Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Characterization of bactenecin : a small antimicrobial cationic peptide Wu, Manhong 1999

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1999-390063.pdf [ 10.28MB ]
Metadata
JSON: 831-1.0089250.json
JSON-LD: 831-1.0089250-ld.json
RDF/XML (Pretty): 831-1.0089250-rdf.xml
RDF/JSON: 831-1.0089250-rdf.json
Turtle: 831-1.0089250-turtle.txt
N-Triples: 831-1.0089250-rdf-ntriples.txt
Original Record: 831-1.0089250-source.json
Full Text
831-1.0089250-fulltext.txt
Citation
831-1.0089250.ris

Full Text

Characterization of Bactenecin: A Small Antimicrobial Cationic Peptide By Manhong W u B. Sc. Fudan University, Shanghai, 1988 M . Sc. University of British Columbia, 1993 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R OF P H I L O S O P H Y in T H E F A C U L T Y OF G R A D U A T E STUDIES (Department of Microbiology and Immunology) We accept this thesis as conforming tdvthe required standard T H E U N I V E R S I T Y OF BRITISH C O L U M B I A September, 1998 © M a n h o n g Wu, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2788) A B S T R A C T Bactenecin is a 12-amino acid-long cyclic peptide from bovine neutrophils. It has the sequence: R L C R I V V I R V C R , and was cyclized by formation of a single disulphide bond. Computer modeling suggested it was amphipathic, with a central hydrophobic ring and positive charges located at both the C - and N-termini. Circular dichroism (CD) spectral studies were consistent with the computer model, in showing that bactenecin is rigid, (3-turn structure regardless of its environment. C D spectral studies showed that two linear variants of bactenecin were more flexible, then adopted random structure in phosphate buffer, a-helical structure in trifluoroethanol and (3-sheet structure in the presence of liposomes. Bactenecin was shown to have moderate antimicrobial activity against three wild-type Gram-negative bacteria, but was inactive against several Gram-positive bacteria. Disruption of the disulphide bond abolished the antimicrobial activity against the wild type Gram-negative bacteria, but had in improved antimicrobial activity against Staphylococcus epidermis and Entercoccus faecalis. Both the native cyclic and linear bactenecins interacted with the outer and cytoplasmic membranes of Escherichia coli differently. Nature bactenecin bound and permeabilized the outer membrane better than the linear variants, which explains its better activity against the wild-type Gram-negative bacteria. However, bactenecin had poor activity in depolarizing the cytoplasmic membrane. Conversely the linear variants demonstrated poor interaction with the outer membrane, yet were effective in depolarizing the cytoplasmic membrane. These results ii suggest that cyclization was important for bactenecin to interact with the outer membrane, and furthermore, that cyclic bactenecin could act by a fundamentally different mechanisms from its linear variants. Bactenecin was proposed to have a mechanism of action which involves other cellular targets than the bacterial membrane. Structure:function studies implied that increasing the number of positive charges, by introduction of additional arginine residues at both the N - and C-termini, resulted in improved activity. Linear variants were more active against Gram-positive bacteria. Introduction of the hydrophobic residue, tryptophan, into the loop of cyclic bactenecin improved the antimicrobial activity and dramatically broadened the antimicrobial spectrum. Two interesting candidates as potential therapeutic agents were identified, B a c 2 A - C N and Bac2R,W. These two peptides had good antimicrobial activity against a wide range of bacteria, including some clinically significant pathogens, and low toxicity. TABLE OF CONTENTS ABSTRACT ii TABLE OF CONTENTS iv LIST OF TABLES vi LIST OF FIGURES vii LIST OF ABBREVIATIONS ix ACKNOWLEDGEMENT xi INTRODUCTION 1 A. Antimicrobial Polycationic Peptides 2 B. Loop-Structured Cationic Peptides and Bactenecin 12 C. Therapeutic Potential 16 D. Aims of This Study 17 MATERIALS AND METHODS 18 A. Bacterial Strains and Growth Conditions 18 B. Chemicals 18 C. Recombinant Expression of Peptides 21 D. Purification of the Recombinant Peptides 25 E. Peptide Preparation 26 F. Structural Studies 30 G. Antimicrobial Activity and Hemolytic Activity 31 H. Membrane Permeabilization Assays 32 RESULTS 35 CHAPTER I: RECOMBINANT EXPRESSION OF CATIONIC PEPTIDES 35 A. Introduction 35 B. Recombinant Expression of Three Small Cationic Peptides in S. aureus 36 C. Production of Bactenecin in E. coli 46 D. Summary 49 CHAPTER TWO: ANTIBACTERIAL MECHANISM OF BACTENECIN 50 A. Introduction 50 B. Antimicrobial Activity 52 C. Secondary Structure by Circular Dichroism 54 D. Interaction with Bacterial cells 54 E. Summary 62 iv CHAPTER THREE: INNER MEMBRANE PERMEABILITY 64 A. Introduction 64 B. Fluorescence Quenching of diSC3-(5) by E. coli DC2 cells 66 C. The relationship of K + concentration outside the cells and the fluorescence intensity in the presence of valinomycin 66 D. Antimicrobial activities in the presence of potassium chloride 72 E. The interaction of adielical cationic peptides with the cytoplasmic membrane... 72 F. The interaction of P-structured cationic peptides with the cytoplasmic membrane 75 G. The interaction of extended structured cationic peptides with the cytoplasmic membrane 78 H . The interaction of looped structured cationic peptides with the cytoplasmic membrane 78 G. Summary 84 CHAPTER FOUR: STRUCTURE AND FUNCTION RELATIONSHIPS 85 A. Introduction 85 B. Linearization 86 C. Positive Charges 90 E. Disulphide bond 98 DISCUSSION 102 A. Overview 102 B. Recombinant Expression of Cationic Peptides 103 C. Antimicrobial Mechanism of Bactenecin 107 D. Design of Novel Peptides with Improved Activity 113 E. Potential Application of Bactenecin 114 F. Future Studies 115 REFERENCES 120 L I S T O F T A B L E S Table I: Loop-Structured Cationic Peptides 12 Table II: Bacterial Strains 19 Table III: Amino Acid Sequences of Bactenecin and its Derivatives 28 Table IV: Amino Acid Sequences of other Cationic Peptides Used in This Study 29 Table V : MICs of Three Partially Purified Cationic Peptides Expressed in S. aureus 46 Table VI: MICs of Bactenecin, its Linear Variant Bac 2S and its Reduced Form Lin-Bac 53 Table VII: MICs of E.coli D C 2 in the Presence and Absence of K + 73 Table VIII: MICs of Linear, Amidated Bactenecin Derivatives 88 Table IX: MICs of Bactenecin Derivatives in the Reduced Form 89 Table X : MICs of Positive-Charge Bactenecin Derivatives 94 Table XI: MICs of Hydrophobicity Bactenecin Derivatives 97 Table XII: Agglutination Activities of Bactenecin and its Derivatives on Human Red Blood Cells 100 Table XIII: Influence of Bactenecin Structural Modifications on Antimicrobial Activity 101 vi L I S T O F F I G U R E S Figure 1: Amino acid sequences of the three recombinantly expressed cationic peptides. 23 Figure 2: Sequences of oligonucleotides 24 Figure 3: Production of protein A/cationic peptide fusion proteins in S. aureus 37 Figure 4: Cleavage of cationic peptides from protein A 39 Figure 5A: Elution profile of the CNBr-digested protein A/apidaecin fusion from a Biogel PI00 Column 40 Figure 5B: Purification of apidaecin on a Biogel P100 column 41 Figure 6A: Elution profile of the CNBr-digested protein A/bactenecin fusion from a Biogel P100 Column. 42 Figure 6B: Purification of bactenecin on a Biogel PI 00 Column 43 Figure 7A: Elution profile of the CNBr-digested protein A/indolicidin fusion from a Biogel PI00 Column 44 Figure 7B: Purification of indolicidin on a Biogel PI00 Column 45 Figure 8: Production of bactenecin in E. coli 48 Figure 9: Amino acid sequence of bactenecin and its linear variants 51 Figure 10A: C D spectra of bactenecin, its reduced form and bac2S in phosphate buffer. 55 Figure 10B: C D spectra of bactenecin, its reduced form and bac2S in the presence of P O P C / P O P G liposomes 56 Figure 10C: C D spectra of bactenecin, its reduced form and bac2S in 60% (v/v) T F E . . .57 Figure 10D: C D spectra of bactenecin, its reduced form and bac2S in l O m M SDS 58 Figure 11: Binding of peptides to LPS as assessed by their ability to displace dansyl polymyxin B from E. coli UB1005 LPS 60 Figure 12: Pep tide-induced outer membrane permeabilization measured by the N P N uptake assay in E. coli U B 1005 61 Figure 13: Fluorescence quenching of diSC3-(5) by log-phase E. coli cells 67 Vll Figure 14: Effect of external KC1 concentration on the fluorescence intensity of 0.4 p M diSC3-(5) incubated with E. coli D C 2 cells and valinomycin 68 Figure 15: Kinetics of fluorescence intensity changes of diSC3-(5) incubated with E. coli D C 2 cells in the presence of valinomycin and KC1 70 Figure 16: The effect of 0.1 M KC1 on the dissipation of cytoplasmic membrane potential caused by gramicidin S 71 Figure 17: Kinetics of fluorescence intensity changes of diSC3-(5) incubated with E. coli D C 2 in the presence of a-helical cationic peptides 74 Figure 18: Effect of a-helical cationic peptides on the fluorescence intensity changes of diSC3-(5) incubated with E. coli D C 2 cells 76 Figure 19: Kinetics of the fluorescence intensity changes of diSC3-(5) incubated with E. coli DC2 in the presence of p-structured cationic peptides 77 Figure 20: Effect of B-structured cationic peptides on fluorescence intensity of diSC3-(5) incubated with E. coli D C 2 cells 79 Figure 21: Kinetics of fluorescence intensity changes of diSC3-(5) incubated with E. coli D C 2 in the presence of extended cationic peptides 80 Figure 22: Effect of extended cationic peptides on the fluorescence intensity of diSC3-(5) incubated with E. coli D C 2 cells 81 Figure 23: Kinetics of fluorescence intensity changes of diSC3-(5) incubated with E. coli D C 2 in the presence of loop-structured cationic peptides 82 Figure 24: Effect of loop-structured cationic peptides on the fluorescence intensity of diSC3-(5) incubated with E. coli D C 2 cells 83 Figure 25: Computer generated model of bactenecin 91 Figure 26: Computer generated model of bacR,P 92 Figure 27: Computer generated model of bacP 117 Figure 28: Computer generated model of bacW 118 Figure 29: Computer generated model of bac2R,W 119 viii L I S T O F A B B R E V I A T I O N S A U : Acid-Urea C D : Circular Dichroism C E M E : cecropin/melittin hybrid peptide (CA1-8M1-18) C F U : colony forming unit CNBr: cyanogen bromide diS-C3-(5): 3, 3-dipropylthiacarbocyanine E D T A : ethylenediamine tetraacetic acid F P L C : fast protein liquid chromatography H E P E S : (N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid) sodium salt HNP: human neutrophil peptide IPTG: isopropyl-P-D-thiogalactoside L B N S : Luria Broth Normal Salt LPS: Polyliposaccaride M I C : Minimum Inhibitory Concentration N M R : nuclear magnetic resonance N P N : 1-N-phenylnaphtylamine O D : optical density P A G E : polyacrylamide gel electrophoresis PCR: polymerase chain reaction POPC: 1 -pamitoyl-2-oleoyl-sn-glycero-3 -phosphocholine P O P G : l-palmitoyl-2-oleoyl-sn-glycero-3-phosphoglycerol ix N A P S : Nucleic Acid/Protein Service unit RP: reverse phase rpm: revolutions per minute SDS: sodium dodecyl sulfate T F A : trifluroacetic acid T F E : trifluroethanol U V : ultraviolet A C K N O W L E D G E M E N T I would like to thank my supervisor Bob Hancock, for his excellent supervising, his patience, and his constant encouragement. I would like to thank the members in Hancock's lab, present and past who were always helpful. In addition, I would like to thank the members of my committee, Drs Tony Warren, Brett Finlay, Bi l l Mohn for their helpful discussion and advice. Finally I would like to thank my parents, my brother Haiying W u and my friends (especially Liping, Liyue, Kewen and Hua) who were always there for me and encouraged me to carry on. The financial support of the British Columbia Science Council and of Micrologix Biotech Inc. was also greatly appreciated. XI I N T R O D U C T I O N The rapid emergence of antibiotic resistance has been of great concern in recent years. Some bacteria have developed resistance to virtually all antibiotics available in the world due to extensive clinical usage of antibiotics (Neu, 1992). Resistance in the Enterococci and Staphylococci is becoming particularly troublesome (McManus, 1997). Efforts have been made to prevent and control the resistance problem, including public health efforts to reduce unnecessary usage of antibiotics in humans and animals, as well as the development of new antimicrobial agents (Cohen, 1992). The latter has been the focus of research scientists interested in this area. Based on our current understanding of the molecular mechanisms of bacterial resistance, researchers continue to improve the existing antimicrobial agents by targeting "resistance factors" (Setti and Micetich, 1998). It seems that bacteria can easily develop resistance by the same mechanisms to the new analogs of existing antibiotics. Therefore, the development of a new class of antibiotics has become increasingly important. Among the possible candidates, a group of antimicrobial cationic peptides has attracted increasing research and clinical interest due to their unique properties. This group of gene-encoded "endogenous peptide antibiotics" (Martin et al., 1995) is widely distributed in nature, and are key components of the non-specific defense systems for plants and animals (Boman, 1995). For conventional antibiotics to reach their targets and kill microorganisms, they have to overcome two major diffusion barriers, the outer and cytoplasmic membranes. Bacteria can easily develop mechanisms to reduce accessibility to these barriers. Cationic peptides have been proposed to exert their antimicrobial 1 activity by perturbing the barrier function of bacterial membranes. Conventional antibiotics function by an entirely different mechanism, in that they usually interfere with either cell wall or macromolecular synthesis (Neu, 1992). Cationic peptides have thus become a new source of potentially novel antimicrobial agents. Therefore, a detailed characterization and understanding of their structures and mode of action, and a correlation of structure with function is essential in assessment of their therapeutic prospects. A. Antimicrobial Polycationic Peptides 1. In Nature Antimicrobial polycationic peptides have been found in a variety of sources, from prokaryotes (bacteria) to eukaryotes (plants and animals, such as amphibians, mammals, insects) (Hancock et al, 1995). Some examples of the best known and studied peptides are: 1) from mammals, classic defensins (Kagan et al., 1994); 2) from amphibians, magainins from the granular glands present in the skin of Xenopus laevis (Duchohier, 1994); 3) from insects, insect defensins (Lehrer, 1993), cecropins from the North American silk moth Hyalophora cecropia (Hultmark et al., 1993), and melittins from the venom of the honey bee Apis mellifera (Dempsey, 1990); 4) from plants, thionins (Bohlmann, 1994); and 5) nisin from bacteria Lactococcus lactis (Jung and Sahl, 1991). In recent years, it has become clear that these endogenous peptide antibiotics constitute part of the first line of host defense (Boman, 1991). Cationic peptides can be induced and synthesized much more rapidly than immuoglobulins upon infection, before 2 the adaptive immune system is activated, and they function without the high specificity and memory of immunoglobulins or immune cells (Boman, 1995; Nissen-Meyer and Nes, 1997). In mammals, antimicrobial peptides are present at high concentrations in phagocytes (e.g. macrophages, neutrophils, N K cells) and mucosal epithelial cells (e.g. Paneth cells). Human defensins constitute about 5-7% of total cellular protein in neutrophils (Ganz, 1987) and participate in the early phases of defense against invading microbes by the oxygen-independent pathway of neutrophils, as an adjunct to the oxygen-dependent pathway (Lehrer, 1991). Besides being antimicrobial, defensins may also play a role in inflammation, tissue injury and repair (Lehrer, 1993). In lower life forms, such as invertebrates, which have no adaptive immunity, cationic peptides are the major defensive system against infection (Boman, 1995). Insects produce cationic peptides in their fat bodies and haemolymph, where they are induced upon bacteria challenge (Boman, 1995; Hoffmann and Hetru, 1992). Frog skin contains abundant antimicrobial peptides to keep wounds safe from infections in contaminated pond water (Zasloff, 1987). Cationic peptides also function to keep the natural flora of bacteria in a steady state, in a variety of different niches such as the skin, the mouth, and the intestine (Boman, 1995). The natural cationic peptides of animals and plants are synthesized ribosomally as precursor and are processed into their mature forms by cleavage of a signal peptide and a prosequence (Zanetti, 1995; Hancock, 1997). 3 2. In Vitro Activity The above-mentioned cationic peptides and many others exhibit potent activities against a rather broad spectrum of microorganisms including bacteria, enveloped viruses, fungi and tumor cells. For example, defensins are active against Gram-positive and Gram-negative bacteria (including spirochetes and mycobacteria), fungi and enveloped viruses (Lehrer et al., 1993). Magainins (Bevin and Zasloff, 1990) also exhibit antifungal and antiviral activities in addition to their anti-bacterial activities. Defensins (Lichtenstein, et al., 1988) and magainin (Cruciani, et al., 1991) have also been shown to be active against tumor cells. Most interestingly, these cationic peptides (e.g. tachyplesin I) not only inhibit the growth of normally susceptible Gram-positive and negative bacteria but also inhibit with equal potency clinically important antibiotic-resistant strains e.g. methicillin-resistant S. aureus (Ohta et al., 1992). However, the spectrum of antimicrobial activity varies. For example, classical defensins (Kagan, et al., 1994) and insect defensins (Hoffmann and Hetru, 1992) are more active against Gram-positive than Gram-negative bacteria. In contrast, cecropins (Boman, 1991) and magainins (Bevin and Zasloff, 1990) exhibited greater potency against Gram-negative bacteria than Gram-positive bacteria. The reasons for this selectivity against Gram-positive or Gram-negative bacteria have not been made clear in the literature to date. 3. Structural Features and Categories Polycationic peptides show significant diversity in their sizes, sequences and structures. They range from 12 to 46 amino acids long with diverse amino acid compositions (Hancock, 1997). Despite their diversity, antimicrobial peptides share 4 common structural features, being virtually all cationic and very often amphiphilic. The cationic antimicrobial peptides have a net charge of at least +2 at neutral p H , usually due to the presence of arginine or lysine residues in their amino acid sequence (Hancock, 1997). Their secondary structures often contain a hydrophobic domain and a hydrophilic domain. The basicity and amphipathicity of cationic peptides are essential for their antimicrobial activities (Saberwal and Nagaraj, 1994; Hancock, 1997). The hydrophilic (positively charged) surface facilitates the interaction of the peptides with the negatively charged bacterial surface, e.g. LPS on the outer membrane of Gram-negative bacteria, teichoic acid on the Gram-positive bacteria, or negatively charged headgroups of the phospholipids in the lipid bilayer (Wade et al., 1992; Piers et al., 1994; Kagan et al., 1994). Some evidence showed that the property of amphipathicity was important to permit the antimicrobial peptides to interact with the membrane and to form channels or pores (Saberwal and Nagaraj, 1994; Maloy and Kari, 1995). Based on their known or assumed secondary structures, polycationic peptides could be categorized into four major groups: 1) P-sheet structures stabilized by two or three disulphide-bridges; 2) amphipathic a-helices often comprising a helix-turn-helix arrangement; 3) extended polyprohelices with a predominance of one or more amino acids like proline, or tryptophan; and 4) loop-structures containing only a single disulphide-bridge. These are described immediately below, except for the loop-structured peptides which are described in section 5. 5 3.1 P-structured peptides. Cysteine-rich cationic peptides usually contain p-strands stabilized by two or three disulphide-bridges. Antimicrobial peptides with six cysteine residues generally have some sequence conservation and have been classified as defensins (Ganz and Lehrer, 1994). There are three subgroups of defensins, classical defensins, P-defensins and insect defensins. They differ from each other in the spacing and connectivity of their six cysteine residues. Classical defensins contain triple-stranded antiparallel P-sheet without the presence of a-helical stretches (Zhang et al., 1992; Pardi et al., 1992). In contrast, insect defensins consist of a short a-helix and an antiparallel P-sheet connected by a loop (Hoffmann and Hetru, 1990). The three dimensional structure of P-defensins is unknown at present. The details of the secondary and tertiary structures of defensins are known from two-dimensional nuclear magnetic resonance, circular dichroism, Fourier transform infra red spectroscopy, and X-crystallography (Bach et al., 1987; Pardi et al., 1988; Selsted and Harwig, 1989; Hi l l et al., 1991; Zhang, et al., 1992;). Human defensin 3 has been shown by X-ray crystallography to be present as basket-like dimer with the top half containing the polar residues and the lower half rich in hydrophobic residues (Hill, 1991). Structural studies on human and rabbit defensins suggest this may be a common structural theme for all defensins (Ganz, et al., 1990). Tachyplesin I (Kawano, et al., 1990) and protegrins (Harwig et al., 1995) which contain two disulphide bonds also adopt an antiparallel P-sheet (so-called hairpin) structure connected by a P-turn. 6 3.2 a-Helical structures Cecropins (Holak et a l , 1988), melittin (Bazzo et al., 1988) and magainins (Marion et al., 1988) are typical examples of a-helical structured peptides. They are devoid of any cysteine residues and adopt a distinct secondary structure upon interacting with hydrophobic environments like membranes. For cecropin A this consists of two a-helical segments joined by a hinge region (Ala-Gly-Pro). The strongly basic N-terminus contains an almost perfect amphipathic helix and the C-terminus contains a more hydrophobic helix (Steiner, 1982; Holak et al., 1988). Melittin adopts a similar helix-bend-helix structure. In contrast to cecropin, its N-terminus is hydrophilic and its C -terminus is hydrophobic. Magainin from amphibian skins also forms an amphipathic a-helix (Marion, 1988). 3.3 Extended structures. The group of antimicrobial peptides that are rich in specific amino acids include Bac5 and Bac7 (Frank et al., 1990), PR-39 (Agerberth et al., 1991), prophenin (Harwig et al., 1995) and indolicidin (Selsted et al., 1992). They all contain a high content of proline, with Bac5 and Bac7 also rich in arginine, prophenin rich in phenylalanine and indolicidin rich in tryptophan. The presumptive three dimensional structures show that Bac5 and Bac7 contain a series of turn/coil-segments interspersed with extended amphipathic structures (Frank et al., 1990). Indolicidin has been proposed to form an extended polyproline type-II helix (Falla et al., 1996). 4. Proposed Mechanism The ability of polycationic peptides to disrupt the cytoplasmic membrane of bacteria has been proposed to represent their mechanism of action (Hancock and Lehrer, 1998). This action is proposed to involve three steps, 1) binding to the cell surface; 2) permeabilization of the outer membrane and then the cytoplasmic membrane; 3) loss of viability of cells involving cell lysis and possible D N A injury (Lichtenstein et al., 1988). The killing is believed to be initiated by the electrostatic interaction of cationic peptides with the negatively charged cell surface. For Gram-negative bacteria, the positively charged domain of the cationic peptides binds to the divalent cation binding sites of LPS (Sawyer et al., 1988; Piers et al., 1994). The displacement of the native cations C a 2 + and M g 2 + disrupts the structures of the outer membrane, due to the bulky size of the cationic peptides. This disruption subsequently results in the self-promoted uptake of cationic peptides (Hancock, et al., 1995). For Gram-positive bacteria, the cell wall contains covalently-bound, negatively charged teichuronic acid and carboxyl groups in the amino acids in the peptidoglycan (Hamoond, 1984), and these are probably the initial binding sites for the cationic peptides. The lethal step of killing is believed to involve the interaction of peptides with the cytoplasmic membrane (Lehrer, 1989; Westerhoff et al., 1989; Skerlavaj et al., 1990). The interaction between the peptides and the cytoplasmic membrane is thought to be determined by factors such as the anionic lipid composition of the bacterial membrane, and by the presence of an electrochemical potential across the membrane. After positively charged cationic peptides bind to the negatively charged head groups of lipid, under the influence of a transmembrane potential (oriented internal negative), the 8 peptides insert into the membrane, and undergo conformation changes to adopt a structured form such as an amphipathic a-helix. In model systems, peptides, like cecropin A (Christensen, et al., 1988), magainins (Duclohier, 1994), melittin (Tosteson, 1981) and defensins (Kagan et al., 1990) then aggregate to form multimers, which allow them to form channels or pores with their hydrophobic faces positioned toward the membrane and their hydrophilic faces oriented towards the interior of these channels or pores. In the cytoplasmic membrane of bacteria, this would result in leakage of protons, causing dissipation of the membrane potential, and leakage of other small compounds, thus causing cell death. After the membrane permeability was changed, Lehrer et al (1989) observed the simultaneous loss of the proton motive force, cessation of biosynthesis of macromolecules like DNA, R N A and protein, and leakage of intracellular contents, which would presumably be responsible for eventual cell death. Although it seems that channel formation is fundamental to the interaction of peptides with the cytoplasmic membrane, it is still not clear i f channel formation is the mechanism by which these peptides kiir microorganisms. The same factors also seem to regulate the selectivity of cationic peptides for bacterial membranes over eukaryotic cell membranes. 1) The compositions of their membranes are quite different with bacterial membranes predominantly containing negatively charged lipids such as phosphatidylglycerol and cardiolipin, while the eukaryotic cell membrane is largely composed of zwitterionic lipids such as phosphatidylcholine and sphingomyelin. 2) Eukaryotic cell membranes are rich in 9 cholesterol, which may inhibit membrane insertion. 3) Bacterial cells have large, interior-negative transmembrane potentials of around -140 m V whereas eukaryotic plasma membranes have membrane potentials of only -20 m V or so. There are a few cationic peptides which have been proposed to function through non-membranous mechanisms, including apidaecin from honeybees (Casteels, et al., 1989), thanatin and attacin. Apidaecin was shown to lack the ability to permeabilize the membrane and the all-D-enantiomer of apidaecin was completely devoid of antibacterial activity (Casteels and Tempst, 1994). Apidaecin has no effect on Gram-positive bacteria and it does not show homology with any other cationic peptides. It was suggested that the action of apidaecin on bacteria involved a stereoselective recognition of a chiral cellular target. Correspondingly, it has been reported that attacin inhibits synthesis of outer membrane proteins in E. coli through specific interference with omp gene transcription (Carlsson, et al., 1991; Hultmark et al., 1983). 5. Structure and Function Relationships A variety of studies have examined the relationship between chemical structure and antimicrobial activity. There are several structural requirements that have been found to be of importance in determining antimicrobial activity. In general, the results clearly indicate a requirement for amphipathic structures and positive charges. Although it is speculated that more positive charges will facilitate the interaction of cationic peptides with the target organism, there is no absolute relationship between the number of positive charges and activity. The position of specific positive charges may be of importance 10 (Blondelle and Houghten, 1991). Increasing the a-helical content appears to increase the antimicrobial activity of a-helical peptides (Andreu et al., 1985; Steiner et al., 1988; Frohlich and Wells, 1991; Blondelle and Houghten, 1991). For the a-helical peptides, enantiomers have equal activity (Besalle et al., 1990; Wade et al., 1990), showing that chirality is not important. For the disulfide-bonded peptides, intra-chain disulphide bonds are essential for antimicrobial and cytotoxic activity (Lehrer, 1993). The disulfide structure of the defensins is essential for structural stability and potency (Ganz, et al., 1987; Selsted et al., 1983). Tachyplesin I is a 17-residue peptide from horseshoe crab hemocytes. It contains two disulfide bonds, and studies showed that the deletion of these disulfide bridges caused a major decrease in all activities (Tamamura et al., 1993; Matsuzaki, et al., 1993). Protegrins are p-sheet structured peptides, containing 16 to 18 residues and two intramolecular disulfide bonds. Both intra-chain disulfide bonds are required for the antiparallel P-sheet conformation in membrane-mimetic environments and for optimal antimicrobial activity, especially in media containing NaCl, with at least one disulphide bond being needed for any activity (Sylvia et al, 1996; Yasin, et al., 1997; Qu, et al., 1997). Deletion of two disulfide bridges in protegrin-1 caused a significant decrease in antibacterial activity against both Gram-negative and Gram-positive bacteria, as well as antiviral activity (Tamamura, et al., 1995). The above results all indicate the importance of the structure maintained by the disulphide bonds for the antimicrobial activity of these peptides. A cationic cluster flanked by a hydrophobic cluster appears to be a common 11 structural feature shared by tachyplesin, protegrin and human defensin (Kini and Evans, 1989; Aumelas, et al., 1996). The above observations and future studies involving more specific changes may provide guidelines for the design of potent synthetic antimicrobial peptides. B. Loop-Structured Cationic Peptides and Bactenecin. Recently, a few cationic peptides with only a single disulphide bond forming a loop structure, have been identified (Morikawa, et al., 1992; Simmaco et al., 1993; Clark et al., 1994; Suzuki, et al., 1995; Fehlbaum, et al., 1996). According to the source from which they were derived, they can be sorted into two groups. Their sequences are shown in Table I. The first group were isolated from frog (Rand) skin, namely the brevinin family (brevinins 1, IE , 2, 2E) (Morikawa, et al., 1992; Simmaco, et al., 1993), esculentin (Simmaco, et al., 1993), the rugosin family (rugosins A , B, C) (Suzuki, 1995), and Table I: Loop-Structured Cationic Peptides Name Source Amino Acid Sequence Brevinin-1 Skin, Rana brevipodaporsa F L P V L A G I A A K V V P A L F C K I T K K C - O H Brevinin-2 Skin, Rana brevipodaporsa G L L D S L K G F A A T A G K V L Q S L L S T A S C K L A K T C - O H Brevinin-IE Skin, Rana esculenta F L P L L A G L A A N F L P K I F C K I T R K C Con't 12 Table I Loop-Structured Cationic Peptides (Con't) Brevinin-2E Skin, Rana esculenta G I M D T L K N L A K T A G K G A L Q S L L N K A S C K L S G Q C Esculentin Skin, Rana esculenta G I F S K L G R K K I K N L L I S G L K N V G K E V G M D V V R T G IDIAGCKIKGEC Ranalexin Skin, Rana catesbeiana F L G G L I K I V P A M I C A V T K K C - O H Rugosin A Skin, Rana rugosa G L L N T F K D W A I S I A K G A G K G V L T T L S C K L D K S C Rugosin B Skin, Rana rugosa S L F S L I K A G A K F L G K N L L K Q G A Q Y A A C K V S K E C Rugosin C Skin, Rana rugosa G I L D S F K Q F A K G V G K D L I K G A A Q G V L S T M S C K L A K T C Thanatin Insect, Podisus maculiventris GSKKPVPI IYCNRRTGKC Q R M B-dodecapeptide (Bactenecin) Bovine neutrophil R L C R I V V I R V C R s-dodecapeptide Sheep myeloid cells RICRIIFLRVCR Cysteine residues involved in disulphide bond formation are underlined. 13 ranalexin (Clark, 1994). Thanatin (Fehlbaum, 1996) was isolated from insects, but shows similarity to with the frog skin peptides. The lengths of these peptides ranges from 20 amino acids to 46 amino acids. Brevinins, rugosins, ranalexin and thanatin share significant amino acid sequence similarity. The sequence of esculentin is not related to any of the known cationic peptides. This group of peptides all contain a single C -terminal intramolecular disulfide bond which forms a cationic loop. Except for thanatin, the loop contains five residues at the C-terminal end. Thanatin has a loop of six residues and 3 extra amino acids at the C-terminus outside the ring. These frog skin peptides exhibit different potencies against bacteria. Most of these peptides have antimicrobial activity against both Gram-negative (E. coli) and Gram-positive (S. aureus) bacteria. Rugosin A , rugosin B and brevinin 1 are selectively active against Gram-positive bacteria. Brevinin IE was shown to have high lytic activity. Thanatin was also shown to be fungicidal, and did not exhibit hemolytic activity. Few studies of the antimicrobial mode of action have been done on these peptides, with the exception of thanatin. Brevinin-2 and esculentin have an N-terminal region which can form an amphiphilic a-helix, and was suggested to be involved in membrane permeabilization (Kreil, 1994). Ranalexin has a structure similar to polymyxin B and was proposed to function like polymyxin B (Clark, 1994). It was suggested that the cytoplasmic membrane is not the target of thanatin, and thanatin is not a pore-forming peptide (Fehlbaum, 1996). A n all-D-enantiomer of thanatin is nearly inactive against Gram-negative bacteria, but maintains activity against some Gram-positive bacteria and 14 all fungi. This suggested that more than one mechanism might be involved in the antimicrobial activity of thanatin. The second group of loop-structured peptides contains bactenecin (also called bovine dodecapeptide) from bovine neutrophils (Romeo et al., 1988) and ovine dodecapeptide from sheep myeloid cells (Bagella, et al., 1995). Both are 12 amino acids long, including four arginine residues, two cysteine residues, and 6 hydrophobic residues. Ovine dodecapeptide has four conserved hydrophobic residue-substitutions compared to bovine dodecapeptide, including one phenylalanine in the ring. These peptides are the smallest known cationic antimicrobial peptides. The cysteine residues, close to either end form a disulphide bond to make a loop. The mode of synthesis of these peptides involves prepropeptides, a precursor which identify these peptides as belonging to the cathelicidin family. Cathelicidins are a novel protein family, which have highly conserved N-terminal preproregions, and variable C-terminal antimicrobial cationic peptide regions (Zanetti, et al., 1995). They have been identified in mammalian myeloid cells, including bovine (Storici et al., 1996), porcine (Tossi et al., 1995), rabbit (Levy, et al., 1993), sheep (Bagella, et al., 1995) and human (Cowland et al., 1995) myeloid cells. Bactenecin was previously found to be active against E. coli and S. aureus, and strongly cytotoxic for rat embryonic neurons, fetal rat astrocytes and human glioblastoma (Radermacher, et al., 1993). However, little is known about its antimicrobial mechanism and whether it shares a common killing mechanism with other antimicrobial peptides, or 15 if it has a distinct mode of action due to its unique compact structure (c.f. the silk moth peptide cecropin which is a 26-amino acid amphipathic a-helix). C. Therapeutic Potential The earliest peptide antibiotics that were used extensively in human medicine (mainly topical therapy) were the gramicidins and polymyxins. The lantibiotic bacteriocin nisin is currently used as food preservative (Delves-Broughton et al., 1996). MSI-78, a 22 residue magainin analogue has recently completed human Phase III clinical trials, showing equivalent efficacy to oral ofloxacin vs. polymicrobic infections of individuals with diabetic foot ulcers (Hancock, 1997). Another promising prospect for cationic peptides would be plant and fish biotechnology in which cationic peptides are engineered by transgenic techniques into host organisms to permit enhanced bacterial disease resistance in crops and fish (Hancock and Lehrer, 1998). There are many interesting features of cationic peptides as potentially novel antibiotics. Besides broad antimicrobial activity, there are two other main antibacterial activities of these peptides, namely their ability to bind to LPS and act as an anti-endotoxin (Gough et al., 1996), and their ability to act synergistically with conventional antibiotics as enhancers. Resistance against antimicrobial peptides may not develop easily. Some bacterial strains that show resistance to cationic peptides have been identified, e.g. Burkholderia cepacia, Proteus sp. and Brucella sp. They have developed resistance intrinsically rather than by exposure to the peptides. Since these cationic peptides are gene-coded, and 16 synthesized as precursors that undergo posttranslational modifications to become active, their production by gene engineering becomes possible. While researchers continue to isolate and characterize new cationic peptides from nature, the study of analogs, as well as studies of structure:function relationships have become the focus in new drug discovery efforts. D . Aims of This Study Bactenecin was selected as the model peptide in this study. The group of loop-structured cationic peptides have not been well studied. The unique structure of bactenecin among known peptides (e.g. small size, single disulphide bond and loop structure) makes it an interesting peptide to study. The general aim of this study was to investigate the relationship between the structure and activity of bactenecin. The mode of action and antimicrobial mechanism of bactenecin was examined and several bactenecin derivatives were designed, aimed at both discovering a superactive candidate and understanding the mode of action. 17 M A T E R I A L S A N D M E T H O D S A . Bacterial Strains and Growth Conditions A l l bacterial strains used in these studies are listed in Table II. Bacterial strains for cloning purposes included E. coli D H 5 a , E. coli BL21, S. aureus K147. A l l strains were grown in Luria broth (LBNS: 10 g/L tryptone and 5 g/L yeast extract, 5 g/L NaCl), or L B N S supplemented with 10 pg/ml chloramphenicol or lOOpg/ml ampicillin. Bacterial strains for antimicrobial activity testing were grown in no salt Luria broth (10 g/L bacto-tryptone and 5 g/L bacto-yeast extract). Streptococcus strains were grown in Todd Hewitt broth (500 g/L beef heart infusion, 20 g/L bacto neopeptone, 2 g/L bacto dextrose, 2 g/L sodium chloride, 0.4 g/L disodium phosphate, 2.5 g/L sodium carbonate). Both L B and Todd Hewitt broth were purchased from Difco Laboratories, Detroit, Michigan. B. Chemicals Polymyxin B and 1-N-phenylnaphtylamine (NPN) were purchased from Sigma (St. Louis, Mo). 3,3-Dipropylthiacarbocyanine (DiS-C3-(5)) was from Molecular Probes (Eugene, Or). Dansyl polymyxin B was synthesized as described previously (Schindler andTeuber, 1975). The lipids l-pamitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) and l-palmitoyl-2-oleoyl-sn-glycero-3-phosphoglycerol (POPG) were purchased from Northern Lipids Inc. (Vancouver, B C , Canada). 18 NO 00 ON NO ON 1 - 1 rt 00 > ON ^ § I H „ COGO fl S c o o s CN CN OO 00 ON Os ON ON oo oo ON ON KS rt <D ID o f GO* 13 13 00 ON rt <o • I—I to (L) c o u cu 5s. • a •S fl" *- fl GO o s i Co l-l fl-CX fl GO fl o fl 5) ^ S3 to co ^ fl OH o GO fl GO l-l <D fl^ 00 o 1 .2 CN u Q i-i <8 fl P-t s D '-+-> a <u o oo fl 00 11 !• 00 S co 00 OO o ^ ON to ^ NO a, i4 N > 00 fl CO 00 00 fl <+H T 3 ON in s •S c s fl S ^ 2 ° "S g % -R Ci_ 00 ^ CN .3 ° S3 ^ CU —< S w ^ OO H « m NO co U U oo o fl 'oo 00 <D l-i fl. X <D fl (D fl O '3 CO oo "rt o £ M * 2 ° o o - to "S 1—1 «I s •a .2 oo fl 4) <2 o oo --j fl -fl. l-l H to 3 cu !< a a to s CJ o o CJ O o -^CN •s? CN §r z "53 S o I s -3 13 o ~ CN ^ _H t 5 2 ' CO [ f l fl* O rfl 3 s ^ m CN ON r-H CN -IB ^ &0 < co oo "rt 5T to s u o o CJ o &, CN •S ^ co U OS CD O i-i =3 O GO ~CD o fi <u t H C<H CD C o o CD O U U H < C '-*-> o CD o O O H fi » o CD "o O O U H B o U fi •c o co CD CD O ' f i • r-H u CD CD o CO c CD 2 • i—i CD "o CO o U ~CD CD co 13 o ;a u CD o i—i 13 o u »5 #g '3 >-CC 1-cu « P9 CO fi 'c3 is cc cu ^ CN S CN U ° ^ O ^ s H h3 < CU e o s o £ CU O <U cu CJ <3 -Ci CU R b a to .3 o C3 0=5 <u R O £ s cu R CJ ON CO < sr3 <u R =o 3 cj CJ O cj vo ON IS S3 < »3 s CJ CJ O cj O Si 3 o C. Recombinant Expression of Peptides C.I. D N A Manipulations General D N A techniques such as plasmid preparations from E. coli, agarose gel electrophoresis, D N A ligation, E. coli DH5a transformation and restriction enzyme digestion were performed as described in Ausubel et al.. (1987) and Sambrook et al. (1989). Slot lysis gel electrophoresis was according to Sekar (1987). D N A restriction and modifying enzymes and D N A ligase (Bethesda Research Laboratories (BRL), Burlington, M A ) were used according to the manufacturers' methods. D N A fragment isolations from agarose gels were performed using the Geneclean kit (Bio 101 Inc., L a Jolla, C A ) according to the manufacturer's protocol. D N A sequencing was done using Taq Dye Deoxy Terminator Cycle Sequencing Kit on an A B I 370A automated D N A sequencer (ABI, Foster City, C A ) . Plasmid D N A for sequencing was purified using Qiagen columns (Qiagen Inc., Chatsworth, C A ) , by following the manufacturer's protocol. C.2. Electroporation The method for electroporation of S. aureus RN4220 was based on that of Kraemer and Landolo (1990). Briefly, cells were grown in 100 ml L B N S broth to an OD 6 0 0=0.3-0.8. Harvested cells were washed twice with 500 m M sucrose and placed on ice for 15-30 min. The cells were finally resuspended in 10 ml of 500 m M sucrose. Then, 140 ng plasmid D N A were mixed with 160 pi cells in a cuvette, and chilled on ice for 1 min. Electroporation was carried out using a BioRad (Mississauga, Ont) Gene Pulser set at 200 ohm, 1.8 kV. Immediately after charging, 1 ml S M M P medium (2.8% 21 (w/v) Bacto Pennassay Broth with 275 m M sucrose, 11 m M maleic acid, 11 m M M g C l 2 and 2.5% (w/v) B S A ) was added and the mixture incubated on ice for 15 min before plating on L B N S plates with 10 pg/ml of chloramphenicol. C.3. Oligonucleotide Synthesis and Purification The amino acid sequences of indolicidin, apidaecin and bactenecin were used to design three oligonucleotide sequences according to the S. aureus codon preference (Wada et al., 1992) (Fig 1 & Fig 2B). Both coding strands and template strands for these cationic peptides were made. The two complimentary strands were annealed by mixing 1 mg/ml D N A of each strand in the annealing buffer (20 m M Tris -HCl p H 7.5, 10 m M MgC12 and 50 m M NaCl). The mixture was heated at 90 °C for 2 min, then cooled down gradually. The annealed double stranded D N A was quantitated by U V absorbance at 280 nm and used directly for ligation. Primers 1 and 2 (Fig. 2C) for D N A sequencing after cloning into pRIT5 were also made according to the upstream and downstream sequences of the multiple cloning site of pRIT5. In order to be able to clone the peptide genes into vector pSP72, a third primer (Fig. 2C) primer 3 was made to incorporate an EcoRV site. The bactenecin peptide gene fragment was obtained by P C R (94 °C lmin, 62 °C 2 min, 72 °C 1-1.5 min 20 cycle, 72 °C 10 min) from vector pRIT5 using primers 2 and 3, and subcloned into pSP72. A l l oligonucleotides were synthesized on an Applied Biosystems Inc., Model 380B D N A synthesizer and purified according to manufacturer's instruction. 22 Apidaecin Ia G N N R P V Y I P Q P R P P H P R I Indolicidin I L P W K W P W W P W R R Bactenecin R L C R I V V I R V C R Figure 1: Amino acid sequences of the three recombinantly expressed cationic peptides. The charged amino acids are in bold, the characteristic amino acids for each specific peptide are underlined. 23 A. Construction of Cationic Peptide Genes for Cloning into pRIT5 A A T T C C A T G Cationic Peptide T A A C T A A G T A A G A G C T C G G G T A C GeneSequence ATT G ATT C ATT C T C G A G C A G C T EcoRI Start Stop Stop Stop Sst I Sail B. Three Cationic Peptide Gene Sequences (Single Strand 5'-3') Apidaecin GGT A A T A A T A G A C C A GTT TAT ATT C C A C A A C C A A G A C C A C C A CAT C C A A G A TTA Indolicidin A T A TTA C C A T G G A A A TGG C C A TGG TGG C C A TGG A G A A G A Bactenecin A G A TTA TGT A G A ATT GTT GTT ATT A G A GTT TGT A G A C. Primers for Sequencing and PCR Primer 1: A C G T A A C G G CTT CAT C C A A A G Primer 2: A G C TAT G A C CAT GAT T A C GCC Primer 3: A A A G A C GAT A T C G G G G A A TTC C Figure 2: Sequences of oligonucleotides. Panel A shows the construction of the oligonucleotides synthesized for cloning into pRIT5. Panel B shows the corresponding cationic peptide gene sequences. Primers 1 and 2 in panel C were designed for sequencing. Primers 2 and 3 were used to engineer an EcoR I site into an EcoR V site so that the cationic peptide genes could also be cloned into pSP72. 24 D. Purification of the Recombinant Peptides D.l. Electrophoresis SDS-PAGE was performed as previously described (Hancock and Carey, 1979). Acid-Urea gels (AU-PAGE) were prepared according to the method of Panyim and Chalkey(1969). D.2. Purification of the Protein A/Cationic Peptide Fusion Protein Two litre LBNS cultures of S. aureus strains harboring the various plasmids, pPA-bactenecin, pPA-apidaecin and pPA-indolicidin, were grown at 37 °C to an OD600=1.5-1.8. Cells were removed by centrifugation, and the supernatant was collected, adjusted to pH 7.6 and passed over an IgG-Sepharose-6 Fast-Flow column (Pharmacia). The column was loaded, washed and fusion protein was eluted according to the manufacturer's protocol. To ensure fusion proteins were eluted correctly, samples were taken to run on SDS-PAGE. Fractions containing pure fusion proteins were pooled, lyophilized and the residue resuspended in 70 % formic acid for the CNBr digestion. D.3. Cleavage of Cationic Peptides from Protein A Cationic peptides were cleaved from protein A by CNBr which cleaved at internal methionines. The cleavage reaction was carried out in 70 % formic acid and 1M CNBr in the dark at room temperature overnight. The reaction was quenched by diluting to 5 % formic acid with distilled water. The sample was then lyophilized. The dried sample was resuspended in 1 % acetic acid for loading onto a PI00 column. AU-PAGE was run to ensure that cationic peptides were cleaved properly from the carrier. 25 D .4 . Purification of Cationic Peptide by P 1 0 0 Column Chromatography The sample after cleavage was applied to a Bio-gel polyacrylamide PI00 column (Biorad). The PI00 column was packed according to the manufacturer's protocol. Two bed-volumes of 1 % acetic acid were passed over the column by gravity feed. One-ml fractions were collected. Fractions were monitored at 280 nm, and samples taken every 5 fractions for analysis by A U - P A G E . Fractions containing peptides were pooled, lyophilized and resuspended in distilled water. D. 5. Expression of Fusion Gene in E. coli One single colony of E. coli harboring plasmid pSP72/bactenecin was picked and grown overnight at 37 °C in L B broth containing 100 pg/ml ampicillin. Fresh L B broth (5 ml) with 100 pg/ml ampicillin was inoculated with the overnight culture and grown for an additional 3-4 hours until O D 6 0 0 = 0.6-1. The culture was then induced with 1 m M IPTG. The cells were harvested 2-3 hours later after induction, and resuspended in 100 pi TE buffer (10 m M Tris-HCl, 1 m M EDTA). A 10 pi sample was removed and mixed with 2-fold concentrated SDS-PAGE sample buffer (4 % SDS, 0.2 M Tris pH 6.8, 20 % glycerol, 8 % p-mecaptoethanol). The mixture was heated at 95 °C for 5 min. A sample of 5 pi was loaded on a 12 % SDS-PAGE to check the production of fusion protein. E. Peptide Preparation E. l . Synthesis of Peptides. Peptides were synthesized by Fmoc (N-(9-fluorenyl) methoxycarbonyl) chemistry by the Nucleic Acid/Protein Service unit at the University of British Columbia using an 26 Applied Biosystems, Inc. (Foster City, CA) Model 431 peptide synthesizer. The amidated peptides were provided by Micrologix Biotech Inc., Vancouver, B.C., Canada. Gramicidin S was purchased from Sigma (St. Louis, MO), and Gram474 and Gram4112 were provided by the Protein Engineering Network of Centres of Excellence, University of Alberta, Edmonton Alberta, Canada. A l l the peptide sequences and their sources are listed in Table III and Table IV. E .2 . Refolding of bactenecin and its derivatives. The purchased bactenecin and its derivatives were in their fully reduced form. After a series of trials to determine the optimal strategy, the disulphide bond was formed by air-oxidation in 0.01 M Tris buffer, pH 7.2, at 23 °C for 24 h. The concentration of peptides was kept below 100 pg/ml in the oxidation buffer to minimize the formation of multimers. After oxidation, the solutions were frozen and lyophilized. The dried sample was resuspended in 0.3 % (v/v) TFA and centrifuged to remove any debris before loading on a column. E . 3 . Peptide Purification A reversed phase Pep RPC HR5/5 (Pharmacia; Quebec, Canada) column was used to purify the disulphide-bonded bactenecin and derivatives from the multimer by-products. The column was equilibrated with 0.3 % (v/v) aqueous trifluoroacetic acid and eluted with a slow gradient of acetonitrile (0.1 % acetonitrile per minute) in 0.3 % TFA at a flow rate of 0.7 ml/min. The absorbance was monitored at 220 nm. Fractions containing the desired product were pooled and lyophilized. Matrix-assisted laser desorption/ionization (MALDI)-mass spectrometry and A U - P A G E (Spiker, 1980) 27 Table III: Amino Acid Sequences of Bactenecin and its Derivatives Peptide Amino Acid Sequence Number of Number of Amino Acids Positive Charges at pH 7 Native Bactenecin R L C R I V V I R V C R 12 4 Linear Bac2A -CN R L A R I V V I R V A R - C O N H , 12 5 BacS -CN R L S R I V V I R V C R - C O N H " 12 5 Bac2S -CN R L S R I V V I R V S R - C O N H , " 12 5 Bac2S R L S R I V V I R V S R 12 4 Positive Charge BacR, P - C N R R C P I V V I R V C R - C O N H , 12 5 BacR, P R R C P I V V I R V C R 12 4 Bac2I-CN R I C R I V V I R C I R - C O N H , 12 5 BacP, 2 R - C N R L C P R V R I R V C R - C O N H , 12 5 Bac3K, P K K C P I V V I R V C K 12 4 Bac3R, P R R R C P I V V I R V C R R 14 6 Bac3R, P , (V) R R R L C P I V I R V C R R 14 6 Bac2R R R L C R I V V I R V C R R 14 6 Hydrophobicity BacP R L C R I V P V I R V C R 13 4 BacW R L C R I V W V I R V C R 13 4 Others BacW, 2R P x R L C R I V W V I R V C R R 15 6 Underlined residues are amino acids different from native bactenecin. 28 Table IV: Amino Acid Sequences of other Cationic Peptides Used in This Study Peptides Amino Acid Sequence Number of Number of Amino Acid Positive Charges CP 26 K W K S F I K K L T S A A K K V V T T A K P L I S S 26 7 CP 27 K W K L F K K I G I G A V L K V L T T G L P A L I S 26 5 (CEME) CP 28 K W K L F K K I G I G A V L K V L T T G L P A L K L T K 28 7 (CEMA) CP 29 K W K S F I K K L T T A V K K V L T T G L P A L I S 26 6 CP 10CN ILPWLWPWWPWRR-CONH2 13 3 CP 11CN ILKKWPWWPWRRK-CONH2 13 5 Gramicidins F P V O L F P V O L 10 2 Gram474 Y P V K L K V Y P V K L K V 14 4 Gram4112 Y P V K L K V Y P L K V K L 14 4 O: Ornithine Residues underlined are D-amino acids. were used to confirm that the disulphide bond was properly formed, and a pure product obtained. E .4. Peptide Concentration Determination Concentrations of gramicidin S and analogs were determined by the dried weight of powder. E.4.a. Amino Acid Analysis Concentrations of bactenecin and its derivatives were determined by amino acid analysis at the NAPS unit at UBC. E. 4.b. Dinitrophenylation. Concentrations of CP26, CP27, CP28, CP29, CP10CN and C P 1 1 C N were estimated by the dinitrophenylation assay, which measures the presence of free amino groups (Bader and Teuber, 1973). F . Structural Studies F . l . Circular Dichroism A Jasco (Japan) J-720 spectropolarimeter was used to measure the circular dichroism (CD) spectra of bactenecins (Falla et al., 1996). The data were collected and analyzed by Jasco software. Liposomes P O P C / P O P G (7:3) were prepared by the freeze-thaw method to produce multilamellar vesicles as described previously (Mayer et al., 1985), followed by extrusion through 0.1 mm double stacked Nuclepore filters using an extruder device (Lipex Biomembranes, Vancouver, B C Canada), resulting in unilamellar liposomes. C D measurements of peptide solutions were taken at a final concentration of 50 p M peptide in 10 m M sodium phosphate buffer, pH7.2. T F E at a final concentration 60 % (v/v), SDS (final concentration 10 mM) or liposomes (final concentration 100 u M lipid) were added and incubated at room temperature for 10 min before the C D measurement. F.2. Computer Modeling Computer modeling of peptides was done using the Insight II program (Biosystems, San Diego, C A ) . The secondary structure was modeled based on energy minimization for 4000 iterations. Three steps of minimization were performed to reach the minimal total energy state: 1) Steepest descent; 2) conjugate gradient; 3) V A 0 9 A . 30 G. Antimicrobial Activity and Hemolytic Activity G . l . Minimal Inhibitory Concentration Determination The minimal inhibitory concentrations (MICs) of peptides were determined by a two-fold microtitre broth dilution method modified from that of Steinberg et al (1997). Using the classical method (Amsterdam, 1991), higher concentrations of peptides tend to precipitate in the L B broth, thus the concentrations of peptides in the sequential wells are not accurate. Also the peptides stick to the most readily available (tissue-culture treated polystyrene) 96-well microtitre plates. Therefore the 2-fold dilution series were performed in Eppendorf tubes (polypropylene). Serial two fold dilutions of peptides ranging from 640 pg/ml to 1.25 pg/ml were made in 0.2% B S A , 0.01% acetic acid buffer. Ten pi of each concentration were added to each corresponding well of a 96-well microtitre plate (polypropylene cluster; Costar Corporation, Cambridge, M A ) . Bacteria were grown overnight and diluted 10"5 into fresh L B broth or Todd Hewitt broth for Streptococcus. One hundred pi of broth containing about 104-105 C F U / m l of tested bacteria were added to each well. Plates were incubated at 37°C overnight. The M I C was taken as the concentration at which greater than 90% of growth inhibition was observed. G.2. Human Red Blood Cells Lysis Assay Human red blood cells (RBC) were freshly collected with heparin, and centrifuged to remove the buffy coat. R B C were then washed in saline 3 times and centrifuged at 1500 xg for 5 min each. The erythrocytes were finally resuspended in saline to packed cell volume at the ratio of 25/1. Serial 2-fold dilutions of test peptides 31 were made in a microtitre plate with 100 pi saline per well. Fifty microlitre of R B C were added. Plates were covered and incubated with rocking at 37 °C. Readings of lysis or agglutination of R B C were taken after 4 and 24 h. Concentrations at which 90% lysis or agglutination was observed were taken as the hemolytic or hemagglutination concentrations. H . Membrane Permeabilization Assays H . l . Isolation of L P S E.coli UB1005 LPS was prepared according to the phenol-chloroform-petroleum ether extraction method (Gerhardt, 1994). Briefly, two six-liter cultures of E. coli UB1005 were grown to late logarithmic phase. Cells were harvested and washed sequentially with 150 ml cold deionized water once, 90 ml cold methanol once and 90 ml cold acetone twice. DNase (100 pg/ml) or RNase (25 pg/ml) were added. The pellet was then air-dried overnight. Dried cells were extracted with 50 ml P C P (mixture of 40 ml phenol, 100 ml chloroform, and 160 ml petroleum ether) twice by homogenizing at maximum speed for 30 s. Supernatants were collected and filtered through Whatman no. 1 paper, and concentrated under a stream of N 2 gas at 50 °C until the LPS precipitated. After centrifugation, the pellet was washed three times with 20 ml methanol, and then dried for 2 hr. The dried LPS was resuspended in 24 ml water and ultracentrifuged at 200,000 x g for 4 hr. The final product was resuspended in water again and lyophilized. A homogenizer was used to resuspend the pellets after each step. H.2. Dansyl-polymyxin B displacement assay The dansyl-polymyxin B displacement assay (Moore et al., 1986) was used to determine the relative binding affinity of peptides for LPS. H.3. Outer Membrane Permeability Assay The ability of peptides to permeabilize the outer membrane was determined by the N P N assay of Loh et al (1984). H.4. Inner Membrane Permeability Assay Cytoplasmic membrane permeabilization was determined by using the membrane potential sensitive carbocyanine dye DiS-C3-(5) (Sim et al, 1974). The mutant E. coli D C 2 with increased outer membrane permeability was used so that DiS-C3-(5) could readily reach the cytoplasmic membrane. Fresh L B medium was inoculated with an overnight culture, and incubated at 37 °C with shaking, and mid-logarithmic phase cells ( O D 6 0 0 n m = 0.5 - 0.6) were collected. The cells were washed with buffer (5 m M H E P E S , 5 m M glucose, p H 7.2) once, then resuspended in the same buffer to an O D 6 0 0 of 0.05. The cell suspension was incubated with 0.4 p M DiS-C3-(5) until DiS-C3-(5) uptake was maximal (as indicated by a stable reduction in fluorescence due to fluorescence quenching as the dye became concentrated in the cell by the membrane potential), and 100 m M K C I was added to equilibrate the cytoplasmic and external K + concentration. One-ml of cell culture was placed in a 1-cm cuvette and the desired concentration of tested peptide was added. The fluorescence reading was monitored by using a Perkin-Elmer model 650-10S fluorescence spectrophotometer (Perkin-Elmer Corp. Norwalk, Conn), with an excitation wavelength of 622 nm and an emission wavelength of 670 nm. 33 The maximal increases in fluorescence due to the disruption of the cytoplasmic membrane by different concentrations of cationic peptides were recorded. A blank with only cells and the dye was used to subtract the background. Control experiments (chapter 3) titrating with valinomycin and K + showed that the increase in fluorescence was directly proportional to the size of the membrane potential and that a buffer concentration of 100 mM KC1 prevented any effects of the high internal K + concentration and corresponding opposing chemical gradient. 34 RESULTS CHAPTER I: RECOMBINANT EXPRESSION OF CATIONIC PEPTIDES A. Introduction Polycationic peptides were originally obtained either by isolation from the host organism or by chemical synthesis. Either method is costly and difficult to scale up. To allow for large scale purification, two bacterial expression systems were developed successfully in Hancock's laboratory. The first system utilized a protein A fusion protein secreted by S. aureus (Piers, 1993). The second one involved a small replication protein RepA from E. coli, a synthetic cellulose binding domain and a hexa histidine domain as fusion partner, and expressed in E. coli (Zhang et al., 1998). Recombinant peptide production will allow future screening of interesting variants created by semi-random mutagenesis, which will facilitate the structure and function relationship studies. It will also allow incorporation of heavy atoms to do 2 D - N M R studies. Three cationic peptides were selected for study due to their unique amino acid sequences (e.g. rich in Trp or Pro) which suggested that they probably have unique structures, namely apidaecin (Casteels, 1989) from honeybees, bactenecin (Remeo 1988) and indolicidin (Selsted, 1992) from bovine neutrophils (Fig 2). Each has its own distinct characteristics. Apidaecin is 18 amino acids long and rich in proline. Bactenecin is 12 amino acids long and contains two cysteines which form one disulfide bond and maintain the peptide in a cyclic structure. Indolicidin is 13 amino acids long and rich in tryptophan. 35 The objectives of the work described in this chapter were to produce these three active cationic peptides in bacterial expression systems and obtain them in relatively pure and biologically active forms. B. Recombinant Expression of Three Small Cationic Peptides in S. aureus Plasmid pRIT5 is a shuttle plasmid between E. coli and S. aureus, allowing cloning procedures to be carried out in E. coli and gene expression in S. aureus. Since the coding sequences for mature apidaecin, indolicidin and bactenecin are small (36-54 nt), they can be made as oligonucleiotides. Three oligonucleotides were designed and synthesized according to the amino acid sequences of peptides; an EcoRI site and a start codon were included at the 5'-end and a Sail site and stop codons were included at the 3'-end (Fig. 2A and 2B). These oligonucleotides were ligated into the polylinker cloning sites between EcoRI and Sail on plasmid pRIT5 and transformed into E. coli D H 5 a . After clones with inserts were obtained, sequencing was done to confirm that they contained the correct sequences of peptide genes. Then plasmids pPA-apidaecin, pPA-indolicidin and pPA-bactenecin were electroporated into S. aureus. Plasmid pRIT5 contained a protein A signal sequence just upstream of the protein A gene, allowing the fusion protein to be exported to the external medium which would reduce proteolytic degradation of the fusion protein. Strains containing the plasmid of interest were grown overnight, samples of the growth supernatants were analyzed by S D S - P A G E (Fig. 3). Protein A has a molecular weight of 32 kD (Fig. 3, lane 1). The fusion protein with the cationic peptide will have higher 36 Figure 3: P roduc t ion o f protein A /ca t i on i c peptide fusion proteins i n S. aureus. S. aureus strains w h i c h harbored pRIT5 w i t h specific inserts were g rown overnight i n L B broth. Fifteen microl i t re o f culture supernatant from each strain were taken and run on S D S - P A G E . Lanes : 1, protein molecular weight markers as indicated; 2, protein A secreted by S. aureus; 3, protein A /ap idaec in fusion protein; 4, protein A/bac tenec in fusion protein; 5, protein A / i n d o l i c i d i n fusion protein. 3 7 molecular weight than protein A. Bands with higher molecular weight fusion proteins of protein A-apidaecin (lane 2), protein A-indolicidin (lane 4) and protein A-bactenecin (lane 3) were detected which showed these three fusion proteins were stably produced and secreted out of S. aureus. After the production of the desired products was confirmed, large scale purifications of fusion proteins were carried out by IgG affinity column chromatography. In order to produce active cationic peptides, they were cleaved from protein A with CNBr which cleaved at methionine residue. After the cleavage, the mixtures were run on acid-urea gels and the cationic peptides apidaecin, indolicidin and bactenecin were detected (Fig 4). Two bands of bactenecin were detected: the upper band was believed, and confirmed by mass spectrometry to be a multimer, and lower band to be intra-chain disulphide-bonded bactenecin. After the cleavage, purification of cationic peptides from CNBr-digested protein A fragments was done by running the CNBr-digested mixture through a size exclusion Bio-gel PI00 column (Fig 5A, 6A, 7A). The elution profiles showed that all three cationic peptides were eluted in the last (third peak), and further A U gel analysis (Fig. 5B, 6B, 7B) showed that the three cationic peptides were reasonably pure after the size exclusion purification. The last step was to determine whether the recombinantly produced cationic peptides were biologically active. MIC tests were done on these partially purified peptides (Table V). The results showed that all three cationic peptides had activity against Gram-negative E. coli and P. aeruginosa and no activity against Gram-positive bacteria. Both bands of bactenecin had moderate activity against the wild type Gram-38 1 2 3 4 Figure 4: Cleavage of cationic peptides from protein A. Cationic peptides were cleaved from protein A by CNBr (described in Material and Methods). After the digestion, samples were run on an A U gel. Lanes: 1, synthetic C E M E as control; 2, bactenecin (upper band is multimer, lower band is the intra-chain, disulphidedoonded product); 3, apidaecin; 4, indolicidin. The arrows indicate cationic peptide bands. 3 9 0.14 0.12 0.08 o oo C N 0.06 -0.04 A 0.02 11 16 21 •Fraction Number 26 31 Figure 5A: Elution profile of the CNBr-digested protein A/apidaecin fusion from a BiogelPlOO Column. After C N B r digestion, the sample was diluted 100 times with distilled water and lyophilized. The dried pellet was dissolved in 1 % acetic acid and applied to a Biogel PI00 column. Peptides were eluted with 1 % acetic acid. 40 Figure 5 B : Purification of apidaecin on a Biogel PI00 column. One mililitre fractions were collected. Five microlitre samples from every second fractions (Fig 5A) were run by acid-urea gel electrophoresis. Lanes: 1, C N B r digested mixture as control; 2-9, fraction 8, 10, 12, 14, 16, 18, 20, 22. The arrows indicate apidaecin bands. 41 0.35 Fraction Number Figure 6 A : Elution profile of the CNBr-digested protein A/bactenecin fusion from a Biogel PI00 Column. After CNBr digestion, the sample was diluted 100 times with distilled water and lyophilized. The dried pellet was resuspended in 1 % acetic acid and applied to a Biogel PI00 column. Peptides were eluted with 1 % acetic acid. 42 *0- i .** Figure 6B: Purification of bactenecin on a Biogel PI 00 Column. One mililitre fractions were collected. Five microlitre samples from each peak (Fig 6A) were run by acid-urea gel electrophoresis. Lanes: 1, control; 2-10, fractions 7, 11, 25, 26, 27, 28, 29, 30, 31. The arrows indicate both the upper band (multimer) and lower band of bactenecin (intra-disulphide-bridged monomer). 43 Figure 7A: Elution profile of the CNBr-digested protein A/indolicidin fusion from a Biogel PI00 Column. After CNBr digestion, the sample was diluted 100 times with distilled water and lyophilized. The dried pellet was resuspended in 1% acetic acid and applied to a Biogel PI00 column. Peptides were eluted with 1% acetic acid. 44 1 2 3 4 5 6 7 8 9 10 9m 11 12 13 14 15 16 17 18 19 20 21 Figure 7B: Purification of indolicidin on a Biogel PI00 Column. One mililitre fractions were collected. Five microlitre samples from every third fraction (Fig 7A) were run by acid-urea gel electrophoresis. Lanes: 1, control; 2-10, fraction 9, 12, 15, 18, 21, 25, 28, 31, 34; 11-20, fraction 44-54. The arrows indicate the indolicidin bands. 45 Table V : MICs of Three Partially Purified Cationic Peptides Expressed in S. aureus MIC pg/ml Apidaecin Indolicidin Bactenecin Species Strains Lower Band Upper Band E.coli UB1005 8 32 32 64 DC2 1 4 8 4 P. aeruginosa K799 64 >64 16 32 Z61 32 n/a 4 2 S. aureus K147 >64 >64 >64 >64 SAP0017 >64 >64 >64 >64 S.epidermidis >64 >64 >64 16 B. subtilis >64 >64 >64 >64 negative bacteria E. coli (32 pg/ml) and P. aeruginosa (16-32 pg/ml). Indolicidin was active against E. coli and its mutant. Apidaecin had the lowest MIC (8 pg/ml) of the three peptides against E. coli. None showed any activity against Gram-positive bacteria. C. Production of Bactenecin in E. coli While the fusion of cationic peptides to protein A proved to be an effective and convenient way to produce pure cationic peptides, there are two major drawbacks to this system: First, S. aureus is a pathogen and is not an industrial standard strain; secondly, the IgG column matrix is expensive. Another alternative would be to utilize E. coli as expression vector and a less expensive cellulose-binding domain as affinity tag. Plasmid pSP72 contained a fusion partner including Rep 21 (a truncated version of Rep A), a synthetic cellulose-binding domain and a hexa histidine domain as well as the prepro region of human defensin 1 (ELNP-1) (Zhang et al., 1998). To transfer the 46 cationic peptide gene from plasmid pRIT5 to pSP72, a primer (primer 3, Fig. 2C) was designed to engineer the EcoRI site to an E.coRV site at the 5'-end. Another primer (primer 2, Fig. 2C), downstream of the polylinker at pRIT5 was also used in the P C R amplification to obtain the fragment of cationic peptide gene from pRIT5. These fragments were digested with appropriate restriction enzymes, and cloned into pSP72. E. coli BL21 harboring the pSP72/bactenecin construct was grown in L B medium with 100 pg/ml ampicillin until the cell density reached an O D 6 0 0 of 0.4, then was induced with 1 m M IPTG and grown for an extra 3-4 hours before harvesting. The expression of bactenecin fusion protein was detected (Fig. 8). With vector alone, lanes 1 and 2 showed the exact same protein patterns with or without IPTG induction. With the bactenecin construct, lane 4 showed extra band (as expected at 15.2 K D a , arrow indicated) with IPTG induction, which wasn't present without IPTG induction (lane 3). Attempts to extract inclusion bodies were made, but insufficient fusion protein was obtained to do cleavage and M I C tests. The E. coli expression system proved to produce significantly more peptide fusion protein than the protein A system (Zhang et al., 1998). While efforts are being made to improve the bacterial expression system in our laboratory, including utilizing a high expression vector pT7-7 (Zhang et al., 1998), the focus of the thesis shifted to the antimicrobial mechanism of bactenecin. 47 1 2 3 4 5 Figure 8: Expression of bactenecin in E. coli. E. coli harboring plasmid pSP72/bactenecin was grown until OD 6oo = 0.6-l , and induced by 1 mM IPTG. After the induction, the culture was grown for additional 3-4 hr. Samples of whole cell lysates were run on SDS-PAGE. Lanes: 1, molecular weight markers; 2, E. coli pSP72 without IPTG induction; 3, E. coli pSP72 with IPTG induction; 4, E. coli pSP72/bactenecin without IPTG induction; 4, E. coli pSP72 with IPTG induction. The arrow indicates the fusion protein. 48 D . S u m m a r y Cationic peptides apidaecin, indolicidin and bactenecin were produced recombinantly and proved to be active against the Gram-negatives E. coli and P. aeruginosa. Prior to this study, only longer (26-28 amino acid) a-helical peptides had been recombinantly produced as fusion proteins, and this study demonstrated the capability of the system for expressing shorter peptides (12-18 amino acids) including one with a disulphide bridge (bactenecin) and two with unusual compositions (indolicidin with >40% tryptophan, and apidaecin with 5 proline residues). Having proven that these peptides could be recombinantly expressed, it was decided to concentrate on one of these shorter peptides bactenecin, and study its mechanisms of its action. 49 C H A P T E R T W O : A N T I B A C T E R I A L M E C H A N I S M OF B A C T E N E C I N A . Introduction Bactenecin was previously found to be active against E. coli and S. aureus (Romeo, et al., 1988), and strongly cytotoxic for rat embryonic neurons, fetal rat astrocytes and human glioblastoma (Radermacher, et a l , 1993). However, little is known about its antimicrobial mechanism and whether it shares a common killing mechanism with many other antimicrobial peptides by perturbing the bacterial membrane, or if it has a distinct mode of action due to its unique compact structure (c.f. the silk moth peptide cecropin which is a 26-amino acid amphipathic a-helix). Its small size and single disulphide bond also makes bactenecin an interesting candidate for research and drug development. The aim of this chapter was to investigate how bactenecin interacts with and kills microoganisms. The amino acid sequence of bactenecin and its linear derivative are shown in Fig. 9. The linear derivative (bac2S) with two cysteine residues replaced by two serine residues, was made to determine the importance of the disulphide bond in bactenecin's antimicrobial activity. The reduced form of bactenecin was also included in this study as a linear version of bactenecin. The identity of these peptides was confirmed by M A L D I mass spectrometry. The M A L D I data showed the molecular weight of the reduced bactenecin as 1486+1 dalton and oxidized bactenecin as 1484+1 dalton, in agreement with formation of one disulphide bond in the latter. 50 Bactenecin: R L C R I V V I R V C R Lin-Bac: R L C R I V V I R V C R B a c 2 S : RLSRIVVIRVSR Figure 9: Amino acid sequence of bactenecin and its linear variants. The cysteine residues that are involved in the disulphide bond formation are underlined. 51 B. Antimicrobial Activity The MICs of bactenecin and its derivatives against a range of bacteria was determined (Table VI) by using a modified broth dilution method. Bactenecin was active against all Gram-negative bacteria tested. It was relatively inactive (MIC=64 pg/ml) against the Gram-positive bacterium S. aureus, in contrast to a previous report (Remeo, et al., 1988). The linear variant bac2S and reduced bactenecin were inactive against wild type Gram negative bacteria. P. aeruginosa Z61 and E. coli D C 2 are outer-membrane-barrier-defective mutants, that have more permeable outer membranes than their parent strains, allowing potentially easier access of the peptides to the cytoplasmic membrane. A l l three bactenecins exhibited equivalent activity against these two mutants. Native bactenecin and bac2S showed somewhat improved activity against the defensin supersusceptible phoP/phoQ mutant of S. typhimurium, while reduced bactenecin was inactive against this mutant. Linearization by changing cysteine to serine or by reduction dramatically changed the antimicrobial activity against two Gram-positive species. The linear bactenecins were active against S. epidermidis and E. facaelis, while the native bactenecin was inactive against S. epidermidis and E. facaelis. Bac2S also showed moderate activity (16 pg/ml) against S. aureus. In order to see i f linear bactenecins are in general more Gram positive specific, six other Gram positive strains were tested. Linear bactenecins were 8x and 2-4x more active against L. monocytogenes and S. mitis respectively than was the native bactenecin. Reduced bactenecin was active against S. pneumoniae, while both bac2S and native bactenecin were relatively inactive against S. 52 O rt CQ 1 PI • i-H o PL, -a <D O S3 CD rt T 3 00 CN O ca PQ > CD p o <D PI CD ' O rt PQ o co u > — 3 CD O S3 T3 CD rt o CD c CD * o rt 6J) u 00 CN O rt PQ NO A CN hi" NO A hi" NO A NO A 00 oo NO A CN cn CN hi" NO A hi" NO A CN cn NO ho T3 CD N 'HI Q. o CD fi <D » O rt _ CD ft +-» O c CD . f i PL, I > CD 'CD rt CO fi +-» 00 CO CD 'o CD CXI OO CN © NO i - H I oo oo hi" NO A NO A CN O Q o CD (D > 00 I 00 »—H w o PI CD CL CD > CO P i CD CO O 3 CD CD, CO PI i<8 CD CD co "rt o CD ft CN u Q ON ON f- 00 in U NO o o cj N 53 , s •§> cu <3 •3 s Cd CO. 1^ ICO cs CJ cu NO CD cn CN ON in CN U O H CO S CU i< 3 I CO NO A IT) o h* NO A h o NO A NO h f NO A NO A NO A h* NO A o <D © © CD 'o co CD ft 00 cu R ^ , O R O 5 to o cu s to ON NO ON u H I* cu •2 R o £ s R C L NO in © NO CN ft l-q NO as O U H < * <u R cu CO 03 CD s CD « •a -a o H _c CD P CD T3 CD s-CD CO CD *rt > pneumoniae. C. Secondary Structure by Circular Dichroism Circular dichroism (CD) spectrometry (Fig 10A) showed that bactenecin in the reduced form and bac2S were present in 10 m M sodium phosphate buffer as unordered structures which had strong negative ellipticities near 200 nm. Native bactenecin had a spectrum resembling oxyribonuclease and nuclease that are short polypeptides, each with a disulphide bond (Venyaminov et al., 1996). The C D spectrum of bactenecin (Fig 10A), demonstrated a negative ellipticity near 205 nm, typical of that seen for a type I B-turn structure (Perczel et al., 1996). In 60 % T F E buffer, in the presence of liposomes and 10 m M SDS, the native bactenecin retained a similar structure (Fig 10B, 10C and 10D). However, in that medium, the reduced form and bac2S exhibited clearly distinct structures from those observed in aqueous solution. In 60% T F E buffer (considered a helix-inducing solvent), these two peptides tended to form a-helical structures (Fig 10B), whereas in the presence of liposomes or l O m M SDS, p-sheet structures were evident (Fig. 10C, 10D). SDS micelles resemble the environment of lipid membranes. D. Interaction with Bacterial cells 1. Interaction with outer membrane of Gram negative cells 1) The binding ofbactenecins to purified LPS from E. coli UB1005 54 4 3 2 Figure 10A: C D spectra of bactenecin, its reduced form and bac2S in phosphate buffer. The concentrations of peptides were 50 p M . C D measurements were taken in 10 m M sodium phosphate buffer (pH 7.0). Bactenecin: ; Reduced bactenecin: — ; Bac2S: -55 Wavelength (nm) Figure 10B: C D spectra of bactenecin, its reduced form and bac2S in 60% (v/v) T F E . The concentrations of peptides were 50 p M . Bactenecin: ; Reduced bactenecin: Bac2S: - - -. Wavelength (nm) Figure IOC: C D spectra of bactenecin, its reduced form and bac2S in 60% (v/v) T F E . The concentrations of peptides were 50 p M . Bactenecin: ; Reduced bactenecin: Bac2S: - -. 4 .4 ' Wavelength (nm) Figure 10D: C D spectra of bactenecin, its reduced form and bac2S in lOmM SDS. The concentrations of peptides were 50 p M . Bactenecin: ; Reduced bactenecin: — ; Bac2S: - -. 58 The M I C results indicated that the interaction of bactenecin with the outer membrane might be critical in causing the differences in antimicrobial activity against Gram negative bacteria among the three bactenecin forms. The first step of cationic peptide antimicrobial action involves the binding of the cationic peptide to the negatively charged surface of the target cells (Hancock et al., 1995). In Gram negative bacteria, this initial interaction occurs between the cationic peptides and the negatively charged LPS in the outer membrane (Falla, 1996; Sawyer, 1988; Piers, 1994). Such binding can be quantified using the dansyl polymyxin B displacement assay. Dansyl polymyxin B is a fluorescently tagged cationic lipopeptide, which is non-fluorescent in free solution, but fluoresces strongly when it binds to LPS. When the peptides bind to LPS , they displace dansyl polymyxin B resulting in decreased fluorescence, which can be assessed as a function of peptide concentration (Fig. 11). Bactenecin was a relatively weak LPS-binder compared to polymyxin B , but it was still better than M g 2 + , the native divalent cation associated with LPS. Most importantly, it seemed that native bactenecin bound to LPS far better than its linear derivative and reduced form, which may partially explain the difference in their activities against Gram negative bacteria. 2) Effect on outer membrane permeability Antimicrobial peptides bind to LPS , displacing bound divalent cations. Due to their bulky nature they disrupt the outer membrane and promote their own uptake across the outer membrane (Sawyer, 1988; Piers, 1994). In order to determine whether better binding ability resulted in better outer membrane permeabilization, an N P N assay was 59 Figure 11: Binding of peptides to LPS as assessed by their ability to displace dansyl polymyxin B from E. coli UB1005 LPS. Dansylpolymyxin B was added to 1ml of 3 pg/ml LPS to a final concentration of l p M which saturated the binding sites on LPS, and the fluorescence sensitivity was adjusted to 90%. The peptides and M g 2 + were titrated in resulting in a decrease in fluorescence due to dansyl polymyxin displacement until no decrease or very small decrease was detected. Polymyxin B: - X - ; MgCl 2 : Bactenecin: - A - ; Bac2S ^ B - ; Bactenecin (reduced): 60 100 0.1 1 10 100 Concentration (ug/ml) Figure 12: Peptide-induced outer membrane permeabilization measured by the N P N assay uptake in E. coli UB1005. Mid-log phase E . coli cells were collected and incubated with N P N in the presence of various concentrations of bactenecin (oxidized), bac2S, or bactenecin (reduced). N P N was taken up into cells when the outer membrane was disrupted by the peptides. The uptake of N P N was measured by an increase in fluorescence. Polymyxin B: Bactenecin: -+-; Bac2S: -o-; Bactenecin (Reduced): - V -61 performed. N P N is a neutral hydrophobic probe that is excluded by an intact outer membrane, but is taken up into the membrane interior of an outer membrane that is disrupted by antimicrobial peptide action. N P N fluoresces weakly in free solution but strongly when it enters the membrane. Fig. 12 shows that polymyxin B permeabilized the outer membrane to 50% of maximal increase in fluorescence arbitrary units at 0.4 pg/ml, while bactenecin, bac2S and linear bactenecin caused half maximal permeabilization at 0.8 pg/ml, 2 pg/ml and 4.5 pg/ml respectively. Native bactenecin thus appeared to be better than the linearized derivative and its reduced form at permeabilizing the outer membrane of E. coli UB1005. E. Summary Bactenecin was a type I P-turn molecule either in an aqueous or in a membrane-like environment. It was active against the Gram-negative bacteria E. coli, P. aeruginosa, and S. typhimurium, but was relatively inactive, or less active, against Gram-positive bacteria compared to its linear and reduced derivatives. Bactenecin bound to negatively charged LPS from the E. coli outer membrane and permeabilized the outer membrane to allow the small hydrophilic N P N molecules into cells. However bactenecin did not appear to be a strong outer membrane permeabilizer compared to polymyxin B. Bactenecin appeared to bind to LPS and permeabilize the outer membrane better than its linear and reduced derivatives, which explained its better antimicrobial activity against Gram-negative bacteria. The disulphide bond seemed to be critical for bactenecin to interact with the outer membrane of Gram-negative bacteria. 62 Although bactenecin interacted with the outer membrane of bacteria by the same mechanism as typical peptides (e.g. defensin, cecropin), how it interacted with and eventually killed bacteria is still unclear. Therefore, Chapter Three focuses on developing a cytoplasmic membrane permeability assay to study the interaction with the cytoplasmic membrane. 63 C H A P T E R T H R E E : INNER M E M B R A N E P E R M E A B I L I T Y A . Introduction It has been proposed that the antibacterial target of cationic peptides is at the cytoplasmic membrane (Duclohier et al., 1989). Cationic peptides are generally able to interact electrostatically with the negatively charged headgroups of bacterial phospholipids and then insert into the cytoplasmic membrane, forming channels or pores which are proposed to lead to the leakage of cell contents and cell death (Christensen et al., 1988; Lehrer, 1989; Wimley, et al., 1994). However there are very few data for peptides pertaining to measurement of the disruption of the cytoplasmic membrane permeability barrier, despite ample evidence that membrane disruption can occur in model membrane systems (Silvestro et al., 1997). Although some authors have utilized measurements of the accessibility of a normally-membrane-impermeable substrate to cytoplasmic B-galactosidase, such assays suffer from using a bulky substrate (ortho nitrophenyl galactoside) (Skerlavaj, 1990; Lehrer, 1989), as well as an inability to dissociate inner membrane from outer membrane permeabilization (Lehrer, 1989). To circumvent this I have adopted an assay involving the membrane potential sensitive dye diSC3-(5) to measure the disruption of the electrical potential gradients across the cytoplasmic membrane in intact bacteria. Bacteria, through either electron transport or A T P hydrolysis, eject protons and maintain an energized membrane or proton motive force comprising an electrical potential gradient (A\\i, oriented internal negative) and a p H gradient (ApH, oriented internal alkaline) (Harold, 1972). Since the cytoplasmic p H of E. 64 coli is maintained at a p H 7.4, at neutral p H in the external medium, by far the major contribution to the protonmotive force is provided by the membrane potential gradient of about -150 m V (Bakker, 1978). The homeostasis between A p H and A\\i is maintained by potassium flux utilizing the large potassium pool inside cells (approximate concentration 100-150 m M K + ) . Permeabilization of the cytoplasmic membrane would allow ions to equilibrate across the cytoplasmic membrane to reduce or destroy the membrane potential gradient. The use of E. coli mutant D C 2 allowed me to perform this assay in the absence of E D T A [required by previous workers who have used similar assays in E. coli (Letellier and Shechter, 1979; Ghazi et al., 1981)]. E. coli D C 2 is an outer membrane defective mutant, which has an outer membrane that is more permeable to hydrophobic, small hydrophilic and polycationic peptide molecules (Richmond, et al., 1976). The fluorescent probe diSC3-(5), which is a caged cation (a positively charged quaternary ammonium buried within a hydrophobic matrix), distributes between cells and the medium depending on the cytoplasmic membrane potential gradient. Once it is inside the cells, it becomes concentrated and self-quenches its own fluorescence (Sim, et al., 1974). The quenching is probably due to the formation of dye aggregates inside cells with reduced fluorescence (Sim, et al., 1974). If peptides form channels or otherwise disrupt the membrane, the membrane potential will be dissipated, and diSC3-(5) will be released into the medium causing the fluorescence to increase, as can be detected by fluorescence spectrometry. In these assays, 0.1M K C I (final concentration) was added to the buffer to balance the chemical potential of K + inside and outside the cells. 65 B. Fluorescence Quenching of diSC3-(5) by E. coli DC2 cells Fig. 13 demonstrates the fluorescence changes of the diSC3-(5) upon addition to mid-log phase E. coli D C 2 cells (OD 6 0 0=0.05) at room temperature. The fluorescence decreased gradually until it stabilized to a steady level after an hour. The fluorescence quenching was induced in E. coli cells with an intact membrane potential gradient (inside negative). The addition of 0.1M KC1 without the presence of valinomycin after the stabilization had no effect on the fluorescence intensity. C. The relationship of K + concentration outside the cells and the fluorescence intensity in the presence of valinomycin Changing the concentration of potassium outside the cells will change the potassium concentration gradient accordingly. The potassium concentration gradient opposes the electrical gradient and thus maintains the membrane potential gradient A\p across the membrane (Letellier and Shechter, 1979). Valinomycin, an ionophore renders the cytoplasmic membrane permeable to potassium (Harold, 1970). Therefore varying the K + concentration in the presence of valinomycin outside cells will change the equilibrium Avj/ accordingly. This was indicated by plotting the fluorescence quenching of diSC3-(5) as a function of K + concentration outside the cells (Fig. 14). As the concentration of K + outside increased, the fluorescence intensity increased proportionally (ie. fluorescence quenching decreased as diSC3-5 left the cells) indicating a decrease of membrane potential. A linear relationship was observed over the range of 10 m M to 100 m M 66 0 10 20 30 40 50 60 Time (minutes) Figure 13: Fluorescence quenching of diSC3-(5) by log-phase E. coli cells. DiSC3-(5) at 0.4 p M was added to log-phase E. coli cells ( O D 6 0 0 = 0.05) resuspended in 5 m M H E P E S and 5 m M glucose (pH 7.2), and incubated at room temperature. 67 100 -90 -External K C I concentration (mM) Figure 14: Effect of external KCI concentration on the fluorescence intensity of diSC3-(5) incubated with E. coli DC2 cells and valinomycin. Log-phase E. coli cells (OD6oo = 0.05) were incubated with 0.4 p M diSC3-(5) until maximal fluorescence quenching was observed and then 1 p M valinomycin was added. Various concentrations of KCI were added to 1 ml aliquots of the mixture to alter the K + gradient and were incubated at room temperature for 10 min before fluorescence readings were taken. 68 external potassium concentrations. At 0.1M KC1, the increase in fluorescent intensity reached its maximum, due to equilibration of the outside and inside K + concentrations. This result proved that the partitioning of diSC3-(5) in the suspension medium and inside cells was membrane potential - dependent and was directly proportional to the size of the membrane potential. To prevent any effects due to the movement of potassium (which could oppose the dissipation of the electrical potential gradient), an external concentration of 0.1 M KC1 was used in the subsequent cytoplasmic membrane permeability assay. As a control, it was shown that 1 p M valinomycin alone did not cause a change in fluorescence intensity, while subsequent addition of 0.1 M KC1 resulted in a very rapid immediate release of diSC3-(5) (Fig. 15). Cationic peptides have been hypothesized to form pores or channels in the membrane and cause the cell to leak out their cytoplasmic contents. When such pores or channels are formed, membrane potential will be dissipated and diSC3-(5) would be redistributed accordingly, causing the fluorescence to increase. The inner membrane permeability assay was developed based on this assumption. Gramicidin S (16 pg/ml) increased the fluorescence intensity by about 30 arbitrary units, while subsequent addition of 0.1 M KC1 resulted in another 30 unit increase (Fig 16). Presumably the first increase in fluorescence was due to redistribution of both protons and K + to establish the membrane potential at a new (lower) level, whereas in the presence of 0.1 M K + which equilibrated the transmembrane K + concentration gradient, the membrane potential was completely dissipated. 69 Figure 15: Kinetics of fluorescence intensity changes of diSC3-(5) incubated with E. coli DC2 cells in the presence of valinomycin and KCI. Log-phase E. coli cells were incubated with 0.4 p M diSC3-(5) until maximal fluorescence quenching was obtained. Valinomycin l p M and 0.1M KCI were added to a 1 ml aliquot when indicated by arrows. 70 Figure 16: The effect of 0.1 M KC1 on the dissipation of cytoplasmic membrane potential caused by gramicidin S. Gramicidin S (16 pg/ml) was added to log-phase E. coli cells that had been preincubated with 0.4 p M diSC3-(5), and an increase of fluorescence intensity was observed. When the fluorescence increase was stable, 0.1M KC1 was added which resulted in an additional fluorescence increase. 71 D. Antimicrobial activities in the presence of potassium chloride Since 0.1M KCI was added to the assay buffer to equilibrate the cytoplasmic and external K + concentration, the effect of 0.1 M K C I on the antimicrobial activity was tested. Table VII shows the minimum inhibitory concentrations (MIC) of a variety of cationic peptides in the presence and absence of K + . It seemed that the activity of the cc-helical cationic peptides was not affected by the presence of K + . Among the (3-structured peptides tested, the analog Gram 474 was not affected, the MICs of gramicidin S and Gram 4112 were increased by 4 and 8 fold respectively. The MICs of the extended structured peptides CP10CN and its variant CP11CN were increased by 2 to 4 fold. The M I C of polymyxin B did not change, but MICs of bactenecin, its reduced form and bac2S were increased by 4 to 8 fold. E. The interaction of a-helical cationic peptides with the cytoplasmic membrane CP26, CP27, CP28 and CP29 belong to the group of amphipathic a-helical peptides (Table IV). CP27 ( C E M E , MBI-27) is a hybrid of cecropin (residues 1-8) and mellitin (residues 1-18) (Wade, et al., 1992). CP28 ( C E M A , MBI-28) is a derivative of CP27 with two extra positive charges at the C-terminal end (Piers, 1994). CP26 and CP29 were also CP27 derivatives designed in our laboratory by Dr. Nedra Karunaratne to have improved a-helicity. Although these four peptides had only 2-4 fold differences in MICs, their abilities to dissipate the membrane potential were quite different (Fig 17). CP26 had the lowest 72 Table VII: MICs of E.coli DC2 in the Presence and Absence of K + M I C tig/ml Peptide K+=0 K+=0.1M a-helical CP26 0.25 0.5 CP27 2 2 CP28 0.5 1 CP29 1 1 ^-structured Gramicidin S 2 8 Gramicidin 474 32 32 Gramicidin 4112 1 8 extended structured CP 10CN 4 16 CP 11CN 2 4 Polymyxin B 0.0156 0.0156 loop-structured Bac 2S 2 16 Bactenecin (reduced) 2 16 Bactenecin 2 8 100 80 A -20 J Time (minutes) Figure 17: Kinetics of fluorescence intensity changes of diSC3-(5) incubated with E. coli DC2 in the presence of a-helical cationic peptides. a-Helical peptides at their MICs were added seperately to respective log-phase E. coli DC2 cells preincubated with 0.4 p M diSC3-(5) in the presence of 0.1 M KC1. The fluorescence changes with time were recorded. CP26: 0.5 pg/ml, CP27: 2 pg/ml, -• - ; CP28: 1 pg/ml, - A - ; CP29: 1 pg/ml, - X - . 74 M I C , yet it failed to dissipate the membrane potential at its M I C . It started to disrupt the membrane potential at 1 pg/ml, two fold higher than its M I C and achieved maximal membrane potential disruption at 8 pg/ml (Fig 18). CP27, CP28 and CP29 had similar MICs and they showed a similar pattern (Fig 17 and 18) in completely dissipating the membrane potential at concentrations within two-fold of their MICs. These fluorescence increases caused by CP27, 28, 29 stabilized within 2 min. CP27 had the highest M I C among these peptides, but it seemed to have the greatest ability to dissipate the membrane potential. F. The interaction of P-structured cationic peptides with the cytoplasmic membrane Gramicidin S is a naturally occurring cyclic peptide antibiotic first isolated from Bacillus brevis (Gause and Brazhnikova, 1944). It has been shown to possess a two-stranded P-sheet structure connected by P-turns (Izumiya, et al., 1979; Rackovsky and Scheraga, 1980). Gram 474 and Gram 4112 are two analogs of gramicidin S synthesized by Dr. Kondajewski and Dr. Hodges in collaboration with Dr. Hancock (Table IV). Gram S and Gram 4112 had similar antimicrobial activity against E. coli, while Gram 474 was 4 times less active (Table VII). At their M I C , all three gramicidin peptides caused rapid membrane depolarization (Fig. 19), although Gram 4112 was not as effective in dissipating membrane potential as the other two. Both gramicidin S and Gram 474 started to cause membrane permeabilization at very low (sub-MIC) concentrations and achieved maximal dissipation of membrane potential at concentrations well below their MICs, at 2 pg/ml 75 90 -Figure 18: Effect of a-helical cationic peptides on the fluorescence intensity changes of diSC3-(5) incubated with E. coli D C 2 cells. Various concentrations of cationic peptides were added to 1 ml log-phase E. coli DC2 cells that had been pre-incubated with 0.4 p M diSC3-(5) in the presence of 0.1 M KC1. Readings were taken when the maximal fluorescence increase was reached typically within 5 min. CP26: CP27: CP28: - A - ; CP29: - X - . Arrows indicated M I C values. 76 Time (minutes) Figure 19: Kinetics of the fluorescence intensity changes of diSC3-(5) incubated with E. coli D C 2 in the presence of p-structured cationic peptides. P-Structured peptides at their MICs were added to log-phase E. coli D C 2 cells that had been pre-incubated with 0.4 p M diSC3-(5) in the presence of 0.1 M KCI . The fluorescence changes with time were recorded. Gramicidin S: 8 pg/ml, G474: 32 p g / m l , G 4 1 1 2 : 8 pg/ml, - A - . 7 7 (Fig. 20). Gram 4112 on the other hand, did not cause maximum membrane permeabilization at its M I C , and achieved this only at >40 pg/ml (ie. >5x MIC). G. The interaction of extended structured cationic peptides with the cytoplasmic membrane Peptide CP10CN, with its carboxyl terminus amidated, is identical to the 13-amino acid antimicrobial peptide indolicidin, present in the cytoplasmic granules of bovine neutrophils (Selsted, et al., 1992). C P 1 1 C N is a derivative of C P 1 0 C N with Pro3 and Trp4 substituted with Lys3, and an additional Arg residue at the C-terminus producing a molecule with a greater positive charge (Falla and Hancock, 1997). At their MICs, both peptides increased membrane permeability rapidly (Fig. 21). However the increase was only 30-35 arbitrary units (cf. CP27). These peptides failed to cause full dissipation of cytoplasmic membrane potential even at concentrations 4 fold the M I C (Fig. 22) and quite different concentration dependences were observed. H. The interaction of looped structured cationic peptides with the cytoplasmic membrane The MICs of bactenecin, reduced bactenecin and bac2S in the presence of 0.1 M KC1 were determined and shown to be 8-16 pg/ml (Table VII). Despite these similar MICs for the 3 peptides against E. coli DC2, the influence of different concentrations of these peptides on the membrane potential was quite different (Fig. 23 and 24). The native bactenecin started to cause the release of probe at half-MIC concentrations, whereas its 78 100 90 H 0 H , , 1 0.1 1 10 100 Peptide concentration (mg/ml) Figure 20: Effect of p-structured cationic peptides on fluorescence intensity of diSC3-(5) incubated with E. coli DC2 cells. Various concentrations of cationic peptides were added to 1 ml log-phased E. coli DC2 cells incubated with 0.4 p M diSC3-(5) in the presence of 0.1 M K C 1 . Readings were taken when the maximal fluorescence increase was reached. Gramicidin S: ^ ; Gram474: Gram 4112: -—-. Arrows indicated the MICs. 79 100 Time (minutes) Figure 21: Kinetics of fluorescence intensity changes of diSC3-(5) incubated with E. coli D C 2 in the presence of extended cationic peptides. Extended peptides at their MICs were added to log-phase E. coli D C 2 cells incubated with 0.4 p M diSC3-(5) in the presence of 0.1 M KC1. The fluorescence changes with time were recorded. CP10CN: 8 pg/ml, CP11CN: 4 pg/ml, CP27: 2 pg/ml, - A - . 80 Peptide concentration (mg/ml) Figure 22: Effect of extended cationic peptides on the fluorescence intensity of diSC3-(5) incubated with E. coli DC2 cells. Various concentrations of cationic peptides were added to 1 ml log-phase E. coli DC2 cells that had been pre-incubated with 0.4 p M diSC3-(5). Readings were taken when the maximal fluorescence increase was reached. CP10CN: CP11CN: Arrows indicate the M I C . 81 Time (minutes) Figure 23: Kinetics of fluorescence intensity changes of diSC3-(5) incubated with E. coli D C 2 in the presence of loop-structured cationic peptides. Loop-structured peptides at their MICs were added to log-phase E. coli DC2 cells that had been pre-incubated with 0.4 p M diSC3-(5) in the presence of 0.1 M KC1. The fluorescence changes with time were recorded. Bactenecin: 8 pg/ml, Bactenecin (reduced): 16 pg/ml, Bac 2S: 16 pg/ml, - A - . 82 Figure 24: Effect of loop-structured cationic peptides on the fluorescence intensity of diSC3-(5) incubated with E. coli D C 2 cells. Various concentrations of cationic peptides were added to 1 ml log-phase E. coli DC2 cells that had been pre-incubated with 0.4 u M diSC3-(5) in the presence of 0.1 M KCI. Readings were taken when the maximal fluorescence increase was achieved. Bactenecin: -—-; Bactenecin (reduced): Bac 2S: Arrows indicate the MICs. 83 linear derivative bac2S and reduced bactenecin showed maximal probe release (dissipation of membrane potential) at concentrations well below the M I C concentration (Fig. 24). Bactenecin caused the permeabilization gradually and very slowly, and even at 64 pg/ml it reached maximum permeabilization at 90 minutes, while reduced bactenecin reached its maximal permeabilizing effect within 4 min (Fig. 23). Bac 2S behaved somewhat like reduced bactenecin. These results agreed with the kinetics observed in a killing assay. G. Summary A n assay was developed to study the interaction of cationic peptides with the cytoplasmic membrane and the importance of transmembrane potential. It seemed that different peptides behaved quite differently to depolarize the membrane potential. There was no consistent correlation between the ability to permeabilize the cytoplasmic membrane and the antimicrobial activity. 84 CHAPTER FOUR: STRUCTURE AND FUNCTION RELATIONSHIPS A. Introduction Cationic peptides share common features of being highly cationic and forming amphipathic structures. Their cationic feature facilitates their interaction with the negatively charged cell surface and their amphipathic structures enable them to be incorporated into the hydrophobic membrane interior to form channels and pores which have been proposed to ultimately cause cell death. However, bactenecin was showed to function in fundamentally different way to its linear variants and other known cationic peptides, like defensin and cecropin, due to its unique structure. Its basicity still allowed it to interact with the negatively charged outer membrane and be translocated across the outer membrane by the self-promoted uptake pathway, the same mechanism as proposed for a-helical and P-structure peptides. However, it presumably used a different mechanism to interact with and translocate across the cytoplasmic membrane, and it may attack an unknown cellular target leading to cell death. The linear variants of bactenecin created either by reduction (Lin-Bac) or serine substitution (Bac2S) lost the ability to interact with the negatively charged outer membrane due to the disruption of the disulphide bond, but were able to interact with the cytoplasmic membrane, possibly by forming channels and pores (although multimers or aggregates have to be involved). The different secondary structure that they adopted when came in contact with the membrane-like environment (e.g. liposome) may account for the difference in the mechanism. 85 The structural features (cationicity, disulphide bond, amphipathicity) of bactenecin and its linear variants seemed to play a key role in the antimicrobial mechanism. Therefore, work in this chapter was designed to study structure:function relationship. Hydrophobicity, positive charges, disulphide-bridging and amphipathicity are also recognized as important factors in the antimicrobial activities of cationic peptides. Analogues were designed to investigate the.effects of modification of these factors on antimicrobial activities. These analogues can be divided into three groups, linearized, more positive in charged, and increased hydrophobicity of the ring. B. Linearization In Chapter I, two linear derivatives of bactenecin, bac2S and reduced bactenecin were described. They were found to have higher selectivity against Gram-positive bacteria, and no activity against wild type Gram-negative bacteria. To further confirm this observation, two more linear derivatives of bactenecin were made, bac2A-CN (with two cys to ala replacements) and bacS-CN (with a single cys-3 to ser-3 replacement). The hydroxyl groups in serine residues and the sulphydral groups in cysteine residues are capable of hydrogen bonding to water and thus are hydrophilic, and would tend to make linear bactenecin more hydrophilic than cyclic bactenecin since the S H groups in native bactenecin are joined to form a disulphide bridge. To maintain the hydrophobicity of linear and native bactenecin as similarly as possible, bac2A-CN had alanine substitutions at both cysteine positions, since alanines are hydrophobic residues. Both bac2A-CN and bacS-CN were amidated at the carboxyl terminal end. Both amidated and non-amidated versions of bac2S were also included in this study. 86 A l l three peptides were found to be more active against Gram-positive bacteria compared to native bactenecin (Table VIII). While native bactenecin was inactive against S. aureus, S. epidermidis, E. facaelis, and S. pneumoniae (MIC > 64 pg/ml), the linear amidated peptides showed dramatically improved antimicrobial activities. For example, B a c 2 A - C N had an M I C of 4 pg/ml against S. aureus, 1 pg/ml against S. epidermidis and 2 pg/ml against E. facaelis. For bac2A-CN and bac2S-CN the activity against L. monocytogenes was increased by a factor of 32, against S. pyogenes and S. mitis by a factor of 8, and against C. xerosis by a factor of 4. B a c 2 A - C N and bac2S-CN had the same antimicrobial pattern, with only a 2 fold difference in their activities against E. facaelis and S. mitis, the difference that is considered not to be a significant difference for M I C assays (i.e. within the variability of the assay). These data indicated that introduction of two O H groups made no difference. On the other hand, bacS-CN with a single Cys 3 to Ser 3 alteration showed slightly poorer activities (2-4 fold), with exception of S. pneumoniae (only 2-fold lower MICs) compared to bac2S-CN. The differences in antimicrobial activities between bac2S and reduced bacetenecin were not as consistent. Compared to bac2S, reduced bactenecin (containing two S H groups) was less active against S. aureus, but was more active against S. pneumoniae, a trend also seen with bacS-CN (with one S H group). In general, amidation of bac2S improved the antimicrobial activities against both Gram positive and Gram-negative bacteria by a factor of 2-8. The antimicrobial activities of the reduced forms of other bactenecin derivatives were also determined (Table IX). Again, the tendency to be Gram-positive selective was 87 observed, yet not as obviously. Most of the linear peptides had decreased or equivalent activities compared with reduced bactenecin, and only 2-4 fold improvements in activities compared to native bactenecin. Bac2R, which has extra positively charged (arginine) Table VIII: MICs of Linear, Amidated Bactenecin Derivatives M I C (pg/ml) Bacteria Species2 B a c 2 A - C N BacS-CN Bac2S-CN Bac 1' 3 Bac 2S 1 B a c - R 1 3 Gram-Negative E. coli 4 4 2 8 32 >64 P. aeruginosa 8 16 16 4 >64 >64 S.typhimurium 32 >64 32 8-16 >64 >64 Gram-Positive S. aureus 4 16 4 64 16 >64 S. epidermidis 1 2 1 >64 8 8 E. facaelis 2 16 4 >64 8 8 L. monocytogenes 0.25 0.5 0.25 8 1 1 C. xerosis 0.25 0.5 0.25 1 1 0.5 B. subtilis 32 >64 16 >64 64 >64 S. pyogenes 2 8 2 >64 >64 4 S. mitis 0.25 0.5 0.125 16 16 16 S. pneumoniae 16 8 16 2 0.5 1 * 1. MICs were also shown in Table VI. Peptide sequences (also see Table III): Bactenecin R L C R I V V I R V C R B a c 2 A - C N R L A R I V V I R V A R - C O N H 2 B a c l S - C N R L S R I V V I R V C R - C O N H 2 Bac2S-CN R L S R I V V I R V S R - C O N H , Bac2S R L S R I V V I R V S R 2. Species and strains are described in Tablell. 3. Bac= Bactenecin Bac-R=Reduced Bactenecin 88 ON OO CU 3 I* o 03 \P4 CN O ICQ CJ IPQ NO A h* No A NO A ho NO A NO A NO A CN ^1-NO A NO A NO NO NO A NO A NO A hi" NO A oo CN cn NO oo NO A CN cn CN cn CO © CN CN CN h f NO A NO A NO A NO A h o CN cn NO CN cn CN CN NO CN cn NO A > CN m CN PH O O O CJ o3 oj 03 o3 CQ CQ CQ CQ OO IW rt CN O 03 CQ ho CN cn NO A un o un CN NO A vo NO A o fe • f l CD O •g CD rt CD cn CD > > "C CD Q fl • i-H o CD c CD pq Cw O 00 o X rt cn o o3 |pq rt cn o 03 pq cn o o3 |PQ_ 00 fl "o3 t-H loo rf l 00 CD •»—i O CD , ^ 00 CN cn CN cn NO A •3-NO A NO A So! s i l o cj NO A NO A NO A NO A NO A 0 0 CN cn un o C3 oo O R I ca »R I* § ll NO NO A NO NO A CN un o NO A hf NO A NO A oo cj s oo cj C3 cj cj R CJ £1 O R O oo s cu NO cn o un CN o NO A h f NO A NO A s «0 cu R Oo" oo" cu a R O £ s cu §,1 oo" CQ PQ CQ' residues at both the N- and C-terminus, seemed to have better (2-4 fold) activity than all the rest. All derivatives were relatively inactive against the wild type Gram-negative bacteria, P. aeruginosa and S. typhimurium, as observed for reduced bactenecin. For E. coli, the derivatives with more positive charges (bac3K,P, bac3K,P,(V), and bac2R) were found to have equivalent antimicrobial activities to native cyclic bactenecin, even though they were linear. BacP, bacW, and bac2R,W were inactive against all three Gram negative bacteria. It seemed that an increase in positive charges, either by amidation or due to the presence of arginine residues, improved the antimicrobial activities for the linear bactenecin derivatives. The effect of positive charges was thus examined further. C. Positive Charges Eight derivatives of bactenecin were designed to investigate the importance of size, type and topology of positive charges in antimicrobial activity. Six of these peptides had more positive charges than did native bactenecin. The amino acid sequences and numbers of positive charge are listed in Table III and reproduced under each table. Native bactenecin is a type I P-turn loop-structure, with two arginine residues at positions 4 and 9 adjacent to the disulphide bond (Fig. 25). In order to facilitate the formation of the disulphide bond, the arginine at position-4 was substituted with a turn-promoting proline residue. In order to maintain the same number of positive charges, leucine at position 2 was substituted by a arginine residue to form bacR,P. BacR,P had positive charges concentrated at the N- and C-termini and a hydrophobic loop (Fig. 26), 90 Figure 25: Computer generated model of bactenecin. This secondary structure model was developed using the program Insight II (Biosystems, San Diego, California) on a Silicon Graphics Indy workstation. The secondary structure was modeled based on energy minimization for 4000 iterations. Three steps of minimization were performed to reach the minimal total energy state: 1) Steepest descent; 2) conjugate gradient; 3) VA09A. Argiriine residues are in red; the cysteine residues are in yellow. Isoleucine and leucine residues are in blue, and valine residues in green; OH group at the C-tenriinus is in white. 91 Figure 26: Computer generated model of bacR,P. This secondary structure model was developed using the program Insight II (Biosystems, San Diego, California) on a Silicon Graphics Indy workstation. The secondary structure was modeled based on energy mmimization for 4000 iterations. Three steps of rninirnization were performed to reach the minimal total energy state: 1) Steepest descent; 2) conjugate gradient; 3) V A 0 9 A . Arginine residues are in red; the cysteine residues are in yellow. Isoleucine and leucine residues are in blue, and valine residues in green; O H group at the C-terminus is in white; proline residue is in blue. 92 and when modeled seemed to be a good amphipathic molecule. Bac3R,P was a derivative of bacR,P with arginine residues added at both the N - and C termini, giving bac3R,P two extra positive charges. Bac3R,P,(V) was made in error during amino acid synthesis with one valine residue in middle of the loop missing. A l l positive charges of native bactenecin are contributed by arginine residues, and not a single lysine residue is present. Bac3K,P was a derivative of bacR,P with three arginine residues replaced by three lysine residues, designed to test the effect of the different basic amino acid side chain. Bac2R was a derivative of bactenecin with two more arginine residues. Bac2I-C N and BacP,2R-CN also had one to two extra positive charges compared to the native bactenecin. Among all bactenecin derivatives, Bac2R seemed to have the best antimicrobial activities against Gram-positive bacteria tested (Table X). Compared to native bactenecin Bac2R had >8 fold improved activities against S. epidermidis, C. xerosis, and L. monocytogenes, and 2-4 fold improved activities against E. facaelis, S. pyogenes and S. mitis. Bac3R,P and bac3R,P(V) were the second best, with only about a two fold increase in MIC over bac2R. The absence of the valine residue in the middle of the loop in bac3R,P(V) seemed to make very little difference. A l l three of the above improved peptides had two extra positive charges outside the loop. 93 CO cu > > •c CD Q _c '3 CD G CD -^ -» O cd PQ CD 60 o I CD > o PH CM O co u cu H PQ oo 8-16 NO >64 >64 00 >64 >64 NO r - H CN Bac2R CN NO oo CN ro <0.125 >64 oo uo © >64 -CN PH CN o cd PQ NO i — ( CN ro >32 NO i — f >32 N/A N/A N/A N/A N/A N/A Bac2I-CN Bac2I-CN T * NO 00 CN ro oo >32 N/A N/A N/A N/A N/A N/A Bac2I-CN 00 i O Bac3R, P, (V) CN oo 00 >64 VO >64 >64 >64 PH Bac3R, CN 00 00 >64 NO r-H CN CO •<*• CO o >64 00 <—i >64 BacR, P >32 NO CN ro >32 >32 NO 00 >64 >64 NO >64 •CN BacR, P-CN NO NO T—H 32-6^ CN ro >64 CN >64 NO CN >64 PH Bac3K, NO >64 CN ro >64 >64 >64 NO oo >64 >64 CN >64 Bac3K, Bacteria Species Gram-Negative E. coli P. aeruginosa S. typhimurium Gram-Positive S. aureus S. epidermidis E. facaelis L. monocytogenes C. xerosis B. subtilis S. pyogenes & mitis S. pneumoniae TI-O N z OH PH Z U rt rt CN O O O cd cd cd Z u I rt CN O cd PO CQ pq pq^ rt rt rt ro co ro CN o o o o cd cd cd cd ~ pq pq pq pq pq > ID 1 o <D CO o CN The introduction of the turn-promoting proline and movement of the arginine from position 4 to position 2 (bacR,P) actually had a negative effect on the antimicrobial activities. The amidation of this peptide led to restoration of its activities, which confirmed that having more positive charges at the N - and C - termini improved the antimicrobial activities. Substitution of the arginine residues with lysine residues did not seem to have any effect, since bac3K,P had the same M I C values as bacR,P against all tested Gram-positive bacteria except for S. mitis against which bac3K,P had a M I C of 2 pg/ml, 8 times lower than bacR,P. Bac2I-CN and bacP,2R-CN were only tested against three Gram-positive bacteria S. aureus, S. epidermidis, and E. facaelis due to the limited amount of peptide obtained. They had improved activities only against S. epidermidis. BacP,2R-CN had two arginine residues inserted in the loop, while bac3R,P had them at the ends. At the same time as having improved MICs against Gram-positive bacteria, bac2R, bac3R,P, and bac3R,P,(V) had the same antimicrobial activities against the Gram-negative bacteria P. aeruginosa, and S. typhimurium as did native bactenecin, with bac2R 2-fold better which was not significant. These three peptides as well as bacR,P-CN had increased activities by a factor of 4 against E. coli. BacR,P-CN, bacR,P, bac2I-CN, and bacP,2R-CN were found to have slightly worse antimicrobial activities against P. aeruginosa and S. typhimurium. BacR,P, bac2I-CN, and bacP,2R-CN had a 2-fold higher antimicrobial activities against E. coli than did native bactenecin. Bac3K,P had the lower activities against all three Gram-negative bacteria, possibly due to the introduction of both lysine and proline residues. 95 In conclusion, introduction of more positive charges increased the antimicrobial activities, but the position of the positive charges was important. For bactenecin, the positions for addition of charged residues which will improve the antimicrobial activity appeared to be at the N - and C-termini, which form the hydrophilic surface of the bactenecin molecule, while the loop which forms the hydrophobic surface was not an appropriate place for charge addition (Figure 25). The type of positive side chain did not seem to be critical. D. Hydrophobicity Two derivatives of bactenecin were made to determine the effect of hydrophobicity on antimicrobial activities, namely bacP and bacW. Hydrophobic residues such as tryptophan and proline were introduced into the loop since the loop moiety comprised the hydrophobic surface of bactenecin. In their model of bactenecin, Romeo et al. (1988) proposed that bactenecin adopted an antiparallel extended structure forming a bend containing three a-carbon atoms (y turn) at position 7. A proline residue was inserted in the middle of the loop, at position 7 (bacP), to promote the turn as well as hydrophobicity. BacW had an added aromatic tryptophan residue instead. BacP remained relatively inactive against the Gram-positive bacteria S. aureus, S. epidermidis, E. facaelis, B. subtilis, S. pyogenes, and S. pneumoniae, and had similar activities to native bactenecin against C. xerosis, L. monocytogenes and S. mitis, while the activities against Gram-negative bacteria were lost (Table XI). On the other hand, bacW not only retained its activities against Gram-negative bacteria, but it also had comparable 96 antimicrobial activities against Gram-positive bacteria to linear bac2A-CN. It seemed that hydrophobicity played a critical role for bactenecin peptides to interact with Gram-positive bacteria, which do not have outer membranes. Therefore, hydrophobicity might Table X I : MICs of Hydrophobicity Bactenecin Derivatives M I C (pg/ml) Bacteria Species BacP BacW Bac2R, W Bac Gram-Negative E. coli 32 8 2 8 P. aeruginosa >64 4 2 4 S. typhimurium >64 4 2 8-16 Gram-Positive S. aureus 64 4 2 64 S. epidermidis 64 2 1 >64 E. facaelis >64 8 2 >64 L. monocytogenes 8 1 0.25 8 C. xerosis 2 0.5 0.25 1 B. subtilis >64 32 64 >64 S. pyogenes 64 2 1 >64 S. mitis 4 1 0.25 16 S. pneumoniae >64 16 8 2 1. Peptide sequences (also see Table III): Bactenecin R L C R I V V I R V C R BacP R L C R I V P V I R V C R BacW R L C R I V W V I R V C R BacW, 2R R R L C R I V W V I R V C R R be important for the interaction of peptides with the cytoplasmic membrane of bacteria. Tryptophan was a better substitution for improving antimicrobial activity than proline. When comparing bac2R and bacW, it was interesting to find that bac2R had decreased antimicrobial activities against Gram-positive bacteria, except C. xerosis and S. mitis, than did bacW, while it had a 4-fold increase in activity against E. coli. These results may imply that the effect of positive charges is more important for peptides to interact with the outer membrane of Gram negative bacteria, and that the effect of hydrophobicity is more important for the interaction with the cytoplasmic membrane. Bac2R and bacW, however, had the same MIC against P. aeruginosa and S. typhimurium. Based on the above results I designed bac2R,W which combined the optimal features of bac2R and bacW. It seemed to be the bactenecin derivative with the highest antimicrobial activities against both Gram-negative and Gram-positive bacteria, produced in this study. It had a 4-fold increase in antimicrobial activities against all three Gram-negative bacteria tested, and had at least a 4 -fold, and up to 32-fold increase in activities against Gram-positive bacteria. E. Disulphide bond Reduced bac2R,W and bacW lost all activities against all three Gram-negative bacteria, just like the reduced bactenecin, which further confirmed the importance of the disulphide bond for the interaction with the outer membrane of Gram-negative bacterial (Table IX). Interestingly, reduction of the disulphide bond of both bacW and bac2R,W also led to the reduction of activities against Gram-positive bacteria to almost the same level as native bactenecin. This then indicated that there were also some structural requirements for the linear peptides to interact with bacteria and that these were clearly different from the structural requirements of the cyclic peptides. 98 F. Agglutination Activities of Bactenecin and its Derivatives Bactenecin and its derivatives did not lyse human red blood cells, but some of them did cause agglutination of these cells (Table XII). Bac2R,W which had the best antimicrobial activities, also was the most active in agglutinating erythrocytes at 32 pg/ml. This value was at least 4-fold higher than the highest of its MICs. BacW and bac2I-CN also exhibited the same agglutination activity. In general, the reduced forms of the cyclic bactenecin derivatives showed 2-8 fold higher agglutination activities than their oxidized equivalents. Reduced bactenecin also caused agglutination of red blood cells at a concentration of 16 pg/ml, 4 times lower than the hemagglutination concentration of native bactenecin (64 pg/ml). Reduced bac2R,W and reduced bacW also caused agglutination at a lower concentrations than did their disulphide-bridged equivalents, at 8 pg/ml (4-fold) and 4 pg/ml (8 fold) respectively. It seemed that the formation of the disulphide bond inhibited the agglutination of human red blood cells by bactenecin peptides. On the other hand, the linear derivatives bac2A-CN, bacS-CN, bac2S-CN, and bac2S did not agglutinate RBC (>64 pg/ml). BacS-CN, which was only two amino acids different from reduced bactenecin, was non-agglutinating. This may argue that the presence of the SH groups in the cysteine residues of reduced bactenecin may be important in the hemagglutinating effect of reduced bactenecin derivatives. The low agglutinating activities of many bactenecin derivatives, especially those that had a broad spectrum of antimicrobial activities, make these peptides interesting and valuable candidates for drug development. 99 Table XII : Agglutination Activities of Bactenecin and its Derivatives on Human Red Blood Cells pg/ml Peptide Oxidized Form Reduced or Linear Form Bac2A-CN >64 BacS-CN 32 Bac2S-CN >64 Bac2S >64 BacR, P-CN >64 N/A BacR, P >64 N/A Bac3K, P >64 4 Bac3R, P 64 4 Bac3R, P, (V) >64 32 Bac2I-CN 32 N/A Bac2R, P-CN >32 N/A BacP 64 32 BacW 32 4 Bac2R 64 4 Bac2R, W 32 8 Bactenecin 64 4 G. Summary Bactenecin derivatives were designed based on the general information obtained in previous studies with other classes of peptides. Their antimicrobial and agglutinating activities were determined. The structural elements that were critical for antimicrobial function were thus examined. Their effects on antimicrobial activity of bactenecin is summarized in Table XIII. More positive charges at the N- and C-termini of the molecules facilitated the interaction between peptides and target cells. Increased 100 hydrophobicity in the loop not only increased the antimicrobial activities, but also broadened the antimicrobial spectrum. The disulphide bond was important for the interaction with the outer membrane of Gram-negative bacteria. Two interesting candidates of bactenecin derivatives with optimized activities were identified, bac2A-CN and bac2R,W. These two peptides had high activities against both Gram-positive and Gram-negative bacteria, and a low hemagglutination effect. Table XIII: Influence of Bactenecin Structural Modifications on Antimicrobial Activity Antimicrobial Activity Gram-negative bacteria Gram-positive bacteria Reduced and Linear - + Linear/Amidated ++ ++++ Positive Charge + + Hydrophobicity + +++ Hydrophobicity/ Positive charge ++ ++++ Modifications that resulted in increased activity are indicated by a "+" sign and the numbers of "+" signs represent the relative improvement. Modifications that resulted in decreased activity are indicated by a "-" sign. 101 D I S C U S S I O N A. Overview Polycationic peptides as "endogenous antibiotics" are potential sources of novel antibiotics due to their unique properties. They are usually amphipathic molecules and tend to carry multiple positive charges in nature, have antimicrobial activity against a wide range of microorganism including bacteria, fungi and enveloped viruses. They were proposed to have a novel action on bacterial cells as membrane-perturbing agents (Lehrer, 1989; Westerhoff et al., 1989; Skerlavaj et al., 1990), which is different from the conventional antibiotics which act by inhibition of either cell wall or macromolecular synthesis (Neu, 1992). Bactenecin is a cationic loop molecule with only 12 amino acids. Being the smallest and simplest disulphide-bonded antimicrobial cationic peptide, bactenecin made an interesting model peptide for this study. Little information is available on its mode of action, and whether it shares a common mechanism with other well-known cationic peptides as a membrane-perturbing agent, or if it has a unique mechanism due to its small size (too short to span the membrane) and unique loop structure. Its small size also makes it a desirable potential therapeutic agent, making easy to produce and manipulate. The disulphide bond could potentially render it protease resistant. Thus the general aim of this study was to investigate the mode of action of bactenecin and structure:function relationships in related peptides. Such a study would provide a better understanding of the nature of the antibacterial properties of bactenecin, and provide the basis for design of a novel peptide with improved activity and a high therapeutic index. 102 With these aims in mind, two main approaches were taken in this study. First, biochemical assays were conducted to assess the antimicrobial mechanism of bactenecin, especially its interaction with the bacterial membranes including the outer membrane of Gram-negative bacteria and the cytoplasmic membrane. Second, analogs were designed with simple amino acid modifications (substitution and/or addition) to examine the effect of key structural elements (including the disulphide bond, positive charge, ring size, hydrophobicity and amphipathicity) on the antimicrobial activity of bactenecin. The toxicity of these analogs was also examined. Besides the main observations, the work in this thesis have raised several issues worthy of discussion and future studies. B . Recombinant Expression of Cationic Peptides The main focus of this study was the antimicrobial mechanism and the structure:function relationship studies, but first I wished to evaluate the possibility that such peptides could be produced recombinantly, making them accessible for commercial exploitation. To produce peptide, there are three ways, from the natural source, by chemical synthesis, and by recombinant expression. Recombinant expression is cheap on a large scale and avoids problems with troublesome amino acids that are difficult to make (e.g. tryptophan and arginine). Recombinant expression also has the potential to be developed as a tool for screening for analogs, to facilitate structure:function studies, and as a tool for incorporating heavy atoms to facilitate 3-dimensional structure studies. The production of bactenecin, as well as two other peptides (indolicidin and apidaecin), was thus attempted by recombinant expression technique. 103 Three small cationic peptides, apidaecin, indolicidin and bactenecin, were recombinantly expressed in S. aureus, and bactenecin was also expressed in an E. coli system. Piers et al (1993) were able to successfully express the 26-28 amino acids alpha-helical peptides C E M E and C E M A in S. aureus. Although the 33 amino acid peptide HNP-1, a (3-sheet structure stabilized by three disulphide-bonds (Piers et al., 1993) was expressed in the linear format, no success in refolding the peptide into its native form was achieved (Hancock, personal communication). Conversely C E M E and C E M A were shown to be chemically and biologically identical to their synthetic counterparts, while HNP-1 produced recombinantly did not show any antibacterial activity. It seemed that peptides with multiple disulphide bonds could not fold correctly in bacteria. It was the objective of the first Chapter of this thesis to extend this technology to smaller peptides (e.g. the 12-amino acid bactenecin, and 13 amino acid indolicidin),. to peptides with highly restricted amino acid compositions (e.g. tryptophan, proline-rich indolicidin and proline, arginine-rich apidaecin), and to loop peptides with a single disulphide bridge (bactenecin). As shown here, I was successful in expressing all 3 peptides tried. There appeared to be no reduction in the amount of the protein Axationic peptide fusion proteins produced, compared to the amount of protein A carrier present in control cells. Two products of bactenecin were obtained recombinantly, and identified as an inter-chain disulphide-bonded multimer and an intra-chain disulphide-bonded monomer. It was not clear whether the disulphide bonds were formed during the recombinant synthesis in bacteria or during the subsequent purification. Only one major fusion protein product was detected, and this seemed to be a monomer (Fig. 3). Since the bactenecin was rather 104 small (less than 1.5 K dalton) compared to its fusion partner (32 K dalton), it might have been difficult for the disulphide bond to be formed at the C-terminus, thus explaining why no dimers or multimers of fusion protein were detected on S D S - P A G E . Therefore I would reason that the oxidation happened during the subsequent C N B r cleavage and purification procedure, and both products (multimer and cyclized monomer of bactenecin) were thus obtained (Fig. 4). As for indolicidin and apidaecin, only one product was obtained. Therefore, the S. aureus system, which permits efficient downstream purification due to the secretion of fusion protein, is acceptable for the production of different types of cationic peptides. A n E. coli system was also adopted in this study to optimize the expression and production of peptides, although an insufficient amount of pure bactenecin was obtained to do M I C tests. Later, my colleagues Drs. Fidai and Falla did obtain pure bactenecin and indolicidin using this system and showed there was no difference in antimicrobial activity compared to chemically made ones (Zhang et al., 1998). The E. coli system has been improved to allow for much higher expression since that time, by utilizing a strong phage T7 promoter and an optimized ribosome binding sequence (Zhang et al., 1998). Apidaecin produced recombinantly showed no activity against Gram-positive bacteria, which agreed with previous studies showing that apidaecin had only activity against Gram-negative bacteria (Casteels et al., 1990). Chemically made indolicidin was shown to be active against E. coli UB1005 (MIC =16 pg/ml), its mutant D C 2 (4 pg/ml) and P. aeruginosa (64 pg/ml), and it had little activity against Gram-positive bacteria (Falla et al., 1996). These results agreed with the data obtained with recombinantly-made 105 indolicidin (Table V) . Recombinantly-made, disulphide-bonded bactenecin had 4-8 fold higher MICs against wild type and mutants of E. coli and P. aeruginosa compared to chemically-synthesized bactenecin. This difference may be due to impurities in the sample, a relatively low efficiency of disulphide bond formation or inaccuracies in determining the exact concentration of bactenecin (this was solved later in the thesis by amino acid analysis to accurately determine peptide composition). Further purification could be carried out by reverse phase Pep HR5 column. Recombinantly-made bactenecin did not exhibit any activity against S. aureus (MIC > 64 pg/ml) and this observation was subsequently confirmed by the chemically-made bactenecin which also had no activity against S. aureus. This result contradicted previous studies (Romeo, 1988) which showed that bactenecin was active with a M I C 1-8 pg/ml against S. aureus. The different medium used in the previous studies for M I C determination may partially account for the difference. L B broth with no salt was used in this study while a Iso-Sensitest broth (Oxoid, U K ) was used for previous studies. Another possibility would be that a mixture of cyclic bactenecin contaminated with linear amidated bactenecin or amidated cyclic bactenecin was used in the previous studies, since the linear amidated bactenecin (Bac2A-NH2 and Bac2S-NH2) was active against S. aureus at a M I C of 4 pg/ml. Unfortunately, attempts to make pure amidated linear bactenecin (Lin Bac-NH2) and cyclic amidated bactenecin (Bac-NH2), by chemical synthesis, to test this hypothesis, failed. In conclusion, even peptides with restricted or short amino acid sequences, or with single disulphide bonds could be recombinantly produced and had equivalent activity against bacteria to chemically synthesized cationic peptides. 106 C. Antimicrobial Mechanism of Bactenecin 1. The Role of Secondary Structure Bactenecin belongs to a group of cationic peptides with only one disulphide bond. Although the four basic arginine residues permit classification of bactenecin as a cationic peptide, it has a unique, characteristic structure, making it distinct from any other known peptides. It is the smallest cationic peptide with only twelve amino acids, and is too small to span bacterial cytoplasmic membrane to form channels, like the amphipathic a-helical peptide or longer and more complex P-structured defensins, unless a multimer is involved. The disulphide bond makes bactenecin a loop molecule with a hydrophobic ring and positive charges, located at the C- and N-termini, which are brought together by the disulphide bond. Therefore it is interesting to investigate the role of these structural features in the antimicrobial mechanism. The disulphide bond was proven to be a key secondary structure element. In this study, it was shown that bactenecin was active against the wild type Gram negative bacteria E. coli, P. aeruginosa and S. typhimurium, whereas the linear unamidated derivative and reduced form were relatively inactive. A l l three forms were equally active against outer membrane barrier defective mutants. This observation indicated that the disulphide bond was important for interaction with the outer membrane as confirmed here. Bactenecin had a better binding ability for LPS than its two linear variants (bac2S and reduced bactenecin) and also permeabilized the outer membrane better, explaining its better activity vs. wild type Gram negative bacteria. Computer modeling of bactenecin with Insightll software (Biosym Technologies Inc., 107 San Diego, CA) indicated that bactenecin was a disulphide-bonded loop molecule comprising a long p-turn of hydrophobic residues and a positively charged face constructed from the C- and N-terminal portions of the molecule (Fig. 25). The model was consistent with the CD spectral studies (Fig. 10), which indicated that bactenecin existed as a rigid P-turn loop molecule regardless of its environment. Such a conformation, may make bactenecin a more amphipathic molecule than the unstructured linear and reduced form which exist in solution as random structures. This could explain why bactenecin interacted better with the negatively charged LPS than its linear and reduced forms. Alternatively since the loop brings together the 2 N-terminal and 2 C-terminal arginines, it could be that the higher local charge density is important for disrupting the divalent cation bridging of LPS at the outer membrane surface. Compared to polymyxin B and two a-helical peptides C E M E and C E M A (Piers, 1994), bactenecin was a relatively weak outer membrane permeabilizer. Bactenecin was too small to span the membrane and form pores or channels unless a multimer is involved. Therefore it was of interest to know how this molecule interacted with the cytoplasmic membrane, which was believed to be the final target of many cationic peptides (Lehrer, 1989). If bactenecin interacted with the cytoplasmic membrane, it should at a minimum make it proton-leaky, and thus dissipate the membrane potential. The cytoplasmic membrane depolarization assay results showed that bactenecin had only minor effects on membrane potential at MIC concentrations, whereas the maximal effect was only seen at 4 fold the MIC. 108 Linear variants of bactenecin adopted (3-structure in the presence of liposome. Despite their equivalent M I C value against E. coli DC2 , the pattern of interaction of bactenecin and its linear variants with the cytoplasmic membrane was quite different. The linear variants dissipated the cytoplasmic membrane potential at concentrations as low as 0.125 pg/ml, and rapid kinetics were observed. This pattern was similar to that seen with the a-helical peptide C E M E , which suggest that they act similarly. Planar bilayer studies showed that the linear variants and the native bactenecin had different channel forming activities, which agreed with the cytoplasmic membrane permeabilization results. The loop bactenecin was poorly active in model membranes, requiring high voltages to initiate conductance events and forming smaller channels compared to its linear variants (reduced bactenecin and Bac2S) (Wu and Hancock, submitted). The different conformations that these bactenecin peptides adopted when interacting with hydrophobic membranes may be responsible for these differences. Linear variants with a more flexible structure were able to interact with the inner membrane in a manner that was fundamentally different from that of the native bactenecin with a rigid cyclic structure. The lethal step for the native bactenecin may be quite different from the model proposed for classical peptides like defensin and cecropin, while linear variants may function by a similar channel-forming mechanism. Linear variants are also too small to span the membrane unless multimers were involved. The possibility that these peptides may change their conformation (disulphide-bonded bactenecin be ing reduced or l inear forms be ing oxidzed) once inside the c e l l can not be excluded. In conc lus ion , secondary structure plays a k e y role i n the interaction o f bactenecin w i t h bacterial cel ls . The results argues for a distinct mechanism o f act ion for this peptide, such as the ones proposed for po lyca t ion ic aminoglycosides (Hancock , 1981). Other mechanisms o f act ion have been suggested for an t imicrobia l peptides, i nc lud ing s t imulat ion o f auto ly t ic enzymes (Chi tn is et a l , 1990), interference w i t h bacterial D N A and/or prote in synthesis (Lehrer, 1989; B o m a n et a l , 1993), i nh ib i t ion o f D N A synthesis leading to f i l iamentat ion (Subbalakshmi et a l . , 1998), or general b i n d i n g to and inh ib i t ion o f cel lular nuc le ic acids (Park et a l . , 1998). 2. Is the Membrane the Only and Final Target of Cationic Peptides? Lehrer et a l . (1989) showed that defensins permeabi l ized both the outer and inner membrane o f E. coli, and that the inner membrane permeabi l iza t ion and c e l l death were two c lose ly l i n k e d events. Therefore they proposed that the inner membrane permeabi l iza t ion was the lethal step o f the k i l l i n g act ion o f cat ionic peptides. Howeve r , a mixture o f H N P 1 - 3 ( l O O p g / m l , at a molar ratios o f 1:1:0.5 instead o f M I C amounts) were used i n this study, this compl ica t ing the results. Later ex v i v o studies were interpreted as indicat ing that the formation o f mul t imer ic pores was responsible for inner membrane permeabi l iza t ion ( K a g a n et a l . , 1990; H i l l et a l , 1991; Schwarz et a l . , 1992; W i m l e y et a l . , 1994; M a t s u z a k i et a l . , 1994; R e x , 1996). The formation o f i o n channels w o u l d cause the leakage o f protons or other ions and depolarize the membrane potential . However , 110 studies regarding the inner membrane activity were mainly done on model membrane. To investigate this event from the microbiological, instead of biophysical point of view, an assay was developed by taking advantage of an outer membrane-defective mutant E. coli D C 2 and a membrane potential indicator DiSC3-(5). The ability of peptides representing different secondary structures to depolarize the membrane were examined. If the final and only target of cationic peptides was the inner membrane, one would expect that the killing concentrations of peptides (MICs) would lead to the complete disruption of the inner membrane. It was surprising to find that there was no correlation between the ability to dissipate the membrane potential and the concentration leading to cell death. At their MICs, different degrees of membrane potential dissipation were observed. For example, the alpha-helical peptide CP26 (Fig. 16) failed to cause any dissipation of the membrane potential, the beta-structured peptide Gram4112 (Fig. 17) and two extended-structured peptide C P 1 1 C N and C P 1 0 C N (Fig. 18) caused partial depolarization, whereas the loop-structured peptide bactenecin (Fig. 19) caused only partial and very slow depolarization. The concentrations required to cause maximum dissipation of membrane potential varied also; some peptide required concentrations much lower than the M I C (e.g., two P-structure peptides Gramicidin S and Gram 474), whereas some required concentrations higher than the M I C (e.g., CP26). Gram474 caused maximum depolarization at 4 pg/ml (8 fold lower than its MIC) , while CP26 did this at 4 pg/ml (8 fold higher than its MIC). These observations suggest that the killing mechanism may not be as simple as previously proposed, and multiple mechanisms may be involved. For peptides like CP26, 111 very few peptide molecules that bind to the inner membrane will form channels or pores. It is proposed that these will flip-flop across the membrane without formal channel formation and in this way get into the cells and cause subsequent events which contribute to the cell death (Wu et al., submitted). For Gram474, dissipation of the membrane potential is clearly not enough to kill the bacterium and other targets must be required. Interestingly, no cell lysis was observed for any of the peptides studied, and there was no change in optical density over time, when the tests were conducted for either CP26 and Gram474. Inhibition of synthesis of macromolecules or precipitation of D N A and R N A , since D N A and R N A molecules are strongly negatively charged polyanions, would be possible mechanisms that might occur after entry of cationic peptides into the cytoplasm. 3. Do A l l Cationic Peptides Interact with the Bacterial Cytoplasmic Membrane by the Same Mechanism? The results of the cytoplasmic membrane permeability studies also raised another question. Do all cationic peptide interact with this membrane by the same mechanism? Different patterns of membrane potential dissipation were observed as discussed above, especially among peptides with different structures. Again CP26 and Gram474 would be two examples that represent the extremes. The extended-structured peptides (CP11CN and CP10CN) as well as bactenecin and its derivatives (bac2S and reduced bactenecin) are only 12-13 amino acids long, and as such they are not long enough to span the membrane. The bilayer membrane of an E. coli cell is approximately 5.4 nm to 6.5 nm thick (Park, 1987). In order for a peptide to span this distance as an a-helix it must be 20 112 amino acids or more in length or as a P-strand 14 amino acids (Eisenberg, 1982). Thus these peptides may interact with the membrane in a way different from the longer peptides (e. g. a-helical peptides). Longer peptides are potentially capable of forming regular channels by multimerization, like the model proposed for defensin and cecropin. In contrast, shorter peptides may just aggregate in the membrane thus disrupting membrane integrity (Wu et al., submitted). D. Design of Novel Peptides with Improved Activity 1. Structure:Function Relationships Model peptides were designed to study the structure:function relationships in bactenecin-related peptides. Three structural elements were examined, namely basicity, hydrophobicity and loop-structure. B y modifying these elements, the antimicrobial activities were either improved or decreased. Computer modeling of the best and worst molecules were conducted in an attempt to explain the results at the level of assumed three dimensional structure. Increasing the numbers of positive charges at the C - or N- terminus either by including more arginine residues, or by amidation of the carboxyl terminus, increased antimicrobial activity. Bac2S-CN had better activity against both Gram-positive and Gram-negative bacteria than bac2S by at least 2 fold, and up to 128 fold (Table VIII). Bac2R with two extra arginines at C - and N-terminus also had better activity compared with the native bactenecin. This could be explained by the molecular modeling of bactenecin (Fig. 25). Cyclic bactenecin is an amphipathic molecule composed of a 113 hydrophobic loop and two positive charged ends that are brought in proximity to one another by the disulphide bridge. When the molecule binds to the negatively charged bacterial surface, the two ends will probably come in contact with the negatively charged bacterial surface first. Therefore increasing the positive charges at the terminus would presumably facilitate the interaction of peptides with the negatively charged cell surface. There were two kinds of residues which were introduced into the middle of the loop at position 7 to promote a turn (and make disulphide bond formation easier), proline (BacP), and tryptophan (BacW). BacP had worse activity, while BacW had improved activity. Fig 27 shows that insertion of the proline residue changed the direction of one arginine residue and the amphipathicity of bactenecin was destroyed, explained the decreased activity. Insertion of a tryptophan residue on the other hand maintained the amphipathic properties of bactenecin (Fig. 28 and 29). Thus, although bactenecin can tolerate larger ring structures, it is very important as to which residue is employed. Introduction of a proline residue at position 4 also decreased the antimicrobial activity against P. aeruginosa and S. typhimurium, although the molecule seemed still to be an amphipathic molecule by computer modeling. This may be because the two arginine residues (position 1 and 13) were brought too close to each other (Fig 25). E. Potential Application of Bactenecin There are several potential advantages of bactenecin and derivatives (Bac2A-NH2 and Bac2R, W) as antibiotics. They are relatively small, only 12-15 amino acids long. They would be easier to manipulate. They demonstrate no toxicity against human red blood cells. Certain peptides e.g. bac2A-CN and bac2R,W have a broad antimicrobial 114 spectrum against both Gram-negative and Gram-positive bacteria, and especially against several clinically important pathogens. Furthermore, the disulphide bond may render the peptides protease resistant, because the secondary structure maintained by the disulphide bond may make the molecule less accessible for proteases. MSI-78 is a 22-residue magainin analogue that has been in human Phase Ilb/III clinical trials. Four applications have been suggested for MSI-78 as a potential human therapeutic agent, "as a broad-spectrum topical agent; as a systemic antibiotic, as a wound healing stimulant, and as an agent against cancer" (Jacob and Zasloff, 1994). B a c 2 A - C N and bac2R,W certainly have the potential to be developed as broad spectrum antimicrobials. Their anti-cancer, and anti-virus activity and other therapeutic function are also worth investigating in future studies. F . Future Studies The optimal use of these peptides as antimicrobials will require full and detailed knowledge of their mode of action and the structural elements important in their activity and specificity (Nissen-Meyer and Nes, 1997). Therefore future studies should still concentrate on these aspects. As this study showed, cyclic bactenecin may have cellular targets (e.g. D N A , protein etc.) other than bacterial membrane. It will be very important to identify these targets and to understand the mode of action in the future. Bactenecin derivatives with both improved activity (bac2A-CN and bac2R,W) and worse activity (bacP and bacP,R) have been identified in this study. Thus the biochemical assays described in this study (e.g. the outer and cytoplasmic membrane permeability assays, dose and time-dependent . 115 kinetics of killing assay) should be applied to both types of derivatives in an attempt to explain these differences. The results obtained will help to further understand the effect of simple structural modifications on the antimicrobial mode of action. Secondary structure studies on these derivatives by circular dichroism in different environment (aqueous, hydrophobic, in the presence of liposome or LPS) and by 2-dimensional N M R should be done in conjunction with the antimicrobial mechanism studies. The linear variants of bactenecin and cyclic bactenecin may have fundamentally different mechanisms, as shown in this study. Thus bac2A-CN, a linear peptide and bac2R,W, a cyclic peptide may also have different modes of action and different approaches may be taken to study their modes of action. The factors that determine the different antimicrobial spectra of bactenecin and its derivatives against Gram-negative and Gram-positive, are poorly understood. This is another aspect that worth looking into. In summary, this study has provided some insight into the mode of action of bactenecin and its structural requirements for activity. There is plenty of work that needs to be done to fully understand the antimicrobial nature of this peptide and to its derivatives and develop a useful antimicrobial agent. 116 Figure 27: Computer generated model of bacP. This secondary structure model was developed using the program Insight II (Biosystems, San Diego, California) on a Silicon Graphics Indy workstation. The secondary structure was modeled based on energy rrunimization for 4000 iterations. Three steps of mirrimization were performed to reach the minimal total energy state: 1) Steepest descent; 2) conjugate gradient; 3) VA09A. Arginine residues are in red; the cysteine residues are in yellow. Isoleucine and leucine residues are in blue, and valine residues in green; O H group at the C-tenninus is in white, proline residue is in blue. 117 Figure 28: Computer generated model of bacW. This secondary structure model was developed using the program Insight II (Biosystems, San Diego, California) on a Silicon Graphics Indy workstation. The secondary structure was modeled based on energy minimization for 4000 iterations. Three steps of minimization were performed to reach the minimal total energy state: 1) Steepest descent; 2) conjugate gradient; 3) VA09A. Arginine residues are in red; the cysteine residues are in yellow. Isoleucine and leucine residues are in blue, and valine residues in green; OH group at the C-terrninus is in white, tryptophan residue is in blue. 118 Figure 29: Computer generated model of bac2R, W. This secondary structure model was developed using the program Insight II (Biosystems, San Diego, California) on a Silicon Graphics Indy workstation. The secondary structure was modeled based on energy nrinimization for 4000 iterations. Three steps of minimization were performed to reach the miriimal total energy state: 1) Steepest descent; 2) conjugate gradient; 3) VA09A. Arginine residues are in red; the cysteine residues are in yellow; isoleucine and leucine residues are in blue, and valine residues in green; OH group at the C-tenninus is in white; tryptophan residue is in blue. 119 R E F E R E N C E S 1. Agerberth B., Lee J. Y., Bergman T., Carlquist M., Boman H. G., Mutt V., (1991). Amino acid sequence of PR-39: Isolate from pig intestine of a new member of the family of proline-arginine rich antibacterial peptides. Eur. J. Biochem. 202: 849-854. 2. Amsterdam D., (1991). In Antibiotics in Laboratory Medicine (Lorian V., ed), pp. 72-78. Williams and Wilkins, Baltimore. 3. Andreu D., Merrifield R. B., Steiner H., and Boman H. G., (1985). N-terminal analogues of cecropin A: synthesis, antimicrobial activity and conformational properties. Biochem. 24: 1683-1688. 4. Angus B. L., Carey A. M., Caron D. A., Kropinski A. M. B., and Hancock R. E. W., (1982). Outer membrane permeability in Pseudomonas aeruginosa: comparison of a wild-type with an antibiotic-supersusceptible mutant. Antimicrob. Agents Chemother. 21:299-309. 5. Aumelas A., Mangoni M., Roumestand C , Chiche L., Despaux E., Grassy G., Calas B., and Chavanieu A., (1996). Synthesis and solution structure of the antimicrobial peptide protegrin-1. Eur. J. Biochem. 237:575-583. 6. Ausubel F. M., Brent R., Kingston R. E., Moore D. D., Seidman J. G., Smith J. A., and Struhl K., (1987). Current protocols in molecular biology. Greene Publishing Associates and Wiley-Interscience, New York, New York. 7. Bach A. C , Selsted M. E., Pardi A., (1988). Two-dimensional NMR studies of the antimicrobial peptide NP-5. Biochem. 26: 4389-4397. 8. Bader J. and Teuber M., (1973). Binding action of polymyxin B on bacterial membranes to the O-antigenic lipopolysaccharide of Salmonella typhimurium. Z. Naturforsch. Teil C. 28: 422-430. 9. Bagella L., Scocchi M., and Zanetti M., (1995). cDNA sequences of three sheep myeloid cathelicidins. FEBS Lett. 376: 225-228. 10. Bakker EP., (1978). Accumulation of thallous ions (T1+) as a measure of the electrical potential difference across the cytoplasmic membrane of bacteria. Biochem. 17(14):2899-2904. 11. Bazzo R., Tappin M. J., Pastore A., Harvey S., Carver J. A., and Campbell I. D., (1988). The structure of melittin: a 1H-NMR study in methanol. Eur. J. Biochem. 173 (1): 139-146. 120 12. Besalle R., Kapitkovsky A . , Gorea A . , Shalit I., and Fridkin M . , (1990). A l l - D -magainin: chirality, antimicrobial activity and proteolytic resistance. F E B S Lett. 274: 151-155. 13. Bevins C . L . , and Zasloff M . , (1990). Peptides from frog skin. Annu. Rev. Biochem. . 59:395-414. 14. Blondelle S. E . , and Houghten R. A . , (1991). Hemolytic and antimicrobial activities of the twenty-four individual omission analogues of melittin. Biochem. 30: 4671-4678. 15. Bohlmann H . , (1994). The role of thionins in plant protection. Crit Rev Plant Sci 13: 1-16. 16. Boman H . G . , (1995). Peptide antibiotics and their role in innate immunity. Annu. Rev. Immunol. 13: 61-92. 17. Boman H . G . , (1991). Antibacterial peptides: key components needed in immunity. Cell 65: 205-207. 18. Boman H . G . , Agerbath B. , and Boman A . (1993). Mechanisms of action on Escherichia coli of cecropin PI and PR-39, two antibacterial peptides from pig intestine. Infect. Immun. 61: 2978-2984. 19. Boman H . G . , and Hultmark D. , (1987). Cell-free immunity in insects. Annu. Rev. Microbiol. 41: 103-126. 20. Casteels P., Ampe C , Riviere L . , Van Damme J., Elicone E . , Fleming M . , Jacobs F. , and Tempst P., (1990). Isolation and characterization of abaecin, a major antibacterial response peptide in the honeybee (Apis mellifera). Eur. J. Biochem. 187: 2387-2391. 21. Carlsson A . , Engstrom P., Palva E . T. , and Bennich H . , (1991). Attacin, an antibacterial protein from Hyalophora cecropia, inhibits synthesis of outer membrane proteins in Excherichia coli by interfering with omp gene transcription. Infec. Immun. 59: 3040-3045. 22. Christensen B. , Fink J., Merrifield R. B. , and Mauzerall D. , (1988). Channel-forming properties of cecropins and related model compounds incorporated into planar lipid membranes. Proc. Natl. Acad. Sci. U S A 85: 5072-5076. 23. Chitnis S. N . , Prasad K . S. N . , and Bhargava P. M . , (1990). Isolation and characterization of auto lysis-defective mutants ofEscherichia coli that are resistant to the lytic activity of seminalplasmin. J. Gen. Microbio. 136: 463-469. 121 24. Clark D . P., Stewart D . , Maloy W. L . , and Zasloff M . , (1994). Ranalexin. A novel antimicrobial peptide from bullfrog (Rana catesbeiana) skin, structurally related to the bacterial antibiotic, polymyxin. J. Biol. Chem. 269 (14): 10849-10855. 25. Cohen M . L . , (1992). Epidemiology of drug resistance: Implications for a post-antimicrobial era. Science 257: 1050-1055. 26. Cowland J. B . , Johnsen A . H . , and Borregaard N . , (1995). hCAP-18, a cathelin/pro-bactenecin-like protein of human neutrophil specific granules. F E B S Lett. 368: 173-176. 27. Cruciani R. A . , Barker J. L . , Zasloff M . , Chen H . C , and Colamonici O., (1991). Antibiotic magainins exert cytolytic activity against transformed cell lines through channel formation. Proc. Natl. Acad. Sci. U S A 88: 3792-3796. 28. Delves-Boughton J. , Blackburn P., Evans R. J. , Hugenholtz J. , (1996). Applications of the bacteriocin, nisin. Antonie Van Leeuwen-hoek 69: 193-202. 29. Dempsey C . E . , (1990). The actions of melittin on membranes. Biochim. Biophys. Acta 1031: 143-161. 30. Duclohier H . , (1994). Anion pores from magainins and related defensive peptides. Toxicology 87: 175-188. 31. Eisenberg D. , Weiss R. M . , and Terwilliger T. C , (1982). The helical hydrophobic moment: a measure of the amphiphilicity of a helix. Nature 299: 371-374. 32. Falla T. J. and Hancock R. E . W. (1997). Improved activity of a synthetic indolicidin analog. Antimicrob. Agents Chemother: 771-775. 33. Falla T. J. , Karunaratne D . N . , and Hancock R. E . W. (1996). Mode of action of the antimicrobial peptide indolicidin. J. Biol. Chem. 271 (32): 19298-19303. 34. Fahrner R. L . , Dieckmarm T., Harwig S. S. L . , Lehrer R. I., and Feigon J., (1996). Solution structure of protegrin 1, a broad spectrum antimicrobial peptide from porcine leukocytes. Chem. Biol. 3: 543-550. 35. Fehlbaum P., Bulet, P., Chernysh S., Briand J. P., Roussel J. P., Letellier L . , Hetru C , and Hoffmann J. A . , (1996). Structure-activity analysis of thanatin, a 21-residue inducible insect defense peptide with sequence homology to frog skin antimicrobial peptides. Proc. Nalt. Acad. Sci. U S A 93: 1221-1225. 36. Fields P. I., Groisman E . A . , and Heffron F. , (1989). A Salmonella locus that controls resistance to microbicidal proteins from phagocytic cells. Science 243: 1059-1062. 122 37. Frohlich D. R., and Wells M . A . , (1991). Peptide amphipathy: a new strategy in design of potential insectidies. Int. J. Pept. Prot. Res. 37 (1): 2-6. 38. Frank R. W. , Gennaro R., Schneider K . , Przybylski M . , and Romeo D. , (1990). Amino acid sequence of two pro line-rich bactenecins. J. Biol. Chem. 265: 18871-18874. 39. Ganz T. , (1987). Extracellular release of antimicrobial defensins by human polymorphonuclear leukocytes. Infect. Immun. 55: 568-571. 40. Ganz T. , Selsted M . E . , and Lehrer R. I., (1987). In Bacteria-Host Cell Interaction (Horwitz M . , and Lovett M . , eds.), pp3-14. Alan R. Liss, Inc., New York. 41. Ganz T. , Selsted M . E . , and Lehrer R. I., (1990). Defensins. Eur. J. Haematol. 44: 1-8. 42. Gause G . F . , and Brazhnikova M . G . , (1944). Nature (London) 154: 703. 43. Ghazi A . , Schechter E . , Letellier L . , and Labedan B . , (1981). Probes of membrane potential in Escherichia coli cells. F E B S 125: 197-199. 44. Gough M . , Hancock R. E . W. , and Kelly N . M . , (1996). Anti endotoxic potential of cationic peptide antimicrobials. Infect. Immun. 64: 4922-4927. 45. Hammond S. M . , Lambert P. A . , and Rycroft A . N . , (1984). The bacteria cell surface. Croom Helm, London. 46. Hancock R. E . W. , and Carey A . M . , (1979). Outer membrane of Pseudomonas aeruginosa. Heat- and 2-mercaptoethanol-modifiable proteins. J. Bacteriol. 140: 902-910. 47. Hancock R. E . W. , (1981). Aminoglycoside uptake and mode of action-with special reference to streptomycin and gentamicin. J. Antimicrob Chemother 8: 429-445. 48. Hancock R. E . W. , Falla T. , and Brown M . H . , (1995). Cationic bactericidal peptide. Adv. Microb. Physiol. 37: 135-175. 49. Hancock R. E . W. (1997). Peptide antibiotics. Lancet 349: 418-422. 50. Hancock R. E . W. , and Lehrer R. I., (1998). Cationic peptides: a new source of antibiotics. Trend in Biotech. 16: 82-87. 51. Harold, F. M . , (1970). Adv. Microbiol. Physiol. 4: 45-104. 123 52. Harwig S. S. L. , Waring A. , Yang H . J., Cho Y . , Tan L. , and Lehrer R. I., (1996). Intramolecular disulfide bonds enhance the antimicrobial and lytic activities of protegrins at physiological sodium chloride concentrations. Eur. J. Bioehcm. 240: 352-357. 53. Harwig S. S. L. , Kokryakov V. N . , Swiderek K. M . , Aleshina G. M . , Zhao C., Lehrer R. I., (1995). Prophenin-1, an exceptionally proline-rich antimicrobial peptide from porcine leukocytes. FEBS Lett. 362: 65-69. 54. Hultmark D., Engstrom A. , Andersson K. , Steiner H. , Bennich H. , Boman H. G., (1983). Insect immunity. Attacins, a family of antibacterial proteins from Hyalophora cecropia. E M B O J. 2: 571-576. 55. Hi l l C. P., Yee J., Selsted M . E., and Eisenberg D., (1991). Crystal structure of defensin HNP-3, an amphiphilic dimer: mechanisms of membrane permeabilization. Science 251: 1481-1485. 56. Holak T. A. , Engstrom A., Kraulis P. J., Lindeberg G., Bennich H. , Jones T. A. , Gronenborn A. M . , and Clore G. M . , (1988). The solution conformation of the antibacterial peptide cecropin A: a nuclear magnetic resonance and dynamical simulated annealing study. Biochem. 27(20): 7620-7629. 57. Hoffmann J. A. , and Hetru C , (1992). Insect defensins: inducible antibacterial peptides. Immun. Today 13 (10): 411-415. 58. Hultmark D., (1993). Immune reactions in Drosophila and other insects: a model for innate immunity. Trends Genet. 9: 178-183. 59. Izumiya N . , Kato T., Aoyaga H. , Waki M . , and Kondo M . , (1979). Synthetic Aspects of Biologically Active Cyclic Peptides: Gramicidin S and Tyrocidines. Halsted Press, New York. 60. Jacob L. , and Zasloff M . , (1994). Potential therapeutic applications of magainins and other antimicrobial agents of animal origin. In: Boman H . G., Marsh J., Goode J. A . (eds). Antimicrobial peptides. Wiley, New York, pp 197-223. 61. Jung G, and Sahl H . G., (eds), (1991). Nisin and novel lantibiotics. Escom, Leiden. 62. Kagan B. L. , Ganz T., and Lehrer R. I., (1994). Defensins: a family of antimicrobial and cytotoxic peptides. Toxicology 87: 131-149. 124 63. Kagan B. L . , Selsted M . E . , Ganz T. , and Lehrer R. I., (1990). Antimicrobial defensin peptides form voltage-dependent ion-permeable channels in planar lipid bilayer membranes. Proc. Natl. Acad. Sci. U S A 87: 210-214. 64. Kennedy E . P., and Rumley M . K . , (1988). Osmotic Regulation of biosynthesis of membrane-derived oligosaccharides in Escherichia coli. J. Bacteril. 170: 2457-2461. 65. Kini M . R., and Evans H . J. , (1989). A common cytolytic region in myotoxins, hemolysins, cardiotoxins and antibacterial peptides. Int. J. Pept. Protein Res. 34: 277-286. • 66. Kraemer G.R., and Landolo J. J. , (1990). Current Microbiology 21: 373-376. 67. Kreiswirth B. N . , Lofdahl S., Bently M . J. , O'Reilly M . , Schlievert P. M . , Bergdoll M . S., and Novick R. P., (1983). The toxic shock syndrome exotoxin structural gene is not detectably transmitted by a prophage. Nature 305: 709-712. 68. Kreil G . , (1994). Antimicrobial peptides from amphibian skin: an overview. In Antimicrobial peptides (Boman H . G . , Marsh J. , and Goode J. A . , eds), pp77-90. Wiley, New York. 69. Lehrer R. I., Barton A . , Daher K . A . , Harwig S. S. L . , Ganz T. , and Selsted M . E . (1989). Interaction of human defensins with Escherichia coli. Mechanism of bactericidal activity. J. Clin. Invest. 84: 553-561. 70. Lehrer RI, Ganz T. , and Selsted M . , (1991). Defensins: endogeneous antibiotic peptides of animal cells. Cell 64: 229-230. 71. Lehrer R. I., Lichtenstein A . K . , and Ganz T. , (1993). Defensins: Antimicrobial and cytotoxic peptides of mammalian cells. Annu. Rev. Immunol. 11: 105-128. 72. Letellier L . and Shechter E . , (1979). Cyanine dye as monitor of membrane potentials in Escherichia coli cells and membrane vesicles. Eur. J. Biochem. 102, 441-447. 73. Levy O., Weiss J., Zarember K . , Ooi C . E . , and Elsbach P., (1993). Antibacterial 15-kDa protein isoforms (pi5s) are members of a novel family of leukocyte proteins. J. Biol. Chem. 268: 6058-6063. 74. Lichtenstein A . K . , Ganz T. , Nguyen T. M . , Selsted M . E . , and Lehrer R. I., (1988). Mechanism of target cytolysis by peptide defensins. Target cell metabolic activities, possibly involving endocytosis, are crucial for expression of cytotoxicity. J. Immunol. 140: 2686-2694. 75. Loh B. , Grant C , and Hancock R. E . W. , (1984). Use of the fluorescent probe 1-N-phenylnaphthylamine to study the interactions of aminoglycoside antibiotics with the 125 outer membrane of Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 26: 546-551. 76. Marion D. , Zasloff M . , and Bax A . , (1988). A two-dimensional N M R study of the antimicrobial peptide magainin 2. F E B S Letters 227: 21-26. 77. Martin E . , Ganz T. , and Lehrer R. I., (1995). Defensisn and other endogenous peptide antibiotics of vertebrates. J. Leukocyte Biol. 58: 128-136. 78. Matsuzaki K., Nakayama M . , Fukui M . , Otaka A . , Funakoshi S., and Fujii N . , (1993). Role of disulfide linkages in Tachyplesin-lipid interactions. Biochem. 32: 11704-11710. 79. Matsuzaki K . , Murase O., Tokuda H . , Funakoshi S., Fujji N . , and Miyajima K. , (1994). Orientational and aggregational states of magainin 2 in phospholipid bilayers. Biochem 33: 3342-3349. 80. Mayer L . D . , Hope M . J. , Cullis P. R., and Janoff A . S., (1985). Solute distributions and trapping efficiencies observed in freeze-thawed multilamellar vesicles. Biochim. Biophys. Acta 817: 193-196. 81. Maloy W. L . , Kari U . P., (1995). Structure-activity studies on magainins and other host defense peptides. Biopolymers 37: 105-122. 82. McManus M . C , (1997). Mechanisms of bacterial resistance to antimicrobial agents. A m . J. Health-Syst Pharm. 54: 1420-1433. 83. Moore R. A . , Bates N . C , and Hancock R. E . W., (1986). Interaction of polycationic antibiotics with Pseudomonas aeruginosa lipopolysaccharide and lipid A studied by using dansyl-polymyxin. Antimicrob. Agents Chemother. 29: 496-500. 84. Morikawa N . , Hagiwara K . , and Nakajima T. , (1992). Brevinin-1 and -2, unique antimicrobial peptides from the skin of the frog, Rana brevipoda porsa. Biochim. Biophys. Res. Commun. 189 (1): 184-190. 85. Neu H . C , (1992). The crisis in antibiotic resistance. Science 257: 1064-1073. 86. Nissen-Meyer J. , Nes I. F . , (1997). Ribosomally synthesized antimicrobial peptides: their function, structure, biogenesis, and mechanism of action. Arch Microb 167: 67-77. 87. Ohta M . , Ito H . , Masuda K . , Tanaka S., Arakawa Y . , Wacharotayankun R., and Kato N . , (1992). Mechanisms of antibacterial action of tachyplesins and polyphemusins, a group of antimicrobial peptides isolated from horseshoe crab hemocytes. Antimicrob. Agent Chemother. 36:1460-1465. 126 88. Pardi A. , Hare D. R., Selsted M . E., Morrison R. D., Bassolino D. A. , Bach A. C , (1987). Solution structures of the rabbit neutrophil defensin NP-5. J Moi . Biol. 201: 625-636. 89. Park C. B., Kim H . S., K im S. C , (1998). Mechanism of action of the antimicrobial peptide buforin II: Buforin II kills microorganisms by penetrating the cell membrane and inhibiting cellular functions. Biochem. Biophys. Res. Commun. 244: 253-257. 90. Park J . T. , (1987). The murein sacculus. In: Escherichia coli and Salmonella typhimurium. Cellular and molecular biology. (Neidhardt F. C. , ed). pp23-55, A S M Washington D . C . 91. Piers K. L. , Brown M . H . and Hancock R. E. W., (1993). Recombinant D N A procedures for producing small antimicrobial cationic peptides in bacteria. Gene 134: 7-13. 92. Piers K . L. , Brown M . H. , and Hancock R. E. W., (1994). Improvement of outer membrane-permeabilization and lipopolysaccharide-binding activities of an antimicrobial cationic peptide by C-terminal modification. Antimicrob. Agents Chemother 38: 2311-2316. 93. Perczel A. , and Hollosi M . , (1996). Turns. In: Circular dichroism and the conformational analysis of biomolecules (Fasman G. D., ed), pp.285-380, Plenum Press New York. 94. Qu X . , Harwig S. S. L. , Oren A. , Shafer W. M . , and Lehrer R. I., (1997). Protegrin structure and activity against Neisseria gonorrhoeae. Infect. Immun. 65: 636-639. 95. Rackovsky S., and Scheraga H. A. , (1980). Proc. Natl. Acad. Sci. U S A 77: 6965-6967. 96. Radermacher S. W., Schoop V . M . , and Schluesener H . J., (1993). Bactenecin, a leukocytic antimicrobial peptide, is cytotoxic to neuronal and glial cells. J. Neuroscience Res. 36, 657-662. 97. Rex S., (1996). Pore formation induced by the peptide melittin in difference lipid vesicle membranes. Biophysical Chem 58(l-2):75-85. 98. Richmond M . H. , Clarke D. C , and Wotton S., (1976). Indirect method for assessing the penetration of beta-lactamase-nonsusceptible penicillins and cephalosporins in Escherichia coli strains. Antimicrob. Agents Chemother. 10: 215-218. 127 99. R o c q u e W . J. , Fe s ik S. W . , H a u g A . and Mcgroa r ty E . J. , (1988). Po lyca t ion ic b ind ing to isolated l ipopolysacchar ide f rom antibiotic-hypersusceptible mutant strains o f Escherichia coli. A n t i m i c r o b . Agents Chemother . 32: 308-313. 100. R o m e o D . , Skerlavaj B . , B o l o g n e s i M . , and Gennaro R . , (1988). Structure and bacter ic idal act ivi ty o f an antibiotic dodecapeptide pur i f ied from bov ine neutrophils. J . B i o l . C h e m . 263: 9573-9575. 101.Saberwal G . , and Nagaraj R . , (1994). C e l l - l y t i c and antibacterial peptides that act by perturbing the barrier function o f membranes: facets o f their conformat ional features, stmcture-function correlations and membrane-perturbing abil i t ies. B i o c h i m . B i o p h y s . A c t a 1197: 109-131. 102.Sambrook, J. , F r i t i s ch E . F . , and Man ia t i s T . , (1989). In M o l e c u l a r c lon ing . A laboratory manual . C o l d Spr ing Harbor Laboratory Press, C o l d Sp r ing Harbor , N e w Y o r k . 103.Sawyer J. G . , M a r t i n N . L . , and H a n c o c k R . E . W . , (1988). Interaction o f macrophage cat ionic proteins w i t h the outer membrane o f Pseudomonas aeruginosa. Infect. I m m u n . 56: 693-698. 104.Schindler P . R . G . , and Teuber M . , (1975). A c t i o n o f p o l y m y x i n B o n bacterial membranes: morpho log ica l changes i n the cytoplasm and i n the outer membrane o f Salmonella typhimurium and Escherichia coli B . A n t i m i c r o b . Agents Chemother . 8: 94-104. 105.Schwarz T. , Z o n g R . T. , and Popescu T. , (1992). K i n e t i c s o f mel i t t in- induced pore format ion i n the membrane o f l i p i d vesicles . B i o c h i m . B i o p h y s . A c t a . 1110: 97-104. 106.Sekar V . , (1987). A rapid screening procedure for the ident i f icat ion o f recombinant bacterial clones. 5:11-13. 107.Selsted M . E . , N o v o t n y M . J. , M o r r i s W . L . , T a n g Y . Q . , S m i t h W . , and C u l l o r J. S. (1992). Indo l i c id in , a nove l bacter ic idal tridecapeptide amide from neutrophils . J. B i o l . C h e m . 267: 4292-4295. 108.Selsted M . E . , Szkla rek D . , and Lehrer R . I., (1983). Pur i f i ca t ion and antibacterial act ivi ty o f an t imicrobia l peptides o f rabbit granulocytes. Infect. I m m u n . 45: 150-154. 109.Selsted M . E . , and H a r w i g S. S., (1989). Determinat ion o f the disulf ide array i n the human defensin H N P - 2 . J. B i o l . C h e m . 264: 4003-4007. H O . S e t t i E . L . , and M i c e t i c h R . G . , (1998). N e w trends i n an t imicrobia l development. Current M e d i c i n a l C h e m . 5: 101-113. 128 l l l .Silvestro L . , Gupta K . , Weiser J. N . , and Axelsen P. H . , (1997). The concentration-dependent membrane activity of cecropin A . Biochem. 36: 11452-11460. 112.Sim P. J. , Waggoner A . S., Wang C . H . , and Hoffman J. F . , (1974). Studies on the mechanism by which cyanine dyes measure membrane potential in red blood cells and phosphatidylcholine vesicles. Biochem. 13 (16): 3315-3329. 113.Simmaco M . , Mignogna G . , Barra D. , Bossa F. , (1993). Novel antimicrobial peptides from skin secretion of the European frog Rana esculenta. F E B S 324(2): 159-161. 114.Skerlavaj B . , Romeo D. , and Gennaro R., (1990). Rapid membrane permeabihzation and inhibition of vital functions of gram-negative bacteria by bactenecins. Infect. Immun. 58: 3724-3730. 115.Smith J. A . , (1994). Neutrophils, host defense, and inflammation: a double-edged sword. J. Leukoc. Biol. 56: 672-683. 116.Sprott G . D . , Koval S. F . , and Schnaitman C. A . , (1994). In Methods for General Molecular Bacteriology (Gerhardt P., Murray R. G . E . , Wood W. A . , Kreig N . R., ed.), pp. 91-92. American Society for Microbiology Washington D C . 117.Spiker S., (1980). A modification of the acetic acid-urea system for use in microslab polyacrylamide gel electrophoresis. Anal. Biochem. 108: 263-265. 118.Steinberg D . A . , Hurst M . A . , Fujii C. A . , Kung A . H . , Ho J. F. , Cheng F. C , Loury D. J . , and Fiddes J. C , (1997). Protegrin-1: a broad-spectrum, rapidly microbicidal peptide with in vivo activity. Antimicrob. Agents Chemother. 41, 1738-1742. 119.Steiner H . , (1982). Secondary structure of the cecropins; antibacterial peptides from the moth Hyalophora cecropia. F E B S Lett. 137: 283-287. 120.Steiner H . , Andreu D. , and Merrifield R. B. , (1988). Binding and action of cecropin and cecropin analogues: antibacterial peptides from insects. Biochim. Biophys. Acta 939: 260-266. 121.Storici P., Tossi A . , Lenarcic B. , and Romeo D. , (1996). Purification and structural characterization of bovine cathelicidins, precursors of antimicrobial peptides. Eur. J. Biochem. 238: 769-776. 122.Subbalakshmi C , and Sitaram N . , (1998). Mechanism of antimicrobial action of indolicidin. F E M S Microbiol. Lett. 160: 91-96. 129 123.Suzuki S., Ohe Y . , Okubo T. , Kakegawa T. , and Tatemoto K . , (1995). Isolation and characterization of novel antimicrobial peptides, rugosins A , B and C , from the skin of the frog, Rana rugosa. Biochim. Biophys. Res. Commun. 212: 249-254. 124. Tamamura H . , Ikoma R., Niwa M . , Funakoshi S., Murakami T. , and Fujii N . , (1993). Antimicrobial activity and conformation of tachyplesin I and its analogs. Chem. Pharm. Bull . 41(5): 978-980. 125. Tamamura H . , Murakami T. , Horiuchi S., Sugihara K . , Otaka A . , Takada W., Ibuka T., Waki M . , Yamamoto N . , and Fujii N . , (1995). Synthesis of protegrin-related peptides and their antibacterial and anti-human immunodeficience virus activity. Chem. Pharm. Bull. 43(5); 853-858. 126. Tamamura H . , Kuroda M . , Masuda M . , Otaka A . , Funakoshi S., Nakashima H . , Yamamoto N . , Waki M . , Matsumoto A . , Lancelin J. M . , Kohda D. , Tate S., Inagaki F. , and Fujii N . , (1993). Biochim. Biophys. Acta 1163: 209-216. 127. Tossi A . , Scocchi M . , Zanetti M . , Storici P., and Gennaro R., (1995). PMAP-37 , a novel antibacterial peptide from pig myeloid cells. c D N A cloning, chemical synthesis and activity. Eur. J. Biochem. 228: 941-946. 128. Vaara M . , (1992). Agents that increase the permeability of the outer membrane. Microbiol. Rev. 56(3): 395-411. 129. Venyaminov S. Y . and Yang J. T. , (1996). Determination of protein secondary structure. In: Circular dichroism and the conformational analysis of biomolecules (Fasman G . D . , ed), pp.69-107, Plenum Press New York. 130. Wada K . , Wada Y . , Ishibashi F . , Gojobori T. , Ikemura T. , (1992). Codon usage tabulated from the GenBank genetic sequence data. Nucleic Acids Research 20 suppl: 2111-2118. 131. Wade D. , Andreu D. , Mitchell S. A . , Silveira A . M . V . , Boman A . , Boman H . G . , and Merrifield R. B . , (1992). Antibacterial peptides designed as analogs or hybrids of cecropins and melittin. Intl. J. Pept. Prot. Res. 40: 429-436. 132. Wade D. , Boman A . , Wahlin B. , Drain C. M . , Andreu D. , Boman H . G . , and Merrifield R. B . , (1990). A l l - D amino acid-containing channel-forming antibiotic peptides. Proc. Natl. Acad. Sci. U S A 87: 4761-4765. 133. Westerhoff, H . V . , Juretic, D. , Hendler, R. W. , Zasloff, M . , (1989). Magainins and the disruption of membrane-linked free-energy transduction. Proc. Natl. Acad. Sci. U S A 86, 6597-6601. 130 134. Wimley W. C , Selsted M . E . , White S. H . , (1994). Interactions between human defensins and lipid bilayers: evidence for the formation of multimeric pores. Protein Sci. 3: 1362-1373. 135. W u M . , and Hancock R. E . W., (1998). Interaction of the Cyclic Antimicrobial Cationic Peptide Bactenecin with the Outer and Cytoplasmic Membranes. Submitted to Journal of Biological Chemistry. 136. W u M . , Maier E . , Benz R., and R. E . W. Hancock, (1998). Mechanism of Interaction of Different Classes of Cationic Antimicrobial Peptides with Planar Bilayers and with the Cytoplasmic Membrane of Escherichia coli. Submitted to Journal of Biological Chemistry. 137. Yasin B. , Lehrer R. I., Harwig S. S. L . , and Wagar E . A . , (1996). Protegrins: structural requirements for inactivating elementary bodies of Chlamydia trachomatis. Infect. Immun. 64: 4863-4866. 138. Zanetti M . , Renato G . , and Romeo D. , (1995). Cathelicidins: a novel protein family with a common proregion and a variable C-terminal antimicrobial domain. F E B S letters 374: 1-5. 139. Zasloff M . , (1987). Magainins, a class of antimicrobial peptides from Xenopus skin: isolation, characterization of two active forms, and partial c D N A sequence of a precursor. Proc. Natl. Acad. Sci. U S A 84: 5449-5453. 140. Zhang L . , Fidai S., Falla T. , W u M . , Burian J., Key W. , and Hancock R. E . W., (1998). Determinants of Recombinant Production of Antimicrobial Cationic Peptides and Creation of Peptide Variants in Bacteria. Biochem. Biophys. Res. Commun. 247: 674-680. 141. Zhang X . L . , Selsted M . E . , and Pardi A . , (1992). N M R studies of defensin antimicrobial peptides. 1. Resonance asignment and secondary structure determination of rabbit nP-2 and human HNP-1. Biochem. 31: 11348-11356. 131 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0089250/manifest

Comment

Related Items