Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author. ANTIMICROBIAL PEPTIDES ISOLATED FROM OVINE BLOOD NEUTROPHILS A thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Biotechnology at Massey University, Palmerston North, New Zealand. Rachel C Anderson 2005 Abstract ABSTRACT The aim of the research presented in this thesis was to investigate the properties of the antimicrobial peptides found in ovine blood, in order to assess their potential as a high-value product. Due to the large number of lambs and sheep that are slaughtered New Zealand (approximately 25 million lamb and 5 million sheep per year), there are considerable volumes of ovine blood available for processing (approximately 40 million litres per year). Currently this blood is dried and sold as a low value product. The first objective of this research was to purify and characterise the antimicrobial peptides isolated from ovine neutrophils. A number of proline/arginine-rich peptides, as well as two small fragments of larger proteins, that displayed antimicrobial activity were identified. The second objective of this research was to investigate the mechanism of action of ovine antimicrobial peptides. For this investigation, three ovine peptides, a-helical SMAP29 and proline/arginine-rich OaBac5mini and OaBac7.5mini, were synthesised. Of these, SMAP29 was the most potent. The three peptides all bound Gram-negative bacterial LPS and caused the outer membrane to be permeabilised. SMAP29 caused significant depolarisation of the cytoplasmic membrane that led to cell lysis. However, the other two peptides only caused slight depolarisation of the cytoplasmic membrane, which indicates that they probably passed through the membrane to interact with the inner cellular contents. The third objective of this research was to investigate the morphological changes to bacterial cells induced by the ovine antimicrobial peptides. Transmission electron microscopy and atomic force microscopy confirmed that SMAP29 caused significant damage to the membranes of bacterial cells and induced cell lysis; whereas, OaBac5mini caused minor alterations to the bacterial membranes but did not induce cell lysis. The fourth objective of this research was to determine the effect of the environmental conditions on the activity of the peptides. The peptides were very stable over a range of pH values and when heated to temperatures up to 80°C. The activity of the peptides decreased slightly in the presence of monovalent cations and was inhibited by the presence of divalent cations. The peptides were significantly more active in combination than individually, and they were strongly synergistic with polymyxin B, a peptide antibiotic. The final objective of this research was to develop a pilot-scale extraction process for the isolation of antimicrobial peptides from ovine blood. The laboratory-scale process was simplified and adapted to design a process that could be used industrially. The crude pilot-plant extract was active against a broad-range of food pathogens and disease causing organisms. The antimicrobial peptides found in ovine blood have the potential to be used as biopreservatives for chilled lamb products, or in a topical cream for cuts and grazes; therefore it is recommended that further research is carried out to investigate the above applications and. if successful, the feasibility of commercialising the technology. iii Acknowledgements ACKNOWLEDGEMENTS First and foremost I would like to thank my supervisors. Or Pak-Lam Yu, thank you for taking a keen interest in my project and for teaching me all I need to know to be a successful researcher. Or Brian Wilkinson, thank you for being around when I needed that extra bit of help or advice. Professor [an Maddox, thank you for joining the team to help me with the preparation of this manuscript. I would also like to thank Professor Robert Hancock and his team for allowing me to visit their laboratory at the Department of Microbiology and Immunology, University of British Columbia, for three months, and for supervising and assisting with my bacterial membrane interaction experiments. This work was made possible by the financial support I received from Meat and Wool New Zealand (formerly MeatNZ), in the form of both a doctoral scholarship and project funding. The project was also partially funded by the Massey University Research Fund (MURF), and my research trip to UBC was funded by the C. Alma Baker Trust. This work was made easier by help [ received from numerous people including the ITE technical staff, especially Anne-Marie Jackson and Mike Sahayam, and the staff of Feilding Lamb Packers, who collected the sheep blood for my experiments. [ also received valuable help from the undergraduate and foreign-intern students that assisted on various parts of this project, including David Houlding (laboratory extraction process), Adi Sugiarto (RP-HPLC), Marie Bourin (crude extract MICs) and Andrew Lister (pilot-scale extractions). I received assistance from Aaron Hicks (Institute of Veterinary, Animal and Biomedical Sciences) to prepare the TEM samples, HortResearch to image the TEM samples, and Associate Professor Richard Haverkamp to image the AFM samples. Finally, [ would like to thank family and friends who helped keep me sane throughout this whole process. Other postgrads, especially Craig, Stephen, Roland and Anna, it always helped to know that there were others who shared the same, or worse, difficulties - Good luck to you all. Regan, thank you for caring enough to wade through this thesis to find the spelling and grammatical mistakes - a best friend who doubles as a proof-reader, what more could I ask for? Dad, I would never have made it this far without the support of you and "The Anderson Trust" - I think I was the best fed undergraduate student in town. And finally, Peter, there are not words to describe how much I appreciate you - I look forward to the future we will spend together. I dedicate this thesis to my mother, who [ know would have been proud. Her encouragement, support and love will be with me always. v Table of Contents TABLE OF CONTENTS Abstract ................................................................................................................................ iii Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v Table of contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi List of figures . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii List of tables .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvi List of abbreviations . . . . . . . . . . . . . . . .. . . . . . . .. . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvi i i List of publications . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . .. . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . xx CHAPTER 1 PROJECT INTRODUCTION 1.1 Reason for the research ................................................................................................ 1 1.2 Project objectives .......................................................... . . ........................ . ................ .... 2 CHAPTER 2 ANTIMICROBIAL PEPTIDES LITERATURE REVIEW 2.1 Introduction . .. ............................ ....... . . ......................................................................... 4 2.2 Antimicrobial peptides ................................................................... . ............................. 5 2.2.1 Animal antimicrobial peptides ................. ............................................................. 5 2.2.2 Plant antimicrobial peptides ........................................................ . . ...................... 11 2.2.3 Microbial antimicrobial peptides ........................................................... ........... ... 15 2.3 The animal immune system ........................................................................................ 18 2.3.1 Innate immunity ................. .. . .......... .................................................................... 19 2.3.2 Adaptive immunity ... . .......... . .......... . ............ . . .. . . . . .. ..... ........................................ 24 2.3.3 Role of antimicrobial peptides in animal immune systems . ................... .............. 26 2.4 Potential applications of antimicrobial peptides .......................................................... 28 2.4.1 Applications for antimicrobial peptides ............................... ......... ....................... 29 2.4.2 Possible applications for ovine blood antimicrobial peptides ............................... 31 2.5 Purification and characterisation of animal antimicrobial peptides .............................. 35 2.5.1 Techniques to purify and characterise antimicrobial peptides .............................. 36 2.5.2 Livestock blood antimicrobial peptides . . ......... .................................................... 39 2.5.3 Ovine antimicrobial peptides .............................................................................. 45 2.6 Mechanism of action of animal antimicrobial peptides ............................................... 46 2.6.1 Techniques to determine mechanisms of action of antimicrobial peptides ........... 47 vi Table of Contents 2 .6 .2 Mechanisms of action .... . .... . ................................ ... .... .. . . . . .. ................................ 48 2 .6 .3 Mechanism of action of ovine antimicrobial peptides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1 2 .7 Morphological changes to microbial cells induced by animal antimicrobial peptides .. 5 1 2 .7 . 1 Techniques to investigate morphological changes ..................... ......... . . .. . . . ...... . . . . 52 2 .7 .2 Morphological changes . . ... ............... . ........ . . . ......... . . . .. . . . .. . . . ........................... . ..... 53 2 .7 .3 Morphological changes induced by ovine antimicrobial peptides . . . . . . . . . . . . . . . . . . . . . . . . 5 3 2 . 8 Effect of environmental condit ions on activity of animal antimicrobial peptides ... ..... 54 2 .8 . 1 Techniques to determine effect of environmental conditions ............... .... . . . ......... 54 2 .8 .2 Effects of environmental condit ions ........... . . . . ...... .. ......... . .......... . . . ...................... 55 2 .8 . 3 Effects of environmental conditions on ovine antimicrobial peptides . . ................ 56 2 .9 Pilot-scale extraction of animal antimicrobial peptides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 2 . 1 0 Conclusions ...... ........ ... . . . ........... . ... .... . ......... .... ....................... . . . ..... .. ............. .. . ..... . . .. 57 CHAPTER 3 MATERIALS AND METHODS 3 . 1 Materials and methods used for peptide purification . .... . ......... .. . .... . . .. ......................... 59 3 . 1 . 1 Crude extraction ... . . ... . .... . ... . .......................... . . ...... ..... . . ..... . ......... . . . . . . . . . .. . .... . . .. . . . 59 3 . 1 .2 Gel electrophoresis ................ . .... . . . . .. . .. ... . . ... . . ............................. ................... . .. . . 60 3 . 1 . 3 Gel filtration .................. . ................. . . . . ......... . . ................. . .................................. 61 3 . 1 .4 Cationic-exchange chromatography .... . . . . . . . . . . . .... ..... . . . . . . . . . . . ......... ... .. ...... . . .. ......... 6 1 3 . 1 . 5 Peptide purification using HPLC ............................ . .... . . . . ... . . . . ... ......................... 62 3 . 1 .6 Radial diffusion plate assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 3 . 1 . 7 Radial diffusion plate assay MIC method . ...................................... ............ . .. . . . ... 63 3 . 1 . 8 Mass spectroscopy ......... . .. ... . . . . . . . . ..... . . ....................... . . . . . . ...... . . .. ...... . . . . . . . . . .. . . . . . ... 64 3 . 1 .9 N -terminal sequencing ... . ..... . . . .... . .............................. .. . . . . . . ........ . .... . . . .. . ..... . . . . ..... 65 3 . 1 . 1 0 Peptide characterisat ion ...... ... . . . . . . . . . ....................................... .... . . . ............. ... . . . .. . 65 3 . 1 . 1 1 Analysis of proline/arginine-rich sequences ... . .......... . . . . .. . . . .. .. .. . . . ..... . . . .. . . ... . ..... ... 66 3 .2 Materials and methods used for mechanism of action tests . . .. .. ..... ....... . . ........ .. . . . . . . . . . . . 66 3 .2 . 1 Peptide synthesis ................................... . ........ .......... ............ .............................. 66 3 .2 .2 Micro-broth dilution MIC method ...... ... . . . ......................................................... .. 67 3 .2 . 3 Circular dichroism spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 3 .2 .4 LPS binding assay . ........ .... . . . ............... . . ........ . . ........... .... ...... .. . . . ........... . .... . ... . .... 69 3 .2 . 5 Outer membrane permeabil isation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 3 .2 .6 Cytoplasmic membrane depolarisation ..... . ........ .. . ............................................... 70 vii Table of Contents 3 .2 .7 Optical density and viable cell counts over time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1 3 .2 .8 Peptide-DNA binding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1 3 .3 Materials and methods used to investigate bacterial cell morphological changes . . . . . . . . 72 3 . 3 . 1 Transmission electron rrUcroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 3 . 3 . 2 AtorrUc force rrUcroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 3 .4 Materials and methods used to assess the effect of condit ions on peptide activity . . . . . . . 74 3 .4. 1 Salt effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 3 .4 .2 Cation effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 3 .4 .3 pH effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 3.4.4 Temperature effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 3 .4 .5 Synergistic effects between test peptides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 3 .4 .6 Synergistic effects between test peptides and com on antibiotics . . . . . . . . . . . . . . . . . . . . . . . 76 3 . 5 Materials and methods used for the pilot-scale extraction of antimicrobial peptides from ovine blood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 3 . 5 . 1 Crude extraction process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 3 . 5 .2 Minimum inhibitory concentrations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 3 . 5 . 3 Transmission electron microscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 3 . 5 .4 Yield calculat ions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 CHAPTER 4 ISOLATION AND CHARACTERISATION OF ANTIMICROBIAL PEPTIDES FROM OVINE NEUTROPHILS 4. 1 Introduction . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 4.2 Extraction of crude antimicrobial solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1 4 .3 Purification of antimicrobial peptides using gel filtration and RP-HPLC . . . . . . . . . . . . . . . . . . . . 83 4.4 Characterisation of OaBac5 and variants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 4.5 Characterisation of truncated OaBac7.5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 4.6 Characterisation ofOaBac l l and truncates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 4 .7 Minimum inh ibitory concentrations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 4 .8 Sequence analysis of proline/arginine-rich peptides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 4.9 Purification of antimicrobial peptides using cationic exchange chromatography and RP- HPLC ............................................................ ............................................................. 98 4. 1 0 Other predicted cathelicidins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 03 4.11 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 vii i Table of Contents CHAPTER 5 SPECTRUM OF ACTIVITY AND BACTERIAL MEMBRANE INTERACTIONS O F SYNTHETIC OVINE CATHELlCIDINS 5.1 I ntroduction .................................... ........................... . ... ............. ............................. 106 5.2 Minimum inh ibitory concentrations .... .... ... . . ........................................................... . 108 5.3 Circular d ichro ism spectroscopy ................................. ............................................. 111 5.4 LPS binding assay ................................ ................................................................ .... 112 5.5 Outer membrane permeabil isation ........ .......................... . .................. ....................... 116 5.6 Cytoplasmic membrane depolarisation ......................... .............................. .. .. . ... ...... 119 5.7 Kill curves .. ............................................................. ................................................ 123 5.8 DNA binding ................. . ............... . ............................ . ............................... ............. 124 5.9 Conc lusions ............................. .... .... .. . ..... .... ............................................................ 126 CHAPTERS MORPHOLOGY OF BACTERIAL CELLS TREATED WITH SYNTHETIC OVINE CATHELlCIDINS 6.1 Introduction ................... . ...... .......... . . ..... .................................................................. 128 6.2 E. coli TEM results ...... ..................................................................... .............. ......... 128 6.3 S. aureus TEM results ............................................... .. ... . ........... . .......... ................... 129 6.4 S. aureus AFM method development ..................... ............. . .................................... 132 6.5 S. aureus AFM results .............................................................................................. 135 6.6 E. coli AFM method development problems .. . ......................................................... 139 6.7 Conclusions ................... ................. ............ .. ................. .......................................... 141 CHAPTER 7 FACTORS AFFECTING THE ANTIMICROBIAL ACTIVITY OF SYNTHETIC OVINE CATHELlC IDINS AGAINST E. COLl0157:H7 7. 1 Introduction .................. .............. ........... . . ............................................................ . . . . 143 7.2 Effect of salt .......... ........... ... ........... .. . ...................................................................... 144 7.3 Effect of metal ions ................................................ ............................. ..................... 145 7.4 Effect of pH ................ ............................. ................................................................ 148 7.5 Effect of temperature ...... ........................ . . . .............................................................. 150 7.6 S ynergy between peptides ............................................ .......... ..................... ............. 151 7.7 Synergy between peptides and known antibiotics .. . . . . . .. . . ... . ..... . . . ... .. . . . .... ..... ...... .... . . . 153 ix Table of Contents 7 .8 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 54 CHAPTER 8 PILOT -SCALE EXTRACTION OF ANTIMICROBIAL PEPTIDES F ROM OVINE BLOOD 8 . 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 57 8 . 2 Crude extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 58 8 . 3 Minimum inhibitory concentrations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . 1 6 1 8 . 4 Transmission electron microscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 63 8 . 5 Y ield calculation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 65 8 .6 Industrial-scale process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 66 8 . 7 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 69 CHAPTER 9 CONCLUSIONS AND RECOMMENDATIONS 9 . 1 Summary of research conclusions . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 1 7 1 9 .2 Recommendations for future research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . 1 74 9 . 3 Final conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 76 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 77 APPENDIX A1 RAW DATA AND CALCULATIONS FROM C HARACTERISATION STUDIES A 1 . 1 Mass Spectra of the purified HPLC peaks . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 97 A 1 .2 Example calculation of confidence intervals from plate assay raw data . . . . . . . . . . . . . . . . . . . . 203 A1 . 3 Raw data, calculated MICs and 95% confidence intervals for the MICs of the purified peptides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 APPENDIX A2 RAW DATA AND CALCULATIONS FROM MECHANISM OF ACTION STUDIES A2. 1 Raw data from the micro-broth dilution MIC method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 A2 .2 Example calculation of the mean MIC and confidence intervals for the mean from the raw data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1 1 A2 .3 Raw Data from the LPS binding assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1 2 A2 .4 Calculation of Imax and I so from LPS binding assay raw data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1 5 A2.5 Raw data from the outer membrane permeabil isat ion assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1 7 x Table of Contents A2.6 Analysis of variance ofNPN uptake data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1 9 A2 .7 Raw data from the outer inner membrane depolarisation assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 A2. 8 Analysis of variance ofDiSC35 release data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222 APPENDIX A3 RAW DATA AND CALCULATIONS FROM EFFECT OF CONDITIONS STUDIES A3 . 1 Raw data of MICs at different salt concentrations . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . 223 A3 .2 Raw data ofMICs at different cation concentrations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224 A3 . 3 Raw data of MICs at different pH values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226 A3 .4 Raw data of MICs after heating to different temperatures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227 A3 . 5 Example calculation of the mean MIC and confidence intervals for the mean from the raw data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228 A3 .6 Analysis of variance of MIC data from different conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229 APPENDIX A4 RAW DATA AND CALCULATIONS FROM PILOT -SCALE EXTRACTION STUDIES A4. 1 Raw data from the micro-broth dilution MIC method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 1 A4.2 Example calculation of the mean MIC and confidence intervals for the mean from the raw data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232 A4.3 Calculation of the settling velocit ies of different types of blood cells . . . . . . . . . . . . . . . . . . . . . . . 233 APPENDIX A5 PEER-REVIEWED PUBLICATIONS A5 . 1 Ovine antimicrobial peptides: new products from an age-old industry . . . . . . . . . . . . . . . . . . . . . . 236 A5 .2 Iso lation and characterisation of proline/arginine-rich cathelicidin peptides from ovine neutrophils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244 A5 .3 Antimicrobial activity and bacterial membrane interaction of ovine-derived cathelicidins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254 A5.4 Investigation of morphological changes to S. aureus induced by ovine-derived antimicrobial peptides using TEM and AFM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259 A5 .5 Factors affecting the antimicrobial activity of ovine-derived cathelicidins against E. coli 0 1 57 :H7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266 xi List of Figures LIST OF FIGURES Figure 2 . 1 - Examples of the four structural classes of cationic ant imicrobial peptides . . . . . . . . . 6 Figure 2 .2 - Examples of the three groups of defensins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Figure 2 .3 - Schematic diagram of a cathelicidin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Figure 2 .4 - Schematic diagram of the gene for the human cathelicidin LL-37 . . . . . . . . . . . . . . . . . . . . 1 1 Figure 2 .5 - Structure of thionins. 'A' shows the secondary structures of thionins with six and eight cysteine residues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 2 Figure 2 .6 - Structure of plant defensins. 'A' shows the secondary structures of plant defensins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 3 Figure 2 . 7 - Structure of lipid transfer proteins. 'A' shows the secondary structures of lipid transfer proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 3 Figure 2 . 8 - Structure of hevein- and knottin-type pept ides. 'A' shows the secondary structures of hevein- and knottin-type peptides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 4 Figure 2 .9 - Structures of four Class I bacterioc ins. Nisin A, epidermin and lacticin 48 1 are class la bacterioc ins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 6 Figure 2 . 1 0 - Schematic diagram showing the principle mechanisms of innate and adaptive immunity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 8 Figure 2. 1 1 - Functions of the epithelia in innate immunity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 9 Figure 2 . 1 2 - Mechanism of phagocytosis and intracellular kill ing of microbes. NO is nitric oxide and ROI is reactive oxygen intermediate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Figure 2 . 1 3 - Functions of the natural kil ler (NK) cells. A) NK cells kill infected host cells and B) NK cells activate macrophages to kil l phagocytosed microbes. I L- 1 2 in interleukin- 1 2 and IFN-y is interferon-y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Figure 2 . 1 4 - Pathways of activation of the complement system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Figure 2 . 1 5 - Types and mechanisms of adaptive immunity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Figure 2 . 1 6 - Specificity and memory in adaptive immunity illustrated by primary and secondary immune response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 5 Figure 2 . 1 7 - Proposed roles of antimicrobial peptides within the innate immune system . . . . . 27 Figure 2 . 1 8 - Schematic diagram of the proposed mechanisms of permeability change o f cytoplasmic membranes caused by antimicrobial peptides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Figure 3 . 1 - Graph of Ln peptide concentration versus clearing size showing the relationsh ip between the line-of-best-fit and the bounds of the 95% confidence intervals for the line . ..... ... ... . . . ......... . . .. . . . . ....... . ... . . . . . ...... . . . . . . . ... . .. . . ....... . . .. . . . . . .... .. . .. .. . . ..... . .. . . 64 xii List of Figures Figure 3 .2 - Graph of Ln peptide concentration versus c learing size showing the relationship between the l ine-of-best-fit and the bounds of the 95% confidence intervals for the line when the bounds do not cross the x-axis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Figure 4 . 1 - Flowchart showing the process used to extract the crude antimicrobial solution from ovine blood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1 Figure 4 .2 - Images of typical stained blood samples during the extraction process . . . . . . . . . . . . 82 Figure 4 .3 - Images of typical plate assay results of neutrophil crude extract against three test organisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 F igure 4.4 - Typical gel filtration chromatograph resulting from the addition of an ovine neutrophil crude extract into a P 1 0 gel filtrat ion column . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 Figure 4 .5 - I mage of a typical SDS-PAGE gel of ovine neutrophil extract gel filtration fractions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 Figure 4 .6 - RP-HPLC chromatograph of the second gel filtration fraction (F2) of the ovine neutrophil crude extract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 Figure 4.7 - Hydrophobicity plots of the prol ine/arginine-rich cathelic idin peptides . . . . . . . . . . . 96 Figure 4 .8 - Polarity plots of the proline/arginine-rich cathel ic idin peptides . . . . . . . . . . . . . . . . . . . . . . . . 97 Figure 4.9 - Ion-exchange chromatograph for the addit ion of the ovine neutrophil crude extract to a weak cationic exchange co lumn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Figure 4. 1 0 - RP-HLPC chromatograph of the cationic fraction ( F3 and F4) of the ovine crude extract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 00 Figure 5 . 1 - Schematic diagram showing the proposed mechanism of action of antimicrobial peptides against Gram-negative bacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 08 Figure 5 .2 - C ircular dichro ism spectra of25mM synthetic ovine antimicrobial peptides . . I I I Figure 5 . 3 - Schematic diagram showing the mechanism invo lved in the lipopolysaccharide binding assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 2 F igure 5 .4 - A typical run showing the changes in the fluorescence of dansyl po lymyxin B due to the addition of SMAP29 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . 1 1 3 Figure 5 . 5 - Lineweaver-Burke plot for a typical run of the SMAP29-LPS binding assay. 1 1 4 F igure 5 .6 - Schematic diagram showing the mechanism invo lved in the I -N-phenylnapthyl- amine (NPN) uptake assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 6 F igure 5 . 7 - Uptake of 1 -N-phenylnapthylamine (NPN) by E. coli UB 1 005 cells caused by synthetic ovine peptides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 8 F igure 5 . 8 - Schematic diagram showing the mechanism involved in the 3 ,3- dipropylthiacarbo-cyanine (DiSC35) assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 20 xiii List of Figures Figure 5.9 - Release of 3,3-dipropylthiacarbocyanine (DiSC35) dye fro m the cytoplasmic membrane of E. coli DC2 cells caused by synthetic ovine peptides . . . . . . . . . . . . . . . . 1 22 Figure 5 . 1 0 - Optical density and viable cell count over time for E. coli 0 I I I treated with synthetic ovine peptides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 24 Figure 5 . 1 1 - DNA gel showing the running pattern of different ratios of DNA and synthet ic ovine peptides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 25 Figure 6. 1 - Transmission electron microscope images taken of E. coli 0 I I I cells treated with SMAP29 and OaBac5 mini for one hour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 3 0 Figure 6.2 - Transmission electron microscope images taken of S. aureus 4 1 63 NCTC cel ls treated with SMAP29 and OaBac5 mini for one hour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 3 1 Figure 6.3 - AFM images of S. aureus NCTC 4 1 63 cells trapped on a polycarbonate membrane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 3 3 Figure 6.4 - AFM images of S. aureus NCTC 41 63 cells grouped together on a polycarbonate membrane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 33 Figure 6 .5 - AFM 3D representation of S. aureus NCTC 4 1 63 cells grouped together on a po lycarbonate membrane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 3 3 Figure 6.6 - AFM images of S. aureus NCTC 4 1 63 treated with 25).!g/mL nisin . . . . . . . . . . . . . . . 1 3 5 Figure 6 .7 - Far away AFM images o f S. aureus NCTC 4 1 63 cells on a g lass slide treated with SMAP29 and OaBac5mini for 30 minutes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 36 Figure 6.8 - Close up AFM images of S. aureus NCTC 4 1 63 cells on a g lass sl ide treated with SMAP29 and OaBac5 mini for 30 minutes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 3 7 Figure 6 .9 - AFM images of E. coli 0 1 1 1 debris after being suspended in dist i l led water. 1 39 Figure 6. 1 0 - AFM images of E. coli 0 1 1 1 cells covered with dried Muel ler-H inton broth. 1 40 Figure 6. 1 1 - AFM images of crystals that formed when E. coli 01 1 1 was suspended in phosphate buffer and dried on a g lass sl ide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 40 Figure 7. 1 - The effect of salt concentrat ion on the minimum inhibitory concentration (MIC) of synthetic ovine peptides against E. coli 01 5 7 : H7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 44 Figure 7.2 - The effect of metal ion concentrations on the minimum inhibitory concentration (MIC) of synthetic ovine peptides against E. coli 0 1 5 7 : H 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 46 Figure 7.3 - The effect of media pH on the minimum inhibitory concentration (MIC) of synthetic ovine peptides against E. coli 0 1 57:H7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 49 Figure 7.4 - The effect of heating on the minimum inhibitory concentration (MIC) of synthetic ovine peptides against E. coli 0 1 57:H7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 50 F igure 7.5 - Diagram of a microtitre plate for a typical synergy test for OaBac5 mini and OaBac7.5mini . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 52 xiv List of Figures Figure 8 . 1 - Flow diagram showing the pilot-scale process used to extract antimicrobial peptides from ovine blood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 59 Figure 8 .2 - Photograph of the pi lot-scale d isk-stack centrifuge used to separate white blood cells from plasma and red blood cells . . . . . . .. . ......... . . .. . ........... .... .. . . . . . . . . . . . . . . . . . . . . 1 60 Figure 8 . 3 - Transmission e lectron rnicroscopy images of control cells treated with 0 .0 1 % acetic acid ( left) and cells treated with ovine neutrophil crude extract (right). 1 64 Figure 8 .4 - Steps in an industrial process to produce a crude antimicrobial extract from ovine blood . .. . . . . . . . . . . . . . . . . ..... . ..... . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 67 xv list of Tables LIST OF TABLES Table 2. 1 - Amino acid sequences of the p-defensins found in l ivestock blood . . . . . . . . . . . . . . . . . . 4 1 Table 2 . 2 - Amino acid sequences of cathelicidins rich in one or more amino ac ids found in l ivestock blood. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Table 2 .3 - Amino ac id sequences of a-helical cathe licidins found in l ivestock blood . . . . . . . 44 Table 2 .4 - Amino ac id sequences of cathelic idins containing disulphide bonds found in l ivestock blood . . . . . . . . . . . .. . .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 Table 3 . 1 - Sequences of ovine ant imicrobial pept ides used for this research . . . . . . . . . . . . . . . . . . . . . 66 Table 3 .2 - Microorganisms used for micro-broth di lution lTllmmUm inhibitory concentration tests and their sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Table 3 .3 - Microorganisms used for crude extract minimum inhibitory concentration tests and their sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . . . . . 78 Table 4. 1 - Antimicrobial activity of typical ovine neutrophil extract gel fi ltration fractions against test organisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 Table 4.2 - Antimicrobial act ivity of RP-HPLC peaks from the second gel filtration fraction of the ovine neutrophil crude extract against test organisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 Table 4.3 - Comparison of masses and N-terminal sequences of Pa and Pc purified from the ovine neutrophil crude extract to known Bac5 peptides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 Table 4.4 - Comparison of mass and N-terminal sequence of Pb purified from the ovine neutrophil crude extract to OaBac7.5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 1 Table 4.5 - Comparison of masses and N-terminal sequences ofPd and Pf purified from the ovine neutrophil crude extract to Bac l l . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 1 Table 4.6 - Minimum inhibitory concentrations of peptides purified from ovine neutrophil extract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 Table 4.7 - I dent ification of repeats in the sequences of the pro line/arg inine-rich cathelic idin peptides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Table 4.8 - Antimicrobial act ivity of typical ovine neutrophil extract cationic-exchange chromatography fractions against test organisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Table 4.9 - P late assay results and molecu lar weights of antimicrobial peptides iso lated from the cationic fraction of ovine neutrophil extract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 0 1 Table 4. 1 0 - Comparison of the N-terminus of cationic Peak 1 8 to the cathel in- l ike precursor of SMAP29 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 02 Table 4. 1 1 - Comparison of the N-terminus of cationic Peak 24 to the signal peptide of T-cell surface g lycoprotein CD4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 02 xvi List of Tables Table 5 . 1 - Sequences of synthetic ovine antimicrobial peptides used for this research . . . . 1 06 Table 5 .2 - Minimum inhibitory concentrations (MIC) of synthetic ovine antimicrobial peptides against various microorganisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 09 Table 5 . 3 - Data collected and calculated for the change in dansyl polymyxin B fluorescence due to the addition of SMAP29 in a typical run . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 4 Table 5 .4 - The abil ity of synthetic ovine peptides to bind to E. coli lipopolysaccharide (LPS) using the dansyl po lymyxin B (DPX) displacement assay . . . . . . . . . . . . . . . . . . . 1 1 5 Table 5 . 5 - Data collected and calculated for change in I -N-phenylnapthylamine (NPN) fluorescence due to the addition of SMAP29 for a typical run . . . . . . . . . . . . . . . . . . . . . . . 1 1 7 Table 5 .6 - Data collected and calculated for change in 3 ,3 -dipropylthiacarbocyanine (DiSC35) fluorescence due to the addit ion of SMAP29 for a typical run . . . . . . . . 1 20 Table 7. 1 - Concentrations of metal ions in lean trimmed, raw lamb meat. . . . . . . . . . . . . . . . . . . . . . . 1 48 Table 7 .2 - Antibiotics used in the synergy tests and their mechanisms of actions . . . . . . . . . . . 1 53 Table 7 .3 - Fractional inhibitory concentrations of synthet ic ovine peptides in combination with common ant ibiotics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 54 Table 8 . 1 - Minimum inhibitory concentrations of ovine neutrophil crude extract from the pilot-scale extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 62 Table 8 .2 - Yields for pilot-scale extractions of antimicrobial peptides from ovine blood . 1 66 xvii List of Abbreviations AFM Bac BMAP BSA CD cDNA CFU ChBac DiSC35 DNA DPX EDTA FIC HLPC I so I FN-y IL- 1 2 Imax LPS LTPs MHB MIC MRSA MAP NF-KB NK cells NMR NO NPN NCLSS xviii LIST OF ABBREVIATIONS atomic force microscopy bactenicin bovine myeloid antimicrobial peptide bovine serum albumin circular dichro ism complementary DN A colony forming units Capra hircus bactenecin 3,3-dipropylthiacarbocyanine deoxyribonucleic acid dansyl po lymyxin B ethy lened iaminetetraacetic acid fractional inhibitory concentration high performance l iquid chromatography concentration of peptide required to displace half the of the maXimum displacement amount of DPX from LPS interferon-y interleukin- 1 2 maximum percentage of DPX that could be displaced from LPS by the peptides lipopolysaccharide l ipid transfer proteins Mueller-Hinton broth minimum inhibitory concentration methic il lin resistant Staphylococcus aureus myeloid antimicrobial peptides nuclear factor KB natural killer cells nuclear magnetic resonance nitric oxide I -N-phenyl-napthylamine National Committee of Laboratory Safety Standards NCPF NCTC OaBac OaDode OD PBSX PMAP PMN RP-HPLC SBD SDS SDS-PAGE SEM SMAP TEM TEMED TFA TFE TLRs TSB National Collection of Pathogenic Fungi National Collection of Type Cultures Ovine aries bactenicin Ovine aries dodecapeptide optical density phosphate buffered saline plus magnesium chloride porcine myeloid ant imicrobial peptide polymorphonuclear leukocytes reverse-phase high performance liquid chromatography sheep �-defensin sodium dodecyl sulphate sodium dodecyl sulfate - polyacrylamide gel electrophoresis scanning electron microscopy sheep myeloid ant imicrobial peptide transmission electron microscopy N,N,N',N'-tetramethylethylenediamine trifluoroacetic acid 2,2,2-trifluroethanol Toll-like receptors tryptic-soy broth List of Abbreviations xix List of Publications LIST OF PUBLICATIONS Most of the research presented in this thesis has been peer-reviewed and published in journals and/or presented at conferences. These publ icat ions are listed below. The full text of the journal art ic les are given in Appendix AS. Journal Articles Anderson RC, and Yu PL. (2003) I so lat ion and characterization of prol ine/arginine-rich cathe licidin peptides from ovine neutrophils. Biochemical and Biophysical Research Communications 3 1 2(4), 1 1 39- 1 1 46. Anderson RC, Wilkinson B, and Yu PL. (2004) Ovine ant imicrobial peptides: new products from an age-old industry. Austra lian Journal of Agricultu ral Research, SS( I ), 69-7S . Anderson RC, Hancock REW, and Yu PL. (2004) Antimicrobial activity and bacterial membrane interaction of ovine-derived cathel icidins. Antimicrobial Agents and Chemotherapy, 48(2), 673-676. Anderson RC, Haverkamp R and Yu PL. (2004) Invest igat ion of morpho logical changes to S. aureus induced by ovine-derived antimicrobial peptides using TEM and AFM. FEMS Microbiology Letters, 240( 1 ), 1 OS - 1 1 O . Anderson RC and Yu PL.(200S) Factors affect ing the antimicrobial activity of ovine-derived cathe l icidins against E. coli 0 I S7 :H7. International Journal of Antimicrobial Agents, 2S(3 ), 20S-2 1 O. Anderson RC and Yu PL. Purification and characterisation o f two protein fragments with ant imicrobial activity from ovine blood, inc luding part of the cathel icidin precursor. (wait ing for Meat and Wool NZ approval to submit) Anderson RC and Yu PL. P i lot-scale extraction and antimicrobial activity of crude extract from ovine neutrophils. (wait ing for Meat and Wool NZ approval to submit) Conference Proceedings Anderson RC, Hancock REW and Yu PL (2003) Mechanism of action of ovine-derived ant imicrobial peptides. New Zealand Institute of Chemistry Conference, 30th Nov- 4th Dec 2003, Nelson, New Zealand. Anderson RC, Wilkinson B and Yu PL (2002) Separation and activity of antimicrobial peptides from ovine blood. American Society of M icrobiology General Meeting, 1 9- 24th May, Salt Lake C ity, Utah, USA. Anderson RC, Wilk inson B and Yu PL (200 1 ) Purificat ion of antimicrobial peptides from sheep' s blood. Proceedings of the Molecules for Life Conference, 6-9th November 200 1 , Napier, New Zealand. Yu PL and Anderson RC (2004) Ovine Antimicrobial peptides: How much do we know? New Zealand M icrobiological Society Annual Conference. 1 7th - 1 9th November 2004, Palmerston North, New Zealand xx Chapter 1 - Project Introduction CHAPTER 1 PROJECT INTRODUCTION 1.1 REASON FOR THE RESEARCH This research project was carried out to investigate the properties of antimicrobial peptides that naturally occur in ovine blood cells. These antimicrobial peptides are of interest because they have the potential to be used in new high-value products for the New Zealand sheep industry. In New Zealand the meat industry is the second largest export earner, after the dairy industry (Statist ics New Zealand, 200 1 ) . Of this, 45% of the earnings come from the sale of sheep meat (Benincasa et ai, 2003) , so the sheep industry is very important for the national economy. Due to the large sheep industry in New Zealand, ovine blood is readily available. Each year approximately 25 mil lion lambs and 5 million sheep are slaughtered nationwide (Benincasa et ai, 2003). During the slaughtering process it is possible to recover 2 .3 litres of blood per adult sheep and 1 . 1 litres of blood per lamb (Fernando, 1 976). This means that approximately 40 million litres of ovine blood could be co llected in New Zealand annually. Currently, ovine blood is not util ised well by the New Zealand sheep industry. Ovine blood is either processed into low-value products, such as dried blood meal, which sells for approximately US$0.35Ikg, or it is discarded as effluent (van Asch, 200 1 ) . It would be more beneficial to the sheep industry, and in turn the New Zealand economy as a whole, if this blood was further processed into high-value products. The ovine blood could be col lected during slaughter, separated and then further processed into a number of high-value products. Such products inc lude blood serum, which is used in laboratories for growth of cell cultures, blood plasma proteins, including serum albumins, fibronectin, transferrin, antibodies, and trypsin, and red blood cell fractions including haemin and amino acids (Ockerman and Hansen, 1 988) . As well as these products from blood plasma and red blood cells, the ant imicrobial peptide� could be extracted from the white blood cells and used in high-value products. These antimicrobial peptides have the potential to be used in a biopreserving solution for chilled lamb products. It is possible that this biopreserving so lution could inhibit the growth of food spoilage organisms and increase shelf-life, and it could inhibit the growth of food-poisoning Chapter 1 - Project Introduction organisms and ensure the safety, of chil led lamb products. The antimicrobial pept ides also have the potential to be used in topical antiseptic creams. It is possible that such creams could be used to protect cuts and grazes from becoming infected, or as treatments for fungal infect ions such as ath lete's foot. 1 .2 PROJECT OBJECTIVES The aim of the research presented in this thesis was to gain an understand ing of the properties of antimicrobial peptides found in ovine blood neutrophils, so that their potential to be uti lised in high-value products could be assessed. Although animal antimicrobial peptides have been studied by numerous research groups, few studies have focussed on ovine antimicrobial peptides, so relat ively little is known about these peptides. In order to learn more about ovine ant imicrobial peptides, five project objectives were created. A literature review was carried out to establish what was a lready known about ant imicrobial peptides in general, as well as ovine antimicrobial peptides specifical ly. This is presented in Chapter 2. From this, areas where knowledge was lack ing were ident ified and the aims of this project were created accordingly. The first object ive of the research presented in this thesis was to purify and identify ant imicrobial peptides from ovine blood. Prior to this research project, a crude extract had been produced from ovine white blood cells that had antimicrobial act ivity (Anderson and Vu, unpublished results). It was assumed that this activity was due to antimicrobial peptides because seven such peptides had been predicted from ovine cDNA; however, only one had been iso lated from ovine blood. The second object ive of the research presented in this thesis was to determine the mechanisms of action of ovine ant imicrobial peptides. The mechanisms of action of antimicrobial peptides are not fully understood; however, the mechanisms used seem to depend on the structural c lass of the peptide. Because the predicted ovine ant imicrobial peptides were from more than one structural c lass, this research aimed to compare the interaction of d ifferent ovine antimicrobial peptides with bacterial membranes. The third object ive of the research presented m this thesis was to invest igate the morphological changes to microbial cel ls induced by ovme antimicrobial peptides. Antimicrobial peptides cause different morphological changes to microbial cells depending on 2 Chapter 1 - Project Introduction their mechanism of action. Therefore, this research aimed to compare the morphological changes induced by different ovine antimicrobial peptides to further understand their mechanisms of act ion. The fourth objective of the research presented in this thesis was to determine the effect of different environmental factors on the activity of ovine antimicrobial peptides. The activities of some antimicrobial peptides are inhibited by high salt concentrations, d ivalent cations and acidic pH values; whereas the activit ies of other peptides are not. The effects of such factors are important as they may limit the types of applications that the ovine antimicrobial peptides may be used in. The fifth object ive of the research presented in this thesis was to determine whether it i s possible to produce an active ant imicrobial extract on a scale larger than that used in the laboratory, using industrial-style equipment. Large-scale extractions of ant imicrobial peptides have not been previously reported. I f the peptides are to be used in commercial products, they need to be able to be isolated cost-effectively. From the results of these five objectives it was hoped that enough information would be gained to decide whether product development to utilise the ovine antimicrobial peptides should be carried out. For this to occur, the antimicrobial peptides need to be robust and active in condit ions that are likely in a product. As well as this, the peptides need to be present in the blood at concentrations high enough that the extraction is economical and the extraction process needs to be easily scaled-up to industrial-style operations. 3 Chapter 2 - Literature Review CHAPTER 2 ANTIM ICROBIAL PEPTIDES LITERATURE REVIEW 2 .1 INTRODUCTION The first objective of this literature review was to examine the �haracteristics of antimicrobial peptides. Antimicrobial peptides are produced by all forms of life to protect themselves against invading microorgansims. There are a num.Qer of structural c lasses of antimicrobial peptide�that were investigated, including those produced by animals, plants and microbes. The second object ive of this literature review was to invest igate the natural role of antimicrobial peptides within the immune system of animals. Firstly, an overview of the animal immune system, inc luding both the innate and adaptive immune systems was carried out. Then, the function that antimicrobial peptides perform within these immune systems was examined. The third objective of this l iterature revIew was to determine possible applications of antimicrobial peptides in order to show the reason for carrying out this research. Antimicrobial peptides are bioactive compounds that could be used in a number of areas including human health, animal health and food preservation. Firstly, the applications developed to date for antimicrobial peptides are reviewed. Then, possible applications for ovine blood antimicrobial peptides are proposed. The fourth objective of this literature review was to investigate in detail the five areas relating to the objectives of this research project. These include p�ification of antimicrobial peptides from blood, mechanism of action of antimicrobial peptides, morphological changes to bacterial cells induced by antimicrobial peptides, effects of environmental conditions on activity of antimicrobial peptides and large-scale extraction of antimicrobial peptides from blood. The current knowledge in each of these areas is summarised and the techniques used to gather this knowledge are analysed. The fmal objective of this review was to identify areas within these five target research fields where knowledge is lacking. From this, the objectives and hypotheses of the project were developed, with the aim of buildirIg on, and extending, the current knowledge in this field, specifically in relation to ovine blood antimicrobial peptides. 4 Chapter 2 - literature Review 2.2 ANTIMICROBIAL PEPTIDES Antimicrobial peptides are small strings of amino acids, fewer than 1 00 residues long, that (inhibit the growth of, and in some cases kil l, microorganiSm� These peptides are produced by all forms of life, including animals, plants and microorganisms, to protect themselves against microbial invasions. Antimicrobial peptides are part of the innate immune system o f the host and they p lay a role in the fIrst line o f defence against microbial invasions. The antimicrobial peptides produ,.cesl by animals and R.ian!S have broad-spectrum activity . - , - - This is necessary because the peptides are requIred to protect the host from a wide range o f microorganisms. The predominant function of antimicrobial peptides i s to inhibit the growth of bacteria; however other organisms such as fungi (Selsted et ai, 1 985 ; AJcouloumre et ai, 1 993 ; Giacometti et ai, 1 999; Newman et ai, 2000), protozoa (Aley et ai, 1 994), amoeba ( Schuster and Jacob, 1 992) , viruses (Lehrer et ai, 1 985 ; Daher et ai, 1 986; Robinson et ai, 1 998; Yasin et ai, 2000; Bastian and Schafer, 200 1 ) and tumour cells (Sheu et ai, 1 985 ; Lichtenstein et ai, 1 986; Winder et ai, 1 998; Johnstone et ai, 2000; Shin et ai, 2000b) are also susceptible to some of these peptides. In most cases, the antimicrobial peptides produced by microorganism� have a n�rrower -­.,.- spectrum of acti,::ity. This is necessary so that the host microbes themselves are not - - susceptible to the peptides. This means that the antimicrobial peptides produced by microbes target particular features of other microbes. The target microbes are usually those that compete with the host for an ecological niche. Antimicrobial peptides can have a variety of different structures and are usually classified - - � . - .,- - according to these structures. In the fo llowing subsections the structural c lasses of - . � antimicrobial peptides prod'uced by animals, p lants and microbes are discussed in succession and the properties of each group of peptides are examined. 2.2 .1 Animal antimicrobial peptides Over the past few decades numerous animal antimicrobial peptides have been isolated and investigated. More than 800 animal antimicrobial peptide sequences are now stored in the antimicrobial peptides online database at http://www.bbcm.univ.trieste. it/�tossi/pag2.htm. There are four main tructural c lasses of ant imicrobial peptides produced by animals. . . Examples of animal antimicrobial peptides from each structural class are given in F igure 2 . 1 . 5 Chapter 2 - Literature Review Although the structures of antimicrobial peptides vary considerably, they have a number of common features. They '!U have separate charged and hydrophobic regions, and they are a l l - cationic, with a posit ive charge of at least two. These properties he lp them to interact with <=>--- � bacterial membranes. �-sheet stabl ised by disulphide bonds ( human defensin-1 ) + + N extended structures rich in one or more amino acids ( indolicid in) r amphipathic a-helix (cecropin-mel itin hybrid ) loop structure (bactenecin dodecapeptide) N Figure 2.1 - Examples of the four stru ctural class� of cationic antimicrobial peptides. The positive residues are marked with a '+' , the N-terminus is marked with an 'N' and the disulphide bonds are shown in red. This fig u re is adapted from that given by Hancock, 1 997. The fust structural c lass of animal ant imicrobial peptides are predominately p-sheets. These peptides have an e�en number of cysteine residues and are s!abilised by disulphide bonds. Members of this structural c lass include the defensins produced by mammals and birds, the insect defensins, big defensin produced by horseshoe crabs, and penaeidins produced by shr imp, which have three disu lphide bonds; and the protegrins produced by pigs, the tachyplesins produced by horseshoe crabs, and androctonin produced by scorpions, which have two disulphide bonds (Dimarcq et al, 1 998). The second structural c lass of animal antimicrobial peptides are amphipathic a-hel ices. - - Members of this structural class include the myeloid antimicrobial peptides ( MAP) produced by mammals, the cecropins produced by insects (Otvos, 2000), and the bombinins, bombinin­ like peptides, magainins, brevinins, dermaseptins and caerins produced by amphibians (Simmaco et al, 1 998). 6 Chapter 2 - literature Review The third structural c lass of animal antimicrobial peptides are extended l inear structures. These peptides are rich in one or more amino acid. Members of this structural c lass include the proline/arginine-rich Bac peptides and the tryptophan-rich indolicidin produced by mammals and the proline-rich, short peptides produced by insects (Otvos, 2002). The final structural c lass of animal antimicrobial peptides are hairpin structures. These peptides contain one disulphide bond to hold them in their loop shape. Members of this structural family include the bactenecin dodecapeptides produced by mammals and thanatin produced by insects (Otvos, 2000) Due to the large number of famil ies of animal antimicrobial peptides, only the families of peptides produced by mammals are reviewed in detail. The antimicrobial peptides produced by mammals fal l into two families, the defensins and the cathelicidins. The characteristics of these families are discussed below. The characteristics of the individual peptides that have been purified from animal blood are discussed later in Section 2 .5 .2 . The first family of mammalian antimicrobial peptides are the defensins. Defensins are also produced by other higher animals such as birds. These peptides are rich in arginine residues and contain six cysteine residues that form three disulphide bonds. These d isulphide bonds stabi lise the peptides which fold into a three-stranded, antiparallel �-sheet structure. The defensins are divided into groups according to the position and connecting pattern of the ------ - - ' cysteine residues. Examples of defensins from each of the groups are given in Figure 2 .2 A B c VVCACRRAKCKPRERRAGFCRIRGRIHPLCCRR 1 I I I I 1 FASCHTNGGICLPNRCPGHMIQIGICFRPRVKCCRSW 1 I I I 1 1 GF?R?L1RR RTCICRCVG Figure 2.2 - Examples of the three g roups of defensins. (A) is rabbit a-defensin NP·1 , (B) is bovine p·defensin BNBD·1 , and (C) is monkey 9-defensin RTD·1 . The l ines connecting the cysteine residues show the disulphide-bond l inking patterns. The first group of mammalian defensins to be isolated were the a-defensins (Selsted et ai, 1 985) . These peptides are 29-35 residues long, and contain six cysteine residues that are l inked in a 1 -6, 2-4, and 3-5 pattern (Selsted and Harwig, 1 989) . The a-defensins are 7 Chapter 2 - Literature Review commonly found in the neutrophils of mammals, where they are synthesised then stQled in the specific granules. a-defensins have been ident ' ified in the neutroph�s o;umans (W�e et ai, 1 989), monkeys (Tang et ai, 1 999a), rabbits (Fuse et ai, 1 993), rats (Eisenhauer et ai, 1 989) , guinea pigs (Selsted and Harwig, 1 987; Yamashita and Saito, 1 989) and hamster (Mak et ai, 1 996). A sub-group of a-defensins has also been iso lated from the gastrointestinal tract of mice (Ouellette and Lualdi, 1 990; Eisenhauer et ai, 1 992; Ouellette et ai, 1 992; Ouellette et ai, 1 999). These intestinal peptides are also known as cryptdins. Another sub-group of a­ defensins has been ident ified in the kidney of rabbits (Bateman et ai, 1 996; Wu et ai, 1 998) . The second group of mammalian defensins, the p_-defensins, are slight ly bigger than the a- � / defensins, with 38-42 residues, and have the cross-liiiking pattern 1 -5 , 2-4, and 3 -6 ( Selsted et ai, 1 993) . p-defensins have N-terminal a-helices that allow them to form dimers. These din1ers then group together to form octamers (Hoover et ai, 2000). The p-defensins are predominately found in the epithelia of the gastrointestinal, resQ.�atory and urogen�tal tracts of mammals. p-defensins have been identified in these locations in humans, (Diamond et ai, 1 99 1 ; Jones and Bevins, 1 993 ; Diamond et ai, 1 993 ; Schnapp et ai, 1 998; Krisanaprakornkit et ai, 1 998; Mathews et ai, 1 999; Bonass et ai, 1 999; J ia et ai, 200 1 ), chimpanzees (Du its et ai, 2000), monkeys (Bals et ai, 200 1 ), mice (Huttner et ai, 1 997; Jia et ai, 2000), rats (J ia et ai, 1 999; Li et ai, 200 1 ), sheep (Huttner et ai, 1 998), goats (Zhao et ai, 1 999), horses (Davis et ai, 2004) and chickens (Lynn et ai, 2004). However, numerous p-defensins have also been found in the neutrophils of cattle (Selsted et ai, 1 993), chickens (Evans et ai, 1 994; Brockus et ai, 1 998), turkey (Evans et ai, 1 994; Brockus et ai, 1 998) and ostrich (Yu et ai, 200 1 ) . There is a third group of defensins, the 8:.defensins, which have been iso lated from Rhesus - -:.. macaque monkeys. The first 8-defensin, R T6:1 , is biosynthesised as two abbreviated a- defensins, each 9 amino acids long and these peptides are stabi lised by three disulphide bonds to form a cyc lic peptide (Tang et ai, 1 999b; Trabi et ai, 200 1 ) . The two 9 an1ino acid peptides that make up RTD- l have been termed R TD- l a and RTD- l b. Two other 8-defensins, RTD-2 and RTD-3, have also been iso lated. RTD-2 and RTD-3 are homodirners of RTD- l b and RTD- l a respectively (Tran et ai, 2002) . However, RTD- l is more abundant in monkey neutrophils. The ratio ofRTD- l , 2, and 3 is 29: 1 :2 . D�ensigs are synthesi�ed as pre,propeptides. The en;piece is a 1 9 amino acid endoplasrpc - - ,I ... .... signal sequence, which is c leaved during translation. nine @-S-@ Jl..mahyllamhionine F ig u re 2.9 - Structures of four Class I bacteriocins. Nisin A, epiderm i n and lacticin 481 are class la bacteriocins. Mersacidin is a class Ib bacteriocin. This image was taken from McAu l iffe et aI, 2001 . The production of lantibiotics is a complicated process (McAuliffe et aI, 200 1 ) . Firstly, the prelant ibiotics are produced on the ribosome. These prepept ides then undergo extensive post­ translat ional modification, inc luding dehydration and cross-linking react ions. Finally, the leader peptide is cleaved to produce the active molecule, which is then secreted. For this reason, lantibiotics require a large number of genes. For example, the most well studied lantibiot ic, nisin, has a gene c luster that includes genes for the prepeptide, enzymes for modifying the amino acids, cleavage of the leader peptide, secretion, immunity and regulation of expression ( Ri ley and Wertz, 2002) . The second c lass of bacterioc ins contain no modified amino ac ids, are heat-stable and are smaller than 1 0kDa (Oscartz and Pisabarro, 200 1 ) . L ike Class I bacterioc ins, the peptides are 1 6 Chapter 2 - Literature Review produced by Gram-positive, lactic acid bacteria. The members of this class are divided into four groups. The first group of c lass II bacteriocins, called c lass Ha bacteriocins, contain a concensus sequence of YGNGV at their N-terminus and are all active against L isteria (McAuliffe et aI, 200 1 ) . These peptides also have sequence similarities in their C-termini (Ennahar et aI, 2000). C lass Ha bacteriocins are between 37 and 58 residues long and they act through formation of pores in the cytoplasmic membrane (Nes and Holo, 2000). The second group of c lass II bacteriocins, called c lass lIb bacteriocins, are also pore forming complexes; however; they require two peptides for their activity (Oscartz and Pisabarro, 200 1 ) . I n some cases the two peptides are active individually but are synergistic when acting together, in other cases both peptides are necessary for activity. The final group of c lass I I bacteriocins, called c lass lIe bacteriocins, includes all c lass 1 1 bacteriocins that do not fal l into the first two groups (Oscartz and Pisabarro, 200 1 ). This includes bacteriocins with one or two cysteine residues and bacteriocins without any cysteine residues. As wel l as the c lass I and I I bacteriocins there are two other classes of bacteriocins produced by Gram-positive bacteria. However, these c lasses have not been studied in as much detail. C lass I I I bacteriocins are heat-labile proteins with masses higher than 30kDa (Oscartz and Pisabarro, 200 1 ). C lass IV bacteriocins are glycoproteins and l ipoproteins that require non­ protein moieties for their activity (Oscartz and Pisabarro, 200 1 ) . L ike Gram-positive bacteria, Gram-negative bacteria produce bacterioc ins. The most wel l studied are the bacteriocins produced by the Enterobacteriaceae family (Oscartz and Pisabarro, 200 1 ) . These peptides are c lassified according to their sizes. Colicins are larger that l OkDa and microcins are smaller than 1 0kDa. Colicins have a narrow spectrum of activity because their activity is mediated by interaction with specific membrane receptors. The Archaea produce a distinct family of antimicrobial peptides. These are known as archaeocins. The only characterised family of these peptides is the halocins which are produced by halo bacteria. These peptides are extremely hardy. They can be desalted, boiled, SUbjected to organic solvents and stored at 4°C for extended periods without detrimental effects on their activity. 1 7 Chapter 2 - Literature Review 2.3 THE ANIMAL IMMUNE SYSTEM Antimicrobial pept ides are one component of the very complex immune system of animals. The immune system is a co llection of cells, t issues and mo lecules that have the physio logical functions of preventing new infections and eradicating established infections. The host defence system uses a number of mechanisms to protect itself from invading microbes. These mechanisms can be split into those that are part of innate immunity and those that are part of adaptive immunity as il lustrated in Figure 2. l O . / Microbe #. I I n nate i m m u n ity I I Adaptive i m m u n ity I ��========� �------�==========�--------� .� Epithelial LW barriers @(J Phagocytes D 1/ Complement B Iymphocytes � O . �T Iymphocytes : . r:H. .. . . � I: -7'0.-----"'- N K cel ls Anti bodies �-/I- Effector T cells �, ____ H_o�_rs ______ 'M/�! ______ �, __________ D_a�y�s __________ � ________ � o 6 1 2 1 3 5 Time after infection --� ..... Figure 2.1 0 - Schematic diagram showing the principle mechanisms of innate and adaptive immunity. This image was taken from Abbas and Lichtman, 2004. The ' ate immune system, which is also cal led the natural or native Immune system, IS always present in healthy individuals. It responds immediately to invading organisms and mediates the init ial protection against infections by blocking entry of microbes and rapidly eliminating microbes that do enter the host tissues. The components of the innate immune system are generic and do not change to match a varying microbial assault. In contrast, the adaptive immune system, which is also called the specific or acquired immune -r--- --- - - system, is not always present; instead it is stimulated by the presence of microbes in the host tissues. This system is much slower than the innate immune system to respond to infections; however, its mechanisms are more effect ive so it is required when the innate immune system is overwhelmed. The adaptive immune system is st imulated by the microbes that invade the 1 8 Chapter 2 - Literature Review host and it adapts to the presence of the particular invaders. Adaptive immunity is relatively new in evo lutionary terms and is present only in vertebrates. The fo llowing sections firstly outline the components and mechanisms of the two immune systems; the innate and the adaptive immune systems. Then the role of animal antimicrobial peptides within these defence systems is examined. 2.3.1 Innate immunity The innate immune system is that which the host is born with and which is always present and - - available at very short notice to protect the host from chal lenges from foreign invaders. There are a number of components that make up the innate immune system including . iD· ..... � Table 2.2 - Amino acid sequences of cathelicidins rich in one or more amino acids found in l ivestock blood. Peptide Source Bac4 cattle Bac5 cattle Bac7 cattle OaBac5 sheep OaBac6 sheep OaBac7.5 Sheep .-- OaBac l 1 sheep ChBac5 goat PR39 pIgS Prophenin pIgS PMAP-23 pIgS indolicidin cattle Sequence RRLHPQHQRFPRERPWPKPLSLPLPRPGPRPWPKPL RFRPPIRRPPIRPPFYPPFRPPIRPPIFPPIRPPFRPPLGPFP Reference Scocchi et ai, 1 998 Gennaro et ai, 1 989 RRIRPRPPRLPRPRPRPLPFPRPGPRPIPRPLPFPRPGPRPI PRPLPFPRPGPRP I PRP Gennaro et ai, 1 989 RFRPPIRRPPIRPPFRPPFRPPVRPPIRPPFRPPFRPPIGPFP RRLRPRHQHFPSERPWPKPLPLPLPRPGPRPWPKPLPLPLPRPGLRPWKPL RRLRPRRPRLPRPRPRPRPRPRSLPLPRPQPRRIPRPILLPWRPPRPI PRPQPQP I PRWL RRLRPRRPRLPRPRPRPRPRPRSLPLPRPKPRPIPRPLPLPRPRPKPI PRPLPLP RPRPRRI PRPLPLPRPRPRPI PRPLPLPQPQPSPIPRPL RFRPPIRRPPIRPPFNPPFRPPVRPPFRPPFRPPFRPPIGPFP RRRPRPPYLPRPRPPPFFPPRLPPRIPPGFPPRFPPRFP AFPPPNFPGPRFPPPNVPGPRFPPPNFPGPRFPPPNFPGPRFPPPNFPGPPFPPP I FPGPWFPPPPPFRPPPFGPPRFPGRR RI IDLLWRVRRPQKPKFVTVWVR I LPWKWPWWPWRR Bagella et ai, 1 995 Huttner et ai, 1 998 Huttner et ai, 1 998 Huttner et ai, 1 998 Shamova et ai, 1 999 Storici and Zanetti, 1 993 Harwig et ai, 1 995 Zanetti et ai, 1 994 Selsted et ai, 1 992 Chapter 2 - Literature Review The second structural class of cathelic idins are a-helical. Seven a-helical cathelicidins have been discovered in livestock. The amino acid sequences of these peptides are given in Table 2 . 3 . Three of these were found in cattle, BMAP27, BMAP28 and BMAP34 (bovine myelo id ant imicrobial peptides with 27, 28 and 34 residues) (Scocchi et ai, 1 997; De Lucca and Walsh, 1 999); two in pigs, PMAP36 and PMAP37 (porcine myeloid antimicrobial peptides with 36 and 37 residues) (Storici et ai, 1 994); and two in sheep, SMAP29/0aMAP29 and SMAP34/0aMAP34 (sheep/ovine aries myelo id ant imicrobial peptides with 29 and 34 residues) (Mahoney et ai, 1 995 ; Bagella et ai, 1 995; Huttner et ai, 1 998) . These a-hel ical cathel icidin peptides are active against a broad spectrum of Gram-positive and Gram-negative bacteria, and some are also fungic idal (BMAP27, BMAP28 and SMAP29). The a-hel ical cathelicidins are haemolytic, but at concentrations much higher than those that are needed for antimicrobial activity. The final structural class of cathelicidins contain disulphide bonds. Seven cathelicidins containing disulphide bonds have been found in the blood of livestock. The amino acid sequences of these peptides are given in Table 2 .4 . One of these is from cattle, bactenecin (Romeo et ai, 1 988) ; one is from sheep, OaDode (Ovine aries dodecapeptide) (Huttner et ai, 1 998); and five are from pigs, PG- I , PG-2, PG-3, PG-4 and PG-5 (protegrins 1 -5) (Kokryakov et ai, 1 993) . Bactenecin and OaDode both contain one disulphide bond. Bactenecin has been shown to be active against Gram-negative bacteria (Romeo et ai, 1 988), but it is more active against Gram-posit ive bacteria (Wu and Hancock, 1 999a). The protegrins contain four cysteine residues that form two disulphide bonds. These peptides are active against Gram-negative and Gram-posit ive bacteria, fungi and enveloped viruses (Kokryakov et ai, 1 993 ; Steinberg et ai, 1 997). This activity is retained at physiological salt concentrations (Bellm et ai, 2000). 43 """ """ Table 2.3 - Peptide BMAP27 BMAP28 BMAP34 PMAP36 PMAP37 SMAP29 SMAP34 . ." Table 2.4 - Peptide bactenecin OaDode r- PG- l PG-2 PG-3 PG-4 PG-5 (") Ami n o acid seq uences of a-hel ical cathelicidins fou nd in l ivestock blood. � III "0 ,... (I) .., Source Sequences References N I r- cattle GRFKRFRKKFKKLFKKLSPVI PLLHLG Skerlavaj et ai, 1 996 it .., III ,... Skerlavaj et ai, 1 996 c cattle GGLRSLGRKILRAWKKYGPI IVPI IRIG .., (I) ;;0 (I) cattle GLFRRLRDSIRRGQQKILEKARRIGERIKDIFRG Scocchi et ai, 1 997 < iii" � pIgS GRFRRLRKKTRKRLKKIGKVLKWIPPIVGSI PLGCG Storici et ai, 1 994 pigs GLLSRLRDFLSDRGRRLGEKI ERIGQKIKDLSEFFQS Storici et ai, 1 994 sheep RGLRRLGRKIAHGVKKYGPTVLRI IRIAG BageUa et ai, 1 995 sheep GLFGRLRDSLQRGGQKI LEKAREIWCKIKDIFRG Mahoney et ai, 1 995 Amino acid sequences of cathelicidins containing disulph ide bonds fou nd in l ivestock blood. Source Sequences References cattle RLCRIVVIRVCR Romeo et ai, 1 988 sheep RICRI IFLRVCR Huttner et ai, 1 998 pIgS RGGRLCYCRRRFCVCVGR Kokryakov et ai, 1 993 pigs RGGRLCYCRRRFCICV Kokryakov et ai, 1 993 pigs RGGGLCYCRRRFCVCVGR Kokryakov et ai, 1 993 pIgS RGGRLCYCRGWICFCVGR Kokryakov et ai, 1 993 pigs RGGRLCYCRPRFCVCVGR Kokryakov et ai, 1 993 Chapter 2 - Literature Review 2.5.3 Ovine antimicrobial peptides As discussed, seven antimicrobial peptides have been predicted from ovine cONA to be .. present in ovine blood (Mahoney et ai, 1 995; Bagella et ai, 1 995 ; Huttner et ai, 1 998). These include four proline/arginine-rich peptides, two a-helical peptides and one peptide containing a disulphide bond. However, of these only one peptide has been iso lated from ovine blood. Four proline/arginine-rich cathelicidins were predicted from ovine cONA. Two variants of - - one of these predicted peptides, OaBac5, have been iso lated from ovine blood (Shamova et ai, - 1 999) . The purified peptides were amidated and differed from the predicted OaBac5 by at ---- � ,. - - -- / least one amino acid in the case of OaBac5a and at least 3 amino aCIds for OaBac5� . ----- . OaBac5a had broad-spectrum activity. Its MIC values were between 0.5 and 2 )lg/rnL against a variety of Gram-negative and Gram-posit ive bacteria. The activity of OaBac5� was not determined in this study as not enough material was available. Two ovine a-helical cathelic idins, SMAP29 and SMAP34 were predicted from ovine cONA. To date neither of these peptides has been isolated from ovine blood, but both have been synthesised and studied. The SMAP29 cathelic idin gene was identified by two research groups simultaneously (Bagella et ai, 1 995); however, Mahoney and coworkers were the first to synthesise and test the activity of this peptide (Mahoney et ai, 1 995) . It is thought that the natural form of SMAP29 would be an amidated, 28 residue peptide, so it is usually synthesised in this form (Mahoney et ai, 1 995; Bagella et ai, 1 995; Skerlavaj et ai, 1 999). Numerous studies have shown that SMAP29 has very potent, broad spectrum activity (Skerlavaj et ai, 1 999; Travis et ai, 2000; Brogden et ai, 200 1 ; Guthrniller et ai, 200 1 ) . SMAP29 has MIC values in the ranges of 0 .5-8 )lg/mL for Gram-negative bacteria, 0 .5-3 )lg/mL for Gram-positive bacteria and 3 - 1 2 )lg/rnL for yeast. Interestingly, SMAP29 is haemolytic against human, but not sheep erythrocytes (Skedavaj et ai, 1 999). However, the concentrations required to lyse human erythrocytes are ten-fo ld higher than those needed for antibacterial activity. Other studies have looked at the relationship between the structure and the function of SMAP29. I t was shown that SMAP29 has a random structure in aqueous solutions but forms an a-helix in so lutions containing sodium dodecyl sulfate (SOS) micelles (Shin et ai, 200 1 b). This same study showed that the N-terminal a-helical region was responsible for the 45 Chapter 2 - Literature Review ant imicrobial activity, whereas the C-terminal hydrophobic region was responsible for the haemolytic activity. The proline residue at position 1 9 was crucial for potent ant imicrobial activity. Another study, focusing on the LPS binding abi lity of SMAP29, showed that it had two LPS binding sites, one in the N-terminal reg ion and one in the C-terminal region and that LPS binds cooperatively (Tack et ai, 200 1 ) . This study also determined that residues 8 to 1 2 are in a helical structure, residues 1 8 and 1 9 form a hinge and residues 20 to 2 8 make up an ordered hydrophobic segment. The other ovine a-he lical pept ide, SMAP34, has not been investigated in the same depth as - r SMAP29. L ike SMAP29, SMAP34 also has broad-spectrum activity, although not as potent. It is also not significant ly affected by high salt concentrations (Travis et ai, 2000; Brogden et al, 200 1 ) . SMAP34 contains a single cysteine residue in the C-terminal region, which, in the case of guinea pig CAP l 1 , promotes homodimerisation by the formation of an intermo lecular d isulphide bond (Hashimoto et ai, 1 993). The study of ovine cDNA uncovered two ident ical genes that encoded the short peptide, OaDode (Qvis aries dodecapeptide) (Huttner et ai, 1 998) . This peptide is homologous to the bovine dodecapeptide bactenecin, with only 4 residues differing (Bagella et ai, 1 995) . Bovine bactenec in is a small ( 1 2 residues), amphipathic loop structure held in place by a disulphide bond. Since OaDode differs by the subst itution of hydrophobic residues for different residues that are also hydrophobic, the structure of OaDode is thought to be similar to that of the bovine dodecapeptide. OaDode has not been isolated or synthesised so its activity has not been examined. However, bovine bactenecin is active against E. coli and S. aureus (Romeo et ai, 1 988). p-defensins have also been identified in sheep; however they are not stored in the blood neutrophils. The two ovine p-defensin genes were first characterised in 1 998 (Huttner et ai, 1 998) and named SBD ! and 2 (sheep beta defensin 1 and 2) . SBD l is expressed extensive ly in ovine epithelial t issues, inc luding the trachea, tongue, rumen, ret iculum, omasum and colon. In contrast, SBD2 is expressed in only the ileum and colon. Both peptides are regulated pre- and post-nataUy (Huttner et aI, 1 998) . 2.6 MECHANISM OF ACTION OF ANIMAL ANTIMICROBIAL PEPTIDES The second object ive of the research presented in this thesis was to investigate the mechanism of action of ovine ant imicrobial peptides. The mechanisms of action of antimicrobial peptides 46 Chapter 2 - Literature Review , are not fully understood. The mechanisms used seem to vary between peptides, and in some cases a single peptide can use different mechanisms depending on the organism it is confronted with. Numerous techniques have been developed to investigate the mechanisms of antimicrobial peptides, especially in relation to the interactions of the antimicrobial peptides with bacterial membranes. The most commonly used o f these techniques were explored. The current knowledge regarding the mechanism of action of animal antimicrobial peptides is also reviewed. There are several theories about the mechanisms used by antimicrobial peptides, so each of these is critically examined and compared. The known details of the mechanism of action of specific animal antimicrobial peptides are also described. Finally, the previous work carried out to determine the mechanism of action of antimicrobial peptides from ovine blood is reviewed. From this, areas where knowledge is lacking are identified, so that the specific objectives of the current research project could be formulated. 2.6.1 Techniques to determine mechanisms of action of antimicrobial peptides A number of assays have been developed to investigate the mechanism of action of antimicrobial peptides. These assays look at each proposed step in the mechanism of action individually. To test whether antimicrobial peptides are able to bind to the outer surface of Gram-negative bacteria, the dansyl polymyxin B d isplacement assay is commonly used (Moore et ai, 1 986; Sawyer et ai, 1 988) . Oansyl polymyxin B binds to the divalent-cation-binding sites of LPS, which results in enhanced fluorescence of the dansyl group. The affinity of antimicrobial peptides for LPS can be determined by measuring their abi lity to reduce the fluorescence by displacing the dansyl polymyxin B . To test whether antimicrobial peptides are able to increase the permeability of the outer membranes of Gram-negat ive bacteria a few different assays have been used. These assays monitor the passage of a normally exc luded molecule across the outer membrane. One assay uses lysozyme, which once taken up causes lysis of the cells (Hancock et ai, 1 98 1 ; Fidai et ai, 1 997). Another assay used I -N-phenyl-napthylamine (NPN), which fluoresces only when it is in a hydrophobic environment (Loh et ai, 1 984; Patrzykat et ai, 2002). To study the interactions of antimicrobial peptides with the cytoplasmic membrane, model membranes are often used. P lanar bilayers are used to measure the increases in conductance 47 Chapter 2 - Literature Review of the ions (Gazit et ai, 1 996), liposomes are used to monitor the movement of fluorescent dyes (Zhang et ai, 2000), and Langmir monolayers are used to measure the surface pressure (Zhang et ai, 2000). These techniques can be used to investigate the effects of different lipid composit ions on the activity of the peptides (Hristova et ai, 1 997). Other studies investigate the interactions of antimicrobial pept ides with the cytoplasmic membrane using intact bacteria. Ones method monitors the hydrolysis of the substrate 0- nitrophenyl galactoside by the cytoplasmic enzyme �-galactosidase. This compound is usually exc luded by the cytoplasmic membrane. Another method uses a florescent dye cal led 3,3-dipropylthiacarbocyanine (DiSC35), which does not fluoresce when bound to the cytoplasmic membrane. However, it is re leased from the membrane and fluoresces when the membrane potential is reduced due to disrupt ion (Wu and Hancock, 1 999b; Zhang et ai, 2000). To study the interactions between antimicrobial peptides and inner cellular contents a few different tests are used. The abi l ity of antimicrobial peptides to inhibit DNA, RNA and protein synthesis is commonly studied by monitoring the cells' incorporation of the rad ioactive precursors [methyPH]thymidine, [5-3H]uridine and L-[2,5-3H]hist idine, respectively (Patrzykat et ai, 2002). The abi lity of the antimicrobial peptides to bind to DNA is determined by investigating the migration of peptide-treated DNA on an agarose gel (Park et ai, 1 998). Peptides with a high affinity for DNA inhibit its migration through the gel at low ratios. 2.6.2 Mechanisms of action The interactions between animal ant imicrobial peptides and microbial cel ls are not fully understood, but a number of models have been proposed. In al l cases the interactions between the peptide and the membranes ofthe target cells play a key ro le. The initial binding of the peptides to the outer cellular surface relies on the electrostatic interactions between the posit ively charged peptides and the negat ively charged molecules on the surface of the cells (Ladokhin and White, 200 1 ) . The se lect ivity of the peptides for prokaryotic cells depends on the lipid composition associated with the membranes of these cells compared to those of eukaryotic ce l ls . Gram-negative bacterial membranes contain LPS and Gram-positive bacterial membranes contain ac id ic po lysaccharides, so in both cases the membranes are negatively charged. In contrast, animal cells are composed of predominately 48 Chapter 2 - Literature Review zwitterionic and sphingomyelin phospholipids, to which antimicrobial peptides have a lower affinity (Oren and Shai, 1 998). In the case of Gram-negative bacteria, once the peptides have bound to the outer surface of the cells, they need to pass through this membrane in order to interact with the cytoplasmic membrane. When antimicrobial peptides bind to the divalent-cation-binding sites on LPS, they displace the native cations. This causes a distortion in the outer membrane structure, which increases the permeabil ity of the membrane. The peptides are then able to pass through the outer membrane. This process is termed "self-promoted uptake" (Fidai et aI, 1 997) Once the peptides pass through the outer membrane, they interact with the cytoplasmic membrane. This can lead to loss of cytoplasmic membrane function, including breakdown in membrane potential, loss of metabolites and ions, and alteration of membrane permeabil ity. Three main mechanisms for the membrane permeation have been suggested (Bals and Wilson, 2003) . These are i l lustrated in Figure 2. 1 8 . a . b 1 . barrel stave model :+ cationic a ntimicrobial peptide ,'W¥! - I " anionic biomembranes , '8 membrane proteins �iMY.I:l�mNtlWj b 2 . aggregate channel model b3. carpet model Figure 2 . 18 - Schematic diagram of the proposed mechanisms of permeability change of cytoplasmic membranes caused by antimicrobial peptides. After electrostatic interactions between the negatively charged bacterial wall and the positively charged peptides (a), the peptides may destabil ise the membrane leading to cell death. Three models for this membrane destabilisation have been proposed ; the barrel slave model (b1 ) , the aggregate channel model (b2) and the carpet model (b3). This figure is adapted from that g iven by Bals and Wilson, 2000. 49 Chapter 2 - Literature Review The first suggested mechanism of cytoplasmic membrane permeation is the barrel and stave model. This model involves the formation of voltage-dependent, transmembrane channels. The non-polar domains of the molecules face the membrane lipids to create a hydrophilic pore that spans the membrane. The barrel and stave mechanism involves four main steps: (1 ) binding of the monomers to the membrane in an amphipathic structure, (2) molecular recognition between membrane-bound peptide monomers that leads to their assembly, (3) insert ion of at least two assembled monomers into the membrane to init iate the formation of a pore, and (4) progressive recruitment of addit ional monomers to increase the pore size (Oren and Shai, 1 998) . There is evidence for this mechanism being used by defensins (Kagan et ai, 1 990; Wimley et ai, 1 994). The second suggested mechanism of cytoplasmic membrane permeation is the aggregate channel model . This model invo Ives the arrangement of the peptides in unstructured clusters in the membrane, which al lows the dynamic formation of pores for short periods of t ime. This model is most likely used by peptides that are too small to span the membrane ( Fidai et ai, 1 997). This can lead to intracellular components leaking out and the peptides entering the intracellular space. Antimicrobial peptides that pass through the cytoplasmic membrane and interact with the inner cellular contents include the dodecapeptide bactenecin (Wu and Hancock, 1 999b), the pro line/arg inine-rich pept ides PR-39 (Boman et ai, 1 993; Cabiaux et ai, 1 994), the two bovine cathelicidins BacS and Bac7 (Skerlavaj et ai, 1 990) and the tryptophan­ rich indo lic idin (Subbalakshmi and S itaram, 1 998; Friedrich et ai, 200 1 ). The third suggested mechanism of cytoplasmic membrane permeation is the carpet model. This invo lves the covering of the microbial cells with the peptide. The integrity of the membrane col lapses via holes that form by the bending of the lipid bi layer back on itself. There are four proposed steps in this model : ( 1 ) preferential binding of positively charged peptide monomers to the negatively charged phospho lipids, (2 ) laying of amphipathic monomers on the surface of the membrane so that the posit ive charges of the basic amino acids interact with the negat ively charged phospho lipid headgroups or water molecules, (3) rotation of the molecule leading to reorientation of the hydrophobic residues towards the hydrophobic core of the membrane, and (4) d isintegrating the membrane by disrupting the bilayer curvature lead ing to micell isat ion (Oren and Shai, 1 998) . The mechanism is used by a-hel ical peptides such as the insect cecropins, the amphibian magainins and dermaseptins 50 Chapter 2 - Literature Review (Oren and Shai, 1 998) and the cathe lic idins LL-37 (Oren et ai, 1 999) and SMAP29 (Schuster and Jacob, 1 992). 2 .6.3 Mechanism of action of ovine antimicrobial peptides Of the seven predicted ovine cathe l icidins that may be present in ovine neutrophils, only SMAP29, and peptides based on SMAP29, have been synthesised and studied in regard to their mechanism of action. An investigation into the mechanisms of action of a derivative of SMAP29 (residues I to 1 8) against yeast showed that the peptide disrupted the structure of the cell membrane by interacting with the lipid bilayer (Schuster and Jacob, 1 992) . Another study showed that ovisprin, an 1 8-residue peptide based on SMAP29, orientates paral lel to the l ipid bilayer. This is consistent with the carpet mechanisms of membrane disruption, which is common for a-helical pept ides. However, in contrast, irnmunoelectron microscopy showed that SMAP29 permeated the outer and inner membranes of E. coli cells almost immediately, and then entered the bacterial cytoplasm (Kalfa et ai, 200 I ) . Therefore, the target of SMAP29 is not fully understood and may be different in d ifferent circumstances. The mechanisms of action of the other predicted ovine cathel ic idins have not been investigated. However, it is l ikely that they would behave similarly to the bovine proline/arginine-rich cathelicid ins and bovine bactenecin. This means that the other ovine cathelic idins probably pass through the inner cellular membrane and interact with the inner cellular contents. 2.7 MORPHOLOGICAL CHANGES TO MICROBIAL CELLS INDUCED BY AN IMAL ANTIMICROBIAL PEPTIDES The third object ive of the research presented in this thesis was to examine the morphological changes to bacterial cells induced by ovine antimicrobial peptides, in order to further understand the mechanism of action of these peptides. Previously, two nucroscopy techniques have been commonly used to investigate morphological changes induced by antimicrobial peptides. These are transmission electron microscopy (TEM) and scanning electron microscopy (SEM). The advantages and disadvantages of each of these techniques are invest igated. As well as this, the potential for a relat ively new technique, atomic force microscopy (AFM), to be useful for this app l icat ion is also examined. 51 Chapter 2 - Literature Review The knowledge gathered about bacterial cell morphological changes induced by antimicrobial peptides is also reviewed. L ike the mechanism of action studies these results were often contradictory which again impl ies that antimicrobial peptides may have different effects depending on the microbial cells being attacked and the environmental cond itions. Final ly, the previous work carried out to determine the effect of ovine ant imicrobial peptides of the morpho logy of microbial cells is reviewed. From this, areas where knowledge was lacking were ident ified, so that the specific objectives of the current research project could be formulated. 2.7.1 Techniques to investigate morpholog ical changes To invest igate the morpho logical changes to microbial cells induced by antimicrobial peptides, two microscopy techniques, TEM and SEM, are commonly used. These techniques are complementary because TEM looks at a cross-section of the sample, and SEM looks at the surface of the sample. Both TEM and SEM use an electron microscope to image the samples. An electron gun produces a stream of monochromatic electrons. This stream is focused to a smal l, thin line by the two condenser lenses, and then the condenser aperture, which removes the high angle electrons, restricts the beam. The beam strikes the sample and then parts of the beam are transmitted. The transmitted beam is focused by the objective lens and further restricted by the objective and selected area apertures. The image is enlarged by the intermediate and projector lens, before it strikes the phosphor image screen and generates light. The darker areas of the image represent those areas of the sample that fewer e lectrons were transmitted through and the lighter areas represent those areas of the sample that more electrons were transmitted through. Another technique, which has not previously been used to image ant imicrobial peptide-treated cells, but may be useful, is AFM. L ike SEM, AFM gives an image of the surface of the sample. The advantage of AFM is that it can be used to image cell changes in real-time (Ahimou et ai, 2003) . This is not possible with SEM because the cells are fixed prior to imaging. AFM is also simpler, more convenient and less costly than SEM, which requires an ultravacuum environment (Amro et ai, 2000). 52 Chapter 2 - Literature Review AFM uses a cantilever with a sharp tip to probe the surface of the sample (Prater et al, ) . The canti lever is between 1 00 and 200/lm long, and the tip is approximately 2 /lm long and less than 1 00A in diameter. When the tip and the sample come into c lose proximity, the force between the two causes the cantilever to bend or deflect. As the tip scans over the sample the movement in the cantilever is detected by using a laser beam that bounces off the sample onto a posit ion-sensitive photodetector. This information is used to generate a map of the surface topography of the sample. The use of AFM to image l iving cells is relatively new (Dufrene, 2002). It has been successfully used to image live bacteria (Boonaert et ai, 2000), fungi (Kasas and Ikai, 1 995 ; Gad and Ikai, 1 995; Boonaert e t ai, 2000; Ahimou et ai, 2003) and animal cells (Kasas et ai, 1 993) . AFM has also been used to image Bacillus subtilis treated with penicill in (Kasas et ai, 1 994) and E. coli treated with cefodizime (Braga and Ricci, 1 998) . 2.7.2 Morpholog ical changes Microscopy techniques are commonly used to confIrm theories and gain further knowledge about the mechanism of action of antimicrobial peptides. For defensins and a-helical peptides, the target is thought to be the cytoplasmic membrane of bacterial cells, so it is expected that these peptides would induce notable differences in the morphology of bacterial cells. Numerous TEM and SEM studies have confirmed that defensins cause damage to the cytoplasmic membranes of both Gram-negative (Sawyer et ai, 1 988; Eisenhauer et ai, 1 989) and Gram positive bacteria (Shimoda et ai, 1 995; Miyakawa et ai, 1 996). S imilarly, a-hel ical peptides also cause visible damage to the cytoplasmic membrane (Tiozzo et ai, 1 998; Oren et ai, 1 999). Similarly, for other antimicrobial peptides the target is thought to be the inner cellular contents, not the cytoplasmic membrane, so it is expected that these cells would not induce damage to the bacterial membranes. Studies have shown that indo licidin-treated E. coli cells (Subbalakshmi and Sitaram, 1 998) and PR-39-treated S. typhimurium (Shi et ai, 1 996a) became fllamentous, which may have been induced by inhibition of DNA synthesis. 2.7.3 Morphological changes induced by ovine antimicrobial peptides Like the mechanism of action studies, the only ovine ant imicrobial peptide that has been studied for its induced morphological changes is SMAP29. SEM showed SMAP29 rapidly 53 Chapter 2 - Literature Review caused alterations, including blebs and blisters, to the surface of both Gram-negative and Gram-positive bacteria (Skerlavaj et ai, 1 999; Saiman et ai, 200 1 ; Arzese et ai, 2003 ) . This is cons istent with the idea that SMAP29 acts on the cytoplasmic membrane of bacterial cells. The morpho logical changes to bacterial cells induced by other ovine cathel icidins have not been investigated. However, since the majority of these peptides are proline/arginine-rich, like porcine PR-39, it is expected that these peptides would also cause the bacteria cells to become filamentous. 2.8 EFFECT OF ENVIRONMENTAL CONDITIONS ON ACTIVITY OF ANIMAL ANTIMICROBIAL PEPTIDES The fourth objective of the research presented in this thesis was to invest igate the effects of environmental conditions on the activity of ovine antimicrobial peptides. I f the ant imicrobial peptides iso lated from ovine blood are to be used as therapeutics agents or biopreservatives, as discussed in Section 2 .4.2, then they will need to be act ive in a variety of environmental conditions. Therefore, techniques for testing factors that may influence the antimicrobial activity of the peptides are investigated. The current knowledge regarding the effects of varIOUS environmental conditions on antimicrobial peptides activity is also reviewed. This includes the effects of physical factors such as pH, monovalent and divalent cation concentration and temperature. The effects of using the antimicrobial peptides in combination compared to alone was also researched. Finally, the previous work carried out to determine the effect of environmental conditions on the antimicrobial activity of ovine antimicrobial peptides was reviewed. From this, areas where knowledge was lacking were identified, so that the spec ific objectives of the current research project could be formulated. 2.8 . 1 Techniques to determine effect of environmental conditions In order to investigate the effect of environmental conditions on the activity of ant imicrobial peptides, a few different techniques have been used. Environmental factors that have been tested for their effects include salt concentration, cation concentration, pH and temperature. The antimicrobial activity of the peptides has also been examined when they are used In combination with each other, or with other antimicrobial compounds (Selsted et ai, 1 985 ). 54 Chapter 2 - Literature Review To test the effect of environmental conditions on antimicrobial peptide activity the growth media are usually altered to the required test conditions. In most studies the MIC of the peptides in different media is determined using either the diffusion plate assay (Harwig et ai, 1 995; Cho et aI, 1 998), or the micro-broth dilution method (Scort et aI, 1 999) . In other studies, the log decrease in viable cells (Selsted et ai, 1 985), or the percentage by which the peptide is inhibited is measured (Lehrer et ai, 1 983) . To test the effect of antimicrobial peptides in combination with each other or with known antimicrobial substances the fraction inhibitory concentration (FIC) has been used (Fidai et ai, 1 997; Scort et ai, 1 999). This method is similar to the micro-broth dilution MIC method, except that it uses checkerboard titrations, where one compound is di luted down the columns and the other is diluted across the rows, in a microtitre plate. The FICs are calculated using the formula: FIC = [A]/MICA + [B]/MICB. MICA and MICB are the MICs of the compounds A and B alone. [A] and [B] are the MICs of compounds A and B when in combination. An FIC of less than 0.5 means the compounds are synergistic, an FIC of unity means the compounds are additive, and an FIC of greater than four means that the peptides are antagonistic . 2.8.2 Effects of environmenta l cond itions The most commonly tested factor for its effect on antimicrobial activity is salt concentration. At a physio logical NaCl concentration of 1 00mM, defensins (Lehrer et ai, 1 983), (Selsted et ai, 1 985), (Aley et ai, 1 994), and some cathelicidins, such as prophenin (Harwig et aI, 1 995), are considerably inhibited. In contrast, other antimicrobial peptides, such as the a-helical cathelicidins (Travis et ai, 2000 ; Zhao et ai, 200 1 ), and the proline/arginine-rich cathe licidins (Shamova et ai, 1 999), are not inhibited in these condit ions. As well as the effect ofNa+, the effect of other cations, particularly the divalent cations Mg2+ and Ca2+, have been investigated. All animal antimicrobial peptides tested are inhibited considerably by divalent cations. This is because these peptides bind to the cation-binding sites of LPS on the surface of Gram-negative bacteria so the peptides have to compete with the ions. In most cases, Ca2+ ions are more inhibitory than Mg2+ ions (Selsted et ai, 1 985 ; Aley et ai, 1 994; Turner et ai, 1 998) . This may be due to Ca2+ being a larger ion than Mg2+. However, in one study Mg2+ ions were more inhibitory than Ca2+ ions (Lawyer et ai, 1 996). It was also shown that the effect of cations is accumulative so it is the overall cation 55 Chapter 2 - Literature Review concentration that is important, not the concentration of the individual cat ions (Cho et ai, 1 998). Another factor that has been investigated for its effect on ant imicrobial activity is pH. The antimicrobial peptides tested all have decreased antimicrobial activity in acidic pH conditions compared to neutral and slightly basic pH conditions (Lehrer et al, 1 983) . The human cathelicidin LL-37 has an a-hel ical structure at neutral and basic pH values but not at ac idic pH values (Johansson et al, 1 998). This conformational change is probably brought about by the acidic side chains being protonated at low pH values and it is this conformat ional change that is probably responsible for the peptide 's change in activity. The effect of temperature on the activity of antimicrobial peptides has not been investigated in detail. It is known that SMAP29 retains its activity after heating to 65°C (Mahoney et ai, 1 995); however, other peptides have not been tested. The effect of antimicrobial peptides in combination with each other and with known ant imicrobial substances is another area that has been studied. Cathe licidin peptides are synergistic, but only in some combinations (Yan and Hancock, 200 1 ). The amphibian peptide magainin 1 1 is synergist ic with �- lactams (Scott et ai, 1 999). 2.8 .3 Effects of environmental cond itions on ovine antimicrobial peptides Only a few studies have been carried out to investigate the effects of condit ions on ovine antimicrobial peptides. I t has been shown that the predicted ovine a-hel ical peptides SMAP29 and SMAP34 (Travis et ai, 2000) and the purified ovine proline/arginine-rich peptide OaBac5a (Shamova et ai, 1 999) are not significantly impaired by high ( 1 00mM) NaCI concentrations. As previously mentioned SMAP29 retains its antimicrobial activity after being heated to 65°C for 30 minutes (Mahoney et ai, 1 995) . The effects of other factors on the activity of ovine antimicrobial peptides, such as cation concentration and pH, have not been investigated. The activity of ovine antimicrobial peptides in combination with each other or with other ant imicrobial substances has also not been examined. 56 Chapter 2 - Literature Review 2.9 PILOT -SCALE EXTRACTION OF ANIMAL ANTIMICROBIAL PEPTIDES The [mal objective of the research presented in this thesis was to determine whether it was possible to extract antimicrobial peptides from ovine blood on a scale larger than that used in the laboratory. No studies have previously been reported on the large-scale extraction of antimicrobial peptides from blood. This is probably because, as d iscussed in Section 2.4. 1 , most research groups focus on applications o f synthetic antimicrobial peptides, not those that can be purified from natural sources. However, animal blood is routinely processed commercially (Ockerman and Hansen, 1 988; van Asch, 200 1 ) . Various products are isolated from blood plasma including serum albumins, fibronectin, transferrins and trypsin; and the red blood cells including haemoglobin protein. Centrifugation and ultrafiltration are often used in commercial operat ions to remove some of these fractions. A similar approach may be feasible for the isolation and fract ionation of proteins from white blood cells. 2 . 1 0 CONCLUSIONS The first objective of this research project was to purify and characterise the antimicrobial substances found in ovine blood. This l iterature review showed that ovine neutrophils are predicted to contain seven cathelic idin peptides. Of these antimicrobial peptides, variants of only one have been previously purified. Therefore, the aim of this research was to purify and characterise the other antimicrobial peptides from ovine blood, as wel l as other components which have ant imicrobial act ivity. The second object ive of this research project was to determine the mechanism of action of ovine antimicrobial peptides. This literature review showed that antin1icrobial peptides can have a number of d ifferent mechanisms of action depending on their structures. Studies with ovine SMAP29 gave contradictory results so the mechanism of action of this peptide is unclear. The mechanisms of action of the other ovine cathelicidins have not been investigated. It was hypothesised that the pro line/arginine-rich cathelicidins wil l interact with the inner cellular contents of the bacterial cel ls, whereas the a-hel ical peptides will target the cytoplasmic membrane. Therefore, the aim of this research was to investigate and compare the mechanisms of action of ovine proline/arginine-rich peptides and a-helical peptides. 57 Chapter 2 - Literature Review The third objective of this research project was to investigate the morpho logical changes to bacterial cel ls induced by ovine antimicrobial peptides. This l iterature review showed that antimicrobial peptides induce different morpho logical changes to microbial cel ls depending on their mechanism of action. Studies with ovine SMAP29 showed that it damages the cel l membranes. The morpho logical changes to microbial cel ls induced by the other ovine cathelicidins have not been investigated. It was hypothesised that the prol ine/arginine-rich cathelicidins would not damage the membranes of the bacterial cel ls, whereas the a-helical peptides would target the cytoplasmic membrane. Therefore, the aim of this research was to investigate and compare the morpho logical changes to bacterial cel ls induced by ovine prol ine/arginine-rich and a-helical peptides. The fourth object ive of this research project was to determine the effect of different environmental factors on the activity of ovine antimicrobial peptides. This literature review showed that some antimicrobial peptides are inhibited by high salt-concentrations, divalent cations, and acidic pH values; whereas other peptides are not. Studies with ovine SMAP29 showed it was not impaired by high salt concentrations or by heating to 65°C. Other conditions have not been tested. The effects of environmental factors on the other ovine cathel icid ins have also not been investigated. Therefore, the aim of this research was to determine the effects of various factors on the activity of ovine prol ine/arginine-rich and a­ hel ical peptides. The [mal objective of this research project was to determine whether it is possible to produce an active antimicrobial crude extract on a larger scale than in the laboratory using industrial­ style equipment. This l iterature review showed that antimicrobial peptides have not been extracted from animal blood in large quantities before. B lood is routinely fractionated on a commercial scale to iso late various high-value proteins. Therefore, the aim of this research was to determine whether antimicrobial peptides can be extracted from ovine blood on a large scale and to develop a process that could be easily operated industrially. 58 Chapter 3 - Materials and Methods CHAPTER 3 MATERIALS AND METHODS 3 . 1 MATERIALS AND METHODS USED FOR PEPTIDE PURIFICATION 3. 1 . 1 Crude extraction Antimicrobial peptides were extracted from neutrophil granules using a process adapted from methods given in the literature (Eisenhauer et ai, 1 989; Borenstein et ai, 1 99 1 ; Selsted et ai, 1 993) . The pooled blood from a number of lambs was collected from Fei lding Lamb Packers (Feilding, New Zealand), mixed with 1 0% (w/v) aqueous sodium citrate in a ratio of 1 0 : 1 to stop coagulation, and stored on ice until used. Before processing, the blood was filtered to remove any solid contaminants such as wool and clotted blood. Two different methods were used to lyse the red blood cells. The original process was based on that used by Selsted's group at the University of California, I rvine (Selsted et ai, 1 993 ) . The blood cells were separated from the blood plasma by centrifugation. The red blood cells were then lysed by adding water ( 3 : 1 v/v) and mixing for ten seconds, before triple-strength PBSX buffer (4 1 1 mM NaCl, 8 . 1 mM KCl, l .SmM MgCb, 24. 3mM Na2HP04 and 4 .SmM KH2P04 (pH 7.4)) was added to restore the ionic balance. The white blood cells were collected by centrifugation. Usually it was necessary to repeat the red blood cell lysis steps at least once to ensure all the red blood cells were removed. Because this extraction process was t ime-consuming and used large amounts of buffer, it was replaced by one based on that of Robert Lehrer' s group at the University of California, Los Angeles (Borenstein et ai, 1 99 1 ) . In this process the blood cells were not separated from the plasma before the red blood cells were lysed. This reduced the amount of centrifugation required. To lyse the red blood cells, 0 .83% (w/v) aqueous ammonium chloride solution was used instead of water. The ammonium chloride solution was mixed with whole blood (2 : 1 v/v) . This was advantageous because it was not necessary to add buffer to stop white blood cell lysis because the white blood cel ls were relat ively stable in the ammonium chloride solution. The red blood cell lysis was also more effective than the previous method, so it was usually not necessary to repeat the lysis steps. 59 Chapter 3 - Materials and Methods After the red blood cell lysis step, the white blood cells were collected by centrifugat ion (Sorvall centrifuge, SS-3 rotor, 700g, I S minutes, 4°C) and resuspended in PBSX buffer ( 1 37mM NaC I, 2 .7mM KC I, O .SmM MgCh, 8 . l mM Na2HP04, l . SrnM KH2P04, pH 7.4) . The cells were stained using a Oiff-Quick staining k it (Baxter Cat. No. B4 1 32- 1 ) and then examined under a microscope to see if any intact red blood cells remained. I f present, the ammonium chloride lysis step was repeated. The white blood cells were disrupted using sonication (MSE sonicator, three times for 30 seconds at 8 microns peak to peak) to release the neutrophil granules. These granules were col lected by high-speed centrifugation (Sorvall centrifuge, SS-34 rotor, 27,000g, 40 minutes, 4°C). Again, the cells were stained with the Oiff-Quick staining kit and examined under the microscope. I f intact white blood cells were st i l l present, sonicat ion was repeated. The antimicrobial peptides were extracted from the granules in 1 0% acetic acid with stirring overnight at 4°C. The so lution containing the peptides was separated from the granu les (Sorvall centrifuge, SS-34 rotor, 27,000g, 20 minutes, 4°C) . This so lution was rotary - evaporated (bath temperature 3S-40°C) and freeze-dried. The extracted solids were then suspended in 0.0 1 % acetic ac id . This so lution is referred to as the crude extract. 3 . 1 .2 Gel electrophoresis SOS-PAGE (sodium dodecyl sulfate - polyacrylamide gel e lectrophoresi J was used to determine the n�mber and approximate size of the proteins and peptides present in the crude extract and the gel fi ltration fractions. To separate peptides and proteins 22% and I S% - - acrylamide gels were used, respective ly. The peptide gel contained 2 . SmL l .SmM Tris-HCl buffer, 1 00)lL 1 0% SOS stock, 7.3 2mL 30% acrylamide stock (30g acrylamide and 0.8g methlenebis acrylamide per 1 00mL dist i l led water), N,N,N',N'- tetramethylethylenediamine (TEMEO) and SO)lL 1 0% ammonium persulphate solution. The protein gel contained 2 .02mL distil led water, 2 . SmL l . SrnM Tris-HCl buffer, 1 00)lL SOS stock, S . 3rnL 30% acry lamide stock, S )lL TEMEO and SO)lL 1 0% ammonium persulphate solution. In both cases the separating gels were covered with a stacking gel to increase the resolut ion. The stack ing gel contained 6. 1 mL d istil led water, 2 .SmL O .SmM Tris-HCI buffer, 1 00)lL SOS stock, 1 . 3mL 30% acrylamide stock, l O)lL TEMEO and SO)lL 1 0% ammonium persulphate solut ion. 60 Chapter 3 - Materials and Methods The samples were diluted with an equal volume of double-strength sample buffer (25mL 0.5M Tris-HCI buffer, 20mL glycerol, 40mL 1 0% SDS, 1 0mL l3-mercaptoethanol and 5mL 0.05% bromophenol blue) and heated at 80°C for three minutes, before they were added to the gels. The gels were run at 20mA per gel for 45 minutes (Hoefer Scientific I nstruments, Mighty Small n, SE 250) with 200mL running buffer (3g Tris, 1 4 .5g glycine and I g SDS in l L of distilled water). The gel was fixed for one hour (25% isopropanol and 1 0% acetic acid in distilled water) , stained for two hours ( 1 .25g Coomassie Brilliant B lue R250 to 242mL distilled water, 242mL methanol and 46mL acetic acid) and destained overnight (7 .5% acetic acid and 5% methanol in distil led water) . The gels were covered in GelAir Cellophane Support (BioRad), which allowed them to dry without cracking. Images of the gels were captured using a digital scanner. 3 . 1 .3 Gel fi ltration Gel filtration was used to separate the components i n the crude neutrophil extract according to their sizes. The extract was passed through a column packed with B io-Gel P l O ( Bio-Rad Laboratories, California). The running buffer (5% acetic acid) was pumped through the column at a flow rate of 20mL/hour. The eluant was passed through a UV detector (LKB Bromma 2 1 38 UVICORD S), and the absorbance was plotted over time using a chart recorder (Sekonic SS-250F). A fraction collector (LKB U ltrorac 7000) was used to collect the eluant in 1 0-minute fractions. The chart recorder was run at a speed of 30mmlhour so that 5mm on the chromatograph chart corresponded to one fraction. The fractions that constituted each peak on the chromatograph were pooled, freeze-dried and dissolved in 0 .0 1 % acetic acid. 3 . 1 .4 Cationic-exchange chromatog raphy As an alternative to gel filtration, cationic-exchange chromatography was used to separate the cationic molecules from the other components present in the crude extract. The extract was passed through a Macro-Prep CM column (B ioRad Laboratories, California) . The cationic compounds bound to the anionic resin, and the non-cationic molecules were washed through with 25mM ammonium acetate. The cationic molecules were e luted with 1 0% acetic acid, freeze-dried and dissolved in 0 .0 1 % acetic acid. This system was set-up in the same way as the gel filtration system already described (Section 3 . 1 .3 ) . 61 Chapter 3 - Materials and Methods 3 .1 .5 Peptide purification us ing HPLC RP-HPLC (reverse-phase high performance l iquid chromatography) was used to separate the components in the active gel filtration and cat ion exchange fract ions. A Waters HPLC system with a Phenomenex Jup iter C- 1 8 co lumn was used to separate the components in the active gel filtration fract ion. A Dionex HPLC system with a Phenomenex Jupiter Proteo column was used to separate the components in the cationic fract ion. In both cases the peptides were separated using an acetonitrile gradient. Buffer A was 5% acetonitrile with 0. 1 % TF A (tritluoroacetic acid), and Buffer B was 95% acetonitrile with 0. 1 % TF A. The elution of the peptides was monitored using a UV detector (230mTI for the Waters system and 2 1 5mn for the Dionex system) . The peptide fractions were co llected manually, freeze-dried, and disso lved in 0.0 1 % acetic ac id. 3.1 .6 Radial diffusion plate assay The radial diffusion plate assay method (Steinberg and Lehrer, 1 997) was used to test if the crude extract, gel filtrat ion fract ions, ion-exchange fract ions, and HPLC peaks displayed ant imicrobial act ivity. The test cultures used were Escherichia coli 0 1 1 1 (Gram-negative bacterium), Staphylococcus aureus NCTC 4 1 63 (Gram-positive bacterium) and Candida albicans 3 1 53A (yeast). These cultures were sourced from the Massey Microbiology Department, the NCTC (National Collect ion of Type Cultures) and the NCPF (National Collection of Pathogenic Fungi), respective ly. The culture was mixed with the underlay agar (1 OmL trypt icase soy broth, 1 09 ultrapure agarose (S igma A-60 l 3) , l L distil led water, pH 7.4), which contained limited nutrients, and was allowed to set. Wells were made in the agar and 5 1-1L of the test sample was added to each well. For each plate both a positive and a negative control were inc luded. The posit ive contro l was a common ant ibiotic ; polymyxin B for Gram-negative bacteria, nisin for Gram­ posit ive bacteria and nystatin for yeast. The negative control was 0.0 1% acetic acid because this was used to dissolve the samples. The p lates were left for one hour to allow the test solut ions to diffuse into the agar. After this, the overlay agar (60g trypt icase soy broth powder, 1 09 u ltrapure agarose, l L dist i l led water), which was r ich in nutrients, was added. The plates were incubated overnight at 37°C. The fo llowing morning the diameters of the c learings around the wells were recorded. 62 Chapter 3 - Materials and Methods 3.1 .7 Rad ia l d iffusion plate assay MIC method The radial diffusion plate assay method was used to determine the mInImUm inhibitory concentrations (MICs) of the purified peptides (Lehrer et ai, 1 99 1 ). This method was based on the radial diffusion plate assay method given in Section 3 . 1 .6 but mUltiple dilutions of each sample were tested. The MICs were also determined in the presence of salt. This was achieved by adding 1 OOmM NaCI to the underlay agar The diameters of the c leared zones were recorded for each peptide concentration. The assay was carried out only once for each peptide due to the limited amount of purified material available. The log of the peptide concentration was plotted against the diameter of the c learing on the p late ( less the size of the well), and a straight line was fitted. L inear regression analysis was carried out using the statistical package GenStat. The "simple linear regression" function, which uses the least-squares method, was used. The Ln peptide concentration was the explanatory variant (x-axis variant) and the c learing size was the response variant (y-axis variant) . From the GenStat outputs the gradient and y-intercept were recorded. From this the MICs were determined by calculating the po int at which the line crossed the x-axis. This was the peptide concentration required to make a clearing size of zero. The limits o f the 95% confidence intervals for the x- intercept were calculated using Equation 3 . 1 (Draper and Smith, 1 98 1 ) . A _ A - t G (Xo - X) (1 - g) Limits of confidence intervals == g(Xo - X) ± -vs� +-- b] Sxx n where: t2s2 g = constant == -- b]Sxx Xo = the estimated x-intercept X = the mean of the x values t = the t-value b] == gradient of the l ine-of-best-fit S2 = estimate of the pooled variance Equation 3.1 Sxx == total corrected sum of the squares for x == (X] -X)2 + (Xl -X)2 + (X} - X)2 + (X4 -X)2 n = the number of x values The relationship between the bounds of the confidence intervals and the l ine-of-best-fit is demonstrated in F igure 3 . 1 . The upper limit of the x-intercept is a lower value than the lower 63 Chapter 3 - Materials and Methods l imit. In some cases the limits of the 95% confidence intervals of the x- intercept could not be calculated. This was because the bounds of the confidence intervals sometimes did not cross the x-axis. An example of this is i l lustrated in Figure 3 .2 E .s QI N .;; Cl C . .::: ca QI (j Ln peptide concentration xlower -- 95% confidence interval lower l imit -- 95% confidence interval upper l imit -- l ine-of-best-fit Figure 3.1 - Graph of Ln peptide concentration versus clearing size showing the relationship between the l ine-of-best-fit and the bounds of the 95% confidence intervals for the l ine. -- 95% confidence interval lower limit --95% confidence interval upper l imit -- line-of-best-fit Figure 3.2 - Graph of Ln peptide concentration versus clearing size showing the relationsh ip between the line-of-best-fit and the bounds of the 95% confidence intervals for the l ine when the bounds do not cross the x-axis. 3.1 .8 Mass spectroscopy Mass spectroscopy was used to determine the molecular weight of the purified peptides. Mass spectroscopy was carried out at the Proteome Analysis Faci l ity, Sydney, Australia, and 64 Chapter 3 - Materials and Methods the School of B iological Sciences, University of Auckland, New Zealand. At the Proteome Analysis Faci l ity a Micromass Q-TOF-MS equipped with a nanospray source was used. The data were manually acquired using borosilicate capillaries over the mlz range 400- 1 6000 Da. At the School of B iological Sciences a Voyager DE PRO MALDI-TOF mass spectrometer from Applied B iosystems was used. The ion-acceleration potential used was 20 kv, and for each sample the " 1 00 shots" mode was used to acquire data within the 500-8000Da range of positive po larity. 3 . 1 .9 N-terminal sequencing N-terminal sequencing was used to determine part of the amino acid sequence of the purified peptides. The N-terminal sequencing was carried out at the Proteome Analysis Facility, Sydney, Australia, and the School of B iological Sciences, University of Auckland, New Zealand. At both facilities automated Edman degradation using an Applied B iosystems Procise Protein Sequencing System was used. In the Edman degradation the N-terminal residue was labelled and cleaved from the peptide. This residue was identified by its retention time using RP-HPLC. The procedure was repeated to identify the next amino acid, and so on for subsequent amino acids. 3 . 1 . 10 Peptide characterisation Online bioinformatics tools were used to characterise the purified peptides by comparing them to other known peptides and proteins. The theoretical molecular weights of the antimicrobial peptides predicted from ovine cDNA were calculated using the ExPASy PeptideMass software (http ://au.expasy.org/tools/peptide-mass.html) . These masses were then compared to the experimentally determined molecular weights of the purified peptides. For the purified cathel icidin peptides, the N-terminal sequences were manually compared to the sequences of the predicted peptides. When the purified peptides were determined to be truncated forms of the predicted cathelicidins, then the peptidase database MEROPS (http://merops.sanger.ac.uk/) was searched to identifY peptidases that could have performed the truncation. For the purified peptides with N-terminal sequences that did not match any of the cathelicidin sequences, the ExP ASy BLAST software (http;llau.expasy.org/tools/blast/) was used to compare the N-terminal sequences of the purified peptides with known proteins and peptides. 65 Chapter 3 - Materials and Methods 3.1 .1 1 Analysis of prol i ne/a rgin ine-rich sequences The sequences of all the known pro line/arginine-rich cathel ic id in peptides were compared to find similarities. The sequence motifs and repeats were detected manually. The hydrophobic ities of the pept ides were compared by generating graphs using the ExP ASy ProScale Kyte/Doolittle hydrophobicity plot tool (http ://au.expasy.org/tools/protscale.html) . S imi larly, the po larit ies of the peptides were compared by generat ing graphs using the ExP ASy ProScale Zimmerman polarity p lot tool (http ://au.expasy.org/too ls/protscale .html) . 3.2 MATERIALS AND METHODS USED FOR MECHANISM OF ACTION TESTS 3.2 .1 Peptide synthesis Three peptides were synthesised based on the known ovine peptides. This was necessary because the time required to purify suffic ient amounts of the pure peptides to carry out a large number of experiments was too long. Between 1 5 and 20mg of each peptide were produced. To do this, N-(9-fluorenyl)methoxy carbonyl so lid-phase peptide synthesis using an Applied Biosystems model 432A synthesiser was carried out at the University of Brit ish Co lumbia Nucleic AcidlProtein Service fac il ity. The purit ies of the peptides were coruumed to be at least 99% using high-performance liquid chromatography (HPLC) and mass spectroscopy analysis. The sequences of the three peptides are given in Table 3 . 1 . The full OaBac5 mo lecule could not be synthesised, due to the large number of pro line residues, so a shortened version (the flIst 25 amino acids) called OaBac5mini was created. OaBac7.5mini was made in its natural form, the C-terminal 29 amino acids of OaBac7. 5 . SMAP29 was made as a 2 8 amino acid ami dated peptide because this is thought to be its natural form (SkerJavaj et ai, 1 999). Table 3.1 - Sequ ences of ovine antimicrobial peptides used for this research Peptide Sequence OaBac5mini RFRPPIRRPPIRPPFRPPFRPPVR -NH2 OaBac7. 5 mini RRIPRPILLPWRPPRPI PRPQPQPIPRWL -NH2 SMAP29 RGLRRLGRKIAHGVKKYGPTVLRI IRIA-NH2 66 Chapter 3 - Materials and Methods 3.2.2 Micro-broth di lution MIC method The minimum inhibitory concentrations (MTCs) of the ovine-derived peptides against a variety of organisms were determined using a modified broth dilution method (Wu and Hancock, 1 999a). This method is based on the standard micro-broth dilution method recommended by the Nat ional Committee of Laboratory Safety and Standards (NCLSS)(Amsterdam, 1 996). The changes to the standard method included: the use of polypropylene plates instead of polystyrene plates - because the cationic peptides bind to polystyrene plates and are inactivated; and using Mueller-Hinton broth (MHB) instead of trypticase-soy broth (TSB) - because the activity of the peptides can be inhibited by the salt in TSB. The test organisms used included eight Gram-negative bacteria, five Gram-positive bacteria and two yeasts. The test organisms and there sources are listed in Table 3 . 2 Table 3.2 - Microorganisms used for micro-broth di lution minimum inhibitory concentration tests and their sources Microorganism Escherichia coli 01 1 1 E. coli UB 1 005 E. coli DC2 E. coli 0 1 57 :H7 Salmonella typhimurium 14028s S. typhimurium MS4252S Pseudomonas aeruginosa P AO 1 P. aeruginosa Z6 1 Staphylococcus aureus NCTC 4 1 63 S. aureus MRSA R 1 4 7 S. aureus 1 056 MRSA S. epidermidis cl inical iso late Enterococcus faecalis A TCC 292 1 2 Candida albicans 1 05 C. albicans 3 1 53A Source Communicable Disease Centre, New Zealand. Department of Microbiology and I mmunology, University of Brit ish Columbia, Canada. Department of Microbiology and I mmunology, University of Brit ish Columbia, Canada. Communicable Disease Centre, New Zealand. Department of Microbio logy and I mmunology, University of Brit ish Columbia, Canada. Department of Microbio logy and I mmunology, University of British Columbia, Canada. Department of Microbio logy and I mmunology, University of British Co lumbia, Canada. Department of Microbio logy and I mmunology, University of British Columbia, Canada. National Type Culture Collection, England. Department of Microbio logy and I mmunology, University of British Columbia, Canada. Institute of Food, Nutrition and Human Health, Massey University, New Zealand. University of British Columbia Children's Hospital, Canada. American Type Culture Collection, USA. Department of Microbiology and I mmunology, University of British Columbia, Canada. I nstitute of Molecular B iosciences, Massey University, New Zealand. 67 Chapter 3 - Materials and Methods For these assays, S igma 96-wel l polypropylene microtitre p lates were used. The peptides were di luted across the rows in two-fold serial d i lutions from 64 to 0.0625Ilg/mL. Each wel l contained 50llL o f peptide d i luted i n a solution containing 0 .2% bovine serum albumin ( BSA) and 0 .0 1 % acetic acid. To each well, 50llL Muel ler-H inton broth (Difco) containing 2 x l 05 cells was added, to give a fmal concentration of 1 x l 05 cel ls. The plates were incubated overnight at 37°C. The next day the we lls that contained culture growth were recorded. The assessment was done visually. If the culture was turbid or contained a pel let, it was considered to have grown. The MIC was defmed as the concentration of the peptide in the last well in which culture growth did not occur. The assay was carried out five times independently. For each run dupl icates of each sample were included. For each test organism in each run, a contro l peptide was also tested. The controls were po lymyxin B for the Gram-negative bacteria, nisin for the Gram-positive bacteria and nystat in for the yeast. The MICs of these peptides were generally consistent across the runs (approximately 90% of the time they were the same). The MICs were reported as the geometric means and the 95% confidence intervals for the means. The geometric mean was used, as opposed to the arithmetic mean, because the data sets were skewed instead of normally d istributed. The use of the geometric mean meant that the larger values had less influence on the mean. The data values were log-transformed, and then the means, standard deviations and 95% confidence intervals of the log-transformed data were determined using the functions "AVERAGE", "STDEV" and "CONFI DENCE" in Microsoft Excel. The means and the 95% confidence l imits were exponential ly-transformed to give the geometric means and 95% confidence intervals of the means of each original data set. The variat ion was asymmetric because of the log-transformation. 3.2.3 C i rcular d ichro ism spectroscopy Circular dichroism (CD) spectroscopy was used to invest igate the secondary structures of the peptides in different solutions. The CD spectra were collected using a Jasco J-8 1 0 spectropolarimeter connected to a Jasco spectra manager using a quartz cell with a path length of 1 mm. The CD spectra were measured at 25°C, between 1 90 and 250 nm at a scanning speed of 50nmlminute. For each pept ide, CD spectra were collected for 251lM peptide in 1 0 mM phosphate buffer (pH 7.2), 50% 2,2,2-trifluroethano l (TFE), and 1 0mM lyso-PC/lyso-PG ( 1 : 1 ) . For each solution ten scans were carried out and averaged. Because the peptide 68 Chapter 3 - Materials and Methods solutions were dynamic each scan was slight ly different. The CD spectra of the solutions without the peptides were subtracted from those with the peptides in solution to eliminate scattering. 3.2.4 LPS bind ing assay The ability of the test peptides to bind to l ipopolysaccharide (LPS) was determined by assessing their abi l ity to displace bound dansyl polymyxin B (DPX) from LPS (Moore et ai, 1 986). The test solution was made by mixing 2 mL of 3 flg/mL LPS in 5mM HEPES buffer (pH 7.2), and 20flL of 1 00flM DPX. The LPS was almost saturated with DPX (>90%). The LPS used was iso lated from E. coli UBI 005 at the Department of M icrobiology and Immunology at the University of British Columbia (Moore et ai, 1 986). The fluorescence was measured using a Perkin E lmer Luminescence Spectrometer (Model LS50B). The excitation and emission wavelengths were 340nrn and 485nrn respectively. The solution was titrated with small, equal amounts of the test peptide and the fluorescence was recorded after each addition. If the peptides were able to displace some DPX from LPS, then the fluorescence decreased. From the data collected the inverse of the percentage of the fluorescence inhibited was plotted against the inverse of the peptide concentration, and a straight line was fitted. From the equation of the l ine it was possible to calculate the maximum percentage of DPX that could be displaced by the peptides I max ( I so=- l Ix-intercept). Three runs were carried out for each peptide. The mean I max and Iso values and the 95% confidence intervals of the means were calculated for each peptide using Microsoft Excel . To determine if there were differences between the means of the different test peptides the Excel Student T -test function was used. A p-value of less than 0.05 was considered significant. 3.2.5 Outer membrane permeabi l isation The abil ity of the test peptides to permeabilise the outer membrane of E. coli UB 1 005 was assessed by measuring their ability to promote uptake of I -N-phenylnapthylamine (NPN) into intact cells (Loh et ai, 1 984). For these tests, E. coli UB 1 005 cultures grown to an optical density of 0 .5 at 600nm were used. The cells were col lected by centrifugation ( Sorvall SS34, 1 0,000rpm, 5 minutes) and washed with 5mM HEPES buffer containing 5flM CCCP 69 Chapter 3 - Materials and Methods (carbonylcyanide m-chlorophenyl-hydrazone) and 5mM glucose, then re-suspended in this buffer to the same optical density. The fluorescence was measured usmg the luminescence spectrometer (Perk in E lmer Luminescence Spectrometer - Model LS50B), with excitation and emission wavelengths o f 350nm and 420nm, respectively. The init ial suspension was made by mixing 1 mL of cell suspension and 20llL of 0.5mM NPN, and the init ial fluorescence was recorded. The suspension was titrated with smal l, equal amounts of the test peptide, and the fluorescence was recorded. I f the test peptides enabled the uptake of NPN into the cells, then the fluorescence increased. As a contro l, the NPN uptake caused by the antibiotic po lymyxin B at a concentration of 4llg/rnL was also determined . The NPN uptake caused by the peptides was expressed as a percentage of the NPN uptake caused by this contro l, and this percentage o f N PN uptake was plotted against the peptide concentration. The assay was carried out three times. The mean NPN uptakes at each concentrat ion and the 95% confidence intervals for the means were calculated using Microsoft Excel and were plotted using SigmaPlot. 3.2.6 Cytoplasmic membrane depolarisation The abil ity of the test peptides to interact with the bacterial cytoplasmic membrane was determined by invest igat ing their abi l ity to depolarise this membrane ( Wu and Hancock, 1 999b) . The fluorescence of the membrane potential-sensitive dye 3,3-dipropylthiacarbo­ cyanine (D iSC3 5) was monitored when added to the E. coli DC2, which is an outer membrane permeable mutant. This strain was used to permit the assay to be performed in the absence of an outer membrane permeabil ising agent like EDT A. The culture was grown to an optical density of 0 .5 at 600nm. The cells were col lected using centrifugation (S ilencer, 3000rpm, 1 0 minutes), and washed with 5mM HEPES buffer containing 5mM glucose, then re-suspended in this buffer to the same optical density. To introduce the dye into the cel ls, 45mL of the cel l suspension was incubated with 451lL of OAmM DiSC35 (fmal D iSC35 concentration of OAIlM) at room temperature with mixing for 30 minutes. To stabil ise the irmer membrane polarity, 5 mL of 1 M KCI was added to the suspension and it was incubated for a further 1 0 minutes at room temperature with mixing. The init ial fluorescence of the suspension containing the cells, DiSC35 and KCI, was measured with the excitation wavelength set to 622nm and the emission wavelength set to 70 Chapter 3 - Materials and Methods 665nm (Perkin Elmer Luminescence Spectrometer - Model LS50B). The suspension was titrated with small, equal amounts of the test peptide, and the fluorescence was recorded. I f the test peptides caused the release ofD iSC35 from the cells then the fluorescence increased. As a control, the DiSC35 release caused by the ant ibiotic gramicidin S at a concentration of 4J.lg/mL was also determined. The DiSC35 release caused by the peptides was expressed as a percentage of the DiSC35 release caused by this control, and this percentage of DiSC35 release was plotted against the peptide concentration. The assay was carried out three times. The means of the three runs and the 95% confidence intervals for the means were calculated using Microsoft Excel and were plotted using S igmaPlot. 3.2.7 Optical density and viable cel l counts over time To determine how quickly the peptide acted on Gram-negative bacteria the viable cell numbers and the culture optical density were monitored over time. Log phase E. coli 0 1 1 1 cells were diluted in MHB to an optical density at 600nm of approximately 0. 1 . The culture was split between 5 cuvettes, each containing 2mL. The cultures were incubated in a water bath at 37°C. At time zero, the peptides were added so that their final concentrations were twice their MICs (4/lg/mL for SMAP29, 4/lg/mL for OaBac5mini, 32/lg/rnL for OaBac7.5mini). As a negative control, one sample contained 1 0J.lL of 0.0 1% acetic acid, and as a positive control, one sample contained 4J.lg/mL po lymyxin B . The optical density at 600nm of each sample was recorded initially and at various t imes up to two hours after the start of the assay. At each t ime point 1 00J.lL was taken from each sample, and ten-fold serial dilutions up to 1 0-7 were made and plated on Brain-Heart Infusion agar. The plates were incubated overnight at 3 7°C, and the plates containing the appropriate dilutions were counted the fo llowing day. 3.2.8 Peptide-DNA binding To determine whether the peptides were able to bind to DNA, the abil ity of the DNA to migrate on an agarose gel was monitored (Park et ai, 1 998). The test solutions were made by mixing 1 00ng of DNA (Lambda DNA Eco markers) with various amounts of test peptide up to 400ng in 20J.lL of binding buffer ( l OmM Tris-HCI (pH8), 20mM KCI, I mM EDTA, I mM DTT, 5% glycerol, 50J.lg/rnL BSA). The solutions were left to react for 1 hour. To trace the DNA, 2J.lL of loading buffer (20% sucrose, 0 . 1 25% bromophenol blue) was added to each 71 Chapter 3 - Materials and Methods sample before they were loaded onto a 0 .75% agarose/TAE gel. The gel was run at 1 08V and 92A for approximately 1 hour. The running buffer used was T AE buffer (0 .04M Tris-acetate, O .OO l M EDTA). The gel was removed from the apparatus and stained with a drop of ethidium bromide in water for 20 minutes. The gel was placed over a fluorescent light to make the bands visible. The gel was photographed with a digital camera to record the results. 3.3 MATERIALS AND METHODS USED TO INVESTIGATE BACTERIAL CELL MORPHOLOGICAL CHANGES 3.3 .1 Transmission electron microscopy Transmission electron microscopy (TEM) was used to investigate the morpho logy of bacterial cells treated with the test peptides. TEM works in a similar way to regular light microscopy, except that instead of using a focussed beam of l ight to see through the samples, a focussed beam of electrons is used. L ight microscopy is l imited, because of the physics of light, to 1 000 times magnification and 0.2f.1m resolution, whereas the electron microscope used for this work was capable of magnifications up to 70,000 times. Log phase cultures of E. coli 0 1 1 1 and S. aureus NCTC 4 1 63 in MH B were split into 1 .5rnL aliquots. The cells were co llected by centrifugat ion ( 1 0,000rpm, 5mins) and re-suspended in peptone water. For each organism 6 samples were prepared; two untreated, two treated with SMAP29 (4f.1g/rnL for both organisms) and two treated with OaBac5 mini (4f.1g/rnL for E. coli 01 1 1 and 64f.1g/rnL for S. aureus NCTC 4 1 63) . The samples were incubated at 37°C for one hour. The cel ls were collected by centrifugation ( 1 0,000rpm, 5mins) to remove the peptone water. To fix the cel ls, 2% glutaraldehyde in O . I M cacodylate was added and the samples were incubated at 4°C for one hour. The cells were col lected by centrifugation ( 1 0,000rpm, 5mins) and washed twice with O. I M phosphate buffer. To postfix the cel ls, 1 % osmium tetraoxide was added and the samples were left at room temperature for one hour. The samples were dehydrated with graded ethano l solutions (30% ethano l for 1 0mins, 60% ethano I for 1 0mins, 90% ethano I for l Omins, 1 00% ethano I for 1 0mins, 1 00% ethano I for 1 hour), embedded in Procure 8 1 2 resin and left to polymerise for 48 hours. Thin slices ( approximately 1 00nm) of the samples were made with a diamond knife and stained with uranyl acetate and lead c itrate on grids. For each sample ten sect ions were cut. Each of the sections was examined with a Phi l ips E M2 0 1 80kV Transmission E lectron Microscope and images were taken with a 35nun camera. 72 Chapter 3 - Materials and Methods 3.3.2 Atomic force microscopy Atomic force microscopy (AFM) was used to investigate the surface morphology of bacterial cells treated with the test peptides. AFM uses a cantilever with a sharp tip to probe the surface of the sample to gather information about its topology. The cantilever is between 1 00 and 200llm long and the tip is approximately 21lm long and less than l OoA in diameter. When the tip and the sample come into close proximity, the force between the two causes the cantilever to bend or deflect. As the tip scans over the sample the movement in the cantilever is detected by using a laser beam that bounces off the sample onto a position-sensitive photodetector. This information is used to generate a map o f the surface topography of the sample. Log phase cultures of E. coli 0 1 1 1 and S. aureus NCTC 4 1 63 in MHB were split into I mL aliquots. The cells were collected by centrifugation ( l O,OOOrpm, 5mins) and resuspended in peptone water. For each organism one sample was untreated, one was treated with SMAP29 (4Ilg/mL for both organisms) and one was treated with OaBac5mini (4Ilg/mL for E. coli 0 1 1 1 and 641lg/mL for S. aureus NCTC 4 1 63 ). These peptide concentrations were four times the MICs for the given organisms. The samples were incubated at 3rc for 30 minutes. The cells were collected by centrifugation ( l O,OOOrpm, 5mins) to remove the peptone water. A number of different methods were tried to secure the cells for AFM imaging. These included passing the samples through polycarbonate membranes (SPI -Pore Filter, 1 3mm d iameter, 0 .81lm pores) to entrap the cells in the pores, and immobilising the cells in agarose gel (5% agarose). The most successful method used involved drying the culture on a glass slide. The treated samples were washed with, and resuspended in, water, and then diluted l OO-fo ld. Approximately l OllL of the sample was dropped onto a glass microscope slide and allowed to dry at room temperature. For each sample duplicate slides were prepared. To image the samples an Asylum Research MFP-3D scanning probe microscope was used. The tips used were either NT-MDT CSG0 1 or Olympus TR400PSA, which typically have spring constants in the range 0 .01 -0.08 N/m. The areas for imaging were chosen randomly from the total sample area. 73 Chapter 3 - Materials and Methods 3.4 MATERIALS AND METHODS USED TO ASSESS THE EFFECT OF CONDITIONS ON PEPTIDE ACTIVITY 3 .4. 1 Salt effects The effect of salt concentration on the antimicrobial act ivity of the peptides was tested by determining the MICs of the peptides against E coli 0 1 57 :H7 in a variety of salt concentrations. The modified micro-broth dilution assay described in Section 3 .2 .2 was used. The MHB used in the assay was altered by the addition of NaCI. The [mal NaC l concentrations desired were 0, 50, 1 00 and 250rnM, so MHB so lutions containing 0, 1 00, 200 and 500rnM NaCl were required, because the media were di luted with equal amounts of the peptide solution in the assay. Three runs each containing dupl icates of each sample were carried out. The geometric means and 95% confidence l imits of the means of the MICs were calculated in the way described in Section 3 .2 .2 . The MIC data from each experiment were analysed to determine if there was a difference between the mean MICs for the different peptides and for the different conditions. The "general analysis of variance" function in statistical programme GenStat was used. The y­ variate was "Ln MIC", the treatment structure was "condition x peptide" and the blocks was "run number". From the output, the F-statist ics for the condition and peptide were calculated. For these calculations the interact ion error term was used instead of the residual error term. This was done because the residual error was likely to be an underestimate of the variabil ity of the data because the data were discrete. Moreover, the use of the interaction error was a more conservative approach. The F-statistics were used to calculate the probability that the means of each data set were the same. 3.4.2 Cation effects The effect of cation concentration on the antimicrobial activity of the peptides was tested by determining the MICs of the peptides against E. coli 0 1 57 :H7 in a variety of cation concentrations. This was accomplished in the same way as described in Section 3 .4 . 1 for determining the effect of the salt concentration on the activity of the peptides . The two monovalent cations, Na+ and K+, and two divalent cations, Ca2+ and Mg2+, were added to the media as chloride salts. The concentrations o f these cations in the assays were 0, 2 , 5 and 1 0 mM. The statistical analysis was carried out using the method described in Section 3 .4. 1 . 74 Chapter 3 - Materials and Methods To directly compare the means of two sets of data the Student' s t-test was used. This was done using the "TTEST" function in Excel for a one-sided, two-sample, equal variance test. 3.4.3 pH effects The effect of pH on the antimicrobial activity of the peptides was tested by determining the MICs of the peptides against E. coli 0 1 57:H7 at a variety of pH values. This was achieved by using the micro-broth dilution MIC method described in Section 3 .2 .2 and altering the pH of the media by the addit ion of concentrated HCI o r NaOH. The peptides were tested in pH conditions that varied from pH5 to pH9. The statistical analysis was carried out using the method described in Section 3 .4. 1 . 3.4.4 Temperature effects The effect of temperature on the antimicrobial activity of the peptides was tested by determining the MIC of the peptides against E. coli 0 1 57 :H7 after they had been heated to a variety of temperatures. The MICs were determined using the micro-broth dilution MIC method described in Section 3 .2 .2 . Before the MIC assay, the peptides were heated for 30 minutes and then cooled. Temperatures of 30, 40, 50, 60, 70 80 and 90°C were used. The samples were also autoclaved ( 1 2 1 QC, 20 minutes). The statistical analysis was carried out using the method described in Section 3 .4 . 1 . 3.4.5 Synergistic effects between test peptides To test whether the peptides worked synergistically against E. coli 0 1 57:H7, checkerboard t itrations were used (Fidai et aI, 1 997). One compound was diluted down the columns and the other was diluted across the rows in a 96-well polypropylene microtitre plate. Each well contained 25jJ.L of one peptide diluted in a solution containing 0.2% bovine serum albumin (BSA) and 0.0 1 % acetic acid, and 25jJ.L of the other peptide diluted in the same solut ion. To each well, 50jJ.L Mueller-Hinton broth (Difco) containing 2x l 05 cells was added, to give a final concentration of 1 x l 05 cells. The plates were incubated overnight at 3 7°C. The next day the wells in which bacterial growth occurred were recorded. The fractional inhibitory concentrations (FICs) were calculated using the formula: FIC = [A]/MICA + [B]/MICB. MICA and MICB were the MICs of peptides A and B alone. These were the lowest concentration of each peptide that could inhibit the growth of the culture when the 75 Chapter 3 - Materials and Methods concentration of the other peptide was zero. [A] and [B] were the MICs of pep tides A and B when in combination. Each pair of [A] and [B] values from the same wel l were used to calculate the FIC. The FIC reported is the lowest FIC value calculated from the p late. Three independent runs were carried out . The FICs were the same for each run so no statist ical analysis was carried out. 3.4.6 Synergistic effects between test peptides and common antibiotics The checkerboard titration method described in Section 3 .4 .5 was used to test whether the test peptides worked synerg istically with common ant ibiot ics against E. coli 0 1 57. The conmlon ant ibiotics tested in combinat ion with the test peptides were po lymyxin B, ampici l l in, kanamyc in A and rifampicin. These antibiotics were chosen because they each have d ifferent mechanisms of action. The FICs of each antibiotic with each test peptide were determined three times in separate runs. The FICs were the same for each run so no statist ical analys is was carried out. 3.5 MATERIALS AND METHODS USED FOR THE PILOT -SCALE EXTRACTION OF ANTIMICROBIAL PEPTIDES FROM OVINE BLOOD 3.5.1 Crude extraction process A number of pilot-scale extraction runs were carried out to determine whether the crude extract could be produced on a larger-scale than that used in the laboratory. For these extractions 1 a-SOL o f blood was used compared to 2-4L in the laboratory extractions. The pilot-scale extract ion of antimicrobial peptides used a modified version of the laboratory extraction described in 3 . 1 . 1 . Full details of this entire process are given in Chapter 8 . The first step in the extraction process was to separate the white blood cells from the rest of the blood. For the pilot-scale extractions the intention was to use the modified lab-scale process based on that of the University of California, Los Angles group (Borenstein et ai, 1 99 1 ) to lyse the red blood cells. However, during experimentation it was d iscovered that it was possible to separate most of the red blood cells fro m the white blood cells using centr ifugation only. A continuous-feed disk-stack centrifuge (Alfa-Laval Cream Separator), usually used for milk separation, was used. When the whole blood was passed through the centrifuge the p lasma came out the liquid stream, most of the red blood cells came out the concentrated so lids stream, and the white blood cells and the rest of the red blood cel ls were 76 Chapter 3 - Materials and Methods pelleted inside the centrifuge. The white blood cells and residual red blood cells were suspended in PBSX buffer ( l 3 7mM NaCl, 2 .7mM KCl, 0 .5mM MgCh, 8 . 1 mM Na2HP04, 1 . 5mM KH2P04, pH 7.4). The contaminating red blood cells were lysed using the ammonium chloride method, and then the white blood cells were collected using centrifugation. Once the white blood cells were separated from the rest of the blood the next step in the extraction process was to isolate the granules, which contained the antimicrobial peptides, from the neutrophils. The original lab-scale process used ultrasonic energy to disrupt the white blood cells and release the neutrophil granules (Eisenhauer et al, 1 989). The granules were then co llected using high-speed centrifugation and dissolved in acetic acid solution. However, due to the large amount of heat generated by sonication, this operation would be impractical for a large-scale process. Therefore, an alternative method to disrupt the white blood cells was required. Mechanical disruption using a bench-top blender was tested and it was found to be as effective as the sonication method. The yield from an extraction using the sonicator was the same as the yield from an extraction using the blender. A further modification was introduced to the pilot-scale extraction process. Instead of resuspending the pelleted white blood cells in PBSX buffer, acetic acid so lution was used. This eliminated the need for the centrifugation step to pellet the granules so they could be resuspended in the acetic acid solution. This meant that the white blood cell debris was not removed prior to the acid extraction step; however, this d id not affect the process. The next step in the process was to extract the antimicrobial peptides from the neutrophil granules. In both the lab-scale and pilot-scale processes the granules were suspended in 1 0% acetic acid and left overnight at 4°C with constant mixing. However, these conditions were not determined experimentally, so further work is required to optimise the extraction. Both the extraction solution and the extraction method should be investigated to determine the most effective method for extracting the antimicrobial peptides. The final step in the process was to convert the product of the acid extraction into a crude antimicrobial extract that was stable for use and storage. This involved the removal of the neutrophil granules and other suspended solids, removal of the acetic acid and concentration of the solution. In both the lab-scale and pilot-scale processes the formulation of the crude extract was carried out in the same way. After centrifugation to remove the solids rotary evaporation was used to remove the acid and freeze-drying was used to remove the water. 77 Chapter 3 - Materials and Methods Then the product was dissolved to form the crude extract. These acid and water removal steps were used because they did not require heating, which may have damaged the product. 3.5.2 Minimum inh ib itory concentrations The minimum inhibitory concentrations (MICs) of the crude extract against a variety of organisms were determined using the modified micro-broth dilution method described in Section 3 .2 .2 . The test organisms included eight Gram-negative bacteria, eight Gram-posit ive bacteria and one yeast . The test organisms and there sources are listed in Table 3 . 3 Table 3.3 - Microorganisms used for crude extract minimum inhibitory concentration tests and their sou rces Microorganism Source Escherichia coli 01 1 1 Communicable Disease Centre, New Zealand. Escherichia coli 01 57 :H7 Communicable Disease Centre, New Zealand. Salmonella enteritidis Institute of Ye teri nary, Animal and Biomedical Sciences, Massey University, New Zealand. Salmonella typhimurium Institute of Techno logy and Engineering, Massey University, New Zealand. Klebsiella pneumoniae Institute of Technology and Engineering, Massey University, New Zealand. Pseudomonas aeruginosa NCTC 6749 National Collection of Type Cultures, England. Pseudomonasfluorescens ATCC 1 3525 American Type Culture Collection, USA. Yersinia enterocolitica A TCC 96 1 0 American Type Culture Collect ion, USA. Staphylococcus aureus NCTC 4 1 63 National Collection of Type Cultures, England. Staphylococcus aureus 1 056 MRSA Institute of Food, Nutrition and Human Health, Massey University, New Zealand. Staphylococcus faecalis NHI 89 National Health I nst itute, New Zealand. Bacillus cereus NCIB 8709 National Collection of Industrial Bacteria, Scotland. Bacillus nato I nstitute of Food, Nutrition and Human Health, Massey University, New Zealand. Listeria monocytogenes 1 08 A Institute of Food, Nutrition and Human Health, Massey University, New Zealand. Listeria monocytogenes NCTC 10884 National Collection of Type Cultures, England. Listeria monocytogenes NCTC 7973 National Collection of Type Cultures, England. Candida albicans 3 1 53A Inst itute of Mo lecular B iosciences, Massey University, New Zealand. 78 Chapter 3 - Materials and Methods Two-fo ld serial dilutions of the extract at concentrations between 1 28 and 0.0625 ).lL crude extractlmL were tested. The MIC was defined as the concentration of the peptide in the last well in which culture growth did not occur. The MIC values could not be expressed in ).lg/mL because the mass of the solids in the crude extract was not comparable to the activity due to the presence of other non-active proteins and peptides in unknown concentrations. Instead, the MICs were described in units/mL, where one unit was defined as the amount required to inhibit 1 05 CFU/mL E. coli 0 1 1 1 . The assay was carried out three times independently. For each run duplicates of each sample were included. The statistical analysis was done as described in Section 3 .2 .2 . 3.5.3 Transmission electron microscopy Transmission electron microscopy (TEM) was used to investigate the morphology of bacterial and yeast cells treated with the test peptides. Log phase cultures of E. coli 0 1 1 1 , S. aureus NCTC 4 1 63 and C. albicans 3 1 53A in MHB are split into 1 mL aliquots. The cells were collected by centrifugation ( l O,OOOrpm, 5mins) and resuspended in peptone water. For each organism there were four samples prepared; two untreated (0.0 1 % acetic acid) and two treated with l O).lL of crude extract. The samples were incubated at 37°C for one hour. The samples were prepared for TEM and photographed as described in Section 3 .3 . 1 . 3.5.4 Yield calculations The yield of units of activity was calculated using Equation 3 .2 for each extraction. This is the number of units of activity produced from each l itre of blood, where one unit is the amount required to inhibit 1 05 CFU/mL E. coli 0 1 1 1 . volume of extract/ / volume per unit yield = ----------'------"------ - Equation 3.2 volume of blood where: volume of extract vo lume per unit vo lume of blood the volume of crude extract produced from the run (mL) the volume of crude extract required for one unit of activity (mL) the volume of blood processed for the run (L) 79 Chapter 4 - Isolation and Characterisation CHAPTER 4 ISOLATION AND CHARACTERISATION OF ANTIMICROBIAL PEPTIDES FROM OVI NE NE UTROPHILS 4. 1 INTRODUCTION The work presented in this chapter was concerned with the first objective of this research project, which was to purify and identify ant imicrobial peptides fro m ovine blood. Prior work showed that it was possible to extract components from ovine neutrophil granules that have broad-spectrum ant imicrobial activity (Anderson and Yu, unpublished results). The first step of the current research was to separate and identify the components in the crude extract to determine the components responsible for the ant imicrobial activity. It was hypothesised that most of the antimicrobial activity would be caused by cathel ic idin peptides. As d iscussed in the l iterature review in Chapter 2 , cathe licidins are a family of gene-encoded cat ionic antimicrobial peptides found only in mammals. These peptides are stored as inactive propeptides in neutrophil granules and are cleaved into active peptides by neutrophil e lastase when required. Cathelicidins all share a strongly conserved N-terminal precursor, similar to that of porcine cathelin (hence the name) . However, the C-terminal regions of these mo lecules, which are bacteric idal, are highly variable. Using this knowledge of the cathel in- l ike domain, eight ovine genes encod ing seven different cathe licidins have been identified : two a-helical peptides, SMAP29 and SMAP34 (also cal led OaMAP28 and OaMAP34); four arg inine/prol ine rich extended structures, OaBac5, OaBac6, OaBac7.5 , and OaBac l l ; and two copies of the dodecapeptide OaDode ( Huttner et ai, 1 998). Unt il now, only two variants of the predicted peptide OaBac5, have been iso lated from ovine neutrophils (Shamova et ai, 1 999). Another two of the predicted peptides, SMAP29 and SMAP34, have been synthesised by several groups, and one of these, SMAP29, has potent, broad-spectrum antimicrobial act ivity (Skerlavaj et ai, 1 999; Travis et ai, 2000). The object ive of the research presented in this chapter was to iso late the components in the ovine neutrophil crude extract that displayed antimicrobial activity in the earl ier study (Anderson and Yu, unpublished result s, 2000). Once iso lated, the mo lecular weights of the compounds were determined by mass spectroscopy, and their N-terminal sequences were determined and compared with those of known peptides and proteins. 80 Chapter 4 - Isolation and Characterisation 4.2 EXTRACTION OF CRUDE ANTIMICROBIAL SOLUTION The process outlined in Figure 4. 1 was used to obtain the crude antimicrobial extract from the ovine blood. This process was developed by combining ideas from numerous processes presented in the l iterature as described in Section 3 . 1 . 1 . Over the course of this research the extraction process was modified several times to make it quicker, easier and more economical. blood + sodium ciliate solution FIL Tr nON I----- "'"" , do', blood + sodium citrate CENTRI{UGATION I----- b lood p i" m" PBSX buffer--.white b lood ce l l s + red b lood ce l l s ,mmoolom ,hl",d, .1 ISOTO+ SHOCK solution ----J � white b lood c el ls + lysed red b lood ce l ls CENTRI{UGATION 1----- ',,,' "d blood "", PBSX buffer--. white b lood ce l l s sONlfAno" neutrophi l granules + white b l ood ce l l d ebris CENTRI{UGATION I • whit, b lood "" � r-- debris 1 0% acet i c acid --. neutrophi l granu les EXT4CTION neutrophi l granules + ant imicrob ia l pept ides i n a cet ic ac id so lutio n CENTR+GATION I----- ",,'coph l ! g""'" ant imicrob ia l pept ides in acet i c acid s o lut ion ROTARY E{APORATION I----- '''' IC "Id ant imi crob ia l pept ides i n water FREE1,DRYI"G I-----w" " dri ed crude extract 0 01 % ,,,t I, "Id � REDlSSfLUTION crude antimicl obi.l l e xtract in 0 .0 1 " • • lcetic acid Figure 4.1 - Flowchart showing the process used to extract the crude antimicrobial solution from ovine blood. Sodium citrate solution contained 1 0% sodium citrate. PBSX buffer contained 1 37mM NaCI, 2.7mM KCI, 0.5mM MgCI2, 8.1mM Na2HP04 and 1 .5mM KH2P04 (pH 7.4). Ammonium ch loride solution contained 0.83% ammonium chloride. 81 Chapter 4 - Isolation and Characterisation Throughout the course of this research the extraction process was carried out approximately fifteen times, with between two and four litres of ovine blood being processed each t ime. To ensure the steps in the process worked effectively, samples were taken after each critical step and examined. The blood cells were stained using a Quick-Diff staining kit then viewed under a microscope. Images of the blood at various stages during the extraction process are given in Figure 4.2 . The whole blood contained numerous small red blood cells and some larger white blood cel ls. If red blood cells were still present in the blood after the lysis step, this step was repeated as outlined in the methodology in Section 3 . 1 . 1 . The white blood cel l disruption step was simi larly repeated if white cel ls were st ill present after the disruption step . .. Figure 4.2 - Images of typical stained blood samples d u ring the extraction p rocess. 'A' shows whole ovine blood, 'B' shows ovine blood after the red blood cell lysis step, and 'C' shows ovi ne blood after the wh ite blood cell disru ption step. After each extraction, the antimicrobial activity of the crude extract was tested to determine whether the extraction was successful. To do this the rad ial diffusion plate assay method was used. This method is described in Section 3 . 1 .6 . The test organisms used were E. coli 0 1 1 1 , S. aureus NCTC 4 1 63 and C albicans 3 1 53 A These organisms were chosen to ind icate the spectrum of activity of the extract because they are a Gram-negative bacterium, Gram­ positive bacterium and yeast respectively. Photographs of a typical set of plate assays for the crude extract are shown in Figure 4.3 . To separate the components in the crude extract, the extract was fractionated using either gel fi ltration or cationic exchange chromatography. The active fractions were determined using the radial diffusion plate assay method, and were further processed using RP-HPLC to separate the ind ividual components. These RP-HPLC fract ions were then subjected to a rad ial diffusion p late assay to determine which ones were active, and these pure act ive components were then analysed. The characterisation of ant imicrobial peptides via the gel 82 Chapter 4 - Isolation and Characterisation fLltrat ion purification method will be discussed first, fo llowed by the characterisation via the cationic exchange method. Figure 4.3 - Images of typical plate assay results of neutrophil crude extract against three test organisms. 'A' shows E. coli 01 1 1 , 'B' shows S. aureus NCTC 4163 and 'C' shows C. albicans 3153A. Wel l ' 1 ' contains the negative control (0.01% acetic acid), '2' contains the positive control ( 1llg/mL polymyxin for E. coli 01 1 1 , 1 1lg/mL nisin for S. aureus NCTC 4163, and 1 0llg/mL nystatin for C. albicans 3153), '3' contains ovine neutrophi l crude extract di luted 1 /10, and '4' ovine neutrophil extract undiluted. The white circles are computer graphics to high light the clearings. 4.3 PURIFICATION OF ANTIMICROBIAL PEPTIDES USING GEL FILTRATION AND RP-HPLC The crude extract was passed through a P l O gel filtration co lumn to separate the components in the so lution according to their s izes using the method described in Section 3 . 1 .3 . A typical chromatograph is shown in Figure 4.4. This P l O resin, which has a molecular weight cut -off of 1 0kDa, was chosen because it was thought that the active components were cathelic idin peptides, the majority of which have a molecular weight between 3 and 5kDa. The running buffer contained 5% acetic acid and no salt. Salt was not added to the buffer because it was too difficult to remove the salt from the solutions after the fractions were concentrated without losing the active components. It was undesirable to keep the salt in the solutions because it reduces the antimicrobial activity of the peptides as demonstrated in Section 7 .2 . The plate assay method was used to determine which gel filtrat ion fractions had ant imicrobial activity. Before testing, each fraction was rotary evaporated to remove the majority o f the acid, freeze-dried to remove the residual ac id and water, and then disso lved in 2-5mL 0.0 1 % acetic acid. The results of a plate assay of the gel filtration fractions from a typical run are given in Table 4. l . The earliest fractions to elute from the gel filtration co lumn (F 1 , F2 and F3) had antimicrobial activity against the three test organisms, E. coli 01 57:H7, S. aureus 83 Chapter 4 - Isolation and Characterisation 1 056 MRSA and C. albicans 3 1 53A, but the fractions that eluted later (F4, F5 and F6) were not active against any of the test organisms. Fraction 2 was the most active fraction, fo llowed by Fraction 1 and then Fract ion 3 . E c o 00 N .... cu Q) () c cu ..c L- a (J) ..c <{ o 1 0 F1 20 30 40 50 Fraction Number Figure 4.4 - Typical gel filtration chromatograph resulting from the addition of an ovine neutrophi l crude extract into a P10 gel filtration column. The running buffer was S% acetic acid, which was pumped through the column at a rate of 20mUhour. Each fraction contained SmL of liquid . Table 4.1 - Antimicrobial activity of typical ovine neutrophil extract gel filtration fractions against test organisms. Test solution Diameters of clearings on plate assays (mm) E. coli 01 57:H7 S. aureus 1 056 MRSA C. albicans 3 1 53A positive control' 1 8 9 1 3 F1 8 7 6 F2 1 2 8 7 F3 6 4 2 F4 no clearing no c learing no c learing F5 no clearing no c learing no c learing F6 no c learing no c learing no c learing negative control" no clearing no c learing no c learing • The positive controls were 1 Jlg/mL polymyxin for E. coli 0157:H7, 1 Jlg/mL nisin for S. aureus 1 056 MRSA, and 1 0Jlg/mL nystatin for C. albicans 31 53A . •• The negative control was 0.01% acetic acid. 84 Chapter 4 - Isolation and Characterisation SDS-PAGE was used to determine the number of components in each gel filtration fraction and their approximate molecular weights according to the method described in Section 3 . 1 .2 . An image of a typical SDS-PAGE gel i s given in Figure 4 . 5 . This showed that F2, the most active fraction, contained molecules of the expected size for cathelicidin peptides (3-5kDa) . Interestingly, F3 and F4 contained components that were larger than those in F2, even though these fractions eluted from the gel filtration column after F2. The P l O resin was supposed to separate components smaller than 1 0kDa, and larger molecules, such as those present in F3 and F4, should have passed straight through the column. The large molecular weights of the mo lecules in F3 and F4 were probably not due to mult imers forming during electrophoresis because the p-mercaptoethanol should have reduced the bonds. These large components may have had reduced migration speeds during gel filtration as a consequence of their interactions with the resin. MWt (kOa) "C � ca "C c::: ca -(/) c::: Q) - 0 � Q. -(.) ca � ->< Q) Q) "C :::J � (J � N M oo:t c::: c::: c::: c::: 0 0 0 0 :;: :;: :;: :;: CJ (.) (.) (.) ca ca ca ca � � � � LL LL LL LL MWt (kOa) fS. 9 14.4 10.7 8.2 25 1 1 .93 (NA, 1 7.7 1 ) 22. 1 5 (NA, 54.56) S. aureus NCTC 0 1 2 .02 (NA, NA) 1 0.67 (NA, 1 8.63) 1 0.20 ( 1 .27, 1 9.47) 4 1 63 1 00 >30 >30 > 1 25 C. albicans 0 >30 1 3 .60 (NA, 2 1 . 92) 44.29 (NA, 77.6 1 ) 3 1 53A l OO >30 1 2.02 (NA, NA) 3 1 .32 (NA, 64. 1 2) The values given in brackets are the limits of the 95% confidence intervals for the MICs. 'NA' means that the limits of the confidence intervals could not be calculated because the bounds did not cross the x­ intercept. Of the three pept ides tested, OaBac5y ( Pc) was the most active against the test bacteria, and OaBac7.5(32-60) was the most active against the yeast. OaBac7.5 (32-60 ) retained its activity against E. coli and C. albicans, but not S. aureus at high salt concentrations, whereas the activity of OaBac5y was impaired against both bacterial strains. OaBac5 (Pa) had similar act ivity to OaBac5y against the bacteria, but its MIC against the yeast could not be evaluated because there was not enough material to test higher concentrat ions. OaBac5 and OaBac5y only differ by one amino acid so it is expected that their activity would be simi lar. OaBac5y has a leuc ine residue instead of an arginine residue in OaBac5, which is a change from a posit ively charged amino ac id to a hydrophobic one . 4.8 SEQUENCE ANALYSIS OF PROLINE/ARG ININE-RICH PEPTIDES All of the antimicrobial peptides purified from the second gel fi ltration fraction of the crude ovine neutrophil extract were proline/arginine-rich cathel icidins cal led bactenecins. As we l l as these, there are a number of other bactenecins that have previously been ident ified in sheep and other ruminants. The sequences o f all the known prol ine/arginine-rich cathel ic idin peptides are given in Table 4.7. These sequences were analysed and compared as described in Section 3 . 1 . 1 1 . 94 (0 c.n Table 4.7 - Identification of repeats in the sequences of the proline/arg inine-rich cathelicid in peptides. Peptide Reference Repeat PR39 purified from ( RXPP ) ( XXPP ) porcine blood ( RXPP ) Bac4 predicted from ( XPXP ) 2 ( XPRP ) 2 bovine cDNA OaBac6 predicted from (XPXP ) 2 ( XPRP ) 2 ovine cDNA Bac5 purified from (XRPP ) 4 bovine blood OaBac5 predicted from ( XRPP ) 4 ovine cDNA ChBac5 purified from ( XRPP ) 4 caprine blood Bac7 purified from LP ( XPRP ) 3 bovine blood OaBac7.5 predicted from LP ( XPRP ) 3 ovine cDNA OaBac l l predicted from LP (XPRP ) 3 ovine cDNA Sequence RRRPRPPYLPRP RPPP FFPP RLPP RIPP GFPP RFPP RFP RRLHPQHQRFPRERP WPKP LSLP LPRP GPRP WPKP L RRLRPRHQHFPSERP WPKP LPLP LPRP GPRP WPKP LPLP LPRP GLRP WKPL RFRPPI RRPP IRPP FYPP FRPP IRPP IFPP IRPP FRPP LGPFP RFRPPI RRPP IRPP FRPP FRPP VRPP IRPP FRPP FRPP IGPFP RFRPPI RRPP IRPP FNPP FRPP VRPP FRPP FRPP FRPP IGPFP RRIRPRPPR LPRP RPRP LP FPRP GPRP I PRP LP FPRP GPRP I PRP LP FPRP GPRP I PRP RRLRPRRPR LP RPRP RPRP RPRS LP LPRP QPRR I PRP I L LP WRPP RPIP RPQP QPIP RWL RRLRPRRPR LP RPRP LPRP KPRP I PRP LP LP LPRP RPRR I PRP I PRP LP LPQP QPSP RPRP RPRS LP LPRP RPKP I PRP LP LPRP RPRP I PRP L The repeats are shown by underlining. The residues that differ from the repeat pattern are indicated by bold-type. Reference Agerberth et ai, 1 99 1 Scocchi et al, . 1 998 Huttner et ai, 1 998 Frank et al. 1 990 Huttner et aI, 1 998 Shamova et aI, 1 999 Frank et al. 1 990 C') :l" III "0 It ., � Huttner et at, 1 998 I iii 0 Di" C!: 0 � Huttner et at, 1 998 III � c- C') :l" III ., III 0 .... Cl) ., iii· III C!: 0 � Chapter 4 - Isolation and Characterisation Manual sequence analysis revealed that the proline/arginine-rich pept ides all contain a repeating motif of four residues and larger repeated sequences made up of a combination o f the smal ler four-residue mot ifs. The mot ifs and repeats are highlighted in Table 4.7. I n each case, the motif contains two pro lines, one arginine and one varying residue. The order of these residues differs between the peptides. The varying residue is usually a hydrophobic amino ac id such as I, L, F and V. The sequences were further analysed by generating hydrophobic ity and po larity p lots using the method described in Sect ion 3 . 1 . 1 1 . The plots are presented in Figure 4 .7 and F igure 4.8 . The three Bac5 peptides contained two copies of a 1 6-residue repeat each made up of four of the four-residue mot ifs. These repeats were evident in the hydrophobic ity and po larity plots of these peptides ( Figure 4.7 and F igure 4 .8) . These plots also i l lustrated that the N-terminal end of the peptides were more polar, and the C-terminal ends were more hydrophobic. � o u 15 -1 o -a. -2 e � -3 PR39 r 4 +------.------� o 20 residue 40 1 -.----------------, � 0 'u :g - 1 .s:::. g- -2 L.. '0 � -3 B� 4 +-----�------� o � o u :g -1 .s:::. g- -2 L.. '0 � -3 Bac7 2 0 40 residue 4 +----.----.---� o 20 40 residue 60 � o u :g -1 .s:::. � -2 '0 � -3 Bac4 J\ 4 +-----�------� o resf8ue 40 OaBacS 4 +------.------� � o u :g -1 .s:::. g- -2 L.. '0 � -3 o 20 40 residue OaBac7.S 4 +----.----.---� o 20 40 res idue 60 >. Oa Bac6 'E 0 /\fv 15 0 -1 .s:::. g- -2 L.. '0 �-3 4 +----.----.---� ,� 0 u :g -1 .s:::. g- -2 -0 � -3 o 20 40 res idue ChBacS 60 4 +-----�------� ,� 0 u :g -1 .s:::. g- -2 L.. '0 � -3 o 20 40 res idue Oa Bac1 1 4 +--.--.--.�,-� o 20 40 60 80 1 00 residue Figure 4.7 - Hydrophobicity plots of the prol ine/argin ine-rich cathelicidin peptides. 96 Chapter 4 - Isolation and Characterisation The hydrophobic and polarity plots also highlighted differences in the sequences of the Bac5 peptides from the different species . Bovine Bac5 only differed from OaBac5 by 5 residues ; however, two of these differing amino acids are arginine residues in the OaBac5 sequence but tyrosine and phenylalanine residues in the Bac5 sequence. This made Bac5 more hydrophobic and less cationic than OaBac5 . Bac5 had comparable activity to OaBac5 against E. coli, but unlike OaBac5 it was not active against S. aureus (Gennaro et ai, 1 989). The differences between OaBac5 and ChBac5 were less significant. These two sequences differ by only two amino ac ids. Only one of the substitutions alters the po larity; an arginine in OaBac5 is replaced with an asparagine in ChBac5 . ChBac5 has similar activity to OaBac5, showing that this substitution does not alter the activity considerably (Shamova et ai, 1 999) . However, like OaBac5a and OaBac5y, ChBac5 is not inhibited by a salt concentration of 1 00 mM NaCl. 40 ...-------------, 30 � .� 20 (5 a. 1 0 P R39 o -I--------,-------j o 20 residue 40 26 -r---------, 21 � 16 · c ro 1 1 'R 6 1 � +----.---� o 20 residue 40 40 .----------, 30 � ·c � 20 o a. 1 0 8ac7 O +------r---r-----i o 20 40 residue 60 40 �------� 30 >. ·c � 20 o a. 1 0 8ac4 o -t------.,------i o 20 residue 40 25 .---------. 20 Oa8ac5 >. 'E 1 5 ro 'R 1 0 5 o -t------.,------i o 20 residue 40 40 --r---------, 30 � ·c � 20 o a. 1 0 Oa8ac7.5 O -t------r--.,...-----i o 20 40 residue 60 40 --r----------, >. 30 =E � 20 o a. 1 0 Oa8ac6 0 +------.--.,...----1 o 20 40 residue 60 25 .------------, 20 Ch8ac5 >. =E 1 5 ro 8. 1 0 5 o +------,-----1 o 20 residue 40 40 .---------. 30 � � 20 o a. 1 0 Oa8ac1 1 o -1-�____,-__r_____r____f o 20 40 60 80 1 00 residue Figure 4.8 - Polarity plots of the proline/arginine-rich cathelicidin peptides. 97 Chapter 4 - Isolation and Characterisation Like the Bac5 peptides, OaBac6 also had two copies of a 1 6-residue repeat. However, its four-residue motif was different to that of the Bac5 peptides. Bovine Bac4 also contained the same 1 6-residue repeat as ovine OaBac6, but it had only one copy compared to OaBac6's two copies. Again, the p lots showed a po lar N-terminus, but the C-terminus was not as hydrophobic as that of the Bac5 peptides. The last three peptides, Bac7, OaBac7.5 and OaBac l l , had the same four-residue mot if as Bac4 and OaBac6, but their repeat pattern was different. The 1 4-residue repeat started with the amino acids leucine and pro line fo l lowed by three sets of the four-residue motif. Bac7 and OaBac7.5 contained three copies and OaBac l l contained s ix copies of this 1 4-residue repeat. In the case of Bac7, the repeats were perfect, whereas for the ovine pept ides the sequence differed from that of the repeat numerous t imes. For OaBac7.5 there were two residues (LP) between the second and third repeats, which made the sequence even less uniform. 4.9 PURIF ICATION OF ANTIMICROBIAL PEPTIDES USING CATIONIC EXCHANGE CHROMATOGRAPHY AND RP-HPLC Cationic exchange chromatography was used as an alternative to the gel fi ltrat ion purificat ion step. This method separated the cationic mo lecules from the neutral and anionic molecules in the crude extract using the method described in Section 3 . 1 .4. The crude extract was added to the cationic co lunm and the non-cationic mo lecules were washed through the co lunm with 25mM ammonium acetate solution. The cationic mo lecules were e luted from the column with 5% acetic ac id. A salt gradient was not used to elute the cationic mo lecules because salt is detrimental to the act iv ity of the peptides as shown in Section 7.2. A chromatograph of a typical run is shown in F igure 4.9. The plate assay method was used to test the act ivity of the fract ions. These results are summarised in Table 4 .8 . The molecules that d id not bind to the co lunm did not have ant imicrobial act ivity, whereas the cat ionic fractions that were eluted from the co lunm were active. 98 E c o CO N +-' CO Q.) () c CO ..c � o en ..c « A ! , o 1 Chapter 4 - Isolation and Characterisation F1 F4 F3 B C F2 ! ! - - 2 3 4 5 6 Time (hours ) Figure 4.9 - Ion-exchange chromatograph for the add ition of the ovine neutroph il crude extract to a weak cationic exchange column. The column was packed with Macro-Prep CM resin. At time 'A' the sample was added , at time 'B' the running buffer was changed from 2SmM ammonium chloride to 10% acetic acid and at time 'C' the running buffer was changed to 20% ethanol. Table 4.8 - Antimicrobial activity of typical ovine neutrophil extract cationic-exchange chromatography fractions against test organisms. Test solution Diameters of clearings on plate assays (mm) E. coli 01 57:07 S. aureus 1 056 MRSA C. albicans 3 1 53A posit ive contro\* 1 5 9 1 1 F l no c learing no c learing no c learing F2 no c learing no c learing no c learing F3 6 6 5 F4 6 5 4 negative control** no c learing no c learing no c learing • The positive controls were 1 J.Lg/mL polymyxin for E. coli 0157:H7, 1 J.Lg/mL nisin for S. aureus 1 056 MRSA, and 1 0J.Lg/mL nystatin for C. albicans 31 53A . .• The negative control was 0.01 % acetic acid. The two cationic fract ions, F3 and F4, were pooled together and RP-HPLC was used to separate them further. For this work a different HPLC technique was used than for the separation of the gel fi ltrat ion fract ion already described. The details of the methods used are given in Section 3 . 1 . 5 . Despite the differences in the two methods, a similar separation pattern was seen and the peak act ivit ies were comparable to those of the proline/arginine-rich peptides already iso lated. 99 Chapter 4 - Isolation and Characterisation One d ifference between the gel filtrat ion F2 and the cationjc fraction was that the components that came through in the vo id vo lume, when separat ing the pro line/arginine-rich peptides using a gradient of 25-30% acetorutrile, also had ant imicrobial activity. To separate these components a less hydrophobic gradient of 0-20% acetonitrile was used. Us ing this grad ient, 36 peaks were separated. The chromatograph of a typical run is shown in Figure 4. 1 0. 500 .------------r3 --------------------------------O=------3 - 4"3�6�� E t: Il') 400 ""'" N - ('Cl 300 Cl) (,) t: ('Cl .Q 200 '- o I/) � 1 00 30 o ��------�---------r--------�--------�----------r_------� 5 1 0 1 5 20 25 30 35 Time (minutes) Figure 4.10 - RP-HLPC chromatograph of the cationic fraction (F3 and F4) of the ovine crude extract. Buffer A contained 5% acetonitrile and 0. 1 % triflu roacetic acid (TFA). Buffer B contained 95% aceton itri le and 0.1% TFA. A grad ient of 0% to 20% of Buffer B over 30 minutes was used. The peaks with red labels had antimicrobial activity; the peaks with blue labels d id not. The p late assay method was used to test the activity of the 36 HPLC peaks c o l lected. The results are summarised in Table 4.9. Of these, rune were active against both bacterial strains (E. coli 0 1 1 1 and S. at/reus NCTC 4 1 63 ). A further four peaks were act ive only against E. coli 0 1 1 1 . None of the peaks d isplayed activity against the yeast (C albicans 3 1 5 3 A). It is unknown whether the all o f the components were inactive against yeast, o r whether t he concentrat ions tested were too low for the activity to be detected. Of the active peaks, s ix were invest igated further using mass spectroscopy. These peaks were chosen for mass spectroscopy because they had good activity against both the Gram- negative and Gram-posit ive test bacteria, and because they were we l l separated from the other peaks in the chromatograph. The determined molecular weights are g iven in Table 4.9. Most of the peaks had low mo lecular weights of 1 -2kDa. Due to l imited resources only two of the peaks, Peak 1 8 and Peak 24, were studied further using N-terminal sequenc ing. 1 00 Chapter 4 - Isolation and Characterisation Table 4.9 - Plate assay resu lts and molecular weights of antimicrobial peptides isolated from the cationic fraction of ovine neutrophil extract. Test solution Diameter of clearings on plate assays (mm) MWt (kDa)* E. coli 01 1 1 S. aureus NCTC 4163 peak 1 1 3 peak 1 2 4 peak 1 3 3 peak 1 4 666.03, 1 024.56 5 3 peak 1 5 1 024.65 4 4 peak 1 6 4 4 peak 1 7 5 4 peak 1 8 3 1 84.5 1 4 4 peak 1 9 1 65 1 .39 4 4 peak 20 4 4 peak 24 1 1 26.06 6 5 peak 32 786.56, 1 578 . 1 3 5 4 peak 35 3 * The molecular weights were obtained experimentally (see Appendix A 1 . 1 for mass spectra). Peak 1 8 was chosen for N-terminal sequencing because it was separated wel l from the other peaks and because it was the only peak of those tested that had a mass in the range expected for antimicrobial peptides (3-5kDa). The N-terrninal sequence of Peak 1 8 was similar to part of the cathelin-like domain of cathe lic idins. Peak 1 8 and SMAP29 prepropeptide are compared in Table 4. 1 0. It was previously thought that the cathelin-l ike precursor had the function of suppressing the antimicrobial activity of the cationic peptide unti l it was required, and that it may be involved in targeting and/or assist ing the fo lding of the antimicrobial peptide (Zanetti et aI, 1 995). A recent study using recombinant human cathelin protein showed that the cathelin domain was able to inhibit the activity of the protease cathepsin L (Zhao et aI, 1 995b). This study also showed that, when c leaved, the cathelin domain displayed antimicrobial activity, whereas the full cathelic idin molecule did not . These fmdings suggested that after proteolytic c leavage the cathelin domain can contribute to host defence by inhibiting bacterial growth and limiting tissue damage mediated by cysteine proteinase. The peptide isolated in this work was only a small fragment of the cathelin domain yet it still inhibited the growth ofthe test bacteria. This shows that the complete cathelin domain is not required for antimicrobial activity. 1 01 o N Ta ble 4.1 0 - Comparison of the N-termi nus of cationic Peak 1 8 to the cathel in-l ike precu rsor of SMAP29. Peptide Source peak 1 8 N-terminal purified from cationic fraction SMAP29 predicted from prepropeptide ovine cDN A Sequence Reference LSLY- EAVLYAVDT This work . . . . . . . . . . . . . . . . . . . . METQRASLSLGRRSLWLLLLGLVLASARAQALS- YREAVLRAVDQLNE Huttner et a� 1 998 KSSEANLYRLLELDPPPKQDDENSNI PKPVSFRVKETVCPRTSQQPAE QCDFKENGLLKECVGTVTLDQVGNNFDITCAEPQSVRGLRRLGRKIAH GVKKYGPTVLRI IRIAG The signal peptide is shown in red, the cathel in-like presequence in blue and the cationic antimicrobial peptide in black. Table 4.1 1 - Com parison of the N-terminus of cationic Peak 24 to the signal peptide of T -cell surface g lycoprotein CD4. Peptide Source peak 24 N-terminal purified from cationic fraction human T -cell surface glycoprotein CD4 precursor Sequence VLQLAL . . . . . . . . . . . . MNRGVPFRHLLLVLQLALLPAATQGKKVVLGKKGDTVELTCTASQKKS IQFHWKNSNQIKILGNQGSFLTKGPSKLNDRADSRRSLWDQGNFPL I I KNLKIEDSDTYICEVEDQKEEVQLLVFGLTANSDTHLLQGQSLTLTLE SPPGSSPSVQCRSPRGKNIQGGKTLSTWTCTVLQNQKKVEFKIDIVVL AFQKASSIVYKKEGEQVEFSFPLAFTVEKLTGSGELWWQAERASSSKS WITFDLKNKEVSVKRVTQDPKLQMGKKLPLHLTLPQALPQYAGSGNLT LAEVNLVVMRATQLQKNLTCEVWGPTSPKLMLSLKLENKEAKVSKREK AVWVLNPEAGMWQCLLSDSGQVLLESNI KVLPTWSTPVQPMAL IVLGG VAGLLLFIGLGIFFCVRCRHRRRQAERMSQIKRLLSEKKTCQCPHRFQ KTCSPI The signal peptide is shown in red and the T-cell surface glycoprotein CD4 is shown in black. Reference This work Maddon et ai, 1 986 Chapter 4 - Isolation and Characterisation The N-terminal sequence of Peak 24 was also determined. This peak was chosen for N­ terminal sequencing because it had the most potent activity of the peaks col lected from the cationic fraction even though it was one of the smallest peaks. The N-terminal sequence of Peak 24 matched part of the signal peptide of the T -cell surface glycoprotein CD4 precursor (Maddon et ai, 1 986). Peak 24 and human T -cell surface glycoprotein CD4 precursor are compared in Table 4. 1 1 . The fact that this fragment of the s ignal peptide displays antimicrobial activity shows that the s ignal peptide itself may also p lay a role in fighting infections. 4.1 0 OTHER PREDICTED CATHELlCIDINS Although numerous peptides were purified, this research did not fmd all seven of the cathelicid in peptides predicted from ovine DNA. Neither of the a-helical peptides, SMAP29 and SMAP34 were purified, nor was the dodecapeptide OaDode, nor one of the proline/arginine-rich peptides, OaBac6. The undetected cathelicidins may have been present in their precursor forms and therefore were inactive, or they may have been in their active forms, but at concentrations too low to be detected by the antimicrobial activity testing method used. Another possibility is that the peptides were destroyed by the extraction process or by proteases present in the crude extract. Alternatively, they may not have been present in the crude extract. This could be the case if the peptides were not constitutively expressed, and instead are expressed only under certain c ircumstances, or if they were expressed in cells other than the neutrophils. One possibi l ity is that the undetected cathe licidins were still attached to their pro sequences, which would make them inactive and, therefore, not detectable with the plate assay. Other groups have reported that it is necessary to add neutrophil elastase to the crude extract to c leave the proregion and release the active peptides (Shamova et ai, 1 999). However, for this research, the c leavage must have been carried out by neutrophil elastase naturally present in the crude extract because the extraction process did not include the addition of this enzyme. In the current work, the addit ion of human neutrophil e lastase (ART Biochemicals) did not increase the activity or show any differences in the SDS-PAGE of the crude extract, so the cathelicidin processing was complete. 1 03 Chapter 4 - Isolation and Characterisation Another possibi lity is that some of the cathe l ic d ins were destroyed during the extraction process, either by the process itself or by proteases present in the crude extract. One way to attempt to increase the number of iso lated cathelic idins wou ld be to bo il the so lut ion or to add protease inhibitors during the crude extraction process to stop peptide degradat ion. However, the neutrophil e lastase present in the sample would be inh ibited so the cathelic id ins would be purified in their comp lete, inactive forms. Therefore, the addition o f neutrophil e lastase would be needed to process the cathelic id ins and release the act ive pept ides. This was undesirable for this project because it aimed to develop ways of using the co mponents present in waste ovine blood, and the add ition of e lastase would be uneconomic in a large-scale process. 4.1 1 CONCLUSIONS The object ive of the work presented in this c hapter was to purify and ident i fy antimicrobial peptides from ovine blood. As hypothesised, the majority of the ant imicrobial act ivit y displayed by the ovine neutrophil crude extract was due to cathel icidin pept ides. Previously seven cathelicidins had been predicted fro m ovine c DNA but only two variants of one of the predicted peptides had been purified from ovine blood. I n this study numerous proline/arginine rich cathe l ic idins peptides were purified, inc lud ing OaBac5, OaBacy, OaBac7.5(32-60) and various truncates of OaBac l l . Many o f the peptides purified were in truncated forms compared to the full pept ides predicted from ovine cDNA. Peptidases capable of carrying out the c leavages were ident ified but these peptidases were not specific to the sites c leaved and could have resulted in many more truncations. Therefore, it is thought that the c leavage may have been carried out by peptidases that have not yet been characterised. Even though many of the purified peptides were shorter than those predicted they st i l l had ant imicrobial act ivit y against the three test organisms. This shows that the fu l l mo lecu les are not necessary for the antin1icrobial activity. This property may contribute to the protection o f the animal against pathogens, because even if the peptides are degraded b y proteases secreted by the pathogens, the resulting truncates may st i l l retain their antimicrobial functio n. Although many o f the predicted cathelic idins that had not previously been purified were iso lated in this study, t here were other predicted peptides that were not iso lated. It is unclear whether these pept ides were degraded during the extraction process or by pro teases in the 1 04 Chapter 4 - Isolation and Characterisation extract, or whether they were not present ill the crude extract because they are not constitutively expressed in the blood. As wel l as these cathel icidin peptides, there were a number of other active components present in the crude extract. One component was identified as a fragment of the cathelin-l ike domain of cathel ic idins. Recent ly, it was shown that the cathel in domain had antimicrobial activity (Zhao et al, 1 995b). However, the peptide isolated in this study was only a small fragment of the cathelin domain, which showed again that the ful l molecule was not necessary for bioactivity. Another of the purified peptides was possibly a fragment of T -cel l surface glycoprotein CD4 precursor. This indicates further that the signal peptides and precursors of immune-related proteins may have a secondary role as antimicrobial agents. 1 05 Chapter 5 - Membrane Interactions CHAPTER 5 SPECTRUM OF ACTIVITY AND BACTERIAL MEMBRANE INTERACTIONS OF SYNTHETIC OVINE CATHELICIDINS 5.1 INTRODUCTION The work presented in this chapter was concerned with the second object ive of this research project, which was to determine the mechanism of action o f ovine ant imicrobial pept ides. It was not practical to purify the necessary amounts of each of the indiv idual peptides from the crude extract because each HPLC run yie lded only approximately 30)..lg of each peptide and S mg of each pept ide was required for these tests. Instead, three peptides, OaBacS mini, OaBac7.Smini and SMAP29, were synthesised using the method described in Section 3 .2 . 1 . The peptide sequences are given in Table S . l . Table 5.1 - Seq uences of synthetic ovine antimicrobial peptides used for th is research. Peptide OaBacSmini (N-terminal of Pa) OaBac7. Smini (Pb) SMAP29 Sequence RFRPPIRRPPIRPPFRPPFRPPVR-NH2 RRI PRPILLPWRPPRPI PRPQPQPI PRWL-NH2 RGLRRLGRKIAHGVKKYGPTVLRI IRIA-NH2 OaBacSmini is a truncated version of OaBacS, comprising the flrst 24 N-terminal residues. OaBacS is a S I -residue, proline/arginine-rich peptide that was inferred based on a cDNA sequence (Huttner et ai, 1 998). S ince then, three variants of OaBacS, as we ll as the original ly predicted molecu le, have been iso lated from sheep neutrophils (Shamova et ai, 1 999)(Sect ion 4.4). As shown in Section 4.8, OaBacS is made up of a 6-residue N-terminus, fo llowed by two copies of a 1 6-residue repeat and a s-residue C-terminus. OaBacSmini, the truncated version of OaBacS, is made up of the six N-terminal residues, one copy of the 1 6-residue repeat, and the fIrst two residues of the second repeat. Due to the large number of pro l ine residues, the full- length molecule could not be synthesised. This truncat ion was chosen because it has been shown that the bovine peptide Bac7 retains its activity when truncated similarly (Sadler et ai, 2002). The second synthesised peptide was OaBac7 .Smini, which is a truncated form of OaBac7 .S . L ike OaBacS, OaBac7.S is also rich in pro line and arginine and was predicted from ovine cDNA ( Huttner et ai, 1 998). However, OaBac7 .S , which was predicted to be 60-residues 1 06 Chapter 5 - Membrane Interactions long, has only been isolated from ovine neutrophil as the 29 amino acid C-terminal peptide, OaBac7.5(32-60) (Section 4 .5) . Therefore, it was this purified form that was synthesised for these tests. The final peptide synthesised for these tests was SMAP29. SMAP29 is a very potent a­ hel ical peptide (Skerlavaj et ai, 1 999) that was predicted from ovine cDNA (Mahoney et ai, 1 995; Bagella et ai, 1 995), but has not yet been isolated from ovine cells. This peptide was chosen for these tests, as a comparison to the other two peptides, because it i s an a-hel ical peptide, whereas the others are proline/arginine-rich peptides. The first objective of the work presented in this chapter was to determine the spectra of activity of the three test peptides. It has been shown previously by other research groups that SMAP29 has potent broad-spectrum activity (Skerlavaj et aI, 1 999). However, the spectra of activity of OaBac5mini and OaBac7.5mini, or their parent molecules, have not been determined before. The second objective of the work presented in this chapter was to investigate the secondary structures of the peptides under different conditions. It was thought that the peptides would have random structures in aqueous so lutions and form amphipathic structures in membrane­ l ike conditions. All cationic antimicrobial peptides have a large number of hydrophobic residues that allow them to fold into structures with separate hydrophobic and cationic regIons. This enables the peptides to insert themselves into bacterial membranes, with their hydrophobic region interacting with the oily membrane interior, while their cationic region interacts with anionic phospholipid head groups (Ganz, 1 999) . The final objective of the work presented in this chapter was to investigate the basis of the synthesised peptides' antimicrobial activities by assessing the way they interact with bacteria. As reviewed in Section 2.6.2, it is believed that the steps involved in the mechanism of action of cationic antimicrobial peptides against Gram-negative bacteria are (Zasloff, 2002) : 1 ) The positively charged peptides interact with negatively charged lipopolysaccharide (LPS) at the divalent cation binding sites on the outer surface of Gram-negative bacteria. 2) The peptides form pores in the outer bacterial membrane, which increases the permeabi l ity of the membrane. 3 ) The peptides pass across the outer bacterial membrane by a process termed "self­ promoted uptake" and interact with the cytoplasmic membrane. 1 07 Chapter 5 - Membrane Interactions 4) The peptides either insert into the cytoplasmic membrane, result ing in the loss o f the proton gradient and leakage of essential molecules, or they translocate across the cytoplasmic membrane and interact with target molecules within the cytoplasm, such as DNA, RNA and proteins. Each step in thjs proposed mechanism of action is illustrated in F igure 5 . 1 . The capabilities of the peptides to carry out each of these steps were evaluated experimentally. outer cell membrane cytoPlasmi:\\ , � membrane� I peptide I F igure 5.1 - Schematic diagram showing the proposed mechanism of action of antimicrobial peptides against Gram-negative bacteria. 5.2 M IN IMUM INHIBITORY CONCENTRATIONS The minimum inhibitory concentrations (MICs) of the three test peptides were determined using the modified micro-broth dilution method described in Section 3 .2 .2 (Wu and Hancock, 1 999a). The test organisms inc luded numerous Gram-negative and Gram-posit ive bacteria as well as two yeast strains. These organisms were chosen because they are either food pathogens or are clinically significant. Five runs were carried out, each containing duplicates of each sample. In all cases, at least seven of the ten recorded values for the MIC of each peptide against each orgarusm were the same. The MIC values varied from the mode values by only a s ingle two-fo ld di lution. The variat ions were seen both between duplicates in the same run and between runs. The MICs are reported as the geometric means plus or minus the standard error in Table 5 .2 . The raw data is given in Appendix A2 . 1 and example calculations are given in Appendix A2.2 . 1 08 Table 5.2 - Minimum inhibitory concentrations (MIC) of synthetic ovine antimicrobial peptides against various microorganisms. Organism MIC (�g/mL) SMAP29 OaBac5mini OaBac7.5mini Gram-negative bacteria Escherichia coli 0 I I I 1 .74 ( 1 .45, 2 .09) 2 . 1 4 ( 1 .87, 2 .46) 1 3 .0 ( 1 0.6, 1 6.0) E. coli UB l O05 (rough K- 1 2 strain) 0. 1 3 (0. 1 2, 0. 1 5) 0. 1 3 (0 . 1 2, 0. 1 5) 6 .96 (5 .8 1 , 8 .35) E. coli DC2 (antibiotic-supersusceptible mutant) 0. 1 1 (0.09, 0. 1 3) 0. 1 2 (0. 1 0, 0. 1 3) 1 . 74 ( 1 .45, 2 .09) E. coli 0 1 57 :H7 2.00 (2.00, 2 .00) 8.00 (6 .53, 9 .80) 32 .0 (32.0, 32.0) Salmonella typhimurium 1 4028s 0.23 (0.20, 0.27) 0.47 (0.4 1 , 0 .53) 29.9 (26. 1 , 34.2) S. typhimurium MS4252S (PhoPQ mutant; defensin 0. 1 2 (0. 1 0, 0. 1 3) 0. 1 3 (0. 1 2, 0. 1 3) 1 .62 ( 1 . 32, 2 .00) supersusceptible) Pseudomonas aeruginosa P AO 1 4 .00 (4.00, 4.00) 4.29 (3 .74, 4.9 1 ) 1 8 .4 ( 1 5 .3 , 22.0) P. aeruginosa Z6 1 (antibiotic-supersusceptible mutant) 0.93 (0. 8 1 , 1 .07) 7.46 (6.52, 8 .55) 32.0 (32.0, 32 .0) Gram-positive bacteria Staphylococcus aureus NCTC 4 1 63 1 .07 (0.94, 1 .23) 29.9 (26. 1 , 34.2) 64.0 (64.0, 64.0) S. aureus MRSA R 1 47 0 .50 (0.50, 0.50) 64.0 (64.0, 64.0) 36.8 (30.7, 44. 1 ) () S. aureus 1 056 MRSA 4.00 (4.00, 4.00) 1 8 .4 ( 1 5 .3 , 22.0) >64 ::T 1\1 'C ..... CD S. epidermidis clinical isolate 0.29 (0.24, 0.34) 1 7 .2 ( 1 5 .0, 1 9.6) 32.0 (32.0, 32 .0) .., en I Enterococcus faecalis A TCC 292 1 2 2 .00 (2.00, 2 .00) 32 .0 (32.0, 32 .0) 59.7 (52 . 1 , 68 .4) 3: CD Yeast 3 er iil Candida albicans 1 05 2 . 1 4 ( 1 . 87, 2 .46) 27.9 (2302, 33 .4) 64.0 (64.0, 64.0) :::l CD 5' C. albicans 3 1 53A 4.29 (3 .74, 4.9 1 ) 1 9 .7 ( 1 6.0, 24.2) >64 ..... CD .., 1\1 n The MICs are expressed as the geometric means of five runs, each with duplicate samples. The values in the brackets are the lower and upper limits of the 95% er. -" 0 0 confidence intervals of the means. The raw data is given in Appendix A2.1 and example calculations are given in Appendix A2.2. :::l <.0 III Chapter 5 - Membrane Interactions The potent, broad-spectrum activity of SMAP29 is already well documented ( Skerlavaj et ai, 1 999; Brogden et ai, 200 1 ), but the activit ies of OaBac5mini and OaBac7.5mini have not been investigated previously. These results confIrmed that SMAP29 had potent activity against all organisms tested (MICs of 0. 1 to 4 .3 �glmL). The proline/arginine-rich peptides were generally less act ive than SMAP29. OaBac5mini had potent activity against the Gram-negative bacteria (0. 1 to 8 !J.glmL) , but comparatively weaker activity against the Gram-posit ive bacteria and the yeast Candida albicans ( 1 7 . 1 to 64 !J.glmL). As expected, the truncated OaBac5mini had similar antimicrobial activity to that previously determined for the purified full peptide in Sect ion 4.7. This result confrrms that the full molecule was not necessary to retain the antimicrobial activity. When compared to results reported for the activity of Bac5, a bovine pro line/arginine-rich peptide analogue of OaBac5, OaBac5mini was more active against both Gram-negative and Gram-posit ive bacteria (Gennaro et ai, 1 989). OaBac5 and Bac5 differ in only fIve of their 43 residues, and are 88% homologous (determined using the online EMBL-EBI C lustalW tool at http://www.ebi.ac.uklclustalw/ index.html). Two of these differing amino ac ids are arginine residues in the OaBac5 sequence, but tyrosine and phenylalanine residues in the Bac5 sequence. This makes Bac5 more hydrophobic and less cat ionic than OaBac5, which may account for the observed differences in activity. OaBac7.5mini had relatively weak activity against al l the test orgamsms except for the antibiotic super-susceptible strains E. coli DC2 (weak outer membrane) and S. typhimurium MS4252S (a phoPQ mutant), and the rough LPS strain E. coli UB 1 005 . The activity of the synthetic OaBac7 .5mini compared to the same peptide purified from ovine blood, shown in Section 4.7, was similar for E. coli 01 1 1 , but greatly reduced for S. aureus NCTC 1 63 and C. albicans 3 1 52A. This may be because different antimicrobial act ivity testing techniques were used. The activity of the complete OaBac7.5 peptide has not been investigated. However, its bovine analogue (64% homologous), Bac7, has similar act ivity against Gram-negative bacteria, but is not as active against Gram-positive bacteria, compared to OaBac7.5mini (Gennaro et ai, 1 989). 1 1 0 Chapter 5 - Membrane Interactions 5.3 CIRCULAR D ICHROISM SPECTROSCOPY The c ircular dichroism (CD) spectra were obtained for each test peptide in three different solutions using the method described in Section 3 .2 .3 . Three d ifferent solutions were used to compare the structures of the peptides in d ifferent condit ions. To represent aqueous, hydrophobic and membrane-mimicking condit ions, phosphate buffer, 50% trifluroethanol (TFE) and liposomes were used respectively. The CD spectra are given in F igure 5 .2 . A B C 2 2 2 > � :!: 0 u E a -o :.= N Q) E .... u 0 0 0 (1) Ol 'S Q) lrl -0 - 't o 0 ::! ..--� -2 -2 -2 1 90 220 250 1 90 220 250 1 90 220 250 Wavelength (nm) Wavelength (nm) Wavelength (nm) Figure 5.2 - Circu lar dichroism spectra of 25mM synthetic ovine antimicrobial peptides. (A) shows SMAP29, (B) shows OaBac5mini , and (C) shows OaBac7.5mini in the solutions 25mM phosphate buffer (-), 50% 2,2,2-trifluroethanol (-) and 1 0mM lysoPCllysoPG (-). The CD spectra of SMAP29 in aqueous buffer had a negative band at approximately 200 nm. This indicates that the structure was random. However, in 50% TFE (hydrophobic condit ions), and 1 0 mM lyso-PC/lyso-PG (anionic membrane-mimick ing conditions), SMAP29 had c lear minima at approximately 206 and 230 nm. These are characterist ics found in a-hel ices. This confirms that SMAP29 adopts an a-helical structure in these environments. These results are very similar to those previously obtained by another research group (Tack et al, 200 1 ). The structures of OaBac5mini and OaBac7.5mini were more diffIcult to assess based on the ir CD spectra. Both peptides appeared to be typical of a polypro line type I I helix under membrane mimicking condit ions with a minimum at 202 nm (Raj and Edgerton, 1 995 ; Falla et ai, 1 996). OaBac5mini had more pronounced spectra than OaBac7 .5mini. This was possibly because OaBac5mini may contain additional contributions from other turn structures, which tend to broaden the minimum that defmes the po lyproline type II he lix. Although the 1 1 1 Chapter 5 - Membrane Interactions structures of these proline/arginine-rich peptides were difficult to determine, the results show that the peptides have different conformations in each of the different solvents. The wavelength shift of the minimum may have been related to the change in so lvent polarity (water->TFE->lipid) amongst the different solutions. 5.4 LPS BINDING ASSAY The first step in the interact ion of antimicrobial peptides with Gram-negative bacteria 1S thought to be the binding of the cationic peptides to the anionic lipid A base of l ipopolysaccharides (LPS) on the surface of the outer membrane of the cells. I t has previously been shown that SMAP29 binds to LPS. SMAP29 has two sites that LPS can bind to cooperatively (Tack et ai, 200 1 ) . However, the LPS binding abilit ies of the other two test peptides have not been investigated. To determine if the test peptides were able to bind to bacterial LPS, their ability to displace dansyl polymyxin B (DPX) from LPS was examined using the method described in Section 3 .2.4. A schematic d iagram that shows the principle of the assay is given in F igure 5 . 3 . DPX is a fluorescing compound that fluoresces only in its bound state. Initially the LPS solution is saturated with DPX so that the maximum amount of DPX is bound and maX1mum fluorescence occurs. When a test peptide is added to this LPSIDPX solution, if the test peptide is able to displace some of the DPX, then the fluorescence decreases because less DPX is in its bound state. 1 ) DPX binds to LPS. 2) Peptide displaces some DPX. Bound DPX fluoresces. Unbound DPX does not fluoresce. peptide l i popolysaccharide (LPS) dansylated polymyxin B (DPX) Figure 5.3 - Schematic diagram showing the mechanism involved in the l ipopolysaccharide binding assay. A graph of the fluorescence over time for a typical LPS binding assay run for SMAP29 is shown in Figure 5 .4. The init ial fluorescence was allowed to stabilise and was recorded. Then I llg of SMAP29 was added to the 2mL vo lume to give a fmal concentration of SMAP29 1 1 2 Chapter 5 - Membrane Interactions of 0.5).!g/mL. When the SMAP29 was flfst added this caused a spike in the fluorescence because of the presence of the pipette tip in the sample. Once the fluorescence stabilised the reading was recorded and then another l /J.g aliquot of SMAP29 was added. This process was repeated six times. 60 � 40 c ID U (/) � o � 20 o '-.0.. o '- 5 1 0 1 5 20 lime (minutes) Figure 5.4 - A typical run showing the changes in the fluorescence of dansyl polymyxin B due to the addition of SMAP29. Each spike represents the addition of an al iquot of 1 J.1g of SMAP29. This increased the total SMAP29 concentration in the sample by O.5J.1g/mL with each addition. The data collected from the experimental runs were then processed and analysed. The fraction of the DPX fluorescence that was inhibited at each concentration of SMAP29 was calculated using Equation 5 . 1 . F = Fint - F n inhib F . 1111 Equation 5 . 1 where Finhib is the fraction of the fluorescence inhibited, Fint is the initial fluorescence reading before any SMAP29 was added, and Fn is fluorescence reading at concentration n. e.g. After the flfst addition of SMAP29 (Table 5 . 3 ) F . . = Ent - F n = 28 .6 - 22.0 = 0.23 lIlillb F . 28.6 lilt The results of the calculations of the fraction of fluorescence inhibited for various SMAP29 concentrations for a typical SMAP29 run are summarised in Table 5 .3 . The inverse of the SMAP29 concentration and the inverse of the fraction of fluorescence inhibited were also calculated. From this, the inverse of the peptide concentration was plotted against the inverse 1 1 3 Chapter 5 - Membrane Interactions of the fraction of fluorescence inhibited according to the Lineweaver-Burke p lot method. The graph of the SMAP29 concentrat ion versus the fraction of fluorescence inhibited and the double rec iprocal plot of these parameters are given in F igure 5 . 5 . Table 5.3 - Data col lected and calculated for the change i n dansyl polymyxin B fluorescence d ue to the addition of SMAP29 in a typical ru n. SMAP29 SMAP29 l ISMAP29 Fluorescence Fraction of l Ifraction total added concentration fluorescence (�g) (�g/mL) concentration reading inhibited inhibited 0 0 .00 28.6 0 .00 0.49 2 .02 22.0 0 .23 4. 33 2 0.99 1 . 01 1 6.4 0.43 2. 34 3 1 .48 0.67 1 5. 1 0.47 2 . 1 2 4 1 . 98 0 .51 1 3.9 0 .5 1 1 . 95 5 2.47 0 .41 1 2 .6 0 .56 1 . 79 6 2 .96 0. 34 1 2.2 0 .57 1 . 74 0 .6 - 5 ,....------------..., '0 0 .5 al � - 4 .0 -.- 2 0 .4 ..c .8 c al O 0. 3 U c c 0 al ._ U t) 0.2 (/) al ro .... .... o � • y = 1 . 5306x + 1 . 1 1 48 R2 = 0.971 8 :::J 0. 1 u:: 0 .0 o +----,--,----,----,r----i 0.0 0. 5 1 . 0 1 . 5 2 .0 2. 5 0. 0 0 .5 1 . 0 1 . 5 2 .0 2 .5 SMAP29 concentration (mg/ m L ) SMAP29 concentration-1 Figure 5.5 - Lineweaver-Burke plot for a typical ru n of the SMAP29-LPS binding assay. From the equation of the line on the double rec iprocal p lot, two parameters, I max and I so were calculated. Imax is the maximum percentage of DPX that could be d isplaced by the peptides and Iso is the concentration of peptide required to achieve ha lf the of its maximum DPX d isplacement. These factors were calcu lated using Equat ion 5.2 and Equation 5.3 respective ly. I = 1 00 max y-int Equation 5 . 2 where lmax i s the maximum percentage of DPX displaced and y-int i s the y-intercept on the Lineweaver­ Burke plot. 1 1 4 Chapter 5 - Membrane Interactions e.g. For the typical SMAP29 run shown in Table 5 . 3 and Figure 5 .5 , the equation of the line was y = 1 .5306x + 1 . 1 1 48 . Therefore: - 1 I ---50 x-int I = 1 00 = 1 00 = 89.7 % max y-int 1 . 1 1 48 Equation 5 .3 where 150 i s the concentration of peptide required to displace half the maximum percentage of DPX displaced and x-int is the x-intercept on the Lineweaver-Burke plot. e.g. For the typical SMAP29 run: . - -(y-int) _ 1 . 1 1 48 - 0 7283 x-mt - - - . gradient 1 .5306 - 1 - 1 I = - = = 1 .3 7 j.!g/mL 50 x-int -0.7283 The Imax and 1 5 0 for each of the test peptides are given in Table 5 .4. These values are the means of three runs and the values in the brackets are the limits of the 95% confidence intervals for the means. The raw data and calculations are given in Appendix A2.3 and Appendix A2 .4, respectively. Table 5.4 - The abil ity of synthetic ovine peptides to bind to E. coli l ipopolysaccharide (LPS) using the dansyl polymyxin B (DPX) d isplacement assay. Peptide SMAP29 OaBac5mini OaBac7.5mini Imax (%) 82.6 (71 .9 , 93.3) 55.3 (48 .8, 61 . 8) 55.2 (34 .5 , 75. 9) lmax is the percentage of DPX displaced relative to the total DPX bound. 150 (!-!g/mL) 0 .99 (0. 90, 1 . 08) 3 . 58 ( 1 .54, 5.62) 4 .84 (3.66, 6. 02) 150 is the concentration of peptide required to give half the maximum DPX displacement. Data are the means of three runs. The values in the brackets are the l imits of the 95% confidence intervals for the means. Of the three peptides, SMAP29 displaced the most DPX and caused 50% of its maximal displacement at the lowest concentration; therefore it had the highest affinity for LPS. For OaBac5mini and OaBac7.5mini, there were no significant differences in the mean Imax and 150 values owing to the large variation between trip licates. The calculations in Appendix 2.4 showed that p-values were 0.4966 and 0. 1 83 1 for I max and 150 respectively. However, the 1 50 of OaBac5mini was lower than that of OaBac7.5mini in each run, so the trends were consistent between runs. 1 1 5 Chapter 5 - Membrane Interactions The greater ability of S MAP29 to bind to LPS, compared to the other two peptides, may be because it contains more cationic re sidues (ten) compared to the other two peptides (eight) . However, this enhanced LPS binding d id not corre late with enhanced act ivity against Gram­ negative bacteria - these bacteria have LPS on their membranes. OaBac 5 m ini had similar activity to SM AP29 against Gram-negative bacteria, even though SMAP29 has a much higher affmity fo r LPS. OaBac5mini was less active than SMAP29 against Gram-positive bacteria, whic h lack LPS on their outer surface. This would suggest therefore that the abi l ity of the peptides to bind to LPS was not an important factor in determining the potency of their antimicrobial act ivity. 5.5 OUTER MEMBRANE PERMEABILlSATION After binding to the LPS on the outer membrane surface, it was expected that the cationic peptides would adopt amphipathic structures to adapt to the spec ific condit ions at the membrane-water interface. The insertion of a large number of positive charges at this interface would disrupt surface electrostatics. This would lead to increased outer membrane permeabi l ity and the passage of peptide mo lecules through this membrane in a process termed "se lf-promoted uptake" ( Hancock, 200 1 b) . To test whether the peptides increased the permeabi l ity of the outer membrane, their abi l ity to induce the uptake of I -N-phenylnapthylamine (NPN) by the bacterial cel ls was monitored using the method described in Section 3 . 2 .5 . A schemat ic diagram i llustrating the principle o f this method i s given i n Figure 5.6. NPN i s a small hydrophobic mo lecule that i s normally excl uded by the outer membrane of Gram- negative bacteria. When NPN partitions into the bacterial outer membrane it fluoresces. This indicates that the permeabilisation of the outer membrane has occ urred. 1 ) N P N excluded . 2) Some N P N taken u p by cel ls. N o fl uo rescence occurs. N P N i nside cel ls fl uoresces . • • peptide - • • Gram-negative bacteria • • • • Figure 5.6 - Schematic diagram showing the mechanism involved in the 1 -N-phenylnapthyl­ amine (NPN) uptake assay. 1 1 6 Chapter 5 - Membrane Interactions For each peptide concentration, the initial fluorescence reading of the E. coli UB I 005 culture mixed with NPN was recorded. This initial reading increased slightly throughout the day as the culture aged and the cell membranes became weaker. After the addition of the ant imicrobial peptide the fluorescence was allowed to stabi lise, then the reading was recorded. The data collected for a typical run with SMAP29 are given in Table 5 .5 . Table 5 .5 - Data collected and calculated for change in 1 -N-phenylnapthylamine (NPN) fluorescence due to the addition of SMAP29 for a typical run. SMAP29 Initial Final Change in NPN uptake' concentration fluorescence fluorescence {l!wmL} reading reading fluorescence (%) 0.0 22.5 22.5 0 .0 0 .0 0. 1 21 . 3 26.7 5.40 3.4 0.2 30.6 55.0 24.40 1 5. 3 0 .3 28. 1 50.2 22. 1 0 1 3. 8 0 .4 28.3 57. 2 28.90 1 8. 1 0.5 50.6 1 1 9 .8 69.20 43.3 1 .0 36.0 1 22 .3 86.30 53.9 2.0 45. 0 1 56.2 1 1 1 .20 69.5 4.0 56.4 1 84 .0 1 27.6 79.8 The percentage of NPN taken up is compared to that caused by 4IJg/mL polymyxin B. Since the readings varied between runs, the change in fluorescence due to the addition of the test peptides was compared to the change in fluorescence due to the addition of 4/-lg/mL polymyxin B using Equation 5 .4 and Equation 5 . 5 . �F pb = F pb(finai ) - F pb(111t iai ) Equation 5 .4 where Fpb(inilial) i s the background fluorescence reading before the polymyxin B was added, and Fpb(final) is the fluorescence measurement after the addition of 41lg/mL polymyxin B. e.g . For the typical run given in Table 5 .5 , the polymyxin B standard was calculated as follows: F = 22 and F . = 1 82 pb(intial) pb(flllai ) A F = F - F · = 1 82 - 22 = 1 60 fluorescence units L.l. pb pb(final ) pb(111tiai) % NPN uptake = �Fn x l00 �Fpb Equation 5 .5 where i\Fpb i s the change in fluorescence caused by the addition of 41lg/mL polymyxin B and i\Fn i s the change in fluorescence caused by the addition of the peptide at concentration n. 1 1 7 Chapter 5 - Membrane Interactions e.g. For an SMAP29 concentrat ion ofO. l llg/rnL in the typical run (Table 5 .5 ) : L1F 5 .4 % NPN uptake = __ " x 1 00 = - x 1 00 = 3 .4% L1F pb 1 60 For a typical SMAP29-NPN assay run, the percentage of NPN uptake compared to that of polymyxin B is summarised in Table 5 .5 . As expected the amount of N PN taken up by the cells increased with increasing peptide concentration. The addit ion of 4Jlg/rnL S MAP29 caused less NP to be taken up into the cells than that caused by the addit ion of 4Jlg/rnL polymyxin B. The NPN uptake caused by each peptide is summarised in Figure 5 .7. The means o f three runs and their assoc iated 95% confidence intervals are plotted. The raw data and calculations are g iven in Appendix A2 .5 . The analysis of variance, which is reported in Appendix A2.6, showed that the probabi l ity of the mean MICs for each peptide being the same and the probabi lity of the mean MIC at different concentrations being the same were both less than 0.000 1 so the results were very significantly different. There was variat ion between the runs because the stabi l ity of the culture was different on each day; however, the trends were the same for each run. 1 20 1 00 80 ...-. � e...- Q) 60 � ro .... a. :::J 40 z a.. z 20 0 o 1 2 3 4 peptide concentration ( Jlg/mL) Figure 5.7 - Uptake of 1 -N-phenylnapthylamine (NPN) by E. coli UB1 005 cells caused by synthetic ovine peptides. The test peptides were SMAP29 (-) , OaBac5mi n i (-) and OaBac7.5min i (-). The amou nt of NPN taken up is g iven as a percentage of the maximum NPN u ptake caused by 41lg/mL polymyxin B. The points show the mean values of three runs and the bars show the 95% confidence intervals for the means. 1 1 8 Chapter 5 - Membrane Interactions All three peptides made the outer membrane of E. coli UB I 005 cells more permeable to NPN . The peptide concentration that caused the half-maximal uptake ofNPN was between 0.4 and 0 .5/-lg/mL for each of the three peptides. For each peptide there was a large increase in permeabil ity caused between peptide concentrations of 0 .2 and l /-lg/mL. At peptides concentrations of 2/-lg/mL the maximum permeability was reached and there appeared to be l ittle further increase. Of the three test peptides, OaBac7 .5mini caused the highest total amount ofNPN to be taken up, which was more than the total uptake caused by the control (4/-lg/mL polymyxin B) ; whereas SMAP29 caused the least amount ofNPN to be taken up. This was unexpected since SMAP29 is a lot more active than OaBac7.5mini against this organism. SMAP29 and OaBac5mini each caused small amounts of uptake of NPN at their MICs of 0. 1 25/-lg/mL, but they required concentrations much higher than their MICs to cause a large increase in membrane permeabi lity. This shows that a large amount of membrane permeabilisation is not necessary for the mechanism of action o f these peptides. These peptides may form transient pores in the outer membrane to allow self-promoted uptake to occur, without requiring a large flux of molecules in and out of the cells to inhibit the orgamsm. In contrast, OaBac7.5mini caused substantial NPN uptake at concentrations well below its MIC of 8/-lg/mL against this organism. This means that the ability for this peptide to increase the permeabil ity of the outer membrane of Gram-negative bacteria does not correlate with its antimicrobial activity. This property of OaBac7 .5mini may be useful as it could be used in conjunction with other active molecules to increase the permeabil ity of the outer membrane to allow the uptake of the other active molecules. 5.6 CYTOPLASMIC MEMBRANE DEPOLARISATION After passing through the outer membrane of Gram-negative cells, antimicrobial peptides are then able to interact with the bacterial cytoplasmic membrane and depolarise and/or traverse this membrane (Hancock and Rozek, 2002). To assess the interaction of the peptides with the cytoplasmic membrane, the fluorescence of the dye 3 ,3-dipropylthiacarbocyanine (DiSC35) was monitored using the method described in Section 3 .2 .6 . A schematic diagram il lustrating the principle of this assay is given in F igure 5 .8 . DiSC35 is a cyanine dye that inserts into the 1 1 9 Chapter 5 - Membrane Interactions cytoplasmic membrane under the influence of the membrane potential gradient and quenches its own fl uorescence. After the addition of a permeabilis ing peptide that disrupts membrane potential, the dye is released. This causes an increase in fluorescence. 1 ) Di SC35 is bound. N o fl u orescence occu rs. 2) Mem brane depolarises and releases Di SC35. Un bound Di SC35 fl uoresces. peptide \ , Figure 5.8 - Schematic diagram showing the mechanism involved in the 3,3-dipropylthiacarbo­ cyanine (OiSC35) assay. For each peptide concentration the initial fluorescence reading of the E. coli DC2 culture mixed with DiSC35 was recorded. This initial reading increased throughout the day as the culture aged and the inner cellular membranes became weaker. After the add it ion of the ant imicrobial peptide the fluorescence was allowed to stabil ise, then the fmal reading was recorded. The data co l lected for a typical run with S MAP29 is given in Table 5 .6 Table 5 .6 - Data collected and calculated for change in 3,3-d ipropylthiacarbocyanine (OiSC35) fl uorescence due to the addition of SMAP29 for a typical run. SMAP29 Initial Final Change in DiSC35 released* concentration fluorescence fluorescence {!!g!mL} reading reading fluorescence (%) 0. 1 5. 1 6.4 1 . 3 1 . 7 0. 1 25 4.9 2 3 . 0 1 8 . 1 2 3 . 5 0 .25 6.3 46 . 2 3 9 . 9 5 1 . 8 0. 5 7.4 5 2 . 2 44.8 5 8 . 2 1 . 0 7.2 6 1 .4 54.2 70.4 2 . 0 7 . 8 5 7 . 2 49.4 64 . 2 3 . 0 8.4 60. 1 5 1 . 7 67. 1 4.0 9.6 64.4 5 4 . 8 7 1 . 2 . The percentage of DiSC35 released is compared to that caused by 41.1g/mL gramicidin S. Since the read ings varied between runs, the change in fluorescence due to the addition of the test peptides was compared to the change in fluorescence due to the add ition of 4�g/mL gramicidin S using Equation 5 . 6 and Equation 5 . 7 . Gramicidin S was chosen for the 1 20 Chapter 5 - Membrane Interactions companson SInce it IS known to cause substantial depolarisation of the cytoplasmic membrane. �F gs = F gs(finai) - F gs(intiai) Equation 5 .6 where FgS(initial) i s the background fluorescence reading before the gramicidin S was added, and FgS(final) is the fluorescence measurement after the addition of 41lg/mL gramicidin S. e.g. For the typical run given in Table 5.6, the gramicidin S standard was calculated as follows: F = 4.5 and F = 8 1 .5 gs(intial) gs(final) A F = F - F . . = 8 1 .5 - 4.5 = 77 fluorescence units L..l. gs gs(finai ) gS(lIltlai) % DiSC 5 released = �F n x l 00 3 �Fgs Equation 5 .7 where �Fgs i s the change in fluorescence caused by the addition of 41lg/mL gramicidin S and �Fn i s the change in fluorescence caused by the addition of the peptide at concentration n. e.g. For an SMAP29 concentration ofO. I J..!g/mL in the typical run (Table 5 .6) : % DiSC 5 = �F n x l 00 = .!.l x 1 00 = 1 . 7% 3 �Fgs 77 For a typical SMAP29 run, the percentage of DiSC35 released compared to that of gramicidin S is summarised in Table 5 .6 . The amount of D iSC35 released from the cells due to the addition of 4/lg/mL SMAP29 was substantially less than that released by the control 4/lg/mL gramicidin S. The results for each test peptide from the D iSC35 assay are shown in Figure 5 .9. The means of three runs and their associated 95% confidence intervals are plotted. The raw data and calculations are given in Appendix A2. 7. The analysis of variance, which is reported in Appendix A2.8, showed that the probability of the mean MICs for each peptide being the same, and the probability of the mean MIC at different concentrations being the same, were both less that 0.000 1 so the results were very significantly different. There was variation between the runs because the stabil ity of the culture was different each day; however, the trends were the same for each run. 12 1 Chapter 5 - Membrane Interactions 1 00 80 ..-.. � 0 60 ----"C ID (J) ca ID 40 ID � I.{') '" 0 20 Cl) (5 0 -20 0 1 2 3 4 peptide c o ncentration ().!g/mL) Figure 5.9 - Release of 3,3-d ipropylth iacarbocya nine (DiSC35) dye from the cytoplasm ic membrane of E. coli DC2 cells ca used by synthetic ovine peptides. The test peptides were SMAP29 (-) , OaBac5mini (-) and OaBac7.5mini (-) . The amount of DiSC35 released by the peptides is given as a percentag e of the maximum DiSC35 released by 4).!g/mL gramicidin S. The points show the mean values of th ree runs and the bars show the 95% confidence intervals for the means. For all three peptides some DiSC3S was released. This means that the peptides caused some depo larisat ion of the cytoplasmic membrane. For this to happen, the peptides must have interacted with the cytoplasmic membrane. This confirms that the peptides traversed the outer membrane, as predicted from the results of the outer membrane permeabi lisat ion assay in Section S .S . SMAP29 caused depolarisation of the cytoplasmic membrane at its MIC of 0 . 1 2S).lg/mL, and had a maximum depo larisat ion o f 83% of that of 4).lg/mL gramic idin S. There was a rapid increase in the amount of depolarisation caused by SMAP29 between the concentrat ions of 0 and O.S).lg/mL, but there was little increase at concentrat ions higher than O .S).lg/mL, which shows that concentrations higher than this have no extra effect. This depo larisat ion of the cytoplasmic membrane caused by SMAP29 at low concentrations indicates that cytoplasmic membrane disruption might be invo lved in S MAP29's mechanism of act ion. In contrast, OaBacSmini and OaBac7.Srnini caused relat ively low cytoplasmic membrane depo larisat ion compared to gramic idin S. OaBacSmini caused very l ittle depo larisat ion at its MIC of O. 1 2S).lg/mL and had a maximum depolarisation of only SO% of that of gramic idin S. 1 22 Chapter 5 - Membrane Interactions OaBac7 .5mini caused similar depolarisation at its MIC of 1 . 75/-Lg/mL compared to SMAP29 at its MIC; however its maximum depolarisation was only 40% of that of gramicidin S. These results suggest that membrane depolarisation is not the mechanism of action used by these prol ine/arginine-rich peptides. Instead it was likely that this depolarisation was an intermediate step in the process. The peptides probably passed through the cytoplasmic membrane to interact with inner cellular contents. This mechanism of action has been displayed by PR-39, a porcine proline/arginine-rich peptide, which kills bacteria by stopping protein and DNA synthesis (Boman et al, 1 993). 5.7 KILL CURVES To further compare the way the peptides interact with Gram-negative bacteria, the optical density and viable cell count of E. coli 0 1 1 1 treated with the test peptides was monitored over time. The method used is described in Section 3 .2.7. The log-phase bacteria cells were diluted to concentrations of approx imately 2 x 1 07 CFU/mL in MHB and were incubated in a water bath at 37°C. The peptides were added to the cultures at twice their MIC concentrations. This experiment was carried only out once because of the large amount of each peptide required and the limited amounts available. The effect of the peptides on the optical density and viable cell count of E. coli 0 1 1 1 over time are shown in F igure 5 . 1 0. The addition of SMAP29 to log-phase E. coli 0 1 1 1 reduced the number of bacterial cells by six-log in the first five minutes. After ten minutes no viable cells remained. The positive control, polymyxin B, also caused a rapid decrease in cell numbers, although not as quickly as SMAP29. Both of these peptides also caused the optical density to decrease slightly over time, which indicates that some cell lysis occurred. However, direct cell lysis was probably not the mechanism of action of SMAP29 because the drop in optical density was not as pronounced as it would be if this was the case (Zabucchi et al, 1 983; Yourassowsky et al, 1 985) . This is consistent with the hypothesis that the higher membrane depolarisation caused by SMAP29, compared to the prol ine/arginine-rich peptides, leads to loss of the proton gradient. This results in leakage of essential molecules and cel l death, but not necessarily complete cell lysis. 1 23 Chapter 5 - Membrane Interactions A B 0 .6 1 . E+09 1 . E+08 E 0 .5 c: 1 . E+07 0 0 0.4 E 1 . E+06 CD -co :3 1 . E+05 Q) 0 .3 (,) LL c: () 1 . E+04 co .0 0.2 ..... 1 . E+03 0 Cl) .0 1 . E+02 <{ 0. 1 1 . E+01 0.0 1 . E+00 0 30 60 90 1 20 0 30 60 90 1 20 Time (minutes) Time (minutes) Figure 5.10 - Optical density and viable cell count over time for E. coli 01 1 1 treated with synthetic ovine peptides. Graph (A) shows the optical density and (B) shows the viable cell count. The samples used were E. coli untreated ( ) and treated with 4IJ.g/mL SMAP29 (-) , 4IJ.g/mL OaBac5mini (-) , 321!g/mL OaBac7.5mini (-) and 4IJ.g/mL polymyxin B (-). I n contrast, OaBac5 mini and OaBac7.5 mini did not induce the death of the bacterial cel ls. However, the add ition o f OaBac5 mini resulted in no change to the optical density and viable cell counts, and this would suggest that this peptide stopped cell division. The cells treated with OaBac7.5mini had less of an increase in optical density and viable cell co unt compared to the untreated cells. This means that this peptide may have part ial ly inhibited the division of the cells. Therefore, these two pro line/arginine-rich peptides were not bactericidal, as was SMAP29, but they were bacteriostatic. This is consistent with the theory that they pass through the cytoplasmic membrane and interact with the inner cellular contents. 5.8 DNA BINDING T o determine whether the test peptides were able t o bind t o DNA, they were mixed with DN A at different ratios, and subjected to electrophoresis on an agarose gel (0. 75%) according to the method described in Section 3 .2 .8 . If the peptides bound to the DNA, they inhibited the migration of the DNA on the gel . I mages of the gels are given in Figure 5 . 1 1 . The lambda DNA used had been cut with the restriction enzyme Eco RI into fragments s ized 2 1 .2, 7 .4, 5 . 8 , 5 . 6, 4 .8, and 3 .5kbp. The two simi larly sized bands, 5.6 and 5 . 8kbp, showed up as one wide band. Al l three test-peptides inhibited the migration of the DNA. This was probably due to the peptides cross-linking the DNA to form larger mo lec ules that could not migrate down the 1 24 Chapter 5 - Membrane Interactions agarose gel. Of the three peptides, OaBacSmini was the most effective at inhibit ing the migration of the DNA. At a DNA:OaBacSmini ratio of 1 : 1 , most of the DNA was bound, and at a ratio of 1 :2 all the DNA was bound. OaBac7 .Smini and SMAP29 required ratios of 1 :4 to bind all of the DNA. DNA:SMAP29 DNA:OaBac5min i DNA:OaBac7 .5min i Lt) Lt) Lt) o N Lt) c:i c:i ...... N � '"'! Lt) 0 0 c:i ...... N . . N Lt) 0 c:i 0 ...... N ':'f: ...... ...... ...... Figure 5.1 1 - DNA gel showing the running pattern of d ifferent ratios of DNA and synthetic ovine peptides. These results indicate that the mechanism of action of the proline/arginine-rich peptides may involve interaction with bacterial DNA. OaBacSmini had a higher affinity for DNA than OaBac7 .Smini, which may account for its lower MIC values shown in Section S .2 . OaBacSmini also had a higher affinity for DNA compared to SMAP29, even though SMAP29 is more cationic than OaBacSmini. This may be because the structure of OaBacSmini allows it to interact more readily with the DNA. These results show that as well as depo larizing the cytoplasmic membrane of bacterial cells, SMAP29 can also interact with DNA. Previous studies have shown conflicting results for the mechanism used by SMAP29. Scanning electron microscopy images of E. coli and S. aureus cells treated with SMAP29 showed cells with numerous blebs and cell debris, which is consistent with the membrane interaction model (Skerlavaj et ai, 1 999). Similarly, confocal fluorescence microscopy showed that SMAP29 accumulated on the plasma membrane of fungal cells, which also suggests this i s the target of the peptide (Lee e t ai, 2002) . I n contrast, immuno-electron microscopy was used to show that SMAP29 rapidly penetrated both the outer and inner membranes of P. aeruginosa and accumulated in the bacterial cytoplasm 125 Chapter 5 - Membrane Interactions (Kalfa et ai, 200 1 ) . It implies that the mechanism of action used by SMAP29 changes depending on the organism and/or the conditions. 5.9 CONCLUSIONS This frrst objective of the work presented in this chapter was to determine the spectra of activity of the three test peptides. The MIC tests confirmed that SMAP29 has potent ant imicrobial activity against Gram-negative and Gram-positive bacteria and yeast. The proline/arginine-rich peptides were less active than SMAP29. OaBac5mini displayed potent activity, similar to SMAP29, against the Gram-negative bacteria, but it had only moderate activity against the Gram-positive bacteria and yeast. OaBac7.5mini was only moderately active against all the organisms tested, except the supersensitive mutants. However, both the ovine-derived Bac peptides were more active than their bovine analogs, Bac5 and Bac7 (Gennaro et ai, 1 989). The second objective of the work presented in this chapter was to investigate the secondary structures of the peptides under different conditions. The c ircular dichroism studies showed that all three peptides had random structures in aqueous solutions, but in membrane-l ike condit ions they formed amphipathic structures. In these conditions SMAP29 was a-helical as expected, and the proline/arginine-rich peptides, like their bovine counterparts, probably adopted a polyproline type 1 1 extended helix structure. The fmal objective of the work presented in this chapter was to investigate the basis of the synthesised peptides ' antimicrobial activities by assessing the way they interact with bacteria. The Gram-negative bacteria mechanism of action studies showed that all three peptides were able to bind to LPS, increase the permeability of the outer membrane and pass across this membrane to interact with the cytoplasmic membrane. SMAP29 caused substantial depolarisation of the cytoplasmic membrane, which indicates that the disruption of this membrane is involved in its mechanism of action. This was confirmed by the kill curve results, which showed that SMAP29 caused a large drop in viable cells and a sl ight decrease in optical density, which demonstrated that cell death had occurred. In contrast, OaBac5min i and OaBac7.5mini caused less cytoplasmic depolarisation and did not cause a drop in the viable cell count. Instead it was thought that these peptides passed across the cytoplasmic membrane and interacted with the inner cellular contents. I t was shown that these peptides 1 26 Chapter 5 - Membrane Interactions were able to bind to DNA, so this may be their target. However, SMAP29 was also able to bind to DNA, which means it may have more than one mechanism of action. These experiments indicate that SMAP29 has a different mechanism of act ion to OaBac5mini and OaBac7.5mini. This means that the innate immune system of sheep contains peptides that use at least two different mechanisms of action ( inner membrane depolarisation and interaction with cytoplasmic contents) to fight an infection. The use of pep tides with different mechanisms may be advantageous to the host animal because if an invading organism is resistant to one mechanism it may still be susceptible to the other, thus providing the animal with a backup mechanism for dealing with invading organisms. The use of peptides with different mechanisms would also make it more difficult for the microorganisms to build up resistance. This also means that if a mixture of antimicrobial peptides is i so lated from ovine blood and used in a commercial product (chilled-meat biopreservative or topical antiseptic cream), peptides uti lising different mechanisms should be present, providing "insurance" for the product's effectiveness against a wide range of pathogens. 1 27 Chapter 6 - Morphological Changes CHAPTER 6 MORPHOLOGY OF BACTERIAL CELLS TREATED WITH SYNTHETIC OVINE CATHELICIDINS 6.1 INTRODUCTION The work presented in this chapter was concerned with the third objective of this research project, which was to investigate the morphological changes to microbial cells induced by ovine antimicrobial peptides. The previous chapter indicated that the prol ine/arginine peptides, OaBac5mini and OaBac7.5mini, had a different mechanism of action to that of a­ helical SMAP29. The aim of the research presented in this chapter was to investigate whether the test peptides induced changes in the bacterial cell morphology and whether these changes confIrmed the mechanisms of action proposed in the previous chapter (Chapter 5) . The results presented in Chapter 5 indicated that SMAP29 caused rapid cell death by depolarising the cytoplasmic membrane, which resulted in leakage o f the cellular contents, and some complete cell lysis. I n contrast, OaBac5mini and OaBac7 .5mini appeared to inhibit cell division by interacting with the inner cellular contents. It was hypothesised, therefore, that SMAP29 would cause the morphology of the cells to change considerably, whereas the proline/arginine-rich peptides would cause little or no change. Of the two prol ine/arginine­ rich peptides, only OaBac5mini was used for these tests because OaBac5mini and OaBac7 .5mini appeared to act on cells in a similar manner. OaBac5mini was chosen because it was the more active of the two peptides. To get a complete view of the morphology of the peptide-treated cells two techniques, transmission e lectron rnicroscopy (TEM) and atomic force rnicroscopy (AFM), were used. As discussed in Section 2 .7 . 1 , TEM gives an electron density image o f cell cross-sections; whereas, AFM gives topological images of the cell surface. Together these techniques should produce a good overall model of the cell morphology, which can be used to confirm the proposed mechanisms of action of the respective peptides. 6.2 E. COLI TEM RESULTS E. coli 0 1 1 1 cells were untreated, or treated with SMAP29 or OaBac5mini for one hour. The method used is described in Section 3 .2 . 1 . The samples were fIxed and cut into thin sections. 1 28 Chapter 6 - Morphological Changes For each treatment duplicate samples were prepared and numerous sections (approximately ten) were cut from each. The sections were examined using TEM and multiple images were taken of each section. The electron density images given in F igure 6. 1 show typical results . The images of the samples treated with SMAP29 were different to those of the untreated cells. The untreated cells were all uniformly shaped, with intact cell walls; whereas many of the SMAP29-treated cells had their cell walls missing. In Figure 6. 1 the green arrows point to cells without cell wal ls, called ghost cells, and the yellow arrows point to cell walls that had separated from the cells . The images show that in many cases the cytoplasmic contents were leaking out of the cells. This is consistent with the results of the previous experiments in Chapter 5 that showed that SMAP29 disrupted the membranes of Gram-negative bacteria. The images of the samples treated with OaBac5mini were similar to those of the untreated cells. However, the OaBac5mini-treated cells often had areas where the cell wall was not attached to the cytoplasmic membrane as shown by the red arrows in Figure 6 . 1 . This is a typical sign that outer membrane permeabi lisation has occurred. None of the OaBac5mini­ treated cells appeared to be leaking their inner cellular contents, so inner membrane disruption had not occurred. This is further evidence that OaBac5mini has a different mechanism o f action t o SMAP29. 6.3 S. AUREUS TEM RESULTS As with the E . coli cells, S. aureus 4 1 63 NCTC cells were a lso untreated, or treated with SMAP29 or OaBac5mini for one hour. After treatment the samples were fixed, cut into sections and examined using TEM as described in Section 3 .3 . 1 . The electron density images given in F igure 6.2 show typical results. The SMAP29-treated S. aureus cells were notably different to the untreated cells. The visible differences were similar to those seen in the SMAP29-treated E. coli culture. There were numerous ghost cells without cell walls, which are shown by the green arrows. Some intact parts of cell walls were also visible as shown by the yellow arrows. These results indicated that SMAP29 acted on Gram-pos it ive cells similarly to the way it acted on Gram-negative cells. 1 29 Chapter 6 - Morphological Changes E. coli 01 1 1 untreated E. coli 0 1 1 1 treated with 4llg/mL SMAP29 E. coli 0 1 1 1 treated with 4/-lg/mL OaBac5mini Figure 6 .1 - Transmission electron microscope images taken of E. coli 01 1 1 cells treated with SMAP29 and OaBac5mini for one hour. The yellow arrows point to cell wal ls that have separated from the cytoplasm, the green arrows point to ghost cells that are not surrounded by a cell wal l , and the red arrows point to cells where the cell wal l is partial ly separated from the rest of the cel l . 1 30 s. aureus 4 1 6 3 NCTC u ntreated S. aureus 4 1 63 NCTC t reated with 4j..lg/m L SMAP29 S . aureus 4 1 63 NCTC treated with 64j..lg/m L Oa 8ac5 m i n i Chapter 6 - Morphological Changes Figure 6.2 - Transmission electron microscope images taken of S. aureus 41 63 NCTC cel ls treated with SMAP29 and OaBac5mini for one hour. The yel low arrows point to cel l walls that have separated from the cytoplasm, the g reen arrows point to ghost cel ls that are not surrounded by a cell wal l and the red arrows point to the septa in the divid ing cells. 1 31 Chapter 6 - Morphological Changes In contrast, the OaBac5mini-treated S. aureus cells did not appear to be damaged as much as were the E. coli cells treated with this peptide. There were no areas where the cel l wal l had separated from the cytoplasmic membrane, but some cell wal l debris was present as shown by the yellow arrows. Compared to the control cells, the OaBac5mini treated culture contained a large number of cells with septa, which are indicated by the red arrows. This may be because the peptide stopped the divis ion of the cells at this po int; whereas, the control cells cont inued to divide normally. This is consistent with the theory that the proline/arginine-rich peptides inhibit cell d ivision as indicated in Chapter 5 . 6.4 S . AUREUS AFM METHOD DEVELOPMENT Before AFM could be used to image the surface of bacterial cel ls, a procedure for preparing the bacterial cells for imaging by AFM had to be developed, because the applicat ion of AFM to image cel ls i s relatively new. As d iscussed in Section 2 . 7 . 1 of the literature review, AFM was an attractive method because, unlike other methods like SEM, it can be used to view both dry cells and also cells in a liquid medium. This latter feature was particularly appealing as it meant that images could be taken in real t ime to fo llow the changes occurring to the cel l morphology when treated with the peptides. Numerous difficulties, notably immobil ization o f cells, were encountered when measuring changes to the morphology o f dried bacteria, so the technique was never employed with wet samples. Three techniques for immobilising bacterial cells were investigated. The first looked at the use of filter membranes, the second the use of agar gel surfaces and the third the use of a glass slide. The filter membrane method was tried first because this was the most common method reported in the l iterature for the immobil isation of cells for AFM imaging ( Kasas and I kai, 1 995) . Unfortunately this method did not work wel l for this appl icat ion. The membranes used to immobil ise the cells were polycarbonate membranes, which are very smooth. Images of S. aureus 4 1 63 NCTC cells trapped on a po lycarbonate membrane are shown in Figure 6 .3 . The left image i s the height image. In this image the darker colours show areas that were lower and the l ighter colours show areas that were higher. Therefore, the dark areas are the pores in the membrane. Three of the pores contained visible cel ls as shown by the arrows, so these areas are l ighter. As an alternative way to display the same information, the right image shows the deflection. In this image the dark areas represent places where the height of the sample was decreasing and the light areas represent places where the height of the sample was increasing as it was scanned by the AFM t ip from left to right. 1 32 Chapter 6 - Morphological Changes 1 0 1 0 8 8 6 400 6 400 E 200 E 200 :i :i 4 0 E 4 0 E c c 2 -200 2 -200 -400 -400 0 0 0 2 4 6 8 10 0 2 4 6 8 1 0 Ilm Ilm Figure 6.3 - AFM images of S. aureus NCTC 4163 cells trapped on a polycarbonate membrane. The image on the left is the height image and that on the right is the deflection image. The arrows point to membrane pores which contain cells. 5 4 3 E :i 2 o o 2 3 Ilm 600 400 200 ... c c 0 -200 -400 4 5 5 4 3 1 20 E :i 1 00 2 E c 80 20 0 0 2 3 4 5 Ilm Figure 6.4 - AFM images of S. aureus NCTC 4163 cel ls grouped together on a polycarbonate membrane. The image on the left is the height image and that on the right is the deflection image. 11lm Figure 6.5 - AFM 3D representation of S. aureus NCTC 4163 cells grouped together on a polycarbonate membrane. 1 33 Chapter 6 - Morphological Changes One problem with the use o f the polycarbonate membranes was that the cells often formed bunches on top of the membrane instead of being entrapped in the pores as shown in F igure 6.4 and Figure 6 .5 . This made it difficult to get complete images because the cells were easily detached from the membrane during imaging . Problems also arose with the ant imicrobial treated cells. Few cells were present on the membranes if they had been treated with the ant in1icrobial peptides prior to membrane entrapment. This was probably because many of the cel ls were disrupted. The weakened cells and cell fragments may have passed through the pores in the membrane instead of being entrapped. This made it difficult to locate treated cells for imaging. When the cells were treated with an antinUcrobial peptide, there were also problems capturing images of the few cells that were trapped in the membrane because they often disappeared during in1aging. For example, in a test carried out using commerc ially-produced nisin as the test peptide, init ially two cells were trapped in the membrane pores but on the second pass only one of the cells was left. These images are shown in Figure 6 .6 . This was probably because the cells were weakened by the peptide, so they were easily pushed through the pores by the AFM tip. As an alternat ive to the membrane entrapment method, another method, which in1mo bil ised the cells on a gel surface, was attempted (Goldman et ai, 1 997). However, this method was not successful because the AFM tip cut into the agar instead of scanning its surface. I f this method is to be used in the future, more work is needed to determine the correct way to prepare the agar surface, and to immobil ise the cells on top of the agar. The fmal technique tested, which invo lved drying the cells on a glass sl ide (Braga and Ricci, 1 998), was the most successful. For this method log-phase cel ls were co l lected by centrifugation, then washed and resuspended in water. A drop of the cel l suspension was put onto a glass slide and allowed to dry at room temperature. This technique gave c lear images of the cells against a smooth background. 1 34 Chapter 6 - Morphological Changes 2 2 1 .5 1 .5 400 40 E 200 E 1 20 ::i. ::i. 0 E 0 E c c 0.5 -200 0.5 -20 -400 -40 0 0 0 0 .5 1 .5 2 0 0.5 1 .5 2 j..lm j..lm 2 2 1 . 5 1 .5 400 40 E I 200 E I 20 ::i. ::i. 0 E 0 E c c 0.5 -200 0.5 -20 -400 -40 0 0 0 0 .5 l . 5 2 0 0.5 1 l .5 2 j..lm j..lm Figure 6.6 - AFM images of S. aureus NCTC 4163 treated with 251lg/mL nisin. Initially two cel ls were present (top images) but after a few minutes only one cell remained (bottom images). The images on the left are the height images and those on the right are the deflection images. 6.5 S. AUREUS AFM RESULTS The glass sl ide technique described above was used to compare the surface topology of untreated S. aureus NCTC 4 1 63 cells with those that were treated with SMAP29 or OaBac5mini. The details of the method are given in Section 3 .3 .2 . For each treatment method two slides were prepared and numerous sections of each sl ide were chosen randomly for imaging. The results are shown as both an overview of numerous cells in Figure 6 .7 (30j..lm x 30j..lm) and as a close up of one or two cells in Figure 6 . 8 (2j..lm x 2j..lm). These images are typical of those collected. The apparent sloping of the left hand side of the cells in Figure 6.8 is an artefact of the method due to contact with the side rather than the point of the tip. 1 35 Chapter 6 - Morphological Changes S. aureus NCTC 41 63 untreated 30 30 25 25 20 1 00 20 200 E 1 5 50 E 1 5 1 00 ::i. ::i. 1 0 0 E 1 0 0 E c c 5 -50 5 - 1 00 0 - 1 00 0 -200 0 5 1 0 1 5 20 25 30 0 5 1 0 1 5 20 25 30 !-lm !-lm S. aureus NCTC 4 1 63 treated with 4!-lgimL SMAP29 20 20 1 5 1 5 1 00 200 E 1 0 50 E 1 0 1 00 ::1. ::1. 0 E 0 E c c 5 5 -50 - 1 00 0 - 1 00 0 -200 0 5 1 0 1 5 20 0 5 1 0 1 5 20 !-lm Ilm S. aure us NCTC 4 1 63 treated with 64!-lgimL OaBac5mini 30 30 25 25 20 1 00 20 200 E 1 5 50 E 1 5 1 00 ::i. ::i. 1 0 0 E 1 0 0 E c c 5 -50 5 - 1 00 0 - 1 00 0 -200 0 5 1 0 1 5 20 25 30 0 5 1 0 1 5 20 25 30 !-lm !-lm Figure 6.7 - Far away AFM images of S . aureus NCTC 41 63 cells o n a glass slide treated with SMAP29 and OaBacSmini for 30 min utes. The images on the left are the heig ht images and those on the rig ht are the deflection images. 1 36 Chapter 6 - Morphological Changes S. aureus NCTC 4 1 63 u ntreated 2 2 1 . 5 1 00 1 . 5 50 E 1 50 E 1 25 ::i. ::i. 0 E 0 E t: t: 0 . 5 0 . 5 -50 -2 5 0 - 1 00 0 -50 0 0.5 1 . 5 2 0 0 . 5 1 . 5 2 Il m Ilm S. aureus NCTC 4 1 63 treated with 4llg/ m L SMAP29 2 2 1 . 5 1 00 1 . 5 50 E 1 50 E 1 25 ::i. :::i 0 E 0 E t: t: 0 . 5 0 . 5 -50 -25 0 - 1 00 0 -so 0 0.5 1 . 5 2 0 0 . 5 1 . 5 2 Il m Ilm S. aureus NCTC 4 1 63 treated with 641lg/ m L OaBac5 m i n i 2 2 1 . 5 1 00 1 . 5 50 E 1 50 E I 25 ::i. :::i 0 E 0 E t: t: 0 . 5 0 . 5 -50 -2 5 0 - 1 00 0 -SO 0 0 . 5 1 . 5 2 0 0 . 5 1 . 5 2 Il m Il m F igure 6.8 - Close up AFM images of S . aureus NCTC 4163 cells on a g lass slide treated with SMAP29 and OaBac5mini for 30 minutes. The images on the left are the height images and those on the right are the deflection images. 1 37 Chapter 6 - Morphological Changes The AFM images of the untreated cells showed that the sl ides were covered with a thin layer o f unknown material. This material could be so lids from the growth media; however, the cells were washed before drying. Directly around the bacterial cells were zones where this material was not present. This can be seen in Figure 6 .8 . The material on the edge of the c lear zones was different to that which covered the rest o f the slide. It was in c lumps that were higher than the thin fi lm covering. These rings of built-up material around the cel ls can be seen in Figure 6.7. It appears that the cells repel led the media so l ids during the drying process but the reason why this occurred is not known. The AFM images of the cells treated with OaBac5rnini were similar to those of the untreated cel ls. The c learings d irect ly around the cells can be seen in Figure 6 .8 and the c lumpy material in rings were both present in F igure 6.7. This is consistent with the idea that the mechanism of action of OaBac5 mini does not disrupt the membranes of the bacterial cells and cause the cellular contents to leak out, thus supporting the results described in Chapter 5 In contrast, the AFM images o f the cells treated with SMAP29 were different to those of the cells that were untreated or treated with OaBac5 mini. When areas were randomly selected for imaging, for the sl ides of the untreated and the OaBac5 mini-treated samples, each image contained numerous cells; whereas for the SMAP29-treated samples, each image only contained a few cells, as seen in Figure 6 .7, and somet imes no cells. SMAP29 may have weakened the ce l l walls and caused the cells to be leaky. This may have made them prone to lysis during the drying process. Unl ike the other samples, which only had built up material in rings around the cel ls, the SMAP29-treated sample contained a number of piles of material. This can be seen in F igure 6 .7 and Figure 6.8 . This material was probably cell debris from lysis of the cells induced by the peptide and the drying process. This is consistent with the proposed mechanism of act ion for SMAP29 in Chapter 5, which indicates that this peptide caused disruption of the cel l membranes. For the SMAP29-treated samples, the unknown material coating the slides came right up to the edges of the cells that were st ill intact; whereas, there were clear rings around cells treated with OaBac5mini and also with the contro ls (Figure 6 .8) . This implies that the SMAP29- treated cells were different to the untreated and OaBac5mini-treated cells; however, the reasons for this are not known. 1 38 Chapter 6 - Morphological Changes 6.6 E. COLI AFM METHOD DEVELOPMENT PROBLEMS A considerable amount of t ime was spent trying to get good AFM images of E. coli cells; however, the AFM imaging of E. coli was not successful due to the nature of this bacterium. The method used to image S. aureus cells on a glass s l ide required the cells to be washed and resuspended in distil led water, but the E. coli cells were unable to withstand the osmotic pressure when suspended in water. E. coli is a Gram-negative bacterium so its cell wall is weaker than that of the Gram-positive S. aureus cells because of the difference in the thickness of the peptidoglycan layer. This resulted in the E. coli cells bursting as shown in F igure 6 .9 . 1 0 1 0 8 8 6 40 6 40 E 20 E 20 ::i. ::i. 4 o § 4 E 0 c 2 -20 2 -20 -40 -40 0 0 0 2 4 6 8 1 0 0 2 4 6 8 1 0 �m �m Figure 6.9 - AFM images of E. coli 011 1 debris after being suspended in d istilled water. The images on the left are the height images and those on the right are the deflection images. To try to overcome this lysis problem, the E. coli culture was resuspended in MHB instead of water. This was not successful because the cells were buried under a layer of dried media solids as shown in Figure 6. 1 0 . From these images it is apparent that the E. coli cells had a much rougher texture than the S. aureus cells. This is also probably due to the differences in the cell wall composition of the two bacteria. As another alternative to attempt to image the E. coli cells, the culture was re suspended in phosphate buffer. The E. coli cells were stable in this solution because the salts reduced the osmotic pressure and stopped cell lysis. It was hoped that the cells would be more visible, than they were in MHB, because the phosphate buffer did not contain numerous large molecules as did the medium. However, when dried, this buffer formed crystals on the sl ide and covered the cells as shown in Figure 6 . 1 1 . 1 39 Chapter 6 - Morphological Changes 1 5 E l O � E � 5 o 5 4 3 2 o o o 5 1 0 1 5 /lITI 2 3 4 /lITI 300 200 1 00 0 E c:: - 1 00 -200 -300 -400 20 300 200 1 00 0 E c:: - 1 00 -200 -300 -400 5 20 1 5 40 E I O � 20 0 E c:: 5 -20 -40 0 0 5 1 0 1 5 20 /lITI 5 4 3 40 E 20 � 2 E 0 c:: -20 -40 0 0 2 3 4 5 /lm Figure 6.1 0 - AFM images of E. coli 01 1 1 cells covered with dried Mueller-Hi nton broth . The images on the left are the height images and those on the right are the deflection images. 1 0 1 0 8 8 300 1 50 6 200 6 1 00 E E � 1 00 � 50 4 0 E 4 0 E c c 2 - 1 00 2 -50 -200 - 1 00 0 -300 0 - 1 50 0 2 4 6 8 1 0 0 2 4 6 8 10 /lm /lm F igure 6.1 1 - AFM images of crystals that formed when E. coli 01 1 1 was suspended in phosph ate buffer and d ried on a glass slide. The images on the left are the height images and those on the right are the deflection images. 1 40 Chapter 6 - Morphological Changes To successfully image the Gram-negative bacteria for the purposes of examining their cell morphology a different method to the glass slide method is required. One possibil ity is the membrane filter method because the media would pass through the pores in the membrane filter. However, E. coli cells are rod shaped so they would not easily be entrapped in the round pores. Another possibility is the gel entrapment method. This is probably the most suitable method because the cells would be stable in the agar, but more work is required to develop this technique. 6.7 CONCLUSIONS The objective of the work presented in this chapter was to investigate the morphological changes to microbial cells induced by ovine antimicrobial peptides. As predicted, the two microscopy techniques showed that SMAP29 caused significant morphological changes in E. coli and S. aureus cells. I n the TEM images of both E. coli and S. aureus cells treated with SMAP29, a large number of cells were missing their cell walls and some of these appeared to be leaking out their cytoplasmic contents. In the AFM images of S. aureus cells treated with SMAP29, there were a lot fewer intact cells than for the control and there was material present that was probably cell debris. These results support the earlier proposed theory that SMAP29 disrupted cell membranes of Gram-negative bacteria, which then weakened the cells and led to leakage of the inner cellular contents (Chapter 5). These results showed that SMAP29 also acted on Gram-positive bacteria in the same was that it acted on Gram-negative bacteria. In contrast, the two microscopy techniques showed that OaBac5mini caused only mmor morphological changes to occur in the E. coli and S. aure-us cells. In the TEM images of the E. coli culture, some of the cells had sections where the cell wall was separated from the rest of the cell, but otherwise the cells appeared to be intact , i .e . , the partial loss of the membrane did not weaken the cells in any way and as a consequence no lysis occurred. In the TEM images of the S. aureus culture, there were a lot more cells with septa compared to the contro l. This may have been because OaBac5mini stopped cell division as predicted from the mechanism of action study in Chapter 5 . The AFM images of the S. aureus culture showed no obvious differences between the OaBac5mini-treated cells and the untreated control cells. These results confirm, as determined earlier, that unlike SMAP29, OaBac5mini does not d isrupt the cell membrane and induce cell lysis. 1 41 Chapter 6 - Morphological Changes TEM provided informative images of the cross-section of bacterial cells; however, AFM did not give as much information about the surface of bacterial cells as was hoped. AFM was technically difficu lt so a lot of time was spent developing the method. Eventually, dry S. aureus ce lls were successfully imaged but E. coli cells were not. It was hoped that AFM could be used to image l ive cells in liquid so that the changes induced by the peptide could be imaged in real time; however, this was not possible. I nstead of AFM, it may be better to use another technique, such as SEM for future work. 142 Chapter 7 - Effect of Conditions CHAPTER 7 FACTORS AFFECTING THE ANTIMICROBIAL ACTIVITY OF SYNTHETIC OVINE CATHELICIDINS AGAINST E. COLI 01 57 :H7 7 .1 INTRODUCTION The work presented in this chapter was concerned with the fourth objective of this research project, which was to determine the effect of different environmental factors on the activity of ovine antimicrobial peptides. In the previous chapters the composition of the antimicrobial peptides iso lated from ovine blood neutrophils and the mechanism of action of the synthetic ovine peptides were described . This work investigated the antimicrobial activity of the peptides in a variety of conditions, to determine what applications the peptides may be suitable for. In conjunction with the separation of the major plasma proteins (serum albumins, transferrins, antibodies etc), the antimicrobial peptides extracted from ovine blood have potential to be utilised as, or in, high-value products as discussed in Section 2 .4 .2 . For example, a mixture of peptides could be appl ied as a topical cream for cuts and grazes, or used as a biopreservative for chilled lamb products. The aim of the work reported in this chapter was to gather information about factors that may enhance or inhibit the activity o f the test peptides. The bactericidal activities of these peptides were compared under various conditions, including the addition of salt or metal ions to the media, altering the pH of the media, and heating the peptides before testing. Each of the factors was examined independently to see whether they had a significant effect on the antimicrobial activity of the peptides. Studies were also carried out to determine if there was synergistic activity between the test peptides, or between the peptides and common antibiotics. Due to the limited amount of synthesised peptides available, only one orgamsm, E. coli 0 1 57 :H7, was used for these tests. This Gram-negative bacterium was chosen because it is a dangerous food-pathogen that has been known to contaminate meat products (Garcia-Olmedo et al, 1 998). Normal E. coli inhabits the intestines of all animals, where it suppresses the growth of harmful bacteria and synthesises vitamins. E. coli 0 1 57:H7 is a pathogenic variety of E. coli that produces large quantities of one or more potent toxins that cause severe damage 143 Chapter 7 - Effect of Conditions to the lining of the intestine. The symptoms of the il lness include severe stomach cramps and watery or bloody diarrhoea. I f the antimicrobial peptides are to be used as an effective meat biopreservative they would need to be active against this organism. 7 .2 EFFECT OF SALT The fust factor that was invest igated to see if it affected the activity of the test peptides was sodium chloride concentration. It is well documented that defensins are salt-sensitive (Evans and Harmon, 1 995). However, some cathelicidins, including SMAP29, appear to be resistant to high salt concentrations (Travis et ai, 2000). The salt-sensit ivity of the three test peptides was established by determining the MIC of each peptide against E. coli 01 57 :H7 in a variety of NaCI concentrations from 0 to 250mM as described in Section 3 .4. 1 . The results are graphed in Figure 7. 1 . The raw data are given in Appendix A3 . 1 , example calculations are given in Appendix A3 . 5 and the statistical analysis is given in Appendix A3 .6 . The analysis of variance determined that there were significant d ifferences between the mean MICs of each peptide (p-value = 0 .00 1 9) , and that there were significant differences between the mean MICs at different salt concentrations (p-value = 0.0024). 70 �--------------------------------------------� _ SMAP29 60 _ OaBac5min i _ OaBac7 .5mini :::J 50 E -- -3 40 u � Cl) "0 30 :.;:::; a. Cl) a. 20 1 0 0 0 50 1 00 250 salt concentration (mM) Figure 7.1 - The effect of salt concentration on the minimum inhibitory concentration (MIC) of synthetic ovine peptides against E. coli 0157:H7. The bars show the geometric means of three runs of duplicate samples and the error bars show the 95% confidence intervals for these means. 1 44 Chapter 7 - Effect of Conditions As expected the MICs for each peptide increased with increasing ionic strength . However, SMAP29 had only a two-fold increase in MIC from 0 to 1 00mM NaCI so its activity was relatively stable in saline conditions. This agrees with previous results, which showed that SMAP29 is active at high salt concentrations (Travis et ai, 2000; Shin et ai, 200 1 a; Shin et ai, 200 1 b) . OaBac5mini and OaBac7.5mini were affected more than SMAP29 by the increased ionic strength. Both peptides had MIC values 1 6-fo ld higher at 1 00mM NaCI, than in the absence of salt. Another study showed that a purified variant of OaBac5 had decreased activity in the presence of 1 00mM NaCI, which is consistent with these results (Shamova et ai, 1 999) . The decrease in activity is probably due to either the Na+ ions competively binding to the bacterial LPS or the c r ions binding to the peptide. Both cases would decrease the binding of the peptide to the outer membrane of the bacteria. If the peptides were to be effective preservatives for chilled lamb products they would need to be active at the salt concentrations present in the meat. The concentration of salt in average, trimmed, raw lamb meat is 2mmolesl l OOg (Chan et ai, 1 995) . For comparison, the 1 00mM solution tested is equivalent to 1 Ommolesll OOg ( l OOmM = 1 00mmoles/L = 1 00mmoles/kg = 1 0mmolesI l 00g) . This means that the peptides are unlikely to be inhibited by the salt present in the meat . 7.3 EFFECT OF METAL IONS The second factor to be investigated for its effect on the antimicrobial activity of the test peptides was metal ion concentration. Other studies have shown that the activity of cationic antimicrobial peptides is decreased even at low divalent cation concentrations (Selsted et ai, 1 985 ; Turner et ai, 1 998). It is believed that the first step in the mechanism of action of cationic antimicrobial peptides against Gram-negative bacteria is the binding of the peptides to the divalent-cation-binding sites of the polyanionic surface LPS (Devine and Hancock, 2002) . The presence of cations in the solution may limit this binding ability by saturating the LPS divalent-cation-binding sites, and thereby decreasing the antimicrobial activity of the peptides. To determine whether the test peptides were affected by the presence of cations, metal ions in the form of their chloride salts were added to the media at concentrations of 1 , 5 and 1 0mM, and the MICs of the peptides were determined. The experimental method is described in Section 3 .4.2 . The results are summarised in F igure 7.2. The raw data are given in Appendix 1 45 Chapter 7 - Effect of Conditions A3 .2, example calculations are given in Appendix A3 . 5 and the statistical analysis is given in Appendix A3 .6. The analysis of variance determined that there were significant differences between the mean MICs at different concentrations for each of the cations tested (p-values were 0.0022, 0.0022, 0.0002 and 0.0000 for Na+, K+, Mg2+ and Ca2+ respectively). There were significant differences between the mean MICs of each peptide in the presence of K+ and cl+ (p-values were 0.003 1 and 0.0268), but not in the presence of Na+ and Mg2+ (p- values were 0 . 1 042 and 0.067 1 ). 80 80 - SMAP29 - SMAP29 - OaBac5mini - OaBac5mini :J - OaBac7.5mini :J - OaBac7 .5mini E 60 E 60 -- -- Cl Cl 2: 2: () 40 () 40 � � Q) Q) "0 "0 a :;:. 20 a. 20 Q) Q) a. a. 0 0 0 2 5 1 0 0 2 5 1 0 Na+ concentration (mM) K+ concentration (mM) 80 80 - SMAP29 - SMAP29 - OaBac5mini - OaBac5mini :J 60 - OaBac7.5mini :J 60 - OaBac7.5mini E E -- -- Cl Cl 2: 2: () 40 () 40 � � Q) Q) "0 "0 :;:. a a. 20 20 Q) al al al Q) a. > > > a. u ·ti U III III III 0 0 0 0 0 0 c: 0 2 5 1 0 0 2 5 1 0 Mg2+ concentration (mM) Ca2+ concentration (mM) Figure 7.2 - The effect of metal ion concentrations on the minimum inhibitory concentration (MIC) of synthetic ovine peptides against E. coli 0157:H7. The bars show the geometric means of three runs of dupl icate samples and the error bars show the 95% confidence intervals for these means. In the presence of the monovalent cations, Na + and K+, the MICs of the test peptides were higher than in their absence. However, the peptides st ill showed moderate antimicrobial activity under these conditions, which shows that the monovalent ions do not complete ly block the binding of the cationic peptides to LPS. 146 Chapter 7 - Effect of Conditions The divalent ions had a larger effect on the activity of the test peptides than the monovalent cations. Mg2+ and Ca2+ ions inactivated the proline/arginine-rich peptides at concentrations of 1 0mM and SmM respectively. SMAP29 was inactive in the presence of 1 0mM of both divalent cations. SMAP29's higher tolerance to divalent cations, compared to the other two peptides, was probably due to its higher affinity for LPS as shown in Section S .4 . I t has previously been shown that, like OaBacSmini and OaBac7. Smini, other cationic antimicrobial peptides are also more sensitive to Ca2+, than Mg2+ ions (Turner et al, 1 998) (Selsted et al, 1 985) . This is probably due to Ca2+ ions having a higher affinity for LPS than Mg2+ ions. This may be because Ca2+ ions are larger than Mg2+ ions and are able to interact more readily with the divalent-cation-binding sites on LPS. For the proline/arginine-rich peptides there were some unexpected results. The mean MIC was significantly higher for OaBacSmini in the presence of SmM compared to 1 0mM Na + (p­ value = 0.0090). The mean MICs of both OaBacSmini and OaBac7 .Smini were significantly higher in the presence ofSmM compared to 1 0mM K+, and in the presence of 2mM compared to SmM Mg2+ (p-value = 0.0090 in both cases). This implies that certain cation concentrations are more inhibitory than others; however, the reason for this is unknown. S ince it is envisaged that the antimicrobial peptides could be used as biopreservatives for chilled lamb products, the concentrations of the cations in fresh lamb meat were sought. The concentrations ofNa+, K+, Mg2+ and Ca2+ in fresh, lean, raw lamb meat are 70, 330, 22 and 1 2mg/ 1 00g respectively (Chan et al, 1 995) . The concentrations of each cation in mmollkg were calculated using Equation 7 . 1 . I . /k I b mg cation/kg raw lamb mmo catlOn g raw am = ---==------'---=- --- g cation/mol e.g. For sodium: I d· /k I b mg sodium/kg raw lamb mmo so lUm g raw am = g sodium/ mol 70mg sodium/kg raw lamb = -��--���----- 22.99g sodium/mol = 3 .04mmol/kg raw lamb Equation 7. 1 The calculated cation concentrations in raw lamb meat are summarized in Table 7. 1 . The divalent cations completely inhibited the ant imicrobial activity at concentrations of l OmM, 1 47 Chapter 7 - Effect of Conditions which corresponds to l Ornrnoles/kg ( l OmM = 1 0rnmoles/L = 1 0rnmoles/kg), and were part ial ly inhib itory at concentrations of 2mM, which corresponds to 2rnrnoles/kg. The concentrations of the individual d ivalent cations in the raw lamb meat are lower than the inhibitory concentrations; however, the inhibitory effect is most likely addit ive so the total divalent cation concentration should be considered . Table 7.1 - Concentrations of metal ions in lean trimmed, raw lamb meat. cation sodium potassium magneslUm calc ium mg/1 00g lamb 70 330 22 1 2 molecular weight 22.99 39. 1 0 24.30 40.08 mmoVkg raw lamb 3.04 8.44 0.9 1 0.30 The locations of the cations within the meat are also important when considering their possible interaction with the peptides and the bacterial LPS. Of the magnesium present in the meat, most of it is situated in the intracellular fluid at concentrations as high as 5 - 1 0mM (Price and Schweigert, 1 987), so it is unlikely to interfere with the act ivity of the peptides on the surface of the meat. Similarly, in animals the majority of calcium is situated within the cells where is acts as an intracellu lar messenger ( Price and Schweigert, 1 987). However, in post-mortem meat free Ca2+ tends to rise and so could inhibit these peptides after the animal has died. 7.4 EFFECT OF PH Another property that was important to investigate for its effect on the activity of the peptides was pH. This was established by determining the MIC of each peptide against E. coli 01 57:H7 at a variety of pH values from 5 to 9 according to the method described in Section 3 .4 .3 . The results are graphed in Figure 7 . 3 . There were significant differences between the mean MICs of each peptide (p-value = 0.0006), but there were no significant differences between the mean MICs at different pHs (p-value = 0.085 1 ). The raw data are given in Appendix A3 .3, example calculations are given in Appendix A3 .5 and the statist ical analysis is given in Appendix A3.6 Two of the test peptides had higher mean MICs in acidic condit ions compared to basic conditions. OaBac5 mini and OaBac7.5 mini had MICs four to eight times greater at acidic pH than at pH8. The activity of these two peptides was inhibited more at pH 6 than pH 5. There was little change in the antimicrobial activity of SMAP29 over the pH range tested. 1 48 20 1 5 .-.. -I E -- � --- 0 1 0 :2: Q) -0 a Q) a. 5 5 6 7 pH 8 Chapter 7 - Effect of Conditions _ SMAP29 _ Oa8ac5mini _ Oa8ac7 .5mi 9 Figure 7.3 - The effect of media pH on the min imum inhibitory concentration (MIC) of synthetic ovine peptides against E. coli 0157:H7. The bars show the geometric means of three runs of duplicate samples and the error bars show the 95% confidence intervals for these means. The changes in activity at different pH values may be due to changes in the charges of the peptides. All three peptides contain numerous arginine residues; however, arginine has a pK of 12 (Stryer, 1 995), so these residues would be positive in a l l the conditions tested. All three peptides also contain a terminal amino group, which has a pK of 8 (Stryer, 1 995) . It is unlikely that the protonation of this group causes the decrease in activity of the proline/arginine-rich peptides, because this also occurs in SMAP29, which is not affected. SMAP29 contains a histidine residue, which has a pK value of 6.5 (Stryer, 1 995); however, the protonation of this residue does not appear to have an effect on the activity of the peptide. The peptides do not contain any other ionisable residues. A decrease in antimicrobial activity at low pH has previously been shown to occur with other cationic antimicrobial peptides (Miyakawa et al, 1 996). LL-37, a human cathelicidin, showed changes in its structure under different pH conditions (Johansson et al, 1 998). At basic pH LL-37 is a-hel ical, but at acidic pH it has a random structure. Such conformational changes to the peptides may be responsible for their decreased activities in acidic pH conditions. Alternatively, the changes in activity at acidic pH may be due to changes in the bacterial membranes, instead of the peptides. A reduction in the negative charge of the bacterial surface would decrease the binding of the cationic peptides. 1 49 Chapter 7 - Effect of Conditions I f the peptides are to be used as a biopreservative they need to operate within the pH spectrum of meat products. The pH of the live animal is 7.2, but this falls to approximately 5 .4 in the dead animal. The acceptable pH range for eating quality lamb is 5 .3-5 .7 (Oppenhei et ai, 2003) . This pH range is more ac idic than the optimum for the proline/arginine-rich peptides; however, the peptides are still act ive in this range. 7 .5 EFFECT OF TEMPERATURE The [mal physical property that was investigated for its impact on the ant imicrobial activity of the peptides was temperature. The effect of temperature on the activity of the peptides against E. coli 0 1 57 :H7 was determined by heating the peptide so lutions for 30 minutes prior to determining the MIC values as described in Section 3 .4.4. The results are summarised in Figure 7 .4. The raw data are given in Appendix A3 .4, example calculations are given in Appendix A3 .5 and the statistical analysis is given in Appendix A3 .6 . There were significant differences between the mean MICs of each peptide (p-value = 0.0000), and there were significant differences between the mean MICs at different temperatures (p-value = 0.0000) . 1 2 _ SMAP29 1 0 _ OaBac5min i _ OaBac7.5min i -----.J 8 E --Cl � -- () 6 � Q) '0 � c.. 4 Q) c.. 2 control 30 40 50 60 70 80 90 1 20 temperatu re (deg C) Figure 7.4 - The effect of heating on the minimum inhibitory concentration (MIC) of synthetic ovine peptides against E. coli 0157:H7. The bars show the geometric means of three runs of duplicate samples and the error bars show the 95% confidence intervals for these means. 1 50 Chapter 7 - Effect of Conditions The peptides were relatively heat-stable. All three peptides retained their antimicrobial activity when heated to temperatures as high as 80°C. However, SMAP29 and OaBac5 had an e ight-fo ld increase in their MICs after being heated to 90°C or 1 20°C, and OaBac7.5mini had a four-fo ld increase in MIC in these conditions. It is common for heat-treatment to reduce the biological activity of peptides and proteins because it can cause them to unfo ld. The test peptides have simple secondary structures. In aqueous solutions the peptides have a random structure, and it is not until they come into contact with membrane-like conditions that they fold into their amphipathic structures, as shown by the CD spectra experiments in Section 5 .2 . The test peptides do not contain any disulphide bonds to stabilise them, which probably accounts for their decrease in activity after heating to high temperatures. The reduction in activity of the peptides at high temperatures may limit their use. If they are to be used as preservatives for chilled lamb products, they would not be heated. However, they may not be suitable for other applications such as preservatives in canned goods, which are heated to high temperatures. 7.6 SYNERGY BETWEEN PEPTIDES Along with exploring the effect of the environmental conditions on the activity of the peptides, their activity in combination with each other, compared to their individual activities, was investigated. Work carried out by other researchers has shown that some pairs of cathelicidins are synergistic, although not in al l combinations (Yan and Hancock, 200 1 ). To determine if the test peptides worked synergist ically together, the fractional inhibitory concentration (FIC) was calculated for each combination of the three peptides. Each pair of peptides was tested in a 96-well microtitre p late, with one peptide diluted down the columns and one across the rows. The details of the method are given in Section 3 .4. 5 . The wells where growth occurred were recorded for each plate. A diagram of a typical FIC plate is given in F igure 7 .5 . 1 51 Chapter 7 - Effect of Conditions ..- � e 4 2 blJ 4 � -- = 2 .s: -� 1 I--= � 0.5 c:J = 0 0.25 c:J .- = .- 0. 1 25 e I£l c:J 0.0625 � � � 0 0 Oa8ac7.5mini MIC OaBac7.Smini concentration (Jlg/mL) 0.5 0.25 X X Fie X X X X 0. 1 25 0 .0625 X X X X X X X X X X o Oa8ac5mini MIC X X X X X Figure 7.S - Diag ram of a microtitre plate for a typical synergy test for OaBacSm ini and OaBac7.Smini . Growth occu rred in the wells containing an 'X'. MIC = minimum i n hibitory concentration . Fie = fractional inhibitory concentration. The FICs were calcu lated using Equation 7.2. FIC = [A] + [B] MIC A MIC B Equation 7 .2 where MICA and M ICB are the MICs of peptides A and 8 alone, and [A] and [8] are the minimum amounts of peptides A and 8 required for inhibition when used in combination. 1 52 e.g. For the typical OaBac5mini and OaBac7.5mini run described in Figure 7. 5 : M I CA = M ICoaBac5111ini = I l-lg/mL MI CB = M ICOaBac7.5mini = 21-lg/mL (MIC of OaBac5mini in the right-hand column where the concentration of OaBac 7.5 is zero) (MIC of OaBac7 .5rnini in the bottom row where the concentration of OaBac5 is zero) The FIC is calculated at the point where the two concentrations required are at their lowest as indicated in F igure 7 .5 . [A] = [OaBac5rnini] = 0.06251-lg/mL [B] = [OaBac7.5rnini] = 0.51-lg/mL FIC = [A] + [B] = 0 .0625 + .2.2 = 0.3 1 25 M IC A MIC B 1 2 Chapter 7 - Effect of Conditions The FIC values indicate whether peptide activities are additive, synergistic or antagonistic. An FIC of unity, i.e. each antimicrobial agent is at half its MIC, indicates that the antimicrobial agents have additive activity. An FIC of O .S or less, i .e . each antimicrobial agent at a quarter of its MIC or less, indicates that the antimicrobial agents have synergistic activity. An FIC of four or higher, i.e. each antimicrobial agent at twice its MIC or higher, indicates that the antimicrobial agents are antagonistic. Three runs were carried out and the results were the same. Each pair of test peptides (SMAP29/0aBacSmini, OaBac7.SminilSMAP29, and OaBacSmini/OaBac7.Smini) had an FIC of 0.3 1 2S . This means that antimicrobial activity was observed when one of the peptides was diluted to one quarter of its MIC and the other peptide was diluted to one sixteenth of its MIC; therefore, they have strong synergistic behaviour. Synergistic interactions of the peptides in vivo would be advantageous and enable the animal' s immunity system to deal with most microbial challenges. Firstly, it would mean that the animal would need to produce less of the peptides to fight microbial invasions than would be needed if a single peptide was used. Secondly, it would be more difficult for microorganisms to build up resistance to the peptides because they would be attacked by more than one peptide with different structures and mechanisms of action at the one time. 7.7 SYNERGY BETWEEN PEPTIDES AND KNOWN ANTIBIOTICS Since the test peptides worked synergist ically with each other, it was decided to test if they were also synergistic in conjunction with common ant ibiotics. The method used is described in Section 3 .4 .6 . For these tests four antibiotics, each with a different mechanism of action, were used. The antibiotics and their mechanisms of action are listed in Table 7.2 . The tests were carried out three times and the results were the same. The FICs for each antibiotic in combination with each test peptide are summarized in Table 7 .3 Table 7 .2 - Antibiotics used in the synergy tests and their mechanisms of actions. Antibiotic polymyxin B ampicil l in kanamycin A rifampicin Mechanism of action disrupts the structure/function of the cell membrane inhibits bacterial cell wall synthesis inhibits protein synthesis inhibits transcription/replication 1 53 Chapter 7 - Effect of Conditions Table 7.3 - Fractional inhibitory concentrations of synthetic ovi ne peptides i n combination with common antibiotics. Antibiotic Fractional inhibitory concentration SMAP29 OaBacSmini OaBac7.Smini polymyxin B 0. 1 875 0 . 1 5625 0.28 1 25 ampic i l lin 1 1 kanamycin A 0.53 1 25 0 .5625 0.625 rifampicin 0 .625 0 .5 0 .53 1 25 Of the antibiot ics tested, only po lymyxin B showed strong synergistic act ivity with al l three ovine-derived pept ides. The FIC values of po lymyxin B in combinat ion with the peptides were low, indicating strong synergistic behaviour. The fractions of the MIC required for activity of po lymyxin B and peptide were one sixteenth and one eighth for SMAP29, one eighth and one thirty-second for OaBac5mini, and one quarter and one thirty-second for OaBac7.5mini. In contrast, the FIC values for ampici llin in combination with the peptides were all one, meaning that the activity was addit ive but not synergist ic . The synergy between po lymyxin B and the ovine peptides may be because polymyxin B is also a pept ide antibiotic that permeabi l ises bacterial membranes in a similar way to the ovine peptides (Moo re et ai, 1 986). Kanamycin A and rifampicin both interact with the inner cellular macromolecules of the bacteria, so the increased permeability of the cel l membranes induced by the ovine peptides may have assisted the uptake of these ant ibiotics, which is a possible exp lanation for the increase in activity observed with these two antibiotics. The mechanism used by ampicil l in, inhibit ing the synthesis of the bacterial cell wall, is unrelated to those used by the peptides, which probably accounts for the lack of synergy shown. 7.8 CONCLUSIONS The objective of the work presented in this chapter was to determine the effect of different environmental factors on the activity of ovine antimicrobial peptides. Due to the ir broad­ spectrum ant imicrobial activity, ovine antimicrobial peptides have the potential to have many useful applications. Before developing appl ications it was necessary to determine what factors may interfere with, or enhance, the activity of the peptides. The activit ies of the test peptides were evaluated in condit ions that mimic the physio logical properties of raw lamb meat derived from wel l rested animals, i .e . acidic pH and presence o f cations. In these conditions the antirrilcrobial activity of the peptides were reduced compared 1 54 Chapter 7 - Effect of Conditions to their activities in low-salt antibiotic testing media (Muel ler-Hinton broth). However, the peptides stil l displayed reasonable antimicrobial activity in the relevant conditions, which indicates that they have the potential to be used as biopreservatives for chilled lamb products. Further testing is required to determine if the peptides are capable of protecting the meat surface from bacterial contamination. While not guaranteeing the success of these tests, the information gathered in this research suggests that the peptides should be suitable for this application. However, these tests were carried out in liquid media and the peptides may behave differently on a meat surface that will vary considerably in moisture content depending on the chilling regime and the time after slaughter that the peptides are applied. I n some processing plants, carcasses are chilled by subjecting them to high velocity air during the initial chilling cycle ; whereas, in other plants the carcasses are intermittently sprayed with chilled, chlorinated water and cooled by evaporative cooling of the added water. The peptides could be applied to carcasses that are to be cooled by blast air as they are leaving the cooling floor; whereas, in the case of the spray-chilled carcasses, the peptides may best be applied to the primals just prior to packaging. More work is needed to evaluate the success of these peptides in a variety of chilling and packaging regimes, to assess their viability as potential shelf life extenders for the country' s chilled meat exports. These experiments also showed that the peptides were able to retain their activity when heated to temperatures up to 80°C, but their activity was reduced if they were heated higher than this. This property would not affect the ability of the peptides to be effective chilled meat preservatives, but it may limit their usefulness in applications such as canned products. The results in this chapter also showed that ovine peptides d isplayed synergistic activity when used in combination. Due to this, it would be best to use a mixture of peptides in a product as opposed to a single purified peptide. This would also have the advantage of reducing the number of downstream purification steps required to create the product, which would make the process more cost effective. It would also mean that lower concentrations of each of the peptides would be necessary compared to the situations where only single peptides are used. This study has demonstrated that synthetic ovine antimicrobial peptides are robust and that they retain their activity over a large variety of conditions. The cathel icidins that can be iso lated from ovine neutrophils show great potential for a range of applications. These applications should be evaluated, so that it can be decided whether to commercialise the 1 55 Chapter 7 - Effect of Conditions production of the peptides and hence the ir use in a range of food products and possibly even pharmaceutical products. 1 56 Chapter 8 - Pilot-Scale Extraction CHAPTER 8 PILOT-SCALE EXTRACTION OF ANTIMICROBIAL PEPTIDES FROM OVINE BLOOD 8.1 INTRODUCTION The work presented in this chapter was concerned with the fifth objective of this research project, which was to determine whether it is possible to produce an active antimicrobial extract on a scale larger than that used in the laboratory, using industrial-style equipment. The results in the previous chapter showed that the synthetic ovine antimicrobial peptides are robust and can exert antimicrobial activity in a variety of conditions. This indicates that the naturally occurring peptides may be able to be used in products such as biopreservatives and antiseptic topical creams. The first objective of the research presented in this chapter was to carry out pilot-scale extractions of antimicrobial peptides from ovine blood to determine whether it is possible to produce an active crude neutrophil extract from ovine blood on a pilot-scale. As some of the laboratory scale operations, such as sonication, were not easily scaled up, it was necessary to investigate alternative processes to derive the active peptides. The second objective was to investigate the activity of the crude neutrophil extract to make sure that the extract produced on the pilot-plant had similar activities to that of the laboratory­ derived samples. The MIC of the crude extract was determined against a wide spectrum of organisms. A number of food pathogens and food-spoilage organisms were used to determine if the extract had the potential to be used as a biopreservative for chil led lamb products. The results in Chapter 4, i .e . the production of the antimicrobial peptides using the laboratory procedure, showed that the antimicrobial activity present in the crude solution extracted from ovine neutrophils was predominately caused by the proline/arginine-rich cathelicidin peptides. These peptides were more active against Gram-negative than Gram-positive bacteria and had poor activity against yeast. It was expected that the crude extract produced by the pilot plant process would have similar activity. To further investigate the activity of the crude extract, TEM was used to see what physical effects the peptides had on microbial cells. The experiments described in Chapter 5 and Chapter 6 showed that the proline/arginine-rich peptides, OaBac5mini and OaBac7.5mini, are 1 57 Chapter 8 - Pilot-Scale Extraction bacteriostat ic, not bactericidal, and that they inhibit bacterial cel l division by interacting with their inner cel lular contents. Therefore, it was expected that the crude extract would not cause notable morphological changes to the cells. The third objective was to more accurately determine the yield of antimicrobial peptides from ovine blood. Yie ld is a vital determinant of commercial process viability. 8.2 CRUDE EXTRACTION A number of extraction runs were carried out using pilot-scale equipment to see if it was possible to produce an active crude extract on a larger scale than was used in the laboratory. Instead of using four litres of blood solution (blood plus sodium citrate solution), which was usual for the laboratory extractions, either 1 0, 20 or 50 litres of blood so lution were processed to more accurately determine the yields. The pilot scale extraction process used for this latter set of experiments is described in Section 3 . 5 . 1 and is summarised in Figure 8 . 1 . There were three main changes in the pilot-scale process from the laboratory process. Firstly an initial centrifugat ion step was used to remove the blood plasma and the majority of the red blood cells from the pelleted white blood cells in two clean streams. This separation was done using a disc-stack centrifuge (Alfa-Laval Cream Separator) that is usually used for separating milk. A photograph of this centrifuge is given in Figure 8 .2 . This centrifugation step had two main advantages. The flIst advantage was that the blood plasma and the red blood cells were separated into two clean streams. Although this research has focussed on utilising the white blood cells, the economic viability of a venture designed to extract the antimicrobial peptides from neutrophils would be great ly enhanced if the plasma and red blood cell streams were also further processed. The plasma in particular has potential to be converted into high-value products, such as blood serum and blood proteins inc luding serum albumins, fibronectin, transferrin, antibodies and trypsin. I n the modified laboratory process the p lasma stream was mixed with the lysed blood cells, which would make further processing of the plasma difficult . 1 58 Chapter 8 - Pilot-Scale Extraction P BSX buffer (2kg) ammon ium chlo ride so lut i on (2 .5kg) 1 0% acet i c ac id (500 g) b l o o d .. so d i u m citrate so l uti o n (10kg) CENTRIFUGATI O N white b lood ce l ls + red b lood cel ls (2 .5kg) RE D B L O O D CE L L L VS IS white blood cel ls + lysed red blood ce l ls (5k g) CE NTRI F U GATI O N plasma (6kg) red b lood ce l ls (3 .5kg) l ysed red b lood ce l l s (4 .9kg) white b lood ce l l s in acet i c ac id so lut ion (600g) B L E�D I N G neutroph i l granu les + white b lood ce l l debr is in acet ic acid s o lut ion (600g) CENTRIF U GATION neutroph i l granu les (1 00g) ant im icrob ia l pept ides in acet i c ac id so lut ion (5OOg) 0 .0 1 % acet i c ac id (24g) ROTARV EVAPORATION ant im icrob ia l pept ides in water (450g) F RE EZE -D RVI N G dr ied crude ext ract « 1 g) R E D I S S O L UT I O N acet i c ac id (50g) water (1 80g) crude a ntim i cro b i a l e xtra ct in 0.0 1% a cetic a cid (24g) Figure 8.1 - Flow d iagram showing the pi lot-scale process used to extract antimicrobial peptides from ovine blood. Sod ium citrate solution contains 10% sodium citrate. PBSX buffer contains 1 37mM NaCI, 2.7mM KCI, 0.5mM MgCI2, 8.1 mM Na2HP04 and 1 .5mM KH2P04 (pH 7.4). Ammonium chloride solution contains 0.83% ammonium chloride. Acetic acid solution contains 1 0% acetic acid. 1 59 Chapter 8 - Pilot-Scale Extraction Figure 8.2 - Photograph of the pi lot-scale disk-stack centrifuge used to separate white blood cel ls from plasma and red blood cel ls. The white blood cells and some of the red blood cel ls were retained inside the centrifuge on the discs. The second advantage of the pilot-scale centrifugation process over the laboratory process was that a lot fewer red blood cells needed to be lysed because the majority were removed from the white blood cell fraction during the centrifugation. This reduced the amount of ammonium chloride so lution required. In the laboratory process an equal amount of ammonium chloride so lution to blood was used, whereas in the pilot-scale process the amount of ammonium chloride solution needed was only a quarter of the volume of the blood processed. This reduction was important because it would reduce the raw material costs of the process and it reduces the amount of ammonium chloride solution that would need to be either recovered or sent to waste. The last course of action would be unfavourable as it would impose undesirable nitrogen loads on the waste treatment system. The second change in the pilot-scale process compared to the laboratory process was the use of a bench-top blender instead of a sonicator to disrupt the white blood cells. This method was used because it disrupted the cells mechanically instead of using ultrasonic waves. Mechanical breaking o f the white blood cells would be more practical in a commercial process because it would not generate as much heat as sonication would on an industrial scale. The third change in the pilot-scale process compared to the laboratory process was that the white blood cells were re suspended directly in acetic acid solution instead of buffer. This had the benefit of eliminating the need to centrifuge the disrupted white blood cells to collect the neutrophil granules prior to suspension in acid extraction solution. 1 60 Chapter 8 - Pilot-Scale Extraction 8.3 MIN IMUM INH IBITORY CONCENTRATIONS The MIC of the crude extract from a typical pi lot-scale extraction run was determined using the method described in Section 3 .5 .2 . The crude extract was tested against a variety of microorganisms, including Gram-negative and Gram-positive bacteria, and yeast. Three runs were carried out, each containing duplicates of each sample. In all cases, at least four of the six recorded values for the MIC of each peptide against each organism were the same. The MIC values varied from the mode values by only a single two-fold dilution. The variations were seen both between duplicates in the same run and between runs. The MICs are reported as the geometric means and the limits of the 95% confidence intervals for the means in Table 8 . 1 . The raw data and example calculations are given in Appendix A4. l . The MIC values were not expressed as a mass per volume (j.1g/mL) because the majority of the components in the crude extract did not contribute to the ant imicrobial activity. There were variations in the composit ion of the crude extract between runs so the mass of solids did not necessarily correlate with the antimicrobial activity. To compare activity between extracts, the MIC values were expressed as units/mL. One unit was defined as the amount required to inhibit E. coli 01 1 1 at a concentration of 1 05 CFU/mL. The crude extract had very broad-spectrum activity. It was equally active against both the Gram-positive and Gram-negative bacteria. This was unexpected because the purification work in Chapter 4 showed that the majority of the crude extract antimicrobial activity was due to the proline/arginine-rich peptides. In Chapter 5 the synthetic proline/arginine-rich peptides tested were more active against Gram-negative than Gram-positive bacteria. The crude extract was also more active against the yeast than expected. The results presented in Chapter 7 showed that the ovine-derived peptides were strongly synergistic. Therefore, the strong antimicrobial activity of the crude extract may be due to the synergistic interactions between the different peptides present. Alternatively, the strong antimicrobial activity may be due to other peptides or proteins present in the crude extract that were not characterised in Chapter 4 1 61 (J) N Table 8.1 - Minimum inhibitory concentrations of ovi ne neutrophil crude extract from the pilot-scale extraction. Organism MIC (J.lL crude/mL total) Gram-negative bacteria Escherichia coli 0 1 1 1 4.0 (4.0, 4.0) Escherichia coli 0 1 57 :H7 4.5 (3 .6, 5 .6) Salmonella enteritidis 4.0 (4.0, 4.0) Salmonella typhimurium 7. 1 (5 .7, 8 .9) Klebsiella pneumoniae 4.5 (3 .6, 5 .6) Pseudomonas aeruginosa 4.0 (4.0, 4.0) Pseudomonas fluorescens 3.2 (2 .0, 5 .0) Yersinia enterocolitica 4.5 (3 .6, 5 .6) Gram-positive bacteria Staphylococcus aureus NCTC 4 1 63 4.0 (4.0, 4.0) Staphylococcus aureus 1 056 MRSA 9.0 (7 .2, 1 1 .3 ) Streptococcus faecalis 4.0 (4.0, 4 .0) Bacillus cereus 4.5 (3 .6, 5 .6) Bacillus nato 5.0 (3 .8 , 6 .7) Listeria monocytogenes 1 08 A 2 .0 (2 .0, 2 .0) Listeria monocytogenes NCTC 1 0884 2.2 ( 1 .8 , 2 .8) Listeria monocytogenes NCTC 7973 4 .0 (4.0, 4.0) Yeast C. albicans 3 1 53A 32 .0 (32.0, 32 .0) 'One unitlmL is defined as the concentration required to inhibit 1 05 CFU/mL E. coli 01 1 1 . MIC (units/mL*) 1 . 0 ( 1 .0, 1 . 0) 1 . 1 (0 .9, l A) 1 . 0 ( 1 .0, 1 . 0) 1 . 8 ( 1 .4, 2.2) 1 . 1 (0.9, l A) 1 . 0 ( 1 .0, 1 . 0) 0.8 (0.5, 1 .2) 1 . 1 (0.9, l A) 1 . 0 ( 1 .0, 1 . 0) 2 . 2 ( 1 .8 , 2 .8) 1 . 0 ( 1 .0, 1 . 0) 1 . 1 (0.9, l A) 1 . 3 (0.9, 1 . 7) 0.5 (0 .5 , 0 .5) 0 . 6 (004, 0 .7) 1 . 0 ( 1 .0, 1 .0) 8.0 (8 .0, 8 .0) The MICs are expressed as the geometric means of five runs, each with duplicate samples. The values in the brackets are the lower and upper limits of the 95% confidence intervals of the means. The raw data are given in Appendix A4.1 and calculations are given in Appendix A4.2 Chapter 8 - Pilot-Scale Extraction Many of the test organisms were food pathogens or food-spoilage organisms. E. coli 0 1 57 :H7, Listeria monocytogenes, Salmonella enteritidis, Salmonella typhimurium, Yersinia enterocolitica, Streptococcus faecalis, Bacillus cereus and Staphylococcus aureus are al l pathogens that can infect food products. Others of the test organisms, such as Pseudomonas jluorescens can be involved in food spoilage. The unexpected activity against Gram-negative bacteria as well as its activity against Gram-positive bacteria reinforced the idea that the crude extract had the potential to be an effective preservative for chilled meat products. Some of the test organisms are c l inically significant. Klebsiella pneumoniae is common in hospitals where it causes pneumonia and urinary tract infections. Pseudomonas aeruginosa is an opportunistic pathogen that causes numerous conditions such as infections of the urinary tract, respiratory system, dermatitis, soft tissue, bone and joints, and gastrointestinal tract. Candida albicans is also an opportunistic pathogen that most commonly infects the oral cavity and intestinal tract. 8.4 TRANSMISSION ELECTRON MICROSCOPY Transmission electron microscopy (TEM) was used to further investigate the activity of the crude extract. Cultures of E. coli 0 1 1 1 , S. aureus NCTC 4 1 63 and C. albicans 3 1 53A were treated with the crude extract and then prepared for TEM according to the method described in Section 3 .5 .3 . For each sample numerous sections were cut and images were taken of random areas within each section. The images presented in Figure 8 .3 are typical for each culture. The majority of the E. coli and a large number of the S. aureus cells that were treated with crude extract either leaked their inner cellular contents or had completely burst open. Although some of the E. coli cells in the control appeared to be damaged, the majority of the cells in both bacteria controls were intact. Therefore, the extensive cell damage seen in the samples treated with the crude extract must be due to the action of the crude extract. Whether the crude extract directly caused this damage, or it weakened the cells so that they lysed during the TEM preparation process, is unknown. The morphological changes observed were more dramatic than those seen with the cultures treated with the individual synthetic ovine peptides in Sections 6 .2 and 6.3 . 1 63 Chapter 8 - Pilot-Scale Extraction E. coli 0 1 1 1 s. aureus NCTC 4 1 6 c. albicans 3 1 53A Figure 8.3 - Transmission electron microscopy images of control cells treated with 0.01 % acetic acid ( left) and cells treated with ovine neutrop hi l crude extract ( right). 1 64 Chapter 8 - Pilot-Scale Extraction The majority of the yeast cells that were treated with the crude extract appeared to be damaged. They were not leaking their inner cellular contents or burst open as was seen with the bacterial cells. Instead they had large areas where the outer membrane appeared to be peeled off the cell wall and air sacs or vacuoles were present. Due to the thick cell wal l of the yeast, the cells held together; however, perturbation of the cell wall and cell membrane allowed fluid or air to accumulate inside the cells. The control cells did not have these ultrastructures so they must have been induced by the crude extract. L ike the MIC results, these TEM results were unexpected. The purification work in Chapter 4 showed that the crude extract activity was predominately due to the proline/arginine-rich peptides, but the mechanisms and microscopy work in Chapter 5 and Chapter 6, using synthetic proline/arginine-rich peptides, showed that these peptides d id not induce cell death and had l ittle effect on cell morphology. Again, this may be due to the peptides acting synergistically, or to the presence of other peptides or proteins that were not characterised. 8.5 YIELD CALCULATION For each of the pilot-scale extraction runs the y ield of units of activity per l itre of blood was calculated using the method described in Section 3 .5 .4. The calculation of the yield for a typical p ilot-scale extraction was done as fo llows: Starting volume F inal volume MIC against E. coli 0 1 1 1 = 1 0L of blood solution (9L of whole blood + 1 L 1 0% sodium c itrate) = 24mL of crude extract = l )lL crude extractl l OO)lL solution = 1 0)lL crude extract/mL solution If 1 unit of activity is described as the amount required to inhibit 1 mL of 1 x l 05 E. coli 0 1 1 1 then for this extract : 1 unit Therefore: Total units extracted = 1 0)lL extract = 24mLI l O)lL = 2400units 1 65 Chapter 8 - Pilot-Scale Extraction Therefore: Y ield = 2400uruts/9L whole blood = 267units/L whole blood The calculated yields of the pilot-scale extractions are summarised in Table 8 .2 . For the pilot­ scale extractions the yield varied between 260 and 300units/L of whole blood. I n comparison, the laboratory process y ield was usually between 300-360units/L of blood. Therefore, the current pilot-scale extraction process is not as effective as the laboratory process, so process optimisation is required to reduce the losses and increase the yield. Table 8.2 - Yields for pilot-scale extractions of antimicrobial peptides from ovine blood . Run Blood volume (L) Crude extract MIC (J!L crude Yield volume (mL) extract/mLl (unitslL blood) 9 24 1 0 267 2 1 8 23 .5 5 26 1 3 1 8 2 7 5 300 4 45 65 5 289 From the MIC results it is apparent that a concentration of 3units/mL would be suffic ient to inhibit most bacteria in vitro. With a yield o f 260-300units/L of blood it would be possible to produce 87- 1 00mL of crude extract at a concentration of 3unit/mL for each l itre of blood processed. However, further experiments are required to determine what concentration o f extract is required in practical applications. 8.6 INDUSTRIAL-SCALE PROCESS The pilot-scale extraction trials showed that it was possible to effect ively extract a crude antimicrobial extract from ovine blood on a scale larger than that done in the laboratory. I f the product were to be made for commercial sale, the process would need to be scaled-up and modified further. The appropriate industrial equipment for each step was investigated. From the original laboratory extraction process, which contained eleven steps, the process was reduced to six necessary steps for an industrial-scale extraction. These steps are outl ined in Figure 8 .4. 1 66 Chapter 8 - Pilot-Scale Extraction blood + sodium citJate solution p lasma WH ITE B L O O D C E L L S E PARATION red b lood ce l l s 1 0 % acet i c ac i d ------. white b lood ce l l s i n a c et i c a c id so lut i on WHITE B L O O D C E L L D I S R U PTION neutroph i l g ranu les + white b lood ce l l de bris i n acet i c ac i d so l ut ion ANTI M I C R O B IAL P E PT I D E E XTRACTI O N n e ut roph i l g ranu les + white b lood ce l l d ebr is + ant im i c rob ia l pept i des i n acet ic a c i d so lut i on i I • neut roph i l g ranu les S O L I D S R E M OVAL r- + ce l l deb ris ant im icrob ia l pept ides i n acet ic ac id so lu t ion ACID R E M OVAL A N D C O N C E NTRATI O N acet ic ac id + water clllde tlntimicrobitll extmct Figure 8.4 - Steps in an industrial process to produce a crude antimicrobial extract from ovine blood. The fIrst step in the process is to remove any solids from the incoming blood stream. Commonly pieces of wool and c lotted blood are present in the collected blood. As in the laboratory, in an industrial process these solids could be removed using a fIlter screen. The next step is to separate the white blood cells from the blood plasma and the red blood cells. In the laboratory process the red blood cells were lysed using ammonium chloride solution, then the white blood cells were collected using centrifugation. I n the pilot-scale process the white blood cells were separated from the plasma and the majority of the red blood cells using centrifugation, then the remaining red blood cells were lysed with ammonium chloride solution, to remove them from the white blood cell fraction. 1 67 Chapter 8 - Pilot-Scale Extraction I deally it would be best to collect the white blood cells in single step. The c lean separation of the neutrophils from the red blood cells was not possible with the pilot-scale centrifuge used for this research because it did not have adjustable rotational speed. However, the calculations in Appendix A4.3 show that it would be possible to design a centrifuge to do this separation. This would be advantageous because it would e liminate the need for a lysis step to remove contaminating red blood cells from the white blood cell fraction. The third step in the process is the disruption of the white blood cells to release the neutrophil granules. I n the laboratory this was achieved using sonication or a bench-top blender. I n an industrial-scale process numerous other mechanical processes could replace this operation (Meat New Zealand, 2003 ), such as an homogeniser that uses high pressure pumping through a t iny orifice or a mechanical grinding technique (Perry and Green, 1 997). The next step in the process is to extract the antimicrobial peptides from the neutrophil granules. In the laboratory this was carried out using 1 0% acetic acid with cont inuous mixing. I nstead of using a mixing vessel that would result in a batch process, there are a number of extraction vessel designs that can be used in a continuous process (Braga and Ricci, 1 998). These are usual ly run counter-currently. This means that the fresh so lution comes in contact with the most spent solids, which increases the total amount extracted. Further experiments are required to determine the type of extraction vessel and most suitable arrangement. After the extraction has been carried out the next step is to remove the granules and cell debris from the solution. In the laboratory this was achieved using centrifugation, but in an industrial process a filter press could be used instead. The insoluble so lids would be retained and the c larified solution containing the ant imicrobial peptides would pass through. The fmal step in the process is to remove the acid and concentrate the solution ready for use. In the laboratory this was carried out using rotary evaporation to remove the ac id and freeze­ drying to remove the water. The dried solids were then resuspended in 0.0 1 % acetic acid for use. As an alternative, in an industrial process diafiltration, which concentrates and washes the suspended proteins/peptides, could be used. This invo lves passing the so lution through an ultrafi ltrat ion membrane with a low molecule weight cut-off The peptides of interest are approximately 3kDa so a cut-off of 0.5kDa would be appropriate in this case. The retained proteins and peptides would then be washed with water, i.e. diafiltration, to remove any salts. 1 68 Chapter 8 - Pilot-Scale Extraction As wel l as the number of steps required, another factor that was reduced from the laboratory process to the industrial-scale process was the number and amounts of chemicals required. The industrial process does not require ammonium chloride solution or PBSX buffer, both of which were used for the lab extractions. This el iminates the need to purchase and dispose of these chemicals. 8.7 CONCLUSIONS The first objective of the research presented in this chapter was to carry out pilot-scale extractions of antimicrobial peptides from ovine blood to determine whether it is possible to produce an active crude neutrophil extract from ovine blood on a pilot-scale. The results showed that an active crude extract could be produced from ovine blood on a pilot-scale. It also showed that the process worked effectively using industrial-style equipment instead of laboratory equipment. The preliminary industrial process design showed that it should be possible to produce an active crude extract on an industrial-scale using six simple unit operations and minimal chemicals. Further pi lot-scale trials are required to determine what equipment and conditions are optimal. The most important simplification to the laboratory-scale process was the white blood cell separation step. In the laboratory process the red blood cells were lysed using ammonium chloride solution, then the white blood cells were isolated using centrifugation. In the pi lot­ scale process the whole blood was centrifuged first and the plasma and the majority of the red blood cells were separated from the white blood cells in two c lean streams. I n an industrial process it will be possible to specify a centrifuge that was able to cleanly separate all the white blood cells from the red blood cell and plasma streams. This would have the advantages of allowing the red blood cells and plasma streams to be further processed into other products and it would eliminate the use of ammonium chloride solution. The second objective the work presented in this chapter was to investigate the activity of the crude neutrophil extract to make sure that the extract produced on the pi lot p lant had similar activities to that of the laboratory derived samples. The MIC results showed the crude extract had potent activity against Gram-negative and Gram-positive bacteria, including numerous food pathogens and c linically s ignificant organisms, and was relatively active against the yeast C. albicans. This was unexpected because the major active molecules in the crude 1 69 Chapter 8 - Pilot-Scale Extraction extract were the proline/arginine-rich peptides that were shown to be more active against Gram-negative than Gram-positive bacteria, and poorly active against yeast. The TEM results showed that the crude extract induced lysis of Gram-posit ive and Gram­ negative bacteria and caused the separation of the membrane from the yeast cel ls. Again these results were unexpected because they did not correspond to the properties of the synthetic pro line/arginine-rich peptides that were evaluated. The proline/arginine-rich peptides were not bactericidal individually and did not induce morpho logical changes on bacterial cel ls. The enhanced activity of the crude extract compared to the individual peptides may have been due to synergistic interactions between the peptides, or due to a component in the crude extract that was not characterised. This shows that it would be better to use the crude extract in a product as opposed to purified peptides. Further purification would be costly on an industrial-scale and would mostly likely reduce the effectiveness of the product . The third objective of the work presented in this chapter was to more accurately determine the yield of antimicrobial peptides from ovine blood. The results of this research also showed that the yield of antimicrobial activity was 260-300 units/L of blood. This means that from each l itre of blood that was processed it was possible to make 87-90mL of crude extract at a concentration three times that required to inhibit 1 05 CFU/mL E. coli 0 1 1 1 . This is a suffic ient concentration to inhibit all the bacteria tested in vitro. The next step should be to develop applications for the crude extract, such as coating for chilled-lamb products or as a component in topical creams. Once it has been determined what concentration of the crude extract would be required for a given application or product, then the cost for producing the product should be estimated. From this, it will be possible to determine whether the process is economically viable. 1 70 Chapter 9 - Conclusions CHAPTER 9 CONCLUSIONS AND RECOMMENDATIONS 9.1 SUMMARY OF RESEARCH CONCLUSIONS The fIrst objective of the research presented in this thesis was to purify and identify antimicrobial peptides from ovine blood. Prior work had shown a crude extract from ovine neutrophils had antimicrobial activity. I t was hypothesised that the activity was due to antimicrobial peptides, because the l iterature review uncovered that seven ovine antimicrobial peptides were predicted from ovine cDNA. However, only one of these had been purifIed from ovine blood. In the research presented in this thesis, a number of peptides responsible for the antimicrobial activity of the crude extract were identified. The majority of these were proline/arginine-rich cathelicidin peptides. OaBac5 and its variant OaBac5y were both characterised. Prior to this work, OaBac5 had been identifIed only from cDNA, and two variants, OaBac5a and OaBac5�, had been isolated from ovine blood. Other proline/arginine-rich peptides were also characterised, including a truncated form of OaBac7.5 and various truncated forms of OaBac l 1 . Both OaBac7.5 and OaBac l 1 were previously predicted from ovine cDNA, but neither of them had been isolated from ovine blood. As well as the proline/arginine-rich cathelicidins, two other novel peptides that displayed antimicrobial activity were iso lated from the crude ovine neutrophil extract. One of these peptides was similar to part of the cathelin domain of the cathelicidin peptides. Previously it was thought that this domain was responsible for suppressing the antimicrobial activity of the peptides unti l they were cleaved for use. Now it appears that this domain may also play a role in inhibiting microorganisms. The other peptide identifIed was part of the signal peptide of the T -cell glycoprotein CD4 precursor. This peptide was very small but it had relatively high antimicrobial activity. The second objective of the research presented in this thesis was to determine the mechanisms of action of ovine antimicrobial peptides. The literature review revealed that antimicrobial peptides had different mechanisms of action depending on their structures. Therefore, this research compared the mechanisms used by peptides from different structural c lasses. 1 7 1 Chapter 9 - Conclusions For this research three synthetic ovine peptides were used. One of these peptides, S M AP29 was a-helical and the other two, OaBac5mini and OaBac 7 . 5 mini, were pro l ine/argin ine-rich peptides. S MAP29 had potent broad-spectrum act ivity. OaBac5 mini and OaBac 7 . 5 mini were less active than S M AP29, but were more active than their bovine ho mo logues. The research presented in this thesis showed that the a-helical peptide had a different mechanism of action to the proline/arginine-rich peptides. The membrane interaction studies showed that all three test-peptides were able to bind to bacterial surface L P S , permeate the outer membrane, and interact with the cytoplasmic membrane of Gram-negative bacteria. However, the way they interacted with the cytoplasmic membrane differed. S M AP29 caused significant depolarisation of the cytoplasmic membrane; whereas, the other two peptides only induced slight depolarisation. Other studies were also carried out to investigate the mechanism of action o f the synthetic ovine peptides. Optical density and viable cel l count studies showed that S MAP29 caused a rapid decrease in the number of viable cel ls and a s light decrease in optical density of the culture. This meant that the depolarisation o f the cytoplasmic membrane caused by S M AP29 was the lethal event and probably led to lysis of the cells. In contrast, OaBac5mini and OaBac 7 . 5 mini did not decrease the number of viable cells or the optical density of the culture ; instead they limited the growth of the cells. This meant that they were bacteriostatic but not bactericidal. It was predicted that the proline/argin ine-rich peptide s inhibited cells by passing across the cytoplasmic membrane and interacting with the inner cel lular contents. Further studies showed that these peptides were able to bind to DNA, so this may be their mechanism of act ion. However, SMAP29 was also able to bind to DNA, which indicates it may have more than one mechanism of inhibiting cells. The third object ive of the research presented ill this thesis was to investigate the morpho logical changes to microbial cel ls induced by ovine antimicrobial pept ides. The literature review showed that ant imicrobial peptides induce different morpho logical changes to microbial ce l ls depending on their mec hanism of action. Therefore, the aim o f this research was to compare the morpho logical changes to bacterial cells induced by the synthetic ovine peptides. The research presented in this thesis showed that S M AP29 induced substant ial morpho logical changes to bacterial cells; whereas, OaBac5 mini did not. Micro scopy studies using TEM 1 72 Chapter 9 - Conclusions confIrmed that SMAP29 did act on the cytoplasmic membrane and induced cell lysis; not only to Gram-negative cells, but also to Gram-positive cells. I n contrast, OaBac5mini did not induce cell lysis, but it did cause the cell membrane to separate from the cytoplasmic membrane in parts of some cells . AFM was also used to image the surface of cells treated with the peptides; however, this technique was not very successful for this application. The fourth objective of the research presented in this thesis was to determine the effect of different environmental factors on the activity of ovine antimicrobial peptides. The literature review showed that some antimicrobial peptides were inhibited by high salt concentrations, divalent cations and acidic pH values, and some peptides were not. The aim of this research was to determine the effects of a variety of conditions on the synthetic ovine peptides. The research presented in this thesis found that the ovine antimicrobial peptides were very robust. The activity of the peptides was reduced in physiological conditions, such as in the presence of cations and at acidic pH; however they were sti l l relatively active. The peptides were also resistant to heat. They showed no decrease in activity when heated to temperatures up to 80°C, and had only a slight decrease in activity when heated to temperatures above this, or when autoclaved. To further investigate the effect of different conditions on the antimicrobial peptides, synergy studies were carried out. This showed that the ovine antimicrobial peptides worked signifIcantly better in combination with each other than alone. This means that it would be better to use a mixture of peptides in a product, as opposed to a single purifIed peptide. This would have the advantage of reducing the number of downstream process purifIcation steps required to create the product, and reducing the concentration of the peptides required, both of which would make the process more cost effective. The fIfth objective of the research presented in this thesis was to determine whether it IS possible to produce an active antimicrobial extract on a scale larger than that used in the laboratory, using industrial-style equipment. The literature search did not fInd any previous cases where antimicrobial peptide had been extracted on a large scale. The research presented in this thesis showed that it was possible to produce the antimicrobial extract on a scale larger than that in the laboratory using equipment similar to that found in industrial processes. A number of successful pilot-scale trials were carried out. The 1 73 Chapter 9 . Conclusions laboratory process was simplified from eleven unit operations to only SIX necessary unit operations. For each pilot-scale extraction the yield was determined. The yields were between 260 and 300 units per l itre of blood, which is sufficient to make 87 to 90mL of crude extract at a concentration three times that required to inhibit 1 05 CFU/mL E. coli 0 1 1 1 from each litre of blood. This concentration was chosen because the crude extract was active against all the food pathogens tested at either equal to, or two-fold higher than, the MIC for E. coli 0 1 1 1 . The MIC and TEM tests carried out with the crude extract confirmed that it would be best to use the peptides in a mixture. The crude extract had higher and more broad-spectrum activity than expected and was able to induce cell lysis, even though it consisted of predominately proline/arginine-rich cathelicidins that do not display these properties in their pure forms. It is unclear whether this was due to the peptides acting synergistically or to other components present in the crude extract that were not characterised. 9.2 RECOMMENDATIONS FOR FUTURE RESEARCH Although the results presented in this thesis yielded a lot of useful information, the results also uncovered more unanswered questions. Areas where further fundamental research on ovine antimicrobial peptides could be carried out include: 1 ) Investigating the reason for the variations in the OaBac5 sequence. To date, four different sequences have been identified. These variations may be due to each animal having mult iple copies of the OaBac5 gene that vary slightly, or they may be due to genetic differences between animals because the blood was col lected from more than one animal. 2) Investigating the reason for OaBac7.5 and OaBac 1 1 truncations. Both of these peptides have been iso lated only in truncated forms. This is probably due to enzymatic c leavage of the peptides during the extraction process, but the enzymes responsible have not yet been identified. 3 ) Investigating the reason why the other predicted ovine cathelicidins peptides, especial ly the potent SMAP29, were not isolated from the neutrophil extract. This may be because these peptides are not expressed constitutively. It is possible that they are produced only when infection occurs. Alternatively, these peptides may be expressed in different parts of the animal instead of the blood. 1 74 Chapter 9 - Conclusions 4) Investigating the reason for the different mechanisms of action used by the different peptides . This may provide insurance against an infect ion, because if one mechanism is not successful, then the other may still be able to inhibit the invading organism. SMAP29 itself may also be able to use more than one mechanism of action, but further work is required to confirm this. 5) Investigating the use of scanning electron microscopy to image the surface of treated cells. This technique should give information similar to that expected from the atomic force microscope; however, it may be less technically difficult. 6) Investigating other properties of the peptides. Some other antimicrobial peptides have proven to have antiviral, anti-tumour and anti-inflammatory activity, so these areas should also be studied. The results presented in this thesis indicated that ovine antimicrobial peptides may be suitable for use in commercial products. Areas where further research into the possible commercialisation of the crude antimicrobial extract could be carried out include: 1 ) Investigating the effectiveness of the crude antimicrobial extract as a biopreservative for chil led lamb products. This research showed that the antimicrobial peptides retained their activity in a variety of conditions, which indicates that they may be useful as a biopreservative. However, further studies are required to determine its efficacy. The best way to apply the extract to the meat cuts should also be tested. 2) I nvestigating the use of the crude antimicrobial extract as a topical cream. This could be an antiseptic cream for cuts and grazes, an anti-fungal cream for infections such as athletes' foot, or an anti-inflammatory cream for conditions such as arthritis. The antimicrobial extract could also be added to existing products, such as lanolin cream, to enhance the product and add value. 3) Investigating the various types of industrial-scale equipment available to carry out each step in the extraction process. Further pilot-scale trials are required to determine the most effective process for producing the crude extract. It is important that the process is optimised so that the maximum amount of product can be produced. More accurate cost analysis will be possible once the best industrial process has been defined. 1 75 Chapter 9 - Conclusions 4) Investigating the possible products that can be produced from the blood plasma and red blood cells fractions. Further process development and cost analysis is also required to include these product streams. 9.3 F INAL CONCLUSION The overall objective of this research project was to gam an understanding of the antimicrobial peptides that occur naturally in ovine blood, so that their potential to be utilised in high-value products could be assessed. This research has shown that the antimicrobial peptides found in the crude neutrophil extract produced from ovine blood have stable, broad­ spectrum activity, and that they have the potential to be effective biopreservatives and topical antiseptic agents. This research also showed that the peptides were significantly more active in combination than alone, which indicates that the use of the peptides as a mixture would be best. Finally, this research showed that is was possible to produce the crude ant imicrobial peptide on a pilot-scale and that the industrial-scale process required would be relatively simple. Therefore, it is recommended that further work be carried out to develop the extraction process further and to invest igate the appl ications of the peptides with the aim of commercialisation in the future. 1 76 References REFERENCES Aarbiou, 1 . , Ertmann, M. , van Wetering, S. , van Noort, P. , Rook, D. , Rabe, K.F . , L itvinov, S .V. , Han, 1 . , van Krieken, 1 .M. , de Boer, W.I . , and H iemstra, P .S. (2002) Human neutrophil defensins induce lung epithelial cell proliferation in vitro. Journal of Leukocyte B iology, 22, 1 67- 1 74. Abbas, A K. and L ichtman, AH. (2004) Basic Immunology - Functions and Disorders of the Immune System Second Edition. Saunders, Philadelphia, USA Agerberth, 8., Charo, 1 . , Werr, 1 . , Idali, F., L indborn, L., Kiessling, R., Jornvall, H., Wigzell, H . , and Gudmundsson, G.H. (2000) The human antimicrobial and chemotactis peptides LL-3 7 and a-defensins are expressed by specific lymphocyte and monocyte populations . Antibacterial Peptides in Lymphocytes, 96(9), 3086-3093 . Agerberth, 8 . , Gunne, H . , Oderberg, 1., Kogner, P . , Boman, H.G. , and Gudmundsson, G.H. ( 1 995) FALL-39, a putative human peptide antibiotic, is cysteine-free and expressed in bone marrow and testis. Proceedings of the National Academy of Science USA, 92, 1 95- 1 99. Agerberth, B. , Lee, 1 .-Y., Bergman, T. , Carlquist, M. , Boman, H.G. , Mutt, V. , and Jornvall, H. ( 1 99 1 ) Amino acid sequence of PR-39. European Journal of B iochemistry, 202, 849- 854. Ahimou, F. , Touhami, A , and Dufrene, YF. (2003) Real-time imaging of the surface topography of living yeast cells by atomic force microscopy. Yeast, 20, 25-30. Alcouloumre, M.S . , Ghannoum, M.A, Ibrahim, AS. , Selsted, M.E ., and Edwards, 1.E.1. ( 1 993) Fungicidal properties of defensin NP- l and activity against Cryptococcus neoformans in vitro. Antimicrobial Agents and Chemotherapy, 3 7( 1 2), 2628-2632. Aley, S .8 . , Zimmerman, M., Hetsko, M., Selsted, M.E. , and Gillin, F.D. ( 1 994) Killing of Giardia lamblia by cryptdins and cationic neutrophil peptides. Infection and Immunity, 62( 1 2) , 5297-5403 . Amro, N.A, Kotra, L .P ., Wadu-Mesthrige, K. , Bulychev, A, Mobashery, S . , and Lui, G.-Y (2000) High-resolution atomic force microscopy studies of th Escherichia coli outer membrane: structural basis for permeability. Langmuir, 1 6, 2789-2796. Amsterdam, D. ( 1 996) Susceptibility testing of antimicrobials in liquid media In: Lorian, V . Antibiotics in Laboratory Medicine. Willimas and Wilkins, B latimore. Andreu, D . , Mitchell, S. A , Silveira, A M. V. , Boman, H. G. , Boman, A, and Merrifield, R. 8 . Antibacteral peptides designed as analogs or hybrids of cecropines and melettin. 92;40:429-436. Ariyapitipun, T., Mustapha, A, and C larke, AD. ( 1 999) Microbial shelf life determination of vacuum-packages fresh beef treated with polylactic acid, lactic acid, and nisin solutions. Journal of Food Protection, 62(8), 9 1 3-920. Arzese, A, Skerlavaj, 8. , Tomasinsig, L . , Gennaro, R., and Zanett i, M. (2003) Antimicrobial activity of SMAP-29 against the Bacteroides fragilis group and clostridia. J Antimicrob Chemother , . Aymerich, T . , Garriga, M. , Ylla, J . , Vallier, J . , Monfort, 1 .M., and Hugas, M. (2000) Application of enterocins as biopreservatives against Listeria innocua in meat products. Journal of Food Protection, 63 (6), 72 1 -726. 1 77 References Bagella, L . , Scocchi, M. , and Zanetti, M. ( 1 995) cDNA sequences of three sheep myeloid cathelicidins. FEBS Letters, 376, 225-228. Bals, R., Goldman, M.J . , and Wilson, l M . ( 1 998) Mouse b-defensin 1 is a salt-sensitive antimicrobial peptide present in epithelia of the lung and urogenital tract . Infection and Immunity, 66(3), 1 225- 1 232. Bals, R. , Lang, c . , Weiner, DJ. , Vogeimeier, c . , Welsch, U. , and Wilson, l M . (200 1 ) Rhesus monkey (Macaca mulatta) mucosal antimicrobial peptides are close homologues of human molecules. Cl inical and Diagnostic Laboratory I mmunology, 8(2) , 370-3 75 . Bals, R. , Wang, X., Zasloff, M.A, and Wilson, lM. ( 1 998) The peptide antibiotic LL- 37/heAP-I8 is expressed in epithelia of the human lung where it has antimicrobial activity at the airway surface. Proceedings of the National Academy of Sc ience USA, 95, 954 1 -9546. Bals, R. and Wi lson, lM. (2003) Cathelicidins - a family of multifuncLional antimicrobial peptides. Cellular and Molecular Life Sciences, 60, 7 1 1 -720. Barbuddhe, S .B . , Malik, S .V.S . , and Bhi legaonkar, K.N. ( 1 999) Growth inhibition of Listeria monocytogenes by commerical nisin and lactic acid in raw buffalo meat mince . Journal of Food Science and Techno logy, 36(4), 320-324. Bastian, A and Schafer, H. (200 1 ) Human alpha-defensin I (HNP-I) inhibits adenoviral infection in vitro . Regulatory Peptides, 1 0 1 ( 1 -3), 1 5 7- 1 6 1 . Bateman, A, MacLeod, R.l, Lembessis, P., Hu, l , Esch, F . , and Solomon, S . ( 1 996) The isolation and characterisation of a novel corticostaLin/defensin-like peptide form the kidney. Journal of B io logical Chemistry, 27 1 ( \ 8), 1 0654- \ 0659. Bechinger, 8 . Structure and function of channel- forming peptides: Magainins, cecropins, melitt in and alamethic in. 97;9 1 56 : 3 : 1 97-2 1 1 . Bellm, L. , Lehrer, R.I . , and Ganz, T. (2000) Protegrins: new antibiotics of mammalian origin. Expert Opinion on Investigational Drugs, 9(8), 1 73 1 - 1 742. Benincasa, M., Skerlavaj , 8 . , Gennaro, R , Pellegrini, A, and Zanetti, M. (2003) In vitro and in vivo antimicrobial activity of two a-helical cathelcidin peptides and of their synthetic analogs. Peptides, 24, 1 723- 1 73 1 . Blondelle, S .E . and Houghten, R.A ( 1 992) Design of model peptides having potent antimicrobial activities. Biochemistry, 3 1 , 1 2688- 1 2694. Bodnar, Rl and Hadj imarkou, M.M. (2003) Endogenous opiates and behavior: 2002. Peptides, 24, 1 24 1 - 1 302. Boman, H.G. , Agerberth, B . , and Boman, A ( 1 993) Mechanisms of action on Escherichia coli of cecropin P 1 and PR-39, two antibacterial peptides from pig intestine. I nfection and I mmunity, 6 1 (7), 2978-2984. Bonass, W.A., H igh, AS., Owen, P.1., and Devine, D.A ( 1 999) Expression of b-defensin genes by human salivary galnds. Oral Microbiology and I mmunology, 1 4, 3 7 1 -374. Boonaert, C.1.P. , Rouxhet, P.G., and Dufrene, Y.F . (2000) Surface properties of microbial cells probed at the nanometre scale with atomic force microscopy. Surface and Interface Analysis, 30, 32-35 . Borenstein, L .A, Selsted, M.E. , Lehrer, R. I . , and Mi l ler, IN. ( 1 99 1 ) Antimicrobial activity of rabbit leukocyte defensins against Treponema pallidum subsp. pallidum. Infection and 1 78 References lnununity, 59(4), 1 359- 1 367 . Braga, P .C . and Ricc i, D. ( 1 998) Atomic force microscopy: Application to investigation of Escherichia coli morphology before and after exposure to cefodizime. Ant imicrobial Agents and Chemotherapy, 42( 1 ), 1 8-22. Brockus, C .W., Jackwood, M . W. , and Harmon, B .G. ( 1 998) Characterization of fJ-defensin prepropeptide mRNA from chicken and turkey bone marrow. Animal Genetics, 29, 283- 289. Broekaert, W.F. , Canunue, B .P .A., De Bolle, M.F.C. , Thevissen, K., De Samblanx, G.W., and Osborn, R. W. ( 1 997) Antimicrobial peptides from plants. Critical Reviews in P lant Sciences, 1 6(3) , 297-323 . Brogden, K.A. , Ackermann, M .R., McCray, P .B.J . , and Huttner, K.M. ( 1 999) Differences in the concentrations of small, anionic, antimicrobial peptides in bronchoalveolar lavage fluid and in respiratory epithelia of pateints with and without cystic fibrous. Infection and lnununity, 67(8), 4256-4259. Brogden, K.A. , Kalfa, V .c., Ackermann, M.R. , Palmquist, D.E., McCray, P.B.J . , and Tack, B .F . ( 200 1 ) The ovine cathelicidin SMAP29 kills ovine respiratory pathogens in vitro in an ovine model of pulmonary infection. Antimicrobial Agents and Chemotherapy, 45 ( 1 ), 3 3 1 -334. Brogden, K. A., Lehrnkuhl, H. D. , and Cutlip, R. e . ( 1 998) Pasteurella haemolytica complicated respiratory infections in sheep and goats. Veterinary Research, 29, 233-254. Cabiaux, Agerberth, B., Johansson, J . , Homble, F., Goorrnaghtigh, E., and Ruysschaert, 1.-M. ( 1 994) Secondary structure and membrane interaction of PR-39, a Pro+Arg-rich antibacterial peptide. European Journal of B iochemistry, 224, 1 0 1 9- 1 027. Carlson, G.P. and Kaneko, J .1. ( 1 973) Isolation of leukocytes from bovine peripheral blood. Proceedings of the Society for Experimental B io logy and Medicine, 1 42, 853-856. Chan, W., Brown, 1. , Lee, S .M. , and Buss, D.H. ( 1 995) Meat, Poultry and Game - Fifth supplement to the fifth edition of McCance and Widdowson 's "The Composition of Foods ". The Royal Society of Chemistry, Cambridge, UK. Chen, 1. , Falla, T.J . , L iu, H. , Hurst, M.A., Fuj i i, c.A., Mosca, D.A., Embree, 1.R., Loury, 0.1., Radel, P.A., Chang, C.C. , Gu, L. , and F iddes, 1.e. (2000) Development of prate grins for the treatment and prevention of roal mucositis: structure-activity relationships of synthetis protegrins analogues. Biopolymers, 55, 88-98. Chertov, 0 . , Michiel, D.F. , Xu, L., Wang, 1.M., Tani, K. , Murphy, W.J. , Long, D.L . , Taub, 0 . 0 . , and Oppenheim, J .J . ( 1 996) Indentification of defensin-l, defensin-2 and CAP371azurocidin as T-cell cehmoattractant proteins released from interleukin-8- stimulated neutrophils. Journal of Biological Chemistry, 27 1 (6), 2935-2940. Cho, Y . , Turner, 1 .S . , Dinh, N.-N., and Lehrer, R.I . ( 1 998) Activity of protegrins against yeast-phase Candida albicans. Infection and lnununity, 66(6), 2486-2493 . Cole, A.M. , Shi, 1 .S . , Ceccarelli, A., Kim, Y.H. , Park, A. , and Ganz, T . (200 1 ) Inhibition of neutrophil elastase prevents cathelicidin activation and impairs clearance of bacteria from wounds. Blood, 97( 1 ), 297-304. Cowland, 1.B . , Johnson, A.H., and Borregaard, N . ( 1 995) hCAP-l8, a cathelinlpro­ bactenecin-like protein of human neutrophil specific granules. FEBS Letters, 368, 1 73 - 1 76. 1 79 References Daher, K., Selsted, M.E. , and Lehrer, R. I . ( 1 986) Direct inactivation of viruses by human granulocyte defensins. Journal of V irology, 60(3), 1 068- 1 074. Dathe, M. , Schumann, M., Wieprecht, T., Winkler, A, Beyermann, M. , Krause, E . , Matsuzaki, K., Osamu, M. , and Birnert, M. ( 1 996) Peptide helicity and membrane surface charge modulate the balance of electrostatic and hydrophobic interaction with lipid bilayers and biological membranes. Biochemistry, 35(38), 1 26 1 2- 1 2622. Dathe, M., Weipresht, T., Niko lenko, H . , Handel, L. , Maloy, W.L . , MacDonald, D .L . , Beyermann, M . , and Bienert, M. ( 1 997) Hydrophobicity, hydrophobic moment and angle subtended by charges residues modulate antibacterial and haemolytic activity of amphipathic helical peptides. FEBS Letters, 403(2), 208-2 1 2 . Davis, E.G. , Sang, Y., and Blecha, F. (2004) Equine fJ-defensin- l :full-length cDNA sequence and tissue expression. Veterinary Immuno logy and Immunopathology, 99, 1 27- 1 32 . De Lucca, AJ. and Walsh, T.J . ( 1 999) Antifungal peptides: novel therapeutic compounds against emerging pathogens. Antimicrobial Agents and Chemotherapy, 43( 1 ), 1 - 1 l . Del Sal, G., Storic i, P. , Schneider, c . , Romeo, D. , and Zanetti, M . ( 1 992) cDNA cloning of the neutrophil bactericidal peptide indolicidin. Biochemical and Biophysical Communications, 1 87( 1 ), 467-472. Devine, D.A. and Hancock, R.E . (2002) Cationic peptides: distribution and mechanisms of resistance. Current Pharmaceutical Deseases, 8(9), 703 - 1 4 . Diamond, G. , lones, D.E. , and Bevins, c .L . ( 1 993 ) Airway epithelial cells are the site of expression of mammalian antimicrobial peptide gene. Proceedings of the National Academy of Science USA, 90, 4596-4600. Diamond, G., Zasloff, M.A, Eck, H., Brasseur, M., Maloy, W.L. , and Bevins, c . L . ( 1 99 1 ) Tracheal antimicrobial peptide, a cysteine-rich peptide from mammalian tracheal mucosa: peptide isolation and cloning of cDNA. Proceedings of the National Academy o f Science USA, 88, 3952-3956. Dimarcq, J . -L. , Bulet, P., Hetru, C., and Hoffmann, J . ( 1 998) Cysteine-rich antimicrobial peptides in invertebrates. Biopolymers, 47, 465-477. Dioguardi, N., Agostoni, A., F iorelli, G., and Lomanto, B . ( 1 963 ) Charcterization of lactic dehydrogenase of normal human granulocytes. Journal of Laboratory and C l inical Medicine, 6 1 (5) , 7 1 3 -723 . Draper, N.R. and Smith, H. ( 1 98 1 ) Applied regression analysis, 2nd Edition. Wiley, New York. Dufrene, Y. F. (2000) Atomic force mucroscopy, a powerful tool in microbiology. Journal of Bacteriology, 1 84( 1 9), 5205-52 1 3 . Duits, L .A., Langermans, J .A.M. , Paltansing, S . , van der Straaten, T. , Vervenne, R.AW., Frost, P .A. , Heimstra, S . , Thomas, A.W. , and Nibbering, P .H . (2000) Expression of beta­ defensin-l in chimpanzee (Pan troglodytes) airways. Journal of Medical Promatology, 29(5), 3 1 8-323 . Eggleton, P., Gargan, R., and Fisher, D. ( 1 989) Rapid method for the isolation of neutrophils in high yield without the use of dextran or density gradient polymers. Journal of I mmunological Methods, 1 2 1 , 1 05- 1 1 3 . Eisenhauer, P.B. , Harwig, S .S .L. , and Lehrer, R . I . ( 1 992) Cryptdins: antimicrobial defensins of the murine small intestine. Infection and I mmunity, 60(9), 3 556-3565. 1 80 References Eisenhauer, P .B . , Harwig, S .S .L . , Szklarek, D . , Ganz, T., Selsted, M.E. , and Lehrer, R.I . ( 1 989) Purification and antimicrobial properties of three defensins from rat neutrophils. Infection and Immunity, 57(7), 202 1 -2027. Ennahar, S. , Sashihara, T. , Sonomoto, K., and Ishizaki, A. (2000) Class IIa bacteriocins: bio!,ynthesis, structure and activity. FEMS Microbiology Reviews, 24, 85- 1 06. Ernst, R.K. , Guina, T. , and Miller, S . 1 . ( 1 999) How intracellular bacteria survive: surface modifications that promote resistance to host innate immune responses. Journal of Infectious D iseases, 1 79(Supp 2), S326-S330. Evans, E .W., Beach, G.G., Wunderlich, 1 . , and Harmon, B.G. ( 1 994) Isolation of antimicrobial peptides from avian heterophils. Journal of Leukocyte B io logy, 56, 66 1 - 665 . Evans, E .W. and Harmon, B .G. ( 1 995) A review of antimicrobial peptides: defensins and related cationic peptides. Veterinary Clinical Pathology, 24(4), 1 09- 1 1 6. Falla, TJ., Karunaratne, D .N., and Hancock, R.E .W. ( 1 996) Mode of action of the antimicrobial peptide indolicidin. Journal of B io logical Chemistry, 27 1 (32), 1 9298- 1 9303 . Fernando, T. ( 1 976) Blood processing - a feasibility study of some alternatives . MIRINZ, Technical Report, Fidai, S. , Farmer, S. W. , and Hancock, R.E. W. ( 1 997) Interaction of cationic peptides with bacterial membranes In: Shafer, W.M. Humana Press Inc, Totowa, N1. F inlay, B .B . and Hancock, R.E .W. (2004) Can innate immunity be enhanced to treat microbial infections? Nature Reviews Microbiology, 2, 497-504. F ischer, P .M. , Krausz, E. , and Lane, D.P. (200 1 ) Cellular delivery of impermeable effector molecules in the form of conjugates with peptides capable of mediating membrane translocation. B ioconjugate Chemistry, 1 2(6), 825-4 1 . F itzGerald, R.1. ( 1 998) Potential uses of caseinophophopeptides. I international Dairy Journal, 8 , 45 1 -457. Frank, R.W., Gennaro, R. , Schneider, K. , Przybylski, M., and Romeo, D. ( 1 990) Amino acid sequences of two proline-rich bactenecins. Journal of B io logical Chemistry, 265(3 1 ), 1 887 1 - 1 8874. Fricker, L.D. ( 1 988) Carboxypeptidase E. Annual Reviews in Physiology, 50, 309-32 1 . Friedrich, c.L. , Rozek, A., Patrzykat, A. , and Hancock, R.E.W. (200 1 ) Structure and mechanism of action of an indolicidin peptide derivative with improved activity against gram-positive bacteria. Journal of B io logical Chemistry, 276(26), 240 1 5-24022. Frohm, M., Agerberth, B., Ahangari, G., Stahle-Backdahl, M., Liden, S., Wigzell, H., and Gudmundsson, G .H . ( 1 997) The expression of the gene coding for the antibacterial peptide LL-3 7 induced in human keratinocytes during infalmmatory disorders. Journal of B iological Chemistry, 24, 1 5258- 1 5263 . Fuse, N . , Hayashi, Y . , Fukata, 1 . , Tominaga, T., Ebisui, 0 . , Satoh, Y. , Isohara, T., Uno, I . , and Imura, H . ( 1 993) Pur(fication and characterisation of new anti-adrenocorticotropin rabbit neutrophil pep tides (defensins) . European Journal of B iochemistry, 2 1 6, 653-659. Gad, M. and Ikai, A. ( 1 995) Method for immobilising microbial cells on gel surface for dynamic AFM studies. Biophysical Journal, 69, 2226-2233 . 1 81 References Gallo, R.L . , Kim, K.J . , Bernfield, M., Kozak, e . A , Zanetti, M. , Merluzzi, L . , and Gennaro, R. ( 1 997) Identification of CRAMP, a cathelin-related antimicrobial peptide expressed in the embryonic and adult mouse. Journal of B io logical Chemistry, 272(20), 1 3088- 1 3093 . Ganz, T. ( 1 999) Defensins and host defense. Science, 286, 420-42 1 . Gao, Y., Lecker, S. , Post, M .J., Hietaranta, A l , L i, J . , Yolk, R. , L i, M. , Sato, K., Saluja, A K. , Steer, M.L . , Goldberg, A.L . , and Simons, M. (2000) Inhibition of ubiquitin­ proteasome pathway-mediated I K13 a degradation by naturally occuring antibacterial peptide. Journal of Clinical Investigation, 1 06(3), 439-448 . Garcia-Graells, C . , Masschalck, B . , and Michiels, e . W. ( 1 999) Inactivation of Escherichia coli in milk by high-hydrostatic-pressure treatment in combination with antimicrobial peptides. Journal of Food Protection, 62( 1 1 ), 1 248- 1 254. Garcia-Olmedo, F, Molina, A, Alamillo, J . M. , and Rodriguez-Palenzuela, P. ( 1 998) Plant defense peptides. Biopolymers (Peptide Science), 98(47), 479-49 1 . Gazit, E., Mil ler, I .R., Biggin , P . e . , Sansom, M.S .P. , and Shai, Y. ( 1 996) Structure and orientation of the mammalian antibacterial peptide cecropin PI within phospholipid membranes. Journal of Molecular B io logy, 258, 860-870. Gennaro, R., Dolzani, L., and Romeo, D. ( 1 983) Potency of bactericidal proteins purified from the large granules of bovine neutrophils. Infection and Immunity, 40(2), 684-690. Gennaro, R., Skerlavaj, B., and Romeo, D. ( 1 989) Purification, composition, and activity of two bactenecins, antibacterial peptides of bovine neutrophils. Infection and Immunity, 57( 1 0), 3 1 42-3 1 46. Gentilucci, L. (2004) New trends in th development of opioid peptide analogues as advanced remideis for pain relief Current Topics in Medicinal Chemistry, 4, 1 9-38 . Giacometti, A , Cirioni, 0 . , Barchiesi, F. , Caselli, F . , and Scal ise, G. ( 1 999) In-vitro activty of polycationic peptides against Cryptosporidium parvum, Pneumocystis carinii and yeast clinical isolates. Journal of Antimicrobial Chemotherapy, 44(3) , 403-406. Giacometti, A, Cirioni, 0 . , Ghiselli, R., Mocchegiani, F., D'Amato, G. , Circo, R., Orlando, F., Skerlavaj, B . , Silvestri, C. , SAba, V . , Zanetti, M. , and Scal ise, G. (2004) Cathelicidin peptide sheep myeloid antimicrobial peptide-29 prevents endotoxin-induced mortality in rat models of septic shock. American Journal of Respiratory and Critical Care Medicine, 1 69, 1 87- 1 94. Goff, lH. , Bhunia, A K. , and Johnson, M.G. ( 1 996) Complete inhibition of low lovels of Listeria monocytogenes on refigerated chicken meat with Pediocin AcH bound to heat­ killed Pediococcus acidilactis cells. Journal of Food Protection, 59( 1 1 ), 1 1 87- 1 1 92. Goldman, M.J. , Anderson, G.M., Stolzenberg, E.D. , Kari, U .P . , Zasloff, M . , and Wilson, lM. ( 1 997) Human beta-defensin-l is a salt-sensitive antibiotic in lung that is inactivated in cysticjibrosis. Cell, 88(4), 553-60. Groisman, E .A. , Parra-Lopez, C., Salcedo, S. , L ips, C.l, and Heffron, F. ( 1 992) Resistance to host antimicrobial peptides is necessary for Salmonella virulence. Proceedings of the National Academy of Sc ience USA, 89, 1 1 939- 1 1 943. Guina, T. , Vi , E .C. , Wang, H. , Hackett, M. , and Mil ler, S . 1 . (2000) A PhoP-regulated outer membrane protease of Salmonella enterica serovar typhimurium promotes resistance to alpha-helical antimicrobial peptides. Journal of Bacteriology, 1 82(1 4), 4077-4086. Guo, L . , L im, K.B., Poduje, e . M . , Daniel, M. , Gunn, J .S . , Hackett, M., and Miller, S . 1 . ( 1 998) 1 82 References Lipid A acylation and bacterial resistance against vertebrate antimicrobial peptides. Cell, 95, 1 89- 1 98. Guthmiller, 1 . M., Vargas, K. G, Srikantha, R. , Schomberg, L. L., Weistroffer, P . L. , McCray, P. B . Jr, and Tack, B. F. (200 1 ) Susceptibilities of oral bacteria and yeast to mammalian cathelicidins. Antimicrobial Agents and Chemotherapy, 45( 1 1 ), 3 2 1 6-32 1 9. Han, 1 .H. (2000) Antimicrobial Food Packaging. Food Technology, 54(3), 56-65. Hancock, R.E. W. ( 1 997) Peptide antibiotics. The Lancet, 349(9049), 4 1 8-422. Hancock, R.E .W. (2000) Cationic antimicrobial peptides: towards clinical applications. Expert Opinion in Investigative Drugs, 9(8), 1 723- 1 729. Hancock, R.E.W. (200 1 a) Cationic peptides: ef ectors in innate immunity and novel antimicrbials. The Lancet Infectious D iseases, 1 , 1 56- 1 64. Hancock, R.E. W. (200 1 b) Cationic peptides: effectors in innate immunity and novel antimicrobials. Lancet Infectious Diseases, 1 (3) , 1 56-64. Hancock, R.E .W. , Raffle, V.1 . , and nicas, T. I . ( 1 98 1 ) Involvement of th outer membrane in gentamicin and streptomycin uptake and killing in Pseudomonas aeruginosa. Antimicrobial Agents and Chemotherapy, 1 9(6), 777-785 . Hancock, R. E . W. and Rozek, A. (2002) Role of membrane on the activities of antimicrobial cationic peptides. FEMS Microbiology Letters, 206, 1 43 - 1 49. Harwig, S .S .L. , Chen, N.P. , Park, A.S.K. , and Lehrer, R.1 . ( 1 993) Purification of cysteine-rich bioactive peptides from leukocytes by continuous acid-urea-polyacrylamide gel electrophoresis. Analytical B iochemistry, 208, 382-3 86. Harwig, S .S .L . , Kokryakov, V.N., Swiderek, K.M. , Aleshina, G.M. , Zhao, C., and Lehrer, R. 1 . ( 1 995) Prophenin-1 , an exceptionally proline-rich antimicrobial peptide from porcine leukocytes. FEBS Letters, 362, 65-69. Hashimoto, Y., Nagaoka, I., and Yamashita, T. ( 1 993) Purification of the antibacterial fragments of guinea-pig major basic protein. Biochimica et Biophysica Acta, 1 23, 236- 242. H irata, M . , Shimomoura, Y., Yoshida, M. , Morgan, 1 .G. , Palings, I . , Wilson, D . , Yen, M.H. , Wright, S .c. , and Larrick, 1 . W. ( 1 994) Characterisation of rabbit cationic protein (CAP 1 8) with lipopolysaccharide-inhibitory activity. Infection and Immunity, 62(4), 1 42 1 - 1 426. Holzapfel, W., Geisen, R. , and Schillinger, U . ( 1 995) Biological preservation of foods with reference to protectve culutres, bacteriocins and food-grade enzymes. International Journal of Food Microbiology, 24, 343-362. Hoover, D .M . , Rajashankar, K.R., B lumenthal, R. , Puri, A. , Oppenheim, J .1 . , CHertov, 0 . , and Lubkowski, 1. (2000) The structure of human beta-defensin-2 shows evidence of higher order oligomerization. Journal of B iological Chemistry, 275(42), 329 1 1 -329 1 8 . Hristova, K., Selsted, M.E., and White, S .H . ( 1 997) Critical role of lipid composition in membrane permeabilization by rabbit neutrophil defensins . Journal of B iological Chemistry, 272(39), 24224-24233 . Huang, H.-J . , Ross, c.R., and B lecha, F . ( 1 997) Chemoattractant properties of PR-39, a neutrophil antibacterial peptide. Journal of Leukocyte B iology, 6 1 , 624-629. Huttner, K .M. , Brezinski-Caliguri, D.J . , Mahoney, M.M. , and D iamond, G. ( 1 998) 1 83 References Antimicrobial peptide expression is developmentally regulated in the ovine gastrointestinal tract. Journal of Nutrit ion, 1 28, 297S-299S. Huttner, K.M., Kozak, c.A., and Bevins, c .L . ( 1 997) The mouse genome encodes a single homolog of the antimicrobial peptide human �defensin J . FEBS Letters, 4 1 3, 45-49. Huttner, KM. , Lambeth, M.R., Burkin, H . R. , B urk in, 0.1., and Broad, T.E. ( 1 998) Localisation and genomic organisation of sheep antimicrobial peptide genes. Gene, 206, 85-9 1 . Hwang, P .M. and Vogel, H.1 . ( 1 998) Structure-function relationships of antimicrobial peptides. Biochemistry and Cell B io logy, 76((2-3 )), 235-246. I keda, Y., Young, L.H. , Scalia, R., Ross, c .R., and Lefer, A.M. (200 1 ) Pr-39, a proline/arginine-rich antimicrobial peptide, exerts cadioprotective effects in myocardial ischemia-reperfusion. Cardiovascular Research, 49, 69-77. Iwata, T., Lee, S., Oishi, I . , Aoyagi, H., Ohno, M., Anzai, K, Kirino, Y., and Sug ihara, G. ( 1 994) Design and synthesis of amphipathic 3 lo-helical peptides and their interactions with phospholipid bilayers and ion channel formation. Journal of B io logical Chemistry, 269( 7), 4928-493 3 . Jia, H .P., Mil ls, 1 .N . , Barahmand-Pour, F. , Nishimura, D. , Mallampal i, R .K, Wang, G. , Wiles, K, Tack, B.F . , Bevins, c.L. , and McCray, P .B.1. ( 1 999) Molecular cloning and characterzation of rat genes encoding homologues of human �defensins. I nfection and Immunity, 67(9), 4827-483 3 . Jia, H .P . , Schutte, B .c . , Schudy, A., L inzmeier, R . , Guthmiller, 1 .M . , Johnson, G .K. , Tack, B .F . , Mitros, 1 .P. , Rosenthal, A., Ganz, T., and McCray, P.B.1 . (200 1 ) Discovery of new human �defensins using a genomics-based approach. Gene, 263, 2 1 1 -2 1 8 . Jia, H .P . , Wowk, S.A., Schutte, B.c., and Lee, S.K. (2000) A novel murine �defensin expressed in tongue, esophagus, and trachea. Journal of B io logical Chemistry, 275(43), 333 1 4-33320. Johansson, 1 . , Gudmundsson, G.H., Rottenberg, M.E . , Brendt, K.D . , and Agerberth, B . ( 1 998) Conformation-dependent antibacterial activity of the naturally occuring human peptide LL-3 7. Journal of B io logical Chemistry, 273(6), 3 7 1 8-3 724. Johnstone, S.A., Gelmon, K, Mayer, L .D. , Hancock, R.E. W., and Bal lay, M .B . (2000) In vitro characterisation of the anticancer activity of membrane-active cationic peptides. I. Peptide-mediated cytotoxicity and peptide-enhanced cytotoxic activity of doxorubicin against wild-type and p-glycoprotein over-expressing tumor cell lines. Anti-cancer Drug Design, 1 5(2), 1 5 1 - 1 60. Jones, D.E. and Bevins, c.L. ( 1 993) Defensin-6 mRNA in human Paneth cells: implications for antimicrobial pep tides in host defense of the human bowel. FEBS Letters, 3 1 5(2), 1 87- 1 92. Juvvadi, P., Vunnam, S . , Yoo, B. , and Merrifield, R.B. ( 1 999) Structure-activity studies of normal and retro pig cecropin-melittin hybrids. Journal of Peptide Research, 53, 244-25 1 . Kagan, B .L . , Se lsted, M .E. , Ganz, T., and Lehrer, R. 1 . ( 1 990) Antimicrobial defensin peptides form voltage-dependent ion-permeable channels in planar lipid bilayer membranes. Proceedings of the National Academy o f Science USA, 87, 2 1 0-2 1 4. Kalfa, V . C, Jia, H . P . , Kundle, R. A., McCray, P. B. Jr, Tack, B . F . , and Brogden, K A. (200 1 ) Congeners of SMAP 29 kill ovine pathogens and induce ultrastructural damage in 1 84 References bacterial cells. Antimicrobial Agents and Chemotherapy, 45( 1 1 ), 3256-326l . Kamysz, W., Okroj, M. , and Lukasiak, J . (2003) Novel properties of antimicrobial peptides. Acta B iochimica Polonica, 50(2), 46 1 -469. Kasas, S. , Fellay, B., and Cargnello, R. ( 1 994) Observations of the action of penicillin on Bacillus subtilis using atomic force microscopy: technique for the preparation of bacteria. Surface and Interface Analysis, 2 1 , 400-40 1 . Kasas, S . , Gotzos, V . , and Cel io, M .R. ( 1 993) Observation of living veils using the atomic force microscope. Biophysical Journal, 64, 539-544. Kasas, S. and Ikai, A ( 1 995) A method for anchoring round shape cells for atomic force microscopy imaging. Biophysical Journal, 68, 1 678- 1 680. Kayer, H. and Meisel, H. ( 1 996) Stimulation of human peripheral blood lymphocytes by bioactive pep tides derivedfrom bovine milk proteins. FEBS Letters, 383 , 1 8-20. Kim, W.1. ( 1 993) Bacteriocins of lactic acid bacteria: their potentials as food preservative. Food reviews International, 9(2), 299-3 1 3 . Kirikae, T. , Hirata, M. , Yamasu, H . , Kirikae, F. , Tamura, H . , Kayama, F . , Nakatsuka, K., Yokochi, T. , and Nakano, M. ( 1 998) Protective effects of human 18-kilodalton cationic antimicobial protein (CAP 1 8)-derived peptide against murine endotoxemia. Infection and Immunity, 66(5), 1 86 1 - 1 868 . Koczulla, A R and Bals, R. (2003) Antimicrobial peptides - Current status and therapeutic potential. Drugs, 63 (4), 389-406. Koczulla, R., von Degenfeld, G., Kupatt, c., Krotz, F., Zahler, S., Gloe, T., Issbrucker, K., Unterberger, P., Zaiou, M . , Lebherz, c., Karl, A , Raake, P . , Pfosser, A, Boekstegers, P. , Welling, M. M. , Welsch, U . , Welsch, U. , H iemstra, P . S . , Hiemstra, P . S . , Vogelmeier, C. , Vogelmeier, c . , Gallo, R L. , C lark, D . P . , Clauss, M. and Bals, R (2003) An angiogenic role for the human peptide antibiotic LL-37IhCAP-18. The Journal of Clinical Investigation, 1 1 1 ( 1 1 ), 1 665- 1 672. Kokryakov, V.N., Harwig, S .S .L ., Panyutich, E.A , Shevchenko, A A , Aleshina, G.M. , Shamova, O.V. , Korneva, H.A, and Lehrer, RI . ( 1 993) Protegrins: leukocyte antimicrobial pep tides that combine features of carticostatic defensins and tachyplesins. FEBS Letters, 327(2), 23 1 -236. Kondejewski, L .H . , Jelokhani-Niaraki, M. , Farmer, S .W., L ix, B . , Kay, C.M., Sykes, B .D . , Hancock, R.E . W. , and Hodges, R S . ( 1 999) Dissociation of antimicrobial and hemolytic activities in cyclis peptide diasteromers by sysematicm alterations in amphipathicity. Journal of B iological Chemistry, 274( 1 9), 1 3 1 8 1 - 1 3 1 92. Korhonen, H. and Philaanto, A (2003) Food-derived bioactive pep tides - opportunities for designing future foods. Current Pharmaceutical Design, 9, 1 297- 1 308. Kris-Etherton, P .M. , Lefevre, M . , Beecher, G .R. , Gross, M.D., Keen, c . L . , and Etherton, T .D . ( 2004) Bioactive compounds in nutrition and health-research methodologies for establisheng biologica function: the antioxidant and anti-injlammatory effects of jlavonoids and atherosclerosis. Annual Reviews in Nutrition, 24, 5 1 1 -538 . Krisanaprakornkit, S . , Weinberg, A , Perez, C .N. , and Dale, B .A ( 1 998 ) Expression of the peptide antibiotic human �defensin 1 in cultures gingival epithelia cells and gengival tissue. Infection and Immunity, 66(9), 4222-4228. Ladokhin, AS. and White, S .H . (200 1 ) Protein chemistry at membrane interfaces: non- 1 85 References additivity of electrostatic and hydrophobic interactions. Journal of Molecular B io logy, 309(3), 543-552. Larrick, J. W., Hirata, M., Balint, RF., Lee, J . , Zhong, J . , and Wright, S .c. ( 1 995) Human CAP 18: a novel antimicrobial lipopolysaccharide-binding protein. I nfection and Immunity, 63 (4), 1 29 1 - 1 297 . Larrick, J . W. , H irata, M. , Zheng, H. , Zhong, J . , Bolin, D . , Cavaillon, J . -M. , Warren, H . S . , and wright, S . C. ( 1 994) A novel granulocyte-derived peptide with lipopolysaccharide­ neutralizing activity. Journal of Immunology; 1 52( 1 ) , 23 1 -240. Larrick, J .W., Morgan, J .G. , Palings, I . , Hirata, M., and Yen, M.H. ( 1 99 1 ) Complementary DNA sequence of rabbit CA P 18 - A unique lipopolysaccharide binding protein. Biochemical and Biophysical Communications, 1 79( 1 ), 1 70- 1 75. Lawyer, c . , Pai, S . , Watabe, M., Borgia, P. , Mashirno, T., Eagleton, L. , and Watabe, K. ( 1 996) Antimicrobial activity of a 13 amino acid tryptophan-rich peptide derived from a putative porcine precursor protein of a novel family of antibacterial peptides. FEBS Letters, 390, 95-98. Lee, D . G., Kim, P. I . , Park, Y. , Park, S .- c . , Woo, E . R., and Hahm, K. S . (2002) Antifungal mechanism of SMAP-29 (1 - 18) isolated from sheep mRNA against Trichosporon beigelii. Biochemical and Biophysical Research Communications, 295, 59 1 -596. Lee, S . , Mihara, H., Aoyagi, H., Kato, T. , Izumiya, N., and Yamasaki, N. ( 1 986) Relationship between antimicrobial activity and amphiphilic property of basic model peptides. Biochimica et Biophysica Acta, 862, 2 1 1 -2 1 9. Lehrer, RI . , Daher, K. , Ganz, T . , and Selsted, M.E. ( 1 985) Direct inactivation of viruses by MCP-1 and MCP-2 natural peptide antibioitcs from rabbit leukocytes. Journal of Virology, 54(2), 467-472. Lehrer, R.I . , Rosenman, M. , Harwig, S .S .L. , Jackson, R., and E isenhauer, P .B . ( 1 99 1 ) Ultrasensitive assays for endogenous antimicrobial peptides. Journal of Immuno logical Methods, 1 3 7, 1 67- 1 73. Lehrer, RI ., Selsted, M.E. , Szklarek, D., and Fleischmann, J . ( 1 983) Antibacterial activity of microbicidal cationic protiens 1 and 2, natural peptide antibiotics of rabbit lung macrophages. I nfection and Immunity, 42( 1 ), 1 0- 1 4 . L i, P . , Chan, H.C. , He, B. , So, S .C. , Chung, Y .W. , Shang, Q. , Zhang, Y.D. , and Zhang, Y.L . (200 1 ) An antimicrobial peptide gene found in the male reproductive system of rats. Science, 29 1 (5509), 1 783- 1 785 . L ichtenstein, A. , Ganz, T . , Selsted, M.E., and Lehrer, R. I . ( 1 986) In-vitro tumor cell cytolysis mediated by peptide defensins of human and rabbit granulocytes. Blood, 68(6), 1 407- 1 4 1 0. L iu, L . and Ganz, T. ( 1 995) The pro region of human neutrophil defensin contains a motif that is essential for normal subcellular sorting. Blood, 85(4), 1 095- 1 1 03 . Loh, B . , Grant, C . , and Hancock, R.E .W. ( 1 984) Use of the fluorescent probe 1 -n­ phenylnaphthylamine to study the interactions of aminoglycoside antibiotics with the outer membrane of Pseudomonas aeruginosa. Antimicrobial Agents and Chemotherapy, 26(4), 546-55 1 . Lupetti, A., Dansei, R , van't Wout, J . W., van Dissel, J . , Senesi, S . , and N ibbering, P . H . (2002) Antibiotic peptides: therapeutic potential for the treatment of Candida infections. 1 86 References Expert Opinion on Investigational Drugs, 1 1 (2), 309-3 1 8 . Lynn, DJ. , H iggs, R. , Gaines, S . , Tierney, J. , lames, T. , L loyd, AT., Fares, M.AM.G., and O'Farrelly, C . (2004) Bioinformatic discovery and initial characterisaion of nine novel antimicrbial peptide genes in the chicken. Immunogentics, 56, 1 70- 1 77. Lysenko, E.S., Gould, J. , Bals, R. , Wilson, J.M., and Weiser, J.N. (2000) Bacterial phosphorylcholine decreases susceptibility to the antimicrbial peptide LL-3 71hCAP 18 expressed in the upper respiratory tract. Infection and I mmunity, 68(3), 1 664- 1 67 1 . Maddon, P.J. , L ittman, D .R. , Godfrey, M. , Maddon, D.E., Chess, L . , and Axel, R. ( 1 986) The isolation and nucleotide sequence of a cDNA encoding the T cell surface protein T4: A new member of the immunoglobulin gene family. Cell, 42, 93- 1 04 . Mahoney, M.M. , Lee, AY. , Brezinski-Caliguri, D . J . , and Huttner, K.M. ( 1 995) Molecular analysis of the sheep cathelin family reveals a novel antimicrobial peptide. FEBS Letters, 3 37, 5 1 9-522. Mak, P. , Wojcik, K. , Thogersen, LB., and Dubin, A. ( 1 996) Isolation, antimicrobial activites, and primary structures of hamster neutrophil defensins. I nfection and I mmunity, 64( 1 1 ), 4444-4449. MaIm, l . , Sorensen, 0 . , Persson, T., Frohm-Nilsson, M., Johansson, 8., Bjartell, A, Lilja, H . , Stahle-Backdahl, M. , Borregaard, N., and Egesten, A (2000) The human cationic antimicrobial protein (hCAP 18) is expressed in the epithelium of human epididymis, is present in seminal plasma at high concentrations, and is attached to !>permatozoa. Infection and Immunity, 68(7), 4297-4302. Mathews, M. , Jia, H.P. , Guthmiller, J .M., Losh, G . , Graham, S . , johnson, G.K. , Tack, 8.F. , and McCray, P.8.J. ( 1 999) Production of J3-defensin antimicrobial pep tides by oral mucosa and salivary glands. Infection and Immunity, 67(6), 2740-2745. Mayo, K.H. , Haseman, J . , Young, H.C., and Mayo, J. W. (2000) Structure-function relationships on novel peptide dodecamers with broad-spectrum bactericidal and endotoxin-neutralizing activites. B iochemical Journal, 349, 7 1 7-728. McAuliffe, 0., Ross, R P., and H ill, C. (200 1 ) Lantibiotics: structure, biosynthesis and mode of action. FEMS Microbiology Reviews, 25, 285-308. Meat and L ivestock Australia (2000) Effect of pH on lamb meat quality. Meat New Zealand (2003) Annual Report. Meisel, H. ( 1 998) Overview on milk protein-derived peptides . I nternational Dairy Journal, 8, 363-373 . Meisel, H . and FitzGerald, R.J. (2003) Biofunctional peptides from milk proteins: mineral binding and cytomodulatory effects. Current Pharmaceutical Design, 9(8), 1 -7 . Miyakawa, Y. , Ratnakar, P . , Rao, AG., Costello, M.L . , Mathieu-Costello, 0 . , Lehrer, R I., and Catanzaro, A. ( 1 996 ) In vitro activity of the antimicrobial peptides human and rabbit defensins and porcine protegrin against Mycobacterium tuberculosis. I nfection and Immunity, 64(3), 926-932 . Moore, R.A., Bates, N.C. , and Hancock, RE. ( 1 986) Interaction of polycationic antibiotics with Pseudomonas aeruginosa lipopolysaccharide and lipid A studied by using dansyl­ polymyxin. Antimicrob Agents Chemother, 29(3 ), 496-500. Morassutti, C., De Amicis, F . , Skerlavaj , 8., Zanetti, M., and Marchetti, S. (2002) Production 1 87 References of a recombinant antimicrobial peptide in transgenic plants using a modified VMA intein expression system. FEBS Lett, 5 1 9( 1 -3 ), 1 4 1 -6. Murakami, M., Ohtake, T., Dorschner, R.A, Schittek, B., Garbe, C., and Gallo, R.L. (2002) Cathelicidin anti-microbial peptide expression in sweat, an innate defense system for th skin. Journal of I nvestigative Dermatology, 1 1 9(5 ), 1 090- 1 095. Nagaoka, I . , Tsutsumi-I shii, Y., Yomogida, S. , and Yamashita, T. ( 1 997 ) Isolation of cDNA encoding guines pig neutrophil cationic antibacterial polypeptide of 1 1 kDA (CAP 1 1) and evaluation of CAP 1 1 mRNA expression during neutrophil maturation. Journal of Bio log ical Chemistry, 272(36), 22742-22750. Nakazato, M., Shiomi, K., Date, Y., Matsukura, S., Kangawa, K., M inamino, N., and Matsuo, H. ( 1 995) Isolation and sequence determination of 6- and 8-kDa precursors of human neutrophil peptides from bone marrow, plasma and peripheral blood neutrophils. Biochemical and Biophysical Communications, 2 1 1 (3) , 1 053- 1 062. Nes, I . F . and Holo, H . (2000) Class 11 antimicrobial peptides from lactic acid bacteria. Biopolymers (Peptide Science), 55 , 50-6 l . Newman, S.L. , Gootee, L. , Gabay, lE. , and Selsted, M.E. (2000) Identification of constituents of human neutrophil azurophil granules that mediate fungistasis against Histoplasma capsula/um. Infection and Immunity, 68( 1 0), 5668-72. Nibbering, P. H . , Wel l ing, M.M., Van den Broek, P.l, Van Wyngaarden, K.E., Pauwels, E.K.1. , and Calame, W. ( 1 998) Radiolabelled antimicrobial peptides for imaging of infections: A review. Nuclear Medicine Communications, 1 9( 1 2), 1 1 1 7- 1 1 2 l . Niyonsaba, F. , Someya, A, H irata, M. , Ogawa, H. , and Nagaoka, I . (200 1 ) Evaluation of the effects of peptide antibiotics human fJ-defensin- 11- 1 and LL-3 7 on histamine release and prostagluandin D2 production from mast cells. European Journal of I mmuno logy, 3 1 , 1 066- 1 075. Ockerman, H. W. and Hansen, C.L. ( 1 988) Animal Bioproduct Processing, Ell is Horwood Ltd, England. Oppenhei, J.1 . , Biragyn, A., Kwak, L. W. , and Yang, D. (2003 ) Roles of antimcirobial peptides such as defensins in innate and adaptive immunity. Annals of the Rheumat ic Diseases, 62 (Suppl 1 1 ), ii 1 7-i i2 1 . Oren, Z. , Lerman, lC. , Gudmundsson, G.H. , Agerberth, B., and Shai, Y. ( 1 999) Structure and organiszation of the human antimicrobial peptide LL-3 7 in phospholipid membranes: relevance to the molecular basis for its non-cell-selective activity. Biochemical Journal, 34 1 , 50 1 -5 1 3 . Oren, Z. and Shai, Y. ( 1 998) Mode of action of linear amphipathic a-helical antimicrobial peptides. Biopolymers, 47, 45 1 -463 . Orzechowski, A, Ostaszewski, P . , Jank, M. , and Berwid, S.1 . (2002) Bioactive substances of plant origin in food - impact on geomics. Reproduction Nutrit ion Development, 42, 46 1 - 477. Oscartz, J .C. and Pisabarro, AG. (200 1 ) Classification and mode of action of membrane­ active bacteriocins produced by gram-positive bacteria. International Microbio logy, 4, 1 3 - 1 9 . Otvos, L. (2000) Antibacterial peptides isolated from insects. Journal of Peptide Sc ience, 6, 497-5 1 1 . 1 88 References Otvos, L. (2002) The short proline-rich antibacterial peptide family. Cellular and Molecular Life Sciences, 59(7), 1 1 3 8-50. Ouellette, A.l., Darmoul, D . , Tran, D., Huttner, K.M., Yuan, 1 . , and Selsted, M.E . ( 1 999) Peptide localization and gene structure of cryptdin 4, a differently expressed mouse paneth cell a-defensin. Infection and Immunity, 67( 1 2) , 6643-665 1 . Ouellette, A.l . and Lualdi, 1 .C. ( 1 990) A novel mouse gene family coding for cationic, cysteine-rich peptides. lournal of B io logical Chemistry, 265( 1 7) , 983 1 -9837. Ouellette, A.l . , Miller, S . I . , Henschen, A .H . , and Selsted, M.E. ( 1 992) Purification and primary structure of murine crypdin-1 a Paneth cell defensin. FEBS Letters, 304(2-3), 1 46- 1 48. Park, c.B. , Kim, H.S. , and Kim, S.c. ( 1 998) Mechanism of action of the antimicrobial peptide buforin II: buforin 11 kills microorganisms by penetrating the cell membrane and inhibiting cellular functions. Biochemical and Biophysical Research Communications, 244( 1 ), 253 -7. Park, K., Oh, D., Shin, S.Y. , Hahm, K.-S. , and Kim, Y. (2002) Structural studies of porcine myeloid antibacterial peptide PMAP-23 and its analogues in DPC micelles by NMR spectroscopy. Biochemical and Biophysical Research Communications, 290( 1 ), 204-2 1 2 . Pathak, N. , Salas-Auvert, R. , Ruche, G., lanna, M. -H. , McCarthy, D . , and Harrison, RG. ( 1 995) Comparsion of the effects of hydrophobicity, amhiphilicity, and a-helicity on the activities of antimicrobial peptides. Proteins: Structure, Function, and Genet ics, 22, 1 85- 1 86. Patrzykat, A., Friedrich, C. L . , Zhang , L . , Mendoza, V., and Hancock, R. E. W. Sublethal concentraions of pleuroc idin-derived antimicrobial peptides inhibit macromolecular synthesis in Escherichia coli. 2002;46: 3 :605-6 1 4. Perez-Paya, E . , Houghten, R.A., and B londelle, S .E . ( 1 995) The role of amphipathicity on the folding, self-association and biological activity of mUltiple subunit small proteins. lournal of Biological Chemistry, 270(3), 1 048- 1 056. Perry, R.H. and Green, D. W. ( 1 997) Perry's Chemical Engineers ' Handbook - Seventh Edition. McGraw-Hill Companies Inc, Australia. Peschel, A. , lack, R. W., Otto, M. , Collins, L . V . , Staubitz, P ., Nicho loson, G., Kalbacher, H . , Nieuwenhuizen, W. F . , lung, G., Tarkowski, A. , van Kessel, K. P . M. , and van Strijp, 1 . A. G. (200 1 ) Staphylococcus aureus resistance to human defensins and evasion of neutrophil killing via the novel virulence factor MprF is based on modification of membrane lipids with L-Iysine. lournal o f Experimental Medicine, 1 93 (9), 1 067- 1 076. Peschel, A. , Otto, M., lack, R.W., Kalbacher, H . , lung, G., and Goetz, F. ( 1 999) Inactivation of the dlt operon in Staphylococcus aureus confers sensitivty to defensins, protegrins, and other antimicrobial peptides. lournal of B io logical Chemistry, 274( 1 3 ), 8405-84 1 0. Popsueva, A.E . , Z inovjeva, M.V. , Visser, 1. W.M. , Zij lmans, M.1 .M. , F ibbe, W.E . , and Belyavsky, A.V. ( 1 996) A novel murine cathelin-like protein expressed in bone marrow. FEBS Letters, 39 1 , 5-8. Powers, 1 .-P.S . and Hancock, R.E.W. (2003) The relationship between peptide structure and antibacterial activity. Peptides, 24, 1 68 1 - 1 69 1 . Prater, C . B . , Maivald, P . G. , Kjoller, K. 1 . , and Heaton. M.G. Probing nano-scale forces with the atomic force microscopy. Di Dig ital I nstruments Promotional Material, Group 1 89 References Scientific Pty Ltd. Price, 1 .F . and Schweigert, B .S . ( 1 987) The Science of Meat and Meat Products, 3rd Edition. Food and Nutrition Press, Inc . , Westport, USA. Qu, X.-D., Harwig, S .S .L . , Oren, A., Shafer, W.M., and Lehrer, R. I . ( 1 996) Susceptibility of Neisseria gonorrhoeae to protegrins. Infection and Immunity, 64(4), 1 240- 1 245. Quintavalla, S . and Vivini, L . (2002) Antimicrobialfood packaging in the meat industry. Meat Sc ience, 62, 3 73-380. Raj, P . and Edgerton, M. ( 1 995) Functional domain of poly-L-proline 11 conformation for candidacidal activity of bactenecin 5. FEBS Letters, 368, 526-530. Rautenbach, M. and Hastings, 1. W. ( 1 999) Cationic peptides with antimicrobial activity - the new generation of antibiotics? Chimica Oggi, Nov/Dec 1 999, 8 1 -89. Riley, M.A. and Wertz, 1.E. (2002) Bacteriocins: evolution, ecology, and applications. Annual reviews in Microbiology, 56, 1 1 7-37 . Robinson, W.E. , McDougaU, B . , Tran, D. , and Selsted, M.E . ( 1 998) Anti-HIV-I activity of indolicidin, an antimicrobial peptide from neutrophils. Journal of Leukocyte Biology, 63 , 94- 1 00. Romeo, D., Skerlavaj, B. , Bolognesis, M., and Gennaro, R. ( 1 988) Structure and bactericidal activity of an antibiotic dodecapeptide purified from bovine neutrophils. Journal of Biological Chemistry, 263(20), 9573-9575. Sadler, K . , Eom, K.D., Yang, 1 .-L. , Dimitrova, Y , and Tarn, 1 .P . (2002) Translocating proline-rich peptides from the antimicrobial peptide bactenecin 7. Biochemistry, 4 1 , 1 4 1 50- 1 4 1 57. Saiman, L . , Tabibi, S. , Starner, T.D. , San Gabriel, P. , Winokur, P .L. , Jia, H . P. , Mccray, P.B. , and Tack, B .F . (200 1 ) Cathelicidin peptides inhibit multiply antibiotic-resistant pathogens from patients with cystic fibrosis. Antimicrobial Agents and Chemotherapy, 45( 1 0), 2838- 2844. Savoini, A., Marzari, R. , Dolzani, L., Serrano, D., Graziosi, G., Gennaro, R., and Romeo, D . ( 1 984) Wide-spectrum antibiotic activity of bovine granulocyte polypeptides. Antimicrobial Agents and Chemotherapy, 26(3 ), 405-407. Sawai, M. V . , Waring, A. 1., Kearney, W. R. , McCray, P. B. Jr, Forsyth, W. R., Lehrer, R. 1 . , and Tack, B . F . (2002) Impact of single-residue mutations on the structure and funtion of ovispirinlnovispirin antimicrobial peptides. Protein Engineering, 1 5( 3 ), 225-232. Sawicki, W. and Mystkowska, E .T. ( 1 999) Contraceptive potential of peptide antibiotics. The Lancet, 353 , 464-465. Sawyer, J .G . , Martin, N.L. , and Hancock, R.E . W. ( 1 988) Interaction of macrophage cationic proteins with the outer membrane of Pseudomonas aeruginosa. I nfection and I nununity, 56(3), 693-698. Scannell, A.G.M. , Ross, R.P., Hill, C., and Arendt, E.K. (2000) An ef ective lacticin biopresevative in fresh pork sausage. Journal of Food Protection, 63(3), 3 70-375. Schnapp, D. , Reid, c . 1 . , and Harris, A. ( 1 998) Localisation of expression of human beta defensins-l in the pancreas and kidney. Journal of Pathology, 1 86, 99- 1 03 . Schuster, F.L. and Jacob, L .S . ( 1 992) Ef ects of magainins on ambea and cyst stages of acanthamoeba-polyphaga. Antimicrobial Agents and Chemotherapy, 3 6(6), 1 263- 1 27 1 . 1 90 References Scocchi, M . , Bontempo, D. , Boscolo , S . , Tomasinsig, L . , Giulotto, E . , and Zanetti, M. ( 1 999) Novel cathelicidins in horse leukocytes. FEBS Letters, 457, 459-464. Scocchi, M . , Skerlavaj , B. , Romeo, D. , and Gennaro, R. ( 1 992) Proteolytic cleavage by neutrophil elastase converts inactive storage proforms to antibacterial bactenecins. European Journal of B iochemistry, 209(2), 589-95. Scocchi, M . , Wang, S . , Gennaro, R., and Zanetti, M. ( 1 998) Cloning and analysis of a transcript derived from two contiguous genes of the cathelicidin family. Biochimica et B iophysica Acta, 1 398, 393-396. Scocchi, M., Wang, S . , and Zanetti, M. ( 1 997) Structural organisation of the bovine cathelicidin gene family and identification of a novel member . FEBS Letters, 4 1 7, 3 1 1 - 3 1 5 . Scott, M.G. , Yan, H . , and Hancock, R.E.W. ( 1 999) Biological propeties of structurally related alpha-helical cationic antimicrobial peptides. Infection and I mmunity, 67(4), 2005-2009. Selsted, M .E . ( 1 997) HPLC methods for purification of antimicrobial peptides In: Shafer, W.M. Humana Press Inc, Totowa, N1. Selsted, M.E. , Brown, D.M., DeLange, R.J., Harwig, S .S .L. , and Lehrer, R. I . ( 1 985) Primary structures of six antimicrobial peptides of rabbit peritoneal neutrophils. Journal of B io logical Chemistry, 260(8), 4579-4584. Selsted, M .E . and Harwig, S .S.L . ( 1 987) Purification, primary structure and antimicrobial activities of a guinea pig neutrophil defensin. Infection and I mmunity, 55(9), 228 1 -2286. Selsted, M.E . and Harwig, S .S .L . ( 1 989) Determination of the desuljide array in the human defensin HNP-2 . Journal of B io logical Chemistry, 264(7), 4003-4007. Selsted, M.E. , Novotny, M.1. , Morris, W.L. , Tang, Y.-Q. , Smith, W., and Cullor, 1 .S . ( 1 992) Indolicidin, a novel bactericidal tridecapeptide amide from neutrophils. Journal of B io logical Chemistry, 267(7), 4292-4295. Selsted, M .E ., Szklarek, D., Ganz, T., and Lehrer, R. I . ( 1 985) Activity of rabbit leukocyte peptides against Candida albicans. I nfection and Immunity, 49( 1 ), 202-206. Selsted, M.E . , Tang, Y.-Q., Morris, W.L., McGuire, P.A. , Novotny, M.1 . , Smith, W. , Henschen, A.H . , and Cullor, 1 .S. ( 1 993) Purification, primary structures, and antibacterial activities of fJ-defensins, a new family of antimicrobial peptides from bovine neutrophilis. Journal of B iological Chemistry, 268(9), 664 1 -6648. Shamova, O.V., Brogden, K.A. , Zhao, c., Nguyen, T., Kokryakov, V.N. , and Lehrer, R. I . ( 1 999) Purification and properties of proline-rich antimicrobial peptides from sheep and goat leukocytes. I nfection and I mmunity, 67(8), 4 1 06-4 1 1 1 . Sheu, M .J.T. , Baldwin, W.W., and Brunson, K.W. ( 1 985) Cytotoxicity of rabbit macrophage pep tides MCP-I and MCP-2 for mouse tumor cells. Antimicrobial Agents and Chemotherapy, 28(5), 626-629. Shi, 1 . , Ross, C .R. , Chengappa, M.M. , and B lecha, F. ( 1 994) Identification of a proline­ arginine-rich antibacterial peptide from neutrophils that is analoguos to PR-39, an antibacterial peptide from the small intestine. Journal of Leukocyte B io logy, 56, 807-8 1 1 . Shi, 1 . , Ross, C.R., Chengappa, M.M. , Syite, M.J . , McVey, D .S . , and B lecha, F . ( 1 996a) Antibacterial activity of synthetic peptide (PR-26) derived from PR-39, a proline­ arginine-rich neutrophil antimicrobial peptide. Antimicrobial Agents and Chemotherapy, 1 9 1 References 40( 1 ), 1 1 5 - 1 2 1 . Shi, 1 . , Ross, C.R., Leto, T., and Blecha, F. ( 1 996b) PR-39, a proline-rich antibacterial peptide that inhibits phagocyte NADPH oxidase activity by binding to Src homology 3 domains of p4 7phox. Proceedings of the National Academy of Sc ience USA, 93, 60 1 4- 60 1 8 . Shimoda, M. , Ohki, K., Shimamoto, Y. , and Kohashi, O. ( 1 995) Morphology of defensin­ treated Staphylococcus aureus. Infection and Immunity, 63(8), 2886-289 1 . Shin, S. Y. , Kang, 1 .H . , Jang, S . Y., Kim, K.L., and Hahm, K.S. (2000a) Structure and antibiotic activity of a porcine myeloid antibacterial peptide, PMAP-23 and its analogues. Journal of Biochemistry and Molecular Biology, 3 3( 1 ), 49-53 . Shin, S .Y. , Kang, S . -W., Lee, D.G., Eom, S .H . , Song, W.K., and Kim, 1 . 1 . (2000b) CRAMP analogues having potent antibiotic activity against bacterial, fungal, and tumor cells without hemolytic activity. Biochemical and Biophysical Communications, 275, 904-909. Shin, S .Y. , Park, E.1 . , Yang, S.T., lung, H .1 . , Eom, S .H . , Song, W.K., Kim, Y. , Hahm, K.S . , and Kim, 1 . 1 . (200 I a ) Structure-activity analysis of SMAP-29, a sheep leukocytes-derived antimicrobial peptide . Biochemical and Biophysical Research Communications, 285(4), 1 046-5 1 . Shin, S .Y. , Park, E.1 . , Yang, S.T., lung, H .1 . , Eom, S .H . , Song, W.K., Kim, Y . , Hahm, K.S. , and Kim, 1 . 1 . (200 1 b) Structure-activity analysis of SMAP-29, a sheep leukocytes-derived antimicrobial peptide . Biochemical and Biophysical Research Communications, 285(4), 1 046- 1 05 1 . Shin, S .Y. , Yang, S.T. , Eom, S .H. , Song, W.K., Kim, Y. , Hahm, K.S . , and I l Kim, l . (200 1 c) Design of dermaseptin S3 analogues having bacterial cell selectivity without mammalian cell toxicity. Protein and Peptide Letters, 8(4), 28 1 -288 . Sieprawska-Lupa, M., Mydel, P . , Krawczyk, K . , Wojcik, K . , Puklo, M. , Lupa, B. , Silberring, 1 . , Reed, M., Pohl, 1 . , Shafer, W., McAleese, F., Foster, T., Travis, 1 . , and Potempa, 1 . (2004) Degradation of human antimicrobial peptide LL-3 7 by Stapylococcus aureus­ derived proteinases. Ant in1icrobial Agents and Chemotherapy, 48( 1 2), 4673-4679. Simmaco, M., Mignogna, G . , and Barra, D. ( 1 998) Antimicrobial peptides from amphibian shin: what do they tell us? Biopolymers, 47, 435-450. Siragusa, G .R. , Cutter, C.N., and Willett, 1 .L. ( 1 999) Incorporation of bacteriocin in plastic retains activity and inhibits surface growth of bacteria on meat. Food Microbio logy, 1 6, 229-235 . SkerJavaj, B . , Benincasa, M. , Risso, A. , Zanetti, M . , and Gennaro, R. ( 1 999) SMAP-29: a potent antibacterial and antifungal peptide from sheep leukocytes. FEBS Letters, 463, 58- 62. Skerlavaj, B., Gennaro, R. , Bagella, L., Merluzz, L., Risso, A., and Zanetti, M. ( 1 996) Biological characterisation of two novel cathelicidin-derived peptides and identification of structural requirements for their antimicrobial and cell lytic activities. Journal of Biological Chemistry, 27 1 (45), 28375-2838 1 . Skerlavaj, B . , Romeo, D. , and Gennaro, R. ( 1 990) Rapid membrane permeabilization and inhibition of vital functions of gram-negative bacteria by bactenecins. Infection and Immunity, 58( 1 1 ), 3724-3730. Sorensen, O.E. , Follin, P. , lohnsen, A.H. , Calafat, 1 . , Tjabringa, G .S . , Hiemstra, P . S., and 1 92 References Borregaard, N. (200 1 ) Human cathelicidin, hCAP-I8, is processed to the antimicrobial peptide LL-3 7 by extracellular cleavage with proteinase 3 . Blood, 97( 1 2), 395 1 -3959. Statistics New Zealand (200 1 ) Exports of major commodities 1 999-2000, http://www. stats.govt.nzlquick-facts/economy/exports-and-imports.htm. Accessed 2005. Steinberg, D.A, Hurst, M.A, Fuj ii, C .A, Kung, AH.C. , Ho, J .F. , Cheng, F.-C. , and Fiddes, J.c. ( 1 997) Protegrins-I : a broad-spectrum, rapidly microbicidal peptide with in vitro activity. Antimicrobial Agents and Chemotherapy, 4 1 (8), 1 73 8- 1 742. Steinberg, D.A and Lehrer, R.I . ( 1 997) Designer assays for antimicrobial peptides In: Shafer, W.M. Humana Press Inc , Totowa, Nl Steinstraesser, L . , Tack, B .F. , Waring, Al, Hong, T. , Boo, L.M. , Fan, M.H., Remick, D J . , Su, G.L. , Lehrer, R.I . , and Wang, S .C. (2002) Activity of novi5pirin G10 against Pseudomonas aeruginosa in vitro and in infected burns. Antimicrobial Agents and Chemotherapy, 46(6), 1 837-44. Stiles, M.E. ( 1 994) Potential for biological control of agents of foodborne disease. Food Research International, 27, 245-250. Storici, P . , Del Sal, G. , Schneider, C. , and Zanetti, M. ( 1 992) cDNA sequence analysis of an antibiotic dodecapeptide from neutrophils. FEBS Letters, 3 1 4, 1 87- 1 90. Storici, P . , Scocchi, M. , Tossi, A, Gennaro, R. , and Zanetti, M. ( 1 994) Chemical synthesis and biological activity of a novel antibacterial peptide deduced from a pig mysloid cDNA . FEBS Letters, 337, 303-307. Storici, P . and Zanetti, M. ( 1 993) A novel cDNA sequence encoding a pig leukocyte antimicrobail peptide with a cathelin-like pro-sequence. Biochemical and B iophysical Communications, 1 96(3), 1 363- 1 368. Stryer, L. ( 1 995) Biochemistry, Fourth Edition. W.H. Freeman and Company, New York. Subbalakshmi, C. and S itaram, N. ( 1 998) Mechanism of antimicrobial action of indolicidin. FEMS M icrobiology Letters, 1 60( 1 ), 9 1 -96. Tack, B.F . , Sawai, M.V. , Kearney, W.R. , robertson, AD., Sherman, M.A, Wang, W. , Hong, T., Boo, L.M. , Huiyuan, W., waring, Al, and Lehrer, R.1. (200 1 ) SMAP-29 has two LPS­ binding sited and a central hinge. European Journal of B iochemistry, 269, 1 1 8 1 - 1 1 89. Takano, T. ( 1 998) Milk derived peptides and hypertension reduction. International Dairy Journal, 8, 375-3 8 1 . Takemura, H. , Kaku, M. , Kohno, S . , Hirakata, Y. , Tanaka, H. , Yoshida, R., Tomono, K. , Koga, H . , Wada, A, Hirayama, T. , and Kamihira, S. ( 1 996) Evaluation of susceptibility of gram-positive and -negative bacteria to human defensins by using radial diffusion assay. Ant imicrobial Agents and Chemotherapy, 40( 1 0) , 2282-2284. Tang, Y.-Q., Yuan, l, Mil ler, S J . , and Selsted, M.E . ( 1 999a) Isolation, characterization, cDNA cloning, and anitmicrobial properties of two distinct subfamilies of alpha-defensins from rhesus macaque leukocytes. I nfection and Immunity, 67( 1 1 ) , 6 1 39-6 1 44. Tang, Y.-Q., Yuan, l , Osapay, G., Osapay, K. , Tran, D., Miller, C.J., Ouellette, Al, and Selsted, M.E. ( 1 999b) A cyclic antimicrobial peptide produced in primate leukocytes by the ligation of two truncated a-defensins. Science, 286, 498-502. Teschemacher, H., Koch, G., and Brantl, V . ( 1 997) Milk protein-derived opioid receptor ligands. Biopolymers, 43, 99- 1 1 7 . 1 93 References Tiozzo, E. , Rocco, G., Tossi, A, and Romeo, D. ( 1 998) Wide-spectrum antibiotic activity of synthetic amphipathic peptides. Biochemical and B iophysical Research Communications, 249, 202-206. Tossi, A., Scocchi, M. , Skerlavaj, B., and Gennaro, R. ( 1 994) Identification and characterisation of primary antibacterial domain in CAP 18, a lipopolysaccharide binding proteinfrom rabbit leukocytes. FEBS Letters, 339, 1 08- 1 1 2 . Tossi, A, Scocchi, M. , Zanetti, M., Storici , P., and Gennaro, R. ( 1 995) PMAP-37, a novel antibacterial peptide from pig myeloid cells - DNA cloning, chemical .synthesis and activity. European Journal of B iochemistry, 228, 94 1 -946. Trabi, M., Schirra, H.J . , and Craik, 0.1. (200 1 ) Three-dimensional structure of R TD- J cyclic antimicrobial defensin from rhesus macaque leukocytes. B iochemistry, 40( 1 4), 42 1 1 - 422 l . Tran, D., Tran, P.E., Tang. Y.-Q., Yuan, 1 . , Cole, T. , and Selested, M.E . (2002) Homodimeric rfrdefensins from Rhesus macaque leukocytes. The Journal of Bio logical Chemistry, 277(5) , 3079-3084. Travis, S. M., Anderson, N.N., Forsyth, W. R., Espiritu, e., Conway, B .O . , Greenberg, E.P. , McCray Jr, P .B. , Lehrer, R. I . , Welsh, M.l, and Tack, B .F . (2000) Bactericidal activity of mammalian cathelicidin-derived peptides. Infect ion and I mmunity, 68(5), 2748-2755 . Turner, J . , Cho, Y . , Dini, N. -N ., Waring, Al, and Lehrer, R. I . ( 1 998) Activities of LL-37, a cathelin-associated antimicrobial peptide of human ne utrophils. Antimicrobial Agents and Chemotherapy, 42(9), 2206-22 1 4. Uzzell , T. , Sto lzenberg, E .D. , Shinnar, AE. , and Zasloff, M. (2003) Hagfish intestinal antimicrobial peptides are anicent cathelicidins. Peptides, 24( 1 1 ), 1 655- 1 667. Valore, E .V., Martin, E., Harwig, S .S .L . , and Ganz, T. ( 1 996) Intramolecular inhibition of human defensin HNP- I by its propoece. Journal of Cl inical I nvestigation, 97(7 ) , 1 624- 1 629. van Asch, K. (200 1 ) Current uses of animal blood in the New Zealand meat industry. Meat Technology Report, I nstitute of Technology and Engineering, Massey University. Verbanac, D., Zanetti, M., and Romeo, D. ( 1 993) Chemotatic and protease-inhibiting activities of anitbiotic peptide precursors. FEBS Letters, 3 1 7(3), 255-258. Weipresht, T., Dathe, M. , Beyermann, M., Krause, E . , Maloy, W.L. , MacDonald, D.L. , and Bienert, M. ( 1 997) Peptide hydrophobicity controls the activity and selectivity of magainin 2 amide in interactions with membranes. Biochemistry, 63(20), 6 1 24-6 1 32 . Wi lde, e.G. , Griffith, l E ., Marra, M.N. , Snable, lL. , and Scott, R .W. ( 1 989) Purification and characterisation of human neutrophil peptide 4, a novel member of the defensin family. Journal of Bio logical Chemistry, 264( 1 9), 1 1 200- 1 1 203 . Wilson, C .L. and Ouellette, A.J. ( 1 999) Regulation of intestinal alpha-defensin activation by the metalloproteinase matrilysin i innate host defense. Sc ience, 2 86(5437), 1 1 3 - 1 1 7. Wimley, W.e., Selsted, M.E. , and White, S .H. ( 1 994) Interactions between human defensins and lipid bilayers: Evidence for formation of multimeric pores. Protein Sc ience , 3(9), 1 362- 1 3 73 . Winder, D. , Gunzburg, W.H . , Erfle, V . , and Salmons, B. ( 1 998) Expression of antimicrobial peptides has an antitumour effect in human cells. Biochemical and B iophysical Communications, 242, 608-6 1 2 . 1 94 References Wu, E .R., Daniel, R., and Bateman, A ( 1 998) RK-2: A novel rabbit kidney defensin and its implications for renal host defense. Peptides, 1 9(5) , 793-799. Wu, M. and Hancock, R.E.W. ( 1 999a) Improved derivatives of bactenecin, a cyclic dodecameric antimicrobial cationic peptide. Antimicrobial Agents and Chemotherapy, 43(5), 1 274- 1 276. Wu, M. and Hancock, R.E.W. ( 1 999b) Interaction of the cyclic antimicrobial cationic peptide bactenecin with the outer and cytoplasmic membrane. Journal of B io logical Chemistry, 274( 1 ), 29-35 . Yamamoto, N . , Ejiri, M. , and Mizuno, S. ( 2003) Biogenic peptides and their potential use. Current Pharmaceutical Design, 9(8), 1 - 1 1 . Yamashita, T. and Saito, K. ( 1 989) Purification, primary structure, and biological activity of guinea pig neutrophil cationic peptides. Infection and Immunity, 57(8) , 2405-2409. Yan, H. and Hancock, R.E . W. (200 1 ) Synergistic interactions between mammalian antimicrobial defense peptides. Antimicrobial Agents and Chemotherapy, 45(5), 1 558- 1 560. Yang, B .O . , Chen, Q., Schmidt, AP. , Anderson, G.M. , Wang, 1 .M. , Wooters, 1 . , Oppenheim, 1 . 1 . , and Chertov, O. (2000) LL-3 7, the neutrophil granule and epithelial cell derived cathelicidin, utilises formyl peptide receptor-like ] (FP RL]) as a receptor to chemoattract human peripheral blood neutrophils, monocytes, and T cells. Journal of Experimental Medicine, 1 92(7), 1 069- 1 074. Yarus, S . , Rosen, J .M. , Cole, AM., and Diamond, G. ( 1 996) Production of active bovine tracheal antimicrobial peptide in milk of transgenic mice. Proceedings of the National Academy of Science USA, 93, 1 4 1 1 8- 1 4 1 2 1 . Yasin, B . , Harwig, S .S .L. , Lehrer, R.I . , and Wagar, E.A ( 1 996) Susceptibility of Chamydis trachomatis to protegrins and defensins. Infection and Immunity, 64(3) , 709-7 1 3 . Yasin, B . , Pang, M. , Turner, 1 .S. , Cho, Y . , Dinh, N.-N., Waring, AJ. , Lehrer, R.I . , and Wagar, E .A. (2000) Evaluation of the inactivation of infectious herpes simplex virus by host-defense peptides. European Journal of Clinical Microbiology and Infectious Diseases, 1 9, 1 57- 1 94. Yourassowsky, E . , van der L inden, M.P . , Lismont, M.J., Crokaert, F., and Glupczynski, Y. ( 1 985) Correlation between growth curve and killing curve of Escherichia coli after a brief exposure to superinhibitory concentrations of ampicillin and piperacillin. Antimicrobial Agents and Chemotherapy, 28(6), 756-760. Yu, P .-L., Deb Choudhury, S., and Ahrens, K. (200 1 ) Purification and characterization of the antimicrobial peptide, ostricacin. Biotechnology Letters, 23 , 207-2 1 0. Zabucchi, G. , Rottini, G.D., Soranzo, M.R., Tedesco, F . , and Patriarca, R. ( 1 983) A new method for assessment of serum-induced damage to E. coli. Journal of Immunological Methods, 5 7, 253-264 . Zanetti, M. , Del Sal, G. , Storici, P . , Schneider, C . , and Romeo, D. ( 1 993) The cDNA of the neutrophil antibiotic Bac5 predicts a pro-sequence homologous to a cysteine proteinase inhibitor that is common to other neutrophil antibiotics. The Journal of B io logical Chemistry, 268( 1 ), 522-526. Zanetti, M., Gennaro, R. , and Romeo, D . ( 1 995) Cathelicidins: a novel protein family with a common pro region and a variable C-terminal antimicrobial domain. FEBS Letters, 1 95 References 374( 1 ) , 1 -5 . Zanetti, M. , L itteri, L . , Ge nnaro , R. , Horstmann, H. , and Romeo, D . ( 1 990) Bactenecins, defense polypeptides of bovine ne utrophils, are generated from precursor molecules stored in the large granules. Journal of Cel l Bio logy, I l l , 1 363- 1 3 7 1 . Zanetti, M. , L itteri, L . , Griffiths, G., Gennaro, R., and Romeo, D . ( 1 99 1 ) Stimulus-induced maturation of probactenecins, precursors of neutrophil antimicrobial polypeptides. Journal of Immunity, 1 46( 1 2) , 4295-4300. Zanetti, M., Storic i, P., Tossi, A. , Scocchi, M., and Gennaro, R. ( 1 994) Molecular cloning and chemical synthesis of a novel antibacterial peptide derivedfrom pig myeloid cells. Journal of Bio logical Chemistry, 269( 1 1 ) , 7855-7858 . Zasloff, M. (2002) Antimicrobial pep tides of multicellular organisms. Nature, 4 1 5(6870), 389-95 . Zhang, L. , Scott, M. G . , Van, H . , Mayer, L . D. , and Hancock, R . E. W. (2000) Interaction of polyphemusin I and structural analogs with bacterial membranes, lipopolysaccharide, and lipid monolayers. Biochemistry; 39, 1 4504- 1 45 1 4. Zhao, c . , Ganz, T. , and Lehrer, R. I . ( 1 995a) The structure of porcine protegrin genes. FEBS Letters, 368, 1 97-202. Zhao, C., Ganz, T. , and Lehrer, R. I . ( 1 995b) Structures of genes for two cathelin-associated antimicrobial peptides: prophenin-2 and PR-39. FEBS Letters, 376, 1 30- 1 34. Zhao, C., Liu, L . , and Lehrer, R . I . ( 1 994) Identification of a new member of the protegrin family by cDNA cloning. FEBS Letters, 346, 285-288 . Zhao, C., Nguyen, T., Liu, L . , Shamova, 0., Brogden, K., and Lehrer, R. I . ( 1 999) Differential expression of caprine j3-defensins in digestive and respiratory tissues. I nfection and Immunity; 67( 1 1 ), 622 1 -6224. Zhao, C.Q., Nguyen, T., Boo, L .M. , Hong, T., Espiritu, C., Orlov, D., Wang, W., Waring, A., and Lehrer, R.I . (200 1 ) RL-3 7, an alpha-helical antimicrobial peptide of the rhesus monkey. Ant imicrobial Agents and Chemotherapy, 45( 1 0), 2695-2702. Zhong, L., Putnam, RJ., Johnson, W.c.J. , and Rao, A.G. ( 1 995) Design and synthesis of amphipathic antimicrobial peptides. I nternational Journal of Peptide Science, 45, 3 3 7- 347. Zimmermann, G.R. , Legault, P . , Selsted, M.E ., and Pardi, A. ( 1 995) Solution structure of bovine neutrophil j3-defensin-12: the peptide fold of the j3-defensins in identical to that of the classical defensins. Biochemistry, 34, 1 3663- 1 367 1 . 1 96 Appendix A 1 - Characterisation Data APPENDIX A1 RAW DATA AND CALCULATIONS FROM CHARACTERISATION STUDIES A1 . 1 MASS SPECTRA OF THE PURIFIED HPLC PEAKS HPLC Pa from gel filtration F2 (OaBacS) I .. Cpa" I O {0 5Q8) \1 1 [b . IO�44J.hJ I ) IC".0 7�O.'4.\ '4\.1 OO,U3.1U3), Cm (I I)) I 51697S .0.01 " 5 1 00 5 1 50 l2J1 80 %005 HPLC Ph from gel filtration F2 (OaBac7.S(32-60» lo,Cpb6c 16 ( 1 010) \11 [E,.8-l559, I C7) (G"O 750,489 n8.0 SO,U3,1U3) Cm ( I 19) 3600 9 .0 00 Tm \\. l � l \5e5 Tor \IS E J :!'>e5 1 97 Appendix A1 - Characterisation Data HPLC Pc from gel filtration F2 (OaBac5y) la,f:'pcbb 18 (I ll()q) \ 1 1 ([,- 108641 11) I He;,. 0 750,;03 CJ9SC,( 51 17 1. 138., '02� O.7(8P · 5117.6, 2819) "',.l '"'' "" .. I " :�'" "j, ll1 W2 t� " � � \JS2 7700.57 .L _'_ __ __ \.::-::;:-__ A.... ___ � ___ .------I 4n.0 33".4 am.1 11200.2 Mau (m/z) Acquired 18.38:00, July 10.2003 O'\SERVlCEHOO703\massyp05_0003 dat ,'1191 36 " "' 08 12100.6 1 2819o( -. . 15001.0 TIll' \IS F.S I 38e< Appendix A 1 - Characterisation Data HPLC PF from gel fi ltration F2 (OaBac l l truncates) 60 " Voyager Spec #1=>BC� ';O.7[BP ::: 3655.7, 292) �'8 3C ,0529 702-- �I' 77 t� 1-\ r ' " \'!::�1 " &8 I 1 ' . 1 � 1;)62 94 '916 i:I 132lj�, 1 1 1 9 1 6' ,13:�'/: 3�\53 32 315 55 4r'" _? 55n c" b58 ... 54 1 ,1\)4 6� l,'ll� 2 25 I iI .4Ili,clf'il <; �-' 10 , 5HJ 85 I @l(.:�l jQ"i� �76!l 4�t I%'S s S56JIi� --36 31 1 1 1 9 1 3 11: �4\ j,l '+.,r' :\Jj�, .. :) \ 6}� "l \.t.. �J4J 59 9366 P egg28J ',1322 , 22 14'\ .�""..,f .... �...... � "' ...... .... ,.... . �'''''''� ....... __ ,.t' " ..... �<.,.. .... . /" ....... ,�, .... "" ."" .... _. o 4990 3399 4 6299.8 9200.2 HPLC P14 from cationic fraction (unidentified) VOyager Spec j/1=>BC"'>N� .ap .. 1024.6, 6838) 1'.4'0 ""'< / .. ' ..... • ..... j,M, . ... I . .. _,.�., ',,", ",,,,�.�:,"' >�' ",'"", ... �: .. " '�'..h.I." '\cqulred 1 2 30GO September 09 2003 ) \SERVICE1090503IMasseyq7 _14_0001 oat o 15001 0 . tllJl 1 99 Appendix A 1 - Characterisation Data HPLC PIS from cationic fraction (unidentified) . -.• \ Acqwed 12 32 00, September 09, 2003 o \sERVICE'D9080J'lMassew7 15 0003 dat HPLC P I8 from cationic fraction (Fragment of Cathelin domain) 200 I f � . AcQuit'ed 12_3300, September 09, 2003 OISERVlCB090803\Masseyq7_'8_0001 dlt VOYllger Spec; 1f1-"BCooNFO,', ... I" . lon.1, 19S01) HPLC P19 from cationic fraction (unidentified) . �" AcqUired 1 2 34 00 September 09 2003 o ',SERV1CE\090e03IMasseyq7 _19_0001 da! Voyager Spec #1=>SC">NFO '. " 1651 4, 10180] Appendix A1 - Characterisation Data HPLC P24 from cationic fraction (fragment of T -cell surface protein CD4 precursor) Voyager Spec #l"'>BC=>NFO . = 11260, 49(64) Acaulred ' 2 3 5 00 Septel'1ber09 20Cl3 201 Appendix A1 - Characterisation Data HPLC P32 from cationic fraction (unidentified) f J . Acquwed 14:2300, September 09 2003 202 Appendix A1 - Characterisation Data A1 .2 EXAMPLE CALCULATION OF CONFIDENCE INTERVALS FROM PLATE ASSAY RAW DATA The calculation of the MIC of Pa (OaBac5) from the raw plate assay data was carried out as described below. The raw data from the plate assay of Pa against E. coli 0 1 1 1 are given below. Di lution 1 1 /2 1 /4 1 /8 Cone (J.lg/mL) 25 .00 1 2 .50 6 .25 3 . 1 3 Clearing (mm) 2 .00 1 .00 0 . 50 0 .00 Ln cone 3.22 2 .53 1 .83 1 . 1 4 The log concentrations and the c learing sizes were entered into the statistics package GenStat. L inear regression analysis was done by entering the c learing size as the response variate and the Ln conc as the constant. The GenStat output is shown below. * * * * * Regression Analysi s * * * * * Response variate : clearing Fitted terms : Constant , In conc * * * Summary of analysis * * * d . f . s . s . Regression 1 2 . 1 1 2 5 0 Res idual 2 0 . 0 7 5 0 0 Total 3 2 . 1 8 7 5 0 m . s . 2 . 1 1 2 5 0 0 . 0 3 7 5 0 0 . 7 2 9 1 7 Percentage variance accounted for 94 . 9 v . r . F pr . 5 6 . 3 3 0 . 0 1 7 Standard error o f observations i s est imated t o be 0 . 1 94 * * * Est imates of parameters * * * Constant In conc estimate - 1 . 1 6 9 0 . 93 8 s . e . 0 . 2 8 9 0 . 12 5 t ( 2 ) t pr . - 4 . 04 0 . 0 56 7 . 51 0 . 0 17 From the GenStat output the y-intercept (Bo), the gradient (B ( ) and the residual ms (S2) were recorded. In this case Bo=- 1 . 1 69 , B (=0.938 and S2=0.03750. The MIC was the point where the c learing size is zero so it is equal to the x-intercept. The x­ intercept was calculated for the equation of the line as follows: y = 0.9838x + - 1 . 1 69 203 Appendix A1 - Characterisation Data when y=O x = 1 . 1 69/0.9838 = 1 .246 (this is the Ln MI C) MIC = exp( 1 .246) = 3 .476 flg/mL The 95% confidence intervals for the x-intercept were calculated using the formula below. . - t " J (X - X) ( 1 - g) Limits of confidence intervals = g(Xo -X) ± - -V S2 0 +--bl Sxx n where: A Xo = the estimated x-intercept X = the mean of the x values t = the t-value b l = gradient of the line-of-best-fit S2 = estimate of the pooled variance Sxx = total corrected sum of the squares for x = (XI - XY + (X2 -XY + (X3 - X)2 + (X4 -X)2 n = the number of x values 204 Appendix A1 - Characterisation Data A1 .3 RAW DATA, CALCULATED MICS AND 95% CONFIDENCE INTERVALS FOR THE MICS OF THE PURIFIED PEPTIDES MIC of Pa E.coli 0 157:"7 Dilution Cone �u�mq Clearing �mm} Ln cone 25 .00 2 .00 3 .22 xmean 2 . 1 8 1 12 1 2 .50 1 .00 2 .53 ymean 0.88 1 /4 6.25 0.50 1 . 83 Sxx 2.40 1 /8 3 . 1 3 0.00 1 . 1 4 ho - 1 . 1 7 Ln mean 1 .25 mean 3 .48 hi 0.94 Ln upper l imit -0. 1 7 upper limit 0.84 s2 0.04 Ln lower l imit 1 .75 lower l imit 5.77 S. aureus 0156 MRSA Dilution Cone �u�mq Clearing �mm} Ln cone 30.00 0.75 3 .40 3 .05 30.00 0.25 3 .40 0.3 1 1 /2 1 5 .00 0.25 2 . 7 1 0.48 1 12 1 5 .00 0.00 2 .7 1 ho - 1 .35 Ln mean 2.49 mean 1 2.02 hi 0.54 Ln upper l imit n/a upper l imit nla! s2 0.08 Ln lower l imit nla lower limit nla C. albicans 3153A Dilution Cone �u�mq Clearing �mm} Ln cone 30.00 0.00 3.40 1 /2 1 5 .00 0.00 2. 7 1 1 /4 7.50 0.00 2 .0 1 1 /8 3 .75 0.00 1 .3 2 ho n/a Ln mean nla mean >30 hi nla Ln upper l imit nla upper limit n/a s2 n/a Ln lower l imit nla lower l imit n/a E.coli 0157:"7 with 100mM NaCI Dilution Cone �u�mq Clearing �mm} Ln cone 25.00 0.00 3 . 22 1 12 1 2 . 50 0.00 2 .53 1 14 6.25 0.00 1 .83 1 /8 3 . 1 3 0.00 1 . 1 4 ho nla Ln mean nla mean >25 hi n/a Ln upper l imit n/a upper limit nla s2 nla Ln lower l imit nla lower l imit nla S. aureus 1056 MRSAwith IOOmM NaCI Dilution Cone �u�mq Clearing �mm} Ln cone 30.00 0.00 3.40 1 /2 1 5 .00 0.00 2.7 1 1 /4 7.50 0.00 2.0 1 1 /8 3 .75 0.00 1 .3 2 ho nla Ln mean n/a mean >30 hi nla Ln upper l imit nla upper limit n/a s2 n/a Ln lower l imit nla lower limit n/a 205 Appendix A 1 - Characterisation Data C. albicans 3 153A with 100mM NaCI Dilution Cone {u�mL} Clearing {mm} Ln cone 30.00 0.00 3 .40 1 /2 1 5 .00 0.00 2 . 7 1 1 /4 7.50 0.00 2.0 1 1 /8 3.75 0.00 1 .32 bo n/a Ln mean n/a mean >30 bi n/a Ln upper l imit n/a upper l imit n/a 52 n/a Ln lower l imit n/a lower limit n/a MIC of Pb E.coIi 0 157:H7 Dilution Cone (u�mL) Clearing (mm) Log cone 30.00 2.25 3 .40 3 .05 xmean I 30.00 I . 75 3 .40 1 .25 ymean 1 /2 1 5 .00 0.75 2 . 7 1 0.48 Sxx 1 /2 1 5 .00 0.25 2 . 7 1 bo -5.39 Ln mean 2.48 mean 1 1 .93 bi 2. 1 7 Ln upper l imit 62.96 upper l imit n/a 52 0. 1 3 Ln lower l imit 2 .87 lower l imit 1 7. 7 1 S. aureus 0 156 MRSA Dilution Cone {u�mL} Clearing {mm} Log cone 30.00 1 .75 3 .40 3 .05 xmean I 30.00 1 .25 3 .40 1 .00 ymean 1 /2 1 5 .00 0.75 2 . 7 1 0.48 Sxx 1 12 1 5 .00 0.25 2 . 7 1 bo -3.43 Ln mean 2 .37 mean 1 0.67 bi 1 .45 Ln upper l imit 4.25 upper l imit 70. 1 5 52 0. 1 3 Ln lower l imit 2 .92 lower l imit 1 8.63 C. albicans 3 1 53A Dilution Cone (u�mL) Clearing (mm) Log cone 30.00 1 .25 3 .40 3 .05 xmean 30.00 0.75 3.40 0.56 ymean 1 /2 1 5 .00 0.25 2.7 1 0.48 Sxx 1 /2 1 5 .00 0.00 2 .7 1 bo -3 .3 1 Ln mean 2 .6 1 mean 1 3 .60 bi 1 .27 Ln upper l imit 4.04 upper limit 56.76 52 0.08 Ln lower l imit 3 .09 lower l imit 2 1 .92 E.coli 0 157: H7 with 100mM NaCI Dilution Cone {u�mL} Clearing {mm} Log cone 30.00 2.25 3.40 3 .05 xmean 30.00 1 . 75 3.40 1 .25 ymean 1 /2 1 5 .00 0.75 2 .7 1 0.48 Sxx 1 /2 1 5 .00 0.25 2.7 1 bo -5.39 Ln mean 2.48 mean 1 1 .93 bi 2. 1 7 Ln upper l imit 62.96 upper l imit n/a 52 0. 1 3 Ln lower l imit 2.87 lower l imit 1 7 . 7 1 206 Appendix A 1 - Characterisation Data S. aureus 1056 MRSA with 1 00mM NaCI Dilution Cone {u�mq Clearing {mm� Log eone 30.00 0.00 3 .40 1 12 1 5 .00 0.00 2 . 7 1 1 14 7.50 0.00 2 .0 1 1 /8 3 .75 0.00 1 .32 bo n/a Ln mean n/a mean >30 bi n/a Ln upper l imit n/a upper limit n/a 52 n/a Ln lower l imit n/a lower limit n/a C. albicans 3153A with 100mM NaCI Dilution Cone {u�mq Clearing {mm� Log eone 30.00 0.75 3 .40 3.05 xmean 30.00 0.25 3.40 0.3 1 ymean 1 12 1 5 .00 0.25 2 . 7 1 0.48 Sxx 1 12 1 5 .00 0.00 2 . 7 1 bo - 1 .35 Ln mean 2 .49 mean 1 2 .02 bi 0.54 Ln upper l imit n/a upper limit n/a 52 0.08 Ln lower l imit n/a lower l imit n/a MIC of Pe E.coli 0157:H7 Dilution Cone {u�mq Clearing {mm� Log eone 1 25 .00 3 .00 4.83 3.79 xmean 1 /2 62.50 2.50 4. 1 4 2.25 ymean 1 /4 3 1 .25 2 .00 3 .44 2 .40 Sxx 1/8 1 5 .63 1 . 50 2.75 bo -0.48 Ln mean 0.67 mean 1 .95 bi 0.72 Ln upper l imit 0.63 upper limit 1 . 87 52 0.00 Ln lower l imit 0.7 1 lower l imit 2 .02 S. aureus 0156 MRSA Dilution Cone (u�mL) Clearing (mm) Log eone 1 25.00 2.50 4.83 1 12 62.50 1 .50 4. 1 4 1/4 3 1 .25 1 .00 3 .44 118 1 5 .63 0.50 2.75 bo -2. 1 8 Ln mean 2.32 mean 1 0.20 bi 0.94 Ln upper l imit 0.24 upper limit 1 .27 52 0.04 Ln lower l imit 2.97 lower l imit 1 9.47 C. albicans 3 153A Dilution Cone {u�mq Clearing {mm� Log eone 1 25.00 1 .75 4.83 4.48 xmean 1 12 1 25.00 1 .25 4.83 1 .00 ymean 1 14 62.50 0.75 4. 1 4 0.48 Sxx 1 /8 62.50 0.25 4 . 1 4 bo -5.47 Ln mean 3.79 mean 44.29 bi 1 .44 Ln upper l imit 5.67 upper limit 289. 1 5 52 0. 1 3 Ln lower l imit 4.35 lower limit 77.61 207 Appendix A 1 - Characterisation Data E.coli 0157: H7 with 100mM NaCI Dilution Cone {uglmL} Clearing {mm} Log cone 1 25.00 2.75 4.83 4.48 xmean 1 25 .00 2.25 4.83 2 .00 ymean 1 /2 62.50 1 . 75 4. 1 4 0.48 Sxx 1 /2 62.50 1 .25 4. 1 4 bo -4. 4 7 L n mean 3 . 1 0 mean 22. 1 5 bi 1 .44 Ln upper l i mit 7.07 upper limit 1 1 8 1 .5 7 s2 0. 1 3 Ln lower l imit 4.00 lower limit 54.56 S. aureus 1 056 MRSAwith IOOmM NaCl Dilution Cone {uglmL} Clearing {mm} Log cone I 1 25.00 0.00 4.83 1 /2 62.50 0.00 4. 1 4 1 /4 3 1 .25 0.00 3 .44 1 /8 1 5 . 63 0.00 2.75 bo nla Ln mean nla mean > 1 25 bi nla Ln upper l imit n/a upper limit n/a s2 n/a Ln lower l imit n/a lower limit n/a C. albicans 3 1 53A with I OOmM NaCl Dilution Cone {u�mq Clearing {mm} Log cone 1 25.00 2.25 4.83 4.48 xmean 1 25 .00 1 .75 4.83 1 . 50 ymean 1 /2 62.50 1 .25 4. 1 4 0.48 Sxx 1 /2 62.50 0.75 4. 1 4 bo -4.97 Ln mean 3 . 44 mean 3 1 .32 bi 1 .44 Ln upper l imit 6.39 upper limit 593 .20 s2 0. 1 3 Ln lower l imit 4. 1 6 lower limit 64. 1 2 208 Appendix A2 - Mechanisms Data APPENDIX A2 RAW DATA AND CALCULATIONS FROM MECHANISM OF ACTION STUDIES A2.1 RAW DATA FROM THE MICRO-BROTH DILUTION MIC METHOD Organism Run M IC !l:!g/mL) Organism Run MIC !l:!g/mL) SMAP29 OaBacSmini OaBac7.Smini SMAP29 OaBacSmini OaBac7.Smini 2 2 1 6 2 32 64 Escherichia coli 2 2 1 6 Staphylococcus 32 64 0 1 1 1 2 2 2 8 aureus NCTC 2 32 64 2 2 2 8 4 1 63 2 32 64 3 1 2 1 6 3 32 64 3 1 2 1 6 3 32 64 4 2 4 1 6 4 32 64 4 2 2 1 6 4 32 64 5 2 2 8 5 1 6 64 5 2 2 1 6 5 32 64 E. coli U B 1 005 1 0. 1 25 0.25 8 S. aureus 1 0.5 64 32 1 0. 1 25 0. 1 25 8 1 0.5 64 32 (rough K- 1 2 2 0. 1 25 0. 1 25 8 MRSA R 1 47 2 0.5 64 32 strain) 2 0. 1 25 0. 1 25 8 2 0.5 64 32 3 0. 1 25 0 . 1 25 8 3 0.5 64 64 3 0.25 0 . 1 25 8 3 0.5 64 64 4 0. 1 25 0 . 1 25 8 4 0.5 64 32 4 0. 1 25 0 . 1 25 8 4 0.5 64 32 5 0. 1 25 0. 1 25 4 5 0.5 64 32 5 0. 1 25 0 . 1 25 4 5 0.5 64 32 E. coli DC2 0. 1 25 0 . 1 25 2 S. aureus 1 056 1 4 1 6 >64 0. 1 25 0 . 1 25 2 1 4 16 >64 (antibiotic- 2 0. 1 25 0 . 1 25 2 MRSA 2 4 16 >64 supersusceptible 2 0. 1 25 0 . 1 25 2 2 4 1 6 >64 mutant) 3 0. 1 25 0. 1 25 2 3 4 16 >64 3 0. 1 25 0 . 1 25 2 3 4 16 >64 4 0.0625 0.0625 1 4 4 16 >64 4 0.0625 0 . 1 25 1 4 4 16 >64 5 0. 1 25 0 . 1 25 2 5 4 32 >64 5 0 . 1 25 0 . 1 25 2 5 4 32 >64 E. co1i 01 57:H7 1 2 8 32 0.25 32 32 1 2 8 32 S. epidermidis 1 0.25 16 32 2 2 8 32 clinical isolate 2 0.25 16 32 2 2 4 32 2 0.25 16 32 3 2 8 32 3 0.25 16 32 3 2 8 32 3 0.25 16 32 4 2 8 32 4 0.25 16 32 4 2 8 32 4 0.25 16 32 5 2 1 6 32 5 0.5 16 32 5 2 8 32 5 0.5 16 32 Salmonella 1 0.25 0.5 32 Enterococcus 1 2 32 64 1 0.25 0.5 32 2 16 64 typhimurium 2 0.25 0.5 32 faecalis ATCC 2 2 32 64 1 40285 2 0.25 0.5 32 292 1 2 2 2 32 64 3 0.25 0.5 32 3 2 32 32 3 0.25 0.5 32 3 2 32 64 4 0.25 0.5 32 4 2 32 64 4 0. 1 25 0.25 1 6 4 2 64 64 5 0.25 0.5 32 5 2 32 64 5 0.25 0.5 32 5 2 32 64 209 Appendix A2 - Mechanisms Data Organism Run MIC (J.1g/mL) Organism Run MIC Il!s/mL) SMAP29 OaBac5mini OaBac7.Smini OaBacSmini OaBac7.Smini s. typhimurium 0 . 1 25 0. 1 25 2 Candida 2 32 64 MS4252S 0 . 1 25 0. 1 25 2 albicans 1 05 2 32 64 (PhoPQ mutant; 2 0 . 1 25 0. 1 25 1 2 4 1 6 64 defensin 2 0 . 1 25 0. 1 25 2 2 2 1 6 64 supersusceptible 3 0. 1 25 0. 1 25 2 3 2 32 64 ) 3 0. 1 25 0. 1 25 2 3 2 32 64 4 0. 1 25 0. 1 25 2 4 2 32 64 4 0. 1 25 0. 1 25 2 4 2 32 64 5 0 . 1 25 0. 1 25 1 5 2 32 64 5 0.0625 0 . 1 25 1 5 2 32 64 Pseudomonas 1 4 4 1 6 C. albicans 1 4 1 6 >64 aeruginosa 1 4 4 1 6 31 53A 1 4 1 6 >64 PA01 2 4 4 1 6 2 4 1 6 >64 2 4 4 1 6 2 4 1 6 >64 3 4 4 32 3 4 1 6 >64 3 4 8 32 3 4 32 >64 4 4 4 1 6 4 4 1 6 >64 4 4 4 1 6 4 4 1 6 >64 5 4 4 1 6 5 8 32 >64 5 4 4 1 6 5 4 32 >64 P. aeruginosa 1 8 32 Z61 (antibiotic- 1 1 8 32 supers usceptible 2 4 32 mutant) 2 8 32 3 8 32 3 8 32 4 0.5 8 32 4 8 32 5 8 32 5 8 32 21 0 Appendix A2 - Mechanisms Data A2.2 EXAMPLE CALCULATION OF THE MEAN MIC AND CON F IDENCE INTERVALS FOR THE MEAN FROM THE RAW DATA Before calculating the mean MIC and confidence intervals the data were log-transformed. This was done to reduce the effect of the outlying data because the data sets were skewed. The mean and standard deviation of the log-transformed data was calculated using the Excel functions "AVERAGE" and "STDEV". The 95% confidence intervals were calculated using the Excel function "CONFIDENCE" where the alpha value was set to 0.05 . The limits of the confidence intervals were calculated by adding and subtracting the confidence interval value to the mean. These values for the log-transformed data were then exponentially-transformed to give the mean and confidence interval limits either side of the mean for the untransformed data. The variance on each side of the mean was different due to the log-transformation. The table below shows the calculated figures for the MIC of SMAP29 against E. coli 0 1 1 1 . Organism E. coli 0 1 1 1 Run 1 I 2 2 3 3 4 4 5 5 mean Ln MIC standard deviation Ln MIC confidence interval Ln MIC upper l imit Ln MIC = (mean Ln MIC + confidence interval Ln MIC) lower limit Ln MIC = (mean Ln M IC - confidence interval Ln MIC) mean MIC = exp(Ln MIC) upper l imit MIC = exp(upper l im it Ln MIC) lower l imit MIC = exp(lower limit Ln MIC SMAP29 Ln MIC 0.693 1 0.693 1 0.693 1 0.693 1 0.0000 0.0000 0.693 1 0.693 1 0. 693 1 0.693 1 0 . 5 545 0.2923 0. 1 8 1 1 0.7357 0 . 3 734 1 . 74 2 .09 1 .45 21 1 Appendix A2 - Mechanisms Data A2.3 RAW OAT A F ROM THE LPS BINDING ASSAY SMAP29 Run 1 Vol added h.d) Cone (llg/mL) 1 /eone Reading 0 0 .00 43.8 1 0.49 2 02 31 . 3 2 0.99 1 . 0 1 25.4 3 1 .48 0.67 22.2 4 1 . 98 0 .51 1 9. 1 5 2 .47 0 .41 1 7. 1 6 2 .96 0 .34 1 5. 5 SMAP29 Run 2 Vol added (Ill) Cone (llg/mL) 1 /eone Read ing 0 0 .00 38.2 1 0.49 2 .02 28. 5 2 0.99 1 . 0 1 24. 2 3 1 .48 0.67 2 1 .2 4 1 . 98 0 .51 1 9.6 5 2.47 0.41 1 7.6 6 2 .96 0.34 1 6.8 SMAP29 Run 3 Vol added hll) Cone (llg/mL) 1/conc Reading 0 0 .00 29.2 1 0 .49 2 .02 20.5 2 0.99 1 . 0 1 1 7.8 3 1 .48 0.67 1 3.6 4 1 . 98 0 .51 1 1 5 2.47 0 .41 9 .8 6 2 .96 0 .34 9 .2 LPS bind ing assay with SMAP29 4.50 4 .00 3.50 3.00 .0 .r:::. 2.50 .E: ::R. 2.00 0 -- ..- 1 . 50 1 . 00 0 .50 0.00 0 .00 212 y = 1 .2793x + 1 . 378 R2 = 0.9978 y = 1 . 1 694x + 1 . 094 1 R2 = 0.9638 y = 1 . 1 523x + 1 . 1 94 R2 = 0. 9979 • Run1 • Run2 A Run3 0 .50 1 . 00 1 . 50 2.00 1 /conc % Inh ib 1 /% Inh ib 0 .00 0 .29 3 . 50 0 .42 2 . 38 0.49 2 . 03 0 .56 1 . 77 0 .61 1 .64 0 .65 1 . 55 % Inh ib 1 /% Inh ib 0 .00 0 .25 3 .94 0 .37 2 .73 0 .45 2 .25 0 .49 2 .05 0 . 54 1 . 85 0 . 56 1 . 79 % Inh ib 1 /% Inh ib 0 .00 0 .30 3 . 36 0 .39 2 . 56 0 .53 1 . 87 0.62 1 .60 0.66 1 . 5 1 0.68 1 .46 2 .50 Appendix A2 - Mechanisms Data OaBae5mini Run 1 Vol added (ll ) Cone (J.19/mL) 1 /eone Read ing % Inhib 1 /% Inhib 0 0 .00 41 0 .00 #DIV/O! 4 1 . 98 0 .51 34. 4 0. 16 6 .21 8 3 .94 0 .25 29 0.29 3 .42 1 2 5 .91 0 . 1 7 27.5 0 .33 3 .04 1 6 7 .86 0. 1 3 26.6 0 .35 2.85 20 9 .80 0. 1 0 25.2 0 .39 2 .59 24 1 1 . 74 0 .09 24 .4 0.40 2.47 OaBae5mini Run 2 Vol added bd) Cone blg/mL) 1 /eone Reading % Inhib 1 /% Inhib 0 0 .00 43.4 0.00 #DIV/O! 4 1 . 98 0 .51 31 . 7 0.27 3 .71 8 3 .94 0 .25 27.6 0 .36 2 .75 1 2 5 .91 0 . 1 7 26. 3 0 .39 2 .54 1 6 7 .86 0. 1 3 25. 1 0 .42 2 .37 20 9 .80 0. 1 0 24.6 0.43 2 .3 1 24 1 1 . 74 0.09 23.6 0.46 2. 1 9 OaBae5mini Run 3 Vol added (lll) Cone (llg/mL) 1 /eone Read ing % Inhib 1 /% Inhib 0 0 .00 35.4 0.00 #DIV/O! 4 1 . 98 0 .51 28. 7 0. 1 9 5.28 8 3 .94 0.25 25.8 0 .27 3.69 1 2 5 .91 0. 1 7 24. 2 0 .32 3. 1 6 1 6 7 .86 0. 1 3 23 .2 0 .34 2 .90 20 9 .80 0. 1 0 21 . 8 0 .38 2 .60 24 1 1 . 74 0.09 20. 1 0.43 2 .3 1 LPS b ind ing assay w ith OaBac5min i 7.00 y = 8. 7626x + 1 . 614 • 6.00 R2 = 0 . 9757 5.00 Y = 6 . 7203x + 1 . 9325 ,Q 4.00 R2 = 0.989 .r:. c � 0 3.00 --..- 2.00 Y = 3. 509x + 1 . 9 173 • Run1 R2 = 0.9959 • Run2 1 .00 & Run3 0.00 0.00 0. 1 0 0.20 0 .30 0.40 0 .50 0 .60 1 /conc 213 Appendix A2 - Mechanisms Data OaBae7.5 mini Run 1 Vol added bd) Cone (llg/mL) 1 /eone Reading % Inh ib 0 0 .00 43.4 0 .00 4 1 . 98 0 .51 36. 9 0 . 1 5 6 2 .96 0. 34 34.6 0 .20 8 3 .94 0.25 33 0.24 1 0 4.93 0.20 31 . 7 0.27 1 2 5 .9 1 0. 1 7 30.6 0.29 25 1 2 . 22 0.08 25.4 0.41 OaBae7.5 mini Run 2 Vol added b.Ll) Cone (llg/mL) 1/eone Read ing % Inh ib 0 0.00 36.9 0 .00 4 1 . 98 0. 5 1 32. 5 0 . 1 2 8 3.94 0.25 30.8 0. 1 7 1 2 5. 9 1 0. 1 7 29.5 0.20 1 6 7 .86 0. 1 3 28.4 0.23 20 9 .80 0. 1 0 27.4 0 .26 24 1 1 . 74 0 .09 26. 5 0 .28 OaBae7.5 mini Run 3 Vol added (I.LI) Cone (llg/mL) 1/eone Reading % Inh ib 0 0 .00 50. 1 0 .00 4 1 . 98 0.51 39.4 0 .21 8 3 .94 0.25 34.8 0 . 3 1 1 2 5 .9 1 0. 1 7 30.6 0 .39 1 6 7 .86 0. 1 3 27.3 0.46 20 9 .80 0. 1 0 25.6 0.49 24 1 1 . 74 0 .09 24.4 0 .5 1 LPS bindi ng assay with OaBae7.5min i 9.00 ,------------------------, 8. 00 7 .00 6.00 y = 1 1 .246x + 2. 8689 R2 = 0. 9826 ,Q ..c 5.00 c � 4.00 .,.... 3.00 2.00 y = 9. 8959x + 1 .6569 1 .00 R2 = 0. 9988 y = 6 . 5852x + 1 . 42 18r--_-, R2 = 0 .9914 • Run1 • Run2 ... Run3 0.00 -\-----r------r------r----,---,------1 0 .00 0. 10 0 .20 0. 30 0 .40 0. 50 0 .60 1/conc 214 1 /% Inhib 6.68 4.93 4. 1 7 3 .71 3 .39 2 .41 1 /% Inhib 8.39 6.05 4.99 4 .34 3 .88 3 .55 1 /% Inhib 4.68 3 .27 2 .57 2 .20 2 .04 1 . 95 Appendix A2 - Mechanisms Data A2.4 CALCULATION OF IMAx AND 1 5 0 FROM LPS BINDING ASSAY RAW DATA Imax and 1 50 were calculated from the equation of the lines in an Excel worksheet using the fo llowing formulas: Imax = the maximum percentage of DPX displaced 1 00 = ---- y-intercept 150 = peptide concentration required to displace half I max - 1 = ---- x-intercept . - y-intercept x-mtercept = ---'-------"-- gradient The mean and standard deviation of Imax and 150 from the three runs were calculated using the Excel functions "AVERAGE" and "STDEV". The results are summarised below. SMAP29 Run y-int gradient x-int Imax 150 1 . 1 9 1 . 1 5 -1 .04 83.75 0 .97 2 1 . 38 1 .28 -1 . 08 72 .57 0 .93 3 1 . 09 1 . 1 7 -0. 94 9 1 .40 1 . 07 mean 82.57 0.99 SD 9.47 0.07 OaBac5mini Run y-int gradient x-int Imax 150 1 .61 8 .76 -0. 1 8 6 1 . 96 5.43 2 1 . 92 3 .51 -0. 55 52. 16 1 . 83 3 1 . 93 6. 72 -0.29 5 1 .75 3 .48 mean 55.29 3.58 SO 5.78 1 .80 OaBac7.5mini Run y-int gradient x-int Imax 150 1 1 .66 9 .90 -0. 1 7 60.35 5 .97 2 2.87 1 1 .25 -0.26 34.86 3 .92 3 1 .42 6.59 -0.22 70 .33 4.63 mean 55.18 4.84 SO 1 8.30 1 .04 215 Appendix A2 - Mechanisms Data The probability that there were no differences in the mean I max and I so values from the difference peptides was calculated using Student's Hest. The "TTEST" function in Excel was used. A single tai led t-test for two samples with unequal variances was used. The results are summarised below. probabi l ity meanSMAP29=meanOaBac5mini probabil ity meanSMAP29=meanOaBac7. 5mini probability meanOaBac5mini=meanOaBac7.5min i 21 6 Imax 150 0.0097 0 .0651 0 . 0523 0 .01 1 5 0 .4966 0. 1 831 Appendix A2 - Mechanisms Data A2.S RAW DATA FROM THE OUTER MEMBRANE PERMEABILlSATION ASSAY SMAP29 Run 1 Vol Added {1!11 Cone Added (mg/mL) Final Cone (l!g/mL) Initial Sam�le Increase % U�take 0 0 0.00 0.0 0.0 0.00 0.00 2 0.2 0.20 20.6 30.4 9.80 6. 1 8 4 0.2 0.39 22. 7 51 . 5 28.80 1 8 . 1 5 1 1 0.49 1 7 . 1 89.4 72.30 45.56 2 1 0.98 18.5 1 1 3.8 95.30 60.05 3 1 .47 20. 1 1 1 6.2 96. 1 0 60.55 1 4 1 .96 1 9 . 0 1 30. 1 1 1 1 . 1 0 70. 0 1 2 4 3.92 20.0 1 32.2 1 1 2.20 70.70 Polj'mj'xin B standard 1 .96 22.6 1 8 1 .3 1 58.70 1 00 00 SMAP29 Run 2 Vol Added (I!I) Cone Added (mg/mL) Final Cone (l!g/mL) Initial Sam�le Increase % U�take 0 0 0.00 0.0 0.0 0.00 0.00 2 0.2 0.20 30.6 55.0 24.40 1 5.25 4 0.2 0.39 28.3 57.2 28.90 1 8 06 1 1 0.49 50.6 1 1 9.8 69.20 43.25 2 1 0.98 45.0 1 56.2 1 1 1 .2 0 69.50 3 1 1 .47 56.4 1 40.2 83.80 52.38 1 4 1 .96 46.6 1 64.2 1 1 7.60 73.50 2 4 3.92 56.4 1 84.0 1 27.60 79 75 Polj'mj'xin B standard 1 . 96 20.0 1 80.0 1 60.00 1 00.00 SMAP29 Run 3 Vol Added {1!11 Cone Added (mg/mL) Final Cone (l!g/mL) Initial Sam�le Increase % U�take 0 0 0.00 0.0 0.0 0.00 0.00 2 0.2 0.20 1 6 . 8 24.7 7.90 4.39 4 0.2 0.39 22.4 46.7 24.30 1 3 50 1 1 0.49 1 9.2 1 09.9 90.70 50.39 2 0.98 39. 2 1 6 1 .9 1 22.70 68. 1 7 3 1 .47 35.4 1 53.3 1 1 7.90 65.50 1 4 1 .96 39.3 1 79.2 1 39.90 77. 72 2 4 3.92 36.8 1 94.1 1 57.30 87.39 Polj'mj'xin B standard 1 .96 37. 3 2 1 7.3 1 80.00 1 00.00 Oa8aeSmini Run 1 Vol Added {1!11 Cone Added (mg/mL) Final Cone (l!g/mL) Intitial Sam�le Increase % U�take 0 1 0.00 0.0 0.0 0.0 0.00 2 0.2 0.20 29.6 46.4 1 6. 8 6.33 4 0.2 0.39 34.3 6 1 . 1 26.8 10.09 1 1 0.49 25.9 1 20.9 95. 0 35.77 2 1 0.98 37.4 272.5 235 . 1 88.52 3 1 1 .47 34.8 244.8 21 0.0 79.07 1 4 1 .96 27. 7 240.5 2 1 2 . 8 80. 1 2 2 4 3.92 39. 3 277.0 237 . 7 89.50 Polj'mj'xin B standard 1 .96 41 . 7 307.3 265.6 100.00 Oa8aeSmini Run 2 Vol Added (1!11 Cone Added (mg/mL) Final Cone (l!g/mL) Intitial Sam�le Increase % U�take 0 1 0.00 0.0 0.0 0.0 0 00 2 0.2 0.20 40.8 50.7 9.9 3.74 3 0.2 0.29 42 . 9 63.4 20. 5 7 75 1 1 0.49 47. 1 1 68.6 1 2 1 . 5 45.92 2 0.98 47. 1 23 1 .5 1 84.4 69.69 3 1 .47 42. 5 277.9 235.4 88.96 4 1 .96 47.6 275.4 227 . 8 86.09 8 3.91 47.4 294 8 247.4 93.50 Polj'mj'xin B standard 1 .96 37.5 302.1 264.6 1 00.00 217 Appendix A2 - Mechanisms Data Oa8ae5mini Run 3 Vol Added !�I) Cone Added (mg/mL) Final Cone (�g/mL) Intitial Sam�le Increase % U�take 0 1 0.00 0.0 0.0 0.0 0.00 2 0.2 0.20 40.6 50.0 9A 4. 1 8 3 0.2 0.29 45.7 78.9 33. 2 1 4 . 76 1 1 OA9 44 A 208.3 1 63.9 72 . 84 2 1 0.98 42.0 1 95 . 1 1 53 . 1 68 04 3 1 A7 33. 1 228A 1 95.3 86.80 4 1 . 96 42 . 1 243.9 20 1 .8 89.69 8 3.91 4 1 . 1 23 1 . 5 1 90A 84.62 Pol�m�xin B standard 1 . 96 35.0 260.0 225.0 1 00.00 Oa8ae7.5mini Run 1 {1 6/7/02} Vol Added (�I) Cone Added (mg/mL) Final Cone (�g/mL) Intitial Sam�le Increase % U�take 0 0 0.00 0 .0 0.0 0.0 0.00 2 0.2 0.20 23. 1 30. 5 7 A 4.40 3 0.2 0.29 1 9. 8 30. 2 1 0A 6. 1 8 1 1 OA9 20A 97.0 76.6 45.54 2 1 0.98 24 . 2 1 34 . 8 1 1 0.6 65.76 3 1 1 A7 25.9 1 56.2 1 30.3 77.47 1 4 1 . 96 3 1 . 2 2 1 0. 0 1 78. 8 1 06.30 2 4 3.92 32A 200.0 1 67.6 99.64 Polymyxin B standard 1 . 96 4 1 . 8 2 1 0.0 1 68.2 1 00.00 Oa8ae7.5mini Run 2 {31 /7/02} Vol Added !�I) Cone Added (mg/mL) Final Cone !�g/mL) Intitial Sam�le Increase % U�take 0 0 0.00 0 . 0 0.0 0.0 0.00 2 0.2 0.20 32.0 37.2 5.2 2.46 3 0.2 0.29 33.9 47A 1 3. 5 6.40 1 1 OA9 40.3 1 39.6 99. 3 47 04 2 1 0.98 45.8 20 1 . 8 1 56.0 73.90 3 1 A7 44A 227.0 1 82.6 86.50 1 4 1 . 96 46. 2 279.0 232.8 1 1 0.28 2 4 3.92 49.8 265.3 21 5.5 1 02 08 Polymyxin B standard 1 .96 29.5 240.6 21 1 . 1 1 00.00 Oa8ae7.5mini Run 3 {518102} Vol Added (�I) Cone Added (mg/mL) Final Cone (�g/mL) Intitial Sam�le Increase % U�take 0 1 0.00 0.0 0.0 0.0 0.00 2 0.2 0.20 4 1 . 7 46.2 4.5 3A5 4 0.2 0.39 48.6 57.8 9.2 7. 1 4 1 1 OA9 47.5 1 07A 59.9 46.29 2 1 0.98 �2 3 1 32.7 90.4 69.85 3 1 1 A7 38.5 1 44.5 1 06.0 8 1 . 94 4 1 1 .96 45.6 1 85.3 1 39.7 1 07.98 8 3.91 42 . 1 1 7 1 .7 1 29.6 1 00. 1 2 Pol�m�xin B standard 1 .96 44. 5 1 73.9 1 29A 1 00.00 The % uptake was the % ofNPN taken up due to the peptide compared to that taken up due to 4J.lg/mL polymyxin B. It was calculated using the fo llowing formula: 0/ k - increase due to sample 1 00 /0 upta e - x increase due to polymyxin B The mean NPN uptake and the standard deviation for each peptide concentration over the three runs was calculated and graphed using SigmaPlot. 2 1 8 Appendix A2 - Mechanisms Data A2.6 ANALYSIS OF VARIANCE OF NPN UPTAKE DATA The NPN-uptake data from each experiment were analysed to determine if there was a difference between the mean NPN-uptake for the different peptides. This was done using the ANOV A function in the statistical package GenStat. The data was entered into GenStat in 4 columns. The first column contained the peptide concentration, the second column contained the run number, the third column contained the peptide number ( l =SMAP29, 2=OaBacSmini, 3=OaBac7.Smini) and the fourth co lumn contained the NPN-uptake. The "general analysis of variance" function was used. The y-variate was "NPN-uptake", the treatment structure was "peptide concentration x peptide" and the blocks was "run". The output below shows the ANOV A results. * * * * * Analysis of variance * * * * * Variate : NPN_Uptake Source of variation d . f . s . s . m . s . v . r . F pr . Run number stratum 2 2 01 . 6 1 1 0 0 . 8 0 2 . 3 0 Run number . *Units * stratum Peptide 2 1 0 2 9 . 57 5 14 . 7 8 1 1 . 7 3 < . 0 0 1 Peptide_concentration 7 8 9 2 1 2 . 6 1 1274 4 . 6 6 2 9 0 . 4 1 < . 0 01 Peptide . Pept ide_ concentrat ion 14 3 1 3 2 . 9 8 2 2 3 . 7 8 5 . 1 0 < . 0 01 Res idual 4 6 2 0 1 8 . 73 4 3 . 8 9 Total 7 1 9 55 95 . 4 9 219 Appendix A2 - Mechanisms Data A2.7 RAW DATA FROM THE OUTER INNER MEMBRANE DEPOLARISATION ASSAY SMAP29 Run 1 Vol Added (�I) Cone Added (mg/mL) Final Cone (�g/mL) Initial Sam�le Increase % Released 0 0 0.00 0.0 0.0 0.00 0.00 1 0.2 0. 1 0 1 0 . 2 27.7 1 7 .48 23.24 1 1 0.49 1 1 . 5 57.6 46 . 1 4 61 .36 2 1 0.98 1 3. 2 72. 1 58.91 78.34 1 4 1 .96 1 5. 5 72.6 57.07 75.89 1 .5 4 2.94 1 1 .2 74. 0 62.84 83.56 2 4 3.92 1 8. 9 83.4 64.47 85.73 Gramicid in S standard 1 .96 1 4 . 3 8 9 . 5 75.20 1 00.00 SMAP29 Run 2 Vol Added (�I) Cone Added (mg/mL) Final Cone (�g/mL) Initial Sam�le Increase % Released 0 0 0.00 0.0 0.0 0.00 0.00 1 0.2 0. 1 0 9.6 28.6 1 8. 99 25.56 1 1 0.49 8.3 67.6 59.28 79.78 2 1 0.98 5.6 7 1 . 7 66.07 88.92 1 4 1 .96 1 1 . 3 63. 2 5 1 . 93 69.89 1 . 5 4 2.94 1 4 . 2 71 .4 57. 1 5 76.92 2 4 3.92 1 5. 1 77.9 62.77 84.48 Gramicidin S standard 1 .96 8.7 83. 0 74.30 1 00.00 SMAP29 Run 3 Vol Added (�I) Cone Added (mg/mL) Final Cone (�g/mL) Initial Sam�le Increase % Released 0 0 0.00 0.0 0.0 0.00 0.00 1 0.2 0. 1 0 1 4.2 34. 5 20.31 2 1 .45 1 0.49 1 5 .6 77.9 62.28 65.77 2 0.98 1 6 . 8 90. 3 73.53 77.65 1 4 1 .96 1 5 . 2 88.0 72.80 76.87 1 . 5 4 2.94 1 7 . 9 89.5 7 1 .64 75.65 2 4 3.92 1 8 . 3 94.6 76.27 80. 54 Gramicid in S standard 1 .96 1 2 . 2 1 06.9 94.70 1 00.00 OaBacSmini Run 1 Vol Added (�I) Cone Added (mg/mL) Final Cone (�g/mL) Initial Sam�le Increase % Released 0 0 0.00 0.0 0.0 0.00 0.00 0.2 0. 1 0 1 5 . 1 24.4 9.28 1 2 . 34 1 0.49 1 6 . 3 27. 4 1 1 . 1 0 1 4.76 2 0.98 1 5 . 9 34.6 1 8. 70 24.87 1 4 1 .96 1 7 . 2 3 1 . 7 1 4.48 1 9.25 1 .5 4 2.94 1 8.2 45.0 26. 82 35.66 2 4 3.92 1 6. 3 45.8 29. 52 39.25 Gramicidin S standard 1 . 96 14.3 89. 5 75.20 1 00.00 OaBacSmini Run 2 Vol Added (�I) Cone Added (mg/mL) Final Cone (�g/mL) Initial Samele Increase % Released 0 0 0.00 0.0 0.0 0.00 0.00 1 0.2 0. 1 0 1 2.2 1 8 . 6 6.37 8.58 1 1 0.49 1 1 .5 26.4 1 4 . 87 20.01 2 1 0.98 1 4 . 3 31 . 1 1 6. 76 22.56 1 4 1 .96 1 6 . 8 42 . 0 2 5. 1 6 33.86 1 .5 4 2.94 1 9 .7 47. 8 28. 1 4 37.88 2 4 3.92 20 . 3 5 1 . 7 3 1 . 43 42.30 Gramicidin S standard 1 . 96 8.7 83. 0 74.30 1 00.00 220 Oa8acSmini Run 3 Vol Added !1!11 0 1 1 2 1 1 .5 2 Gramicidin S standard Oa8ac7.Smini Run 1 Vol Added (I!I) 0 1 1 2 1 1 .5 2 Gramicidin S standard Oa8ac7.Smini Run 2 Vol Added !1!11 0 1 1 2 1 1 .5 2 Gramicidin S standard Oa8ae7.Smini Run 3 Vol Added !1!11 0 1 1 2 1 1 . 5 2 Gramicidin S standard Conc Added (mg/mL) Final Conc !l!g/mL) 0 0.00 0.2 0. 1 0 1 0.49 1 0.98 4 1 . 96 4 2.94 4 3.92 1 .96 Conc Added (mg/mLI Final Conc (l!g/mL) 0 0.00 0.2 0. 1 0 1 0.49 1 0.98 4 1 .96 4 2.94 4 3.92 1 .96 Cone Added (mg/mL) Final Cone !l!g/mL) 0 0.00 0.2 0. 1 0 1 0.49 0.98 4 1 .96 4 2.94 4 3.92 1 .96 Cone Added (mg/mL) Final Cone !1!9/mL) 0 0.00 0.2 0. 1 0 1 0.49 1 0.98 4 1 .96 4 2.94 4 3.92 1 .96 Appendix A2 - Mechanisms Data Initial Samele Increase % Released 0.0 0.0 0.00 0.00 1 6. 9 23.2 6.28 6.63 1 7. 8 33. 1 1 5.32 1 6. 1 8 20. 3 44.7 24.41 25.78 2 1 . 4 49. 3 27.93 29.49 1 9.6 56.2 36.60 38.65 20. 1 57.6 37.48 39. 58 1 2 . 2 1 06.9 94.70 1 00.00 Initial Samele Increase % Released 0.0 0.0 0.00 0.00 1 8. 5 1 8.6 0.05 0.07 1 5.4 37 . 8 22.39 29.78 1 9. 5 46.2 26.66 35.45 20.4 54.9 34.49 45.86 2 1 . 1 58.9 37.79 50.25 20.6 57.2 36.60 48.67 1 4 . 3 89. 5 75.20 1 00.00 Initial Samele Increase % Released 0.0 0.0 0.00 0.00 1 7. 5 1 7.6 0 . 1 0 0. 1 3 1 8.6 45. 1 26.51 35.68 1 9. 2 43.5 24.27 32.67 1 8. 9 61 . 5 42.60 57. 34 20.5 52.6 32. 1 1 43. 2 1 2 1 . 6 58.2 36.59 49. 25 8.7 83. 0 74.30 1 00.00 Initial Samele Increase % Released 0.0 0.0 0.00 0.00 1 9. 2 1 9. 2 0.00 0.00 20.6 53. 5 32.92 34.76 2 1 . 8 38. 7 1 6. 92 1 7. 87 2 1 . 5 70.2 48.70 5 1 . 43 22.4 7 1 . 6 49.23 5 1 . 98 22.9 74.3 5 1 . 42 54. 30 1 2 . 2 1 06.9 94.70 1 00.00 The % released was the % of D iSC35 released due to the peptide compared to that released due to 4llg/mL gramicidin S. I t was calculated using the fo llowing formula: % released = increase due to sample x 1 00 increase due to gramicidin S The mean D iSC35 released and the standard deviation for each peptide concentration over the three runs was calculated and graphed using SigmaPlot. 221 Appendix A2 - Mechanisms Data A2.8 ANALYSIS OF VARIANCE OF DISC35 RELEASE DATA The DiSC35-release data from each experiment were analysed to determine if there was a difference between the mean DiSC35 released for the different peptides. This was done using the ANOV A function in the statistical package GenStat. The data was entered into GenStat in 4 columns. The first column contained the peptide concentration, the second co lumn contained the run number, the third co lumn contained the peptide number ( 1 =SMAP29, 2=OaBac5mini, 3=OaBac7.5mini) and the fourth column contained the D iSC35 released. The "general analysis of variance" function was used. The y-variate was "DiSC35-release", the treatment structure was "peptide concentration x peptide" and the blocks was "run". The output below shows the ANOV A results. * * * * * Analys is of variance * * * * * Variate : DiSC3 5 Source of variation d . f . s . s . m . s . v . r . F pr . Run stratum 2 6 3 . 0 5 3 1 . 53 1 . 6 2 Run . *Units* stratum Peptide 2 1 53 0 8 . 66 7 6 54 . 3 3 3 94 . 2 5 < . 0 0 1 peptide_conc 6 2 7 8 0 6 . 0 3 4 6 3 4 . 34 2 3 8 . 7 0 < . 0 0 1 Peptide . peptide_ conc 12 4 92 5 . 6 1 4 1 0 . 47 2 1 . 14 < . 0 0 1 Residual 4 0 7 7 6 . 6 0 1 9 . 42 Total 62 4 8 8 7 9 . 95 222 Appendix A3 - Conditions Data APPENDIX A3 RAW DATA AND CALCULATIONS FROM EFFECT OF CONDITIONS STUDIES A3.1 RAW DATA OF MICS AT DIFFERENT SALT CONCENTRATIONS Run MIC (llg/mL) NaCI conc (mM) SMAP29 OaBacSmini OaBac7.Smini 0 0 .5 1 2 0 0 .5 2 0 2 0 .5 2 0 2 1 2 0 3 0 .5 1 2 0 3 0 .5 1 2 50 1 1 2 8 50 1 1 4 8 50 2 4 8 50 2 4 8 50 3 1 4 8 50 3 1 4 8 1 00 1 6 32 1 00 1 1 6 1 6 1 00 2 1 6 32 1 00 2 1 6 32 1 00 3 2 1 6 32 1 00 3 1 6 32 250 4 32 64 250 2 32 64 250 2 4 32 64 250 2 4 64 64 250 3 4 32 64 250 3 4 32 64 223 Appendix A3 - Conditions Data A3.2 RAW OAT A OF MICS AT DIFFERENT CATION CONCENTRATIONS Na+ conc (mM) Run MIC (llg/mL) SMAP29 OaBacSmini OaBac7.Smini 0 1 0 .5 1 2 0 1 0 .5 0 .5 2 0 2 0 .5 1 1 0 2 0 .5 1 2 0 3 0 .5 1 2 0 3 0 .5 1 2 2 1 8 8 8 2 1 4 8 8 2 2 8 8 8 2 2 8 8 8 2 3 8 8 8 2 3 8 8 8 5 1 8 32 8 5 1 8 32 8 5 2 4 32 8 5 2 8 1 6 8 5 3 8 32 8 5 3 8 32 8 1 0 1 8 1 6 1 6 1 0 1 4 1 6 8 1 0 2 4 32 16 1 0 2 4 1 6 1 6 1 0 3 4 1 6 1 6 1 0 3 4 1 6 16 K+ conc (mM) Run MIC (llg/mL) SMAP29 OaBacSmini OaBac7.Smin i 0 1 0 .5 1 2 0 1 0 .5 0 .5 2 0 2 0 .5 1 1 0 2 0 .5 1 2 0 3 0 .5 1 2 0 3 0 .5 1 2 2 1 1 1 6 1 6 2 1 0 .5 1 6 1 6 2 2 1 1 6 8 2 2 1 1 6 1 6 2 3 1 1 6 1 6 2 3 1 1 6 1 6 5 1 4 32 32 5 1 2 1 6 1 6 5 2 4 32 32 5 2 4 32 32 5 3 4 32 32 5 3 4 32 32 1 0 1 4 32 32 1 0 1 4 1 6 1 6 1 0 2 4 1 6 1 6 1 0 2 4 1 6 1 6 1 0 3 4 1 6 1 6 1 0 3 4 1 6 1 6 224 Appendix A3 - Conditions Data Mg2+ conc (mM) Run MIC h!g/mL) SMAP29 OaBacSmini OaBac7.Smini 0 1 0 .5 1 2 0 1 0 .5 0 .5 2 0 2 0 .5 1 1 0 2 0 . 5 1 2 0 3 0 .5 1 2 0 3 0 . 5 1 2 2 1 8 32 64 2 1 8 64 32 2 2 8 64 64 2 2 8 64 64 2 3 8 64 64 2 3 8 64 64 5 1 1 6 32 32 5 1 32 32 32 5 2 1 6 64 64 5 2 1 6 32 32 5 3 1 6 32 32 5 3 1 6 32 32 1 0 1 >64 >64 >64 1 0 1 >64 >64 >64 1 0 2 >64 >64 >64 1 0 2 >64 >64 >64 1 0 3 >64 >64 >64 1 0 3 >64 >64 >64 Ca2+ conc (mM) Run MIC (!!g/mL) SMAP29 OaBacSmini OaBac7.Smini 0 1 0 . 5 1 2 0 1 0 . 5 0 .5 2 0 2 0 .5 1 1 0 2 0 .5 1 2 0 3 0 .5 1 2 0 3 0 . 5 1 2 2 1 1 6 64 64 2 1 1 6 64 64 2 2 1 6 64 64 2 2 1 6 32 64 2 3 1 6 64 64 2 3 1 6 64 64 5 1 1 6 >64 >64 5 1 1 6 >64 >64 5 2 32 >64 >64 5 2 1 6 >64 >64 5 3 1 6 >64 >64 5 3 1 6 >64 >64 1 0 1 >64 >64 >64 1 0 1 >64 >64 >64 1 0 2 >64 >64 >64 1 0 2 >64 >64 >64 1 0 3 >64 >64 >64 1 0 3 >64 >64 >64 225 Appendix A3 - Conditions Data A3.3 RAW DATA OF MICS AT DIFFERENT PH VALUES pH Run MIC blg/mL) SMAP29 OaBacSmini OaBac7.Smini 5 4 8 5 1 1 4 8 5 2 4 8 5 2 4 8 5 3 8 8 5 3 4 4 6 0 .5 8 1 6 6 0 .5 8 1 6 6 2 0 .5 8 1 6 6 2 0 .5 8 1 6 6 3 0. 5 8 1 6 6 3 1 1 6 1 6 7 1 4 8 7 1 2 4 7 2 1 2 4 7 2 1 2 4 7 3 2 4 7 3 2 4 8 0 .5 2 8 0 .5 1 2 8 2 0 .5 1 2 8 2 1 1 2 8 3 0 .5 1 4 8 3 0 .5 2 2 9 1 0 . 5 2 2 9 1 0. 5 4 2 9 2 0 . 5 2 2 9 2 0 .5 2 2 9 3 1 2 2 9 3 0 .5 2 4 226 Appendix A3 - Conditions Data A3.4 RAW DATA OF MICS AFTER HEAT ING TO DIFFERENT TEMPERATURES temperature Run M IC h�g/mq SMAP29 Oa8acSmini OaBac7.Smin i not heated 1 0 .5 2 2 not heated 1 0 .5 1 4 not heated 2 0.5 1 2 not heated 2 0 .5 1 2 not heated 3 0 .5 1 2 not heated 3 0 .5 1 2 30 1 0 .5 1 2 30 1 0 .5 1 2 30 2 0 .5 2 4 30 2 0 .5 1 2 30 3 0 .5 1 2 30 3 0 .5 1 2 40 1 0 .5 1 2 40 1 0 .5 1 2 40 2 0 .5 1 2 40 2 1 1 2 40 3 0 .5 1 2 40 3 0 .5 1 2 50 1 0 .5 2 4 50 1 0 .5 1 2 50 2 0 .5 1 2 50 2 0 .5 1 2 50 3 0 .5 1 2 50 3 0 .5 1 2 60 1 0 .5 2 2 60 1 0 .5 1 2 60 2 0 .5 1 4 60 2 0 .5 1 2 60 3 0 .5 1 2 60 3 0 .5 1 2 70 1 0 .5 1 2 70 1 0 .5 1 2 70 2 0 .5 1 2 70 2 0 .5 1 2 70 3 0 .5 1 2 70 3 0 .5 1 2 80 1 0 .5 2 2 80 1 0 .5 1 4 80 2 0 .5 1 2 80 2 0 .5 1 2 80 3 0 .5 1 2 80 3 1 1 2 90 1 4 8 8 90 1 4 8 1 6 90 2 2 8 8 90 2 4 4 8 90 3 4 8 8 90 3 4 8 8 1 2 1 1 4 8 1 6 1 2 1 1 4 8 8 1 2 1 2 4 8 8 1 2 1 2 4 8 8 1 2 1 3 4 8 8 1 2 1 3 4 8 8 227 Appendix A3 - Conditions Data A3.5 EXAMPLE CALCULATION OF THE MEAN MIC AND CONF IDENCE INTERVALS FOR THE MEAN FROM THE RAW DATA Before calculating the mean MIC and confidence intervals the data were log-transformed. This was done to reduce the effect of the outlying data because the data sets were skewed. The mean and standard deviation of the log-transformed data was calculated using the Exce I functions "AVERAGE" and "STDEV". The 95% confidence intervals were calculated using the Excel function "CONFIDENCE" where the alpha value was set to 0.05 . The limits of the confidence intervals were calcu lated by adding and subtracting the confidence interval value to the mean. These values for the log-transformed data were then exponentially-transformed to give the mean and confidence interval limits either side of the mean for the untransformed data. The variance on each side of the mean was different due to the log-transformation. The table below shows the calculated figures for the MIC of SMAP29 with no salt. NaCI cone (mM) o o o o o o Run 2 2 3 3 mean Ln M IC standard deviation Ln M IC confidence interval Ln M IC upper l imit Ln M IC = (mean Ln M IC + confidence interval Ln MIC) lower l imit Ln M IC = (mean Ln M IC - confidence interval Ln MIC) 228 mean MIC = exp(Ln MIC) upper l imit MIC = exp(upper limit Ln MIC) lower l imit M IC = exp(lower l imit Ln M IC SMAP29 Ln M IC -0 .6931 5 -0 .6931 5 -0.693 1 5 o -0.6931 5 -0.6931 5 -0.57762 0.282976 0.226424 -0.35 1 2 -0.80405 0 .56 1 23 1 0 .703844 0.44751 4 Appendix A3 - Conditions Data A3.6 ANALYSIS OF VARIANCE OF MIC DATA FROM DIFFERENT CONDITIONS The MIC data from each experiment were analysed to determine if there was a difference between the mean MICs for the different peptides and for the different conditions. This was done using the ANOV A function in the statistical package GenStat . The data was entered into GenStat in 4 columns. The first column contained the condition (eg NaCI concentration), the second column contained the run number, the third column contained the peptide number ( l =SMAP29, 2=Oa8ac5mini, 3=Oa8ac7.5mini) and the fourth column contained the Ln MIC. The "general analysis of variance" function was used. The y-variate was "Ln MIC", the treatment structure was "condition x peptide" and the blocks was "run". The output below shows the ANOV A results for the data collected at different salt concentrations. * * * * * Analysi s of variance * * * * * variate : Ln MIC Source of variation d . f . s . s . m . s . v . r . F pr . Run stratum 2 0 . 2 8 0 2 6 0 . 14 0 1 3 3 . 8 3 Run . *Units* stratum Salt Cone 3 8 7 . 5 7 5 91 2 9 . 1 9 1 9 7 7 9 7 . 9 0 < . 0 0 1 - Peptide 2 71 . 6 2 7 54 3 5 . 8 1 3 7 7 9 7 8 . 8 9 < . 0 0 1 Salt_Cone . Peptide 6 1 0 . 1 5 624 1 . 6 9 2 7 1 4 6 . 2 7 < . 0 0 1 Res idual 5 8 2 . 12 2 0 0 0 . 0 3 6 5 9 Total 71 1 7 1 . 7 6 1 9 5 From the output, the F-statistics for the condition and peptide were calculated. For these calculations the interaction error term was used instead of the residual error term. This was done because the residual error was likely to be an underestimate of the variability of the data because the data was discrete, so using the interaction error was a more conservative approach. The calculation of the F-statistics for the salt concentration data are given below. F- . . = m.s. peptide = 35 .8 1 377 = 2 1 1 5765 statistic e tid • p p e m.s salt_conc .peptide 1 .6927 1 F . . - m.s. salt conc _ 29. 1 9 1 97 - 1 7 24570 -statistIc salt Cone - • - • m.s salt_conc.peptIde 1 .6927 1 The F-statistics were used to calculate the probably of the means of each data set being the same. To do this, the Genstat code below was used. 229 Appendix A3 - Conditions Data "ca l c pvalue = CUF ( Fstat ; d . f . factor ; d . f . interac t i on ) " where Fstat was the calculated F -statistic, d. f. factor was the degrees of freedom of the factor (eg peptide, salt concentration) and d .f. interaction was the degrees of freedom of the interaction term For the salt concentration data the calculations of the p-values are given below. Probability that there are no differences between the mean MICs between peptides = "calc pvalue = CUF ( 2 1 . 1 5 7 6 5 ; 2 i 6 ) " = 0.00 1 9 1 5 Probability that there are no differences between the mean MICs between salt cones = "calc pva lue = CUF ( 1 7 . 2 4 5 7 0 ; 3 ; 6 ) " = 0.002357 The F-statistics and p-values for each set of conditions are summarised in the table below. Cond ition Factor salt cone peptide condition Na+ cone peptide condition K+ cone peptide condition Mg2+ cone peptide condition Ca2+ cone peptide condition pH peptide condition peptide temperature dT con l ion 230 m.s. d .f. factor factor 35.81 377 2 29. 1 91 97 3 4.93 1 32 2 25.67754 3 27.032 1 5 2 27.30352 3 6. 59288 2 64.42297 3 6.45274 2 70.29737 3 32 .7349 2 4.5643 4 23.68752 2 1 1 .07338 8 m.s. d .f. F-statisitic p-value interaction interaction 1 .69271 6 21 . 1 577 0.00 1 9 1 .69271 6 1 7 .2457 0.0024 1 .461 38 6 3 .3744 0. 1 042 1 .461 38 6 1 7 . 5707 0.0022 1 . 53256 6 1 7 .6386 0 .0031 1 . 53256 6 1 7 .81 56 0.0022 1 . 50364 6 4.3846 0 .0671 1 . 50364 6 42 .8447 0. 0002 0 .91 864 6 7 .0242 0.0268 0 .91 864 6 76.5233 0.0000 1 .5054 8 21 . 7450 0.0006 1 .5054 8 3 .0320 0 .0851 0. 1 3531 1 6 1 75 .061 1 0 .0000 0. 1 3531 1 6 81 .8371 0 .0000 Appendix A4 - Pilot-Scale Data APPENDIX A4 RAW DATA AND CALCULATIONS FROM P ILOT-SCALE EXTRACTION STUDIES A4.1 RAW DATA FROM THE MICRO-BROTH DILUTION MIC METHOD Organism MIC hlL crude extractlmL) Run 1 Run 2 Run 3 Escherichia coli 01 1 1 4 4 4 4 4 4 Escherichia coli 01 57:H7 4 4 8 4 4 4 Salmonella enteritidis 4 4 4 4 4 4 Salmonella typhimurium 8 8 4 8 8 8 Klebsiella pneumoniae 8 4 4 4 4 4 Pseudomonas aeruginosa 4 4 4 4 4 4 Pseudomonas fluorescens 4 4 4 4 4 Yersinia enterocolitica 8 4 4 4 4 4 Staphylococcus aureus NCTC 4163 4 4 4 4 4 4 Staphylococcus aureus 1 056 MRSA 8 8 1 6 8 8 8 Streptococcus faecalis 4 4 4 4 4 4 Bacillus cereus 4 8 4 4 4 4 Bacillus nato 4 4 4 4 8 8 Listeria monocytogenes 1 08 A 2 2 2 2 2 2 Listeria monocytogenes NCTC 1 0884 2 2 4 2 2 2 Listeria monocytogenes NCTC 7973 4 4 4 4 4 4 Candida albicans 31 53A 32 32 32 32 32 32 231 Appendix A4 - Pilot-Scale Data A4.2 EXAMPLE CALCULATION OF THE MEAN MIC AND CONFIDENCE INTERVALS FOR THE MEAN FROM THE RAW DATA Before calculating the mean MIC and confidence intervals the data were log-transformed. This was done to reduce the effect of the outlying data because the data sets were skewed. The mean and standard deviation of the log-transformed data was calcu lated using the Excel functions "AVERAGE" and "STDEV". The 95% confidence intervals were calculated using the Excel function "CONFIDENCE" where the alpha value was set to 0.05. The limits of the confidence intervals were calculated by adding and subtracting the confidence interval value to the mean. These values for the Jog-transformed data were then exponentially-transformed to give the mean and confidence interval limits either side of the mean for the untransformed data. The variance on each side ofthe mean was different due to the log-transformation. The table below shows the calculated figures for the MIC of the crude extract against E. coli 01 57:H7. Organism E. coli 01 57:H7 Run 2 2 3 3 mean Ln M IC standard deviation Ln M IC confidence interval Ln M IC upper l imit Ln M IC = (mean Ln M IC + confidence interval Ln MIC) lower l imit Ln MIC = (mean Ln MIC - confidence interval Ln MIC) 232 mean MIC = exp(Ln MIC) upper limit MIC = exp(upper l im it Ln MIC) lower l imit MIC = exp(lower l imit Ln MIC Crude extract Ln MIC 1 . 39 1 .39 2 .08 1 .39 1 . 39 1 . 39 1 . 50 0 .28 0 .23 1 .73 1 .28 4 .5 3 .6 5 .6 Appendix A4 - Pilot-Scale Data A4.3 CALCULATION OF THE SETTLING VELOCITIES OF DIFFERENT TYPES OF BLOOD CELLS To determine whether it is possible to separate the white blood cells from the red blood cells using a centrifugation step only, data about the various blood cell types were required. The table below shows the sizes, and the figure below shows the densities of the different cell types. Blood cell sizes and d istribution of various types of wh ite blood cells. Cell type Cell diameter (!lm) Portion of total leukocytes (% ) monocytes 1 2- 1 9 3 -8 lymphocytes 8- 1 2 20-25 basophils 7- 1 0 0 .5- 1 neutrophils 1 0- 1 2 60-70 eosinophils 1 2- 1 5 2-4 erythrocytes 6-8 n/a p latelets 2-4 n/a These data were taken from the websites of the Deparbnent of Anatomy and cel l biology, Indiana University School of Medicine (http://anatomy.iupui.edu/courses/histo_D5021D502f03/f03_labs/Lab71 Lab7f03.hbnl) and the Wadswroth Centre Clinical Chemistry and Haemotology (New York State Department of Health) (http://www.wadsworth.org/chemheme/heme/microscope/seg.hbn) . • monocytes lymphocytes . basophi ls • neutrophi Is I-] • eosinophils E • erythrocytes :l c Q) u 1 .06 1 .07 1 .08 1 .09 density (g/m1) 1 . 1 0 Graph showing the density distribution and relative amounts of the d ifferent cells present in b lood. This image was adapted from that g iven in the OptiPrep catalogue (www.sh iyaku­ dai ichi .jp/catalog-pub/shiyaku/optiprep/C09.pdf). 233 Appendix A4 - Pilot-Scale Data To determine the settling veloc ity of the so lids in a centrifuge the equat ion below was used (Meat and Livestock Australia, 2000). Where Vs is the settling velocity of the solid (m/s), G is the acceleration due to the centrifugal gravity (m/s2), ps is the density of the solid (g/ml), PL is the density of the l iquid (g/ml), d is the diameter of the solid particles (mm) and Jl is the viscosity of the l iquid (Pa.s). In the case of separation of the different blood cells from the blood plasma, G and J.l are constant. This means that the settling velocities of the various cell types are dependent on their densities and their diameters according to the equation below. Where Vs is the settl ing velocity of the solid (m/s), ps is the density of the solid (g/ml), PL is the density of the liquid (g/mL) and d is the diameter of the solid particles (mm). e.g. For a neutrophil of average size and density: Vs CX: (Ps - PL ) d 2 cx: ( 1 .088 - 1 ) 1 1 2 cx: 1 0 .6 This value was calculated for the average density as well as the minimum and maximum for each type of blood cell. The results are summarised in the table below. Comparison of the component proportional to the settl ing velocities inside a centrifuge for different blood cell types. Cell type Minimum (ps-pdd2 Average (ps-pdd2 Maximum (ps-pdd2 monocytes 6.3 27 .2 6 1 .2 lymphocytes 2.6 5 .6 7 .7 basophils 7.8 1 2 . 7 1 8 .2 neutrophils 7.2 1 5 . 5 25 . 3 eosinophils 7.5 1 3 . 8 2 1 .6 erythrocytes 3 .5 5 .2 6 .8 S ince the densities and sizes of the various types of blood cells are different, their settling veloc ities due to centrifugal gravity are also different. Although the red blood cells are the densest they are also the smallest of the blood cells, which means that their settling velocity 234 Appendix A4 - Pi lot-Scale Data would be slower than most of the other cell types. Of the white blood cells, lymphocytes have a settling veloc ity range similar to that of the red blood cells. This means that it would not be possible to separate these two types of cells by centrifugation. However the white blood cells of interest, the neutrophils, have a settling velocity range that is higher than that of the red blood cells. 235 Appendix A5 - Publications APPENDIX AS PEER-REVIEWED PUBLICATIONS A5.1 OVINE ANTIMICROBIAL PEPTIDES: NEW PRODUCTS FROM AN AGE­ OLD INDUSTRY Anderson RC, Wilkinson B, and Yu PL. (2004) Ovine antimicrobial peptides: new products from an age-old industry. Australian Journal of Agricultural Research, 55(1 ), 69-75 . 236