measurement; when color is calculated from the spectral distributions of an object’s reflectance, the illuminant, and standard observer’s trichromatic (x, y, z) color–matching functions, it is considered an objective measurement (MacDougall, 1994). A color, when detected by the eye, is a combination of several factors, which include three attributes intrinsic to any color—hue, chroma, and value (Hedrick et al., 1989). Hue can be perceived as yellow, green, blue or red; chroma describes the intensity of a fundamental color with respect to the amount of white light combined with it; and value represents the overall light reflectance or brightness of the color (Hedrick et al., 1989). The color of animal flesh is the result of the pigments present in the muscle, which consists mainly of two proteins: hemoglobin, a blood pigment,

PAGE 154

136 and myoglobin, a muscle pigment (Cornforth, 1994; Hedrick et al., 1989). The differences observed in muscle colors are the result of varying amounts of each pigment. For example, poultry dark meat contains higher myoglobin content than white meat, explaining why the color of one area contrasts strongly with the other (Hedrick et al., 1989). Rhee and Ziprin (1987) reported that dark meat contained about 0.54 mg.g of total pigments, while the white meat contained 0.16 mg.g of total pigments and zero myoglobin. The amount of total pigments in the muscle gives poultry its typical gray–white to dull–red color. Several factors affect the color of animal flesh, including bacterial growth, type of packaging used, pH of the meat, metmyoglobin reducing activity, retail lighting conditions, and the effects of exogenous reductants or antimicrobial agents (Cornforth, 1994). The need for testing the effects of increasing level of AIT and days of storage on the color of chicken breast is based on the importance that consumers place on the appearance of a food and the fact that antimicrobials and bacterial growth influence the final color of the muscle. The color measurements obtained in this study, are based on the 1976 CIE L* a* b* Space Color System (or CIELAB) adapted from the Munsell color model and later adopted by the Commission Internationale d’Eclairage (CIE). The L*, a*, and b* values are each represented by an axis. The central vertical axis represents L* values (degree of lightness) whose values run from black (L* value = 0) to absolute white (L* value = 100). The color axes are located perpendicularly to the central vertical axis and represent a* and b* values (Figure

PAGE 155

137 15). The color axes are based on the fact that a color has to be either a hue of red or green, and a hue of blue or yellow, but never red and green or blue and yellow because these colors oppose each other. Both a* and b* values run from +120 to –120. On the a* axis, positive values indicate amounts of red while negative values indicate amounts of green. In the case of the b* axis, positive values indicate amounts of yellow while negative values indicate amounts of blue. For both a* and b* axes, zero indicates a neutral gray. To better understand the results of the colorimetric analysis presented herein, a framework of the CIELAB color model system was added (Figure 15). The framework shows how the L*, a*, and b* values are arranged and how they combine to form the desired color. L* value The L* value is an indication of the degree of lightness, whose values vary from 0 (black) to 100 (absolute white). Table 11 presents the analysis of variance for L* values measured from chicken breasts that had been exposed to 0, 100, 300 and 600 L of vaporized AIT per 100 g of meat for up to 7 days of storage. The results indicate that the level of AIT used, the number of days that meat was exposed to AIT vapors (days of storage) and the interaction of AIT levels and days of storage had a significant impact on the L* value of the chicken breast. Table 12 shows the average treatment means for L* value of chicken breast exposed to AIT vapors and stored for up to 7 days at 4 C. Figure 16 shows how

PAGE 156

138 Figure 15 Framework of a Color Model System, showing the L*, a*, and b* values in the form of three axes (L* a* b* Color Solid. Copyright 2000 HunterLab Associates, Inc. Reprinted with permission of the publisher) 1 1 Hunter Associates Laboratory. 11491 Sunset Hills Road, Reston, VA, 20190. Phone: (703) 471–6870. http://www.hunterlab.com

PAGE 157

139 Table 11 Analysis of variance for L* values of skinless, boneless chicken breasts stored at 4 C and exposed to allyl isothiocyanate vapors (AIT) for 7 days Source df SS Pr > F Model Level of AIT 3 156.3763 <0.0001 Days of storage 3 569.4142 <0.0001 Level of AIT * Days of storage 9 207.7835 <0.0001 Error 80 223.8437 Total 95 857.4177

PAGE 158

140 the level of AIT impacted L* values of chicken breast as the storage time increases. The initial average L* value of the control chicken breast (exposure to 0 L of AIT) was found to be 58.33, which was similar to the initial L* value of chicken breasts exposed to 100 L, 300 L, and 600 L of AIT (58.33 vs. 57.74, 56.82, and 56.48, respectively). Differences in L* value due to exposure to AIT vapors started to become significant after three days of storage. At day 3, the L* value for the control sample was 56.89, which was significantly lower than those observed in chicken breasts exposed to 100 L, 300 L, and 600 L of AIT (56.89 vs. 60.53, 61.97, and 61.76, respectively). This pattern of significance has persisted after 5 days of storage, in which L* values of control samples were significantly lower than those that were exposed to any level of AIT (56.17 vs. 61.18, 61.31, and 60.59). After 7 days of storage, a significant difference was observed in the L* values of chicken breasts exposed to higher levels of AIT. Control samples continued to have a lower L* value than those samples that were exposed to any level of AIT (55.10 vs. 61.08, 62.62, and 63.48). No significant difference was observed when the level of AIT, on day 7, was increased from 100 L to 300 L (61.08 vs. 62.62, respectively). However, increasing AIT levels to 600 L caused a significant increase in L* values of chicken breast (63.48 vs. 62.62 and 61.08).

PAGE 159

141 Table 12 Treatment means for L* values of skinless, boneless chicken breasts stored at 4 C and exposed to allyl isothiocyanate vapors (AIT) for 7 days Days of Storage (L* value) * Treatments 0 3 5 7 0 L AIT/100 g sample 58.33 a w 56.89 ab w 56.17 b w 55.10 b w 100 L AIT/100 g sample 57.74 a w 60.53 b x 61.18 b x 61.08 b x 300 L AIT/100 g sample 56.82 a w 61.97 b x 61.31 b x 62.62 b xy 600 L AIT/100 g sample 56.48 a w 61.76 bc x 60.59 b x 63.48 c y * Data shown are average means of six replicate trays containing 3 halved chicken breasts a–cValues within the same row with the same superscript are not significantly different (p>0.05) w–y Values within the same column with the same subscript are not significantly different (p>0.05)

PAGE 160

142 Several antimicrobial agents are known to impact, in one way or another, the color of muscle foods. Mendona et al. (1989) observed that when organic acids was added to fresh, vacuum packaged pork, the L* value decreased over time. In the same way, Papadopoulos et al. (1991b) reported that 2% or more of sodium lactacte reduced L* values of cooked, beef top rounds stored under refrigeration. Sulfites, ascorbate (vitamin C) and –tocopherol (vitamin E) are also known to protect muscle color when the product is stored for prolonged periods of time (Cornforth, 1994). When the same level of AIT was compared over time, it was observed that L* values increased steadily regardless of the level used. The pattern of the L* value reduction with AIT over time is well demonstrated in Figure 17. The average initial L* value of the chicken breasts that were not exposed to AIT was 58.33. After three days of storage a slight, but not significant, reduction in L* value was observed (58.33 vs. 56.89). As time progressed, L* values reduced even more and the differences seen in color became significant. At day 5, the average L* value of the control sample was 56.17, which was similar to the L* value measured at day 3 (56.89) but significantly lower than the initial L* value (58.33). A similar pattern occurred at day 7 of storage, in which the L* value was significantly lower than the initial L* value but similar those from days 3 and 5 (55.10 vs. 58.33, 56.89, and 56.17, respectively).

PAGE 161

143 54.0055.0056.0057.0058.0059.0060.0061.0062.0063.0064.000100200300400500600AIT LevelL* value Day 0 Day 3 Day 5 Day 7 Figure 16 Mean L* values of halved chicken breasts stored at 4 C for 7 days while being exposed to allyl isothiocyanate vapors (AIT)

PAGE 162

144 When chicken breasts were exposed to 100 L of AIT, a significant difference in L* values was seen after only three days of storage. However, contrary to the pattern observed with the control samples, L* values from chicken breasts exposed to AIT increased with time. Due to this difference in patterns, the L* value of control samples at the end of seven days of storage were significantly lower than those from samples that were exposed to AIT (Figure 17). The L* values observed in the chicken breasts exposed to 100 L of AIT after three, five and seven days of storage were all significantly similar among each other, but higher than the initial L* value (60.53, 60.53, and 61.18 vs. 57.74). Exposure to a 300 L of AIT did not change the pattern seen with exposure to 100 L. L* values of chicken breasts exposed to both levels of AIT behaved in the same way. As a matter of fact, throughout the seven days of storage, no significant differences in L* value were observed when samples were exposed to either level of AIT. L* values of samples after three, five, and seven days of storage were not significantly different (61.97, 61.31, and 62.62, respectively); nonetheless, they were significantly higher than the L* value obtained at day zero (56.82). Delaquis et al. (1999) observed that AIT present in horseradish essential oil did not change the degree of lightness in pre–cooked roast beef stored for 15 days. As the level of AIT increased to 600 L, the differences in L* values over time increased. After seven days of storage the L* value of chicken breasts was

PAGE 163

145 54.0055.5057.0058.5060.0061.5063.0064.5001234567Storage DayL* Value 0 AIT 100 AIT 300 AIT 600 AIT Figure 17 Mean L* values of halved chicken breasts exposed to allyl isothiocyanate vapors (AIT) for 7 days at 4 C

PAGE 164

146 similar to that of day 3 but significantly higher than that found on day 5 (63.48 vs. 61.76 and 60.59, respectively). The L* value of chicken breasts were also significantly higher after seven days when compared to that observed at day zero (63.48 vs. 56.48). The initial L* value was found to be significantly lower than those obtained at any given storage day (56.48 vs. 61.76, 60.59, and 63.48). Based on the differences observed in the L* values of the chicken breasts exposed to increasing levels of AIT, regression models were tested in an attempt to find a model that best describes the behavior of the L* value over time (Table 13). With the regression models presented in Table 13, it was possible to calculate a series of prediction regression equations for each model studied (Table 14). As with the psychrotrophic organism population, the regression models presented in Table 14 should be analyzed in conjunction with Figure 16 and Figure 17 because, for this set of data, the model that best fits the treatment means does not necessarily best represent the behavior of the bacterial population. To combine the results presented in the Table 12 and Figure 16 and Figure 17 is especially important in the case of L* values because of the intrinsic characteristics of the chicken breast and the equipment used to measure the L* values. The flesh of the chicken breast does not have a uniform color throughout

PAGE 165

147 Table 13 P–values associated with linear, quadratic and cubic effects of allyl isothiocyanate vapors (AIT) on L* values of skinless, boneless chicken breasts stored at 4 C for 7 days Days of Storage P–values 0 3 5 7 Model 1 Linear effect of AIT 0.0031 <0.0001 0.0095 <0.0001 Quadratic effect of AIT 0.2572 <0.0001 0.0017 0.0025 Cubic effect of AIT 0.9348 0.0194* 0.0175* 0.0336* Model 2 Linear effect of AIT 0.0025 <0.0001 0.0190 <0.0001 Quadratic effect of AIT 0.2454 <0.0001 0.0042 0.0047 Model 3 Linear effect of AIT 0.0025 — — — *Cubic effect of AIT does not reflect data properly; therefore Model 2 was used as the best fitted model Table 14 Prediction regression equations for the linear and quadratic curves for L* values of skinless, boneless chicken breasts exposed to allyl isothiocyanate vapors (AIT) for 7 days Days of Storage R 2 Regression Equation 0 0.34 L* value = 58.101 – 0.0030 (AIT) 3 0.77 L* value = 57.3596 + 0.0266 (AIT) – 0.00003 (AIT) 2 5 0.44 L* value = 57.0806 + 0.0286 (AIT) – 0.00003 (AIT) 2 7 0.63 L* value = 56.0366 + 0.0380 (AIT) – 0.00004 (AIT) 2

PAGE 166

148 the entire surface, which contributes to the large variation in values obtained. In addition to that, the portable, hand–held colorimeter used to measure the L* values, is fitted with a small aperture (10 mm). An aperture this small gives different values during continuous measurements because of the wide range in translucency in the meat that results from differences in the extent of protein denaturation during post–mortem glycolysis, as affected by pH and temperature during chilling (MacDougall, 1994). Thus, it is important for the person taking the measurements to be able to distinguish changes in values that are due to color and appearance from those that the result of variation in translucency. In order to minimize values that do not reflect well the real color of the meat, two to four areas within each sample should be measured (MacDougall, 1994). In this study, three areas were measured for each one of the three chicken breasts per tray, giving a total of 9 measured areas per experimental unity. Table 15 shows the linear, quadratic and cubic effects of storage day on the L* values observed in untreated and AIT treated chicken breasts. Table 16 presents the prediction regression equations, as well as the coefficient of correlation (R 2 ) for the best fitted model. The regression models described on Table 15 should be analyzed in conjunction with Figure 16. The effect of storage day on the L* values of chicken breasts was different according to the level of AIT used. When chicken breasts were not exposed to.

PAGE 167

149 Table 15 P–values associated with linear, quadratic and cubic effects of days of storage on L* values of skinless, boneless chicken breasts exposed to allyl isothiocyanate vapors (AIT) Levels of AIT (L AIT/100 g sample) P–values 0 100 300 600 Model 1 Linear effect of Day 0.0030 <0.0001 <0.0001 <0.0001 Quadratic effect of Day 0.9677 0.0256 0.0204 0.2278 Cubic effect of Day 0.8399 0.9167 0.0577 0.0203* Model 2 Linear effect of Day 0.0024 <0.0001 <0.0001 <0.0001 Quadratic effect of Day 0.9669 0.0221 0.0284 0.2793 Model 3 Linear effect of Day 0.0019 — — <0.0001 *Cubic effect of Day does not reflect data properly; therefore Model 2 was used as the best fitted model Table 16 Prediction regression equations for the linear and quadratic curves for L* values of skinless, boneless chicken breasts stored for 7 days while exposed to allyl isothiocyanate vapors (AIT) AIT (L AIT/100 g sample) R 2 Regression Equation 0 0.36 L* value = 58.3263 – 0.45269 (Day) 100 0.60 L* value = 57.0189 + 1.8375 (Day) – 0.15548 (Day) 2 300 0.62 L* value = 57.7524 + 1.2457 (Day) – 0.11027 (Day) 2 600 0.53 L* value = 57.2300 + 0.8946 (Day)

PAGE 168

150 AIT (control samples), storage day had a linear effect in the measured L* values. Addition of 100 L and 300 L of AIT to the trays containing chicken breasts, accounted for the quadratic effect of storage day on the L* values observed in this study. A two–fold increase in the level of AIT (from 300 L to 600 L) resulted in the linear effect of storage day on the L* values seen herein. In summary, exposure to AIT vapors significantly increased the lightness of the chicken breasts over time. Chicken breasts that were exposed to 100 L, 300 L, and 600 L of AIT had a consistently lighter muscle color than the control samples. After 7 days of storage, exposure to 600 L of AIT further increased the degree of lightness of the product. Changes in the degree of lightness in muscle are partially the result of unusual binding of water in the muscle. If there is large amount of free water in the intercellular space of the muscle tissue, i.e., between the muscle cells instead of within them, the reflection of light is altered. This excessive amount of extracellular water reflects well the light while limiting the absorption capacity reducing the intensity of the color of the muscle. The result is that the muscle appears pale to the eye and unappealing to the consumer. Similar to exposure to AIT, storage of chicken breasts at 4 C significantly affected the color of the muscle. As storage time increased, so did the degree of lightness indicating that the longer the muscle is kept at the retail case, the lighter it will be. Degree of lightness is important and must be taken into account when choosing an antimicrobial agent because of the influence muscle color has on the

PAGE 169

151 consumer. Studies on pale, soft, exudative (PSE) cuts of meat have shown that muscle colored pale gray to white is not appealing to the consumer and it is often regarded as a low quality product. a* value The a* value is an indication of redness; positive values indicate amounts of red while negative values indicate amounts of green. Table 17 shows the analysis of variance for a* values measured from chicken breasts that had been exposed to 0, 100, 300 and 600 L of vaporized AIT per 100 g of meat for up to 7 days of storage. The results indicate that the level of AIT used, the number of days that meat was exposed to AIT vapors (days of storage) and the interaction of AIT levels and days of storage had a significant impact on the degree of redness of the chicken breast. Table 17 Analysis of variance for a* values of skinless, boneless chicken breasts stored at 4 C and exposed to allyl isothiocyanate vapors (AIT) for 7 days Source df SS Pr > F Model Level of AIT 3 6.4930 <0.0001 Days of storage 3 6.9045 <0.0001 Level of AIT * Days of storage 9 2.5197 <0.0001 Error 80 7.0020 Total 95 22.9193 Table 18 gives the average treatment means for a* value of chicken breast exposed to AIT vapors and stored for up to 7 days at 4 C. Figure 18 shows the effect of AIT on a* values of chicken breast throughout the 7–day storage period.

PAGE 170

152 Table 18 Treatment means for a* values of skinless, boneless chicken breasts stored at 4 C and exposed to allyl isothiocyanate vapors (AIT) for 7 days Days of Storage (a* value) * Treatments 0 3 5 7 Comparison within rows 1 0 L AIT/100 g sample 0.445 a w 0.361 a w 0.253 a w 0.448 a w 100 L AIT/100 g sample 0.365 a w .761 b x .548 b x .443 b x 300 L AIT/100 g sample 0.430 a w .398 b y .220 b x .501 b x 600 L AIT/100 g sample 0.300 a w .226 b y .278 b x .228 b x * Data shown are average means of six replicate trays containing 3 halved chicken breasts a–bValues within the same row with the same superscript are not significantly different (p>0.05) w–y Values within the same column with the same subscript are not significantly different (p>0.05)

PAGE 171

153 -1.00-0.80-0.60-0.40-0.200.000.200.400.600100200300400500600AIT Levela* value Day 0 Day 3 Day 5 Day 7 Figure 18 Mean a* values of halved chicken breasts stored at 4 C for 7 days while being exposed to allyl isothiocyanate vapors (AIT)

PAGE 172

154 The initial average a* value of the control chicken breast (exposure to 0 L of AIT) was 0.445, which was similar to the initial a* value of chicken breasts exposed to 100 L, 300 L, and 600 L of AIT (0.445 vs. 0.365, 0.430, and 0.300, respectively). Following the same pattern of the L* values, differences in the degree of redness due to exposure to vapors of AIT started to become significant after three days of storage. Also the a* values for AIT treated chicken breasts became negative, indicating higher levels of green in the muscle. At day 3, the a* value for the control sample was 0.361, which was significantly higher than those observed in chicken breasts exposed to 100 L, 300 L, and 600 L of AIT (0.361 vs. –0.761, .398, and –0.226, respectively). The degree of redness of the control samples continued to be significantly higher than those of AIT treated samples. With the exception of day 3 (–0.761 for 100 L vs. –0.398 for 300 L, and –0.226 for 600 L of AIT) no significant differences in a* value were observed among chicken breasts exposed to 100 L, 300 L, and 600 L of AIT on days 0, 5 and 7 of storage. On the contrary, a* values for the control samples were significantly higher than those observed in the AIT treated samples after three, five and seven days of storage. Figure 18 gives a good illustration of the sample’s behavior according to the level of AIT used and storage day. Similar decrease in the degree of redness as storage time increased was reported in the literature for other muscle foods and antimicrobial agents. Banks et al. (1998) and Boles and Shand (1999) reported the degree of redness of restructured turkey and pork decreased with retail storage. The reason for this

PAGE 173

155 decrease is believed to be due to a reduction in the levels of oxymyoglobin present in the muscle as the food “ages” in the retail case. Exposure to AIT and storage day affected the degree of redness of chicken breasts. Delaquis et al. (1999) observed that the allyl isothiocyanate in horseradish essential oil increased the degree of redness in pre–cooked roast beef after 3 and 15 days of storage. In the study herein, when the same level of AIT was compared over time, it was observed that a* values for AIT treated samples decreased after 3 days of storage and remained the same (Table 18). The a* values for the control samples did not significantly change during the 7 days storage period (0.455 for day 0, 0.361 for day 3, 0.253 for day 5, and 0.448 for day 7). Chicken breasts exposed to 100 L of AIT had an initial a* value of 0.365, which significantly declined to –0.761 after three days of storage. No significant differences were observed in a* values of the 100 L chicken breasts thereafter. The pattern of reduction of the a* value after 3 days of storage was found in the chicken breasts treated with 300 L and 600 L of AIT. The initial a* value of 300 L AIT treated breasts was 0.430. At day 3, a* value was significantly reduced to .398, not differing after that (–0.220 for day 5, and –0.501 for day7). The initial a* value of the chicken breasts treated with 600 L of AIT was 0.300, which was significantly higher than a* values for days 3, 5, and 7 of storage (–0.226, –0.278, and –0.228, respectively). The effects of storage time on the control and the AIT treated samples can be clearly seen in Figure 19.

PAGE 174

156 -1.00-0.80-0.60-0.40-0.200.000.200.400.600123456Storage Daya* Value 7 0 AIT 100 AIT 300 AIT 600 AIT Figure 19 Mean a* values of halved chicken breasts exposed to allyl isothiocyanate vapors (AIT) for 7 days at 4 C

PAGE 175

157 Table 19 P–values associated with linear, quadratic and cubic effects of allyl isothiocyanate vapors (AIT) on a* values of skinless, boneless chicken breasts stored at 4 C for 7 days Days of Storage P–values 0 3 5 7 Model 1 Linear effect of AIT 0.3273 0.1070 0.0829 0.0739 Quadratic effect of AIT 0.7043 <0.0001 0.0181 0.0030 Cubic effect of AIT 0.4325 <0.0001* 0.0003* 0.0620 Model 2 Linear effect of AIT 0.3226 0.3211 0.1956 0.0917 Quadratic effect of AIT 0.7017 0.0042 0.0732 0.0047 Model 3 Linear effect of AIT 0.3128 — 0.2190 — *Cubic effect of AIT does not reflect data properly; therefore Model 2 was used as the best fitted model Table 20 Prediction regression equations for the linear and quadratic curves for a* values of skinless, boneless chicken breasts exposed to allyl isothiocyanate vapors (AIT) for 7 days Days of Storage R 2 Regression Equation 0 — a* value = 0.38500 3 0.35 a* value = 0.1003 – 0.0046 (AIT) + 0.000006 (AIT) 2 5 0.20 a* value = .19833 7 0.38 a* value = 0.2959 – 0.00548 (AIT) + 0.00007 (AIT) 2

PAGE 176

158 The P–values associated with the linear, quadratic and cubic effects of AIT on the degree of redness at a given storage day are listed on Table 19. At day 0 and 5, AIT had no linear, quadratic or cubic effect on the degree of redness of the chicken breast. As storage time increased from 0 to 3, and 7, AIT exerted a quadratic effect on the degree of redness of the muscle at each given day (0.0042, and 0.0047, respectively). The effects of AIT on a* values of chicken breasts can be seen in Figure 18. The day 0 curve showed almost no variation in trend, exhibiting a straight horizontal line. Day 5, on the contrary, had a sharp decrease in the degree of redness when chicken breasts were exposed to 100 L of AIT, followed by a slight increase with 300 L and again a slight increase. The results obtained accounted for a cubic effect of AIT on the degree of redness instead of a quadratic. A higher number of replicates could allow for a better fitting of the regression models. The high variation in the colorimetric values obtained during this study, were responsible for the trends in L* a* and b* values observed herein. Problems already described in the L* value section of this study such as the small aperture of the hand–held colorimeter and the normal variation of the color of the surface of muscle contributed for the variability observed in the a* value measurements. The lack of a specific pattern in the response analyzed herein has been seen in other studies. Papadopoulos et al. (1991b), reported that the L* values of cooked, beef round tops fluctuated during storage, resulting in no apparent pattern over time.

PAGE 177

159 Based on the regression models described in Table 19, prediction regression equations and the coefficient of correlation (R 2 ) were calculated for the effect of AIT on the degree of redness within 7 days of storage (Table 20). Table 21 shows the linear, quadratic and cubic effects of storage day on the degree of redness observed in the untreated and AIT treated chicken breasts. Table 22 presents the prediction regression equations, as well as the coefficient of correlation (R 2 ) for the model that best fits the trends observed with each sample. Again, the regression models described in Table 21 should be analyzed in conjunction with Figure 19. The effect of storage day on a* values of chicken breasts varied according to the level of AIT used. Storage day had a linear effect on the degree of redness on the control samples and on those exposed to 300 L and 600 L of AIT. The lower level of AIT, 100 L, accounted for a quadratic effect of storage day on the degree of redness observed on chicken breasts. Overall, exposure to vapors of AIT and storage time had a small, but significant, effect on the color of the chicken breast. Exposure to 100 L, 300 L, and 600 L of AIT reduced the degree of redness in the chicken breasts. A lower degree of redness indicates a green discoloration of the muscle. Green discoloration of fresh meat can be induced by oxidizing bacteria and by hydrogen sulfide–producing bacteria. If even a very small percentage of the total microflora is composed of Pseudomonas mephitica, the myoglobin present in the muscle is converted to sulfmyoglobin, which gives the typical greenish color to

PAGE 178

160 Table 21 P–values associated with linear, quadratic and cubic effects of days of storage on a* values of skinless, boneless chicken breasts exposed to allyl isothiocyanate vapors (AIT) Levels of AIT (L AIT/100 g sample) P–values 0 100 300 600 Model 1 Linear effect of Day 0.7581 <0.0001 0.0002 0.0084 Quadratic effect of Day 0.2688 <0.0001 0.0815 0.0710 Cubic effect of Day 0.4233 0.0285* 0.0581 0.7920 Model 2 Linear effect of Day 0.7561 <0.0001 0.0003 0.0069 Quadratic effect of Day 0.2644 <0.0001 0.1012 0.0646 Model 3 Linear effect of Day 0.7576 — 0.0005 0.0100 *Cubic effect of Day does not reflect data properly; therefore Model 2 was used as the best fitted model Table 22 Prediction regression equations for the linear and quadratic curves for a* values of skinless, boneless chicken breasts stored for 7 days while exposed to allyl isothiocyanate vapors (AIT) AIT (L AIT/100 g sample) R 2 Regression Equation 0 — a* value = 0.37625 100 0.73 a* value = 0.3327 – 0.4795 (Day) + 0.05389 (Day) 2 300 0.43 a* value = 0.2787 – 1.2034 (Day) 600 0.26 a* value = 0.17834 – 0.07644 (Day)

PAGE 179

161 fresh muscle products (Cornforth, 1994). According to the author, hydrogen peroxide (H 2 O 2 ) of bacterial source or derived from endogenous muscle reactions can also cause oxidation of the myoglobin to green cholemyoglobin. The reason for the decrease in the degree of redness in the AIT treated samples is most likely the result of the production of cholemyoglobin in the muscle. Cholemyoglobin is produced when bacterial or endogenous H 2 O 2 binds to the active site (6 th site) of a denatured myoglobin molecule. Production of sulfmyoglobin can also be accounted for the reduced degree of redness observed in the chicken breast. Binding of the sulfur group (SH) of the AIT molecule to the active site of a denatured myoglobin causes similar greenish discoloration observed when H 2 O 2 binds to a denatured myoglobin molecule. b* value The b* value is an indication of yellowness; positive values indicate amounts of yellow while negative values indicate amounts of blue. Table 23 gives the analysis of variance for b* values of chicken breasts that had been exposed to 0, 100, 300 and 600 L of vapors of AIT per 100 g of meat for up to 7 days of storage. The results indicate that the level of AIT used, the number of days that meat was exposed to AIT vapors (days of storage) and the interaction of AIT levels and days of storage had a significant impact on the b* value of the chicken breast.

PAGE 180

162 Table 23 Analysis of variance for b* values of skinless, boneless chicken breasts stored at 4 C and exposed to allyl isothiocyanate vapors (AIT) for 7 days Source df SS Pr > F Model Level of AIT 3 280.1336 <0.0001 Days of storage 3 239.1061 <0.0001 Level of AIT * Days of storage 9 70.9710 <0.0001 Error 80 81.2308 Total 95 671.4417 Table 24 shows the average treatment means for b* value of chicken breasts exposed to AIT vapors and stored for up to 7 days at 4 C. Figure 20 shows the effect of AIT on the degree of yellowness of chicken breast throughout the 7–day storage period. The initial average b* value of the control chicken breast (exposure to 0 L of AIT) was 1.50, which was similar to the initial b* value of chicken breasts exposed to 100 L, and 600 L of AIT (1.50 vs. 1.91, and 2.36, respectively), but significantly different than the initial b* value of samples exposed to 300 L of AIT (2.96). As the chicken breasts aged in the retail case, the degree of yellowness changed. After three days of storage, the b* values of the AIT treated samples markedly increased as compared to the control, untreated sample. Chicken breasts exposed to 600 L of AIT vapors had significantly higher b* values than those exposed to 0, 100 L, and 300 L of AIT (7.67 vs. 3.70, 5.28, and 6.11, respectively). Exposure to 100 L or 300 L of AIT significantly increased degree

PAGE 181

163 Table 24 Treatment means for b* values of skinless, boneless chicken breasts stored at 4 C and exposed to allyl isothiocyanate vapors (AIT) for 7 days Days of Storage (b* value) * Treatments 0 3 5 7 0 L AIT/100 g sample 1.50 a w 3.70 b w 3.00 b w 2.61 ab w 100 L AIT/100 g sample 1.91 a wx 5.28 b x 6.00 bc x 6.50 c x 300 L AIT/100 g sample 2.96 a x 6.11 b x 6.53 b xy 7.06 b x 600 L AIT/100 g sample 2.36 a wx 7.67 b y 8.10 b y 10.14 c y * Data shown are average means of six replicate trays containing 3 halved chicken breasts a–c1Values within the same row with the same superscript are not significantly different (p>0.05) w–y Values within the same column with the same subscript are not significantly different (p>0.05)

PAGE 182

164 0.002.004.006.008.0010.0012.000100200300400500600AIT Levelb* value Day 0 Day 3 Day 5 Day 7 Figure 20 Mean b* values of halved chicken breasts stored at 4 C for 7 days while being exposed to allyl isothiocyanate vapors (AIT)

PAGE 183

165 of yellowness of the chicken breasts as compared to the control samples (5.28, and 6.11 vs. 3.70, respectively), but no significant differences were observed between the two concentrations of AIT. The differences in b* values among treatments, became even higher at day 5 and day 7. At day 5, control samples had significantly lower b* values than AIT treated chicken breasts (3.00 vs. 6.00 for 100 L, 6.53 for 300 L, and 8.10 for 600 L of AIT). Similarly to day 3, no significant differences in the degree of yellowness were observed among samples exposure to the intermediate levels of AIT (6.00 for 100 L and 6.53 for 300 L of AIT). Exposure to 600 L significantly decreased the b* values of samples when compared to exposure to 0 or 100 L of AIT; no significant differences were observed among exposure to 300 L and 600 L of AIT. The effects of exposing chicken breasts to AIT vapors on the degree of yellowness can be seen in Figure 20. The trends observed in the figure, clearly indicate that samples stored for three or more days, had a higher degree of yellowness than the fresh product. Delaquis et al. (1999) observed that horseradish essential oil, which is mainly composed of allyl isothiocyanate decreased the degree of yellowness in pre–cooked roast beef after 3 and 15 days of storage. When AIT levels were kept constant and average treatment means for each storage day were compared, significant differences in the degree of

PAGE 184

166 0.002.004.006.008.0010.0012.0001234567Storage Dayb* Value 0 AIT 100 AIT 300 AIT 600 AIT Figure 21 Mean b* values of halved chicken breasts exposed to allyl isothiocyanate vapors (AIT) for 7 days at 4 C

PAGE 185

167 yellowness were observed throughout the 7–day period (Table 24). Figure 21 shows very clearly the effect of storage time on the degree of yellowness of the chicken breasts. The degree of yellowness of the control samples significantly increased at day 3 and day 5 as compared to day 0, but it returned to the same level after 7 days of storage (1.50 vs. 3.70, 3.00, and 2.61, respectively). When chicken breasts were exposed to 100 L of AIT, storage day had a higher effect on the degree of yellowness of the sample than it had on the controls. Degree of yellowness significantly increased from 1.91 on day 0, to 5.28 at day 3, 6.00 at day 5 and 6.50 on day 7. Similarly, the b* values of samples exposed to 300 L of AIT significantly increased from 2.96 at day 0, to 6.11 on day 3, 6.53 on day 5 and 7.06 on day 7. The most dramatic increase in the degree of yellowness was observed when samples were exposed to 600 L of AIT. The degree of yellowness of chicken breasts at day 3, 5 and 7 was significantly higher than the values obtained when samples were fresh (7.67, 8.10, and 10.14 vs. 2.36). The results presented in Table 24 and Figure 21 certainly indicate that exposure to the vapors of AIT increases the degree of yellowness of the chicken breasts. Moreover, as the concentration of AIT and storage time increases, so does the degree of yellowness. It is important to remember, that samples were continuously exposed to the AIT vapors throughout the entire 7–day period of storage; samples were not simply exposed to AIT and then stored.

PAGE 186

168 Table 25 P–values associated with linear, quadratic and cubic effects of allyl isothiocyanate vapors (AIT) on b* values of skinless, boneless chicken breasts stored at 4 C for 7 days Days of Storage P–values 0 3 5 7 Model 1 Linear effect of AIT 0.0192 <0.0001 <0.0001 <0.0001 Quadratic effect of AIT 0.0069 0.2399 0.0255 0.0693 Cubic effect of AIT 0.3812 0.1832 0.0119* 0.0020* Model 2 Linear effect of AIT 0.0183 <0.0001 <0.0001 <0.0001 Quadratic effect of AIT 0.0064 0.2488 0.0477 0.1383 Model 3 Linear effect of AIT — <0.0001 — <0.0001 *Cubic effect of AIT does not reflect data properly; therefore Model 2 was used as the best fitted model Table 26 Prediction regression equations for the linear and quadratic curves for b* values of skinless, boneless chicken breasts exposed to allyl isothiocyanate vapors (AIT) for 7 days Days of Storage R 2 Regression Equation 0 0.42 b* value =1.4039 + 0.0080 (AIT) – 0.00001 (AIT) 2 3 0.68 b* value = 1.8177 + 0.0014 (AIT) 5 0.68 b* value = 3.5535 + 0.0161 (AIT) – 0.00001 (AIT) 2 7 0.73 b* value = 3.8746 + 0.0108 (AIT)

PAGE 187

169 Table 25 shows the linear, quadratic, and cubic effects of AIT on the degree of yellowness of chicken breasts over time. Exposure to 0 or to 300 L of AIT had a quadratic effect on the b* values of the chicken breasts. The quadratic effect was most likely due to the slight increased b* values obtained when samples were exposed to 300 L of AIT on day 0 of storage and to the sharp increase in b* value when samples were exposed to 100 L on day 5, as compared to the control sample (Figure 20). Exposure to AIT vapors had a linear effect on the degree of yellowness on days 3 and 7. This means that for each increase in level of AIT on day 3 and day 7, there was an increase in the b* value. Table 26 presents the prediction regression equations, as well as the coefficient of correlation (R 2 ) for the model that best describes the trend of the degree of yellowness of chicken breasts according to AIT level and storage time. Table 27 presents the linear, quadratic and cubic effects of storage day on the degree of yellowness of chicken breasts exposed to 0, 100 L, 300 L and 600 L of AIT. Storage day had a quadratic effect on b* values of chicken breasts exposed to 0, 100 L, and 300 L of AIT and had a linear effect on the degree of yellowness when samples were exposed to 600 L of AIT. When the trends in b* values for the control and AIT samples is analyzed in Figure 21 is analyzed, the quadratic and linear effects of AIT is clearly observed. Exposure to 600 L of AIT had a linear effect on the degree of yellowness of chicken breasts, meaning that for each increase in storage day, there is an

PAGE 188

170 Table 27 P–values associated with linear, quadratic and cubic effects of days of storage on b* values of skinless, boneless chicken breasts exposed to allyl isothiocyanate vapors (AIT) Levels of AIT (L AIT/100 g sample) P–values 0 100 300 600 Model 1 Linear effect of Day 0.0322 <0.0001 <0.0001 <0.0001 Quadratic effect of Day 0.0006 0.0004 0.0174 0.0399 Cubic effect of Day 0.1145 0.3451 0.3774 0.0742 Model 2 Linear effect of Day 0.0381 <0.0001 <0.0001 <0.0001 Quadratic effect of Day 0.0008 0.0004 0.0166 0.0504 Model 3 Linear effect of Day — — — <0.0001 Table 28 Prediction regression equations for the linear and quadratic curves for b* values of skinless, boneless chicken breasts stored for 7 days while exposed to allyl isothiocyanate vapors (AIT) AIT (L AIT/100 g sample) R 2 Regression Equation 0 0.48 b* value = 1.5824 + 0.9513 (Day) – 0.1176 (Day) 2 100 0.88 b* value = 1.9537 + 1.3693 (Day) – 0.1041 (Day) 2 300 0.73 b* value = 3.0193 + 1.2479 (Day) – 0.0978 (Day) 2 600 0.76 b* value = 3.0839 + 1.0635 (Day)

PAGE 189

171 increase in b* value on samples exposed to 600 L of AIT. Storage day had a quadratic effect on the b* values of samples exposed to 100 L and 300 L because the degree of yellowness sharply increased from day 0 to day 3, plateauing thereafter. Control samples showed an increase in b* values after 3 days of storage followed by a slight decrease in values. Table 28 gives the prediction regression equations and coefficient of correlation (R 2 ) for the model that best describe the trend of the degree of yellowness of chicken breasts according to storage day and AIT level. Overall, exposure to AIT vapors and storage of the samples at 4 C in a retail case significantly increased the degree of yellowness of chicken breasts. Chicken breasts that were exposed to 100 L, 300, and 600 L of AIT had a consistently higher b* values than the control samples. The difference between AIT treated samples and the control increased with storage. The degree of yellowness of chicken breasts exposed to 600 L of AIT on day 0 was 57% higher than the control, while it was 107% higher on day 3, 170% higher on day 5, and finally 288% higher than the control at the end of 7 days of storage time. It is important to consider the effects that AIT and storage time have on the degree of yellowness of the chicken breasts because color is known to influence a consumer’s concept of freshness, i.e., consumers assume that a product is fresh if it has a certain color. Even if the product is not spoiled, acceptance will be reduced if an antimicrobial agent drastically changes the “fresh color” of the muscle (Pearson, 1994). In the case of poultry, particularly

PAGE 190

172 breast meat, emphasis on color is quite different than for red meats. A desirable chicken breast is characterized by a lack of pigmentation (Pearson, 1994). Thus, changes in color of the muscle should, ideally, be kept to a minimum. The increased degree of yellowness observed in the AIT treated samples was most likely caused by the destruction of the heme group of the myoglobin. When the globin molecule is absent and the heme group is destroyed, the muscle acquires a yellowish color due to the presence of free porphyrins. The fact that chicken breast muscles are practically devoid of pigments creates a problem with large variability in measurements, as already stated. Any small increase or decrease in one of the pigments significantly changes the final color of the muscle. Sometimes the problem with uneven colors is enhanced because poultry bones contain the pigments—hemoglobin and myoglobin, which may be released during processing and/or freezing (Pearson, 1994). Conclusion Unfortunately there is little room for complacency and mistakes in the food industry. Major bacterial foodborne outbreaks seen throughout the past five years, including Hudson Foods in 1997 and ConAngra Beef Company in 2002, have proved that faulty food safety measures have cost the lives of several human beings as well as the life of the company involved (Knight and Pierce, 1997; Henkel, 1998; Veneman and Murano, 2002). Current studies should focus on the multiplicity of sub–lethal stressors or inhibitors that are commonly experienced by psychrotrophic organisms and

PAGE 191

173 Campylobacter jejuni during food manufacture, distribution and storage. The methods used must be able to be carried out on an industry–type setting. The intervention steps must be easy to adopt and adapt to a variety of settings, be reasonably priced, consistent in terms of its effect on the bacterium, and safe, not only to consumers but also for the personnel involved in the process. Finally, the use of allyl isothiocyanate has been shown to inhibit the growth of Campylobacter jejuni and to retard the growth of spoilage microorganism, but its use in the food industry as a inhibitor of microbial growth must take into consideration the enormous risk and poisonousness of handling such a chemical, especially when handling large quantities of isothiocyanates. Allyl isothiocyanate exerted variable bacteriostatic and bacteriocidal effects toward Campylobacter and psychrotrophic organisms on chicken breasts stored for up to seven days at 4 C. Campylobacter organisms were strongly inhibited, while growth of psychrotrophic organisms was reduced as storage time increased. The partial resistance of spoilage psychrotrophic organisms to AIT may be advantageous, since spoilage organisms can inhibit the growth of undesirable pathogenic bacteria in poultry products by out competing them. One of the objectives of this study was to maintain product safety and extend shelf life through elimination or inhibition of spoilage and pathogenic organisms. Based on the results obtained herein, exposure of chicken breasts to vapors of AIT allowed for the accomplishment of this objective.

PAGE 192

174 The use of AIT as an antimicrobial agent and storage time did not affect the pH of the chicken breast. On the contrary, color of the muscle was significantly affected by storage time and levels of AIT. Consumer purchases are highly influenced by the appearance of product, and color is the primary criterion by which consumers evaluate meat quality and acceptability. Furthermore, color is usually considered the most important sensory characteristic in the aspect of a food, especially of animal origin. Exposure to vapors of AIT, as well as storage time, increased the degree of lightness and degree of yellowness of the muscles while it decreased the degree of redness. Growth of certain species of bacteria has long been suspected to influence meat color. Since identification of the bacteria commonly grouped as psychrotrophic organisms was not performed in this study, it is not possible to determine a single reason for muscle discoloration observed in this study. However, the increase degree of lightness is believed to be the result of an increased amount of free water located between the muscle cells. Also, the production of cholemyoglobin due to binding of endogenous or bacterial hydrogen peroxide to the active site of the myoglobin molecule and the production of sulfmyoglobin caused by the binding of a sulfur group to the myoglobin results in a decreased degree of redness in the muscle. Furthermore, the total destruction of the heme group of the myoglobin molecule with the release of free porphyrins causes an increase in the degree of yellowness. All

PAGE 193

175 those factors combined could have accounted for the differences in the L*, a*, and b* values observed in the chicken breasts that have been exposed to AIT. Allyl isothiocyanate proved that it has the potential for being a useful antimicrobial agent to the poultry industry when applied to packaged products. However, a less toxic form of AIT, such as horseradish distillates, should be used. Future studies should be conducted to determine if other sources of AIT available in the market could exert the same antimicrobial effect on poultry products without the extreme toxicity to humans. Also, residual AIT should be determined by gas chromatography throughout the storage period and identification of the psychrotrophs present in the poultry should be performed to give a better understanding of the effects of AIT in poultry muscle.

PAGE 194

CHAPTER 4 EFFECT OF ACID–SHOCK ALONE AND IN COMBINATION WITH PORCINE BILE AND MUCIN ON CAMPYLOBACTER JEJUNI STORED AT 4C FOR THREE MONTHS Introduction Campylobacter jejuni is a gram–negative, microaerophilic foodborne enteropathogen, being one of the bacterial pathogens that have their incidence monitored by the Centers for Disease Control and Prevention (CDC) through the Emerging Infections Program Foodborne Diseases Active Surveillance Network (FoodNet). According to the CDC’s 2001 preliminary data on the incidence of foodborne illnesses in the United States, Campylobacter was the second leading cause of gastroenteritis being responsible for 13.8 cases of illnesses per 100,000 persons (Centers for Disease Control and Prevention, 2002). An estimated 2.4 million persons are affected with Campylobacter alone each year. Campylobacter is commonly found in the intestinal tracts of humans and animals, poultry being the main reservoir for some species such as C. jejuni. It is believed that about 80% of all poultry sold for human consumption in the United States is contaminated with Campylobacter (Centers for Disease Control and Prevention, 2000). It is generally accepted that low storage temperatures (4 C) allow Campylobacter to remain viable in the food product during storage. Given the 176

PAGE 195

177 importance of water and food, especially poultry, as vehicles of C. jejuni along with the extensive use of low temperatures for food preservation, studies examining the survival of C. jejuni stored at 4 C for prolonged periods of time and its capability to withstand the low acidity of the stomach when inside a host are of extreme importance. The hypothesis of this study was that environmental factors such as acid shock, and the presence of bile and mucin might trigger the release of Campylobacter jejuni from a period of dormancy after a prolonged storage on a nutrient–deprived environment allowing a monitoring of viability through viable plate counts. Material and Methods Selective Culture Media and Agar Most of the methodology used in this study was similar to those described in the Bacteriological Analytical Manual (Food and Drug Administration, 1998). Campylobacter Enrichment Broth (Acumedia Manufacturers, Inc., a subsidiary of Idexx Laboratories, Baltimore, Maryland, 21211, cat. No. 7526) containing 5% lysed horse blood (Lampire Biological Laboratories, Pipersville, PA, 18947) was used as enrichment medium. Frozen horse lysed blood was purchased from Lampire Biological Laboratories. Upon arrival, blood was thawed and 50 mL aliquots were aseptically transferred to 50 mL sterile conical centrifuge tubes. Individual tubes containing horse blood were kept frozen at –8 C until needed. During medium

PAGE 196

178 preparation, blood was removed from freezer, thawed and aseptically added to the autoclaved enrichment broth. Plates used to culture Campylobacter were made of Karmali medium (Oxoid, Inc. Nepean, Ontario, Canada, K2G 1E8, cat. No. CM0935B). Unless stated otherwise, medium did not contain any supplement such as antibiotics or horse blood. Plates were pre–poured at the Oral Biology laboratory at the University of Florida’s Dental School at the beginning of the study and stored in the dark at 4 C until needed. Campylobacter Enrichment Broth tubes and Karmali plates were incubated in a microaerophilic environment, i.e., 5% O 2 , 10% CO 2 and 85% N 2 , using a gassed jar system. Two 3.5 L Campy Gas Generating Gas Pak envelopes (Oxoid, Inc. Nepean, Ontario, Canada, K2G 1E8) were added to each 7 L BBL gas jar and one 2.5 L Campy Gas Generating Gas Pak (Oxoid, Inc. Nepean, Ontario, Canada, K2G 1E8) was added to each small 2.5 L BBL gas jar (BBL). Unused jars were cleaned and stored opened with the screw clamps lids stored separately. Bacterial Culture A freeze–dried culture of Campylobacter jejuni subsp. jejuni (ATCC 29428) was purchased from ATCC (American Tissue Culture Collection, Manassas, Virginia, 20110). Upon receiving, culture was propagated according to ATCC’s instructions. The freeze–dried pellet was hydrated in 6 mL of Fluid Thioglycollate. Several drops of the suspension were used to inoculate slants and plates of Trypticase Soy Agar supplemented with 5% lysed horse blood.

PAGE 197

179 Test tube, slants and plates were incubated at 37 C, in anaerobe jars fitted with Campy–Paks, for 48–72 hours. Identification of Campylobacter colonies was based on colony morphology, wet mount, and catalase and oxidase tests. Typical colonies were stored on ceramic beads (Protect Beads, Med–Ox Diagnostics Inc., Ogdensburg, NY, 13669, cat. No. STCTS70A) at –70 C until needed. Porcine Bile Juice Collection and Preparation The porcine bile juice was collected from pigs slaughtered at Nettles Sausage Company (Route 28, Box 97, Lake City, FL, 32055). Immediately after slaughtering, the entire gall bladder was removed from the animal. Gall bladders were placed in sterile bags and placed on ice. Removal of the gall bladders was performed as aseptically as possible. After collection, all gall bladders were transported on ice to the Microbiology Laboratory of the Department of Animal Sciences at the University of Florida. Using a sterile scalpel, the gall bladders were cut open and their contents poured into a large clean glass container. Bile juice from all gall bladders was pooled and thoroughly combined. Small aliquots of the bile juice were transferred to clean Tupperware–type plastic containers and frozen. Bile juice was lyophilized in the Food Science and Human Nutrition Department at the University of Florida. Samples required an average of 48 hours to totally dry out. Bile juice samples were again pooled, placed in sterile plastic bags and ground by hand while still inside the bags. The dry, ground samples were

PAGE 198

180 maintained frozen in labeled sealed plastic bags, inside a large desiccator until needed for media preparation. Treatment Preparation All solutions and culture media used in this study were prepared freshly at the University of Florida Meat Science laboratory. The pH of 0.1% peptone water was adjusted to 3, 4, 5, 6, or 7.4 with 1 N hydrochloric acid and 1 N sodium hydroxide. The initial idea was to acidify the enrichment broth instead of peptone water. This was not possible due to the fact that the hydrochloric acid when added to the medium curdled the blood rendering the medium useless. Fresh 0.1 M phosphate buffer was prepared by dissolving 77.4 mL of 1 M dibasic sodium phosphate (Na 2 HPO 4 , Fisher Scientific, cat. No. S369 500) and 22.6 mL of 1 M monobasic sodium phosphate (NaH 2 PO 4 , Fisher Scientific, cat. No. S374-500) into 1 L of diionized water. The experimental media consisted of Karmali base medium (Oxoid, Inc. Nepean, Ontario, Canada, K2G 1E8, cat. No. CM0935B) supplemented with 0, 6, 7, 8, or 9 mg freeze–dried porcine bile juice per 1 mL of medium; 0, 0.05, 0.1, 0.15, or 0.2 mg porcine gastric mucin (Sigma–Aldrich, St. Louis, MO, 63118) per 1 mL of medium; or a combination of both. A total of 25 different media formulations were obtained and they are described below (Table 29). Two days prior to the beginning of the trial, 90 test tubes containing 10 mL of Enrichment Broth supplemented with 5% lysed blood and filter–sterilized

PAGE 199

181 Table 29 Composition of experimental media. All supplements were added to Karmali base medium without antibiotics Supplement Medium number Porcine bile juice Porcine gastric mucin 1 — — 2 6 mg/mL — 3 7 mg/mL — 4 8 mg/mL — 5 9 mg/mL — 6 — 0.025 mg/mL 7 — 0.05 mg/mL 8 — 0.1 mg/mL 9 — 0.2 mg/mL 10 6 mg/mL 0.025 mg/mL 11 7 mg/mL 0.025 mg/mL 12 8 mg/mL 0.025 mg/mL 13 9 mg/mL 0.025 mg/mL 14 6 mg/mL 0.05 mg/mL 15 7 mg/mL 0.05 mg/mL 16 8 mg/mL 0.05 mg/mL 17 9 mg/mL 0.05 mg/mL 18 6 mg/mL 0.1 mg/mL 19 7 mg/mL 0.1 mg/mL 20 8 mg/mL 0.1 mg/mL 21 9 mg/mL 0.1 mg/mL 22 6 mg/mL 0.2 mg/mL 23 7 mg/mL 0.2 mg/mL 24 8 mg/mL 0.2 mg/mL 25 9 mg/mL 0.2 mg/mL

PAGE 200

182 (25 mm low–protein binding filters, Millipore, Burlington, MA, 01803, cat. No. SLGPR25LS) Campylobacter Selective Supplement (Bolton’s, Medox diagnostics, Ogdensburg, NY, 13669, cat. No. VL220) were inoculated with a loopful of Campylobacter jejuni subsp. jejuni ATCC 29428 and incubated microaerophilic at 37 C for 2 hours and then at 42 C for 46 hours. The Campylobacter Selective Supplement contains 10 mg cefoperazone, 10 mg vancomycin, 10 mg trimethoprim, and 25 mg cycloheximide. Each vial of supplement was filter–sterilized and added to 500 mL of sterile basal medium. On the first day of the trial, a small aliquot was transferred from each test tube to a disposable spectrophotometer cuvette and absorbance at 525 nm was measured and recorded. One milliliter of sample from each tube was plated on labeled pre–poured Karmali medium plates and incubated microaerophilic at 42 C for 48 hours to check for growth. The remaining of the sample (approximately 9.7 mL per test tube) was transferred to a sterile centrifuge tube and centrifuged at 5,000 rpm for 10 minutes. Supernatant was discarded and pellet was resuspended in 10 mL of phosphate buffer. Sample was centrifuged once more, supernatant was discarded and pellet was resupended in 5 mL of phosphate buffer and then transferred to 84 glass bottles containing 45 mL of sterile phosphate buffer each. Bottles were labeled and stored at 4 C in a small refrigerator in the microbiology laboratory (Figure 22, Step 1).

PAGE 201

183 Sample Storage All eighty–four 100 mL glass bottles containing 50 mL of Campylobacter jejuni in phosphate buffer were labeled and stored in a small refrigerator in the Meat Science Microbiology Laboratory. The temperature of the refrigerator was kept at 4 C throughout the study. Five bottles labeled from 1–5 were removed from the refrigerator twice a week and analyzed for Campylobacter jejuni colony forming units (Figure 22, Step 2). After microbial analysis, all five bottles were returned to the refrigerator for further usage. Two additional bottles were removed from the refrigerator on a weekly basis and all contents of the bottles were prepared for transmission electron microscopy analysis. Two more bottles were also removed from the refrigerator on a weekly basis, their contents were centrifuged for 5 minutes, resuspended in 1 mL of phosphate buffer containing 15% glycerol and frozen in 2.5 mL mini–centrifuge bottles at –70 C. Those samples were kept only as backup samples in the case there was a need for checking the cells at a later time. Microbial Analysis Samples from five labeled bottles were analyzed on days 1, 3, 8, 10, 15, 17, 22, 24, and 29. One milliliter of sample from each bottle was transferred to a test tube containing 9 mL of 0.1% peptone water from which 10 to 10 serial dilutions were prepared. The serial dilution was adjusted accordingly over time

PAGE 202

184 Step 1 Step 5 Step 2 Step 3 Step 4 Figure 22 Diagram of the experimental design

PAGE 203

185 as cell growth diminished each passing week. Initially, higher dilutions were prepared to assess the initial bacterial load in each bottle. A sterile cotton swab was dipped into the 10 to 10 serial dilution test tubes and fully saturated with the sample solution. The cotton swab was then used to spread one microliter of sample onto pre–poured Karmali agar plates for detection of Campylobacter jejuni colony forming units. All plates were done in duplicate. Again, plating of serial dilution was adjusted as needed throughout the study. Karmali plates were placed inside several anaerobe jars and brought to microaerophilic conditions with one 2.5 L or two 3.5 L Campy Gas Paks according to the size of the anaerobe jar. Jars were incubated at 42 C for 48 hours. Identification of Campylobacter cells were performed based on colony appearance and morphology, wet mount, catalase and oxidase tests. Campylobacter jejuni is catalase and oxidase positive. Campylobacter colonies on agar appear as small round, thick whitish translucent colonies. When a wet mount slide is made and examined on a phase–contrast microscope fitted with a 100X oil immersion objective, Campylobacter appears as long (1.5–5 m long) curved cells and demonstrate wiggly or corkscrew–type motility. Old, stressed cells appear as static long cells or they might have a coccoid shape. To perform the catalase test, a loopful of growth from a colony was placed in a drop of 3% hydrogen peroxide (H 2 O 2 ). If bubbles were released from the drop immediately after adding the cells, cells were considered catalase positive. To perform the oxidase test, a loopful of growth was rubbed on an oxidase test dry slide.

PAGE 204

186 Campylobacter jejuni cells were plated on Karmali plates for as long as colony forming units (cfu/mL) were seen in the plates. When direct plating was no longer supporting colony growth, cells were enriched for 48 hours at 42 C in Campylobacter Enrichment Broth prior to plating on Karmali plates (Figure 22, Step 3). Enrichment of samples continued until no colony growth was achieved. Samples were stored at 4 C for period of 3 months prior to being exposed to the acid shock treatment. Transmission Electron Microscopy Analyses Negative staining Fresh Campylobacter jejuni culture suspended in phosphate buffer (pH 7.4) was negatively stained with 1% aqueous uranyl acetate solution. A drop of bacterial culture was placed on parafilm, and a Formvar–coated (0.25% solution in ethylene dichloride, Electron Microscopy Sciences, Fort Washington, PA, 19034, cat. No. 15810) 300–mesh copper grid (Ted Pella, Inc., Redding, CA, 96049) was laid on top of the droplet (Formvar side down). By laying the grid on top of the drop containing the bacterial culture, only the organisms that have flagella and are still motile will swim and attach to the grid. After 5 minutes, the grid was dried with Whatman No. 1 filter paper and washed twice with distilled water, drying in between washes. One drop of 1% uranyl acetate solution (Electron Microscopy Sciences, Fort Washington, PA, 19034, cat. No. 22400) was added to the grid. After 1 minute, the grid was dried with filter paper and air–dried, covered in a Petri dish, for 15 minutes. Samples were examined with a

PAGE 205

187 Zeiss transmission electron microscope (EM 10C–A, Carl Zeiss, Oberkochen, West Germany, D–7082) at 80kv. Specimen preparation for sectioning Two 50–mL phosphate buffer bottles containing Campylobacter jejuni that had been stored at 4 C were opened at days 1, 3, 8, 15, 22 and 29 and their contents prepared for transmission electron microscopy (Figure 22, Step 5). Contents of each bottle were transferred to two 40–mL sterile centrifuge vials and centrifuged at 5,000 rpm for 10 minutes. Supernatants were discarded and pellets were resuspended in 1 mL of phosphate buffer and contents from each centrifuge vial were combined and transferred to a mini–centrifuge vial. Vials were refrigerated until the following day when specimen preparation was performed. Prior to fixation, samples were centrifuged at 8,000 rpm for 3 minutes to assure that a pellet of significant size was obtained. Samples for electron microscopy were fixed in 2% glutaraldehyde (made from 8% glutaraldehyde, Tousimis Research Corporation, Rockville, MD, 20847), 1% osmium tetroxide (Ted Pella, Inc., Redding, CA, 96049, cat. No. 18456) and 1% uranyl acetate (Electron Microscopy Sciences, Fort Washington, PA, 19034, cat. No. 22400). Following fixation, samples were dehydrated with ethanol and acetone and embedded in Spurr resin. Because Campylobacter jejuni cells were pelleted, the mini–centrifuge vial was vortexed after addition of each chemical. The process of vortexing resuspends the cells increasing their exposure to the chemicals.

PAGE 206

188 Furthermore, in order to avoid losing cells during removal of each chemical, the mini–centrifuge vial was centrifuged at the end of each step. The specimen preparation was divided in three steps: fixation, dehydration, and embedment. The fixation process consisted of a sequence of steps (sequential fixation). First, sample was fixed with 2% glutaraldehyde (0.25 mL 8% glutaraldehyde + 0.75 mL 0.2 M cacodylate buffer (made from cacodylic acid, sodium salt, Fisher Scientific, cat. No. BP325–50) for 30 minutes. At the end of 30 minutes, sample was washed in 3 changes of 0.1 M cacodylate buffer, i.e., sample was left in cacodylate buffer for 10 minutes at each change. When all glutaraldehyde was removed from sample a second fixation was performed with 1% osmium tetroxide (OsO 4 ) (0.25 mL 4% osmium tetroxide + 0.75 mL 0.2 M cacodylate buffer). Sample was submerged in osmium tetroxide for one hour at room temperature. After one hour, sample was washed in 3 changes of distilled water, submerged in distilled water for 10 minutes at each change. Finally, sample was fixed with 1% aqueous uranyl acetate for 30 minutes at room temperature. The fixed sample was then dehydrated in 25% ethanol for 10 minutes followed by 50% ethanol for 10 minutes, 75% ethanol for 5 minutes, 95% ethanol for 5 minutes, twice in 100% ethanol for 5 minutes, 100% acetone for 15 minutes and 100% acetone for 30 minutes. The dehydrated fixed sample was then embedded by infiltrating it with 30% Spurr resin for 1 hour followed by infiltration in 70% Spurr resin overnight.

PAGE 207

189 On the following day, the 70% resin was removed and sample was embedded in two changes of 100% Spurr resin for 1 hour each. The micro–centrifuge tube was placed in 60 C oven for 36 hours for resin polymerization to occur. The Spurr resin used during embedding was freshly made as needed with 62% noneyl succinic anhydride (NSA) (Electron Microscopy Sciences, Fort Washington, PA, 19034, EM grade, cat. No. 19050), 23% vinyl cyclohexene dioxide (ERL 4206) (Electron Microscopy Sciences, Fort Washington, PA, 19034, cat. No. 15000), 12% epoxy resin diglycidyl ether of polypropylene glycol (DER 736) (Electron Microscopy Sciences, Fort Washington, PA, 19034, cat. No. 13000), 0.95% 2–dimethylaminoethanol (DMAE) (Polysciences, Inc., Warrington, PA, 18976, cat. No. 1458), and 0.95% soya lecithin (Eckerd Drug Company, Clearwater, FL, 34618, item 594788). Addition of lecithin softens the resin, which in turn facilitates sectioning with a glass knife. After 36 hours in the oven, the mico–centrifuge tube was slashed; the polymerized resin block was removed from the vial and stored for sectioning. Sectioning methodology A polymerized resin block containing the fixed Campylobacter jejuni was carefully trimmed to remove excessive plastic with of a jeweler’s saw. The trimmed block was mounted on a stub and affixed with a strong adhesive (Super Glue). The stub was placed in the 60 C oven for 5 minutes. The block was placed in a block–trimming holder and placed under a dissecting microscope. Thin slices of plastic were removed with the help of a

PAGE 208

190 double–edge razor blade. The sides of the block were cut into an oblique shape in order to obtain a large base and a narrow top. The width and the length of the block face were measured with a ruler to make sure that the top was about 0.5 mm wide and no longer than 1 mm. Narrow tops provide better slices, but caution should be exercised when trimming the top to avoid removing excessive amounts of sample. Any plastic left in the block face was carefully removed. This is important because when the block is placed in the ultramicrotome, the knife will be slicing mainly sample, not just plastic. If the sample was too small, a razor blade was not used to trim the block face. Instead, the block was placed in a LKB 8800 Ultratome III microtome (Ultrotome Nova, LKB–Produkter AB, Stockholm–Bromma 1, Sweden, cat. No. 2188–011) and the extra plastic was trimmed off with a glass knife, thus avoiding excessive loss of sample material. The trimmed block was placed in an ultramicrotome chuck and clamped securely to it. The block chuck was then clamped into the specimen arm of the ultramicrotome in a way that the block face was vertically positioned, i.e., the long side of the block was in a vertical position. For sectioning, glass knives were produced with glass strips (6.4 x 25 x 400 mm, Alkar glass strips ultramicrotomy, Ted Pella, Inc., Redding, CA, 96049, cat. No. 8700) and a knife breaker (LKB 7800B Knifemaker, LKB–Produkter AB, Stockholm–Bromma 1, Sweden). A freshly made glass knife was placed in the ultramicrotome’s knife holder and the fluorescent light was turned on. The glass knife was secured in the knife holder by turning the holder’s securing screw.

PAGE 209

191 The knife’s angle was adjusted to 4–5. As needed, the knife holder was moved sideways, so the best part of the knife was used for trimming the block face. While looking from the side of the ultramicrotome, the knife holder stage was brought close to the block. When block and knife were relatively close to one another, the specimen arm was manually lowered so that half of the block face was below the knife’s edge while the other half was above it. The block was trimmed very carefully until a slice of the sample material was obtained. When the block face was totally flat, it was polished with the polishing knob until the block face was really shiny. Polishing is necessary to make the block face shiny and smooth. In addition to polishing, the block face was checked for knife marks. Knife marks appear as streaks in the block face. If a knife mark was observed, the block face was thoroughly polished until no more mark was seen. After polishing, the sides of the block, as well as the top and bottom, were trimmed. The width of the block face was checked once more to make sure that it was a little less than 0.5mm. The block was then ready to be sectioned. For sectioning, a new glass knife was made and a plastic glass knife boat (Ted Pella, Inc., Redding, CA, 96049, cat. No. 123–3) was affixed to it with molten dental wax. Knife and boat were placed in the ultramicrotome and secured tightly. With the help of an incandescent backlight, the specimen arm was moved until a shadow was seen in the block face. The shadow represents the distance between the block face and the knife. As the knife gets closer, the

PAGE 210

192 shadow gets smaller. Depending on the type of the light used in the ultramicrotome, either a bright reflection or a dark shadow is cast by the knife’s edge onto the block face when the two are very close together. Using shadow as an indication of the degree of closeness is very important because the knife must be as close as possible to the block face without touching it, which would damage the knife before thin sections of samples were obtained. The polishing step is very critical for the appearance of the shadow since poorly polished block face often results in terrible shadows. If the shadow cast on the block face was not even, it was adjusted by turning the knife sideways. If the right side of the shadow was higher than the left side, it means that the knife was closer to the left side than to the right side since the lower the shadow, the closer the knife is to the block. Once the shadow was very small and even, the incandescent light was turned off and the fluorescent light was turned back on. At this point the knife boat was filled up with distilled water and the sample started to be sectioned. Thicknesses of the sections were adjusted as needed based on the color of the sections seen through the ocular lenses. Acceptable thickness of the sections was in the 60 to 100 nm range. Sections were stretched with chloroform vapors prior to being transferred to a grid. Five to six ultra–thin sections were then placed on a Formvar–coated (0.25% solution in ethylene dichloride, Electron Microscopy Sciences, Fort Washington, PA, 19034, cat. No. 15810–25) 100–mesh copper grid (Ted Pella, Inc., Redding, CA, 96049).

PAGE 211

193 Section staining procedure Grids containing samples were post–stained with 5% uranyl acetate (Electron Microscopy Sciences, Fort Washington, PA, 19034, cat. No. 22400) followed by lead citrate. The lead citrate solution was made according to Reynolds (1963) and it contained 3.25% lead nitrate (Fisher Scientific, cat. No. L62–100), 4.28% sodium citrate (Polysciences, Inc., Warrington, PA, 18976, cat. No. 02121), 19.47% sodium hydroxide and 73% distilled water. Uranyl acetate is an excellent general stain producing high contrast and it is very suitable for high–resolution work. The uranyl acetate strongly stains nucleic acids for which it has high affinity, bonding to them probably via their phosphate groups. It also stains proteins, the free amino groups of which may be involved in the bonding process. The lead citrate solution also produces high contrast and stains many cellular and tissue components (Lewis and Knight, 1977). Each grid was floated on top of a drop of 5% uranyl acetate and stained for 15 minutes. Grids were then rinsed with distilled water and dried with a filter paper (P5 filter paper, Fisher Scientific, cat. No. 09–801B). When grid was dry, it was floated over a drop of lead acetate for 4 minutes. Grids were rinsed once more with distilled water and allowed to dry before being transferred to the electron microscope. Samples were examined with a Zeiss transmission electron microscope (EM 10C–A, Carl Zeiss, Oberkochen, West Germany, D–7082) at 80kv. All reagents used in this study were ultrapure TEM grade.

PAGE 212

194 Acid–Shock Treatment At the end of three months of storage, Campylobacter cells that had been stored at 4 C in phosphate buffer were acid–shocked (Figure 22, Step 4). Two samples (samples 2 and 4) containing Campylobacter jejuni cells that were used for the weekly plating were removed from the refrigerator and vortexed. One milliliter of bacterial solution was transferred to a test tube containing 9 mL of 0.1% peptone water adjusted to pH 3, 4, 5, 6, and 7.4 (Figure 21, Step 4). Test tubes were vortexed and placed in a 37 C water bath for 3 hours. A glass filtering tower system (Microana glass filter SPT 25mm, Fisher Scientific, cat. No. 09753E) was fitted with a sterile black polycarbonate 0.2mm, 25mm diameter membrane filter (Fisher Scientific, cat. No. GTBP02500). After 3 hours of incubation at 37 C in acidic peptone water, the contents of each test tube were filtered under vacuum through the filtering system. The glass tower was rinsed with 10 mL of 0.1% peptone water at pH 7.4. The membrane filter was removed from the filtering system with sterile tweezers and transferred to a test tube containing 10 mL of Campylobacter Enrichment Broth supplemented with 5% lysed horse blood. All broths were incubated at 42 C under microaerophilic conditions for 48 hours. Filtering system was cleaned and sanitized with 70% alcohol and sterile distilled water between samples. In addition to samples 2 and 4, the acid–shock treatment was performed on fresh Campylobacter cells. A loopfull of Campylobacter was suspended in 9 mL of sterile phosphate buffer. One milliliter of the bacterial solution was

PAGE 213

195 transferred to a test tube containing peptone water adjusted to pH 3, 4, 5, 6, and 7.4, incubated at 37 C for 3 hours and filtered through a 0.2 m membrane filter. The filter was transferred to the Enrichment broth supplemented with 5% lysed horse blood and incubated for 48 hours at 42 C under microaerophilic conditions. Incubated broths were plated onto Karmali plates and onto modified Karmali agar medium containing porcine bile, mucin or a combination of the two. A sterile cotton swab was dipped into the broth and used to streak two half–plates of each experimental media. Plates were incubated at 42 C under microaerophilic conditions and checked for growth after 48 hours of incubation. Results and Discussion Microbial Culturability Figure 23 shows the growth curves for all five Campylobacter jejuni samples. All samples contained Campylobacter jejuni ATCC 29428 cells that had been stored in phosphate buffer at 4 C. In the present study, cfu–based viability assessments were implemented in order to obtain readily interpretable estimates of potential inoculum levels of the bacteria during storage at low temperature. Furthermore, plate count is the generally accepted standard methodology for measuring cell viability (Rollins and Colwell, 1986). A number of methods based on metabolic activity (Rollins and Colwell, 1986, Federighi et al., 1998), maintenance of DNA (Lzaro et al. 1999), protein synthesis (Cappelier et al.,

PAGE 214

196 01234567048121620242832Storage dayCampylobacter jejuni counts (Log 10 cfu/ml) Sample 1 Sample 2 Sample 3 Sample 4 Sample 5 Figure 23 Number of Campylobacter jejuni cells in stationary phosphate buffer stored aerobically at 4 C. Arrow represents beginning of enrichment of samples 1, 2 and 4 in Campylobacter Enrichment Broth supplemented with 5% horse blood prior to plating. Data represents mean duplicate of Log 10 colony forming units/mL (cfu) that had been plated in Karmali medium and incubated for 48 hours in microaerophilic conditions

PAGE 215

197 2000), cellular respiration (Hazeleger et al., 1998) and maintenance of morphological structures (Hazeleger et al., 1995) have been proposed and tested to assess the viability of long–term stored cells, but so far none has been agreed upon as being the most suitable for C. jejuni. Although the viable but nonculturable state has been studied in C. jejuni (Rollins and Colwell, 1986; Federighi et al., 1998; Lzaro et al., 1999; Rowan, 1999), conflicting results have been obtained concerning their existence and the infectivity of putative viable but nonculturable forms in animal models (Saha et al., 1991; Cappelier et al., 1999). As expected, bacterial counts for Campylobacter jejuni decreased with storage time since the bacteria were deprived of nutrients and stored at a temperature below their minimal growth temperature. According to Hazeleger et al. (1998) and Chan et al. (2001), temperatures around 31 C were found to be the minimal temperature C. jejuni requires for growth. However, oxygen consumption, catalase activity, ATP generation, and protein synthesis can still occur in C. jejuni cells at temperatures as low as 4 C. Chemotaxis and aerotaxis are also observed at low temperatures. The initial bacterial load (Figure 23) in each sample was around 2x10 6 cfu/mL, with the exception of sample 2 which started with a slightly lower load (2x10 5 cfu/mL). Although the initial load was somewhat similar, the growth behavior experienced by each sample varied considerably over time.

PAGE 216

198 Interestingly, all samples could be divided into 3 groups: Group 1 containing samples 3 and 5; Group 2 containing samples 1 and 4; and Group 3 containing only sample 2. Bacterial counts for Group 1 (samples 3 and 5) decreased slowly over 31 days of plating. At day 3, a slight increase in numbers was seen, followed by a steady 1–log decrease per week. Since the counts never reached undetectable numbers during the experimental period, enrichment of samples was not necessary. Harvey et al. (1996) observed that plate counts of Campylobacter declined rapidly with storage time, but eventually stabilized, suggesting that although cells decline in culturability, they might retain some metabolical activity and viability after a period of adaptation. According to Harvey et al. (1996), this may reflect passage through a dormant stage of the bacteria, where specific changes in protein expression involving the synthesis of novel proteins and down–regulation of others occur. Samples 1 and 4 (Group 2) behaved very differently than samples from Group 1. The number of Campylobacter cells was similar for the first 3 days and decreased by approximately 5 logs after samples had been stored for only 1 week. At day 8, Campylobacter counts increased for 2 days and then started to decline on a steady pattern until day 22, when plating in solid medium was not sufficient to support growth of cells. At day 22, samples were enriched in Campylobacter Enrichment Broth supplemented with 5% lysed horse blood prior to being plated in solid medium. As samples were enriched, a 3–log increase in cell counts was observed for both samples at day 24. Enrichment allowed an

PAGE 217

199 extra 5 and 7 days of plating, for samples 2 and 4 respectively, before numbers declined again (Figure 23). Finally, sample 2 (Group 3) behaved in a totally unexpected manner. The initial bacterial load was slightly lower than Groups 1 and 2 and cell numbers declined sharply after 3 days reaching undetectable levels after only a week of storage (Figure 23). Since this sample was behaving so differently than the others, bacterial counts were monitored for another 2 weeks before enrichment of sample was performed. At day 22, sample 2 was enriched in Campylobacter Enrichment Broth for 48 hours prior to plating. Despite the fact that cell numbers for sample 2 declined at a rapid rate on the first week of storage, reaching undetectable levels, bacterial counts had a similar pattern as samples from Group 1 (samples 3 and 5) having a 2–log increase after enrichment. After three months of storage at 4 C, a sample from Group 2 (sample 4), Group 3 (sample 2) and a fresh culture of Campylobacter jejuni were selected for the acid–shock treatment. The fresh Campylobacter jejuni culture was used as a control sample. Those samples were chosen because the objective of the experiment was to determine the effects of acid shock and plating on media containing bile and mucin on resuscitation of cells that were no longer culturable, even after enrichment. Since samples 2 and 4 had reached undetectable levels after enrichment, they could be used to test the hypothesis that cells that are no longer culturable would became culturable after being exposed to an acid treatment.

PAGE 218

200 Furthermore, those samples were less likely to show a “tail–section” growth after exposing them to an acid–shock. In 1999, Rowan observed that some enteric pathogens frequently exhibited an initial exponential kill where the number of culturable cells decreased drastically followed by a “tail section” indicating that some of the microorganisms present in the bacterial population are more resistant to the environmental stress than others. This “tail–section” growth is very important for the food industry because the infective dose for some of the most important foodborne pathogens, such as Listeria monocytogenes and Escherichia coli O157–H7 might be as low as 50 cells. The infective dose for Campylobacter is believed to be higher than E. coli. However, less than 5 X 10 2 cfu of C. jejuni has been shown to be sufficient to cause illness in an average person (Jeffrey et al., 2000). The samples chosen allowed for a study of the effects of acid shock on C. jejuni populations that had been severely stressed by prolonged storage and starvation. With the exception of sample 2, all samples followed a similar trend, wherein after 3 days of storage, ability to grow on Campylobacter culture medium was reduced until cells were 8 days old. From day 8 to day 10, cell population increased in numbers before viability resumed its declining pattern. Several authors have described the growth pattern of Campylobacter species over a prolonged period of time as constant loss of viability throughout the storage period regardless of storage temperature (Rollins and Colwell, 1986; Jones et al., 1991; Hazeleger et al., 1995; Federighi et al., 1998; Lzaro et al., 1999).

PAGE 219

201 In several of those studies, viability increased only after an enrichment step was added to the plating methodology. Nonetheless the recovery in viability without enrichment of samples seen in the present study is not a common finding. Chan et al. (2001) studied 19 field isolates of C. jejuni stored for 14 days at 4 C. Three out of the 19 strains revealed similar survival curves as the ones observed in the present study. The viable counts declined after 2 days of storage, then started to rise reaching their peak at day 6, decreasing the counts afterwards. Based on the survival curves, the authors concluded that the rise in viable counts was solely due to an adaptation of the cells to low temperature. This explanation however does not take into consideration the countless studies published on the inability of Campylobacter to grow at such low temperature (Rollins and Colwell, 1986; Hazeleger et al., 1998). Furthermore, if cells were becoming adapted to low temperatures, viable counts should not decline as time elapsed since deprivation of nutrients was not an issue in their study—bacteria were stored on a nutrient–rich environment. As already stated, it has been extensively published in the literature that Campylobacter cells do not grow at temperature as low as 4 C but they survive. This assumption arises from the fact that no signs of viability are observed when cells are plated on agar medium but metabolic and physiologic signs such as oxygen consumption, catalase activity, ATP generation, chemotaxis, aerotaxis and protein synthesis are still observed (Hazeleger et al., 1998). If campylobacters cannot grow at 4 C, why did the viable counts increase in this

PAGE 220

202 study as well as in the study conducted by Chan et al. (2001)? As stated and observed by Hazeleger et al. (1998), Campylobacter, as a species, exhibit extensive genotypic and phenotypic variability, and its survival has been shown to vary tremendously among different strains. It might be possible that only a few strains could grow at 4 C while the majority of strains could only survive. It is not believed that the increase in viability observed in four out of the five samples analyzed in this study was due exclusively to an adaptation to low temperatures. Most likely the increase in bacterial population observed from day 8 to day 10 was caused by nutrients that became available in the medium. The fast decline in bacterial numbers from day 3 to 8 was due to a rapid killing of bacteria. As they died, a large amount of nutrients that were inside the cells dispersed into the medium, becoming a source of nutrients for the remaining cells that were more adapted to the low temperature. Therefore, the nutrients served as fuel for the surviving bacteria. This would aid in explaining the increase in cell numbers after such drastic drop in viable counts. One of the characteristics of the logarithmic phase of a bacterial curve is the exponential growth of the organism. During this phase, microorganisms grow and divide at their maximum rate. Eventually, population growth ceases and the stationary phase begins, followed by the death phase. It is believed that microbial populations end the logarithmic phase when nutrient is severely depleted and when there is a large accumulation of toxic waste products (Prescott et al., 1999).

PAGE 221

203 Actually, accumulation of toxic waste seems to be the limiting factor for the growth of many microaerophilic and anaerobic cultures, which are largely sensitive to peroxides and reactive oxygen compounds released into the medium. In the case of Campylobacter, a consequence of their aerobic metabolism is the formation of these toxic products (Grant et al., 1996). After the stationary phase, microbial populations enter the death phase, which like the exponential phase, is usually logarithmic because a constant proportion of cells die every day. Although the death phase is characterized by its exponential trend, the speed in which cells die decreases after the population has been drastically reduced, probably due to survival of resistant cells (Prescott et al., 1999). Acid–Shock Treatment The results in this study as related to acid shock indicated that other factors should be taken into consideration when observing the effects of low acidic environments on the survival of Campylobacter jejuni. The interaction among different components of the gastrointestinal tract, the general homeostasis of the system and the gastrointestinal transit of fluids and food play a role in providing a unique environment to the invading bacteria. All these aspects of the human gastrointestinal system must work together in an efficient manner. Koo et al. (2001) at Mississippi State University designed a mechanical gastrointestinal model to study the survival of Vibrio vulnificus on the human gastrointestinal tract. The gastrointestinal model contains a gastric compartment, an intestine compartment and a peristaltic and a cassette pump as well as an

PAGE 222

204 Isotemp immersion circulator. Simulated intestinal fluid, hydrochloric acid (HCl), sodium bicarbonate (NaHCO 3 ) and bile solution are introduced slowly into the system simulating the environment that a bacterium would encounter as it reaches the human gastrointestinal tract (Koo et al., 2001). When working with the model, the researchers observed that gastric emptying rate might be more important in Vibrio vulnificus infections than acid tolerance. If gastric emptying occurs in less than 30 minutes, viable cells are readily delivered to the small intestine where they can multiply. Experimental Media The in vitro responses observed in this study with regards to the bacterial response to media containing bile, mucin or a combination of both are not consistent with the initial hypothesis of the study. The bacteria would normally encounter these compounds when entering a host. Various factors or conditions, including pH, temperature, anaerobiosis, presence of bile, prolonged storage, and osmolarity, are known to affect a wide variety of bacterial functions (Pace et al., 1997). All of these conditions may be encountered in a host by the bacterial pathogen. In addition, many compounds specific to the host may serve as signals to the bacteria. Jones et al. (1991) observed that the definition of non–culturability depends largely on the efficiency of the media at recovering injured or dormant organisms. One of the hypotheses of the present study stated that recovery of C. jejuni stored at low temperature for a prolonged period of time would be

PAGE 223

205 possible if porcine gastric mucin was added to the agar culture medium. Mucin fucose residues are thought to be chemotactic for C. jejuni (Meinersmann et al., 1991; Hugdahl et al., 1988). Like any other enteric pathogen, Campylobacter must penetrate the protective mucus biofilm that overlies the gastrointestinal epithelium before it can adhere or invade mucosal cells and cause infection. Sylvester et al. (1996) found that C. upsaliensis adhered to purified human small intestine mucin in vitro. According to the researchers, specific adherence to mucin receptors by organisms having the peculiar corkscrew motility, typical of Campylobacter species may actually enhance penetration of the mucus barrier by allowing the bacteria to dock in place before they bore through the mucus to reach the underlying epithelial cells. Bradshaw et al. (1994) observed that pure cultures of oral bacteria, which depend on the ability to utilize glycoproteins and mucins for growth, were able to grow in complex medium only if 0.25% of gastric porcine mucin was added to the medium. Purified gastric mucin of porcine source was selected for this study because of the availability and practicability of such compound. Since the whole idea of the study was to improve the methodology for culturing C. jejuni in the laboratory, it was important to use a source of mucin that was of easy access as to facilitate its use as a new medium supplement. Purified porcine gastric mucin can be purchased at a very reasonable price from several major chemical suppliers. Furthermore, swine (or hog) gastric mucin has been extensively used as model substrates for studies on the role of resident microflora in the

PAGE 224

206 catabolism of complex, endogenous molecules because its olygosaccharide structure is similar to that of many human glycoproteins (Ktyi, 1990; Corfield et al., 1992; Bradshaw et al., 1994). Sylvester et al. (1996) also observed in their study that C. upsaliensis binds phosphatidylethanolamine and gangliotetraosylceramide. The gangliotetraosylceramide present in the small intestine epithelium derives from the cleavage of sialic acid residues of membrane Gm1 ganglioside by neuraminidases of the normal flora. The researchers concluded that both components of the intestinal epithelium are accessible to Campylobacter and that the bacterium has the ability to interact with them in vivo. An artifact of current microbiological practice is to measure the effects of inimical processes using laboratory rich culture media where the challenge microorganisms are commonly assessed during their exponential growth phase. To exemplify the richness of culture media and laboratory settings, Archer and Hopkins stated: “although life in a Petri plate with a few simple sugars and a handful of amino acids might not seem like much, to a bacterium it’s a night in Trump Tower with a breakfast in bed” (quoted by Rowan, 1999). In reality, bacteria must adapt to multifaceted hostile environments, whether in food or during passage through the low pH environments of the stomach. This is why the Campylobacter cells in this study were first starved and then acid–shocked prior to plating in rich culture medium. The starvation assures that the cells are not in exponential growth phase and the acid–shock

PAGE 225

207 mimics one of the hostile environments that cells must undergo when inside the host. Bacteria in the stationery growth phase are more resistant to hurdles than bacteria in the exponential phase. Pace et al. (1997) found that the addition of bile or bile deoxycholic acid to nutrient–deprived Vibrio parahaemolyticus cultures led to an increase in the direct viable counts as well as colony counts among virulent strains. Fox et al. (1996) observed that several species of Helicobacter (H. hepaticus, H. bilis, H. pullorum, and H. canis) were able to grow on medium containing 1.5% desiccated ox bile. Electron Micrographs The basic shape of C. jejuni cells was helical (Figure 24, Figure 25, Figure 26, and Figure 27) and cells showed a range of cell lengths and diameters (Figure 28). The general spiral shape characteristic of Campylobacteraceae was observed as well as the presence of a single, polar flagellum, which could be observed at one or both ends. The helical Campylobacter morphology observed in this study was similar to those reported in the literature (Rhoades, 1954; Rollins and Colwell, 1986; Moran and Upton, 1987). Figure 28 shows negative stained Campylobacter cells at day one. The electron micrographs reveal the characteristic spiral shape of the cell and the polar flagella. The flagella in Campylobacter were quite resistant, remaining intact and attached to the cell even during fixation procedures for transmission electron microscopy. However, it is important to note that not all cells retained the

PAGE 226

208 Figure 24 Helically shaped C. jejuni on Day 1. 25,200x (scale bar = 1 m) Figure 25 Campylobacter jejuni on Day 1 presenting different morphological shapes. 31,600x (scale bar = 1 m)

PAGE 227

209 Figure 26 Spiral C. jejuni. Showing flagellum (F), flagellum insertion point (FIP), polar membrane (Pm), cell membrane (CM), outer membrane (OM), and periplasmic space (PS). 63,400x (scale bar = 0.5 m) F PM FIP OM F CM PS Figure 27 Coccoid and spiral shaped C. jejuni on Day 22. Showing flagellum (F). 29,900x (scale bar = 1 m)

PAGE 228

210 F F A B D F C Figure 28 Different sizes of negatively stained flagellated (F) C. jejuni. A, 17,900x (scale bar = 1 m); B, 6,100x (scale bar = 1 m); C, 6,100x (scale bar = 1 m); D, 11,600x (scale bar = 1 m)

PAGE 229

211 flagellum during staining and fixation, or during prolonged periods of storage. The Campylobacter cells shown on the negative staining electron micrographs retained their flagella because of the preparation method employed. A 300–mesh copper grid was positioned on the top of a drop containing the bacteria sample. Only cells that have flagella are able to swim to the top of the drop and stick to the Formvar on the grid. The vast majority of cells that were unable to move remained at the bottom of the drop. The cell wall over most of the surface of the cell was typical of Gram–negative bacteria. Figure 26 shows the bacterial cell wall with two electron–dense layers (cell membrane and outer membrane) separated by an electron–transparent layer. The outer electron–dense layer corresponds to the outer membrane; the inner electron–dense layer represents the cell membrane, whereas the electron–transparent layer is the periplasmic space. Figure 26, Figure 30, and Figure 31 show the presence of a polar membrane underlying the cytoplasmic membrane at the Campylobacter cell poles. Interestingly, the polar membrane was visible only at the ends that had a flagellum; non–flagellated poles ended with a round shape and had no polar membrane. Similar description of polar membrane has been described in the literature for all species of Aquaspirilum and in other genera of helical bacteria, such as Spirillum, Oceanospirillum, Ectothiorhodospira, Rhodospirillum and even

PAGE 230

212 Figure 29 Spiral C. jejuni. 22, 800x (scale bar = 1 m) FIP OM CM F F PM Figure 30 Polar end of C. jejuni. Showing flagellum (F), flagellum insertion point (FIP), polar membrane (PM), outer membrane (OM), and cell membrane (CM). 124,600x (scale bar = 0.1 m)

PAGE 231

213 PM PM Figure 31 Polar membrane (PM) of helical C. jejuni. 101,500x (scale bar = 0.2 m)

PAGE 232

214 Campylobacter (Krieg, 1984). However, there are no electron micrographs in the current literature showing the polar membrane in Campylobacter jejuni. The electron micrographs of sectioned C. jejuni showing the polar membrane in Figure 30 and Figure 31 are remarkably identical with those presented by Murray and Birch–Andersen (1963), and Krieg (1984) of Aquaspirillum serpens. The polar membrane is considered an unusual elaboration of the plasma membrane that is attached to the plasma membrane by bar–like links of varying small dimensions (Murray and Birch–Andersen, 1963). As can be seen in Figure 26 and Figure 30, the polar membrane is located, most commonly, in the region surrounding the polar flagella. Occasionally however, the polar membrane was located at a non–flagellated end (Figure 31). The presence of a polar membrane has also been shown in Aquaspirillum fasciculus and Oceanospirillum spp. (Krieg, 1984). Ng et al. (1985) reported the presence of “thick bands” at the flagellar poles of Campylobacter cells analyzed by scanning and transmission electron microscopy. Based on the electron micrographs presented on their study, it is clear that the thick bands they observed were in fact the polar membranes described in this study and the one by Murray and Birch–Andersen (1963). Murray and Birch–Andersen (1963) reported that the polar membrane works as a washer or a disc, with a hole in the center somehow related to the region where the flagellum enters the cell. The function of a bushing (if the

PAGE 233

215 researcher meant a bushing, not really a washer) is to give support to coupling of different parts of a structure, especially when there is any type of movement involved in this junction, like in the case of a flagellum rotating inside a central rod (White, 1995). Bushings in flagellar region have been described and their function is to protect the insertion point of the flagellar motor with the outer and cell membranes. Instead of comparing the polar membrane with a washer, it would be more appropriate to compare the polar membrane to a stator. The bacterial flagellar system has been largely investigated and most of its general structure has been defined. However, based on the literature reviewed, there is not an agreement on the presence and location of a stator in the flagellar system. This study provides a possible explanation for the function of the polar membrane in flagellated bacteria. The polar membrane could work as a stator of the flagellar motor, being the stationary part of that gives support to the rotating structures. This assumption is based on the statement of White (1995), in which he cited that the flagellar motor needed to have both a rotor and a stator to function: because of the large mass of the [flagellar] filament and the significant viscous drag that it encounters, the stator must be attached to a structural element sufficiently massive so as the preclude the stator from spinning in the membrane while the filament barely moves. It is usually assumed that whatever functions as [sic] the stator is firmly anchored to the peptidoglycan. (White, 1995, Page 9) According to Smibert (1984), the outer membrane of Campylobacter cells is double layered, loosely fitted over the cell wall, and has a wavy morphology.

PAGE 234

216 The cytoplasmic membrane is thicker at the polar region as compared to the body portion of the cell because of the presence of the polar membrane, which can be seen at both ends of the cell. The general structure, as well as the true role of the polar membrane in Campylobacter cells, remains to be investigated. Figure 26 and Figure 30 show the flagellum insertion point. As it can be seen, there was a tendency for a rather characteristic shape of the polar end of the cell that has a flagellum. When the flagellum was inserted at the polar end of the cell, a depression was formed; non–flagellated polar ends were blunt and rounded. Nevertheless, they contained the polar membrane on both sides of the end. Murray and Birch–Andersen (1963) observed that some polar ends of the Spirillum serpens exhibited similar depression. However, the actual exit of flagella was not necessarily from the center of the depression, but could be on the side. A very fortunate image of the flagellum insertion point show quite clearly that the flagellum passes through the cell wall and through the plasma membrane (Figure 30). As expected, samples revealed the expected gradual transition from spiral shape to a predominantly coccoid form that is usually seem in both Spirillum and Vibrio species (Fearnley et al., 1996; Harvey et al., 1996; Lee et al., 1997). However, transition to a coccoid form over the period of one month was not as dramatic as cited by several authors. The speed at which Campylobacter cells become coccoid vary largely among different strains. Ogg (1962) observed that 100% of the cells

PAGE 235

217 of two strains of C. fetus (formerly Vibrio fetus) were coccoid after 96 hours, whereas less than 0.1% of two other strains were spherical in shape. Contrary to reports from some researchers, cell morphology did not change as dramatically over the course of the study (Ogg, 1962; Rollins and Colwell, 1986; Moran and Upton, 1987). The main difference observed over time was the increased numbers of ghost and dying cells, not the large number of coccoid cells as depicted in the literature. One possible reason for the large number of coccoid cells seen in transmission electron micrographs could be that spiral cells were sectioned crosswise, thus resulting in a misleading coccoid cell. The occurrence of coccoid cells in the genus Campylobacter has been known for some time. However, the intermediate steps that the cell undergoes during the morphological transformation are not well known. Under unfavorable conditions, typical spiral Campylobacter cells undergo a morphological transformation becoming coccoid in appearance (Rollins and Colwell, 1986; Harvey et al., 1996; Lee et al., 1997). Ng et al. (1985) reported ring– or donut–shaped cells of Campylobacter and suggested that those forms were part of the progressive morphological change associated with the Campylobacter species. The reasons for this morphological change have been extensively investigated and it was believed to include light, medium components, atmospheric oxygen tension, and toxic oxygen derivatives in the medium (Rollins and Colwell, 1986; Moran and Upton, 1987; Fearnley et al., 1996; Harvey

PAGE 236

218 et al., 1996). However, results have indicated that many individual factors have no influence on cell transformation. The viability of the coccoid cells has been also investigated. Several techniques were used to check culturability and viability, but results were not conclusive. For some researchers, coccoid cells are merely a degenerative or dormant state. It is important to understand the causes of this morphological transition and loss of culturability on laboratory media because there are evidences that virulence of Campylobacter is enhanced with cellular changes. Harvey et al. (1996) reported that the ability to invade and enter mammalian enterocytic cells was enhanced after C. jejuni had gone through morphological transformation. The researchers concluded that environmentally stressed cells might become less detectable by standard microbiological techniques while at the same time become a greater health risk. The present study provides a possible explanation of how the coccoid cells could arise. Figure 32 show the various morphological forms observed in Campylobacter jejuni cells stored in phosphate buffer for 29 days at 4 C. It is believed that a shortening and rounding of the helical cell, until a spherical shape is obtained, leads to the formation of coccoid cells. During the transition from one shape to another, club–shaped cells are formed due to the thickening and rounding of one of the ends. According to Williams and Rittenberg (1956), and

PAGE 237

219 E F C D A B CM CM Figure 32 Morphological transformation of helical C. jejuni to coccoid. Arrow shows duplicated cell membrane. Figures A and B show helically–shaped bacterium. Figure C, D, and E show club–shaped bacterium. Figure F shows coccoid bacterium. A, 18,300x; B, 28,800x; C, 36,600x; D, 28,300x E, 28,300x; F, 28,300x (scale bars = 1 m)

PAGE 238

220 CM Figure 33 Duplicated cell membrane (CM) of C. jejuni during transition from spiral to coccoid shape. 55,700x (scale bar = 0.5 m)

PAGE 239

221 Williams and Rittenberg (1957), this is one of the approaches helical Spirillum bacteria employ to form coccoid cells. Several electron micrographs of what seem to be individual intermediate stages of the morphological transformation that occurs in the Campylobacter genus have been published in the literature. However, those intermediate stages had not been combined in a single study in an attempt to show the transformation process. Analysis of the electron micrographs reported by Moran and Upton (1987) and Ng et al. (1985) showed some club–shaped Campylobacter cells, similar to the ones reported in this study. The mode of formation of coccoid cells is not completely known. Much of the work on the morphological changes of helically shaped bacteria has been done on Spirochetes, Vibrio, Oceanospirillum, and Aquaspirillum. As with Campylobacter, most of the helically shaped bacteria transform into “spherical bodies” under unfavorable conditions and as culture ages (Williams and Rittenberg, 1956; Williams and Rittenberg, 1957; Canale–Parola, 1984; Krieg, 1984; Johnson and Faine, 1984). However, the spherical forms of the bacteria are often regarded as nonviable. Figure 32 shows that regardless of the bacterial shape, the cell wall has the same general structural features. However, during what is believed to be the intermediate steps of the morphological transformation, a short length of duplicated cell membrane can be seen (Figure 32, images D and E; Figure 33).

PAGE 240

222 Similar duplication of plasma membrane was observed in Vibrio marinus and streptomycete (Moore and Chapman, 1959; Felter et al., 1969). Williams and Rittenberg (1956) reported several possible ways that a helically shaped Spirillum lunatum could change to a coccoid form of that formation of coccoid forms—by contraction of the entire cell, protuberance formation; fusion of entwined forms, and ring formation. A year later, Williams and Rittenberg (1957) described the modes of formation of spherical forms in Spirillum ehrenberg. Contraction of the entire cell is the easiest mode of transformation. The helical cell undergoes gradual shortening and rounding until a coccoid body is formed (Williams and Rittenberg, 1956; Williams and Rittenberg, 1957). The sequential steps of the transformation of helical C. jejuni to a coccoid cell (Figure 32) agree with the description of cell contraction. Therefore, it is possible that this is how Campylobacter changes its morphology from spiral to coccoid when under stress. Clark–Walker (1969) observed that during the morphological transformation, the helical Spirillum itersonii started to present a regional swelling, mostly at the end but sometimes in the middle of the cell. Shortly after the swelling had started, cells completely lost the helical form assuming a spherical shape. When morphological transformation occurs by protuberance formation, the helical cell become shorter and thicker and a protuberance arises from the center or from each end of the cell. The protuberances enlarge and eventually

PAGE 241

223 merge into a single round body as the helical cell is absorbed (Williams and Rittenberg, 1956; Williams and Rittenberg, 1957). Fusion of entwined forms is another way that coccoid cells can be formed. In this case, two individual cells twist about the other end until they fuse. The cells become shorter and thicker and a protuberance develops at the point of fusion. This gradually enlarges and absorbs the organisms forming the coccoid cell. When the two cells fuse, a rearrangement of the chromatic material occurs (Williams and Rittenberg, 1956; Williams and Rittenberg, 1957). Coccoid cell transformation by ring formation mode occurs when a cell bends, as if about to divide. However, instead of dividing, a ring is formed in the central portion of the cell with the two ends crossing each other. Eventually, these two end portions are absorbed and the ring gradually closes with the formation of a spherical body (Williams and Rittenberg, 1956; Williams and Rittenberg, 1957). The formation of degenerative coccoid cells is believed to occur in a similar manner, in which the cell starts to enlarge on one end forming a club shape. The cell wall starts to separate from the cytoplasm and it begins to stretch out. As the cell wall stretches, it loses its integrity and cytoplasmic material leaks out causing the cell to quickly degenerate. This degenerative state can be accompanied by bleb formation (Ng et al., 1985) or not. By the end of the morphological and physiological transformations, the cell becomes a round, empty cytoplasm with a stretched cell membrane.

PAGE 242

224 Baker and Park (1975) reported that formation of coccoid cells of Vibrio NCTC 4716 was the result of attack of the peptidoglycan by an autolytic endo–N–acetylhexosaminidase and an autolytic endopeptidase. When the peptidoglycan loses its integrity, the cells become spherical and the coiled layers of the cell wall straighten and expand away from the cytoplasmic membrane. When this happens, the cell wall pulls the remaining peptidoglycan to which they are linked, resulting in a non–viable coccoid cell. Williams and Rittenberg (1956), and Williams and Rittenberg (1957) were also able to demonstrate that the coccoid forms of Spirillum lunatum and Spirillum ehrenberg are not a degenerative state of the bacterium. Upon transfer into fresh medium, the coccoid cells “germinated” by unipolar or bipolar growth of a helical cell, with the former being absorbed into the developing helical cell. Rollins and Colwell (1986) noticed that C. jejuni cells that had been held at 4 C in microcosms showed a transition from the typical spiral shape to a coccoid form and that this transition occurred faster if cells were held at 37 C. They observed through scanning and transmission electron microscopy that the majority of coccoid cells maintained an apparently intact, though asymmetric, membrane structure. Also, examination of the nonculturable, predominately coccoid forms by electron microscopy, showed that cell shape and size varied significantly. It is important to note, however, that the remarks made by Rollins and Colwell (1986) regarding a large variability of sizes and shape of cells can be

PAGE 243

225 misleading. During sample examination, a mixture of different shapes was observed. However, the shape of a bacterium does not necessarily represent its real form, but the angle at which the bacterium was sectioned. A spiral shaped bacterium can appear as a half–moon shaped cell or a club–shaped or even a donut–shaped; it all depends on the position of the cell in the sample and the angle of sectioning. The variety of shapes obtained according to the angle of sectioning is more profound in bacteria with spiral shape such as Campylobacter when compared to bacteria that the whole cell is located in only one plane such as bacilli or coccoid cells because the larger the number of planes of a cell, the larger the variability of shapes obtained during sectioning. Rollins and Colwell (1986) noticed that transmission electron microscopy of the Campylobacter cells often revealed a condensed cytosol in these cells. The authors were not sure if the condensed cytosol represented a survival mechanism or an artifact of fixation. A continuum of morphological types, with predominance of spiral and coccoid forms, depending on the stage of growth was also observed. Most of the work published on the morphological transformation of Campylobacter is based on the changes that occurred in a short period of time, i.e., one week. If the majority of changes in bacterial shape occurs during the first few days of aging, it is common to extrapolate the data and assume that all cells will undergo such transformation as time passes. Nevertheless, this assumption did not hold true in this study because a large number of spiral cells were

PAGE 244

226 observed after 29 days of storage under refrigeration temperature in a nutrient–deprived environment. The large number of spiral cells observed after such a prolonged storage time (29 days of storage) is in accordance with the recovery of cells by plating in solid medium observed in the present study. The growth curves showed on Figure 23 indicate that Campylobacter jejuni stored at refrigeration temperature (4 C) in a nutrient–deprived environment for 29 days can survive and remain culturable using standard microbiological methods. Examination of more than 40 electron micrographs and analysis of 12 to 15 thin sections of duplicate samples per day of storage of C. jejuni lead to the conclusion that the number of coccoid cells increased slightly with storage time, whereas the number of helical cells decreased. However, the reduction in numbers of spiral cells was not as considerable as previously expected. The numbers of spheroplasts observed in the samples increased as the storage time increased. At the end of 29 days, a large number of cells were spheroplasts. On day one, transmission electron microscopy of Campylobacter cells stored in phosphate buffer at 4 C revealed mainly helical and coccoid cells. A few ghost cells or spheroplasts were observed, but the number was small. The vast majority of cells presented the characteristic helical shape of Campylobacter. Furthermore, a few long helical shaped cells (4 to 5 times longer than typical cells) were observed.

PAGE 245

227 After 8 days of storage, the numbers of ghost cells and spheroplasts increased. A large number of helical and coccoid cells were observed, in similar proportion. Several parts of loose membrane were present in the samples, probably from cells that underwent plasmolysis. A problem with quantifying the proportion of coccoid and helical cells to ghost cells and loose membranes is the lack of a standardized methodology. During preparation of the samples for electron microscopy, samples are centrifuged several times. Upon centrifugation, cells segregate in different layers with intact cells combining at the bottom of the centrifuge tube, followed by ghost cells and membranes. Therefore, depending on the location the sections were taken, more of one type of cell will be observed than others. On day 15, helically shaped and coccoid cells were equally observed. Some of the spiral cells had pieces of flagella attached to the polar end of the cell. A large number of cells showed a less dense cytoplasm as compared to the cells from the previous days. The number of spheroplasts remained high and the membrane of some cells appears to be detaching from the cytosol. At the end of 29 days of storage, the number of helicall–shaped cells decreased slightly while the number of coccoid cells increased. However, a considerable amount of spiral cells were present in the samples. The presence of shorter and thicker spiral cells was noticed. Again, a large number of spheroplasts were observed. Jones et al. (1991) observed a large number of bacterial debris among spiral cells in a bacterial suspension that was 2 weeks old.

PAGE 246

228 In addition, most of the spiral cells showed protoplast shrinkage away from the outer envelope, mostly at the polar ends. Figure 34 shows Campylobacter cells from late stationary phase. Baker and Park (1975) published electron micrographs of Vibrio cells that have a remarkably identical morphology with the electron micrographs of C. jejuni presented in this study. Thin sections of the coccoid cells showed the outer membrane detaching from the cell (Figure 35, DM). Some of the large coccoid cells were found to contain more than one cell units inside the cell wall (Figure 35, 2), similarly to the findings of Felter et al. (1969) with V. marinus. In the work of Rollins and Colwell (1986), occasional spheroplasts (a bacterial cell whose cell wall is absent or deficient, causing it to have a spherical form) and “ghost” cells could be seem and these forms were believed to be nonviable. Williams and Rittenberg (1957) described the formation of extremely large and delicate Spirillum cells called “giant cells.” Apparently, in spite of the “giant cell” bearing similarities to large bodies observed with other bacteria, occurrence of this type of cell is spontaneous, meaning that no special agent is needed to induce such formation. Figure 35 (S) shows one spheroplast identical to that of Desulfovibrio aestuarii presented by Levin and Vaughn (1968). Spontaneous spheroplast formation was observed in Desulfovibrio aestuarii as the culture aged (Levin and Vaughn, 1968). Spheroplasts originate when a lateral bleb is formed, expanded

PAGE 247

229 Figure 34 Campylobacter jejuni on Day 22 of storage. Showing spiral and coccoid cells, and spheroplasts. 3,900x (scale bar = 1 m) DM S 2 Figure 35 Campylobacter jejuni spheroplast (S), cell with detaching membrane (DM) and coccoid cell containing two cell units (2). 18,700x (scale bar = 1 m)

PAGE 248

230 and pulls the cytoplasmic membrane out from the protoplasm, until it acquires a spherical shape. Advanced stages include separation of the cell wall from the cytoplasmic membrane. Spontaneously formed spheroplasts, i.e., formation not induced by glycine or penicillin, are believed to be non–viable and loss of viability seems to precede spheroplast formation (Levin and Vaughn, 1968). Felter et al. (1969) observed a large number of giant cells, packed with ribosomes, nuclear material, and membranous structures in old cultures of Vibrio marinus. Some large cells consisted of two cells enclosed by a single cell wall (Figure 35, 2), each bounded by a separate and intact cell membrane. Conclusion Campylobacter is a sui generis organism because it is very sensitive to a variety of environmental stresses, yet it is the second leading cause of gastroenteritis in the United States of America (Centers for Disease Control and Prevention, 2002). An estimated 2.4 million persons are affected with Campylobacter alone each year. The initial average population of C. jejuni was approximately 2x10 6 cfu/mL. The bacterial population decreased over 31 days of storage in a nutrient–deprived environment at refrigeration temperature (4 C). Enrichment of the samples with Campylobacter Enrichment broth after 22 days of storage increased culturability of cells. The loss of culturability observed in the study

PAGE 249

231 herein was similar to those reported in the literature by other researchers (Rollins and Colwell, 1986; Hazeleger et al., 1995; Federighi et al., 1998; Lzaro et al., 1999). Contrary to the hypothesis of the study, treatment of the cells with various acidic pH (3, 4, 5, and 6) after 3 months of storage did not induce the recovery of nutrient–deprived C. jejuni thus indicating that other factors should be taken into consideration when observing the effects of low acidic environments on the survival of C. jejuni. Furthermore, recovery of C. jejuni was not possible when bile juice or mucin was incorporated into the solid media, regardless if cells were acid–shocked prior to plating. Jones et al. (1991) indicated that the definition of non–culturability depends on the efficacy of the culture media used to recover the injured and dormant organisms. Therefore, if the culturability of organisms decreases as the cells ages, other media should be used to try to recover the aged cells, which most likely have different nutritional requirements. The optimum pH for Campylobacter spp. is around 7.4 (Hunt and Abeyta, 1995), however the organism must be able to thrive in the deleterious effects of the acidic environment it encounters when passing through the gastrointestinal tract. Most of the information available on the effects of pH on the growth and survival of Campylobacter jejuni relates to its sensitivity to withstand extreme levels. So far, there is no information on the impact an acid shock on the recovery of cells that were starved and kept under refrigeration temperature (4 C) for a prolonged period of time. The hypothesis of this study was that encounter with acidic pH (3, 4, 5, and 6) could serve as a predictor of future

PAGE 250

232 exposure to optimal conditions for colonization and pathogenesis inside a host, thus triggering survival mechanisms on the bacterium. If these mechanisms were activated, starved cells that could not be cultured under standard microbiological methods, would became culturable and their recovery in growth medium would be possible. Another environmental stress encountered by bacteria when in the gastrointestinal tract is bile juice. Bile is usually added to selective bacterial culture medium as a bacterial growth inhibitor. High levels of bile salts are toxic to bacteria, but low levels of bile might be important in regulating bacterial physiology and host–pathogen interaction (Pace et al., 1997). Several studies have been published on bacterial tolerance to bile salts (Christopher et al., 1982; Flahaut et al., 1996). Pace et al. (1997) reported that bile could enhance growth of nutrient–deprived Vibrio parahaemolyticus and enhance its virulence. They also reported that C. jejuni cultured with bile was able to adhere and invade epithelial cells at greater level as compared to that cultured without bile. Bertchinger (cited by Carvalho et al., 1997) observed that as much as 21% of the bile samples from chickens contained Campylobacter spp. The second hypothesis of this study was that the incorporation of 6, 7, 8, and 9 mg/mL of lyophilized bile juice into the solid medium would increase the growth of C. jejuni, allowing its recovery after 3 months of storage under nutrient–deprived environment. In addition to bile juice, as the bacterium continues its passage through the gastrointestinal tract, it encounters mucus in the small intestine. Sylvester et

PAGE 251

233 al. (1996) observed that C. upsaliensis adheres to mucin and that mucin fucose seems to be a chemotactic factor to C. jejuni. Furthermore, Ktyi (1990), and Mantle and Rombough (1993) observed that C. jejuni, utilize mucin as a source of energy and depend on it for growth. In the study herein, it was believed that the addition of 0.025, 0.05, 0.1, and 0.2 mg/mL gastric swine mucin to solid medium would result in recovery of starved and aged C. jejuni. As expected, the electron microscopy analysis revealed the morphological transformation from helical to coccoid shape that C. jejuni undergo as the cells age. In addition, thin sections of bacterial solutions showed a variety of cell forms, indicating that the cells in a bacterial population are heterogeneous in age and physiological state. The process by which the cells change their morphology is not well documented in the literature and no work that demonstrates the sequential changes in cell morphology has been published. The present study reported a possible sequence of morphological shapes that C. jejuni might undergo during the process of aging and transformation from a spiral to a coccoid form. Future studies on the recovery of nutrient–deprived C. jejuni should focus on the interaction of the gastrointestinal compounds and on gastrointestinal transit. An option would be to include a mechanical gastrointestinal model, such as the one being used in Mississippi State University, to the research design (Koo et al., 2001).

PAGE 252

CHAPTER 5 SUMMARY Two studies were conducted to examine five different factors that have the potential for impacting the growth and culturability of Campylobacter jejuni. The first study analyzed the effects of allyl isothiocyanate (AIT) and storage time on the Campylobacter and psychrotrophic organism population, muscle color and pH of the meat homogenate. The second study examined the effects of storage time, acid shock, addition of bile juice, and addition of mucin to the solid medium on the growth and recovery of nutrient deprived C. jejuni. Campylobacter are small, microaerophilic, spiral bacteria that inhibit the lower gastrointestinal system of warm–blooded animals. Campylobacter prefer to grow at neutral pH, but must nevertheless contend with acid encounters when passing through the gastrointestinal tract. Campylobacter do not multiply in foods stored at refrigeration temperature for prolonged periods of time, but they survive, usually outlasting the shelf life of the product (Hunt and Abeyta, 1995). The reservoir for Campylobacter species in nature is enormous since this organism has been revealed to be quite ubiquitous. Campylobacter is a pathogenic or commensal organism in cattle, sheep, swine, fowls, dogs, cats and monkeys, and it has been isolated from both fresh and sea water (Blaser, 1980). Infections with Campylobacter cause campylobacteriosis, a severe but not fatal type of 234

PAGE 253

235 gastroenteritis. Although campylobacteriosis does not usually lead to death and it is self–limiting, it has been estimated that 500 people with Campylobacter infections die each year, possibly due to neurological complications associated with the disease such as Guillain–Barr syndrome and Miller Fisher Syndrome (Dreesen, 1997; Lindsay, 1997; Ang et al., 2001). Poultry and poultry products are most often incriminated as the cause of campylobacteriosis (Blaser, 1980; Grant et al., 1980; Blaser et al., 1982; Blaser, 1982; Oosterom et al., 1983; Genigeorgis et al., 1986; Harris et al., 1986; Tauxe et al., 1987; Annan–Prah and Janc, 1988; Clark and Bueschkens, 1988; Franco, 1988; Boer and Hahn, 1990; Stern et al., 1992a; Brian and Doyle, 1995; Dreesen, 1997). According to the Centers for Disease Control and Prevention, about 80% of all poultry sold for human consumption in the year 2000 in the United States was contaminated with Campylobacter (Centers for Disease Control and Prevention, 2000). Microbial growth in fresh meat and poultry is the primary factor associated with spoilage, and loss of meat quality and money (Brewer et al., 1995). Most spoilage organisms in poultry are psychrotrophs whereas the most common pathogens are Campylobacter and Salmonella. It has been shown that prevention and control of Campylobacter infection at the animal level is hard and sometimes impractical, thus control measures must be taken at processing level. Previous studies on the use of AIT demonstrated that this antimicrobial agent is effective against a wide variety of microorganisms (Walker et al., 1937;

PAGE 254

236 Carter et al., 1963; Isshiki et al., 1992; Tsunoda, 1994; Shofran et al., 1998, Ward et al., 1998; Cherry, 1999; Delaquis et al., 2002). Although isothiocyanates are known to have antimicrobial properties, few studies were published on the effect of isothiocyanates on shelf stability of foods, especially poultry. Exposure of chicken breasts to vapors of 100, 300 and 600 L of AIT while stored at 4 C significantly affected the growth and survival of Campylobacter and psychrotrophic organisms. Campylobacter organisms were strongly inhibited by the vapors, while growth of psychrotrophic organisms were reduced as storage time increased. The partial resistance of spoilage psychrotrophic organisms to AIT may be advantageous, since spoilage organisms can inhibit the growth of undesirable pathogenic bacteria in poultry products by out competing them. The pH of the chicken breast was not affected by exposure to AIT or storage time. On the contrary, exposure of chicken muscle to vapors of AIT significantly impacted the color of the product. Chicken breasts exposed to AIT were lighter, had higher degree of yellowness and reduced degree of redness when compared to unexposed samples. Allyl isothiocyanate is a useful antimicrobial agent for the poultry industry when applied to packaged products. However, a less toxic form of AIT, such as horseradish distillates, should be used. Future studies should be conducted to determine if other sources of allyl isothiocyanates available in the market could exert similar antimicrobial effect on poultry products without the dangers of handling such a toxic compound.

PAGE 255

237 There is little information on how pH affects Campylobacter species. Furthermore, there is no evidence in the literature that an acid shock treatment, i.e., exposure of Campylobacter to very low pH for a specific amount of time, was attempted as a way of recovering starved Campylobacter cells. The acidic pH of the stomach of warm–blooded animals could serve as a predictor of future exposure to optimal conditions for colonization and pathogenesis inside a host, thus triggering survival mechanisms on the bacterium. Another factor encountered by bacteria upon entrance in the gastrointestinal tract of warm–blooded animals is the presence of bile salts, which have the ability to solubilize membranes leading to the destruction of the cell. Therefore, cells that have the ability to resist the harmful effects caused by bile salts seem to be better suited for passage through the small intestine. Several studies have been published on bacterial tolerance to bile salts (Christopher et al., 1982; Flahaut et al., 1996). Bile is usually added to selective bacterial culture medium as a bacterial growth inhibitor. High levels of bile salts are toxic to bacteria, but low levels of bile might be important in regulating bacterial physiology and host–pathogen interaction (Pace et al., 1997). Pace and colleagues showed that bile can enhance growth of nutrient–deprived Vibrio parahaemolyticus and enhance its virulence. The researchers also reported that C. jejuni cultured with bile was able to adhere and invade epithelial cells at greater level as compared to that cultured without bile.

PAGE 256

238 Carvalho et al. (1997) studied the incidence of C. jejuni in the viscera, gallbladder and bile of broilers and found that Campylobacter could survive and maybe multiply in high concentrations of bile. In another study, Bertchinger (cited by Carvalho et al., 1997) observed that as much as 21% of the bile samples from chickens contained Campylobacter spp. In addition to bile juice and bicarbonate, bacteria encounter mucus in the small intestine. Ktyi (1990), and Mantle and Rombough (1993) observed that several virulent and non–virulent enteric bacteria, including C. jejuni, utilize mucin as a source of energy and depend on it for growth. Sylvester et al. (1996) observed that C. upsaliensis adheres to mucin and that mucin fucose seems to be a chemotactic factor to C. jejuni. The initial average population of C. jejuni in the acid shock study was approximately 2x10 6 cfu/mL. The bacterial population decreased over 31 days of storage in a nutrient–deprived environment at refrigeration temperature. The addition of an enrichment step after 22 days of storage increased culturability of cells. The loss of culturability observed in the study herein was similar to those reported in the literature (Rollins and Colwell, 1986; Hazeleger et al., 1995; Federighi et al., 1998; Lzaro et al., 1999). The acid shock treatment after 3 months of storage did not induce the recovery of nutrient–deprived C. jejuni, thus indicating that other factors should be taken into consideration when observing the effects of low acidic environments on the survival of C. jejuni. In addition, the in vitro response

PAGE 257

239 observed with regards to the bacterial response to media containing bile, mucin or a combination of both was not as expected. Plating organisms in media containing bile and/or mucin did not allow for recovery of C. jejuni. As expected, C. jejuni presented a morphological transformation from spiral to coccoid shape as cells aged. The process by which the cells change their morphology is not well documented in the literature and no work that demonstrates the sequential changes in cell morphology has been published. The present study reported a possible sequence of morphological shapes that C. jejuni might undergo during the process of aging and transformation from a spiral to a coccoid form. Some researchers consider the coccoid form of Campylobacter a survival state, which allows the bacteria to survive and remain viable during stressful situations such as lack of nutrients, low osmolality, low temperatures, etc. (Rollins and Colwell, 1986; Reezal et al., 1998). Nevertheless, even if these dense coccoid cells are still viable their recovery and culture in the laboratory is difficult when conventional culture techniques are employed (Reezal et al., 1998). Future studies on the recovery of nutrient–deprived C. jejuni should focus on the interaction of the gastrointestinal compounds and on gastrointestinal transit. An option would be to include a mechanical gastrointestinal model to, such as the one developed by the Mississippi State University (Koo et al., 2001).

PAGE 258

APPENDIX A P–VALUES ASSOCIATED WITH THE EFFECTS OF ALLYL ISOTHIOCYANATE AND DAYS OF STORAGE ON NUMBER OF PHYSCHROTROPHIC ORGANISMS AND COLOR OF CHICKEN BREASTS

PAGE 259

241 Table 30 P–values* associated with the combination of effects of allyl isothiocyanate vapors (AIT) and storage on total psychrotrophic counts of skinless, boneless chicken breasts stored at 4 C** Trt 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 — 0.207 0.152 0.037 0.040 0.198 0.164 0.000 0.000 0.215 0.001 <.000 <.000 0.721 0.000 0.140 2 0.207 — 0.864 0.227 0.001 0.980 0.895 0.010 <.000 0.981 0.036 0.001 <.000 0.107 0.007 0.555 3 0.152 0.864 — 0.276 0.000 0.883 0.968 0.014 <.000 0.846 0.053 0.001 <.000 0.075 0.012 0.638 4 0.037 0.227 0.276 — 0.000 0.234 0.264 0.249 <.000 0.221 0.770 0.075 <.000 0.020 0.467 0.610 5 0.040 0.001 0.000 0.000 — 0.001 0.000 <.000 0.158 0.001 <.000 <.000 <.000 0.087 <.000 0.004 6 0.198 0.980 0.883 0.234 0.001 — 0.914 0.010 <.000 0.961 0.038 0.001 <.000 0.102 0.008 0.567 7 0.164 0.895 0.968 0.264 0.000 0.914 — 0.013 <.000 0.876 0.049 0.001 <.000 0.082 0.010 0.619 8 0.000 0.010 0.014 0.249 <.000 0.010 0.013 — <.000 0.009 0.262 0.521 <.000 0.000 0.491 0.099 9 0.000 <.000 <.000 <.000 0.158 <.000 <.000 <.000 — <.000 <.000 <.000 <.000 0.002 <.000 0.000 10 0.215 0.981 0.846 0.221 0.001 0.961 0.876 0.009 <.000 — 0.034 0.001 <.000 0.112 0.007 0.544 11 0.001 0.036 0.053 0.770 <.000 0.038 0.049 0.262 <.000 0.034 — 0.059 <.000 0.000 0.537 0.361 12 <.000 0.001 0.001 0.075 <.000 0.001 0.001 0.521 <.000 0.001 0.059 — <.000 <.000 0.143 0.023 13 <.000 <.000 <.000 <.000 <.000 <.000 <.000 <.000 <.000 <.000 <.000 <.000 — <.000 <.000 <.000 14 0.721 0.107 0.075 0.020 0.087 0.102 0.082 0.000 0.002 0.112 0.000 <.000 <.000 — <.000 0.085 15 0.000 0.007 0.012 0.467 <.000 0.008 0.010 0.491 <.000 0.007 0.537 0.143 <.000 <.000 — 0.179 16 0.140 0.555 0.638 0.610 0.004 0.567 0.619 0.099 0.000 0.544 0.361 0.023 <.000 0.085 0.179 — *<.000 = <0.0001 **Treatment Combinations: 1) 0 day — 0 L/100g AIT; 2) 0 day — 100 L/100g AIT; 3) 0 day — 300 L/100g AIT; 4) 0 day — 600 L/100g AIT; 5) 3 days — 0 L/100g AIT; 6) 3 days — 100 L/100g AIT; 7) 3 days — 300 L/100g AIT; 8) 3 days — 600 L/100g AIT; 9) 5 days — 0 L/100g AIT; 10) 5 days — 100 L/100g AIT; 11) 5 days — 300 L/100g AIT; 12) 5 days — 600 L/100g AIT; 13) 7 days — 0 L/100g AIT; 14) 7 days — 100 L/100g AIT; 15) 7 days — 300 L/100g AIT; and 16) 7 days — 600 L/100g

PAGE 260

242 Table 31 P–values* associated with the combination of effects of allyl isothiocyanate vapors (AIT) and storage on L* values of skinless, boneless chicken breasts stored at 4C** Trt 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 — 0.544 0.122 0.059 0.141 0.025 0.000 0.000 0.028 0.004 0.002 0.021 0.001 0.005 <.000 <.000 2 0.544 — 0.344 0.196 0.383 0.005 <.000 <.000 0.108 0.000 0.000 0.004 0.007 0.000 <.000 <.000 3 0.122 0.344 — 0.725 0.941 0.000 <.000 <.000 0.503 <.000 <.000 0.000 0.078 <.000 <.000 <.000 4 0.059 0.196 0.725 — 0.671 <.000 <.000 <.000 0.750 <.000 <.000 <.000 0.156 <.000 <.000 <.000 5 0.141 0.383 0.941 0.671 — 0.000 <.000 <.000 0.458 <.000 <.000 0.000 0.067 <.000 <.000 <.000 6 0.025 0.005 0.000 <.000 0.000 — 0.139 0.203 <.000 0.499 0.421 0.945 <.000 0.568 0.033 0.003 7 0.000 <.000 <.000 <.000 <.000 0.139 — 0.832 <.000 0.416 0.494 0.157 <.000 0.359 0.502 0.120 8 0.000 <.000 <.000 <.000 <.000 0.203 0.832 — <.000 0.547 0.636 0.228 <.000 0.480 0.378 0.078 9 0.028 0.108 0.503 0.750 0.458 <.000 <.000 <.000 — <.000 <.000 <.000 0.269 <.000 <.000 <.000 10 0.004 0.000 <.000 <.000 <.000 0.499 0.416 0.547 <.000 — 0.897 0.544 <.000 0.916 0.140 0.019 11 0.002 0.000 <.000 <.000 <.000 0.421 0.494 0.636 <.000 0.897 — 0.462 <.000 0.815 0.177 0.026 12 0.021 0.004 0.000 <.000 0.000 0.945 0.157 0.228 <.000 0.544 0.462 — <.000 0.615 0.039 0.003 13 0.001 0.007 0.078 0.156 0.067 <.000 <.000 <.000 0.269 <.000 <.000 <.000 — <.000 <.000 <.000 14 0.005 0.000 <.000 <.000 <.000 0.568 0.359 0.480 <.000 0.916 0.815 0.615 <.000 — 0.114 0.014 15 <.000 <.000 <.000 <.000 <.000 0.033 0.502 0.378 <.000 0.140 0.177 0.036 <.000 0.114 — 0.373 16 <.000 <.000 <.000 <.000 <.000 0.003 0.120 0.078 <.000 0.019 0.026 0.003 <.000 0.014 0.373 — *<.000 = <0.0001 **Treatment Combinations: 1) 0 day — 0 L/100g AIT; 2) 0 day — 100 L/100g AIT; 3) 0 day — 300 L/100g AIT; 4) 0 day — 600 L/100g AIT; 5) 3 days — 0 L/100g AIT; 6) 3 days — 100 L/100g AIT; 7) 3 days — 300 L/100g AIT; 8) 3 days — 600 L/100g AIT; 9) 5 days — 0 L/100g AIT; 10) 5 days — 100 L/100g AIT; 11) 5 days — 300 L/100g AIT; 12) 5 days — 600 L/100g AIT; 13) 7 days — 0 L/100g AIT; 14) 7 days — 100 L/100g AIT; 15) 7 days — 300 L/100g AIT; and 16) 7 days — 600 L/100g

PAGE 261

243 Table 32 P–values* associated with the combination of effects of allyl isothiocyanate vapors (AIT) and storage on a* values of skinless, boneless chicken breasts stored at 4C** Trt 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 — 0.640 0.930 0.398 0.627 <.000 <.000 0.000 0.265 <.000 0.000 <.000 1.000 <.000 <.000 0.000 2 0.640 — 0.704 0.704 0.984 <.000 <.000 0.000 0.515 <.000 0.001 0.000 0.640 <.000 <.000 0.000 3 0.930 0.704 — 0.448 0.690 <.000 <.000 0.000 0.304 <.000 0.000 <.000 0.930 <.000 <.000 0.000 4 0.398 0.704 0.448 — 0.719 <.000 0.000 0.002 0.785 <.000 0.003 0.001 0.398 <.000 <.000 0.002 5 0.627 0.984 0.690 0.719 — <.000 <.000 0.000 0.527 <.000 0.001 0.000 0.627 <.000 <.000 0.000 6 <.000 <.000 <.000 <.000 <.000 — 0.036 0.002 <.000 0.215 0.002 0.005 <.000 0.660 0.131 0.002 7 <.000 <.000 <.000 0.000 <.000 0.036 — 0.317 0.000 0.382 0.299 0.484 <.000 0.792 0.546 0.322 8 0.000 0.000 0.000 0.002 0.000 0.002 0.317 — 0.006 0.063 0.969 0.763 0.000 0.208 0.111 0.992 9 0.265 0.515 0.304 0.785 0.527 <.000 0.000 0.006 — <.000 0.006 0.002 0.265 0.000 <.000 0.006 10 <.000 <.000 <.000 <.000 <.000 0.215 0.382 0.063 <.000 — 0.058 0.117 <.000 0.540 0.785 0.064 11 0.000 0.001 0.000 0.003 0.001 0.002 0.299 0.969 0.006 0.058 — 0.733 0.000 0.194 0.103 0.961 12 <.000 0.000 <.000 0.001 0.000 0.005 0.484 0.763 0.002 0.117 0.733 — <.000 0.337 0.194 0.770 13 1.000 0.640 0.930 0.398 0.627 <.000 <.000 0.000 0.265 <.000 0.000 <.000 — <.000 <.000 0.000 14 <.000 <.000 <.000 <.000 <.000 0.066 0.792 0.208 0.000 0.540 0.194 0.337 <.000 — 0.733 0.211 15 <.000 <.000 <.000 <.000 <.000 0.131 0.546 0.111 <.000 0.785 0.103 0.194 <.000 0.733 — 0.113 16 0.000 0.000 0.000 0.002 0.000 0.002 0.322 0.992 0.006 0.064 0.961 0.770 0.000 0.211 0.113 — *<.000 = <0.0001 **Treatment Combinations: 1) 0 day — 0 L/100g AIT; 2) 0 day — 100 L/100g AIT; 3) 0 day — 300 L/100g AIT; 4) 0 day — 600 L/100g AIT; 5) 3 days — 0 L/100g AIT; 6) 3 days — 100 L/100g AIT; 7) 3 days — 300 L/100g AIT; 8) 3 days — 600 L/100g AIT; 9) 5 days — 0 L/100g AIT; 10) 5 days — 100 L/100g AIT; 11) 5 days — 300 L/100g AIT; 12) 5 days — 600 L/100g AIT; 13) 7 days — 0 L/100g AIT; 14) 7 days — 100 L/100g AIT; 15) 7 days — 300 L/100g AIT; and 16) 7 days — 600 L/100g

PAGE 262

244 Table 33 P–values* associated with the combination of effects of allyl isothiocyanate vapors (AIT) and storage on b* values of skinless, boneless chicken breasts stored at 4C** Trt 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 — 0.484 0.014 0.143 0.000 <.000 <.000 <.000 0.011 <.000 <.000 <.000 0.060 <.000 <.000 <.000 2 0.484 — 0.074 0.439 0.002 <.000 <.000 <.000 0.642 <.000 <.000 <.000 0.233 <.000 <.000 <.000 3 0.014 0.07 — 0.306 0.210 0.000 <.000 <.000 0.943 <.000 <.000 <.000 0.547 <.000 <.000 <.000 4 0.143 0.439 0.306 — 0.024 <.000 <.000 <.000 0.274 <.000 <.000 <.000 0.672 <.000 <.000 <.000 5 0.000 0.002 0.210 0.024 — 0.007 <.000 <.000 0.236 0.000 <.000 <.000 0.065 <.000 <.000 <.000 6 <.000 <.000 0.000 <.000 0.007 — 0.159 <.000 0.000 0.222 0.034 <.000 <.000 0.039 0.003 <.000 7 <.000 <.000 <.000 <.000 <.000 0.159 — 0.008 <.000 0.848 0.468 0.001 <.000 0.502 0.107 <.000 8 <.000 <.000 <.000 <.000 <.000 <.000 0.008 — <.000 0.005 0.053 0.462 <.000 0.047 0.293 <.000 9 0.011 0.064 0.943 0.274 0.236 0.000 <.000 <.000 — <.000 <.000 <.000 0.500 <.000 <.000 <.000 10 <.000 <.000 <.000 <.000 0.000 0.222 0.848 0.005 <.000 — 0.360 0.000 <.000 0.389 0.072 <.000 11 <.000 <.000 <.000 <.000 <.000 0.034 0.468 0.053 <.000 0.360 — 0.008 <.000 0.956 0.371 <.000 12 <.000 <.000 <.000 <.000 <.000 <.000 0.001 0.462 <.000 0.000 0.008 — <.000 0.007 0.076 0.000 13 0.060 0.233 0.547 0.672 0.065 <.000 <.000 <.000 0.500 <.000 <.000 <.000 — <.000 <.000 <.000 14 <.000 <.000 <.000 <.000 <.000 0.039 0.502 0.047 <.000 0.389 0.956 0.007 <.000 — 0.343 <.000 15 <.000 <.000 <.000 <.000 <.000 0.003 0.107 0.293 <.000 0.072 0.371 0.076 <.000 0.343 — <.000 16 <.000 <.000 <.000 <.000 <.000 <.000 <.000 <.000 <.000 <.000 <.000 0.000 <.000 <.000 <.000 — *<.000 = <0.0001 **Treatment Combinations: 1) 0 day — 0 L/100g AIT; 2) 0 day — 100 L/100g AIT; 3) 0 day — 300 L/100g AIT; 4) 0 day — 600 L/100g AIT; 5) 3 days — 0 L/100g AIT; 6) 3 days — 100 L/100g AIT; 7) 3 days — 300 L/100g AIT; 8) 3 days — 600 L/100g AIT; 9) 5 days — 0 L/100g AIT; 10) 5 days — 100 L/100g AIT; 11) 5 days — 300 L/100g AIT; 12) 5 days — 600 L/100g AIT; 13) 7 days — 0 L/100g AIT; 14) 7 days — 100 L/100g AIT; 15) 7 days — 300 L/100g AIT; and 16) 7 days — 600 L/100g

PAGE 263

REFERENCES Aizawa, S., C. S. Harwood, and R. J. Kadner, 2000. Signaling components in bacterial locomotion and sensory reception. Journal of Bacteriology 182(6):1459. Alderton, M. R., V. Korolik, P. J. Coloe, F. E. Dewhirst, and B. J. Paster, 1995. Campylobacter hyoilei sp. nov., associated with porcine proliferative enteritis. International Journal of Systematic Bacteriology 45(1):61–66. Altekruse, S. F., D. A. Street, S. B. Fein, and A. S. Levy, 1996. Consumer knowledge of foodborne microbial hazards and food–handling practices. Journal of Food Protection 59(3):287–294. Ang, C. W., M. A. de Klerk, H. P. Endtz, B. C. Jacobs, J. D. Laman, F. G. A. van der Mech, and P. A. van Doorn, 2001. Guillain–Barr syndrome– and Miller Fisher syndrome–associated Campylobacter jejuni lipopolysaccharides induce anti–GM 1 and anti–GQ 1b antibodies in rabbits. Infection and Immunity 69(4):2462. Annan–Prah, A., and M. Janc, 1988. Chicken–to–human infection with Campylobacter jejuni and Campylobacter coli: biotype and serotype correlation. Journal of Food Protection 51(7):562–564. Baker, D. A., and R. W. A. Park, 1975. Changes in morphology and cell wall structure that occur during growth of Vibrio sp. NCTC 4716 in batch culture. Journal of General Microbiology 86:12–28. Banks, W. T., C. Wang, and M. S. Brewer, 1998. Sodium lactate/sodium tripolyphosphate combination effects on aerobic plate counts, pH and color of fresh pork Longissimus muscle. Meat Science 50:499. Barnes, E. M., 1976. Microbiological problems of poultry at refrigerator temperatures – a review. Journal Sci. Food Agric. 27:777. 245

PAGE 264

246 Barnes, E. M., and M. J. Thornley, 1976. The spoilage flora of eviscerated chickens stored at different temperatures. Journal of Food Technology 1:1139. Bean, N. H., J. S. Goulding, C. Lao, and F. J. Angulo, 1996. Surveillance for foodborne–disease outbreaks—United States, 1988–1992. Morbidity and Mortality Weekly Report. CDC Surveillance Summaries. Oct 25, 45(5):1–66. Bean N. H., P. M. Griffin, J. S. Goulding, and C. B. Ivey, 1990. Foodborne disease outbreaks, 5–year summary, 1983. Journal of Food Protection 53(8):711–728. Benjamin, J., S. Leaper, R. J. Owen, and M. B. Skirrow, 1983. Description of Campylobacter laridis, a new species comprising the nalidixic acid resistant thermophilic Campylobacter (NARTC) group. Current Microbioloby 8:231–238. Beery, J. T., M. B. Hugdahl, and M. P. Doyle, 1988. Colonization of gastrointestinal tracts of chicks by Campylobacter jejuni. Applied and Environmental Microbiology 54(10):2365–2370. Berry, E. D., and C. N. Cutter, 2000. Effects of acid adaptation of Escherichia coli O157:H7 on efficacy of acetic acid spray washes to decontaminate beef carcass tissue. Applied and Environmental Microbiology 66(4):1493–1498. Blais, B. W., and L. M. Phillippe, 1999. Growth of Campylobacter jejuni in various media under different atmospheric conditions. Journal of Rapid Methods and Automation in Microbiology 7:291–300. Blaser, M. J., 1980. Campylobacter fetus subspecies jejuni: the need for surveillance. The Journal of Infectious Diseases, 141(5):670. Blaser, M. J., 1982. Campylobacter jejuni and food. Food Technology, 36:89–92. Blaser, M. J., P. Checko, C. Bopp, A. Bruce, and J. M. Hughes, 1982. Campylobacter enteritis associated with foodborne transmission. American Journal of Epidemiology, 116(6):886–894. Blaser, M. J., F. M. LaForce, N. A. Wilson, and W. L. L. Wang, 1980. Reservoirs for human campylobacteriosis. The journal of Infectious Diseases 141(5):665–669.

PAGE 265

247 Boer, E. De, and M. Hahn, 1990. Cross–contamination with Campylobacter jejuni and Salmonella spp. From raw chicken products during food preparation. Journal of Food Protection 53(12):1067–1068. Boles, J. A., and P. J. Shand, 1999. Effects of raw binder system, meat cut and prior freezing on restructured beef. Meat Science 53:233. Bolton, F. J., and D. Coates, 1983. Development of a blood–free Campylobacter medium: screening tests on basal media and supplements, and the ability of selected supplements to facilitate aerotolerance. Journal of Applied Bacteriology 54:115–125. Bolton, F. J., D. Coates, P. M. Hinchliffe, and L. Robertson, 1983. Comparison of selective media for isolation of Campylobacter jejuni/coli. Journal of Clinical Pathology 36:78–83. Booth, I. R., 1998. The bacteria strike back. The Biochemist April:8–11. Brabban, A. D., and C. Edwards, 1995. The effects of glucosinolates and their hydrolysis products on microbial growth. Journal of Applied Bacteriology 79:171–177. Bradshaw, D. J., K. A. Homer, P. D. Marsh, and D. Beighton, 1994. Metabolic cooperation in oral microbial communities during growth on mucin. Microbiology 140:3407. Brewer, M. S., B. K. Rostogi, L. Argoudelis, and G. K. Sprouls, 1995. Sodium lactate/sodium chloride effects on aerobic plate counts and color of aerobically packed ground pork. Journal of Food Science 60:58–62. Brunton, L. L., 1996. Agents affecting gastrointestinal water flux and motility; emesis and antiemesis; bile acids and pancreatic enzymes. Pages 917–936 in J. G Hardman, L. E. Limbird, P. B. Molinoff, R. W. Ruddon, and A. G. Gilman (ed) Goodman’s & Gilman’s The Pharmacological Basis of Therapeutics. Bryan, F. L., and M. P. Doyle, 1995. Health risks and consequences of Salmonella and Campylobacter jejuni in raw poultry. Journal of Food Protection 58(3)326–344. Buchanan, R. F, and N. F. Gibbons, 1974. Bergey’s Manual of Determinative Bacteriology. 8 th Edition. The Williams & Wilkins Co. Baltimore.

PAGE 266

248 Buswell, C. M., Y. M. Herlihy, L. M. Lawrence, J. T. M. McGuiggan, P. D. Marsh, C. W. Keevil, and S. A. Leach, 1998. Extended survival and persistence of Campylobacter spp. in water and aquatic biofilms and their detection by immunofluorescent–antibody and –rRNA staining. Applied and Enviromental Microbiology 64(2):733–741. Butzler, J., 1984. Campylobacter Infection in Man and Animals. CRC Press, Inc. Boca Raton, FL. Campbell, D. F., R. W. Johnston, G. S. Campbell, D. McClain, and J. F. MacAluco, 1983. The microbiology of raw, eviscerated chickens: a ten–year comparison. Poultry Science 62:437. Canale–Parola, E., 1978. Motility and chemotaxis of spirochetes. Annual Reviews in Microbiology 32:69. Canale–Parola, E., 1984. The spirochetes. Pages 38–70 in Bergey’s Manual of Systematic Bacteriology. Volume 1. Krieg, N. R., and J. G. Holt (Ed.). The Williams & Wilkins Co., Baltimore, MD. Cappelier, J. M., J. Minet, C. Magras, R. R. Colwell, and M. Federighi, 1999. Recovery in embryonated eggs of viable but nonculturable Campylobacter jejuni cells and maintenance of ability to adhere to HeLa cells after resuscitation. Applied and Environmental Microbiology 65(11):5154–5157. Cappelier, J. M., A. Rossero, and M. Federighi, 2000. Demonstration of a protein synthesis in starved Campylobacter jejuni cells. International Journal of Food Microbiology 55:63–67. Caragay, A. B., 1992. Cancer preventive foods and ingredients. Food Technology 46(April):65–68. Carter, G. A., J. L. Garraw, D. M. Spencer, and R. L. Wain, 1963. Investigation on fungicide. VI. The antifungal activity of certain dithiocarbamic and hydroxydithioformic acid derivatives. Ann. Appl. Biol. 51:135–151. Carvalho, A. C. F. B, R. P. Schocken–Iturrino, and L. F. S. A. M. Cama, 1997. Isolation of Campylobacter jejuni from viscera and bile secretion of broiler chickens with diarrhea. Revista de Microbiologia 28:125.

PAGE 267

249 Cason, J. A., J. S. Bailey, N. J. Stern, A. D. Whittemore, and N. A. Cox, 1997. Relationship between aerobic bacteria, Salmonella, and Campylobacter on broiler carcasses. Poultry Science 76:1037–1041. Centers for Disease Control and Prevention, 2002. Preliminary FoodNet data on the incidence of foodborne illnesses – selected sites, United States, 2001. Morbidity and Mortality Weekly Report 51:321–344. Centers for Diseases Control and Prevention, 2000. Campylobacter infections. [Online] Available FTP://www.cdc.gov/ncidod/dbmd/diseaseinfo/ campylobacter_t.htm [2002, April 30] Chan, K. F., H. L. Tran, R. Y. Kanenaka, and S. Kathariou, 2001. Survival of clinical and poultry–derived isolates of Campylobacter jejuni at a low temperature (4C). Applied and Environmental Microbiology 67(9):4186–4191. Cherry, J. P., 1999. Improving the safety of fresh produce with antimicrobials. Food Technology 53(11):54–59. Christopher, F. M., G. C. Smith, and C. Vanderzant, 1982. Examination of poultry giblets, raw milk and meat for Campylobacter fetus subsp. jejuni. Journal of Food Protection 45:260–262. Clark, G. S., 1992. Allyl isothiocyanate. Perfect Flavors 17:107–109. Clark, A. G., and D. H. Bueschkens, 1988. Horizontal spread of human and poultry–derived strains of Campylobacter jejuni among broiler chicks held in incubators and shipping boxes. Journal of Food Protection, 51(6):438–441. Clark–Walker, G. D., 1969. Association of microcyst formation in Spirillum itersonii with the spontaneous induction of a defective bacteriophage. Journal of Bacteriology 97(2):885–892. Cooper, S., 2001. Helical growth and the curved shape of Vibrio cholera. FEMS Microbiology Letters 198:123–124.

PAGE 268

250 Corfield, A. P., S. A. Wagner, J. R. Clamp, M. S. Kriaris, and L. C. Hoskins, 1992. Mucin degradation in the human colon: production of sialidases, sialate O–acetylesterase, N–acetylneuraminate lyase, arylesterase, and glycosulfatase activities by strains of fecal bacteria. Infection and Immunity 60(10):3971–3978. Cornforth, D., 1994. Color—its basis and importance. Pages 34 in Quality Attributtes and their Measurement in Meat, Poultry and Fish Products. Advances in Meat Research Series, Volume 9. A. M. Pearson and T. R. Dutson (Eds.). Blackie Academic & Professional, Glasgow, UK. Corry, J. E., D. E. Post, P. Colin, and M. J. Laisney, 1995. Culture media for the isolation of Campylobacters. International Journal of Food Microbiology 26(1):43–76. Cousin, M. A., J. M. Jay, and P. C. Vasavada, 1992. Psychrotrophic microorganisms. Pages 153–168 in Compendium of Methods for the Microbiological Examination of Foods. 3 rd Edition. American Public Health Association, Washington, DC. Cunningham, F. E., 1979. Shelf–life ad quality characteristics of poultry parts dipped in potassium sorbate. Journal of Food Science 44:863–864. Daikoku, T., S. Suzuki, S. Oka, and K. Takama, 1989. Profiles of enterotoxin and cytotoxin production in Campylobacter jejuni and C. coli. FEMS Microbiology Letters 49:33–36. Delaquis, P. J., and G. Mazza, 1995. Antimicrobial properties of isothiocyanates in food preservation. Food Technology 49(11):73–84. Delaquis, P. J., and P. L. Sholberg, 1997. Antimicrobial activity of gaseous allyl isothiocyanate. Journal of Food Protection 60(8):943–947. Delaquis, P. J., S. M. Ward, R. A. Holley, M. C. Cliff, and G. Mazza, 1999. Microbiological, chemical and sensory properties of pre–cooked roast beef preserved with horseradish essential oil. Journal of Food Science 64(2):519–524. Denton, J. H., 2002. Performance standards for Campylobacter are not warranted. Food Technology 56(7):104.

PAGE 269

251 Dickins, M. A., S. Franklin, R. Stefanova, G. E. Schutze, K. D. Eisenach, I. Wesley, and M. D. Cave, 2002. Diversity of Campylobacter isolates from retail poultry carcasses and from humans as demonstrated by pulse–filed gel electrophoresis. Journal of Food Protection 65(6):957–962. Dillon, S. T., E. J. Rubin, I. M. Yabubovich, C. Pothoulakis, J. T. Lamont, L. A. Feig, and R. J. Gilbert, 1995. Involvement of Ras–related Rho proteins in the mechanisms of action of Clostridium difficile toxin A and toxin B. Infection and Immunity 63:1421–1426. Dodd, C. E. R., R. L. Sharman, S. F. Bloomfield, I. R. Booth, and G. S. A. B. Steward, 1997. Inimical processes: bacterial self destruction and sub–lethal injury. Trends in Food Science Technology 8:238–241. Dreesen, D. W., 1997. Has the time come to give more concern to Campylobacter? Poultry Times, November 3:18. Englert, S. I., 1991. Avicultura: tudo sobre raas, manejo, alimentao e sanidade. 6 Edio. Livraria e Editora Agropecuria Ltda, Guaba, RS. Etoh, Y., F. E. Dewhrist, B. J. Paster, A. Yamamoto, and N. Goto, 1993. Campylobacter showae sp. nov., isolated from the human oral cavity. International Journal of Systematic Bacteriology 43(4):631–639. Euzby, J. P., 1997. List of bacterial names with standing in nomenclature: a folder available in the Internet (ftp://ftp.cict.fr/pub/bacterio/). International Journal of Systematic Bacteriology 47:590. Fearnley, C., R. Ayling, S. Cawthraw, and D. G. Newell, 1996. The formation of viable but nonculturable C. jejuni and their failure to colonise one–day–old chicks. Pages 101–104 in Campylobacters, Helicobacters, and Related Organisms. Newell, D. G., J. M. Ketley, and R. A. Feldman (Ed.). Plenum Press. New York, N.Y. Federighi, M., J. L. Tholozan, J. M. Cappelier, J. P. Tissier, and J. L. Jouve, 1998. Evidence of non–coccoid viable but non–culturable Campylobacter jejuni cells in microcosm water by direct viable count, CTC–DAPI double staining, and scanning electron microscopy. Food Microbiology 15:539–550. Felter, R. A., R. R. Cowell, and G. B. Chapman, 1969. Morphology and round body formation in Vibrio marinus. Journal of Bacteriology 99(1):326.

PAGE 270

252 Fernandez, H., and J. Toro, 1998. The relationship of Campylobacter jejuni subsp. jejuni enterotoxigenic and the increase of cAMP and electrolyte changes in the rat intestine. Rev med Chil 126(8):919–923. Flahaut, S., A. Hartke, J. Giard, A. Benachour, P. Boutibonnes, and Y. Auffray, 1996. Relationship between stress response towards bile salts, acid and heat treatment in Enterococcus faecalis. FEMS Microbiology Letters 138:49–54. Fleet, G. H., 1999. Microorganisms in food ecosystems. International Journal of Food Microbiology 50:101–117. Florin, I., and F. Antillon, 1992. Production of enterotoxin and cytotoxin in Campylobacter jejuni strains isolated in Costa Rica. Journal of Medical Microbibiology 37(1):22–29. Food and Drug Administration, 1998. Campylobacter. Pages 7.01–7.24 in Bacteriological Analytical Manual. 8 th Ed. Revision A. AOAC International. Gaithersburg, MD. Forrest, J. C., E. D. Aberle, H. B. Hedrick, M. D. Judge, and R. A. Merkel, 1975. Properties of Fresh Meat. W. H. Freeman. San Francisco, CA. Foster, J. W., 1995. Low pH adaptation and the acid tolerance response of Salmonella typhimurium. Critical Reviews in Microbiology 21(4):215–237. Fox, J. G., 1982. Campylobacteriosis: a “new” disease in laboratory animals. Laboratory Animal Sciences, 32(6):625–637. Fox, J. G., T. Chilvers, C. S. Goodwin, N. S. Taylor, P. Edmonds, L. I. Sly, and D. J. Brenner, 1989. Campylobacter mustelae, a new species resulting from the elevation of Campylobacter pylori subsp. mustelae to species status. International Journal of Systematic Bacteriology 39:301. Fox, J. G., R. Drolet, R. Higgins, S. Messier, L. Yan, B. E. Coleman, B. J. Paster, and F. E. Dewhirst, 1996. Helicobacter canis isolated from a dog liver with multifocal necrotizing hepatitis. Franco, D. A., 1988. Campylobacter species: considerations for controlling a foodborne pathogen. Journal of Food Protection, 51(2):145–153.

PAGE 271

253 Friedman, M., P. R. Henjka, and R. E. Mandreall, 2002. Bactericidal activities of plant essential oils and some of their constituents against Campylobacter jejuni, Escherichia coli, Listeria monocytogenes, and Salmonella enterica. Journal of Food Protection 65(10) 1545–1560. Gahan, C. G. M., B. O’Driscoll, and C. Hill, 1996. Acid adaptation of Listeria monocytogenes can enhance survival in acidic foods and during milk fermentation. Applied and Environmental Microbiology 62(9):3128–3132. Gebhart, C. J., P. Edmonds, G. E. Ward, H. J. Kurtz, and D. J. Brenner, 1985. “Campylobacter hyointestinalis” sp. nov.: a new species of Campylobacter found in the intestines of pigs and other animals. Journal of Clinical Microbiology 21:715–720. Genigeorgis, C., M. Hassuneh, and P. Collins, 1986. Campylobacter jejuni infection on poultry farms and its effect on poultry mat contamination during slaughtering. Journal of Food Protection 49(11)895–903. George, H. A., P. S. Hoffman, R. M. Smibert, and N. R. Krieg, 1978. Improved media for growth and aerotolerance of Campylobacter fetus. Journal of Clinical Microbiology 8:36–41. Goossens, H., and J. Butzler, 1992. Isolation and identification of Campylobacter spp. Pages 93–109 in Campylobacter jejuni: Current Status and Future Trends. I. Nachamkin, M. J. Blaser, and L. S. Tompkins (Ed.). American Society for Microbiology, Washington, DC. Goossens, H., L. Vlaes, I. Galand, C. Van den Borre, and J. P. Butzler, 1989. Semisolid blood–free selective–motility medium for the isolation of campylobacters from stool specimens. Journal of Clinical Microbiology 27:1077–1080. Graevenitz, A. V., 1990. Revised nomenclature of Campylobacter laridis, Enterobacter intermedium, and “Flavobacterium branchiophila.” International Journal of Systematic Bacteriology 40(2):221. Grant, K. A., J. H. Dickinson, D. Purdy, and S. F. Park, 1996. Molecular characterization of oxidative stress defence system in Campylobacter spp. Pages 109 in Campylobacters, Helicobacters, and Related Organisms. Newell, D. G., J. M. Ketley, and R. A. Feldman (Ed.). Plenum Press. New York, N.Y.

PAGE 272

254 Grant, C. C. R., M. E. Konkel, W. Cieplak, Jr., and L. S. Tompkins, 1993. Role of flegalle in adherence, internalization, and translocation of Campylobacter jejuni in nonpolarized and polarized epithelial cell cultures. Infection and Immunity 61(5)1764. Grant, I. H., N. J. Richardson, and V. D. Bokkenheuser, 1980. Broiler chickens as potential source of Campylobacter infections in humans. Journal of Clinical Microbiology 11(5):508–510. Grau, F. H., 1986. Microbial ecology of meat and poultry. Pages 1 in Meat, and Poultry Microbiology. Advances in Meat Research Series, Volume 2. Eds. A. M. Pearson and T. R. Dutson. AVI, Westport, Connecticut. Greenberg, E. P., 1997. Quorum sensing in gram–negative bacteria. ASM News 63(7)371–377. Greenberg, E. P., and E. Canale–Parola, 1977. Relationship between cell coiling and motility of spirochetes in viscous environments. Journal of Bacteriology 131(3):960. Guerry, P., R. A. Alm, M. E. Power, and T. J. Trust, 1992. Molecular and structural analysis of Campylobacter flagellin. In Campylobacter jejuni: Current Status and Future Trends. I. Nachamkin, M. J. Blaser, and L. S. Tompkins (Ed.). American Society for Microbiology, Washington, D.C. Harris, N. V., N. S. Weiss, and C. M. Nolan, 1986. The role of poultry and meats in the etiology of Campylobacter jejuni/coli enteritis. American Journal of Public Health, 76(4):407–411. Harvey, P., C. Fearnley, D. Newell, M. Hudson, and S. Leach, 1996. Coccal cell switching and the survival and virulence of C. jejuni at high oxygen tensions. Pages 115 in Campylobacters, Helicobacters, and Related Organisms. Newell, D. G., J. M. Ketley, and R. A. Feldman (Ed.). Plenum Press. New York, N.Y. Harvey, S. M., 1980. Hippurate hydrolysis by Campylobacter fetus. Journal of Clinical Microbiology 11(4):435–437.

PAGE 273

255 Hazeleger, W. C., J. D. Janse, P. M. F. J. Koenraad, R. R. Beumer, F. M. Rombouts, and T. Abee, 1995. Temperature–dependente membrane fatty acid and cell physiology changes in coccoid forms of Campylobacter jejuni. Applied and Environmental Microbiology 61(7):2713–2719. Hazeleger, W. C., J. A. Wouters, F, M., and T. Abee, 1998. Physiological activity of Campylobacter jejuni far below the minimal growth temperature. Applied and Environmental Microbiology 64(10):3917–3922. Hbert, G. A., D. G. Hollis, R. E. Weaver, M. A. Lambert, M. J. Blaser, and C. W. Moss, 1982. 30 years of Campylobacters: biochemical characteristics and a biotyping proposal for Campylobacter jejuni. Journal of Clinical Microbiology 15(6):1065. Hedrick, H. B., E. D. Aberle, J. C. Forrest, M. D. Judge, and R. A. Merkel, 1989. Principles of Meat Science. 3 rd Edition. Kendall/Hunt Publishing Company, Duduque, IA. Henkel, J., 1998. Irradiation: a safe measure for safer food. FDA Consumer. May–June. [Online] Available FTP:// http://www.cfsan.fda.gov/ ~dms/fdirrad.html [2002, May 29] Heredia, N. L., G. A. Garca, R. Luvanos, R. G. Labb, and J. S. Garca–Alvarado, 1997. Elevation of the heat resistance of vegetative cells and spores of Clostridium perfringens type A by sublethal heat shock. Journal of Food Protection 60(8):998–1000. Heredia, N. L., R. G. Labb, and J. S. Garca–Alvarado, 1998. Alteration in sporulation, enterotoxin production, and protein synthesis by Clostridium perfringens type A following heat shock. Journal of Food Protection 61(9):1143. Hinton Jr., A., J. A. Cason, and K. D. Ingram, 2002. Enumeration and identification of yeasts associated with commercial poultry processing and spoilage of refrigerated broiler carcasses. Journal of Food Protection 65(6):993–998. Hoffman, C., and A. C. Evans, 1911. The use of spices as preservatives. J. Ind. Eng. 3:835.

PAGE 274

256 Hugdahl, M. B, J. T. Berry, and M. P. Doyle, 1988. Chemotactic behavior of Campylobacter jejuni. Infection and Immunity 56:1560–1566. Hunt, J. M., and C. Abeyta, 1995. Campylobacter. Pages 7.01–7.27 in FDA Bacteriological Analytical Manual. 8 th Edition. AOAC International, Gaithersburg, MD. Hunt, J. M., C. Abeyta, and T. Tran, 1998. Campylobacter. Pages 7.01–7.24 in FDA Bacteriological Analytical Manual. 8 th Edition (Revision A). AOAC International, Gaithersburg, MD. Institute of Food Technologists, 2002. IFT expert report on emerging microbiological food safety issues: implications for control in the 21 st century. Institute of Food Technologists. Feb 20. Atlanta, GA. Islam, M. N., R. J. H. Gray, and J. N. Geiser, 1978. Development of antimicrobial agents for the extension of poultry shelf–life. Poultry Science 57(5):1266–1271. Isshiki, K., K. Tokura, R. Mori, and S. Chiba, 1992. Preliminary examination of allyl isothiocyanate vapor for food preservation. Biosci. Biotech. Biochem. 56:1476–1477. Izat, A. L., and F. A. Gardner, 1988. Incidence of Campylobacter jejuni in processed egg products. Poultry Science, 67:1431–1435. Jay, J. M., 1992. Campylobacteriosis Pages 597 in Modern Food Microbiology. 4 th Edition. Chapman & Hall, New York, NY. Jay, J. M., 2002. A review of aerobic and psychrotrophic plate count procedures for fresh meat and poultry products. Journal of Food Protection 65(7): 1200–1206. Jeffrey, J. S., A. Hunter, and E. R. Atwill, 2000. A field–suitable, semisolid aerobic enrichment medium for isolation of Campylobacter jejuni in small numbers. Journal of Clinical Microbiology 38(4):1668. Johnson, R. C., and S. Faine, 1984. Familly II. Leptospiraceae. Pages 62–70 in Bergey’s Manual of Systematic Bacteriology. Volume 1. Krieg, N. R., and J. G. Holt (Ed.). Williams & Wilkins.

PAGE 275

257 Jones, D. M., E. M. Stucliffe, and A. Curry, 1991. Recovery of viable but nonculturable Campylobacter jejuni. Journal of General Microbiology 137:2477–2482. Joseph, R., and E. Canale–Parola, 1972. Axial fibrils of anaerobic spirochetes: ultrastructure and chemical characteristics. Archives Mikrobiology 81:146–168. Joseph, R., S. C. Holt, and E. Canale–Parola, 1973. Peptidoglycan of free–living anaerobic spirochetes. Journal of Bacteriology 115(1):426–435. Kaijser, B., and R. J. Meinersmann, 1996. Immunity to and vaccines for Campylobacters p. 547–551 in Campylobaters, Helicobacters and Related Organisms. D. G. Newell, J. M. Ketley, and R. A. Feldman (Ed.). Plenum Press, New York, N.Y. Karlyshev, A. V., M. V. McCrossan, and B. W. Wren, 2001. Demonstration of polysaccharide capsule in Campylobacter jejuni using electron microscopy. Infection and Immunity 69(9)5921–5924. Kawakishi, S., and T. Kaneko, 1985. Interaction of oxidized glutathione with allyl isothiocyanate. Phytochemistry 24(4):715–718. Kawakishi, S., and T. Kaneko, 1987. Interaction of proteins with allyl isothiocyanate. J. Agric. Food Chem. 35:85–88. Kempton, A. G., and S. R. Bobier, 1970. Bacterial growth in refrigerated, vacuum–packed luncheon meats. Canadian Journal of Microbiology 16:287. Ktyi, I., 1990. Binding of enteric bacteria to hog gastric mucin. Acta Microbiologica Hungarica 37(1):45–53. Kiehlbauch, J. A., D. J. Brenner, M. A. Nicholson, C. N. Baker, C. M. Patton, A. G. Steigerwalt, and I. K. Wachsmuth, 1991. Campylobacter butzleri sp. nov. isolated from humans and animals with diarrheal illness. Journal of Clinical Microbiology 29:376–385. Kim, Y. S., E. S. Ahn, and D. H. Shin, 2002. Extension of shelf life by treatment with allyl isothiocyanate in combination with acetic acid on cooked rice. Journal of Food Science 67(1):274.

PAGE 276

258 King, E. O., 1957. Human infections with Vibrio fetus and a closely related vibrio. Journal of Infectious Diseases 101:119–128. Kita, E., D. Oku, A. Hamuro, F. Nishikawa, M. Emoto, Y. Yagyu, N. Katsui, and S. Kashiba, 1990. Hepatotoxic activity of Campylobacter jejuni. Journal of Medical Microbiology 33(3):171–182. Knight, J., and J. Pierce, 1997. USDA announces recall of additional Hudson frozen ground beef – Release No. 0276.97. [Online] available http://www.usda.gov/news/ releases/1997/08/0276 [2002, July 21] Knchel, S., and G. Gould, 1995. Preservation microbiology and safety: Quo Vadis? Trends in Food Science & Technology 6:127–131. Koch, A. L., 2000. The bacterium’s way for safe enlargement and division. Applied and Environmental Microbiology 66(9):3657–3663. Kolm, R. H., H. U. Danielson, Y. Zhang, P. Talalay, and B. Mannervik, 1995. Isothiocyanates as substrates for human glutathione transferases: structure–activity studies. Journal of Biochemistry 311:453–559. Koo, J., D. L. Marshall, and A. DePaola, 2001. Antacid increases survival of Vibrio vulnificus phage in a gastrointestinal model. Applied and Environmental Microbiology 67(7):2895–2902. Krieg, N. R., 1984. Genus Oceanospirillum. Pages 73–110 in Bergey’s Manual of Systematic Bacteriology. Volume 1. Krieg, N. R., and J. G. Holt (Ed.). Williams & Wilkins. Kyung, K. H., and H. P. Fleming, 1996. Antimicrobial activity of sulfur compounds derived from cabbage. Journal of Food Protection 60(1):67–71. Lzaro B., J. Crcamo, A. Audcana, I. Perales, and A Fernndez–Astorga, 1999. Viability and DNA maintenance in nonculturable spiral Campylobacter jejuni cells after long–term exposure to low temperatures. Applied and Environmental Microbiology 65(10):4677–4681. Lee, S. G., A. Kim, and Y. C. Ha, 1997. Successful cultivation of a potentially pathogenic coccoid organism with tropism of gastric mucin. Infection and Immunity 65(1):49–54.

PAGE 277

259 Lee, A., J. L. O’Rourke, P. J. Barrington, and T. J. Trust, 1986. Mucus colonization as a determinant of pathogenicity in intestinal infection by Campylobacter jejuni: a mouse cecal model. Infection and Immunity 51:536–546. Levin, R. E., and R. H. Vaughn, 1968. Spontaneous spheroplast formation by Desulfovibrio aestuarii. Canadian Journal of Microbiology 14:1271. Lewis, P. R., and D. P. Knight, 1977. Cytological staining methods in electron microscopy. Pages 25–76 in Staining Methods for Sectioned Material. Elsevier Biomedical Press, Amsterdam, The Netherlands. Lindsay, J. A., 1997. Chronic sequelae of foodborne disease. Emerging Infectious Diseases 3(4):443–452. Loesche, W. J., R. J. Gibbons, and S. Socransky, 1965. Biochemical characteristics of Vibrio sputorum and relationship to Vibrio bubulus 91 and Vibrio fetus. Journal of Bacteriology 89:1109. MacDougall, D. B., 1994. Colour of Meat. Pages 79–93 in Quality Attributtes and their Measurement in Meat, Poultry and Fish Products. Advances in Meat Research Series, Volume 9. A. M. Pearson and T. R. Dutson (Ed.). Blackie Academic & Professional, Glasgow, UK. Mantle, M., and C. Rombough, 1993. Growth in and breakdown of purified rabbit small intestinal mucin by Yersinia enterocolitica. Infection and Immunity 61(10):4131. Marron, L., N. Emerson, C. G. M. Gahan, and C. Hill, 1997. A mutant of Listeria monocytogenes L028 unable to induce an acid tolerance response displays diminished virulence in a murine model. Applied Environmental Microbiology 63:4945. Marshall, B. J., and C. S. Goodwin, 1987. Revised nomenclature of Campylobacter pyloridis. International Journal of Systematic Bacteriology 37:68. Marshall, B. J., H. Royce, D. I. Annear, C. S. Goodwin, J. W. Pearman, J. R. Warren, and J. A. Armstrong, 1984. Original isolation of Campylobacter pyloridis from human gastric mucosa. Microbios Lett. 25:83–88.

PAGE 278

260 Mazza, G., 1984. Volatiles in distillates of fresh, dehydrated and freeze dried horseradish. Canadian Institute of Food Science Technology Journal 17:18–23. McClung, C. R., D. G. Patriquin, and R. E. Davis, 1983. Campylobacter nitrofigilis sp. nov., a nitrogen–fixing bacterium associated with roots of Spartina alterniflora Loisel. International Journal of Systematic Bacteriology 33:605–6012. McMeekin, T. A., 1975. Spoilage association of chicken breasts muscle. Applied Microbiology 29(1):44. Mead P., L. Slutsker, A. Dietz, L. McCaig, J. Bresee, C. Shapiro, P. Riffin, and R. Tauxe, 1999. Foor-related illness and deaths in the United States. Emerging Infectious Diseases 5(5): 607-625. Meinersmann, R. J., W. E. Rigsby, N. J. Stern, L. C. Kelley, J. E. Hill, and M. P. Doyle, 1991. Comparative study of colonizing and noncolonizing Campylobacter jejuni. American Journal of Veterinary Research 52(8):1518–1522. Mendelson, N. H., 1982. Bacterial growth and division: genes, structures, forces, and clocks. Microbiology Reviews 46:341–375. Mendona, A. F., R. A. Molins, A. A. Kraft, and H. W. Walker, 1989. Microbiological, chemical and physical changes in fresh, vacuum–packaged pork treated with organic acids and salts. Journal of Food Science 54:18–21. Misawa, N., K. Hirayama, K. Itoh, and E. Takahashi,1995. Detection of alpha– and beta–hemolytic activity from Campylobacter jejuni. Journal of Clinical Microbiology 33(3):729–731. Montrose, M. S., S. M. Shane, and K. S. Harrington, 1985. Role of litter in the transmission of Campylobacter jejuni. Avian Diseases 29(2):392–399. Moore, R. T., and G. B. Chapman, 1959. Observations of the fine structure and modes of growth of a streptomycete. Journal of Bacteriology 78:878.

PAGE 279

261 Moran, A. P. and M. E. Upton, 1987. Factors affecting production of coccoid forms by Campylobacter jejuni on solid media during incubation. Journal of Applied Bacteriology 62:527–537. Murray, P. R., E. J. Baron, M. A. Pfaller, F. C. Tenover, and R. H. Yolkem, 1995. Helicobacter. Pages 492–498 in Manual of Clinical Microbiology. 6 th Edition. American Society for Microbiology, Washington, DC. Murray, R. G. E., and Birch–Andersen, 1963. Specialized structure in the region of the flagella tuft in Spirillum serpens. Canadian Journal of Microbiology 9:393–409. Musk, S. R., and I. T. Johnson, 1993. The clastrogenic effect of isothiocyanates. Mutation Research 300:111–117. Neill, S. D., J. N. Campbell, J. J. O’Brien, S. T. C. C. Weatherup, and W. A. Ellis, 1985. Taxonomic position of Campylobacter cryaerophila sp. nov. International Journal of Systematic Bacteriology 35:342–356. Ng, L., R. Sherburne, D. E. Taylor, and M. E. Stiles, 1985. Morphological forms and viability of Campylobacter species studied by electron microscopy. Journal of Bacteriology 164(1):338–343. Nurmi, E., and M. Rantala, 1973. New aspects of Salmonella infection in broiler production. Nature 241:210–211. Ogg, J. E., 1962. Studies on the coccoid form of ovine Vibrio fetus I. Cultural and serologic investigations. American Journal of Veterinary Research. March:354. On, S. L. W., B. Bloch, B. Holmes, B. Hoste, and P. Vandamme, 1995. Campylobacter hyointestinalis subsp. lawsonii subsp. nov., isolated from the porcine stomach, and an emended description of Campylobacter hyointestinalis. International Journal of Systematic Bacteriology 45(4):767–774. Oosterom, J., S. Notermans, H. Karman, and G. B. Engels, 1983. Origin and prevalence of Campylobacter jejuni in poultry processing. Journal of Food Protection 46(4):339–344. Pace, J. L., T. Chai, H. A. Rossi, and X. Jiang, 1997. Effect of bile on Vibrio parahaemolyticus. Applied Environmental Microbiology 63(3):2372–2377.

PAGE 280

262 Papadakis, S. E., S. Abdul–Malek, R. E. Kamdem, and K. L. Yam, 2000. A versatile and inexpensive technique for measuring color of foods. Food Technology 54(12):48. Papadopoulos, L. S., R. K. Miller, G. R. Acuff, C. Vanderzant, and H. R. Cross, 1991a. Effect of sodium lactate on microbial and chemical composition of cooked beef during storage. Journal of Food Science 56:341–347. Papadopoulos, L. C., R. K. Miller, L. J. Ringer, and H. R. Cross, 1991b. Sodium lactate effect on sensory characteristics, cooked meat color and chemical composition. Journal of Food Science 56:621. Pearson, A. M., 1994. Introduction to quality attributes and their measurement in meat, poultry and fish products. Pages 1–33 in Quality Attributtes and their Measurement in Meat, Poultry and Fish Products. Advances in Meat Research Series, Volume 9. A. M. Pearson and T. R. Dutson (Ed.). Blackie Academic & Professional, Glasgow, UK. Perez–Perez, G. I., D. N. Taylor, P. D. Echeverria, and M. J. Blaser, 1992. Lack of evidence of enterotoxin involvement in pathogenesis of Campylobacter diarrhea, p. 184–192. In I. Nachamkin, M. J. Blaser, and L. S. Tompkins (ed.) Campylobater jejuni: current status and future trends. American Society for Microbiology, Washington, D.C. Prescott, L. M., J. P. Harley, and D. A. Klein, 1999. Microbial growth. Pages 113–134 in Microbiology. 4 th edition. WCB/McGraw–Hill. New York, N.Y. Reezal, A., B. McNeil, and J. G. Anderson, 1998. Effect of low–osmolarity nutrient media on growth and Culturability of Campylobacter species. Applied and Environmental Microbiology 64(12):4643–4649. Reynolds, E. S., 1963. The use of lead citrate at high pH as an electron–opaque stain in electron microscopy. Journal of Cell Biology 17:208. Rhee, K. S., and Y. A. Ziprin, 1987. Lipid oxidation in retail beef, pork and chicken muscles as affected by concentrations of heme pigments and non–heme iron and microsomal enzymic lipid peroxidation activity. Journal of Food Biochemistry 11:1–.

PAGE 281

263 Rhoades, H. E., 1954. The illustration of the morphology of Vibrio fetus by electron microscopy. American Journal of Veterinary Research. October:630–633. Rhoades, R. A., and G. A. Tanner, 1995. Medical Physiology. Pages 530–570. Little, Brown and Company. New York, NY. Rollins, D. M., and R. R. Colwell, 1986. Viable but nonculturable stage of Campylobacter jejuni and its role in survival in the natural aquatic environment. Applied and Environmental Microbiology 52(3):531–538. Rose, B. E., W. E. Hill, R. Umholtz, G. M. Ransom, and W. O. James, 2002. Testing for Salmonella in raw meat and poultry products collected at federally inspected establishments in the United States, 1998 through 2000. Journal of Food Protection 65(6):937–947. Rowan, N. J., 1999. Evidence that inimical food–preservation barriers alter microbial resistance, cell morphology and virulence. Trends in Food Science & Technology 10:261–270. Rowe, M. T., and R. Kirk, 1999. An investigation into the phenomenon of cross–protection in Escherichia coli O157–H7. Food Microbiology 16(2):157–164. Rowland, A. C., G. H. K. Lawson, and Maxwell, 1973. Intestinal adenomatosis in the pig: occurrence of a bacterium in affected cells. Nature 243:417–. Ruiz–Palacios, G. M., J. Torres, N. I. Torres, E. Escamilla, B. R. Ruiz–Palacios, and J. Tamayo, 1983. Cholera–like enterotoxin produced by Campylobater jejuni. Lancet ii:250–253. Ruzal, S. M. and C. Sanchez–Rivas, 1994. Physiological and genetic characterization of the osmotic stress response in Bacillus subtilis. Canadian Journal of Microbiology 40:140–144. Saha, S. K., S. Saha, and S. C. Sanyal, 1991. Recovery of injured Campylobacter jejuni cells after animal passage. Applied and Environmental Microbiology 57(11):3388.

PAGE 282

264 Snchez, M. X., W. M. Fluckey, M. M. Brashears, and S. R. McKee, 2002. Microbial profile and antibiotic susceptibility of Campylobacter spp. and Salmonella spp. in broilers processed in air–chilled and immersion–chilled environments. Journal of Food Protection 65(6):948–956. Sandstedt, K., and J. Ursing, 1991. Description of Campylobacter upsaliensis sp. nov. previously known as the CNW group. Syst. Appl. Microbiology 14:39–45. SAS Institute, 1990. SAS user guide. 6 th ed. SAS: Cary, NC. Schleyer, M., R. Schmidt, and E. P. Baker, 1993. Transient specific and extremely rapid release of osmolytes from growing cells of Escherichia coli K exposed to hypoosmotic shock. Archives Microbiology 160:424. Schoeni, J. L., and M. P. Doyle, 1992. Reduction of Campylobacter jejuni colonization of chicks by cecum–colonizing bacteria producing anti–C. jejuni metabolites. Applied Environmental Microbiology 58:664–670. Schoeni, J. L., and A. C. L. Wong, 1994. Inhibition of Campylobacter jejuni colonization in chicks by defined competitive exclusion bacteria. Applied and Environmental Microbiology 60(4):1191. Schulze, F. I. Hanel, and E. Borrmann, 1996. Detection of a cytolethal distending toxin in campylobacters of human and animal origin p. 631–635 in Campylobaters, Helicobacters and Related Organisms. D. G. Newell, J. M. Ketley, and R. A. Feldman (Ed.). Plenum Press, New York, N.Y. Schulze, F. I. Hanel, and E. Borrmann, 1998. Formation of cytotoxins by enteric Campylobacter in humans and animals. Zentralblatt fur Bakteriologie 228:225–236. Shane, S. M., 1997. Campylobacteriosis Pages 235–245 in Diseases of Poultry. 10 th Edition. Clanek, B. W., H. J. Barnes, C. W. Beard, L. R. McDougald, and Y. M. Saif (Ed.). Iowa State University Press, Ames, IA. Shane, S. M., M. S. Montrose, and K. S. Harrington, 1985. Transmission of Campylobacter jejuni by the housefly (Musca domestica). Avian Diseases 29(2):384–391.

PAGE 283

265 Sharma, R., F. Fernadez, M. Hilton, and U. Schumacher, 1997. The influence of diet on the mucin carbohydrates in the chick intestinal tract. Cellular and Molecular Life Sciences 53:935–942. Shofran, B. G., S. T. Purrington, F. Breidt, and H. P. Fleming, 1998. Antimicrobial properties of sinigrin and its hydrolysis products. Journal of Food Science 63(4):621–624. Silliker, J. H., R. P. Elliot, A. C. Baird–Parker, F. L. Bryan, J. H. B. Christian, D. S. Clark, J. C. Olson, Jr., and T. A. Roberts, 1980. Microbial Ecology of Foods. Academic Press, Inc., New York, N.Y. Simon, G. L., and S. L. Gorbach, 1986. The human intestinal microflora. Digestive Diseases and Sciences 31(Supplement 9):147S–162S. Skerman, V. B. D., V. McGowan, and P. H. A. Sneath, 1980. Approved lists of bacterial names. International Journal of Systematic Bacteriology 30(1):225–420. Skirrow, M. B., 1977. Campylobacter enteritis: a “new” disease. British Medical Journal 2:9. Skirrow, M. B., and J. Benjamin, 1980. Differentiation of enteropathogenic Campylobacter. Journal of Clinical Pathology 32:1122. Skirrow, M. B., and M. J. Blaser, 1992. Clinical and epidemiologic considerations, Pages 3–8 in Campylobacter jejuni: Current Status and Future Trends. I. Nachamkin, M. J. Blaser, and L. S. Tompkins (Ed.). American Society for Microbiology, Washington, D.C. Smibert, R. M., 1974. Genus II. Campylobacter. Pages 207–211 in Bergey’s Manual of Determinative Bacteriology. 8 th Edition. R. F. Buchanan and N. F. Gibbons (Ed.). Williams & Wilkins. Smibert, R. M., 1978. The genus Campylobacter. Annual Reviews in Microbiology 32:673–709. Smibert, R. M., 1984. Genus Campylobacter. Pages 111 in Bergey’s Manual of Systematic Bacteriology. Volume 1. Krieg, N. R., and J. G. Holt (Ed.). Williams & Wilkins.

PAGE 284

266 Snedecor, G. W., and W. G. Cochran, 1989. Statistical methods. 8 th ed. Iowa State University Press. Ames, IA. Soerjadi, A. S., G. H. Snoeyenbos, and O. M. Weinack, 1982. Intestinal colonization and competitive exclusion of Campylobacter fetus subsp. jejuni in young chicks. Avian Diseases 26(3):520–524. Sofos, J. N., 1994. Microbial growth and its control in meat, poultry, and fish. Pages 359 in Quality Attributtes and their Measurement in Meat, Poultry and Fish Products. Advances in Meat Research Series, Volume 9. A. M. Pearson and T. R. Dutson (Ed.). Blackie Academic & Professional, Glasgow, UK. Sperandio, V., J. L. Mellies, W. Nguyen, S. Shin, and J. B. Kaper, 1999. Quorum sensing controls the expression of the type III secretion gene transcription and protein secretion in enterohemorrhagic and enteropathogenic Escherichia coli. Procedures Natural Academy of Science 96(26):15196–15201. Stanley, J., A. P. Burnens, D. Linton, S. L. W. On, M. Costas, and R. J. Owen, 1992. Campylobacter helveticus sp. nov., a new thermophilic species from domestic animals: characterization and cloning of a species–specific DNA probe. Journal of General Microbioliology 138:2293. Stern, N. J., M. R. S. Clavero, J. S. Bailey, N. A. Cox, and M. C. Robach, 1995. Campylobacter spp. in broilers on the farm and after transport. Poultry Science 74:937–941. Stern, N. J., C. M. Patton, M. P. Doyle, C. E. Park, and B. A. McCardell, 1992a. Campylobacter. Pages 475–495 in Compendium for the Microbiological Examination of Foods. 3 rd Edition. American Public Health Association, Washington, DC. Stern, N. J., B. Wojton, and K. Kwiatek, 1992b. A differential–selective medium and dry ice–generated atmosphere for recovery of Campylobacter jejuni. Journal of Food Protection 55(7):514–517. Stonnet, V., C. Harmant, C. Bizet, and J. Guesdon, 1995. Characterization of the type strain of Campylobacter coli, CIP 70.80, by plasmid typing. International Journal of Systematic Bacteriology 45(3):592.

PAGE 285

267 Sylvester, F. A., D. Philpott, B. Gold, A. Lastovica, and J. F. Forstner, 1996. Adherence to lipids and intestinal mucin by a recently recognized human pathogen, Campylobacter upsaliensis. Infection and Immunity 64(10):4060–4066. Tang, C. S., 1974. Benzyl isothiocyanate as a natural occurring papain inhibitor. Journal of Food Science 39:94–96. Tang, C. S., 1973. Localization of benzyl glucosinolates and thioglucosidase in Carica papaya fruit. Phytochemistry 12:769–773. Tanner, A. C. R., S. Badger, C. H. Lai, M. A. Listgarten, R. Visconti, and S. Socransky, 1981. Wollinella gen. nov., Wollinella succinogenes (Vibrio succinogenes Wolin et al.) comb. nov., and description of Bacteroides gracilis sp. nov., Wolinella recta sp. nov., Campylobacter concisus sp. nov., and Eikenella corrodens from humans with periodontal disease, International Journal of Systematic Bacteriology 31:432–. Tauxe, R. V., D. A. Pegues, and N. Hargrett–Bean, 1987. Campylobacter infections: the emerging nation pattern. American Journal of Public Health, 77(9):1219–1221. Taylor, D. E., and P. Courvalin, 1988. Mechanisms of antibiotic resistance in Campylobacter species. Antimicrobial Agents and Chemotherapy, 32(8):1107–1112. Totten, P. A., C. L. Fennell, F. C. Tenover, J. M. Wezenberg, P. L. Perine, W. E. Stamm, and K. K. Holmes, 1985. Campylobacter cinaedi (sp. nov.) and Campylobacter fennelliae (sp. nov.): two new Campylobacter species associated with enteric disease in homosexual men. Journal of Infectious Diseases 151:131–139. Troller, J. A., and V. N. Scott, 1992. Measurement of water activity (A w ) and acidity. Pages 135–151 in Compendium of Methods for the Microbiological Examination of Foods. 3 rd Edition. American Public Health Association, Washington, DC. Tsunoda, K., 1994. Effects of gaseous treatment with allyl isothiocyanate on the control of microbial growth on a wood substrate. Journal of Antibact. Agents. 22:145–148.

PAGE 286

268 Tunnicliff, R., 1914. An anaerobic vibrio isolated from a case of acute bronchitis. Journal of Infectious Diseases 15:350. U. S. Department of Health and Human Services, 1999. Food Code. Public Health Service, Food and Drug Administration, Washington, CD. Ursing, J. B., H. Lior, and R. J. Owen, 1994. Proposal of minimal standards for describing new species of the family Campylobacteraceae. International Journal of Systematic Bacteriology 44(4):842–845. Vandamme, P., M. I. Daneshvar, F. E. Dewhirst, B. J. Paster, K. Kersters, H. Goossens, and C. W. Moss, 1995. Chemotaxonomic analyses of Bacteroides gracillis and Bacteroides ureolyticus and reclassification of B. gracilis as Campylobacter gracilis comb. nov. International Journal of Systematic Bacteriology 45:145–152. Vandamme, P., E. Falsen, R. Rossau, P. Segers, R. Tytgat, and J. De Ley, 1991. Revision of Campylobacter, Helicobacter, and Wolinella taxonomy: emendation of generic descriptions and proposal of Arcobacter gen. nov. International Journal of Systematic Bacteriology 41(1):88. Vandamme, P., and J. De Ley, 1991. Proposal for a new family, Campylobacteraceae. International Journal of Systematic Bacteriology 41(3):451–455. Vandamme, P., L. J. van Doorn, S. T. Al Rashid, W. G. V. Quint, J. Van Der Plas, V. L. Chan, and S. L. W. On, 1997. Campylobacter hyoilei Alderton et al. 1995 and Campylobacter coli Vron and Chatelain 1973 are subjective synonyms. International Journal of Systematic Bacteriology 47:1055–1060. Veneman, A. M., and E. Murano, 2002. Press conference with secretary of agriculture Ann M. Venema and undersecretary for food safety Elsa Murano regarding the ConAngra Beef Company recall – Friday, July 19, 2002 – Release No. 0297.02. [Online] available http://www.usda.gov/ news/releases/2002/07/0297.htm [2002, July 21] Vron, M., and R. Chatelain, 1973. Taxonomic study of the genus Campylobacter Sebald and Vron and designation of the neotype strain for the type species Campylobacter fetus (Smith and Taylor) Sebald and Vron. International Journal of Systematic Bacteriology 23:122–134.

PAGE 287

269 Villarreal, L., N. L. Heredia, and S. Garca, 2000. Changes in protein synthesis and acid tolerance in Clostridium perfringens type A in response to acid shock. International Microbiology 3:113. Wadstrom, T., S. B. Baloda, K. Krovacek, A. Faris, S. Bengtson, and M. Walder, 1983. Swedish isolates of Campylobacter jejuni/coli do not produce cytotonic or cytotoxic enterotoxins. Lancet ii:911. Walker, J. C., C. C. Lindegren, and F. M. Bachman, 1925. Further studies on the toxicity of juice extracted from succulent onion scales. Journal of Agricultural Research. 30:175–187. Walker, J. C., S. Morell, and H. H. Foster, 1937. Toxicity of mustard oils and related sulfur groups to certain fungi. American Journal of Botany 24:536–541. Wallis, M. R., 1994. The pathogenesis of Campylobacter jejuni. British Journal of Biomedical Science 51(1)57–64. Wang, W. L., N. W. Luechtefeld, L. B. Reller, and M. J. Blaser, 1980. Enriched Brucella medium for storage and transport of cultures of Campylobacter fetus subsp. jejuni. Journal of Clinical Microbiology 12(3):479–480. Ward, S. M., P. J. Delaquis, R. A. Holley, and G. Mazza, 1998. Inhibition of spoilage and pathogenic bacteria on agar and pre–cooked roast beef by volatile horseradish distillates. Food Research International 31(1):19–26. Wassenaar, T. M., 1997. Toxin production by Campylobacter spp. Clinical Microbiology Review, 10(3):466. Wassenaar, T. M., B. A. M. van der Zeust, R. Ayling, and D. G. Newell, 1993. Colonization of chicks by motility mutants of Campylobacter jejuni demonstrates the importance of flagellin A expression. Journal of General Microbiology 139:1171. White, D., 1995. The physiology and biochemistry of prokaryotes. Pages 7. Oxford University Press, Inc. New York, NY. Whitehouse, C.A., P. B. Balbo, E. C. Pesci, D. L. Cottle, P. M. Mirabito, and C. L. Pickett, 1998. Campylobacter jejuni cytolethal distending toxin causes a G2–phase cell cycle block. Infection and Immunity 66(5):1934–1940.

PAGE 288

270 Wierup, M., M. Wold–Troell, E. Nurmi, and M. Hokkinen, 1988. Epidemiological evaluation of the salmonella–controlling effect of a nationwide use of a competitive exclusion culture in poultry. Poultry Science 67:10263. Williams, M. A., and S. C. Rittenberg, 1956. Microcyst formation and germination in Spirillum lunatum. Journal of General Microbiology 15:205–209. Williams, M. A., and S. C. Rittenberg, 1957. A taxonomic study of the genus Spirillum ehrenberg. International Bulletin of Bacteriological nomenclature and taxonomy 7(2):49. Yamanaka, K., L. Fang, and M. Inouye, 1998. The CspA family in Escherichia coli: multiple gene duplication for stress adaptation. Molecular Microbiology 27(2):247–255. Zaritsky, A., and N. H. Mendelson, 1984. Helical macrofiber formation in Bacillus subtilis: inhibition by penicillin G. Journal of Bacteriology 1558(3):1182–1187.

PAGE 289

BIOGRAPHICAL SKETCH Indau Ieda Giriboni de Mello was born on November 17, 1971, in Itapetininga, So Paulo, Brazil. She obtained her elementary school degree in public schools of the state of So Paulo. Thereafter, she completed middle school at the Colgio da Polcia Militar in 1985 and high school at the Colgio Integrado Objetivo in 1988. From 1990 to 1994, Indau Ieda attended the Faculdade de Zootecnia e Engenharia de Alimentos at the Universidade de So Paulo, Pirassununga, Brazil, where she obtained a Bachelor of Science degree with a major in animal sciences. In 1994, Indau Ieda was granted a scientific initiation research assistantship from Fundao de Amparo Pesquisa do Estado de So Paulo (FAPESP), a Brazilian Government Research Support Foundation, for conducting research in the area of poultry production. In the fall semester of 1995, Indau Ieda was granted a two–year research assistantship from the Department of Animal Sciences at the University of Kentucky to work on her master’s program. In the summer of 1997, she obtained a Master of Science degree with emphasis in poultry nutrition. For outstanding academic achievement, the author became a member of Gamma Sigma Delta, the honor society of agriculture, in 1997. 271

PAGE 290

272 In the fall semester of 1997, the author received a four–and–half–years research assistantship from the Department of Animal Sciences at the University of Florida to work on her doctoral program. In the summer of 1999, Indau Ieda was admitted to candidacy and in the fall semester of 2002 she received her Doctor of Philosophy degree with emphasis in food microbiology. After graduation, she plans to work in the United States of America in the area of microbiology for a few years before returning to Brazil, her home country.