BIOCONTROL OF FOODBORNE BACTERIAL

PATHOGENS USING IMMOBILIZED BACTERIOPHAGES

A Thesis

Presented to

The Faculty of Graduate Studies

of

The University of Guelph

by

HANY EL-SAID MOHAMAD ANANY

In partial fulfillment of requirements

for the degree of

Doctor of Philosophy

August, 2010

©HanyAnany, 2010 Library and Archives Bibliotheque et 1*1 Canada Archives Canada Published Heritage Direction du Branch Patrimoine de I'edition

395 Wellington Street 395, rue Wellington Ottawa ON K1A 0N4 OttawaONK1A0N4 Canada Canada

Your file Votre reference ISBN: 978-0-494-67847-3 Our file Notre reference ISBN: 978-0-494-67847-3

NOTICE: AVIS:

The author has granted a non­ L'auteur a accorde une licence non exclusive exclusive license allowing Library and permettant a la Bibliotheque et Archives Archives Canada to reproduce, Canada de reproduire, publier, archiver, publish, archive, preserve, conserve, sauvegarder, conserver, transmettre au public communicate to the public by par telecommunication ou par I'lnternet, preter, telecommunication or on the Internet, distribuer et vendre des theses partout dans le loan, distribute and sell theses monde, a des fins commerciales ou autres, sur worldwide, for commercial or non­ support microforme, papier, electronique et/ou commercial purposes, in microform, autres formats. paper, electronic and/or any other formats.

The author retains copyright L'auteur conserve la propriete du droit d'auteur ownership and moral rights in this et des droits moraux qui protege cette these. Ni thesis. Neither the thesis nor la these ni des extra its substantiels de celle-ci substantial extracts from it may be ne doivent etre imprimes ou autrement printed or otherwise reproduced reproduits sans son autorisation. without the author's permission.

In compliance with the Canadian Conformement a la loi canadienne sur la Privacy Act some supporting forms protection de la vie privee, quelques may have been removed from this formulaires secondaires ont ete enleves de thesis. cette these.

While these forms may be included Bien que ces formulaires aient inclus dans in the document page count, their la pagination, il n'y aura aucun contenu removal does not represent any loss manquant. of content from the thesis.

••I Canada ABSTRACT

BIOCONTROL OF FOODBORNE BACTERIAL PATHOGENS USING

IMMOBILIZED BACTERIOPHAGE

Hany ElSaid Mohamad Anany Advisor: Dr. Mansel W. Griffiths

University of Guelph, 2010

The goal of the present research was to develop a simple technique to immobilize isolated lytic phages and explore the potential use of these immobilized phages to control certain foodborne pathogens in a real food system. More than one hundred phages were isolated from different environmental samples against different strains of four major foodborne pathogens; E. coli 0157:H7, Salmonella, Listeria monocytogenes and

Shigella. A turbidimetric method in high throughput format using the Bioscreen C was used to monitor phage lytic activity and determine the host range of the isolated phages.

This enabled identification of isolated phages with similar characteristics and the selection of twelve good phage candidates for biocontrol purposes. These phages were characterized by TEM, restriction endonuclease pattern, one-step growth curve, BIM

(bacteriphage insensitve mutant) development, cross infectivity and determination of their stability and infectivity under different conditions. A novel Shigella phage, OSboM-

AG3, was isolated and its genome sequenced. Its genome did not show any homology to any reported virulent or lysogenic genes and it was considered as a member of the "T4 superfamily". Phage cocktails made from the isolated phages were very effective for specific control of the four target pathogens in both broth media and contaminated food

stored under different environmental conditions. The phages were immobilized using positively charged carrier substrates, which allowed specific binding of phages through their heads and leaving tail fibers free to interact with bacteria. These immobilized phages retained infectivity and immobilized Listeria and E. coli phage cocktails were able to control the growth of L. monocytogenes and E. coli 0157:H7, respectively, in food under different temperatures and packaging conditions. ACKNOWLEDGEMENTS

In the first place, I would like to express my sincerest gratitude to Allah, who provided me the blessing to complete this work. He showed me numerous times that HIS support and guidance exist in every step of the way.

Also, I would like to express my deepest gratitude to my advisor, Dr. Mansel W.

Griffiths, for his supervision, guidance, support, kindness and patience from the very early stage of this research as well as giving me extraordinary experiences throughout the work. Above all and the most needed, he provided me with so much encouragement and support in various ways on both the scientific and personnel levels. His truly scientist nature has made him as a constant source of ideas and passions in science, which exceptionally inspire and enrich my growth as a student, a researcher and a scientist want to be. I am indebted to him more than he knows. One simply could not wish for a better or kind supervisor!

I am heartily thankful to my advisory committee members, Dr. Andrew Kropinski and Dr. Parviz Sabour for their invaluable advice, support and words of encouragement.

They gave me from their precious times to gain from their exceptional experience in phage biology and genomic studies. Really, without their help, I could not have completed this research. I also would like to thank Dr. Robert Pelton and Dr. Micheal

Brook for their collaboration in this research and providing the positively charged cellulose membranes and silica beads. Many thanks should also go to Dr.Yi-Min She for performing the mass spectrometry analysis of O SboM-AG3 proteins.

This work has benefited from the insights, directions and help of many people:

i H.-W. Ackermann, Lynn Mclntyre, Luba Brovko, Ann Blake, Haifeng Wang, Kieth

Warriner, Milena Corredig, Robert Harris, Sandy Smith, Mona Tolba and Erika Lingohr.

Each of them provided me with invaluable comments, advices and help which substantially improved the finished product.

I am grateful to my friends and lab mates from Dr. Griffiths's lab; they made our lab such a convivial place to work, providing me with the excellent atmosphere for doing my research. In particular, I would like to thank Tarek El-Arabi for his friendship, help and support in the past four years. My special thanks go to all my Egyptian friends in

Guelph, especially Mumdooh Ahmed and his respectful family, for their support they offered to my family and to me during my study. I would like also to thank the Egyptian government and Sentinel bioactive network for the financial support throughout my PhD program.

My deepest gratitude goes to my family for their love and support throughout my life and considerably extended academic career. I am indebted to my father, El-Said

Anany and my mother, Khyria Ahmed for their care, love and support. I have no enough words that can fully describe what they have done for my little family and for me. They are simply perfect parents! My brothers, Bassem and Ahmad have always been a consistent source of encouragement and support. Finally, yet importantly, I am so grateful to my soul mate and my wife, Nahed Anany, she has made her love and support available in a number of ways throughout my life, especially during my PhD study. This dissertation is simply impossible without her! I would like also to thank my beautiful kids, Omar, Aly and Yara for their love and prayers for me. They were the motive force that inspired me throughout this research.

ii TABLE OF CONTENTS

ACKNOWLEDGEMENTS I

LIST OF TABLES VII

LIST OF FIGURES IX

CHAPTER 1: INTRODUCTION 1 1.1 RESEARCH INTRODUCTION 1 1.2 FOODBORNE ILLNESSES AND PATHOGENIC BACTERIA 2 1.2.1 Escherichia coli 0157:H7 4 1.2.2 Salmonella spp 6 1.2.3 Shigella spp 9 1.2.4 Listeria monocytogenes 10 1.3 OVERVIEW OF BACTERIOPHAGE 13 1.3.1 Bacteriophage Discovery 13 1.3.2 Bacteriophage Biology 14 1.3.3 Bacteriophage Taxonomy 15 1.3.4 Lytic and Lysogenic Pathway of Bacteriophage 18 1.4 BACTERIOPHAGE AS BIOCONTROL TOOLS FOR FOODBORNE PATHOGENS 24 1.4.1 An overview and considerations for phage application in food 24 1.4.2 Examples for the experimental proof of concept for phage application as a biocontrol agent 35 1.5 IMMOBILIZATION OF PHAGES ON SOLID SURFACES: A NEW PROPOSED TECHNIQUE FOR FOOD APPLICATION 42 1.6 RESEARCH OBJECTIVES. 52

CHAPTER 2: ISOLATION AND CHARCTERIZATION OF LYTIC BACTERIOPHAGES AGAINST IMPORTANT FOODBORNE PATHOGENS ...54 2.1 ABSTRACT 54 2.2 INTRODUCTION 55 2.3 MATERIALS AND METHODS 57 2.3.1 Bacteria and Bacteriophage 57

in 2.3.2 Enrichment and isolation of phages 57 2.3.3 Purification of phages 60 2.3.4 Propagation and stock preparation 60 2.3.5 Host Range Determination using Bioscreen C 61 2.3.6 Phage DNA Isolation and Restriction Endonuclease Digestion 62 2.3.7 Transmission Electron Microscopy 63 2.3.8 Stability of Phages under Different Temperatures and pH's 63 2.3.9 Infectivity of Phages under Different Environmental Conditions 64 2.3.10 Determination of the Frequency of Emergence of Bacteriophage Insensitive Mutants (BIM) and Lysogenic Potential 65 2.3.11 Cross Infectivity 65 2.3.12 Determination of the Phage Genome Size using PFGE 66 2.3.13 One-step growth curve 66 2.3.14 Statistical analysis 67 2.4 RESULTS 67 2.4.1 Isolation of phages 67 2.4.2 Host range pattern and determination of the identical isolated phages using Bioscreen C 68 2.4.3 Characterization of the selected phages 78 2.4.3.1 Morphology 78 2.4.3.2 One-step growth curve 86 2.4.3.3 Restriction enzyme digestion patterns and determination of genome sizes ..86 2.4.3.4 Effect of different temperatures and pH's on viability of the selected phages 90 2.4.3.5 Effect of different environmental conditions on the infectivity of the selected phages 95 2.4.3.6 Cross infection 95 2.4.3.7 Emergence of bacteriophage insensitive mutants (BIM) and lysogenic ability 96 2.5 DISCUSSION 101

CHAPTER 3: SEQUENCING AND GENOME ANALYSIS OF SHIGELLA PHAGE (

IV 3.3.1 Bacteria and Bacteriophage 112 3.3.2 Phage purification, DNA isolation and sequencing 113 3.3.3 Genome annotation 114 3.3.4 Proteome analyses 115 3.3.5 Genome sequence 116 3.4 RESULTS 116 3.4.1 General features of the OSboM-AG3 genome 116 3.4.2 Identification and analysis of open reading frames (ORFs) 117 3.4.3

CHAPTER 4: USE OF A COCKTAIL OF PHAGES TO CONTROL FOODBORNE PATHOGENS IN LIQUID MEDIA AND IN A REAL FOOD SYSTEM 149 4.1 ABSTRACT 149 4.2 INTRODUCTION 150 4.3 MATERIALS AND METHODS 153 4.3.1 Bacteria and Bacteriophage 153 4.3.2 Effect of different multiplicity of infection values on the growth of the target pathogens as determined using a Bioscreen C Microbiology Plate Reader 155 4.3.3 Bacterial challenge test 156 4.3.4 Potential of the phage cocktails to control Listeria monocytogenes and E. coli 0157:H7infbod 156 4.3.5 Effect of Listeria phage cocktail on biofilm formation by Listeria monocytogenes 158 4.3.6 Statistical analysis 160 4.4 RESULTS 160 4.4.1 Effect of different multiplicity of infection values on the growth of the target pathogens as measured using the Bioscreen C 160 4.4.2 Bacterial challenge test 166 4.4.3 Potential use of the phage cocktails to control L. monocytogenes and E. coli 0157:H7infood '. 169 4.4.4 Effect of Listeria phage cocktail on biofilm formation by L. monocytogenes ...175 4.5 DISCUSSION 177

v CHAPTER 5: IMMOBILIZATION OF PHAGE COCKTAILS ON CHARGED CELLULOSE MEMBRANES AND THEIR BIOCONTROL APPLICATION ...183 5.1 ABSTRACT 183 5.2 INTRODUCTION 184 5.3 MATEREIALS AND METHODS 187 5.3.1 Bacteria and Bacteriophage 187 5.3.2 Immobilization of phages on surface modified silica particles 189 5.3.3 Immobilization of phages on positively charged and unmodified cellulose membranes 190 5.3.4 Investigating the overall charge difference between phage head and tail structures 192 5.3.5 Transmission Electron Microscopy 193 5.3.6 Effect of dryness on stability of phages 193 5.3.7 Potential application of the immobilized phage cocktails on cellulose membranes to control foodborne pathogens on meat surface 194 5.3.8 Statistical analysis 197 5.4 RESULTS 197 5.4.1 Immobilization of phages on surface modified silica particles 197 5.4.2 Immobilization of phages on positively charged cellulose membranes 201 5.4.3 Investigating the overall charge difference between phage head and tail structures 206 5.4.4 Effect of drying on the stability of phages...... 206 5.4.5 Potential application of the immobilized phage cocktails on positively charged cellulose membranes to control foodborne pathogens on meat surfaces.... 210 5.5 DISCUSSION 218

CHAPTER 6: CONCLUSIONS AND FUTURE DIRECTIONS 226 6.1 Thesis summary and general conclusion 226 6.2 Future research 233

REFERENCES 236

APPENDIX 262

VI LIST OF TABLES

Table 2.1. Bacterial strains that were used for phage isolation and propagation 59

Table 2.2. Host range pattern of the isolated Salmonella phages using Bioscreen C. Similar host range patterns were grouped together 71

Table 2.3. Host range pattern of the isolated Listeria phages using Bioscreen C. Similar host range patterns were grouped together 72

Table 2.4. Host range pattern of the isolated E. coli phages using Bioscreen C. Similar host range patterns were grouped together 74

Table 2.5. Host range pattern of the isolated Shigella phages using Bioscreen C. Similar host range patterns were grouped together '. 75

Table 2.6. Host range pattern of selected phages 80

Table 2.7. Selected isolated phages and their susceptible bacterial hosts used for propagation 84

Table 2.8. Approximate dimensions, family and morphologically related phages for the selected phages 84

Table 2.9. Latent period and burst size of the phages that represent different morphotypes among the finally selected phages 87

Table 2.10. Estimated size of genomic DNA of the finally selected phages based on the analysis of the PFGE results 90

Table 2.11. Effect of different aeration condition on the infectivity of the selected phages 97

Table 2.12. Effect of pH, salinity and temperature on the infectivity of the selected phages 98

Table 2.13. Cross infectivity of the selected phages to other bacterial hosts 99

Table 2.14. Bacteriophage insensitive mutant development of the selected phages and their cocktails on their propagating hosts 100

vii Table 3.1. General features of putative ORFs of OSboM-AG3 and homology to proteins in the databases 120

Table 3.2. Genes which constitute the core genome of the T4-like phages are determined by CoreGenes comparisons of T4, 44RR2.8t, RB43, RB49 and P-SSM2 136

Table 3.3. Number of the transmembrane domains found in the genome of OSboM-AG3 using Phobius and TMHMM software packages. Bold ORF's are suspected holin genes 139

Table 3.4. Putative promoters in the genome of OSboM-AG3 phage 140

Table 3.5. Putative rho-independent terminators in the genome of OSboM-AG3 phage predominantly discovered and verified using MFOLD 140

Table 3.6. Location of the tRNA genes in the OSboM-AG3 genome, their cognate amino acids and anticodons detected using Aragorn 141

Table 3.7. Comparison of the tRNA codon usage in OSboM-AG3 and its host Shigella boydii. 141

Table 3.8. Summary of the open reading frames identified by the mass spectrometry analysis of CsCl-purified Vi-I like OSboM-AG3 phage 144

Table 4.1. Phages and bacterial strains that used for phage propagation and biocontrol experiments 154

Table 5.1. Phages and bacterial strains that used for phage propagation, immobilization and biocontrol experiments , 189

Table 5.2. Amounts of surface modification agent (APTS: 3-aminopropyl triethoxysilane) added to silica particles during preparation method and the produced mobility 1 199

viii LIST OF FIGURES

Figure 1.1. Morphotypes of bacteriophages 17

Figure 1.2. Life cycle depicting the lytic and lysogenic pathways of a typical bacteriophage when it infects bacterial cell 23

Figure 2.1. Representative results from Bioscreen C after adding phage to a tested bacterial strain 69

Figure 2.2. Representative transmission electron micrograph of some isolated Salmonella phage groups that showed similar host range pattern 76

Figure 2.3. Electrophoresis on 1.0 % agarose of AccI restriction enzyme digests of DNA from representative Salmonella phages that showed similar host range patterns 76

Figure 2.4. Representative transmission electron micrographs of some isolated Listeria phage groups that showed similar host range pattern... 77

Figure 2.5. Electrophoresis on 1.0% agarose of EcoRI restriction enzyme digests of DNA of representative Listeria phages that showed similar host range patterns 77

Figure 2.6. Representative transmission electron micrographs of the different morphotypes of the twelve selected phages 85

Figure 2.7. One-step growth curves of the phages that represent different morphotypes among the finally selected phages 87

Figure 2.8. Restriction fragment produced from digestion of Salmonella phage genomic DNA with endonuclease AccI and Ndel 88

Figure 2.9. Restriction fragment produced from digestion of Shigella phage genomic DNA with endonuclease AccI and Ndel 88

Figure 2.10. Restriction fragment produced from digestion of genomic DNA of E. coli phages with endonuclease EcoRV and Sspl 89

Figure 2.11. Restriction fragment produced from digestion of Listeria phages genomic DNA with three restriction endonucleases 89

ix Figure 2.12. Stability of Salmonella phages stored for 24 hours under different environmental conditions 93

Figure 2.13. Stability of Listeria phages stored for 24 hours under different environmental conditions 93

Figure 2.14. Stability of Shigella phages stored for 24 hours under different environmental conditions 94

Figure 2.15. Stability of E. coli phages stored for 24 hours under different environmental conditions 94

Figure 3.1. Genetic and physical map of phage OSboM-AG3 prepared using DNAPlotter with proteins that shows homology with T4 subfamily indicated in the outer rim 119

Figure 3.2. SDS-PAGE of OSboM-AG3 143

Figure 3.3. Electron micrograph of the Vi-I like

Figure 4.1. Effect of different MOI's ofSalmonella phage cocktail on the growth of three Salmonella strains 162

Figure 4.2. Effect of different MOI's of Listeria phage cocktail on the growth of three Listeria strains 163

Figure 4.3. Effect of different MOI's of Shigella phage cocktail on the growth of two Shigella strains 164

Figure 4.4. Effect of different MOI's of E. coli phage cocktail on the growth of two E. coli strains 165

Figure 4.5. Effect of Salmonella, Listeria, Shigella and E. coli phage cocktails on their corresponding susceptible hosts incubated at 4°C in TSB 167

Figure 4.6. Effect of Salmonella, Listeria, Shigella and E. coli phage cocktails on their corresponding susceptible hosts incubated at 25°C in TSB 168

Figure 4.7. Effect of Listeria phage cocktail on the growth of Listeria monocytogenes (C391) in RTE oven roasted turkey breast incubated aerobically at 25°C and 4°C for nine days 171

x Figure 4.8. Effect of Listeria phage cocktail on the growth of Listeria monocytogenes (C391) in RTE oven roasted turkey breast incubated under vacuum condition at 25°C and 4°C for nine days 172

Figure 4.9. Effect of Listeria phage cocktail on the growth of Listeria monocytogenes (C391) in RTE oven roasted turkey breast incubated under modified atmospheric packaging (MAP) condition at 25°C and 4°C for nine days 173

Figure 4.10. Effect of E. coli phage cocktail on the growth of E. coli 0157:H7 (C899) in raw beef meat incubated aerobically at 25°C, 10°C and 4°C for 2, 9 and 15 days 174

Figure 4.11. Effect of Listeria phage cocktail on biofilm formation by different concentrations of two strains of Listeria monocytogens 176

Figure 5.1. Number of infective phage on APTS modified silica particles after an overnight incubation and washing 199

Figure 5.2. Mobility of APTS modified silica particles with and without the addition of phage 200

Figure 5.3. Transmission Electron Microscope (TEM) images of AG 11 and AG3 specifically immobilized through their heads on cationic, APTS-modified silica particles, sample D .200

Figure 5.4. Reduction in the E. coli phage cocktail titre after removing both positively charged and unmodified cellulose membranes (log PFU/ml) 203

Figure 5.5. Number of phage plaques developed under positively charged and unmodified cellulose membranes on a lawn of E. coli 057:H7 strain 203

Figure 5.6. Bioluminescent signal from E. coli 0157:H7 (Lux) cells grown with starting inoculum of around 103 CFU/ml in the presence of positively charged and unmodified cellulose membrane treated with different concentrations of E. coli phage cocktail. Phage-free membranes were considered as control 204

Figure 5.7. Bioluminescent signal from E. coli 0157:H7 (Lux) cells grown with starting inoculum of around 105 CFU/ml in the presence of positively charged and unmodified cellulose membrane treated with different concentrations of E. coli phage cocktail. Phage-free membranes were considered as control 205

Figure 5.8. Deposition of negatively charged and positively charged gold nanoparticles on SboM-AG3 phage tail fibers and EcoM-AG2 phage head 207

XI Figure 5.9. Effect of air drying at 25°C and 37°C on the stability of phages of different morphotypes 209

Figure 5.10. Effect of adding polysacharrides at different concentrations on the stability of wild type T4 phage to the air drying effect 209

Figure 5.11. Effect of lyophilization, as a method of drying, on the stability of T4 phage with and without polysaccharides 210

Figure 5.12. Effect of the immobilized Listeria phage cocktail on growth of Listeria monocytogenes C391 on RTE oven roasted turkey breast samples incubated aerobically at 25°C, 10°C and 4°C 213

Figure 5.13. Effect of the immobilized Listeria phage cocktail on growth of Listeria monocytogenes C391 on RTE oven roasted turkey breast samples incubated under modified atmospheric packaging conditions at 25°C, 10°C and 4°C 214

Figure 5.14. Effect of the immobilized Listeria phage cocktail on growth of Listeria monocytogenes C391 on RTE oven roasted turkey breast samples incubated under vacuum packaging conditions at 25°C, 10°C and 4°C 215

Figure 5.15. Effect of the immobilized E. coli phage cocktail on growth of E. coli 0157:H7 (amp::/wx) C918 on raw beef samples incubated aerobically at 25°C, 10°C and 4°C 216

Figure 5.16. Bioluminescence activity of E. coli 0157:H7 (amp::lux) on the surface of raw beef incubated for one week at 10°C and 4°C arid then at 30°C for 16 hours ..217

Figure 5.17. Model showing modes of electrostatic interaction between phage and charged silica particles 219

xii Chapter 1: INTRODUCTION

1.1 RESEARCH INTRODUCTION

Recent years have witnessed a large number of foodborne outbreaks in many countries. Even though different policies are applied to ensure high hygiene and

sanitation standards, the pathogens can not always be eradicated from the finished product or food processing environment. The potential of the final product to become

contaminated during storage, slicing, preparation and display equipment is high. On the

other hand, the ability of the pathogen to survive and/or grow under unfavorable conditions and the development of resistant strains with new virulence factors, represent

a formidable challenge to food processing industries in marketing safe food products. For these reasons, thinking of safe alternative ways to mitigate and/or replace the existing protocols to control foodborne pathogens has been initiated. The idea of destroying pathogens by nature's own method was very attractive. In this context, bacteriophages

(phages) have emerged as a new biocontrol tool as it holds enormous possibilities as a

safe weapon for fighting infectious diseases. The specificity of interaction of phage to its host cell could be exploited to control pathogenic bacteria without affecting the viability of other microorganisms in the habitat. Phage biocontrol strategies for food preservation have the advantages of being self-perpetuating, highly discriminatory, natural, safe to human and cost effective. However, the use of phages as biocontrol agents is complex and many factors such as a limited host range, phage-resistant mutants, and the potential for the transduction of undesirable characteristics from one bacterial strain to another may influence its efficacy. Therefore, it is critical to develop highly virulent, lytic, broad

1 spectrum, stable and non-transducing phage before its widespread approval as a food preservative.

The available data clearly indicate that phage can contribute to reduce the number of many pathogens either in laboratory media or in the real food system but to different degrees, which was unexpected. This may be due to the effect of food matrices on the activity of the virus, or the effect of diffusion, which in turn affects the concentration of the phage and/or the interaction between virus and its host. The proposed methods for application of phages in the processing facilities have some limitations and might be one of the reasons for consumers' objections to the adoption of this technology in food.

Hence, novel techniques for application should be innovated. Oriented immobilization of phage on cheap, solid substrates like cellulose, which is associated with many food product packages, and application of the whole system over the surface of food products, may be a good alternative to provide persistent and effective control of potential pathogens. Moreover, it would minimize direct addition of phages and satisfy the actual demand of consumers for healthier foods that contain fewer additives. In this context, the existing immobilization protocol for T4 phage has required phage to be fully sequenced, identified and mutated to get the recombinant phage that can be immobilized on cellulose, which is time and cost consuming. Therefore, more research is required to establish an alternative protocol to immobilize non-sequenced phages.

1.2 FOODBORNE ILLNESSES AND PATHOGENIC BACTERIA

Recently, greater attention is being given to the emergence and re-emergence of foodborne pathogens that have a big impact on the public health (Hagens and Loessner,

2 2007, Rees and Dodd, 2006, Kothary and Babu, 2001). Foodborne illnesses can cause, in

addition to common acute symptoms such as diarrhea, fever, abdominal pain, nausea and

vomiting, severe diseases that may lead to death even at low infectious doses such as in

immunocompromised individuals (Bell and Kyriakides, 2002b).

It was reported that up to 30% of the population in developed countries would get

sick from the food and water they consume each year (World Health Organization, 2000).

In the United States, the Centers for Disease Control and Prevention (CDC) estimated

that approximately 76 million cases of food-related illness (resulting in 5,000 deaths and

325,000 hospitalizations) occurred each year (Mead, 1999). In a more recent report, acute

foodborne illnesses cost the United States around $152 billion a year (Scharff, 2010). It

cost Canada around CAN$ 1.33 billion in 1985 and £ 1.9 billion annually in the U.K

(Snowdon et al, 2002). However, data from the developing countries where people are

more exposed to foodborne illnesses were not available (Stein et al, 2007). Thus, the

incidence and cost of foodborne illnesses prompted an urgent call to improve methods for minimizing or preventing the occurrence of pathogens in food products. Moreover, the

increasing demand for minimally processed and organic foods requires the development

of natural antimicrobials to control bacterial contamination. Currently, different chemicals and antimicrobial compounds are applied to achieve this goal, however the worldwide increase in bacterial resistance and potential side affect of these antimicrobials have stimulated researchers to find alternative ways to combat foodborne pathogens and

spoilage organisms (Mclntyre et al, 2007, Hudson et al, 2005, Rees and Dodd, 2006,

Hagens and Loessner, 2010).

Although foodborne illness may result from various pathogenic microorganisms,

3 bacterial pathogens are the etiological agents in the majority of recent reported outbreaks.

In the 2006 annual report of the C-EnterNet project in Canada, three bacterial pathogens and one parasite were reported to be responsible for 82% of the studied outbreak cases in the Waterloo region (Public Health Agency of Canada, 2007). In Ontario alone, eight enteric pathogens (Campylobacter, Salmonella, verotoxin-producing E. coli, Yersinia,

Shigella, Hepatitis A, Listeria and Clostridium botulinum) were reported to be responsible for 44,451 sporadic cases from 1997 to 2001 (Lee and Middleton, 2003). The major cause of 74.0% of these outbreaks was foodborne contamination. Moreover, in this study fresh produce, poultry and other meat products were identified as the main vehicle of these pathogens. In a recent report for the Foodborne Diseases Active Surveillance

Network (FoodNet) of CDC's emerging infections program, there were 15.19 cases of

Salmonella and 0.34 of Listeria cases per 100,000 people in the United States during

2009, comparing to the national health targets of 6.8 and 0.25 respectively (Centers for

Disease Control and Prevention, 2010).

E. coli 0157:H7, Salmonella, Listeria monocyogenes and Shigella are among the most important foodborne bacterial pathogens that have been associated with recent outbreaks of foodborne illness (Centers for Disease Control and Prevention, 2005).

1.2.1 Escherichia coli 0157:H7

Escherichia coli strains are one of the predominant Gram-negative bacteria in both the human and animal gut. E. coli cells are facultative anaerobes and non-spore forming rods. The species belongs to the family Enterobacteriaceae. The presence of somatic (O), flagellar (H) and capsular (K) antigens are used to differentiate serotypes of this species (Feng, 2001). Generally, E. coli strains are harmless to their host; however,

4 some pathogenic strains have emerged which cause diseases to humans and animals.

These strains have been grouped based on their unique virulence factors. The pathogenic groups have been recognized as enteropathogenic (EPEC), enteroinvasive (EIEC), enterotoxigenic (ETEC), enterohemorrhagic (EHEC) and diffusely adherent (DAEC)

(Nataro and Kaper, 1998, Willshaw et al, 2000). Even among healthy individuals, the fatality rate of EHEC infection compared to other E. coli infections is high (Hagens and

Loessner, 2010). E. coli 0157:H7 is one EHEC strain that has generated great interest around the world as it is exhibiting the highest morbidity and mortality rates amongst all pathogenic E. coli strains. Moreover, it has been described as the cause of many sporadic and outbreak-associated hemorrhagic colitis cases (Reiss et al, 2006). The infectious dose of E.coli 0157:H7 is as few as 10 cells and in severe infection, damage may occur to the kidneys resulting in a potentially fatal haemolytic uremic syndrome (HUS), which is most commonly observed in young children (O'Flynn et al, 2004). The principal reservoir of E. coli 0157:H7 is ruminants, in particular cattle. Reported outbreaks are associated with different food sources of bovine origin such as undercooked beef, raw milk, as well as cold sandwiches, water, unpasteurized apple juice, sprouts and leafy green vegetables(Feng, 1995). Dry fermented salami, which is processed to create unfavourable conditions for microbial growth and survival, has also been identified as a vehicle for E. coli 0157:H7 infections (Conedera et al, 2007). Due to recent E. coli 0157:H7 outbreaks linked to contaminated lettuce and other leafy greens, FDA is going to announce mandatory rules (as opposed to voluntary recommendations) which will set enforceable standards for production and packaging of fresh produce to ensure reduction of the associated illness (Falkenstein, 2010).

5 In Canada the annual cases of verotoxigenic E. coli, which includes the E .coli

0157:H7 serotype, numbered between 1200 and 1700 cases from 1990 to 1996 (4.1 cases

per 100,000 Canadians). Moreover, the average annual number of cases in Ontario is

estimated in the same study to be 492 cases between 1994 and 1998 (Williams et al,

2000). In 2006, the Public Health Agency of Canada reported 35 human cases of E. coli

0157:H7 (7.3 cases per 100,000 persons) in the Waterloo region (Public Health Agency

of Canada, 2007). In the United States, in 2007, one million pounds of ground beef were

recalled due to E. coli 0157:H7 contamination (United States Department of Agriculture

Food Safety and Inspection Service, 2008). E. coli 0157:H7 can be controlled mainly by

the application of effective cleaning and hygiene procedures during the whole food

production chain, starting from the raw materials to the retail or catering outlet. The

treatment and formulation of the whole food product can play an important role to create unfavourable conditions for E. coli 0157:H7 growth to prevent harmful effects on the

consumers and maintain food safety (Bell and Kyriakides, 2002d).

1.2.2 Salmonella spp.

Salmonella spp. have been considered to be one of the most important etiological

agents of foodborne illness around the world. Since its discovery in 1885, Salmonella has been related to many outbreaks of foodborne diseases in different countries (Bell and

Kyriakides, 2002a). Salmonella spp. are Gram negative, facultative anaerobic and usually motile small rods. The surface antigens and phage typing has led to the current recognition of almost 2400 serotypes under 2 species of Salmonella, S. enterica and S.

bongori. Most foodborne serotypes belong to S. enterica (e.g. Salmonella enterica subsp. enterica serotype Typhimurium, which is commonly mentioned as Salmonella

6 Typhimurium) (D' Aoust, 2001). Salmonella strains are included in chicken and turkey

microbiota. Therefore, during slaughter of colonized animals, meat is regularly

contaminated and eggs may carry the bacterium as well (Hagens and Loessner, 2010).

Salmonellosis results from the ingestion of Salmonella-contaminated food products such as poultry and dairy products. The symptoms range from mild to severe

gastroenteritis to enteric fever. In more severe cases, the organism can migrate into the blood stream and/or the lymphatic system and result in bacteraemia or septicaemia which

may result in serious disorders such as osteomyelitis, cardiac inflammation and/or neural

disorders (D'Aoust, 1994). The development of multiple-antimicrobial resistant strains of

Salmonella, increases the severity of the problem. For instance, Salmonella Typhimurium

DTI04 in Europe and North America was reported to be resistant to at least five antibiotics, enhancing its role in many recent outbreaks of severe gastrointestinal

infections. Furthermore, novel resistant serotypes with no recognized phage type have emerged and reported to be associated with foodborne illness (Besser et al, 2000, Davis et al, 2007). Salmonella Typhimurium DTI04 outbreaks have been associated with various food items such as unpasteurized dairy products, pork sausage, chicken, meat paste and fresh apple cider. It represented about 38% of the human reported case of

salmonellosis in 2000 in Canada (Dore et al, 2004). Hydrolyzed vegetable protein (HVP) has been identified recently as another vehicle for Salmonella contamination after U.S.

Food and Drug Administration found Salmonella Tennessee in this flavor enhancer which is used in wide variety of processed food products such as hot dogs, sauces, dips and dressings (The U.S. Food and Drug Administration, 2010b). Another food ingredient, black pepper was associated with a recent outbreak in the United states when 245 people

7 have been infected with Salmonella Montevideo due consumption of food products contain this ingredient (The U.S. Food and Drug Administration, 2010a).

Salmonellosis was one of the three most commonly reported enteric diseases in the Sentinel Site 1 of Canada's C-EnterNet Project with an incidence rate of 36% of all enteric diseases reported (Public Health Agency of Canada, 2007). A trend of increased

incidence rates was found among infants and the elderly for salmonellosis. Swine and dairy cattle were found to be reservoirs for Salmonella. However, raw chicken was found to be the source for contamination at home. The United States Centre for Disease Control and Prevention reports approximately 40,000 cases of Salmonella yearly, although the

actual number might be higher due to the minor and unreported cases (Centers for

Disease Control and Prevention, 2006). Additionally, it was estimated that around 98,204 raw meat and poultry products were involved in the Salmonella outbreaks in the U.S.

from 1998 to 2000 (Rose et al, 2002). One recent well-known recall, due to Salmonella contamination, occurred in 2006 when Cadbury had to recall approximately one million chocolate bars costing the company around £20 million (Food Standards Agency UK,

2006). Moreover, several brands of pistachio and pistachio products have been recalled in the United states in 2009 due to potential Salmonella contamination (The U.S. Food and

Drug Administration, 2010d). In one of the worst known outbreaks in the United Sates, nine deaths and 714 persons infected with Salmonella Typhimurium have been recently reported due to consumption of contaminated peanut products which led to recall around

19 human and pet food products of different brands (The U.S. Food and Drug

Administration, 2010c). In addition to utilizing in-process strict hygiene systems and procedures, formulation of the food products to create growth-limiting conditions could

8 be applied to control Salmonella and prevent its adverse effect (Bell and Kyriakides,

2002a).

1.2.3 Shigella spp.

Shigella is another important foodborne pathogen belonging to the

Enterobacteriaceae family. It is a Gram-negative, non-motile facultative anaerobic rod.

According to genetic studies, it is more related to Escherichia than Salmonella (Jay,

2000). The biochemical and serological characteristics identify four main species under the genus Shigella : S. sonnei, S. flexneri, S. boydii and S. dysenteriae. Although all the

species have the ability to cause human disease, only the first three species are generally regarded as foodborne pathogens. Humans are the main reservoir of infection and the

infectious dose may be as low as 10 CFU (Kothary and Babu, 2001, Sutherland and

Varnam, 2002). Shigellosis is a foodborne disease that is developed within 12- 48 hours after the ingestion of Shigella contaminated food. Shigella colonizes and penetrates the colon epithelial cells. The initial symptoms include fever, aches, fatigue and loss of appetite, which may be associated with watery diarrhea that may manifest into bloody stools or dysentery. In certain severe cases, fatal hemolytic-uremic syndrome (HUS) may develop due to the production of Shiga toxin (Acheson, 2001).

Although shigellosis is more common in tropical developing countries where hygiene standards are low, it is also reported in developed countries. It has been estimated that Shigella causes around 450,000 illnesses, 6000 hospitalization and 70 deaths each year in the United States (Mead, 1999). Recently, the problem has increased due to the high prevalence of antimicrobial resistant strains among the isolated Shigella strains in the U. S. (Sivapalasingam et al, 2006). It was reported that the main source of

9 Shigella-related outbreaks is through direct or indirect human fecal contamination. Many foods have the potential to be contaminated so a wide range of food products has been implicated as vehicles of shigellosis transmission. Generally, where hygiene standard are poor, foods that receive significant handling during preparation such as salads, soft cheese, vegetables and meat products are at the greatest risk of contamination by Shigella

(Acheson, 2001). A recent Shigella sonnei outbreak was reported among air travelers who departed from Hawaii on 12 flights dispersed to Japan, Australia, 22 US states, and

American Samoa (Gaynor et al, 2009). It was estimated that 300-1500 passengers were infected and the raw carrot served onboard was considered as the likely vehicle of infection. This outbreak illustrates the risk of rapid, global spread of illness from a single point of contamination. In order to control this pathogen, several approaches can be taken such as; minimize hands-on procedures; exclude potential execretors of Shigella from handling food and ensure good standards of personal hygiene (Sutherland and Varnam,

2002).

1.2.4 Listeria monocytogenes

The genus Listeria currently consists of six species: L. monocytogenes, L. ivanovii, L. innocua, L. welshimeri, L. seeligeri, and L. grayi . They are Gram-positive, facultative anaerobic, non-spore forming and motile species. Although there are very few reports for L. ivanovii and L. seeligeri to infect human, L. monocytogenes is regarded as the major cause of illness in human among all six Listeria species. It is an opportunistic pathogen that is able to colonize the intestine and cause listeriosis (Donnelly, 2001).

Listeriosis usually causes flu-like symptoms including fever, muscle aches, and gastrointestinal symptoms such as nausea or diarrhoea, while in high risk groups,

10 including pregnant women, neonates, elderly and immunocompromised persons, the

disease may lead to fatal disorders such as meningitis, bacteraemia and abortion in

pregnant women (Swaminathan and Gerner-Smidt, 2007). L.monocytogenes contains 14

serotypes, but l/2a, l/2b and 4b are the most common serotypes associated with

foodborne illness (Donnelly, 2001). This pathogen is considered to be important

foodborne pathogen due to its ability to survive and grow over a wide range of

temperatures from -1.5°C to 50°C and pH ranges of 4.3 to 9.6; it is resistant to high salt

concentrations up to 25.5%; and it can tolerate freezing and drying; all of which indicate

that it may survive and contaminate most refrigerated food and ready-to-eat food

products (Donnelly, 2001). Therefore, it is able to continue growing under refrigeration

conditions and can become endemic in cold storage facilities. Indeed once it formed the

biofilm, this organism can develop resistance to standard cleaning agents (Frank and

Koffi, 1990). Although listeriosis is a rare disease when compared with other foodborne

pathogens, it has a high mortality rate of approximately 30% of its cases especially in

high-risk groups. It is also found ubiquitously in the environment and can be transferred

to people through contamination of food products (Hagens and Loessner, 2007,

Swaminathan and Gerner-Smidt, 2007).

The Center for Disease Control and Prevention reported that there were about

2,500 confirmed cases of severe listeriosis each year in the United States; of these about

500 die (Centers for Disease Control and Prevention, 2005). In Canada, the incidence rates were estimated to be 3.4 cases per million persons in 1998 and 2.7 cases per million persons in 1999 (Food Safety Network, 2003). The number of cases of listeriosis has

stabilized or is on the rise in Europe, after having undergone a steep decline in the first

11 part of the last 20 years (Hagens and Loessner, 2010). L. monocytogenes is usually killed

during pasteurization or other heat treatments. This means that the highest risk foods are

those that are not heated before consumption. Several large outbreaks of listeriosis have

been related to different contaminated fresh and ready-to-eat food products including

dairy products, meat, egg products, vegetables and seafood (Farber and Peterkin, 1991).

The contamination occurred in different stages during food production and processing via

ingredients, factory workers, contaminated faulty production equipment, or the factory

environment. In a very recent study, slicers, conveyor belts and sandwich ingredient

were found to be among the major sources for L. monocytogenes contamination in one of

Swiss sandwich-producing plants. Moreover, it was found that certain strains persisted

for more than nine months on slicers and conveyor belts, which increased the chance of

sandwich contamination (Blatter et al, 2010). Recalls of the contaminated products are

costly to the food industry because of the product loss and loss of customer confidence in these products (Swaminathan and Gerner-Smidt, 2007). An obvious recent example happened in 2008 when a massive recall of Maple Leaf Foods was launched after Listeria

was detected in some of the ready-to-eat products at one of the company's plants in

Ontario, Canada due to contaminated slicing machines. This contamination led to 57 confirmed cases of listeriosis across the country, of those cases there were 23 deaths

(Public Health Agency of Canada, 2010). Maple Leaf Foods estimated that the recall directly cost the company at least CAD$20 million plus around CAD$ 29 million for

settlement of class action suit. This is in addition to further losses expected due to lost

sales and advertising to rebuild its image. It was reported that eliminating L. monocytogenes from most food is difficult as it invades the food-processing environment

12 better than other pathogens. However, it is possible to control its number in food to minimize the associated hazard to consumers. This can be accomplished by applying strict hygiene and sanitation procedures throughout the whole food production process, in addition to controlling the formulation, treatment and storage of the food products to create growth limiting conditions for L. monocytogenes (Bell and Kyriakides, 2002c).

1.3 OVERVIEW OF BACTERIOPHAGE

1.3.1 Bacteriophage Discovery

Bacteriophages (phages) are bacterial viruses that only infect and multiply within their specific hosts, disrupt bacterial metabolism and cause the bacterium to lyse. They were discovered a long time ago but the history of bacteriophage discovery has been the subject of lengthy debates. Ernest Hankin, a British bacteriologist, reported in 1896 that the waters of the Ganges and Jumna rivers in India had marked antibacterial action against Vibrio cholerae and that ingestion of the water of these rivers prevented spread of cholera epidemics. He suggested that an unidentified substance (which passed through fine porcelain filters and was heat labile) was responsible for this phenomenon and for limiting the spread of cholera epidemics. The same observation was made two years later by the Russian bacteriologist Gamaleya, while working with Bacillus subtilis (Deresinski,

2009). However, none of these investigators further explored their findings until

Frederick Twort (1915), a British pathologist, and a French-Canadian bacteriologist;

Felix d'Herelle (1917) at the Pasteur Institute in Paris independently reported isolating filterable entities that could destroy bacterial cultures and produce small clear areas on bacterial lawns. D'Herelle called them "bacteriophages" (Wikipedia, 2010, Summers,

13 2005, Sulakvelidze et al, 2001, Summers, 2001).

1.3.2 Bacteriophage Biology

Phages are the largest group of viruses, utilizing species in the Bacteria and

Archaebacteria as hosts and they measure 20 to 200 nanometeres (Ackermann and

DuBow, 1987). They are the most abundant form of life on the planet; there are an estimated 1031 phages in the biosphere (Kutter and Sulakvelidze, 2005). Like other viruses, phages are infectious particles that have at least two components, genome surrounded by protein subunits that form capsid (Ackermann, 2003). It was suggested that the capsid plays three important roles in the phage life cycle: protecting the phage genome (e.g., from DNA-degrading enzymes) until finding the right host; effecting phage adsorption to a susceptible bacterium; and the subsequent delivery (uptake) of the phage genome into the cytoplasm of the now-infected bacterium (Gill and Abedon, 2003).

Some of the protein subunits of the capsid also play a role in packaging the genome, adsorbing to the host cell, and injecting the genome into the bacterial host (Maloy et al,

1994).

The capsid encloses a single copy of the genome which is usually one molecule of either double-stranded DNA (dsDNA), single-stranded DNA (ssDNA), double stranded

RNA (dsRNA), or single stranded RNA (ssRNA) (Guttman et al, 2005). Some phages have an extremely small genome, for example, E. coli phage R17, which only contains 4 genes and has around 3600 bases. Others are relatively large, for example E. coli phage

PB51 possesses a genome which is around 2.5 X 105 bases in length and encodes for over

240 genes (Birge, 1994).

Many but not all phages have tails attached to the phage head. The tail is a hollow

14 tube through which the nucleic acid passes during infection. The size of the tail can vary and some phages do not have a tail structure. In some phages, the tail is surrounded by a contractile sheath which contracts during infection of the bacterium. At the end of the tail, the more complex phages like T4 have a base plate and one or more tail fibers attached to it. Tail fibers contain proteins that recognize molecules on the surface of bacterial cell walls and limit their ability to attach to non-specific cells. Not all phages have base plates and tail fibers (Ackermann, 2005, Ackermann and DuBow^ 1987).

Although most of the tailed phages have no lipid in their structure, about 30% of tailed phages are readily inactivated by lipid solvents (acetone, chloroform, ether, toluene). However, the sensitivity to lipid solvents does not necessarily prove the presence of lipids in tailed phages (Ackermann, 1999). Sensitivity to ultraviolet (UV) light varies from one phage to another; single stranded DNA phages are more sensitive than other phages. Most tailed phages are stable at pH range from 5 to 9 and are inactivated by heating at 60°C for 30 min (Ackermann, 2007). Myoviruses with large heads are apparently more sensitive to freezing and thawing than other types. Tailed phages are best preserved by lyophilization or in liquid nitrogen after addition of 15-50% glycerol, but some are quickly inactivated under these conditions. Storage at 4°C is a good alternative for most phage preservation (Guttman et al, 2005, Puapermpoonsiri et al, 2010).

1.3.3 Bacteriophage Taxonomy

In a recent survey, at least 5568 bacterial viruses have been examined by the electron microscopy since 1959. About 96.2% of the examined phages are tailed (about

5360 phages) and only 208 phages (3.7%), are polyhedral, filamentous, or pleomorphic

15 (PFP) (Ackermann, 2007). The reported non-tailed phages belong to 17 families or

"floating genera" and most of them represent small virus families that are restricted to uncommon hosts, such as mycoplasmas and Archaebacteria (Ackermann, 2009). In the same survey, phages have been reported in 10 archeal and 144 eubacterial genera, enterobacteria genera have the most phage observations with around 906 phages

(Ackermann, 2007).

Tailed phages show a great variation in DNA content and composition, dimensions and fine structure, and physiology; for example, DNA sizes vary between 17 and over 700 kb and tail lengths range from 10 to 800 nm (Ackermann, 2003). They belong to the order Caudovirales which contains three families; Myoviridae (long, contractile tails; 24.5%), Siphoviridae (long, non-contractile tails; 61%) and Podoviridae

(short, non-contractile tails; 14%) which correspond to the morphological groups A, B and C, respectively, as shown in Figure 1.1. (Ackermann, 2007, Maniloff and

Ackermann, 1998, Ackermann, 1996, Ackermann, 2001).

These families include 18 genera named after their respective type viruses. The genera are defined based on a set of partially overlapping criteria relating to the genome and deoxyribonucleic acid (DNA) packaging such as presence ofpac sites and cohesive ends, circular permutations and terminal repeats, terminal proteins, DNA or ribonucleic acid (RNA) polymerases, or inclusion of unusual DNA bases into the phage genome

(Maniloff and Ackermann, 1998). Although it is of little taxonomical value, tailed phages of each family may be divided into three morphotypes, corresponding to phages with isometric, moderately elongated or very long heads (Ackermann, 2009). As more tailed

16 QQQ £ »,fl «,¥ r.,0 x\ EH n

EO

A2 83 C2 D4

F4 El O

F3

E2 »3 f f Bsjf CJ"

Si n G2

Figure 1.1. Morphotypes of bacteriophages. Types A, B and C corresponding to the

three families of order Caudovirales. They possess an isocahedral head with

a tail that is either long and contractile, long and non-contractile, or short

and non-contractile, respectively. Types D, E, F and G phages are

polyhedral, filamentous, or pleomorphic in morphology. Capsid shape and

fine structure were used to subdivide these types (Ackermann, 2001).

17 phages are being identified, tailed phage taxonomy is still at its beginning and more phage genera and species are likely to be defined in the future. Currently, there are six genera in the family Myoviridae, eight in the family Siphoviridae, and four in the family

Podoviridae (Ackermann, 2009, Fauquet and Fargette, 2005). Recently, the proteome analysis of 102 Myoviridae phages revealed that this family should contain three new subfamilies; Peduovirinae, Teequatrovirinae and Spounavirinae and new eight independent genera (Lavigne et al, 2009).

1.3.4 Lytic and Lysogenic Pathway of Bacteriophage

The phage life cycle can be one of two types, called lytic and lysogenic (Guttman et al, 2005). The choice between both types depends on the relative expression rates of phage repressor encoded by ell gene (promoting lysogeny) and cro protein, capable of turning off repressor gene expression and starting the lytic pathway (Campbell, 1994).

Lysogenic phages infect cells and incorporate their nucleic acid into the genome of the host cell or exist as an episomal element, leading to a permanent association as a prophage with the cell and all its progeny. During lysogeny, phages neither produce virions nor lyse bacteria. The phage is called temperate and the cells that harbor a prophage are known as lysogenic. Lysogenic relationship between temperate phage and its host bacterium provides a safe home to the temperate phage genome, blocks replication of non-virulent homologous phages and has the potential to alter the phenotype of the host cell; lysogenic (phage) conversion (Gill and Abedon, 2003). The lysogenic host bacterium may carry prophage for many generations until it is reactivated and produce new copies of phages that lead to lysis and release of progeny phages. The mechanisms of reactivation vary between phages, but are usually triggered when the host

18 cell is placed under adverse environmental conditions (Strauch et al, 2007).

Lytic phages, also called vegetative or productive, infect bacterial cells without

integrating their nucleic acid into the genome of the host. Inhibition of host-specific

synthesis occurs upon host cell infection by the phage. The host metabolic energy is then

subverted to the production of phage progeny. The lytic cycle, results in the lysis of the bacteria associated with the release of several phage particles. The new phages produced by the host bacterium spread to infect other cells. The time for whole cycle and the number of produced phage depend upon the phage type (Guttman et al, 2005,

Ackermann, 2003).

The typical lytic cycle of phages consists of the following sequential steps:

Adsorption or infection. Infection with tailed phages starts when specialized adsorption structures, such as fibers or spikes, bind to specific surface molecules on their target bacteria. Replication will only proceed if the cell contains specific receptor sites for the phage. Many types of molecules on the host cell's surface may serve as specific phage receptors. Proteins used by phages as receptors generally have important roles for the normal functioning of the bacterial cell. The nature of the bacterial receptor varies for different bacteria which are not only located on the cell wall but also on flagella, pili, capsules, or the plasma membrane (Lindberg, 1973). The environment is an important factor in the adsorption process and some cofactors may be required to enhance the adsorption. The most frequently required cofactors are Ca++ ions, followed by Mg++ ions

(Brussow and Kutter, 2005, Ackermann and DuBow, 1987). There are different factors that control the likelihood of phage attack on bacteria such as phage diffusion rate and phage density. Moreover, phage adsorption to its host is a function of phage-bacteria

19 chemical and physical interaction (Gill and Abedon, 2003).

Penetration or Injection. In most phage groups, only the phage nucleic acid enters

the host and the shell remains outside. When T4 phage contacts with outer membrane

receptors, conformational changes in the phage structure are initiated that lead to the

contraction of the tail sheath, which forces the hollow inner tube into the cell. The short-

tail fibers help anchor the baseplate to the cell surface receptors. In the meanwhile, the

baseplate shifts from a hexagon to a star-shaped structure. At this time, the whole tail

structure shrinks and widens, bringing the internal pin-like tube in contact with the outer

membrane of the bacterial cell and phage enzymes located on the tail tip degrade the

bacterial cell wall. As the tail tube punctures the outer and inner membranes of the cell,

the viral DNA is injected through the tail tube into the host cell's cytoplasm

(Kostyuchenko et al, 2003, Kostyuchenko et al, 2005). Tail contraction may be triggered

by a variety of agents or factors, such as alcohol, Cd(CNs), freezing and thawing,

formalin, H2O2, pH changes, urea, or sonication (Ackermann and DuBow, 1987). In

Siphoviridae phages, such as phage A., the tail sheath does not contract during DNA

injection. On the other hand, filamentous DNA phages of E. coli seem to enter the cell by being drawn into the inner membrane of the cell envelope while being uncoated; the

DNA is released intracellularly as the coat protein dissociates into subunits which remain

in the membrane (Gottesman and Oppenheim, 1994).

When the phage DNA enters the cell, it is subject to host exonuclease and restriction

enzymes. For that reason, many phages circularize their DNA quickly by sticky ends or terminal redundancies or by linear end protection. Some phages have other methods to

inhibit host nucleases or use an odd nucleotide in their DNA for protection (The National

20 Center for Biotechnology Information (NCBI), 2010)

The Latent Period. Immediately after the entry of the viral chromosome, the genes expressed early code for proteins which are needed to replicate the phage genome and which modify the cellular machinery so that the synthetic capacity of the cell is subverted to the reproduction of the phage. These early gene products are rarely found in the completed phage. Among the early proteins produced are a repair enzyme to repair the hole in the bacterial cell wall, a DNAase enzyme that degrades the host DNA into precursors of phage DNA, and a virus specific DNA polymerase that will copy and replicate phage DNA. During this period energy-generating and protein-synthesizing abilities of the bacterial cell are continue, but they have been subverted by the virus. The result is the synthesis of a number of copies of the phage DNA. Each of these copies can now be used for transcription and translation of a second set of proteins called the late proteins that make up the capsomeres and the various components of the tail assembly.

Lysozyme is also a late protein that will be packaged in the tail of the phage and be used to escape from the host cell during the last step of the replication process (Guttman et al,

2005,Birge, 1994)

Maturation or Morphogenesis. It is the period during which the new phage components are assembled into virons. Phage assembly is the last step before release; however, transcription, replication, and morphogenesis are often more or less occurring simultaneously, particularly in the small cubic and filamentous phages. Assembly can occur spontaneously or with the help of specific enzymes. The DNA is packaged into preassembled protein shells called procapsids. In most phages, their assembly involves complex interactions between specific scaffolding protein and the major head structural

21 proteins. Before or after the packaging, the head expand and becomes more stable, with increase of internal volume for the DNA. In tailed phages the head and tails are assembled by separate pathways and joined together after DNA encapsidation (Kutter et al, 2005, Ackermann and DuBow, 1987, Maloy et al, 1994).

Lysis or Release. Phages are liberated by lysis. Lysozyme lyses the cell wall, liberating infectious phages that are capable of infecting new susceptible host cells, and starting the cycle over again (Guttman et al, 2005, Ackermann, 2003). Phages are released into the surrounding medium by lysis of the host cell. The number of phages produced depends on the phage type and the physiology of the host cell. The tailed phages use two enzymes for the lysis of the host cell; lysin, an enzyme capable of degrading the cell wall peptidoglycan and holin, another enzyme that assembles pores in the inner membrane to let the lysin reach the peptidoglycan layer. These enzymes disrupt the cell membrane and cell wall, causing the cell to burst, and phages are released into the surrounding medium. The tailless phages encode a variety of single protein lysis- precipitating proteins that sabotage the host peptidoglycan-processing enzymes by different techniques (Guttman et al, 2005, Maloy et al, 1994, Gottesman and Oppenheim,

1994).

The latent period is the time interval from infection to the release of new phages and during the rise period phage titres increase to a maximum. The number of new phages per infected cell is called the "burst size" (Guttman et al, 2005). The life cycle of bacteriophages is illustrated in Figure 1.2.

22 Phage DNA O O Cell Division

induction of lytic cycle by excisionof phage chromosome from bacterial chromosome

Figure 1.2. Life cycle depicting the lytic and lysogenic pathways of a typical

bacteriophage when it infects bacterial cell.

23 1.4 BACTERIOPHAGE AS BIOCONTROL TOOLS FOR

FOODBORNE PATHOGENS

1.4.1 An overview and considerations for phage application in food

Phages can infect and multiply within their specific hosts. Host specificity is generally observed at strain level, species level, or, more rarely, at genus level. This specificity led to the idea of using phages for directed targeting of dangerous bacteria

(Rees and Dodd, 2006). Phages have been employed in human and veterinary medicine to control bacterial infections after Felix d'Herelle proved their effectiveness in 1919.

D'Herelle used phages to treat bacillary dysentery in the first attempt to use bacteriophages therapeutically. Later, he succeeded in using phages to reduce mortality due to cholera in India. The patients were given cholera-specific phage by mouth and by water after pouring phage stocks into the drinking water supplies. D'Herelle observed that the severity and duration of cholera symptoms in patients were reduced. D'Herelle and several members of des Enfants-Malades hospital in Paris ingested the phage preparation, in order to confirm its safety before administering it to the patients.

D'Herelle also injected his family as well as his colleagues to evaluate the safety of this treatment (D'Herelle, 1926).

However, this path was initially abandoned with the discovery of antibiotics and as a result of the conflicting results of phage treatments (Summers, 2001). In Eastern

Europe and the Soviet Union, however, research and application of phage therapy in human medicine continued. Phage therapy is currently used in this region to treat bacterial infections in humans, and is used as a complement to conventional antibiotics

24 (Kutter etal, 2010).

The recent growing concerns of antimicrobial resistance have allowed for research

in phage therapy to regain its vitality, especially in the food industry where phages have emerged as a novel new biotechnology to control bacterial contamination in food. This process has been termed biocontrol (Rees and Dodd, 2006, Hagens and Loessner, 2007).

In the early 80s of the last century, the results of one experiment in the western world directed the researchers to reexamine the therapeutical value of phages; The results were that a mixture of two phages protected calves and piglets against enteropathogenic strains of E. coli. The treated calves and piglets had much lower numbers of E. coli in their alimentary tract than untreated ones (Smith and Huggins, 1983). The study revealed that the use of phage cocktails against target bacterium may be a necessary prerequisite for a successful phage therapy and biocontrol (Strauch et al, 2007).

The life cycle of bacteriophage proves to be advantageous for food safety applications. Lytic phages have the ability to attach to bacteria and integrate into their metabolic system, while utilizing its host resources to reproduce. The release of new phage leads to lysis of the bacterial cell. Lysogenic phages have the ability to remain dormant within their host and transfer genes from one bacterium to another, potentially allowing for the development of more virulent and resistant pathogens (lysogenic conversions) (Hagens and Loessner, 2010, Greer, 2005, Sulakvelidze and Barrow, 2005).

A good example for pathogenicity associated with lysogenic conversion is Vibrio cholerae, where ctxA and ctxB genes on the integrative phage CTX O encode the cholera toxin, CTX (Waldor and Mekalanos, 1996). Shiga-like toxin (STX) producing E. coli is considered as another important foodborne pathogen known for a phage-dependent

25 virulence phenotype where in many cases, the implicated stxl and stx2 genes for the

toxin are encoded on temperate phages integrated into the host genomes (O'Brien et al,

1984). Therefore, the possibility for lysogenic conversion minimizes the usefulness of

temperate phages for biocontrol purposes. Moreover, temperate phages generally have

narrower host ranges than virulent ones (Hagens and Loessner, 2010). Thus, virulent

(strictly lytic) phages are the obvious choice for food safety applications (Greer, 2005).

Historically, most phage work in the biocontrol research area has been done in

liquids and usually with high concentration of pure target bacteria (Rees and Dodd,

2006). In liquid environments, thermal motion-driven particle diffusion and mixing due

to either fluid flow or active swimming (ex. bacterial motility) will increase the

likelihood of phages to hit and infect susceptible host bacteria (Hagens and Loessner,

2010, Murray and Jackson, 1992). When it comes to a food application, one might face

two major obstacles. First, a significant portion of targeted foods is solid rather than

liquid in nature. Second, bacterial contamination would likely occur at very low numbers

due to the expected high hygiene standards in place (Rees and Dodd, 2006). So it is

important to understand that a significantly high number of phages is required (threshold

of approximately 1X10 PFU/ml) to ensure sufficiently rapid contact and infection of the few targeted bacterial cells present. In other words, low numbers of phages are

unlikely to infect low numbers of bacteria simply because phages and bacteria are hardly

likely to meet together. The concentration of bacterial host is not a limiting factor if the

critical concentration threshold of phage numbers is reached and is able to cover the

entire available space of the targeted food matrix (Hagens and Loessner, 2010).

Experimental verification of this claim has been done recently when a Salmonella phage

26 (P7) was incubated with its respective host at 24°C for up to 2 h in LB broth at varying ratios of phage and host cell concentrations, and the surviving host cells were counted

(Bigwood et al, 2009). It was observed that inactivation of Salmonella by P7 seemed to be independent of the host concentration, with nearly complete inactivation occurring at a phage concentration of around 5 x 108 PFU/ml.

In other words, the requirement of a minimum bacterial density as a prerequisite for successful phage biocontrol is not universally accepted (Kasman et al, 2002). This was again supported by studies on the control of spoilage bacteria on meat surfaces, which suggest that phages can be effective biocontrol agents when the population of host cells is as low as 46 CFU/cm (Greer, 1988). Generally, it is recommended when investigating an antimicrobial agent that has the ability to kill the target organism as a post-lethality treatment, that is, following a bacterium-killing processing step such as cooking, to lower the amount of bacteria during artificial contamination (Scott et al,

2005). Another important factor that should be considered during experimental setup is the incubation temperature (Hagens and Loessner, 2010). Efficacy of the phages to control target pathogens should be tested both at higher than normal storage temperature, which provides good growth conditions for the undesired contaminants, and under recommended storage conditions.

In some reports of the use of phage for biocontrol of foodborne pathogens, the ratio of phages to host cells is described as multiplicity of infection 'MOI' (O'Flynn et al,

2004). However, the number of phages that infected bacterial cells has to be used in MOI calculations, not the number of adsorbed phages to the cell or, number of added phages to the food (Kasman et al, 2002). It was suggested that PFU/CFU ratio could be considered

27 as a more descriptive term in food application where there may be physical barriers preventing or slowing phage adsorption (Bigwood et al, 2009, Whichard et al, 2003). The exact phage concentration that needs to be used in a given application will depend on several factors: surface micro-structure which affects phage diffusion rates and accessibility of target bacteria; the amount of fluid that is available which affects phage diffusion; and the target reduction levels required (Hagens and Loessner, 2010). It could be suggested that, very high phage concentrations might be applied to eliminate foodborne pathogens without the necessity that the bacterium grows and replicates during the phage application (Strauch et al, 2007).

The inactivation results of foodborne pathogens in some reports using phage application may be due to lysis from without (Delbruck, 1940). Lysis from without occurs where host cells adsorbing numerous phage particles are inactivated rapidly in the absence of phage replication. In E. coli phage T4, lysis from without is mediated by a lysozyme on the base plate (Abedon, 1999). It occurs when more than 100 phages are adsorbed on the bacterial cell, then swelling and bulging of the membrane occurs within

5-10 min after infection. Finally, this is followed by the formation of holes through which cytoplasmic contents may escape (Tarahovsky et al, 1994).

Phages intended to be used for applied purposes have to meet some demands.

First, it should be determined if their DNA carry any genes coding for virulence factors like toxins or not, so complete genome sequences should be known (Rees and Dodd,

2006). A second phenomenon that should be kept in mind when selecting candidate phages is generalized transduction, which is a process where host DNA is packaged into phage heads, rather than phage DNA. This might lead to introduction of new genes into

28 the recipient bacterium (Ikeda and Tomizawa, 1965). Distribution of a virulence- associated genome region via transduced DNA has been reported for several pathogens

(Cheetham and Katz, 1995). So, only phages not able to transduce non-viral (i.e. bacterial) DNA should be used for biocontrol purposes.

In addition, selected phages should have a broad host range by infecting large number of the target species and/or genus (Hagens and Loessner, 2010). Narrow host range may present a problem for biocontrol purposes, as in some species there are numerous sub-types that all need to be controlled. Therefore, an effective phage should have a 'Goldilocks' host range, not too narrow and not too broad. Felix 01 is a perfect example; it lyses 96-99.5% of Salmonella serovars. However, narrow host range limitation would be overcome by using phage cocktails (Mclntyre et al, 2007). Stability at different storage and application conditions is another important aspect that should be defined (Strauch et al, 2007). Indeed, it is important to test phages for durability within the intended-use environment, which requires investing a great deal in further phage characterization (Gill and Hyman, 2010). From the economical point of view, abilities to be propagated in non-pathogenic hosts and large scale commercial production are typical criteria for phages that are considered for biocontrol of pathogens in food (Hagens and

Loessner, 2010). Treatment of food products will need large volumes of phage lysates containing high numbers of effective phages and these preparations should be purified to remove endotoxins and undesirable cell debris. This is in addition to showing no adverse effect upon oral feeding (Gill and Hyman, 2010).

As they are bacterial viruses, infection of mammalian cells is unlikely. All available evidence indicates that their oral consumption is entirely harmless to humans.

29 In fact, oral toxicity tests on rats given phages against Listeria monocytogenes at a dose of 2X10 PFU/kg body weight per day showed no signs of abnormality with regards to histological changes, morbidity, or mortality (Carlton et al, 2005). Similar results were found in a human study with E. coli T4 phages supplemented in drinking water (Bruttin and Briissow, 2005). Although the used phages were able to infect commensal E. coli strains in vitro, they seemed to have little effect on the E. coli occurring in the gut ecological systems of the human volunteers. It was suggested that the commensal E. coli population lives in niches not easily accessible to phages (Briissow and Kutter, 2005).

Individuals with HIV and other immunodeficiency diseases and healthy volunteers have also been intravenously injected with purified phages (e.g. OX 174) without any apparent side effects (Atterbury, 2009). Indeed, early phage therapy pioneers demonstrated safety by ingesting preparations themselves (D'Herelle, 1926). Moreover, thousands of people have received phage therapy in Eastern countries, especially the former Soviet Union and

Poland with great success in treating the causal agents (Kutter et al, 2010). The phages used not only were administered orally or superficially, but also were injected intramuscularly, intravenously, and even into the pericardium and carotid artery

(Ackermann and DuBow, 1987, Kutateladze and Adamia, 2008).

Phages are also natural components of the microflora and are found ubiquitously.

They are commonly isolated from soil, water, food, sewage, and from different environments containing their bacterial hosts (Briissow and Kutter, 2005). It was reported that freshwater environments contain up to 109 phages per milliliter, and up to 107 phage- like-particles per milliliter were found in marine surface systems. Similar numbers have been reported for terrestrial ecosystems such as topsoil (Rohwer and Edwards, 2002).

30 They are detectable from the farm to the retail outlet and are remarkably stable in all

environments (Greer, 2005). Phages have been isolated from a number of foods, like

lettuce, pork, oysters, mussels, mushrooms, turkey, chicken, cheese, yoghurt, buttermilk and beef and so are being ingested by everyone every day (Hudson et al, 2005). E. coli phages have been isolated from fresh chicken, pork, ground beef, mushrooms, lettuce, raw vegetables, chicken pie, and delicatessen food, with numbers up to 104 PFU/gram

(Allwood et al, 2004). Moreover, Campylobacter phages have been isolated from chicken at levels of 4 x 106 PFU/gm (Atterbury et al, 2003a), and Brochothrix thermosphacta phages have been reported in beef (Greer, 1983). In addition, fermented foods were found to have high numbers of those phages infecting the fermentation flora. For example, one

study described 26 different phages isolated from commercial cabbage (Sauerkraut) fermentation plants (Lu et al, 2003). Swiss Emmental cheese samples yielded phages active against Propionibacterium freudenreichii at levels ranging from 14 to 7 x 105

PFU/gm (Gautier et al, 1995). Sixty-one phages infecting Streptococcus thermophilus and Lactobacillus delbrueckii subsp. Bulgaricus have been isolated from Argentinean dairy plants samples at numbers of up to 109 PFU/ml (Suarez et al, 2002). Moreover, phages were seen as 'green' and environmentally friendly by many people (Fox, 2005).

Phages may also be considered as a natural alternative to chemical preservatives

(Mclntyre et al, 2007). Although phages are and will be present forever in human nutrition, consumer perception of adding viruses to foods will be the most critical hurdle which will need to be overcome in order to use phages on a broad basis for biocontrol of bacterial pathogens within food (Strauch et al, 2007).

As a biocontrol agent in food, phage application strategies for processing facilities

31 should be optimized to be the most convenient, most economical, and least invasive to the process itself. Hagens and Loessner discussed in detail the current different proposed methods of the industrial application of phages in food in their recent phage review

(Hagens and Loessner, 2010). Application can be done at different or even multiple points in,the food processing facility, where bacterial contamination is freshest, thereby enhancing the killing efficiency and so reducing the potential for bacterial evolution to phage resistance. Phage application could be useful at all stages of production in the classic 'farm to fork' approach throughout the entire food chain (Garcia et al, 2008).

Phages can be added by dipping, spraying or as a liquid to a large volume of food material. These methods may not serve as the ideal first choice as the means of phage application as they could be wasteful and potential decay of the phage particles could happen as a consequence of inclusion of other materials within the wash fluid such as bleach. Moreover, if the washing fluids themselves are suitable for bacterial growth, then the potential for bacterial evolution of phage resistance might exist. When phages are added directly to a batch of food, two major problems may be detected; dilution of phages and evolution of bacterial resistance. The dilution problem could be overcome by adding large numbers of phages or by applying phages before the mixing or disruption of food materials, such as spraying carcasses before processing. On the other hand the evolution concern can be addressed by regular disinfection of the equipment using highly efficient disinfectant (Hagens and Loessner, 2010).

There are many advantages of phages over traditional antimicrobials such as antibiotics and sanitizers. The most obvious ones are: phages target specific bacteria so there is no adverse effect on the natural microflora; no serious side effects on humans

32 have been detected; easy and cheap production; self replicating so there is no need to carry out repeat dosing. On the other hand, they have some drawbacks of being limited in host range, the risk for the development of resistant mutants and the potential for the transduction of virulent characters from one bacterial strain to another. In addition, the effectiveness of using of phage for bacterial control depends on the likelihood that phage and susceptible bacteria are in the same place. Another important drawback is that most research to date has involved in vivo experiments with artificially inoculated foods away from the real commercial environments (Hagens and Loessner, 2010, Rees and Dodd,

2006, Greer, 2005, Hanlon, 2007).

However, the advantages of phages for food applications outweigh their disadvantages, for instance, spontaneously occurring phage-resistant mutants are not likely to significantly influence treatment efficacy and the complex phage resistance mechanisms common in bacteria can be overcome by screening for broad host range phages and use of phage cocktails (Hagens and Loessner, 2010). In addition, increasing the concentration of the applied phages will increase the likelihood of meeting between phages and target bacteria (Greer, 2005, Garcia et al, 2008). Based on these scientific considerations and the recent multiple approvals of phage biocontrol products for use in foods, several companies throughout North America and Europe are conducting research on the use of bacteriophage as a biocontrol agent in food and starting to have commercial products in the market (Garcia et'al, 2008). For instance, LMP-102™ is a bacteriophage mixture produced by Intralytix Inc. in the U.S. that targets Listeria monocytogenes in ready-to-eat meat products and has no preservatives or allergens and most importantly, does not change the taste, color, odor or quality of the meat product. This product has

33 gained the approval of the FDA to be used as a safe food additive on ready-to-eat meat and poultry food products prior to packaging (www.intralytix.com), (www.fda.gov/).

Another phage preparation comprising a single lytic Listeria phage; LISTEXTM PI00

(www.ebifoodsafety.com), has even received the highly desirable GRAS (generally recognized as safe) status for its use in all food products by EPA (Enironmental

Protection Agency). E. coli and Salmonella phage preparations are also offered

(www.omnilytics.com); and were approved for being sprayed on cattle and chickens respectively, prior to slaughter of the animals to decrease pathogen transfer to meat

(Sulakvelidze and Barrow, 2005). Moreover, phage preparations active against

Pseudomonas putida that were developed for treatment of tomato and pepper against bacterial spot diseases (www.omnilytics.com), have been approved for use by the US

Environmental Protection Agency (EPA) (Balogh et al, 2010). The US EPA has also recently approved the use of an anti-£. coli 0157:H7 phage product to be sprayed or used as a wash on cattle hides prior to slaughter (Hagens and Loessner, 2010). The recent phage research development can lead to the appearance of more phage products in the near to mid-term future.

As Canada has a harmonized registration system with that of the United States, the approval of phage application in food by FDA would ultimately encourage Health

Canada and particularly the Pest Management Regulatory Agency (PMRA) to consider it as an approved biopesticide that can be used to enhance food safety in Canada. However, due to the co-evolution of phages and their bacterial hosts, the registration process needs to allow for the acceptance of new phage on a continuous basis so that they could be approved without going through the whole registration procedure again.

34 Companies who seek permission for commercial use of phages as a biocontrol

means should consider some safety issues. For instance, phage preparations have to be

tested for the absence of the pathogen, toxins and/or virulence factors after purification.

Also, a monitoring system for development of phage resistant bacterial cells would be

sensible to ensure the effectiveness of the preparation. Another important point, genome

analysis and bioinformatic studies should be done to ensure the absence of virulence

genes and any genes that might lead to mutation of the virulent phages to be temperate

variants and lysogenize a pathogen by horizontal gene transfer. Moreover, presence of the phage on the surface of a food might change its microflora, so changes in the microflora

composition of a food should be intensively analyzed (Strauch et al, 2007).

It is important to mention that, it is never envisaged that phage would totally replace standard preservative and cleaning agents. Phage application should be

considered as one approach in the hurdle technology in combination with different

existing methods (Garcia et al, 2008, Leverentz et al, 2003, Martinez et al, 2008, Roy et al, 1993).

1.4.2 Examples for the experimental proof of concept for phage application

as a biocontrol agent

The application of phages to reduce pathogenic bacteria during the preharvest and postharvest stages of food production has shown promise for safer food production

(Strauch et al, 2007). Lytic phages have been used successfully to specifically eradicate

zoonotic pathogens from living animals, decontaminate carcass meat or disinfect the

surfaces of ready-to-eat products (Hagens and Loessner, 2010, Rees and Dodd, 2006,

Strauch et al, 2007). In spite of the reduced activity in the real food system compared to

35 liquid media, phages have been used in a somewhat successful way in controlling some foodborne pathogens such as E. coli 0157:H7, Salmonella and Listeria monocytogenes

(Hagens and Loessner, 2007, O'Flynn et al, 2004, Greer, 2005, Carlton et al, 2005).

Moreover, the recent FDA approval of phage preparations as food additives for preservation has also triggered the search for new applications for these natural bacterial killers (FDA, 2006).

Several studies have described the biocontrol of Listeria in foods using Listeria phages. The effectiveness of a range of different phages to remove Listeria from stainless steel and polypropylene surfaces was investigated (Roy et al, 1993). It was found that phage treatment alone was able to achieve approximately a 3 log units drop in cell number. In the same study, the phages used were also evaluated for their ability to tolerate inactivation by a quaternary ammonium compound (QUATAL) used for cleaning, and it was found that they were not inactivated by concentrations up to 50 ppm.

A combination of phage and 40 ppm QUATAL resulted in a 5 log unit reduction in levels of surface attached Listeria. Hibma et al. succeeded to isolate a phage that was specific for L-forms of Listeria where cell wall structure is either deficient or absent and used this phage to control biofilm formation by this bacterium (Hibma et al, 1997). The phage was as successful as lactic acid (130 ppm) at inactivating pre-formed L-form biofilms on stainless steel, both reduced viable cell numbers by 3-log units over a 6 h period.

Listeria phage cocktails alone and in combination with nisin were tested on honeydew melon and apple slices (Leverentz et al, 2003, Leverentz et al, 2004). A reduced activity has been noticed in the apple slices, which might be due to phage instability. In honeydew melon, a reduction of 2 to 4.6 log units of the bacteria was

36 observed using phage alone and a cumulative effect was detected when nisin was added.

Another study describes the use of the broad host range lytic Listeria phage, PI00 to control Listeria monocyotgenes in soft cheese (Carlton et al, 2005). Bioinformatic analysis of the total genome sequence of PI00 did not reveal any similarity to any genes or proteins believed to play a role in the pathogenicity or virulence of L. monocytogenes.

The efficacy of the phage in controlling artificial contamination during manufacture was evaluated, showing that depending on dose and treatment regimen the contamination could be reduced below levels of detection for the entire ripening time. No cells were able to re-grow at higher doses indicating that the killing effect was complete only at the highest used doses. In another recent study, PI00 and A511 Listeria phages were used to control L. monocytogenes in different liquid and solid ready-to-eat foods (Guenther et al,

2009). It was reported that in liquid foods, such as chocolate milk and mozzarella cheese brine, bacterial counts rapidly dropped below the level of direct detection after storing at

6°C for 6 days. While on solid foods (hot dogs, sliced turkey meat, smoked salmon, seafood, sliced cabbage, and lettuce leaves), phages could reduce bacterial counts by up to 5 log units under the same conditions. In addition, application of higher doses of

o phages (3x10° PFU/gm) was more effective than lower doses. Another study which used artificially contaminated salmon as a model to test the ability of Listeria phages to control

Listeria monocyotgenes, recorded a 3 log unit reduction in the target pathogen when a high titre of phages was applied (Hagens and Loessner, 2007). In another study, Listeria phages were used in combination with a proactive culture {Lactobacillus sakei TH1) to reduce Listeria monocytogenes on sliced cooked ham (Hoick and Berg, 2009). Phages alone caused a ten-fold reduction of L. monocytogenes in the tested samples, while using

37 phages and proactive culture resulted in 100-fold reduction after 14- 28 days of storage.

In a recent report, the influences of Listeria phage PI00 dose, contact time, and storage temperature on the listericidal activity of the phage to reduce the L. monocytogenes loads on the surface of fresh channel catfish fillet have been studied (Soni et al, 2010). It was found that the phage contact time of 30 min reduced L. monocytogenes more than 1 log

CFU/gm, whereas 15 min contact time with phage yielded less than 1 log CFU/gm reduction in L. monocytogenes loads on catfish fillet treated samples. The overall reduction was detected in the treated samples over a 10-day shelf life at 4°C or 10°C. On the other hand, one study reported that addition of Listeria phage LH7 had no significant effect on the presence of two strains of Listeria monocytogenes on artificially inoculated beef stored for 4 weeks at 4°C (Dykes and Moorhead, 2002). After analyzing the data, it was suggested that the reason for these negative results may be the usage of insufficient phage concentration per sample (it was around 0.5 - 1 phage per square centimeter)

(Hagens and Loessner, 2010).

Similar promising results have been obtained when phages were used to control growth of Salmonella in food (Rees and Dodd, 2006). Cheddar cheese made from raw and pasteurized milk was artificially inoculated with Salmonella Enteritidis at a level of

104 CFU/ml of milk during the manufacture and Salmonella phage SJ2 was added at a

fixed ratio and samples were taken over a period of 99 days at 8°C (Modi et al, 2001).

Salmonella did not survive in the pasteurized cheeses after 89 days, whereas they were present at around 50 CFU/ml in raw milk cheeses. However, in contrast, all the untreated cheeses showed levels of 103 CFU/gm after 99 days. In another trial, a significant reduction was reported in the number of Salmonella Enteritidis PT4 recovered from

38 artificially contaminated chicken skin samples after immersion in a suspension containing a cocktail of three lytic Salmonella phages and stored for 9 days at 5°C (Fiorentin et al,

2005). In a similar experiment, artificially contaminated chicken carcasses with

Salmonella were used as a model for surface disinfection using phages (Atterbury, 2009).

It was found that a cocktail of phages could reduce Salmonella recovery by more than

1000-fold compared with untreated controls. In another study, biocontrol of Salmonella

Enteritidis on artificially inoculated melon and apple slices stored at different storage temperatures using a cocktail of four Salmonella lytic phages has been reported

(Leverentz et al, 2001). The phage cocktail reduced Salmonella populations by approximately 3.5 log units on honey dew melon slices stored at 5 and 10°C and by approximately 2.5 log units on slices stored at 20°C. Whereas it could not reduce

Salmonella contamination in the apple slices, which might be due to the instability of phages at low pH of apple slices.

Broad host range FelixOl Salmonella phage and a related phage variant were used to control Salmonella Typhimurium DTI04 on chicken frankfurters (Whichard et al,

2003). It was revealed that under certain environmental conditions a reduction of

Salmonella cells by 1.8 and 2.1 log units could be achieved. Application of high ratios of phages to host cells (PFU/CFU of 100 to 1,000) reduced Salmonella Enteritidis cells recovered from chicken skin after 48 hours by 2 logs. Using a higher ratio of phage (10 ) also eliminated other Salmonella strains possibly by a simple lysis mechanism (lysis from without) (Goode et al, 2003). Phages infecting Salmonella Typhimurium PT160 were also added at different concentrations to various densities of host bacteria inoculated onto raw and cooked beef. Significant host inactivation of the order of 2-3 log units at 5°C

39 and >5.9 log units at 24°C was achieved (Bigwood et al, 2008). On the other hand, one recent study evaluated the ability of two broad host range Salmonella bacteriophages

(SSP5 and SSP6) to control Salmonella Oranienburg in vitro and on experimentally contaminated alfalfa seeds showed in incomplete lysis during in vitro treatment and moreover, no significant reduction in the viable Salmonella population in treated seed

(Kocharunchitt et al, 2009). Use of low MOI (-70) might be the reason for these negative results.

Studies on the biocontrol of pathogenic E. coli in different foods have shown encouraging results. O'Flynn et al. reported that treatment with phage cocktail of three E. coli 0157:H7- specific phages eliminated E. coli 0157:H7 from seven of nine artificially contaminated beef surfaces examined after incubation at 37°C for one hour, and the two remaining samples had a very low bacterial count. In the meanwhile, 5 logs reduction in the cell number was noticed when the experiment was done in the broth culture (O'Flynn et al, 2004). This might illustrate that effect of food matrices on the efficiency of phage biocontrol applications and that culture broth studies are of limited use for phage applications in food matrices (Rees and Dodd, 2006). In another recent study, bacteriophage cocktail, (ECP-100) containing three Myoviridae phages lytic for

Escherichia coli 0157:H7 was evaluated for its ability to reduce three virulent strains of the host bacterium on artificially inoculated hard surfaces (glass coverslips and gypsum boards), tomato, spinach, broccoli, and ground beef (Abuladze et al, 2008). ECP-100 significantly reduced the number of viable E. coli 0157:H7 organisms on the tested hard surfaces and four food samples. The observed reductions of the target bacteria from hard surfaces ranged from 85% to 100% while that of food samples were from 94% to 100%

40 based on the concentration of phage cocktail added. The same phage cocktail (EPC-100) was used again to control E. coli 0157:H7 on contaminated fresh-cut iceberg lettuce and cantaloupe. Significant reductions of the target cells were observed on both treated food products after incubation at 4°C for 2 and 7 days, respectively (Sharma et al, 2009).

Most of the reported work with Shigella phages was directed to evaluate their use as therapeutic agents (Sulakvelidze et al, 2001, Kutter et al, 2010, Kutateladze and

Adamia, 2008). Results of using Shigella phages to control Shigella dysentery in rabbits and humans with bacillary dysentery encouraged D'Herelle to focus on using phages as therapeutic and biocontrol agents (Merril et al, 2003). Another earlier experiment showed interesting findings when mice infected intracerebrally with Shigella dysenteriae were rescued by administering phage into the peritoneal cavity (Dubos et al, 1943). The survival of untreated animals was 3.6%, whereas the survival of phage-treated animals was 72%. Shigella phages were successfully used for prophylaxis of bacterial dysentery

(Sulakvelidze et al, 2001). In one study, it was found that the combination of phages and antibiotics was effective in treating bacterial dysentery cases, whereas using antibiotics alone was ineffective (Miliutina and Vorotyntseva, 1993). Recently, Shigella phages active against Shigella flexneri and Shigella sonnei have been formulated in the tablet form and now commercially available for treatment and prophylaxis of dysentery in all age groups and high-risk groups (http://www.biochimpharm.ge/). Shigella is commonly spread through fecal-oral transmission and this spread may be through intermediary contaminated food product (Acheson, 2001, Sivapalasingam et al, 2006). Therefore, use of phages on these food materials or to clean the contaminated surfaces in the food processing plants, can control spread of this pathogen.

41 1.5 IMMOBILIZATION OF PHAGES ON SOLID SURFACES: A

NEW PROPOSED TECHNIQUE FOR FOOD APPLICATION

Immobilized biologically active materials are of great importance to industry and research. The selection of the immobilization method and support depend on the nature of the bioactive material and on the application itself. Microorganisms have been immobilized for applications in different areas, for instance in food biotechnology (e.g.

Acetobacter in vinegar production) and in waste water treatment (trickling filter). Several techniques have been proposed for immobilization of microorganisms such as, physical adsorption, gel entrapment, covalent binding to support matrix (Knaebel et al, 1997,

Selvaraj et al, 1997, Jirku, 1999). Hydrogels such as calcium-alginate, potassium pectate and gelatin are frequently used to immobilize cells through entrapment (Pan et al, 2010).

Cost and instability of hydrogels may limit their use on an industrial scale. Surface immobilization of microorganisms by adsorption has the advantage of being very simple to carry out and there is no transfer barrier as in the entrapment method (Klein and Ziehr,

1990). On the other hand, the relative interaction between the support and the cells is low resulting in the release of the immobilized material to the surrounding medium. Covalent binding between a functional group of the cells, generally amino acids, and support matrix has been reported as another method of immobilization (Jirku, 1999). This provided a strong binding and overcame the diffusion problem of simple adsorption technique, but some chemicals which might be used to create covalent cross-linking, such as glutaraldehyde, could result in the loss of the activity and viability of the cells.

The bacteriocin, pediocin (ALTA 2351) was incorporated in cellulose matrix

films, and these bioactive films were evaluated to control growth of Listeria innocua and

42 Salmonella spp. and extend shelf life of artificially contaminated ready-to-eat ham slices

(Santiago-Silva et al, 2009). Reduction of two log CFU/gm of L. innocua after storage

for 15 days at 12°C has been reported. While only 0.5 log CFU/gm reduction of

Salmonella occurred after 12 days of storage at the same temperature. Another

bacteriocin, nisin, was immobilized at different concentrations into palmitoylated

alginate-based films or in activated alginate beads and the antimicrobial efficiency has

been examined using artificially contaminated beef muscle slices and ground beef with

Staphylococcus aureus and stored at 4°C for 7 and 14 days respectively (Millette et al,

2007). It was found that sliced beef covered with film containing 500 or 1000 IU/mL of

nisin, showed S. aureus reduction of 0.91 and 1.86 log CFU/cm2, respectively. While

alginate beads containing the same amounts of nisin caused 1.77 and 1.93 log CFU/gm

reduction of S. aureus counts in ground beef. In an earlier study, nisin has been also

incorporated into a polyethylene-based plastic film and tested to inhibit surface growth of

bacteria on meat (Siragusa et al, 1999) . Treated beef carcass samples with Brochothrix

thermosphacta wrapped with this nisin impregnated plastic films, showed lower counts of

this spoilage bacterium after storage at 4°C and 12°C for 20 days when compared to

control vacuum-packaged samples.

Properties of phages in specifically interacting and lysing their host bacteria make

them another ideal bioactive material that can be used in the immobilized form to

increase their application in research and industrial fields. Encapsulation can be

considered as one of the interesting techniques for phage immobilization that can be used

to broaden phage application areas. In a recent US patent, phages have been absorbed

onto a solid matrix (such as skim milk powder, soya protein powder and whey protein

43 powder) and then dried by heating under vacuum (Murthy, 2008). The adsorbed phages were embedded in a solid support such as microbeads, cellulose-based material, and carbohydrate-based material. It was suggested that these immobilized phages may be encapsulated and incorporated into a capsule or tablets to be protected from the physico- chemical stresses of its environment. The release rate was related to the material used for encapsulation. It was claimed that this innovation can be used for phage therapy applications in human, veterinary and aquaculture in addition to agricultural applications.

For instance; it can be used as food additive for fish, livestock, birds and poultry to aid in reducing the shedding of target pathogens and could be used as an oral, topical or nasal medication for humans.

The electrospinning process for the encapsulation and immobilization of T7, T4, X phages in electrospun polymer nanofibres, was demonstrated as a potential technique for phage immobilization (Salalha et al, 2006). The encapsulated phages managed to survive the electrospinning process while maintaining their infectivity. These immobilized phages were able to infect their target bacterial host after dissolving the polymer fibres and releasing them from the nanofibers. The potential of nanoencapsulating of a broad lytic phage (O-PVP-SEl) in water-in-oil-in-water (W/O/W) multiple emulsion has been also investigated (Costa et al, 2009). In a similar approach, phages against

Staphylococcus aureus or Pseudomonas aeruginosa have been encapsulated into biodegradable polyester microspheres with only a partial loss of lytic activity after being frozen in liquid nitrogen and lyophilized for over 72 hours (Puapermpoonsiri et al, 2009).

The produced microspheres were designed to have an appropriate size and density to facilitate inhalation via a dry-powder inhaler and thus, can be used in the controlled

44 delivery of phages for the treatment of bacterial lung infections. In all the above mentioned techniques, phages are enclosed in protective material and they need to be released to be able to infect target pathogens, so these techniques could be used in phage therapy applications where phages need to overcome some environmental barriers before reaching sites of the target pathogen.

Various studies reported other techniques for phage immobilization and the potential use of the developed phage-based biosorbent as biosensors to detect, concentrate and identify target bacteria. Salmonella have been captured from mixed cultures using Salmonella-specific phage passively immobilized on polystyrene supports

(Bennett, 1997). Different polystyrene surfaces, microplates and dipsticks were dipped

into a 5 x 1010 PFU/mL phage solution; this was followed by washing in order to remove

unbound phage, and blocking the remaining adsorption sites. These biosorbents were then incubated with mixed bacterial cultures containing Salmonella cells, and their

capture ability of Salmonella cells was assessed either by PCR or by epifluorescence

microscopy. In spite of the capability of this protocol to separate Salmonella from mixed

culture, high detection limits (107 CFU/ml) and low capture efficiency (1%) were

reported. The wrong orientation of the phage was suggested to be the reason for observed

results. The tail must be free to specifically bind to the host's receptor. Chemical

biotinylation of the phage head was another approach to immobilize phages where they

were immobilized on the surface of streptavidin-coated magnetic beads (Sun, 2001). The

developed biosorbent was used to capture target cells of recombinant bioluminescent

Salmonella Enteritidis by magnetic separation. Although approximately 20% of the target

cells could be recovered when 2 X 106 CFU/ml was used, this system could not capture

45 low numbers of cells in the food sample. The orientation and/or the inactivation of the virus due to chemical reaction may have played a role in the low capture efficiency.

Phages have been immobilized and employed as recognition receptors in a number of biosensors. Physical adsorption has been used to immobilize IG40 filamentous phages on gold surfaces of quartz crystal microbalance (QCM) biosensor and used as recognition elements to detect P-galactosidase from E. coli (Nanduri et al, 2007a).

Filamentous phage Lm P4:A8, expressing the scFv antibody to virulence factor actin polymerization protein (ActA) on its surface, was immobilized to the surface plasmon resonance (SPR) sensor surface through physical adsorption and used for detection of

Listeria monocytogenes (Nanduri et al, 2007b). However, a high detection limit was reported using this approach. Another filamentous phage specific to Salmonella

Typhimurium has been physically adsorbed to the surface of magnetoelastic sensor and used as a biorecognition element for detection of its host cells at different concentrations

(Lakshmanan et al, 2007). Staphylococcus lytic phages were immobilized on the gold surface of surface plasmon resonance (SPR) sensor via direct physical adsorption

(Balasubramanian et al, 2007). This system was able to detect 104 CFU/ml of

Staphylococcus aureus.

On the other hand, Podoviridae Salmonella phage P22 was covalently immobilized to glass substrates, forming a monolayer that was able to detect Salmonella

Typhimurium with the aid of enzyme-linked immunosorbent assay and atomic force microscopy (Handa et al, 2008). Moreover, wild type T4 tailed phage has been immobilized on modified gold surfaces of SPR sensor through hydrogen bonding (Singh et al, 2009). The gold surfaces were activated with glutaraldehyde after modification by

46 hydrophilic interaction with sugars (dextrose and sucrose) and through thiol linkage with the amino acids histidine and cysteine. The attachment density of phages was 37-fold higher when compared with simple physisorbtion and as a result, the capture efficiency of this system for E. coli host cells improved by 9-fold compared with physically adsorbed phages. In addition, electrochemically modified screen printed electrode (SPE) microarrays have been developed to covalently immobilize T4 phages and use these phages as recognition receptors for the detection of E. coli through impedance measurements (Shabani et al, 2008). These novel techniques provided specific and direct detection of bacteria using phage as a probe and it can be applied to detect other bacteria by using appropriate phages (Zourob and Ripp, 2010).

Site-specific immobilization of phages can be suggested to overcome the

orientation problem of immobilized phages and as a consequence increase the capture

efficiency to the target bacteria. Phage display technology, in which foreign gene

fragments encoding a polypeptide is inserted into the phage genome through fusion with one of the coat protein genes in order to be expressed on the phage's surface, facilitates the production of recombinant phages that are able to display different proteins on their heads (Paschke, 2006). This technique has been applied in order to create oriented

immobilized T4 phages on cellulose membranes and streptavidin-coated magnetic beads

(Tolba et al, 2010). Biotin carboxyl carrier protein gene (BCCP) or the cellulose binding

module gene (CBM) has been fused with the T4 small outer capsid protein gene (SOC)

and as a result, the developed recombinant phages were able to express the respective

ligand on the phage head. This allowed T4 phage to be immobilized on the corresponding

solid substrate through its head, leaving the tail free for interaction with the bacterial host.

47 Both developed recombinant phages were able to specifically capture and infect the host bacterium. Around eight hundred E. coli cells have been detected within 2 hours by using

BCCP-T4 phages and real-time PCR. In addition, the capture efficiency of these genetically biotinylated T4 bacteriophages onto streptavidin-coated gold surfaces was found to be improved 15 times when compared to the simple physisorption of the wild- type phage (Gervais et al, 2007). Monitoring the cell growth by measuring impedance showed that immobilized phages lyse their target cells and could inhibit or slow cell growth.

Although, the genetic modification of phage gave well-oriented phage particles with high capture efficiency, this protocol is initially laborious, time consuming and costly even with T4 phage which is one of the best-known phages with a fully sequenced genome (Zourob and Ripp, 2010). In addition, it was reported that the developed recombinant T4 phages have lower burst size and longer latent periods when compared with the wild type T4 (Tolba et al, 2010). The ability to detect E. coli B using wild type

T4 phage and recombinant BCCP-T4 and CBM-T4 that were immobilized on nano- aluminum fiber-based filter (Disruptor™), streptavidin-coated magnetic beads and microcrystalline cellulose, respectively has been investigated (Minikh et al, 2010). It was found that immobilized recombinant phages lost their infectivity as compared with free phage and could not detect the target bacterium. Therefore, more research is required to establish alternative protocols to site specifically immobilize phages for which the structure is well known as well as non-sequenced and newly isolated phages.

It was reported that electrostatic interactions between the charged proteins and the charged membranes might play a role in the separation of proteins through ultrafiltration

48 membranes (Bhushan and Etzel, 2007). Positively-charged ultrafiltration membranes substantially improved the separation of proteins that have a small difference in molecular weight (such as pMactoglobulin and glycomacropeptide). The application of charged membranes may also be used to improve high-performance tangential flow filtration (HPTFF) for protein purification (Van Reis et al, 1999).

The charge difference approach might prove a simple approach to enhance oriented immobilization of different phages without going through the genetic modification protocol. Even for the deeply studied T4 phage and as far as we know, the isoelectric points (pKa) of capsid and tail fibers as single units have not been determined and as a result there is no available report for the overall charge at a given pH for these components. However it was reported that the net charge on most viruses is negative and the whole T4 phage (capside, tail and fibers) has an isoelectric point close to 4 (Archer and Liu, 2009). The pKa values for major T4 head proteins were found to have a range between 4.62 and 6.63, while a range between 5.21 and 9.76 was reported for tail fiber proteins (Cummings et al, 1970, Showe and Onorato, 1978, Karam et al, 1994). Based on these values, it was suggested that capsids acquire a negative overall charge above pH 4

(Archer and Liu, 2009). In an early study, T7 phage head was suggested to be primarily responsible for the overall negative charge of the phage and the tail fibers could be positively charged (Serwer and Hayes, 1982). In a recent study, it was found that T4 heads were aggregated at pH 5.6 and 7.5 on aminosilanized substrate, where capsids behave as negtively charged entities and electrostatically attracted to a positively charged surface (Archer and Liu, 2009). Lower aggregation of T4 phage heads happened with slightly positive and negative surfaces due to weaker electrostatic interaction and

49 repulsion forces that occurred respectively. This lower aggregation suggested that adsorption of T4 capsids might not be purely electrostatic, but rather a combination of various types of interactions. It was also reported that presence of divalent counterions could facilitate the interaction and overcome electrostatic repulsion (Archer and Liu,

2009, Pastre et al, 2003).

Cellulose is the most abundant natural polymer on earth and can be considered as an attractive matrix for immobilization mainly because of its combination of excellent physical properties and low price. Moreover, cellulose can be easily manufactured according to the application purpose, for instance, it has been modified to carry positive and negative charges by introducing positive diethylaminoethyl (DEAE) or negative carboxymethyl (CM) into cellulose matrix and applied to adsorb different proteins

(Mahiout et al, 1997). Cellulose is commercially available in many different forms, such as cotton wool, filters, beads, powders, fibers, hydro gel, membranes and sheets of defined porosity and as a consequence, it can be involved in various applications and products (Kim et al, 2006). Cellulose was applied in several medical applications such as covers in experimental micro-nerve surgery and as artificial blood vessel interpositions with inner diameter of about 1 mm (Klemm et al, 2001). Nitrogen has been continuously removed from seawater using immobilized nitrifier and denitrifier bacteria on macro- porous cellulose carrier (Sakairi et al, 1996). Organic carbon, ammonium ion, and phosphate ion have been removed from water by immobilized Rhodobacter capsulatus on cellulose beads (Sawayama et al, 1998). Moreover, glucoamylase enzyme was immobilized on bacterial cellulose beads and was applied successfully in manufacture of crystalline glucose or glucose syrup (Wu and Lia, 2008). In the food industry, some

50 classes of antimicrobial compounds were immobilized on cellulose materials to develop antimicrobial packaging. Bacteriocins fixed on cellulose casing inhibit the growth of

Listeria monocytogenes on ham, turkey breast meat and beef (Ming et al, 1997, Zhu et al,

2005). In addition, antimycotic agents have been incorporated into edible cellulose film to increase shelf life of produce and cheese items (Cutter, 2002).

51 1.6 RESEARCH OBJECTIVES

Based on the above description, the purpose of this research was to develop a simple technique to immobilize isolated lytic phages and explore the potential use of these immobilized phages to control some foodborne pathogens in food. Using pathogenic E. coli strains, Salmonella spp., Listeria monocytogenes and Shigella spp. as model foodborne bacteria, this doctoral project had the following objectives:

1. Isolation and characterization of effective and stable lytic phages against the

targeted foodborne enteric pathogens from different environmental samples.

2. Investigate the possibility of using the isolated phages as biocontrol agents for

targeted pathogens in liquid media and in food system.

3. Study the charge difference between phage head and tail fibers and use this

feature for oriented immobilization of phages on cellulose-based materials

using physical adsorption.

4. Examine the performance of the immobilized phages to control the growth of

their bacterial host in laboratory media and in ready-to-eat foods.

In the next chapters, the efforts undertaken in the present thesis towards the development of a novel technique for immobilization of phages in order to control foodborne pathogens are presented. Chapter 2 shows the isolation and characterization of lytic phages against targeted bacteria. The most potent and stable phages with broad host range were selected and characterized. One of those isolated was found to be unique and, to the best of our knowledge, the first time to be isolated, so sequencing and bioinformatics studies have been performed on this phage. Virulence genes were not

52 detected in the genome of this phage and hence it can be applied safely in food. This will be discussed in Chapter 3. Chapter 4 presents the evaluation of using cocktail of these phages to control their host bacteria in broth media and in a real food system. Different storage and packaging conditions have been applied to optimize the best conditions for application. In chapter 5, a novel technique for oriented immobilization of phages was established. Investigation of the charge difference between T4 heads and tail fibers and immobilization of the isolated phages on cellulose membranes based on electrostatic interaction are presented in this chapter. Charged silica, cellulose membranes and gold nanoparticles were used to investigate the role of charge differences in enhancing the number of the immobilized phages in the right orientation. Testing infectivity of the immobilized phages was conducted to ensure the right phage orientation by using different techniques. This chapter also shows the application of the developed immobilized biocontrol system to control growth of E. coli 0157:H7 and Listeria monocytogenes in real food matrices under different storage and packaging conditions.

53 \ Chapter 2: ISOLATION AND CHARCTERIZATION OF LYTIC BACTERIOPHAGES AGAINST IMPORTANT FOODBORNE PATHOGENS

2.1 ABSTRACT

In this study, approximately one hundred bacteriophages (phages) were isolated from different environmental sources against various strains of four major foodborne pathogens including pathogenic E. coli, Salmonella, Listeria and Shigella species. The

Bioscreen C was used for the first time as a high throughput turbidimetric assay to monitor the lytic activity and host ranges of the isolated phages and to establish the similarity of the isolated phages. The Bioscreen C results demonstrated that many of the isolated phages had a broad host range while others were very limited in the number of strains that they could infect. Some of the isolates had identical host range profile which was confirmed by their restriction endonuclease patterns and TEM images. Three isolates with the broadest host range against each genus were selected for further characterization and application. These phages exhibited different morphologies: with one being identified as a member of the family Siphoviridae and eleven were Myoviridae phages.

However, they gave distinct restriction endonuclease patterns and were varied in their

stability at different tested pH and temperatures. Most of them were infective under different environmental conditions and showed no or low frequency of resistant mutants.

Four ambivalent phages were detected among the selected twelve phages. Therefore, these isolated phages can be considered as useful candidates for biocontrol of the targeted pathogen.

54 2.2 INTRODUCTION

Despite application of a variety of preventive hygiene and sanitation standards during food processing, there are still many reported outbreaks of foodborne illness

(Hagens and Loessner, 2007, Kothary and Babu, 2001). These outbreaks have significant health and economic impact worldwide (Mead, 1999).

Pathogenic Escherichia coli, Salmonella, Shigella and Listeria monocytogenes are common etiological agents of bacterial foodborne illness. E. coli 0157:H7 is one of the enterohemorrhagic (EHEC) strains that have the highest morbidity and mortality rates amongst all pathogenic strains of this bacterium. It has been documented as the cause of many sporadic and outbreak-associated hemorrhagic colitis cases for which very low infectious doses as few as 10 cells, are required (Reiss et al, 2006). Salmonella spp. result in salmonellosis that has symptoms ranging from mild to severe gastroenteritis to enteric fever and in more severe cases results in bacteraemia or septicaemia (D'Aoust, 1994).

Shigella causes shigellosis that develops within 12-48 hours after the ingestion of

Shigella contaminated food. The symptoms range from fever, aches, fatigue and watery diarrhea to bloody stools or dysentery and in severe cases, fatal haemolytic-uremic syndrome (HUS) may develop due to the production of Shiga toxin (Acheson, 2001).

Ingestion of food contaminated with Listeria monocytogenes causes another important foodborne illness; listeriosis which presents as flu-like symptoms that may develop to meningitis, bacteraemia and abortion in high-risk individuals (Swaminathan and

Gerner-Smidt, 2007).

The growing concerns surrounding the development of antimicrobial resistance against these pathogens have generated renewed interest in novel alternatives such as

55 phage therapy. In the food industry phages have emerged as a novel biotechnology tool to control bacterial contamination in food in a process termed "biocontrol" (Hagens and

Loessner, 2010). Strongly lytic phages are the obvious choice for food safety applications as they are able to lyse specific bacterial pathogen; resulting in the generation of more phage particles that can infect more bacterial cells (Greer, 2005). Pathogen-specific phages are abundant in environmental sources such as sewage, soil, as well as in feed and food (Brussow and Kutter, 2005). Phages have been used to control foodborne pathogens such as E. coli 0157:H7, Enterbacter (now known as Cronobacter) sakazakii,

Salmonella, Listeria and Campylobacter, and for decontamination of carcasses and surfaces of ready-to-eat products (Hagens and Loessner, 2010, Greer, 2005, Sulakvelidze and Barrow, 2005).

Phage identification, differentiation and characterization by host range testing, one-step growth curve and lysis experiments, transmission electron microscopy (TEM) and random fragment length polymorphism (RFLP) are important steps that are required before they can be used to control foodborne pathogens (Hagens and Loessner, 2010).

Although the isolation step is critical, it is time-consuming and laborious to characterize the host range of the isolated plaques and determine the extent of duplication. The

Bioscreen C (optical density monitoring device) was used to determine the effect and activity of various antimicrobial compounds by measuring the turbidity of bacterial suspensions (Pietiainen et al, 2009, Si et al, 2006). Thus, this method may also be applicable to monitor the lytic activity of phage. Therefore, the objective of this study was to isolate stable lytic phages, active against selected pathogens, and use a turbidimetric method, based on the Bioscreen C, as a rapid and relatively simple

56 alternative to traditional lysis and host range testing protocols in order to differentiate between the isolated phages. In addition, this method was used to select candidates with the broadest host range for future application. The selected phages were then subjected to further microbiology and molecular characterization in order to validate their usefulness as biocontrol agents.

2.3 MATERIALS AND METHODS

2.3.1 Bacteria and Bacteriophage

Tryptic Soy Broth (TSB), Tryptic Soy Agar (TSA), and Tryptose Soft Agar (TSB

+ 0.4% agarose) (Difco Laboratories, Detroit, MI) were used in this study to grow the host bacteria and to isolate and propagate the phages. In the case of Listeria phages, after sterilization, all the media were supplemented with filter-sterilized 1.25mM CaC^ per liter of the medium (Fisher Chemicals, Mississauga, ON, Canada). Strains of E. coli,

Salmonella, Shigella and Listeria spp. were selected from the Canadian Research

Institute for Food Safety (CRIFS) Culture Collection at the University of Guelph (Table

2.1) and were used for phage isolation and propagation. Pure cultures were obtained from

-80°C frozen stocks and maintained at 4°C on TSB until use. Cultures were re-streaked every month to maintain cell viability.

2.3.2 Enrichment and isolation of phages

In order to isolate the phages, different samples (rinse water, sewage and fecal samples) collected from local meat, poultry and wastewater treatment plants and the

Arkell Poultry Research Farm, University of Guelph (Guelph, ON, Canada) were tested

57 for the presence of phages.

Each sample was enriched in an equal volume of TSB medium and 100 ul of overnight culture of a mixture of selected strains of the targeted genus (Table 2.1); for fecal samples 10 grams were enriched in 20 ml of medium. The mixture was incubated for 16-20 h with gentle shaking at 30°C. After incubation, the suspensions were then centrifuged at 4000 x g for 15 min at 4°C (Beckman Avanti J-20 XPI, Beckman Coulter

Inc., Mississauga, ON, Canada) and the supernatant was carefully transferred to another tube and filtered through a 0.45 um sterile disposable filter (Mandel Scientific, Guelph,

ON, Canada) prior to storage at 4°C.

Detection of phage activity was performed using the spot-test technique

(Sambrook et al, 1989). Briefly, 100 ul (200 ul in case of Listeria) of bacterial host overnight culture were added to 4 ml of molten TSB containing 0.4% agarose (cooled to approximately 50-55°C before being used) and mixed. The soft agar was poured immediately onto TSA (1.5% agar) plates and allowed to solidify for 15 min. Phage activity was tested in the enriched samples by spotting 10 (0.1 from each previously prepared supernatant on the top soft agar and allowing it to dry for 20 min before incubation for 16-20 h at 25°C. After incubation, the plates were examined for the presence of complete or partial lysis zones. These zones were then removed from the

TSA plates by cutting the soft layer from the plate using a sterile wire loop and placing them separately in 1ml of k-Ca2+ phage buffer (k buffer: 2.5g/L MgS04. 7H20; 0.05g/L gelatin; 6mL/L 1M Tris buffer; pH 7.2). Following autoclaving at 121°C for 15 minutes, filter-sterilized CaCl2.2H20 was added to X buffer to a final concentration of 5 mM. The tubes were held at room temperature overnight to let the phages diffuse out of the soft

58 agar. The mixture then was filtered through a 0.45 urn membrane filter to purify the phages.

Table 2.1. Bacterial strains that were used for phage isolation and propagation

CRIFS culture Genus Species/Serovar Strain Designation collection number Escherichia coli C899 0157:H7,ATCC 43888 Escherichia coli C761 0126:H8, EC 910061 Escherichia coli C772 0113:H21 EC 910020 Escherichia coli C758 0153:H25 EC 910010 Escherichia coli C760 022:H8 EC 910018 Salmonella Typhimurium C1077 DT104-SA2002-5807 Salmonella Typhimurium C435 SA 941256 Salmonella Enteritidis C417 En-2588 Salmonella Heidelburg C434 SA 941270 Salmonella Brandenburg C384 S-474 Shigella boydii C 865-2 _ Shigella sonnei C 866-2 _ Shigella flexneri C 869-2 1:C4 Shigella sonnei C 870-2 1:D6 Listeria grayi C837 . Listeria ivannovii C877 - Listeria ivannovii C370 - Listeria innocua C399 . Listeria innocua C505 . Listeria monocytogenes C391 4b Listeria monocytogenes C519 l/2b Listeria monocytogenes C375 l/2a Listeria monocytogenes C561 3a

59 2.3.3 Purification of phages

The isolated phages were purified using the soft agar overlay method described by

Sambrook et al. (1989). A series of 10-fold dilutions were prepared from each phage lysate, and lOOul from each dilution and 100 ul of overnight host bacteria (200 ul in case of Listeria) were mixed and left for 15 min at 37°C. To this suspension, 4ml of molten soft agar were added at 45°C. The mixture was poured onto TSA plates and the plates were incubated at 25°C for 16-20 h. Plaques from different hosts, of varying sizes and different morphology, were picked from the overlay plates and placed separately in 1ml

^-Ca2+ phage buffer and left overnight to allow the phages to diffuse into the buffer. The picking of a single plaque from the soft agar was repeated three successive times to ensure the purity of the selected phages. The titre of the phage was determined after a 10- fold dilution of the isolated phage and tested by the previously described soft agar overlay technique. The phage suspensions were stored at 4°C for propagation.

2.3.4 Propagation and stock preparation

Phage stocks were prepared by the soft agar overlay method as previously described in which 100 ul of the phage suspension were mixed with 100 ul (200 ul in case of Listeria) of an overnight culture of the original bacterial host and incubated for 30 min at 30°C to allow attachment. Following this step, 4 ml of molten TSB containing

0.4% agarose were added, mixed and poured onto TSA plates. After solidification, the plates were incubated at 25°C for 16-20 h. For each phage, 10 plates were prepared in this way. Phages were collected by adding 3 ml of A,-Ca phage buffer to each plate and the top layer of soft agarose from all plates were scraped off using sterile hockey sticks.

60 All the scraped top agarose layers were transferred into two 50 ml tubes using sterile pipettes. The remaining agarose and phages were washed from the plates with another lml of the buffer for each plate and added to the collected agarose. The tubes were placed on ice for 30 min. The mixture was gently vortexed (Vortex-Genie-2; Scientific

Industries, Inc., Bohemia, NY) and centrifuged at 7000 x g for 20 min at 4°C. The supernatant was transferred to another tube and filtered through a 0.45 |j.m membrane filter. The titre of each phage stock was determined by preparing serial 10-fold dilutions and tested using the previously described overlay method. The phage lysate was stored at

4°C.

2.3.5 Host Range Determination using Bioscreen C

The host range of all the isolated phages on selected strains of their bacterial hosts

(20 strains for Salmonella and E. coli phages while 24 and 29 strains were used with

Listeria and Shigella phages, respectively) (Canadian Research Institute for Food Safety

(CRIFS) Culture Collection, University of Guelph) was determined by measuring the optical density (OD) of the tested bacteria in the presence of phage using the Bioscreen C

Microbiology Plate Reader (Labsystems, Helsinki, Finland). The following experimental parameters were used for all experiments: single, wide band (wb) wavelength; 25°C incubation temperature; 5 min preheating time; kinetic measurement; measurement time

24 hours; reading every 20 min and medium intensity shaking for 10 s before measurements. Fifty microliters of the phage lysate were transferred to each of the 100 wells of the sterilized honeycomb plates of the Bioscreen C reader (Fisher Scientific,

Mississauga, ON), and then each of the wells was inoculated with 125ul of the diluted overnight host bacterial culture (around 10 CFU/ml). The multiplicity of infection (MOI)

61 was around 100. The control wells contained either phage only, phage buffer only or bacteria with phage buffer with the same volume as tested wells. All samples were tested in triplicate. The OD data were analyzed using the Bioscreen C data processing software version 5.26 (Labsystems, Helsinki, Finland) to determine the detection time (time required for each test well to increase by 0.3 OD units). Detection times (h:min) were converted to decimal values, averaged and the mean control detection time was subtracted from all test data for each isolate tested and expressed as detection time difference (DT diff). Instead of having positive and negative results and based on this time difference, we proposed that the lytic activity of the phages can be classified as; (N): in which phage did not cause any delay in the tested bacterial growth and the growth curve was similar to that of the control; (D): phage cause a delay of tested bacterial growth by less than five hours; DT < 5 hrs, (D+): phage cause a delay of the tested bacterial growth by 5 or more hours; DT > 5 hrs, and (C): in which phage caused complete inhibition of the bacterial growth after 24 hours.

2.3.6 Phage DNA Isolation and Restriction Endonuclease Digestion

The Midi Lambda DNA purification kit was used for phage DNA purification according to the instructions of the manufacturer (Qiagen, Chatsworth, CA). After washing with 70% ethanol, each DNA pellet was resuspended in 20ul TE buffer (lOmM

Tris, ImM EDTA, pH 8.0) and stored at -20°C. The purified DNA samples were digested with different restriction enzymes (PstI, Ndel, Ecorl, EcoRV, Sspl, Dral, AccI and

Hindlll) according to the supplier's recommendations (New England BioLabs, Pickering,

ON, Canada). DNA fragments were separated by running 20 uL of the digested DNA on

1% agarose gel (Bio-Rad Laboratories Limited, Mississauga, ON, Canada) containing

62 ethidium bromide (0.5ug/ml) in lx Tris-acetate-EDTA buffer at 90 V using Bio-Rad agarose gel electrophoresis system model Power PAC 200 (Bio-Rad Laboratories

Limited, Mississauga, ON, Canada). The DNA fragment pattern was compared visually in order to identify the similarities and the differences between the isolated phages.

2.3.7 Transmission Electron Microscopy

The morphology of the isolated phages was examined by electron microscopy. To prepare the phages for electron microscopy, 1ml of high titre stocks was centrifuged at

16000 x g for 1 h at 4°C (Beckman J-20 centrifuge, Beckman Coulter Inc., Mississauga,

ON, Canada) and washed twice using phage buffer. The supernatant was discarded and the pellet was gently resuspended in 20 ul of X,-Ca+2 buffer. Five microliters of the resuspended phages were applied onto 300-mesh copper grids coated with formvar and allowed to stand for 2 min. The excess liquid was drawn off by filter paper and the remaining phages were negatively stained with 2% uranyl acetate for 30s and then the excess liquid was drawn off again by blotting with filter paper. Finally the samples were examined in a LEO 912AB electron microscope (Energy filtered TEM, EFTEM, LEO

912ab model operated at 100 kv, Zeiss, Germany).

2.3.8 Stability of Phages under Different Temperatures and pH's

To test the pH stability of the phage, lOOul of the phage lysate were added to

900ul of buffer solutions with different pH's (4.0, 5.0, 9.0, and 10.0), along with the control (pH 7.0), and then incubated at 25°C for 18 h. For the temperature treatments, phage lysates of known titres were incubated at different temperatures; -20°C, 4°C, 37°C, and 42°C. Each treatment was performed in triplicate. After incubation, ten-fold serial

63 dilutions for each treatment were made and lOOul of each were mixed with lOOul of overnight culture bacterial host (200ul in case of Listeria), incubated for 15 min at 37°C then 4 ml of overlay media were added to the mixture and spread over TSA plates. Plates were incubated at 25°C and the PFU/ml was determined the next day. The average of the triplicate counts was taken and the log unit reduction in the phage titre was calculated.

2.3.9 Infectivity of Phages under Different Environmental Conditions

The infectivity of the selected phages was examined by spotting 20 ul from each phage over a lawn of its bacterial host on TSA plates of different pH (4.5, 6.0, 7.0 and

9.0) or containing different NaCl concentrations (5%, 10% and 15%). After diffusion, the plates were incubated at 25°C. The effect of incubation at different temperatures (4°C,

10°C and 25°C) and atmosphere (anaerobic, MAP and vacuum) was also examined by

spotting 20 ul of tested phage over a lawn of its bacterial host on normal TSA plates. The modified atmosphere was composed of approximately 1 % O2, 13 % CO2 and 86 % N2, which was generated by using an AnaeroGen sachet in a sealed anaerobic jar (Oxoid,

Fisher Scientific, Mississauga, ON, Canada). The vacuum packaging was performed using a Komet vacuum packaging machine (Stuttgart- W, Kornbergstr 27-29, Germany) at a vacuum of 1.0 bar. The plates were placed in sterile polyethylene bags of 65 um thickness (Seward Laboratory Systems Inc., Bohemia, NY) before being placed in the vacuum machine. The development of lytic areas was examined daily for 2 weeks. The whole experiment was repeated three times.

64 2.3.10 Determination of the Frequency of Emergence of Bacteriophage

Insensitive Mutants (BIM) and Lysogenic Potential

The frequency of bacteriophage insensitive mutant (BIM) development was determined as described previously (O'Flynn et al, 2004, O'Flaherty et al, 2005). Phage lysate (100 ul) was added to 100 ul of its bacterial host culture at a multiplicity of infection (MOI) of 10 and incubated for 15 min at 37°C. Then, 4 ml of overlay agarose medium were added and poured over TSA plates. The plates were incubated for 24 h at

25°C. Both single phages and a cocktail of phages were used. Any developing colonies were counted, and the BIM frequency (number of surviving colonies divided by the original bacterial titer) was calculated. All experiments were performed in triplicate. To examine the lysogenic ability of the selected phages, the developed colonies from the

BIM experiment were picked up and subcultured five successive times on TSA plates to remove any residual phages. The produced separate colonies were induced in broth tubes by 0.5 mg/1 mitomycin C, heat shock or by UV for different times. The tubes were then centrifuged at 5000 x g for 5 minutes and the supernatant was filtered through a 0.45 um membrane filter. The obtained lysates were tested for their ability to cause lysis on the bacterial host.

2.3.11 Cross Infectivity

The ability of the selected phages to infect different species from other genera was examined. Twenty microliters of each tested phage were spotted over a lawn of different bacterial species of genera other than the bacterial host genus. The development of a lytic area in the plates was examined after incubation at 25°C for 18 h. Samples of

65 phage were tested in triplicate and the whole experiment was repeated two times.

2.3.12 Determination of the Phage Genome Size using PFGE

Genome sizes of the phages were determined by pulsed-field gel electrophoresis

(PFGE) as described previously (Lingohr et al, 2009). Phage DNA embedded in 1.4%

Seakem Gold agarose (Mandel Scientific, Guelph, ON) was subjected to electrophoresis

in 0.5 x TBE buffer at 14 °C for 18 h, using a Chef DR-III Mapper electrophoresis

system (Bio-Rad, Mississauga, ON), with pulse times of 2.2 - 54.2 s, at 6 V/cm. Low range PFG marker and phage lambda DNA concatemers (NewEngland Biolabs, NEB,

Canada) were used as size standards. The bands were visualized under UV transillumination after staining with ethidium bromide. PFGE results were analyzed using

BioNumerics software (Applied Maths, Inc., Austin, TX).

2.3.13 One-step growth curve

Burst size and latent period for the selected phages were determined by a one-step

growth experiment with some modifications from that described early (Ellis and

Delbruck, 1939). Phage was added to its host bacterium at a MOI of around 0.1 and

incubated in a water bath at 30°C for 5 min. Then, 1ml was removed and added to lOOul

of chloroform and mixed very well. One hundred microliters of this mixture were added to lOOul of the host bacterial culture with 4 ml of overlay media and poured onto TSB

agar plates to determine the degree of adsorption of phage to bacterial cells. After 5.5

min, lOOul were transferred to a tube containing 9.9 ml of fresh TSB and then 1 ml was transferred from this tube to 9 ml TSB and the tube was gently shaken before transferring

lml to a further tube containing 9 ml TSB. All three tubes were incubated in a water bath

66 at 30°C. Beginning from 6 minutes after addition of phage to its host and at 5 min intervals for 180 min, the diluted suspensions were plated by removing 100 ul of the suspension and adding 100 ul of the indicator bacterium and 4 ml of overlay media. The mixture was briefly mixed and poured onto TSA agar plates. The number of plaques was determined after 24 h ul of incubation at 25°C. The relative burst size was determined according to the equation:

Relative burst size = [(Final titre - Initial titre) / initial titre].

The relative burst size at different times was plotted against time to determine the latent period and burst size.

2.3.14 Statistical analysis

The statistical analysis of the experimental data was accomplished with SigmaPlot

Version 10.1 (Systat Software Inc., Chicago). A one-way analysis of variance (ANOVA) was performed. Statistical differences between the means were indicated by P < 0.05.

2.4 RESULTS

2.4.1 Isolation of phages

Testing the collected environmental samples (rinse water, fecal samples, and sewage) against Salmonella, Listeria, Shigella and E. coli strains (identified in Table 2.1) resulted in the isolation of 28, 24, 24 and 31 phages of various plaque morphologies, respectively. While it was anticipated that several of these phages were likely to be identical based on indistinguishable plaque morphologies, all isolated phages were subjected to host range determination and Bioscreen C characterization.

67 2.4.2 Host range pattern and determination of the identical isolated phages

using Bioscreen C

Stock lysates of the isolated E. coli and Salmonella phages were tested on 20 selected strains of E. coli and Salmonella, while Shigella and Listeria isolated phages were tested using 24 and 29 strains of Shigella and Listeria, respectively. Based on the obtained detection time difference values, some tested phages produced a growth curve were similar to that of the control (N), while others caused a complete inhibition of the growth for 24 hours (C). On the other hand, some tested phages caused a delay in the bacterial growth to reach the detection limit (OD = 0.3) and the bacteria started to grow again before the end of the experiment (D and D+). Figure 2.1 represents an example for all these cases.

Salmonella phages were not able to cause complete killing of all tested bacterial strains; however, several of them were able to cause a delay in time for bacterial growth to reach the detection limit by more than 5 h (D+) when compared with the control (Table

2.2). Salmonella phages AG1 to AG20, AG24 and AG26 were unable to infect only 8 of the 20 Salmonella strains examined, while AG 14, AG 16 and AG28 were unable to infect

9 strains. Listeria phages AG1 to AG5 were found to have the narrowest host range pattern of the tested isolated Listeria phages (Table 2.3). They were able to infect only 3

Listeria strains out of the 29 examined. On the other hand, Listeria phage AG8 was able to infect all the examined Listeria strains with varying degree of activity, causing complete growth inhibition of 24 strains. Listeria phage AG 13 and AG20 were unable to infect only one and three strains of the examined Listeria strains, respectively.

68 1.2 -i -•— control

|r> -•- Phage 1 (N) ~ 1 nn**"^^1^ Phage 2 (D) -*- Phage 3 (D+) -*- Phage 4 (C) | °-6 1 °-4 ° 0.2

Un n i i i i i

% V ** Hours

Figure 2.1. Representative results from Bioscreen C after adding phage to a tested bacterial strain. Patterns of the growth curves of E. coli 0126:H8 in the presence and absence of different phages; phage 1: growth similar to control (N), phage 2 and 3 cause detection time difference of < 5 and > 5 hours, respectively (D and D+), Phage 4 causes complete lysis of the bacterial cells (C) after 24 h incubation with the tested strain at 25°C.

For the isolated E. coli phages, the narrowest host range phage was AG28 which was able to infect only two strains of the twenty used E. coli strains (Table 2.4). E. coli phages AG21, AG22, AG23, AG24 and AG34 were able to infect only five E. coli

strains, while AG25 and AG27 showed lytic activity only against four E. coli strains among all the examined strains. E. coli phages AG5, AG6 and AGIO had the broadest host range, being able to infect 15 E. coli strains. By comparison, E. coli phage AG5 caused complete growth inhibition of 13 strains among the sensitive 15 E. coli strains.

Shigella phages showed the greatest variation in the host range patterns when

69 compared with Salmonella, E. coli and Listeria phages (Table 2.5). The narrowest host range phages among the isolated Shigella phages were AG 13 and AG22, which were found to only infect three and five strains, respectively, among the 24 examined. On the other hand, the broadest host range isolated Shigella phages was Shigella phage AGIO, which was able to infect 20 Shigella strains; with the growth of 11 strains being completely inhibited.

Some of the Salmonella and Listeria phages which gave identical host range patterns were randomly selected and examined morphologically by TEM and their genomic material was subjected to digestion by different restriction enzymes. Phages that represent the broadest and narrowest host range patterns were selected. All the examined phages had icosahedral heads and possessed contractile or non-contractile tails. These phages essentially represent four different morphotypes (Figure 2.2 and 2.4). A number of the examined phages were members of the Siphoviridae family as designated by the presence of a flexible long tail and the absence of a contractile sheath, such as Salmonella phages AG11, AG25 and AG27 and Listeria phages AG11, AG19, AG22, AG23 and

AG24. Salmonella phages AG1, AG4, AG6, AGIO, AG13, AG20 and AG26 and Listeria phages AG1 to AG5 had contractile, nonflexible tails and therefore belonged to the family Myoviridae. Moreover, the members of each tested group of Salmonella phages gave identical restriction digestion patterns when digested by restriction endonuclease enzyme AccI (Figure 2.3). In addition, restriction digestion patterns using EcoRI were identical among phages of each examined group of the selected Listeria phages (Figure

2.5).

70 Table 2.2. Host range pattern of the isolated Salmonella phages using Bioscreen C. Similar host range patterns were grouped together.

Salmonella strains with CRIFS culture collection number and serotype PHAGE

D D+ D D D D D D D D+ D D D D D D D D D D D D+ D D D D D D D D D D D D+ D D D D D D D D D D D D+ D D D D D D D D D D D D+ D D D D D D D D D D D D+ D D D D D D D D D D D D+ D D D D D D D D D D D D+ D D D D D D D D D D D D+ D D D D D D D D D D D D+ D D D D D D D D D D D D+ D D D D D D D D D D D D+ D D D D D D D D D D D D+ D D D D D D D D D D D D+ D D D D D D D D D D D D D D D D D D D D D D D+ D D D D D D D D D D D D+ D D D D D D D D D D D D+ D D D+ D D D D D D D D D+ • D D D D D D D D D D+ D D D D D D D D D D D+ D+ D+ D D+ D+ D+ D+ D+ D+ • D D D D+ D+ D+ D D • D D+ D D D D+ D+ IH D D D D+ D D D D+ D+ D+ D D HUH D D+ D D D+ D+ m- D+ D D mmiml D D+ D D D+ D+ D+ D+ D D ^^i D D C : Complete inhibition of the bacterial growth D+: More than 5 hours delay in bacterial growth to reach log phase compared to control (i.e. Detection time difference > 5 hours) D : Less than 5 hours delay in bacterial growth to reach log phase compared to control (i.e. Detection time difference < 5 hours) N : No effect of phage on bacterial growth (growth like control) Table 23. Host range pattern of the isolated Listeria phages using Bioscreen C. Similar host range patterns were grouped together.

Listeria species with CRIFS saifare eolleetioi number PHAGE en* C505 C.J70 077 C8J7 075 CJ91 C561 (519 If 2P it 4P 5P 6P iimocuu innoeua ivanovii ivuimvii gruyi monocytogenes muttvcytogaies nwHotylogenes mtntiKylogenef> nmtwcyltt^eiiat monocytogenes monocytogenes monocytogenes monocytogenes monocytogenes AGS C C D C C C C C C C C C C D C

AG 13 C C D C C C C C C C C C C D C

AG 20 C c D C C C c C C C D+ D+ D D c

AG 18 C c D C C C c C C c C C C D c

A(J 15 C c D C C C c C C c C C C D c

AG 9 c c D C C C c C C c C C D+ D c

AG 11 c c D C C C c C C c C C C D c AG 19 c c D C C C c C C c C C C D c AG 22 c c D C C C c C C c C C C D c AG 23 c c D C C C c C C | C C C C D c AG 2.1 c c D C C C c C c c C C C D c

AG 21 c c D C C C c C c D D D D D c

AG 10 c c D C N C c C c D D D D D c

AG 17 c c D C C c c C c N D D N D c

AG 6 c c D C C c c C c D D D N N c

AG 16 c c D C N c c C c N N N N D c

AG 7 c c D C N c c C c N N N N N c AG 12 c c D C N c c C c N N N N N c

AG 14 c D+ D C N N c D c N N N N N N

AG1 N N D D+ H N N N D N N N N N N AG 2 N N D D+ N N N N D N N N N N N AG 3 N N D D+ N N N N D n N N N N N N AG 4 N N D D+ N N N N D N N N N N N AGS N N D D+ \ N N N D ^ N N h N N Continued... Listeria species with CRIFS culture collection number PHAGE 7P 8P 9P 10P IIP 12P UP UP 15P 16P 17P ISP 19P 20P Monocytogenes monocytogenes nwnocyUigenes monocytogenes monocytogenes numocytogenes monocytogenes monovytogt'nes monocytogenes monocytogenes monocytogenes monocytogenes monocytogenes inottocytogenes AGS C C C C C C C C m D D C C C

AG 13 C C C c C C D+ c D N D D D+ C

AG 20 C C D c C C N c D N N D D C

AG 18 c C C c C C N c N N N D D+ C

AG 15 c C C c C C N c N D D N N C

AG«> c C D c C C N c D+ N N N N C

AG 11 c C C c C C N c N N N N N C AG 19 c C C c C C N c N N N N N C AG 22 c C C c c C N c N N N N N c AG 23 c C C c c C N c N N N N N c AG 24 c C c c c C N c N N N N N c

AG 21 c c N c c C N c N N N D N c

AG 1(1 c c D c c C N c N N N N N D+

AG 17 c c N c c C N c N N N N N C

AG 6 D+ c N c c N N c N N N N N c

AG 16 c c N c c C N c N N N N N c

AG 7 c c N c c C N c N N N N N c AG 12 c c N c c c N c N N N N N c

AG 14 N c N c N N N c N N N N N c

AG1 N N N N N N N N N N N N N N AG 2 N N N N N N N N N N N N N N AG 3 N N N N N N N N N N N N N N AG 4 N N N N N N N N N N N N N N AGS N N N N N N N N N N N N N N C : Complete inhibition of the bacterial growth D+: More than 5 hours delay in bacterial growth to reach log phase compared to control (i.e. Detection time difference > 5 hours) D : Less than 5 hours delay in bacterial growth to reach log phase compared to control (i.e. Detection time difference < 5 hours) N : No effect of phage on bacterial growth (growth like control) Table 2.4. Host range pattern of the isolated E. coli phages using Bioscreen C. Similar host range patterns were grouped together.

E. coli strains with CRIFS culture coOkcttooon mmmitosr and strain designation

PHAGE 1761 £465 l'*!« C8'J4 fOU C75H (4fc4 t^53 (ll'> I "-, i ( "M C-I7II C7<*J C7 CV5 (674 (^W) CV>

0126:H8 0126:HS 0157:H7 0157: H7 0157:H7 0153:H25 0153:H25 O103:H2 O103:H2 O103:H2 OllliNM- 0111:NM 0132:NM 0132:NH OII5:H28 0I15:H18 02:H5 02:Hl 022 OM3

AG 5 C C DT C C D+ C C C C | C N C C N N C C N N AC 6 C C DT DT D+ D+ C C DT C C N C c N N c C N N AGIO C DT DT DT c DT C C D+ C | C N D+ c N N c DT N N 1 AC 2 LH C DT DT DT DT C D+ DT D. 1 DT N C c N N DT N N N

AG 3 DT c DT DT D+ N C C DT c I DT N c c N N C DT N N AG 7 C ]> DT DT DT N D+ C DT c D+ N c c N N C DT N N AGS c DT DT DT DT N C c C c DT N DT c N N C DT N N AG 9 1> C DT DT DT N c c DT c l>f N C c N N C DT N N

AG 16 c 1> DT c c D N c C c D+ N C c N N C D~ N N AG I? DT D \> D+ c DT C c DT c DT N C c H N H N N N AG1 c C DT D+ C N N c DT D- DT N C c N N C D+ N N AG 4 c c D- DT D+ N N c DT c C N C c N N C DT N N AG 11 c DT DT C C N N c DT c c N I> DT N N c DT N N AG 14 c DT DT C C N fj c c c DT N C c N N c D- N N AG 17 c DT DT C c N N c c c IK N c c N N c DT N N AG 18 c DT D+ c c N N c c c DT N c c N N c D- N N AG 19 c D DT c D N N c C c DT N c c N N D+ DT N N

\c.to N H c DT DT C DT N N N N N DT DT DT C N N C N

\C 31 N N DT DT DT C DT N N N N N N N N D+ N N DT N

AG 21) N N D N N N N N D c N H D> DT N N X DT D N

AG2» D N N N N N N W N N N N N D+ DT C N N~ DT DT

AG 12 N N DT C D+ D+ C N N N N N N N N N Dt H a N u; \i N N D- LK C DT C N N N N N N N N N C N N N AG 2-1 D a N N N « N N N N N N N N DT C N N c DT AG 21 N N N N N N DT N N N N N N N DT C N N c DT

AG 22 N N N N N N • tf N N N N N N DT DT c H N c D- AG 23 N N N N N N DT N N N N » N N DT N N DT c DT AG 26 N N N N N DT N N N N N N N N DT c N N c D+

AG 25 N N N N N N H N N N N N N N DT c S M D+ DT AG 27 N N N N N N N H N N N N N N D+ c N N c DT

AG 28 N N N N N D^ N N N 1 N N N N N N N N D N C : Complete inhibition of the bacterial growth D+: More than 5 hours delay in bacterial growth to reach log phase compared to control (i.e. Detection time difference > 5 hours) D : Less than 5 hours delay in bacterial growth to reach log phase compared to control (i.e. Detection time difference < 5 hours) N : No effect of phage on bacterial growth (growth like control) Table 2.5. Host range pattern of the isolated Shigella phages using Bioscreen C. Similar host range patterns were grouped together.

Shigella species with CRIFS culture collection number

^^^^^^™ ^^^ ^^^ ^^: C ^c^ ^^^ ZZ^J ^^^^^^E^ Z ^^^~ ^^^ ^Dt K| ^^^^ __2^_ ^^^^ __2^__ ^^^^ ^Dt^ ^^^^rv^ Dt D c D C D C ^^^^^^^^ c ^^^^ i>^n ^^^ ^^^1 D C C Dt Dt HBH__EI_ lit c D ^^^ ^C^ I Dt 1 c Dt | 1 C C 1 D C Dt Dt HH^^^^Z ^^^ ^^^ ^Dt 1 c Dt | C Dt Dt Dt ^^^ ^^^ c 1 c C 1 D C C Dt ^^^ ^C^ ^Dt L_L_ Dt | C Dt Dt C HH^^^^ C c ^^^ ^^^ 1)^1 ^^^^^ Dt Dt ^^^^^ _^^_ ^^^™ ^H^^^^^ Dt c c ^C^ r>t ^^^^^_ ^Dt ^^^ ^^^T ^[^^ ^^^^^ ^^^^^ C c c C ^C^ ^C^ ^^^ ^^^^ p^^^^: ^^^ ^Dt^ ^^C^ _D+_ ^^^^^^™ ^^^™ Dt ^^^ ^^^ ^^^^^_ ^I> ^^j ^^_ ^^^T ^[^^ ^Dt^ ^^C^ ^^^ c ^^^ ^C^ ^C^ ^C^ ^^^ ^^^^ ^^^^H ^^_ ^Dt^ ^^C^ ^Dt^ ^^^ ^^^ ^^^ ^^^ ^^^ ^^^ ^^^3 ^^^ ^^^ ^[^^ ^^^^^ C ^c^ ^^^ ^^^ ^C^ c ^^^B ^^^^^^ ^^^ ^[^^ ^^^^^ MMH^^^Z C D ^^^ ^^_ Dt Dt C c c c C c ^^^ D C ^^^ ^^^ ^c^ ^^^ ^^^ ^^^ _^_J I Dt ^I>^ Dt Dt ^c^ ^^^ c c __^L_ ^c^ ^^^ ^^^ ^^^ D Dt l>* ^^^ ^^^ ^^_ ^c^ ^^^3 ^Dt ^[>^ ^^^^ __2>__, ^••^11 D D c D c ^B™~r™ D •!• D+ t> D ^^_ D Dt ^H__L_ D ^^^ ^c^ ^c^ ^^^ ^^_ D __^1__

^c^ ^c^ 1^^ _Dt_ c c ^^^I)t D C : Complete inhibition of the bacterial growth D+: More than 5 hours delay in bacterial growth to reach log phase compared to control (i.e. Detection time difference > 5 hours) D : Less than 5 hours delay in bacterial growth to reach log phase compared to control (i.e. Detection time difference < 5 hours) N : No effect of phage on bacterial growth (growth like control) Figure 2.2. Representative transmission electron micrograph of some isolated Salmonella phage groups that showed similar host range pattern. A is a representative image for morphological structure of isolated phages AG 11, 25and 27 while B is a representative image for isolated phages phages AG 1, 4, 6, 10, 13, 20 and 26.

Ml 2 3456789 10 M 10.0Kb

3.0Kb

1.0Kb

0.5Kb

Figure 2.3. Electrophoresis on 1.0 % agarose of AccI restriction enzyme digests of DNA from representative Salmonella phages that showed similar host range patterns. Lanes 1 to 3 are for phages AG 11, 25 and 27. Lanes 4 to 10 are for phages AG 1, 4, 6, 10, 13, 20 and 26 respectively. M lanes are for 1 Kb ladder.

76 B

Figure 2.4. Representative transmission electron micrographs of some isolated Listeria phage groups that showed similar host range pattern. A is a representative image for morphological structure of isolated phages AG 1 to 5 while B is a representative image for isolated phages AG 11, 19, 22, 23 and 24.

M12 3456 7 89 10

1.0Kb

0.5Kb

Figure 2.5. Electrophoresis on 1.0% agarose of EcoRI restriction enzyme digests of DNA of representative Listeria phages that showed similar host range patterns. Lanes 1 to 5 are for phages AG 1 to 5. Lanes 4 to 10 are for phages AG 11, 19, 22, 23 and 24 respectively. M lane is for 1 Kb ladder.

77 Based on the host range results, the three broadest host range phages against each targeted genus were selected for further characterization (Table 2.6). The phages were named according to the recent recommendations for phage nomenclature by the

International Committee on Taxonomy of Viruses (ICTV), by putting the first letter of the genus, first two letters of species, first letter of the phage family, first letters of the authors' last names and isolation number (Andrew Kropinski and Rob Lavigne, personal communication). Table 2.7 shows the selected phages and their bacterial host that was used for their isolation and propagation.

2.4.3 Characterization of the selected phages

2.4.3.1 Morphology

Figure 2.6 shows the common morphology of the twelve selected phages by electron microscopy of negatively-stained preparations. Morphologically, the examined phages were related to previously reported phages in the literature (Table 2.8). All of them except SenS-AGll phage had icosahedral heads with contractile tails and therefore were designated to members of the family Myoviridae. SenS-AGl 1 phage was a member of the Siphoviridae family as it had an icosahedral head with a non-contractile tail and its morphotype was related to Jersey phage of Salmonella Paratyphi B (Ackermann and

Gershman, 1992). The selected E. coli phages AG2, AG3 and AGIO were morphologically similar and were considered to be T even-like phages (Miller et al,

2003). Similarly, the morphology of Listeria phages AG8, AG13 and AG20 are related and were found to be morphologically similar to A511 Listeria phage (Zink and

Loessner, 1992). Salmonella phages AG6 and AG 16 and Shigella phage AG3 possessed

78 the same morphotype, which was related to Vil phage of Salmonella Typhi (Zink and

Loessner, 1992). SsoM-AG8 and SsoM-AGlO phages were morphologically related to

Sfv phage of Shigella flexneri (Allison et al, 2003). The dimension of the heads ranged from 40 to 105 nm, while the tail dimensions were from 67 to 213 nm and diameters from

8tol8nm(Table2.8).

79 Table 2.6. Host range pattern of selected phages. a) Salmonella phages

Salmonella host StyM- SenS- StyM- Salmonella host StyM- SenS- StyM- (Identification number) AG6 AG11 AG16 (Identification number) AG6 AG11 AG16 S. Typhimurium(C1077) D D D+ S. Schwarzengrund (C 431) D+ D+ N S. Typhimurium (C435) D D D+ S. Braenderup(C410) N N N S. Brandenburg (C384) D D D S. Tennessee (C408) N N N S. Brandenburg (C428) D D D+ S. Thompson (C412) N N N S. Enteritidis (C721) D D D+ S. Choleraesuis (C402) N N N S. Enteritidis (C417) D D D+ S. Ohio (C409) N N N S. Heidelberg (C390) D D D+ S. Mantis (C413) N N N S. Heidelberg (C434) D D D+ S. Montevideo (C419) N N N S. Saintpaul (C433) D D N S. Berta(C416) D D D+ S. Bredeney (C432) N N N S. Panama (C424) D N D+

C : Complete inhibition of the bacterial growth D+: More than 5 hours delay in bacterial growth to reach log phase compared to control (i.e. Detection time difference > 5 hours) D : Less than 5 hours delay in bacterial growth to reach log phase compared to control (i.e. Detection time difference < 5 hours) N : No effect of phage on bacterial growth (growth like control)

80 b) Listeria phages

Listeria species LinM- LmoM- LmoM- Listeria species LinM- LmoM- LmoM- (Identification number) AG8 AG13 AG20 (identification number) AG8 AG13 AG20 L. innocua C399 C C C L. monocytogenes (7P) C C C L. innocua C505 C C C L. monocytogenes (8P) C C C L. ivanovii C370 D D D L. monocytogenes (9P) c c D L. ivanovii C877 C C C L. monocytogenes (10P) c c C L. grayi C873 C C C L. monocytogenes (IIP) c c C L. monocytogenes (C375) C C C L. monocytogenes (12P) c c C L. monocytogenes (C391) C C C L. monocytogenes (13P) c D+ N L. monocytogenes (C561) C C C L. monocytogenes (14P) c C C L. monocytogenes (C519) C C C L. monocytogenes (15P) D+ D D L. monocytogenes (IP) C C C L. monocytogenes (16P) D N N L. monocytogenes (2P) C C D+ L. monocytogenes (17P) D D N L. monocytogenes (3P) C C D+ L. monocytogenes (18P) C D D I. monocytogenes (4P) C C D I. monocytogenes (19P) C D+ D Z,. monocytogenes (5P) D D D Z,. monocytogenes (20P) C C C I. monocytogenes (6P) C C C C : Complete inhibition of the bacterial growth D+: More than 5 hours delay in bacterial growth to reach log phase compared to control (i.e. Detection time difference > 5 hours) D : Less than 5 hours delay in bacterial growth to reach log phase compared to control (i.e. Detection time difference < 5 hours) N : No effect of phage on bacterial growth (growth like control) c) E. coli phages

E.coli strains EcoM- EcoM- EcoM- E.coli strains EcoM- EcoM- EcoM- (Identification number) AG2 AG3 AG10 (Identification number) AG2 AG3 AG10 E. coli 0126:H8 (C761) C D+ D+ E.coli 0111:NM(C764) C D+ D+ E. coli 0126:H8 (C465) D+ C C E.coli Olll:NM(C470) N N N E.coli 0157:H7(C899) D+ D+ D+ E.coli O132:NM(C790) D+ C C E.coli 0157:H7(C894) D+ D+ D+ E.coli 0132:NM(C791) C C C E. coli O157:H7(C700) C D+ ' D+ E.coli 0115:H28(C771) N N N E.coli 0153:H25(C758) D+ D+ N E.coli 0115:H18(C466) N N N E. coli 0153:H25 (C464) C C C E. coli 02:H5 (C675) C D+ C E.coli O103:H2(C753) C D+ C E. coli 02:H1 (C674) D+ N D+ E. coli O103:H2(C729) D+ D+ D+ E coli 022 (C760) N N N E.coli O103:H2(C754) C D+ C E.coli 0113 (C772) N N N C : Complete inhibition of the bacterial growth D+: More than 5 hours delay in bacterial growth to reach log phase compared to control (i.e. Detection time difference > 5 hours) D : Less than 5 hours delay in bacterial growth to reach log phase compared to control (i.e. Detection time difference < 5 hours) N : No effect of phage on bacterial growth (growth like control) d) Shigella phages

Shigella species SboM- SsoM- SsoM- Shigella species SboM- SsoM- SsoM- (Identification number) AG3 AG8 AG10 (Identification number) AG3 AG8 AG10 S. boydii (C865) D+ D+ D+ S. boydii (79-M09) C C C S. sonnei (C866) C D+ C S. boydii (74-3594) D D+ D S. flexneri (CS69) C D C S. boydii (84-11 \9) N N N S. sonnei (C870) C C C S. boydii (83-578) D+ C C S.jlexneri (61-1186) N D N S. boydii (99-4528) D+ C D+ S.jlexneri (71-2747) C C C S. dysenteriae (04-3380) C D C S. flexneri (04-3435) C D N 5*. dysenteriae (91-3501) D+ C D+ S. flexneri (95-3239) c C C S. dysenteriae (53-4738) N N D+ S. flexneri (05-3605) c N C S. dysenteriae (52-2050) C C C S. flexneri (86-3239) N N N S. dysenteriae (69-2387) D+ N N S. boydii (74-1789) D+ C C S. dysenteriae (94-3065) C D+ D+ S boydii (74-4334) D+ c D+ S. dysenteriae (79-8006) D+ D+ D+ C : Complete inhibition of the bacterial growth D+: More than 5 hours delay in bacterial growth to reach log phase compared to control (i.e. Detection time difference > 5 hours) D : Less than 5 hours delay in bacterial growth to reach log phase compared to control (i.e. Detection time difference < 5 hours) N : No effect of phage on bacterial growth (growth like control) Table 2.7. Selected isolated phages and their susceptible bacterial hosts used for propagation CRIFS culture collection Phage Bacterial Host number StyM-AG6 Salmonella Typhimurium C 1077 SenS-AG 11 Salmonella Enteritidis C435 StyM-AG16 Salmonella Typhimurium C417 LinM-AG8 Listeria innocuoa C505 LmoM-AG13 Listeria monocytogenes C391 LmoM-AG20 Listeria monocytogenes C519 EcoM-AG2 E. co/z 0126:H8 C761 EcoM-AG3 E. coli 0126:H8 C761 EcoM-AGlO E. coli 0157:H7 C899 SboM-AG3 Shigella boydii C 865-2 SsoM-AG8 Shigella sonnei C 866-2 SsoM-AGlO Shigella sonnei C 866-2

Table 2.8. Approximate dimensions, family and morphologically related phages for the selected phages.

Head Tail dimensions (nm) Phage Dimension Family Similar to Length Diameter (nm) StyM-AG6 61 90 14 Myoviridae Vil like SenS-AG 11 49 117 8 Siphoviridae Jersey like StyM-AG16 90 120 18 Myoviridae Vil like LinM-AG8 84 196 15 Myoviridae A511 Like LmoM-AG13 90 195 16 Myoviridae A511 Like LmoM-AG20 86 213 19 Myoviridae A511 Like EcoM-AG2 105 107 19 Myoviridae T-even like EcoM-AG3 102 103 17 Myoviridae T-even like EcoM-AGlO 105 109 17 Myoviridae T-even like SboM-AG3 83 110 14 Myoviridae Vil like SsoM-AG8 66 109 13 Myoviridae SfV like SsoM-AGlO 40 67 12 Myoviridae SfV like

84 Figure 2.6. Representative transmission electron micrographs of the different morphotypes of the twelve selected phages. Morphotype A is for E. coli phages AG2, AG3 and AGIO. Listeria phages AG8, AG13 and AG20 are of morphotype B. Morphology of Salmonella phages AG6 and AG 16 and Shigella phage AG3 is represented by image C. Morphology of the other Shigella phages (AG8 and AGIO) is represented by image D. All these phages have contractile tails while Salmonella phage AG11 has a noncontractile tail (image E).

85 2.4.3.2 One-step growth curve

Five phages (EcoM-AG2, LmoM-AG13, SsoM-AG8, SsoM-AG3 and SenS-

AG11) that represent the five common morphotypes among the twelve selected phages were selected for the determination of latent periods and burst sizes from their one-step growth curves (Figure 2.7 and Table 2.9). SsoM-AG8 phage has the lowest relative burst size among the tested Myoviridae phages (91) with a latent period of 43 min, while

LmoM-AG13 phage has the highest at 175 with a latent period of 72 min. The examined

Siphoviridae phage SenS-AGl 1 has a burst size of 72 and latent period of 23 min.

2.4.3.3 Restriction enzyme digestion patterns and determination of genome

sizes

The genetic material of all isolated phages was purified and digested using different restriction endonucleases. Digestion with several restriction enzymes indicated that the examined phages had double-stranded DNA. A distinct restriction digestion pattern was developed for each examined phage. DNA of Salmonella and Shigella, phages was digested by AccI and Ndel and gave distinct restriction digestion patterns (Figure 2.8 and 2.9). EcoM-AG2 phage DNA was digested with EcoRV, while distinct patterns were developed for EcoM-AG3 and EcoM-AGlO phages when digested with Sspl (Figure

2.10). Hindlll and EcoRI were used to differentiate between Listeria phages (Figure 2.11 a). No restriction sites for these two enzymes were detected in DNA of LmoM-AG13 phage, but it was digested by AccI (Figure 2.11 b). Based on the analysis of the pulsed field gel electrophoresis data, EcoM-AGlO phage had the largest genome size among the tested phage, with a DNA size of around 184 kb, while SenS-AGl 1 phage had the smallest genome of size of around 26 kb (Table 2.10).

86 200.00 -, -•- EcoM-AG2 -•—LmoM-AG13 SsoM-AG8 «> 160.00 J******' f-' "35 ****** t&•-* • -*-SboM-AG3 12 120.00 -*—SenS-AG11 3 f If > 80.00 ^tt****B**I *f dat i (5 40.00

U.UU HWWIWIWIWKWS S liJH w MSww'Wwi wiw ~^ i i i i i i i i i i 6 30 60 90 140 190 Minutes (post infection)

Figure 2.7. One-step growth curves of the phages that represent different morphotypes among the finally selected phages.

Table 2.9. Latent period and burst size of the phages that represent different morphotypes among the finally selected phages.

Phage Latent period (min) Burst size EcoM-AG2 67 142 LmoM-AG13 72 175 SsoM-AG8 43 91 SboM-AG3 57 152 SenS-AGl 1 23 72

87 Ml 2 3 M 5 6 7 M

10.0 kb

3.0 kb M.M

1.0 kb

0.5 kb

Figure 2.8. Restriction fragment produced from digestion of Salmonella phage genomic DNA with endonuclease AccI (lanes 1 to 3) and Ndel (lanes 5 to 7). Lanes 1 and 5 are for AG6, lanes 2 and 6 are for AG 11 and lanes 3 and 7 are for AG 16. Lane M is for 1 kb ladder.

M123M567M

10.0 kb *••. —, ^S mmm ZlZ 5.0 kb -

1.5 kb •<

0.5 kb ••

Figure 2.9. Restriction fragment produced from digestion of Shigella phage genomic DNA with endonuclease AccI (lanes 1 to 3) and Ndel (lanes 5 to 7). Lanes 1 and 5 are for AG3, lanes 2 and 6 are for AG8 and lanes 3 and 7 are for AGIO. Lane M is for 1 kb plus ladder.

88 M123M5 6 7M

10.0 kb •c=5 5 |: 3.0 kb •*-*

1.0 kb

0.5 kb

Figure 2.10. Restriction fragment produced from digestion of genomic DNA of E. coli phages with endonuclease EcoRV (lanes 1 to 3) and Sspl (lanes 5 to 7). Lanes 1 and 5 are for AG2, lanes 2 and 6 are for AG3 and lanes 3 and 7 are for AGIO. Lane M is for 1 kb ladder.

A) B) M123M567M M1M 10.0 kb 10.0 kb

3.0 kb 3.0 kb

1.0 kb • 1.0 kb

0.5 kb • 0.5 kb

Figure 2.11. Restriction fragment produced from digestion of Listeria phages genomic DNA with three restriction endonucleases. A) EcoRI (lanes 1 to 3) and Hindlll (lanes 5 to 7). Lanes 1 and 5 are for AG8, lanes 2 and 6 are for AG 13 and lanes 3 and 7 are for AG20. B) Lane 1 is for AccI with AG20. Lanes M are for 1 kb ladder.

89 Table 2.10. Estimated size of genomic DNA of the finally selected phages based on the analysis of the PFGE results.

Phage DNA size (kb)

EcoM-AG2 168 EcoM-AG3 173 EcoM-AGlO 184 SboM-AG3 165 Sso.M-AG8 90 SsoM-AGlO 84 LinM-AG8 134 LmoM-AG13 133 LmoM-AG20 133 StyM-AG6 173 SenS-AGll 27 StyM-AG16 162

2.4.3.4 Effect of different temperatures and pH's on viability of the selected

phages

The selected phages were stored at different temperatures (-20, 4, 25 and 42°C) and pH (4.0, 5.0, 9.0 and 10.0) and the log unit reductions in their titre were determined by the overlay technique. Some phages kept their activity after being stored under extreme environments without a significant log unit reduction of the PFU/ml, while others were significantly affected by these environments. There was no significant reduction in the titre of all tested phages when stored at 4°C for 24 h. Salmonella phages

SenS-AGl 1 and StyM-AG16 were tolerant to pH 5.0, while only StyM-AG16 could also resist pH 9.0 with no significant log unit reduction in phage count (Figure 2.12). At pH

90 4.0 and pH 10.0 a significant reduction of the number of all tested Salmonella phages was observed, with log unit reduction ranging from around 0.5 to around 3.5 log units at pH

10.0 and from around 1 to around 3.5 log units at pH 4.0. SenS-AGll and StyM-AG16 phages could be stored at -20°C without any significant reduction in activity, while the

SenS-AG6 phage count was reduced by about 7.5 log unit at the same temperature.

Moreover, SenS-AGll and StyM-AG16 phage showed less than 0.5 log unit reductions in their counts at 42°C. The effect of 25°C was not significant on all Salmonella phages, however the StyM-AG6 titre was reduced by around one log unit at 37°C. All the tested

Listeria phages were sensitive to the examined acidic and alkaline pH with the exception of LmoM-AG20 phage, which was resistant to pH 9.0 and 10.0 with no significant reductions in its count (Figure 2.13). Freezing at -20°C caused a significant reduction in all tested phage counts of between 2.9 to 3.7 log units. Incubation at 25°C did not significantly affect any of the examined Listeria phages. For Listeria phages; LmoM-

AG13 and LmoM-AG20, phage counts were reduced by only 0.4 and 0.9 log units, respectively, when kept at 37°C, while AG8 did not show any significant reduction at the same temperature. LinM-AG8, LmoM-AG13 and LmoM-AG20 phages were sensitive to

42°C with count reductions of 5.6, 3.9 and 1.2 log units, respectively.

Highly significant reductions in the counts of tested Shigella phages (SboM-AG3,

SsoM-AG8 and SsoM-AGlO) have been observed when kept at pH 4.0 for 24 h (Figure

2.14). The log unit reductions were 8.9, 8.4, 6.7 log units, respectively. Less acidic pH

(pH 5.0) resulted in a lesser effect on these phages with count reductions ranging from around 0.5 to 1.2 log units. AG8 showed no significant reduction in the phage number at pH 9.0. SsoM-AGlO phage was the most resistant among the tested Shigella phages at

91 pH 10.0 with only a 0.7 log unit reduction in its titre. Freezing affected the count of

SboM-AG3 significantly but only resulted in around a 0.5 to 1.0 log unit reduction in

SsoM-AG8 and SsoM-AGlO phage numbers. Approximately a one log unit reduction in

SboM-AG3 and SsoM-AGlO phage numbers were observed when these phages were kept at 25 and 37°C, while SsoM-AG8 phage was stable at these two temperatures without any significant reduction in phage titre. All the three tested Shigella phages were sensitive to 42°C with reductions in activity ranging from around 1.5 to 3.3 log units. For

E. coli phages, the titre of EcoM-AG2 phage was reduced significantly at all the tested acidic and alkaline pH (Figure 2.15). EcoM-AG3 phage was resistant to pH 9.0 and 10.0, while EcoM-AGlO phage could only tolerate pH 9.0 with more than 2 log units reduction being found at pH 10.0. On the other hand, EcoM-AGlO phage was found to be tolerant to pH 5.0 with no significant reduction in its number. EcoM-AG3 phage was sensitive to acidic environments with 8.2 and 2.5 log unit reductions at pH 4.0 and pH 5.0, respectively. Although AG2 and AG3 phage were stable at 25°C, only AGIO titre was significantly reduced by around 1.8 log units at this temperature. With the exception of

AG3, 42°C caused a significant reduction to the tested E. coli phages by around 2 and 6.5 log units in phage counts of AG2 and AGIO phages, respectively. Keeping phages at -

20°G significantly reduced the titre of all the tested E. coli phages.

92 QStyM-AG6 BSenS-AG11 PStyM-AG16

pH 10.0 (-)20°C 25°C Treatment

Figure 2.12. Stability of Salmonella phages stored for 24 hours under different environmental conditions.

• Lit* LJLmoM-AG13 DLmoM-AG20

PH5.0 pH9.0 pH 10.0 (-)20°C 25°C Treatment

Figure 2.13. Stability of Listeria phages stored for 24 hours under different environmental conditions.

93 • SboM-AG3 • SSOM-AG8 DSsoM-AGIO

pH9 0 pH 10.0 (-)20°C 25°C Treatment

Figure 2.14. Stability of Shigella phages stored for 24 hours under different environmental conditions.

DECOM-AG2 • ECOM-AG3 • EcoM-AGIO

pH4.0 pH5.0 pH9.0 pH 10.0 (-)20°C 25°C 37°C 42°C 4°C Treatment Figure 2.15. Stability of E. coli phages stored for 24 hours under different environmental conditions.

94 2.4.3.5 Effect of different environmental conditions on the infectivity of the

selected phages

The infectivity of the selected isolated phages against their propagating host strains was examined under different environmental conditions by the spot test technique.

Different aeration conditions did not inhibit the lytic activity of the tested phages on their hosts, complete or partial lysis was detected (Table 2.11). However, zones showing turbidity with slight lysis were developed with EcoM-AG3 and EcoM-AGlO phages with

E. coli 0126:H8 and E. coli 0157:H7 as hosts, respectively, under MAP conditions. The infectivity of Shigella phages was not affected by any of the tested aeration conditions.

Interestingly, an anaerobic atmosphere enhances the intensity of the lytic area of StyM-

AG6 phage. Table 2.12 shows the effect of different pH, salinity and temperatures on the infectivity of all selected phages. At low pH (pH 4.5), LinM-AG8, SsoM-AGlO, EcoM-

AG3 and EcoM-AGlO phages were not able to infect their hosts. Moreover, EcoM-AG3 phage could not infect its host at pH 9.0. Overall, pH 4.5 and pH 9.0 decreased the lytic activity of all the tested phages. At 5% NaCl, E. coli phages resulted in either turbid lytic areas (EcoM-AG2 and EcoM-AG3 phages) or no lysis (EcoM-AGlO phage) on lawns of their respective hosts. Only Listeria spp. were able to grow at 10 % NaCl and the Listeria phages showed turbid lysis on lawns of their corresponding hosts. On the other hand, all the tested phages were able to infect their hosts at different incubation temperatures that support the growth of these bacteria.

2.4.3.6 Cross infection

The selected isolated phages were tested against genera other than those used for

95 propagation (Table 2.13). StyM-AG16 phage was able to infect and produce incomplete but marked lysis on E. coli 0157:H7 (C899). In addition, the three E. coli phages were able to infect Shigella sonnei (C866).

2.4.3.7 Emergence of bacteriophage insensitive mutants (BIM) and lysogenic

ability

The frequency of the development of BIM strains against the selected isolated phages was determined after incubation of single phages or phage cocktails with their propagating host at 25°C (Table 2.14). The developed colonies were further tested for

lysogenicity by the tested phages. All the tested bacterial strains were able to develop resistant mutants against either their specific single phages or cocktail of them, except the tested Shigella strains, which could not develop any colonies after incubation with

Shigella phages. EcoM-AG2 phage exhibited around 10-fold less BIM frequency when compared with EcoM-AG3 phage on E. coli 0126:H8 (C761) strain. However, a cocktail of the three E. coli phages caused a significant reduction of the BIM frequency of the tested E. coli 0157:H7 (C899) strain. StyM-AG6 phage caused significantly high BIM

frequency of Salmonella Typhimurium (CI 077) when compared to the effect of the other

Salmonella phages with their hosts. Using a cocktail of the three Salmonella phages

caused a significant reduction of the emergence of resistant mutants of the same

Salmonella strain when compared with those developed by using StyM-AG6 phage

alone. Interestingly, a low BIM frequency was observed for Listeria innocua (C505) with

LinM-AG8 phage. On the other hand, LmoM-AG13 and LmoM-AG20 phages when used

with L. monocytogenes (C391 and C519) resulted in around 3.22 x 10"4 and 4.62 x 10"4

CFU/ml of BIM, respectively. However, it is worth noting that there was no significant

96 difference between the BIM frequencies developed in L. monocytogenes (C391) after addition of LmoM-AG13 phage alone or in the presence of a combination of the three

Listeria phages. Induction with mitomycin C, heat shock or ultraviolet did not cause any release of prophages from any of the tested resistant colonies.

Table 2.11. Effect of different aeration condition on the infectivity of the selected phages

Aerobic Anaerobic MAP Vacuum Salmonella StyM-AG6 1+ 2+ 1+ 1+ SenS-AGll 2+ 2+ 1 + 2+ StyM-AG16 2+ 2+ 1 + L 2+ J Listeria LinM-AG8 1+ 1+ 1+* 2+ LmoM-AG13 2+ 2+ 2+ 2+ LmoM-AG20 2+ 1+ 2+ 1+ Shigella SboM-AG3 2+ 2+ 2+ 2+ SsoM-AGlO 2+ 2+ 2+ 2+ SsoM-AG8 2+ 2+ 2+ 2+ E.coli EcoM-AG2 1+ 1 + 1+ 1 + EcoM-AG3 2+ 1 + ± 1 + EcoM-AGlO 1+ 1 + ± 1 + 2+ = complete lysis ± = Turbid with slight lysis 1 + = incomplete but marked lysis * Weak bacterial growth

97 Table 2.12. Effect of pH, salinity and temperature on the infectivity of the selected phages

pH4.5 pH6 pH7 pH9 5%NaCl(a) 10%NaCl(a) 15%NaCl(a) 4°C(a) 10oC(a) 25°C Salmonella StyM-AG6 1+ 1+ 1+ 1+ 1+ NA NA 1+* 1 + 1+ SenS-AGll 2+ 2+ 2+ 1+ 2+ NA NA 2+* 2+ 2+ StyM-AG16 2+ 2+ 2+ 1+ 1+ NA NA 1+* 2+ 2+ Listeria LinM-AG8 0 ± 1+ ± 1+ ± NA 1+ 2+ 1+ LmoM-AG13 1+ 1+ 2+ 1+ 1+ ± NA 2+ 2+ 2+ LmoM-AG20 ± ± 2+ ± 1+ ± NA 2+ 2+ 2+ Shigella SboM-AG3 ±* 2+ 2+ 1+ 1+* NA NA NA 2+* 2+ SsoM-AGlO 0 2+ 2+ 2+ 1+ NA NA 1+* 2+* 2+ SsoM-AG8 ± 2+ 2+ 1+ 1+ NA NA 1+* 2+* 2+ Exoli EcoM-AG2 ± 1+ 1+ ± ± NA NA 1+ 1+ 1+ EcoM-AG3 0 ± 2+ 0 0 NA NA 1+ 2+ 2+ EcoM-AGlO 0 ± 1+ ± ± NA NA 1+* 1 + 1+ 2+ = complete lysis ± = Turbid with slight lysis NA = No bacterial growth 1+ = incomplete but marked lysis 0 = No lysis area * Weak bacterial growth

StyM- SenS- StyM- LinM- LmoM- LmoM- SboM- SsoM- SsoM- EcoM- EcoM- EcoM- AG6 AG11 AG16 AG8 AG13 AG20 AG3 AG10 AG10 AG2 AG3 AG10 Salmonella Typhimurium 0 0 0 0 0 0 0 0 0 (C1077) Salmonella Typhimurium 0 0 0 0 0 0 0 0 0 (C435) Salmonella Enteritidis 0 0 0 0 0 0 0 0 0 (C417) L. innocua 0 0 0 0 0 0 0 0 0 (C505) L. monocytogenes 0 0 0 0 0 0 0 0 0 (C391) L. monocytogenes 0 0 0 0 0 0 0 0 0 (519) Shigella boydii 0 0 0 0 0 0 0 0 0 (C865) Shigella sonnei 0 0 0 0 0 0 2+ 1+ 1+ (C866) E. coli 0157:H7 0 0 1+ 0 0 0 0 0 0 (C899) E. coli 0126:H8 0 0 0 0 0 0 0 0 0 (C761) 2+ = complete lysis 1 + = incomplete but marked lysis 0 = no lysis area Table 2.14. Bacteriophage insensitive mutant development of the selected phages and their cocktails on their propagating hosts.

Phage Host BIM frequency (CFU/ml) EcoM-AG2 £.co//0126:H8(C761) 6.41 x 10"5 EcoM-AG3 E. coli 0126:H8 (C761) 2.41 x 10"4 EcoM-AGlO E.coli 0157:H7(C899) 3.52 xlO'4 E. coli phage cocktail E.coli 0157:H7(C899) 4.75 x lO-5 (AG2+AG3+AG10) Salmonella Typhimurium StyM-AG6 1.46 xlO"2 (CI 077) Salmonella Enteritidis SenS-AGll 1.43 xlO"4 (C417) Salmonella Typhimurium StyM-AG16 1.89 xlO"4 (C435) Salmonella phage Salmonella Typhimurium cocktail 1.28 xlO"3 (CI 077) AG6+AG11+AG16 * SboM-AG3 Shigella boydii (C865-2) NA * SsoM-AG8 Shigella sonnei (C866-2) NA * SsoM-AGlO Shigella sonnei (C866-2) NA Shigella phage cocktail Shigella boydii (C865-2) NA* AG3+AG8+AG10 LinM-AG8 Listeria innocuoa (C505) 1.79 xlO"6 Listeria monocytogenes LmoM-AG13 3.22 xlO"4 (C391) Listeria monocytogenes LmoM-AG20 4.62 xlO-4 (C519) Listeria phage cocktail Listeria monocytogenes 3.11 xlO"4 AG8+AG13+AG20 (C391) * NA: No development of BIM colonies by direct plating.

100 2.5 DISCUSSION

Phages have emerged as a novel biotechnological approach to control bacterial contamination in food (Hagens and Loessner, 2007). Strong, stable, lytic phages are the obvious choice for food safety applications (Greer, 2005). The first step toward successful biocontrol of foodborne pathogens by phages is to isolate and characterize useful phage candidates for this process.

In this investigation, more than one hundred phages of different plaque morphologies and sizes targeting four foodborne pathogens, E. coli 0157:H7,

Salmonella, Listeria monocyogenes and Shigella, have been isolated from different environmental samples. This large number of isolates was consistent with the fact that phages are now acknowledged as the most abundant and diverse microorganisms on earth and can be isolated from environments where their host can grow (Kutter and

Sulakvelidze, 2005). However, it was anticipated, as more than one strain was used in the enrichment process, that some would be identical, as they would likely infect more than one of the bacterial hosts used and therefore be isolated from more than one overlay.

Therefore, all plaques from individual hosts were harvested, purified and tested by a turbidimetric method using the Bioscreen C. This automated turbidimetric approach provided both host range and quantitative lysis data simultaneously; allowing the determination of the host range pattern for each isolated phage. The isolated phages showed different abilities to infect the tested bacterial strains, which were signified by different degrees of lysis, which appeared as a delay or inhibition of the growth of the tested bacteria to reach the detection limit (OD = 0.3). Many of the isolated phages were

101 identified to have a broad host range while others were highly specific and infected only a limited number of the tested bacteria. The differences in susceptibility of the tested bacterial strains to the different isolated phages may be due to development of resistance mechanisms against these phages through abortive infection, variation in membrane phage receptors (adsorption blocking) and/or a restriction endonuclease modification system (Petty et al, 2007).

The Bioscreen C results have also revealed that some of the isolated phages in this study were identical in their host range patterns which might mean that they most likely represent multiple re-isolations of the same phage on different hosts. Some of these phages against Salmonella and Listeria species were selected for study by TEM and restriction endonuclease digestion pattern. The phages with similar host range patterns did not show an obvious difference in morphology and/or restriction endonuclease patterns with different enzymes. The Bioscreen C turbidimetric assay gave more rapid and useful quantitative lytic activity and host range information, making it an attractive and rapid alternative to agar-based overlay methods. Moreover, morphology can not differentiate between the isolated phages in most cases, so this approach also saved time in identification of the identical newly isolated phages without the need to perform more time-consuming studies such as restriction endonuclease analysis. Turbidity measurement of bacteria using the Bioscreen C has been used in several studies, including a study to determine of the effect of different phage MOI values on the growth of host bacteria

(Atterbury, 2007). However, it is the first time, to the best of our knowledge, that this technique has been applied to ascertain the host range of phages and to identify identical isolated phages.

102 In the selection of phages for potential use in biocontrol of the target pathogen, three broad host range phages for each targeted pathogen were selected for further study.

A broad host range is one of the most important criteria that should be present in lytic phages intended to be used for biocontrol. Felix 01 is a perfect example; it lyses 96-

99.5% of Salmonella serovars (Whichard et al, 2003). Narrow host range may present a problem for biocontrol purposes, as in some species there are numerous sub-types that all need to be controlled. However, narrow host range limitation could be overcome by using phage cocktails (Mclntyre et al, 2007).

The morphology of the selected phages revealed that eleven phages were among the family Myoviridae and only one belonged to the family Siphoviridae. In a recent survey, 96 % of the phages examined since 1959 were tailed and belonged to three families, the Myoviridae, Siphoviridae, and Podoviridae (Ackermann, 2009). However, of these phages 61% and 24.5%, respectively, were Siphoviridae and Myoviridae, which means that more phages with non-contractile were isolated. In our study, fewer

Siphoviridae phages were found among the selected phages than Myoviridae. They were morphologically related to five phage species; T4, Jersey, Vil, A511 and Sfv. Moreover, the Vil-like phage, SboM-AG3, was reported for the first time among Shigella phages

(H-W. Ackermann, personal communication). The one-step growth curve data for the five common morphotypes among the twelve selected phages revealed that some phages had high burst sizes and a long latent period such as LmoM-AG13 phage while others had a low burst size associated with shorter latent period. This is consistent with the suggestion that there is a correlation between the latent period and burst size, the delay in the lysis would result in increasing burst size due to the increase in the length of the

103 progeny-producing period (Abedon, 1989). On the other hand, it was suggested that short latent period phages have the advantage of being able to more quickly infect a greater number of cells than those possessing a longer latent period (Abedon, 1989, Lenski and

Levin, 1985). The selected phages gave distinct restriction digestion patterns with different enzymes which revealed that the genomic materials of all of the selected phages was comprised of dsDNA (Jamalludeen et al, 2007). The smallest genome size among the selected, isolated phages was for the Siphoviridae phage SenS-AGll which was

26.97 Kb. Similar values were detected during screening of Siphoviridae phages isolated from a marine environment and these were shown to have a genome size ranging from 20 to 55Kb (Steward et al, 2000).

Stability under different storage and application conditions is another important aspect that needs to be defined for successful use of phages in food and/or the environment (Strauch et al, 2007). The tested phages showed a great variation in their sensitivity to different temperatures and pH. All the selected isolated phages were stable at 4°C which is consistent with previous reports that described this temperature as the best storage temperature for phages (Brussow and Kutter, 2005, Ackermann, 2003).

Interestingly, only one of the tested phages, EcoM-AGlO phage, was not stable at room temperature and its count was reduced by almost two log units after being kept at 25°C for only 24 h. The tested higher temperatures, 37 and 42°C caused a significant reduction of most isolated phages, which might be due to the effect of these temperatures on phage proteins. Most tailed phages are inactivated at 60°C for 30 min (Ackermann, 2007).

However, three tested phages, EcoM-AG3, SenS-AGll and StyM-AG16, were resistant to high temperatures and therefore are suitable for application on food products and or

104 surfaces at temperatures higher than those used for cooling and/or chilling. Similarly, phage (p29 of Bacillus subtilis HSR was found to be active at both 37 and 42°C (Larcom and Thaker, 1977). Although some phages were resistant to freezing at -20°C, but as most phages are sensitive to freezing and thawing (Guttman et al, 2005), it was recommended that protectants such as glycerol or dimethyl sulphoxide (DMSO) should be added to maintain viability during freezing (Sambrook et al, 1989). However, in another study, no significant drop in titre was observed for A. phage stock held at -70°C in the absence of any protectant (Jepson and March, 2004).

All of the tested phages were highly sensitive to overnight exposure to pH 4.0.

Similar results were obtained when the stability of nine E. coli phages were tested in an acidic environment, when all phages were highly susceptible to pH from 1.0 to 4.0

(Jamalludeen et al, 2007). Acidic environments affect the infectivity of Bacillus phages and it was suggested that these conditions result in denaturation of phage proteins and consequently loss of phage viability (Hazem, 2002). Generally, it was reported that most tailed phages are stable at pH range from 5.0 to 9.0 (Ackermann, 2007). This was consistent with the results presented in this study. Moreover, EcoM-AG3 and LmoM-

AG20 phages were found to be resistant to a higher alkaline environment, pH 10.0, which would consequently increase the area of application of these phages.

Testing the ability of phages to infect their hosts under different environmental conditions is another important step to confirm that they can be used as biocontrol agents

(Hagens and Loessner, 2010). It was found in this study that, with few exceptions, whenever the host can grow, the phages were able to cause lysis of varying extent, which enhances the possibility of using these phages to control their susceptible hosts in a broad

105 range of food products manufactured and packed under different environmental conditions. Nevertheless, pH 4.5, pH 9.0 and 5 % NaCl resulted in inhibition of the infectivity of five phages. This might be due to the effect of these conditions on the adsorption of the phages on its host membrane and/or stability of these phages. It was reported that there was a reduction in activity of Listeria phages to control Listeria monocytogenes on apple slices due to the low pH of this food product that might affect the stability of the phages (Leverentz et al, 2003, Leverentz et al, 2004). In the present study, the selected Listeria phages were able to produce turbid but marked lysis of the tested Listeria monocytogenes at about 10 % NaCl concentration, indicating that these phages can be used to infect their hosts in high-salt food products that might be contaminated with Listeria monocytogenes.

In the current study, three E. coli phages and one Salmonella phage were considered as ambivalent by being able to infect, beside their original host bacteria,

Shigella sonnei (C866-2) and E. coli 0157:H7 (C899), respectively. With their broad host range criteria, these ambivalent phages would have the potential to be used for biocontrol to not only prevent growth of strains of their own host genus but also strains from another genus; presumably through interaction with similar phage receptors on both genera. Ambivalent phages were reported earlier when AR1 and LG1 E. coli phages were able to lyse Proteus mirabilis, Shigella dysenteriae, and two Salmonella strains

(Goodridge et al, 2003). Several phages (including phages U2 and LB related to T-even phages of E. coli) were reported to be able to grow both on E. coli K12 and on some

Salmonella strains (Krylov et al, 2006). Interestingly, no resistant mutants emerged against Shigella phages which suggests that these phages have the ability to overcome

106 different host resistant mechanisms. Although other tested bacteria were able to develop

BIM when treated with their infective phages, but for most of them the frequency of BIM was remarkably low and should not hinder the use of these phages as biocontrol agents.

Previous report on the characteristics of BIM which developed against E. coli 0157:H7 phages showed that they had a smaller, more coccoid cellular morphology than the parental strain and moreover they reverted to phage sensitivity within 50 generations

(O'Flynn et al, 2004). In addition to that, it was claimed that phages can evolve in the same way as bacteria to overcome the different resistant mechanisms that can be developed in their bacterial hosts (Petty et al, 2007). In a recent review, various phages have been reported to counter-attack and circumvent these resistance mechanisms by different strategies in order to thrive in most environments (Labrie et al, 2010). In addition, using a phage cocktail could help to significantly reduce the frequency of emergence of BIM as appeared in the current study with the tested E. coli 0157:H7 and

Salmonella Typhimurium strains. Previous reports showed that using phage cocktails resulted in significant reductions in some foodborne pathogens such as Listeria monocytogenes, Salmonella and E. coli 0157:H7 (O'Flynn et al, 2004, Leverentz et al,

2004, Fiorentin et al, 2005).

In conclusion, a number of strong, stable, lytic phages have been isolated and characterized that can be considered as good candidates for biocontrol purposes in food applications. A fast and reliable high throughput assay indicating the specificity and lytic activity of the isolated phages was developed for the first time using the Bioscreen C to select phages possessing the broadest host range and identification of identical isolates.

One novel Vil-like phage was reported for the first time to infected Shigella spp. The

107 selected phages against each targeted pathogen showed different host range patterns and

stability criteria, so using cocktails of these phages would increase the area of application

with different food products, with different packaging conditions and control a broader range of the targeted pathogens. Future work is still needed to optimize the phage

cocktail concentration that can be used to cause complete inhibition of the targeted pathogens and optimize their use in both artificial media and real food systems.

108 Chapter 3: SEQUENCING AND GENOME ANALYSIS OF

SHIGELLA PHAGE (

3.1 ABSTRACT

Sequencing of bacteriophage genomes is considered as an important preliminary step to ensure their safety prior to food applications. The lytic bacteriophage, OSboM-

AG3 (or AG3), targeting the important foodborne pathogen, Shigella, was isolated from sewage and found to be active against Shigella boydii. It was morphologically similar to phage Vil of Salmonella enterica serovar Typhi and a series of phages of Vibrio cholerae and Rhizobium meliloti. This is the first time that a Vi-specific phage for Shigella has been described. The complete genome of OSboM-AG3 was obtained using 454 pyrosequencing and was analyzed. The genome consisted of 158,006 bp with a G + C content of 50.4% and was terminally redundant and circularly permuted. Two hundred and sixty potential open reading frames (ORFs) were identified and annotated, but only

107 possessed homologs in the current NCBI non-redundant database. Fifty six of these proteins showed homology to similar proteins from members of the Teequatrovirinae.

The AG3 genome included four tRNA genes, eight putative rho-independent terminators and eight putative promoters. There was no evidence for the presence of virulence and lysogenic-associated genes. In conclusion, the genome analysis of OSboM-AG3 revealed the possibility of safe application of this phage for biocontrol purposes.

109 3.2 INTRODUCTION

Shigella species sonnei, flexneri, and boydii are generally considered among the

most important foodborne pathogens (Kothary and Babu, 2001, Sutherland and Varnam,

2002). Ingestion of S7i/ge//a-contaminated food causes shigellosis; a disease that develops

within 12-48 hours. Fever, aches, fatigue and loss of appetite are the initial symptoms,

which may be associated with watery diarrhea that, in turn, may develop into bloody

stools or dysentery. Fatal hemolytic-uremic syndrome (HUS) may also develop in certain

severe cases, due to the production of Shiga toxin (Acheson, 2001). Shigella-related

outbreaks occur through direct or indirect human fecal contamination and have been reported in both developed and developing countries wherever poor hygiene standards

occur (Sivapalasingam et cil, 2006). Food products such as salads, soft cheese, vegetables

and meat products are usually reported as being associated with these outbreaks

(Acheson, 2001).

Lytic phages have been applied successfully to control the growth of various

foodborne pathogens including Shigella (Hagens and Loessner, 2010). They are able to

attach to bacteria and integrate into their metabolic system, while utilizing their host's resources to reproduce and the release of new phage leads to lysis of the bacterial cell

(Kutter and Sulakvelidze, 2005). As phages are becoming recognized as fascinating tools to control pathogens in food, phage genomics will play an increasingly important role to ensure that potentially harmful phages are avoided or re-designed without undesirable features (Garcia et al, 2008). For instance, genomic data could ensure the absence of virulence genes and any genes that might lead to lysogenization of a pathogen (Strauch et

110 al, 2007). As DNA sequencing methodologies advance, the number of sequenced phage genomes has increased exponentially (Clokie and Kropinski, 2009). To date, 579 phage sequences are available in the NCBI database, which has allowed detailed comparative analyses to be performed in order to obtain more coherent classification schemes and provide insights into evolutionary processes (Ackermann, 2003, Lavigne et al, 2009, The

National Center for Biotechnology Information (NCBI), 2010, Briissow and Hendrix,

2002, Hendrix et al, 2003, Lavigne et al, 2008).

One of the important phage groups are those phages that target Vi or virulence- associated polysaccharide capsular antigen as a receptor (Pickard et al, 2008). Those viruses, which infect Salmonella Typhi, are distinguished into seven types (Vi-I to Vi-

VII) and are used for phage typing for this bacterium. All Vi phages have isometric heads and either contractile, or long or short noncontractile tails. As a result, they belong to the

Myoviridae (Vi-I), Siphoviridae (Vi-II), and Podoviridae (Vi-III to VII), respectively

(Ackermann, 1970). Phages belonging to Vi-I have a quite characteristic morphology that was reported also in Vibrio cholerae and Rhizobium meliloti viruses (Ackermann, 1970,

Werquin et al, 1988, Guidolin et al, 1984). Phages of this group possess a neck, a collar and a contractile tail. They have a thin baseplate, which is connected to a very ramified structure consisting of either short tail fibers or spikes. Vi-II phages are lambda-like but have a base plate with three spikes, while Vi-III to VII viruses were found to have a very short tail terminating in a base plate with at least two spikes (Ackermann, 1970).

Although Vi phage were first described over 60 years ago, little is known about their molecular structure (Pickard et al, 2008). To the best of our knowledge, Vil has not been sequenced or reported to infect Shigella. Moreover, it probably represents a novel genus

111 and is thus taxonomically interesting (H. W. Ackermann, personal communication).

According to the current classification of phages by the International Committee on

Taxonomy of Viruses ICTV, tailed phages belong to the order Caudovirales, which contains three families; Myoviridae (long, contractile tails; 24.5%), Siphoviridae (long, non-contractile tails; 61%) and Podoviridae (short, non-contractile tails; 14%)

(Ackermann, 2007). Sequencing of phages and analysis of the sequence data will help to improve phage classification. Recently, the proteomic analysis of 102 Myoviridae phages revealed the establishment of three new subfamilies; Peduovirinae, Teequatrovirinae and

Spounavirinae and new eight independent genera (Lavigne et al, 2009).

In the previous chapter we isolated three strongly lytic, phages that could be considered as good candidates for Shigella control. One of them, OSboM-AG3 (or AG3), was morphologically similar to the Vi-I phages. Therefore, the objectives of this study were to determine the genomic sequence of this phage, decipher the sequence data and assess its structural functions in relation to other existing phages in the GenBank. This would aid in a better understanding of the molecular structure of this phage and the interesting features that might be useful in its taxonomical classification and application for Shigella control.

3.3 MATEREIALS AND METHODS

3.3.1 Bacteria and Bacteriophage

Tryptic Soy Broth (TSB), Tryptic Soy Agar (TSA), and Tryptose Soft Agar

(TSB+ 0.4% agar) (Difco Laboratories, Detroit, MI) were used to grow the host bacteria and to propagate the phage. The bacterial strain Shigella boydii C865-2 was used in this

112 study for phage propagation (Canadian Research Institute of Food Safety, University of

Guelph, ON, Canada) according to the same technique mentioned in chapter 2. OSboM-

AG3 was previously isolated from sewage samples from a local wastewater treatment plant (Guelph, ON) and characterized using a variety of techniques, including transmission electron microscopy (TEM). For genomic sequencing, OSboM-AG3 lysate was prepared in liquid culture and syringe filtered using 0.45-um-pore-size surfactant- free cellulose acetate filters (Millipore, Nepean, ON).

3.3.2 Phage purification, DNA isolation and sequencing

DNA extraction from purified phage was performed by a method modified from

Sambrook et al 1989. Briefly, crude phage lysate was clarified from bacterial debris by centrifugation at 14,000 x g for 20 min at 4°C (Beckman Coulter Inc., Mississauga, ON,

Canada). Nucleic acids in the supernatant were digested with pancreatic DNase 1, and

RNase, each to a final concentration of lOug/ml (Sigma-Aldrich, Oakville, ON) and phage particles were precipitated in the presence 10% w/v (final concentration) PEG-

8000 and 1 M NaCl at 4°C overnight. The precipitated phage particles were recovered by centrifugation at 14000 x g for 20 min at 4°C, and then resuspended in TM buffer (10 mM Tris-HCl, pH 7.8, 1 raM MgS04). The phage suspension was purified by separation on a self-generating cesium chloride (CsCl) gradient (1.5 g/ml CsCl, run at 21,000 x g at

4°C for 24 h) in a fixed angle, Beckman SW 90Ti rotor. Following purification by a second passage through a CsCl gradient for another 24 hours, the phage was dialyzed against two changes of 2 L each of 1 x lOmM TE buffer (pH 8.0), using dialysis cassettes at 3500 MWCO (Thermo Scientific, Fisher Scientific, Mississauga, ON), and stored as a high-titer purified phage stock at 4°C until use.

113 To extract the DNA, 0.5M EDTA (pH 8.0) was added to purified phage lysate to a final concentration of 20mM. Proteinase K (Sigma-Aldrich, Oakville, ON) and 10%

(W/V) SDS solution at a final concentration of 50 |Jg/ml and 0.5% W/V, respectively, were added to lyse phage protein coats. The mixture was incubated at 56°C from one hour and up to overnight until the mixture gained viscosity and appeared fluid upon inversion. After incubation, phage nucleic acid was purified in a three-stage organic extraction by mixing phage lysate with an equal volume of phenol:chloroform:isoamyl alcohol (25:24:1, V/V). This was followed by centrifugation for 5 min at 13,000 x g. The collected DNA in the aqueous phase DNA was precipitated by adding sodium acetate to a final concentration of 0.3 M and adding two volumes of 100% ethanol. The precipitated

DNA was collected by centrifugation at 15,000 x g for 15 min at 4°C, washed with 70% ethanol, and air dried before being resuspended in 10 mM Tris-HCl (pH 7.5) and stored at 20°C. DNA concentration was determined by sample absorbance at 260 nm (DU640 spectrophotometer; Beckman Coulter, Mississauga, ON, Canada).

DNA of AG3 was subjected to pyrosequencing (454 technology) at the McGill

University and Genome Quebec Innovation Centre (Montreal, QC, Canada).

3.3.3 Genome annotation

Prior to annotation, the genome was opened immediately upstream of the rllA gene so that it could be directly analyzed and compared with the sequence of other related phages. The genome was annotated with the help of Dr. Andrew M. Kropinski

(Laboratory for Foodborne Zoonoses, Guelph, ON) using a variety of online tools

(http://molbiol-tools.ca). The GC content and presence of direct repeats in AG3 genome

114 were analyzed using DNAMAN (Lynnon Corp., Vaudreuil-Dorion, QC, Canada). The genome was initially subjected to automated annotation using AutoFACT (Koski et al,

2005), following which all open reading frames (ORFs) were confirmed using Kodon total genome and sequence analysis software, version 2.0 (Applied Maths Inc., TX).

Genes were identified from among the predicted coding sequences (CDSs) based on the presence of ATG, GTG, CTG or TTG start codons, followed by at least 30 additional codons, and an upstream sequence resembling the following ribosome-binding site,

GGAGGT (Shine and Dalgarno, 1974, Shine and Dalgarno, 1975). The BLASTP algorithm was used to determine the similarity to described proteins in the global database (available through the National Center for Biotechnology Information [NCBI] at http://www.ncbi.nlm.nih.gov. Phage-encoded tRNA genes were identified with Aragorn, using the default parameters (Laslett and Canback, 2004). DNAMAN was used to determine the codon usage information of both AG3 and its bacterial host Shigella boydii.

Promoters were identified based on sequence homology to the consensus E. coli promoter, TTG AC A (Nis-ig) TATAAT, immediately upstream of an annotated gene

(Ermolaeva et al, 2000). Rho-independent terminators were discovered by examining the secondary structure of the DNA adjacent to polyT sequences using MFOLD (Zuker,

2003). Genomic comparisons at the proteomic level were made using CoreGenes (Zafar et al, 2002, Kropinski et al, 2009). Transmembrane domains were predicted using

TMHMM v2.0 and Phobius (Krogh et al, 2001, Kail et al, 2007).

3.3.4 Proteome analyses

Phage proteins were analyzed using standard methods (Sambrook et al, 1989).

The intact phage particles were lysed using Laemmli sample buffer (4% SDS, 20%

115 glycerol, 10% 2-mercaptoethanol, 0.004% bromphenol blue, 0.125 M Tris HC1) and boiled for 5 min. The solubilized proteins were subsequently separated by a 12.5% SDS- polyacrylamide gel electrophoresis, and stained with SimplyBlue SafeStain (Invitrogen

Canada, Burlington, ON). The gel data were analyzed using BioNumerics software

(Applied Maths). The six most intense phage bands were excised and subjected to mass spectrometric analysis at the Mass Spectrometry Facility, by Dr. Yi-Min She, Queen's

University (Kingston, ON, Canada).

3.3.5 Genome sequence

The annotated genome sequence for the Shigella phage

3.4 RESULTS

3.4.1 General features of the 3>SboM-AG3 genome

The phage genome sequence was received as a single contig representing the consensus from 19 X coverage. It contains apparently no frameshifts as indicated by

BLASTX analysis and is 158,006 bp long with a G+C content of 50.4 %. Since pulsed- field gel electrophoresis indicated that the phage genome is a single double-stranded linear DNA molecule of approximately 164.99 kb (Chapter 2), the sequence data suggests that this phage possesses terminally redundant termini. As with other phages containing rft4i?-encoding genes, the AG3 genome was opened just upstream of the rllA homolog for annotation purposes.

116 3.4.2 Identification and analysis of open reading frames (ORFs)

The genome was analyzed using AutoFACT complemented by Kodon and

BLAST analyses. A total of 260 putative ORFs were identified in the genome (Figure

3.1; Table 3.1). A total of 146,356 nucleotides (92% of the genome) were involved in

coding for putative proteins. Of these, 107 ORFs were identified as possessing homologues with proteins in the nonredundant NCBI database, while the rest were

considered to be hypothetical proteins unique to this phage. Bacterial toxin and integrase

homologs were absent in the AG3 genome. Four different start codons were used; ATG,

GTG, CTG and TTG at frequencies of 95.4%, 3.2%, 0.9% and 0.5%, respectively. More

detailed analysis using CoreGenes shows that this phage shares 46 homologs with phage

P-SSM2, 57 with T4, 60 with Aehl, and interestingly 69 with Delftia acidovorans phage

OW-14. Table 3.2 shows the 36 proteins which are conserved in Enterobacteria phages

T4 (NC_000866), RB49 (NC_005066), RB43 (NC_007023), Aeromonas phage 44RR2.8t

(NC_005135) and Prochlorococcus phage P-SSM2 (NC_006883). Of these, 31 were

identified in OSboM-AG3 using CoreGenes parameters (default BLAST score = 75), while four more were tentatively identified using PSI-BLAST (Altschul and Koonin,

1998).

Packaging and morphogenesis homologs. Sequence-based predictions identified many genes as being involved in AG3 packaging and morphogenesis. Terminase large

and small subunit homologs were determined to be gene 196 and 198. Genes for major head, prohead core, portal vertex proteins'were identified as genes 185, 186 and 191, respectively. In addition, genes for prohead protease and head completion protein were identified as genes 188 and 36, respectively. Genes 37 and 84 were for baseplate

117 assembly while, genes 210 and 212 showed sequence similarity to tail spike proteins. Tail tube and base plate initiator gene was identified to be gene 73. Genes 192, 195 and 199 were predicted to be genes for tail sheath and tube. Gene 6 was considered to be similar to a putative tail protein of Enterobacteria phage N4.

Nucleotide metabolism and DNA replication. Several genes encoding proteins directly involved in nucleotide metabolism were identified in the AG3 genome: exonuclease (gene 26), putative thymidylate synthase (gene 66), alpha and beta subunits of ribonucleotide reductase (gene 76) and putative nicotinamide phosphoribosyl transferase (gene 149). No similarities with kinase and or deoxyribonuclease were detected. More gene similarity was detected with DNA replication genes in the database: a DNA ligase (gene 45), a primase/helicase (gene 59), primase subunits (gene 102), DNA polymerase accessory proteins (genes 163, 164 and 165), DNA polymerase (gene 236) and RNA-DNA and DNA-DNA helicase (gene 168).

118 • CDS • GC content • CC skew+ • CC skew-

Figure 3.1. Genetic and physical map of phage OSboM-AG3 prepared using DNAPlotter (Carver et al, 2009) with proteins that shows homology with T4 subfamily indicated in the outer rim.

119 Table 3.1. General features of putative ORFs of

Similarities/homologies to Coordinates Length ORF Strand Predicted function genes or gene (bp) (bp) products of other phages NP_899390.1 rllA [Vibrio orflOOl 1..2757 2757 + R1IA protein phage KVP401 NP_899391.1 rllB Protector orf002 2789..4351 1563 + R11B protein from prophage-induced early lysis [Vibrio phage K.VP40] orfl)03 4403..4708 306 + hypothetical protein orf004 4677..5102 426 + hypothetical protein conserved hypothetical YP717859.1 gpl89 [Synech- orf005 5132..5524 393 + protein ococcus cyanophage syn9] YP950495.1 32 kDa protein orf006 5503..6312 810 + putative tail protein [Enterobacteria phage N4] orfl)08 6315..6545 231 + hypothetical protein NP046614.1 histone-like putative histone-like prokaryotic DNA-binding orf009 6634..7146 513 + protein protein family [Bacillus phage SPBc2] hypothetical membrane orf009.1 7196..7393 198 + protein orfOlO 7436.J924 489 + hypothetical protein NP943884.1 hypothetical conserved hypothetical orfOll 7921..8502 582 + protein Aehlp006 protein [Aeromonas phage Aehl] YP656238.1 gp39plus60 topoisomerase 11, large orf014 8552..10465 1914 + DNA topoisomerase II large subunit subunit [Aeromonas phage 25] YP_053029.1 hypothetical conserved hypothetical orfl)15 10467..10913 447 + protein VP5_gp35 [Vibrio protein phage VP51

120 C-terminus of YP001497977.1 gp52 topoisomerase II, hypothetical protein orf017 10906.. 12246 1341 + medium subunit NY2AB781L [Paramecium bursaria Chlorella virus NY2A1 hypothetical membrane orfl)18 12289..12576 288 + protein ABY40532.1 putative conserved hypothetical anaerobic dehydrogenase orf020 12579.. 12890 312 + protein [Burkholderia phage Bups Phill NP049871.1 Arn.3 Arn.3 conserved conserved hypothetical orf021 12894.. 13304 411 + hypothetical protein protein [Enterobacteria phage T41 orfl)22 13355..13597 243 + hypothetical protein YP002003706.1 Tk.4 conserved hypothetical orf024 13594..14193 600 + Tk.4 protein protein [Escherichia phage rv51 NP943887.1 DexA orf026 14194..14826 633 + DexA exonuclease exonuclease A [Aeromonas phage Aehl] YP224039.1 hypothetical conserved hypothetical orf027 14823..15128 306 + protein BPKS7gp 19 membrane protein [Salmonella phage KS7] orf028 15193..15534 342 + hypothetical protein YP006862.1 putative putative serine/threonine protein orf029 15534.. 16094 561 + serine/threonine phosphatase [Enterobacteria protein phosphatase phage T5] conserved hypothetical YP001468615.1 gp45 orf030 16129..16542 414 + protein [Listeria phage A511 ] orf031 16539..16877 339 + hypothetical protein

121 orfl)32 16948..17262 315 + hypothetical protein Cd, allosteric enzyme activated by HM-dCTP NP_932553.1 Cd orf033 17262.. 17768 507 + and dCTP for dUMP [Aeromonas phage 44RR2.8t] synthesis hypothetical membrane orf034 17779.. 18186 408 + protein YP512307.1 hypothetical conserved hypothetical protein PhiV10p52 orf035 18189..18401 213 + protein [Enterobacteria phage phiVlOl YP_717682.1 head Gp4, head completion completion protein orft)36 18398..19012 615 - protein [Synechococcus cyanophage syn9] YP195236.1 putative tail Gp48, putative tail tube associated base plate orfl)37 19063..20031 969 + tube associated base protein, gp48 [Synechococcus plate protein phage S-PM2] NP891704.1 baseplate Gp53, baseplate wedge orfl)38 20043..20597 555 + wedge subunit [Enterobacteria subunit phage RB49] YP_214250.1 hypothetical conserved hypothetical protein PSSM2018 orfD39 20594..21982 1389 + protein [Prochlorococcus phage P- SSM21 orfD40 21994..23940 1947 + hypothetical protein YP_214238.1 T4-like loader Gp59, T4-like loader of gp41 DNA helicase orfl)41 23941..24606 666 - ofgp41 DNA helicase [Prochlorococcus phage P- SSM2] NP_943926.1 MobE homing putative homing orf042 24599..25294 696 - endonuclease [Aeromonas endonuclease phage Aehl] orf043 25291..25629 339 . hypothetical protein

122 hypothetical membrane orf044.1 25595..25852 258 - protein hypothetical membrane orfl)44 25854..26087 234 - protein NP_899305.1 DNA ligase orf045 26056..27522 1467 - gp30 DNA ligase [Vibrio phage KVP401 hypothetical membrane orfl)46 27578..27730 153 - protein hypothetical membrane orf047.1 27727..27894 168 - protein hypothetical membrane orf047 27891..28076 186 - protein hypothetical membrane orf048 28076..28618 543 - protein e.6 conserved NP_932411.1 e.6 orf049 28661..29269 609 - hypothetical protein [Aeromonas phage 44RR2.8t] orf050 29285..29626 342 _ hypothetical protein orf051 29681..30031 351 _ hypothetical protein orf053 30033..30245 213 _ hypothetical protein hypothetical membrane orf054 30245..30361 117 - protein NP899614.1 hypothetical conserved hypothetical orfl)55 30358..31551 1194 - protein KVP40.0369 [Vibrio protein phage KVP40] orf057 31679..31999 321 hypothetical protein orrQ58 32015..32350 336 _ hypothetical protein YP_ 195170.1 DNA primase- Gp41, DNA primase- helicase subunit gp41 orf059 32413..33840 1428 - helicase subunit [Synechococcus phage S- PM2] orfD60 33847..34176 330 _ hypothetical protein NP_861734.1 UvsX RecA- UvsX RecA-like orf061 34154..35239 1086 - like recombination protein recombination protein [Enterobacteria phage RB69] orf062 35224..35766 543 _ hypothetical protein

123 ZPJ)0369821.1 probable putative dUTP dUTP diphosphatase Cj 1451 orf063 35766..36320 555 - diphosphatase [Campylobacter upsaliensis RM31951 orfi%5 36317..36886 570 _ hypothetical protein putative thymidylate YP223948.1 thymidylate orfl)66 36883.37929 1047 - synthase synthase [Phage phiJLOOl] YP001595836.1 conserved hypothetical orfi)67 37929..38591 663 - hypothetical protein protein [Pseudomonas phage YuA] orf069 38666.39532 867 _ hypothetical protein orf070 39709.39996 288 _ hypothetical protein orfl)71 40017..40766 750 _ hypothetical protein YP001469476.1 gp2 DNA Gp2 DNA end orft)72 40829..41533 705 - end protector protein protector protein [Enterobacteria phage Phi 1 ] NP944078.1 gp54 baseplate Gp54 baseplate tail orf073 41587..42531 945 + tail tube initiator [Aeromonas tube initiator phage Aehl] YP214565.1 gp32 Gp32 T4-like ssDNA orf074 42558..43589 1032 - [Prochlorococcus phage P- binding protein SSM4] orfl)75 43688.-43927 240 _ hypothetical protein Gp33 T4-like late YP214562.1 gp33 orfD76 43935..44180 246 - promoter transcription [Prochlorococcus phage P- accessory protein SSM41 orf077 44173..44298 126 _ hypothetical protein hypothetical membrane orfl)78 44405..44716 312 - protein conserved hypothetical YP002003 844.1 gp5 orfD79 44716..45306 591 - protein [Mycobacterium phage Nigel] orfl)80 45358..45849 492 _ hypothetical protein orfD81 45827..46351 525 _ hypothetical protein Gp26 baseplate hub NP944077.1 gp26 baseplate orf082 46551..47207 657 + subunit hub subunit [Aeromonas

124 phage Aehl] YP001595278.1 gp5 Gp5 baseplate hub baseplate hub subunit and tail orf084 47718..49328 1611 + subunit and tail lysozyme [Enterobacteria lysozyme phage JS98] YP_195113.1 baseplate putative Gp25, wedge subunit gp25 orf085 49401..49781 381 + baseplate wedge [Synechococcus phage S- subunit PM2] orf086 49778..50104 327 _ hypothetical protein orf087 50134..50595 462 _ hypothetical protein NP944996.1 Putative phage orf088 50701..50925 225 - putatitve glutaredoxin glutaredoxin [Enterobacteria phage Felix 01] YP 717864.1 ribonucleotide NrdB, ribonucleotide reductase subunit B orf090 50935..52038 1104 - reductase subunit beta [Synechococcus cyanophage syn9] orfD91 52040..52333 294 _ hypothetical protein NrdA, ribonucleoside- YP214439.1 NrdA orf093 52404..54683 2280 - diphosphate reductase, [Prochlorococcus phage P- alpha subunit SSM21 NP_597931.1 PhoH-like orf095 54769..55611 843 - PhoH-like protein protein [Roseobacter phage SlOll YP802996.1 putative site- putative homing specific intron-like DNA orfD96 55663..56355 693 - endonuclease endonuclease [Enterobacteria phage RB32] YPJ)01111074.1 gp43, conserved hypothetical bacteriophage-acquired orf097 56442..57236 795 - protein protein [Burkholderia phage phiE202] orf098 57306..57605 300 _ hypothetical protein orfl)99 57602..57790 189 . conserved hypothetical YP 001469590.1

125 protein hypothetical protein philp247 [Enterobacteria phage 011] orflOO 57845..58333 489 _ hypothetical protein orflOl 58529..58735 207 _ hypothetical protein YP_717862.1 DNA primase Gp61 DN A primase orfl02 58732..59790 1059 - subunit [Synechococcus subunit cyanophage syn9] CAE 14763.1 unnamed conserved hypothetical protein product [Leptospira orfl03 59790..60401 612 - protein biflexa temperate bacteriophage LE11 NP891656.1 hypothetical MobD.6 conserved orfl04 60459..60812 354 - protein RB49p085 hypothetical protein [Enterobacteria phage RB49] orfl05 60863..61102 240 _ hypothetical protein orfl07 61112..61681 570 _ hypothetical protein orfl09 61739..63460 1722 _ hypothetical protein orfllO.l 63481..63621 141 . hypothetical protein orfllO 63627..64118 492 _ hypothetical protein orf111 64194..64529 336 _ hypothetical protein orfll2 64552..64827 276 _ hypothetical protein orfll3 64935..65573 639 _ hypothetical protein NP944779.1 Phage Vs.l conserved conserved protein orfll4 65573.-65962 390 - hypothetical protein [Enterobacteria phage Felix oil orfll4.1 66087..66221 135 _ hypothetical protein orf 115 66287..66565 279 _ hypothetical protein NP049846.1 NrdA.l NrdA.l conserved conserved hypothetical orfll6 66546..66875 330 - hypothetical protein protein [Enterobacteria phage T4] hypothetical membrane orfll7 66872..67036 165 - protein orfll9 67039..69375 2337 . Gp46 recombination AAR90909.1 gp46

126 endonuclease subunit recombination endonuclease subunit [Aeromonas phage 65] YP_214377.1 T4-like Gp47 recombination endonuclease orf 121 69377..70492 1116 - protein subunit [Prochlorococcus phage P- SSM21 YP_214376.1 T4-like sigma Gp55, T4-like sigma factor, late transcription orfl22 70479..71270 792 - factor involved in late [Prochlorococcus phage P- transcription SSM21 YP_418157.1 putative putative Ribonuclease orf 123 71282..71722 441 - Ribonuclease HI HI [Pseudomonas phage EL] orf 124 71855..72622 768 + hypothetical protein NP_542515.1 ATP- putative ATP- orf 125 72619..74337 1719 - dependent helicase dependent helicase [Halorubrum phage HF2] YP001429759.1 DNA- putative DNA-binding orf 127 74466..74744 279 - binding HU protein [Bacillus protein phage 0305phi8-36] orf12 8 74838..75086 249 _ hypothetical protein orfl29 75083..75361 279 _ hypothetical protein orfl30 75363..76046 684 _ hypothetical protein orfl31 76104..77009 906 _ hypothetical protein orf 132 77061..77375 315 _ hypothetical protein orf 133 77375..78061 687 _ hypothetical protein YPJ64323.1 hypothetical conserved hypothetical orfl34 78141..78590 450 - protein F116p59 protein [Pseudomonas phage Fl 161 hypothetical membrane orf 135 78624..78782 159 - protein orf 136 78784..79071 288 _ hypothetical protein orf 137 79068..79442 375 _ hypothetical protein orf 138 79502..79741 240 _ hypothetical protein orf 139 79838..80122 285 _ hypothetical protein

127 NP_817639.1 gp50 conserved hypothetical orfl41 80224..80550 327 - [Mycobacterium phage Bxz2] protein (using PSI Blast iteration 2) YP239095.1 hypothetical conserved hypothetical orfl44 80593..82794 2202 - protein RB430RF 119c protein [Enterobacteria phage RB43] orfl46 82787..82957 171 _ hypothetical protein orfl47 82959..83318 360 _ hypothetical protein YP024539.1 putative putative nicotinamide nicotinamide phosphoribosyl orfl49 83361..85004 1644 - phosphoribosyl transferase [Staphylococcus transferase phage K] conserved hypothetical YP_950526.1 gp48 orfl52 85115..85909 795 - protein [Enterobacteria phage N4] YP_239311.1 hypothetical conserved hypothetical orfl53 85890..86291 402 - protein XPXV15_gp43 protein [Xanthomonas phage Xp 15] YP001949934.1 conserved hypothetical hypothetical protein orfl55 86352..86723 372 - protein RSLl_gp059 [Ralstonia phage RSL1] YP_001949933.1 conserved hypothetical hypothetical protein orfl58 86768..88786 2019 - protein RSLl_gp058 [Ralstonia phage RSL1] YP001950110.1 conserved hypothetical hypothetical protein orfl60 88902..90122 1221 - protein RSLl_gp235 [Ralstonia phage RSL1] NP944984.1 Phage conserved hypothetical hypothetical protein orfl61 90192..91061 870 - protein [Enterobacteria phage Felix on RegA translational YP_214692.1 RegA orfl62 91078..91542 465 - repressor protein [Prochlorococcus phage P-

128 SSM4]

Gp62 clamp loader YP717829.1 clamp loader subunit, DNA orfl63 91572..91994 423 - subunit [Synechococcus polymerase accessory cyanophage syn9] protein Gp44 clamp loader NP861749.1 gp44 clamp subunit, DNA loader subunit, DNA orfl64 91999..92988 990 - polymerase accessory polymerase accessory protein protein [Enterobacteria phage RB69] Gp45 sliding clamp, YP214389.1 T4-like sliding orfl65 93070..93738 669 - DNA polymerase clamp [Prochlorococcus accessory protein phage P-SSM2] orfl66 94079..94456 378 + hypothetical protein YP214374.1 UvsW RNA-DNA and DNA- orfl68 94447..95955 1509 - [Prochlorococcus phage P- DNA helicase UvsW SSM21 YP 195231.1 putative conserved hypothetical orfl69 95984..96730 747 - exonuclease [Synechococcus protein phage S-PM2] putative DNA YP214674.1 UvsY orfl70 96730..97182 453 - repair/recombination [Prochlorococcus phage P- protein UvsY SSM41 YP001595274.1 gp3 tail Gp3 tail completion completion and sheath orfl71 97225.-97725 501 - and sheath stabilizer stabilizer protein protein [Enterobacteria phage JS98] orfl72 97755..98405 651 + hypothetical protein NP944879.1 Phage conserved hypothetical hypothetical protein orfl74 98407..99135 729 - protein [Enterobacteria phage Felix on orfl75 99174..99341 168 _ hypothetical protein NP_899401.1 conserved conserved hypothetical orfl76 99383..99805 423 - hypothetical protein [ Vibrio protein phage KVP40]

129 orfl77 99811..100131 321 . hypothetical protein orfl78 100191..100472 282 _ hypothetical protein orfl79 100575..100811 237 _ hypothetical protein orfl80 100820.. 101260 441 , hypothetical protein orfl81 101273..101500 228 _ hypothetical protein orfl82 101561..101761 201 _ hypothetical protein orfl83 101821..102153 333 _ hypothetical protein orfl84 102246..102473 228 _ hypothetical protein YP_003358893.1 gp23 Deftia Gp23, major head phage phiW-14]; Motif: orfl85 102566..103888 1323 - protein PHA02541, major capsid protein NP_899608.1 gp22 [Vibrio phage KVP40] & Gp22, prohead core orfl86 103980.. 104846 867 - YP 195141.1 prohead core protein protein gp22 [Synechococcus phage S-PM21 NP_899607.1 gp21 [Vibrio phage K.VP40] & Gp21, prohead core orfl88 104892.. 105560 669 - YPJ 95140.1 prohead scaffold and protease protease gp21 [Synechococcus phage S-PM2] ABC95188.1 GP68-prohead conserved hypothetical core protein orfl89 105568..105873 306 - protein [Stenotrophomonas phage SMB141 orfl90 105884.. 106051 168 » hypothetical protein YP 214665.1 gp20 [Prochlorococcus phage P- Gp20 portal vertex orfl91 106089..107780 1692 - SSM4]& YP717798.1 protein of head portal [Synechococcus cyanophage syn9] YP214362.1 T4-like tail orfl92 107847..108380 534 - Gpl9 tail tube protein tube protein [Prochlorococcus phage P-SSM2] &

130 YP_717797.1 tail tube [Synechococcus cyanophage syn9] YP_001469499.1gpl8tail sheath protein [Enterobacteria Gpl8 tail sheath orfl95 108459..110354 1896 - phage Phil] & NP891725.1 protein tail sheath protein [Enterobacteria phage RB49] YP195134.1 terminase large subunit gpl7 [Synechococcus Gp 17 terminase DN A phage S-PM2] & orfl96 110407..112617 2211 - packaging enzyme, YP_214662.1 gpl7 large subunit [Prochlorococcus phage P- SSM4] YP001595292.1 gpl6 terminase DNA packaging Gpl6 terminase DN A enzyme small subunit orfl98 112598..113299 702 - packaging enzyme [Enterobacteria phage JS98] small subunit &NP899600.1 terminase DNA packaging enzyme small subunit [Vibrio phage KVP40] YP_239193.1 gpl5 proximal tail sheath stabilization Gpl5 proximal tail [Enterobacteria phage RB43] orfl99 113302..113997 696 - sheath stabilization &YPJ717778.1 tail sheath protein stabilizer [Synechococcus cyanophage syn9] YP195128.1 neck protein gpl4 [Synechococcus phage S-PM2]&YP_2143 53.1 T4- orf200 114000..114650 651 - Gpl4 neck protein like neck protein [Prochlorococcus phage P- SSM21 orOOl 114710..114922 213 + hypothetical protein orf202 114950.. 115702 753 . Gpl3 neck protein ABC95179.1 GP13-neck

131 protein [Stenotrophomonas phageSMB14]& YP_214352.1 T4-likeneck protein [Prochlorococcus phage P-SSM2] & YP_ 195127.1 neck protein gpl3 [Synechococcus phage S-PM21 NP_757861.1 hypothetical conserved hypothetical orf203 115692..116030 339 - protein MYPE4760 membrane protein [Mycoplasma penetrans HF-2] orf204 116014..116262 249 _ hypothetical protein YP_717772.1 structural protein [Synechococcus cyanophage syn9]& conserved hypothetical orf206 116313..121151 4839 - YP 195118.1 hypothetical protein protein S-PM2p084 [Synechococcus phage S- PM2] CA078738.1 tailspike protein [Salmonella phage conserved hypothetical orf207 121250..123034 1785 - Det7]& YP_002003781.1 protein gpl7 [Enterobacteria phage EcoDSll YP001039674.1 tail fiber- like protein [Erwinia amylovora phage Era 103] & orf210 123236.. 125497 2262 - tailspike protein YP002003 543.1 putative tail fiber protein [Escherichia phage rv51 YP_002003781.1 gpl7 [Enterobacteria phage orf212 125698..127719 2022 - tailspike protein EcoDSl] & YP001742077.1 maturation/adhesion protein [Salmonella phage E1 ]

132 YPJ)02003 830.1 gpl7 [Klebsiella phage Kll]& conserved hypothetical YP214526.1 phage tail orf213 127766..130825 3060 - protein fiber-like protein [Prochlorococcus phage P- SSM21 orf214 130878..132089 1212 _ hypothetical protein orf215 132092..132946 855 _ hypothetical protein YPJ 95114.1 baseplate wedge subunit gp6 Gp6 baseplate wedge [Synechococcus phage S- orf217 132927.. 134708 1782 - subunit PM2]&YP_214640.1 gp6 [Prochlorococcus phage P- SSM41 orf218 135058..135630 573 + hypothetical protein orf219 135688..135834 147 + hypothetical protein NP944054.1 hypothetical conserved hypothetical orf220 137998..138195 198 + protein Aehl pi 76 protein [Aeromonas phage Aehl] orf223 139058..139234 177 + hypothetical protein orf224 139577..140230 654 + hypothetical protein orf225 140287.. 140796 510 + hypothetical protein orf226 140879..142135 1257 + hypothetical protein orf227 142281..142739 459 + hypothetical protein orf229 142767..143138 372 _ hypothetical protein YP001039890.1 hypothetical protein conserved hypothetical orf230 143135..143521 387 - phi3396_03 [Streptococcus protein phage phi3396] (using PSI Blast interation 2) YP001936049.1 conserved hypothetical orf232 143570..144127 558 - hypothetical protein BPs 121 protein [Mycobacterium phage BPsl orf233 144223.. 14443 5 213 _ hypothetical protein orf234 144478.. 145050 573 . conserved hypothetical YP 001294624.1

133 protein hypothetical protein ORF031 [Pseudomonas phage PA11] YP_195168.1 DNA Gp43 DNA polymerase gp43 orf236 145130..148129 3000 + polymerase [Synechococcus phage S- PM21 orf237 148192..148533 342 + hypothetical protein

YP_002003602.1 conserved hypothetical hypothetical protein orf239 148530..149309 780 + protein rv5_gpl00 [Escherichia phage rv51 orf240 149319.. 149624 306 + hypothetical protein SegD homolog NP_899393.1 SegD [Vibrio orf242 149624.. 150310 687 + (homing endonuclease) phage KVP401 orf243 150303..151220 918 + hypothetical protein hypothetical membrane orf244 151223..151420 198 + protein conserved hypothetical YP001468492.1 gpll2 orf245 151417..151797 381 + protein [Listeria phage A511 ] hypothetical membrane orf246 151778..151963 186 + protein orf247 151960..153186 1227 + hypothetical protein . hypothetical membrane orf248 153223..153390 168 + protein orf249 153473..153685 213 + hypothetical protein orf250 153682..154068 387 + hypothetical protein NP944185.1 hypothetical conserved hypothetical orf251 154065..155177 1113 + protein Aehlp307 membrane protein [Aeromonas phage Aehl] NP944184.1 hypothetical conserved hypothetical orf252 155196..155765 570 + protein Aehlp306 membrane protein [Aeromonas phage Aehl] orf253 155770..155898 129 + hypothetical protein orf255 155898..156323 426 + hypothetical protein orf256 156386..156748 363 + hypothetical protein

134 orf257 156748.. 157200 453 + hypothetical protein orf258 157197..157397 201 + hypothetical protein orf259 157369..157542 174 + hypothetical protein orf260 157553..157900 348 + hypothetical protein

135 Table 3.2. Genes which constitute the core genome of the T4-like phages are determined by CoreGenes comparisons of T4, 44RR2.8t, RB43, RB49 and P-SSM2. The presence of these homologs in OSboM-AG3 is indicated in the right hand column.

Product Gene Homolog in OSboM-AG3 DNA primase gp61 + Helicase gp41 + DNA polymerase gp43 + Translation repressor protein regA + Clamp loader gp44 + Sliding clamp gp45 + Recombination endonuclease gp46 + Recombination endonuclease gp47 + RNA polymerase sigma factor gp55 + Tail completion and sheath stabilizer protein gp3 + Head completion protein gp4 + Baseplate wedge gp53 + Baseplate hub subunit and tail lysozyme gp5 + Baseplate wedge subunit gp6 + Baseplate wedge subunit gp8 + Fibritin Wac + Neck protein RP'13 + Neck protein gpl4 + Tail sheath stabilizer and completion protein gp!5 + Terminase, small subunit gpl6 + Terminase, large subunit gpl7 + Tail sheath protein gpl8 + Tail tube protein gpl9 + Continued...

136 Product Gene Homolog in

137 Homing Endonucleases. The AG3 genome contains three copies of sequences

homologous to homing endonuclease (HNH endonucleases); genes 42, 96 and 242.

Lysis. No holin or lysin-encoding genes were detected in the genome. However,

transmembrane domains were identified in some of the AG3 gene products (Table 3.3).

Protein products of genes 44.1, 78, 117, 244, 246 and 251 were found to contain more than two transmembrane domains and could possibly code for the holin.

Regulatory sequences. Based on sequence homology to the consensus Sigma 70

promoter, TTGACA (N15.18) TATAAT, eight promoters were identified (Table 3.4).

Also, eight terminators were identified with the characteristic T-rich extensions on the 3'

end (Table 3.5). Among the analyzed genes, orfl22 and orf076 were found to be

homologous to late transcription genes of T4; gp55 and gp33, respectively. The former protein is involved in the shift to late transcription in coliphage T4.

Codon usage and tRNA. The AG3 genome was found to contain four tRNA

genes for three amino acids, namely, serine (TGA and GCT), asparagine (GTT), and tyrosine (GTA) (Table 3.6). All lie between genes 220 and 223 and are transcribed in the

same direction as those genes. The GC content of these tRNAs ranged from 53.9 to

56.2%. Adding these tRNAs to the ORFs resulted in a total coding capacity for the AG3 phage genome of 146,481 bp or 92.7% of the genome codes for products. A comparison

of the codon usage pattern of the phage and its host (Shigella boydii) showed some

interesting features (Table 3.7). Both have nearly the same trend of using the detected tRNA codons.

138 Table 3.3, Number of the transmembrane domains found in the genome of

Number of detected Coordinate ORF Predicted product transmembrane domains (bp) using Phobius and TMHMM orf009.1 7196..7393 hypothetical membrane protein 1* orf018 12289..12576 hypothetical membrane protein 1 conserved hypothetical membrane orf027 14823..15128 1 protein orf034 17779.. 18186 hypothetical membrane protein 1 orf044.1 25595..25852 hypothetical membrane protein 2 orfl)44 25854..26087 hypothetical membrane protein 1

orf046 27578..27730 hypothetical membrane protein 1

orf047.1 27727..27894 hypothetical membrane protein 1 orfl)47 27891..28076 hypothetical membrane protein 1 orit)48 28076..28618 hypothetical membrane protein 1 orf054 30245..30361 hypothetical membrane protein 1 orfl)78 44405..44716 hypothetical membrane protein 2 orni7 66872..67036 hypothetical membrane protein 2 orfl35 78624..78782 hypothetical membrane protein 1 conserved hypothetical membrane orf203 115692.. 116030 1 protein orf244 151223..151420 hypothetical membrane protein 2 orf246 151778..151963 hypothetical membrane protein 2 orf248 153223..153390 hypothetical membrane protein 1 conserved hypothetical membrane orf251 154065..155177 3 protein conserved hypothetical membrane orf252 155196..155765 1 protein 2 transmembrane domains with TMHMM

139 Table 3.4. Putative promoters in the genome of OSboM-AG3 phage

Promoter Coordinates Sequence P009 6563..6590 TTttCAtttccagcattcggtgTATAAT P067 complement^ 8622..3 8648) TTcAatagatgaaggggcTGcTATAAT P069 complement 9601 ..39630) TTtACAtttatgaaaaatgcagtaTATtAT P100 complement^ 8420. .58446) TTGAaAaagactttactcttcaATAAT P138 complement(79794..79821) TTtACttcttcaataactcacgTATAAT P170 complement(97212..97241) TTGAttctaatccatgaaatacggTATAAT P206 complement(121176..121203) TTGgCtgtatactaaatatctcTATAAT P237 148126..148155 TTGAttcaataaaccaaagggggaTATAAT Sigma70 TTGACA(N 15-18)TATAAT consensus

Table 3.5. Putative rho-independent terminators in the genome of OSboM-AG3 phage predominantly discovered and verified using MFOLD

Terminator Position Sequence number T01 Complement (42526..42550) Gccctcttcggagggctttttattt T02 42535..42555 Gccctccgaagagggcttgtt T03 18421..18442 Cccgctccgttggggcgggttt T04 Complement (49767..49791) Ggacgaaattctaattcgtcctttt T05 Complement (114921..114944) Ggcgggttcccccgcccttctttt T06 114929.. 114947 Ggcgggggaacccgccttt T07 Complement (142732..142760) Ggggcttaatgcccctttttgttattctt T08 142746..142768 Ggggcattaagcccctttctttt

140 Table 3.6. Location of the tRNA genes in the OSboM-AG3 genome, their cognate amino acids and anticodons detected using Aragorn.

GC tRNA tRNA Amino Length Anticodon Codon Strand content number location acid (nt) (%) 137688.. 1 Ser TGA UCA + 80 56.2 137767 138197.. 2 Asp GTT AAC + 76 53.9 138272 138530.. 3 Tyr GTA UAC + 84 56.0 138613 138924.. 4 Ser GCT AGC + 89 56.2 139012

Table 3.7. Comparison of the tRNA codon usage in OSboM-AG3 and its host Shigella boydii.

tRNA codon OSboM-AG3 Shigella boydii UAC 14% 18% AAC 71% 55% UAG 7% 10% AGC 37% 30%

141 Selecting the codons which are used at a frequency >30% and show a >1.5 fold increase in use, we could identify six codons which are significantly overrepresented in the phage genes (phenylalanine [UUC], isoleucine [AUG], proline [CCU], lysine [AAG], aspartic acid [GAC] and arginine [CGU]), yet no phage-specified tRNAs exist.

3.4.3

Separation of the cesium chloride purified AG3 phage particles on SDS-PAGE revealed 15 protein bands after one hour of destaining (Figure 3.2). The molecular weights of the observed bands were 12.7, 15.6, 17.7, 21.9, 27.2, 32.4, 34.3, 35.7, 37.3,

44.6, 60.4, 67.9, 76.5, 107.5 and 126.3 kDa based on comparison to size markers. Six of the most intense bands (named A to F) were excised and analyzed by QqTOF mass spectrometry (Appendix A 3.1). The observed masses of the protein bands B to F were obviously similar to the predicted sizes by SDS-PAGE (Table 3.8). However, the predicted size of protein band A (177.1 kDa) was higher than the observed one (126.3 kDa), suggesting a truncated form at the C-terminus of the protein. On the other hand, the highest sequence coverage was achieved for protein band D (major head protein, 44.6 kDa), which was 64.1%, while the lowest one was 16.4 % and was found to be for protein band E (tail membrane protein, 21.9 kDa) due to less trypsin cleavage sites. Based on the sequence identification by mass spectrometry, the protein bands C and F were two components (sheath, tube) of tail proteins (products of genes 195, and 192, respectively), while the protein band D was assigned to the major head protein (product of gene 185)

(Figure 3.3). In addition, the large protein bands A and B were identified as conserved hypothetical phage proteins (product of genes 206 and 213) that were found to be similar to those reported in cyanophages and Klebsiella.

142 M 1

200 KDa

50 KDa

15 KDa

Figure 3.2. SDS-PAGE of

143 Table 3.8. Summary of the open reading frames identified by the mass spectrometry analysis of CsCl-purified Vi-I like OSboM-AG3 phage

Bands No. of Molecular weight (KDa) Sequence on the ORF; gene unique coverage SDS- description SDS- Predicted peptides MS (%) PAGE PAGE (sequence) identified 206; conserved A 126.3 177.1 177.2 36 39.6 hypothetical protein 213; conserved B 107.5 106.8 106.9 25 42.6 hypothetical protein 195; Gpl8 tail C 67.9 68.1 68.2 18 43.1 sheath protein 185;Gp23, D major head 44.6 47.8 48.0 23 64.1 protein 006; putative E 21.9 27.5 27.6 6 16.4 tail protein; 192; Gp 19 tail F 17.7 19.7 19.8 12 55.9 tube protein

144 A

Figure 3.3. Electron micrograph of the Vi-I like OSboM-AG3 phage and SDS-PAGE gel examination of purified phage particles and relationship of the resulting protein bands to mass spectrometric analysis of individual polypeptides. Six bands have been analyzed by mass spectrometry and, if possible, their sites within the phage particle have been highlighted. Proteins A and B are hypothetical proteins, their functions and positions on the phage particles are unknown.

145 3.5 DISCUSSION

Vil-like phages were reported to be found among isolated phages against

Salmonella Typhi, Vibrio cholerae and Rhizobium meliloti (Ackermann, 1970, Werquin

et al, 1988, Guidolin et al, 1984). To the best of our knowledge,

Vil-like phage characterized for Shigella species. Although Vi phage were first described

over 60 years ago (Ackermann, 1970), their genome structure has generally not been

investigated. The only other fully characterized Vi-specific phage, is Salmonella phage

El (Pickard et al, 2008), which is a member of the Siphoviridae, with a genome of 45.4

kb. The OSboM-AG3 genome was found to be a double-stranded linear DNA of

approximately 158 kb and based on the sequencing results; it is circularly permuted and

terminally redundant.

The BLASTP analyses revealed sequence similarity to a number of phages,

including members of the "T4 superfamily". More detailed analysis using CoreGenes

showed that this phage shared homologs with P-SSM2, T4, Aehl, and interestingly 69

with Delftia acidovorans phage 0W-14 (Miller et al, 2003, Kropinski and Warren, 1970,

Sullivan et al, 2005). Phages related to coliphage T4 belong to the most ecologically and

genomically diverse group of viruses, a fact long recognized by phage biologists. These

phages have been identified in Japanese rice fields (Jia et al, 2007), marine systems (Filee

et al, 2006, Filee et al, 2005, Comeau et al, 2010), and in soils from Lithuania (Klausa et

al, 2003), Bangladesh and Switzerland (Zuber et al, 2007). Fully sequenced members of

the "T4 superfamily" includes viruses which infect Enterobacteria, Aeromonas, Vibrio,

Prochlorococcus and Synechococcus species, in other words members of the phyla

146 Proteobacteria and Cyanobacteria. These viruses contain genomes of 161 (Aeromonas

phage 25) to 252 kb (Prochlorococcus phage P-SSM2) harbouring 240-330 protein-

encoding genes (Sullivan et al, 2005). All sequenced "T4-like" genomes contain a set of

33-35 genes that have resisted divergence throughout evolution (Nolan et al, 2006, Petrov

et al, 2006). It was reported that there are 36 proteins which are conserved in

Enterobacteria phages T4 (NC_000866), RB49(NC_005066), RB43 (NC_007023),

Aeromonas phage 44RR2.8t (NC005135) and Prochlorococcus phage P-SSM2

(NC006883) (Nolan et al, 2006, Petrov et al, 2006). They include morphogenesis

protein, DNA packaging proteins, as well as proteins involved in nucleotide metabolism,

DNA replication and recombination, and transcriptional control. These conserved

proteins were identified in

gp34 of the long tail fiber protein. These results clearly indicate that ®SboM-AG3 is part

of the "T4 superfamily".

Based on sequence analysis of the major head gene, members of the "T4

superfamily" were classified into T-even, exo-T-even, pseudo-T-even and thermo-T-even

(Desplats and Krisch, 2003), but since these groups, and indeed, the "T4 superfamily"

possess no taxonomic status, Lavigne et al. reexamined the classification of these viruses

using whole genome proteomic comparisons (Lavigne et al, 2009). Their results suggest

that many, but not all, of the T4-like phages belong to a new subfamily, the

Teequatrovirinae, which currently contains two genera (T4-like and KVP40-like) and

several subgroups. And as a result of this study we consider OSboM-AG3 to be a member of this new subfamily.

The presence of T4 gp55 (orfl22) and gp33 (orf076) homologs in the phage

147 genome suggests that, like T4, late transcriptional control is exerted by alteration in promoter recognition by modifying the host RNA polymerase, but with OSboM-AG3 there are no homologs of the T4 gpAlt (early transcription), or gpMotA and gpAsiA

(middle transcription) (Miller et al, 2003). On the other hand, the lack of an apparent gp34 homologue is interesting. The proximal portion of tail fibres of T4-like phages are

composed of two molecules of gp34 with the distal portion made up of a dimmer of gp37.

This structure is lengthened by the addition of gp36 and further modified with gp35

(Miller et al, 2003). The only possible similar gene cluster in OSboM-AG3 is orf206,

207, 210, 212 and 213. Interestingly orf272 shows sequence similarity to a maturation/adhesion protein (YP_ 001742077) from Salmonella Vi-typing phage El

(Pickard et al, 2008), which is particularly notable in the N-terminal region. Most importantly because of the focus of this research, the genome of this phage does not contain any genes that are related to bacterial toxins and/or lysogeny. Similar results were obtained when the sequencing data analysis of Listeria phage PI 00 revealed that it does not contain any homology to genes or proteins which are known or suspected to be involved in toxicity, pathogenicity or antibiotic resistance (Carlton et al, 2005).

In conclusion, ®SboM-AG3 was identified to be a member of the "T4 superfamily" and, to be more precise, to the newly proposed subfamily,

Teequatrovirinae. Along with its strong lytic activity and broad host range (Chapter 2), it does not have any virulent and/or lysogenic genes. This would make this phage a good candidate for safe application to control the foodborne pathogen Shigella spp. (Strauch et al, 2007).

148 Chapter 4: USE OF A COCKTAIL OF PHAGES TO

CONTROL FOODBORNE PATHOGENS IN LIQUID

MEDIA AND IN A FOOD SYSTEM

4.1 ABSTRACT

This study investigated the potential of different phage cocktails to combat various targeted pathogens in both liquid media and in a real food system. First, the

Bioscreen C was used to determine the optimal MOI value that can be used for this objective. An MOI value of around 105 was necessary to completely inhibit the growth of the targeted hosts for 5 days. Phage cocktails at this high MOI were added to their corresponding strains of Salmonella, Listeria monocytogenes, Shigella and Escherichia

coli 0157:H7 and incubated at 4 and 25°C. Pathogen populations were determined at regular intervals throughout the incubation period. Under these conditions, non-detectable growth was observed for tested strains of Salmonella, L. monocytogenes and Shigella sonnei during nine days. Application of Listeria and E. coli phage cocktails to ready-to- eat and raw meat products, respectively, under different packaging conditions resulted in a reduction of the host pathogens below the assay detection limit when the meats were

stored at 4°C. However, significant reductions of L. monocytogenes were noticed after 9 days under vacuum and modified atmospheric packaging conditions with 2.0 and 0.9 log unit reductions, respectively, being achieved during storage at 25°C. The Listeria phage

cocktail resulted in significant differences in biofilm formation of two L. monocytogenes

strains. These results indicate that these phages may be useful for achieving safe food.

149 4.2 INTRODUCTION

Phages have been used safely in human and veterinary medicine to treat bacillary dysentery and to reduce mortality due to cholera in India since Felix d'Herelle proved their effectiveness in 1919 (D'Herelle, 1926). Although, this path was initially abandoned with the discovery of antibiotics, research and application of phage therapy in human medicine continued in Eastern Europe and the Soviet Union. Phage therapy is currently used in this region to treat bacterial infections in humans, and is used as a complement to conventional antibiotics (Kutter et al, 2010). The incidence and cost of foodborne illnesses prompted an urgent call to improve methods for minimizing or preventing the occurrence of pathogens in food products. Moreover, the increasing demand for minimally processed and organic foods requires the development of natural antimicrobials to control bacterial contamination. Currently, different chemicals and antimicrobial compounds are applied to achieve this goal, however the worldwide increase in bacterial resistance and potential side affect of these antimicrobials have recently attracted many researchers in western countries to re-think the use of phage to combat foodborne pathogens and spoilage organisms (Mclntyre et al, 2007, Hudson et al,

2005, Hagens and Loessner, 2010).

Lytic (virulent) phages either in a single form or in cocktails have been used successfully to control several foodborne pathogens (Hagens and Loessner, 2007,

O'Flynn et al, 2004, Greer, 2005, Carlton et al, 2005). They have been used to specifically eradicate pathogens from living animals, decontaminate carcass meat and to disinfect the surfaces of ready-to-eat products (Hagens and Loessner, 2010, Strauch et al,

2007). The recent FDA approval of phage preparations as food additives for preservation

150 of ready-to-eat meat has also triggered the search for new applications for these natural bacterial killers (FDA, 2006). Using phage cocktails may be a necessary requirement in order to achieve significant reduction in the targeted pathogen and to reduce the chance of resistance development (Strauch et al, 2007). In addition, a significantly high number of phages is required to ensure sufficiently rapid contact and infection of the few targeted bacterial cells present (Hagens and Loessner, 2010). It was recommended that the efficacy of phages to control target pathogens in foods has to be tested both at higher than normal"storage temperatures, which provide good growth conditions for the undesired contaminants, and at recommended storage conditions (Hagens and Loessner, 2010).

Salmonella, Listeria monocytogens, Escherichia coli 0157:H7 and Shigella strains are common foodborne pathogens and account for most of the recent reported outbreaks (Centers for Disease Control and Prevention, 2005). Raw beef and ready-to-eat meat products are the important vehicles for E. coli 0157:H7 and L. monocytogenes infections (Feng, 1995, Farber and Peterkin, 1991). Around 545,000 pounds of beef products were recalled recently in the United States due to E. coli 0157:H7 contamination that resulted in 26 illnesses and 19 hospitalizations and five who developed hemolytic uremic syndrome ((Centers for Disease Control and Prevention,

2009). In a recent outbreak in Canada, consumption of contaminated RTE meat products with L. monocytogenes led to 57 confirmed cases of listeriosis across the country, of those cases there were 23 deaths with an estimated economical loss for the producing company of CAD$ 50 million (Public Health Agency of Canada, 2010). As a result, it is very important to investigate the behavior of phages to control both pathogens in these food products.

151 Biofilm formation is an important problem in the food industry because it may represent an important source of contamination for food materials coming into contact with biofilm-containing areas and causing food spoilage or transmission of diseases

(Bonaventura et al, 2008). Once formed, biofilm allows pathogens to persist in the food environment for prolonged periods and resist treatment with antimicrobial and sanitizing agents (Folsom and Frank, 2006). L. monocytogenes strains were reported to produce multicellular biofilms through their ability to adhere to food-contact surfaces such as polystyrene, glass and stainless steel (Blackman and Frank, 1996). Although there are many promising studies on the use of phages to control biofilm formation by pathogenic bacteria such as Pseudomonas aeruginosa (Knezevic and Petrovic, 2008)and

Staphylococcus epidermidis (Curtin and Donlan, 2006), only two published studies have investigated the effect of phages on biofilm formation by L. monocytogenes (Roy et al,

1993, Hibma et al, 1997). It was suggested that phages could eradicate biofilm through three types of phage hydrolytic enzymes; first, enzymes that are able to hydrolyze exopolysaccharides that are involved in the structural integrity of the biofilm, second, enzymes that can degrade the protective layer (capsular material) that can be found around the bacterial host and inhibit phage attack and third, endolysin that is involved in the lysis of the already infected bacterial cells (Abedon, 2010).

As a biocontrol agent in food, phage lytic activity should be examined in both artificial media and in foods to optimize phage concentration and conditions for successful results (Hagens and Loessner, 2010). Therefore, this part of the research work examined the possibility of using cocktails of lytic phages described in Chapter 2 to control the targeted pathogens in both liquid media and food under different conditions.

152 4.3 MATERIALS AND METHODS

4.3.1 Bacteria and Bacteriophage

Tryptic Soy Broth (TSB), Tryptic Soy Agar (TSA), Tryptose Soft Agar (TSB +

0.4% agarose) and Listeria Selective Agar Base (Oxford Formulation) (Difco

Laboratories, Detroit, MI) were used in this study to grow the host bacteria, propagate the phages and count L. monocytogens after challenge experiments. Ampicillin (100 ug/ml) was added to TSB and TSA to grow and count the ampicillin-resistant E. coli 0157:H7 strain. For propagation and titre determination of Listeria phages, after sterilization, all the media were supplemented with filter-sterilized CaCb to a final concentration of

1.25mM. The previously isolated and characterized phages in Chapter 2 and their susceptible strains of E. coli, Salmonella, Shigella and Listeria spp. (obtained from the

Canadian Research Institute for Food Safety (CRIFS) Culture Collection at the University of Guelph; Table 3.1) were used in this study for phage propagation and biocontrol experiments. Only for the meat challenge experiment, E. coli 0157:H7 (ampy.lux) (C918) was used to artificially inoculate raw meat samples. Pure cultures were obtained from

-80°C frozen storage and maintained at 4°C on TSB until use. Cultures were re-streaked every month to maintain cell viability. Propagation of the phages with high titre was performed as previously described (Chapter 2). The three phages against each target genus were mixed together to form the phage cocktail. The titre of each phage cocktail stock was determined by preparing serial 10-fold dilutions and tested using the previously described overlay method. The phage cocktail lysate was stored at 4°C.

153 Table 4.1. Phages and Bacterial strains that used for phage propagation and biocontrol experiments

CRIFS culture Phage Bacterial Host Strain Designation collection number Salmonella DTI04 - SA2002- StyM-AG6 C 1077 Typhimurium 5807 Salmonella SenS-AGll C417 En- 2588 Enteritidis Salmonella StyM-AG16 C435 SA 941256 Typhimurium LinM-AG8 Listeria innocua C505 . Listeria LmoM-AG13 LJH391 4b monocytogenes Listeria LmoM-AG20 C519 l/2b monocytogenes SboM-AG3 Shigella boydii C 865-2 SsoM-AG8 Shigella sonnei C 866-2 SsoM-AGlO Shigella sonnei C 866-2 0126:H8, EC EcoM-AG2 E. coli 0126:118 C761 910061 0126:H8,EC EcoM-AG3 E coli 0\26:m C761 910061 0157:H7,ATCC EcoM-AGlO E. coli 0157:117 C899 43888

154 4.3.2 Effect of different multiplicity of infection values on the growth of the

target pathogens as determined using a Bioscreen C Microbiology

Plate Reader

The effect of different multiplicity of infection (MOI) values (ratio of the phage numbers to the bacterial numbers measured at the beginning of the experiment) for each phage cocktail on the growth of the susceptible target host bacteria was determined by measuring the optical density (OD) of the tested bacteria in the presence of different concentrations of corresponding phage cocktail using a Bioscreen C Microbiology Plate

Reader (Labsystems, Helsinki, Finland). The following experimental parameters were used for all experiments: Single, wide band (wb) wavelength; 25°C incubation temperature; 5 min preheating time; kinetic measurement; measurement time 120 hours; reading every 30 min and medium intensity shaking for 10s before measurements. Fifty microliters of the phage cocktail lysate, were transferred to each of the 100 wells of the

sterilized honey comb plates of the Bioscreen C reader (Fisher Scientific, Mississauga,

ON), and then each of the wells was inoculated with 125 ul of the diluted overnight host bacterial culture (around 10 CFU/ml). The phage cocktails were added at different concentrations to achieve MOI values of around 10"1, 10, 103 and 105. The control wells contained either phage, TSB, phage buffer only or bacteria with phage buffer with the same volume as tested wells. All samples were tested in triplicate. The means of the developed triplicate OD data were calculated and a growth curve was developed for each

MOI value used with each host bacterium.

155 4.3.3 Bacterial challenge test

Only one strain from each target genus was involved in the challenge test in TSB

media to determine the effect of phages on bacterial numbers. These strains are

Salmonella Enteritidis (C417), L. monocytogenes (C391), Shigella sonnet (C866-2) and

E. coli 0157:H7 (C899). Challenge tests were performed using a fresh overnight culture

of the appropriate strain, which was then diluted in fresh medium to around 103 CFU/ml.

Phage cocktail was added to achieve a MOI of approximately 105. The mixture was then

incubated at 4°C and 25°C. Phage-free cultures and bacteria-free culture (containing

either only phage or TSB) were also included as controls. The change in the bacterial

numbers was determined by taking aliquots of the culture suspensions after incubation for

1, 3, 7 and 9 days and making serial dilutions in sterile 0.85% saline solution (Sigma-

Aldrich, Oakville, ON). Each dilution (100 ul) was plated onto TSA plates in duplicate

and incubated for 24 hours at 37°C. Colonies were counted and used to calculate the

bacterial numbers (CFU/ml) in each sample. Bacterial counts were transformed to logio

values.

4.3.4 Potential of the phage cocktails to control Listeria monocytogenes and E.

coli 0157:H7 in food

Listeria and E. coli phage cocktails were applied to ready-to-eat oven roasted turkey breasts and raw beef, respectively, in the presence of their respective hosts. These

food products were selected as an example of foods frequently found to be contaminated with L. monocytogenes and E. coli 0157:H7. L. monocytogenes (C391) and E. coli

0157:H7 (ampr.lux) (C918) strains were used as phage indicator strains in this

156 experiment. The latter was used in order to enable direct plating and reduce problems arising from contamination of the plates by background flora. Meat samples were purchased on the first day of the experiment from local grocery stores and initially

screened for contamination with the targeted bacteria. Twenty five grams of the meat

samples were sliced, weighed aseptically and then placed in sterile Petri-dishes and pre- equilibrated to the desired temperature. An overnight culture of each bacterium was diluted to approximately 10 CFU/ml. A 100 |j.l portion was spotted onto the surface of each piece of meat and the bacterial cells were allowed to attach for 10 min at room temperature. This was followed by addition of 100 ul of phage cocktail to achieve

PFU/CFU ratio of 105. For the controls, the same volume of phage buffer was used

instead of the phage cocktail suspension. For Listeria-containing samples, meat was

incubated at 4°C and 25°C for nine days under aerobic, modified atmosphere packaging

(MAP) and vacuum conditions. The modified atmosphere was composed of approximately 1 % O2, 13 % CO2 and 86 % N2, which was generated by using an

AnaeroGen sachet in a sealed anaerobic jar (Oxoid, Fisher Scientific, Mississauga, ON,

Canada). The vacuum packaging was performed using a Komet vacuum packaging machine (Stuttgart- W, Kornbergstr 27-29, Germany) at a vacuum of 1.0 bar. The meat

samples were packed in sterile polyethylene bags of 65 um thickness (Seward Laboratory

Systems Inc., Bohemia, NY) before being placed in the vacuum machine. Counts were performed to determine the host number after 1, 3, 7 and 9 days. E. co//-contaminated meat samples were incubated at 4°C, 10°C and 25°C for fifteen, nine and two days, respectively, under aerobic conditions only. Viable E. coli 0157:H7 counts were monitored after 1, 3, 7, 9, 12 and 15 days for 4°C incubated samples, after 1, 3, 7 and 9

157 days for 10°C incubated samples and after 6 , 12, 24 and 48 hours for 25°C incubated samples. In both cases, the bacterial viable counts (CFU/gm) were initially determined immediately after addition of bacteria and phage. Triplicate samples were prepared for each sampling time. At each sampling time, the meat was transferred to a Whirl-Pak bag

(Nasco, Fort Atkinson, WI) containing 225ml of 0.85% saline solution and the sample stomached for 1 min at 200 rpm (Seward Blender Stomacher 400 circulator, Seward

Laboratory Systems Inc., NY). The liquid portion was transferred to a sterile centrifuge tube and serially diluted in sterile 0.85% saline solution. An aliquot (100 ul) was plated in duplicate on Oxford agar plates and TSA + Ampicillin plates for L. monocytogenes and

E. coli 0157:H7 counts, respectively, and incubated at 37°C for 48 hours in the case of L. monocytogenes until typical Listeria colonies could be enumerated and for 24 hours in case of E. coli 0157:H7. For E. coli 0157:H7, the developed colonies with bioluminescence were the only ones that were counted using NightOwl II LB 983

(Molecular Imager; E.G. and G. Berthold, Munich, Germany). Colonies were counted and used to calculate the bacterial numbers (CFU/gm) in each sample. Bacterial counts were transformed to logio values. The growing colonies, if any, were randomly re- isolated and tested for phage susceptibility. For this purpose, 10 colonies were picked from each plate of the targeted bacteria and sub-cultured for three successive times.

These were then used as targets for the corresponding phage cocktails in a spot test (see

Chapter 2) to assess their sensitivity to the phage through detection of lytic zones.

4.3.5 Effect of Listeria phage cocktail on biofilm formation by Listeria

monocytogenes

To study the effect of phages on biofilm formation by two strains of L.

158 monocytogenes (C391 and C519), quantification of biofilm production in plastic microtitre plates containing Listeria strains mixed with the phage was performed based on the previously described methods (Knezevic and Petrovic, 2008, Djordjevic et al,

2002). The wells of a Corning Costar sterile 96-well, flat-bottomed polystyrene microtiter plate (Corning Inc., Corning, NY; not prepared by the manufacturer for tissue culture work) were filled with 100 ul of different concentrations of L. monocytogenes strains prepared in TSB from overnight fresh cultures. The employed concentrations were approximately 102, 104, 106 and 10 CFU/ml. Appropriate Listeria phage cocktail

Q dilutions in phage buffer (100 ul) were added into the wells to obtain around 10 PFU/ml in each well. The negative control wells contained TSB or phage buffer only, while the positive controls contained the same concentrations of bacteria with addition of phage buffer instead of phage cocktail lysate. Each treatment was tested in three wells. The microtiter plate was incubated at 30°C for 48 hours. After overnight incubation, the contents of the plate were then poured off and the wells washed three times with 250 ul of phosphate buffered saline (PBS) (Sigma-Aldrich, Oakville, ON) and left to dry. The remaining attached bacteria were fixed with 250 ul of absolute methanol for 15 min. The fixative was removed and the plates were air-dried. The developed biofilm was stained by adding 220 ul of (0.1% w/v) crystal violet (Gram-color staining set for microscopy,

Fisher Scientific, Mississauga, Canada) to each well followed by incubation for 2-3 min at room temperature. Excess stain was later rinsed off and wells were washed twice with sterile distilled water. After every wash, the microtiter plate was inverted on absorbent tissue paper to drain out all the liquid. The dye bound to the adherent cells was resolubilized by addition of 220 ul of destaining solution (ethanol:acetone 80:20 v/v) to

159 each well and the plate was incubated at room temperature for 15 min. The absorbance of each well was measured at 530 nm using a Victor2 1420 Multilabel Counter (Wallac Oy,

Turku, Finland).

4.3.6 Statistical analysis

The statistical analysis of the experimental data was accomplished with SigmaPlot

Version 10.1 (Systat Software Inc., Chicago). A one-way analysis of variance (ANOVA) was performed. In all cases, statistical differences between the means were indicated by

P<0.05.

4.4 RESULTS

4.4.1 Effect of different multiplicity of infection values on the growth of the

target pathogens as measured using the Bioscreen C

The Bioscreen C Microbiology Plate reader was used to determine the effect of different MOI values on the growth of ten bacterial hosts by measuring the optical density of the cell suspensions over a five day period of incubation at 25°C. For all tested bacterial strains, adding phage cocktail to a MOI value of approximately 105 resulted in complete inhibition of the bacterial growth throughout the whole experiment. For the

Salmonella phage cocktail, MOI values of around 103 and 105 were required to completely inhibit growth of the tested Salmonella Typhimurium (C435) and (CI077) strains, respectively (Figure 4.1). However, a MOI of 0.1 and 10 caused a delay of S.

Typhimurium C435 to enter log phase by more than 24 hours. Growth of S. Enteritidis

(C417) was completely inhibited by the addition of the Salmonella phage cocktail at a

160 MOI greater than 0.1, the latter caused a delay of the growth of the bacteria to reach log phase by more than 24 hours. For all the tested Listeria spp., different concentrations of

Listeria phage cocktail resulted in complete inhibition of the growth for five days (Figure

4.2). At a MOI of around 0.1, Shigella boydii (C865) started to grow only after about 48 hours and, moreover, the same turbidity as the control sample was not observed even after five days (Figure 4.3a). However, higher MOI values caused complete inhibition of the bacterial growth. On the other hand, Shigella sonnei (C866) growth was inhibited by all of the tested MOI values during the incubation period (Figure 4.3b). All the MOI values tested, except an MOI of 105, could not completely inhibit the growth of both tested E. coli strains for the whole incubation period (Figure 4.4). However, they did delay the time to enter log phase. A MOI of 105 was able to inhibit completely the growth of the tested strains for five days.

161 a)

1.6-1 — MOI = 0.1 ^ 1.4- ——MOI = 10

t 1.2- ——MOI = 100000 'St 1- Control (A § 0.8- •D s °-6- •e 0.4- " 0.2-

0 24 48 72 96 120 Hours

b)

1.6 I — MOI = 0.1 ^ 1.4- ^^^ MOI = 10 " MOI = 1000 — MOI = 100000 £ 1- ~™~~~~ Control (A S 0.8- XI 1 ""

* 0.2- () 24 48 72 96 120 Hours

c)

1.6 -i ——MOI = 0.1 ^ 1.4- MOI = 1000 — MOI = 100000 £ 1- m Control § 0.8- XI s °6- '% 0.4- O 0.2- (f^ 0- ) 24 48 72 96 120 ( Hours

Figure 4.1. Effect of different MOI's of Salmonella phage cocktail on the growth of three Salmonella strains; a) Salmonella Typhimurium CI077, b) Salmonella Enteritidis C417 and c) Salmonella Typhimurium C 435.

162 a)

11 —-MOI = 0.1 —MOI = 10 MOI = 1000 £ 0.8- •-»•«' —-MOI = 100000 Control c 0) 2 0.4. S 5 °2-

0 24 48 72 96 120 Hours

b)

1.2 i ^—MOI = 0.1 MOI = 10 ***-.,_! MOI = 1000 £. 0.8- MOI = 100000 Control | 0.6- S °-4- 5" 0.2- [ 0- () 24 48 72 96 120 Hours

c)

MOI = 0.1 1 -| MOI = 10 $ 0.8- —^ MOI = 1000 f* __ —— MOI = 100000 2" » 0.6- C d) 2 0.4- n u 8 0-2-

0- () 24 48 72 96 120 Hours

Figure 4.2. Effect of different MOI's of Listeria phage cocktail on the growth of three Listeria strains; a) Listeria innocua C505, b) Listeria monocytogenes C391 and c) Listeria monocytogenes C 519.

163 a)

1.4 i ^—MOI = 0.1 MOI = 10

^•"•*~~* — MOI = 100000 Control g 0.8 • d> r^ •D 0.6- | 0.4- O 0.2- 1 ^y—"*

0 24 48 72 96 120 Hours

b)

1.4 -MOI = 0.1 -MOI = 10 1.2 MOI = 1000 f 1 -MOI = 100000 -Control £ Oft ¥>c 0) TJ 0.6 0.4 tical 8 0.2

24 48 72 96 120 Hours ure 4.3. Effect of different MOI's of Shigella phage cocktail on the growth of two Shigella strains; a) Shigella boydii C865 and b) Shigella sonnei C866.

164 a)

1.8- — MOI = 0.1 -~ 1.6- |1.4- y^ MOi = 1000 y/ —MOI = 100000 g 1- f ^^, •» Control •§ 0.8- 8 °-6- '§. 0.4- ff 0.2- 0- () 24 48 72 96 120 Hours

b)

1.6-I — MOI = 0.1 ^ 1.4- — MOI = 10 " MOI = 1000 8 saw** * ——MOI = 100000 £ 1. Control (0 S 0.8- •o e o.4-

Figure 4.4. Effect of different MOI's of E. coli phage cocktail on the growth of two E. coli strains; a) E. coli 0126:H8 (C761) and b) E. coli 0157:H7 (C899).

165 4.4.2 Bacterial challenge test

To investigate the ability of each phage cocktail to lyse its susceptible host in vitro, challenge trails were performed by addition of the phage cocktail to a diluted fresh, overnight culture. A MOI of around 105, which was the optimum MOI value determined from the previous experiment as it resulted in complete inhibition of the growth of all tested strains, was chosen to perform the bacterial challenge test. Four representative bacterial strains were selected for testing efficacy of their corresponding phage cocktail to inhibit their growth at 4°C and 25°C for nine days. At 4°C, adding Salmonella, Shigella and Listeria phage cocktails resulted in non-detectable growth by direct plating of their host bacteria for nine days (Figure 4.5a, b and c). Although non-detectable growth of E. coli 0157:H7 (C899) was observed only at 1 and 3 days, this strain was able to grow in the presence of E. coli phage cocktail after incubation for 7 and 9 days. However, the counts were significantly reduced by 1.7 and 1.3 log units, respectively, when compared with the control (Figure 4.5d). On the other hand, there was no significant effect of

Salmonella phage cocktail on the growth of Salmonella Enteritidis (C417) when the mixture was incubated at 25°C (Figure 4.6a). An approximate 0.5 log unit reduction was detected after 1 and 3 days. Listeria monocytogenes (C391) growth was below the detection level for 6 days at the same temperature but was able to grow again when examined after nine days with around 3.7 log units reduction compared to the control

(Figure 4.6b). For the tested Shigella and E. coli strains, no growth was detected after all sampling times except after 3 days for E. coli 0157:H7 strain, when counts of around 2 log CFU/ml were detected (Figure 4.6C and d).

166 a) b)

4.50 • -•—Treatment -^—Treatment 4.60 • 4.00 • ^— Control * Control 4.00

,. >- i _ 3.50' 3S0 _l —- ' E J 5 3.00- u. | 3.00 ^ 2.50- u. O) 0 £2.60 "* ZOO- o \ 2.00 1.50 \ 1.60 \ 0 13 7 9 0 1 3 7 9 Days Days

c) d)

-•—Treatment 4.00 i 4.5- —•—Treatment I -*- Control —*— Control 4.0- 3.50 t 1— I— 1 _ 3.5- j 3.00 \ _l E \ 3 IO- 3 "- 250 \ II. \

$«• O) O) ° 2.00 \ O J 2.0- \ \ >* 1.5- 1.50 \ jS ,\ \ ^s 0 1 3 7 9 0 13 7 9

Days Days

Figure 4.5. Effect of Salmonella, Listeria, Shigella and E. coli phage cocktails on their corresponding susceptible hosts incubated at 4°C in TSB; a) Salmonella Entritidis (C417), b) L. monocytogenes (C391), c) Shigella sonnei (C866-2) and d) E. coli 0157:H7(C899). a) b)

-Treatment 11.00 -I -^-Treatment -Control •^—Control 9.00 9.00 _l F 700 'J 7.00 _u>. u_ O o 500 — 5.00- O—) o> o o _1 _1 3.00 3.00

< 1.00 <.: 3 0 13 7 9

Days Days

c) d)

11.00 i 11.00 -I -•—Treatment -•—Treatment -*— Control -•—Control 9.00 • 9.00' _1 _i E ~ 7.00 3 7.00 • 3 u. li. o O — 5.00. ~ 5.00 CD Ol O o _l 3.00' 3.00

0 1 3 7 9 0 1 3 7 9

Days Days

Figure 4.6. Effect of Salmonella, Listeria, Shigella and E. coli phage cocktails on their corresponding susceptible hosts incubated at 25°C in TSB; a) Salmonella Enteritidis (C417), b) L. monocytogenes (C391), c) Shigella sonnei (C866-2) and d) E. coli 0157:H7(C899). 4.4.3 Potential use of the phage cocktails to control L. monocytogenes and E.

coli 0157:H7 in food

In order to examine the efficacy of the Listeria and E. coli phage cocktails to control their host pathogens in a real food system, one L. monocytogenes and one E. coli strain were used to artificially contaminate RTE and raw meat samples, respectively, and then the corresponding phage cocktails were added. The samples were then incubated under different temperatures and packaging conditions for up to 15 days in some cases.

The count of the surviving bacteria was determined at regular intervals and compared to the control, which had no phage treatment. Figures 4.7, 4.8 and 4.9 show the effect of addition of Listeria phage cocktail to contaminated RTE meat samples incubated under aerobic, vacuum and MAP conditions, respectively, at 25 and 4°C. Phage addition with incubation at cold temperature (4°C) resulted in inhibition of L. monocytogenes to undetectable levels by direct plating under all packaging conditions. When meat samples were stored aerobically at 25°C, significant reductions of the bacterial count were detected after 1, 3 and 7 days with reductions of around 2, 1.5 and 1.4 log units, respectively, compared to the level in the controls (Figure 4.7a). However, the count was nearly the same as that of the control after nine days. Adding Listeria phage cocktail to vacuum packed meat samples resulted in a significant reduction in the recovered L. monocytogenes cells by around two log units by the end of the experiment when compared to control counts (Figure 4.8a). MAP with phage cocktail treatment resulted in a significant reduction of L. monocytogenes cells in the treated meat samples by around one log unit after nine days compared with the control count (Figure 4.9a). For aerobically incubated E. coli 0157:H7 spiked samples, E. coli phage cocktail only

169 resulted in reduction of the bacterial count by less than one log unit after 12 hours

incubation at 25°C (Figure 4.10a). It was also noticed that there was no significant difference between the bacterial count in the treated samples and control samples by the end of the incubation period. When the meat samples were incubated at 10°C, significant reductions in the recovered E. coli cells from the contaminated samples have only been noticed after seven and nine days when compared with the control (Figure 4.10b). Non- detectable levels by direct plating of the recovered E. coli cells from the treated samples were obtained following incubation at 4°C for 12 days; however, after 15 days counts of the order of 1.7 CFU/gm were detected (Figure 4.10c). When randomly selected recovered colonies were picked from Oxford and TSA + ampicillin agar plates and tested against Listeria and E. coli phage cocktails, they were all sensitive to the effect of the phages and showed clear lysis.

170 a)

11.00 n -Treatment

-Control 9.00 H

3 u. O D) 5.00 H O

3.00 H

S 1.00 3 Days

b)

4.00- -Treatment -Control 3.50-

E 3.00- •5* o Z50- O) O 2.00- _l

1.50-

S 1.00- 3 Days

Figure 4.7. Effect of Listeria phage cocktail on the growth of L. monocytogenes (C391) in RTE oven roasted turkey breast incubated aerobically at 25°C (a) and 4°C (b) for nine days.

171 a)

11.00-J —•—Treatment * Control 9.00- ^~^ ? 5* 7.00- U. o W 5.00- o -I 3.00-

0 1 3 7 9 Days

b)

3.00 -Treatment -Control 2.60

S> 2.20

1.80

1.40

S 1.00 3 Days

Figure 4.8. Effect of Listeria phage cocktail on the growth of L. monocytogenes (C391) in RTE oven roasted turkey breast incubated under vacuum condition at 25°C (a) and 4°C (b) for nine days.

172 a)

9.00- • Treatment * Control

^ 7.00- f E 5>

"- 5.00 • o O) o "* 3.00-

0 1 3 7 9 Days

b)

3.50- —•—Treatment * Control

3.00- ?

i? 2.50 •

O o 1.50-

0 1 3 7 9 Days

Figure 4.9. Effect of Listeria phage cocktail on the growth of L. monocytogenes (C391) in RTE oven roasted turkey breast incubated under modified atmospheric packaging (MAP) condition at 25°C (a) and 4°C (b) for nine days.

173 a)

11.00- —•—Treatment —•—Control 9.00- ? f 7.00- LL. O — 5.00 • O o _l 3.00-

0 6 12 24 48 Hours

b)

5.00-1 —•—Treatment ^^ —*—Control

•=- 4.00- I T 3 "r 3.00-

O -1 2.00-

0 1 3 7 9 Days

c)

3.00-1 -Treatment -Control •—. 2.50 I I -I •5* U- 2.00 O O) o -1 1.50-

S 1.00

ure 4.10. Effect of E. coli phage cocktail on the growth of E. coli 0157:H7 (C899) in raw beef meat incubated aerobically at 25°C (a), 10°C (b) and 4°C (c) for 2, 9 and 15 days, respectively.

174 4.4.4 Effect of Listeria phage cocktail on biofilm formation by L.

monocytogenes

The effect of Listeria phage cocktail on biofilm formation in polystyrene microtitre plates by different concentrations of two L. monocytogenes strains (C519 and

C391) was investigated by the crystal violet staining technique. The strains were allowed to grow in the presence and absence of a cocktail of Listeria phages for two days at 30°C.

Then the biofilm was stained with crystal violet and the optical density of the developed solution after destaining was measured. The observed results revealed that Listeria phage cocktail was able to significantly reduce biofilm formation by two strains regardless of the starting bacterial concentration (Figure 4.11). Generally, for the non-phage treated controls, it was found that starting with higher concentrations (106 and 108 CFU/ml) of L monocytogenes (C391), more dense biofilms were formed (as indicated by higher

OD530nm) than for L. monocytogenes (C519) at the same concentrations. A similar trend was observed at lower bacterial concentrations, around 104 CFU/ml, where L monocytogenes (C519) formed denser biofilm when compared with L. monocytogenes

(C391). Under these conditions, it was noted that the phage cocktail effect was greater in the case of L. monocytogenes (C519) strain when the MOI was around 105. On the other hand, a MOI of approximately 10 and 10 resulted in a greater reduction in biofilm formation of L. monocytogenes (C391) strain.

175 a)

0.2 i • Control p • Treatment

Ihl1.00E+02 1.00E+04 1.00E+06 l1.00E+08 CFU/mL CFU/mL CFU/mL CFU/mL Initial bacterial concentration

b)

• Control ••• H^H • Treatment

•1.00E+04 1.00E+0•6 1.00E+0I8 CFU/mL CFU/mL CFU/mL Initial bacterial concentration

Figure 4.11. Effect of Listeria phage cocktail on biofilm formation by different concentrations of two strains of L. monocytogenes; C519 (a) and C391 (b).

176 4.5 DISCUSSION

Virulent phages have been successfully used to control growth of different strains of several foodborne pathogens in both artificial media and different food products

(Strauch et al, 2007). The use of phage cocktails to control the four targeted foodborne pathogens in this study, Salmonella, L. monocytogenes, Shigella and E. coli 0157:H7, have been reported in many studies (O'Flynn et al, 2004, Sulakvelidze et al, 2001,

Atterbury, 2009, Leverentz et al, 2003, Leverentz et al, 2004, Guenther et al, 2009,

Fiorentin et al, 2005, Leverentz et al, 2001, Goode et al, 2003, Abuladze et al, 2008,

Sharma et al, 2009). In the first part of this study, the effect of different MOI values of the various phage cocktails was investigated by measuring the optical density of the indicator bacterial cells using the Bioscreen C. The results revealed that using a high

MOI value of around 105 would guarantee complete inhibition of the tested bacteria for five days. Although phage count was not determined after each treatment, these data might suggest that lysis from without could be the proposed mechanisms for this reduction due to the application of the high MOI value. This phenomenon occurs when host cells adsorb more than 100 phages on their surfaces and the production of phage lysozyme leads to holes being formed in the bacterial cell wall through which cytoplasmic contents may escape (Abedon, 1999, Tarahovsky et al, 1994). Our data are consistent with many published studies where high MOI values caused optimum control of the target pathogen. For example, using a phage cocktail of three E. coli phages at high

MOI of around 106 resulted in control of £ coli 0157:H7 in broth culture (O'Flynn et al,

2004). In another study, a MOI of around 103 of ECP-100 E. coli phage cocktail resulted in 99.99% reduction in the recovered E. coli 0157:H7 cells from contaminated food

177 surfaces (Abuladze et al, 2008). Furthermore, using a high MOI value of approximately

10 also resulted in the elimination of Salmonella strains from melon and raw and cooked meat (Leverentz et al, 2001, Bigwood et al, 2008). On the other hand, in some tested strains; all the used MOI's were able to cause complete inhibition of the growth, such as

Listeria spp. This might be due to of the use of more than one phage, which might reduce the development of resistant mutants and result in complete control of the host pathogen.

It was stated that the use of phage cocktails against a target bacterium may be a necessary prerequisite for a successful phage therapy and biocontrol (Strauch et al, 2007).

The Bioscreen C measured only the turbidity of the tested bacteria, so counting of the recovered viable cells would provide a better indication of the effect of the phages on the growth of the host bacteria. Adding phage cocktail with incubation at 4°C reduced the growth of Salmonella, L. monocytogenes and Shigella below the detectable level by direct plating throughout the incubation period. However, E. coli 0157:H7 was present at levels of 1.7 and 2.2 log CFU/ml after 7 and 9 days, respectively, which might be due to the very slow metabolic rate at this temperature leading to the delay of the development of resistant mutants. On the other hand, the Salmonella phage cocktail did not inhibit completely the growth of the tested Salmonella strain as other phage cocktails did with their corresponding hosts at room temperature. Resistant L. monocytogenes strains might be developed after nine days incubation with phage cocktail at the same temperature.

Similarly, the increase in population of E. coli 0157:H7 (C899) strain after three days of about two log CFU/ml might be due to the development of resistant mutants against one of the used phages. However, the subsequent growth inhibition might be the result of sensitivity of these mutants to the other two phages used in the cocktail. The development

178 of phage resistant colonies in this study at 25°C was not considered to be a problem that can hinder the future use of phage in food safety applications as there is no realistic possibility for food products that can be contaminated with these targeted pathogens to be stored at room temperature for more than a few hours. There is no doubt that successful use of phages as biocontrol agents will depend on the emergence or persistence of phage- resistant mutants. So using a cocktail of broad host range phages and treatment of the products immediately before packaging and shipment were recommended to avoid the development of these mutants in the production environment (Guenther et al, 2009).

Historically, most phage work in the biocontrol area has been done in liquids with high concentrations of a monoculture of the target bacterium (Hagens and Loessner,

2010), but it is crucial to test their efficacy in actual foods as it is difficult (if not impossible) with relatively low amounts of the target pathogen to predict the behavior of the phages with its host in a food matrix. It was shown that different food matrices resulted in varied levels of phage activity against L. monocytogenes (Guenther et al,

2009). Hence, this study used RTE and raw meat as models to test the activity of Listeria and E. coli phage cocktails to control the growth of L. monocytogenes and E. coli

0157:H7, respectively. In our study we usually stared with a PFU/CFU ratio of approximately 10 by adding a high titre of phage cocktail (10 PFU/ml) to a low titre of host bacteria (103 CFU/ml). Although it might not be the case in practice, the starting bacterial concentration approximates the infectious dose for the targeted bacteria in the contaminated food and at the same time was suitable to monitor the bacterial growth by direct plating. Increasing the concentration of the applied phages (threshold of approximately 108 PFU/ml) was recommended to increase the likelihood of interaction

179 between phages and target bacteria and to be able to cover the entire available space of the targeted food matrix (Hagens and Loessner, 2010, Greer, 2005, Garcia et al, 2008).

Our results have shown that the combination of phage application and a 4°C storage temperature resulted in a reduction in numbers of recovered cells of the target from the treated samples to undetectable levels after nine days' storage under different packaging conditions. However at room temperature, which supports the faster growth of the target bacteria and represents temperature abuse for most foods, phage treatment with vacuum or MAP conditions resulted in a significant reduction of L. monocytogenes recovered from the treated samples when compared to controls. Vacuum and MAP packaging conditions have been reported to be used as additional barriers for the growth of different pathogens (Phillips, 1996, Huss, 1997). Phages have been used with other hurdles such as bacteriocin (Leverentz et al, 2003) and quaternary ammonium compound (QUATAL)

(Roy et al, 1993) to reduce L. monocytogenes populations. Our results strongly support the notion that phage application should be applied as one component of hurdle approach to maintain food safety (Garcia et al, 2008, Leverentz et al, 2003, Martinez et al, 2008,

Roy etal, 1993).

Testing phage sensitivity of the randomly selected recovered colonies from the treated samples revealed that we could not isolate any bacteria that were insensitive to either Listeria or E. coli phage cocktail. This finding indicated that the bacteria remaining in the phage-treated meat might have not acquired phage resistance but rather escaped contact with the phage particles after application. Other studies also did not detect development of resistance against phages used in biocontrol experiments, for example, in studies on L. monocytogenes in RTE meat stored for up to 13 days (Guenther et al, 2009)

180 and in cheese stored for over three weeks (Carlton et al, 2005), Campylobacter jejuni on chicken skin after 10 days (Atterbury et al, 2003b) and Salmonella Enteritidis on fresh cut fruit during a seven-day incubation period (Leverentz et al, 2001). On the other hand, another study found that the developed phage resistant strains can be reverted when phages are removed from the growth environment (O'Flynn et al, 2004). It is important to mention that some of the experimental conditions used in this study were not particularly applicable to those found in practice. For instance, RTE and raw meat are not normally stored at room temperature for long periods. However, this study attempted to investigate whether phage cocktails are able to control both existing and re-growing cells of the host pathogen that might persist for a long time. This would open up other applications (e.g. phage therapy).

Control of biofilm formation is another promising area for phage application. The

Listeria phage cocktail used in this study was able to reduce biofilm formation by two tested L. monocytogenes strains on polystyrene microtitre plates. We suggested that the presence of the phages with the bacterial cells from the beginning might have reduced the number of the cells that were available to form biofilm. Furthermore, it was suggested that phages can control biofilm formation through the action of phage hydrolytic enzymes on the exopolysaccharide matrix of the biofilm (Abedon, 2010). Our biofilm results were consistent with an earlier study when a cocktail of Listeria phage suspensions at concentrations of around 3.5 x 10 PFU/ml were at least as efficient as a 20 ppm solution of a quaternary ammonium compound (QUATAL) in reducing adherence of L. monocytogenes to stainless-steel and polypropylene surfaces (Roy et al, 1993). A synergistic effect has been noticed when both were used together. In another study,

181 modified Listeria phage was able to prevent biofilm formation by cell wall-deficient L. monocytogenes (L-forms) on stainless steel(Hibma et al, 1997). However, more detailed studies using larger number of L. monocytogenes strains of different serotypes are required to confirm that phages can prevent and remove already existing biofilm.

Moreover, it was shown that the crystal violet technique could not be used to monitor the killing effect of the phages as cells with no metabolic activity can still contribute to the total amount of biomass in the biofilm (Romanova et al, 2007). Therefore, more accurate methods need to be developed to accurately investigate the efficacy of the phages to control biofilm formation.

In conclusion, this study suggested that using a lytic phage cocktail is a promising specific, effective and environmentally friendly way to control different foodborne pathogens and produce safer food. Using these cocktails at high MOI decreased the development of resistant mutants and inhibited the growth of targeted foodborne pathogens in both artificial broth media and in foods. Moreover, they were able to reduce biofilm formation by two strains of L. monocytogenes, which opens another window for application in the food industry. Using other control methods along with the phage treatment in a hurdle technology approach would enhance the safety of the treated food products and environment.

182 Chapter 5: IMMOBILIZATION OF PHAGE COCKTAILS

ON CHARGED CELLULOSE MEMBRANES AND THEIR

BIOCONTROL APPLICATION

5.1 ABSTRACT

The ability of phages to specifically interact with and lyse their host bacteria

make them an ideal bioactive material that can be used in the immobilized form to

increase their application in research and industrial fields. Silica particles and cellulose

membranes with modified surface charges were used as supportive carriers for the tested

phages. It was found that the amount of the infective phages that could be immobilized

increased with the increase of the overall surface positive charge of the silica and TEM

images showed that phages were bound through their heads to these positively charged

silica particles. Similarly, the amount of the infective phages immobilized on the

positively charged cellulose membranes was higher than those on unmodified ones.

When Zeta potential was measured, there was a difference between T4 head and tail

components, which revealed that the heads exhibit an overall net negative charge while

the tail fibers possess an overall net positive charge. Immobilized Listeria and E. coli

phage cocktails were able to control the growth of L. monocytogenes and E. coli

0157:H7 in RTE and raw meat samples, respectively, under different temperatures and packaging conditions. Lyophilization can be used as a drying method in developing these

bioactive membranes. In conclusion, utilizing the charge difference between phage heads

This work has been partially published as: Immobilization of bacteriophages on modified silica particles. (2010) Biomaterials volume 31, issue 7, pages 1904-1910.

183 and tails may provide a simple technique to enhance oriented immobilization of different phages.

5.2 INTRODUCTION

Recently, phage have received regulatory approval for use as a safe food additive to enhance food safety in certain food products (Mclntyre et al, 2007). Given that phage can be efficacious for biocontrol of foodborne pathogens, application strategies should be optimized to be the most convenient, most economical, and least invasive to the process itself (Hagens and Loessner, 2010). Interventions can be performed at different or even multiple points in the food processing facility in order to enhance the control process and so reduce the potential for development of phage resistance. It should be mentioned that phage application could be useful at all stages of production in the classic 'farm to fork' approach throughout the entire food chain (Garcia et al, 2008). It has been suggested that phages can be added by dipping, spraying or as a liquid to a large volume of food material. These methods may not be ideal as they could be wasteful and potential inactivation of the phage particles could happen as a consequence of inclusion of other materials within the wash fluid, such as sanitizers. Moreover, if the washing fluids themselves are suitable for bacterial growth, then the potential for bacterial evolution of phage resistance might exist. When phages are added directly to a batch of food, two major problems may be encountered: dilution of phages and evolution of bacterial resistance. Addition of large numbers and volumes of phages, using phage cocktails and regular disinfection of the equipment using effective protocols might help to overcome these problems (Hagens and Loessner, 2010). These problems may also be overcome by

184 using immobilized phage for application. This will ensure that phages are applied and retained near to the surface that is being treated, thereby avoiding excessive phage wastage. This approach was identified by S. Abedon as an efficient strategy of application (Abedon, 2010).

Immobilized biologically active materials are of great importance to industry and research. The selection of the immobilization method and support depends on the nature of the bioactive material and on the application itself (Knaebel et al, 1997, Selvaraj et al,

1997, Jirku, 1999). Surface immobilization of microorganisms by adsorption has the advantage of being very simple to carry out and there is no transfer barrier as in the entrapment (encapsulation) approach for immobilization (Klein and Ziehr, 1990). The potential use of a phage-based biosorbent to detect, concentrate and identify target bacteria has been reported in several studies (Zourob and Ripp, 2010). In one approach, physical adsorption has been used recently to immobilize filamentous and Podoviridae phages on gold surfaces of surface plasmon resonance (SPR) sensors and glass substrates in order to be used as a recognition element for many targeted pathogens (Nanduri et al,

2007a, Balasubramanian et al, 2007, Handa et al, 2008). In an earlier study, Salmonella cells were captured from food matrices using Salmonella-specific phage passively immobilized on polystyrene but this resulted in a low capture efficiency (Bennett, 1997).

Furthermore, chemical biotinylation of the phage head has been used to immobilize phages on streptavidin-coated magnetic beads, but again this system only captured low numbers of Salmonella Enteritidis cells in foods (Sun, 2001). The orientation and/or the inactivation of the virus may have played a role in the low capture efficiency. Site- specific immobilization of phages was suggested as a way to orientate the immobilized

185 phages in such a way as to have tail fibers free and, thereby, increase the efficiency of capture of the target bacteria. The genetic modification of wild type T4 phage through a phage display technique has been used to give well-oriented T4 phage particles when immobilized on cellulose membranes or streptavidin-coated magnetic beads, which resulted in a high capture efficiency for E. coli cells (Tolba et al, 2010). However, this protocol was initially laborious and costly even with T4 phage, which is one of the best characterized phages with a fully sequenced and annotated genome. In addition, these modified phages exhibited a lower burst size and less infectivity than wild type T4 (Tolba et al, 2010, Minikh et al, 2010). Therefore, more research is required to establish alternative simple protocols to immobilize phages in the right orientation that can be applied to newly isolated phages with known structures but not fully sequenced.

One approach to immobilization may be to use electrostatic interaction between the phage and the support matrix. Interestingly, it was reported that the net charge on most viruses is negative and the whole T4 phage (capsid, tail and fibers) has an isoelectric point close to 4 (Archer and Liu, 2009). Furthermore, the same study suggested that capsids acquire a negative overall charge above pH 4. In an earlier study, the T7 phage head was suggested to be responsible for the overall negative charge of the phage and the tail fibers could be positively charged (Serwer and Hayes, 1982).

Cellulose is the most abundant natural polymer on earth and can be considered as an attractive matrix for immobilization mainly because of its combination of excellent physical properties and low price. Moreover, cellulose can be easily manufactured and modified according to the application purpose (Mahiout et al, 1997). Different bioactive materials such as; nitrifier and denitrifier bacteria (Sakairi et al, 1996), Rhodobacter

186 capsulatus (Sawayama et al, 1998), glucoamylase enzyme (Wu and Lia, 2008),

bacteriocins (Ming et al, 1997, Zhu et al, 2005) and antimycotic agents (Cutter, 2002)

have been immobilized on cellulose and were applied successfully in several medical,

environmental and food applications.

In this current study, we postulate that the inherent charge characteristics of

phages can be utilized to anchor them in the right orientation to appropriately modified

surfaces, which could then serve as supported anti-bacterial agents. We report below the

study of the charge difference between phage heads and tails directly and indirectly and

the possibility of using charge difference to immobilize two different phage cocktails on

cellulose membranes in order to control E. coli 0157:H7 and Listeria monocytogenes in

raw and ready-to-eat meat, respectively.

5.3 MATEREIALS AND METHODS

5.3.1 Bacteria and Bacteriophage

Tryptic Soy Broth (TSB), Tryptic Soy Agar (TSA), Tryptose Soft Agar (TSB +

0.4% agarose) and Listeria Selective Agar Base (Oxford Formulation) (Difco

Laboratories, Detroit, MI) were used in this study to grow the host bacteria, propagate the

phages and count Listeria monocytogenes after challenge experiments. The synthetic

minimal medium M9A [10% salt mixture (7% Na2HP04, 3% of KH2P04, 0.5% NaCI, 1%

NH4C1), 0.4% glucose, 0.1% CaCl2 (0.01M) and 0.1% MgS04 (0.1M)] supplemented

with 5% of a mixture of Casamino acids (Fisher Scientific, Mississauga, ON, Canada)

was used for production of T4 heads, tail and tail fiber components from a T4 mutant.

Ampicillin (100 ug/ml) was added to TSB and TSA to grow and count ampicillin-

187 resistant E. coli 0157:H7 {amp::lux) strain (C918). For propagation and titre determination of Listeria phages, after sterilization, all the media were supplemented with filter-sterilized 1.25mM CaCk per liter of the medium. The previously isolated and characterized phages described in Chapter Two and their susceptible strains of E. coli,

Salmonella, Shigella and Listeria spp. (selected from the Canadian Research Institute for

Food Safety (CRIFS) Culture Collection at the University of Guelph) (Table 5.1) were used in this study for phage propagation, immobilization and biocontrol. Wild type T4 phage (ATCC 11303B4) was used in the dryness experiment. E. coli B (ATCC 11303) was used for its propagation. Mutant T4 strains (10" and 23" ) and their propagating suppressor host bacteria {E. coli CR63) were kindly provided by Dr. Fumio Arisaki,

Department of Life Science, Graduate School and School of Bioscience and

Biotechnology, Tokyo Institute of Technology, Tokyo, Japan. E. coli 0157:H7

{amp::lux) (C918) was used to test infectivity of immobilized E. coli phages and to artificially inoculate raw meat samples. Pure cultures were obtained from -80°C frozen storage and maintained at 4°C on TSB. Cultures were re-streaked every month to maintain cell viability. Propagation of the phages with high titre was performed as previously described (Chapter 2). The three phages against L. monocytogenes and E. coli

0157:H7 were mixed together to form the phage cocktail that was used in the immobilized form for the biocontrol experiment. The titre of each phage cocktail stock was determined by preparing serial 10-fold dilutions and tested using the previously described overlay method. The phage cocktail lysate was stored at 4°C.

188 Table 5.1. Phages and bacterial strains that used for phage propagation, immobilization and biocontrol experiments

CRIFS culture Phage Bacterial Host Strain Designation collection number Salmonella SenS-AG 11 C417 En-2588 Enteritidis LinM-AG8 Listeria innocua C505 . Listeria LmoM-AG13 LJH391 4b monocytogenes Listeria LmoM-AG20 C519 l/2b monocytogenes SboM-AG3 Shigella boydii C 865-2 0126:H8, EC EcoM-AG2 E. coli 0126:H8 C761 910061 0126:H8, EC EcoM-AG3 E. coli 0126:H8 C761 910061 0157:H7,ATCC EcoM-AGlO E.coli 0157:H7 C899 43888

5.3.2 Immobilization of phages on surface modified silica particles

Silica particles (around 200 nm diameter) with surfaces of varying positive charge were synthesized and provided by Dr. Michael Brook, McMaster University (Hamilton,

ON, Canada). Briefly, the silica surface was cationically modified by adding various concentrations of surface modification agents, aminopropyl-triethoxysilane (APTS)

(Sigma-Aldrich, Oakville, ON). The electrophoretic mobility values were measured at

25°C using a Brookhaven Zeta Potential Analyzer (Holtsville, NY). The standard deviation of 10 measurements was used as an estimate of experimental variability. Three phages belonging to the Myoviridae (EcoM-AG2, SboM-AG3, LinM-AG8) and one

189 member of the Siphoviridae (SenS-AGl 1) were used in order to examine the binding of phages to the modified silica particles. Silica particles (10 ul, density of solution 0.1 g/ml of water) were added to 50 |j.l phage (of around 109 PFU/ml) in phage buffer (440 ul) in triplicate, and then the mixture was incubated overnight at room temperature with shaking. The tubes were centrifuged at 11,180 x g for 1 min (Beckman J-20 centrifuge,

Beckman Coulter Inc., Mississauga, ON, Canada). The pellet was re-suspended in phage buffer (500 ul) and centrifuged again for 1 min at 11,180 x g. The washing step was repeated 15 times. After washing, 500 ul of phage buffer were added to the pellet and swirled carefully and the suspension was used to make serial dilutions to determine the titre of phages immobilized on the pellet using the overlay technique. The electrophoretic mobility values of the silica particles with phages were also measured.

5.3.3 Immobilization of phages on positively charged and unmodified

cellulose membranes

Positively charged cellulose membranes were synthesized and provided by Dr.

Robert Pelton, McMaster University (Hamilton, ON). Regenerated cellulose membranes were purchased as dialysis membranes from Spectrum Laboratories Inc. (Spectra/Por® 4 product No: 132709, 12kDa MWCO Spectrum Laboratories, Inc., Rancho Dominguez,

CA). The membranes were cut into round pieces with a diameter of 4 cm. The obtained membranes were boiled for four to six hours in distilled water (during which time, the water was changed at least three times) to remove any preservative. Afterwards, the cellulose membranes were thoroughly rinsed with water and stored in distilled water at

4°C. The membranes were modified by adding 0.5 wt% polyvinylamine polymer (Sigma-

Aldrich, Oakville, ON) to provide the positive charge to the cellulose membrane surface.

190 A phage cocktail, consisting of the three E. coli 0157:H7 phages, was added to both the positively charged membranes and unmodified membrane. Five milliliters from the phage cocktail at different concentrations (around 103, 105, 107 and 109 PFU/ml) were added to each membrane and incubated overnight at 4°C with gentle shaking. The produced phage-carrying membranes were labeled as M-3, M-5, M-7 and M-9, according to the titre of the phage cocktail initially added to them. The membranes were removed and the amount of the remaining phages in the supernatant was determined by serial dilution and overlay technique as described previously to determine the amount of phages deposited on the membranes. The membranes were washed three times in 5 ml phage buffer. The infectivity of phages on the membranes was determined by two approaches; overlay technique and bioluminescence assay. In the overlay technique, the membranes were placed over a top agar layer inoculated with the host bacterium (E. coli 0157:H7 ampr.lux) and incubated overnight at 25°C before counting the plaques developed underneath the membranes. The other approach involved detecting the effect of these membranes on the bioluminescence produced by a luminescent strain of E. coli 0157:H7

{ampr.lux). Both the positive and normal membranes were added to 4 ml of broth medium inoculated with two different concentrations of the indicator bacteria (around 103 and 105 CFU/ml) in a Corning Costar sterile 6-well flat-bottomed polystyrene microtitre plate (Corning Inc., Corning, NY; not prepared by the manufacturer for tissue culture work). Each treatment was performed in triplicate along with controls of bacteria, media and phage buffer only. The plates were incubated at 25 °C and the bioluminescence was measured every hour for 12 hours by using a Victor2 1420 Multilable Counter (Wallac

Oy, Turku, Finland).

191 5.3.4 Investigating the overall charge difference between phage head and tail

structures

The study of the overall charge difference between phage heads and tail

components was investigated by two approaches; indirect and direct. For the indirect one, positively and negatively charged gold nanoparticles were synthesized and provided by

Dr. Robert Pelton, McMaster University, Hamilton, Ontario, Canada. One hundred

microliter of the charged gold particles were mixed with 100 ul of EcoM-AG2, SboM-

AG3, LinM-AG8 or SenS-AGll phage. The mixtures were incubated with gentle

shaking at room temperature for 24 hours before examining by TEM to determine the

sites of deposition of charged gold nanoparticles on the tested phages. The direct

approach involved the preparation of T4 heads and tail with tail fibers by using T4

mutants 10" and 23", respectively, and then the Zeta potential was measured to determine the overall charge on each preparation. The T4 mutant strains were first propagated in the

suppressor E. coli host strain, CR63, to obtain the infectious phage using the same propagation protocol mentioned in Chapter 2. The infectious phages were then used to produce non-infectious capsids and tail with tail fibers in the non-suppressor host E. coli

strain, E. coli B, modified from a previous procedure (Archer and Liu, 2009). The E. coli

B cells were grown in 1 liter of M9A broth until an ODeoo of around 0.2 was obtained and then they were infected with 10" or 23" T4 phage mutant strains at a MOI of around 1.

The cells were superinfected with the same strains after 5 and 10 min. The whole mixture was incubated for 3 hours at 37°C with gentle shaking. Cells were spun down and resuspended in 10 ml phage buffer. Chloroform (1 ml), DNasel (20 ug/ml) and RNase I

(50 ug/ml) (Invitrogen Canada, Burlington, ON) were then added to the cell suspension

192 and shaken at 37°C for 1 hr. The cell debris was removed after spinning at 5000 x g for

30 min. The produced supernatant was washed three times by centrifugation at 16000 x g for 1 h at 4°C to precipitate the particles and resuspend in 500 ul Milli-Q sterile water.

The overall charge of the prepared heads and tail with tail fibers preparations was determined by using a Zeta Potential Analyzer (Zetasizer Nano, Malvern Instruments,

Worcestershire, UK). Two separate preparations were made from each mutant and five

Zeta potential measurements were done for each preparation.

5.3.5 Transmission Electron Microscopy

The phages with cationic silica particles and gold nano particles were examined using a LEO 912AB electron microscope (Energy filtered TEM, EFTEM, LEO 912ab model operated at 100 kv, Zeiss, Germany). Five microliters of the phage/silica or phage/gold nanoparticles suspension were applied onto 200-mesh copper grids coated with formvar and allowed to stand for 1 min. The excess liquid was drawn off by filter paper and the remaining phages were negatively stained with 2% uranyl acetate for 30 s and then the excess liquid was drawn off again using filter paper. Finally, the samples were examined by transmission electron microscopy.

5.3.6 Effect of dryness on stability of phages

The stability of phages after drying was investigated using five different phages; four Myoviridae phages (Wild type T4, EcoM-AG2, SboM-AG3, LinM-AG8) and one

Siphoviridae phage (SenS-AGll). Two hundred microliters from each phage lysate of know titre were left to be air dried for 2 hours or until completely dried at 25°C and

37°C. The dried phage particles were reconstituted in 2 ml phage buffer for 30 min before

193 counting the amount of active phages in the produced phage lysate by the overlay technique and log reductions of phage counts were calculated. Wild type T4 phage was used for further experiments to examine the effect of i) leaving dried phage particles with phage buffer for a longer time, ii) the addition of polysaccharides and iii) lyophilization in the presence and absence of polysaccharides on the phage desiccation stability. The T4 air-dried particles were incubated with the phage buffer for 18 hours and the log reduction in the phage count was calculated and compared with that determined after leaving phage particles with buffer for only 30 min. Two hundred microliters of T4 phage with and without maltose (0.1 %, 0.5 %, 1 %, 2 %, 5 %), trehalose (lOOmM), or starch

(0.3 %, 1.5 %) were air-dried until complete dry and then reconstituted in 2 ml of the phage buffer for 18 hours. The log reduction in phage count was calculated and compared with that obtained after air drying T4 without addition of any polysaccharides.

Lyophilization was used as another approach for drying T4 phages. Two hundred microliters of T4 phage lysate with and without 0.5% maltose, lOOmM trehalose or 0.3 % soluble starch were incubated at -80°C for 30 minutes and then freeze-dried with a Lyph- lock 6-liter freeze-dryer (Labconco, Kansas City, MO) for 22 h at a vacuum pressure of

1.5 x 10"' Pa. The freeze-dried phage particles were reconstituted in 2 ml of the phage buffer overnight and the log reduction in the phage titre was calculated. Each treatment was done in triplicate.

- 5.3.7 Potential application of the immobilized phage cocktails on cellulose

membranes to control foodborne pathogens on meat surface

Listeria and E. coli phage cocktails were immobilized on cellulose membranes as previously described in section 5.3.3 but around 109 PFU/ml were used as the initial

194 phage titre added to the positively charged membranes. The immobilized phages were applied to ready-to-eat, oven-roasted turkey breasts and raw beef samples inoculated with

L. monocytogenes (C391) and E. coli 0157:H7 (ampr.lux) (C918) strains, respectively.

The luminescent strain of E. coli 0157:H7 was used to enable direct plating and reduce problems posed by contamination of the plates with background microflora. Meat samples were purchased on the initial day of the experiment from local grocery stores and initially screened for contamination with the targeted bacteria. Twenty five grams of the meat samples were sliced and weighed aseptically and then placed in sterile Petri-dishes and pre-equilibrated to the desired temperature. An overnight culture of each strain was diluted to approximately 10 CFU/ml. A 100 ul portion was spotted onto the surface of each piece of meat and allowed to attach for 10 minutes at room temperature. This was followed by addition of the matrix containing immobilized phages to cover the contaminated surface of the meat. The phage-free positively charged and unmodified membranes were used as controls. For Zister/a-containing samples, meat was incubated at 25°C, 10°C, 4°C and for 2, 4 and 15 days, respectively, under aerobic, modified atmospheric packaging (MAP) and vacuum conditions. The modified atmosphere was composed of approximately 1 % O2, 13 % C02 and 86 % N2, which was generated by using an AnaeroGen sachet in a sealed anaerobic jar (Oxoid, Fisher Scientific,

Mississauga, ON, Canada). The vacuum packaging was done using a Komet vacuum packaging machine (KOMET Maschinenfabrik GmbH, Plochingen, Germany) at a vacuum of 1.0 bar. The meat samples were packed in sterile polyethylene bags of 65 urn thickness (Seward Laboratory Systems Inc., Bohemia, NY) before being placed in the vacuum machine. On the other hand, E. co//-containing meat samples were incubated

195 only aerobically under the same temperatures and incubation periods. Counts were performed for both hosts after 6, 12, 24 and 48 hours for samples incubated at 25°C, 1, 2,

3 and 4 days for 10°C samples and 1, 3, 6, 9, 12 and 15 days for those incubated at 4°C.

In all cases, the bacterial viable counts (CFU/gm) were determined immediately after addition of bacteria and phage. Triplicate samples were prepared for each sampling time.

At each sampling time, the membrane was removed and the meat was transferred to a

WhirlPak bag (Nasco, Fort Atkinson, WI) containing 225 ml of 0.85 % saline solution and the sample was stomached for 1 min at 200 rpm (Seward Blender Stomacher 400 circulator, Seward Laboratory Systems Inc., NY). The liquid portion was transferred to a sterile centrifuge tube and serially diluted in sterile 0.85 % saline solution. One hundred microliters were plated in duplicate on Oxford agar plates and TSA + ampicillin plates for L. monocytogenes and E. coli 0157:H7 counts, respectively, and incubated at 37°C for 48 hours in the case of L. monocytogenes until typical Listeria colonies could be enumerated and for 24 hours for the E. coli 0157:H7 strain. The developed colonies of E. coli 0157:H7 with bioluminescence were counted using a NightOwl II LB 983 Molecular

Imager (Berthold Technologies GmbH & Co. KG, Bad Wildbad, Germany). Colonies were counted and used to calculate the bacterial numbers (CFU/gm) in each sample.

Bacterial counts were transformed to logio values. In another approach to check the efficiency of the immobilized E. coli phages to control growth of E. coli 0157:H7, the immobilized phages were added to the artificially contaminated raw meat samples and incubated at 4°C or 10°C for one week then at 30°C for 16 hours. Afterwards, the treated samples along with the controls covered with phage-free positively membranes were examined using the NightOwl to detect the bioluminescence emitted from their surfaces.

196 5.3.8 Statistical analysis

The statistical analysis of the experimental data was accomplished with SigmaPlot

Version 10.1 (Systat Software Inc., Chicago, IL). A one-way analysis of variance

(ANOVA) was performed. In all cases, statistical differences between the means were indicated by P< 0.05.

5.4 RESULTS

5.4.1 Immobilization of phages on surface modified silica particles

The surface of highly porous silica particles was modified by APTS polymer to carry a cationic charge and used as support carriers for the tested phages. The surface positive charge on the APTS-modified particles increased with increasing amounts of the surface modification reagent in the reaction mixture and the naked silica particles, which are anionic in nature, were thus transformed to highly positively charged particles (Table

5.2). Hence, the used silica particles in this experiment ranged from highly cationic through neutral to highly anionic (samples A to F) as shown by zeta potential measurements. The surface coverage reached a maximum by sample C, however, the presence of additional amines in samples D to F did not lead to observable additional charge. Four phages with different phage morphology (described in Chapter 2) were > immobilized on silica via electrostatically-facilitated physisorption. All four were adsorbed on the modified silica particles during an overnight incubation and their activity/infectivity after washing was used as a measure of success of the binding of phage to the surface. The adsorption of the four phages on APTS surfaces appeared to be sigmoidal, with an increase in adsorption between sample A and C, between which the

197 particle charge changed from negative to positive (Figure 5.1). SboM-AG3 phage showed low levels of binding to native (anionic) silica, and to silica bearing low quantities of amine, its absorptiveness increased by about 3.5 log PFU/ml when the amount of amine on the surface was increased (Figure 5.1). SenS-AGll phage showed a comparable 3.6 log PFU/ml increase in binding with increases in surface positive charge, whereas the adsorption EcoM-AG2 and LinM-AG8 phages only increased by about 2 log PFU/ml on these surfaces.

The mobilities of the APTS-modified particles with the adsorbed phages were examined and compared with the mobility values of the modified particles without the phages. As can be seen in Figure 5.2, the charges of the silica particles with and without phage binding changed from negative to positive with increasing amount of amine on the surface, and followed a sigmoidal pattern. However, the overall change in mobility was less than that of the phage-free modified silica particle itself. EcoM-AG2 and SenS-

AG11 phages had slightly higher end charges than SboM-AG3 and LinM-AG8, which were only about 0.3 to 0.4 units below them. TEM images of the adsorbed SboM-AG3 and SenS-AGll phages showed them to be immobilized through their heads to the

APTS-modified silica particles sample D (Figure 5.3).

198 Table 5.2. Amounts of surface modification agent (APTS: 3-aminopropyl triethoxysilane) added to silica particles during preparation method and the produced mobility

Amount of APTS Mobility Sample STD (mmol/g SiOa) (lO'mW1) Pure Silica 0 -3.75 0.12 A 0.04 -2.63 0.08 B 0.43 0.31 0.02 C 2.14 3.69 0.08 D 4.27 3.43 0.06 E 8.55 3.3 0.05 F 12.82 2.7 0.2

Figure 5.1. Number of infective phage on APTS modified silica particles after an overnight incubation and washing.

199 >—Phage-free particles 6 ^coM-AG2 SenS-AG11 4H —• —*~___ s-SboM-AG3 2 __-* <- LinM-AG8 ^y~~~ 5 0 o TA S-2 M ^/

•A

1 -6 Pure A B c D E F Silica Silica samples

Figure 5.2. Mobility of APTS modified silica particles with and without the addition of phage.

* %

•.;?,}

IMB^^MHH B Figure 5.3. Transmission Electron Microscope (TEM) images of A: AG11 and B: AG3 specifically immobilized through their heads on cationic, APTS-modified silica particles, sample D

200 5.4.2 Immobilization of phages on positively charged cellulose membranes

Different concentrations of an E. coli phage cocktail that contains EcoM-AG2,

EcoM-AG3 and EcoM-AGlO phage were incubated with both cellulose membranes that

had modified positively charged surfaces and unmodified ones. Significant reductions of the initially added phage titre occurred (around 2 to 2.5 log units) after interaction with

the positively charged membranes M-9 and M-5, while the presence of the unmodified

M-9 and M-5 membranes caused around a 0.5 log unit reduction in phage titre (Figure

5.4). Reduction of the phage count could not be determined for both modified and

unmodified M-3 membranes as it was below the detection limit of the overlay technique.

The membranes were washed three times in phage buffer to remove non-specifically bound phages before being used to cover a lawn of the indicator bacteria. Phage-treated,

positively-charged cellulose membranes M-9, M-7 and M-5 developed around 646, 544

and 107 plaques, respectively, which were significantly higher than those developed

under the phage-treated unmodified ones (Figure 5.5). In another approach to detect the

infectivity of the immobilized phages on both positively charged and unmodified membranes, two different concentrations of E. coli 0157:H7 (amp::lux) were added to

the membranes and their effect on bioluminescence signal development was determined.

Treatment of cellulose membranes with higher phage cocktail concentrations (M-9, M-7

and M-5) resulted in nearly complete inhibition of the bioluminescence development

from both used concentrations of bacteria (Figure 5.6 and 5.7). A marked reduction in the bioluminescence signal was observed with the positively charged M-3 sample after 10,

11 and 12 hours when the initial bacterial titre was around 103 CFU/ml (Figure 5.6a).

While the same sample (M-3) but using unmodified membrane showed signal

201 development similar to that of the control at the same initial bacterial concentration

(Figure 5.6b). Furthermore, when an inoculum of around 105 CFU/ml was used, the M-3 positively charged membrane sample showed nearly the same bioluminescence pattern as that of the control for eight hours then the signal decrease until it reached nearly the same values of samples with higher phage concentrations, M-9, M-7 and M-5 (Figure 5.7a). On the other hand, the M-3 unmodified membrane resulted in a bioluminescence pattern very similar to that of the control (Figure 5.7b).

202 11.00 n • Original phage lysate • Unmodified membrane 9.00 a Positive membrane

£ M-9 M-7 JM-L5 M-3 Phage-treated cellulose membrane samples

Figure 5.4. Reduction in the E. coli phage cocktail titre after removing both positively charged and unmodified cellulose membranes (log PFU/ml).

800 -, • Positive membrane

700 D Unmodified membrane

M-9 M-7 M-5 M-3 Phage-treated cellulose membrane samples

Figure 5.5. Number of phage plaques developed under positively charged and unmodified cellulose membranes on a lawn of E. coli 057:H7 strain.

203 a)

M-3 5000 M-5 3 -I WI-7 K, 4000 M-9 c 3000 Control a> u « 2000 E 2o 1000 m o •*- 8 9 10 Time (hours)

b)

5000n —•-M-3 ^•M, 3 -•-M-5 £ 4000- M-7 "W a> i -H-M-9 o c 3000- J* —*—Control u / (0 /. 7 ® 2000- /y SMxZ 1 y 2 1000- r " ^^ Bi o

> =^— K UH I—a—i 1 6 7 8 9 10 11 12 Time (hours)

Figure 5.6. Bioluminescent signal from E. coli 0157:H7 (amp::lux) cells grown with starting inoculum of around 10 CFU/ml in the presence of a) positively charged and b) unmodified cellulose membrane treated with different concentrations of E. coli phage cocktail. Phage-free membranes were considered as control.

204 a)

14000 -I M-3 M-5 M-7 M-9 Control

7 8 9 Time (hours)

b)

14000 M-3 M-5 M-7 M-9 Control

7 8 Time (hours)

Figure 5.7. Bioluminescent signal from E. coli 0157:H7 {amp::lux) cells grown with starting inoculum of around 105 CFU/ml in the presence of a) positively charged and b) unmodified cellulose membrane treated with different concentrations of E. coli phage cocktail. Phage-free membranes were considered as control.

205 5.4.3 Investigating the overall charge difference between phage head and tail

structures

Two approaches were followed to investigate the charge difference between heads

and tail components of the phages. When positively and negatively charged gold nanoparticles were added to the same four phages used in the silica particle experiment,

examination by TEM did not give high quality images for most of the tested phages and

showed inconclusive results for the others. However, as can be seen in figure 5.8, for

SboM-AG3 and EcoM-AG2 phages, there was a slight precipitation of negatively

charged particles on tail fibers and positively charged ones on the heads were observed, respectively. In an alternative approach, T4 head and tail components were purified and the overall charge of each preparation was determined. It was found that the overall

charge of the products of the 10" T4 mutant, which was supposed to compose mainly of

heads and tail fibers was around -5.31 ± 0.67 mV and that of 23" mutant, which was tail and tail fibers, was around 1.80 ± 0.19 mV.

5.4.4 Effect of drying on the stability of phages

One member of the Siphoviridea and four Myoviridae phages were used to

investigate the effect of drying on stability of phages. There was a significant difference between the log unit reductions of phage SenS-AGl 1 and the four myoviruses that were

air-dried at 25°C and 37°C (Figure 5.9). SenS-AGl 1 phage was only reduced by around

0.47 log units when air dried and stored at 37°C, while only a 0.04 log unit reduction in numbers was obtained at 25°C.

206 m Ilk

iJKkMx

B

Figure 5.8. Deposition of negatively charged (A) and positively charged (B) gold

nanoparticles on SboM-AG3 phage tail fibers and EcoM-AG2 phage head,

respectively.

207 Wild type T4, EcoM-AG2, SboM-AG3 and LinM-AG8 phages had log unit reductions ranging from 2.8 to 4 at 25°C. However, the reduction in phage titre ranged from 3.29 log unit for T4 phage to 4.18 log units for phage SboM-AG3 at 37°C.

Increasing the time of the reconstitution of the dried T4 phage with phage buffer for up to

18 hours resulted in about a 2.96 ± 0.27 log unit reduction, which was not significantly different from 30 min reconstitution time that resulted in around 3.05 ± 0.16 log unit reduction. Addition of 0.5 % maltose and 0.3 % starch caused a significant reduction in the number of the inactivated phages by about 0.72 and 0.76 log units, respectively, when compared with that of the untreated phage (Figure 5.10). Interestingly, increasing maltose concentration to 5 % and lOOmM trehalose caused a greater reduction in phage titre by about 1.9 log unit. When lyophilization was used to dry the phages, the T4 phage count was not significantly reduced after reconstitution in phage buffer. Adding polysaccharides did not show a significant difference in the log unit reduction when compared to untreated lyophilized T4 phage (Figure 5.11).

208 5.00 n

-.n 25°C I 37°C 25°C | 37°C 25°C | 37°C 25°C | 37°C EcoM-AG2 SboM-AG3 LinM-AG8 SenS-AG11 Phages

Figure 5.9. Effect of air drying at 25°C and 37°C on the stability of phages of different morphotypes.

y y y y y j» y .* \>- «>- *>- - ^ \* \0 <& # \« \" K \*

Figure 5.10. Effect of adding polysacharrides at different concentrations on the stability of wild type T4 phage to the air drying effect.

209 1.00 -i 0 l_ *J '•5 0) O) (0 If I"t 3 0.50- u3 •o i_ O) o "* 0.00 • 1 I II II T4 T4 + Maltose T4 + Trehalose T4 + Soluble (0.5%) (100mM) starch (0.3%)

Figure 5.11. Effect of lyophilization, as a method of drying, on the stability of T4 phage with and without polysaccharides.

5.4.5 Potential application of the immobilized phage cocktails on positively

charged cellulose membranes to control foodborne pathogens on meat

surfaces

Listeria and E. coli phage cocktails immobilized on positively charged cellulose membranes were examined to control growth of L. monocytogenes and E. coli 0157:H7

(amp::lux) in ready-to-eat (RTE) and raw meats, respectively, under different storage temperatures and packaging conditions. The bacterial count was determined at regular time intervals. In general, it was found that phage-free positively charged membranes had no effect on the growth of the bacteria in all tested cases when compared to phage-free unmodified membranes. When ZJ^ma-contaminated RTE oven roasted turkey breast meat samples were incubated aerobically at 25 °C in the presence of immobilized Listeria

210 phage cocktail, the Listeria count was reduced by less than one log unit after 6, 12 and 48 hours and by around 1.4 log units after 24 hours when compared to the number of the recovered cells from the control samples (Figure 5.12a). At 10°C, count reductions of less than one log unit were detected for samples incubated aerobically for four days

(Figure 5.12b). Interestingly, the growth of L. monocytogenes was below the detection limit by direct plating of the treated samples that were incubated for six days at 4°C

(Figure 5.12c). However, 0.5, 0.6 and 1.2 log unit reductions were detected after 9, 12 and 15 days, respectively, under the same conditions. Incubating treated meat samples under modified atmosphere packaging (MAP) conditions at 25 °C showed reductions of less than one log unit after six hours and more than a log unit for the rest of the sampling times (Figure 5.13a). The highest reduction was after 24 hours when a 1.37 log unit reduction in count was observed. Incubation at 10°C under the same conditions resulted in reducing the growth to an undetectable level after one day and about a 0.8 log unit reduction after four days. Reduction of L. monocytogenes growth to below the detection limit was also noticed when MAP packed meat samples treated with the immobilized

Listeria phage cocktail were incubated at 4°C for one day (Figure 5.13c). Approximately

2.2, 1.48 and 1.7 log unit reductions in Listeria counts were also observed after 9, 12 and

15 days, respectively. Packing treated RTE meat samples under vacuum conditions at

25°C in the presence of the immobilized Listeria phage cocktail resulted in significant reductions in populations of Listeria at all sampling times with around one log unit reduction after 6 and 12 hours and 1.79 and 1.63 log unit reductions after 24 and 48 hours, respectively (Figure 5.14a). At 10°C, the number of recovered cells was below the detection limit after one day and the count was reduced by 2.71 log units after 4 days

211 (Figure 5.14b). While incubation at 4°C, undetectable levels of L. monocytogenes were found in the turkey breast after 3 days and further incubation to 12 and 15 days produced a 4 log unit reductions in pathogen levels (Figure 5.14c).

The immobilized E. coli phage cocktail resulted in a non-significant reduction in the number of the recovered E. coli 0157:H7 {ampr.lux) cells at all sampling times from artificially inoculated raw beef samples incubated aerobically at 25°C (Figure 5.15a).

However, storing meat samples at 10°C resulted in a reduction of less than a log unit after four days (Figure 5.15b). The most obvious effect was noticed at 4°C where the number of target bacteria was reduced significantly by about one log unit after 6 and 9 days and below the detection limit by direct plating after 12 and 15 days (Figure 5.15c). To visually investigate the ability of the immobilized E. coli phage cocktail to control growth of E. coli 0157:H7 {amp::lux) on raw meat surfaces, images were taken for the bioluminescence developed on the meat surface after incubation at 10°C and 4°C. As can be seen in figure 5.16, the presence of immobilized phage cocktail resulted in reduction of bioluminescence activity on the treated meat surfaces.

212 a)

11.00 -Immobilized phage -Positively membrane 9.00 Unmodified membrane E f 7.00 u O) 5.00 O _l

3.00-

S1.00 6 12 24 48 Time (hours)

b)

9.00-I —•—Immobilized phage —a—Positively membrane Unmodified membrane _ 7.00-

•5" fc 5.00. '* ^^^ o J-; "" 3 00- fV^^

DO D1 D2 D3 D4 Time (Days)

c)

-Immobilized phage -Positive membrane Unmodified membrane 5.00 f,0H §> 3.00

<1.00 D3 D6 D9 D15 Time (Days)

Figure 5.12. Effect of the immobilized Listeria phage cocktail on growth of Listeria monocytogenes C391 on RTE oven roasted turkey breast samples incubated aerobically at a) 25°C, b) 10°C and c) 4°C. Phage-free positively charged and unmodified cellulose membranes were used as controls.

213 a)

-Immobilized phage -Positive membrane Unmodified membrane 9.00 n

g 7.00 •5* 3 u 5.00 o 3.00

£1.00 6 12 24 48 Time (hours)

b)

—•— Immobilized phage • Positive membrane 6.00- Unmodified membrane I4"' g.3.00-

2.00-

0 D1 D2 D3 D4 Time (Days)

c)

-Immobilized phage 6.00 -Positive membrane Unmodified membrane 5.00 E TO Z> 4.00 ti. o g> 3 00 _l

2.00 • i *1.00 D3 D6 D9 D12 D15 Time (Days)

Figure 5.13. Effect of the immobilized Listeria phage cocktail on growth of Listeria monocytogenes C391 on RTE oven roasted turkey breast samples incubated under modified atmospheric packaging conditions at a) 25°C, b) 10°C and c) 4°C. Phage-free positively charged and unmodified cellulose membranes were used as controls.

214 a)

-Immobilized phage -Positive membrane Unmodified membrane 11

E 9 3 i M o 3

S 1 12 Time (hours)

b)

« Immobilized phage 8.00-I • Positive membrane 7.00 • ,.'" Unmodified membrane

E 6.00 • r* = 5.00 • o ^^ o -1 3.00' 2.00' K^^/ 0 D1 D2 D3 D4 Time (Days)

C)

-Immobilized phage 8.00 - Positive membrane 7.00 Unmodified membrane E 6.00 | 500 O) 4 00

5 ,m 3.00 2.00 <1.00 Time (Days)

Figure 5.14. Effect of the immobilized Listeria phage cocktail on growth of Listeria monocytogenes C391 on RTE oven roasted turkey breast samples incubated under vacuum packaging conditions at a) 25°C, b) 10°C and c) 4°C. Phage-free positively charged and unmodified cellulose membranes were used as controls.

215 a)

9.00 i —•—Immobilized phage - —"—Positive membrane Unmodified membrane 7 00 • I y £ 5.00-

"* 3.00- J,

0 6 12 24 48 Time (hours)

b)

5.00 -Immobilized phage 4.50 -Positive membrane 4.00 Unmodified membrane 3.50 3.00 CFU/g m 2.50

Lo g 2.00 1.50 < 1.00 DO D1 D2 D3 D4 Time (days)

c)

-Immobilized phage -Positive membrane Unmodified membrane

D3 D6 D9 D15 Time (days)

Figure 5.15. Effect of the immobilized E. coli phage cocktail on growth of E. coli 0157:H7 (amp::/wx) C918 on raw beef samples incubated aerobically at a) 25°C, b) 10°C and c) 4°C. Phage-free positively charged and unmodified cellulose membranes were used as controls.

216 5.16. Bioluminescence activity of E. coli 0157:H7 (amp::lux) on the surface of raw beef incubated for one week at 10°C (a and b) and 4°C (c and d) and then at 30°C for 16 hours. The inoculated meat samples were covered by immobilized E. coli phage cocktail on positively charged cellulose membranes (b and d) and phage-free unmodified cellulose membranes (a and c). The blue color represents low degree of bioluminescence while red represents the highest level of bioluminescence.

217

i 5.5 DISCUSSION

Regulatory acceptance of the use of phage to control foodborne pathogens has triggered the search for new applications for these natural bacterial killers using different strategies in order to improve consumer and industry acceptance of the technology. One strategy is to immobilize phages on to a carrier material, which has been used successfully in detection and phage therapy applications (Zourob and Ripp, 2010,

Kropinski, 2006). Developing a technique for achieving oriented immobilization of phages through their heads would enhance their capture efficiency by reducing the amount of non-specific binding to the matrix.

In this study, the charge difference within the external phage structure provides a strategy to immobilize phage on surfaces through electrostatic interactions. In order to investigate that, modified silica particles that carry different amounts of surface charge were used as supportive carriers for morphologically different phages. It was anticipated that, when the tail fibers are positively charged, phage should preferentially physisorb

'tail down' on anionic surfaces (Fig. 5.17A), whereas a cationic surface should preferentially lead to 'head down' phage association with the surface, leaving the bacteria-binding tails freely accessible to interact with bacterial cell phage-receptors

(Figure. 5.17B). Phage binding to the surface would continue until saturation was achieved (Figure 5.17C). This hypothesis was supported by the results of this study, which showed that an increase in the overall surface positive charge of the silica beads led to an increase in phage binding until the silica surface was saturated with amine groups (positive charges), at which point the amount of active phage on the material

218 became nearly constant. Studying the change in the surface charge of the silica particles carrying phages indicated that the surface positive charge is diluted to a small degree by the presence of the associated phage, which might suggest that the adsorption of phage was driven by stabilization of surface charge. However, the type of phage used seems to have little influence on the overall charge of the phage/silica complexes.

non-specific adsorption

Figure 5.17. Model showing modes of electrostatic interaction between phage and

charged silica particles; A: tail down and non-specific adsorption, B: head

down, C: head down saturated surface

On the other hand, the presence of infective phages on the slightly positive and negative silica particles suggested that adsorption of phages in the right orientation might not be purely electrostatic, but rather a combination of various types of interactions. In spite of the poor quality of the TEM images of the phage/silica combination, which might be due to the chemicals involved in the silica synthesis and modification, two phages were found to be attached to the positively charged silica beads from their head. This also might support the idea that the tested phages might have anionic head groups that were

219 associated with the cationic silica particle. This is consistent with recent reports, which

showed that T4 phage heads were aggregated at pH 5.6 and pH 7.5 on aminosilanized

substrates, where capsids behave as negatively charged entities and electrostatically

attracted to a positively charged surface (Archer and Liu, 2009). Serwer and Hayes in

1982 also suggested that the T7 phage head carries an overall negative charge and tail

fibres might be positively charged (Serwer and Hayes, 1982).

When an E. coli phage cocktail was immobilized on modified and unmodified

cellulose membranes using the charge difference approach, the initial number of phages

attracted to the positively charged membranes was higher than observed with unmodified

surfaces, and this was most apparent with M-9 and M-5 phage treated membranes.

Furthermore, these modified membranes developed a higher number of plaques over a

lawn of the indicator bacteria when compared with unmodified surfaces. The

development of plaques underneath the unmodified membranes indicated that portion of

the immobilized phages in the right orientation was simply due to physical adsorption

without totally depending on the charge. However, these results indicated that the charge

modification obviously enhanced the binding of more infective phages (i.e. in the right

orientation) among the total number of phages immobilized. The charge effect was

obvious when the positively charged membranes treated with a low E. coli phage cocktail

•a

concentration (around 10 PFU/ml) were able to reduce the bioluminescence signal from

E. coli 0157:H7 {amp::lux) when compared to the effect of using an unmodified

membrane. The nearly similar inhibitory effect achieved when positively charged and unmodified membranes were exposed initially to a high phage titre suggests that the

amount of the immobilized phages in the right orientation on unmodified membranes was

220 enough to provide sufficient MOI to achieve an equivalent level of cell lysis. The used phage cocktail in this experiment was reported to be effective to control the growth of E. coli 0157:H7 at low MOI values (Chapter 4). This emphasized that charge directed

immobilization might be helpful in less virulent phages. Physical adsorption has been used successfully to immobilize IG40 and Lm P4:A8 filamentous phages on gold

surfaces of a quartz crystal microbalance (QCM) biosensor and a surface plasmon resonance (SPR) sensor surface to detect E. coli and L. monocytogenes, respectively

(Nanduri et al, 2007a, Nanduri et al, 2007b). Another filamentous phage specific to

Salmonella Typhimurium has been physically adsorbed to the surface of a magnetoelastic

sensor and used as a biorecognition element for detection of its host cells at different concentrations (Lakshmanan et al, 2007).

In light of the results obtained, we investigated the charge difference between phage heads and tail structures. Unfortunately, using charged, gold nanoparticles to determine the charge difference of heads and tail components of phage did not give conclusive results, which may be due to the effect on the quality of the images, integrity of tested phages and adsorption of the nanoparticles on phage surfaces. This might have been due to the different chemicals used in the production of the charged gold nanoparticles. On the other hand, Zeta potential results revealed that the overall charge of the T4 heads + tail fibers preparation was negative, while a preparation containing only tails and tail fibers was positively charged. The pKa values for major T4 head proteins were found to have a range between 4.62 and 6.63, while that of the tail sheath was 4.80 with a range between 5.21 and 9.76 being reported for tail fiber proteins (Cummings et al, 1970, Showe and Onorato, 1978, Karam et al, 1994). These earlier data and the

221 obtained Zeta potential results in this study supported our conclusion that phage heads exert a net negative charge, which can interact electrostatically with positively charged surfaces. Furthermore, bacterial cell surfaces possess a net negative electrostatic charge on the outer cell envelope due to the ionization of phosphoryl and carboxylate groups which are exposed to the extracellular environment (Wilson et al, 2001). This would suggest that the presence of positively charged phage tail fiber proteins would potentially help in the initial attraction of the phage to its host cell wall before receptor interaction.

These modified cellulose membranes that contain immobilized phages were able to control the growth of L. monocytogenes and E. coli 0157:H7 in meat samples incubated at different temperatures and under different packaging conditions. The presence of the immobilized Listeria phage cocktail in MAP and vacuum packed RTE meat samples resulted in obviously higher reduction in the growth of the target strain in these samples than the reduction observed for aerobically stored samples. A new proposed Health Canada Listeria policy in ready-to-eat foods (http://www.hc-sc.gc.ca/fn- an/consultation/init/_listeria/draft-ebauche-eng.php) recommends the application of post- process lethality treatment that can achieve more than 1-log reduction in L. monocytogenes numbers and/or an antimicrobial agent or process resulting in no more than a 2-log increase of the L. monocytogenes during the shelf-life of the RTE food product. The results of the current study indicate that immobilized Listeria phages can be employed in RTE meats packaged under vacuum or MAP to limit growth of L. monocytogenes to less than 1 log-cycle in RTE foods throughout their shelf-lives. This confirms that immobilized phage technology could be used to comply with the revised

Health Canada policy.

222 An immobilized E. coli phage cocktail produced a reduction of E. coli 0157:H7 to undetectable levels when meat was stored at 4°C for 12 and 15 days. The bioluminescence activity of this strain on the meat surface was also remarkably reduced in the presence of the immobilized phage cocktail. This could be explained due to either killing activity of this phage cocktail or due to the development of phage-resistant bacteria which were not able to express most of their genes and as a result might become less virulent. O'Flynn indicated that the developed phage-resistant E. coli 0157:H7 strains are less virulent and had different morphology than the original sensitive strain

(O'Flynn et al, 2004).

Although immobilized bacteriocin and antimycotic agents on cellulose membranes have been used to inhibit the growth of Listeria monocytogenes on ham, turkey breast meat and beef (Ming et al, 1997, Zhu et al, 2005) and to increase shelf life of produce and cheese items (Cutter, 2002), it is the first time, to the best of our knowledge, that phages immobilized on cellulose membranes have been used to control foodborne pathogens in foods. Although there is no available information on the principle of manufacture, the Eliava Institute in Georgia is using a nearly similar application strategy in phage therapy of infected wounds by using PhagoBioDerm (Phage

International, USA), in which phages are impregnated on a biodegradable, polymer-based membrane along with ciprofloxacin, benzocaine and alpha-chymotrypsin (Jikia et al,

2005). This product was able to heal wounds infected with multidrug-resistant

Staphylococcus aureus within seven days. Phages have also been immobilized on nylon strips and this product was found to be effective against most of the major epidemic methicillin-resistant Staphylococcus aureus strains (http://www.newsmedical.net/?id=

223 8938). Furthermore, it is the first time, to the best of our knowledge to use the charge difference approach to adsorb phages in the right orientation.

In order to assist the commercial manufacture of this bioactive membrane, this study investigated the effect of dryness and the use of different ways of drying on phage stability. Interestingly, the tested Siphoviridae phage was significantly more tolerant to the effect of air-drying than the tested members of the Myoviridae. This might be explained due to the effect on the contractile sheath, which plays a critical role in the infectivity of Myoviridae phages (Guttman et al, 2005). Increasing the time of exposure of the air-dried phage with phage buffer before counting did not help to reactivate the T4 phage. In other words, this study suggests that desiccated phages irreversibly lose infectivity. Adding polysaccharides such as 0.5 % maltose or 0.3% starch enhanced significantly the tolerance of T4 phage to the air drying effect. This is consistent with previous reports that mentioned the usefulness of different polysaccharides in enhancing the stability of various biologically active compounds against dessication (Leslie et al,

1995, Esimbekova et al, 2007, Izutsu et al, 2004). This study found that drying using freeze drying or lyophilization could be considered as an efficient method to dry phages with little decrease in the phage activity, so would help in developing these bioactive membranes. The same technique has been also used for encapsulation of Staphylococcus aureus and Pseudomonas aeruginosa phages without affecting their lytic activity

(Puapermpoonsiri et al, 2010, Puapermpoonsiri et al, 2009)

Although further studies are needed to develop a final commercial product, these results suggest that it is possible to use the charge difference between phage heads and tail fibers to specifically immobilize phages through their heads leaving tail fibers free to

224 capture target bacteria and result in infection. The developed bioactive membranes were able to successfully reduce L. monocytogenes and E. coli 0157:H7 strains in a real food system. This proposed approach can help to broaden phage application not only to enhance food safety but also in many fields such as phage therapy in humans and animals.

225 Chapter 6: CONCLUSIONS AND FUTURE DIRECTIONS

6.1 Thesis summary and general conclusion

Bacteriophages (phages, bacterial viruses) infect members of the superkingdom

Bacteria. There are an estimated 10 phages in the biosphere, making them the most

abundant form of life on the planet (Kutter and Sulakvelidze, 2005). As viruses, they do not possess their own metabolism, but must rely on the cellular mechanisms of a host cell

in order to complete their life cycle. The specificity of interaction of a phage with its host

cell can be exploited to detect and control pathogenic bacteria. In this matter, many

studies have been described in the literature regarding application of lytic phages to

control different foodborne pathogens in various food products and environments.

Immobilization of phages might be an alternative way of phage application to the targeted surfaces and it would generally broaden the application of phages in areas other than the food industry. Therefore, this research focused on the isolation of stable lytic phages and development of a protocol to immobilize these phages on membranes for application in foods.

More than one hundred phages were isolated from different environmental

samples against different strains of four major foodborne pathogens; Escherichia coli

0157:H7, Salmonella, Listeria monocyogenes and Shigella. Traditional methods used for phage specificity testing including soft agar overlay and spot testing are very laborious and time-consuming. We used a turbidimetric method, the Bioscreen C system and associated Biolink software, in high throughput format to monitor the lytic activity of the isolated phages on different strains of food borne pathogens. This approach provided both

226 host range and quantitative lysis data simultaneously, which resulted in the rapid development of a host range pattern for each isolated phage. Based on these results, the three broadest host range phages were selected against each targeted pathogen. They were of different morphology and related morphologically to previously isolated phage

species, such as T4, A511, Jersey, Vil and Sfv. When representatives of these five common morphotypes were tested by a one-step growth curve experiment, the results revealed that all of them had either high burst size with long latent period or lower burst

size with short latent period. The best temperature for the isolated phage storage was 4°C.

One phage, EcoM-AGlO, was affected by storage at 25°C by showing around 2 log units reduction in its count after 24 h. Some of the isolated phages were found to be resistant to the tested high temperatures and had a pH stability range from pH 5 to pH 9. Moreover, two of the isolated phages (EcoM-AG3 and LmoM-AG20) were stable at higher alkaline pH. Defining these stability criteria is essential to define the area of application of these phages for biocontrol purposes. Some environmental conditions such as acidic pH and high salinity might affect the stability and/or adsorption of the tested phages on their hosts as revealed by the inhibition of lysis on lawns of these bacteria under these conditions. However, most of the tested phages were able to produce lysis on their

susceptible hosts under a variety of environmental conditions, which would increase the possibility of application of these phages on different food products packed and manufactured under different environments. Interestingly, four ambivalent phages;

EcoM-AG2, EcoM-AG3, EcoM-AGlO and StyM-AG16, were detected among the twelve

selected phages, which were able to infect E. coli 0157:H7, Shigella sonnei and

Salmonella Typhimurium. The isolated phages in this study with their broad host range

227 and ambivalency would be very useful for biocontrol in the food industry. Surprisingly, the use of the Shigella phages did not result in the appearance of any bacteriophage insensitive mutants (BIM's) when tested on Shigella strains. However, most of the other selected phages resulted in the development of low frequency of BIM in their susceptible hosts.

Sequencing of phages is a highly recommended step for detection of pathogenic and/or lysogenic genes in phages intended to be used for food application and for seeking governmental approval. In order to prove the feasibility of this concept, this study selected one of the isolated phages to be sequenced. One of the isolated phages, SboM-

AG3, is similar to phage Vil of Salmonella Typhi and a series of phages of Vibrio cholerae and Rhizobium meliloti. This is the first time that a Vi-type phage for Shigella has been described and the first Vil-like phage that has been sequenced. Sequencing helped to define its taxonomical position among phages and also to examine the presence of virulence and/or lysogeny-promoting genes in its genome. Fifty six of the identified protein products of SboM-AG3 phage genome showed homology to similar proteins from members of the Teequatrovirinae (Lavigne et al, 2009). From the taxonomic point of view, these results clearly indicate that SboM-AG3 phage is part of the "T4 superfamily".

Most importantly, the genome did not contain any homologue which would indicate that the phage is temperate. This confirmed the importance of phage sequencing and genome analysis before commercial application.

This part of the research provided characterization information for the isolated selected phages and provided evidence that using cocktails of these phages would be a good strategy when controlling their hosts in food. Using phage cocktails may result in a

228 reduction in the appearance of resistant mutants and an increase in the range of host bacteria that could be effectively infected. Therefore, checking the ability of these cocktails to control targeted pathogens in artificial media and in real food systems was necessary to determine their infectivity in different environments. Again, the turbidimetric method involving the Bioscreen C was used to determine the effect of different multiplicity of infection (MOI) values on the growth of different strains of the targeted bacterial hosts. As was expected, using a relatively high MOI value (of around

105) resulted in complete inhibition of the growth of all ten strains tested. However, in some cases as with the three tested Listeria strains, all the MOI values used were effective against these strains and caused complete inhibition of the growth for 5 days.

These results supported the use of phage cocktails in future experiments with a MOI value of around 105 to achieve successful biocontrol.

The four phage-cocktails were used against their respective hosts in TSB medium and the recovered bacterial cells were enumerated at regular intervals during the incubation period to determine the efficacy of the phage cocktails. At 4°C, phage cocktails reduced the growth of Salmonella, L. monocytogenes and Shigella below the detectable level by direct plating during nine days of incubation. The Salmonella phage cocktail could not inhibit growth of the tested Salmonella strain when grown at 25°C.

Interestingly, L. monocytogenes phage-resistant mutants were developed only after nine days incubation at the same temperature. However, the E. coli 0157:H7 strain was inhibited for one day and then was able to grow by about 2 log CFU/ml before the growth was inhibited again after 7 and 9 days. Once more, these results supported the possibility of using these phage cocktails for biocontrol.

229 To investigate the behavior of the Listeria and E. coli phage cocktails in real food situations, a high titre of phage cocktail (10 PFU/ml) was added to ready-to-eat and raw meat samples spiked with a low concentration of host bacteria (around 103 CFU/ml).

Incubation of the treated samples at 4°C under different packaging conditions reduced populations of the tested L. monocytogenes and E. coli 0157:H7 strains to undetectable levels. Moreover, adding more hurdles such as MAP and vacuum packaging with phage treatment resulted in significant reductions in the recovered L. monocytogenes cells from the treated samples when compared to controls at 25°C. Interestingly, random selection of the recovered bacterial strains did not detect any resistant mutants, which indicated that the bacteria remaining in the phage-treated meat might have not acquired phage resistance but rather escaped contact with the phage particles after application.

As biofilm formation is one of the major sources of food contamination, the last experiment described in Chapter 2 was to investigate the efficacy of a phage cocktail to control biofilm formation. Although a more appropriate technique than crystal violet staining should be used in the future to confirm the activity of the biofilm cells, the preliminary results revealed that the Listeria phage cocktail was able to significantly reduce biofilm formation by two L. monocytogenes strains in a polystyrene microtitre plate.

This part of the research demonstrated that cocktails of virulent phages such as those isolated in this study could be very effective for specific control of foodborne pathogens in both broth media and contaminated food especially if there is another hurdle(s) applied to ensure complete killing of target pathogens and, thus, enhance food safety. However, in these studies the phages were added directly to the food, which might

230 elicit consumer concerns about adding phages to their food. Also, in order to cover the whole surface of the food with phages to increase the chances of meeting with target host bacteria, a large number of phages is needed to be added to the food surface, which results in wastage and increased costs. Therefore, a second strategy for application of phages by using cellulose membranes carrying these phages in the right orientation was proposed.

The possibility of using electrostatic binding through charge difference to enhance the specific binding of phages through their heads and leaving tail fibers free for infecting targeted bacteria was studied in the last part of the present research. Modified silica particles that carry different intensities of surface charge were used as supportive carriers for four morphologically different phages. Interestingly, the number of the infective phages present on the surface increased with an increase of the overall surface positive charge of the silica particles. Moreover, TEM images showed that phages were bound through their heads to the positively charged surface. This led to the suggestion that these phages might have anionic head groups that were able to bind electrostatically to the cationic silica particles, leaving the positively charged tail fibers free to interact with bacterial cells. Similar results were obtained when positively charged cellulose membranes were used as the support matrix for binding E. coli phage cocktail. The number of the infective phages immobilized on the positively charged cellulose membrane was higher than those on unmodified ones. In both cases, it was clear that other forces played a role in binding the tested phages in the right orientation as indicated by the activity of immobilized phages on silica particles with slightly positive or negative charges and the unmodified cellulose membrane. The Zeta potential difference between

231 T4 head and tail components were measured and it was confirmed that the head had an overall negative charge while the tail fibers were net positively charged. All these results

supported the hypothesis of using charge difference between phage heads and tail fibers to develop immobilization strategies that resulted in selective orientation whereby there were greater numbers of infective phages among the total immobilized phage population.

Therefore, the last component of this thesis was to examine the ability of these bioactive membranes to control foodborne pathogens in real food systems. Immobilized Listeria and E. coli phage cocktails were able to control the growth of L. monocytogenes and E. coli 0157:H7 in RTE and raw meat samples, respectively, under different temperatures and packaging conditions. However, the effect of drying on phages was deemed to be important for commercial production of these bioactive membranes. The tested

Siphoviridae phage (SenS-AGll) was more tolerant to desiccation than the tested members of the Myoviridae (EcoM-AG3, LinM-AG8, SboM-AG3). Moreover, adding

0.5 % maltose or 0.3 % starch increased the stability of T4 phages to drying. Freeze- drying or lyophilization proved less harmful to the phage than air drying and so may be used for developing these bioactive membranes.

In conclusion, this thesis describes the successful isolation of lytic phages that can be considered as potential candidates for biocontrol applications. Moreover, it describes the development of a novel, simple, oriented phage immobilization technique that may be applied to newly isolated phages without the need for laborious and time consuming genetic modification. Based on this research, phage applications can be developed that may alleviate consumer concerns about consumption of "viruses" in their food and that can be used to broaden phage applications.

232 6.2 Future research

The major limitation to this immobilization approach for phages is that it depends on the electrostatic binding and physical adsorption of phages to the cellulose membranes, which can be easily affected by the external environment and the techniques used in the manufacture and application of these membranes. This project provided a proof of concept of using charge difference to direct phages to be immobilized in higher amounts through their heads to positively charged supportive materials. Hence, studies are still needed to investigate other immobilization techniques such as electrospinining

(Salalha et al, 2006) to hold the phages tightly to the membranes and only use the charge difference concept for the initial steps of deposition of phages on the membranes in the right orientation. Actually, substantial work is required before commercialization of this product. For instance, a greater understanding of the factors controlling the charge difference between phage heads and tail fiber proteins is required to extend the future application of this technique to all phages. Furthermore, optimization of the cellulose membrane, degree of modification by positively charged polymer and the amount of the initially added phages to this modified membrane would be beneficial to increase efficiency of these bioactive membranes to control target pathogen. Moreover, the possibility of using more than one phage cocktail to control the growth of more than one pathogen would also increase the value of these membranes. This should be followed by testing the stability and infectivity of these bioactive membranes under different conditions. Application of these membranes under practically relevant conditions is crucial and necessary before routine use of these bioactive membranes. Due to the diverse composition of the food matrices and nature of application areas, each type of food and

233 area of application represent a challenge that has to be investigated on a case-by-case basis. In addition, lower starting concentration of the targeted pathogen than used in this study to contaminate food could be used to simulate the amounts that might exist after food processing under the hygienic standards. In this case, more sensitive detection technique should be involved, such as real time quantitative PCR, to monitor the change in the bacterial count. Furthermore, the proposed technique of immobilization could be used to immobilize phages on gold surfaces of a quartz crystal microbalance (QCM) biosensor and the surface plasmon resonance (SPR) sensor surface to detect various pathogenic bacteria.

Regarding the immobilized bioactive material, phages, due to perceived consumer concerns and to facilitate governmental approval for food and therapy applications, future work should be directed, in my opinion, to investigate phage purification techniques since

CsCb and glycerol density gradients purification approaches are too costly. The recent introduction of column separation using CIM monolith technology (http:// www.biaseparations.com/sp/481 /cim-monolith-tecnnology-features-and-benefits) might provide a fast effective and inexpensive method to purify industrial levels of specific phages. Furthermore, studying the virulence of the developed phage-resistant bacterial strains would help to understand the side effects of phage applications. Most importantly, complete characterization by sequencing is a critical requirement if we are targeting governmental approval for food application or human and animal treatment. This will confirm the presence or absence of any pathogenic or lysogeny-associated (integrase, repressor) genes in the used phage genomes to ensure that this product will not add more problematic strains to the currently emerging pathogenic ones. In addition, all phages

234 should be tested for their ability to transduce their host bacteria. Moreover, testing consumption of these phages by animals would provide more value to the safe application of these phages. Finally, investigating the possibility of using some phage products such as holin, lysin or hydrolyase enzymes as an alternative to whole phage for biocontrol purposes would minimize the requirements for governmental approval of this organism as a biocontrol tool.

Although, phages have been used as therapeutic agents for many years in Eastern

Europe, the development of phage applications is in its infancy in the western world.

Therefore, most of the information about phages and their taxonomy, mechanisms of interaction with different hosts, overcoming resistance mechanisms of their hosts and different strategies of application are not yet completely studied. Isolation and extensive characterization of more phages in addition to more application research would definitely help to understand this new emerging biocontrol tool and apply it to mitigate the action of different foodborne pathogens and enhance food safety.

In addition, specific viruses have been identified to be able to infect parasitic protozoa such as Giardia spp. (Wang and Wang, 1991), so isolation of these viruses and testing their killing ability of these parasites would give another potential bioactive material that can be immobilized and used to enhance food and water safety.

235 REFERENCES

Abedon, S.T. (2010) Bacteriophages and Biofilms: Ecology, Phage Therapy, Plaques (In press).

Abedon, S.T. (1999) Bacteriophage T4 resistance to lysis-inhibition collapse. Genetics Research 74: 1-11.

Abedon, S.T. (1989) Selection for bacteriophage latent period length by bacterial density: A theoretical examination. Microb Ecol 18: 79-88.

Abuladze, T., Li, M., Menetrez, M.Y., Dean, T., Senecal, A., and Sulakvelidze, A. (2008) Bacteriophages Reduce Experimental Contamination of Hard Surfaces, Tomato, Spinach, Broccoli, and Ground Beef by Escherichia coli 0157:H7. Appl Environ Microbiol 74: 6230-6238.

Acheson, D.W.K. (2001) Shigella. In: Guide to Foodborne Pathogens. R. Labbe, and S. Garcia (eds.) John Wiley and Sons, New York, NY. pp. 193-200.

Ackermann, H. (2009) Phage classification and characterization. In: Bacteriophages. M.R.J. Clokie, and A.M. Kropinski (eds.) Humana Press, pp. 127-140.

Ackermann, H.W. (2007) 5500 Phages examined in the electron microscope. Archives of Virology 152: 227-243.

Ackermann, H.W. (2005) Bacteriophage classification. In: Bacteriophages : biology and applications. E. Kutter, and A. Sulakvelidze (eds.) CRC Press, Boca Raton, FL. pp. 67.

Ackermann, H.W. (2003) Bacteriophage observations and evolution. Res Microbiol 154:245-251.

Ackermann, H.W. (2001) Frequency of morphological phage descriptions in the year 2000. Archives of Virology 146: 843-857.

Ackermann, H.W. (1999) Tailed bacteriophages: the order Caudoviales. Adv. Virus Res. 51: 135-201.

236 Ackermann, H.W. (1996) Frequency of morphological phage descriptions in 1995. Archives of Virology 141: 209-218.

Ackermann, H.W. (1970) Ultrastructure of Vi phages I to VII of Salmonella typhi. Canadian Journal of Microbiology 16: 411.

Ackermann, H.W., and DuBow, M.S. (1987) Viruses of Prokaryotes: General properties of bacteriophage. In: CRC Press, Boca Raton.

Ackermann, H.W., and Gershman, M. (1992) Morphology of phages of a general Salmonella typing set. Res Virol 143: 303-310.

Allison, G.E., Angeles, D.C., thi Huan, P., and Verma, N.K. (2003) Morphology of temperate bacteriophage SfV and characterisation of the DNA packaging and capsid genes: the structural genes evolved from two different phage families. Virology 308: 114- 127.

Allwood, P.B., Malik, Y.S., Maherchandani, S., Vought, K., Johnson, L.A., Braymen, C, et al (2004) Occurrence of Escherichia coli, Noroviruses, and F-specific coliphages in fresh market-ready produce. J Food Prot 67: 2387-2390(4).

Altschul, S.F., and Koonin, E.V. (1998) Iterated profile searches with PSI- BLAST—a tool for discovery in protein databases. Trends Biochem Sci 23: 444-447.

Archer, M.J., and Liu, J.L. (2009) Bacteriophage T4 nanoparticles as materials in sensor applications: variables that influence their organization and assembly on surfaces. Sensors 9: 6298-6311.

Atterbury, R.J. (2007) Bacteriophage therapy to reduce Salmonella colonization of broiler chickens. Applied and environmental microbiology 73: 4543.

Atterbury, R.J. (2009) Bacteriophage biocontrol in animals and meat products. Microbial Biotechnology 2: 601-612.

Atterbury, R.J., Connerton, P.L., Dodd, C.E., Rees, C.E., and Connerton, I.F. (2003a) Isolation and characterization of Campylobacter bacteriophages from retail

237 poultry. Appl Environ Microbiol 69: 4511-4518.

Atterbury, R.J., Connerton, P.L., Dodd, C.E.R., Rees, C.E.D., and Connerton, I.F. (2003b) Application of host-specific bacteriophages to the surface of chicken skin leads to a reduction in recovery of Campylobacter jejuni.

Balasubramanian, S., Sorokulova, LB., Vodyanoy, V.J., and Simonian, A.L. (2007) Lytic phage as a specific and selective probe for detection of Staphylococcus aureus: A surface plasmon resonance spectroscopic study. J. Biosens. and Bioelect. 22: 948-955(6).

Balogh, B., Jones, J.B., Iriarte, F.B., and Momol, M.T. (2010) Phage therapy for plant disease control. Curr Pharm Biotechnol 11: 48-57(10).

Bell, C, and Kyriakides, A. (2002a) Salmonella: A Practical approach to the organism and its control in foods. CRC Press, New York, Washington, DC.

Bell, C, and Kyriakides, A. (2002b) Foodborne pathogens hazards, risk analysis and control. CRC Press, New York, Washington, DC.

Bell, C, and Kyriakides, A. (2002c) Listeria monocyotogenes. In: FoodBorne Pathogens Hazards, risk ananlysis and control. C.W. Blackburn, and P.J. McClure (eds.) CRC Press, New york, Washington, DC. pp. 279-306.

Bell, C, and Kyriakides, A. (2002d) Pathogenic Escherichia coli. In: Foodborne Pathogens Hazards, risk analysis and control. C.W. Blackburn, and P.J. McClure (eds.) CRC Press, New York, Washington, DC. pp. 279-306.

Bennett, A.R. (1997) The use of bacteriophage-based systems for the separation and concentration of Salmonella. Journal of applied microbiology 83: 259.

Besser, T.E., Goldoft, M., Pritchett, L.C., Khakhria, R., Hancock, D.D., Rice, D.H., et al (2000) Multiresistant Salmonella Typhimurium DTI04 infections of humans and domestic animals in the Pacific Northwest of the United States. Epidemiology and Infection 124: 193.

238 Bhushan, S., and Etzel, M.R. (2007) Charged ultrafiltration membranes for protein separation. ASABE Annual International Meeting.

Bigwood, T., Hudson, J.A., Billington, C, Carey-Smith, G.V., and Heinemann, J.A. (2008) Phage inactivation of foodborne pathogens on cooked and raw meat. Food Microbiol 25: 400-406.

Bigwood, T., Hudson, J.A., and Billington, C. (2009) Influence of host and bacteriophage concentrations on the inactivation of foodborne pathogenic bacteria by two phages. FEMS Microbiol Lett 291: 59-64.

Birge, E.A. (1994) Bacterial and bacteriophage genetics. In: Springer-Verlag, New York, NY.

Blackman, I.C., and Frank, J.F. (1996) Growth of Listeria monocytogenes as a biofilm on various food-processing surfaces. J Food Prot 59: 827-831(5).

Blatter, S., Giezendanner, N., Stephan, R., and Zweifel, C. (2010) Phenotypic and molecular typing of Listeria monocytogenes isolated from the processing environment and products of a sandwich-producing plant. Food Control (In Press).

Bonaventura, G.D., Piccolomini, R., Paludi, D., D'Orio, V., Vergara, A., Conter, M., et al (2008) Influence of temperature on biofilm formation by Listeria monocytogenes on various food-contact surfaces: relationship with motility and cell surface hydrophobicity. J Appl Microbiol 104: 1552-1561.

Brussow, H., and Hendrix, R.W. (2002) Phage genomics: small is beautiful. Cell 108: 13-16.

Brussow, H., and Kutter, E. (2005) Phage ecology. In: Bacteriophages : biology and applications. E. Kutter, and A. Sulakvelidze (eds.) CRC Press, Boca Raton, FL. pp. 67.

Bruttin, A., and Brussow, H. (2005) Human volunteers receiving Escherichia coli phage T4 orally: A safety test of phage therapy. Antimicrobial agents and chemotherapy 49: 2874.

239 Campbell, A. (1994) Comparative Molecular biology of lambdoid phages. Annu Rev Microbiol 48: 193-222.

Carlton, R.M., Noordman, W.H., Biswas, B., de Meester, E.D., and Loessner, M.J. (2005) Bacteriophage PI 00 for control of Listeria monocytogenes in foods: Genome sequence, bioinformatic analyses, oral toxicity study, and application. Regulatory Toxicology and Pharmacology 43: 301-312.

Carver, T., Thomson, N., Bleasby, A., Berriman, M., and Parkhill, J. (2009) DNAPlotter: circular and linear interactive genome visualization. Bioinformatics 25: 119- 120.

Centers for Disease Control and Prevention (2010) Preliminary FoodNet Data on the Incidence of Infection with Pathogens Transmitted Commonly Through Food 10 States, 2009 [WWW document]. URL http://www.cdc.gov/mmwr/preview/mmwrhtml/ mm5914a2.htm.

Centers for Disease Control and Prevention (2009) Multistate Outbreak of E. coli 0157:H7 Infections Associated with Beef from Fairbank Farms [WWW document]. URL http://www.cdc.gov/ecoli/2009/.

Centers for Disease Control and Prevention (2006) Salmonellosis [WWW document]. URL http://www.cdc.gOv/ncidod/dbmd/diseaseinfo/salmonellosis_g.htm# How%20common%20is%20salmonellosis.

Centers for Disease Control and Prevention (2005) Listeriosis [WWW document]. URL http://www.cdc.gov/ncidod/dbmd/diseaseinfo/listeriosis_g.htm.

Cheetham, B.F., and Katz, M.E. (1995) A role for bacteriophages in the evolution and transfer of bacterial virulence determinants. Mol Microbiol 18: 201-208.

Clokie, M.R., and Kropinski, A.M. (2009) Bacteriophages. Methods and Protocols, Molecular and Applied Aspects. Humana Press.

Comeau, A.M., Arbiol, C, and Krisch, H.M. (2010) Gene network visualization and quantitative synteny analysis of more than 300 marine T4-like phage scaffolds from

240 the GOS metagenome. Mol Biol Evol.

Conedera, G., Mattiazzi, E., Russo, F., Chiesa, E., Scorzato, I., Grandesso, S., et a\ (2007) A family outbreak of Escherichia coli 0157 haemorrhagic colitis caused by pork meat salami. Epidemiology and infection 135: 311.

Costa, C.I., Azeredo, J., Balcao, V.M., Castro, L.M., Santos, S., Teixeira, J.A., et a\ (2009) A multiple emulsion formulation of bacteriophage encapsulated in lipid nanovesicles. Book of Abstracts of MicroBiotec 09.

Cummings, D.J., Kusy, A.R., Chapman, V.A., DeLong, S.S., and Stone, K.R. (1970) Characterization of T-Even Bacteriophage Substructures: I. Tail Fibers and Tail Tubes. J Virol 6: 534-544.

Curtin, J.J., and Donlan, R.M. (2006) Using bacteriophages to reduce formation of catheter-associated biofilms by Staphylococcus epidermidis. Antimicrob Agents Chemother 50: 1268-1275.

Cutter, C.N. (2002) Microbial control by packaging: A review. Crit Rev Food Sci Nutr42:151.

D' Aoust, J.Y. (2001) Salmonella . In: Guide to Foodborne Pathogens. R. Labbe, and S. Garcia (eds.) John Wiley and Sons, New York. pp. 163-191.

D'Aoust, J.Y. (1994) Salmonella and the international food trade. International journal of food microbiology 24: 11.

Davis, M.A., Besser, T.E., Eckmann, K., MacDonald, J.K., Green, D., Hancock, D.D., et al (2007) Multidrug-resistant Salmonella Typhimurium, Pacific Northwest, United States. Emerging infectious diseases 13: 1583.

Delbruck, M. (1940) The growth of bacteriophage and lysis of the host. The Journal of General Physiology 23: 643-660.

Deresinski, S. (2009) Bacteriophage therapy: exploiting smaller fleas. Clinical infectious diseases 48: 1096-101.

241 Desplats, C, and Krisch, H.M. (2003) The diversity and evolution of the T4-type bacteriophages. Res Microbiol 154: 259-267.

D'Herelle, F. (1926) The bacteriophage and its behavior. Williams & Wilkins Company, Baltimore, MD, USA.

Djordjevic, D., Wiedmann, M., and McLandsborough, L.A. (2002) Microtiter Plate assay for assessment of Listeria monocytogenes biofilm formation. Appl Environ Microbiol 68: 2950-2958.

Donnelly, C.W. (2001) Listeria monocytogenes: a continuing challenge. Nutrition Reviews 59: 183.

Dore, K., Buxton, J., Henry, B., Pollari, F., Middleton, D., Fyfe, M, et al (2004) Risk factors for Salmonella Typhimurium DT 104 and non-DT 104 infection: a Canadian multi-provincial case-control study. Epidemiology and infection 132: 485.

Dubos, R.J., Straus, J.H., and Pierce, C. (1943) The multiplication of bacteriophage in vivo and its protective effect against an experimental infection with Shigella dysenteriae. The Journal of Experimental Medicine 78: 161-168.

Dykes, G.A., and Moorhead, S.M. (2002) Combined antimicrobial effect of nisin and a listeriophage against Listeria monocytogenes in broth but not in buffer or on raw beef. Int J Food Microbiol 73: 71-81.

Ellis, E.L., and Delbriick, M. (1939) The growth of bacteriophage. The Journal of General Physiology 22: 365-384.

Ermolaeva, M.D., Khalak, H.G., White, O., Smith, H.O., and Salzberg, S.L. (2000) Prediction of transcription terminators in bacterial genomes. J Mol Biol 301: 27- 33.

Esimbekova, E.N., Kratasyuk, V.A., and Torgashina, I.G. (2007) Disk-shaped immobilized multicomponent reagent for bioluminescent analyses: correlation between activity and composition. Enzyme Microb Technol 40: 343-346.

242 Falkenstein, D. (2010) FDA to announce new produce safety rules. [WWW document]. URL http://www.foodpoisonjournal.com/2010/04/articles/ foodborne-illness-outbreaks/fda-to-announce-new-produce-safety-rules/.

Farber, J.M., and Peterkin, P.I. (1991) Listeria monocytogenes, a foodborne pathogen. Microbiol Mol Biol Rev 55: 476-511.

Fauquet, CM., and Fargette, D. (2005) International Committee on Taxonomy of Viruses and the 3,142 unassigned species. Virology Journal 2: 64.

FDA (2006) FDA approval of Lwter/a-specific bacteriophage preparation on ready-to-eat (RTE) meat and poultry products [WWW document]. URL http://www .cfsan.fda.gov/~dms/opabacqa.html.

Feng, P. (2001) Escherichia coli. In: Guide to Foodborne Pathogens. R. Labbe, and S. Garcia (eds.) John Wiley and Sons, New York, NY. pp. 143-162.

Feng, P. (1995) Escherichia coli serotype 0157:H7: novel vehicles of infection and emergence of phenotypic variants. Emerging infectious diseases 1: 47.

Filee, J., Bapteste, E., Susko, E., and Krisch, H.M. (2006) A Selective barrier to horizontal gene transfer in the T4-Type bacteriophages that has preserved a core genome with the viral replication and structural genes. Mol Biol Evol 23: 1688-1696.

Filee, J., Tetart, F., Suttle, C.A., and Krisch, H.M. (2005) Marine T4-type bacteriophages, a ubiquitous component of the dark matter of the biosphere. Proceedings of the National Academy of Sciences of the United States of America 102: 12471-12476.

Fiorentin, L., Vieira, N.D., and Barioni Junior, W. (2005) Use of lytic bacteriophages to reduce Salmonella Enteritidis in experimentally contaminated chicken cuts. Revista Brasileira de Ciencia Avicola 7: 255-260.

Folsom, J.P., and Frank, J.F. (2006) Chlorine Resistance of Listeria monocytogenes biofilms and relationship to subtype, cell density, and planktonic cell chlorine resistance. J Food Prot 69: 1292-1296(5).

243 Food Safety Network (2003) Listeria fact sheet [WWW document]. URL http://www.foodsafetynetwork.ca/en/article-details.php?a=3&c=14&sc=112&id=439.

Food Standards Agency UK (2006) Cadbury recall update 1 August 2006 [WWW document]. URL http://www.food.gov.uk/news/newsarchive/2006/aug/cadbury.

Fox, J.L. (2005) Therapy with phage: Mirage or potential barrage of products? ASM News 71: 453-455.

Frank, J.F., and Koffi, R.A. (1990) Surface-adherent growth of Listeria monocytogenes in associated with increased resistance to surfactant sanitizers and heat. J Food Prot. 53:550-554(7)

Garcia, P., Martinez, B., Obeso, J.M., and Rodriguez, A. (2008) Bacteriophages and their application in food safety. Letters in Applied Microbiology 47: 479-485.

Gautier, M., Rouault, A., Sommer, P., and Briandet, R. (1995) Occurrence of Propionibacterium freudenreichii bacteriophages in Swiss cheese. Appl Environ Microbiol 61: 2572-2576.

Gaynor, K., Park, S.Y., Kanenaka, R., Colindres, R., Mintz, E., Ram, P.K., et a\ (2009) International foodborne outbreak of Shigella sonnei infection in airline passengers. Epidemiol Infect 137: 335-341.

Gervais, L., Gel, M., Allain, B., Tolba, M., Brovko, L., Zourob, M., et a\ (2007) Immobilization of biotinylated bacteriophages on biosensor surfaces. Sensors Actuators B:Chem 125: 615-621.

Gill, J.J., and Hyman, P. (2010) Phage choice, isolation, and preparation for phage therapy. Curr Pharm Biotechnol 11: 2-14(13).

Gill, J., and Abedon, S.T. (2003) Bacteriophage ecology and plants [WWW document]. URL (http://www.apsnet.org/online/feature/phages/).

Goode, D., Allen, V.M., and Barrow, P.A. (2003) Reduction of experimental Salmonella and Campylobacter contamination of chicken skin by application of lytic

244 bacteriophages. Appl Environ Microbiol 69: 5032-5036.

Goodridge, L., Gallaccio, A., and Griffiths, M.W. (2003) Morphological, host range, and genetic characterization of two coliphages. Appl Environ Microbiol 69: 5364- 5371.

Gottesman, M., and Oppenheim, A. (1994) Lysogeny and Prophage. In: Encyclopedia of Virology. R.G. Webster, and A. Graof (eds.) Academic Press., New York. pp. 814-824.

Greer, G.G. (2005) Bacteriophage control of foodborne bacteria. Journal of Food Protection 68: 1102.

Greer, G.G. (1983) Psychrotrophic Brocothrix thermosphacta bacteriophages isolated from beef. Appl Environ Microbiol 46: 245-251.

Greer, G.G. (1988) Effects of phage concentration, bacterial density, and temperature on phage control of beef spoilage. J Food Sci 53: 1226-1227.

Guenther, S., Huwyler, D., Richard, S., and Loessner, M.J. (2009) Virulent bacteriophage for efficient biocontrol of Listeria monocytogenes in ready-to-eat foods. Appl Environ Microbiol 75: 93-100.

Guidolin, A., Morelli, G., Kamke, M., and Manning, P.A. (1984) Vibrio cholerae bacteriophage CP-T1: characterization of bacteriophage DNA and restriction analysis. J Virol 51: 163-169.

Guttman, B., Raya, R., and Kutter, E. (2005) Basic phage biology. In: Bacteriophages : biology and applications. E. Kutter, and A. Sulakvelidze (eds.) CRC Press, Boca Raton, FL. pp. 28.

Hagens, S., and Loessner, M.J. (2010) Bacteriophage for biocontrol of foodborne pathogens: calculations and considerations. Current Pharmaceutical Biotechnology 11: 58-68.

245 Hagens, S., and Loessner, M. (2007) Application of bacteriophages for detection and control of foodborne pathogens. Appl Microbiol Biotechnol 76: 513-519.

Handa, H., Gurczynski, S., Jackson, M.P., Auner, G., Walker, J., and Mao, G. (2008) Recognition of Salmonella Typhimurium by immobilized phage P22 monolayers. Surf Sci 602: 1392-1400.

Hanlon, G.W. (2007) Bacteriophages: An appraisal of their role in the treatment of bacterial infections. Int J Antimicrob Agents 30: 118-128.

Hazem, A. (2002) Effects of temperatures, pH-values, ultra-violet light, ethanol and chloroform on the growth of isolated thermophilic Bacillus phages. Microbiologica 25: 469-474.

Hendrix, R.W., Hatfull, G.F., and Smith, M.C.M. (2003) Bacteriophages with tails: chasing their origins and evolution. Res Microbiol 154: 253-257.

Hibma, A.M., Jassim, S.A.A., and Griffiths, M.W. (1997) Infection and removal of L-forms of Listeria monocytogenes with bred bacteriophage. Int J Food Microbiol 34: 197-207.

Hoick, A., and Berg, J. (2009) Inhibition of Listeria monocytogenes in cooked ham by virulent bacteriophages and protective cultures. Appl Environ Microbiol 75: 6944-6946.

Hudson, J.A., Billington, C, Carey-Smith, G., and Greening, G. (2005) Bacteriophages as biocontrol agents in food. J Food Prot 68: 426-437.

Huss, H.H. (1997) Control of indigenous pathogenic bacteria in seafood. Food Control 8: 91-98.

Ikeda, H., and Tomizawa, J. (1965) Transducing fragments in generalized transduction by phage PI: I. Molecular origin of the fragments. J Mol Biol 14: 85-109.

Izutsu, K., Aoyagi, N., and Kojima, S. (2004) Protection of protein secondary structure by saccharides of different molecular weights during freeze-drying. Chem

246 PharmBull52: 199-203.

Jamalludeen, N., Johnson, R.P., Friendship, R., Kropinski, A.M., Lingohr, E.J., and Gyles, C.L. (2007) Isolation and characterization of nine bacteriophages that lyse 0149 enterotoxigenic Escherichia coli. Vet Microbiol 124: 47-57.

Jay, J.M. (2000) Modern Food Microbiology. Gaithersburg, Md: Aspen Publishers.

Jepson, CD., and March, J.B. (2004) Bacteriophage lambda is a highly stable DNA vaccine delivery vehicle. Vaccine 22: 2413-2419.

Jia, Z., Ishihara, R., Nakajima, Y., Asakawa, S., and Kimura, M. (2007) Molecular characterization of T4-type bacteriophages in a rice field. Environ Microbiol 9: 1091-1096.

Jikia, D., Chkhaidze, N., Imedashvili, E., Mgaloblishvili, I., Tsitlanadze, G., Katsarava, R., et al (2005) The use of a novel biodegradable preparation capable of the sustained release of bacteriophages and ciprofloxacin, in the complex treatment of multidrug-resistant Staphylococcus awrews-infected local radiation injuries caused by exposure to Sr90. Clinical & Experimental Dermatology 30: 23-26.

Jirku, V. (1999) Whole cell immobilization as a means of enhancing ethanol tolerance. Journal of Industrial Microbiology and Biotechnology 22: 147-151.

Kail, L., Krogh, A., and Sonnhammer, E.L.L. (2007) Advantages of combined transmembrane topology and signal peptide prediction. The Phobius web server. Nucl Acids Res 35: W429-432.

Karam, J.K., Drake, J.W., and Kreuzer, K.N. (1994) Molecular Biology of Bacteriophage T4. ASM, Washington, DC.

Kasman, L.M., Kasman, A., Westwater, C, Dolan, J., Schmidt, M.G., and Norris, J.S. (2002) Overcoming the Phage Replication Threshold: a Mathematical Model with Implications for Phage Therapy. J Virol 76: 5557-5564.

247 Kim, J., Yun, S., and Ounaies, Z. (2006) Discovery of cellulose as a smart material. Macromolecules 39: 4202-4206.

Klausa, V., Piesiniene, L., Staniulis, J., and Nivinskas, R. (2003) Abundance of T4-type bacteriophages in municipal wastewater and sewage. Ekologija (Vilnius) 1: 47- 50.

Klein, J., and Ziehr, H. (1990) Immobilization of microbial cells by adsorption. J Biotechnol 16: 1-15.

Klemm, D., Schumann, D., Udhardt, U., and Marsch, S. (2001) Bacterial synthesized cellulose-artificial blood vessels for microsurgery. Progress in Polymer Science 26: 1561-1603.

Knaebel, D.B., Stormo, K.E., and Crawford, R.L. (1997) Immobilization of bacteria in macro- and microparticles. In: Bioremediation Protocols, pp. 67-78.

Knezevic, P., and Petrovic, O. (2008) A colorimetric microtiter plate method for assessment of phage effect on Pseudomonas aeruginosa biofilm. J Microbiol Methods 74:114-118.

Kocharunchitt, C, Ross, T., and McNeil, D.L. (2009) Use of bacteriophages as biocontrol agents to control Salmonella associated with seed sprouts. Int J Food Microbiol 128: 453-459.

Koski, L., Gray, M., Lang, B.F., and Burger, G. (2005) AutoFACT: An Automatic Functional Annotation and Classification Tool. BMC Bioinformatics 6: 151.

Kostyuchenko, V.A., Chipman, P.R., Leiman, P.G., Arisaka, F., Mesyanzhinov, V.V., and Rossmann, M.G. (2005) The tail structure of bacteriophage T4 and its mechanism of contraction. Nat Struct Mol Biol 12: 810-813.

Kostyuchenko, V.A., Leiman, P^G., Chipman, P.R., Kanamaru, S., van Raaij, M.,J., Arisaka, F., et al (2003) Three-dimensional structure of bacteriophage T4 baseplate. Nat Struct Mol Biol 10: 688-693.

248 Kothary, M.H., and Babu, U.S. (2001) Infective dose of foodborne pathogens in volunteers: A review. J Food Saf 21: 49-73.

Krogh, A., Larsson, B., von Heijne, G., and Sonnhammer, E.L.L. (2001) Predicting transmembrane protein topology with a hidden markov model: application to complete genomes. J Mol Biol 305: 567-580.

Kropinski, A.M.B., and Warren, R.A.J. (1970) Isolation and Properties of a Pseudomonas acidovorans Bacteriophage. J Gen Virol 6: 85-93.

Kropinski, A.M. (2006) Phage Therapy - Everything Old is New Again. Can J Infect Dis Med Microbiol. 17: 297-306.

Kropinski, A.M., Borodovsky, M., Carver, T.J., Cerde'no-Tjrraga, A.M., Darling, A., Lomsadze, A., et al (2009) In silico identification of genes in bacteriophage DNA. Bacteriophages: 57-89.

Krylov, V., Miller, S., Rachel, R., Biebl, M., Pleteneva, E., Schuetz, M., et al (2006) Ambivalent bacteriophages of different species active on Escherichia coli K12 and Salmonella sp. strains. Russian Journal of Genetics 42: 106-114.

Kutateladze, M., and Adamia, R. (2008) Phage therapy experience at the Eliava Institute. Medecine et Maladies Infectieuses 38: 426-430.

Kutter, E., and Sulakvelidze, A. (2005) Introduction. In: Bacteriophages : biology and applications. E. Kutter, and A. Sulakvelidze (eds.) CRC Press, Boca Raton, FL. pp. 1. Kutter, E., Raya, R., and Carlson, K. (2005) Molecular mechanisms of phage infection. In: Bacteriophages : biology and applications. E. Kutter, and A. Sulakvelidze (eds.) CRC Press, Boca Raton, FL. pp. 163. Kutter, E., De Vos, D., Gvasalia, G., Alavidze, Z., Gogokhia, L., Kuhl, S., et al (2010) Phage therapy in clinical practice: treatment of human infections. Curr Pharm Biotechnol 11: 69-86(18).

Labrie, S.J., Samson, J.E., and Moineau, S. (2010) Bacteriophage resistance mechanisms. Nat Rev Micro 8: 317-327.

249 Lakshmanan, R.S., Guntupalli, R., Hu, J., Kim, D., Petrenko, V.A., Barbaree, J.M., et al (2007) Phage immobilized magnetoelastic sensor for the detection of Salmonella Typhimurium. J Microbiol Methods 71: 55-60.

Larcom, L.L., and Thaker, N.H. (1977) The effects of temperature and ultraviolet irradiation on multiplication of bacteriophage phi29. Biophys J 19: 299-306.

Laslett, D., and Canback, B. (2004) ARAGORN, a program to detect tRNA genes and tmRNA genes in nucleotide sequences. Nucl Acids Res 32: 11-16.

Lavigne, R., Darius, P., Summer, E.J., Seto, D., Mahadevan, P., Nilsson, A.S., et al (2009) Classification of Myoviridae bacteriophages using protein sequence similarity. BMC Microbiology 9: 224.

Lavigne, R., Seto, D., Mahadevan, P., Ackermann, H., and Kropinski, A.M. (2008) Unifying classical and molecular taxonomic classification: analysis of the Podoviridae using BLASTP-based tools. Res Microbiol 159: 406-414.

Lee, M.B., and Middleton, D. (2003) Enteric illness in Ontario, Canada, from 1997 to 2001. J Food Prot 66: 953-961.

Lenski, R.E., and Levin, B.R. (1985) Constraints on the coevolution of bacteria and virulent phage: A model, some experiments, and predictions for natural communities. AmNatl25:585.

Leslie, S., Israeli, E., Lighthart, B., Crowe, J., and Crowe, L. (1995) Trehalose and sucrose protect both membranes and proteins in intact bacteria during drying. Appl Environ Microbiol 61: 3592-3597.

Leverentz, B., W.S., C, Z., A., W.J., J., Y., F., M.J., C, et al (2001) Examination of bacteriophage as a biocontrol method for Salmonella on fresh-cut fruit: A model study. J Food Prot 64: 1116-1121(6).

Leverentz, B., Conway, W.S., Camp, M.J., Janisiewicz, W.J., Abuladze, T., Yang, M., et al (2003) Biocontrol of Listeria monocytogenes on fresh-cut produce by treatment with lytic bacteriophages and a bacteriocin. Appl Environ Microbiol 69: 4519-4526.

250 Leverentz, B., Conway, W.S., Janisiewicz, W., and Camp, M.J. (2004) Optimizing concentration and timing of a phage spray application to reduce Listeria monocytogenes on honeydew melon tissue. J Food Prot 67: 1682-1686(5).

Lindberg, A. (1973) Bacteriophage receptors. Annual Review of Microbiology 27:205-241.

Lingohr, E., Frost, S., and Johnson, R.P. (2009) Determination of bacteriophage genome size by pulsed-field gel electrophoresis. Bacteriophages: 19-25.

Lu, Z., Breidt, F., Plengvidhya, V., and Fleming, H.P. (2003) Bacteriophage ecology in commercial sauerkraut fermentations. Appl Environ Microbiol 69: 3192-3202.

Mahiout, A., Matata, B.M., Vienken, J., and Courtney, J.M. (1997) Ex vivo complement protein adsorption on positively and negatively charged cellulose dialyser membranes. J Mater Sci Mater Med 8: 287-296(10).

Maloy, S.R., Cronan, J.E., and Freifelder, D. (1994) Microbial Genetics. Jones and Bartlett publishers, London.

Maniloff, J., and Ackermann, H.W. (1998) Taxonomy of bacterial viruses: establishment of tailed virus genera and the other Caudovirales . Archives of Virology 143:2051-2063.

Martinez, B., Obeso, J.M., Rodriguez, A., and Garcia, P. (2008) Nisin- bacteriophage crossresistance in Staphylococcus aureus. Int J Food Microbiol 122: 253- 258.

Mclntyre, L., Hudson, J.A., Billington, C, and Withers, H. (2007) Biocontrol of bacteria: past, present and future Strategies. Food New Zealand: 25-32.

Mead, P.S. (1999) Food-related illness and death in the United States. Emerging infectious diseases 5: 607.

Merril, C.R., Scholl, D., and Adhya, S.L. (2003) The prospect for bacteriophage therapy in western medicine. Nat Rev Drug Discov 2: 489-497.

251 Miliutina, L.N., and Vorotyntseva, N.V. (1993) Current strategy and tactics of etiotropic therapy of acute intestinal infections in children. Antibiot Khimioter 38: 46-53.

Miller, E.S., Kutter, E., Mosig, G., Arisaka, F., Kunisawa, T., and Ruger, W. (2003) Bacteriophage T4 Genome. Microbiol Mol Biol Rev 67: 86-156.

Millette, M., Le Tien, C, Smoragiewicz, W., and Lacroix, M. (2007) Inhibition of Staphylococcus aureus on beef by nisin-containing modified alginate films and beads. Food Control 18: 878-884.

Ming, X., Weber, G.H., Ayres, J.W., and Sandine, W.E. (1997) Bacteriocins applied to food packaging materials to inhibit Listeria monocytogenes on meats. J Food Sci 62: 413-415.

Minikh, O., Tolba, M., Brovko, L.Y., and Griffiths, M.W. (2010) Bacteriophage- based biosorbents coupled with bioluminescent ATP assay for rapid concentration and detection of Escherichia coli (In press).

Modi, R., Y., H., A., H., and M.W., G. (2001) Effect of phage on survival of Salmonella Enteritidis during manufacture and storage of cheddar cheese made from raw and pasteurized milk. J Food Prot 64: 927-933(7).

Murray, A.G., and Jackson, G.A. (1992) Viral dynamics: A model of the effects size, shape, motion and abundance of single-celled planktonic organisms and other particles. MAR ECOL PROG SER 89: 103-116.

Murthy, K. (2008) Stabilized bacteriophage formulations. (US Patent)

Nanduri, V., Balasubramanian, S., Sista, S., Vodyanoy, V.J., and Simonian, A.L. (2007a) Highly sensitive phage-based biosensor for the detection of |3-galactosidase. Anal Chim Acta 589: 166-172.

Nanduri, V., Bhunia, A.K., Tu, S., Paoli, G.C., and Brewster, J.D. (2007b) SPR biosensor for the detection of L. monocytogenes using phage-displayed antibody. Biosensors and Bioelectronics 23: 248-252.

252 Nataro, J.P., and Kaper, J.B. (1998) Diarrheagenic Escherichia coli. Clin Microbiol Rev 11: 142.

Nolan, J., Petrov, V., Bertrand, C, Krisch, H., and Karam, J. (2006) Genetic diversity among five T4-like bacteriophages. Virology Journal 3: 30.

O'Brien, A.D., Newland, J.W., Miller, S.F., Holmes, R.K., Smith, H.W., and Formal, S.B. (1984) Shiga-like toxin-converting phages from Escherichia coli strains that cause hemorrhagic colitis or infantile diarrhea. Science 226: 694-696.

O'Flaherty, S., Ross, R.P., Meaney, W., Fitzgerald, G.F., Elbreki, M.F., and Coffey, A. (2005) Potential of the polyvalent mti-Staphylococcus bacteriophage K for control of antibiotic-resistant staphylococci from hospitals. Appl Environ Microbiol 71: 1836-1842.

O'Flynn, G., Ross, R.P., Fitzgerald, G.F., and Coffey, A. (2004) Evaluation of a cocktail of three bacteriophages for biocontrol of Escherichia coli 0157:H7. Appl Environ Microbiol 70: 3417-3424.

Pan, H., Wenna, B., Zhipeng, X., Jianguo, Z., and Yongquan, L. (2010) Immobilization of Escherichia coli cells with cis-epoxysuccinate hydrolase activity for d- tartaric acid production. Biotechnol Lett 32: 235-241.

Paschke, M. (2006) Phage display systems and their applications. Applied microbiology and biotechnology 70: 2.

Pastre, D., Pietrement, O., Fusil, S., Landousy, F., Jeusset, J., David, M., et al (2003) Adsorption of DNA to mica mediated by divalent counterions: A theoretical and experimental study. Biophys J 85: 2507-2518.

Petrov, V.M., Nolan, J.M., Bertrand, C, Levy, D., Desplats, C, Krisch, H.M., et al (2006) Plasticity of the gene functions for DNA replication in the T4-like phages. J Mol Biol 361: 46-68.

Petty, N.K., Evans, T.J., Fineran, P.C., and Salmond, G.P.C. (2007) Biotechnological exploitation of bacteriophage research. Trends Biotechnol 25: 7-15.

253 Phillips, C.A. (1996) Review: Modified atmosphere packaging and its effects on the microbiological quality and safety of produce. Int J Food Sci Tech 31: 463-479(17).

Pickard, D., Thomson, N.R., Baker, S., Wain, J., Pardo, M, Goulding, D., et al (2008) Molecular characterization of the Salmonella enterica serovar Typhi Vi-typing bacteriophage El. J Bacteriol 190: 2580-2587.

Pietiainen, M., Francois, P., Hyyrylainen, H., Tangomo, M., Sass, V., Sahl, H., et al (2009) Transcriptome analysis of the responses of Staphylococcus aureus to antimicrobial peptides and characterization of the roles of vraDE and vraSR in antimicrobial resistance. BMC Genomics 10: 429.

Puapermpoonsiri, U., Ford, S.J., and van der Walle, C.F. (2010) Stabilization of bacteriophage during freeze drying. Int J Pharm 389: 168-175.

Puapermpoonsiri, U., Spencer, J., and van der Walle, C.F. (2009) A freeze-dried formulation of bacteriophage encapsulated in biodegradable microspheres. European Journal of Pharmaceutics and Biopharmaceutics 72: 26-33.

Public Health Agency of Canada (2010) Update to 2008 Listeria monocytogenes case numbers [WWW document]. URL http://www.phac-aspc.gc.ca/alert-alerte/listeria/ listeria_20100413-eng.php.

Public Health Agency of Canada (2007) National Integrated Enteric Pathogen Surveillance Program (C-EnterNet) [WWW document]. URL http://www.phac-aspc.gc. ca/publicat/2007/c-enternet06/areport06_e.html.

Rees, C.E., and Dodd, C.E. (2006) Phage for rapid detection and control of bacterial pathogens in food. Advances in applied microbiology 59: 159-186.

Reiss, G., Kunz, P., Koin, D., and Keeffe, E.B. (2006) Escherichia coli 0157:H7 infection in nursing homes: review of literature and report of recent outbreak. Journal of the American Geriatrics Society 54: 680-684.

Rohwer, F., and Edwards, R. (2002) The phage proteomic tree: a genome-based taxonomy for phage. J Bacteriol 184: 4529-4535.

254 Romanova, N.A., Gawande, P.V., Brovko, L.Y., and Griffiths, M.W. (2007) Rapid methods to assess sanitizing efficacy of benzalkonium chloride to Listeria monocytogenes biofilms. J Microbiol Methods 71: 231 -237.

Rose, B.E., Hill, W.E., Umholtz, R., Ransom, G.M., and James, W.O. (2002) Testing for Salmonella in raw meat and poultry products collected at federally inspected establishments in the United States, 1998 through 2000. J Food Prot 65: 937-947.

Roy, B., Ackermann, H.W., Pandian, S., Picard, G., and Goulet, J. (1993) Biological inactivation of adhering Listeria monocytogenes by listeriaphages and a quaternary ammonium compound. Appl Environ Microbiol 59: 2914-2917.

Sakairi, M.A.C., Yasuda, K., and Matsumura, M. (1996) Nitrogen removal in seawater using nitrifying and denitrifying bacteria immobilized in porous cellulose carrier. Water Science and Technology 34: 267-274(8).

Salalha, W., Kuhn, J., Dror, Y., and Zussman, E. (2006) Encapsulation of bacteria and viruses in electrospun nanofibres. Nanotechnology 17: 4675.

Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989) Molecular cloning: a laboratory manual. Cold Spring Harbor, NY.

Santiago-Silva, P., Soares, N.F.F., Nobrega, J.E., Junior, M.A.W., Barbosa, K.B.F., Volp, A.C.P., et a\ (2009) Antimicrobial efficiency of film incorporated with pediocin (ALTA® 2351) on preservation of sliced ham. Food Control 20: 85-89.

Sawayama, S., Rao, K.K., and Hall, D.O. (1998) Immobilization of Rhodobacter capsulatus on cellulose beads and water treatment using a photobioreactor. J Ferment Bioeng 86: 517-520.

Scharff, R.L. (2010) Health-related costs from foodborne illness in the United States [WWW document]. URL http://www.producesafetyproject.org/admin/assets/ files/Health-Related-Foodborne-Illness-Costs-Report.pdf-1 .pdf.

Scott, V.N., Swanson, T.A., Freier, T.A., Payton, P., Jr., Sveum, W.H., Hall, P.A., et a\ (2005) Guidelines for conducting Listeria monocytogenes challenge testing of foods.

255 Food Prot. Trends 25: 818-825.

Selvaraj, P.T., Little, M.H., and Kaufman, E.N. (1997) Biodesulfurization of flue gases and other sulfate/sulfite waste streams using immobilized mixed sulfate-reducing bacteria. Biotechnol Prog 13: 583-589.

Serwer, P., and Hayes, S.J. (1982) Agarose gel electrophoresis of bacteriophages and related particles. I. Avoidance of binding to the gel and recognizing of particles with packaged DNA. Electrophoresis 3: 76-80.

Shabani, A., Zourob, M., Allain, B., Marquette, C.A., Lawrence, M.F., and Mandeville, R. (2008) Bacteriophage-modified microarrays for the direct impedimetric detection of bacteria. Anal Chem 80: 9475-9482.

Sharma, M., Patel, J.R., Conway, W.S., Ferguson, S., and Sulakvelidze, A. (2009) Effectiveness of bacteriophages in reducing Escherichia coli 0157:H7 on fresh-cut cantaloupes and lettuce. Journal of food protection 72: 1481-1482.

Shine, J., and Dalgarno, L. (1975) Terminal-sequence analysis of bacterial ribosomal RNA. European Journal of Biochemistry 57: 221-230.

Shine, J., and Dalgarno, L. (1974) The 3'-terminal sequence of Escherichia coli 16S ribosomal RNA: complementarity to nonsense triplets and ribosome binding sites. Proceedings of the National Academy of Sciences of the United States of America 71: 1342-1346.

Showe, M.K., and Onorato, L. (1978) Kinetic factors and form determination of the head of bacteriophage T4. Proc Natl Acad Sci U S A 75: 4165-4169.

Si, W., Gong, J., Tsao, R., Zhou, T., Yu, H., Poppe, C, et a\ (2006) Antimicrobial activity of essential oils and structurally related synthetic food additives towards selected pathogenic and beneficial gut bacteria. J Appl Microbiol 100: 296-305.

Singh, A., Glass, N., Tolba, M., Brovko, L., Griffiths, M., and Evoy, S. (2009) Immobilization of bacteriophages on gold surfaces for the specific capture of pathogens. Biosensors and Bioelectronics 24: 3645-3651.

256 Siragusa, G.R., Cutter, C.N., and Willett, J.L. (1999) Incorporation of bacteriocin in plastic retains activity and inhibits surface growth of bacteria on meat. Food Microbiol 16: 229-235.

Sivapalasingam, S., Nelson, J.M., Joyce, K., Hoekstra, M, Angulo, F.J., and Mintz, E.D. (2006) High prevalence of antimicrobial resistance among Shigella isolates in the United States tested by the National Antimicrobial Resistance Monitoring System from 1999 to 2002. Antimicrobial Agents Chemother 50: 49-54.

Smith, H.W., and Huggins, M.B. (1983) Effectiveness of phages in treating experimental Escherichia coli diarrhoea in calves, piglets and lambs. J Gen Microbiol 129: 2659-2675.

Snowdon, A., Buzby, J.C., and Roberts, T. (2002) Epidemiology, cost, and risk of foodborne disease. In: Foodborne diseases. D. Cliver, and H. Riemann (eds.) Elsevier Press, Amsterdam, The Netherlands, pp. 31-51.

Soni, K., Nannapaneni, R., and Hagens, S. (2010) Reduction of Listeria monocytogenes on the surface of fresh channel catfish fillets by bacteriophages Listex P100. Foodborne Pathog Dis 7: 427-434.

Stein, C, Kuchenmuller, T., Hendrickx, S., Pruss-Uestun, A., Wolfson, L., Engels, D., et al (2007) The global burden of disease assessments - Who is responsible? PLoSNeglTropDisl:el61.

Steward, G.F., Montiel, J.L., and Azam, F. (2000) Genome size distributions indicate variability and similarities among marine viral assemblages from diverse environments. Limnol Oceanogr 45: 1697-1706.

Strauch, E., Hammerl, J., and Hertwig, S. (2007) Bacteriophages: New Tools for Safer Food? Journal fur Verbraucherschutz und Lebensmittelsicherheit 2: 138-143.

Suarez, V.B., Quiberoni, A., Binetti, A.G., and Reinheimer, J.A. (2002) Thermophilic lactic acid bacteria phages isolated from Argentinian dairy industries. J FoodProt65: 1597-1604(8).

257 Sulakvelidze, A., and Barrow, P. (2005) Phage therapy in animals and agribusiness. In: Bacteriophages: biology and application. E. Kutter, and A. Sulakvelidze (eds.) Boca Raton, Florida, pp. 335-380.

Sulakvelidze, A., Alavidze, Z., and Morris, J.G., Jr. (2001) Bacteriophage therapy. Antimicrob Agents Chemother 45: 649-659.

Sullivan, M.B., Coleman, M.L., Weigele, P., Rohwer, F., and Chisholm, S.W. (2005) Three Prochlorococcus cyanophage genomes: signature features and ecological interpretations. PLoS Biol 3: el44.

Summers, W.C. (2005) Bacteriophage research: Early history. In: Bacteriophages : biology and applications. E. Kutter, and A. Sulakvelidze (eds.) CRC Press, Boca Raton, FL. pp. 5

Summers, W.C. (2001) Bacteriophage therapy. Annu Rev Microbiol 55: 437-451.

Sun, W. (2001) Use of bioluminescent Salmonella for assessing the efficiency of constructed phage-based biosorbent. Journal of industrial microbiology biotechnology 27: 126.

Sutherland, J., and Varnam, A. (2002) Enterotoxin-producing Staphylococcus, Shigella, Yersinia, Vibrio, Aeromonas and Plesiomonas. In: Foodborne Pathogens. CRC press, Washington, DC. pp. 384-415.

Swaminathan, B., and Gerner-Smidt, P. (2007) The epidemiology of human listeriosis. Microb Infect 9: 1236-1243.

Tarahovsky, Y.S., Ivanitsky, G.R., and Khusainov, A.A. (1994) Lysis of Escherichia coli cells induced by bacteriophage T4. FEMS Microbiol Lett 122: 195-199.

The National Center for Biotechnology Information (NCBI) (2010) . URL http:// www.ncbi.nlm.nih.gov/genomes/genlist.cgi ?taxid=10239&type=6&name=Phages.

The U.S. Food and Drug Administration (2010a) FDA Update on the investigation into the Salmonella Montevideo outbreak [WWW document]. URL

258 http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm203344.htm.

The U.S. Food and Drug Administration (2010b) Hydrolyzed Vegetable Protein Product Recalls [WWW document]. URL http://www.fda.gov/Safety/Recalls/Major ProductRecalls/HVP/default.htm.

The U.S. Food and Drug Administration (2010c) Peanuts product recalls. [WWW document]. URL http://www.fda.gov/Safety/Recalls/MajorProductRecalls/Peanut/default .htm.

The U.S. Food and Drug Administration (2010d) Pistachio product recalls. [WWW document]. URL http://www.fda.gov/Safety/Recalls/MajorProductRecalls/ Pistachio/default.htm.

Tolba, M, Minikh, O., Brovko, L.Y., Evoy, S., and Griffiths, M.W. (2010) Oriented immobilization of bacteriophages for biosensor applications. Appl Environ Microbiol 76: 528-535.

United States Department of Agriculture Food Safety and Inspection Service (2008) FSIS recalls [WWW document]. URL http://www.fsis.usda.gov/Fsis_Recalls/

Van Reis, R., Brake, J.M., Charkoudian, J., Burns, D.B., and Zydney, A.L. (1999) High-performance tangential flow filtration using charged membranes. J Membr Sci 159: 133-142.

Waldor, M.K., and Mekalanos, J.J. (1996) Lysogenic conversion by a filamentous phage encoding cholera toxin. Science 272: 1910-1914.

Wang, A.L., and Wang, C.C. (1991) Viruses of parasitic protozoa. Parasitology Today 7: 76-80.

Werquin, M., Ackermann, H., and Levesque, R.C. (1988) A study of 33 bacteriophages of Rhizobium meliloti. Appl Environ Microbiol 54: 188-196.

Whichard, J.M., Namalwar, S., and William, P.F. (2003) Suppression of Salmonella growth by wild-type and large-plaque variants of bacteriophage Felix 01 in

259 liquid culture and on chicken frankfurters. J Food Prot 66: 220-225(6).

Wikipedia (2010) Bacteriophage [WWW document]. URL http://en.wikipedia .org/wiki/Bacteriophage#History.

Williams, R.C., Isaacs, S., Decou, M.L., Richardson, E.A., Buffett, M.C., Slinger, R.W., et a\ (2000) Illness outbreak associated with Escherichia coli 0157 : H7 in genoa salami. Can Med Assoc J 162: 1409-1413.

Willshaw, G., Cheasty, T., and Smith, H.R. (2000) Escherichia coli. In: The microbiological safety and quality of food. B.M. Lund, T.C. Baird-Parker, and G.W. Gould (eds.) Aspen Inc, Maryland, USA.

Wilson, W.W., Wade, M.M., Holman, S.C., and Champlin, F.R. (2001) Status of methods for assessing bacterial cell surface charge properties based on zeta potential measurements. J Microbiol Methods 43: 153-164.

World Health Organization (2000) WHO responds to new challenges in food safety. [WWW document]. URL http://www.who.int/inf-pr-2000/en/pr2000-04.html.

Wu, S., and Lia, Y. (2008) Application of bacterial cellulose pellets in enzyme immobilization. J Molec Catal B 54: 103-108.

Zafar, N., Mazumder, R., and Seto, D. (2002) CoreGenes: A computational tool for identifying and cataloging "core" genes in a set of small genomes. BMC Bioinformatics 3: 12.

Zhu, M., Du, M., Cordray, J., and Ahn, D.U. (2005) Control of Listeria monocytogenes contamination in ready-to-eat meat products. Comprehensive Reviews in Food Science and Food Safety 4: 34-42.

Zink, R., and Loessner, M.J. (1992) Classification of virulent and temperate bacteriophages of Listeria spp. on the basis of morphology and protein analysis. Appl Environ Microbiol 58: 296-302.

260 Zourob, M, and Ripp, S. (2010) Bacteriophage-Based Biosensors. In: Recognition Receptors in Biosensors, pp. 415-448.

Zuber, S., Ngom-Bru, C, Barretto, C, Bruttin, A., Brtissow, H., and Denou, E. (2007) Genome Analysis of Phage JS98 Defines a Fourth Major Subgroup of T4-Like Phages in Escherichia coli. J Bacteriol 189: 8206-8214.

Zuker, M. (2003) Mfold web server for nucleic acid folding and hybridization prediction. Nucl Acids Res 31: 3406-3415.

261 APPENDIX

A 3.1. Mass spectroscopy data for six isolated proteins from OSboM-AG3

1054.674 1200 1069.509

1000 1288.687 2148.993 2075.992! 1471.704 2206.097 ~ 800 2061.971 1482.819 972.494 2011.962 2294.155 1505.689 '« 600 1962.055: 1628.883 : 2433.232 1736.897 2544.272 180918?: 400 1886:991 2697.276 2812.350 2984.467 200 3185.559 3275.631 3491.653

llj|MIMfcdMilJ^.il4- ii ..[ pliltl». I.LJ li^ll.l iI Ji.l^hi.ip*liiif i.I. , i .MII^N iJi ik l f ll ( 1000 1500 2000 2500 3000 3500 4000 mJz. aaiu

MALDI QqTOF mass spectrum of the tryptic digest of protein band A at 126 kDa £91

t~' 1—» I—' h-> h-'I—'I—' h-'I—"OOOOOOOOOOOOOOO 05 OlOi^rt-loWPPOlDiflCOJ-JOlUifcililOMMHOO ft H^-JOU)OHDK)Uiajhl*.-JOLJOiaMU'a)HtvlO e ft s ft ft ft o •

ON List of the peptides identified by mass spectrometry (Band-A-126 kDa)

Mass m/z [MH+] error Peptide Sequence (Meas.) (calc.) (Da) 972.494 972.494 0.000 1035-1042 VYPFFSGR 1054.674 1054.677 -0.003 24-32 ILFRPAPIK 1069.509 1069.506 0.003 1081-1089 FNTGDNVFR 1288.687 1288.690 -0.003 1069-1080 GVFAVPQNETVK 1471.704 1471.702 0.002 1099-1112 SADDTLTNAEAVHK 1482.819 1482.813 0.006 329-341 FVAVMKPGISYVR(Met-OX) 1505.689 1505.681 0.008 9-19 RPYWDDWNPEK 1586.846 1586.854 -0.008 1123-1136 TYVNTRVLGYTATK 1628.883 1628.885 -0.002 344-357 RIENLGEELVTIDK 1736.897 1736.897 0.000 726-741 IDAVYLADNGVFNVAR 1747.808 1747.806 0.002 249-263 YGYDEEVADFVELAK 1809.837 1809.836 0.001 610-627 NASGADVTSNFTLDGGQR 1880.907 1880.906 0.001 37-51 ELNQMQTIFQDQLEK(Met-OX) 1962.055 1962.066 -0.011 396-414 FLNASSAVQATALLISAER List of the peptides identified by mass spectrometry (Band-A-126 kDa) (continued...)

m/z [MH+] Mass error Peptide Sequence (Meas.) (calc.) (Da) 2011.962 2011.972 -0.010 1081-1098 FNTGDNVFRLTDSPVDSK 2061.971 2061.972 -0.001 842-860 LIDDLSADWVGSIDTDNGR 2075.992 2075.999 -0.007 966-982 VWEPHGAGGVWWGYRYR 2148.993 2149.003 -0.010 920-936 LNPTTDYWFENYYVAPR 2206.097 2206.099 -0.002 360-380 DTDTLNNNPVAVATGNYLVSK 2294.155 2294.174 -0.019 760-778 LYELMIPPYTANIDDIQIR (Met-OX) 2347.098 2347.090 0.008 269-288 IQSMVTQSTYNILEDSMAQR (Met-OX) 2425.256 2425.263 -0.007 999-1019 TTTTLTGEQIVETQVIPYMRK 2433.232 2433.237 -0.005 358-380 ARDTDTLNNNPVAVATGNYLVSK 2446.233 2446.239 -0.007 79-101 FTLAGGSEFTDLEGISELYVLGK 2544.272 2544.287 -0.015 242-263 IDLVLSRYGYDEEVADFVELAK 2633.236 2633.259 -0.013 289-310 TYETNGDYNVSTHQIDLREHLK 2690.376 2690.384 -0.008 1020-1042 TNINFEATGLRPFTRVYPFFSGR 2697.276 2697.274 0.002 209-233 VTETIVTETEDESLFSNAQGTPNSK 2812.350 2812.364 -0.014 579-604 TVTEVTETVTFTSAASVQLNNHDGYK List of the peptides identified by mass spectrometry (Band-A-126 kDa) (continued...)

Mass m/z [MH+] error Peptide Sequence (Meas.) (calc.) (Da) 2984.467 2984.467 0.000 699-725 ITSGASDTDMVRPNTAIVLD AEYYLPR (Met-OX) 2990.492 2990.515 -0.023 495-524 ITLNAAGAGSISAPLGYSFSPEFSLYSAAK 3185.559 3185.568 -0.009 176-204 GVYFVRGMFLDVEAATLIVDNASNSTSHR (Met-OX) 3275.631 3275.637 -0.006 892-919 EISVRQDYATTTINVNPYAV FNWEGFLK 3402.759 3402.780 -0.021 234-263 APGAHRLRIDLVLSRYGYDE EVADFVELAK 3491.653 3491.664 -0.011 799-827 ISNVEYYTSLSQLESSAMTQQVFDPITGNPR (Met-OX) 3606.731 3606.758 -0.027 861-891 LRPFVQQNAVDLTPVGWNNV QDGMVVCNYTR 1187.588 10001

8001 1215.583

1135.635 1747.823 2944.573 = 6001 1061.551 3000.501 1902.012 3199.624 989.553 1426.683 096.060 14001 3284.701 2315.229 1445.802 3355.701 2388.231 3565.675 200 2752.354 4052.093 2982.990I

f'-TT'T . I .J r\ . Mr. •) \- . 1000 1500 2000 2500 3000 3500 4000 4500 m/z. amu

MALDI QqTOF mass spectrum of the tryptic digest of protein band B at 108 kDa Protein identification results by MS and MS/MS (Band-B-108 kDa)

Name: Orf00213; conserved hypothetical phage protein [phiSboM-AG3] MW: 106909.85 Da Sequence coverage: 42.6%

001 MANKPTQPVF PLGLVAEEQS TLAGILNTGT IEHGPDAVLT LPEGNASAGL PSSVRYNADS 061 DEFEGFYENG GWLPLGGGGI RWEALPHAST ATLTEGRGYL VDNSTGVSTV VFPSPTRIGD 121 SVTVCDLYGK FSLYPLTIDP NGHPMYGSVE PMTLSTDSVS ATFTWSGDAR GWIVTAGVGL 181 GQGRVYSRTI FTETVASDTA QVTLTTQPSI VDVYVDGKRL LESKYSLNGF NVDFSPSIPS 241 GSELQVIQYV PIQLGDGSGG SGGGTVITWI YNSGSAVGGE TEIELDVDAE DVSEIFIDGS 301 RQQKGLGFTY DSVTKIITLA DELEAGDEW WINGDPTLY NQIDRTPNEV ARSVNVPNSQ 361 VILSSDTITK LDGKTVIYDV VAQKIWGLPS GIPTGASIVS VSGSNLSYAP GNVWPLLPA 421 PGSKDALEAY KGELLAGNTG LVGANAVVVT PQGTLAEMQY YVTPEQFSHL VTAGEYVDEN 481 TDFTLAVQGA VDYAASHPGV IVRGTEKVYG VGRVLVTVGV KVIDGLKLKC IVANTDTLLY 541 SFVDTGHTDL QIRNCILNGN NNTRKGIIVS GVIRATIEKN YVYGLDGTGE AYGIRIGTTS 601 TTSMNINNKI SENVIEMPTD PWAGTGNYAI CGIGMIGQIT SLYGGLDTNA GVPLFPSTIT 661 LRDTIIEGNF ISGGTHGVQG LGLFRTLITK NHIIGNTHRN INLSPNCQRV NVVGNLLIDG 721 GSSGVNVAWG CRWINISGNH IQTSTAAVSP SDDAAIQLYK GVDQCTVSGN TILGDWKYSV 781 YMGAGVTNVS VNANGLFAGS LASIAVESDW VLTADYPLAI YSSSRNPNTT PIAGDTGNIN 841 IGGNAYGAGS CAIYLAATNN KAMYNVNIHD EVINSVTSRP HVVYAYDAGT LMTDGSLTNI 901 AARGATTSKY YLSRGRGAFN VIRDVTALDD PKGEVTVSGG TPSAVFGPNL YIASGTITDF 961 TGAQSGDIIN LRMGDGVVLT HNSTVMRLKG GVNATASGGL AIMTLQRRAG IWFEMSRNF List of the peptides identified by mass spectrometry (Band-B-108 kDa)

Mass m/z [MH+] error Peptide Sequence (Meas.) (calc.) (Da) 776.434 776.441 -0.007 917-923 GAFNVIR 989.553 989.564 -0.009 915-923 GRGAFNVIR 1061.551 1061.560 -0.009 691-699 NHIIGNTHR 1135.635 1135.636 -0.001 375-384 TVIYDVVAQK 1187.588 1187.594 -0.006 305-315 GLGFTYDSVTK 1215.582 1215.590 -0.008 700-709 NINLSPNCQR 1426.683 1426.688 -0.005 118-130 IGDSVTVCDLYGK 1445.882 1445.884 -0.002 508-521 VYGVGRVLVTVGVK 1455.884 1455.889 -0.005 566-579 GIIVSGVIRATIEK 1747.823 1747.828 -0.005 580-595 NYVYGLDGTGEAYGIR 1902.012 1902.018 -0.006 353-370 SVNVPNSQVILSSDTITK 2096.060 2096.066 -0.006 98-117 GYLVDNSTGVSTVVFPSPTR 2315.229 2315.245 -0.016 353-374 SVNVPNSQVILSSDTITKLDGK 2388.231 2388.231 -0.000 663-685 DTIIEGNFISGGTHGVQGLGLFR 2752.354 2752.361 -0.007 530-553 CIVANTDTLLYSFVDTGHTDLQIR List of the peptides identified by mass spectrometry (Band-B-108 kDa) (continued...)

Mass m/z [MH+] error Peptide Sequence (Meas.) (calc.) (Da) 2944.573 2944.589 -0.016 663-690 DTIIEGNFISGGTHGVQGLG LFRTLITK 3000.501 3000.506 -0.005 733-760 WINISGNHIQTSTAAVSPSDDAAIQLYK 3199.624 3199.626 -0.004 189-218 TIFTETVASDTAQVTLTTQPSIVDVYVDGK 3226.518 3226.532 -0.014 580-609 NYVYGLDGTGEAYGIRIGTTSTTSMNINNK(Met-OX) 3284.701 3284.690 0.011 316-345 IITLADELEAGDEVVVVINGDPTLYNQIDR 3355.701 3355.727 -0.016 189-219 TIFTETVASDTAQVTLTTQPSIVDVYVDGKR 3565.685 3565.698 -0.013 826-861 NPNTTPIAGDTGNINiGGNAYGAGSCAIYLAATNNK 3858.981 3859.100 -0.019 385-424 IWGLPSGIPTGASIVSVSGSNLSYAPGNVVVPLLPAPGSK 3982.990 3983.003 -0.013 933-972 GEVTVSGGTPSAVFGPNLYIASGTITDFTGAQSGDIINLR 4052.093 4052.082 0.011 316-352 IITLADELEAGDEVVVVINGDPTLYNQIDRTPNEVAR 1422.777

1520.660

1859.990 1702.927 1120.597

3324.755 3227.561 2813.471 358.259 3632.834 4410.238 2924.420 in, llilll, K- i-"*!"'^ "4 1000 1500 2000 2500 3000 3500 4000 4500 m/z, gjijy

MALDI QqTOF mass spectrum of the tryptic digest of protein band C at 68 kDa Protein identification results by MS and MS/MS (Band-C-68 kDa)

Name: Orf00195; Gpl8 tail sheath protein [phiSboM-AG3] MW: 68215.33 Da Sequence coverage: 43.1 %

001 MATQSFSVAP SVQWTERDAT LQTSPSVVVQ GATVGKFQWG EAELPVLVTG 051 GETGLVKKFF KPNDATATDF LVIADFLSYS SVAWVTRVVG PAARNAVTKG 101 QTAILIRNKL DFETASPSAS ITWTGRYAGS LGNDVAINVC DAAGFPTWEF 151 RNNFAYAPQA GEYHIVIVDK VGRITDSSGA VGQVDRISVS GTATGAGSIS ^ 201 VAGEDVAYTD TDTPATLATK IGTALTALTD VYSSVWKSN TVTVTHKAIG 10 251 PQTVTAIVPD ANGLTATAVT TTVGASGSII EKYELMQATQ GSKKSDGSNA 301 YFKDVINDTS NWVYTFATTL AAGVTELEGG VDDYTGNRVA AIEALNNAEA 351 YDAKPVFAFC EELIEQQTLI DLSTERKDTV SFVSPLRDW VGNRGREMED 401 VVAWRESLVRDSSYFFMDDN WAYVYDKYND KMRWIPACGG TAGVWARSIE 451 IAGIYKSPAF HNRGKYNNYN RMAWSASSDE RAVLYRNQIN SIVTFSNEGI 501 VLYGDKTGLT RPSAFDRINV RGLFIMAEQN IAAIAKYYLG ENNDEFTRSL 551 FSNAVRPYIR QLANMGAIYD GQVKCDADNN TADIIAANQM VAGIWLKPEY 601 SINWVYLDFA AVRPDMEFSE IETGGGIVAA S List of the peptides identified by mass spectrometry (Band-C-68 kDa)

Mass m/z [MH+] error Peptide Sequence (Meas.) (calc.) (Da) 993.561 993.562 -0.002 448-456 SIEIAGIYK 1120.597 1120.600 -0.003 378-387 DTVSFVSPLR 1220.636 1220.638 -0.002 507-517 TGLTRPSAFDR 1422.777 1422.785 -0.008 549-560 SLFSNAVRPYIR 1520.660 1520.665 -0.005 537-548 YYLGENNDEFTR 1605.864 1605.867 -0.003 522-536 GLFIMAEQNIAAIAK (Met-OX) 1702.927 1702.935 -0.008 507-521 TGLTRPSAFDRINVR 1837.024 1837.032 -0.008 221-238 IGTALTALTDVYSSVVVK 1859.990 1859.997 -0.007 378-394 DTVSFVSPLRDVVVGNR 2149.064 2149.071 -0.011 152-170 NNFAYAPQAGEYHIVIVDK 2211.125 2211.129 -0.004 487-506 NQINSIVTFSNEGIVLYGDK 2230.164 2230.175 -0.011 37-57 FQWGEAELPVLVTGGETGLVK 2358.259 2358.270 -0.011 37-58 FQWGEAELPVLVTGGETGLVKK 2813.471 2813.483 -0.012 482-506 AVLYRNQINSIVTFSNEGIV LYGDK 2924.420 2924.432 -0.012 537-560 YYLGENNDEFTRSLFSNAVRPYIR List of the peptides identified by mass spectrometry (Band-C-68 kDa) (Continued..)

Mass m/z [MH+] error Peptide Sequence (Meas.) (calc.) (Da) 3227.561 3227.580 -0.019 187-220 ISVSGTATGAGSISVAGEDVAYTDTDTPATLATK 3324.775 3324.790 -0.015 248-282 AIGPQTVTAIVPDANGLTATAVTTTVGASGSIIEK 4410.238 4410.217 0.021 339-377 VAAIEALNNAEAYDAKPVFAFCEELIEQQTLIDLSTERK 1123.538 1216.629 1963.982 J2047.975 ! 2077.951

3059.606

4180.866

4056.936 3936.994 3424.653 4314.029

Krrh'. ••TrrrFrL]-"^'-r^\' 1000 1500 2000 2500 3000 3500 4000 4500 5000 mfe. mm

MALDI QqTOF mass spectrum of the tryptic digest of protein band D at 45 kDa Protein identification results by MS and MS/MS (Band-D-45 kDa)

Name: Orf00185; Gp23, major head protein [phiSboM-AG3] MW: 47974.42Da Sequence coverage: 64.1%

001 MAKKLVTEQM REQWLPVLQK ESESIQPLSA ENVAVRLLQN QAEWNAKNLG 051 ESDAPGSVNN SVGKWQPVLI DMAKRLAPIN IAMDFFGVQP LSGPDGQIFA 101 LRARQGVGDS SNTQQSRKEL FMQEADSGYS GDGTVQAGDP SGFTQAEIEG 151 SGAGVTTIGK GMPTTDAELL GTTTNPWARV GITVQKATVT AKSRGLYADY 201 SHELRQDMMA IHGEDVDSIL SDVMVTEIQA EMNREFIRTM NFSAVRFKKF 251 GANGWDIST DISGRWALEK WKYMTFMLEV EANGIGVDTR RGKGNRVLCS 301 PNVASALAMS GMLDYAPALQ ENTKLAIDPT GQTFAGVLSN GMRVYIDPYA 351 VAEYITLAYK GATALDAGIF FAPYVPLEMY RTQGETTFSP RMAFKTRYGI 401 CANPFVQIPA NQDPQVYVTA DGIAQDSNPY FRKGLIKGLF List of the peptides identified by mass spectrometry (Band-D-45 kDa)

Mass m/z [MH+] error Peptide Sequence (Meas.) (calc.) (Da) 925.548 925.546 -0.008 239-246 TMNFSAVR 1123.538 1123.539 -0.001 382-391 TQGETTFSPR 1200.644 1200.645 -0.001 65-74 WQPVLIDMAK 1216.629 1216.639 -0.010 65-74 WQPVLIDMAK (Met-OX) 1323.628 1323.633 -0.005 195-205 GLYADYSHELR 1486.743 1486.747 -0.004 235-246 EFIRTMNFSAVR (Met-OX) 1491.706 1491.714 -0.008 105-118 QGVGDSSNTQQSRK 1566.764 1566.766 -0.002 193-205 SRGLYADYSHELR 1607.801 1607.802 -0.001 250-265 FGANGVVDISTDISGR 1735.884 1735.897 -0.013 249-265 KFGANGVVDISTDISGR 1963.982 1963.991 -0.009 325-343 LAIDPTGQTFAGVLSNGMR 1992.030 1992.036 -0.006 344-360 VYIDPYAVAEYITLAYK 2047.975 2047.975 0.000 161-179 GMPTTDAELLGTTTNPWAR(Met-OX) 2077.951 2077.957 -0.006 273-290 YMTFMLEVEANGIGVDTR 2318.154 2318.152 0.002 361-381 GATALDAGIFFAPYVPLEMYR List of the peptides identified by mass spectrometry (Band-D-45 kDa) (Continued..)

m/z [MH+] Mass Peptide Sequence (Meas.) (calc.) error (Da) 2773.406 2773.418 -0.012 161-186 GMPTTDAELLGTTTNPWARVGITVQK 2903.503 2903.512 -0.009 76-102 LAPINIAMDFFGVQPLSGPDGQIFALR 3043.595 3043.618 -0.023 75-102 RLAPINIAMDFFGVQPLSGPDGQIFALR 3059.606 3059.613 -0.007 75-102 RLAPINIAMDFFGVQPLSGPDGQIFALR(Met-OX) 3422.667 3422.672 -0.005 361-391 GATALDAGIFFAPYVPLEMY RTQGETTFSPR (Met-OX) 3936.994 3937.009 -0.015 325-360 LAIDPTGQTFAGVLSNGMRVYIDPYAVAEYITLAYK 4056.936 4056.954 -0.018 398-433 YGICANPFVQIPANQDPQVYVTADGIAQDSNPYFRK 4180.866 4180.877 -0.011 119-160 ELFMQEADSGYSGDGTVQAGDPSGFTQAEIEGSGAGVTTI GK (Met-OX) 1393.838

3156.769 1000

2248.267 800

600

c 0 - 400

-J 200

lflu'f--f-lV" 1'j '•' fiill ill.,, 1000 2500 3000 3500 4000 ra&. aimi

MALDI QqTOF mass spectrum of the tryptic digest of protein band E at 22 kDa Protein identification results by MS and MS/MS (SboM-E-22 kDa)

Name: Orf00006; putative tail protein; [phiSboM-AG3] MW: 27617.86Da Sequence coverage: 16.4%

001 MPTITVLVAP EWRNKPETE RLHTVTGTAK GWEKTSLNQD PDEILTECKG 051 LDALLTKSNL QADGVTKVDP TKPIGFEISY EIHDPSAVLT TGLTITPATA 101 GGEVGQVVEL LATVAPANAT YQGVNWYSGD ITKAVHIGGG KFKLLAPGSV 151 TVYGVTIEGG HTDSTVITVA GALALSTDLA ASQDVTAGAD ATFTIAATGG 201 TTPYTYAWYF SDVPGGEGVV IDAGANATAA TASLVITAVD AADEGEYWCV 251 VEDADGHSVT STRCEMAVV List of the peptides identified by mass spectrometry (AboM-E-22 kDa)

Mass m/z [MH+] error Peptide Sequence (Meas.) (calc.) (Da) 1393.838 1393.841 -0.003 2-14 PTITVLVAPEVVR 1762.812 1762.816 -0.004 35-49 TSLNQDPDEILTECK 1776.829 1776.832 -0.003 34-48 TSLNQDPDEILTECK (Cys-acrylamide) 1781.943 1781.950 -0.007 15-30 NKPETERLHTVTGTAK 2248.267 2248.266 0.001 2-21 PTITVLVAPEVVRNKPETER 3156.769 3156.774 -0.005 2-30 PTITVLVAPEVVRNKPETER LHTVTGTAK 930.466 1677.892 2000

971.486 1600-

1086.560

11200 1139.516 2826.321 1368.669 2810.330; c 2696.446 c 800

2020.077 400- 00 836.41 4353316 ty.,^1 Jv-A, U4Ik- , ) . i ,i-1, i. 1000 1500 2000 2500 3000 3500 4000 4500 cafe, saw

MALDI QqTOF mass spectrum of the tryptic digest of protein band F at 18 kDa Protein identification results by MS and MS/MS (Band-F-18 kDa)

Name: OrfD0192; Gpl9 tail tube protein; [phiSboM-AG3] MW: 19782.17 Da Sequence coverage: 55.9%

001 MATVNEFRAA MSRGGGVQRQ HRWRVTVNFP AFVAGADTIR DVSLLAVTTN 051 TPTGQLGEIL VPWGGRELPF PGDRRFEALP ITFINVVNNS AYNAFEVWQQ 101 C1NGSNSNRA AANPDDYFRD VIMELLDAND NVTKTWTLQG GWPQNLGQLE 151 LDMSAMDSYT QFTVDLRYFQ AISDKSL List of the peptides identified by mass spectrometry (Band-F-18 kDa)

Mass m/z [MH+] error Peptide Sequence (Meas.) (calc.) (Da) 836.419 836.426 -0.007 2-8 ATVNEFR 930.466 930.468 -0.002 67-74 ELPFPGDR 971.486 971.483 0.003 168-175 YFQAISDK 1086.560 1086.569 -0.009 67-75 ELPFPGDRR 1139.516 1139.512 0.004 110-119 AAANPDDYFR 1368.669 1368.669 0.000 2-13 ATVNEFRAAMSR (Met-OX) 1677.892 1677.896 -0.004 25-40 VTVNFPAFVAGADTIR 2020.077 2020.076 0.001 23-40 WRVTVNFPAFVAGADTIR 2694.446 2694.446 0.000 41-66 DVSLLAVTTNTPTGQLGEILVPWGGR 2810.330 2810.330 0.000 110-134 AAANPDDYFRDVIMELLDANDNVTK 2826.321 2826.325 -0.004 110-134 AAANPDDYFRDVIMELLDANDNVTK (Met-OX) 4353.316 4353.324 -0.008 25-66 VTVNFPAFVAGADTIRDVSLLAVTTNTPTGQLGEILVPWGGR