Clemson University TigerPrints
All Dissertations Dissertations
5-2011 APPLICATION OF NANOTECHNOLOGY FOR TARGETED DELIVERY OF ANTIBACTERIAL ENZYMES AND FOR ENZYME-BASED COATINGS ON MEDICAL DEVICES AND IMPLANTS Rohan Satishkumar Clemson University, [email protected]
Follow this and additional works at: https://tigerprints.clemson.edu/all_dissertations Part of the Biomedical Engineering and Bioengineering Commons
Recommended Citation Satishkumar, Rohan, "APPLICATION OF NANOTECHNOLOGY FOR TARGETED DELIVERY OF ANTIBACTERIAL ENZYMES AND FOR ENZYME-BASED COATINGS ON MEDICAL DEVICES AND IMPLANTS" (2011). All Dissertations. 740. https://tigerprints.clemson.edu/all_dissertations/740
This Dissertation is brought to you for free and open access by the Dissertations at TigerPrints. It has been accepted for inclusion in All Dissertations by an authorized administrator of TigerPrints. For more information, please contact [email protected]. APPLICATION OF NANOTECHNOLOGY FOR TARGETED DELIVERY OF ANTIBACTERIAL ENZYMES AND FOR ENZYME- BASED COATINGS ON MEDICAL DEVICES AND IMPLANTS
A Dissertation Presented to the Graduate School of Clemson University
In Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy Bioengineering
by Rohan Satishkumar May 2011
to Dr. Alexey Vertegel, Committee Chair Dr. Naren Vyavahare Dr. Ken Webb Dr. Antony Guiseppi-Elie
i
ABSTRACT
The frequency of S. aureus infection and subsequent biofilm formation associated with vascular catheterization has been increasing in recent years and often begins as a local colonization at the site of the catheter insertion. Antimicrobial enzymes and peptides, which are effective against a broad range of pathogens and low rates of resistance, have attracted attention as promising alternative candidates in treatment of infections caused by antibiotic resistant bacteria. The use of nanoparticles as carriers for enzymes, in addition to their size, charge, high surface area per volume etc. offers targeted delivery of enzymes to pathogenic bacteria. We proposed to use nanoparticles as surfaces for targeted delivery of antibacterial enzymes and as
„surrogate‟ surface coatings on indwelling central venous catheters (CVCs) to inhibit bacterial colonization and subsequent biofilm formation.
It was shown that nanoparticle charge can be used to enhance delivery and increase bactericidal activity of an antimicrobial enzyme, lysozyme. In the case of bacterial lysis assay with a Gram-positive bacterium Micrococcus lysodeikticus, activity of lysozyme conjugated to positively charged nanoparticles was approximately twice as high as that of free lysozyme. This was believed to occur through charge-directed targeting of enzyme-nanoparticle conjugates to negatively charged bacterial cell walls through enhanced electrostatic interactions. In a clinically more relevant model, we studied antimicrobial activity of lysostaphin against S. aureus for both lysostaphin-coated and lysostaphin-antibody coated nanoparticle conjugates at different enzyme: antibody: nanoparticle ratios. At the highest antibody loading, bacterial lysis rates for antibody- lysostaphin-coated samples were significantly higher than for plain lysostaphin-coated samples and free enzyme due to multiple-ligand directed targeting of antibody molecules to bacterial cell walls (p<0.05).
ii
We also performed in vitro experiments to evaluate the inhibition of bacterial colonies adhering to a surface. Bacterial infections by S. aureus strains are among the most common postoperative complications in surgical hernia repair with synthetic polymer meshes. Colony counting data from the broth count (model for bacteria in wound fluid) and wash count (model for colonized bacteria) for the enzyme-coated samples showed significantly decreased number of colony forming units (CFU) when compared to uncoated samples (p< 0.05). A pilot in vivo study showed a dose-dependent anti-S. aureus efficacy of lysostaphin-coated meshes in a rat model.
Finally, we observed that that coating of nanoparticles overall did not significantly improve binding yield, leaching, durability and antibiofilm activity of enzymes adsorbed on catheter segments (p>0.05). Alternatives to coating catheter surfaces using covalent chemistry through functional groups on nanoparticles either directly or through appropriate crosslinkers could result in significantly higher enzyme loadings, better stability and long term durability for future applications. The approach developed here is universal and can potentially be used for treatment of other medical device-associated infections. Moreover, use of antibacterial enzyme-
NP conjugates can eventually be expanded for intravenous administration, which will further broaden its range of application.
iii
DEDICATION
I would like to dedicate this work to my parents who have always been there for me with their unwavering support, guidance and strength. Thank you for all your love and prayers!
iv
ACKNOWLEDGEMENTS
I would like to sincerely thank my advisor, Dr. Alexey Vertegel, for his guidance and patience through the past 5.5 years of my graduate studies at Clemson University. His high expectations, continuous support and mentorship are greatly appreciated. I am grateful to the other committee members, Dr. Naren Vyavahare, Dr. Antony Guiseppi-Elie and Dr. Ken Webb for their time and active support in the completion of this dissertation. I would like to thank Dr.
Dan Simionescu and Dr. Martine Laberge for their support and guidance from the very beginning of my graduate study.
A special thanks to the all the members of the Nanobiomaterials laboratory for being there during my study at Clemson, sharing all those great moments and engaging in all the discussions with each and every one of you has been memorable. I would like to acknowledge
Gary Thompson, Sriram Sankar and Vladimir Reukov specifically for their help and support. I would also like to thank Sravya Durugu for her encouragement and constant support. Dr. David
Bruce and Iris Cordova (BET study), Dr. Meredith Morris (Flow cytometry), Dr. Terry Bruce
(Confocal microscope), Dr. Joann Hudson, Donald Musvee and Dayton Cash (SEM imaging),
Dr. Yuliya Yurko and Dr. Todd Heniford ( In vivo studies at Carolinas Medical Center). To all the staff in the Bioengineering department for their help and cheerful disposition – Thank you so much! I would like to personally thank all my friends and colleagues who have studied with me during my years at Clemson for their help and support.
Finally, I would like to acknowledge the financial support and assistance provided by
Carolinas Medical Center, ViMedrx and the Department of Bioengineering.
v
TABLE OF CONTENTS
TITLE ...... i
ABSTRACT ...... ii
DEDICATION ...... iv
ACKNOWLEDGEMENTS...... v
TABLE OF CONTENTS ...... vi
LIST OF FIGURES ...... ix
LIST OF TABLES ...... xiv
CHAPTER ONE ...... 1
INTRODUCTION ...... 1
CHAPTER TWO ...... 4
LITERATURE REVIEW ...... 4
2.1. Bacteria ...... 4
2.2. Antibiotic resistance ...... 10
2.3. Antimicrobial proteins and peptides ...... 16
2.4. Enzyme immobilization ...... 24
2.5. Nanotechnology for enzyme immoblization ...... 32
2.6. Nanoparticles ...... 35
2.7. Enzyme-nanoparticle conjugates as antibacterials – Targeted delivery ...... 41
2.8. Enzymes as antibacterial coatings on medical devices and implants...... 43
CHAPTER 3 ...... 45
SPECIFIC AIMS AND SIGNIFICANCE ...... 45
3.1. Clinical significance ...... 45
3.2. Specific aims ...... 48
vi
CHAPTER 4 ...... 53
CHARGE-DIRECTED TARGETING OF ANTIMICROBIAL ENZYME-NANOPARTICLE CONJUGATES ...... 53
4.1. Introduction ...... 53
4.2. Materials and methods ...... 54
4.2.2. Methods ...... 56
4.3. Results ...... 63
4.4. Discussion ...... 73
CHAPTER 5 ...... 77
ANTIBODY-DIRECTED TARGETING OF LYSOSTAPHIN ADSORBED TO POLYLACTIDE NANOPARTICLES INCREASES ITS ANTIMICROBIAL ACTIVITY AGAINST S.AUREUS IN VITRO ...... 77
5.1. Introduction ...... 77
5.2. Materials and methods ...... 79
5.3. Results and Discussion ...... 85
CHAPTER 6 ...... 100
EVALUATION OF ANTIMICROBIAL ACTIVITY OF LYSOSTAPHIN-COATED HERNIA REPAIR MESHES ...... 100
6.1. Introduction ...... 100
6.2. Materials and methods ...... 102
6.3. Results ...... 110
6.4. Discussion ...... 120
CHAPTER 7 ...... 125
APPLICATION OF ENZYMES AS ANTIBACTERIAL COATINGS ON PLAIN AND NANOPARTICLE COATED CATHETER SEGMENTS ...... 125
7.1. Introduction ...... 125
vii
7.2. Materials and methods ...... 127
7.3. Results and Discussion ...... 139
7.4. Conclusion ...... 163
CHAPTER 8 ...... 165
CONCLUSIONS AND FUTURE RECOMMENDATIONS ...... 165
8.1 Conclusions ...... 165
CHAPTER 9 ...... 169
CHAPTER 10 ...... 177
References ...... 177
viii
LIST OF FIGURES Page
2.1 A) Biofilm growing on a creek and B) S .aureus biofilm growing on a catheter segment...... 5
2.2 Bacterial cell walls for Gram positive and Gram negative bacteria…………………………..7
2.3 Alternatives to antibiotics - AQUACEL® Ag...... 14
2.4 Schematic of affinity enzyme immobilization through the Avidin-Biotin system…………..30
2.5 Nanotechnology for enzyme immobilization - Basic structures of nanoscale biocatalysts.....34
2.6 Reaction scheme showing different covalent immobilization techniques of proteins on polymer nanoparticles through different functional groups ………………...... 38
4.1 Schematic of lysozyme conjugation to nanoparticles...... 60
4.2 Efficiency of covalent attachment of lysozyme to polystyrene latex nanoparticles for samples prepared by (A) direct conjugation and (B) conjugation via PEG spacers...... 64
4.3 Results of bacterial lysis assay analyzing effect of enzyme activity for samples directly conjugated or through PEG spacers purified by centrifugation or dialysis...... 66
4.4 Results analyzing effect of control nanoparticles on bacterial lysis…………………………68
4.5 Tapping mode AFM images comparing the degree of aggregation for samples purified by dialysis and centrifugation...... 69
ix
List of Figures (continued…………)
4.6 Lysozyme activity assay with low molecular weight substrate PNP-(GlnAc)5
(A) Direct attachment and (B) attachment via PEG spacer……………………………………...70
4.7 Cartoon showing charge-directed targeting of lysozyme to bacterial cells………………….74
5.1 Schematic showing adsorption of lysostaphin to PLA nanoparticles………………………..86
5.2 Binding yield of A) Alexafluor 594 labeled lysostaphin to PLA NPs alone or in the presence of S. aureus antibody and B) Alexafluor 350 labeled S. aureus antibody on the PLA NPs……………………………………………………………………………………..87
5.3 Fraction of leached lysostaphin activity for samples with highest lysostaphin and antibody coatings………………………………………………………………………………...92
5.4 Bacterial lysis assay comparing activity of lysostaphin conjugated to NPs and free enzyme…………………………………………………………………………………………...96
5.5 (A) Kinetic constant of bacterial lysis plotted as function of adsorbed lysostaphin molecules per NP and (B) Kinetic constant of bacterial lysis plotted as function of adsorbed antibody molecules per NP...... 98
6.1 Binding yield for Alexa 594 labeled lysostaphin on Ultrapro mesh………………………..110
x
List of Figures (continued….)
6.2 Standard curve plotting the mean rate of lysis (OD per min) against the different
known lysostaphin concentrations……………………………………………………………..112
6.3 Leaching of adsorbed lysostaphin as a function of time for mesh samples treated by different initial enzyme concentrations…………………………………………………………113
6.4 Turbidimetric enzyme activity assay for the different lysostaphin-coated hernia mesh samples………………………………………………………………………………………….114
6.5 Fraction of leached sample activity for the aliquots collected at different time intervals compared against total mesh activity for different coating concentrations…………...116
7.1 Cartoon showing a single lumen catheter segment before and after coating with nanoparticles……………………………………………………………………………………141
7.2 Binding yield comparing lysostaphin and DispersinB on plain and NP coated catheter segments either alone or through coadsorption………………………………………………...142
7.3 SEM images of plain and PLA NP coated Polyurethane catheter segments at different magnifications…………………………………………………………………………………..145
List of Figures (continued….)
xi
7.4 Activity assay data comparing the kinetic constant of bacterial lysis against the highest and lowest lysostaphin coating concentrations for plain and NP coated catheter segments……………………………………………………………………………….148
7.5 Activity assay data comparing the kinetic constant of hydrolysis against the highest and lowest DisperinB coating concentrations for plain and NP coated catheter segments..…...150
7.6 Leaching of enzyme as a function of time for plain and NP coated catheter segments at different Coating concentrations for lysostaphin and DispersinB ………………………… .151
7.7 Direct cytotoxicity results of MTT assay that shows the percentage of cell viability after
48 h of all the catheter samples coated with the highest coating concentration that were placed in contact with 3T3 fibroblasts………………………………………………………….154
7.8 Percentage of Biomass on catheters for catheter samples coated with highest enzyme concentration…………………………………...... 156
7.9 Digital camera images showing biofilm mass on control and sample catheter segments………………………………………………………………………………………...158
7.10 Mean rate constants of bacterial lysis for lysostaphin coated on both plain and
NP coated catheter samples…………………………………………………………………….160
xii
List of Figures (continued……………)
7.11 Mean rate constants of hydrolysis for DispersinB coated on plain and NP coated catheter samples………………………………………………………………………………...162
xiii
LIST OF TABLES Page
2.1 List of some of the antibacterial peptides and proteins and their enzymatic mode of action……………………………………………………………………………………………22
4.1 Properties of Polystyrene nanoparticles with different functional groups…………………..56
4.2 Initial rates of the bacterial lysis study comparing samples conjugated directly or through PEG spacers purified by dialysis and centrifugation……………………………………67
4.3 Rates of hydrolysis using low molecular weight substrate PNP-(GlnAc)5 comparing activity of lysozyme conjugated directly or through PEG spacers………………………………71
4.4 Zeta potential analaysis………………………………………………………………………72
5.1 Characteristics of samples with different nanoparticle (NP): lysostaphin (LS)
: antibody (Ab) ratios…………………………………………………………………………….89
5.2 Size and zeta potential of all enzyme-NP conjugates and controls...... 90
5.3 Bacterial turbdimetric assay comparing activity of enzyme-NP conjugates against free enzyme………………………………………………………………………………………93
6.1 Specific enzyme activity of samples with different enzyme coating concentrations………115
List of Tables (continued…………)
xiv
6.2 Effect of lysostaphin coating concentration on log reduction of colonies for the broth supernatant and wash count……………………………………………………………………118
6.3 In vivo rat study results……………………………………………………………………..120
xv
CHAPTER ONE
INTRODUCTION
My graduate research at the department of Bioengineering in Clemson focused on the specific application of Nanotechnology for targeted delivery of antibacterial enzymes and for enzyme-based coatings on medical devices and implants. Using different model studies, further discussed in various chapters of my dissertation, the goal of my research was to use nanotechnology to address specific issues and drawbacks involved in the delivery of antibacterial enzymes as infections and as coatings on implants to prevent bacterial colonization and biofilm formation. Chapter 2 of this dissertation presents a detailed literature review on the current problems associated with treatment of nosocomial infections with antibiotics specifically focusing on multidrug resistant S. aureus, enzymes and peptides as alternatives to antibiotics and its drawbacks and finally the need for enzyme immobilization using nanotechnology through various techniques.
Chapter 3 lists the specific aims and goals of our research project by using model studies to test our hypothesis and fundamentally address specific issues involved in the application of enzymes for - a) targeted delivery using polymer nanoparticles (NPs) and b) as coatings on hernia repair polymer meshes and Central venous catheter segments.
Our first model study in Chapter 4 aimed at comparing the influence of nanoparticle charge on antibacterial activity of enzyme-nanoparticle conjugates.
Specifically, we covalently attached a model antibacterial enzyme hen egg lysozyme to two types of polystyrene latex nanoparticles: positively charged, containing aliphatic
1 amine surface groups, and negatively charged, containing sulfate and chloromethyl surface groups. We tested lysozyme activity against its substrate, Micrococcus lysodeikticus and compared antibacterial activity of free enzyme against enzyme conjugated to NPs. We further evaluated the method of enzyme purification and effect of PEG spacers on antibacterial activity of enzyme-NP conjugates
In a clinically more relevant model (Chapter 5), we studied whether co-attachment of different ligands such as an antibody (against the pathogen) along with antibacterial enzymes to NPs enhances in vitro antimicrobial activity of the enzyme. We compared the anti S. aureus activity of lysostaphin adsorbed on the surface of PLA (poly (lactic acid)) nanoparticles to that of the free enzyme at different initial enzyme: NP molar ratios. A rabbit polyclonal S. aureus antibody was coadsorbed on PLA nanoparticles along with lysostaphin to evaluate the effect of delivery based on antibody-directed targeting.
Bacterial infections by antibiotic-resistant S. aureus strains are known to be one of the most common postoperative complications in surgical hernia repair. In our third model study, we evaluated the in vitro antimicrobial susceptibility of S. aureus to lysostaphin- coated hernia repair polymer meshes and study the effect of enzyme coating concentration on antimicrobial activity. In vivo antimicrobial efficacy of lysostaphin-coated meshes was further evaluated by our clinical collaborators at Carolinas Medical Center in a rat hernia repair model (Chapter 6).
It is well known that bacteria can either colonize the implant surface or exist in a planktonic free floating form. Both these chapters (4 and 5) provided insight into taking advantage of special properties of NPs such as size, charge, surface area etc. to improve
2 functionality of enzyme-NP conjugates, specifically towards targeting to planktonic cell suspensions.
Colonization is the first stage of device associated-infection. Bacteria in colonies
(biofilms) present more danger than planktonic bacteria and are also less susceptible to treatment by antimicrobial agents. Although coating medical devices with plain antibacterial enzymes could work just as effectively, we hypothesized that coating catheter surfaces with nanoparticles initially may be advantageous due to larger surface area for effective enzyme coating, higher stability, durability and low enzyme leaching. Thus the goal of our final model study was the characterization and comparison of the in vitro performance for plain and NP coated catheter segments using antimicrobial and antibiofilm enzymes (Chapter 7). Coating was achieved via adsorption of two model enzymes, DispersinB (antibiofilm) and lysostaphin (antimicrobial) on the surface of plain and poly ((lactic acid)) nanoparticle coated polyurethane catheter segments. A series of experiments evaluating enzyme binding efficiency, activity, leaching, cytotoxicity, antibiofilm efficacy and durability were performed to compare plain and NP coated catheter segments.
Chapters 8 and 9 provide a conclusion to our studies and future recommendations to design more effective antibacterial surface coatings on devices and implants. By manipulating the available technology today, we do not just limit ideas only to antibacterial coatings but also hopefully expand to a wider range of biomedical applications.
3
CHAPTER TWO
LITERATURE REVIEW
2.1. Bacteria
Bacteria were one of first types of cell to evolve in nature. They have been described as a „prokaryotic‟ organism because their cells lack a membrane bound nucleus and their genetic material is typically a single circular chromosome located in the cytoplasm in an irregularly shaped body called the nucleoid [1]. Despite their small size, bacteria have an enormous range of metabolic capabilities, and a lot of them can be found in some of the extreme environments on earth. All of the various surface components of a bacterial cell are important in its ecology since they mediate contact of the bacterium with its environment. It must use its own surface components to assess the environment and respond ways that supports its own existence and survival in that environment. The surface properties of a bacterium are determined by the molecular composition of its membrane and cell wall. In most prokaryotes, the primary function of the cell wall is to protect the cell from internal turgor pressure caused by the much higher concentrations of proteins and other molecules inside the cell compared to its external environment.
Most bacteria also contain some sort of a polysaccharide layer outside of the cell wall or outer membrane. In a general sense, this layer is called a capsule. A true capsule is a discrete detectable layer of polysaccharides deposited outside the cell wall. A type of capsule found in bacteria called a glycocalyx is a thin layer of polysaccharide fibers which is mostly observed on the surface of cells growing in nature. Capsules are known to protect bacteria from engulfment by phagocytes and from attack by antimicrobial agents
4
[2]. In nature, and in many medical conditions, colonies of bacteria live in a biofilm, primarily composed of an outer slimy layer.
These films can range from a few micrometers to up to tens of centimeters in thickness. Bacteria living in biofilms display a complex arrangement of cells and extracellular components, forming secondary structures such as microcolonies, through which there are networks of channels to enable better diffusion of nutrients [3, 4]. Also, bacteria communicate with each other through an internal cell signaling process called
Quorum sensing, thus regulating a host of cell metabolic activities, similar to a communication network. In natural environments, such as soil or the surfaces of plants, the majority of bacteria are bound to surfaces in biofilms [5]. Biofilms are also important for chronic bacterial infections and infections of implanted medical devices, as bacteria protected within these structures are much harder to kill than individual bacteria [6].
A B
Figure 2.1. A) Biofilm growing on a creek (http://toxics.usgs.gov) and B) S .aureus biofilm growing on a catheter segment ((http://en.wikipedia.org)
5
2.1.1. Bacterial cell wall
The bacterial cell wall has been one of the important clinical targets for antibiotic agents since the first use of penicillin in World War II. Bacterial cell walls are made of peptidoglycan (also called murein), which is made from polysaccharide chains cross- linked by unusual peptides containing D-amino acids [7]. Bacterial cell walls are different from the cell walls of plants and fungi which are made of cellulose and chitin, respectively[8]. Although the primary function of the cell wall is to provide a rigid exoskeleton for protection against both mechanical and osmotic lysis [9], it also serves as an attachment site for proteins that interact with the bacterial environment.
Based on the structure of their cell wall, bacteria are classified into two types – a)
Gram-positive and b) Gram-negative bacteria. The names originate from the reaction of cells to the Gram stain, a test long-employed for the classification of bacterial species.
Gram-positive bacteria have a thick mesh-like cell wall made of peptidoglycan, which stain purple and Gram-negative bacteria have a thinner layer, which stain pink. Although the vast majority of bacteria adhere to the color differentiation of the Gram stain, some bacteria do not; these are called Gram-variable bacteria [10].
The cell wall of Gram-positive bacteria is a peptidoglycan macromolecule with attached accessory molecules such as teichoic acids, teichuronic acids, polyphosphates, or carbohydrates [9, 11]. Gram-positive bacteria have much thicker peptidoglycan layer than their Gram- negative counterparts and no external outer membrane. They lack a distinct periplasmic space, which is usually defined as the region between the cytoplasmic membrane and the outer membrane. The glycan strands of the cell wall consist of the repeating disaccharide N-acetylmuramic acid-(b1-4) - N-acetylglucosamine (MurNAc-
6
GlcNAc) [12, 13]. Glycan strands vary in length and are estimated to contain 5 to 30 subunits, depending on the bacterial species [14-16].
The muramic acid is a unique substance associated with bacterial cell walls and these chains are cross-linked by short polypetide chains consisting of both L- and D- aminoacids. Wall peptides are cross-linked with other peptides that are attached to a neighboring glycan strand [17-20], thereby generating a three-dimensional molecular net- work that surrounds the cell and provides the desired exoskeletal function [21]. The teichoic acids are polyols consisting predominantly of glycerol, ribitol and mannitol, covalently linked to the peptidoglycan through phosphodiester bonds and can be substituted by sugars, aminosugars or D-alanine residues. These anionic polymers account for 10–60% (by weight) of the bacterial cell wall, with the relative amount depending on the culture conditions [22, 23].
7
Figure 2.2. - Bacterial cell walls for Gram positive and Gram negative bacteria
(figure modified from Cabeen et al; 2005 [24].
Although they are structurally diverse[25, 26] ,the negative charge of these anionic polymers mainly originates from phosphate or carboxyl groups, and they may also elaborate acidic side chains containing glycerol, phosphate, organic acids (e.g. pyruvic and succinic acid), or sulphates. They play a very crucial role in the binding of divalent cations, balance of metal ions for membrane functionality, folding of extracellular metallo-proteins and formation of a barrier to prevent diffusion of nutrients and metabolites [23, 27-30]. In addition, the peptidoglycan is a well-known target for almost all clinically useful antibiotics that inhibit bacterial cell wall synthesis [31]. Being an essential and unique cell wall component of virtually all bacteria and not eukaryotic cells, it is an excellent target for recognition by the innate immune system of the host [31, 32].
The cell walls of Gram negative bacteria are composed of three layers, an inner bilayer phospholipid membrane, a thin peptidoglycan layer, and the outer bilayer membrane, composed of lipopolysaccharides [31, 32]. Along with the plasma membrane and the cell wall (outer membrane, peptidoglycan layer, and periplasm) constitute the
Gram-negative envelope [33, 34].The envelope of E-coli and other Gram-negative bacteria contains two distinct lipid bilayers that are separated by the peptidoglycan-containing periplasmic space that represent the inner membrane (IM) and the outer membrane (OM) respectively [35, 36]. The inner membrane is composed of phospholipids in both its inner and outer leaflets as well as integral and peripheral membrane proteins. It carries out a variety of functions typically assigned to both the plasma membrane and specific organelles in higher organisms that includes cell signaling, biosynthesis, electron transport
8 and ATP synthesis [37] . The OM is a unique asymmetric lipid bilayer consisting of an inner face of phospholipids and an outer face of lipopolysaccharide (LPS). The outer membrane (OM) plays an important role in nutrient uptake but in addition provides the organism with a remarkable permeability barrier, conferring resistance to a variety of agents including antibiotics [38].
2.1.2 Staphylococcus aureus
Staphylococcus aureus is facultative anaerobic Gram-positive coccus which occurs singly, in pairs, and in irregular clusters. It is nonmotile, non-spore forming, catalase and coagulase positive bacteruim. The term Staphylococcus is derived from the Greek term staphyle, meaning "a bunch of grapes." It can colonize and infect both healthy, immunologically competent people in the community and hospitalized patients with decreased host defenses [39]. It‟s repertoire of virulence factors is extensive, with both structural and secreted products playing a role in the pathogenesis of infection. Up- regulating the production of virulence factors enables it to persist in the blood stream, adhere and colonize the skin and mucosa, evade the host immune response, form protective biofilms through quorum sensing and also develop resistance against a wide range of antibiotics.
S. aureus has an intrinsic ability to form biofilms on damaged tissue, prosthetic materials such as medical devices and heart valves, thereby making it much more difficult to treat than planktonic bacteria. The biofilm matrix is usually composed of biopolymers such as polysaccharides, proteins and lipids which build a firm consistency thus providing protection against the immune cells and penetration of antimicrobial agents [40]. It is also
9 known that cells living in biofilms display a complex arrangement of various extracellular components and form secondary structures such as microcolonies and small-colony variants (SCVs) that demonstrate almost complete resistance against conventional antibiotics [41].
S. aureus biofilm formation involves two stages – initial adhesion of cells to a surface mediated by a number of surface proteins that serve as an anchor, followed by production of various extracellular factors that enable cell multiplication and maturation into a structured community. „Quorum Sensing‟ or intracellular signaling between cells via the accessory gene regulatory (agr) system has also been strongly implicated in the pathogenesis of biofilm-associated S. aureus infections [42]. The (agr) quorum-sensing system decreases the expression of several cell surface proteins and increases expression of many secreted virulence factors based on bacterial density [43]. It regulates the detachment of cells from the established biofilm and then allows them to colonize new sites or even enter the blood stream. The rates of infections caused by both community and hospital-acquired strains are increasing steadily and subsequent treatment of these infections is also becoming more difficult because of the increasing prevalence of multidrug-resistant strains [44].
2.2. Antibiotic resistance
Antibiotics have traditionally been used in modern medicine as chemotherapeutic agents to treat infections thus becoming indispensable in the modern health care system.
However the rates of multi-antibiotic resistance among bacteria that infect wounds are constantly on the rise and surgical site infections have become a substantial burden of
10 disease for patients and health services [45-48]. The US Centers for Diseases Control and
Prevention (CDC) estimates that about 500,000 surgical site infections occur annually in the United States [48-51]. There are several factors contributing to surgical site infections namely the inoculum of bacteria introduced into the wound during the procedure, the unique virulence of contaminants, the microenvironment of each wound, and the integrity of the patients host defense mechanisms [48, 50].Consequently, rapid control of wound infections and monitoring of prophylactic and therapeutic strategies have recently been proposed [49, 51] . Antibiotic resistance in nosocomial infections is thus an ever- increasing problem.
Antibiotics have been shown to act in different ways, either by targeting the pathogen„s physiology directly or by disrupting the cellular structure of bacterial cells.
There are 4 major modes of action: (1) interference with cell wall synthesis, (2) inhibition of protein synthesis, (3) interference with nucleic acid synthesis, and (4) inhibition of a metabolic pathway. Antibacterial drugs that work by inhibiting bacterial cell wall synthesis include the b-lactams, such as the penicillins, cephalosporins, and carbapenems
[52, 53]. They inhibit synthesis of the bacterial cell wall by interfering with the enzymes required for the synthesis of the peptidoglycan layer [52]. Macrolides, aminoglycosides and tetracyclines produce their antibacterial effects by inhibiting protein synthesis by binding to different subunits of the Ribosome [52, 53]. Fluoroquinolones exert their antibacterial effects by disrupting DNA synthesis and causing lethal double-strand DNA breaks during DNA replication [54]. Antibacterial drugs such as folic acid analogues inhibit different metabolic pathways. Disruption of the bacterial membrane is
11 characteristic of polymyxins that exert their inhibitory effects by increasing bacterial membrane permeability, causing leakage of bacterial contents [55].
Several mechanisms of antimicrobial resistance occur in different bacterial genera.
First, the organism may acquire genes encoding enzymes, such as β-lactamases, that destroy the antibacterial agent before it can have an effect. Susceptible bacteria can acquire resistance to an antimicrobial agent via new mutations [56]. Such spontaneous mutations may cause resistance by altering the target protein to which the antibacterial agent binds by modifying or eliminating the binding site (e.g., change in penicillin-binding protein 2b in pneumococci, which results in penicillin resistance), upregulating the production of enzymes that inactivate the antimicrobial agent (e.g., erythromycin ribosomal methylase in staphylococci) and altering an outer membrane protein channel that the drug requires for cell entry or by upregulating pumps that expel the drug from the cell [56]. Therefore strains of bacteria carrying resistance conferring mutations are selected by continuous antibiotic use, which kills the susceptible strains but allows the newly resistant strains to survive and grow.
Bacteria also develop resistance through the acquisition of new genetic material from other resistant organisms. Mutation and selection, together with mechanisms of genetic exchange, enable many bacterial species to adapt quickly to the introduction of antibacterial agents into their environment [57]. Resistance to multiple antibiotics is increasingly reported in a number of Gram-negative pathogens, especially the
Enterobacteriaceae [58-61], E-coli [62, 63], Pseudomonas aeruginosa [64-66]etc.
Multidrug resistance occurs as a result of accumulation of multiple mutations and/or resistance genes, though single mutations can also promote multidrug resistance
12
Also, drugs come with adverse effects and antibiotics are no exception. Common side-effects are gastrointestinal symptoms, skin rashes, and thrush. A common problem in clinical practice is in determining the required antibiotic to treat infection in a patient who has been labeled as being allergic to the antibiotic. In many cases, such patients are prescribed antibiotics that are less effective or more toxic, have a broader spectrum, or are more expensive than the drug of choice for their condition. The occurrence of overgrowth, or infection, as a direct result of antibiotic consumption is also less well understood by physicians. An antibiotic will inhibit, or kill, a whole range of bacteria, which creates space on mucosal and other surfaces for other organisms to proliferate [67]. These survivors might show various rates of acquired resistance, are more difficult to treat and almost impossible to eradicate in some patients [68].
2.2.1 Alternatives to antibiotics
There is therefore an urgent need for the use of alternative approaches to antibiotics. Widespread application of an effective topical antimicrobial agent substantially reduces the microbial load on an open wound surface and reduces the risk of infection [69-71]). However the selection of topical antimicrobial therapy should be based on the agent's ability to inhibit the microorganisms recovered from wound cultures and monitoring of the nosocomial infections acquired in hospitals and medical centers. For e.g. several topical antimicrobials have also been used for topical burn therapy, including gentamicin sulfate (0.1% water-soluble cream), betadine (10% povidone-iodine ointment), bacitracin-polymyxin ointment, and nitrofurantoin [72].
13
Fig 2.3. Images of AQUACEL® Ag - ionic silver (Convatec, NJ, USA). This material provides an effective barrier to bacterial penetration to help reduce infection. AQUACEL® Ag is indicated for use on acute and chronic wounds including burns, surgical wounds, diabetic foot ulcers, pressure ulcers and leg ulcers. The power of ionic silver in AQUACEL® Ag kills a broad spectrum of wound pathogens in the dressing that can cause infection - including
Staphylococcus aureus, methicillin-resistant Staphylococcus aureus (MRSA), vancomycin-resistant Enterococcus (VRE), and other pathogens (Figure source : http://www.convatec.com).
However, these compounds are no longer used extensively because significant resistance has developed and/or they have been shown to be toxic or ineffective at controlling localized burn wound infections [73]. Toxicity from silver is observed in the form of argyria or skin discoloration, usually when there is a large amount of silver ions are used in the wound dressing. However, a delayed wound healing response has also been observed in a clinical setting. It has been shown that exposure of human dermal fibroblasts and epidermal keratinocytes to silver sulfadiazine in vitro produced a significant reduction in cell proliferation and increased cytotoxicity [74, 75]. It has also been reported to inhibit collagen synthesis in human rheumatoid synovial cells in culture [76].
14
Novel vaccines were developed as they do not suffer the problem of resistance instead work to enhance the body's natural defenses. Nevertheless, it is possible that new strains may evolve that escape immunity induced by vaccines. Vaccines also do not guarantee complete protection from a disease. This may be due to a lowered immunity in general (diabetes, steroid use, HIV infection) or because the host's immune system does not have B-cells capable of generating antibodies to that antigen.
Bacteriophage therapy has been recently studied as an emerging alternative to antibiotics [77]. Phages were found to possess several potential disadvantages when compared with antibiotics. Although they are specific for the target bacterium, which reduces the disruption to the natural flora of the host, the need for isolation of a phage specific to the target bacteria leads to the increased costs and development time. Also, phages that are injected into the bloodstream are immunogenic [78]. Some of them are cleared very quickly and, after a certain period, antibodies against the phages are produced by the body [78]. Funding for phage therapy research and clinical trials is also generally insufficient and difficult to obtain, since it is a lengthy and complex process to patent bacteriophage products. Phage therapy has been used publicly for many years; therefore, the technique itself would not be patentable. Additionally, it is unlikely that it would be possible to patent individual phage, and the wide range of phage targeting each species of bacteria would mean that isolating a range of unique phage would be relatively straightforward.
15
2.3. Antimicrobial proteins and peptides
Use of proteins and peptides as antimicrobial agents is inspired from nature and has recently attracted much attention as an antibiotic-free approach to treat bacterial infections [79-81]. Naturally occurring antimicrobial peptides and enzymes have are characterized by higher activity, broad antimicrobial spectrum, and the low rate of resistance to these agents by bacteria. They may be expressed constitutively or sometimes inducibly in response to certain specific pathogenic challenges [82].
The current promising candidates of antibacterial enzymes are functionally classified as either hydrolases or oxidoreductases [83]. Hydrolases are bacteriolytic enzymes that function by degrading key structural components of the cell walls of bacteria. Peptidoglycan is the major structural component of bacterial cell walls.
Degradation of bacterial cell walls results in lysis due to rupture of osmotic balance in the cell. These enzymes are further grouped into three different classes –1) N- acetylhexosaminidases that catalyze the cleavage of the β (l-4)-glucosidic linkages in the carbohydrate backbone of the peptidoglycan, 2) N-acetylmuramyl-L-alanine amidases that catalyze the cleavage between the carbohydrate moiety and the peptide moiety of the peptidoglycan, and 3) Endopeptidases that hydrolyze the peptide crosslinks of the peptidoglycan [84]. Lysostaphin is an antibacterial enzyme which specifically cleaves cross-linked pentaglycine bridges in the peptidoglycan of S. aureus, thereby hydrolyzing the cell wall and lysing the bacteria.
Oxidoreductases on the other hand do not possess any antimicrobial activity by themselves. However they exert their effect through the generation of reactive species that are known to be toxic to bacteria. Glucose Oxidase, Lactoperoxidase and
16
Myeloperoxidase are some well-studied oxidoreductase systems [83, 85, 86]. These enzymes (Peroxidases) are synthesized and secreted by ductal epithelial cells of the mammary gland and other exocrine glands. They also probably provide protection against microbial infection to the mucosal surfaces of eye, nose, mouth, trachiobronchial tree and intestinal tract [87]. Both salivary peroxidase and myeloperoxidase, the two peroxidase enzymes in human mixed saliva, have been purified but only for research purposes. Large- scale purification from human saliva or from human polymorphonuclear leukocytes is difficult and far too expensive for commercial purposes. Therefore, for commercial purposes bovine lactoperoxidase (purified from milk or colostrum) has been used because this enzyme is structurally and catalytically very close to human salivary peroxidase [87].
The lactoperoxidase system has been incorporated into commercial products such as tooth pastes and mouthwashes like Biotene®, BioXtra® and Oralbalance® [87]. It is also well understood that bacteriolytic enzymes usually offer protection against only a few bacteria, owing to their specificity. The oxidoredctuase systems may affect a broader spectrum of microorganisms but suffer from the drawback of being dependent on the availability of substrates for the generation of antimicrobial reaction products and toxicity towards eukaryotic cells.
2.3.1 Antimicrobial peptides (AMPs)
These are a group of short 10–40-residue polypeptides that are part of the innate immune system in many organisms including bacteria, insects, plants, and vertebrates[88].
In mammals, these peptides are present mainly in phagocytic cells and in mucosal epithelial cells. In addition to their antimicrobial role, antimicrobial peptides (AMPs) also
17 serve as important molecules involved in inflammation, immune activation, and wound healing [89-91].
In humans, three distinct groups of antimicrobial peptides are usually distinguished: defensins, cathelicidins, and histatins [92, 93]. Histatins will not be considered here because their primary function is antifungal, rather than antibacterial. A characteristic feature of defensins (molecular weight 3-5 kDa) is the presence of six cysteine residues forming three intramolecular disulfide bridges. Defensins are further subdivided into - and -defensins based on the distribution of the cysteines and linkages of disulfide bridges. Cathelicidins are a family of proteins containing an N-terminal cathelin domain and a C-terminal cationic antimicrobial domain that becomes active after cleavage [94]. Only one human cathelicidin is known. It is a cationic 18 kDa protein precursor, hCAP18. Its mature antibacterial peptide named LL-37 is liberated through cleavage by elastase and proteinase 3 [95]. LL-37 has a broad-spectrum antimicrobial activity against a variety of Gram-positive and Gram-negative bacterial, fungal, and viral pathogens.
Amphipathicity (presence of both hydrophilic and hydrophobic groups) and net charge are characteristics conserved among many antimicrobial peptides [88]. All antimicrobial peptides bear a large positive charge, which results in their preferential targeting to bacteria, rather than to eukaryotic cells. It is generally understood that the bactericidal activity of antimicrobial peptides is due to permeation of bacterial cell membranes [88, 96]. It is also likely that antimicrobial enzymes and peptides may use more than one mechanism of action, such as destabilization of the cell membrane combined with inhibition of one or more intracellular targets [97]. This “multitarget”
18 mechanism hypothesizes that highly charged cationic peptides and enzymes can bind to anionic molecules within the cytoplasm, such as nucleic acids or enzymes with ionic surfaces, thus interfering with processes in which these molecules are involved.
2.3.2 Synergy
Interactions between antimicrobial enzymes and peptides may be synergistic, additive, or antagonistic in nature. Because of their different modes of action, antimicrobial enzymes and peptides may act cooperatively by simultaneously attacking essential structures in microorganisms [98]. Synergy among individual peptides, was first observed with frog peptides, including members of the dermaseptin family [99] and the a- helical peptides magainin and PGLa [100], and between b-defensins and the cationic protein BPI [101].
The synthetic peptide LL-37 demonstrated antibacterial activity against a number of Gram-negative and Gram-positive organisms including P. aeruginosa and was synergistic with lactoferrin and lysozyme [102]. A synergistic antibacterial effect between
LL-37, β-defensins, and lysozyme was initially described by Nagaoka et al. [103]and later comprehensively investigated by Chen et al [104]. They observed that the antibacterial agents- hBDs (human beta defensins), LL-37 and lysozyme exhibit antimicrobial activities synergistically against invading microorganisms, in particular S. aureus that frequently colonizes the skin [104]. The synergistic effect found was strong enough to speculate that defensins may not be able to act as antibacterial molecules by themselves, but only in synergism with cathelicidins [103]. In another study, synergism in various combinations of six antimicrobial factors, including lysozyme, lactoferrin, LL-37, HBD-1, HBD-2, and
19 secretory leukocyte protease inhibitor (SLPI), had been evaluated by Welsh and co- authors [105]. In this work, synergy of lysozyme with LL-37 has been confirmed, while no synergy has been observed in the cases of lysozyme combined with either HBD-1 or
HBD-2.
Thus synergistic interactions appear to be important evolutionary survival mechanisms that can be cleverly manipulated in the design of antibacterial combinations.
Combining multi-targeting functions through different peptides and proteins simultaneously ensures their antimicrobial action to a wide range of microorganisms. Due to widespread antibiotic resistance, such multiple combinations, similar to that of an antibiotic cocktail, are often required to overcome infectious diseases. Implementing such synergistic principles into the design of next generation therapeutics could prove to be an important determinant in the effectiveness of using an antibiotic-free approach.
2.3.3 Antibiofilm activity
The biofilm matrix is composed of an assortment of cells and various extracellular polymers of different polysaccharides such as glucose, dextrans, cellulose, alginate etc. and also of some glycoproteins [40]. Antimicrobial enzymes have been used in the inhibition of biofilm formation. However due to the heterogeneity of these polysaccharides, a combination of enzymes is necessary to achieve a synergistic effect for biofilm inhibition. For instance, Pectinex UltraSP (a mixture of enzymes containing cellulose, pectinase etc.) has been used for enzymatic degradation of S. aureus, P. aeruginosa, P. fluorescens and S. epidermis biofilms with S. aureus being the most sensitive [106].
20
Some of these polysaccharide-hydrolyzing enzymes are only known for dispersion and inhibition of biofilms but do not have a significant bactericidal activity. For example
DispersinB (patented by Kane Biotech Inc.), a naturally occurring N- acetylglucosaminidase isolated from A. actinomycetemcomitan is a highly active and stable glycoside hydrolase that specifically glycosidic linkages of poly-b-1, 6-N- acetylglucosamine in polysaccharide adhesins of bacteria needed for biofilm formation
[107, 108]. As DispersinB can only inhibit or disperse bacterial biofilms without affecting their growth, it would be necessary to combine it with another antimicrobial in order to inhibit the growth and proliferation of biofilm-embedded bacteria [107, 109]. When used in combination with glucose oxidase; it increased the sensitivity of S. epidermidis biofilm to glucose oxidase [110]. Johansen et al also similarly showed that combining oxidoreductases such as glucose oxidase and lactoperoxidase with polysaccharide- hydrolyzing enzymes resulted in bactericidal activity as well as removal of the biofilm thus citing another example of a synergistic effect [106]. Apart from the exopolysaccharide layer, extracellular DNA has also been as a target used to inhibit biofilm formation. For e.g Bovine DNase1 was shown to suppress P. aeruginosa,
Streptococcus intermedius and S. mutans biofilm formation [111]. The antibiofilm activity of these DNases involves degradation of extracellular DNA, which is an essential component of biofilms.
21
Peptide and proteins Source Proposed antibacterial mechanism References
Lysozyme Human, Chicken Enzymatic lysis of peptidoglycan layer; Non enzymatic mode of action [112, 113] Sequestering iron essential for microbial respiration ; Directly microbicidal due to its N-terminal cationic Lactoferrin Human, Bovine fragment “lactoferricin.” [114, 115]
Mammals; Cell membrane and intracellular targets, inhibits α and β defensins macromolecular synthesis [116, 117] analogues in insects and fungi Membrane permeabilization LL-37 Human [118, 119] Inhibits macromolecular synthesis, Ca2_-calmodulin Indolicidin Bovine interaction [120, 121] Membrane permeabilization; binds to lipid 2 of peptidoglycan layer; reduction of ATP; collapse of Nisin Bacteria pH gradient [122, 123]
Magainin Frog Membrane permeabilization [124, 125]
Very potent, membrane permeabilization Protegrin Human, Porcine [126] Cecropin A Membrane destabilizing Silk moth [127, 128]
Table 2.1. below shows list of some of the antibacterial peptides and proteins
2.3.4 Challenges associated with application of antibacterial proteins and peptides
Overall great enthusiasm exists regarding the prospect of developing antimicrobial proteins and peptides as a new generation of „antibacterial agents‟ for treatment of implant-associated infections. In spite of this, use of enzymes and peptides in a clinical setting is associated with several potential drawbacks, which includes 1) high
22 manufacturing cost, especially of peptides, 2) short half life in vivo, 3) loss of activity in physiological conditions (especially at high ionic strength, which may interfere with their charge-directed targeting and mechanism of action), 4) unwanted systemic reactions
(aggregation, immunoreactivity), 5) interference with normal bacterial flora and even 6) site-specific delivery that may arise when attempts are made to use them for targeted delivery as antibacterial agents. It is also known that small molecules (lesser than 70kDa) such as antimicrobial enzymes and peptides have a relatively short half-life in blood after intravenous administration. Their rapid clearance from circulation may reduce its efficacy as an antimicrobial. Also, some of the enzymes such as lysostaphin and DispersinB are derived from bacterial species and could potentially show immunogenicity in humans which further stimulates its clearance from the blood stream. This is mainly due to the induction of a specific humoral response [129, 130]. Antibodies from the host might neutralize these enzymes and compromise the efficacy of the drug and there could also be other effects related to cross-reactivity with other autologous proteins. Thus, their short circulating half-life cannot be effectively countered by increasing the amount or frequency of dosage. There are also a number of MMPs (Matrix Metalloproteinases), Serine proteases etc. that could also result in protein degradation through hydrolysis. More expensive strategies such as synthesis of recombinant protein that is resistant to proteolytic degradation have been reported [131].
Several research-related issues also must be resolved for better understanding of the fundamental aspects of the processes associated with clinical applications- 1)
Development of standardized techniques to assess the activity of enzymes and peptides 2)
Targeting of agents to the site of action. 3) Addressing toxicity in more detail, for e.g. the
23 positive charge of AMPs may result in their strong interactions with anionic components of both bacterial and host cells. 4) Understanding of the role and expression of antimicrobial enzymes and peptides in both health and disease. Focusing on these issues is expected to lay the foundation for the future applications of enzymes and peptides as highly effective antibacterial agents for treatment of device-related infections.
2.4. Enzyme immobilization
Immobilization of enzymes on surfaces is of great importance for a wide range of biomedical applications because of their recyclability, decrease in systemic toxicity, carrier specificity and improvement in enzyme stability [132]. Enzymes are highly specific and efficient biocatalysts with a wide range of biotechnological and biomedical applications [133, 134]. There are however quite of a few issues that need to be addressed towards their application for treatment of bacterial infections. These include substrate and/ or site specificity, systemic toxicity and loss of enzymatic activity in vivo due to low stability. In order to overcome these problems, numerous efforts have been devoted to the development of several immobilization procedures. Presently, immobilized proteins/enzymes are used routinely in the medical field, such as in the diagnosis and treatment of various diseases. For example, immobilized antibodies, receptors, or enzymes are used in biosensors and ELISA for the detection of various bioactive substances in the diagnosis of disease states; antibody directed drug delivery using colloidal systems [135-
137].
Immobilization can be defined as the attachment of molecules to a surface resulting in reduction or loss of mobility. In fact sometimes, immobilization may lead to
24 partial or complete loss of protein activity, due to random orientation and structural deformation [138]. There is a multipoint attachment between enzyme molecules and the surface that reduces unfolding, and hence improves stability [139]. In order to retain biological activity, enzymes should preferably be attached onto surfaces without significantly affecting their conformation and function.
Generally, the choice of a suitable immobilization strategy is determined by the physicochemical and chemical properties of both surface and enzyme. The stability of an immobilized enzyme is dictated by many factors such as the number of bonds formed between the enzyme and carrier, the nature of the bonds (i.e. covalent, non-covalent or different types of covalent or non-covalent bonds), the degree of confinement of enzyme molecules in/on the carrier, the microenvironment of the enzyme and carrier, and the immobilization conditions [138]. For e.g. nonporous materials, to which enzymes are attached to the surfaces, are subject to minimum diffusion limitation while enzyme loading per unit mass of support is usually low. On the other hand, porous materials can afford high enzyme loading, but suffer a much greater diffusional limitation of substrate.
Many surface immobilization techniques have been developed in the past years, which are mainly based on the following three mechanisms: physical, covalent, and bioaffinity immobilization [138]. It is unlikely that one immobilization technique will be considered optimal for all proteins. Also, the most suitable strategy for protein immobilization is also particularly dependent upon the biomedical application. Hence, several unsolved challenges are involved in immobilization techniques. Correspondingly, many methods have been developed to improve the enzyme stability during immobilization. Engineering the microenvironment of the enzyme molecules has been
25 increasingly used to improve enzyme stability. For e.g. in the case of covalent enzyme immobilization, alteration of microenvironment can be easily achieved by quenching the remaining excess active functionalities using blocking agents [140], which can be simply classified into two groups: small molecules (e.g. amino acids or other amines), and macromolecules (e.g. bovine serum albumin, gelatin, polyethyleneimine (PEI) [141] , and polyethylene glycol (PEG)). These blocking agents (or quenching agents) can easily react with the active functionalities, such as epoxy rings or aldehyde groups, of the commonly used carriers without the use of additional activating chemistry.
Other methods of immobilization include entrapment via inclusion of an enzyme in a polymer network (gel lattice) such as an organic polymer or a silica sol-gel, or a membrane device such as a hollow fiber or a microcapsule. Immobilization in silica sol gels prepared by hydrolytic polymerization of tetraethoxysilane, in the presence of the enzyme [142] and has been used for the immobilization of a wide variety of enzymes
[143]. Also, it has been shown that cross-linking of dissolved enzymes via reaction of surface NH2 groups with a bifunctional chemical cross-linker, such as glutaraldehyde, results in insoluble cross-linked enzymes with retention of catalytic activity [144].
However, this method of producing cross-linked enzymes had several drawbacks, such as low activity retention, poor reproducibility, low mechanical stability, and difficulties in handling the immobilized enzyme. Some methods of immobilization are described below.
2.4.1. Physical adsorption
The type of intermolecular forces that take part in the adsorption process will depend on the particular protein and surface involved. Protein adsorption on surfaces is predominantly governed by electrostatic, hydrophobic interactions and to certain extent
26 even hydrogen bonding. The resulting layer is likely to be heterogeneous and randomly oriented, since each protein molecule can form many contacts in different orientations for minimizing repulsive interactions with the substrate and previously adsorbed proteins
[145].
The arrival of protein at the interface is assumed to be driven solely by diffusion processes, which are dependent on bulk concentration and diffusion coefficient [145]. The particular surface chemistry, intermolecular forces between adsorbed molecules, solvent– solvent interactions, topology and morphology of the protein-adsorbent interface also further dictates the adsorption kinetics. Protein adsorption process includes the redistribution of charged groups in the interfacial layer, hydrophobic interactions between protein and material‟s surface, and structural rearrangements of protein molecules [145].
Drawbacks of the adsorption are random orientation resulting in significant loss of activity and weak attachment followed by subsequent protein desorption from the surface.
The structure of a protein is closely related to its function therefore unspecific protein adsorption and surface-induced conformational changes are important issues in biocompatibility of materials [146]
2.4.2. Covalent Immobilization
This method confers the immobilized enzyme with greater stability against environmental changes such as pH and temperature, as well as higher selectivity toward the substrates. However, the coupling protocols could involve complicated synthetic procedures, which are often carried out under harsh experimental conditions, and with use of toxic and expensive reagents. Furthermore, functional groups of the enzyme are used to
27 link the solid supports, resulting in a significant loss in enzymatic activity [147]. Covalent bonds are mostly formed between side-chain-exposed functional groups of proteins with suitably modified supports, resulting in an irreversible binding and producing a high surface coverage [148]. The functional groups on the support are generated by chemical treatment, and several pretreated surfaces are already commercially available.
Covalent attachment is permanent, leaving no unbound material after clean-up. It may prevent elution of bound protein during storage, thus increasing shelf life. Covalent coupling on surfaces enables a uniform coating procedure [149, 150]. As a consequence of having reactive groups over the entire surface of the material, it is possible to completely cover the surface with protein. Protein coverage is also more easily controlled by covalent coupling, especially when the desired quantity of adsorbed protein is low. Achieving the correct spatial orientation for the bound protein can be difficult via physical adsorption.
Covalent attachment, on the other hand, can orient the molecule properly, if the correct coupling chemistry is chosen, improving the activity of the bound proteins and resulting in lower reagent consumption [149, 150].
Because active enzymes necessarily have very flexible conformation, a spacer between the enzyme and the carrier surface is often used to improve the enzyme activity expression[151]. The obtained immobilized enzymes with spacers could have higher activity due to the favorable environment created by the hydrophilic spacer. Most probably, the presence of the spacer is able to endow the enzyme more conformational flexibility, which is usually a prerequisite for higher activity. Controlled modification of the enzyme molecules with the use of a bi-functional cross-linker have also been recently used to adjust the number of bonds and the spacer formed between the carrier and enzyme
28
.To preserve the activity of biomolecules upon immobilization, several techniques have been directed towards site-specific covalent immobilization [152, 153]. For antibodies, the carbohydrate moieties located in the Fc region were targeted to form aldehyde groups, using periodate for site-specific coupling to solid supports [154, 155].
2.4.3. Bioaffinity Immobilization
Biochemical affinity reactions are also used for immobilization of proteins, providing an important advantage over other surface immobilization techniques.
Moreover, not only oriented and homogeneous attachment is obtained, but it is also possible to detach proteins and make repeated use of the same surface. An example is described below.
Avidin-Biotin interaction
This approach takes advantage of one of the strongest noncovalent bonds (Kd =
1015 M-1) and allowing the use of harsh conditions during biochemical assays [156]. The specificity of the interaction permits uniformly oriented protein immobilization [157].
Avidin is a tetrameric glycoprotein soluble in aqueous solutions and stable over wide pH and temperature ranges. It can bind up to four molecules of biotin. This bond formation is very rapid and unaffected by pH, temperature, organic solvents, enzymatic proteolysis, and other denaturing agents [157].
Streptavidin is a closely related tetrameric protein, with similar affinity to biotin, but differing in other aspects, such as molecular weight, amino acid composition, and pI.
The properties of both avidin and streptavidin have been improved using chemical and recombinant methods providing enhanced stability and/or controlled biotin binding [157].
Biotin or vitamin H is a naturally occurring vitamin found in all living cells. Only the
29 bicyclic ring is required to be intact for the interaction with avidin; the carboxyl group on the valeric acid side chain is not involved and can be modified to generate biotinylation reagents used for conjugation with proteins [158]. Since biotin is a small molecule, its conjugation to macromolecules does not affect conformation, size, or functionality [158].
Biotinylation reagents can be classified depending on their reactivity toward different functional groups. The NHS ester of biotin is the most commonly used biotinylation reagent to target amine groups [159].
Both biotin and avidin/streptavidin may be attached to a variety of substrates.
Direct immobilization of avidin occurs either via adsorption or via covalent coupling. It has been reported that covalent attachment of avidin onto carboxyl-derivatized polymers produces a higher coating density than immobilization via a biotinylated interlayer [160].
However, coupling via carbodiimide chemistry may affect its binding activity. A typical multilayer is composed by biotin directly immobilized and avidin creating a secondary layer for binding biotinylated molecules. This approach is generally preferred due to the higher organization (biotin/avidin/biotin) obtained in comparison to that of the direct immobilization of avidin [161].
30
Figure 2.4. Schematic of the Streptavidin-biotin system: Binding of biomolecules to the particle surface: (Source: http://www.microparticles.de/news_2005.html- accessed - 20/03/2011).
2.4.4. Surface modification
Spontaneous adsorption of proteins from biological fluids onto synthetic materials such as biomaterials and biomedical devices may induce undesirable reactions of the body to the foreign materials, such as immune responses, blood coagulation, or bacterial adhesion [162, 163]. Due to the diversity of the interactions between proteins and surfaces, a preferred strategy for blocking the adsorption of proteins is to immobilize polymers in the form of well-solvated brushes (e.g., poly (ethylene glycol), PEG) on the surface.
Pegylation of a surface can be achieved by adsorbing PEG-containing block copolymers such as polyethylene oxide/polypropylene oxide [164] or poly(d,l-lactide)/ polyethylene oxide [160] on the surface, surface grafting using established strategies
[165]depositing PEG-like coatings using a radiofrequency glow discharge technique [166] or forming self-assembled monolayers. The polymer layer shields the surface, introducing a high activation barrier for the proteins to adsorb. Poly (ethylene glycol) (PEG) has been shown to successfully confer protein resistance to a variety of surfaces [167, 168]. Both
PEG chain length and surface coverage have been demonstrated to play a key role in imparting protein resistance, with PEG chains of typically 1-10 kDa molecular weight providing protein resistance. The protein resistance by PEG chains has been associated with two main mechanisms, steric repulsion and a hydration or water structuring layer
[167].
31
Attachment via PEG spacers also minimizes lateral interactions of immobilized protein with the surface [169, 170]. The spacer groups are thought to permit a degree of freedom to the reagent moiety separating it from the particle surface, thereby lending enhanced specificity [167]. In particular, nanoparticles whose surfaces were modified by the incorporation of poly(ethylene glycol) (PEG) during nanoparticle formulation either through covalent attachment of PEG to surface functional groups or through physical adsorption of PEG to the surface, have show a decreased uptake by phagocytic cells and an increased circulation time to effectively target diseased cells [171, 172]. PEG molecules on the surface of a polymeric nanoparticle can reduce the adsorption of opsonins and other serum proteins [172]. PEGylated polycyanoacrylate particles transferred more effectively across the blood brain barrier than with poloxamine coating
[173].PEG density and configuration were shown to be important for this [174].
Additives such as polyethylene glycol (PEG), polyvinyl alcohol and albumin, can have a stabilizing effect even on sol-gel entrapped enzymes. For example, Zanin and co- workers [175] compared three different methods – physical binding, covalent attachment and gel entrapment, in the presence and absence of PEG 1450 – for the immobilization of
Candida rugosa lipase. Immobilization yields varied from 3 to 32%, the most active biocatalyst resulting from the encapsulation in the presence of PEG.
2.5. Nanotechnology for enzyme immoblization
Nanotechnology also seems to be as a promising alternative to overcome the problems of the administration of peptides and proteins and also stabilizing these biomolecules before targeted delivery. Various nanostructures, generally providing a large
32 surface area for the immobilization of enzyme molecules, have been actively developed for enzyme stabilization. The coupling of biomolecules and materials at the nanoscale has the potential to revolutionize many fields of science and technology potentially having a very significant impact on biomedical technology. Both nanomaterials and biomolecules lie in the same nanometer length scale and can thus complement each other in hybrid systems. Protein based nanostructures are expected to offer some additional advantages and play a key role in the development of multifunctional materials and devices for nanotechnological applications. [176-179].
Many methods have been developed to incorporate enzymes into nanostructures.
Surface attachment was employed in many of the earlier studies and is often the method of choice [180-182]. Another way to develop nanoscale biocatalysts is to entrap enzymes within nanopores. Mesoporous silica gels have been used to host enzymes through physical adsorption [183] or chemical binding. Using carefully designed synthetic routes, enzymes were also entrapped within the cores of discrete polymeric articles. More sophisticated structures, such as porous materials hosting enzyme-carrying nanoparticles and cross-linked enzyme aggregates [185], have been developed by applying enzyme modification and fabrication procedures. Thus the functionalization and modification of nanomaterials using enzymes have led towards research and development for biomedical applications.
However there are drawbacks to the handling of nanomaterials, which presents certain health and environmental concerns. Carbon nanotubes, which exist in powder form when in a pure state, share these concerns when used as enzyme supports due to toxicity
[180-182]. Some of these problems can be overcome by using one-dimensional
33 nanomaterials, such as polymeric nanofibers [186]. The surface area: volume ratio of nanofibers is also high, representing two-thirds that of nanoparticles of the same diameter when considering equivalent amounts of material. Nanofibers, however, have the benefit of being much easier to produce and handle. They can be applied in the form of coils, sheets or dispersed fibers and can also be attached to the surface of other materials or blended with them, thus offering very flexible design of reactors [186].
In the case of surface attachment, smaller particles can provide a larger surface area for the attachment of enzymes, leading to higher enzyme loading per unit mass of particles [187]. A growing interest has been shown in using nanoparticles as carriers for enzyme immobilization [187-190]. The effective enzyme loading on nanoparticles could be achieved up to 10% wt due to a large surface area per unit mass of nanoparticles [188].
Figure 2.5. Nanotechnology for enzyme immobilization (a) Nanoparticles with surface-attached enzymes. (b) Nanofibers carrying enzymes. (c) Nanoporous matrix with entrapped enzymes. (d) Carbon nanotube–enzyme hybrid materials –
Figure taken from Wang et al; 2006 [191].
34
2.6. Nanoparticles
The use of nanoparticles as enzyme supports was first reported in the late 1980s
[192, 193]. Therefore, nanoparticles may appear as alternative carriers considering the design of more sophisticated systems including multifunctional types of device.
Nanoparticles are of great scientific interest as they are effectively a bridge between bulk materials and atomic or molecular structures. Size-dependent properties are observed in case of nanoparticles when compared to the bulk material. Since they have a very high surface area to volume ratio it provides a tremendous driving force for diffusion, especially at elevated temperatures.
Advantages of using nanoparticles over other „nanomaterials‟ and formulations are that they (e.g. polymer nanopaticles) are cheaper materials than lipids (e.g liposomes), and that polymers offer wider chemical engineering solutions. In addition, the stability of polymer particles can be much better controlled than the stability of liposomes upon slight modifications of the formulation. Intrinsic properties of nanoparticles, such as size, surface charge density, surface chemistry etc can also be used to add functionality to their conjugates with biomolecules.
For example, drugs, proteins and peptides can be incorporated into nanospheres that are composed of a biodegradable polymer, and this allows for the timed release of the drug as the nanospheres degrade [194] The circumstances that cause the particle to degrade can be adjusted by varying the nature of the chemical bonding within the particle.
For example, when acid-labile bonds are used, the particles degrade in acidic microenvironments, such as would exist in tumour cells or around a site of inflammation
[195], and so this approach allows site-specific delivery. In other recent studies, polymeric
35 nanoparticles were labeled on their outer surfaces with a viral peptide sequence that promotes the permeation of substances through cell membranes [196]. These peptide- derivatized nanoparticles passed through cell membranes, and were incorporated into living cells at much higher levels than nanoparticles without the surface-bound peptide.
Similarly Gold nanoparticles (20 nm) modified with shells of bovine serum albumin (BSA) were conjugated to various cellular targeting peptides to provide functional nanoparticles that penetrate the biological membrane and target the nuclei [197,
198]. Hybrids of silver nanoparticles with amphiphilic macromolecules exhibit antimicrobial properties and are used to treat bacterial infections [198]. Covalent coupling of proteins to metal nanoparticles is performed using appropriate capping agents such as citrate and thiol groups. This method has been applied in coating of colloidal gold with thiol containing proteins such as immunoglobulins and serine albumins which have cysteine residues that are accessible for coupling [199].
2.6.1. Polymer nanoparticles
The term `polymer colloid' refers to a suspension or dispersion of polymeric microspheres having a diameter in the order of sub-micron to several microns. The dispersion medium is generally water. A dispersion whose medium is not water is referred to as a nonaqueous dispersion. `Latex' originally means a dispersion of microspheres from natural rubber. The stability of the polymer latex itself is decided by the contribution of three factors: electrostatic repulsive forces, van der Waals' attractive forces and steric repulsive forces among the particles [200].
Polymer colloids have a low viscosity and high fluidity compared to solutions containing the same amount of solid. The viscosity of polymer colloids is a universal
36 function of apparent volume fraction of the microspheres. The apparent volume fraction of microspheres can be changed by environmental conditions such as pH and temperature for some polymer colloids [201]. In dispersion, the microspheres can move macroscopically through the medium by gravity, electric field, etc and microscopically via Brownian motion. These movements keep a fresh interface between the microspheres and the medium. Particles having soft layers on the surface allow water to penetrate the layer and meet less resistance when they move through water [202]. Polymer particle dispersions are an alternative to be able to offer a large surface and they can be used as the carrier if easily recovered from the medium.
Polymer latexes, usually with antibodies or antigens attached to their surfaces, can be used in the diagnostic tests. There are many factors that dictate the use of latex particles than the use of any other solid support for agglutination reactions. These include monodispersity (for detection of agglutination using light scattering techniques), colloidal stability and hydrophobicity of the particles. Biological molecules adsorb strongly to hydrophobic surfaces and this can be controlled by time, temperature, pH and protein concentration [149]. An antibody is an asymmetric molecule and its antigen-binding fragment (Fab) is located in two arms of Y-shaped molecules. Antibodies must be immobilized in a manner in which the Fab fragment does not have any steric hindrance and can bind to the directed antigen. When an antibody is immobilized in the opposite orientation, it causes not only lowering of reactivity but also nonspecific agglutination, since the fragment Fc of an antibody adsorbs to certain proteins [149]. Furthermore, surface activity could be enhanced if the antibodies are covalently coupled, by reducing the rearrangement of the protein molecules during and after adsorption [203-205]. Specific
37 covalency between protein and functionalized latex is achieved by treatment of protein– latex system with surfactants under appropriate conditions able to remove all the physically adsorbed protein from the particle surface [150].
2.6.2. Protein-nanoparticle conjugates
In recent years, much attention has been paid to the use of enzymes conjugated to polymer colloids in biomedical applications [206-208]. Such colloids are easy to produce and can be prepared with a high degree of monodispersity [209-211]. Colloidal nanoparticles provide an almost ideal remedy to the issues encountered in the optimization of immobilized enzymes: minimum diffusional limitation, maximum surface area per unit mass, and high enzyme loading [187]. The attachment of molecules to latex particles can be achieved through physical adsorption or covalent coupling.
Polymer engineering has facilitated the synthesis of latex particles with surface reactive groups such as carboxylate, aldehyde, chloromethyl, and hydroxyl groups to enable covalent coupling of protein molecules to the particles[166, 212, 213] Some of these groups must be activated prior to protein immobilization. A number of special linkers can be used to convert one surface functional group on a microsphere to another.
For example, amino microspheres can be converted to carboxylic particles by reacting with succinic anhydride [214]. Carboxylic particles can be converted to amino microspheres through water-soluble carbodiimide-mediated attachment of a diamine
[215].
38
Fig 2.6. Commonly used covalent coupling procedures for peptides and proteins to the surface of nanoparticles with reactive surface groups.
There are also major drawbacks of any immobilization protocol based on the use of highly reactive and unstable intermediates. The balance between covalent coupling and unproductive side reactions depends to some extent on the activation reaction conditions.
The ionic composition of the reaction, the pH, and the buffer type can greatly enhance or inhibit the total binding of protein to particles. Sufficient amounts of functional groups should be present to provide adequate coupling of the protein. The covalent attachment of the IgG molecule to carboxylated particles improves the immunoreactivity of antibodies when compared with physical adsorption and also maintaining its immunoreactivity after long periods of storage [150].
2.6.3. Stabilization using blocking agents and surfactants
There also other techniques in order to improve stability and reduce non specificity of the protein-nanoparticle conjugates. The colloidal suspension can be treated with either a second protein or a surfactant that can act as a stabilizer by covering all the unoccupied
39 parts of the surface and thereby reducing non specificity [216]. Bovine serum albumin
(BSA), a globular protein has been commonly used to improve the colloidal stability of protein coated microspheres. The BSA molecule is a highly-charged protein at physiological pH. It helps in the electrostatic stabilization of the protein-nanoparticle conjugate [217, 218]. Moreover, BSA is easily obtainable in significant amounts, which is why it is commercially available at reasonably low prices.
There are two different methods for preparing latex–antibody complexes with coadsorbed BSA: (1) sequential adsorption where in a first step the antibody adsorption is carried out and after centrifugation the conjugate is resuspended in a solution of BSA at constant concentration [219]; (2) competitive adsorption whereby the adsorption of both proteins (including BSA) occurs in a single step [217, 218]. For competitive adsorption, the method becomes more complicated because of the two types of proteins in the dispersion medium. Peula et al. observed that the colloidal stabilization of IgG-covered particles appears when the coverage of co-adsorbed BSA is high and at pH 7 and 9, and is not stable at pH 5 (pI of albumin) [217]. It was also indicated from the same study that it is necessary to find the adequate equilibrium between the amount of IgG that produces a good immunological response and the amount of BSA responsible for the colloidal stability. These ways of stabilizing antibody–latex particles may have some disadvantages, e.g., only antibody–latex complexes with low antibody coverage can be stabilized or co- adsorption with inactive proteins may involve partial or complete displacement of the preadsorbed molecules of antibody.
The use of surfactants (e.g. Tween) as stabilizer molecules can also have also have desired effects on nonspecific agglutination. They mainly stabilize protein-coated particles
40 by means of steric forces, although electrostatic repulsions can also be generated by ionic surfactants. The surfactant concentration requires careful optimization because excess surfactant could also inhibit interaction with the desired antigen/substrate. However, this strategy to preserve the colloidal stability of particles presents some disadvantages, since such molecules can also desorb the previously adsorbed protein [220].
2.7. Enzyme-nanoparticle conjugates as antibacterials – Targeted delivery
The stratum corneum layer of the epidermis is a strict barrier allowing limited penetration of particulate materials. This aspect has potential to serve as a novel route for drug delivery and has attracted enormous pharmaceutical research interests [221]. While intact skin is known to act as a barrier to nanoparticles, skin wounds may give rise to easier translocations. Dressings and bandages embedded or coated with antibacterial protein-nanoparticle conjugates could be potentially used for prevention of sepsis of severe skin wounds like burns. Close contact may allow nanoparticle-conjugates to penetrate through compromised skin barrier and gain access to the dermal capillaries.
Recently Solid lipid nanoparticles (SLN) are gaining increasing importance as pharmaceutical formulations [222, 223]. Due to their small particle size, chemistry and consequent high surface area, they possess strong adhesive properties. Incorporation of antibacterials into the solid lipid matrix offers protection against chemical degradation of the active compound, as well as allowing either immediate or sustained release, depending upon the application [224]. Sustained release is important with antibacterials that are irritating at high concentrations or when a supply to the skin over a prolonged period of time is desired, whereas immediate release can be useful to improve penetration.
41
Antibacterial activity of nisin-loaded polylactide nanoparticles against
Lactobacillus delbrueckeii has also been evaluated [225]. In most of these cases, the role of nanoparticles was to provide a sustained release of the loaded drug; however, targeted delivery of antibiotics to infected macrophages and to liver has been utilized using nanoparticles and liposomes [226] . The above targeting was made possible because of the increased uptake of polymeric nanoparticles by reticuloendothelial system.
More recently, specific targeted delivery of polymeric nanoparticles either unloaded or loaded by an antimicrobial drug, to bacteria had been studied. The antimicrobial effect of insoluble cross-linked quaternary ammonium polyethylenimine
(PEI) nanoparticles incorporated at 1% w/w in a resin composite was assayed and resulted in complete inhibition of growth S. aureus, S. epidermidis, P. aeruginosa, E. faecalis, and
E. coli (p < 0.0001) [227]. Mannosylated or galactosylated polystyrene latex nanoparticles were shown to result in agglutination of mutant strains of E. Coli overexpressing mannose and galactose surface receptors, respectively [228, 229]. These studies have been thus far restricted to model E. coli mutants; however, the authors envision development of nanoparticles coated by oligosaccharides that will specifically recognize desired bacterial strains and kill them by sticking to bacterial cell walls and causing bacterial agglutination.
Effect of antibody-directed targeting to Gram-positive Listeria Monocytogenes has recently been studied by the same group for lysozyme attached to polystyrene latex nanoparticles simultaneously with anti- L. Monocytogenes antibody [230]. The authors observed significantly (p<0.05) higher antibacterial activity for nanoparticle conjugates containing both lysozyme and the antibody, than for free lysozyme and enzyme- nanoparticle conjugates in the absence of the antibody at higher concentrations.
42
2.8. Enzymes as antibacterial coatings on medical devices and implants
Bacterial adherence to an implant is a known precursor to prosthetic infection
[231, 232]. Antimicrobial coating of medical devices/implants has recently emerged as a potentially effective method for preventing infections [233-236]. Several attempts have been undertaken to immobilize enzymes and peptides on biomedical devices and implants to prevent device-related infections. For a catheter surface, the coating may be retained by either physical adsorption or covalent attachment of the enzyme.
Wilcox et.al immobilized synthetic antimicrobial peptide melimin on contact lenses by physical adsorption, and observed significantly decreased bacterial colonization
[237]. In another embodiment, antimicrobial protein lactoferrin was also adsorbed on contact lenses by the same group of authors and was also found to prevent bacterial colonization [238]. Balaban et.al found that adsorbed dermaseptin to Dacron vascular grafts showed resistant to bacterial infection and demonstrated in vivo efficacy in a rat model [239]. Langer et al covalently attached a hybrid synthetic AMP, Cecropin-Melittin, to model surfaces in a highly ordered manner, and reported enhanced antibacterial activity in comparison to physically adsorbed and randomly immobilized peptides. Shah et.al performed demonstrated significant antimicrobial activity against S. aureus for intravenous catheters coated by physically adsorbed lysostaphin, and observed high efficacy [240]. DispersinB, a naturally occurring enzyme in bacteria, Actinobacillus actinomycetemcomitans, coated on polymer well plates, rods and catheters has been shown to prevent the formation of Staphylococcus epidermis biofilms [241].
Despite such growing interest, a combination of enzymes coated on implant surfaces to prevent bacterial colonization and subsequent biofilm formation has not been
43 studied. DispersinB, is an enzyme that is specifically active against Staph species, however when used alone it can only inhibit and disperse biofilms, but not in killing of the bacteria, which could still persist in the blood-stream to cause infection. So, it would be necessary to combine enzymes such as DispersinB (antibiofilm) with antimicrobial enzymes and peptides such as lysozyme, lysostaphin, LL-37 etc. to not only prevent colonization but also lyse bacteria embedded in biofilms. The application of enzymes and peptides has been shown to work synergistically against a broad spectrum of bacteria from various studies [102-104]. But a systematic study on combination of different enzymes and/or peptides coated on surfaces at different coating concentrations working synergistically has not been performed. Also, aspects of enzyme leaching, cytotoxicity and long term durability on devices for antimicrobial applications have not been studied in detail. With this in mind, the final goal of our project aimed to address through nanotechnology, all issues involved in coating of enzymes (alone or in combination) on catheters by evaluating their in vitro performance through simple model studies.
44
CHAPTER 3
SPECIFIC AIMS AND SIGNIFICANCE
3.1. Clinical significance
Central venous catheters are commonly used modality in hospitals and clinics for administration of fluids, chemotherapeutic agents, central venous pressure monitoring, extracorporeal treatment regimen etc. thus serving a vital role as both a therapeutic and diagnostic tool for monitoring of health in critically ill patients. Nearly 5 million CVCs are inserted annually [242]. However, central venous catheterizations are associated with nearly 90% of nosocomial infections and are responsible for nearly 400,000 cases per year in the United States [242]. Although the incidence of local or bloodstream infections
(BSIs) associated with peripheral venous catheters is usually low, serious infectious complications produce considerable annual morbidity because of the frequency with which such catheters are used. The magnitude of the potential for CVCs to cause morbidity and mortality resulting from infectious complications has been estimated in several studies [243]. The attributable cost per infection is estimated to be ~ $45000 [244] and the annual cost of patient treatment with CVC-associated BSIs has gone up to $2.3 billion [245].
Finding an effective treatment is also difficult as S. aureus has the ability to form biofilms on devices, which makes it even more impenetrable for antimicrobial agents. The antibiotic concentrations required to kill bacteria in the biofilm is a thousandfold higher than that needed to kill the planktonic cells of the same species [246]. Recent prophylactic strategies have included precoating devices with different antibiotics [247], however the
45 occurrence of a number of multi-drug resistant strains as spurred the need for alternatives to antibiotic treatment. [248].
Use of enzymes and peptides as antimicrobial coatings on surfaces has prevented device-associated infections [79-81]. The use of nanoparticles (NPs) as surfaces for enzyme immobilization offers several important advantages over the use of purely enzyme based coatings. Inherent physiochemical properties of NPs. can be used to improve activity of the enzymes immobilized on them [249]. Although coating medical devices and implants with plain antimicrobial enzymes could work just as effectively, coating catheter surfaces with enzyme-NP conjugates may be advantageous due to larger surface area for effective enzyme coating, higher stability, durability and lower enzyme leaching. However it is important to characterize enzyme-NP conjugates by studying individually whether aspects of nanoparticle size, surface charge density etc. could help improve efficacy of enzymes for antibacterial applications.
Our studies involve simple models for studying efficacy of enzyme-NP conjugates to planktonic bacteria or cell suspensions, simultaneously against that of free enzymes.
Also, these studies will help optimize and standardize a protocol that can be used for the preparation and synthesis of enzyme-NP conjugates for multiple enzymes for different applications. For e.g. by studying the binding yields of enzymes at different NP: enzyme ratios, we can understand the binding affinities for each of the enzymes to the surface of nanoparticles. Specific antibacterial activity assays can quickly be used to measure the activity of conjugated enzymes to NPs. Here it is important to clarify the difference in using the terms „activity‟ and efficacy of enzyme-NP conjugates. Activity of the enzyme refers to a specific assay that characterizes the catalytic function of the enzyme against its
46 substrate. Efficacy of enzyme-NP conjugates refers specifically to the targeting and antibacterial activity of the conjugates against bacterial cell suspensions.
High antibacterial efficacy of the conjugates against planktonic cells does not imply high efficacy against the same bacteria in colonies, as different factors may control activities in these cases. Therefore, it is necessary to perform in vitro experiments in order to evaluate the inhibition and disruption of bacteria grown on a surface. Various combinations of enzymes have been studied for the disruption and destruction of bacterial biofilms [106]. This requires a minimum of two enzymes, one enzyme for disruption of the biofilm matrix and a second enzyme with bactericidal activity. Conjugation of these enzymes to the surface of the same catheter could address the problems associated with this application. We chose to attach enzymes to the surfaces of PLA nanoparticles, characterize enzyme-NP conjugates and coat catheter segments at different concentrations, simultaneously coating catheters with free enzymes and then evaluating the inhibition of
S. aureus biofilms. Thus adsorption of two enzymes to the same surface could result in a synergistic effect that could cause inhibition and lysis of the biofilm-infected device and subsequently avoid the need for surgical removal of the infected device.
Since our system will function as a blood contacting device in vivo, the sample coated catheters will encounter various blood plasma proteins [235]. If binding yield and activity of enzyme-NP conjugates coated on catheters is significantly lower than free enzyme coatings or enzyme leaching from the surface of the NPs is a problem, then an alternative strategy will be precoating the catheter segments with NPs before enzyme adsorption . Then, the goal of our final study will be comparison of in vitro performance for plain and NP coated catheter segments coated by antimicrobial enzymes.
47
3.2. Specific aims
Specific aim I. Effect of charge-directed targeting on antibacterial activity using enzyme-nanoparticle conjugates of lysozyme to Micrococcus lysodeikticus
Hypothesis: Use of antimicrobial enzymes attached to nanoparticles is of special interest alternative to antibiotics to treat microbial infections. Special properties of the carrier can be used to improve targeting and activity of the conjugated enzyme. Here, we decided to show in a model system that nanoparticle charge can be used to enhance delivery of lysozyme to the cells of its natural substrate, Micrococcus lysodeikticus and significantly improve its lytic activity. Hen egg lysozyme was covalently attached to two types of polystyrene latex nanoparticles: positively charged, containing aliphatic amine surface groups, and negatively charged, containing sulfate and chloromethyl surface groups. Covalent conjugation of lysozyme to each of these groups was achieved either by direct conjugation or by using PEG spacers. Our study aimed at comparing the lytic activity of these enzyme-nanoparticle conjugates with that of free enzyme using either bacterial or low molecular weight substrates. The effect of attaching PEG spacers and method of enzyme purification from nanoparticles (through dialysis and centrifugation) on enzyme activity was also studied [250].
Specific aim II. Effect of antibody-directed targeting on antibacterial activity using enzyme-nanoparticle conjugates of lysostaphin to Staphylococcus aureus
Hypothesis: Our previous study clearly illustrated that physiochemical properties of nanoparticles can be used to help improve targeting of enzymes to bacterial cells.
Lysozyme shows little or no antibacterial activity by itself against S .aureus [251].
48
However, lysostaphin is an antibacterial enzyme which specifically cleaves cross-linked pentaglycine bridges in the peptidoglycan of S. aureus, thereby hydrolyzing the cell wall and lysing the bacteria [252]. We decided to study and compare the antimicrobial activity of lysostaphin adsorbed on the surface of PLA (poly (lactic acid)) nanoparticles and compare the activity of immobilized lysostaphin to that of the free enzyme. Enhanced antimicrobial efficacy against Gram-positive Listeria monocytogenes has been previously demonstrated for lysozyme co-immobilized with anti- L. monocytogenes antibody on polystyrene latex NPs for applications in storage and food packaging [230]. However, a systematic study of activity of targeted enzyme-NP conjugates with different enzyme and antibody loadings against clinically relevant pathogens has not been performed previously.
With this in mind, the goal of the present work was a detailed study of antimicrobial activity of targeted (lysostaphin-antibody-NPs) and untargeted (lysostaphin-NPs) conjugates against S. aureus at different enzyme and antibody ratios.
Specific aim III. Evaluation of antimicrobial activity of lysostaphin coated hernia repair meshes
Hypothesis: The goal of the earlier proposed work was to initially show a proof of concept where we could achieve enhanced antibacterial efficacy through targeted delivery of enzymes using polymer nanoparticles to cell suspensions (model for planktonic or free floating bacteria) in vitro (Specific aim I and II). Bacteria can either colonize the implant surface or exist in a planktonic free floating form. Hence we had initially proposed to use enzymes and enzyme-NP conjugates as surface coatings for indwelling medical devices and implants to inhibit bacterial adhesion and subsequent colonization. But as mentioned earlier, high activity of enzymes of conjugates enzymes to bacterial cell suspensions does
49 not imply high activity against colonized bacteria, as different factors may control activities in these cases.
Therefore, it was necessary to perform in vitro experiments in order to evaluate the inhibition bacteria adhering to a surface and then expand these studies to test and compare the antimicrobial activity and antibiofilm activity of devices coated by enzymes and enzyme-NP conjugates. However, the studies required to characterize and evaluate enzymes and enzyme-NP conjugates coated on implant surfaces are slightly different compared to studies characterizing enzymes conjugated to nanoparticles. Therefore, before embarking on this study, we decided to study and characterize antimicrobial activity of enzyme adsorption to implants and also evaluate inhibition of bacterial colonization.
The objective of this study was to evaluate the in vitro antimicrobial antimicrobial susceptibility of S. aureus to lysostaphin-coated hernia repair polymer meshes and study the effect of enzyme coating concentration on antimicrobial activity. A pilot in vivo study was also performed by our collaborators at Carolinas Medical Center to study dose- dependence on antimicrobial activity of lysostaphin-coated meshes using in a rat hernia repair model.
Specific aim IV. Application of enzymes as antibacterial coatings on plain and nanoparticle coated catheter segments
Hypothesis: We had initially proposed to study and characterize the in vitro antimicrobial and antibiofilm efficacy of Central venous catheters segments coated with enzyme-NP conjugates and free enzymes. Different coating concentrations were estimated in order to optimize the bound weight of enzyme-NP per specific surface area to the bound
50 concentration of plain enzyme for the catheter segment from the same batch. Synthesis of enzyme-NP conjugates were done via coadsorption of two model enzymes, DispersinB
(antibiofilm) and lysostaphin (antimicrobial) on the surface of Poly ((lactic acid))nanoparticles. However, we were not able to achieve higher loading via adsorption of enzyme-NP conjugates on the surface of 1 cm catheter segments compared to plain enzyme adsorption. We hypothesized that coating by enzymes interfered with nanoparticle attachment and lead to reduced adsorption.
We therefore proposed an alternative strategy, which involved: a) Initial coating of the catheter segments by nanoparticles and b) Coating of a series of different initial enzyme concentrations (same set) on the surface of both plain and NP coated catheter segments. Through this method, we also did not had to optimize enzyme adsorption protocols for NPs separately, achieve high monolayer coatings individually and overall economically it was a more feasible option.
We performed a series of experiments to study and characterize enzyme adsorption to plain and Poly ((lactic acid))nanoparticles coated Polyurethane catheter segments. PLA nanoparticle coating on Polyurethane catheter segments were achieved via adsorption and surface area for plain and NP coated catheter segments was characterized using BET surface are analysis and SEM imaging. The enzymes used for adsorption were the same set of enzymes used for characterization earlier, lysostaphin (antimicrobial and antibiofilm) and DispersinB (antibiofilm). Binding yield of lysostaphin and DispersinB to plain and NP coated segments either alone or coadsorption (in the presence of the other) was studied for a series of different coating concentrations through BCA and fluorescence analysis. Adsorbed lysostaphin and DispersinB activity on catheter segments was done
51 through turbidimetric assay (antibacterial activity) and colorimetric assay respectively.
Enzyme leaching studies were performed using 2% BSA in phosphate buffered saline
(PBS) as a model system. Cytotoxicity of the catheter segments were evaluated using
MTS assay with 3T3 Mouse fibroblasts as a cell model for wound healing. Antibiofilm efficacy of the coated catheter segments was evaluated using a methylene blue staining method to quantify the amount of biomass (biofilm) adhered to the surface of the catheter.
Finally we studied long term durability testing of the coated catheter segments to evaluate its shelf life stability (equivalent of 4 months) for both lysostaphin and DispersinB adsorbed on plain and NP coated segments. Activity of enzymes was monitored similarly using a turbidity assay (lysosyaphin) and colorimetric assay (DispersinB) as described previously.
We expect to prevent bacterial colonization and eventual biofilm formation through surface coating of the catheters with enzyme-nanoparticle conjugates and plain enzymes but not expected to treat S. aureus associated-blood stream infections. Long term durability of NP coated catheters is expected due to enhanced stability of adsorbed enzyme compared to free enzyme at same coating concentrations. The approach to be developed here is universal: it can potentially be used for treatment of other medical device-associated infections.
52
CHAPTER 4
CHARGE-DIRECTED TARGETING OF ANTIMICROBIAL ENZYME- NANOPARTICLE CONJUGATES
4.1. Introduction
Antibiotics once were regarded as a universal antimicrobial weapon. However, many bacteria have developed antibiotic-resistant strains, and the number of such resistant species is growing quickly [253]. Alternative approaches to treat bacterial infections are urgently needed in healthcare facilities worldwide. Use of enzymes as antimicrobial agents is nature-inspired and has recently attracted much attention as an antibiotic-free approach to treat bacterial infections [254]. The use of antimicrobial enzymes covalently attached to nanoparticles is of special interest because of enhanced stability of protein- nanoparticle conjugates and the possibility of targeted delivery.
In recent years, much attention has been paid to the use of proteins conjugated to polymer colloids in biomedical applications [206-208, 255, 256] . Such colloids are easy to produce and can be prepared with a high degree of monodispersity [210, 211, 257, 258].
Colloidal nanoparticles provide an almost ideal remedy to the usually contradictory issues encountered in the optimization of immobilized enzymes: minimum diffusional limitation, maximum surface area per unit mass, and high enzyme loading [187]. Different properties of nanoparticles, such as size, surface charge density or density of the surface reactive groups can also be used to add functionality to their conjugates with biomolecules. In spite of tremendous interest in potential applications of protein-nanoparticle conjugates, systematic studies of the effect of the properties of nanoparticles on the function of the attached proteins have rarely been conducted.
53
Here, we study the effect of nanoparticle charge on bactericidal activity of an antimicrobial enzyme, lysozyme. Hen egg white lysozyme (HEWL) belongs to a subfamily of c-type lysozymes (chicken or conventional). HEWL has molecular weight of
14.3 kDa and consists of 129 amino acid residues. Among these, 17 are positively charged
(6 lysines containing free amine groups and 11 arginins) and 9 are negatively charged (2 glutamate and 7 asparatate residues). HEWL natural substrate is peptidoglycan murein, the major component of the bacterial cell wall. Hydrolysis of murein results in bacterial cell lysis [259]. Gram-positive bacteria are generally more sensitive to lysozyme than Gram- negative bacteria because the latter possess a protective lipopolysaccharide outer membrane that prevents access of lysozyme to its substrate, murein.
In this study, we covalently attached hen egg lysozyme to two types of polystyrene latex nanoparticles: positively charged, containing aliphatic amine surface groups, and negatively charged, containing sulfate and chloromethyl surface groups. Covalent coupling to each of these groups was achieved either by direct conjugation or by using polyethyleneglycol (PEG) spacers. Our study aimed at comparing the activity of these protein-nanoparticle conjugates with that of free enzyme using either bacterial or low molecular weight substrates.
4.2. Materials and methods
4.2.1. Materials
Hen egg lysozyme and lyophilized Micrococcus lysodeikticus were purchased from
Sigma Aldrich (St. Lous, MO). Bis-butyraldehyde PEG cross-linker and NH2-PEG-COOH
54 cross-linker were purchased from Nektar Therapeutics, Inc. (San Carlos, CA). All other chemicals were purchased from Sigma Aldrich.
Protein concentrations for binding efficiency studies were determined using BCA and Micro BCA assay kits purchased from Pierce Biotechnology (Rockford, IL).
Colorimetric and turbidometric activity assays were performed using a Synergy microplate reader (Bio-Tek Instruments, Inc., Winooski, VT). Spectrum/Por Biotech
Cellulose ester membranes with 5 kDa and 100 kDa (MCWO) from Spectrum laboratories
Inc. (Rancho Dominguez, CA) were used for dialysis.
Protein-nanoparticle conjugates were characterized by zeta potential measurements using a ZetaPlus Analyzer (Brookhaven Instruments Co, Holtsville, NY). The samples containing protein-nanoparticle conjugates were diluted 1:100 in 6 mM potassium phosphate buffer to obtain the appropriate particle concentration. Data presented are the average of the measurements of three sample replicates each reproduced four times. The protein-nanoparticle conjugates were characterized using Dimension III Atomic Force
Microscope (AFM) with Nanoscope 4 controller (Vecco Instruments, Santa Barbara, CA).
AFM experiments were conducted in tapping mode in ambient conditions using rectangular cantilevers backside-coated with aluminum with a spring constant of ~40 N/m
(Mikromasch Inc., Chapel Hill, NC). For sample preparation, a silicon substrate surface was sonicated in reagent-grade ethanol and then rinsed copiously with HPLC-grade water.
A 10-15 L drop of nanoparticle suspension was placed on the substrate surface, incubated for 3 min, and blown off with blast of compressed nitrogen.
55
4.2.2. Methods
4.2.2.1. Preparation of protein-nanoparticle conjugates.
The polystyrene nanoparticles used in all experiments were purchased from
IDC.(Interfacial dynamics Corporation, Eugene, OR) and contained monodisperse particles with a diameter close to 20 nm. Positively-charged nanoparticles surface- modified with aliphatic amine groups and negatively-charged nanoparticles surface- modified with chloromethyl and sulfate groups were used in this study.
Sample Mean Solids, Particles Surface area Surface No. of diameter % per mL of per functional density of functional of suspension group functional groups particles groups, per (nm) cm-2 particle
Aliphatic 15 2 12 amine 20 2.4 3.1·10 65 Å /NH2 15.3·10 2850 particles
15 2 12 R-CH2Cl 20 4.1 9.3·10 4828 Å /R- 2.1·10 26 12 particles CH2Cl 4.5·10 56 2 - 2194 Å / R-SO3
The properties of the nanoparticles, as provided by the manufacturer, are highlighted in
Table 4.1. above.
Polystyrene nanoparticles are supplied as a stable suspension in solution containing a proprietary surfactant. To remove the surfactant, this suspension was purified by dialysis using a membrane with the 5 kDa molecular weight cutoff. Approximately 1 mL of the suspension was diluted two-fold by 2 mM MES buffer and loaded into the dialysis tubing. The tubing was then clamped on both ends using dialysis membrane closures and placed in a 5 L reservoir containing 2 mM MES buffer. The dialysis was
56 performed overnight at room temperature; the entire dialysate was stirred continuously and changed after 2-4 hours, 6-8 hours and 10-14 hours. Sample recovery was done by removing it from the tubing with a micropipette. The volume of dialysate did not change significantly after dialysis. Dialyzed sample was then sonicated for 30 min to destroy aggregates.
4.2.2.2. Protein conjugation to aminated nanoparticles
In the case of direct conjugation of lysozyme to the amine-modified nanoparticles
(Figure 4.1.A), dialyzed nanoparticles were again diluted two-fold in 2 mM MES buffer
(pH=5.0) to the final concentration of ~7.8·1014 particles per mL. An 8 wt. % solution of glutaralehyde in 2 mM MES buffer was then added to the diluted nanoparticle suspension to achieve a final glutaraldehyde concentration of 1 wt. %. A 100 mM solution of sodium cyanoborohydride, Na[BH3(CN)] (Caution: foul odor; should be used in a chemical hood only), in 2 mM MES was added to the sample immediately after glutaraldehyde. Final concentration of sodium cyanoborohydride in the suspension was 10 mM. The amine groups on the surface of the particles react with the aldehyde moieties to form Schiff‟s base intermediates, which are then selectively reduced by sodium cyanoborohydride.
After 7 h incubation at room temperature, the unreacted glutaraldehyde and
Na[BH3(CN)] were separated by dialysis using a procedure similar to that described above. A cellulose ester membrane with a molecular weight cutoff of 5 kDa was used for dialysis. Six lysozyme solutions with concentrations of 2.94, 29.4, 73.6, 147.2, 220.8 and
294.4 g/mL were prepared in 2 mM MES buffer at pH 5.0 from a 5 mg/mL lysozyme stock solution. These concentrations were chosen so that to achieve initial lysozyme-to-
57 nanoparticle ratios of 1:1, 10:1, 25:1, 50:1, 75:1, and 100:1 enzyme molecules per nanoparticle, respectively.
100 L of purified nanoparticle suspension (~6.2·1013 nanoparticles) was added to
0.5 mL of each of the protein solutions. The mixtures were incubated overnight at room temperature with gentle stirring. The protein-nanoparticle mixture was then treated with
0.1% Tween 20 for 2 h to remove any physically adsorbed enzyme. The unbound protein along with the excess Tween 20 was then separated by either centrifugation or dialysis. In the case of centrifugation, the suspensions were centrifuged at 13,000 g for 75 minutes.
The pellet was resuspended in 2mM MES buffer and centrifuged again. The supernatant was collected in both steps to quantify the unbound protein. Separation by dialysis was done similarly to dialysis of free nanoparticles described above. A cellulose ester membrane with a MWCO of 100 kDa was used for dialysis to separate any unbound lysozyme (molecular weight 14.3 kDa).
Bis-butyraldehyde PEG spacers (O=CH-CH2-CH2-CH2-(PEG)n-CH2-CH2-CH2-
CH=O, MW 3,700) were also used for conjugation of lysozyme to aliphatic amine particles (Figure 4.1.B). Since attachment via PEG spacers minimizes interactions of immobilized protein with the support [170, 260], lysozyme conjugated to nanoparticles via
PEG spacers was expected to show higher enzymatic activity. The conjugation protocol, enzyme and nanoparticle concentrations, and purification methods were similar to those described above for the glutaraldehyde-based attachment. Final concentration of bis- butyraldehyde PEG spacers was 25 mg/mL in all experiments. It should be noted that the use of butyraldehyde chemistry enables protein conjugation specifically via its N-terminus if the reaction is performed at pH 5.0 [261].
58
4.2.2.3. Protein conjugation to chloromethylated nanoparticles
Direct coupling of lysozyme to chlomethylated latex nanoparticles occurs via the reaction of the latex CH2Cl- groups with the amine groups of lysozyme to give a secondary amine (Figure 4.1.C). Conjugation in 2 mM MES buffer (pH = 5.0) was performed overnight at room temperature. Six lysozyme solutions with the concentrations of 8.82, 88.2, 220.5, 441, 661.5 and 882 g/mL were prepared in 2 mM MES buffer at pH
5.0 from a 5 mg/mL lysozyme stock solution. The dialyzed chloromethylated particles suspension was diluted two-fold in 2 mM MES buffer, and 80 L of this suspension
(~1.9·1014 particles) was added to 0.5 mL of each of the protein solutions. Lysozyme concentrations were chosen to achieve initial lysozyme-to-nanoparticle ratios of 1:1, 10:1,
25:1, 50:1, 75:1, and 100:1 protein molecules per nanoparticle, respectively. To remove any physically adsorbed lysozyme, samples were treated with 0.1% Tween 20, and the protein-nanoparticle conjugates were separated by either centrifugation or dialysis.
For conjugation via a spacer, a bifunctional NH2-(PEG)n-COOH (MW 3,400) spacer was employed. Dialyzed nanoparticle suspension was mixed with the spacer solution in
2 mM MES buffer. Final concentration of nanoparticles was ~1.9·1015 mL-1; and final concentration of the NH2-(PEG)n-COOH spacer was 5 mg/mL. The nanoparticle/spacer mixture was incubated overnight at room temperature with gentle stirring. The amine group of the spacer reacts with the chloromethyl group on the nanoparticles (Figure 4D).
The excess spacer was then separated by dialysis using 5 kDa MWCO membrane.
1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDAC) chemistry
(Hermanson. 1996) was used to conjugate lysozyme to the PEGylated nanoparticles. First,
25 mg/mL EDAC solution in 2 mM MES buffer was added to the dialysed nanoparticles
59 suspension (final EDAC concentration was 5 mg/mL). After 10 min incubation, nanoparticles with activated carboxyl groups were added to lysozyme solutions.
Lysozyme and nanoparticle concentrations were the same as used for direct attachment to chloromethylated particles. After treatment with Tween 20, unbound lysozyme, as well as excessive reagents and reaction products, were separated by either centrifugation or dialysis with a 100 kDa MWCO membrane.
Figure 4.1. Schematic of protein conjugation to nanoparticles. (A) Direct conjugation to aminated nanoparticles. (B) Conjugation to aminated nanoparticles via bis-butyraldehyde PEG spacer. (C) Direct conjugation to chloromethylated nanoparticles. (D) Conjugation to chloromethylated nanoparticles via NH2-PEG-
COOH spacer.
60
4.2.2.4. Binding yield study
The number of bound lysozyme molecules per nanoparticle was determined as follows.
Samples with initial lysozyme concentrations corresponding to 1, 10, 25, 50, 75, and 100 molecules per nanoparticle were prepared. After the conjugation step, samples were treated with Tween 20 to remove any non-covalently adsorbed protein and centrifuged at
13,000 g for 75 min. The clear supernatant was then collected and BCA protein assay was used to determine the concentration of the unbound protein in the supernatant. Lysozyme standards were prepared by diluting 5 mg/mL lysozyme stock solution in 2 mM MES buffer containing 0.1% of Tween 20. The amount of bound protein was found from the difference between the initial and unbound protein concentrations. Each of the experiments on determination of binding efficiency was reproduced three times with independently synthesized samples.
4.2.2.5.Enzyme activity assay with bacterial substrate
Protein-nanoparticle conjugates prepared from the suspension with the initial 25:1 lysozyme:nanoparticle ratio were used for all enzyme activity assays and for zeta-potential measurements. A turbidometric enzyme activity assay was performed by monitoring the decrease in optical density of an aqueous suspension of a natural substrate for lysozyme,
Micrococcus lysodeikticus, in potassium phosphate buffer. Micrococcus lysodeikticus was preferred to other Gram-positive species because it is routinely used as a substrate for assaying lysozyme activity (Shugar 1952). A 6 mM solution of potassium phosphate was prepared by dissolving 0.898 g of potassium dihydrophosphate in 1 L of deionized water.
The pH of the solution was adjusted to 6.2 using a 0.1 M solution of potassium hydroxide.
A 0.018 wt. % suspension of Micrococcus lysodeikticus was prepared by suspending 3.75
61 mg of lyophilized cells in 20 mL of 6 mM potassium phosphate buffer solution. Protein- nanoparticle conjugates were then added to the bacterial suspension; net concentration of lysozyme in all final mixtures was adjusted to 8.3 g/mL. Samples were incubated in a 12- well microplate at room temperature, and the decrease in optical density at 405 nm was monitored as a function of time. Initial rates of the reactions were calculated from the slopes of the linear initial parts of the kinetic curves. Two independently synthesized samples were studied for each type of nanoparticles. Control experiments were performed with free lysozyme at the same enzyme concentration as in the samples. All kinetic experiments were reproduced in triplicate.
4.2.2.6. Enzyme activity assay with a low molecular weight substrate
Lysozyme activity was also determined colorimetrically using a neutral oligosaccharide substrate for lysozyme, p-nitrophenyl penta-N-acetyl-ß-chitopentaoside, PNP-(GlcNAc)5, and was based on a coupled reaction with NAHase [262]. A 10 mM solution of PNP-
(GlcNAc)5 at pH 6.2 was prepared by dissolving 1.5mg/mL PNP-(GlcNAc)5 in 6 mM potassium phosphate buffer. Lysozyme cleaves the first glycosidic bond of PNP-
(GlcNAc)5, giving the monochitoside, PNP-GlcNAc. NAHase [purchased as a suspension in 2.5 M (NH4)2SO4 containing 1.2 mg protein/mL and diluted 1:100 in 6 mM phosphate buffer] then catalyzes the release of the chromogen, p-nitrophenol, from the PNP-GlcNAc.
Addition of 50 mM Na2CO3 stops the reaction and yields a yellow color from the released p-nitrophenolate. For activity measurements, 50 L sample aliquots were incubated with
150 L of PNP-(GlcNAc)5 and 10 µL of NAHase at 37C in a 96 microplate well. Six reaction wells were prepared for each sample. Every 30 minutes, 100 µL of a 50 mM
62
Na2CO3 was added to one of the wells, and the absorbance at 405 nm was measured using a microplate reader, while the reaction was allowed to continue in the other wells. Six time points were taken for each sample. From these points, initial rates were determined and used as a measure of enzyme activity. Two independently synthesized samples were studied for each type of nanoparticles. Control experiments were performed with free lysozyme at the same enzyme concentration as in the samples. All kinetic experiments were reproduced in triplicate.
4.3. Results
4.3.1. Yield of binding.
Figure 4.2 shows efficiency of binding of lysozyme to polystyrene latex nanoparticles without PEG spacers. In the case of attachment via PEG spacers, binding yields are similar to those shown in Fig. 2. Binding yield in the case of aminated particles is in both cases lower than that for their negatively charged counterparts. We speculate that this difference is primarily due to electrostatic effects. The initial step of covalent attachment of lysozyme to nanoparticles involves its physical adsorption (Suen and Morawetz 1985;
Sarobe et al. 1998; Molina-Bolivar and Galisteo-Gonzales 2005; Giacomelli et. al 2000).
This adsorption is primarily determined by the interactions of the hydrophobic parts of lysozyme molecules with the hydrophobic surface of polystyrene latex. However, it can be expected that more positively charged lysozyme molecules (pI=10.6) can be adsorbed to a negatively charged surface, than to a positively charged surface. This difference in the initial physical adsorption results in different binding efficiency. Notably, a theoretical monolayer of lysozyme on a 20 nm spherical particle calculated from the dimensions of
63 the lysozyme molecule (2.5 nm X 3.5 nm X 3.5 nm) should consist of ~120 molecules per nanoparticle [263].
Figure 4.2. Efficiency of covalent attachment of lysozyme to polystyrene latex nanoparticles for samples prepared by (A) direct conjugation and (B) conjugation via PEG spacers. Black squares correspond to aminated nanoparticles and red circles correspond to chloromethylated nanoparticles, respectively.
At high initial lysozyme concentrations, the number of attached lysozyme molecules exceeds the number of chloromethyl groups available for attachment (see Table
4.1). However, it should be noted that lysozyme tends to form oligomers (dimers and tetramers) at pH>4.0 [264]. The degree of oligomerization can also be affected by interaction with polystyrene latex particles. The observed excessive conjugation is consistent with the attachment of oligomers. Possible presence of physically adsorbed lysozyme that cannot be removed by treatment with 0.1% Tween 20 also cannot be completely ruled out. It should be noted, however, that according to the literature data
(Peula et al. 1995), treatment by Tween 20 removes 75 - 85 % of physically adsorbed
64 protein from polystyrene latex microspheres. We observed binding yields of ~80% in the case of lysozyme attachment to chloromethylated nanoparticles after treatment with
Tween 20. It is thus unlikely that non-covalent adsorption of lysozyme plays an important role in the observed phenomenon.
4.3.2. Bacterial turbidometric activity assay
Protein-nanoparticle conjugates prepared from the suspension with the initial 25:1 lysozyme: nanoparticle ratio were used for all enzyme activity assays. A bacterial turbidometric activity assay was performed by monitoring the decrease in optical density of an aqueous suspension of Micrococcus lysodeikticus. Lysozyme cleaves the β (1-4)- glycosidic bond between N-acetylmuramic acid (MurNAc) and the N-acetyl-D- glucosamine (GlcNAc) in the polysaccharide that forms the backbone of the bacterial cell walls. This cleavage is the first step in the lysis of bacterial cells. The degree of lysis is measured as reduction in turbidity of the bacterial suspension (Shugar 1952). We monitored the decrease in optical density at 405 nm for two sets of samples: a) samples purified by centrifugation and b) samples purified by dialysis. To take into account possible precipitation of bacteria from the suspension, the turbidity of the bacterial suspension alone was monitored in a control experiment.
65
A B
C D
Figure 4.3. Results of bacterial lysis assay. (A) Direct attachment of lysozyme to nanoparticles; samples purified by centrifugation, (B) attachment via PEG spacer; samples purified by centrifugation, (C) Direct attachment of lysozyme to nanoparticles; samples purified by dialysis, and (D) attachment via PEG spacer; samples purified by dialysis. Black squares – lysozyme attched to positively charged nanoparticles; red circles – lysozyme attached to negatively charged nanoparticles, blue triangles – free lysozyme.
In the first set of samples, the final step of purification of protein-nanoparticle conjugates was centrifugation. Figure 4.3 compares the rates of bacterial lysis for
66 lysozyme directly conjugated to nanoparticles to that of free lysozyme. The initial rate of the lysis reaction was calculated from the slopes of the linearized kinetic curves. We observed significantly faster bacterial degradation for lysozyme conjugated to aminated nanoparticles (Table 4.2). Bacterial lysis rates for lysozyme conjugated via PEG spacer were not significantly different from those observed for lysozyme directly attached to the same type of nanoparticles.
Initial rate from linearized slopes for Sample sample purified by dialysis centrifugation
Free lysozyme 4.90.1 · 10-4
Lysozyme directly conjugated to aminated 8.90.2 · 10-4 7.00.1 · 10-4 nanoparticles Lysozyme conjugated to aminated 9.30.2 · 10-4 6.50.1 · 10-4 nanoparticles via PEG spacer Lysozyme directly conjugated to 2.70.1 · 10-4 2.40.1 · 10-4 chloromethylated nanoparticles Lysozyme conjugated to chloromethylated 2.40.1 · 10-4 2.50.1 · 10-4 nanoparticles via PEG spacer
Table 4.2. Initial rates of the bacterial lysis study for samples purified by dialysis and centrifugation
In an another study, we also studied the effect of controls, positively charged aliphatic amine and negatively charged chloromethylated nanoparticles on lytic activity of
Micrococcus lysodeikticus.. No detectable lytic activity was observed for any of the controls. Figure 4.4 compares the rates of bacterial lysis for both types of polystyrene latex nanoparticles along with control bacterial cell suspension in the same study
67
Figure 4.4. Results of bacterial lysis assay. Pink circles – Aliphatic amine nanoparticles; Light Blue squares – Chloromethyl nanoparticles, Yellow squares –
Control bacterial cells.
We also studied the effect of purification method on the activity of the conjugates.
The unbound protein was separated from protein-nanoparticle conjugates by either dialysis or centrifugation. Figure 4.3 (B) shows results of bacterial lysis assay for samples purified by centrifugation, and Figure 4.3 (A) shows the data for samples purified by dialysis. Table 4.2 summarizes results of bacterial lysis assay. As can be seen from Table
4.2, dialyzed samples always show slightly higher bacterial lysis rates than those purified by centrifuging. A possible explanation of the effect of purification method is higher degree of nanoparticle aggregation in the case of centrifugation. As a result of aggregation, some lysozyme molecules can be trapped inside the aggregates and blocked from the interaction with their substrate. Figure 4.5 compares representative AFM images of lysozyme conjugated to aminated nanoparticles and purified by dialysis (Fig. 4.3A) or
68 centrifugation (Fig. 4.3B). As can be seen from Fig.4.5, much larger aggregates are observed in the case of samples purified by centrifugation. Since samples purified by dialysis showed higher activity than those purified by centrifugation, dialysis has been used as the primary purification method in all further experiments.
A B
Figure 4.5. Tapping mode AFM images comparing the degree of aggregation for samples purified by dialysis and centrifugation. Scan size is 5 m x 5 m and Z scale is 100 nm. (A) Lysozyme conjugated to aliphatic amine nanoparticles after purification by dialysis and (B) lysozyme conjugated to aliphatic amine nanoparticles after purification by centrifugation.
4.3.3. Enzyme activity assay with a low molecular weight substrate
PNP-(GlcNAc)5 is a chromogenic pentachiteoside that serves as an alternative substrate for lysozyme [265, 266]. The color is produced by free p-nitrophenol (PNP), liberated from PNP-(GlcNAc)5 through a coupled reaction of lysozyme and NAHase. First,
PNP-(GlcNAc)5 is cleaved by lysozyme at the first glycosidic linkage, releasing monochiteoside PNP-GlcNAc [266]. The presence of NAHase is essential for the assay to promote the liberation of free p-nitrophenol. Addition of Na2CO3 at the final step
69 increases the pH of the reaction mixture, stops the reaction and reveals the yellow color of
PNP.
Figure 4.6. Lysozyme activity assay with low molecular weight substrate
PNP-(GlnAc)5. (A) Direct attachment and (B) attachment via PEG spacer. Black squares – lysozyme attached to positively charged nanoparticles; red circles – lysozyme attached to negatively charged nanoparticles, blue triangles – free lysozyme.
Figure 4.6 shows the results of the assay with PNP-(GlcNAc)5. In this case, free lysozyme shows higher activity than lysozyme conjugated to both types of nanoparticles.
The difference between the activities of lysozyme attached to aminated and chloromethylated nanoparticles is much less pronounced than in the case of bacterial lysis assay. To obtain numerical values of kinetic constants, kinetic curves were empirically linearized in coordinates Log [Optical Density] – Time.
70
Sample Initial rate from linearized slopes Free enzyme 5.10.5 · 10-3
Lysozyme directly conjugated to aminated nanoparticles 4.40.3 · 10 -3
Lysozyme conjugated to aminated nanoparticles via PEG 4.90.3 · 10-3 spacer Lysozyme directly conjugated to chloromethylated 4.30.4 · 10-3 nanoparticles Lysozyme conjugated to chloromethylated nanoparticles 4.4 0.3 · 10 -3 via PEG spacer
Table 4.3. Initial rates of enzyme activity with low molecular weight substrate for all samples
4.3.4. Zeta potential analysis
Zeta potential studies were performed to characterize electrostatic properties of colloidal suspensions used in this study. Table 4.4 shows zeta potential values averaged over 4 independent experiments for each of the samples. As expected, Micrococcus lysodeikticus in aqueous suspension has highly negative zeta potential. Highly positive zeta potentials are observed for aminated nanoparticles; zeta potential values slightly decrease upon direct coupling of lysozyme and further decrease upon protein conjugation via bis-butyraldehyde spacers.
It is currently not clear why conjugation of positively charged lysozyme to positively charged nanoparticles reduces their zeta-potential. One possible explanation of this fact is increase of hydrodynamic resistance of the conjugates. Presence of a “fluffy” lysozyme layer may decrease electrophoretic mobility of nanoparticles. Since the absolute value of zeta-potential is proportional to the electrophoretic mobility, this will result in the decrease of zeta-potential. This explanation is in agreement with the observed higher drop of zeta-
71 potential for the case of lysozyme attached to nanoparticles via PEG spacers, in which lysozyme has even more spatial freedom. Notably, even in the latter case zeta potential remains highly positive and significantly exceeds that of free lysozyme. Stronger electrostatic interactions resulting in more effective delivery can thus be expected between enzyme-nanoparticle conjugates and Micrococcus lysodeikticus.
Sample Zeta potential, mV Bare aminated nano-particles (control) + 37.3 1.1 Lysozyme directly conjugated to + 31.5 1.7 aminated nanoparticles Lysozyme conjugated to aminated + 27.2 1.7 nanoparticles via PEG spacer Bare chloromethylated nanoparticles - 43.3 1.5 (control) Lysozyme directly conjugated to - 32.0 1.6 chloromethylated nanoparticles Lysozyme conjugated to - 25.0 1.4
chloromethylated nanoparticles via PEG spacer
Bacterial substrate - 27.8 1.1 Lysozyme + 8.84 1.2
Table 4.4 - Results of zeta potential measurements.
Bare chloromethylated nanoparticles have highly negative zeta potential. Its absolute value decreases upon direct coupling of lysozyme and further decreases upon coupling of lysozyme via PEG spacers, consistent with the hypothesis of decreased electrophoretic mobility of the particles upon protein attachment. Yet for all protein conjugates with chloromethylated particles zeta potential remains highly negative hindering their interactions with the bacteria.
72
4.4. Discussion
Electrostatic interactions play an important role in targeting lysozyme to its substrate bacteria [267-271]. Lysozyme has pI of 10.6 [271] and therefore bears significant positive charge (+8 or +9 per molecule) at neutral pH [272]. Most bacterial cell walls, on the other hand, are negatively charged due to the presence of teichoic, lipoteichoic, and teichuronic acids in the cell membrane [273, 274].
Previous results indicate that positive charge is essential for lysozyme‟s antibacterial function but not to activity towards low molecular weight substrates.
Acetylation of all six free lysine residues of lysozyme abolished action towards
Micrococcus lysodeikticus but did not affect the cleavage of the tetramer of N- acetylglucosamine, (GlnAc)4, from chitin [275]. T4 bacteriophage lysozyme lost > 90% of its native activity in bacterials lysis assay with Micrococcus lysodeikticus when adsorbed onto 9 nm silica nanoparticles, which are strongly negatively charged [276]. At the same time, lysozyme adsorbed onto similarly negatively charged silica nanoparticles (4, 20, and
100 nm in diameter) retained ~ 80% of its native activity when assayed with a neutral substrate, PNP-(GlnAc)5 [263].
The results of this study indicate that nanoparticle charge can be used to control bacteriolytic activity of lysozyme. Lysozyme attached to positively charged nanoparticles shows significantly higher activity in bacterial lysis assay than free lysozyme. The observed effect is probably due to the differences in targeting of the enzyme to negatively charged bacterial cell walls (Figure 6) because similar activities are observed for lysozyme attached to both positively and negatively charged nanoparticles when assayed with a neutral low molecular weight substrate, PNP-(GlnAc)5. These activities are somewhat
73 lower than that of free enzyme, consistent with the previous data [276-279] on decreased activity of immobilized enzymes due to conformational changes arising from the interactions with nanoparticle surface and partial screening of the active site by the nanoparticle.
Figure 4.7 shows charge-directed targeting of lysozyme to bacterial cell walls
The hypothesis of electrostatic contribution to targeting nanoparticles to bacteria is confirmed by zeta-potential measurements, which show negative zeta potential for
Micrococcus lysodeikticus suspension and lysozyme conjugated to negatively-charged nanoparticles, but a positive zeta potential for free lysozyme and lysozyme attached to positively-charged nanoparticles. Notably, zeta potential for lysozyme attached to positively charged nanoparticles is more positive than that of free lysozyme, in good agreement with the observed higher activity of the former. On the other hand, electrostatic interaction is probably not the only factor that mediates targeting of lysozyme to bacterial
74 cells because lysozyme attached to negatively-charged particles also shows some bacteriolytic activity in spite of the fact that these nanoparticles should be electrostatically repelled from bacterial cell walls.
Another important result of this study is the development of a protein attachment protocol that minimizes nanoparticle aggregation. Aggregation often has been observed previously [280-282]and presents a serious problem for the use of nanoparticles in biomedical applications because of the uncontrollable size of the aggregates and broadening of the distributions of their functional properties. Here, we observed that samples with higher degree of aggregation show decreased enzymatic activity of the immobilized protein. The protocol developed in this work uses small amount of surfactant,
Tween 20, to remove non-covalently bound protein and employs dialysis for purification from unbound proteins and small molecules. Optimization of the concentrations of the components enabled us to significantly reduce degree of aggregation, as can be seen from representative AFM images in Figure 4.5.
We also studied the effect of attachment via PEG spacers on enzymatic activity of lysozyme. No significant difference in activity has been observed for lysozyme attached via a spacer compared to lysozyme directly attached to nanoparticle surface. This result was somewhat contradictory to our expectations because previous research [170, 260] demonstrated a beneficial effect of attachment via spacer on activity of immobilized enzymes. This effect was attributed to reduced interaction of the enzyme with the surface of its support due to the presence of the spacer. However, lysozyme is known to be a
“rigid” enzyme [283], which reluctantly changes its structure. In addition, we use Tween
20, which blocks the hydrophobic polystyrene surface and produces oligoethyleneglycol-
75 terminated surface. Thus, interaction of lysozyme with the nanoparticle surface is likely to be reduced even in the case of direct conjugation due to the presence of Tween 20. This agrees well with the small drop in enzymatic activity of the protein-nanoparticle conjugates, as compared to free lysozyme, in the assay with PNP-(GlnAc)5. Therefore, use of PEG spacers does not have a considerable effect on enzymatic activity of lysozyme.
In conclusion, nanoparticle charge has been used to controllably change activity of a covalently immobilized protein. Enhanced charge-directed targeting of lysozyme- nanoparticle conjugates to bacteria was achieved using positively charged aliphatic amine nanoparticles as the protein carriers. This result can be useful for development of antibacterial, antibiotic-free protein-nanoparticle conjugates.
76
CHAPTER 5
ANTIBODY-DIRECTED TARGETING OF LYSOSTAPHIN ADSORBED TO POLYLACTIDE NANOPARTICLES INCREASES ITS ANTIMICROBIAL ACTIVITY AGAINST S.AUREUS IN VITRO
5.1. Introduction
Staphylococcus aureus is a versatile and virulent Gram-positive pathogen that is frequently found among clinical isolates from patients in the United States. It is capable of causing a wide range of infections, being one of the major causes of bacteremia and is associated with significant mortality [39]. It produces a variety of virulence factors, with both structural and secreted products playing a role in the pathogenesis of infection. The rates of infections caused by both community and hospital-acquired strains are increasing steadily and treatment of these infections through conventional antibiotic therapy is becoming more and more difficult because of the increasing prevalence of multidrug- resistant strains [44]. The decreased effectiveness of current antibiotic treatments due to resistance has spurred the search for alternative approaches to antimicrobial therapy.
Currently, much attention is paid towards the use of natural antimicrobial proteins and peptides as antibacterials to overcome the problems associated with antibiotic resistance [284]. Naturally occurring antimicrobial peptides and enzymes are characterized by high activity, broad antimicrobial spectrum, and low rates of resistance [285, 286].
Lysostaphin is an antimicrobial enzyme produced by Staphylococcus staphylolyticus. It belongs to the class of Endopeptidases, which specifically cleave cross-linked pentaglycine bridges in the peptidoglycan of S. aureus, thereby hydrolyzing the cell wall and lysing the bacteria [287].
77
In spite of the great enthusiasm regarding the prospect of developing antibacterial enzymes and peptides as a new generation of antibacterial agents, they are still at an early stage of technologic maturation, and many hurdles have yet to be overcome [288].
Because of their peptidic nature, they could present the following problems: 1) short half life in vivo, 2) loss of activity in physiological conditions (especially at high ionic strength, which may interfere with their charge-directed targeting and mechanism of action), 3) toxicity and immunogenicity, 4) poor storage stability, and 5) interference with normal bacterial flora that may arise when attempts are made to use those peptides as antibacterial agents (3). Availability of a technology for targeted delivery of antimicrobial peptides could provide a valuable contribution towards addressing most if not all of these problems. Most importantly, targeted delivery can improve efficacy and reduce systemic toxicity of antimicrobial peptides by reducing their action against non-specific targets.
Although much work has been done recently on design of drug carriers for site- specific delivery, studies of targeted drug delivery have been thus far primarily focused on anticancer drugs [289] and certain other therapeutics targeting the host cells and tissues
[290, 291]. Much less attention has been paid up to date to studies of targeted delivery of antimicrobial agents. Recently, mannosylated or galactosylated polystyrene latex nanoparticles were shown to result in agglutination of mutant strains of E. Coli overexpressing mannose and galactose surface receptors, respectively [228, 229]. Targeted photothermal lysis of pathogenic bacteria such as Salmonella and Pseudomonas aergusinosa has been studied using antibody-conjugated gold nanoparticles and nanorods, respectively [292, 293]. Our group has previously studied charge-directed targeting of lysozyme conjugated to nanoparticles and demonstrated enhanced antimicrobial activity
78 against Micrococcus lysodeikticus for enzyme conjugated to positively charged nanoparticles [250, 294]. Enhanced efficacy against Gram-positive Listeria
Monocytogenes has also been demonstrated for lysozyme co-immobilized with anti- L.
Monocytogenes antibody on polystyrene latex NPs [295, 296]. However, a systematic study of activity of targeted enzyme-NP conjugates with different enzyme and antibody loadings against clinically relevant pathogens has not been performed previously. With this in mind, the goal of the present work was a detailed study of antimicrobial activity of targeted (lysostaphin-antibody-NPs) and untargeted (lysostaphin-NPs) conjugates against
S. aureus at different enzyme and antibody ratios.
5.2. Materials and methods
Staphylococcus aureus (ATCC 27660) was purchased from American Type Culture
Collection (Manassas, VA). Lysostaphin (26 kDa) isolated from Staphylococcus staphylolyticus was purchased from AMBI products LLC (Lawerence, NY). Poly (DL- lactide) and Bovine Serum Albumin (BSA) were purchased from Sigma Aldrich (St.
Louis, MO). Immunogen-specific Rabbit Polyclonal antibody to Staphylococcus aureus
(ATCC 27660) formulated in a phosphate saline buffer (0.01M, pH 7.2) was purchased from ViroStat (Portland, ME). Alexa Fluor 350 and Alexa Fluor 594 Carboxylic acid,
Succinimidyl ester dyes used for fluorescent labeling were purchased from Invitrogen,
Molecular Probes (Carlsbad, CA). Microsep Centrifugal Devices with a modified
Polyethersulfone membrane (low protein binding; MWCO - 3K and 30K) used for protein purification were purchased from Pall Life Sciences (Ann Arbor, MI). Fluorescence and turbidimetric activity assays were performed using a Synergy microplate reader (Bio-Tek
79
Instruments, Inc., Winooski, VT). Particle sizing and Zeta potential analysis were done using a 90plus Particle Size Analyzer (Brookhaven Instruments Co, Holtsville, NY).
5.2.1. Preparation of nanoparticles
Polylactide nanoparticles were prepared using a nanoprecipitation method based on solvent diffusion. Briefly, Poly(DL-lactide) (Mw 97,500) was dissolved in 25 ml of acetone at a concentration of 8 mg/ml and this solution was then dispersed in an aqueous phase (200 ml) containing Pluronic F68 (5 mg/ml) under sonication for 15 min. The acetone solution was added into the aqueous phase dropwise at a uniform rate. The resulting suspension containing the nanoparticles was then purified from excess of polymer and Pluronic by triple centrifugation followed by redispersion in DI water.
Centrifugation was performed at 3,000 rcf for 1 hr at 25°C using Allegra™ 64R
Centrifuge (Beckman Coulter, USA). Nanoparticle suspension in DI water was stored in refrigerator at 4°C. Concentration of PLA in the suspension was determined using gravimetric analysis. The number of nanoparticles per ml of the suspension was then calculated using the size of nanoparticles determined from DLS experiments (see below) and assuming PLA density of 1.25 g/cm3 [297].
5.2.2. Fluorescent labeling of proteins
Alexa Fluor succinimidyl ester dyes (Alexa Fluor 350 and Alexa Fluor 594) were used for fluorescent labeling of the antibody and lysostaphin, respectively. The dyes react with the primary amines of proteins at pH 7.5–8.5 yielding stable conjugates [298]. Approximately
1 μg of Alexa Fluor 350 (ε 19,000) was dissolved in 100 μl of dimethylformamide and 25
μl of the solution was then added to 2 ml of a 1 mg/ml solution of S. aureus antibody in 10
80 mM PBS (10 mM phosphate; 140 mM NaCl, 3 mM KCl). A similar procedure was performed for labeling of lysostaphin using Alexa Fluor 594 (ε 73,000). The samples were incubated at 25°C for 2 h, diluted two fold in 10 mM PBS and then centrifuged at 3,000 rcf for 1 h using a Nanosep® membrane centrifugal device (MWCO 3K and 30K for lysostaphin and anti-S. aureus antibody, respectively) to remove any unbound dye. As per the manufacturer‟s instructions, the molecular weight cutoff (MWCO) of approximately 6 times less than the molecular weight of the protein to be retained was chosen for purification to achieve maximum sample recovery.
5.2.3. Protein adsorption to PLA nanoparticles
Adsorption of Alexa 594 labeled lysostaphin with or without Alexa 350 labeled S. aureus antibody to PLA nanoparticles was studied at different NP: lysostaphin and NP: lysostaphin: antibody ratios. The ratios used for the preparation of lysostaphin- nanoparticle conjugates corresponded to 1300, 2700 and 5400 lysostaphin molecules per nanoparticle. For protein adsorption, 900 μl each of lysostaphin solutions with concentrations of 5.68, 11.79 and 23.59 μg/ml were prepared in 1 mM PBS buffer from a
1.25 mg/mL lysostaphin stock solution and combined with 100 μl of the NP stock solution containing 9.8*1011 particles per ml.
The ratios for targeted lysostaphin-antibody-nanoparticle conjugates consisted of
1300: 450; 2700: 900 and 5400: 1800 lysostaphin and antibody molecules per nanoparticle, respectively. Three protein solutions containing both lysostaphin and antibody (900 μl each) were prepared in 1 mM PBS buffer from lysostaphin and antibody stock solutions. Lysostaphin and antibody concentrations for each of the ratios consisted of 5.68 : 10.9 μg/ml for sample 1300: 450; 11.79 : 21.9 μg/ml for sample 2700: 900; and
81
23.59 : 43.8 μg/ml and 87.6 μg/ml antibody for sample 5400: 1800. 100 μl of the nanoparticle stock solution (containing 9.8*1011 particles per ml) was added to each of the protein solutions to make up the final volume of 1 ml.
All sample mixtures were then incubated for 6 h in a shaker at 150 rpm at 20°C.
The samples were then centrifuged at 3,000 rcf for 1 h to remove any unbound enzyme and antibody. The supernatant was collected and the pellet was resuspended in 1 ml of 10 mM PBS and stored at 4 °C. The concentration of unbound enzyme and antibody in each of the supernatants was determined from fluorescence measurements. Enzyme and antibody solutions with known concentrations were used as the standards. The concentrations of bound enzyme and antibody in the resuspended pellets were found from the difference between the initial concentration and concentration in the supernatant. The signal-to-background ratio was ~ 500: 1 for the fluorescently labeled lysostaphin (with or without antibody) and ~ 1000:1 for fluorescently labeled antibody. Protein adsorption experiments were reproduced in triplicates.
A second batch of protein-NP conjugates was also synthesized using newly prepared NPs to verify batch-to-batch reproducibility of antibacterial activity. Unlabeled proteins were used for the preparation of the second batch; otherwise all synthesis procedures were identical to those described above. Particle sizing and zeta potential measurements were also performed for the second batch.
5.2.4. Particle Size Measurements and Zeta potential analysis
Stock solutions containing nanoparticles or protein-nanoparticle conjugates were diluted
1:1000 in 10 mM PBS in a disposable cell and particle size was measured using dynamic light scattering with a 90Plus particle size /zeta potential analyzer (BrookHaven
82
Instruments Corporation, Holtsville, NY). Before particle sizing, the samples were sonicated for 20 min to break up any loosely-held agglomerates. Particle size was expressed as a combined mean diameter in nm (+/- SD). For zeta potential analysis, stock solutions were diluted 1:100 in 1 mM PBS in a disposable cuvette and zeta potential was measured using the 90 Plus analyzer. Zeta potential was expressed in millivolts (mV) as a combined mean of (+/- SD).
5.2.5. Bacterial turbidimetric activity assay
A cell suspension of S. aureus was prepared by inoculating 10 ml of 30% (w/v) tryptic soy with 100 µl of S. aureus culture prepared according to ATCC instructions. The procedure used was similar to the one employed by Schindler and Schuhardt [299]. The cells were incubated at 37C for 18 - 24 h under gentle shaking to grow to a mid-log phase. The cells were then pelleted at 2,000 rpm for 10 min and washed twice with 10 mM
PBS. The cells were then resuspended in PBS to prepare bacterial suspension with optical density (OD600) of 0.56 at 600 nm (1 cm light path). The turbidity assay was performed by adding 100 µl of the diluted sample solution in a 96 well microplate followed by 100µl of thus prepared S. aureus cell suspension. Lysostaphin concentration was fixed for all samples and consisted of 0.83 g/ml. The decrease in OD600 was monitored at 37°C in 1 min intervals for a total period of 3 h. These kinetic data were fitted exponentially, in assumption of the first order kinetics with respect to bacteria. The mean kinetic constants determined from these exponential fits were used as the measure of lysostaphin activity in a particular sample. Control experiments were performed with free lysostaphin at the same
83 enzyme concentration. All kinetic experiments were reproduced in triplicate for two different batches of samples synthesized independently (n=6).
5.2.6. Fractional analysis of enzyme activity in leached samples
To determine whether antibacterial activity of the enzyme-NP conjugates comes primarily from adsorbed or leached enzyme, we performed fractional analysis of enzyme activity. Two samples corresponding to the highest lysostaphin and antibody coatings
(1:5400:0 and 1:5400:1800 NP: lysostaphin: antibody ratios) were prepared using the procedure described above. The samples were then incubated in PBS buffer at 37C with gentle shaking. 100 l of each sample was taken at each of the different time points (1, 6 and 24 h) and centrifuged at 3,000 rcf for 1 h followed by collection of the supernatants. A cell suspension of S. aureus was grown to a mid-log phase similarly to the procedure described previously, centrifuged at 2,000 rpm for 10 min, washed twice and resuspended in 10 mM PBS. 100 µl of the cell suspension was added to wells of a 96 well plate containing either 100 µl of the total sample or 100 µl of the supernatant collected at a particular time point (leached aliquot). The samples were incubated with the cell suspension at 37ºC under continuous shaking, and the rate of bacterial lysis was monitored for 3 h, similarly to the procedure described in the previous section. From these experiments kinetic constants for bacterial lysis were determined using the exponential fits and used as the measure of leached lysostaphin activity at different time points. All kinetic experiments were reproduced in triplicates.
5.2.7. Statistical analysis
84
Data are presented as means±SD. Paired t-test and one-way analysis of variance
(ANOVA) was performed for the binding efficiency data of proteins on NPs and kinetics constants in the bacterial lysis assay. The null hypothesis was that the means are not significantly different. Tukey‟s test was used to compare different samples in the event that the ANOVA null hypothesis was not true. Data were analyzed using Origin 6 Data
Analysis and Graphing Software (OriginLab Corporation, MA, USA). A p value of <0.05 was considered statistically significant.
5.3. Results and Discussion
5.3.1. Characterization of enzyme-NP conjugates
Lysostaphin and anti S. aureus antibody was adsorbed onto biodegradable PLA
NPs (Figure 5.1). We found that adsorption was preferred method of immobilization in this case because it resulted in strong protein attachment and high activity without the need for additional chemical modification steps. Also, adsorption can be carried out in milder chemical conditions, generally resulting in better protein stability and is more cost effective. Adsorption has been successfully used previously to co-immoblize two proteins on PLA NPs for in vivo applications [300]. Leaching studies discussed in detail in the following section were performed to demonstrate that major part of antibacterial activity is associated with the enzyme conjugated to NPs. Samples with several enzyme : antibody : nanoparticle ratios were prepared (Table 5.1). The initial protein concentrations were chosen to correspond to approximately 0.5, 1, and 2 monolayers for lysostaphin and antibody on the NPs, respectively.
85
Figure 5.1. Schematic of the synthesis of protein-NP conjugates.
The following assumptions were used when calculating amount of a protein required for a monolayer coating. The average dimensions of the lysostaphin molecule are
4.5 nm x 5.5 nm x 8.5 nm [301]. The average surface area occupied by an adsorbed lysostaphin molecule based on the smallest dimensions was therefore assumed to be 25 nm2. The size of uncoated PLA NPs measured using dynamic light scattering was found to be 145 nm (Table 5.2). A theoretical monolayer of lysostaphin on the surface of a 145 nm
PLA nanoparticle calculated from the smallest dimensions of the lysostaphin molecule should consist of ~2,700 molecules per NP. The average dimensions of an IgG antibody molecule reported in the literature vary greatly from 14 nm x 8.5 nm x 4 nm to 23 nm x 23
86 nm x 4.4 nm [302]. We therefore used hydrodynamic radius, Rh, of IgG molecule for estimation of its theoretical monolayer coating. Rh can be determined more accurately and its value reported for IgG molecule in aqueous solution, Rh=5.3 nm, is consistent in the literature [303]. A theoretical monolayer of antibody calculated in the assumption of dense hexagonal packing of the spheres with the radius of 5.3 nm would consist of ~900 molecules per NP.
A
87
B
Figure 5.2. Binding yield of adsorption on PLA NPs for (A) AlexaFluor 594-labeled lysostaphin (Black squares – Plain lysostaphin and Red circles – lysostaphin coadsorbed in the presence of antibody (n = 3)), and (B) Alexa Fluor 350 labeled anti-S. aureus antibody (n = 3)
Binding yields for lysostaphin consisted of 40-44% for plain lysostaphin-coated samples and 29-38% when lysostaphin was co-adsorbed with the antibody (Table 5.1).
Binding yield for the antibody consisted of 15-23% for different ratios. Overall, amount of adsorbed lysostaphin and the antibody increased linearly with the increase of the concentration of the corresponding protein (Figure 5.2 A, B). For lysostaphin adsorption at the highest initial enzyme concentrations (which corresponded to two monolayers of lysostaphin), coating by the adsorbed enzyme approached theoretically predicted monolayer and consisted of 85% and 77 % of a monolayer for plain and antibody-coated samples, respectively (Table 5.1). At the same time, antibody coating at its highest initial concentration consisted of ~30% of theoretical monolayer. This is an assumption that is based on theoretical dimensions of lysostaphin and antibody molecules adsorbed on
88 nanoparticles. However, the actual monolayer coating could be higher as enzymes are known to unfold upon conjugation resulting in a slightly higher specific surface area occupied by each enzyme molecule.
Initial Ratio, protein Number of theoretical Binding yield Fraction of protein molecules per monolayers assuming of protein, % monolayer adsorbed, nanoparticle all added protein %
were adsorbed
1NP : 1300 LS 0.5 LS 402 LS 201 LS
1NP : 2700 LS 1 LS 445 LS 455 LS
1NP : 5400 LS 2 LS 423 LS 856 LS
1NP :1300 LS : 450 Ab 0.5 LS, 0.5 Ab 293 LS; 234 Ab 151 LS; 122 Ab
1NP :2700 LS : 900 Ab 1 LS, 1 Ab 383 LS; 177 Ab 391 LS; 177 Ab
1NP :5400 LS : 1800 2 LS, 2 Ab 392 LS; 151 Ab 763 LS; 302 Ab Ab
Table 5.1. Characteristics of samples with different nanoparticle (NP): lysostaphin (LS): antibody (Ab) ratios. Binding yield of protein on NPs. Data are presented as meanSD
(n=3).
Overall lower binding yields for antibody compared to lysostaphin could be attributed to stronger electrostatic interaction between the highly positively charged lysostaphin molecules (pI = 10.6) and negatively charged PLA NPs (zeta potential -48 mV). For comparison, pI of IgG lies in the range 6.4-9.0 and because of the polyclonal
89 nature of antibody, molecules with different pIs from this range are expected to be present in the solution [304].
Size was determined by light scattering for bare NPs and protein-NP conjugates
(Table 5.2). The mean diameter of the enzyme-NP conjugates increased significantly
(p<0.05) upon adsorption of lysostaphin with or without antibody to the surface of NPs.
Zeta potentials for all conjugates were highly negative (<-30 mV) indicating that the conjugates, similarly to bare NPs, were negatively charged. The mean values from two different batches are tabulated in Table 5.2 below.
Ratio, protein Number of Size (nm) Zeta potential (mV) molecules per theoretical (Mean SD) (Mean SD) nanoparticle monolayers if all added protein molecules were adsorbed
1NP : 1300 LS 0.5 LS 177.4 ± 6.8 -39.7 ± 11.6
1NP : 2700 LS 1 LS 182.6 ± 6.8 -39.6 ± 13.2
1NP : 5400 LS 2 LS 187.3 ± 4.7 -42.6 ± 10.3
1NP :1300 LS : 450 Ab 0.5 LS, 0.5 Ab 182.7 ± 1.8 -45.3 ± 1.9
1NP :2700 LS : 900 Ab 1 LS, 1 Ab 193.3. ± 7.8 -41.1 ± 4.1
1NP :5400 LS : 1800 Ab 2 LS, 2 Ab 180.9 ± 9.8 -33. ± 1.4
Uncoated NPs 148.1 ± 4.1 -49.5 ± 2.1
Table 5.2. Characteristics of samples with different nanoparticle (NP):lysostaphin
(LS):antibody (Ab) ratios - Size and zeta potentials for all samples.
90
5.3.2. Fractional enzyme activity from leaching study
It is important to know whether activity of enzyme coated NPs is due to immobilized or leached lysostaphin. Fluorescence measurements were found to be not sensitive enough to detect enzyme leached from the NPs into the supernatant. Therefore, turbidimetric bacterial lysis assay was employed to determine the fraction of total activity contributed by leached lysostaphin. This method was found to be more sensitive because even extremely small concentrations of lysostaphin can cause significant lysis of bacteria in S. aureus suspension.
Figure 5.3 shows the percent of leached enzyme activity for the aliquots collected at different time points compared against total activity for two samples with highest lysostaphin loading. These results indicate that less than 10% of enzyme was desorbed from the NPs after 24 h incubation in PBS buffer. Shah et al. studied the antimicrobial activity of lysostaphin adsorbed on two different plastic surfaces, polystyrene (well plates) and fluorinated ethylene-propylene polymer (a Teflon-like material used in Angiocath catheters) [240]. Similarly to our findings, they observed that the antibacterial activity was primarily due to immobilized and not leached enzyme.
91
Figure 5.3. Fraction of leached lysostaphin activity for aliquots collected at different time intervals compared against total enzyme-NP activity for samples with highest lysostaphin and antibody coatings. Black squares – NP: lysostaphin
(1:5400:0) and red circles - NP: lysostaphin: antibody (1:5400:1800); n=3
5.3.3. Bacterial lysis assay – Enzyme activity
Enzyme activity was measured as reduction in turbidity of S. aureus suspension in
PBS buffer by monitoring the decrease in optical density at 600 nm. Lysostaphin concentration was 0.83 µg/ml in all experiments. Enzyme concentration was thus fixed in all kinetic experiments; however the number of NPs (and, where applicable, the antibody concentration) was different because of different number of lysostaphin molecules adsorbed per NP for different ratios.
92
Final Final Kinetic Significance Significance Lysostaphin antibody constant of compared compared concentration concentr bacterial lysis against free against Sample ( µg/ml) ation in (n=6; 3 lysostaphin; enzyme:NP assay replicates from n=6; analogue (µg/ml) 2 different paired t- lacking the batches) test) antibody; n = 6; paired t-test NP:lyso (1:1300) 0.83 - p < 0.05 0.007 ± 2E-03
NP:lyso (1:2700) 0.83 - 0.0136 ± 1E-03 p < 0.05
NP:lyso (1:5400) 0.83 - 0.0147 ± 2E-03 p < 0.05
NP:lyso:Ab (1:1300:450) 0.83 1.27 0.0088 ± 1E-03 p > 0.05 p > 0.05
NP:lyso:Ab (1:2700:900) 0.83 0.68 0.0175 ± 6E-03 p < 0.05 p > 0.05
NP:lyso:Ab (1:5400:1800) 0.83 0.6 0.032 ± 7E-03 p < 0.05 p < 0.05
Free lysostaphin 0.83 - 0.009 ± 1E-03 -
Table 5.3 - Bacterial turbdimetric assay results for all the samples.
Figure 5.4 below shows results of the activity assay for two batches of enzyme-NP conjugates and free enzyme (n=6). Note: The high error bars in NP-enzyme samples are also indicative of interference of NPs during absorbance reading at 600 nm. Assuming first order kinetics with respect to bacteria, the time dependencies were fitted exponentially and the mean kinetic constants of the first order reaction were calculated from the exponential fits and used as the measure of antimicrobial activity. Table 5.3 shows results of the analysis of kinetic data obtained from two different batches (n=6; 3 replicates from each batch). Comparison of the activities for two sets of the samples with the lysostaphin loadings (1:1300:0 and 1:1300:450; 1:2700:0 and 1:2700:900) showed no significant difference (p>0.05, paired t-test; n=6, see Table 5.3), despite a higher mean rate of bacterial lysis for antibody coated samples. Activity of the sample with the lowest
93 lysostaphin loading that lacked the antibody (1:1300:0) was found to be significantly lower than that of the free enzyme (p<0.05, paired t-test; n=6, see Table 5.3). For its corresponding analogue that contained the co-adsorbed antibody (1:1300:450), activity was not found to be significantly different from that of the free enzyme (p>0.05, paired t- test; n=6, see Table 5.3).
However, enhanced activity compared to the free enzyme was observed for both enzyme-coated and enzyme-antibody coated nanoparticles for all other ratios (p<0.05, paired t-tests; n=6, Figs. 5.4B and 5.4C, Table 5.3). The rate of bacterial lysis for the highest antibody coating was (corresponding to an initial bilayer of molecules to NP) was the highest amongst all samples and significantly different compared to its corresponding analogue without antibody (Table 5.3). It had a nearly fourfold faster mean rate of bacterial lysis compared to free enzyme at the same concentration. These experiments demonstrate the possibility of antibody-directed targeting for samples with the highest antibody coating concentration. We hypothesize that co-immobilization of lysostaphin and antibody on nanoparticles leads to enhanced binding to bacteria through antibody-antigen interactions resulting in higher local concentrations of lysostaphin and therefore leading to increased bacterial lysis rates. However, enhanced potency can only be achieved at higher antibody and lysostaphin loadings warranting further studies of the mechanism of the observed phenomenon. Results of this model study suggest the possibility of regulating properties of enzyme-NP conjugates by optimizing the ratio between the functional enzyme and the targeting antibody attached to a nanoparticle. Targeting can be enhanced by increasing the number of antibody molecules per NP (Figure 5.5B), even though total antibody concentration in the assay decreases for samples with higher lysostaphin and
94 antibody loading (Table 5.3). Overall, bacterial lysis rates increased with the increased number of antibody molecules per NP (Figure 5.5B). This increase was found significant for all samples with different antibody: NP ratios (p<0.05, One way ANOVA).
An unexpected result was the observation of enhanced activity of lysostaphin- coated nanoparticles compared to the free enzyme. In spite of the fact that lysostaphin concentration in the assay was kept constant, bacterial lysis rates increased with the increased number of lysostaphin molecules per NP (Figure 5.5A). This increase was found to be statistically significant for all samples with different enzyme:NP ratios (p<0.05, One way ANOVA). For two samples with higher lysostaphin loading (1:2700:0 and 1:5400:0), activity was significantly higher than that of the free enzyme (p<0.05, paired t-test, n=6), while for sample with the lowest lysostaphin loading activity was significantly lower than that of the free enzyme (p<0.05, n=6).
95
A B
C
Figure 5.4. Results of bacterial lysis assay. The data are the means of absorbance for six replicates from two batches along with the standard deviations for each time point. Blue triangles correspond to untreated bacterial suspension
(negative control). Black squares – lysostaphin adsorbed on PLA nanoparticles; red circles – lysostaphin and antibody co-adsorbed on PLA nanoparticles, and green triangles – free lysostaphin. (A) – NPs with low lysostaphin and antibody coatings (1:1300:0 and 1:1300:450); (B) – NPs with intermediate lysostaphin and antibody coatings (1:2700:0 and 1:2700:900); and (C) – NPs with high lysostaphin and antibody coatings (1:5400:0 and 1:5400:1800). High error bars in NP-enzyme samples are also indicative of interference of NPs during absorbance reading at
600 nm.
96
In general, partial loss of enzymatic activity is expected for surface-immobilized enzymes including enzyme-nanoparticle conjugates [263, 277]. In our case, decrease of enzymatic activity was only observed for the conjugates with the lowest lysostaphin loading, while enhanced activity was observed for the other ratios. We have previously observed enhanced enzymatic activity for lysozyme conjugated to positively-charged polystyrene latex NPs. This effect was attributed to charge-directed targeting to negatively charged bacteria (M. Lysodeikticus).
We considered the possibility that enhanced activity of lysostaphin-coated NPs observed in the present work was also due to charge-directed targeting since lysostaphin is positively charged and its attachment to nanoparticles could have lead to the formation of positively charged conjugates. However, all conjugates were found to have highly negative zeta-potential (Table 5.2). Also of note is that charge-directed targeting reported by our group [250] was only observed in an environment with low ionic strength (6 mM phosphate buffer, no saline) and was not noticeable at physiologically relevant ionic strength (PBS buffer, ~150 mM saline). On the contrary, lysostaphin-coated NPs prepared in the present work were robust and showed enhanced activity compared to the free enzyme in PBS buffer containing 150 mM saline. Our hypothesis is therefore that enhanced activity of lysostaphin-coated NPs in the absence of co-immobilized antibody occurs due to multiple-ligand targeting. The mature form of lysostaphin consists of two domains, the endopeptidase domain responsible for its catalytic activity, and the C- terminal cell wall-targeting domain (CWT) [305]. Lysostaphin-NP conjugates studied here carry from several hundred to several thousand lysostaphin molecules per nanoparticle
(Figure 5.5A). Therefore, each particle presents a large number of CWT domains available
97 for binding to bacteria, which can lead to considerably higher binding constant for interaction between the NP conjugates and bacteria, compared to that for interaction between free lysostaphin and bacteria.
A
B
98
Figure 5.5. (A) Kinetic constant of bacterial lysis plotted as function of adsorbed lysostaphin molecules per NP for lysostaphin:NP conjugates lacking the antibody
(mean±SD; n=6). (B) Kinetic constant of bacterial lysis plotted as function of adsorbed antibody molecules per NP (mean±SD; n=6). Note that total antibody concentration in the assay decreases with the increase of the number of antibody molecules per NP (see Table 5.3).
Interestingly, enzyme and enzyme/antibody- coated nanoparticles remain negatively charged. This finding can have important implications because toxicity of many antimicrobial peptides to eukaryotic cells can potentially be reduced by their conjugation to negatively charged carriers. High toxicity of antimicrobial peptides remains one of the most important concerns in their in vivo applications [306]. Use of negatively charged conjugates is expected to reduce non-specific attachment of antimicrobial peptides to slightly negatively charged surface of eukaryotic cells therefore reducing their toxicity.
In conclusion, we demonstrated that lysostaphin-coated nanoparticles show enhanced antimicrobial activity against S. aureus at physiologically relevant conditions.
We also showed that co-adsorption of anti-S. aureus antibody to lysostaphin-coated nanoparticles leads to further increase of their antimicrobial activity. Results of this study indicate that enhanced antimicrobial activity of enzyme-nanoparticle conjugates is due to targeted delivery to bacteria.
99
CHAPTER 6
EVALUATION OF ANTIMICROBIAL ACTIVITY OF LYSOSTAPHIN-COATED HERNIA REPAIR MESHES
6.1. Introduction
The development of an incisional hernia is a common complication after abdominal surgery which results in 90, 000 ventral hernia repair surgeries per year in the
United States [307]. The implantation of a prosthetic mesh is a well-established procedure for reconstructing or reinforcing the abdominal wall and has been shown to decrease the rate of hernia recurrence [308-310]. However, one of the most common problems associated with the use of prosthetic meshes is bacterial infection. The incidence rate for mesh-related infection has been reported to vary between 1% and 18% in different clinical studies [311-313].
Mesh infection is associated with significant morbidity and is increased considerably in patients with diabetes, immunosuppression, and obesity [314]. Infection usually occurs in the early postoperative period and while its incidence can be decreased by the use of antibiotics [315], no significant improvement has been made in the reduction of mesh infection for more than 20 years. Current conventional therapy has proven to be largely unsuccessful due to implantation of a foreign body, wide tissue dissection and the development of a number of multi-drug resistant strains. In a study of mesh-related infections following incisional herniorrhaphy, 63% of the microorganisms isolated were
Methicillin-resistant Staphylococcus aureus (MRSA)[313].
Bacterial adhesion and proliferation to implanted biomaterial surfaces is a key step in the pathogenesis of infection [316-318]. Multiple experimental and clinical studies have
100 been focusing on the use of different antimicrobial agents immobilized on the surface or released from the bulk of an implant [319]. Increasingly, attention is being paid to surface coatings on implants that contain active compounds that may either kill directly on contact or via leaching over time. Naturally occurring antimicrobial peptides and enzymes have recently attracted much attention because of their high activity, broad antimicrobial spectrum, and the low rate of antimicrobial resistance.
Balaban et.al found that adsorption of an antibacterial peptide, dermaseptin to
Dacron vascular grafts resulted in increased resistance to bacterial colonization and demonstrated its in vivo efficacy in a rat model [239]. Shah et.al demonstrated significant antimicrobial activity against S. aureus for intravenous catheters coated by physically adsorbed lysostaphin, and observed high efficacy in vitro [240]. Lysostaphin possesses extremely high activity against a variety of staphylococcal strains including MRSA [240].
Based on literature data lysostaphin has a reported minimum inhibitory concentration at
which 90% of the strains are inhibited [MIC90] of 0.001 to 0.064 µg/ml [320]. This is comparatively much lower than MIC90 for most broad spectrum antibiotics and drugs that are antimicrobial against S. aureus. For comparison, MIC90 for vancomycin, which is currently recommended for treatment of methicillin-resistant S. aureus infections, is 2
µg/ml [321]. Lysostaphin could therefore be superior to other antibacterial agents in hernia repair-related infections given that up to 90% of mesh-related infections result from S. aureus. Here, we evaluate the in vitro antimicrobial susceptibility of S. aureus to lysostaphin-coated hernia repair polymer meshes and study the effect of enzyme coating concentration on antimicrobial activity
101
6.2. Materials and methods
Staphylococcus aureus (ATCC 32370) was purchased from American Type Culture
Collection (Manassas, VA). Lysostaphin (25 kDa; LSPN-50) isolated from
Staphylococcus staphylolyticus was bought from Ambi Products LLC (Lawerence, NY).
Bovine Serum Albumin (BSA) was bought from Calbiochem (San Diego, CA). Ultrapro
(Lightweight Polypropylene mesh with Monocryl weave) from Ethicon Inc, Johnson and
Johnson (Langhorne, PA) and PariatexTM Composite mesh (Polyester mesh with resorbable film) from Covidien (Mansfield, MA) was donated by Carolinas Medical
Center (Charlotte, NC). Alexa Fluor 594 Carboxylic acid, Succinimidyl ester dyes used for fluorescent labeling were purchased from Invitrogen, Molecular Probes (Carlsbad,
CA). Microsep Centrifugal Devices with a modified Polyethersulfone membrane (low protein binding; MWCO - 3K) used for enzyme purification were purchased from Pall
Life Sciences (Ann Arbor, MI). All other reagents were purchased from Fisher Scientific
(Pittsburgh, PA), and used without further purification unless otherwise specified. Type 1
Clear Borosilicate glass vial (25 mL) was with attached screw caps was bought from
VWR (West Chester, PA).
6.2.1 Fluorescent labeling of lysostaphin
Alexa Fluor 594 dye was used for fluorescent labeling of lysostaphin due to its enhanced resistance to photobleaching [298]. Fluorescent labeling is considered to be an acceptable method for monitoring protein concentration, which is assumed to be proportional to the fluorescent signal [322]. Labeling of lysostaphin by Alexa Fluor 594 was performed according to the manufacturer‟s instructions. Approximately 5 μg of Alexa Fluor 594 (ε
102
73,000) was dissolved in 500 μL of dimethylformamide and 100 μL of this solution was added to 2 mL of a 1 mg/mL solution of lysostaphin in 10mM PBS buffer (10 mM phosphate; 140 mM NaCl, 3 mM KCl). The enzyme was incubated with dye at 30°C for 2 h, then diluted two-fold in 10 mM PBS buffer and centrifuged at 3,500 g for 1 h using a
Microsep membrane centrifugal device (MWCO 3K) to remove any unbound dye followed by resuspension of the concentrated labeled enzyme in PBS buffer.
6.2.2. Measurement of the specific surface area of the mesh samples
The specific surface area determined for one of the pieces by Brunauer-Emmett-Teller
(BET) analysis of N2 adsorption at 77 K in a Micromeritics ASAP 2020 (Micromeritics
2 Instrument Corporation, Norcross, GA) was found to be 5.5 m /g. Prior to N2 adsorption, the mesh samples were degassed overnight at 80 °C. Surface area for each of the mesh pieces used in the following experiments was calculated based on the weight of the piece.
The average weight of a 1 × 1 cm2 Ultrapro mesh piece was 6.9±1.0 mg.
6.2.3. Preparation of lysostaphin-coated meshes
Ultarpro, Ethicon Inc., mesh (Lightweight Polypropylene mesh with Monocryl weave) was cut into ~ 1 × 1 cm pieces in a laminar flow hood under sterile conditions prior to physical adsorption of enzyme. Initial enzyme concentrations of 0.01, 0.025, 0.05, 0.1,
0.25, and 0.5 mg/ml in PBS buffer were prepared from a 1 mg/mL stock solution of Alexa
Fluor 594-labeled lysostaphin. The initial fluorescence intensities of the enzyme sample solutions (1 mL) were measured in a 12 well plate (Nunc Inc.) using a microplate reader
(Ex 594 nm; Em 625 nm) and then added to 25 mL sterile glass vials. Using a pair of sterile tweezers, mesh pieces were gently placed into each of the vials containing the
103 enzyme solutions and incubated overnight at room temperature with gentle shaking (100 rpm). The enzyme solution over the mesh was then collected and stored for fluorescence measurements, and the mesh was gently washed 2 times with 1 mL of PBS buffer. The wash solution was also collected and used in determination of enzyme binding yield. In order to remove any loosely adsorbed enzyme, 1 mL of 0.1 (v/v %) Tween 20 solution
(non ionic surfactant) was then added to each of the glass vials followed by incubation for
3 h. This surfactant solution was also collected and used in the determination of the amount of desorbed enzyme and the mesh samples were then washed with copious amount of PBS buffer.
The concentration of unbound enzyme in each of the supernatants and wash solutions was determined from fluorescence measurements. Initial enzyme solutions with known concentrations were used as the standards. The concentration of the unbound/desorbed enzyme at each step was then calculated and subtracted from initial concentration of enzyme present in the initial solution. The difference corresponded to the bound enzyme concentration on the mesh. Adsorption experiments were reproduced in triplicates. Note: Unlabeled lysostaphin was used in all other experiments except binding efficiency studies.
6.2.4. Leaching studies
Leaching of non-specifically bound lysostaphin from the surface can occur both in vitro and in vivo. To model in vivo conditions, we studied enzyme leaching from lysostaphin-coated meshes in the presence of BSA. The samples were incubated with 1 mL of 2 % (w/v) BSA solution in PBS buffer for 24 h at 37C. We found that fluorescence measurements lacked sensitivity to determine concentration of leached enzyme in the
104 samples. We therefore estimated enzyme concentration based on its activity. For kinetic measurements, 0.1 ml aliquots from each of the samples were taken at different time points (5, 15 and 24 h) and added to a 96 well plate. Lysostaphin solutions with known enzyme concentrations of 0.0122 – 25 µg/ml were used as standards for each of the different time points. A 100 µl cell suspension of S. aureus in PBS buffer with optical density (OD600) of 0.3 at 600 nm (1 cm light path) was then added to each of the different wells and the rate of bacterial lysis was monitored continuously for 4 h at 37ºC.
Enzymatic activity for standards and unknown samples was measured simultaneously in the same microplate. The initial rate of the reaction for each of the standards and unknowns was calculated from the linearized slopes. The standard curve was obtained by plotting the initial rate against the enzyme concentration. From the standard curve, the unknown enzyme concentrations were determined and the amount of leached enzyme for each of the samples was plotted as a function of time. All kinetic experiments were reproduced in triplicates.
6.2.5. Turbidimetric assay
A cell suspension of S. aureus was prepared by inoculating 10 mL of 30% (w/v) tryptic soy with 100 µL of S. aureus culture prepared according to ATCC instructions. The procedure used was similar to the one employed by Schindler and Schuhardt [299]. The cells were incubated at 37C for 18 - 24 h under gentle shaking to grow to a mid-log phase. The cells were then pelleted at 3,000 g for 10 min and washed twice with 10 mM
PBS. The cells were then resuspended in PBS to prepare bacterial suspension with optical density (OD600) of 0.55 at 600 nm (1 cm light path). 1 mL of the suspension containing
~107 CFU (colony forming units) was added to 25 mL glass vials containing the mesh
105 samples. The samples were incubated with the suspension at 37ºC under continuous shaking, and the rate of bacterial lysis was monitored for 5 h by taking 0.2 mL aliquots and measuring the OD600 in a 96 well plate at different time intervals. Enzyme activity was expressed in units (one unit of lysostaphin activity is defined as a decrease of 0.01 absorbance units at 600 nm, according to the manufacturer‟s specifications). The vials containing the cell suspension were continuously shaken in order to maintain contact with the meshes. All kinetic experiments were reproduced in triplicates.
6.2.6. Fractional analysis of enzyme activity in leached samples
To determine whether antibacterial activity of the coated meshes comes primarily from immobilized or leached enzyme, we performed fractional analysis of enzyme activity. A set of mesh samples with the same initial coating concentrations was prepared in duplicates using the procedure as described previously. All the samples were incubated in
25 ml glass vials containing 0.5 mL of 2 % (w/v) BSA solution in PBS buffer for 24 h at
37C. From one set of the samples, the meshes were removed at different time intervals
(1, 4 and 24 h) and a corresponding set containing mesh samples of identical coating concentrations were left intact.
A cell suspension of S. aureus was grown to a mid-log phase similar to the procedure described previously, centrifuged at 3,000 g for 10 min, washed twice with 10 mM PBS and resuspended in PBS to a final optical density (OD600) of ~0.9 at 600 nm. 0.5 ml of this cell suspension was added to the 25 ml glass vials containing each set of the samples and the final OD of this mixture measured was 0.55 at 600 nm. The samples were incubated with the cell suspension at 37ºC under continuous shaking, and the rate of bacterial lysis was monitored for 2 h by taking 0.2 mL aliquots from the reaction mixture
106 and measuring the OD600 in a 96 well plate at different time intervals. This procedure was done for identical sets of the samples at the different time intervals to compare the enzyme activity of the leached sample against total activity of the mesh sample incubated in BSA.
6.2.7. Colony counting
The sample meshes were placed in sterile glass vials and challenged with an inoculum of 1 ml of S. aureus suspension in tryptic soy broth containing 6 x 107 CFU for 24 h at 37 ºC.
After the incubation, the mesh pieces were retrieved and the broth supernatant was collected for subsequent analysis. Each mesh piece was then washed once vigorously with
1 mL of PBS buffer using a pipette followed by vortex washing for 5 minutes and collected for subsequent colony counting analysis. Serial dilutions and spot plating were then performed for the broth supernatant (broth count) and the pooled wash (wash count) on agar nutrient plates. The wash count was used to dislodge and quantify the loosely attached bacteria. Bacterial counts were quantified as the number of CFU per mL. All counts were performed in triplicates and results were calculated as mean log reduction
±SD.
6.2.8. In vivo Rat Model
All in vivo studies were performed by our collaborators, Dr. Yuliya Yurko, Dr. Amy
Lincourt, and Dr. Todd Heniford at Carolinas Medical Center along with Dr. John Shipp at ViMedrx. A total of 40 male Lewis rats (225-275g, Charles River Laboratories,
Raleigh, NC) were used. All animal experiments were approved by the Institutional
Animal Care and Use Committee (CMC) and performed in accordance with NIH guidelines. Rats were randomly assigned to one of the ten groups (n=4 each group) as
107 follows. Groups A-F had polypropylene mesh implanted as an onlay to the anterior abdominal wall fascia. Group A received mesh with no lysostaphin and no S. aureus inoculum. Group B received no lysostaphin and S. aureus inoculum of 5x105 CFU. Group
C received mesh-bound lysostaphin (Initial coating concentration - 50 µg/ml) and no S. aureus inoculum Group D received mesh-bound lysostaphin (Initial coating concentration – 50 µg/ml) and S. aureus inoculum of 5x105 CFU. Group E received mesh-bound lysostaphin (Initial coating concentration – 100 µg/ml) and no S. aureus inoculum. Group F received mesh-bound lysostaphin (Initial coating concentration – 100
µg/ml) and S. aureus inoculum of 5x105 CFU. Groups G-J had polyester mesh implanted as an onlay to the anterior abdominal wall fascia. Group G received mesh with no
Lysostaphin and no S. aureus inoculum. Group H received no Lysostaphin and S. aureus inoculum of 5x105 CFU. Group I received mesh-bound Lysostaphin (Initial coating concentration – 100 µg/ml) and no S. aureus inoculum. Group J received mesh-bound
Lysostaphin (Initial coating concentration – 100 µg/ml) and S. aureus inoculum of 5x105
CFU.
6.2.9. Surgical Procedures
Surgical anesthesia was induced and maintained with inhaled isofluorane. The abdominal wall was shaved, prepared first with Betadine®, then isopropyl alcohol 70% (v/v), and draped in sterile fashion. A 1 cm midline vertical incision was made through dermis and subcutaneous pockets was then created bilaterally over the anterior abdominal fascia. A
3cm x 3cm square piece of sterile polypropylene mesh was placed on the fascia and secured with eight stitches of 4-0 Prolene suture (Ethicon Inc.). For groups assigned to receive the bacterial innoculum, a 1 cc suspension of 5 x 105 CFU S. aureus was applied
108 directly on the mesh using a sterile syringe. The incision was closed using vicryl suture and reinforced with skin staples. Topical Bitter Orange (ARC Laboratories, Atlanta, GA) was applied over the closure to dissuade wound disruption. All animals received bupranorphine (0.03mg/kg) immediately after surgery and every 12 hours thereafter for the next 48 hours.
Mesh explantation was performed on day 7 after implantation. General anesthesia was induced with isofluorane and the animals were euthanased by intracardiac pentobarbitol injection. The skin around the original incision was prepared in the same fashion as above and opened sharply and widely to allow full exposure of the graft implant. The entire abdominal wall including the mesh interface was excised in en-block fashion. For quantitative bacterial culture analysis the mesh was carefully separated from the muscle, agitated and washed five times with sterile phosphate buffer solution. This washings were serially diluted up to the final concentration of 10-8 and each dilution was plated on tryptic soy agar and allowed to incubate overnight at 37º C. Bacterial colonies were then counted.
6.2.10. Statistical analysis
Data are presented as means±SD. One-way analysis of variance (ANOVA) was performed for colony counts from the broth supernatant and wash to compare the samples with different initial enzyme concentrations. The null hypothesis was that the means are not significantly different. Tukey‟s test was used to compare different samples in the event the
ANOVA null hypothesis was not true. Data were analyzed using Origin 8 Data Analysis and Graphing Software (OriginLab Corporation, MA, USA). A p value of <0.05 was considered statistically significant.
109
6.3. Results
6.3.1. Binding yield of lysostaphin on Ultrapro
Binding yield was calculated based on the amount of fluorescently labeled enzyme adsorbed on mesh and corresponded to the difference in fluorescence intensities of initial enzyme and the supernatant solutions as described above. Figure 6.1 shows the binding yield for Alexa Fluor-labeled lysostaphin on a 1×1 cm Ultrapro mesh. The amount of adsorbed enzyme per unit BET surface area (µg/cm2) is plotted as a function of the initial enzyme concentration. Based on the above described experimental conditions, 18 to 40% of lysostaphin initially present in the solution remained bound to the mesh after treatment by Tween 20 solution.
0.40
0.35
) 0.30 2
0.25
0.20 Bound (ug/cm Bound 0.15
0.10
0.05
0.00 0 100 200 300 400 500 600 Initial lysostaphin concentration (ug/ml)
Figure 6.1. Binding yield for Alexa 594 labeled lysostaphin on Ultrapro (1 × 1 cm) after overnight adsorption at room temperature (n = 3).
110
The degree of surface coverage on the mesh can be predicted from the binding yield of enzyme. Average dimensions of the lysostaphin molecule are 4.5 nm X 5.5 nm X
8.5 nm. The average surface area occupied by an adsorbed protein molecule based on the smallest dimensions is 25.02 nm2. The maximum theoretical monolayer of lysostaphin adsorbed on a 1×1 cm piece of Ultrapro mesh calculated using the above surface area per molecule and the BET specific surface area of the mesh (3.8 x 1016 nm2) consisted of
1.5x1015 molecules. We observed approaching a theoretical 50%, 100%, and 150% monolayer coating of adsorbed lysotaphin at initial enzyme concentrations of 100, 250, and 500 g/mL, respectively. Typically, the amount of adsorbed enzyme on a surface increases sharply at low initial enzyme concentration, and reaches a plateau at higher concentrations, thereby approaching a certain limiting value for adsorption [323]. In our case, adsorption did not reach saturation even at the highest coating concentration of 0.5 mg/mL suggesting the possibility of multilayer adsorption at higher concentrations.
6.3.2. Lysostaphin leaching
Enzyme leaching could be a potential problem in the case of non-covalent adsorption as the functional protein may be replaced on the surface by more abundant nonfunctional ones in vivo. Also, the Ultrapro mesh is constructed of a monofilament lightweight large porous polypropylene with pores larger than 3 mm which could result in initial confinement of the enzyme followed by release in vivo [324]. In our study, we assessed lysostaphin leaching in the presence of 2 % (w/v) BSA as a model for the most abundant protein in the abdominal fluid. Figure 6.2 shows the standard curve representing the rate of lysis for the standards with known enzyme concentrations. From this standard
111 curve, the unknown enzyme concentrations were determined and the amount of leached enzyme for each of the samples was plotted as a function of time.
Figure 6.2. Standard curve plotting the mean rate of lysis (OD per min) against the different known lysostaphin concentrations (µg/ml) (n=3).
The samples designated based on their initial lysostaphin concentrations of
10µg/ml, 25µg/ml, 50 µg/ml, 100µg/ml, 250µg/ml and 500µg/ml were L-10, L-25, L-50,
L-100, L-250 and L-500 respectively. We observed an overall (0.01- 3) % leaching of enzyme based on the initial coating concentration from all the studied samples after 24 h incubation in BSA. Bacterial lysis was observed in all mesh samples including those that leached very low amounts of lysostaphin at the different time points, suggesting that small concentrations of enzyme could be effective in inhibiting growth of cells. In vitro antimicrobial activity against S. aureus was evaluated for all of the above samples (L-10 through L-500) using turbidity assay and colony counting method. Figure 6.3 below shows leaching for all different coating concentrations after 24 h.
112
Figure 6.3. Leaching of adsorbed lysostaphin as a function of time for mesh samples treated by different initial enzyme concentrations. Black squares -
10µg/ml, Red circles -25µg/ml, Green triangles - 50µg/ml, Blue inverted triangles -
100µg/ml, Maroon diamonds - 250µg/ml and Pink arrows - 500µg/ml. Leaching studies were performed in the presence of 2 % (w/v) BSA in PBS buffer (pH = 7.4) at 37C.
6.3.3.Turbidimetric antibacterial activity assay
Activity assay provides the most direct measure of an enzyme‟s functionality. Qualitative evaluation of lysostaphin activity was done by monitoring the cell lysis of a S. aureus cell suspension [325]. The degree of cell lysis is directly proportional to the decrease in OD600 of the bacterial suspension.
113
Figure 6.4. Turbidimetric enzyme activity assay for the different lysostaphin- coated hernia mesh samples. Black squares - 10µg/ml, Red circles -100µg/ml,
Blue triangles - 500µg/ml and Green inverted triangles - Unocoated mesh; (n=3); p < 0.05 (one way ANOVA; n=3).
We monitored the decrease in optical density of the bacterial suspension and compared the rate of lysis for the samples with different enzyme coatings. Figure 6.4 shows the time course of bacterial degradation for three of the samples along with the appropriate controls. The vials containing the cell suspension were continuously shaken in order to maintain constant contact with the meshes.
114
Initial coating Sample concentration Enzyme activity (µg/ml) (Units/ml)1
(mean±SD)
L - 10 10 2.97 ± 0.8
L - 25 25 3.08 ± 0.8
L - 50 50 3.78 ± 1.1
L - 100 100 4.02 ± 0.01
L - 250 250 5.92 ± 0.3
L - 500 500 6.98 ± 0.3
P - 100 100 6.98 ± 0.3
Table 6.1 - Specific enzyme activity of samples with different enzyme coating concentrations
6.3.4. Leached enzyme activity assay
It is important to know whether activity of coated meshes is due to immobilized or leached enzyme. A turbidity assay that compared the rates of leached sample against that of the total activity of enzyme-coated mesh at different time points (incubated in bovine serum albumin) was studied to calculate the fraction of activity contributed by lysostaphin leaching from the mesh surface into the cell suspension. Enzyme leaching during the first
5 h (which was the duration of the turbidity assay) never exceeded 1%. However, extremely small concentrations of lysostaphin leaching from the mesh can cause
1 Enzyme activity was expressed in units (one unit of lysostaphin activity is defined as a decrease of 0.01 absorbance units at 600 nm, according to the manufacturer‟s specifications).
115 significant lysis to its natural substrate, i.e. cell walls of S. aureus. Fractional analysis of enzyme activity was calculated for the aliquots collected at different time intervals and compared against total mesh activity for different coating concentrations for 24 h in 2 %
BSA. Figure 6.5 below shows the results for all different coating concentrations.
Figure 6.5. Fraction of leached sample activity for the aliquots collected at different time intervals compared against total mesh activity for different coating concentrations for 24 h in 2 % BSA. Black squares – 10 µg/ml, Red circles -50
µg/ml, Green triangles – 50 µg/ml, Blue inverted triangles – 100 µg/ml, Light blue diamonds - 250µg/ml and Pink arrows – 500 µg/ml
It was found that lytic activity for these samples is almost entirely due to enzyme leaching from the mesh and lysing bacteria in the surrounding supernatant rather than adsorbed lysostaphin (data not shown). Interestingly, Shah et al. observed that the antimicrobial activity of lysostaphin coated on two different plastic surfaces, polystyrene
(well plates) and FEP (fluorinated ethylene-propylene) polymer, a Teflon-like material used in Angiocath catheters was primarily due to immobilized but not leached enzyme
116
[240]. These authors used a methodology similar to that reported here. One possible explanation is difference in materials used in our and Shah‟s studies. Ultrapro, a lightweight polypropylene macroporous mesh (with a monocryl weave) containing an absorbable component of poliglecaprone is different in density, weave, porosity (3 mm) and relative hydrophobicity. It has more „yarn-like‟ properties with its higher specific surface area compared to „flat‟ surfaces such as polystyrene well plates and the lumenal sides of catheters made up of FEP polymer. Therefore, larger amounts of lysostaphin can be adsorbed to and consequently leached from the Ultrapro mesh leading to increased fraction of activity coming from the leached enzyme.
6.3.5. Colony counting
Log reduction in bacteria is estimated as logarithm of ratio of initial bacterial colonies to average final surviving colonies after treatment. This has been used for both broth supernatant and wash counts. Table 6.2 shows the log reduction in colonies recovered from the broth supernatant and wash counts for all the different initial coating concentrations. Broth count is a model for activity against the bacteria present in the wound fluid, while the wash count models activity against colonized bacteria present on the surface of the mesh, as initial bacterial attachment plays an important role in infection.
An averaged 1.6 ± 0.4 log reduction in bacterial counts was observed in the case of broth supernatant recovered from all the lysostaphin-coated mesh samples after 24 h incubation with bacteria.
117
Initial coating Log10 reduction in Log10 reduction in Sample concentration Colonies for Broth Colonies for Wash (µg/ml) Supernatant Count (Mean ± SD) (Mean ± SD) L - 10 10 1.2 ± 0.2 1.5 ± 0.4
L - 25 25 1.3 ± 0.2 2 ± 0.3
L - 50 50 1.8 ± 0.8 2 ± 0.4
L - 100 100 1.8 ± 0.6 1.6 ± 0.2
L - 250 250 2.3 ± 0.5 2 ± 0.3
L - 500 500 1.6 ± 0.4 1.6 ± 0.2
P - 100 - 100 1.3 ± 0.6 1.5 ± 0.3
(Polyester)
Table 6.2 - Effect of lysostaphin coating concentration on log reduction of colonies for the broth supernatant and wash count; (n =3).
There was no significant difference between samples with different initial lysostaphin coating concentrations (p > 0.05; one way ANOVA); however, observed log reduction was significantly higher for the all lysostaphin-coated samples when compared against uncoated mesh samples (p<0.05; Tukey‟s test). Similarly, an average 1.8 ± 0.36 log reduction in bacterial counts was observed from the pooled wash count for all the samples that were not significantly different from each other (p>0.05; one way ANOVA).
In contrast, uncoated mesh samples showed hardly any antimicrobial activity and the bacteria thrived in their presence as indicated by a comparable surviving bacterial count with plain control cells. Bacterial adherence to the mesh is a known precursor to prosthetic
118 infection [326, 327]. A significantly less number of CFU were recovered from the wash counts for lysostaphin-coated samples compared to the uncoated samples (p < 0.05). It is thus evident from the observed data that lysostaphin-coated meshes are able to both efficiently kill bacteria in solution and possibly inhibit surface colonization of bacteria.
6.3.6. In vivo rat model trial
All twenty rats survived to the time point of mesh extraction at 7 days. Groups that did not have S. aureus inoculum (Groups A, C, E, G, I) had sterile cultures at the time of extraction. None of lysostaphin treated controls had impaired healing. Two of four rats with polyester and no lysostaphin and one of four with polypropylene and no lysostaphin had wound complications. All of the rats receiving a S. aureus inoculum and no lysostaphin had positive mesh cultures at seven days. Animals treated with L-50 and inoculated with 5 x 105 CFU of S. aureus had positive mesh cultures. The L-100 group with 5 x 105 CFU S. aureus inoculum (Groups F and J) had negative cultures (Table 6.3).
119
S. aureus Lysostaphin Inoculum Time at Positive Group Mesh Type Dose (CFU) Extraction Cultures 0/4 A Polypropylene None None 7 days B Polypropylene None 5 x 105 7 days 4/4 C Polypropylene L-50 None 7 days 0/4 D Polypropylene L-50 5 x 105 7 days 4/4 E Polypropylene L-100 None 7 days 0/4 F Polypropylene L-100 5 x 105 7 days 0/4 G Polyester None None 7 days 0/4 H Polyester None 5 x 105 7 days 4/4 I Polyester L-100 None 7 days 0/4 J Polyester L-100 5 x 105 7 days 0/4
Table 3: In vivo rat study results
6.4. Discussion
Bacterial infection at the site of implanted medical devices presents a serious and ongoing problem in the biomedical arena. Mesh implantation has reduced the rate of recurrent hernia but has led to an increased rate of bacterial infections. The incidence is reported to be as high as 8% following repair of incisonal hernias [311, 314, 328].
Although mesh-related infections occur relatively infrequently compared to other device- related infections, it can cause significant morbidity resulting in non-wound healing, recurrent hernia, and need for reoperation and mesh excision. Considering that nearly one million inguinal and incisional hernia repairs are done every year, this is a real and significant medical issue.
In our current study, we evaluate the in vitro antimicrobial susceptibility of S. aureus to lysostaphin-coated hernia repair meshes and study the effect of different enzyme
120 coating concentrations on antimicrobial activity. Lysostaphin is a glycylglycine endopeptidase isolated from Staphylococcus staphylolyticus which specifically cleaves pentaglycine cross bridges found in the staphylococcal peptidoglycan [329]. It kills S.
aureus within minutes (MIC at which 90% of the strains are inhibited [MIC90], 0.001 to
0.064 µg/ml) [320]. We observed that the antimicrobial activity of the meshes occurs due to extremely low concentrations of lysostaphin leaching from the surface into the supernatant. These concentrations are higher than its minimum inhibitory concentration
(MIC) and enzyme release from the mesh into surrounding tissue or fluids can be expected to counter any initial elevated infection risk immediately post-surgery or implantation.
A 6-h post implantation “decisive period” has been identified during which prevention of bacterial colonization mediated by adhesion is critical to the long term success of an implant [330]. Concentrations of lysostaphin that leached from the mesh in the first 5 h incubation in BSA varied from 0.0026 to 0.5 µg/ml depending upon the initial coating concentration. This is followed by a sustained release of enzyme at the same concentration that could be effective against occurrence of a latent infection. Passive coatings that aim to reduce bacterial adhesion are not as effective as “active” coatings that are designed to allow release of antibacterial agents immediately following the implantation [330]. The antimicrobial activity of the lysostaphin-coated meshes suggests that such enzyme releasing surfaces could be efficient at actively resisting bacterial adhesion and preventing subsequent colonization of implant.
From the colony counting data, the antimicrobial activity was not significantly concentration dependent in the range of L-25 to L- 500 (25 to 500µg/ml) as all the six samples reduced bacterial titers to the same level. This agrees well with the previously
121 reported data [240]. They found that lysostaphin adsorbed onto two different plastic surfaces i.e. polystyrene and FEP polymer showed killing that was not concentration dependent where all coating concentrations (0.1, 1 and 10 mg/ml) reduced the bacterial count to the same level. The overall 1.6 (± 0.421) log reduction in the broth count versus
1.8 (± 0.36) log reduction in the wash count for the coated meshes emphasized the slightly higher activity against adherent bacteria on the mesh surface, possibly due to a higher local concentration of lysostaphin near the mesh surface. Bacterial adhesion after 24 h of incubation in a 107 CFU/ml of S. aureus suspension is greatly reduced by enzyme-coated meshes compared to uncoated sample.
The results of the in vivo study show that none of the controls had bacterial contamination at 7 days. Treatment with L-50 was insufficient to clear all of bacterial innoculum, as evidenced by positive cultures in all rats in Group D. Interestingly, when doubling the treatment concentration to L-100, despite the presence of a foreign body, the entire bacterial load was cleared. Although the colony count data in the in vitro studies did not demonstrate significant difference in bacteriocidal activity of lysostaphin at different concentrations, the antibacterial efficacy of lysostaphin appears to be dose- related in vivo. The dramatic antimicrobial activity in L-100 treated group demonstrates great potential for lysostaphin in the clinical setting for preventing Staphylococcal related prosthesis infections.
Based on our in vivo findings, we decided to characterize and test the P-100 Polyester mesh (PariatexTM) sample with the same set of in vitro experiments as the Ultrapro
Polypropylene mesh.
122
Our preliminary binding yield study showed a twofold decrease in the total amount of adsorbed lysostaphin per mg of a 1 x 1 cm piece of Polyester mesh compared to its
Polypropylene counterpart with the same dimensions (data not shown). The P-100
Polyester mesh sample showed significantly higher leaching (~ 4%) in 24 h compared to the L-100 Ultrapro sample which showed negligible leaching. This was further confirmed with an activity assay of the P-100 Polyester sample that showed an almost twofold higher enzyme activity (7 ± 0.3 units per ml) calculated from the initial rate of the reactions from linearized slopes compared to the L-100 Polypropylene sample (Table 1). This can be attributed to the difference in rigidity, composition and hydrophobicity between the two mesh types. PariatexTM is a composite mesh made up of multifilament Polyester mesh and absorbable hydrophilic film made up of Collagen, Polyethylene Glycol and Glycerol while
Ultrapro is a lightweight polypropylene mesh (with a monocryl weave) containing an absorbable component of poliglecaprone with pores between 0.1 and 3 mm [331].
Hydrophilicity and low surface tension due to Polyethylene Glycol in the film could result in minimizing and weakening enzyme adsorption to Polyester meshes. We observed an average of 1.3 ± 0.6 log reduction for broth supernatant and 1. 46 ± 0.3 log reduction in the bacterial wash count for the P-100 Polyester sample after 24 h incubation. The L-100
Polypropylene sample also had a similar (1.56 ± 0.2 log reduction) in the wash count despite higher amount of lysostaphin adsorbed to the mesh surface (twofold higher) and negligible leaching of enzyme in vitro.
In conclusion we have demonstrated the antimicrobial activity of lysostaphin- coated hernia repair mesh against S. aureus at different coating concentrations. Our early in vivo trial did not show impaired wound healing in the lysostaphin treated groups;
123 furthermore these data demonstrate that lysostaphin has significant promise in preventing prosthetic infection when used at dose of 100 µg/ml. Utilizing such an antibiotic-free approach, one can control the release of enzyme locally, in the area of implant infection, by using an appropriate coating concentration that would result in lower systemic toxicity and higher antibacterial efficacy. Further work is on the way to confirm our findings in the clinical arena. Surface-coatings using antibacterial enzymes could be a ground breaking addition to the field of hernia repair and other areas of surgery in which prosthetic materials, or even biologic materials, are implanted.
124
CHAPTER 7
APPLICATION OF ENZYMES AS ANTIBACTERIAL COATINGS ON PLAIN AND NANOPARTICLE COATED CATHETER SEGMENTS
7.1. Introduction
Medical devices such as intravascular and urinary catheters are routinely employed in healthcare sector for hemodialysis, drainage of urine, administration of IV fluids, medication, oxygen, anesthetic agents and monitoring of blood and intracranial pressure[332, 333]. Unfortunately, these catheters are also a major cause of nosocomial infections that occur in more than two million hospitalizations in the US annually, with major medical remediation costs Consequently, catheter-related infections are notoriously difficult to treat via conventional antibiotic therapy, with associated mortality rates ranging from 12% to 25% [334]. Extended hospital stays therefore calls for an active intervention on the part of the healthcare personnel, and has driven the estimated annual healthcare cost arising from these catheter-related biofilm infections to more than nine billion dollars [334].
Catheters are generally infected by microorganisms that are typically present in the surrounding site of insertion, usually as part of natural flora in the skin; contaminate the catheter along the outer surface [335]. They can also be contaminated in their lumenal compartments through fluid flow from contaminated infusate or from parts of the implant that are improperly sterilized [335]. Sometimes, percutaneous devices such as catheters sometimes are „marsupialized‟, where basal cells from the epidermis near the edge of the wound may migrate down the edge of the dermis to surround the implant forming a pouch without any surrounding tissue integration [336]. The space between the implant and the
125 epidermis then becomes filled with cell debris which then is a permanent source for onset of infection because of the lack of a natural mechanism for rejuvenation or cleaning [336].
In addition, upon insertion in the host, the surface of the catheter becomes „passivated‟ by a layer of blood proteins (due to contact with device) that assists attachment of planktonic bacteria. Eventually, the final stage in bacterial colonization leads to the formation of biofilms on the surface of the device [3, 6, 41, 231, 232]. Various clinical studies have been focusing on the use of different antibiotics immobilized on the surface or released from the bulk of an implant for prevention of biofilm formation [337, 338]. However, treatment with conventional antibiotics frequently fails because bacteria develop multi- drug resistance and bacteria growing in biofilms are more resistant than planktonic cells.
Staphylococcus aureus is a Gram-positive, biofilm-forming bacterium that is regarded as a leading cause of nosocomial infections and accounting (together with Staphylococcus epidermidis) for more than one-half of prosthetic device-associated infections[339].
Thus, there is an urgent need for development of alternative treatment strategies that could overcome the problems associated with the use of conventional antibiotic therapy.
Increasingly, attention has been paid specifically to enzymes and peptides as surface coatings to prevent device-associated infections [106, 109, 240, 340, 341]. For a catheter surface, the coating may be retained by either physical adsorption or covalent attachment of the enzyme. The use of nanoparticles as enzyme supports since the 80s has thereafter sustained a continued interest for its application modern biotechnology [187-190]. Higher surface area for enzyme loading, better stability and long term durability are some of the expected outcomes of using nanoparticles as coatings on catheter segments for enzyme immobilization. Hence we propose to use poly (lactic acid) nanoparticles as „surrogate‟
126 surface coatings for enzyme adsorption on indwelling medical devices such as Central venous catheters (CVCs). In this preliminary study, we propose to coat DispersinB and lysostaphin to the surface of Polyurethane catheter segments at different coating concentrations through single adsorption and coadsorption (two enzymes), characterize and evaluate the in vitro performance of plain and nanoparticle coated catheter segments.
7.2. Materials and methods
Materials: Staphylococcus aureus (ATCC 27660) was purchased from American Type
Culture Collection (Manassas, VA). Lysostaphin (25 kDa; LSPN-50) isolated from
Staphylococcus staphylolyticus was bought from Ambi Products LLC (Lawerence, NY).
DispersinB (41 kDa) produced by Aggregatibacter actinomycetemcomitans was bought from Kane Biotech Inc. (Manitoba, Canada). Bovine Serum Albumin (BSA) was bought from Calbiochem (San Diego, CA). Polyurethane Catheter tubing (Single lumen) was purchased A.P. Extrusion (Salem, NH). Antibiotic–antimycotic solution, heat inactivated fetal bovine serum, Dulbecco‟s modified eagle medium and Alexa Fluor 594 Carboxylic acid, Succinimidyl ester dyes used for fluorescent labeling were purchased from
Invitrogen, Molecular Probes (Carlsbad, CA). PBS Tablets (#524650) were purchased from Cabiochem (La Jolla, CA). BCA and micro BCA reagent kits were purchased from
Pierce Biotech Inc (Rockford, IL). BBLTM Trypticase Soy Broth was purchased from
Becto, Dickinson and Company (Sparks, MD). Methylene Blue was obtained from the
Department of Biological Sciences at Clemson University courtesy of John Abercombie.
Poly (DL-lactide; average Mw (97500), Pluronic F68 (MW = 8.4 kDa) and 4-nitrophenyl-
N-acetyl- D-glucosaminide (substrate for DispersinB) was purchased from Sigma Aldrich
127
® (St. Lous, MO). CellTiter 96 AQueous One Solution Cell Proliferation Assay (MTS) for cell viability studies was purchased from Promega Corporation (Madison, WI). Type 1
Clear Borosilicate glass vial (8 mL) was with attached screw caps was bought from VWR
(West Chester, PA). A microsep Centrifugal Devices with a modified Polyethersulfone membrane (low protein binding; MWCO - 3K) used for enzyme purification were purchased from Pall Life Sciences (Ann Arbor, MI). All other reagents were purchased from VWR (West Chester, PA) and Fisher Scientific (Pittsburgh, PA), and used without further purification unless otherwise specified. Binding efficiency, Turbidimetric activity and Colorimetric assays were performed using a Synergy microplate reader (Bio-Tek
Instruments, Inc., Winooski, VT). Particle sizing and Zeta potential analysis of nanoparticles (NPs) was done using a 90plus Particle Size Analyzer (Brookhaven
Instruments Co, Holtsville, NY). Scanning electron microscopy was done at the Electron microscopy facility at the Advanced Materials Research Laboratory (Pendleton, SC).
7.2.1. Preparation of nanoparticles
Polylactide nanoparticles were prepared using a nanoprecipitation method based on solvent diffusion. Briefly, Poly(DL-lactide) (average Mw (97500) was dissolved in 30 ml of acetone at a concentration of 8 mg/ml and this solution was then dispersed in an aqueous phase (200 ml) containing Pluronic F68 (5 mg/ml) under sonication ((Bransonic
5510, Branson, MO) for 30 min. The acetone solution was added dropwise at a uniform rate into the aqueous phase. The resulting suspension containing the nanoparticles was then purified three times in DI water by ultracentrifugation (Allegra™ 64R Centrifuge,
Beckman Coulter, USA) at 3000g for 1 hr at 25°C to remove any excess polymer and
Pluronic. Nanoparticle suspensions in DI water were stored in refrigerator at 4°C. The
128 number of nanoparticles in the final suspension was then determined using gravimetric analysis.
7.2.2. Particle sizing
Stock solutions containing nanoparticles were diluted 1:1000 in DI water in a disposable cell and particle size was measured using dynamic light scattering with a 90Plus particle size /zeta potential analyzer (BrookHaven Instruments Corporation, Holtsville, NY).
Before particle sizing, the samples were sonicated for 30 min to break up any loosely-held agglomerates. Particle size was expressed as a combined mean diameter in nm (+/- SD).
7.2.3. Preparation of NP coated catheter segments
Single lumen Polyurethane catheter tubing (Outer diameter (0.125 inches) and Inner diameter (0.63 inches) purchased from A.P.Extrusion was used for all the adsorption studies. 1 cm segments were cut off from the main tubing in a laminar flow hood under sterile conditions and autoclaved at 121 °C. Using a pair of sterile tweezers, the catheter segments were gently placed in 8 ml glass vials containing 0.5 ml of the concentrated nanoparticle solution and incubated overnight at room temperature with gentle shaking (50 rpm). Control catheter segments were placed in glass vials containing 0.5 ml of DI water.
The samples were then washed gently with DI water three times and air dried at RT for 6 h before protein adsorption experiments. The average weight of a 1 cm catheter segment was 77.5 mg calculated from combined weight of 6 catheter segments weighed together.
7.2.4. Surface characterization of catheter segments with scanning electron microscopy
129
NP coated catheter and uncoated samples were cut longitudinally to 0.2 cm segments and gently adhered horizontally to aluminum specimen support stubs using double-sided carbon tape. Specimens were then sputter-coated with Platinum and examined using a SU
6600 field emission SEM equipped with variable pressure. The instrument was operated at
10 kV accelerating voltage with a 10-12 mm working distance and the images were captured digitally.
7.2.5. Enzyme adsorption on plain and NP coated catheter segments
Initial enzyme concentrations of 0.025, 0.05, 0.1, 0.25, and 0.5 mg/ml in 1mM PBS buffer were each prepared from a 1 mg/mL stock solution for lysostaphin and DispersinB respectively. Using a pair of sterile tweezers, catheter segments (plain and NP coated) were gently placed into each of the vials containing 0.5 ml of enzyme solutions and incubated overnight at room temperature with gentle shaking (100 rpm). The sample solutions were removed and the catheter segments were gently flushed with 5 ml of PBS buffer five times and air dried for 1 h. The enzyme solution above the catheter segment was then collected and stored for binding efficiency measurements. The concentration of unbound enzyme in the supernatants was determined using BCA assay. Initial enzyme solutions with known concentrations were used as the standards. The concentration of the unbound enzyme was calculated and subtracted from initial concentration of enzyme present in the initial solution. The difference corresponded to the bound enzyme concentration on the catheter segment. Adsorption experiments were reproduced in triplicates.
7.2.6. Enzyme coadsorption on plain and NP coated catheter segments
130
For coadsorption studies, we used fluorescently labeled lysostaphin (AlexaFluor 594) of enzyme concentrations - 0.025, 0.05, 0.1, 0.25, and 0.5 mg/ml in 1mM PBS buffer and correspondingly mixed with 0.025, 0.05, 0.1, 0.25, and 0.5 mg/ml of DispersinB to make up the final volume of 0.5 ml. Alexa Fluor 594 dye was used for fluorescent labeling of lysostaphin as described in previous sections [298] . Catheter segments (plain and NP coated) were placed into each of the vials containing 0.5 ml of enzyme mixture solutions and incubated overnight at room temperature with gentle shaking (100 rpm). The sample solutions were removed and the catheter segments were gently flushed with 5 ml of PBS buffer five times and air dried for 1 h. The enzyme solution above the catheter segment was similarly collected and stored for binding efficiency measurements. The concentration of total unbound enzyme in the supernatants was determined using BCA assay. Initial enzyme mixture solutions with known concentrations were used as the standards. The concentration of unbound enzyme was calculated and subtracted from initial concentration of total enzyme present in the initial mixture solution. The difference corresponded to the total bound enzyme concentration on the catheter segment for both lysostaphin and
DispersinB.
The concentration of unbound lysostaphin in the supernatant was determined from fluorescence measurements. Fluorescence intensities of the supernatant and standards (100
µl) were measured in a 96 well plate using a microplate reader (Ex 594 nm; Em 625 nm).
The concentration of unbound lysostaphin was calculated and subtracted from initial concentration of enzyme present in the initial solution. The difference corresponded to the bound lysostaphin concentration on the catheter segment. The difference of total bound concentration of both enzymes (BCA assay) and bound lysostaphin (fluorescence) corresponded to the bound concentration of DispersinB on the catheter segment.
Coadsorption experiments were also reproduced in triplicates.
131
7.2.7. Activity assay for adsorbed enzymes
Lysostaphin: Turbidity assay was used to assess the activity of lysostaphin adsorbed to
Polyurethane segments. A cell suspension of S. aureus was prepared by inoculating be 10 mL of 30% (w/v) tryptic soy with 100 µL of S. aureus culture prepared according to
ATCC instructions as described previously. The cells were incubated at 37C for 18 - 24 h under gentle shaking to grow to a mid-log phase. The cells were centrifuged at 2,000 g for
10 min and washed twice with 10 mM PBS. The cells were then resuspended in PBS to prepare a bacterial suspension with optical density (OD600) of ~ 0.4 at 600 nm (1 cm light path). 0.5 mL of the bacterial suspension was be added to 8 mL glass vials containing the catheter segments. The samples were be incubated with the suspension at 37ºC under continuous shaking, and the rate of bacterial lysis was be monitored for 1 h by taking 0.1 mL aliquots and measuring the OD600 in a 96 well plate. The vials containing the cell suspension will be continuously shaken in order to maintain contact with the catheter segments. All kinetic experiments were reproduced in duplicates.
DispersinB: Activity of adsorbed DispersinB was measured using a colorimetric assay
[107]. DispersinB activity was quantitatively assessed using UV-visible (UV-VIS) spectroscopy by measuring the absorbance at 400 nm of the p-nitrophenolate reaction product resulting from enzymatic hydrolysis of the substrate 4-nitrophenyl-N-acetyl- D- glucosaminide. 0.5 ml of 5 mM substrate in 50mM Sodium Phosphate buffer was added to the vials containing the catheter segments and the enzymatic reaction was carried out for 1 h at 37°C. After 1 h, the reaction was quenched by adding 50 µl of 10 N NaOH and 100 µl aliquots were collected from each sample and the absorbance was measured at 400 nm.
All experiments were reproduced in duplicates.
132
7.2.8. Enzyme leaching studies
Lysostaphin: Leaching of non-specifically bound enzymes from the catheter surface can occur both in vitro and in vivo. To model in vivo conditions, we studied enzyme leaching from enzyme coated catheter segments in the presence of BSA. The samples were incubated with 0.5 mL of 2 % (w/v) BSA solution in PBS buffer for 72 h at 37C. We found that fluorescence measurements lacked sensitivity to determine concentration of leached enzyme in the samples. We therefore estimated enzyme concentration based on its activity. For kinetic measurements, 50 µl aliquots from each of the samples were taken at different time points (6, 24, 48 and 72 h) and added to a 96 well plate.
A 125 µl cell suspension of S. aureus in PBS buffer with optical density (OD600) of
0.4 at 600 nm (1 cm light path) was then added to each of the different wells and the rate of bacterial lysis was monitored continuously for 2 h at 37ºC. Lysostaphin solutions with known enzyme concentrations of 0.0071 – 7.14 µg/ml were used as standards for the assay. Enzymatic activity for standards and unknown samples was measured simultaneously in the same microplate. The initial rate of the reaction for each of the standards and unknowns was calculated from the linearized slopes. The standard curve was obtained by plotting the initial rate against the enzyme concentration. From the standard curve, the unknown enzyme concentrations were determined and the amount of leached enzyme for each of the samples was plotted as a function of time. All kinetic experiments were reproduced in duplicates.
DispersinB:
Similarly, leaching of DispersinB was studied modeling in vivo conditions, in the presence of 2 % (w/v) BSA solution in PBS buffer for 72 h at 37C. For kinetic
133 measurements, 50 µl aliquots from each of the samples were taken at different time points
(6, 24, 48 and 72 h) and added to a 96 well plate. A 100 µl solution of 4-nitrophenyl-N- acetyl- D-glucosaminide (5 mM) was added to each of the different wells and the absorbance at 400 nm of the p-nitrophenolate reaction product resulting from enzymatic hydrolysis of the substrate was monitored continuously for 1 h at 37ºC.
DispersinB solutions with known enzyme concentrations of 0.083 – 8.3 µg/ml were used as standards for the assay. Enzymatic activity for standards and unknown samples was measured simultaneously in the same microplate. The initial rate of the reaction for each of the standards and unknowns was calculated from the slopes. The standard curve was obtained by plotting the initial rate against the enzyme concentration.
From the standard curve, the unknown enzyme concentrations were determined and the amount of leached enzyme for each of the samples was plotted as a function of time. All kinetic experiments were reproduced in duplicates.
The different samples designated based on their initial coating concentrations of
10µg/ml, 25µg/ml, 50 µg/ml, 100µg/ml, 250µg/ml and 500µg/ml were L-25, L-50, L-100,
L-250, L-500 for lysostaphin and D-25, D-50, D-100, D-250 and D-500 for DispersinB coated on uncoated catheter segments. Similarly, samples were designated as NPL-25,
NPL-50, NPL-100, NPL-250, NPL-500 and NPD-25, NPD-50, NPD-100, NPD-250,
NPD-500 for lysostaphin and DispersinB adsorbed on NP coated catheter segments respectively.
7.2.9. Cell viability using MTS assay
® The CellTiter 96 AQueous Assay uses the novel tetrazolium compound (3-(4,5- dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium,
134 inner salt; MTS) and the electron coupling reagent, phenazine methosulfate (PMS). MTS is chemically reduced by cells into formazan, which is soluble in tissue culture medium
[342]. The measurement of the absorbance of the formazan can be carried out using 96 well microplates at 492nm. The assay measures dehydrogenase enzyme activity found in metabolically active cells. Since the production of formazan is proportional to the number of living cells, the intensity of the produced color is a good indication of the viability of the cells. MTS solutions were prepared according to the manufacturer‟s instructions.
Using the methods outlined in ISO standard 10993-5 as a guideline, mouse 3T3 fibroblasts (donated by Cassie Gregory; Department of Bioengineering, Clemson
University) were previously cultured in Cell culture flasks in DMEM supplemented with
1% Fetal Bovine Serum (FBS) and 1% antibiotic (Penicillin/Streptomycin). The cells were trypsinized, centrifuged for 5 minutes at 1000 rpm with the medium aspirated and the cell pellet resuspended in fresh media at a concentration ~10,000 cells/mL. Aliquots (100 µL) were then pipetted into individual wells of a 24-well plate with fresh media. Cells were allowed to grow to confluence for 24 h at 37°C in a humidified 5% CO2 environment.
Cell viability was checked under a microscope and the media replaced every 12 h. The samples were then cut into 0.3 cm pieces to ensure specimen coverage of approximately one-tenth of the cell layer surface (ISO 10993-5). Samples were then gently placed on the cell layer in the center of 24-well plates using a pair of sterile forceps. Subsequently, plates were incubated for 48 h at 37°C in a humidified 5% CO2 environment with media being changed every 12 h. Care was taken during mesh placement in the wells to prevent unnecessary movement, which can lead to physical trauma to the cells and dislodgement
135 of cells from the mesh. Sterile Latex pieces were used as positive controls for assessment of cytotoxicity (ISO 10993-5) while cells in media alone were used as negative controls.
After 48 h, samples were gently removed using a pair of sterile forceps and 50 µl of MTS reagent was added to every well in one of the 24-well plates in a ratio of 1:5
(MTS reagent : volume of media in the well). The plate was incubated at 37 °C for 4 h in a humidified, 5% CO2 atmosphere. 100 µl from each well was then transferred to wells of a 96-well plate and the absorbance was recorded at 490 nm (OD490) and 650 nm (OD650) using a plate reader. Raw quantitative data generated by the microplate reader were analyzed by first subtracting the difference of OD490 and OD650 reading to remove any background interference. The resultant mean OD readings from each well were calculated as an average of two replicates for each sample and expressed as a percentage of the OD values observed relative to OD values measured in the negative control wells (negative control – cells in media).
7.2.10 Antibiofilm assay
Briefly, S. aureus culture was grown overnight in 10 ml of TSB (Bidifico) supplemented with 1% D-glucose (Tokyo Kasei, Japan). The overnight bacterial cultures was then centrifuged twice, resuspended in TSB and spot plating on Agar was done to determine the number of colonies in the active culture. 1 ml containing ~ 106 CFUs in TSB supplemented with1% D-glucose was added to wells of a 12-well tissue culture plate
(Costar Inc., Corning, N.Y.) containing the coated catheter segments. The 12-well plate was incubated at 37°C with shaking at 100 rpm for 12 h and then removed from the shaker and supplemented with either 200 µl of additional fresh TSB to compensate for medium evaporation. These plates were then be incubated at 37°C for an additional 24 h.
136
After ~36 h of growth, the media was removed from various wells, washed thrice with phosphate-buffered saline (PBS) to remove any non-adherent bacteria. The assay plates were allowed to dry at 37°C for approximately 2 hr. The wells containing the segments were then stained with 0.5 ml of 1% Methylene Blue for 30 min and then gently washed with PBS three times; air dried for 1 h and destained with 30% acetic acid solution. The eluted dye solution was then transferred to wells of a 96 well plate and the absorbance was recorded at 660 nm to evaluate the inhibition of S. aureus biofilms.
7.2.11. Accelerated Shelf life Study
Accelerated aging testing is performed on packaged medical devices such as catheters and implants to ascertain its shelf life and document expiration dates. We modeled a study based on long term stability testing of our enzyme coated catheter segments by measuring enzyme activity after storage. Initial lysostaphin and DispersinB concentration of 0.5 mg/ml in 1mM PBS buffer was each prepared from a 2 mg/mL stock solution and using a pair of sterile tweezers, catheter segments (plain) were gently placed into vials containing 0.5 ml of enzyme solutions and incubated overnight at room temperature with gentle shaking (100 rpm). The sample solutions were removed and the catheter segments were gently flushed with 5 ml of PBS buffer five times and air dried for
1 h. The samples were then incubated in a -80 °C freezer overnight in glass vials sealed with parafilm. Samples were then freeze-dried for 24 h in a Freeze Dry System/Freezone
4.5 from Labconco (Kansas City, MO) courtesy of Dr. Ken Webb (Clemson University).
Samples were then hermetically sealed using a heat press in special peelable high barrier pouches (# TPC-1475B) from Oliver-Tolas Healthcare Packaging (Grand Rapids, MI).
137
The samples were then incubated at 60 °C for 1 week in an Isotemp Oven; Model 516G from Fisher Scientific (Ann Arbor, MI).
After 1 week, the samples were removed and placed in wells of a 24 well plate.
For lysostaphin activity, 0.5 mL of a previously purified bacterial suspension with optical density (OD600) of ~ 0.38 at 600 nm (1 cm light path) was added to each of the wells containing the catheter segments. The samples were incubated with the suspension at 37ºC under continuous shaking along with appropriate controls, and the rate of bacterial lysis was be monitored for 3 h by taking 0.1 mL aliquots every hour and measuring the OD600 in a 96 well plate. The vials containing the cell suspension were continuously shaken in order to maintain contact with the catheter segments. Activity of DispersinB adsorbed on catheter segments was monitored using the colorimetric assay as described previously by measuring the absorbance at 400 nm of the p-nitrophenolate reaction product resulting from enzymatic hydrolysis of the substrate 4-nitrophenyl-N-acetyl-D-glucosaminide. 0.5 ml of 5 mM substrate in 50mM Sodium Phosphate buffer was added to the vials containing the catheter segments and the enzymatic reaction was carried out for 1 h at
37°C. After 1 h, the reaction was quenched by adding 50 µl of 10 N NaOH and 100 µl aliquots were collected from each sample and the absorbance was measured at 400 nm.
All experiments were reproduced in triplicates.
7.2.12. Statistical analysis
Data are presented as means±SD. One-way analysis of variance (ANOVA) and Paired„t‟ test was performed for comparing coated and uncoated samples with different initial enzyme concentrations. The null hypothesis was that the means are not significantly different. Data were analyzed using Origin 6.0 and 8.0 Data Analysis and Graphing
138
Software (OriginLab Corporation, MA, USA). A p value of <0.05 was considered statistically significant.
7.3. Results and Discussion
7.3.1. Enzyme Binding yield
Based on previous preliminary data, which demonstrated high antibacterial activity of enzyme-NP conjugates against planktonic cells (Chapters 4 and 5), we proposed to compare the loading of enzymes and enzyme-NP conjugates on catheter segments and subsequently evaluate antimicrobial and antibiofilm activity of the catheter segments. We decided to adsorb both DispersinB and lysostaphin to the surface of PLA nanoparticles either alone or through coadsorption. After synthesis and characterization of enzyme-NP conjugates, we incubated them with catheter segments and compared the binding efficiency of plain enzyme- and enzyme-NP conjugate-coated catheter segments.
Higher binding yield, better enzyme stability and long term durability were some of the expected outcomes using this approach. However, we observed two- to five-fold less amount of DispersinB adsorbed in the case of enzyme-NP conjugate-coated catheter segment compared to the plain enzyme-coated catheter segment and thirty-fold less amount of DispersinB adsorbed in case of samples coadsorbed with lysostaphin (data not shown).
We hypothesized that coating by enzymes makes NPs more hydrophilic leading to decreased NP attachment to the catheters. Thus, we decided to modify our approach to achieve better coating by enzyme-NP conjugates. We opted to first coat the surface of the catheter segment with PLA nanoparticles and then adsorb enzymes at different initial
139 concentrations either alone or through coadsorption. This approach eliminates possible error in determination of enzyme concentration in enzyme-NP conjugates as we are using the same initial enzyme concentrations when coating catheters by either plain enzyme or enzyme-NP conjugates.
We made certain theoretical approximations. A greater surface area is expected upon coating PLA nanoparticles on the surface of a Polyurethane catheter segment. The average diameter of a PLA NP calculated from previous DLS experiments was 181 ± 29 nm (mean from 3 different batches) depending upon the batch of NPs used for coating experiments. Based on the dimensions of a 1 cm PU catheter segment, with ID and OD of
0.317 cm and 0.158 cm, the theoretical surface area calculated assuming a hollow cylinder is 1.614 cm2. The number of nanoparticles that could saturate the surface of a 1 cm PU catheter segment with a diameter of 180 nm is 1.5 x 109 particles assuming that the surface of the catheter segment is smooth and devoid of any topography. The surface area occupied by each spherical NP adsorbed on the surface is 4πr2. Theoretically, an enzyme molecule that has an average hydrodynamic radius of „r‟ occupying the surface of the catheter segment will have a surface area of r2. However, if it were to adsorb on the surface of a catheter segment that is completely saturated with a monolayer of NPs, it could theoretically occupy a surface area of 4 πr2 per NP on the catheter (Figure 7.1). This increase in surface area and high surface energy of a nanoparticle could contribute to better adsorption and subsequently higher effective loading based on the same initial coating concentration of enzyme for NP coated catheter segments. Previously, Crisante et al investigated the antibiotic release of nano-structured polymer systems in carboxylated polyurethane [343]. They hypothesized that since nanoparticles are characterized by a high
140 surface/volume ratio; their presence in the polymer could result in a greater surface area available for the antibiotic adsorption.
Figure 7.1. An illustration for a model single lumen catheter segment before and after coating with nanoparticles.
Binding yield was calculated based on the total amount of enzyme adsorbed per mg of plain or PLA NP coated PU catheter segment. Figure 7.2 shows the binding yield for lysostaphin and DispersinB on plain or NP coated 1 cm PU catheter segment at different initial enzyme coating concentrations. Figure 7.2A and 7.2B specifically corresponds to enzyme adsorption for lysostaphin and DispersinB individually and Figure
7.2C and 7.2D correspond to adsorption of each enzyme in the presence of the other
(coadsorption). The amount of adsorbed enzyme (µg) per mg of catheter segment is plotted as a function of the initial enzyme concentration.
141
A B
C D
Figure 7.2 A). Binding yield for lysostaphin on plain and NP coated PU catheter segment; B). Binding yield for DispersinB on NP coated PU catheter segment C).
Binding yield for lysostaphin on plain and NP coated PU catheter segment when coadsorbed with DispersinB at the same coating concentration. D). Binding yield for DispersinB on plain and NP coated PU catheter segment when coadsorbed with lysostaphin at the same coating concentration; (Mean ± SD; n = 3). Black squares represent plain catheter segment while red circles represent NP coated catheter segment
142
Based on the above described experimental conditions, ~11 to 23% of lysostaphin initially present in the solution remained bound to the catheter segment for plain catheter samples and ~ 8 to 29% was bound to NP coated catheter segments depending upon the initial coating concentration. Similarly, ~12 to 30% and ~ 20 to 35 % of DispersinB initially present in the solution remained bound to the catheter segment for plain and NP coated catheter segments respectively. Generally, the amount of enzyme bound to catheter segment increased linearly on increasing the enzyme coating concentration for plain and
NP coated catheter segments (Figure 7.2A and 7.2B). Plain coated catheter segments showed a slightly higher adsorbed lysostaphin concentration per weight than NP coated catheter segments for the samples, while the amount of DispersinB bound to catheter segment was not significantly different for both types of catheter segments (p > 0.05).
In case of the coadsorption study, both lysostaphin and DispersinB were simultaneously adsorbed at different initial concentrations and the amount of each enzyme was calculated using both BCA analysis and fluorescence. As described in the methods section, the amount of bound lysostaphin was calculated via fluorescence labeling and the difference between the total protein content (BCA analysis) and bound lysostaphin was used to calculate amount of DispersinB bound on the catheter segment. Binding yields for lysostaphin consisted of ~2 to 10% and ~8 to 13 % per total weight of the protein mixture in the initial solution for plain and NP coated catheter segments respectively. Similarly, the binding yield for DispersinB corresponded to ~0.1 to 68% and ~1.5 to 42% per total protein weight for plain and NP coated catheter segments respectively when co-adsorbed with the lysostaphin (Figure 7.2C and 7.2D).
143
We expected a higher surface area for enzyme adsorption for NP coated catheter segments after precoating plain Polyurethane catheter segments with PLA nanoparticles.
BET specific surface analysis study comparing both plain and NP coated Polyurethane segments was inconclusive for all sets of samples because specific surface area for all samples was below BET detection limit (data not shown). Figure 7.3A-H below shows a cross section of the outer surface of plain and NP coated Polyurethane catheter segments taken at different magnifications taken on SU-6600 field emission SEM equipped with variable pressure. These images clearly show that the sample catheter segments did not have a monolayer of nanoparticle coating on the surface as expected. There was still a large portion of the catheter segment that was bare and devoid of any nanoparticles.
Lysostaphin and DispersinB molecules probably bound more on the bare portion rather than NP coated part of the polyurethane catheter segment just to due to higher available surface area for enzyme adsorption.
144
A B
C D
E F
G H
Figure 7.3. SEM images of plain (A, C, E, G) and PLA NP (B, D, F, H) coated
Polyurethane catheter segments at different magnifications taken on SU-6600 field emission SEM equipped with variable pressure..
145
The degree of surface coverage on the catheter segment can be predicted from the binding yield of each either of the enzymes when adsorbed alone or in the presence of the other. Average dimensions of the lysostaphin molecule are 4.5 nm X 5.5 nm X 8.5 nm
[301]. The average surface area occupied by an adsorbed protein molecule based on the smallest dimensions is 25.02 nm2. Similarly, the average dimensions of DispersinB molecule are 4.1 nm X 8.6 X 18.1 nm [344]. The average surface area occupied by adsorbed DispersinB molecule based on smallest dimensions is 35.26 nm2. The maximum theoretical monolayer of lysostaphin adsorbed on a 1 cm catheter segment calculated using the above surface area per molecule and the theoretical surface area of catheter segment of 1.614 cm2 consisted of 6.5 X 1012 molecules. Similarly, for DispersinB, the number of molecules that could occupy the surface of the catheter segment for a theoretical monolayer is 4.6 X 1012. Based on these theoretical considerations, the binding yield data for lysostaphin and DispersinB adsorbed alone or through coadsorption on plain and NP coated catheter segments suggests multilayer protein adsorption, from ten to several hundred layers of molecules, depending upon the coating concentration.
Multilayer adsorption on polymer surfaces has been observed with different proteins interacting with a variety of surfaces [344-347]. The proteins in direct physical contact with the adsorbing surface or the layer of water molecules present around the surface will be strongly affected by surface properties [348, 349]. These surface properties include wetability, topography, chemical composition etc. Due to these surface characteristics, the interaction between a protein molecule and the surface can induce conformational changes on the protein, resulting in a different surface that would influence interaction with the surrounding protein molecules in the bulk solution and so
146 on. The adsorption isotherms (Figure 7.2A and 7.2B) from a single protein solution do not reach saturation upon increasing the protein concentration providing additional argument in favor of multilayer adsorption hypothesis.
Another possible explanation is that adsorbed proteins actually form a monolayer but real specific surface area of the catheter segment is significantly higher than that estimated from purely geometric considerations. In fact, BET using Kr adsorption performed in this study has a very low detection limit of 0.05 m2 (500 cm2) of the surface:
Thus, even if the catheter‟s actual specific surface area exceeds that calculated from geometric dimensions by the factor of 200-300, it would still be undetectable using BET method. Of note, alternative BET approach that uses N2 instead of Kr is even less sensitive and requires 0.5-1 m2 of the surface at the lower detection limit.
Protein adsorption studies from solutions containing a mixture of proteins imply some competitive adsorption, where different proteins with different characteristics will compete with each other for the available surface area for adsorption [241]. Aside from the size, isoelectric point, molecular weights, secondary structures of the protein etc surfaces features also influences interactions between different proteins. We observed a decrease in amount of each of individually adsorbed protein (lysostaphin and DispersinB) when adsorption occurs from a mixture of protein solutions. At the same time, the total amount of bound protein from a mixture of proteins is significantly higher (p < 0.05) than total protein adsorbing from a single solution for both plain and NP coated catheter segments.
7.3.2. Adsorbed enzyme activity assay
147
Activity of adsorbed lysostaphin on plain and NP coated catheter segments was done by monitoring the cell lysis of a S. aureus cell suspension [325] as described previously. We monitored the decrease in optical density of the bacterial suspension and compared the rate of lysis for the samples with the highest and lowest initial enzyme coating concentration (25µg/ml and 500µg/ml). Figure 7.4A shows the mean rate constants of bacterial lysis and standard deviations for these samples (n=2) for both plain and NP coated catheter samples.
*
Figure 7.4. Activity assay data shown as bar graphs comparing the kinetic constant of bacterial lysis against the highest and lowest lysostaphin coating concentrations for plain and NP coated catheter segments. Black bar – NPL-25 represnts NP coated catheter segment with initial lysostaphin concentration of 25
µg/ml, Red bar – NPL-500 represents NP coated catheter segment with initial lysostaphin concentration 500 µg/ml; Green bar – L-25 represents plain catheter segment with initial lysostaphin concentration of 25 µg/ml and Pink bar – L-500
148 represents plain catheter segment with initial lysostaphin concentration 500 µg/ml;
(Mean ± SD; n= 2; * - p > 0.05)
We found that the initial rate of cell wall lysis was not significantly concentration dependent for the two sets of initial concentrations (p>0.05 –„Paired t test‟) and not significantly different for all samples of plain and NP coated catheter segments (p>0.05 –
One way ANOVA). A similar trend in adsorbed lysostaphin activity was also seen in samples coadsorbed with DispersinB at the same coating concentrations (data not shown).
It is well known that lysostaphin is active against S. aureus at extrememly low concentrations [320]. The results obtained here are in strong agreement with data obtained by Shah et al, who also evaluated lysostaphin activity after physical adsorption to catheter segments [240]. They also found that the antibacterial activity of lysostaphin was not dependent upon coating concentrations as they found all three concentrations used (0.1, 1 and 10 mg/ml) showed the same antistaphylococcal activity by inhibiting bacterial growth to the same level. This is in contrast to our experiments with lysostaphin-coated hernia repair meshes (Polypropylene weave – Ultrapro) where we observed a correlation between lysostaphin coating concentration and lytic activity of all the samples. This can again be attributed to differences in material composition and porosity of the two types of samples as mentioned previously.
We used a colorimetric assay to test activity of DispersinB adsorbed to catheter segments with a synthetic substrate, 4-nitrophenyl-N-acetyl-β-glucosaminide in a procedure similar to the one used by Donelli et.al [107]. We monitored the optical density
(OD400) of p-nitrophenolate reaction product resulting from the enzymatic hydrolysis of the substrate by comparing the rates of enzymatic hydrolysis for the samples with the
149 highest and lowest initial enzyme coating concentration (25 µg/ml and 500 µg/ml). It is well known that most biofilm-producing staphylococcal strains produce a linear PNAG and it has been demonstrated previously that DispersinB is able to degrade N- acetylglucosamine-containing polysaccharides [107, 350, 351]. Figure 7.5 shows the mean rate constants of hydrolysis and standard deviations for these samples (n=2) for both plain and NP coated catheter samples. We found the rate of hydrolysis of substrate was significantly concentration dependent for the two sets of initial concentrations (p<0.05 –
„Paired t test‟) for both plain and NP coated catheter segments. However, we did not see a significant difference between plain and NP coated catheter segments (p > 0.05; one way
ANOVA). A similar trend in adsorbed DispersinB activity was also seen in samples coadsorbed with lysostaphin at the same coating concentrations (data not shown).
* ¥
Figure 7.5. Activity assay data shown as bar graphs comparing the kinetic constant of substrate hydrolysis via colorimetric assay against the highest and lowest DispersinB coating concentrations for plain and NP coated catheter segments. Black bar – NPD-25 represnts NP coated catheter segment with initial
150
DispersinB concentration of 25 µg/ml, Red bar – NPD-500 represents NP coated catheter segment with initial DispersinB concentration 500 µg/ml; Green bar – D-
25 represents plain catheter segment with initial DispersinB concentration of 25
µg/ml and Pink bar – D-500 represents plain catheter segment with initial
DispersinB concentration 500 µg/ml; (Mean ± SD; n= 2; ¥ - p < 0.05; * - p > 0.05;)
7.3.3. Enzyme leaching results
We assessed lysostaphin and DispersinB leaching in the presence of 2 % (w/v)
BSA, as a model for the most abundant protein in blood plasma. Figure 7.6 shows leaching for all samples with different coating concentrations for lysostaphin and
DispersinB adsorbed to plain and NP coated catheter segments. We chose to quantify leaching using activity assays for the individual enzymes rather than BCA analysis or fluorescent labeling owing to low sensitivity of the two latter techniques. The unknown enzyme concentrations for each of the samples were determined from a standard curve that plotted rate of the reaction (enzymatic lysis for lysostaphin and enzymatic hydrolysis for DispersinB) against a series of known enzyme concentrations and the amount of leached enzyme for each of the samples was plotted as a function of time.
A B
151
C D
Figure 7.6. Leaching of enzyme as a function of time for plain and NP coated catheter segments treated by different initial enzyme concentrations. A.)
Lysostaphin leaching from plain catheter segment B). Lysostaphin leaching from
NP coated catheter segment C). DispersinB leaching from plain catheter segment and D). DispersinB leaching from NP coated catheter segment. Black squares –
25 µg/ml, Red circles -50 µg/ml, Green triangles – 100 µg/ml, Blue inverted triangles - 250µg/ml and Light Blue diamonds - 500µg/ml. Leaching studies were
152 performed in the presence of 2 % (w/v) BSA in PBS buffer (pH = 7.4) at 37C.
(Mean ± SD; n= 2).
We observed an overall < 0.2 % enzyme leaching from all the studied samples after 72 h incubation in BSA (Figure 7.6 A-D). In case of coadsorption of lysostaphin and
DispersinB on plain and NP coated catheter segments, we observed an almost identical trend of < 0.2% leaching of enzyme for all the samples (data not shown). Again, the data obtained in our leaching study reflected a similar trend in the results obtained by authors at
Biosynexus Inc [240]. They observed almost negligible leaching after coating lysostaphin on a Teflon-like FEP polymer used in angiocath catheter segments and Polystyrene well plates.
7.3.4. Cell viability study
Of particular importance in assessing cellular response and biocompatibility is quantifying the prospective biomaterial‟s inherent cytotoxicity, or its ability to induce cell death [352]. In vitro models may provide answers regarding several aspects, including cell viability (cytotoxicity of implant material or extractables), proinflammatory effects and alteration of cellular response to different signaling molecules like cytokines [353].
Fibroblasts were chosen as an appropriate cell model as they play a crucial role in the inflammatory response [354]. This study aimed to evaluate the cellular response to enzyme coated catheter segments using an in vitro approach and methods in accordance with the International Organization for Standardization‟s (ISO) standard number 10993-5.
In particular, it is important to evaluate the cytotoxic effects of the two adsorbed enzymes, lysostaphin and DispersinB. Both enzymes are produced by different bacterial species.
Lysostaphin is isolated from culture filtrates of S. Staphylolyticus and DispersinB is
153 isolated from A. actinomycetemcomitans and could show potential cytotoxcity, especially if small amounts of enzyme leached from the catheter surface during in vivo trials.
The results of cytotoxicity tests on extracts by MTS assay are shown in Fig.7.7.
Raw quantitative data generated by the microplate reader were analyzed by first subtracting the difference of OD490 and OD650 reading to remove any background interference. The resultant mean OD readings from each well were calculated as an average of two replicates for each sample and expressed as a percentage of the OD values observed relative to OD values measured in the negative control wells (negative control – cells in media).
Figure 7.7. Direct cytotoxicity results of MTT assay that shows the percentage of cell viability after 48 h of all the catheter samples coated with the highest coating concentration that were placed in contact with 3T3 fibroblasts (n=2). Coating concentration for all samples was 500µg/ml. Black Bar – L-500, Red bar – D-
500, Yellow bar – LD-500, Wine Red – NPL-500, Grey Bar –NPD-500, Dark Blue bar – NPLD-500, Pink Bar – Uncoated cathter segment, Green Bar – PLA NP
154 coated catheter segment, Light Blue Bar – Negative control (Untreated
Fibroblasts) and Orange Bar - Positve control (Latex samples)
The metabolic activity of the fibroblasts is directly proportional to the cell viability. Overall, the cell viability in the negative controls (Plain cells in media) did not differ significantly from any of the test samples including uncoated or coated either by lysostaphin coated or DispersinB or via coadsorption of two enzymes (p > 0.05; p =
0.31718, One way ANOVA). There was also no significant difference between coated plain and NP coated catheter segments (p > 0.05; p = 0.14232, Paired‟t‟ test). Again, we did not see a significant difference between plain and NP coated segments. The positive control (Latex fragments) was shown expectedly to be highly cytotoxic showing ~ 14 % cell viability post 48 h incubation with the cells. These results demonstrated that the enzyme coated catheter segments were not cytotoxic in vitro in chosen experimental conditions.
7.3.5. Antibiofilm assay
DispersinB combined with various antimicrobials has been shown to work synergistically to inhibit the growth and proliferation of biofilm-embedded bacteria such as Staphylococcus epidermis and Staphylococcus aureus [107, 109, 110]. Figure 7.8 shows the percentage of biomass for the samples coated with the highest enzyme concentration – 0.5 mg/ml. We tested the in vitro antibiofilm activity of samples coated with DispersinB or lysostaphin alone or as a combination for both plain and NP coated catheter segments. We wanted to test synergy between lysostaphin (antibacterial) and
DispersinB (antibiofilm) and study if we could completely inhibit adhesion and growth for samples coadsorbed with both enzymes.
155
Figure 7.8. Biofilm inhibition assay of catheter samples done by Methylene Blue staining for catheter samples coated with highest enzyme concentration. Black
Bar – NPL-500, Red bar – NPD-500, Blue bar – NPLD-500, Green Bar – L-500,
Brown Bar –D-500, Light brown bar –LD-500 and Dark Blue Bar – Uncoated catheter segment; (Mean ± SD ; n=3).
We chose methylene blue as our dye for biofilm staining over crystal violet because in our preliminarily studies we observed significant non-specific binding of crystal violet to Polyurethane catheter segments. Percentage of biomass on the catheter segment was calculated from the following formula – ((Absorbance (660) of sample) /
(Absorbance (660) of control))*100
We observed an average of 26 ± 13 % biomass on NP coated catheter segments adsorbed with lysostaphin compared to plain lysostaphin samples that showed an average of 13 ± 1
% biomass and this was not significantly different (p>0.05; Paired „t‟ test) compared to samples coated with DispersinB for plain (12 ± 4)% and NP coated catheter samples (15 ±
4)% biomass. At coating concentrations of 0.5 mg/ml, it seems like both DispersinB and
156 lysostaphin inhibit biofilm formation on catheter surfaces and there was no significant difference in amount of biomass on catheters even after coadsorption of the two enzymes
(p> 0.05; One way ANOVA).
Polysaccharide intercellular adhesin (PIA), composed of poly-β-1,6-linked-N- acetylglucosamine, a sticky extracellular polysaccharide, is the main constituent of S. aureus and S. epidermis biofilm matrix [40]. PIA is produced by the intercellular adhesion (ica) operon present in nearly all S. aureus strains. DispersinB specifically hydrolyses the glycosidic linkages of poly-β-1,6-N-acetylglucosamine without inhibiting the growth of bacteria [108]. It is a stable glycoside hydrolase that functions in a narrow pH range and effective against both Gram positive and Gram negative bacteria.
DispersinB can only inhibit or disperse Staphylococci biofilms without affecting their growth, so it would be necessary to combine it with lysostaphin in order to inhibit the growth and cause lysis of biofilm-embedded bacteria. Previously, authors at Kane Biotech have shown synergy and in vitro and in vivo efficacy of DispersinB® and Triclosan combination against S.aureus and S.epidermis biofilms [109].
However, lysostaphin by itself is known to have some biofilm inhibiting activity.
Kokai-Kun et al have previously shown inhibition of S. aureus biofilms by lysostaphin in vitro on artificial surfaces and in vivo using a catheterized mouse model [341]. The authors have also observed that lysostaphin not only killed the staphylococci in the biofilm, but also appeared to prevent formation of the EPS matrix needed for biofilm formation on the artificial surfaces. The exact mechanism by which lysosytaphin disrupts biofilm formation remains unclear but the authors suggest that disruption occurs through the rapid lysis of the sessile staphylococci, which may be sufficient to destabilize the
157 entire biofilm matrix in a manner that allows cell detachment from surfaces [162]. We also similarly observed inhibition of S. aureus biofilm after coating lysostaphin on plain and
NP coated catheter segments. We hypothesize that the planktonic bacteria in the media
(during growth of biofilm) are directly interacting with the enzyme coating on our catheter segment. Our preliminary binding yield study suggests the possibility of multiple layers of enzyme coating on the catheter surface. These layers of lysosyaphin molecules are either directly interacting with planktonic bacterial cells or lysostaphin molecules slowly leach into the surrounding environment and lyse the bacteria thereby preventing subsequent attachment and biofilm formation. The latter however seems unlikely since data from the leaching study shows lysostaphin leaching (Fig 7.6) to be less than 1 % after 72 h of incubation in BSA.
A B
Figure 7.9. Images taken on a digital camera courstey department of
Bioengineering; A). Control – Uncoated catheter PU segment. B). Lysostaphin and DispersinB coated PU catheter segment.
7.3.6. Accelerated shelf life study
158
Biomedical devices and implants coated with antimicrobial enzymes in storage may change as they age, but they are considered to be stable as long as they retain specific activity as per manufacturer's specifications. The number of days that the product remains stable at the recommended storage conditions is referred to as the shelf life. Shelf life is commonly estimated using two types of stability testing: real-time stability tests and accelerated stability tests [355-357] In the former, a product is stored at recommended storage conditions and its shelf life is monitored while in the latter, a product is stored at elevated stress conditions using temperature, humidity, pH or a combination. In practice, evaluators use both real-time stability tests and accelerated stability tests. However, real- time stability testing can take a longer time to complete and so accelerated tests are often used as temporary measures to expedite processes in product development [356].
Temperature is the most common acceleration factor used for chemicals, pharmaceuticals, and biological products and accelerated aging testing is based on a thermodynamic temperature coefficient formulated by Von't Hof stating that for every 10 C degree rise in temperature, the rate of chemical reaction will at least double [358]. According to FDA regulations and package testing industry, this is useful in defining and justifying accelerated aging testing services [355]. An appropriate temperature for the accelerated shelf life testing must be chosen to avoid failure conditions such as deformation due to polymer melting. In our chosen study, the catheter segments are made up of Polyurethane which has a melting temperature of 240 C.
Based on the shelf life testing of lysostaphin coated hernia-repair meshes performed by
Sriram Sankar in collaboration with CMC (data not shown), we decided to choose a testing temperature of 60 C and a time period of 1 week for storage which is equivalent to
159
4 months of shelf life at room temperature. Before studying long term enzyme stability on catheter segments, it is necessary to lyophilize the enzyme to improve its stability.
Lyophilization gives the opportunity to help avoid enzyme denaturation caused by heating, by maintaining the samples frozen throughout the drying process. The first step in the lyophilization process is to freeze all the water molecules present in the sample and then it is placed in a lyophilizer and gradually heated in vacuum where all the water sublimates as vapor. The samples were frozen overnight at -80 C, lyophilized for 24 h and then sealed in air tight pouches (blasted with nitrogen) to prevent entry of air and moisture.
We studied accelerated shelf life testing for plain and NP coated catheter samples adsorbed with the highest coating concentration of 0.5 mg/ml of lysostaphin and
DispersinB using activity assays as described preciously. Turbidity assay was used to quantify the activity of lysostaphin and colorimetric assay was used to test activity of
DispersinB. We also tested activity of lysostaphin samples coadsorbed with DispersinB at the same coating concentration and vice versa. Figure 7.10 below shows the kinetic constant for bacterial hydrolysis for lysostaphin coated samples for plain and NP coated samples after 1 week at 60 C.
160
*
Figure 7.10. – Comparing the mean rate constants of bacterial lysis for both plain and NP coated catheter samples after 1 day and post 1 week at 60C (equivalent
4 months at RT). Red bar – NPL-500 represents NP coated catheter segment with initial lysostaphin concentration 500 µg/ml; Green bar – L-500 represents plain catheter segment with initial lysostaphin concentration of 500 µg/ml; (Mean ± SD; n= 2; * - p > 0.05)
We found that the initial rate of bacterial lysis for all the samples was almost equal to the lytic activity assayed before shelf life testing (p>0.05 – One way ANOVA) for plain and
NP coated samples at coating concentration of 0.5 mg/ml. Again, no significant difference between samples coated only with lysostaphin and coadsorbed with DispersinB (data not shown). These results seem to suggest that lysostaphin retained all of its activity after 1 week at 60 C (equivalent of 4 months at RT). We had also quantified lysotaphin activity before storage at 60 C, after a lyophilization time period of ~ 24 h and found there was no difference in its lytic activity (data not shown). This was done in order to demonstrate if
161 the process of freeze drying the catheter samples could affect enzyme activity before checking its stability for accelerated shelf life testing.
Figure 7.11. Comparing the mean rate constants of enzymatic hydrolysis for both plain and NP coated catheter samples after 1 day, post lyophilization (1 day) and post 1 week at 60C (equivalent 4 months at RT). Red bar – NPD-500 represents
NP coated catheter segment with initial DispersinB concentration 500 µg/ml;
Green bar – D-500 represents plain catheter segment with initial DispersinB concentration of 500 µg/ml; (Mean ± SD; n= 2 for 1st two sets and n = 3 for accelerated aging study )
In case of samples coated with DispersinB, we found that the rate of hydrolysis for uncoated catheter segment was not significantly different (p>0.05; n =3; One way
ANOVA) for both plain and NP coated catheter segments indicating that DispersinB possibly lost its activity during the accelerated testing study (Figure 7.11). But, we did observe ~ 90% percentage drop in DispersinB activity on catheter segments (compared to
162 its activity before freeze-drying) via the colorimetric assay after a 24 h lyophilization period suggesting that DispersinB (unlike lysostaphin) did not retain its hydrolytic activity after the freeze-drying step. During lyophilization process, stresses associated with freezing and drying cycle can sometimes lead to aggregation of the protein molecules with a consequent loss in enzyme activity. This happens because solutes (including salts in solution) may reach concentrations as high as 50 times their initial concentration due to separation of ice, leading to increased molecular interactions in the freeze-dried state [359,
360]. In order to prevent loss of enzyme activity during lyophilization, enzymes may be modified chemically or stabilizers may be added to the buffer solution [342, 344, 361,
362]. For, e.g. addition of cryoprotectants or stabilizers like Sucrose might help prevent denaturation of the functional enzymes such as DispersinB during the lyophilization process [363].
7.4. Conclusion
In conclusion, adsorbed enzyme activity for both lysostaphin and DispersinB, enzyme leaching, cytotoxicity and antibiofilm activity were not significantly different for plain and NP coated catheter segments. Our study showed that coating of nanoparticles on catheters overall did not significantly improve performance of enzymes on the catheters when adsorbed alone or through coadsorption. This is in part, possibly due to lack of a significantly higher surface area available for enzyme binding due to presence of nanoparticles, as it appears that enzymes were bound to the bare part of the catheter segment. We were unable to achieve good coating of PLA NPs on catheter surface, despite using excess of the NPs (based on theoretical surface area of catheter) in the initial
163
NP suspension. Furthermore, expected improvement in stability has not been achieved, again likely because most of the enzyme was present as adsorbed not on nanoparticles but on the bare part of the catheter segment.
Our accelerated shelf life testing showed lysostaphin retaining all of its activity after freeze-drying cycle and shelf life study, while DispersinB lost most of its activity during the lyophilization process and all of its activity after the shelf life study. Further work involving addition of stabilizers or cryoprotectant to preserve enzyme activity, especially for a temperature sensitive enzyme such as DispersinB may be performed.
However, based on the overall similar antibiofilm activity of the segments coated by lysostaphin, we conclude that use of lysostaphin may be preferable instead of DispersinB because of its higher stability.
164
CHAPTER 8
CONCLUSIONS AND FUTURE RECOMMENDATIONS
8.1 Conclusions
One of the most serious complications in the use of prosthetic devices such as Central venous catheters is device-associated infections and S.aureus is most frequently encountered nosocomial pathogen in this setting. However, the emergence of antibiotic resistant strains such as MRSA, has spurred the need for alternative sources of treatment.
Coatings of devices and implants using antibacterial enzymes and peptides have numerous advantages including low rates of resistance, broad spectrum of action, short treatment time and synergism (between peptides). Our study proposed to use nanoparticles as coatings on biomedical devices for higher enzyme loading, low leaching and long term durability. Based on the results from this study, the following conclusions can be drawn regarding the application of nanotechnology for targeted delivery of antibacterial enzymes to planktonic cell suspensions and as potential „surrogate surfaces‟ for enzyme adsorption on devices to prevent bacterial colonization and subsequent biofilm formation:
Intrinsic properties of nanoparticles such as charge and surface charge density can
be used to improve targeting of enzymes to bacterial cell walls. Lysozyme
covalently conjugated to positively charged nanoparticles (modified with amine
groups) shows better lytic activity to Micrococcus lysodeikticus than free enzyme
and lysozyme conjugated to negatively charged nanoparticles (modified with
chloromethyl and sulfate groups) through charge directed targeting to negatively
charged bacterial cell walls.
165
High surface to volume ratio enables attachment of multiple ligands on the surface
of nanoparticles including a specific antibody along with the corresponding
antibacterial enzyme to improve targeting to bacterial pathogens. Lysostaphin
adsorbed to Poly (lactic) acid nanoparticles in the presence and absence of a S.
aureus antibody showed significantly better activity due to multiple-ligand
targeting and antibody-directed targeting compared to free enzyme against
S.aureus.
Coating of antibacterial enzymes (lysostaphin) to synthetic hernia-repair meshes
through physical adsorption can prevent bacterial colonization and implant
associated infection. Our data specifically showed that enzyme leaching surfaces
could be effective in inhibiting bacterial growth and is significantly dependent
upon coating concentration in vivo.
We did not observe a significant difference in enzyme loading, leaching,
cytotoxicity, antibiofilm activity and long term durability for enzymes (lysostaphin
and DispersinB) coated on plain and PLA nanoparticle coated Polyurethane
catheter segments. This was in part due to the absence of a significant monolayer
of nanoparticle coating on the surface of the catheter segment.
8.2 Future recommendations
Our studies aimed at using nanotechnology for targeted delivery of antibacterial enzymes and as surface coatings for enzyme-based coatings on catheter segments. However, while we showed that nanoparticles can be used to improve targeted delivery of antibacterial enzymes to bacterial cells, we were not able to significantly improve characteristics of enzyme coated catheters after precoating catheters with nanoparticles. PLA nanoparticles
166 did not strongly bind to the surface of Polyurethane catheter segment despite using several fold excess of nanoparticles in the initial NP suspension, suggesting that nonspecific adsorption might not be an appropriate method and PLA not the most appropriate material for surface coating on Polyurethane. Stability of DispersinB could be improved during the lyophilization process by addition of cryoprotectants although the theory behind coating multiple enzymes including lysostaphin and DispersinB to achieve a synergistic interaction seems a little unlikely. Lysostaphin, by itself seems to have antibiofilm activity against S. aureus specifically and is more stable as an enzyme; hence coating of catheters with lysostaphin alone could probably be just as effective. However, DispersinB could be used in combination with other antimicrobials in inhibition of growth and biofilm formation for other slime producing bacterial strains other than S. aureus. Addition of cryoprotectants or stabilizers could prevent loss of activity during the lyophilization process [363]. Before choosing an appropriate stabilizer, it is important that it does not react with the functional enzyme and salt concentration in the buffer be low during the adsorption process.
167
APPENDICES
168
CHAPTER 9
Appendix A
Structure of tween 20
169
Appendix B
Structure of p-nitrophenyl penta-N-acetyl-ß-chitopentaoside
170
Appendix C
Structure of Alexafluor succidimyl ester
171
Appendix D
Structure of Pluronic F68
172
Appendix E
Digital camera images showing reduction of turbidity of S. aureus cell suspension
(clearing) through bacterial lysis by ultrapro samples coated with lysostaphin at different coating concentrations compared to uncoated samples.
173
Appendix F
Tapping mode AFM images showing topography and height of a polyurethane catheter segment.
153.89 nm
20
15
Z[nm] 10
5
0 200nm 0 200 400 600 800 1000
0.00 nm X[nm]
174
Appendix G
Structure of 4-nitrophenyl-N-acetyl- D-glucosaminide
175
Appendix H
Schematic of MTS assay
176
CHAPTER 10
References
1. Thanbichler, M., S.C. Wang, and L. Shapiro, The bacterial nucleoid: a highly organized and dynamic structure. J Cell Biochem, 2005. 96(3): p. 506-21. 2. Kachlany, S.C., et al., Genes for tight adherence of Actinobacillus actinomycetemcomitans: from plaque to plague to pond scum. Trends Microbiol, 2001. 9(9): p. 429-437. 3. Donlan, R.M., Biofilms: microbial life on surfaces. Emerg Infect Dis, 2002. 8(9): p. 881-90. 4. Branda, S.S., et al., Biofilms: the matrix revisited. Trends Microbiol, 2005. 13(1): p. 20-6. 5. Davey, M.E. and A. O'Toole G, Microbial biofilms: from ecology to molecular genetics. Microbiol Mol Biol Rev, 2000. 64(4): p. 847-67. 6. Donlan, R.M. and J.W. Costerton, Biofilms: survival mechanisms of clinically relevant microorganisms. Clin Microbiol Rev, 2002. 15(2): p. 167-93. 7. van Heijenoort, J., Formation of the glycan chains in the synthesis of bacterial peptidoglycan. Glycobiology, 2001. 11(3): p. 25R-36R. 8. Koch, A.L., Bacterial wall as target for attack: past, present, and future research. Clin Microbiol Rev, 2003. 16(4): p. 673-87. 9. Salton, M.R.J.T.b.c.e.a.h.p. and p. 1–22., The bacterial cell envelope—a historical perspective,M. Ghuysen and R. Hakenbeck (ed.), ed1994 Amsterdam, The Netherlands: Elsevier Science BV. p. 1–22. 10. Beveridge, T.J., Mechanism of gram variability in select bacteria. J Bacteriol, 1990. 172(3): p. 1609-20. 11. Hancock, I.C., Bacterial cell surface carbohydrates: structure and assembly. Biochem Soc Trans, 1997. 25(1): p. 183-7. 12. Ghuysen, J.M. and J.L. Strominger, Structure of the Cell Wall of Staphylococcus Aureus, Strain Copenhagen. I. Preparation of Fragments by Enzymatic Hydrolysis. Biochemistry, 1963. 2: p. 1110-9.
177
13. Ghuysen, J.M. and J.L. Strominger, Structure of the Cell Wall of Staphylococcus Aureus, Strain Copenhagen. Ii. Separation and Structure of Disaccharides. Biochemistry, 1963. 2: p. 1119-25. 14. Snowden, M.A. and H.R. Perkins, Peptidoglycan cross-linking in Staphylococcus aureus. An apparent random polymerisation process. Eur J Biochem, 1990. 191(2): p. 373-7. 15. Henze, U., et al., Influence of femB on methicillin resistance and peptidoglycan metabolism in Staphylococcus aureus. J Bacteriol, 1993. 175(6): p. 1612-20. 16. Glauner, B., J.V. Holtje, and U. Schwarz, The composition of the murein of Escherichia coli. J Biol Chem, 1988. 263(21): p. 10088-95. 17. Tipper, D.J., Structures of the cell wall peptidoglycans of Staphylococcus epidermidis Texas 26 and Staphylococcus aureus Copenhagen. II. Structure of neutral and basic peptides from hydrolysis with the Myxobacter al-1 peptidase. Biochemistry, 1969. 8(5): p. 2192-202. 18. Tipper, D.J. and M.F. Berman, Structures of the cell wall peptidoglycans of Staphylococcus epidermidis Texas 26 and Staphylococcus aureus Copenhagen. I. Chain length and average sequence of cross-bridge peptides. Biochemistry, 1969. 8(5): p. 2183-92. 19. Tipper, D.J. and J.L. Strominger, Mechanism of action of penicillins: a proposal based on their structural similarity to acyl-D-alanyl-D-alanine. Proc Natl Acad Sci U S A, 1965. 54(4): p. 1133-41. 20. Tipper, D.J. and J.L. Strominger, Biosynthesis of the peptidoglycan of bacterial cell walls. XII. Inhibition of cross-linking by penicillins and cephalosporins: studies in Staphylococcus aureus in vivo. J Biol Chem, 1968. 243(11): p. 3169-79. 21. Strominger, J.L. and J.M. Ghuysen, Mechanisms of enzymatic bacteriaolysis. Cell walls of bacteri are solubilized by action of either specific carbohydrases or specific peptidases. Science, 1967. 156(772): p. 213-21. 22. Ellwood, D.C. and D.W. Tempest, Influence of growth environment on the cell wall anionic polymers in some Gram-positive bacteria. J Gen Microbiol, 1969. 57(3): p. xv.
178
23. Archibald, A.R., J. Baddiley, and N.L. Blumsom, The teichoic acids. Adv Enzymol Relat Areas Mol Biol, 1968. 30: p. 223-53. 24. Cabeen, M.T. and C. Jacobs-Wagner, Bacterial cell shape. Nat Rev Micro, 2005. 3(8): p. 601-610. 25. Naumova, I.B. and A.S. Shashkov, Anionic polymers in cell walls of gram-positive bacteria. Biochemistry (Mosc), 1997. 62(8): p. 809-40. 26. Araki, Y. and E. Ito, Linkage units in cell walls of gram-positive bacteria. Crit Rev Microbiol, 1989. 17(2): p. 121-35. 27. Navarre, W.W. and O. Schneewind, Surface proteins of gram-positive bacteria and mechanisms of their targeting to the cell wall envelope. Microbiol Mol Biol Rev, 1999. 63(1): p. 174-229. 28. Baddiley, J., Teichoic acids in cell walls and membranes of bacteria. Essays Biochem, 1972. 8: p. 35-77. 29. Fischer, W., Lipoteichoic acids and lipoglycans., in In Bacterial Cell Wall J.M.G.R. Hakenbeck, Editor (1994), Elsevier: Amsterdam. p. pp. 199–215. 30. Munson, R.S.G., L, Teichoic acid and peptidoglycan assembly in Gram-positive organisms, in In Biology of Carbohydrates, , V.G.P. Robbins., Editor (1981), Wiley: New York. p. vol. 1, pp. 91–121. 31. R., D.R.J.a.D., The bacterial cell: peptidoglycan., in Molecular Medical Microbiology,, S.M. (ed.), Editor 2001, Academic Press: London. p. pp. 137–153. 32. Schleifer, K.H. and O. Kandler, Peptidoglycan types of bacterial cell walls and their taxonomic implications. Bacteriol Rev, 1972. 36(4): p. 407-77. 33. Beveridge, T.J. and L.L. Graham, Surface layers of bacteria. Microbiol Rev, 1991. 55(4): p. 684-705. 34. Beveridge, T.J., Ultrastructure, chemistry, and function of the bacterial wall. Int Rev Cytol, 1981. 72: p. 229-317. 35. Raetz, C.R. and C. Whitfield, Lipopolysaccharide endotoxins. Annu Rev Biochem, 2002. 71: p. 635-700. 36. Duong, F., et al., Biogenesis of the gram-negative bacterial envelope. Cell, 1997. 91(5): p. 567-73.
179
37. Doerrler, W.T., Lipid trafficking to the outer membrane of Gram-negative bacteria. Mol Microbiol, 2006. 60(3): p. 542-52. 38. Nikaido, H., Prevention of drug access to bacterial targets: permeability barriers and active effluxScience, 1994. 264.(5157): p. 382 - 388. 39. Fowler, V.G., et al., Staphylococcus aureus bacteremia after median sternotomy - Clinical utility of blood culture results in the identification of postoperative mediastinitis. Circulation, 2003. 108(1): p. 73-78. 40. Gotz, F., Staphylococcus and biofilms. Molecular Microbiology, 2002. 43(6): p. 1367-1378. 41. Gordon, R.J. and F.D. Lowy, Pathogenesis of methicillin-resistant Staphylococcus aureus infection. Clinical Infectious Diseases, 2008. 46: p. S350-S359. 42. Yarwood, J.M., et al., Quorum sensing in Staphylococcus aureus biofilms. Journal of Bacteriology, 2004. 186(6): p. 1838-1850. 43. Yarwood, J.M. and P.M. Schlievert, Quorum sensing in Staphylococcus infections. Journal of Clinical Investigation, 2003. 112(11): p. 1620-1625. 44. Corey, G.R., Staphylococcus aureus Bloodstream Infections: Definitions and Treatment. Clinical Infectious Diseases, 2009. 48: p. S254-S259. 45. Fung, H.B., J.Y. Chang, and S. Kuczynski, A practical guide to the treatment of complicated skin and soft tissue infections. Drugs, 2003. 63(14): p. 1459-80. 46. Horan, T.C., et al., CDC definitions of nosocomial surgical site infections, 1992: a modification of CDC definitions of surgical wound infections. Infect Control Hosp Epidemiol, 1992. 13(10): p. 606-8. 47. Jawhara, S. and S. Mordon, In vivo imaging of bioluminescent Escherichia coli in a cutaneous wound infection model for evaluation of an antibiotic therapy. Antimicrob Agents Chemother, 2004. 48(9): p. 3436-41. 48. Kirkland, K.B., et al., The impact of surgical-site infections in the 1990s: attributable mortality, excess length of hospitalization, and extra costs. Infect Control Hosp Epidemiol, 1999. 20(11): p. 725-30. 49. Bratzler, D.W. and P.M. Houck, Antimicrobial prophylaxis for surgery: an advisory statement from the National Surgical Infection Prevention Project. Clin Infect Dis, 2004. 38(12): p. 1706-15.
180
50. Burke, J.P., Infection control - a problem for patient safety. N Engl J Med, 2003. 348(7): p. 651-6. 51. Pharmacists., A.S.o.H.-S., ASHP therapeuticguidelines on antimicrobial prophylaxis in surgery. Am J Health Syst Pharm, 2003. 56: p. 1839–88. 52. McManus, M.C., Mechanisms of bacterial resistance to antimicrobial agents. Am J Health Syst Pharm, 1997. 54(12): p. 1420-33; quiz 1444-6. 53. Neu, H.C., The crisis in antibiotic resistance. Science, 1992. 257(5073): p. 1064- 73. 54. Drlica, K. and X. Zhao, DNA gyrase, topoisomerase IV, and the 4-quinolones. Microbiol Mol Biol Rev, 1997. 61(3): p. 377-92. 55. Storm, D.R., K.S. Rosenthal, and P.E. Swanson, Polymyxin and related peptide antibiotics. Annu Rev Biochem, 1977. 46: p. 723-63. 56. Raad, I., A. Alrahwan, and K. Rolston, Staphylococcus epidermidis: emerging resistance and need for alternative agents. Clin Infect Dis, 1998. 26(5): p. 1182-7. 57. Tenover, F.C., Mechanisms of antimicrobial resistance in bacteria. Am J Infect Control, 2006. 34(5 Suppl 1): p. S3-10; discussion S64-73. 58. Leblebicioglu, H., et al., Surveillance of antimicrobial resistance in gram-negative isolates from intensive care units in Turkey: analysis of data from the last 5 years. J Chemother, 2002. 14(2): p. 140-6. 59. Sahm, D.F., et al., Multidrug-resistant urinary tract isolates of Escherichia coli: prevalence and patient demographics in the United States in 2000. Antimicrob Agents Chemother, 2001. 45(5): p. 1402-6. 60. Schumacher, H., J. Scheibel, and J.K. Moller, Cross-resistance patterns among clinical isolates of Klebsiella pneumoniae with decreased susceptibility to cefuroxime. J Antimicrob Chemother, 2000. 46(2): p. 215-21. 61. Ungheri, D., E. Albini, and G. Belluco, In-vitro susceptibility of quinolone- resistant clinical isolates of Escherichia coli to fosfomycin trometamol. J Chemother, 2002. 14(3): p. 237-40. 62. Poole, K., Outer membranes and efflux: the path to multidrug resistance in Gram- negative bacteria. Curr Pharm Biotechnol, 2002. 3(2): p. 77-98.
181
63. Ridley, A. and E.J. Threlfall, Molecular epidemiology of antibiotic resistance genes in multiresistant epidemic Salmonella typhimurium DT 104. Microb Drug Resist, 1998. 4(2): p. 113-8. 64. Gales, A.C., et al., Characterization of Pseudomonas aeruginosa isolates: occurrence rates, antimicrobial susceptibility patterns, and molecular typing in the global SENTRY Antimicrobial Surveillance Program, 1997-1999. Clin Infect Dis, 2001. 32 Suppl 2: p. S146-55. 65. Hanberger, H., et al., Surveillance of antibiotic resistance in European ICUs. J Hosp Infect, 2001. 48(3): p. 161-76. 66. Tacconelli, E., et al., Multidrug-resistant Pseudomonas aeruginosa bloodstream infections: analysis of trends in prevalence and epidemiology. Emerg Infect Dis, 2002. 8(2): p. 220-1. 67. Waaij, D.V.d., Colonisation resistance of the digestive tract – mechanism and clinical consequences. Nahrung, 1987. 31: p. 507–517. 68. Dancer, S.J., The problem with cephalosporins. Journal of Antimicrobial Chemotherapy, 2001. 48(4): p. 463-478. 69. Heggers, J.P., et al., The effectiveness of processed grapefruit-seed extract as an antibacterial agent: II. Mechanism of action and in vitro toxicity. J Altern Complement Med, 2002. 8(3): p. 333-40. 70. Monafo, W.W. and M.A. West, Current treatment recommendations for topical burn therapy. Drugs, 1990. 40(3): p. 364-73. 71. Murphy, K.D., J.O. Lee, and D.N. Herndon, Current pharmacotherapy for the treatment of severe burns. Expert Opin Pharmacother, 2003. 4(3): p. 369-84. 72. Palmieri, T.L. and D.G. Greenhalgh, Topical treatment of pediatric patients with burns: a practical guide. Am J Clin Dermatol, 2002. 3(8): p. 529-34. 73. Cooper ML. Laxer JA, H.J., The cytotoxic effects of commonly used topical antimicrobial agents on human fibroblasts andkeratinocytes. J Trauma, 1991. 31: p. 775-784. 74. McCauley, R.L., et al., In vitro toxicity of topical antimicrobial agents to human fibroblasts. The Journal of surgical research, 1989. 46(3): p. 267-74.
182
75. Kuroyanagi, Y., E. Kim, and N. Shioya, Evaluation of a synthetic wound dressing capable of releasing silver sulfadiazine. The Journal of burn care & rehabilitation, 1991. 12(2): p. 106-15. 76. Goldberg, R.L., S.R. Kaplan, and G.C. Fuller, Effect of heavy metals on human rheumatoid synovial cell proliferation and collagen synthesis. Biochemical pharmacology, 1983. 32(18): p. 2763-6. 77. Mcvay, C.S., M. Velasquez, and J.A. Fralick, Phage therapy of Pseudomonas aeruginosa infection in a mouse burn wound model. Antimicrob Agents Chemother, 2007. 51(6): p. 1934-1938. 78. Capparelli, R., et al., Experimental phage therapy against Staphylococcus aureus in mice. Antimicrob Agents Chemother, 2007. 51(8): p. 2765-2773. 79. Marr, A.K., W.J. Gooderham, and R.E. Hancock, Antibacterial peptides for therapeutic use: obstacles and realistic outlook. Curr Opin Pharmacol, 2006. 6(5): p. 468-72. 80. Jenssen, H., P. Hamill, and R.E. Hancock, Peptide antimicrobial agents. Clin Microbiol Rev, 2006. 19(3): p. 491-511. 81. Hancock, R.E. and H.G. Sahl, Antimicrobial and host-defense peptides as new anti-infective therapeutic strategies. Nat Biotechnol, 2006. 24(12): p. 1551-7. 82. Hancock, R.E.W. and M.G. Scott, The role of antimicrobial peptides in animal defenses. Proceedings of the National Academy of Sciences of the United States of America, 2000. 97(16): p. 8856-8861. 83. Claus Crone Fuglsang, C.J., Stephan Christgau and Jens Adler-Nissen, Antimicrobial enzymes: Applications and future potential in the food industry. Trends in Food Science & Technology December 1995. 6: p. 390-396. 84. Vollmer, W., et al., Bacterial peptidoglycan (murein) hydrolases. Fems Microbiology Reviews, 2008. 32(2): p. 259-286. 85. Abu-Soud, H.M. and S.L. Hazen, Nitric oxide is a physiological substrate for mammalian peroxidases. Journal of Biological Chemistry, 2000. 275(48): p. 37524- 37532.
183
86. Lindgren, A., et al., Biosensors based on novel peroxidases with improved properties in direct and mediated electron transfer. Biosensors & Bioelectronics, 2000. 15(9-10): p. 491-497. 87. O‟Brien, P.J., Peroxidases. Chemico-Biological Interactions 2000. 129: p. 113– 139. 88. Brogden, K.A., Antimicrobial peptides: Pore formers or metabolic inhibitors in bacteria? Nature Reviews Microbiology, 2005. 3(3): p. 238-250. 89. Yang, D., et al., Mammalian defensins in immunity: more than just microbicidal. Trends Immunol, 2002. 23(6): p. 291-6. 90. Yang, D., et al., Multiple roles of antimicrobial defensins, cathelicidins, and eosinophil-derived neurotoxin in host defense. Annu Rev Immunol, 2004. 22: p. 181- 215. 91. Koczulla, A.R. and R. Bals, Antimicrobial peptides: current status and therapeutic potential. Drugs, 2003. 63(4): p. 389-406. 92. Zasloff, M., Antimicrobial peptides in health and disease. New England Journal of Medicine, 2002. 347(15): p. 1199-1200. 93. Zasloff, M., Antimicrobial peptides of multicellular organisms. Nature, 2002. 415(6870): p. 389-395. 94. Zanetti, M., R. Gennaro, and D. Romeo, Cathelicidins - a Novel Protein Family with a Common Proregion and a Variable C-Terminal Antimicrobial Domain. Febs Letters, 1995. 374(1): p. 1-5. 95. Sorensen, O.E., et al., Human cathelicidin, hCAP-18, is processed to the antimicrobial peptide LL-37 by extracellular cleavage with proteinase 3. Blood, 2001. 97(12): p. 3951-3959. 96. Hancock, R.E.W. and R. Lehrer, Cationic peptides: a new source of antibiotics. Trends in Biotechnology, 1998. 16(2): p. 82-88. 97. Powers, J.P. and R.E. Hancock, The relationship between peptide structure and antibacterial activity. Peptides, 2003. 24(11): p. 1681-91. 98. F. Niyonsaba, H.O., Protective roles of the skin against infection: Implication of naturally occurring human antimicrobial agents β-defensins, cathelicidin LL-37 and lysozyme. J. Derm. Sci., 2005. 40: p. 157-168.
184
99. Mor, A., K. Hani, and P. Nicolas, The Vertebrate Peptide Antibiotics Dermaseptins Have Overlapping Structural Features but Target Specific Microorganisms. Journal of Biological Chemistry, 1994. 269(50): p. 31635-31641. 100. Matsuzaki, K., Molecular action mechanisms and membrane recognition of membrane-acting antimicrobial peptides. Yakugaku Zasshi-Journal of the Pharmaceutical Society of Japan, 1997. 117(5): p. 253-264. 101. Levy, O., et al., Individual and Synergistic Effects of Rabbit Granulocyte Proteins on Escherichia-Coli. Journal of Clinical Investigation, 1994. 94(2): p. 672-682. 102. Bals, R., et al., The peptide antibiotic LL-37/hCAP-18 is expressed in epithelia of the human lung where it has broad antimicrobial activity at the airway surface. Proc Natl Acad Sci U S A, 1998. 95(16): p. 9541-9546. 103. Nagaoka, I., et al., Synergistic actions of antibacterial neutrophil defensins and cathelicidins. Inflamm Res, 2000. 49(2): p. 73-9. 104. Chen, X., et al., Synergistic effect of antibacterial agents human beta-defensins, cathelicidin LL-37 and lysozyme against Staphylococcus aureus and Escherichia coli. J Dermatol Sci, 2005. 40(2): p. 123-32. 105. Singh, P.K., et al., Synergistic and additive killing by antimicrobial factors found in human airway surface liquid. American Journal of Physiology-Lung Cellular and Molecular Physiology, 2000. 279(5): p. L799-L805. 106. Johansen, C., P. Falholt, and L. Gram, Enzymatic removal and disinfection of bacterial biofilms. Applied and Environmental Microbiology, 1997. 63(9): p. 3724- 3728. 107. Donelli, G., et al., Synergistic activity of dispersin B and cefamandole nafate in inhibition of staphylococcal biofilm growth on polyurethanes. Antimicrobial Agents and Chemotherapy, 2007. 51(8): p. 2733-2740. 108. Izano, E.A., H. Wang, C. Ragunath, N. Ramasubbu, and J. B. Kaplan, Detachment and killing of Aggregatibacter actinomycetemcomitans biofilms by DispersinB and SDS. J. Dent. Res, 2007a. 86: p. 618-622. 109. Darouiche, R.O., et al., Antimicrobial and antibiofilm efficacy of triclosan and DispersinB (R) combination. Journal of Antimicrobial Chemotherapy, 2009. 64(1): p. 88-93
185
110. Izano, E.A., et al., Poly-N-acetylglucosamine mediates biofilm formation and antibiotic resistance in Actinobacillus pleuropneumoniae. Microbial Pathogenesis, 2007. 43(1): p. 1-9. 111. Eckhart, L., et al., DNase1L2 suppresses biofilm formation by Pseudomonas aeruginosa and Staphylococcus aureus. British Journal of Dermatology, 2007. 156(6): p. 1342-1345. 112. Proctor, V.A. and F.E. Cunningham, The chemistry of lysozyme and its use as a food preservative and a pharmaceutical. Crit Rev Food Sci Nutr, 1988. 26(4): p. 359- 95. 113. Laible, N.J. and G.R. Germaine, Bactericidal activity of human lysozyme, muramidase-inactive lysozyme, and cationic polypeptides against Streptococcus sanguis and Streptococcus faecalis: inhibition by chitin oligosaccharides. Infect Immun, 1985. 48(3): p. 720-8. 114. Arnold, R.R., et al., Bactericidal activity of human lactoferrin: differentiation from the stasis of iron deprivation. Infect Immun, 1982. 35(3): p. 792-9. 115. Arnold, R.R., M.F. Cole, and J.R. McGhee, A bactericidal effect for human lactoferrin. Science, 1977. 197(4300): p. 263-5. 116. Lehrer, R.I., et al., Interaction of human defensins with Escherichia coli. Mechanism of bactericidal activity. J Clin Invest, 1989. 84(2): p. 553-61. 117. Huttner, K.M. and C.L. Bevins, Antimicrobial peptides as mediators of epithelial host defense. Pediatr Res, 1999. 45(6): p. 785-94. 118. Murakami, M., et al., Cathelicidin anti-microbial peptide expression in sweat, an innate defense system for the skin. J Invest Dermatol, 2002. 119(5): p. 1090-5. 119. Bals, R., et al., Human beta-defensin 2 is a salt-sensitive peptide antibiotic expressed in human lung. J Clin Invest, 1998. 102(5): p. 874-80. 120. Hsu, C.H., et al., Structural and DNA-binding studies on the bovine antimicrobial peptide, indolicidin: evidence for multiple conformations involved in binding to membranes and DNA. Nucleic Acids Res, 2005. 33(13): p. 4053-64. 121. Falla, T.J., D.N. Karunaratne, and R.E. Hancock, Mode of action of the antimicrobial peptide indolicidin. J Biol Chem, 1996. 271(32): p. 19298-303.
186
122. Ruhr, E. and H.G. Sahl, Mode of action of the peptide antibiotic nisin and influence on the membrane potential of whole cells and on cytoplasmic and artificial membrane vesicles. Antimicrob Agents Chemother, 1985. 27(5): p. 841-5. 123. Breukink, E. and B. de Kruijff, The lantibiotic nisin, a special case or not? Biochim Biophys Acta, 1999. 1462(1-2): p. 223-34. 124. Zasloff, M., B. Martin, and H.C. Chen, Antimicrobial activity of synthetic magainin peptides and several analogues. Proc Natl Acad Sci U S A, 1988. 85(3): p. 910-3. 125. Matsuzaki, K., Magainins as paradigm for the mode of action of pore forming polypeptides. Biochim Biophys Acta, 1998. 1376(3): p. 391-400. 126. Heller, W.T., et al., Multiple states of beta-sheet peptide protegrin in lipid bilayers. Biochemistry, 1998. 37(49): p. 17331-8. 127. Gazit, E., et al., Structure and orientation of the mammalian antibacterial peptide cecropin P1 within phospholipid membranes. J Mol Biol, 1996. 258(5): p. 860-70. 128. Hultmark, D., et al., Insect immunity: isolation and structure of cecropin D and four minor antibacterial components from Cecropia pupae. Eur J Biochem, 1982. 127(1): p. 207-17. 129. De Groot, A.S. and D.W. Scott, Immunogenicity of protein therapeutics. Trends in immunology, 2007. 28(11): p. 482-90. 130. Barbosa, M.D. and E. Celis, Immunogenicity of protein therapeutics and the interplay between tolerance and antibody responses. Drug discovery today, 2007. 12(15-16): p. 674-81. 131. McPhee, J.B., Scott, M.G. & Hancock, R.E.W. , Design of host defence peptides for antimicrobial and immunity enhancing activities. . Comb. Chem. High Throughput Screen., 2005. 8: p. 257–272. 132. Liang, J.F., Y.T. Li, and V.C. Yang, Biomedical application of immobilized enzymes. J Pharm Sci, 2000. 89(8): p. 979-90. 133. Schmid, A., et al., Industrial biocatalysis today and tomorrow. Nature, 2001. 409(6817): p. 258-68.
187
134. Zhu, G. and P. Wang, Polymer-enzyme conjugates can self-assemble at oil/water interfaces and effect interfacial biotransformations. J Am Chem Soc, 2004. 126(36): p. 11132-3. 135. Newman, J.D. and S.J. Setford, Enzymatic biosensors. Mol Biotechnol, 2006. 32(3): p. 249-68. 136. Seurynck-Servoss, S.L., et al., Immobilization strategies for single-chain antibody microarrays. Proteomics, 2008. 8(11): p. 2199-210. 137. Garnett, M.C., Targeted drug conjugates: principles and progress. Adv Drug Deliv Rev, 2001. 53(2): p. 171-216. 138. Cao, L., Immobilised enzymes: science or art? Curr Opin Chem Biol, 2005. 9(2): p. 217-26. 139. Melik-Nubarov, N.S., et al., [Correlation of enzyme thermostability and its surface hydrophobicity (using a modified alpha-chymotrypsin as an example)]. Mol Biol (Mosk), 1990. 24(2): p. 346-57. 140. O Abian, l.W., C. Mateo, G. Fernandez-Lorente, J.M. Palomo, R. Fernandez- Lafuente, J.M. Guisan, D. Re, A. Tam and M. Daminatti,, Preparation of artificial hyper-hydrophilic micro-environments (polymeric salts) surrounding enzyme molecules. new enzyme derivatives to be used in any reaction medium. J Mol Catal B: Enzymatic 2002. 19 . p. 295–303. 141. J.M. Guisan, P.P.S., R. Fernandez-Lafuente, G. Fernandez-Lorente, C. Mateo, P.J. Halling, D. Kennedy, E. Miyata and D. Re, Preparation of new lipase derivatives with high activity-stability in anhydrous media: adsorption on hydrophobic supports plus hydrophilization with polyethylenimine. J Mol Catal B: Enzymatic, 2001. 11: p. 817– 824. 142. S. Braun, S.R., R. Zusman, D. Avnir and M. Ottolenghi Biochemically Active Sol- Gel Glasses: The Trapping of Enzymes". Materials Lett.. 1990. 10: p. 1-5 143. S. Braun, S.S., S. Rappoport, D. Avnir and M. Ottolenghi . , Biocatalysis by Sol- Gel Entrapped Enzymes. J. Non-Cryst. Solids, 1992. 147(148): p. 739-743. 144. Quiocho, F.A.R., F. M., Biochemistry, 1966. 5 p. 4062-4076. 145. Malmsten, Biopolymers at Interfaces 2003 New York: Marcel Dekker.
188
146. Denis F A, H.P., Sutherland D S, et al., Protein adsorption onmodel surfaces with controlled nanotopography and chemistry.. Langmuir, 2002. 18(3): p. 819―828. 148. Sheldon, R.A., Enzyme immobilization: The quest for optimum performance. Advanced Synthesis & Catalysis, 2007. 349(8-9): p. 1289-1307. 149. Kawaguchi, H., Functional polymer microspheres. Progress in Polymer Science, 2000. 25(8): p. 1171-1210. 150. Molina-Bolivar, J.A. and F. Galisteo-Gonzalez, Latex immunoagglutination assays. Journal of Macromolecular Science-Polymer Reviews, 2005. C45(1): p. 59-98. 151. Yodoya, S., et al., Immobilization of bromelain onto porous copoly(gamma- methyl-L-glutamate/L-leucine) beads. European Polymer Journal, 2003. 39(1): p. 173- 180. 152. Hahn, R., et al., Directed immobilization of peptide ligands to accessible pore sites by conjugation with a placeholder molecule. Analytical Chemistry, 2003. 75(3): p. 543-548. 153. Soellner, M.B., et al., Site-specific protein immobilization by Staudinger ligation. Journal of the American Chemical Society, 2003. 125(39): p. 11790-11791. 154. Nisnevitch, M. and M.A. Firer, The solid phase in affinity chromatography: strategies for antibody attachment. Journal of Biochemical and Biophysical Methods, 2001. 49(1-3): p. 467-480. 155. Oates, M.R., et al., Kinetic studies on the immobilization of antibodies to high- performance liquid chromatographic supports. Bioconjugate Chemistry, 1998. 9(4): p. 459-465. 156. Smith, C.L.M., G. S.; Nguyen, G. H, Top. Curr. Chem, 2006. 261: p. 63-90. 157. Kim, S. and J. Lee, Folate-targeted drug-delivery systems prepared by nano- comminution. Drug development and industrial pharmacy, 2011. 37(2): p. 131-8. 158. M. Wilchek, E.A.B., Methods Enzymol., 1990. 184: p. 5-13. 159. Luo, S.F. and D.R. Walt, Avidin Biotin Coupling as a General-Method for Preparing Enzyme-Based Fiber-Optic Sensors. Analytical Chemistry, 1989. 61(10): p. 1069-1072.
189
160. Vermette, P., et al., Immobilization and surface characterization of NeutrAvidin biotin-binding protein on different hydrogel interlayers. Journal of Colloid and Interface Science, 2003. 259(1): p. 13-26. 161. Sun, X.L., et al., Design and synthesis of biotin chain-terminated glycopolymers for surface glycoengineering. Journal of the American Chemical Society, 2002. 124(25): p. 7258-7259. 162. Ratner, B.D., New Ideas in Biomaterials Science - a Path to Engineered Biomaterials. J Biomed Mater Res, 1993. 27(7): p. 837-850. 163. Ostuni, E., et al., A survey of structure-property relationships of surfaces that resist the adsorption of protein. Langmuir, 2001. 17(18): p. 5605-5620. 164. Szleifer, I., Protein adsorption on tethered polymer layers: effect of polymer chain architecture and composition. Physica A, 1997. 244(1-4): p. 370-388. 165. Morra, M., On the molecular basis of fouling resistance. Journal of Biomaterials Science-Polymer Edition, 2000. 11(6): p. 547-569. 166. McPherson, T., et al., Prevention of protein adsorption by tethered poly(ethylene oxide) layers: Experiments and single-chain mean-field analysis. Langmuir, 1998. 14(1): p. 176-186. 167. Lez, M.J.a.F.G.-G., J Macromolecular Science-C Polymer Reviews, 2005. 45: p. 59–98. 168. Masschalck, B. and C.W. Michiels, Antimicrobial properties of lysozyme in relation to foodborne vegetative bacteria. Crit Rev Microbiol, 2003. 29(3): p. 191-214. 169. Chinn, M.K., et al., The effect of polyethylene glycol on the mechanics and ATPase activity of active muscle fibers. Biophys J, 2000. 78(2): p. 927-939. 170. Lee, C.S. and B.G. Kim, Improvement of protein stability in protein microarrays. Biotechnology Letters, 2002. 24(10): p. 839-844. 171. Gref, R., et al., The Controlled Intravenous Delivery of Drugs Using Peg-Coated Sterically Stabilized Nanospheres. Adv Drug Deliv Rev, 1995. 16(2-3): p. 215-233. 172. Mosqueira, V.C.F., et al., Interactions between a macrophage cell line (J774A1) and surface-modified poly(D,L-lactide) nanocapsules bearing poly(ethylene glycol). Journal of Drug Targeting, 1999. 7(1): p. 65-78.
190
173. Calvo, P., et al., Long-circulating PEGylated polycyanoacrylate nanoparticles as new drug carrier for brain delivery. Pharmaceutical Research, 2001. 18(8): p. 1157- 1166. 174. J. K. Gbadamosi, A.C.H.a.S.M.M., FEBS Lett, 2002. 532: p. 338. 175. C. M. F. Soares, O.A.D.S., H. F. De Castro, and G.M.Z. F. F. De Moraes, Appl. Biochem. Biotechnol. 2004. 113: p. 307 – 319. 176. Niemeyer, C.M., Functional hybrid devices of proteins and inorganic nanoparticles. Angew Chem Int Ed Engl, 2003. 42(47): p. 5796-800. 177. Katz, E. and I. Willner, Integrated nanoparticle-biomolecule hybrid systems: synthesis, properties, and applications. Angew Chem Int Ed Engl, 2004. 43(45): p. 6042-108. 178. Chan, W.C. and S. Nie, Quantum dot bioconjugates for ultrasensitive nonisotopic detection. Science, 1998. 281(5385): p. 2016-8. 179. Bruchez, M., Jr., et al., Semiconductor nanocrystals as fluorescent biological labels. Science, 1998. 281(5385): p. 2013-6. 180. Besteman, K., et al., Enzyme-coated carbon nanotubes as single-molecule biosensors. Nano Letters, 2003. 3(6): p. 727-730. 181. Mitchell, D., et al., Lipoxins inhibit Akt/PKB activation and cell cycle progression in human mesangial cells. American Journal of Pathology, 2004. 164(3): p. 937-946. 182. Rege, K., et al., Enzyme-polymer-single walled carbon nanotube composites as biocatalytic films. Nano Letters, 2003. 3(6): p. 829-832. 183. Takahashi, H., et al., Catalytic activity in organic solvents and stability of immobilized enzymes depend on the pore size and surface characteristics of mesoporous silica. Chemistry of Materials, 2000. 12(11): p. 3301-3305. 184. Kim, J. and J.W. Grate, Single-enzyme nanoparticles armored by a nanometer- scale organic/inorganic network. Nano Letters, 2003. 3(9): p. 1219-1222. 185. Lee, J., et al., Simple synthesis of hierarchically ordered mesocellular mesoporous silica materials hosting crosslinked enzyme aggregates. Small, 2005. 1(7): p. 744-753. 186. Jia, H.F., et al., Enzyme-carrying polymeric nanofibers prepared via electrospinning for use as unique biocatalysts. Biotechnology Progress, 2002. 18(5): p. 1027-1032.
191
187. Jia, H.F., G.Y. Zhu, and P. Wang, Catalytic behaviors of enzymes attached to nanoparticles: The effect of particle mobility. Biotechnology and Bioengineering, 2003. 84(4): p. 406-414. 188. Caruso, F. and C. Schuler, Enzyme multilayers on colloid particles: Assembly, stability, and enzymatic activity. Langmuir, 2000. 16(24): p. 9595-9603. 189. Daubresse, C., et al., Enzyme immobilization in reactive nanoparticles produced by inverse microemulsion polymerization. Colloid and Polymer Science, 1996. 274(5): p. 482-489. 190. Schuler, C. and F. Caruso, Preparation of enzyme multilayers on colloids for biocatalysis. Macromolecular Rapid Communications, 2000. 21(11): p. 750-753. 191. J. Kim, J.W.G., Ping Wang. , Nanostructures for Enzyme Stabilization. Chem. Eng. Sci. , 2006. 61(3): p. 1017-1026. 192. Matsunaga, T. and S. Kamiya, Use of Magnetic Particles Isolated from Magnetotactic Bacteria for Enzyme Immobilization. Applied Microbiology and Biotechnology, 1987. 26(4): p. 328-332. 193. Y.L. Khmelnitsky, I.N.N., R. Montcheva, A.I. Yaropolov, A.B. Belova, A.V. Levashov and K. Martinek,, Surface-modified polymeric nanogranules containing entrapped enzymes: a novel biocatalyst for use in organic media. Biotechnol Tech, (1989). 3: p. 275–280. 194. Lee, K.E., B.K. Kim, and S.H. Yuk, Biodegradable polymeric nanospheres formed by temperature-induced phase transition in a mixture of poly(lactide-co-glycolide) and poly(ethylene oxide)-poly(propylene oxide)-poly(ethylene oxide) triblock copolymer. Biomacromolecules, 2002. 3(5): p. 1115-1119. 195. Murthy, N., et al., A novel strategy for encapsulation and release of proteins: Hydrogels and microgels with acid-labile acetal cross-linkers. Journal of the American Chemical Society, 2002. 124(42): p. 12398-12399. 196. Lewin, M., et al., Tat peptide-derivatized magnetic nanoparticles allow in vivo tracking and recovery of progenitor cells. Nat Biotechnol, 2000. 18(4): p. 410-414. 197. Aymonier, C., et al., Hybrids of silver nanoparticles with amphiphilic hyperbranched macromolecules exhibiting antimicrobial properties. Chemical Communications, 2002(24): p. 3018-3019.
192
198. Tkachenko, A.G., et al., Multifunctional gold nanoparticle-peptide complexes for nuclear targeting. Journal of the American Chemical Society, 2003. 125(16): p. 4700- 4701. 199. Antibody-Label Conjugates in Lateral-Flow Assays in Forensic Science and Medicine Drugs of Abuse 2005, Humana Press p. 87-98. 200. McGuiggan, J.N.I.a.P.M., Forces Between Surfaces in Liquids. 201. Newman, D.J., Henneberry, H., and Price, C.P, Ann. Clin. Biochem, 1992. 29: p. 22. 202. Malcata, F.X., Reyes, H.R., Garcı´a, H.S., Hill, C.G., Jr., and Amundson, C.H, J. Am. Oil. Chem. Soc, 1990. 67 p. 870. 203. Norde, W., et al., Protein Adsorption at Solid Liquid Interfaces - Reversibility and Conformation Aspects. Journal of Colloid and Interface Science, 1986. 112(2): p. 447- 456. 204. Norde, W., Adsorption of Proteins from Solution at the Solid-Liquid Interface. Advances in Colloid and Interface Science, 1986. 25(4): p. 267-340. 205. Limet, J.N., et al., Particle Counting Immunoassay .4. Use of F(Ab')2 Fragments and N-Epsilon-Chloroacetyl Lysine N-Carboxy-Anhydride for Their Coupling to Polystyrene Latex-Particles. Journal of Immunological Methods, 1979. 28(1-2): p. 25- 32. 206. Shirahama, H. and T. Suzawa, Surface Characterization of Soap-Free Carboxylated Polymer Lattices. Polymer Journal, 1984. 16(11): p. 795-803. 207. Rembaum, A., et al., Functional Polymeric Microspheres Based on 2- Hydroxyethyl Methacrylate for Immunochemical Studies. Macromolecules, 1976. 9(2): p. 328-336. 208. Lukowski, G., et al., Acrylic-Acid Copolymer Nanoparticles for Drug Delivery .1. Characterization of the Surface-Properties Relevant for Invivo Organ Distribution. International Journal of Pharmaceutics, 1992. 84(1): p. 23-31. 209. Sakota, K. and T. Okaya, Preparation of Cationic Polystyrene Latexes in Absence of Emulsifiers. Journal of Applied Polymer Science, 1976. 20(7): p. 1725-1733. 210. Ceska, G.W., Effect of Carboxylic Monomers on Surfactant-Free Emulsion Copolymerization. Journal of Applied Polymer Science, 1974. 18(2): p. 427-437.
193
211. Ceska, G.W., Carboxyl-Stabilized Emulsion Polymers. Journal of Applied Polymer Science, 1974. 18(8): p. 2493-2499. 212. Yamazaki, A., et al., Modification of liposomes with N-substituted polyacrylamides: identification of proteins adsorbed from plasma. Biochimica Et Biophysica Acta-Biomembranes, 1999. 1421(1): p. 103-115. 213. Prime, K.L. and G.M. Whitesides, Adsorption of Proteins onto Surfaces Containing End-Attached Oligo(Ethylene Oxide) - a Model System Using Self- Assembled Monolayers. Journal of the American Chemical Society, 1993. 115(23): p. 10714-10721. 214. Gounaris, A.D. and G.E. Perlmann, Succinylation of Pepsinogen. Journal of Biological Chemistry, 1967. 242(11): p. 2739-&. 215. Kurzer, F. and Douraghi.K, Advances in Chemistry of Carbodiimides. Chemical Reviews, 1967. 67(2): p. 107-&. 216. Okubo, M., Yamamoto, Y., Uno, M., Kalnei, S., and Matsumoto, T, Colloid Polym Sci, 1987. 265: p. 1061. 217. Peula, J.M. and F.J. Delasnieves, Adsorption of Monomeric Bovine Serum-Albumin on Sulfonated Polystyrene Model Colloids .3. Colloidal Stability of Latex-Protein Complexes. Colloids and Surfaces a-Physicochemical and Engineering Aspects, 1994. 90(1): p. 55-62. 218. Peula, J.M., et al., Electrokinetic Characterization and Colloidal Stability of Polystyrene Latex-Particles Partially Covered by Igg/a-Crp and M-Bsa Proteins. Colloids and Surfaces a-Physicochemical and Engineering Aspects, 1994. 92(1-2): p. 127-136. 219. Puig, J., Ferna´ndez-Barbero, A., Bastos-Gonza´lez, D., Serra-Domenech, J., and and R. Hidalgo-A´ lvarez, in In Surface Properties of Biomaterials, R.a.B. West, G.,, Editor 1994, Butterworth-Heinemann Oxford. 220. Poot, A., et al., Detection of Surface-Adsorbed (Lipo)Proteins by Means of a 2- Step Enzyme-Immunoassay - a Study on the Vroman Effect. J Biomed Mater Res, 1990. 24(8): p. 1021-1036. 221. Shim, J., et al., Transdermal delivery of mixnoxidil with block copolymer nanoparticles. Journal of Controlled Release, 2004. 97(3): p. 477-484.
194
222. Ronde, T.d.V.a.H.A.G.d., Preparation and structure of a water-in-oil cream containing lipid nanoparticles. J. Pharm. Sci., 1995. 84 p. 466–472. 223. Müller, S.A.W.a.R.H., A novel sunscreen system based on tocopherol acetate incorporated into solid lipid nanoparticles (SLN). Int. J. Cosmet. Sci, 2001. 23: p. 233– 243. 224. Jenning, V., M. Schafer-Korting, and S. Gohla, Vitamin A-loaded solid lipid nanoparticles for topical use drug release properties. Journal of Controlled Release, 2000. 66(2-3): p. 115-126. 225. S. Salmaso, N.E., A. Bertucco, A. Lante, P. Caliceti, , Nisin-loaded poly-l-lactide nano-particles produced by CO2 anti-solvent precipitation for sustained antimicrobial activity. Int. J. Pharm, 2004. 287: p. 163-173. 226. Pinto-Alphandary, H., A. Andremont, and P. Couvreur, Targeted delivery of antibiotics using liposomes and nanoparticles: research and applications. International Journal of Antimicrobial Agents, 2000. 13(3): p. 155-168. 227. Beyth, N., et al., Surface antimicrobial activity and biocompatibility of incorporated polyethylenimine nanoparticles. Biomaterials, 2008. 29(31): p. 4157- 4163. 228. Pengju G. Luo, T.-R.T., Liangwei Qu, Yi Lin, E. Caldwell, R.A. Latour, F. Stutzenberger, Ya-Ping Sun, Quantitative Analysis of Bacterial Aggregation Mediated by Bioactive Nanoparticles. J. Biomed. Nanotech, 2005. 1: p. 291-296. 229. Liangwei Qu, P.G.L., S.Taylor, Yi Lin, Weijie Huang, Tzeng-Rong J. Tzeng, F. Stutzenberger, R.A. Latour, Ya-Ping Sun, , Visualizing Adhesion-Induced Agglutination of Escherichia coli with Mannosylated Nanoparticles. J. Nanosci. Nanotech, 2005. 5(2): p. 319-322. 230. Hua Yang, L.Q., A. Wimbrow, Xiuping Jiang, Ya-Ping Sun,, Enhancing antimicrobial activity of lysozyme against Listeria monocytogenes using immunonanoparticles. J. Food Protection, 2007 70: p. 1844-1849. 231. Francolini, G.D.P.S., Polymer designs to control biofilm. Reviews in Environmental Science and Bio/Technology . 2003(2): p. 307–319. 232. Danese, P.N., Antibiofilm approaches: Prevention of catheter colonization. Chemistry & Biology, 2002. 9(8): p. 873-880.
195
233. Elliott, T.S.J., Role of antimicrobial central venous catheters for the prevention of associated infections. Journal of Antimicrobial Chemotherapy, 1999. 43(4): p. 441-446. 234. Gnass, S.A., et al., Prevention of central venous catheter-related bloodstream infections using non technologic strategies. INFECTION CONTROL AND HOSPITAL EPIDEMIOLOGY, 2004. 25(8): p. 675-677. 235. Rupp, M.E. and R. Craig, Prevention of central venous catheter-related bloodstream infections. Infections in Medicine, 2004. 21(3): p. 123-127. 236. Samuel, U. and J.P. Guggenbichler, Prevention of catheter-related infections: the potential of a new nano-silver impregnated catheter. International Journal of Antimicrobial Agents, 2004. 23: p. S75-S78. 237. Willcox, M.D.P., et al., A novel cationic-peptide coating for the prevention of microbial colonization on contact lenses. Journal of Applied Microbiology, 2008. 105(6): p. 1817-1825. 238. Williams, T.J., M.D.P. Willcox, and R.P. Schneider, Interactions of bacteria with contact lenses: The effect of soluble protein and carbohydrate on bacterial adhesion to contact lenses. Optometry and Vision Science, 1998. 75(4): p. 266-271. 239. Giacometti, A., et al., RNA III inhibiting peptide inhibits in vivo biofilm formation by drug-resistant Staphylococcus aureus. Antimicrobial Agents and Chemotherapy, 2003. 47(6): p. 1979-1983. 240. Shah, A., J. Mond, and S. Walsh, Lysostaphin-coated catheters eradicate Staphylococccus aureus challenge and block surface colonization. Antimicrobial agents and chemotherapy, 2004. 48(7): p. 2704-7. 241. Kaplan, J.B., et al., Enzymatic detachment of Staphylococcus epidermidis biofilms. Antimicrobial agents and chemotherapy, 2004. 48(7): p. 2633-6. 242. Chugh, T.D. and Z.U. Khan, Intravascular device-related infections: antimicrobial catheters as a strategy for prevention. Journal of Hospital Infection, 2001. 49(1): p. 1- 3. 243. Mermel, L.A., Prevention of intravascular catheter-related infections. Annals of internal medicine, 2000. 132(5): p. 391-402.
196
244. Rello, J., et al., Evaluation of outcome of intravenous catheter-related infections in critically ill patients. American journal of respiratory and critical care medicine, 2000. 162(3 Pt 1): p. 1027-30. 245. LA., M., Correction: catheter related bloodstream-infections. . Ann Intern Med 2000: p. 133:395. 246. Ceri, H., et al., The Calgary Biofilm Device: new technology for rapid determination of antibiotic susceptibilities of bacterial biofilms. Journal of clinical microbiology, 1999. 37(6): p. 1771-6. 247. C, C., MRSA and MRSE: is there an answer. Clin Microbiol Infect 2000, 2000. 6(2): p. 17-22. 248. Bratzler, D.W., P.M. Houck, and S.I.P. Guidelines, Antimicrobial prophylaxis for surgery: An advisory statement from the National Surgical Infection Prevention Project. Clinical Infectious Diseases, 2004. 38(12): p. 1706-1715. 249. Vertegel, A.A., V. Reukov, and V. Maximov, Enzyme-nanoparticle conjugates for biomedical applications. Methods in molecular biology, 2011. 679: p. 165-82. 250. Satishkumar, R. and A. Vertegel, Charge-directed targeting of antimicrobial protein-nanoparticle conjugates. Biotechnology and bioengineering, 2008. 100(3): p. 403-12. 251. Bera, A., et al., Influence of wall teichoic acid on lysozyme resistance in Staphylococcus aureus. Journal of Bacteriology, 2007. 189(1): p. 280-283. 252. Wu, J.A., et al., Lysostaphin disrupts Staphylococcus aureus and Staphylococcus epidermidis biofilms on artificial surfaces. Antimicrobial Agents and Chemotherapy, 2003. 47(11): p. 3407-3414. 253. Wright, G.D., Bacterial resistance to antibiotics: Enzymatic degradation and modification. Advanced Drug Delivery Reviews, 2005. 57(10): p. 1451-1470. 254. Wenk, C., Recent advances in animal feed additives such as metabolic modifiers, antimicrobial agents, probiotics, enzymes and highly available minerals - Review. Asian-Australasian Journal of Animal Sciences, 2000. 13(1): p. 86-95. 255. Suzawa, T., H. Shirahama, and T. Fujimoto, Adsorption of Bovine Serum-Albumin onto Homo-Polymer and Co-Polymer Lattices. Journal of Colloid and Interface Science, 1982. 86(1): p. 144-150.
197
256. Suzawa, T., H. Shirahama, and T. Fujimoto, Effect of Urea on the Adsorption of Bovine Serum-Albumin onto Polymer Lattices. Journal of Colloid and Interface Science, 1983. 93(2): p. 498-503. 257. Bale, M.D., et al., Influence of Copolymer Composition on Protein Adsorption and Structural Rearrangements at the Polymer Surface. Journal of Colloid and Interface Science, 1989. 132(1): p. 176-187. 258. Sakota, K. and T. Okaya, Effect of Hydrophilic Nature of Growing Radicals on Formation of Particles in Preparation of Soap-Free Carboxylated Polystyrene Latexes. Journal of Applied Polymer Science, 1976. 20(12): p. 3265-3274. 259. Masschalck, B. and C.W. Michiels, Antimicrobial properties of lysozyme in relation to foodborne vegetative bacteria. Critical Reviews in Microbiology, 2003. 29(3): p. 191-214. 260. Chinn, M.K., et al., The effect of polyethylene glycol on the mechanics and ATPase activity of active muscle fibers. Biophysical Journal, 2000. 78(2): p. 927-939. 261. Kinstler, O.B., et al., Characterization and stability of N-terminally PEGylated rhG-CSF. Pharmaceutical Research, 1996. 13(7): p. 996-1002. 262. Klaeger, A.J., et al., Clinical application of a homogeneous colorimetric assay for tear lysozyme. Ocular Immunology and Inflammation, 1999. 7(1): p. 7-15. 263. Vertegel, A.A., R.W. Siegel, and J.S. Dordick, Silica nanoparticle size influences the structure and enzymatic activity of adsorbed lysozyme. Langmuir, 2004. 20(16): p. 6800-6807. 264. Gottschalk, M. and B. Halle, Self-association of lysozyme as seen by magnetic relaxation dispersion. Journal of Physical Chemistry B, 2003. 107(31): p. 7914-7922. 265. Coonrod, J.D., R. Varble, and K. Yoneda, Mechanism of Killing of Pneumococci by Lysozyme. Journal of Infectious Diseases, 1991. 164(3): p. 527-532. 266. Nanjo, F., K. Sakai, and T. Usui, Para-Nitrophenyl Penta-N-Acetyl-Beta- Chitopentaoside as a Novel Synthetic Substrate for the Colorimetric Assay of Lysozyme. Journal of Biochemistry, 1988. 104(2): p. 255-258. 267. Cutinell.C and F. Galdiero, Ion-Binding Properties of Cell Wall of Staphylococcus Aureus. Journal of Bacteriology, 1967. 93(6): p. 2022-&.
198
268. Zschornig, O., et al., Modulation of lysozyme charge influences interaction with phospholipid vesicles. Colloids and Surfaces B-Biointerfaces, 2005. 42(1): p. 69-78. 269. Laible, N.J. and G.R. Germaine, Bactericidal Activity of Human Lysozyme, Muramidase-Inactive Lysozyme, and Cationic Polypeptides against Streptococcus- Sanguis and Streptococcus-Faecalis - Inhibition by Chitin Oligosaccharides. Infection and Immunity, 1985. 48(3): p. 720-728. 270. Price, J.A.R. and R. Pethig, Surface-Charge Measurements on Micrococcus- Lysodeikticus and the Catalytic Implications for Lysozyme. Biochimica Et Biophysica Acta, 1986. 889(2): p. 128-135. 271. Laible, N.J. and G.R. Germaine, Adsorption of Lysozyme from Human Whole Saliva by Streptococcus-Sanguis-903 and Other Oral Microorganisms. Infection and Immunity, 1982. 36(1): p. 148-159. 272. Haynes, C.A., E. Sliwinsky, and W. Norde, Structural and Electrostatic Properties of Globular-Proteins at a Polystyrene Water Interface. Journal of Colloid and Interface Science, 1994. 164(2): p. 394-409. 273. vanderWal, A., et al., Electrokinetic potential of bacterial cells. Langmuir, 1997. 13(2): p. 165-171. 274. Poortinga, A.T., R. Bos, and H.J. Busscher, Electrostatic interactions in the adhesion of an ion-penetrable and ion-impenetrable bacterial strain to glass. Colloids and Surfaces B-Biointerfaces, 2001. 20(2): p. 105-117. 275. Davies, R.C. and Neuberge.A, Modification of Lysine and Arginine Residues of Lysozyme and Effect on Enzymatic Activity. Biochimica Et Biophysica Acta, 1969. 178(2): p. 306-&. 276. Bower, C.K., et al., Activity losses among T4 lysozyme charge variants after adsorption to colloidal silica. Biotechnology and Bioengineering, 1999. 64(3): p. 373- 376. 277. Kondo, A., et al., Kinetic and Circular-Dichroism Studies of Enzymes Adsorbed on Ultrafine Silica Particles. Applied Microbiology and Biotechnology, 1993. 39(6): p. 726-731. 278. Norde, W. and A.C.I. Anusiem, Adsorption, Desorption and Readsorption of Proteins on Solid-Surfaces. Colloids and Surfaces, 1992. 66(1): p. 73-80.
199
279. Czeslik, C. and R. Winter, Effect of temperature on the conformation of lysozyme adsorbed to silica particles. Physical Chemistry Chemical Physics, 2001. 3(2): p. 235- 239. 280. Connolly, S., S. Cobbe, and D. Fitzmaurice, Effects of ligand-receptor geometry and stoichiometry on protein-induced aggregation of biotin-modified colloidal gold. Journal of Physical Chemistry B, 2001. 105(11): p. 2222-2226. 281. Costanzo, P.J., T.E. Patten, and T.A.P. Seery, Protein-ligand mediated aggregation of nanoparticles: A study of synthesis and assembly mechanism. Chemistry of Materials, 2004. 16(9): p. 1775-1785. 282. Sergeev, B.M., et al., Synthesis of protein A conjugates with silver nanoparticles. Colloid Journal, 2003. 65(5): p. 636-638. 283. Pincus, M.R., S.S. Zimmerman, and H.A. Scheraga, Structures of Enzyme- Substrate Complexes of Lysozyme. Proceedings of the National Academy of Sciences of the United States of America, 1977. 74(7): p. 2629-2633. 284. Zaiou, M., Multifunctional antimicrobial peptides: therapeutic targets in several human diseases. Journal of Molecular Medicine-Jmm, 2007. 85(4): p. 317-329. 285. Marr, A.K., W.J. Gooderham, and R.E. Hancock, Antibacterial peptides for therapeutic use: obstacles and realistic outlook. Curr Opin Pharmacol, 2006. 6(5): p. 468-72. 286. Jenssen, H., P. Hamill, and R.E. Hancock, Peptide antimicrobial agents. Clinical microbiology reviews, 2006. 19(3): p. 491-511. 287. Schindle.Ca and Schuhard.Vt, Purification and Properties of Lysostaphin - Alytic Agent for Staphylococcus Aureus. Biochimica Et Biophysica Acta, 1965. 97(2): p. 242- &. 288. Zhang, L. and T.J. Falla, Antimicrobial peptides: therapeutic potential. Expert opinion on pharmacotherapy, 2006. 7(6): p. 653-63. 289. Maximov, V.R.a.A.V., Targeted delivery of therapeutic enzymes. J. Drug Delivery Sci. Tech., 2009. 19(5): p. 311-320. 290. Torchilin, V.P., Targeted pharmaceutical nanocarriers for cancer therapy and imaging. The AAPS journal, 2007. 9(2): p. E128-47.
200
291. Elbayoumi, T.A. and V.P. Torchilin, Liposomes for targeted delivery of antithrombotic drugs. Expert opinion on drug delivery, 2008. 5(11): p. 1185-98. 292. Norman, R.S., et al., Targeted photothermal lysis of the pathogenic bacteria, Pseudomonas aeruginosa, with gold nanorods. Nano letters, 2008. 8(1): p. 302-6. 293. Wang, S., et al., Rapid colorimetric identification and targeted photothermal lysis of Salmonella bacteria by using bioconjugated oval-shaped gold nanoparticles. Chemistry, 2010. 16(19): p. 5600-6. 294. Abdalla, M.O., et al., Enhanced noscapine delivery using uPAR-targeted optical- MR imaging trackable nanoparticles for prostate cancer therapy. Journal of controlled release : official journal of the Controlled Release Society, 2011. 149(3): p. 314-22. 295. Hua Yang, L.Q., A. Wimbrow, Xiuping Jiang and Ya-Ping Sun, Enhancing antimicrobial activity of lysozyme against Listeria monocytogenes using immunonanoparticles. . J. Food Protection 2007. 70: p. p. 1844-1849. 296. Araki, K., et al., Molecular disruption of NBS1 with targeted gene delivery enhances chemosensitisation in head and neck cancer. British journal of cancer, 2010. 103(12): p. 1822-30. 297. Garlotta, D., A literature review of Poly(Lactic acid) J. Polymers And The Environment 2001. 9(2): p. 63-84. 298. Berlier, J.E., et al., Quantitative comparison of long-wavelength Alexa Fluor dyes to Cy dyes: Fluorescence of the dyes and their bioconjugates. Journal of Histochemistry & Cytochemistry, 2003. 51(12): p. 1699-1712. 299. Schindler, C.A., and V. T. Schuhardt, Lysostaphin: a new bacteriolytic agent for the staphylococci. Proc. Natl. Acad. Sci. USA, 1964. 51: p. 414-421. 300. Lamalle-Bernard, D., et al., Coadsorption of HIV-1 p24 and gp120 proteins to surfactant-free anionic PLA nanoparticles preserves antigenicity and immunogenicity. Journal of controlled release : official journal of the Controlled Release Society, 2006. 115(1): p. 57-67. 301. Lu, J.Z., et al., Cell wall-targeting domain of glycylglycine endopeptidase distinguishes among peptidoglycan cross-bridges. The Journal of biological chemistry, 2006. 281(1): p. 549-58.
201
302. Hook, F., Rodahl M, Brzezinski P, Kasemo B, Energy Dissipation Kinetics for Protein and Antibody−Antigen Adsorption under Shear Oscillation on a Quartz Crystal Microbalance. Langmuir : the ACS journal of surfaces and colloids, 1998. 14: p. 729- 734. 303. Armstrong, J., Wenby RB, Meiselman HJ and Fisher, TC, The Hydrodynamic Radii of Macromolecules and Their Effect on Red Blood Cell Aggregation. Biophys J. , 2004. 87(6): p. 4259–4270. 304. Li G, S.R., Conlan B, Gilbert A, Roerth P, Nair H. , Purifification of human immunoglobulin G: a new approach to plasma fractionation. . Vox Sanguinis 2002(83): p. 332-338. 305. Grundling, A. and O. Schneewind, Cross-linked peptidoglycan mediates lysostaphin binding to the cell wall envelope of Staphylococcus aureus. J Bacteriol, 2006. 188(7): p. 2463-72. 306. Oren, Z. and Y. Shai, Mode of action of linear amphipathic alpha-helical antimicrobial peptides. Biopolymers, 1998. 47(6): p. 451-63. 307. Mudge, M. and L.E. Hughes, Incisional Hernia - a 10 Year Prospective-Study of Incidence and Attitudes. British Journal of Surgery, 1985. 72(1): p. 70-71. 308. Burger, J.W.A., et al., Long-term follow-up of a randomized controlled trial of suture versus mesh repair of incisional hernia. Annals of Surgery, 2004. 240(4): p. 578-583. 309. Luijendijk, R.W., et al., A comparison of suture repair with mesh repair for incisional hernia. New England Journal of Medicine, 2000. 343(6): p. 392-398. 310. Hesselink, V.J., et al., An Evaluation of Risk-Factors in Incisional Hernia Recurrence. Surgery Gynecology & Obstetrics, 1993. 176(3): p. 228-234. 311. Heniford, B.T., et al., Laparoscopic repair of ventral hernias nine years' experience with 850 consecutive hernias. Annals of Surgery, 2003. 238(3): p. 391-399. 312. Heniford, B.T., et al., Laparoscopic ventral and incisional hernia repair in 407 patients. Journal of the American College of Surgeons, 2000. 190(6): p. 645-650. 313. Cobb, W.S., et al., Incisional herniorrhaphy with intraperitoneal composite mesh: A report of 95 cases. American Surgeon, 2003. 69(9): p. 784-787.
202
314. Falagas, M.E. and S.K. Kasiakou, Mesh-related infections after hernia repair surgery. Clinical Microbiology and Infection, 2005. 11(1): p. 3-8. 315. Cainzos, M.A., Antibiotic prophylaxis. New Horizons-the Science and Practice of Acute Medicine, 1998. 6(2): p. S11-S19. 316. Sugarman, B., Infections and Prosthetic Devices. American Journal of Medicine, 1986. 81(1A): p. 78-84. 317. Young, E.J. and B. Sugarman, Infections in Prosthetic Devices. Surgical Clinics of North America, 1988. 68(1): p. 167-180. 318. Gristina, A.G., et al., Adhesive Colonization of Biomaterials and Antibiotic- Resistance. Biomaterials, 1987. 8(6): p. 423-426. 319. I. Francolini, G.D.a.P.S., Polymer designs to control biofilm growth on medical devices. Reviews in Environmental Science and Bio/Technology 2003. 2: p. 307–319. 320. Kusuma, C.M. and J.F. Kokai-Kun, Comparison of four methods for determining lysostaphin susceptibility of various strains of Staphylococcus aureus. Antimicrobial Agents and Chemotherapy, 2005. 49(8): p. 3256-3263. 321. Appleman, M.D. and D.M. Citron, Efficacy of vancomycin and daptomycin against Staphylococcus aureus isolates collected over 29 years. Diagnostic Microbiology and Infectious Disease, 2010. 66(4): p. 441-444. 322. Westermeier, R. and R. Marouga, Protein detection methods in proteomics research. Bioscience Reports, 2005. 25(1-2): p. 19-32. 323. Hlady, V., Buijs, J and Jennissen, HP, Methods for Studying Protein Adsorption. Methods Enzymol, 1999(309): p. 402–429. 324. Klosterhalfen, B., K. Junge, and U. Klinge, The lightweight and large porous mesh concept for hernia repair. Expert Review of Medical Devices, 2005. 2(1): p. 103-117. 325. Iversen, O.J. and A. Grov, Studies on Lysostaphin - Separation and Characterization of 3 Enzymes. European Journal of Biochemistry, 1973. 38(2): p. 293- 300. 326. An, Y.H. and R.J. Friedman, Concise review of mechanisms of bacterial adhesion to biomaterial surfaces. Journal of Biomedical Materials Research, 1998. 43(3): p. 338- 348.
203
327. Harrell, A., Novitsky, YW, Kercher, KW, Foster, M, Burns, JM, Kuwada, TS and Heniford, BT, In vitro infectability of prosthetic mesh by methicillin-resistant Staphylococcus aureus. Hernia 2006. 10: p. 120–124. 328. Heniford, B.T., Laparoscopic ventral and incisional hernia repair in 407 patients - Reply. Journal of the American College of Surgeons, 2000. 190(6): p. 650-650. 329. Schindler CA, S., VT, Lysostaphin: a new bacteriolytic agent for the staphylococci. Proc Natl Acad Sci USA 1964. 51: p. 414-421. 330. Hetrick, E.M. and M.H. Schoenfisch, Reducing implant-related infections: active release strategies. Chemical Society Reviews, 2006. 35(9): p. 780-789. 331. Klosterhalfen, B., K. Junge, and U. Klinge, The lightweight and large porous mesh concept for hernia repair. Expert Review of Medical Devices, 2005. 2(1): p. 103-17. 332. Lai, K.K. and S.A. Fontecchio, Use of silver-hydrogel urinary catheters on the incidence of catheter-associated urinary tract infections in hospitalized patients. Am J Infect Control, 2002. 30(4): p. 221-5. 333. Lee, M. and J.F. Donovan, Laparoscopic omentectomy for salvage of peritoneal dialysis catheters. Journal of endourology / Endourological Society, 2002. 16(4): p. 241-4. 334. Crnich, C.J. and D.G. Maki, The promise of novel technology for the prevention of intravascular device-related bloodstream infection. II. Long-term devices. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America, 2002. 34(10): p. 1362-8. 335. C Crnich, N.S.a.D.M., Infections of implanted medical devices. Antibiotics and Chemotherapy In: R, ed. R.G. D Finch, Norby and R Whitley. Vol. . 2001, London: Harcourt Publishers. 336. von Recum, A.F., Applications and failure modes of percutaneous devices: a review. Journal of biomedical materials research, 1984. 18(4): p. 323-36. 337. Kwok, C.S., T.A. Horbett, and B.D. Ratner, Design of infection-resistant antibiotic-releasing polymers. II. Controlled release of antibiotics through a plasma- deposited thin film barrier. Journal of controlled release : official journal of the Controlled Release Society, 1999. 62(3): p. 301-11.
204
338. Kwok, C.S., et al., Design of infection-resistant antibiotic-releasing polymers: I. Fabrication and formulation. Journal of controlled release : official journal of the Controlled Release Society, 1999. 62(3): p. 289-99. 339. Jabra-Rizk, M.A., et al., Effect of farnesol on Staphylococcus aureus biofilm formation and antimicrobial susceptibility. Antimicrobial agents and chemotherapy, 2006. 50(4): p. 1463-9. 340. Balaban, N., et al., A chimeric peptide composed of a dermaseptin derivative and an RNA III-inhibiting peptide prevents graft-associated infections by antibiotic- resistant staphylococci. Antimicrobial agents and chemotherapy, 2004. 48(7): p. 2544- 50. 341. Kokai-Kun, J.F., T. Chanturiya, and J.J. Mond, Lysostaphin eradicates established Staphylococcus aureus biofilms in jugular vein catheterized mice. Journal of Antimicrobial Chemotherapy, 2009. 64(1): p. 94-100 342. SN, L.J.a.T., The stabilisation of proteins bysugars. . J Biol Chem 256(14): p. 7193–7201. 343. Crisante, F., et al., Antibiotic delivery polyurethanes containing albumin and polyallylamine nanoparticles. European journal of pharmaceutical sciences : official journal of the European Federation for Pharmaceutical Sciences, 2009. 36(4-5): p. 555- 64. 344. LL, L.J.a.L., Preferential solvent interaction between proteins and polyethylene glycols. . J Biol Chem 1981. 256(2): p. 625–631. 345. Muraoka, T., et al., Mimicking multipass transmembrane proteins: synthesis, assembly nd folding of alternating amphiphilic multiblock molecules in liposomal membranes. Chemical Communications, 2011. 47(1): p. 194-6. 346. Sahin, E. and K.L. Kiick, Macromolecule-induced assembly of coiled-coils in alternating multiblock polymers. Biomacromolecules, 2009. 10(10): p. 2740-9. 347. Feng, X., et al., Controlled self-assembly of C3-symmetric hexa-peri- hexabenzocoronenes with alternating hydrophilic and hydrophobic substituents in solution, in the bulk, and on a surface. J Am Chem Soc, 2009. 131(12): p. 4439-48.
205
348. Zhou, R., P. Wang, and H.C. Chang, Bacteria capture, concentration and detection by alternating current dielectrophoresis and self-assembly of dispersed single-wall carbon nanotubes. Electrophoresis, 2006. 27(7): p. 1376-85. 349. Crespo-Biel, O., et al., Supramolecular layer-by-layer assembly: alternating adsorptions of guest- and host-functionalized molecules and particles using multivalent supramolecular interactions. J Am Chem Soc, 2005. 127(20): p. 7594-600. 350. Ramasubbu, N., et al., Structural analysis of dispersin B, a biofilm-releasing glycoside hydrolase from the periodontopathogen Actinobacillus actinomycetemcomitans. J Mol Biol, 2005. 349(3): p. 475-86. 351. Izano, E.A., et al., Detachment and killing of Aggregatibacter actinomycetemcomitans biofilms by dispersin B and SDS. Journal of dental research, 2007. 86(7): p. 618-22. 352. Schmalz, G., Concepts in biocompatibility testing of dental restorative materials. Clinical oral investigations, 1997. 1(4): p. 154-62. 353. Lomas, R.J., et al., An evaluation of the capacity of differently prepared demineralised bone matrices (DBM) and toxic residuals of ethylene oxide (EtOx) to provoke an inflammatory response in vitro. Biomaterials, 2001. 22(9): p. 913-21. 354. Ross, R., The fibroblast and wound repair. Biological reviews of the Cambridge Philosophical Society, 1968. 43(1): p. 51-96. 355. Guidelines for submitting documentations for the stability of human drugs and biologics, FDA., Editor 1987: Rockville (MD). 356. Y, T., Shelf life determination based on equivalence assessment. . J Biopharm Stat 2003. 13: p. 431-449. 357. R., M., Estimating degradation in real time and accelerated stability tests with random lot-to-lot variation: a simulation study. J Pharm Sci 2002. 91: p. 893-899. 358. Hemmerich, K.J., General Aging Theory and Simplified Protocol for Accelerated Aging of Medical Devices Medical Plastics and Biomaterials 1998. 5(4): p. 16–23. 359. Heller M, C.J.a.R.T., Protein formulation and lyophilisation cycle design: prevention of damage due to freeze-concentration induced phase separation. . Biotech Bioeng 1999. 63((2)): p. 166–174
206
360. Jammel F, B.R., Mauri F and Kalonia D, Investigation of physicochemical changes to L-asparaginase during freeze-thaw cycling. . J Pharm Pharmacol 1997. 49: p. 472– 477. 361. SN, A.T.a.T., Stabilisation of protein structure by sugars. . Biochemistry . 1982. 21: p.:6536–6544 362. SN, A.T.a.T., Preferential interactions of proteins with salts in concentrated solutions. Biochemistry 21: p. 6545–6552. 363. John F. Carpenter, S.J.P., Thomas J. Anchordoguy, and Tsutomu Arakawa, Interactions of Stabilizers with Proteins During Freezing and Drying, in Formulation and Delivery of Proteins and Peptides, J.L.C.a.R. Langer, Editor 1994, American Chemical Society. p. pp 134–147.
207