Chlorine Dioxide for the Prevention of Biomaterial-Associated Infections

Item Type text; Electronic Dissertation

Authors Powis, Samantha

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 07/10/2021 21:03:55

Link to Item http://hdl.handle.net/10150/194369 1

CHLORINE DIOXIDE FOR THE PREVENTION OF

BIOMATERIAL-ASSOCIATED INFECTIONS

Samantha Powis

A Dissertation Submitted to the Faculty of the

GRADUATE INTERDISCIPLINARY PROGRAM IN BIOMEDICAL ENGINEERING

In Partial Fulfillment of the Requirements

For the Degree of

DOCTOR OF PHILOSOPHY

In the Graduate College

THE UNIVERSITY OF ARIZONA

2005 2

THE UNIVERSITY OF ARIZONA GRADUATE COLLEGE

As members of the Dissertation Committee, we certify that we have read the dissertation

Samantha Powis

Chlorine Dioxide for the Prevention of Biomaterial-Associated Infections and recommend that it be accepted as fulfilling the dissertation requirement for the

Degree of Doctor of Philosophy

______Date: 9-21-05 Stuart K. Williams

______Date: 9-21-05 Edward Mueller

______Date: 9-21-05 James B. Hoying

______Date: 9-21-05 Mark Riley

______Date: 9-21-05 Srini Raghavan

Final approval and acceptance of this dissertation is contingent upon the candidate’s submission of the final copies of the dissertation to the Graduate College.

I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement.

______Date: 9-21-05 Dissertation Director: Stuart K. Williams 3

STATEMENT BY AUTHOR

This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgement of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his or her judgement the proposed use of the materials is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.

SIGNED: Samantha Powis 4

ACKNOWLEDGEMENTS

While only one name appears on this document, as with any endeavor of this magnitude there is always a team of people supporting the project.

I would like to thank my committee members, Dr. Stuart Williams, Mr. Edward Mueller, Dr. James Hoying, Dr. Mark Riley, and Dr. Srini Raghavan, for their insight and enthusiasm towards my research. Your suggestions have led me down some interesting and rather unexpected pathways – thank you!

To all of the people who have assisted in experiments and helped to train me in various scientific methods: Alice Stone, Faith Rice, Leigh Kleinert, Ned Kirkpatrick, and Vangie Patula – thank you for lending your expertise to my experiments, and for your continual patience and understanding.

To the Biomedical Engineering Interdisciplinary Program: Charlotte Todd, Lori Taylor, and Debbi Howard – thank you for your assistance in all administrative matters.

My family: Alison Powis, Keith Powis, Tracey Powis, and June Calvert - you may be on the other side of the world, but when I needed you, you were always here. I am truly blessed to have you in my life.

My friends: Brooke McGuire, Linda Scheu, Kristen O’Halloran, Susan Knopf, and Pamela Davies – you held me together and believed in me. I wouldn’t have made it without you ladies.

My mentor: Stuart Williams – for pushing me to be the best scientist that I can be! The support, guidance and opportunities that you provided have been invaluable – I cannot thank you enough.

DEDICATION

For Ronald Davidson Calvert

July 30, 1916 – August 16, 2002

For introducing me to science;

For your love and absolute faith;

You have always been here for me.

This is for you. 6

TABLE OF CONTENTS

LIST OF FIGURES ...... 9

LIST OF TABLES...... 13

ABSTRACT...... 15

1. PREFACE...... 17 Definitions...... 20

2. INTRODUCTION: STERILIZATION TECHNOLOGIES FOR BIOMATERIALS 23 Sterilization by Heat ...... 23 Steam Sterilization...... 23 Dry Heat Sterilization ...... 24 Mechanisms of Biocidal Activity ...... 24 Procedure for Sterilization by Heat...... 25 Limitations and Disadvantages of Sterilization by Heat...... 28 Radiation Sterilization ...... 28 Electron Beam Sterilization ...... 28 Gamma Radiation ...... 29 Mechanisms of Biocidal Activity ...... 29 Procedure for Radiation Sterilization...... 29 Limitations and Disadvantages of Radiation Sterilization...... 30 Chemical Agents...... 33 Ethylene Oxide...... 33 Low temperature steam and formaldehyde...... 33 Mechanisms of Biocidal Activity ...... 33 Procedure of Sterilization with Chemical Agents...... 34 Limitations and Disadvantages of Sterilization with Chemical Agents ...... 36 Advances in Sterilization Technologies...... 36 Optimization of Ethylene Oxide Sterilization ...... 41 Hydrogen Peroxide Gas Plasma...... 41 Sterilization with Ozone ...... 45 Chlorine Dioxide...... 50 Chemical and physical properties of chlorine dioxide...... 50 Production of chlorine dioxide...... 54 Antimicrobial properties of chlorine dioxide...... 55 Toxicity of chlorine dioxide and its metabolites...... 56 Applications for chlorine dioxide ...... 60 Limitations and Disadvantages of Chlorine Dioxide...... 62 Research Plan...... 63 7

TABLE OF CONTENTS - continued

3. LOW CONCENTRATION CHLORINE DIOXIDE FOR THE STERILIZATION OF BIOMATERIALS...... 64 Introduction...... 64 Materials and Methods...... 66 Results...... 72 Discussion...... 86

4. AN ASSESSMENT OF THE CYTOTOXICITY OF POLYMERS STERILIZED WITH CHLORINE DIOXIDE ...... 91 Introduction...... 91 Materials & Methods ...... 93 Results...... 100 Discussion...... 112

5. INTRODUCTION: ANTIMICROBIAL BIOMATERIALS FOR THE IN SITU PREVENTION OF INFECTION ...... 118 Process of biomaterial-associated infections ...... 118 Antimicrobial biomaterials ...... 122 Physicochemical modifications ...... 122 Incorporation of antibiotics...... 124 Incorporation of biocides ...... 125 Silver...... 127 Chlorine dioxide generating material...... 128 Humidity activation ...... 128 Light activation ...... 129 Research Plan...... 132

6. ADHESION AND COLONIZATION OF STAPHYLOCOCCUS EPIDERMIDIS ON THE SURFACE OF A CHLORINE DIOXIDE-GENERATING MATERIAL ...... 133 Introduction...... 133 Materials & Methods ...... 135 Results...... 140 Discussion...... 155

7. ANTIMICROBIAL EFFICACY OF A CHLORINE DIOXIDE GENERATING BIOMATERIAL IN VITRO AND IN VIVO...... 161 Introduction...... 161 Materials and Methods...... 163 Results...... 170 Discussion...... 187

TABLE OF CONTENTS - continued

8. INTRODUCTION: THE DEVELOPMENT OF MICROBIAL RESISTANCE TO ANTIBIOTICS AND BIOCIDES ...... 192 Antibiotics...... 192 Biocides...... 193 Acquired microbial resistance...... 193 Mechanisms of antibiotic resistance ...... 194 Mechanisms of biocide resistance ...... 195 A correlation between antibiotic and biocide resistance?...... 201 Resistance to chlorine dioxide ...... 202 Research Plan...... 203

9. ACQUIRED MICROBIAL RESISTANCE TO CHLORINE DIOXIDE...... 204 Introduction...... 204 Materials and Methods...... 208 Results...... 213 Discussion...... 230

10. CONCLUSIONS...... 236

APPENDICES ...... 253 Description and operation of a chlorine dioxide generator...... 253 Measuring chlorine dioxide concentration using iodometric titration...... 263 Standardization of 0.01N sodium thiosulfate solution...... 270

REFERENCES ...... 272

LIST OF FIGURES

Figure 2.1: Diagrammatic representation of a steam sterilization system ………… 26

Figure 2.2: Gamma radiation (60Co) sterilization system …………………...….…. 31

Figure 2.3: Example of an ethylene oxide sterilization cycle …………………….. 37

Figure 2.4: Example of a low temperature steam formaldehyde sterilization cycle ……………………………………………………… 39

Figure 2.5: Prototype microwave sterilizer/desorber ……………………………... 42

Figure 2.6: Example of a sterilization cycle with hydrogen peroxide …………… . 46

Figure 2.7: Different STERRAD™ models ……………………..…………….….. 48

Figure 2.8: Schematic diagram of ozone production by corona discharge .………. 51

Figure 3.1: Chlorine dioxide sterilization system……………………………...….. 67

Figure 3.2: Results from the viability counts assay for experiments with high relative humidity chlorine dioxide ………………………………. 79

Figure 4.1: Images depicting the numerical scale of cytotoxicity ……...……….… 98

Figure 4.2: Cytotoxicity values for the extract dilutions from polyurethane …………………………………………………….……. 101

Figure 4.3: Cytotoxicity values for the extract dilutions for polyvinyl chloride …………………………………………………….. 103

Figure 4.4: Cytotoxicity values for the extract dilutions for high density polyethylene …………………………………………….. 105

Figure 4.5: Cytotoxicity values for the extract dilutions for organo-tin stabilized polyvinyl chloride ……………………………… 107

Figure 4.6: Normalized absorbance measurements from the MTT assay…………. 110

LIST OF FIGURES – continued

Figure 5.1: Molecular sequences in bacterial attachment, adhesion, and colonization to a material surface ………………………………… 120

Figure 5.2: Schematic of the generation of ClO2 from a humidity-activated formulation ……………………………………….. 130

Figure 6.1: Zone of inhibition (ZOI) assay assessing bactericidal activity of ClO2-generating material …………...……………………… 141

Figure 6.2: SDS-PAGE gels indicating the incubation time required to achieve equilibrium in protein adsorption …………………………. 143

Figure 6.3: Congo red agar assay …………………………………………………. 145

Figure 6.4: 111In labeled bacterial adhesion assay …………………………….…… 148

Figure 6.5: Scanning electron microscopy images of bacterial adhesion ………… 150

Figure 6.6: Live/Dead® BacLight™ bacterial viability assay ……………….……. 152

Figure 7.1: System used to quantitatively assess ClO2 production ……………..… 164

Figure 7.2: Production of ClO2 in vitro …………………………………………… 171

Figure 7.3: Zone of inhibition (ZOI) assay assessing bactericidal activity of ClO2-generating material ………………………………….. 173

Figure 7.4: Visual assessment of material samples in situ ………………………... 175

Figure 7.5: Histological assessment of peritoneal wall tissue with hematoxylin and eosin …………………………………..……………. 177

Figure 7.6: Immunohistochemical staining of macrophages with RAM-11 ……… 180

Figure 7.7: Macrophage densities in tissue contacting control and ClO2-generating material samples …………………………………….. 182

Figure 7.8: Scanning electron microscopy images of the surfaces of explanted material samples ………………………………………… 185

LIST OF FIGURES – continued

Figure 8.1: Cell structure for gram-positive bacteria and gram-negative bacteria ……………………………………………….. 196

Figure 8.2: Illustration of the main types of bacterial efflux pumps ……………… 199

Figure 9.1: Zone of inhibition (ZOI) assay assessing the bactericidal activity of a ClO2-generating material against SP030724 …………………………………………… 206

Figure 9.2: Schematic diagram representing the modified ZOI (M-ZOI) assay …………………………………………. 209

Figure 9.3: ZOI formed for: (A) 10-16 with 10-16 supernatant; (B) 10-16 with SP030724 supernatant ……………..………………… 214

Figure 9.4: Degree of molecular hydrophobicity of the molecule(s) of interest in the SP030724 supernatant ……………………………… 216

Figure 9.5: Temperature stability of the molecule(s) in the SP030724 supernatant ……………………………………………...… 218

Figure 9.6: Separation of SP030724 supernatant into solutions with different molecular weights ………………………………………...… 221

Figure 9.7: ZOI assay assessing the bactericidal activity of a ClO2-generating material against colonies isolated from the ZOI when a ClO2-generating material is tested against 10-16 with SP030724 supernatant …………………….. 223

Figure 9.8: Assessing the filtered and non-filtered SP030724 supernatant ……………….…………………………………………… 225

Figure 9.9: Susceptibility of SP030724 to increasing concentrations of ClO2 ………………………………………………... 228

Figure 10.1: Schematic diagram of the control systems required in a chlorine dioxide sterilization system …………………………….. 242

Figure A.1: Schematic of ClO2 generator ………………………………………..... 255

LIST OF FIGURES – continued

Figure A.2: Impinger system for measuring [ClO2] < 100ppm …………….……… 265

LIST OF TABLES

Table 2.1: Oxides frequently referred to as chlorine oxide ……………...……….. 53

Table 2.2: Microorganisms inactivated by chlorine dioxide …………………..…. 57

Table 3.1: Calculated flow rate for the digital flow meter compared to the digital setting ……………………………………………………. 73

Table 3.2: Verification of the accuracy of a digital hygrometer ……………..…… 74

Table 3.3: Chlorine dioxide concentration measured from the generator and chamber outputs, and compared with the chlorine dioxide concentration measured within the chamber via three sample ports ……………………………………….. 75

Table 3.4: Chlorine dioxide concentrations measured from the generator and chamber outputs during experiments …………………… 76

Table 3.5: Indicator broth assay for experiments with high relative humidity chlorine dioxide ………………………….. 78

Table 3.6: Indicator broth assay for experiments with low relative humidity chlorine dioxide ………………………….. 82

Table 3.7: Viability counts assay for experiments with low relative humidity chlorine dioxide …………………………... 83

Table 3.8: The delay in detecting viable spores ………………………………….. 84

Table 3.9: Summary of environmental parameters required for ClO2 sterilization ……………………………………………………… 85

Table 4.1: Environmental conditions for sterilization with ClO2 chosen for the assessment of ClO2 sterilization on a selection of polymers ……………………………………………….…. 94

Table 4.2: Scale used for visual assessment of cytotoxicity …………………..…. 97

Table 7.1: Colony forming unit (CFU) counts of viable bacteria recovered from explanted material samples ……………………….….. 184

LIST OF TABLES – continued

Table A.1: Recommended working range for the generator parameters ……………………………………………..…… 259

ABSTRACT

Biomaterial-associated infections remain a significant complication of medical implants. Of the different strains of bacteria associated with nosocomial infections, 70%

are resistant to at least one of the drugs used for treatment (Bren 2002). In 2000 the

Center for Disease Control ranked microbial agents as the 4th leading actual cause of

death in the United States of America (Mokdad et al. 2004).

In an effort to improve the prevention and treatment of infections, this research

has three objectives: the development of an alternative sterilization method for medical devices; assessing a new antimicrobial material for the prevention of infections in situ;

and assessing mechanisms of acquired microbial resistance. The biocide being

investigated in this body of work is chlorine dioxide gas.

While multiple sterilization methods are available, there are limitations to all of

these technologies. For example, chemical sterilization can leave residues on the surface

of the material. These residuals can be toxic, causing sensitization reactions when the

materials are implanted in the body (Dolovich et al. 1984; Marshall et al. 1985; Chapman

et al. 1986; Dolovich et al. 1987). Research has shown that materials sterilized with

increasing concentrations of the chemical sterilizing agent have increasing quantities of

residuals (Lyarskii et al. 1984). The studies presented here will ascertain the

environmental parameters required for sterilization of biomaterials with low

concentrations of chlorine dioxide gas and assess polymers sterilized using these

conditions for the cytotoxicity of possible chemical residuals. 16

Investigations into preventing biomaterial-based infections in situ have focused on changing the biomaterial properties. Materials with altered physicochemical characteristics to prevent bacterial adhesion have been developed, and antibiotics and silver have been incorporated into the biomaterials to inhibit bacterial colonization.

Unfortunately, the rapid depletion of incorporated antimicrobial agents, altered bactericidal activity in vivo, and the development of antibiotic resistance, have all limited the effectiveness of current technologies. In these studies a chlorine dioxide generating material was assessed using in vitro and in vivo assays.

While assessing the bactericidal efficacy of a selection of chlorine dioxide generating materials, a spontaneous bacterial mutant with a reduced susceptibility to chlorine dioxide was isolated. The final section of this work will investigate a potential mechanism of resistance to chlorine dioxide.

1. PREFACE

The journey toward discovering the mechanism underlying disease and methods

of treatment has been fraught with teachings that were often forgotten or disbelieved,

leaving them to be rediscovered in times that were, on occasion, more accepting of the

theories being proposed. Girolamo Fracastoro (1478-1553) first described the

mechanisms behind disease in the 16th century, proposing that infection was caused by

the passage of minute entities capable of self-multiplication between people (Fracastoro

1546). Unfortunately, it is his poem about a hero named Syphilus, who was punished

with a “dread and odious” disease, for which Fracastoro is actually remembered. In

1625, Sir Francis Bacon stated that “Removing that which causes putrefaction, does prevent and avoid putrefaction”, and listed ten methods or substances for prevention and

treatment (Bacon 1635). Two centuries later, however, it was Louis Pasteur whose

hypotheses and experiments actually inspired the discovery of microbial disease and the

preservation of food and beverage (c.f. pasteurization). The work of Pasteur inspired

Joseph Lister, who selected phenol as the disinfectant to prevent infection, and became an

avid spokesman for the substance. While other research had preceded many of his

theories and alternative disinfectants had been discovered, Lister’s success stemmed from

his willingness to believe in his product. In 1876, Lister lectured to the International

Medical Congress in the United States of America. In the audience was Joseph

Lawrence, the Missouri physician who developed Listerine, and Robert Johnson, the New 18

York pharmacist who produced the gauze soaked in phenol that Lister was using, which became the founding product for Johnson & Johnson (Johnson & Johnson 1998).

While the application of chemicals to kill microorganisms first occurred in 1676, albeit unknowingly, by Anton van Leeuwenhoek (van Leeuwenhoek 1684), it was the late 19th century before research into the standardization of the disinfection process began

(Koch 1881; Kronig and Paul 1897; Rideal and Walker 1903; Chick and Martin 1908).

Similarly, an understanding about the application of heat as a method for achieving

antimicrobial activity began in the late 18th century (Spallanzani 1776), even though it

has been utilized since biblical and medieval times. The advent of radiation and its

effects on microorganisms occurred in rapid succession in the late 19th and early 20th centuries (Downes and Blunt 1877; Ward 1893; McCulloch 1945).

Despite the fact that sterilization and disinfection processes designed to eliminate or inhibit the growth of microorganisms have been described for centuries, they remain imperfect. This is as related to the innate ability of microorganism survivors to adapt to environmental challenges, as it is to the methods currently available. Since the introduction of materials into the body (biomaterials), and the realization of the role they play in the development of implant-associated infection (Elek and Conen 1957), the requirements of the sterilization process have expanded beyond the destruction of microorganisms to incorporate the impact on the materials, and the effects in vivo.

This body of work addresses three areas: the development of a new method for the sterilization of biomaterials that utilizes the biocide chlorine dioxide; an assessment of a chlorine dioxide generating material for the prevention of biomaterial-associated 19

infections in situ; and an investigation into the mechanisms of acquired microbial

resistance to chlorine dioxide. Five specific aims and hypotheses, pertaining to the

aforementioned areas, have been investigated in this body of work:

Specific Aim 1: To establish the environmental conditions necessary for sterilization of

Bacillus atrophaeus spore strips using low concentrations of chlorine dioxide gas.

Hypothesis #1. There is a relationship between the exposure time and concentration of

chlorine dioxide required to achieve sterilization and disinfection. Lower concentrations of chlorine dioxide will require longer exposure times to achieve sterilization and disinfection, than higher concentrations of chlorine dioxide.

Specific Aim 2: To determine whether chlorine dioxide sterilization of polymers results in cytotoxic residuals associated with the material. Hypothesis #2. Chlorine dioxide sterilization will not cause cytotoxic chemical residues associated with sterilized polymers.

Specific Aim #3: To investigate microbial adhesion and colonization on the surface of a chlorine dioxide generating material in vitro. Hypothesis #3. The chlorine dioxide generated by the material will cause bactericidal activity in adhered bacteria.

Specific Aim 4: To assess the efficacy of a chlorine dioxide generating material in in vitro and in vivo environments. Hypothesis #4. The chlorine dioxide generating material will cause sustained bactericidal activity in both in vitro and in vivo environments. The chlorine dioxide concentrations generated by the material will not cause an adverse response to the local host tissue in vivo.

Specific Aim #5: To investigate a potential mechanism of resistance to chlorine dioxide 20

of a Staphylococcus epidermidis mutant with a reduced susceptibility to the bactericidal effects of chlorine dioxide. Hypothesis #5: The mechanism of resistance to chlorine

dioxide for the Staphylococcus epidermidis mutant is through the over-production of a

molecule that is inhibiting the effects of chlorine dioxide.

Definitions

The following is a list of official definitions from the Food and Drug

Administration and the US Environmental Protection Agency for terminology used

within, or pertaining to, this body of work:

Antibiotic: An organic chemical substance produced by microorganisms that has the

capacity in dilute solutions to destroy or inhibit the growth of bacteria and other

microorganisms.

Antimicrobial: The property of any pesticide to prevent, destroy, or mitigate any

bacteria, pathogenic fungi, or viruses.

Bactericide: An agent that kills bacteria. This term is applied to chemical agents that

kill all bacteria, both pathogenic and nonpathogenic, but not necessarily bacterial spores.

It is different from germicide in that it does not include fungi, viruses, or other

microorganisms that are not bacteria.

Biocide: A chemical or physical agent that kills all living organisms, pathogenic and

nonpathogenic.

Biologic indicator (BI): A standard preparation of bacterial spores on or in a carrier

utilized to demonstrate whether sterilizing conditions have been met. 21

Decontamination: Disinfection or sterilization of infected articles to make them suitable for use.

Disinfectant: A chemical agent that eliminates a defined scope of pathogenic organisms, but not necessarily all microbial forms.

Disinfection: The destruction of pathogenic and other forms of microorganisms by thermal or chemical means.

Germicide: An agent that destroys microorganisms, including pathogenic organisms.

Inactivation: Removal of the activity of a microorganism by killing or inhibiting reproductive or enzyme activity.

Infection: The growth of microorganisms in a host.

Nosocomial: Hospital-acquired infections.

Pasteurization: A process developed by Louis Pasteur of heating milk, wine, or other liquids to 60-100°C for approximately 30 minutes to reduce significantly or kill the number of pathogenic and spoilage organisms.

Pathogen: Any disease-producing microorganism.

Spore (endospore): The dormant state of an organism, typically a bacterium or fungus, which exhibit a lack of biosynthetic activity and reduced respiratory activity, and are

capable of surviving in unfavorable environments. Spores have an increased resistance to

heat, radiation, dessication, and various chemical agents, relative to vegetative cells.

Sporicide: An agent that destroys microbial spores; in particular, a chemical substance

that kills or inactivates bacterial spores. 22

Sterilant: An agent that destroys all viable forms of microbial life to achieve sterilization.

Sterile: Free from living microorganisms; in practice described as a probability function, such as the probability of a surviving microorganism being one in one million.

Sterility assurance level (SAL): Because absolute sterility cannot be ensured, the SAL is an indication that no greater than a predetermined number of viable microorganisms can exist on a product, based in the intended use of the product and the risk determined to be acceptable.

Sterilization: A process intended to remove or destroy all viable forms of microbial life, including bacterial spores, to achieve an acceptable sterility assurance level.

Vegetative cells: Microbial cells that are in the growth and reproductive phase of the growth cycle.

Viable: Describes microorganisms capable of reproduction under favorable conditions

Virucide: An agent that destroys or inactivates viruses to make them noninfective, especially a chemical substance used on living tissue. The word virucide is a misnomer, since “cide” means kill, and the virus, by itself, is not a living entity. Viruses are inactivated, rather than killed. 23

2. INTRODUCTION: STERILIZATION TECHNOLOGIES FOR

BIOMATERIALS

Sterilization or disinfection of biomaterials prior to their application is essential for minimizing the risks associated with introducing bacteria into the host environment.

-6 Sterilization is defined as a process that provides an acceptably low probability (10 ) that any microorganism will survive the method of treatment, while disinfection refers to the destruction of all pathogenic microorganisms with the exception of bacterial spores. The standard sterilization procedures for biomaterials can be either physical or chemical, and are described as follows.

Sterilization by Heat

Sterilization by heat is considered to be the most reliable and economical form of sterilization for materials that can withstand the extreme temperatures required. Elevated temperatures may be used in conjunction with other antimicrobial agents, creating synergistic sterilization methodologies (Olson 1997; Russell et al. 1997). Sterilization by heat for medical products utilizes either steam or dry heat sterilization.

Steam Sterilization

Steam sterilization utilizes temperatures ≥ 121°C (121-134°C) and requires

exposure times ≥ 15 minutes to ensure sterility. The duration of the exposure time

ensures that all surfaces of the product have reached the minimum required temperature.

For steam sterilization to be effective, removal of the air from the chamber is essential for penetration of moisture into the material. 24

Dry Heat Sterilization

Relative to sterilization with steam, dry heat sterilization requires higher

temperatures (160-180°C) and longer exposure times to achieve sterilization. However,

dry heat sterilization does successfully penetrate materials that are impermeable to water.

Despite being called dry sterilization, a range of humidity levels are used, from very low

to just below saturation levels of moisture. Generally the relative humidity is set between

20-40%.

Mechanisms of Biocidal Activity

Exposure to the environmental conditions necessary for sterilization by heat

causes physical or chemical changes in cells that results in the denaturation of major cell

components, including proteins and nucleic acids. The process of dentauration involves

the loss of protein functionality as the tertiary structure of the protein is altered as bonds

between polypeptide chains are disrupted. While steam sterilization causes the

denaturation of nucleic acids (Brannen 1970), it is the coagulation of the cytoplasmic

proteins and inactivation of the cell enzymes that are believed to be primarily responsible

for the bactericidal action of steam sterilization. The presence of moisture alters the

mechanisms of biocidal activity by heat. The denaturation of proteins that occurs during steam sterilization is replaced by alternative mechanisms of protein inactivation, including oxidative damage, when the relative humidity is low during dry heat sterilization (Gardner and Peel 1986).

There are variations in the susceptibility of proteins and microbial species to the effects of different types of heat. For example, Bacillus stearothermophilus spores are 25

more resistant to heat and high relative humidity, while Bacillus atrophaeus spores are

more susceptible. Conversely, Bacillus stearothermophilus spores are more susceptible

to dry heat, and Bacillus atrophaeus spores more resistant.

Procedure for Sterilization by Heat

Within a heat sterilization cycle, it is the temperature and relative humidity that

influence the exposure time required for sterilization to be achieved. There are multiple

temperature-time combinations that will result in sterilization. Since different materials

have different tolerances to extreme heat, the properties of the material will often dictate

which temperature-time conditions are used for sterilization.

The steam sterilization cycle has three phases: increasing the chamber

temperature to the required level while removing the air from the chamber (pre-vacuum

phase), exposure of the materials to the required sterilization temperature for the

specified time, and finally venting, cooling, and drying of the chamber. The dry heat

sterilization cycle is similar, except that the pre-vacuum phase is not required.

The chamber into which the materials to be sterilized are placed is constructed to withstand the pressure increases that are required to elevate the temperature of the steam

(steam sterilization) and insulate against temperature losses. Once the sterilization cycle commences the door to the chamber is sealed so that it cannot be opened until the cycle is

either completed or terminated. Temperature, time, and pressure are monitored both as a

safety precaution, and to ascertain whether the environmental conditions associated with

each given cycle are sufficient to cause sterilization. A sterilization chamber is depicted

in Figure 2.1. 26

Figure 2.1: Diagrammatic representation of a steam sterilization system (reprinted from

Principles and Practice of Disinfection, Preservation, and Sterilization, 3rd Edition,

Blackwell Publishing) 27

Limitations and Disadvantages of Sterilization by Heat

The primary limitation of sterilization by heat is alterations to the material properties caused by exposure to elevated temperatures and/or moisture. Exceeding the glass transition temperature (Tg) of many synthetic polymers can cause physical and/or chemical alterations, potentially altering the mechanical and/or biological properties of the material.

The degree of hydrophilicity of the material will also impact the effectiveness of steam sterilization. The combination of high relative humidity and pressure may cause polymers with a high degree of hydrophilicity to absorb water resulting in hydrolytic cleavage of polymer bonds, which causes microcavitation and subsequent degradation of the polymer.

Radiation Sterilization

While there are multiple forms of radiation, only gamma rays and electrons are suitable for sterilization purposes. Alpha particles, although capable of causing the ionization reactions required for antimicrobial activity have limited penetration into surfaces, while neutrons, which can penetrate deeply into materials, induce radioactivity.

Electron Beam Sterilization

Electron beam sterilization is a form of ionizing radiation that was first used at the

Ethicon Coporation in 1957 (Dempsey and Thirucote 1989). An accelerator produces a beam of electrons that cause the ionization of cellular components. The materials to be sterilized are placed in the pathway of the electron beam for different exposure times, which are dependent upon the time needed to accumulate the required dose (25kGy). 29

Electron beam sterilization has limited penetration distance into materials, and as such it

can only be used to sterilize small, thin products.

Gamma Radiation

Gamma radiation is the most common form of radiation sterilization. The source of gamma radiation comes from the decay of radioisotopes Cobalt-60 (60Co) or Caesium-

137 (137Ce). Although 167Ce has a longer half-life than 60Co, larger quantities are

required to achieve the same dose, and as such 60Co is generally used. The shorter half-

life associated with 60Co results in longer exposure times being required to achieve the

minimum dose (25kGy) as the radioisotope decays. This means that the radioisotope in

the sterilization unit needs to be replenished regularly.

Mechanisms of Biocidal Activity

Radiation causes ionization of cell components, particularly deoxyribonucleic

acids (DNA), which contains the genetic code for every cell. Ionizing radiation produces

photons, which, when they interact with the biological molecules of the bacteria, produce

charged particles. These particles interact with other cell components to produce free

radicals and activated molecules, which in turn cause lesions in the DNA strands and in

the nitrogenous bases.

Procedure for Radiation Sterilization

With either form of radiation sterilization, the materials to be sterilized need to be

exposed to a dose of at least 25kGy. Electron beam and gamma radiation have very

different methods of producing ionizing radiation. Electron beam radiation uses machine

generated electrons to produces an ionizing effect. The electron beam is produced by an 30

accelerator and does not utilize radioisotopes. Gamma radiation is produced from the

decay of the radioisotope, which is exposed to the materials to be sterilized for the time

necessary to achieve the required dose. As the radioisotope decays, the processing time

necessary to achieve a dose of 25kGy increases. When not in use, the stainless steel

storage racks containing the radioisotope are lowered into a deep (20-25 feet) water pool, preventing radiation from reaching the surface. Figure 2.2 depicts a gamma sterilization system. The presence of moisture and oxygen will influence the production of free radicals during the sterilization cycle, increasing damage to sterilized material samples

(Cheng and Kerluke 2003).

Limitations and Disadvantages of Radiation Sterilization

Sterilization with radiation results in the degradation and/or increased cross-

linking of many types of polymers. The chemical structure of the polymer determines

which reaction occurs (Chapiro 1962). Polymers formed by exothermic reactions tend to

cross-link, while those formed by endothermic reactions to degrade upon exposure to

radiation. If cross-linking predominates, then increasing the dose of radiation leads to an

increase in the molecular weight of the polymer being sterilized. Conversely, polymers

that tend to degrade when sterilized with high-energy radiation will exhibit a decrease in

molecular weight as the radiation dose increases (Bruck and Mueller 1988). Sterilization

by radiation may also lead to the production of toxic gas and the formation of free

radicals that may contribute to material degradation. Additives in polymers, such as

antioxidants, initiators and catalyzing agents, may also be affected by high-energy

radiation and interact with the polymer chains, affecting the material properties. 31

Figure 2.2: A gamma radiation (60Co) sterilization system (reprinted from Introduction to

Sterilization and Disinfection, Gardner, J. F. and M. M. Peel, Sterilization by Ionizing

Chemical Agents

With the advancement of medical devices to include materials that cannot

withstand the high temperatures associated with sterilization by heat, methods of cold

sterilization have been developed. In addition to the aforementioned radiation

sterilization, procedures that utilize chemical agents in low temperature environments

have been developed. The two common forms of gas sterilization for biomaterials are

ethylene oxide gas and a low temperature steam and formaldehyde process.

Ethylene Oxide

Ethylene oxide (C2H4O) is a highly permeable, colorless gas with a boiling point of 10.7°C at ambient pressure. It is highly explosive at concentrations exceeding 3.6%

(by volume) in air and as such is generally stored at 12% (by weight) with 88%

fluorocarbon-12 (CCl2F2). Ethylene oxide reacts with water to form ethylene glycol and

has 1.5 times the density of air.

Low temperature steam and formaldehyde

The low temperature steam and formaldehyde process (Alder et al. 1966) was designed primarily for the sterilization of cytoscopes and heat-sensitive equipment.

Formaldehyde is not available as a pure gas or liquid, but is generated by heating formalin, which contains 37% (w/v) formaldehyde.

Mechanisms of Biocidal Activity

Ethylene oxide and formaldehyde are both alkylating and radiomimetic agents

(Ehrenberg et al. 1974). The process of alkylation involves the addition of hydrocarbon

groups to reactive amino (NH2), sulphydryl (SH), hydroxyl (OH) or carboxyl (COOH) 34

groups, which are associated with proteins, and to imino (NH) groups, which are

associated with nucleic acid bases. The latter reaction, with amino acids, is the proposed

mechanism of biocidal activity for ethylene oxide (Ross 1953; Gunther 1980), with the

ring nitrogen atoms of purine and pyrimidine bases forming critical alkylation sites

(Bruch 1973; Phillips 1977). Formaldehyde reacts with cell DNA during the replication

process as the double-stranded helix is unwinding (Vologodskii and Frank-Kamenetskii

1975; Shikama and Miura 1976). The antimicrobial action of formaldehyde is

accelerated through the addition of saturated steam at temperatures between 70°C and

75°C.

Procedure of Sterilization with Chemical Agents

The antimicrobial action of ethylene oxide is dependent upon achieving an

adequate concentration of chemical vapor in an environment with the appropriate

temperature and relative humidity, and maintaining these parameters for the required

exposure time. Concentrations of at least 450mg/l, at temperatures between 45°C and

60°C with a relative humidity of at least 33% are required to cause sporicidal activity with ethylene oxide. In commercial sterilization systems, the environmental conditions used are ethylene oxide concentrations between 700-1000mg/l, at temperatures between

45°C and 60°C, and with a relative humidity of at least 70%. Using these conditions, an exposure time of 4-6 hours is required to achieve sterilization. The increased concentrations used in the commercial sterilizer are necessary to overcome penetration barriers and to allow for absorption of the chemical gas into the packaging material. The importance of moisture in sterilization with chemical agents has been established, 35

although the mechanism remains unclear. Gunther (1980) proposed that the formation of

a film of water on the surface of the microbes causes an increase in the local

concentration of the chemical agent, while also providing a medium for the ionization of

the reactive groups. Alternatively, moisture may be necessary to dissolve any organic or crystalline matter protecting some spores (Doyle and Ernst 1967).

The biocidal action of formaldehyde is slow at room temperature, but increases

when combined with elevated temperatures and increased relative humidity.

Concentrations of at least 5mg/l of formaldehyde are required to achieve sterilization,

with below atmospheric pressures in the sterilization chamber assisting with penetration

into the materials being sterilized. Exposure times are generally between 40 minutes and

2 hours when the temperature is between 70°C and 75°C.

Sterilization with ethylene oxide or low temperature steam formaldehyde require

that the sterilization chamber be 90kPa below atmospheric pressure to increase the rate of

penetration of moisture and sterilizing agents into the materials being sterilized. The

temperature of the chamber is increased, as is the relative humidity by passing steam

through the chamber. The sterilizing agent is then added for the required exposure time.

For sterilization with ethylene oxide, the chamber is then evacuated and returned to

atmospheric pressure, while the low temperature steam and formaldehyde sterilization process requires that the chamber be evacuated and the sterilized articles dried at 95kPa

below atmospheric pressure. For both methods of sterilization, the materials need to be

aerated at elevated temperatures for at least 12 hours to remove chemical residuals. 36

Examples of an ethylene oxide and a low temperature steam formaldehyde sterilization

cycle are provided in Figure 2.3 and Figure 2.4, respectively.

Limitations and Disadvantages of Sterilization with Chemical Agents

A common limitation of sterilization with both ethylene oxide and low

temperature steam and formaldehyde is the duration of the sterilization cycle. Both forms

of chemical sterilization are too slow for sterilization of equipment that is to be used the same day. The presence of chemical residues associated with materials sterilized using

ethylene oxide is also an issue. Research has indicated that many polymers will readily

absorb ethylene oxide, as well as its secondary by-products ethylene chlorohydrin and

ethylene glycol that form in the presence of water. These chemical residues can cause

toxic side effects (Stanley et al. 1971; McGunnigle et al. 1975) and hypersensitivity

reactions in an in vivo environment (Dolovich et al. 1984; Marshall et al. 1985; Chapman

et al. 1986; Dolovich et al. 1987). Formaldehyde residues occur in the form of

monomeric formaldehyde and as absorbed paraformaldehyde, with the concentration of

each dependent on the type of material being sterilized (Handlos 1977; Handlos 1979;

Vink 1986). The levels of residual formaldehyde associated with polymers is generally

reported to be significantly lower that those associated with ethylene oxide (Gibson et al.

1968).

Advances in Sterilization Technologies

Each of the aforementioned technologies used for the sterilization of biomaterials

have limitations. As such, in the past decade, research has been conducted for both the

improvement of current technologies and the development of new technologies for the 37

Figure 2.3: An example of an ethylene oxide sterilization cycle. The stages of the cycle

are: (i) Air removal; (ii) Leak test; (iii) Steam injection; (iv) Conditioning; (v) Injection

of sterilizing injection; (vi) Exposure; (vii) Removal of the sterilizing agent; (viii)

Flushing; (xi) Air break (reprinted from Principles and Practice of Disinfection,

Preservation, and Sterilization, 3rd Edition, Russell, A.D, et al., Gaseous Sterilization

Figure 2.4: An example of a low temperature steam formaldehyde sterilization cycle.

The stages of the cycle are: (i) Air removal; (ii) Formaldehyde injection; (iii) Steam injection; (iv) Evacuation; (v) Repeated pulses (ii-iv); (vi) Exposure; (vii) Flushing; (viii)

Air break (reprinted from Principles and Practice of Disinfection, Preservation, and

with permission from Blackwell Publishing) 40

sterilization of biomaterials.

Optimization of Ethylene Oxide Sterilization

While the limitations associated with sterilization by heat and radiation are innate to their mechanisms of antimicrobial activity, the principle limitation associated with sterilization by chemical agents is the presence of the chemical residues associated with the material. The development of technologies to improve the ethylene oxide sterilization process has focused on methods to minimize these residuals. Research has indicated that the absorption of ethylene oxide by polymers is caused by an interaction between the molecules of the sterilizing agent and the material being sterilized (Gibson

1987). A theory explored by Matthews et al. (1994) suggested that if the chemical bonds between the ethylene oxide and the polymer substrate could be interrupted, the rate of gas desorption from the material would be increased. Microwave radiation was used to provide an energy source to interrupt the bond between the sterilizing agent and the polymer substrate in a combined sterilization/desorption unit (Figure 2.5), and was found to increase the elimination of ethylene oxide from the polymer materials by up to 400%

(Samuel and Matthews 1985; Gibson 1987). The mechanism of action is believed to be linked to the disruption of the hydrogen bonds between the sterilizing agent and the polymer substrate by the microwaves, causing a reduction in the immobilized ethylene oxide molecule bound to the polymer substrate, with a subsequent increase in the rate of desorption (Matthews et al. 1989).

Hydrogen Peroxide Gas Plasma

Early studies assessing the process of hydrogen peroxide gas plasma found that it 42

Figure 2.5: A prototype microwave sterilizer/desorber. (1)100L stainless steel chamber;

(2) Inlet ports; (3) Outlet ports; (4) Rotary basket for (for sterilized equipment); (5) Wall bearings; (6) Microwave-transparent sheet; (7,8) 750W microwave generators; (9,10)

Launch antennae; (11) Infrared thermometer (reprinted from Journal of Hospital

Infection, 15, Matthews et al., Sterilization of implantable devices, p191-215, copyright

1994, with permission from Hospital Infection Society)

presented a level of antimicrobial activity that exceeded that of plasma and hydrogen peroxide alone (Jacobs and Lin 1987). While the exact mechanism of sporicidal action of hydrogen peroxide gas plasma remains unknown, the dissociation of hydrogen peroxide as the plasma phase commences results in the formation of reactive molecules (hydrogen peroxide, free radicals, UV radiation), which significantly contribute to the biocidal action of hydrogen peroxide gas plasma (Venngopalan and Shih 1981). The formation of these free radicals is similar to the secondary reactions associated with gamma irradiation

(Silverman 1983). The free radicals from the hydrogen peroxide gas plasma have significantly lower energies associated with them (1-10eV versus >1MeV). The lower energies of the hydrogen peroxide gas plasma derived free radicals results in only the surface properties of the materials, rather than the bulk properties, being affected. Once the plasma phase is completed, the reactive species recombine to form stable, nontoxic compounds, predominantly water and oxygen.

In 1993, a commercial hydrogen peroxide gas plasma sterilization system was approved by the Food and Drug Administration (FDA) to be used as a sterilization method for medical devices. The STERRAD™ sterilization system was designed and developed by Advanced Sterilization Products, a Johnson and Johnson company. The sterilization cycle begins with a vacuum being drawn on the sealed chamber. A 59% aqueous solution of hydrogen peroxide is added into the chamber and vaporized to provide a minimum concentration of at least 6.0mg/l. Once the pressure in the chamber is reduced again, the low-temperature gas plasma is created through the application of

400W of radiofrequency energy. As the plasma is generated, the hydrogen peroxide 45

vapor dissociates, producing reactive species that have antimicrobial properties. Once

the exposure time (45-75 minutes at 50°C) has been completed, the radiofrequency

energy is turned off and the chamber is returned to atmospheric pressure. An example of

a hydrogen peroxide sterilization cycle is depicted in Figure 2.6.

Independent studies assessing the effects of hydrogen peroxide gas plasma

sterilization on biomaterials found that, while cosmetic changes may occur with some

polymers and metals, the functionality of the materials was generally maintained,

although some sulfide-based polymers did exhibit embrittlement (Feldman and Hui

1997). Comparative studies assessing resterilization of metallic instruments with

STERRAD™ found no evidence of corrosion or loss of functionality, while resterilization

with steam caused material degradation (Geiss et al. 1994; Timm and Gonzales 1997).

The mechanical properties of polymers were not significantly altered by resterilization with hydrogen peroxide gas plasma, but were significantly affected by resterilization with gamma radiation (Jacobs and Lin 1996). Sterilization of narrow lumen endoscopes using the different STERRAD™ models (Figure 2.7) found that the STERRAD200™ model exhibited the greatest efficacy, potentially because of the different injection site and the higher concentration of hydrogen peroxide used in the sterilization cycle (6.3mg/l versus

6mg/l) (Okpara-Hofmann et al. 2005).

Sterilization with Ozone

Ozone (O3) is produced by the dissociation of oxygen into atomic oxygen in an

energizing environment that allows for recombination of the atoms to form O3. Ozone is

an unstable allotrope of oxygen and a potent oxidizing agent that achieves its biocidal 46

Figure 2.6: An example of a sterilization cycle with hydrogen peroxide. The stages of the cycle depicted are: (i) Air removal; (ii); Hydrogen peroxide injection; (iii) Diffusion; (iv)

Gas plasma; (v) Aeration (reprinted from Principles and Practice of Disinfection,

Figure 2.7: Examples of the different STERRAD™ models: (A) STERRAD50™; (B)

STERRAD100/100S™; (C) STERRAD 200™ (reprinted from Journal Hospital Infection,

59, Okpara-Hofmann et al., Comparison of low-temperature hydrogen peroxide gas plasma sterilization for endoscopes using various Sterrad™models, p280-285, copyright

2005, with permission from the Hospital Infection Society) 49

(A) (B) (C) 50

activity through the denaturation of cell membranes. Ozone is generally produced using a corona discharge (Figure 2.8), by passing dry air or O2 between two electrodes that are separated by a dielectric material. The concentrations of ozone produced are dependent upon whether the feed gas is air, producing 1-3% O3, or pure O2, producing 2-6% O3

(Rice 1986). Alternatively, ozone can be generated by photochemical, electrolytical or radiochemical methods (Masschelein 1982; Langlais et al. 1991). Historically, ozone has been utilized for the decontamination of water (Rice 1999) and food processing (Rice et al. 1997). However, in 2003 a commercial system developed by TSO3 Inc. was approved

by the FDA for the sterilization of selected biomaterials. The TSO3 system has an

average cycle time of 4.5 hours, and runs at a temperature of approximately 32°C. The

materials that can be sterilized using the TSO3 are limited to some reusable medical

devices that are in contact with patients for less than 24 hours, including stainless steel

instruments (TSO3 2005).

Chlorine Dioxide

Chlorine dioxide (ClO2) was first prepared in 1802 by Chenevix (Sidgwick 1951)

and its composition analyzed by Davy, who independently prepared chlorine dioxide in

1811 (Davy 1811). The names chlorine oxide, anthium dioxide, chlorine (IV) oxide,

chlorine peroxide, chloroperoxyl, and chloryl radical, have also been used to describe

chlorine dioxide.

Chemical and physical properties of chlorine dioxide

Chlorine dioxide is a yellow-green gas with an odor that is similar to chlorine, and

is one member of a series of chlorine oxides (Table 2.1) that all have distinct properties. 51

from Ozone in Water Treatment: Applications and Engineering, Chapter II: Fundamental

Aspects, p103-109, by Bruno Langlais, David A. Reckhow, Deborah R. Brink (Editors).

Reproduced by permission of Routledge/Taylor & Francis Group, LLC) 52

Oxide Molecular formula Chlorine monoxide Cl2O Chlorine dioxide ClO2 Chlorine peroxide Cl(O2) (different structure to ClO2) Chlorine trioxide ClO3 Chlorine tetroxide ClO4 Chlorine heptoxide Cl2O7

Table 2.1: Oxides frequently referred to as chlorine oxide, of which chlorine dioxide is a member

At atmospheric pressure chlorine dioxide freezes at -59°C and boils at 11°C (Windholz

1983). It has a relative density in liquid of approximately 1.6 (water = 1) (Cheesman

1930; Gordon 1993) and is stable at concentrations of less than 10% in air at atmospheric

pressure. At higher concentrations chlorine dioxide is unstable and its decomposition can

be catalyzed by light. For this reason, chlorine dioxide gas cannot be compressed for

storage in gas tanks, with or without other gases (Aieta and Berg 1986; Gordon 1993).

The ultraviolet absorption spectrum of chlorine dioxide, both in aqueous solution and as a

gas, exhibits a broad band near 360nm (Gordon et al. 1972), which provides an accurate

method for measuring chlorine dioxide concentration.

Production of chlorine dioxide

There are many chemical reactions that can be used to produce chlorine dioxide.

As such, the method chosen is usually dependent upon the final application of the

chlorine dioxide. For the production of large quantities of chlorine dioxide, the reduction

of sodium chlorate by an acid is usually used (Gordon et al. 1972). When smaller

quantities are required, or the purity of the chlorine dioxide produced is of concern, the oxidation of sodium chlorite by an acid is used. Chlorine dioxide can also be generated electrolytically, by passing nitrogen dioxide gas through a column of sodium chlorite

(Hutchinson and Derby 1947).

Today, one of the most common methods of generating chlorine dioxide is through the oxidation of chlorine (gas or aqueous solution) and sodium chlorite (Aieta and Berg 1986) by the following reaction.

2NaClO2 + Cl2 → 2ClO2 + 2NaCl [Equation 2.1] 55

For the production of chlorine dioxide gas without chlorine as a by-product, the oxidation

of sodium chlorite by persulfate is frequently used (Equation 2.2) (Gordon et al. 1972;

Masschelein and Rice 1979).

2NaClO2 + Na2S2O8 → 2ClO2 +2Na2SO4 [Equation 2.2]

Antimicrobial properties of chlorine dioxide

Even though the biocidal action of chlorine dioxide was first documented during

the 1930’s (Schaufler 1933; Kovtunovitch and Chernaya 1936), the exact mechanism of

the biocidal action of chlorine dioxide has not been identified. Studies assessing the specific interactions of chlorine dioxide with select biomolecules have established that chlorine dioxide reacts with the amino acids cysteine, tryptophan, and tyrosine (Noss et al. 1983; Noss et al. 1986), and also with free fatty acids (Ghanbari et al. 1982).

Investigation into the effects of chlorine dioxide on the physiological functions of microorganisms established the primary mode of action was not the inactivation of dehydrogenase enzymes, the protein synthesizing complex, or DNA (Bernarde et al.

1967; Roller et al. 1980). Research conducted by Olivieri et al. (1985) found that the nucleic acid of poliovirus remained infectious, despite complete inactivation of the virus.

This supports the hypothesis that the mechanism of action of chlorine dioxide involves interactions with proteins (viral proteins, in the case of antiviral activity), rather than nucleic acids. Aieta and Berg (1986) suggested that chlorine dioxide may cause the disruption of the organisms outer membrane, or cause a gross change in membrane permeability, while White (1972) reported the abrupt inhibition of protein synthesis when chlorine dioxide was present. 56

Studies assessing the antimicrobial activity of chlorine dioxide against bacteria in

the late 1940’s determined that the bactericidal activity of chlorine dioxide was

unaffected by pH variations between 6 and 10 (Ridenour and Ingols 1947; Ridenour and

Armbruster 1949), while the inactivation of viruses was more efficient at a higher pH

(Alvarez and O'Brien 1982; Sansebastiano et al. 1983). In 1949 Ridenour and

Armbruster showed that the bactericidal activity of chlorine dioxide was unaffected by

low temperatures. Rosenblatt et al. (1987) outlined the importance of humidity in

optimizing sporicidal activity with chlorine dioxide, analogous to sterilization with

ethylene oxide. Chlorine dioxide has been shown to inactivate numerous microorganisms

(summarized in Table 2.2), including species of bacteria, viruses, endospores, and

protozoa. Furthermore, Brown et al. (1982) established that at low concentrations,

chlorine dioxide causes moderate to substantial inactivation of prions.

Toxicity of chlorine dioxide and its metabolites

Chlorine dioxide gas is a mucous membrane irritant and can cause pulmonary

edema when high concentrations are inhaled (White 1972). The occupational exposure

limits to chlorine dioxide gas have been set at 0.1ppm for up to 8 hours, with a 0.3ppm,

15 minute reference period (American Conference of Governmental Industrial Hygenists

1990). Most of the studies investigating the toxicity of chlorine dioxide have assessed

the ingestion of chlorine dioxide as a liquid and the chlorine dioxide metabolites, chlorite,

chloride and chlorate.

Moore and Calabrese (1982) and Abdel-Rahman et al. (1980; 1982; 1985)

investigated hematological changes in mice, rats and chickens that had been treated with 57

Bacteria Endospores Viruses Protozoa Aerobacter aerogenes Bacillus subtilis Enteroviruses Cryptosporidium (Ridenour and Armbruster 1949) sp. (Sansebastiano et al. 1986) parvum oocytes Eberthella typhosa (Rosenblatt et al. 1985; HIV (Peeters et al. 1989; Korich (Ridenour and Armbruster 1949) Jeng and Woodworth (Farr and Walton 1993) et al. 1990) Escherichia coli 1990a; Botzenhart et al. Poliovirus Giardia lamblia 1993) (Ridenour and Armbruster 1949; (Alvarez and O'Brien (Korich et al. 1990) Scatina et al. 1985; Fujioka et al. Clostridium 1982; Sansebastiano et al. Streptomyces 1986; Harakeh et al. 1988; Botzenhart perfringenes 1983; Fujioka et al. 1986) et al. 1993) griseus (Fujioka et al. 1986) Rotavirus (Whitmore and Denny Klebsiella pneumoniae (Berman and Hoff 1984; Clostridium 1992) (Harakeh et al. 1985; Harakeh et al. Chen and Vaughn 1990) 1988) sporogenes Legionella pneumophila (Rosenblatt and Knapp 1988) (Botzenhart et al. 1993; Walker et al. 1995) Pseudomonas aeruginosa (Ridenour and Armbruster 1949; Harakeh et al. 1988) Salmonella paratyphi B (Ridenour and Armbruster 1949) Salmonella typhimurium (Harakeh et al. 1988) Staphylococcus aureus (Ridenour and Armbruster 1949) Streptococcus faecalis (Fujioka et al. 1986) Streptococcus pyogenes (Harakeh et al. 1988)

Table 2.2: Microorganisms inactivated by chlorine dioxide

chlorine dioxide in drinking water up to concentrations of 1000mg/l. Aside from a transient increase in serum glutathione and acute indications of anemia, which resolved with longer-term exposure, there was little difference between treated animals and controls. Abdel-Rahman et al. (1982) also found that chlorine dioxide and its metabolites chlorite and chlorate, are eliminated from the rat’s system more rapidly than chlorine, predominantly via the urine. Studies assessing the effects of ingested chlorite and chlorate solutions showed a decrease in red blood cell counts and evidence of morphological changes in erythrocytes (Heffernan et al. 1979; Abdel-Rahman et al. 1980;

Moore and Calabrese 1982). The depletion of glutathione from erythrocytes when incubated with chlorite has been demonstrated, while continued exposure to chlorine dioxide results in the recovery of the initially exhausted glutathione levels (Abdel-

Rahman et al. 1985; Abdel-Rahman and Scatina 1985). The ingestion of chlorine dioxide, chlorite and chlorate solutions by human subjects affirmed the safety and tolerance of these solutions (Lubbers et al. 1982).

Ingestion of 10mg/l solutions of chlorine dioxide or chlorite, but not chlorate, caused an increased turnover of intestinal epithelium in a rat animal model (Abdel-

Rahman et al. 1985). The ingestion of chlorine dioxide also caused an inhibition of thyroid function in rats (Orme et al. 1985), monkeys (Bercz et al. 1982) and pigeons

(Revis et al. 1986). The effect was reversible when the animals were no longer ingesting chlorine dioxide (Bercz et al. 1982), and analysis of human serum for indications of thyroid effects found no evidence of reduced thyroid activity with ingestion of chlorine dioxide (Bercz et al. 1982). 59

Teratogenic studies in animals of chlorine dioxide and chlorite found no

significant association between exposure to chlorine dioxide or chlorite and abnormal

embryonic development (Couri et al. 1982; Moore and Calabrese 1982; Suh et al. 1983;

Skowronski et al. 1985; Carlton et al. 1991). A retrospective study of in utero exposure

of human neonates to chlorine dioxide treated water (0.5ppm) or chlorinated water was

performed (Tuthill et al. 1982). The infants were born in one of two towns in

Massachusetts, where the water was either treated with chlorine dioxide or chlorine. The

results from the study indicated that there were no significant differences in the rates of

jaundice, birth defects, and fetal or neonatal mortality, between the communities. There

was a positive association between premature births and the chlorine dioxide treated

water, but this was not statistically significant when the age of the mother was controlled.

To date, the carcinogenicity of chlorine dioxide has not been assessed. Research does suggest that chlorite is not carcinogenic (Yokose et al. 1987). Results from studies assessing the mutagenicity of chlorine dioxide and its metabolites have been less consistent. An AMES mutagenicity test by Jeng and Woodworth (1990b), performed on chlorine dioxide and extracts isolated from a chlorine dioxide gas exposed oxygenator, was negative. Extracts and test solutions from treating fish cubes for 5 minutes with aqueous solutions of chlorine dioxide with concentrations of 20ppm or 200ppm found no mutagenic activity (Kim et al. 1999). Conversely, recent studies assessing the genotoxicity of chlorine dioxide have found that water sterilized with aqueous chlorine dioxide concentrations ranging from 0.61-1.95mg/l and exposure times of 20 days can cause damage to mussel cell DNA (Bolognesi et al. 2004). However, the damage is 60

hypothesized to be due to the formation of chlorinated by-products rather than the

disinfection process. Using gas chromatography/mass spectroscopy, chemical analysis of

raw water treated with 1.8mg/l of aqueous chlorine dioxide showed the presence of

potential mutagenic compounds, although the analysis was qualitative, rather than

quantitative. However, mutagenic analysis of the raw water disinfected with chlorine dioxide showed no significant difference between chlorine dioxide treated and raw water samples (Guzzella et al. 2004).

Applications for chlorine dioxide

While chlorination has been the standard in the United States of America for

water purification, concerns have been expressed about the reaction of chlorine with

selected compounds, such as phenols, in the raw water. The treatment of certain water

with chlorine can result in the formation of trihalomethanes (THMs) and dioxins, which are toxic. Unlike chlorine, chlorine dioxide does not produce THMs or chlorophenols when used to treat water (Symons 1976; US Environmental Protection Agency 1977) and it does not produce a disagreeable odor or taste. Treatment of water with chlorine dioxide eliminated phenols (Burttschell et al. 1959) and cyanide (Cocheci et al. 1971), while enhancing the precipitation of iron and manganese (Enger 1960). In 1944, the first water treatment plant in the United States of America to utilize chlorine dioxide in their treatment process opened in Niagra Falls, New York (White 1972). The treatment of water with chlorine dioxide is generally achieved by the in situ generation of chlorine dioxide through the reaction of aqueous sodium chlorite with chlorine or an acid and sodium hypochlorite. Since this method leaves chlorite and chlorate residues in the 61

water, direct injection of chlorine dioxide into the water stream has been proposed as an

alternative method of water treatment (Gordon and Rosenblatt 1995).

Chlorine dioxide has been employed in the food industry for the disinfection and

sanitization of equipment and materials, and to disinfect fruit and vegetables. It is also

used to bleach and mature flour, sterilize spices, and extend the shelf life of tomatoes

(Masschelein and Rice 1979). Chlorine dioxide has been utilized to control and eliminate

odors (Masschelein and Rice 1979; Engelhard Corporation 1997), and in the textile

industry for the bleaching of pulp. One of the most recent applications of the

antimicrobial properties of chlorine dioxide involved its utilization for the

decontamination in the Hart Senate Office Building after it received a letter contaminated

with anthrax in 2001 (CBS News 2002). In 2002, the Brentwood postal plant was also

decontaminated using chlorine dioxide after two letters containing anthrax spores passed

through the facility, causing the death of two postal workers dues to inhalation of the spores (Washington Post 2002).

Several different liquid chlorine dioxide products are commercially available for sanitization and disinfection processes. These formulations of chlorine dioxide generally require the addition of an “activator” to a “stabilized” chlorine dioxide solution. Aceptrol

(Engelhard Corporation, Iselin, NY), advertised as a hard surface sanitizer, is comprised of solid sodium chlorite capsules with a proprietary activator that produce chlorine dioxide when added to water. The Alcide Corporation (Redmond, WA) manufactures several products that produce chlorine dioxide in situ when sodium chlorite and lactic acid are mixed. These products are designed to treat surfaces or animals directly to 62

remove bacteria, viruses, fungus and protozoa (Kross 1995). Similarly, International

Dioxide, a Dupont Company (Kingstown, RI), has a selection of chlorine dioxide solutions and onsite generators for personal and industrial applications. Allergan

Incorporated (Irvine, CA) has patented technology for the use of stabilized or electrolytically prepared chlorine dioxide as a disinfectant for contact lenses (Dziabo et al. 1994; Dziabo and Ripley 1994; Dziabo and Ripley 1995). Tristel (Tristel Company

Limited, Cambridgeshire, England), a highly concentrated liquid chlorine dioxide solution, has been tested as a disinfectant for endoscopes, and was found to be effective against both spores and pathogenic microorganisms (Ayliffe 2000).

Limitations and Disadvantages of Chlorine Dioxide

Since chlorine dioxide not been utilized as a sterilizing agent for medical devices, its established limitations and disadvantages have focused on its primary utilization for the decontamination of water. Producing large amounts of chlorine dioxide using

chlorite is expensive and the formation of chlorite and chlorate as residual by-products

necessitates additional processing to remove these chemical residuals from the water.

The aforementioned instability of chlorine dioxide gas at high concentrations and

consequent inability to be stored as a compressed gas (Aieta and Berg 1986; Gordon

1993), results in the requirement that chlorine dioxide be produced at the site of use. This

introduces additional labor, training, and equipment costs. Since chlorine dioxide

solutions are unstable they must be stored without residual air space to prevent the

accumulation of chlorine dioxide gas. The decomposition of chlorine dioxide is also 63

accelerated by sunlight, and as such, chlorine dioxide solutions need to be protected from light.

Research Plan

The current standard methods for the sterilization of biomaterials all have

limitations associated with them. While the antimicrobial properties of chlorine dioxide

have been established and it has been utilized as a liquid disinfectant, chlorine dioxide

gas has not been assessed as a chemical sterilizing agent for biomaterials. The following

specific aims were designed to assess the applicability of using low concentrations of

chlorine dioxide gas a sterilizing agent for biomaterials. These studies have been

presented in Chapter 3 and Chapter 4.

Specific Aim 1: To establish the environmental conditions necessary for sterilization of

Bacillus atrophaeus spore strips using low concentrations of chlorine dioxide gas

Hypothesis #1. There is a relationship between the exposure time and concentration of

3. LOW CONCENTRATION CHLORINE DIOXIDE FOR THE

STERILIZATION OF BIOMATERIALS

Introduction

Sterilization of biomaterials is essential for minimizing the risk of developing implant-associated infections. There are numerous methods used for the sterilization of biomaterials, including steam, radiation, and a selection of chemical sterilizing agents, including ethylene oxide. While each of these methods has an established efficacy for sterilizing biomaterials, there are limitations associated with each method, including changes to the material properties (steam; radiation), high costs (radiation), and the presence of chemical deposits on the biomaterials (ethylene oxide).

Gas sterilization using ethylene oxide can result in the presence of residual ethylene oxide, and its secondary products, ethylene chlorohydrin and ethylene glycol, that remain on the material following sterilization. These chemical residues can cause toxic side effects (Stanley et al. 1971; McGunnigle et al. 1975) and hypersensitivity reactions in an in vivo environment (Dolovich et al. 1984; Marshall et al. 1985; Chapman et al. 1986; Dolovich et al. 1987). Research has shown that the level of residual chemicals associated with the surface of exposed materials is dependent on the concentration of ethylene oxide gas used, with exposure to higher gas concentrations resulting in higher levels of contaminating chemical residuals (Lyarskii et al. 1984).

Chlorine dioxide (ClO2), a powerful oxidizing agent with established biocidal

affects, represents a viable alternative for the sterilization of biomaterials. ClO2 has been 65

widely utilized for the decontamination of water since it was shown that it does not form

trihalomethanes (Symons 1976; US Environmental Protection Agency 1977).

Antimicrobial concentrations of ClO2 can be produced at room temperature and

atmospheric pressure (Kowalski 1998). It is highly soluble in water and has been investigated as a liquid decontaminant for reusable medical devices, such as endoscopes

(Ayliffe 2000; Coates 2001). As yet, there are no reports of ClO2 gas having been

employed commercially as a sterilizing agent. There are differing opinions as to the

physiological effects of ClO2, and its derivatives chlorite, chlorate, and chloride ions, with some research indicating that they have low toxicity (Bercz et al. 1982; Lubbers et al. 1982) and are not mutagenic (Kim et al. 1999), while other results suggest that they can be toxic (Moore and Calabrese 1982; Bolognesi et al. 2004; Guzzella et al. 2004) and mutagenic (Zoeteman et al. 1982). However, preliminary assessment of the ClO2 residuals associated with a limited selection of biomaterials suggest that the residuals do not illicit a hypersensitivity response and are not mutagenic (Jeng and Woodworth

1990b).

The sporicidal activity of high concentrations of ClO2 gas has been established

(Jeng and Woodworth 1990a; Han et al. 2003) using ClO2 concentrations between 2324-

9960ppm, and a maximum exposure time of 120 minutes. It is the intent of this current

research to ascertain the optimal environmental parameters (ClO2 concentration, exposure time, and humidity) required to achieve sporicidal activity in Bacillus atrophaeus spores using low concentrations of ClO2 (25-1500ppm) and maximum exposure times

comparable to those used for ethylene oxide sterilization of biomaterials (4-6hrs). 66

Materials and Methods

Chlorine dioxide sterilization system: A chlorine dioxide sterilization system

comprised of a ClO2 generator and a sterilization chamber was constructed and validated

(Figure 3.1). The ClO2 generator was used to produce ClO2 concentrations ranging from

25ppm to 1500ppm with the relative humidity set to ≤ 25% or ≥75%. Chlorine dioxide

was generated based on the chemical reaction between sodium chlorite (NaClO2) (Sigma

Aldrich, Milwaukee, WI) and sodium persulfate (Na2S2O8) (Sigma Aldrich, Milwaukee,

WI) (Equation 3.1). The concentration of ClO2 produced was controlled by the

concentration of the NaClO2 solution and the rate at which the NaClO2 solution was

added to the Na2S2O8. A full standard operating procedure on the chlorine dioxide

generator is provided in the Appendices.

2NaClO2 + Na2S2O8 → 2ClO2 + 2Na2SO4 [Equation 3.1]

To measure the ClO2 concentration produced by the generator, a sample of gas

was taken from the output of the ClO2 generator and bubbled through a solution of 1%

potassium iodide (KI) (Sigma Aldrich, Milwaukee, WI) and glacial acetic acid

(Mallinckrodt Bakers Inc., Phillipsburg, NJ). This solution was titrated with a 0.001N

sodium thiosulfate (Na2S2O3) (Sigma Aldrich, Milwaukee, WI) solution and the concentration of ClO2 being generated was calculated. The standard operating

procedures for measuring the concentration of ClO2 being produced by the generator are

outlined in the Appendices.

The ClO2 produced by the generator formed the input into the sterilization

chamber. The chamber was maintained at atmospheric pressure and room temperature. 67

Figure 3.1: Chlorine dioxide sterilization system, comprised of a ClO2 generator and a

sterilization chamber. The ClO2 generator can produce high- and low-relative humidity

(HRH and LRH) ClO2. The ClO2 output from the generator forms the input into the sterilization chamber (23cm×24.5cm×25.5cm). Bacillus atrophaeus spore strips

(0.5cm×4cm) were placed within the chamber and exposed to ClO2 under select environmental conditions to determine the optimal environmental conditions required to achieve sterilization with each concentrations of ClO2 tested. 68

ClO2 Measurement of Humidity output [ClO2] (≥100ppm) controls

Measurement of [ClO2] (<100ppm)

NaClO2 solution

Na2S2O3 solution

ClO2 Chamber input output

Bacillus atrophaeus spore strips

Fan Sample ports 69

Mixing of the gas in the chamber was achieved using a fan attached to a magnetic stir

bar, placed in the bottom of the chamber near the ClO2 inflow. The chamber was placed

on a magnetic stir plate, allowing for continual mixing of the gas in the chamber without

the presence of electronic components.

Verification of the ClO2 sterilization system: The actual flow rate produced by the

digital flow pump, which was responsible for controlling the rate of addition of NaClO2 to Na2S2O8, was confirmed. The pump was programmed for different flow rates for set

periods of time and the actual flow rates calculated (n = 3 measurements). The accuracy

of the digital hygrometer was tested using a wet and dry bulb thermometer as a standard

(n = 3 measurements). Both the digital hygrometer and the wet and dry bulb

thermometer were exposed to low and high relative humidity conditions and the readings tabulated. Prior to testing the sporicidal properties of the different ClO2 concentrations,

the concentration of gas in the chamber was measured to ensure that the gas was at

equilibrium in the chamber. Samples of ClO2 gas at concentrations of 100ppm, 500ppm,

and 1500ppm were taken from three different positions in the chamber using the septum-

sealed sample ports on the side of the chamber (n = 3 measurements). The

concentrations measured within the chamber were compared with the ClO2 concentration

measured at the chamber output. During the sporicidal experiments, the ClO2 concentration and the humidity in the chamber were monitored by regularly testing the

ClO2 gas from the output of the chamber.

Biological indicator: Bacillus atrophaeus spore strips (formerly Bacillus subtilis; derived from ATCC #9372) in Tyvek pouches were purchased from Raven Biological 70

Laboratories (Omaha NE). Each strip had an average population of 2.0 × 106 colony forming units (CFUs). All spore strips used in these experiments came from a single lot

(Lot 1162312). Spore strips were either kept dry or were humidified prior to ClO2 exposure. To humidify the spore strips, they were sealed in a glass jar with wet gauze and maintained at room temperature for 24 hours.

Assessing the sporicidal activity of ClO2: Chlorine dioxide concentrations ranging

from 25ppm to 1500ppm were produced, with relative humidity levels set to either ≤ 25%

or ≥ 75%. Dry and humidified Bacillus atrophaeus spore strips (n = 6) were placed in the

sterilization chamber and exposed to a set concentration of ClO2 gas for time points

ranging from 0.5hour to 6 hours. Both high and low relative humidity conditions

(environmental and spore strips) were tested to assess the necessity of humidity to

achieve sporicidal activity using ClO2. After being exposed to ClO2, the samples were divided into two groups for analysis.

Indicator broth analysis: Using aseptic techniques, samples were removed from the

Tyvek pouch and placed in an indicator broth of tryptic soy broth (TSB) modified with

Bromocresol purple (TSB-BP) (Raven Biological Laboratories, Omaha NE) and incubated at 37°C for 14 days. The TSB-BP was assessed for bacterial growth over the

14-day incubation period. Positive and negative controls of the indicator broth and

Bacillus atrophaeus spore strips were also assessed.

Viability counts analysis: Under aseptic conditions, each sample was cut into small

strips, placed in 5ml of sterile deionized water, and blended using a tissue homogenizer.

The pulp was collected and the homogenizer was rinsed with 5ml sterile deionized water 71

(×2), and the water added to the pulp (total volume = 15ml). The pulp mix was heat shocked at 80-85°C for 10 minutes and then cooled for 10 minutes in 4°C water. The sample was vortexed for 20 seconds and serial dilutions were made. The original solution and the dilutions were plated on tryptic soy agar (TSA) plates in triplicate and incubated overnight at 37°C. Counts of colony forming units (CFUs) were made and the total number of viable bacteria was determined for each spore strip. Positive control samples of the Bacillus atrophaeus spore strips were analyzed to determine the recovery efficiency of the bacteria from the strips. All CFU counts were normalized to the CFU counts for the positive control spore strips.

Statistics: Data is expressed as the mean ± standard error mean. 72

Results

ClO2 sterilization system verification: The flow rate from the digital flow meter used

to set the rate at which NaClO2 was added to Na2S2O8 was determined. The calculated

flow rate was between 104-110% of the set digital flow rate (Table 3.1). The accuracy of

the digital hygrometer was assessed using a wet and dry bulb thermometer. The digital

hygrometer readings were between 74-110% of the wet and dry bulb thermometer

readings (Table 3.2).

ClO2 production: The gas samples obtained from each of the three sample ports in the

sterilization chamber (Figure 3.1) exhibited ClO2 concentrations that were 91-98% of the

ClO2 concentrations measured from the generator output and 100-103% of the ClO2 concentrations measured at the chamber output (Table 3.3), indicating that the ClO2 concentration within the chamber was at equilibrium. Table 3.4 summarizes the measured experimental conditions utilized in assessing the sporicidal activity of ClO2.

The ClO2 concentrations produced by the generator and the measured environmental

parameters from the chamber output (ClO2 concentration, humidity), relative to the

selected ClO2 set points, were monitored. ClO2 concentrations produced by the generator

were 99-108% of the ClO2 set point concentrations. ClO2 concentrations measured at the

chamber output were 90-104% of the concentrations measured from the generator sampling port. Measurements of the relative humidity of the ClO2 gas showed that low relative humidity levels were between 20-24%, and high relative humidity levels were between 80-87%. 73

Flow Meter Setting (µl/min) Calculated Flow Rate (µl/min) 10.1 10.5±0.23 20.3 21.1±0.35 30 32.6±0.34 40.2 44.4±0.32

Table 3.1: Calculated flow rate for the digital flow meter compared to the digital setting. 74

Relative Humidity Setting Digital Hygrometer Dry/Wet Bulb Thermometer Ambient Humidity 21±0.41% 25% High Humidity (100%) 100% 100% Chamber setting: ≤ 25% 23±1.47% 31% Chamber setting: ≥ 75% 87±0.71% 79%

Table 3.2: Verification of the accuracy of a digital hygrometer for the measurement of relative humidity using a dry and wet bulb thermometer. 75

[ClO2] Setpoint [ClO2] output from [ClO2] output from Average [ClO2] from (ppm) generator chamber sample ports (ppm) (ppm) (ppm) 100 101±0.9 92±1.8 92±1.2 500 488±9 463±6.8 476±6.6 1500 1494±20.5 1353±21.4 1354±12.3

Table 3.3: Chlorine dioxide concentration measured from the generator and chamber outputs, and compared with the chlorine dioxide concentration measured within the chamber via three sample ports. 76

[ClO2] Setpoint [ClO2] output from generator [ClO2] output from chamber (ppm) (ppm) (ppm) 25 27±0.5 23±0.9 50 53±1.1 46±1.4 100 107±2.3 104±2.0 200 211±1.3 200±4.4 500 510±2.7 465±6.2 1000 992±6.2 897±11 1500 1500±6.5 1365±14

Table 3.4: Chlorine dioxide concentrations measured from the generator and chamber outputs during experiments. For low relative humidity experiments (≤25%) the relative humidity of the chlorine dioxide gas was measured to be between 20-24%. For high relative humidity experiments (≥75%) the relative humidity of the chlorine dioxide gas was measured to be between 80-87%.

Sporicidal activity of high relative humidity (HRH) ClO2 (RH ≥ 75%): Dry and humidified Bacillus atrophaeus spores were exposed to HRH ClO2. The indicator broth

assay showed increasing sporicidal activity with increased ClO2 concentration and/or increased exposure time for both dry and humidified spores (Table 3.5). Dry spores remained viable at concentration/exposure time combinations that caused sterilization of

humidified spore strips, suggesting that moisture is important for sporicidal activity.

Results from the viability count assay indicate an increase in sporicidal activity with

increased exposure time for both dry and humidified spores (Figure 3.2). Figure 3.2 also

shows that as concentration increases, the rate of sporicidal activity increases. The

steeper gradients seen in the Figure 3.2B, relative to Figure 3.2A, also indicate a decrease

in the time required to achieve sporicidal activity for humidified spores, relative to dry

spores. Based upon these results, a range of sterilization conditions for dry and

humidified spore strips with HRH ClO2 have been identified (Table 3.9A).

Sporicidal activity of low relative humidity (LRH) ClO2 (RH ≤ 25%): Dry and

humidified Bacillus atrophaeus spores were exposed to LRH ClO2. Since research by

other investigators has established that moisture is important for successfully achieving

sporicidal activity, the conditions of sterilization tested with LRH ClO2 began with the

exposure times required to achieve sterilization with HRH ClO2 for each concentration of

ClO2 tested. When sterilization was achieved, or a 6 hour exposure time was reached,

whichever occurred first, no further testing at that ClO2 concentration was done. Results

from the indicator broth assay indicate an increase in sporicidal activity with increased 78

(A) Time 0.5 1 2 4 6 ClO (hrs) 2 (ppm) % Sterile Samples (Post-ClO2 Exposure) 25 0 0 0 0 0 50 0 0 0 0 75 100 0 0 0 75 100 200 0 0 25 100 - 500 0 25 75 100 - 1000 0 100 - - - 1500 0 100 - - -

(B) Time 0.5 1 2 4 6 ClO (hrs) 2 (ppm) % Sterile Samples (Post-ClO2 Exposure) 25 0 0 0 0 100 50 0 0 0 100 - 100 0 25 75 100 - 200 50 100 - - - 500 100 - - - - 1000 - - - - - 1500 - - - - -

Table 3.5: Indicator broth assay for experiments with high relative humidity chlorine dioxide and: (A) Dry Bacillus atrophaeus spores; (B) Humidified Bacillus atrophaeus spores.

- indicates the experimental condition was not tested 79

Figure 3.2: Viability counts assay for experiments with high relative humidity chlorine dioxide and: (A) Dry Bacillus atrophaeus spores; (B) Humidified Bacillus atrophaeus

spores

(A) 120 25ppm 100 50ppm 100ppm 200ppm 80 500ppm 1000ppm 1500ppm 60

40 % Viable Spores 20

01234567

ClO2 Exposure Time (hrs)

(B) 120

25ppm 100 50ppm 100ppm 200ppm 80 500ppm

% Viable Spores 20

01234567

ClO2 Exposure Time (hrs) 81

ClO2 concentration and/or increased exposure time for humidified spores (Table 3.6).

Sterilization of humidified spores with LRH ClO2 required higher ClO2 concentrations

and longer exposure times, relative to sterilization of humidified spores with HRH ClO2.

Dry spore strips could not be sterilized with LRH ClO2 for the concentration/exposure

time combinations assessed, again suggesting that moisture enhances sporicidal activity.

Results from the viability counts assay indicated an increased rate in sporicidal activity

with increased exposure time for both dry and humidified spores (Table 3.7). A range of

sterilization conditions for humidified spore strips with LRH ClO2 have been established

(Table 3.9B).

Detection of spore viability with indicator broth assay: There were instances where

the spore strips had to be incubated for several days before the indicator broth assay

could detect viable spores (Table 3.8). For experimental conditions where higher levels

of humidity were utilized (humidified spores and HRH ClO2), an increase in the time

required to detect viable spores was often observed, particularly as ClO2 concentration

and exposure time increased.

(A) Time 0.5 1 2 4 6 ClO (hrs) 2 (ppm) % Sterile Samples (Post-ClO2 Exposure) 100 - - - - 0 200 - - - 0 0 500 - - - 0 0 1000 - 0 0 0 0 1500 0 0 0 25* 0

(B) Time 0.5 1 2 4 6 ClO (hrs) 2 (ppm) % Sterile Samples (Post-ClO2 Exposure) 25 - - - - 0 50 - - - 0 0 100 - - - 75 100 200 - 75 100 - - 500 100 - - - -

Table 3.6: Indicator broth assay for experiments with low relative humidity chlorine dioxide and: (A) Dry Bacillus atrophaeus spores; (B) Humidified Bacillus atrophaeus

spores. Dry Bacillus atrophaeus spores could not be sterilized with low relative humidity

chlorine dioxide gas, using the selected chlorine dioxide concentrations and exposure

times.

- indicates the experimental condition was not tested

* No viable spores detected on one dry spore strip 83

(A) Time 0.5 1 2 4 6 ClO (hrs) 2 (ppm) % Viable Spores (Post-ClO2 Exposure) 25 - - - - - 50 - - - - - 100 - - - - <0.04 200 - - - <0.04 <0.04 500 - - - <0.04 0* 1000 - <0.04 <0.04 <0.04 <0.04 1500 - <0.04 <0.04 <0.04 <0.04

(B) Time 0.5 1 2 4 6 ClO (hrs) 2 (ppm) % Viable Spores (Post-ClO2 Exposure) 25 - - - - <0.04 50 - - - <0.04 0 100 - - - 0* 0 200 - 0* 0 - - 500 0 - - - - 1000 - - - - - 1500 - - - - -

Table 3.7: Viability counts assay for experiments with low relative humidity chlorine dioxide and: (A) Dry Bacillus atrophaeus spores; (B) Humidified Bacillus atrophaeus

spores.

- indicates the experimental condition was not tested

* No bacteria were detected using viability counts assay, although results from the indicator broth assay indicated that viable spores were present. 84

(A) Time 0.5 1 2 4 6

ClO (hrs) 2 Days required to indicate presence of viable spores (D: Dry; H: Humidified) (ppm) D H D H D H D H D H 25 1 1 1 1 2 2 2 2 4 S 50 1 2 2 3 2 3 8 S 3 - 100 2 2 3 4 4 2 2 S S - 200 3 6 2 S 5 - S - - - 500 2 S 2 - 2 - - - - - 1000 2 - S 2 ------1500 2 - S 2 ------

(B) Time 0.5 1 2 4 6

ClO (hrs) 2 Days required to indicate presence of viable spores (D: Dry; H: Humidified) (ppm) D H D H D H D H D H 25 ------2 50 ------2 - 2 100 ------3 2 S 200 - - - 2 - S 2 - 2 - 500 - S - - - - 2 - 2 - 1000 - - 2 - 2 - 2 - 3 - 1500 - - 2 - 2 - 2 - 3 -

Table 3.8: The delay in detecting viable spores using the indicator broth assay for dry spores and humidified spores after exposure to: (A) High relative humidity chlorine dioxide; (B) Low relative humidity chlorine dioxide.

S indicates that sterilization of the spore strips had occurred (i.e. no change in the indicator broth was observed for any of the samples)

- indicates the experimental condition was not tested 85

(A) [ClO2] Dry Spores: Humidified Spores: (ppm) Sterilization Exposure Time Sterilization Exposure Time (hours) (hours) 25 > 6 6 50 > 6 4 100 6 4 200 4 1 500 4 0.5 1000 1 < 0.5 1500 1 < 0.5

(B) [ClO2] Dry Spores: Humidified Spores: (ppm) Sterilization Exposure Time Sterilization Exposure Time (hours) (hours) 25 > 6 > 6 50 > 6 > 6 100 > 6 6 200 > 6 2 500 > 6 0.5 1000 > 6 < 0.5 1500 > 6 < 0.5

Table 3.9: Summary of environmental parameters required for ClO2 sterilization of: (A)

Dry and humidified Bacillus atrophaeus spores with high relative humidity chlorine dioxide; (B) Dry and humidified Bacillus atrophaeus spores with low relative humidity chlorine dioxide. Dry Bacillus atrophaeus spores could not be sterilized with low relative

humidity chlorine dioxide at the concentrations and exposures times tested (25-1500ppm,

0.5-6 hours). 86

Discussion

A chlorine dioxide sterilization system, comprising a ClO2 generator and

sterilization chamber, was designed and used to generate low concentrations of ClO2. A series of experiments were conducted to determine the sterilization conditions required to achieve sterilization of Bacillus atrophaeus spore strips under both low and high relative humidity conditions.

Dry and humidified spores were exposed to the selected environmental conditions. Humidification of the spores prior to being exposed to ClO2 allowed for

sterilization of spore strips at a concentration of 25ppm in 6 hours with high relative

humidity (≥ 75%) ClO2. Increasing the concentration of HRH ClO2 reduced the exposure

time required, with spore strips successfully sterilized in 30 minutes at a ClO2 concentration of 500ppm. Sterilization of dry spore strips could be achieved with HRH

ClO2 at a concentration of 100ppm and an exposure time of 6 hours. Increasing the ClO2 concentration to 1000ppm reduced the exposure time required for sterilization to 1 hour.

The sterilization conditions for dry spore strips with HRH ClO2 consistently required

higher ClO2 concentrations and longer exposure times than were required for humidified spore strips. Since previous research has determined that sporicidal activity is dependent

on humidification of the spores (Gilbert et al. 1964; Dadd et al. 1985), it is likely that

exposure to the high-humidity environment during the sterilization process caused sufficient humidification of the dry spores, such that sterilization could be achieved.

Using ClO2 in a low-relative humidity (<25%) environment resulted in an

increase in the ClO2 concentration and exposure times required to achieve sterilization of 87

humidified spore strips. For example, the lowest ClO2 concentration resulting in

sterilization with LRH ClO2 was 100ppm with an exposure time of 6 hours, as compared

to 25ppm with HRH ClO2 and the same exposure time. The difference in the

environmental conditions necessary to achieve sterilization of humidified spore strips

with LRH ClO2, is likely due to the fact that the low humidity of the sterilization

environment causes the spores to become desiccated, such that the lower concentrations

of LRH ClO2 (25-100ppm) cannot achieve sterilization of the spore strips. Sterilization

of dry spore strips with LRH ClO2 could not be achieved for the tested ClO2 concentrations (25-1500ppm with 0.5-6 hour exposure times).

While the specific mechanisms of sporicidal activity of ClO2 have not yet been

established, research has suggested that ClO2 causes the degradation of tyrosine (Noss et

al. 1986), an amino acid that is a major component of both the outer coat and membrane

of spores (Goldman and Tipper 1978). Damage to the membrane of spores has

previously been cited as a potential mechanism of the sporicidal activity of ClO2 (Young and Setlow 2003). Elevated relative humidity has been determined as an essential factor for achieving sporicidal activity (Gilbert et al. 1964; Dadd et al. 1985) with speculation that humidification causes increased susceptibility of spores to damage by chemical agents, although how moisture contributes to the effectiveness of the chemicals has yet to be ascertained. Since humidification is essential for germination, the process whereby spores become vegetative cells, which are more susceptible to the antimicrobial effects of chemical agents, one theory is that the elevated humidity may initiate the germination process. However, the presence of nutrients (amino acids, sugars or purine nucleosides) 88

is required for germination to occur, so even when spores become humidified they are not

necessarily actively undergoing germination. An alternative theory is that humidification

of the outer coat of the spore may cause a localized increase in the concentration of ClO2 within the spore, causing degradation of the outer coat and increased exposure of the spore membrane to the oxidizing effects of ClO2. Alternatively, the presence of moisture

may be required for the oxidative action of ClO2 against protein residues to occur.

Gilbert et al. (1964) suggested that reduced humidity may cause an absence of water

molecular bridges in proteins, removing possible alkylation sites for ethylene oxide, and

inhibiting the sporicidal activity. Jeng and Woodworth (1990a) suggest that absorption

of water by the humidified spore strips causes localized areas of moisture in the samples.

Since ClO2 is highly soluble in water, these localized areas of moisture may lead to

concentrated areas of ClO2, and may thus contribute to the greater efficiency in sporicidal

activity seen with humidified samples.

Interestingly, the viability count analysis of the experiments conducted with dry

spores found that at the concentration and exposure times that caused sterilization with

HRH ClO2, the percentage of viable spores remaining on the spore strips for LRH ClO2 was very low, <0.04%. This suggests that while elevated humidity is important in achieving sterilization of spores, it is not essential for ClO2 gas to cause sporicidal

activity. This hypothesis may be also supported by the evidence that one dry spore strip

was sterilized using LRH ClO2 (1500ppm with a 4 hour exposure time) (Table 3.6A).

Some of the samples from the indicator broth assay exhibited a delayed response,

whereby the detection of viable bacteria required the incubation of the spore samples in 89

the TSB-BP indicator solution for several days. An increase in the delayed response was

observed as ClO2 concentration and/or exposure time increased. It is likely that there

were too few viable spores initially present for the indicator broth to detect. While the

indicator broth is sensitive to viable spores, a threshold number of metabolically active

cells need to be present in the indicator broth for it to indicate that viable cells are

present. Germination of the spores, followed by proliferation of vegetative Bacillus

atrophaeus, may have increased the number of viable cells to a detectable level.

There was variability between the results from the indicator broth and viability

counts assays for the conditions that resulted in sterilization. The indicator broth assay is

more sensitive to low numbers of viable spores than the viability counts assay, allowing

for a more accurate measurement of sterilization. The viability counts assay provides

data concerning the rate of sporicidal activity for the different chlorine dioxide gas concentrations and exposure times, rather than an accurate assessment of conditions required to achieve sterilization. As such, the conditions for which sterilization (i.e. no

detectable viable spores) is said to have occurred are based on the results from the

indicator broth assay analysis.

While sterilization of Bacillus atrophaeus spore strips with ClO2 has been

demonstrated by other investigators (Jeng and Woodworth 1990a; Han et al. 2003), these

studies were conducted using high concentrations of ClO2 (2324-9960ppm) and short exposure times. Previous research has indicated that a contributing factor to the presence of chemical residues on the surface of biomaterials post-sterilization is using high concentrations of the chemical agent for the sterilization procedure (Lyarskii et al. 1984). 90

While some research has suggested that the decomposition products associated with

chlorine dioxide are not harmful (Bercz et al. 1982; Lubbers et al. 1982), these products

have not been fully tested in association with biomaterials. It is the intent of the research

presented here to assess the environmental conditions required to achieve sterilization

using low concentrations of ClO2 gas and exposure times no longer than the current

standards for sterilization with ethylene oxide. The results from these experiments

suggest that sterilization of biomaterials could potentially be achieved using ClO2 at

lower concentrations and with shorter exposure times than those currently required for

ethylene oxide sterilization. Lower concentrations of ClO2 will reduce the risk of

residuals remaining on the sterilized material and decrease high concentration exposure

for users. The results from these experiments show that low concentrations of ClO2 are

an effective sterilizing agent under several different environmental conditions (humidity,

ambient pressure and temperature), allowing for versatility in the materials that can be sterilized. 91

4. AN ASSESSMENT OF THE CYTOTOXICITY OF POLYMERS STERILIZED

WITH CHLORINE DIOXIDE

Introduction

Sterilization with chemical agents has expanded the variety of materials that can

be utilized in medical devices. This is particularly true for the selection of polymers

utilized, since many are sensitive to the effects of sterilization with heat and radiation.

However, polymers sterilized with chemicals frequently absorb and retain the

sterilization agent. For example, residual ethylene oxide and its reaction products,

ethylene chlorohydrin and ethylene glycol, can contaminate materials after sterilization with ethylene oxide gas. These chemical residues can cause toxic side effects (Stanley et al. 1971; McGunnigle et al. 1975) and hypersensitivity reactions in vivo (Dolovich et al.

1984; Marshall et al. 1985; Chapman et al. 1986; Dolovich et al. 1987), and have even been linked to mortality in young children (Stanley et al. 1971). Post-sterilization processing, such as aeration at elevated temperatures to increase the rate of chemical desorption from the polymers, have been employed to limit the presence of chemical residues associated with sterilized materials.

In Chapter 3, chlorine dioxide (ClO2) gas was investigated for its potential as a

new method of chemical sterilization for biomaterials. Research has shown that

sterilization with high concentrations of chemical agents increases the levels of chemical

residues associated with the sterilized material (Lyarskii et al. 1984). While the potential

for utilizing high concentrations of ClO2 as a sterilizing agent has been established (Jeng

and Woodworth 1990a; Han et al. 2003), the studies in Chapter 3 were designed to 92

ascertain the optimal environmental parameters for sterilization with low concentrations

of ClO2 (≤1500ppm; c.f. sterilization with ethylene oxide requires concentrations

between 390,000-560,000ppm). Several different combinations of environmental

parameters were determined to be sufficient for the sterilization of biomaterials with ClO2 gas. Of the environmental conditions that achieved sterilization, several combinations of concentration, exposure time, and humidity were selected for the following experiments.

The environmental conditions included both higher ClO2 concentrations with short exposure times, and lower ClO2 concentrations with longer exposure times, at both low

and high relative humidity. The following studies were designed to assess whether the

sterilization of polymers with different ClO2 sterilization conditions results in cytotoxic

concentrations of chemical residuals, and to determine if post-sterilization processing of

ClO2-sterilized polymers is required. 93

Materials & Methods

These studies, designed to assess the effects of sterilization with low

concentrations of ClO2 on polymers, were based on the in vitro cytotoxicity tests outlined in ISO 10993-5:1999. These experiments used chemical residues extracted from the materials as an assessment of cytotoxicity.

Sterilization conditions: From the sterilization conditions determined for humidified spores in Chapter 3, a low concentration/long exposure time, and a high concentration/short exposure time was selected for both high relative humidity and low relative humidity ClO2. The combinations of ClO2 sterilization conditions and pre-

sterilization sample preparation are outlined in Table 4.1. As a control, one group of the

material samples was sterilized using ethylene oxide (UMC, Tucson AZ).

Materials: Polyvinyl chloride (PVC) (Nalge Nunc International Corp., Rochester, NY)

and polyurethane (PU) (Nalge Nunc International Corp., Rochester, NY) were selected to be tested for the presence of residuals after sterilization with ClO2. While both materials

are commonly used in the medical field, PVC was specifically selected since it readily

absorbs ethylene oxide during chemical sterilization (Bruch 1973; McGunnigle et al.

1975). The material samples, purchased as polymer tubing, were prepared such that each sample had a volume of 0.4cm3. Organo-tin stabilized PVC (Sn-PVC) (USP, Rockville,

MD) and high-density polyethylene (HDPE) (USP, Rockville, MD) were selected as

positive and negative control materials respectively. The control materials came as

sheets and were prepared such that each sample had a surface area of 0.5cm2. Three samples of each type of material were used for each sterilization condition. 94

Gas Concentration Exposure Time Environmental Sample Preparation (ppm) (hours) Humidity 25 6 >75% Humidified 500 0.5 >75% Humidified 100 6 >75% Dry 1000 1 >75% Dry 100 6 <25% Humidified 500 0.5 <25% Humidified

Table 4.1: Environmental conditions for sterilization with ClO2 used for the assessment of ClO2 sterilization on a selection of polymers 95

Cell line: A human fibroblast cell line, MRC-5, (ATCC: CCL-171) was established

using ATCC complete growth media (Minimum essential medium (Eagle), 2mM L- gluatime, Earle’s BSS adjusted to contain 1.5g/L sodium bicarbonate, 0.1mM non- essential amino acids, 1.0mM sodium pyruvate (90%), FBS (10%)). The maximum number of passages of the cells used for these experiments was 28.

Experimental procedure: Dry or humidified samples were placed in the sterilization chamber and exposed to a set concentration of ClO2 gas for the time period specified.

Humidification of the samples was achieved by sealing the samples in a glass jar with

wet gauze for 24 hours at room temperature. After being exposed to ClO2, the samples

were aerated for 24 hours at room temperature and atmospheric pressure, maintaining

sterility. Each material sample was then placed in 6ml of MRC-5 cell culture medium

and incubated at 37°C in an incubator shaker (Model G25, New Brunswick Scientific

Company Inc., Edison, NJ) for 48 hours. Hereafter, this solution will be referred to as the

extract cell culture medium. Dilutions of the extract cell culture medium (100%, 75%,

50%, 25%) were prepared. MRC-5 cells were passed into 12 well plates (6.4×104 cells/cm2) and cultured for 24 hours at 37°C. The cell media was removed and 2ml of the

extract cell culture medium dilutions added to the cells. The cells were incubated at 37°C

with the extract cell culture medium for 72 hours. Control wells containing only MRC-5

fibroblasts and cell culture medium were prepared for each experiment.

Morphometric assessment of cytotoxicity: After 24, 48 and 72 hours, the cells were

visually assessed for cytotoxic effects. Indicators of cytotoxic effects included cell debris

(indicating cell lysis), detachment, vacuolization, and cell morphology. The level of 96

cytotoxicity was graded using the scale outlined in Table 4.2 and a cytotoxicity value

determined for each combination of polymer sample and sterilization condition, and at

each dilution of the extract cell culture medium. Each culture was visually assessed at

three random locations and a cytotoxicity value assigned to the test sample.

Representative images that correspond to the scale described in Table 4.2 are shown in

Figure 4.1. After 72 hours, an MTT assay was performed

MTT assay: This assay was used to provide a quantitative assessment of cell viability.

The extract cell culture medium was aspirated from each well and the cells were rinsed

with DCF-PBS. The DCF-PBS was removed and 280µl of a 5mg/ml MTT solution was added to each well. The cells were incubated at 37°C for 1-4 hours before the MTT

solution was aspirated from each well. 2ml of DMSO was added to lyse the cells which

were incubated in the DMSO at 37°C for ≥ 5 minutes. The absorbencies of the solutions

were measured (×3) using a Biophotometer (Eppendorf, Hamburg, Germany) at a

wavelength of 562nm. The absorbance values for each of the test conditions were

normalized to the absorbance values calculated for the fibroblasts in the control wells.

Statistics: Values are presented as the mean ± standard error mean. 97

Cytotoxicity Value Interpretation 0 Not cytotoxic: few indications of cell death 1 Mildly cytotoxic: more indicators for viable cells than cell death 2 Moderately cytotoxic: approx. equal indicators for viable cells and cell death 3 Severely cytotoxic: more indicators for cell death

Table 4.2: Scale used for morphometric assessment of the cytotoxicity associated with the extracts from the different sterilization conditions for each of the polymer materials assessed

Figure 4.1: Images depicting the numerical scale of cytotoxicity (outlined in Table 4.2).

For each cytotoxicity value assigned, there was a range in the observed level of cytotoxicity: (A1&A2) Non-cytotoxic (Value: 0); (B&B2) Mildly cytotoxic (Value: 1);

(C&C2) Moderately cytotoxic (Value: 2); (D&D2) Severely cytotoxic (Value: 3)

(A1) (A2)

(B1) (B2)

(C1) Cell debris (C2) Cell debris

(D1) (D2)

100

Results

Morphometric assessment of cytotoxicity: For each time point (24, 48, and 72 hours)

cytotoxicity values were assigned for each sterilization condition and polymer sample

combination, and for each extract cell culture medium dilution tested. An average

cytotoxicity value was calculated. The results are represented in Figure 4.2 - Figure 4.5.

For all of the sterilization conditions tested there was an overall decrease in the level of

observed cytotoxcity as the incubation time of the MRC-5 fibroblasts in the extract cell

culture medium increased. A steady decrease in cytotoxicity values from 24 to 72 hours

was exhibited for all of the sterilization conditions tested on PU and HDPE (negative

control). For most of the sterilization conditions tested on PVC, the extract cell culture medium dilutions of 100% and 75% exhibited a transient increase in cytotoxicity at 48 hours, which decreased by 72 hours. All of the sterilization conditions tested on Sn-PVC

(positive control) exhibited a cytotoxicity value of 3 for the extract cell culture medium dilutions of 100% and 75%. As the incubation time of the MRC-5 fibroblasts with the extract cell culture medium increased, the cytotoxicity value decreased for the lower dilution concentrations (50% and 25%), indicating that reduced cytotoxic effects were observed. Comparing between the different dilutions of extract cell culture medium for all of the polymers and all of the sterilization conditions tested, a reduction in the level of cytotoxicity was detected after a 24 hour incubation time as the extract cell culture medium became more dilute. After an incubation time of 72 hours, sterilization of humidified polymer samples with high relative humidity ClO2 at a concentration of

25ppm for 6 hours had the lowest cytotoxicity values. 101

Figure 4.2: Cytotoxicity values for extract dilutions from polyurethane sterilized using

the following environmental conditions: (A) 25ppm/6hrs (HRH-Humidified samples);

(B) 500ppm/0.5hrs (HRH-Humidified samples); (C) 100ppm/6hrs (HRH-Dry samples);

(D) 1000ppm/1hr (HRH-Dry samples); (E) 100ppm/6hrs (LRH-Humidified samples); (F)

500ppm/0.5hrs (LRH-Humidified samples); (G) Ethylene oxide. Legend indicates dilution of extract cell culture medium. 102

(A) 100% (B) 100% 3 75% 3 75% 50% 50% 25% 25%

2 2 Cytotoxicity ValueCytotoxicity ValueCytotoxicity 1 1

0 0 24 48 72 24 48 72 Time (hours) Time (hours)

2 2 Cytotoxicity Value Cytotoxicity Value 1 1

0 0 24 48 72 24 48 72 Time (hours) Time (hours)

(E) 100% (F) 100% 3 75% 3 75% 50% 50% 25% 25%

2 2 Cytotoxicity ValueCytotoxicity ValueCytotoxicity 1 1

0 0 24 48 72 24 48 72 Time (hours) Time (hours)

3 (G) 100% 75% 50% 25%

1 Cytotoxicity ValueCytotoxicity

0 24 48 72 Time (hours) 103

Figure 4.3: Cytotoxicity values for extract dilutions for polyvinyl chloride sterilized using

the following environmental conditions: (A) 25ppm/6hrs (HRH-Humidified samples);

(B) 500ppm/0.5hrs (HRH-Humidified samples); (C) 100ppm/6hrs (HRH-Dry samples);

(D) 1000ppm/1hr (HRH-Dry samples); (E) 100ppm/6hrs (LRH-Humidified samples); (F)

500ppm/0.5hrs (LRH-Humidified samples); (G) Ethylene oxide. Legend indicates dilution of extract cell culture medium.

104

(A) 100% (B) 100% 3 75% 3 75% 50% 50% 25% 25%

2 2 Cytotoxicity ValueCytotoxicity ValueCytotoxicity 1 1

0 0 24 48 72 24 48 72 Time (hours) Time (hours)

2 2 Cytotoxicity ValueCytotoxicity ValueCytotoxicity 1 1

0 0 24 48 72 24 48 72 Time (hours) Time (hours)

(E) 100% (F) 100% 3 75% 3 75% 50% 50% 25% 25%

2 2 Cytotoxicity ValueCytotoxicity ValueCytotoxicity 1 1

0 0 24 48 72 24 48 72 Time (hours) Time (hours)

(G) 100% 3 75% 50% 25%

2 Cytotoxicity Value 1

0 24 48 72 Time (hours) 105

Figure 4.4: Cytotoxicity values for extract dilutions for high density polyethylene sterilized using the following environmental conditions: (A) 25ppm/6hrs (HRH-

Humidified samples); (B) 500ppm/0.5hrs (HRH-Humidified samples); (C) 100ppm/6hrs

(HRH-Dry samples); (D) 1000ppm/1hr (HRH-Dry samples); (E) 100ppm/6hrs (LRH-

Humidified samples); (F) 500ppm/0.5hrs (LRH-Humidified samples); (G) Ethylene oxide. Legend indicates dilution of extract cell culture medium.

106

(A) 100% (B) 100% 3 75% 3 75% 50% 50% 25% 25%

2 2 Cytotoxicity ValueCytotoxicity ValueCytotoxicity 1 1

0 0 24 48 72 24 48 72 Time (hours) Time (hours)

2 2 Cytotoxicity Value Cytotoxicity Value 1 1

0 0 24 48 72 24 48 72 Time (hours) Time (hours)

(E) 100% (F) 100% 3 75% 3 75% 50% 50% 25% 25%

2 2 Cytotoxicity Value Cytotoxicity Value 1 1

0 0 24 48 72 24 48 72 Time (hours) Time (hours)

(G) 100% 3 75% 50% 25%

2 Cytotoxicity ValueCytotoxicity 1

0 24 48 72 Time (hours) 107

Figure 4.5: Cytotoxicity values for extract dilutions for organo-tin stabilized polyvinyl chloride sterilized using the following environmental conditions: (A) 25ppm/6hrs (HRH-

Humidified samples); (B) 500ppm/0.5hrs (HRH-Humidified samples); (C) 100ppm/6hrs

(HRH-Dry samples); (D) 1000ppm/1hr (HRH-Dry samples); (E) 100ppm/6hrs (LRH-

Humidified samples); (F) 500ppm/0.5hrs (LRH-Humidified samples); (G) Ethylene

oxide. Legend indicates dilution of extract cell culture medium.

108

(A) 100% (B) 100% 3 75% 3 75% 50% 50% 25% 25%

2 2 Cytotoxicity ValueCytotoxicity ValueCytotoxicity 1 1

0 0 24 48 72 24 48 72 Time (hours) Time (hours)

2 2 Cytotoxicity ValueCytotoxicity ValueCytotoxicity 1 1

0 0 24 48 72 24 48 72 Time (hours) Time (hours)

(E) 100% (F) 100% 3 75% 3 75% 50% 50% 25% 25%

2 2 Cytotoxicity ValueCytotoxicity ValueCytotoxicity 1 1

0 0 24 48 72 24 48 72 Time (hours) Time (hours)

(G) 100% 3 75% 50% 25%

2 Cytotoxicity ValueCytotoxicity 1

0 24 48 72 Time (hours) 109

MTT assay: After 72 hours of incubation with the extract cell culture medium, an MTT assay was performed on the MRC-5 fibroblasts. The absorbance values measured for each sterilization condition was normalized to the absorbance value calculated for the control cells (MRC-5 fibroblasts with cell culture medium only) for that experiment. The values calculated for each of the extract cell culture medium dilutions are represented in

Figure 4.6. The results indicate that the residuals associated with the ClO2 sterilized

polymers caused a greater reduction in cell viability, relative to the residuals associated

with the samples sterilized using ethylene oxide. Of the different ClO2 sterilization

conditions tested, sterilization of humidified PU and PVC polymer samples with high

relative humidity ClO2 at a concentration of 500ppm for 0.5 an hour usually resulted in

the highest cell viability. However, sterilization of HDPE samples with ethylene oxide

resulted in maximum cell metabolic activity, although of the ClO2 sterilization conditions

tested, sterilizing humidified samples with high relative humidity ClO2 at a concentration of 500ppm for 30 minutes resulted in maximum cell viability. 110

Figure 4.6: Normalized absorbance measurements from the MTT assay for each

sterilization condition tested at the following extract cell culture medium dilutions:

(A) 100%; (B) 75%; (C) 50%; (D) 25%. Legend indicates sterilization conditions. 111

3 3 (A) (B)

2 2 562nm 562nm

Absorbance 1 Absorbance 1

0 0 PU PVC HDPE Sn-PVC PU PVC HDPE Sn-PVC Polymers Polymers

3 3 (C) (D)

2 2

562nm 562nm

Absorbance 1 Absorbance 1

0 0 PU PVC HDPE Sn-PVC PU PVC HDPE Sn-PVC Polymers Polymers

EtO 25ppm/6hrs (HRH-H) 500ppm/0.5hrs (HRH-H) 100ppm/6hrs (HRH-D) 1000ppm/1hr (HRH-D) 100ppm/6hrs (LRH-H) 500ppm/0.5hrs (LRH-H)

112

Discussion

With the sterilization of polymers with chemical agents comes the concern of

absorption of the sterilizing agent by the material, the presence of secondary reaction

products, and whether the concentrations of these residuals are cytotoxic. Utilization of

high concentration of chemical agents for sterilization increases the presence of these

residuals (Lyarskii et al. 1984). In Chapter 3, several combinations of environmental

parameters that allowed for sterilization with low concentrations of chlorine dioxide

(ClO2) gas were determined. The experiments in this chapter were designed to assess the

effects of concentration, exposure time, and humidity on the cytotoxicity of chemical

residuals associated with polymers after being sterilized with ClO2 gas. Polyurethane

(PU) and polyvinyl chloride (PVC) were selected as test materials since these polymers

are commonly utilized as biomaterials. PVC was also selected on the basis that it readily

absorbs and retains ethylene oxide gas during sterilization (Bruch 1973; McGunnigle et

al. 1975). High-density polyethylene (HDPE) was selected as a negative control material

for cytotoxicity and organo-tin stabilized PVC (Sn-PVC) was selected as a positive

control material. Several combinations of concentration, exposure time, and humidity for

ClO2 sterilization were tested, and ethylene oxide sterilization was used as a control

sterilization process.

The results from the morphometric assessment of cytotoxicity, based on cell

morphology, indicate that for all of the polymers tested, sterilization of humidified

polymer samples with high relative humidity ClO2 at a concentration of 25ppm for 6 hours results in the lowest levels of observed cytotoxicity. This corresponds with the 113

observations of Lyarskii et al. (1984) that sterilization with low concentrations of

chemical sterilization agents reduces the presence of chemical residues associated with

the materials. By morphometric assessment alone, sterilization of polymers with

ethylene oxide consistently resulted in high levels of cytotoxicity, relative to the ClO2 sterilized polymers. This result is contradictory to the results observed with the MTT assay, which indicate that more metabolically active cells are associated with the extracts from humidified PU and PVC polymer samples sterilized with high relative humidity

ClO2 at a concentration of 500ppm for 0.5 an hour. One explanation for the differences

observed between the results from the two assays is that the cytotoxicity values assigned

during the morphometric assessment incorporates evidence of cell death as well as the

presence of viable cells. If indications of cell death remain prevalent after 72 hours, even

if the number of viable cells had increased, then the assigned cytotoxicity value reflects

this fact. The MTT assay only provides an indication of cell viability, and not a measure

of the cell death that may have occurred in the prior 72 hours. Also, the MTT assay is

only an indirect measurement of cell viability. The assay actual measures mitochondrial

activity in cell, considered to be proportional to cell viability. However, research has

indicated that exposure of pulmonary epithelial cells to ethylene oxide causes increased

the mitochondrial activity (Seguy et al. 1994). The increased activity would skew the

results of the MTT assay to suggest that there were more viable cells present. The

difference observed in the results from these two assays could be overcome by

performing a quantitative assay, such as a Live/Dead assay at 24, 48 and 72 hours, which

would account for both viable and non-viable cells. 114

The results from the MTT assay indicate that, of the ClO2 sterilization conditions

tested, sterilization of humidified PU and PVC samples with high relative humidity ClO2 at a concentration of 500ppm for 0.5 an hour results in the greatest cell metabolic activity. However, when sterilizing HDPE the greatest cell viability was observed with ethylene oxide, although of the ClO2 sterilization conditions, humidified HDPE samples

with high relative humidity ClO2 at a concentration of 500ppm for 0.5 an hour results in

the greatest cell metabolic activity. One explanation for this aberrance concerns the

material properties of the different polymers and the potential influence of these

properties on the residuals that remain associated with the material. If more ethylene oxide residuals were associated with the HDPE, and this caused the aforementioned increase in the mitochondrial activity, then this would influence the results observed with the MTT assay.

Interestingly, sterilization of humidified polymer samples with low relative humidity ClO2 at a concentration of 500ppm for 0.5 an hour resulted in a reduction in cell

viability as compared with sterilization with high relative humidity ClO2 with the same

concentration and exposure time. The high solubility of ClO2 in water may account for

this apparent discrepency. ClO2 is highly soluble in water, and as such the water molecules associated with the humidified polymer samples may have readily absorbed the ClO2 from the low relative humidity atmosphere. The diffusion of ClO2 from a high relative humidity environment into the water molecules associated with the humidified polymers may have been slower since the ClO2 was already in a high humidity

environment. 115

The results from both methods of analysis indicate the effects of diluting the

extract cell culture medium on the cytotoxicity observed for all of the polymers and the

sterilization methods tested. The reduced cytotoxicity observed at the lower dilutions

emphasizes the importance of the initial volume of cell culture medium used to perform

the extraction of the chemical residuals. A survey of the literature indicated that the

material surface area/extraction fluid volume ratio used for chemical residual extracts is

on the order of 0.1-5cm2/ml (Oshima and Nakamura 1994; Bumgardner et al. 2002;

Pariente et al. 2002; Cao et al. 2005). While the size of the PU and PVC polymer

samples used for these experiments were calculated to have equal volumes (0.4cm3), the range of surface area/fluid volume was 1.6-3.1cm2/ml, which is within the range of the

aforementioned published conditions. The cytotoxicity data from the diluted extract cell

culture medium is perhaps most significant when considering the volume of fluid that the

polymers may be exposed to in an in vivo environment. Chemical residuals associated

with polymers that are exposed to high volumes of fluid would be significantly diluted,

and as such the importance of the cytotoxicity observed at less dilute extract

concentrations is possibly less significant if the residuals exhibit inconsequential

cytotoxic effects at more dilute concentrations.

The observed decreases in cytotoxicity values as the incubation time increases,

suggest that the extracted chemical residues associated with the sterilized polymers are being degraded into non-toxic compounds over time. One hypothesis for the improved cytotoxicity values is that ClO2 is the primary residual causing the initial cytotoxic

effects, which is then being degraded into non-toxic concentrations of chlorite, chlorate 116

and chloride. Alternatively, toxic concentrations of chlorite, chlorate or chloride may be present initially, but are degraded by the cells during normal metabolic functions. The chemicals that the MRC-5 fibroblasts are being exposed to as a result of the ClO2 sterilization of polymers can be determined with further investigation. Extracting the chemical residuals into distilled water under the same environmental conditions used for the cytotoxicity assay and using mass spectroscopy could elucidate the chemicals to which the fibroblasts are initially exposed.

The results from these cytotoxicity experiments indicate that sterilization of polymers with ClO2 does result in the presence of chemical residuals associated with the

materials post-sterilization. Humidification of the materials and sterilization in a high

relative humidity environment results in the least cyototoxicity associated with extracts

from the material. Whether low concentrations of ClO2 and longer exposure times, or

high concentrations of ClO2 and shorter exposure times decreases the cytotoxicity

response is dependent on the polymer being sterilized. Post-sterilization processing for materials sterilized with ClO2, beyond aeration for 24 hours at atmospheric pressure, is

required to improve the cell viability to levels that are, at a minimum, comparable to

those observed with ethylene oxide sterilization. Aeration under pressure, temperature

and/or increased airflow, similar to post-sterilization processing performed for ethylene

oxide sterilized materials, may accelerate the removal of chemical residuals associated

with the ClO2 sterilized polymers. While the results from both cytotoxicity assays

indicate that post-sterilization processing of ClO2 sterilized polymers is required, the results from the morphometric analysis indicate that the initial cytotoxic effects of 117

residuals associated with polymers sterilized with ClO2 are less severe than those associated with ethylene oxide sterilization. This suggests that if the concentration of the extracts associated with ClO2 sterilized polymers can be reduced such that the long-term cell viability is improved, sterilization with ClO2 gas could be a practical alternative for ethylene oxide sterilization. 118

5. INTRODUCTION: ANTIMICROBIAL BIOMATERIALS FOR THE IN SITU

PREVENTION OF INFECTION

The prevalence of bacterial infections in association with biomaterials remains one of the leading causes of medical device failure. Opportunistic staphylococci, enterococci, Enterobacteriaceae and Candida sp. commonly colonize blood vessel catheters, cerebrospinal shunts, arterial grafts, cardiac valves, total artificial hearts, and total joint replacements (Schierholz and Beuth 2001). Enterococci and

Enterobacteriaceae also commonly infect urinary catheters. Statistics indicate that coagulase-negative staphylococci (CNS), such as Staphylococcus epidermidis, are responsible for 50-70% of catheter infections (Archer 1995), 40-50% of cardiac valve infections (Li et al. 2000), and 20-50% of joint replacement infections (Gentry 1997). In

2000, the Center for Disease Control ranked microbial agents as the fourth leading actual cause of death in the United States of America (Mokdad et al. 2004).

Process of biomaterial-associated infections

The process of infection begins with the adherence of bacteria to the material, described as a two-phase procedure, with an initial and reversible physical phase, followed by an irreversible cellular and molecular phase (Marshall et al. 1971; Marshall

1985). The initial attraction between the bacteria and the biomaterial surface is promoted by nonspecific physicochemical interactions, such as van der Waals forces, surface tension, electrostatic interactions and surface hydrophobicity (Ferreiros et al. 1989; Fleer and Verhoef 1989; Kojima et al. 1990; Christensen 1994; Bisno 1995). The adhesion 119

process is mediated by the interactions of specific bacterial adhesins, such as staphylococcal surface proteins (SSPs), SSP-1 and SSP-2 (Timmerman et al. 1991;

Veenstra et al. 1996), capsular polysaccharide adhesin (PS/A) (Tojo et al. 1988), and host or material derived receptors. Colonization of the material by the bacteria begins 30-60 minutes after the initial exposure. Initially, bacteria grow in a single layer, contacting the material. However, further accumulation results in the formation of multiple layers of bacteria and the eventual formation of a biofilm (bacterial cells embedded in a polysaccharide matrix that is produced by the bacteria) (Figure 5.1). This matrix affords the sessile bacteria additional protection against host defense mechanisms and antimicrobial agents (Costerton et al. 1999; Xu et al. 2000). The persistence of the infection is also dependent on the fact that sessile bacteria have a different physiology from planktonic bacteria, with an altered growth rate that makes them less susceptible to the effects of growth rate dependent antibiotics (O'Gara and Humphreys 2001). Results from other studies have shown that the minimum bactericidal concentration (MBC) for several antibiotics is up to 256 times higher for sessile bacteria, relative to planktonic strains (Chuard et al. 1991).

The current method of treatment of biomaterial-associated infections is the administration of systemic antibiotics. However, the effectiveness of this form of treatment is compromised by the formation of bacterial biofilms on the material surface

(Drenkard 2003). As such, the development of antimicrobial biomaterials to prevent bacterial infections from occurring has become a focus in biomaterial research. 120

Figure 5.1: Molecular sequences in bacterial (B) attachment, adhesion, aggregation and dispersion relative to a material surface (reprinted from Science, 237, Gristina, A.G.,

Antimicrobial biomaterials

The development of biomaterials to prevent implant-associated bacterial infections has focused on both modifications to material surface properties to limit bacterial adhesion, and the incorporation of antimicrobial agents into or onto

biomaterials.

Physicochemical modifications

The surfaces of most biomaterials are physicochemically active, and therefore

modulate molecular interactions at the material interface. Altering the surface properties

of materials to inhibit the adhesion of bacteria has been one approach towards preventing biomaterial-associated infections from developing. Surface free energy, surface charge

(Hogt et al. 1985), surface texture (Locci et al. 1981; An et al. 1995), and hydrophobicity

(Reynolds and Wong 1983; Pringle and Fletcher 1986) have all been identified as important material surface properties that influence bacterial interactions. Metal surfaces tend to have high surface energies, are negatively charged, and are hydrophilic. Polymers generally have low surface energy, are less electrostatically charged, and are hydrophobic.

The surface free energy of a material is comprised of three components, dispersion, polarity and hydrogen bonding. Of the three components, polarity has been determined to most influence the adhesion of bacteria to the material surface (Kitano et al. 1996). Decreasing the polarity of the material results in a reduction in the number of bacteria adhered to the surface (Kitano et al. 1996). Irregularities in the surface of biomaterials have been found to enhance bacterial adhesion (Baker and Greenham 1988; 123

McAllister et al. 1993). Quirynen et al. (1993) observed that 25 times more bacteria adhered to rough surfaces relative to smooth surfaces. The effects of surface texture on

bacterial adhesion is believed to stem from increased surface area associated with

increased roughness, with depressions providing favorable colonization sites (Baker and

Greenham 1988).

Modifying the surface properties of biomaterials in order to inhibit bacterial

adhesion, while an ideal approach, is complicated by the different properties among

bacteria species and strains. Surface hydrophobicity of bacteria will influence whether

the bacteria will adhere to a material surface and the extent of adhesion that occurs.

Bacteria with hydrophobic properties will preferentially adhere to hydrophobic material

surfaces, while bacteria with hydrophilic characteristics prefer hydrophilic material surfaces (Hogt et al. 1983; Satou et al. 1988). Hydrophobic bacteria also adhere more readily than hydrophilic bacteria (van Loosdrecht et al. 1987; Galliani et al. 1994).

Gottenbos et al. (2001) found that gram-negative and gram-positive bacteria will adhere to and colonize the surface of positively and negatively charged surfaces differently.

While both gram-negative and gram-positive bacteria adhere more rapidly to positively charged surfaces, only gram-positive bacteria will successfully colonize on such surfaces.

Adhesion to negatively charged surfaces is substantially slower, but both gram-negative and gram-positive bacteria exhibit exponential growth once adhered.

An additional complication associated with using physicochemical modifications of material surfaces to prevent bacterial adhesion is that materials become coated with host proteins once in situ, effectively altering the surface properties of the material. 124

Whether protein adsorption inhibits or promotes bacterial adhesion is dependent upon the protein, its configuration when it adsorbs to the material surface, and the strain of bacteria the material is being tested against. The physicochemical characteristics of the surface of the material can influence the tertiary structure of the protein as it adsorbs to the material surface (Kitano et al. 1996), thereby affecting its ability to enhance or inhibit bacterial adhesion.

Incorporation of antibiotics

Since antibiotics are used systemically to treat biomaterial-associated infections, incorporation of antibiotics into the biomaterials to provide a localized delivery of the drug is an appropriate approach to prevent biomaterial-associated infections. Methods to incorporate antibiotics into biomaterials have included passive and ionic adsorption, covalent binding, and incorporation during the preparation of the biomaterial.

Passive adsorption of antibiotics is generally considered ineffective, since the drug is rapidly eluted from the material once implanted (Avramovic and Fletcher 1991;

Goeau-Brissonniere et al. 1994). Similarly, while ionically bound antibiotics provide temporary protection against infection, they are rapidly depleted (Avramovic and

Fletcher 1991; Goeau-Brissonniere et al. 1994) and below minimal inhibition concentrations (MIC) are then released (Avramovic and Fletcher 1991; Strachan et al.

1991; Gahtan et al. 1995; Vicaretti et al. 1998; van de Belt et al. 2001). This raises concerns since exposure to below MICs creates an optimal environment for the development of antibiotic resistant strains of bacteria (van de Belt et al. 1999). Covalent binding of the antibiotic to the surface of the biomaterial will increase the duration of 125

release of the drug (Duran 2000). This method of attaching the antibiotic requires that

the coating process be mild enough, such that the antimicrobial agent retains its activity.

Incorporation of antibiotics during processing of the raw biomaterial requires that the

antibiotics be compatible with the polymer solvent and not affect the solubility of the

polymer. Gentamicin has been successfully incorporated into polyurethane using this

method (Kwok et al. 1999b), as have several different antibiotics into bone cement

(PMMA) (Hendriks et al. 2004). Preventing the rapid release of antibiotics from the

biomaterials is one of the hurdles associated with most of these methods of antibiotic incorporation. Approaches to control the release of the antibiotics from the material have included the addition of thin polymer films (Kwok et al. 1999a) or hydrogel coatings

(Changez et al. 2004; Ginalska et al. 2005) on to the surface of the material to sustain the delivery of antibiotics.

The prophylactic utilization of antibiotics for the prevention rather than the treatment of infections has caused concern about the development of antibiotic resistant strains of bacteria (Neu 1992; Anderson 1999). One approach to combating these concerns has been to utilize combination therapies, rather than single antimicrobial agents

(Raad and Hanna 1999; Duran 2000). An alternative approach has been the development of controlled release polymers that only release the antibiotics in the presence of

environmental signals indicative of infection (Suzuki et al. 1998; Tanihara et al. 1999).

Incorporation of biocides

The incorporation of biocides into general materials is not a recent development.

Biocides have been integrated into materials such as industrial polymers (Chemical 126

Industries Association 2003) and surfaces maintained in aquatic environments (Alfred

Wegener Institute 2005). However, with the exception of silver (discussed below), the

incorporation of biocides into biomaterials is a recent occurrence. Research assessing

polymers incorporated with quaternary ammonium compounds (QACs) indicate that antimicrobial activity in vitro can be achieved (Kenawy el et al. 1998; Flemming et al.

2000), as can iodine complexed into central venous catheters (Jansen et al. 1992).

Chlorhexidene gluconate, when incorporated into silicone (Whalen et al. 1997) and coated onto catheters (Burrows and Khoury 1999), has demonstrated reduced bacterial adhesion in vitro and bactericidal activity in vivo, respectively. The utilization of combined biocide therapies, with the incorporation of both chlorhexidine and silver sulfadiazine into catheters, has been assessed (Raad and Hanna 1999; Brun-Buisson et al.

2004; Rosato et al. 2004; Jaeger et al. 2005; Ostendorf et al. 2005). While most studies observed a reduction in catheter colonization (Brun-Buisson et al. 2004; Jaeger et al.

2005; Ostendorf et al. 2005), the incidence of blood stream infections was similar to control groups (Brun-Buisson et al. 2004; Ostendorf et al. 2005). A catheter with three biocides (chlorhexidene, silver sulfadiazine and triclosan) incorporated into the material has also been assessed, and found to exhibit a broad spectrum of resistance against microbial colonization (Gaonkar et al. 2003).

As with the prophylactic administration of antibiotics via biomaterials, the incorporation of biocides into materials carries the risk of biocide resistance. While there is a mistaken impression that biocide resistance cannot occur, concerns about biocide resistance were initially expressed as early as 1887, when it was discovered that bacteria 127

could develop resistance to increasing concentrations of biocides (Kossiakoff 1887).

Mechanisms of resistance to chlorhexidine, triclosan and QACs have already been

established (Levy 2002).

Silver

Silver was first utilized as an antimicrobial agent in the medical field in 1981 in

the form of 0.5% silver nitrate (Hartford 1981), which was succeeded by 1% silver

sulfadiazine as complications of the silver nitrate became apparent (Harvey 1980). While the precise mechanism of action of silver against bacteria is not known, it has been established that silver needs to be in ionic form to achieve bactericidal activity (Goetz

1943). Research indicates that silver ions have a strong affinity for deoxyribonucleic acid

(DNA) and ribonucleic acid (RNA) (Fox et al. 1969; Modak and Fox 1973; Fox and

Modak 1974). Silver interacts with the hydrogen bonds that bind the DNA strands, a

process that may interfere with the separation of the DNA, inhibiting bacterial proliferation (Fox et al. 1969). Silver ions have also been attributed with the disruption of cell respiration and energy metabolism (Bragg and Rainnie 1974; Cowlishaw et al.

1982; Schierholz et al. 1998b), and membrane damage (Rochart and Uzdins 1947;

Coward et al. 1973), resulting in cell death. Research has also indicated that at the concentrations required to cause bactericidal activity, silver ions have a minimal effect on eukaryotic cells (Ricketts et al. 1970; Berger et al. 1976; Fox 1983).

Metallic silver has been incorporated into various biomaterials, including orthopedic and dental implants (Becker and Spadaro 1978; Webster et al. 1981; Aydin et al. 1997), vascular catheters and cuffs (Mermel et al. 1993; Raad et al. 1996), and 128

prosthetic heart valves (Tweden et al. 1997a; Tweden et al. 1997b; Illingworth et al.

1998; Langanki et al. 1998), with variable antimicrobial activity observed. While the

bactericidal activity of silver in an in vitro environment has been established (Adams et

al. 1999; Klueh et al. 2000), its bactericidal activity in vivo is more complicated.

Achieving and maintaining the concentrations of free silver ions necessary for

bactericidal activity is hampered by interactions with proteins, which renders them

ineffective against microbial organisms (Schierholz et al. 1999).

Chlorine dioxide generating material

Chlorine dioxide (ClO2) is a halogen-releasing biocide, commonly used in the food and water industry for decontamination and bleaching purposes. It has also been

employed in aqueous form for the decontamination of reusable medical equipment, such

as endoscopes (Ayliffe 2000; Coates 2001). While the specific mechanisms of

antimicrobial activity of ClO2 remain unclear, its bactericidal activity is believed to stem

from the fact that as a strong oxidizing agent, ClO2 causes the oxidation of electron-rich

amino acids, resulting in protein degradation (Noss et al. 1983; Noss et al. 1986).

Bernard Technologies Incorporated (Chicago, IL) has developed a selection of polymer

materials that generate ClO2 gas when they are exposed to specific environmental

conditions.

Humidity activation

The humidity activated ClO2-generating polymers are a blend of polyethylene and ethylvinvyl alcohol, coextruded with microspheres. The microspheres are made from a

hydrophilic material and have been coated with sodium chlorite (NaClO2) and an acid 129

precursor. As the environmental humidity increases, the hydrophilic microspheres attract

water molecules, which convert the acid precursor into an acid that reacts with the

sodium chlorite and produces ClO2. The ClO2 diffuses out of the bulk of the polymer,

creating a ClO2 microenvironment around the material (Figure 5.2).

Light activation

A light activated formulation of ClO2-generating polymer developed by Bernard

Technologies Incorporated (Chicago, IL) utilizes the photoxidation of sodium chlorite to produce chlorine dioxide when stimulated by visible and near ultraviolet light. Sodium chlorite is crystallized onto a modified titanium dioxide surface, which, when exposed to light with the appropriate activation wavelength, produces electrons that oxidize the sodium chlorite to form chlorine dioxide.

Pare Technologies Inc. (Tucson, AZ) is investigating alternative technologies of light activation for ClO2 productoin, utilizing photoacid generating (PAG) chemistries.

PAG chemicals have been utilized in the areas of 3D-microfabrication, ultra high-density optical data storage, biological imaging, and the controlled release of biological agents

(Zhou et al. 2002). The generation of acid occurs when the PAG chemistries absorb photons from an applied light source, releasing protons. In terms of generating ClO2, these protons can cause the oxidation of sodium chlorite, and the subsequent production of ClO2. Pare Technologies Inc. has been investigating several different PAG chemistries

manufactured by AZ Microsystems (Tucson, AZ) for their ability to produce ClO2 when

incorporated into a hydrogel matrix.

130

Figure 5.2: Schematic of the generation of ClO2 from a humidity-activated formulation 131

132

Research Plan

A reliable method for the in situ prevention of biomaterial-based infections has yet to be established. The following specific aims were designed to assess the applicability of using a chlorine dioxide generating material to prevent biomaterial infections. These studies have been presented in Chapter 6 and Chapter 7, respectively.

Specific Aim #4: To assess the efficacy of a chlorine dioxide generating material in in vitro and in vivo environments. Hypothesis #4. The chlorine dioxide generating material will cause sustained bactericidal activity in both in vitro and in vivo environments. The chlorine dioxide concentrations generated by the material will not cause an adverse response to the local host tissue in vivo.

133

6. ADHESION AND COLONIZATION OF STAPHYLOCOCCUS EPIDERMIDIS

ON THE SURFACE OF A CHLORINE DIOXIDE-GENERATING MATERIAL

Introduction

The prevalence of infections associated with medical devices continues to escalate, despite advances in biomaterial research. Biomaterial-associated infections begin with the adhesion of bacteria to the material surface, a process that is mediated by the presence of host proteins that coat the material surface once it is in situ (An and

Friedman 1998). After adhesion occurs, proliferation and colonization of the bacteria on the material surface results in the formation of a biofilm, which further enables the bacteria to evade both host immune responses and systemic antibiotic therapies

(Costerton et al. 1999; Xu et al. 2000).

Measures to prevent bacterial infections associated with biomaterials have included inhibiting bacterial attachment and colonization (Gottenbos et al. 2003), the incorporation of antibiotics, silver, or antibiotic-silver combinations into the material

(Schierholz et al. 1997; Schierholz et al. 1998a), and the prophylactic administration of antibiotics. However, with the continuing emergence of antibiotic resistant strains of bacteria, the practice of administering antibiotics prior to the onset of infection is becoming less popular (Sieradzki et al. 1998). There has also been a recent trend of incorporating biocides into a variety of materials to reduce cross-contamination between patients (Tambyah et al. 1999; Fraise 2002). 134

Chlorine dioxide (ClO2) gas is a biocide with established antimicrobial properties,

although its specific mechanisms of antimicrobial activity remain unclear. A polymer

material that generates ClO2 when exposed to threshold levels of relative humidity has

been developed (Bernard Technologies Incorporated, Chicago, IL) and has the potential

to be used as an antimicrobial surface coating for biomaterials. These experiments were

designed to assess the bactericidal activity of the ClO2-generating material in vitro. Since

microbial adhesion is the first step towards the development of infection, the adhesion

and viability of Staphylococcus epidermidis on the surface of a ClO2-generating material

were assessed. These experiments were conducted using control and ClO2-generating material samples with albumin and fibronectin adsorbed onto the surface, since bacterial adhesion to the surface of materials is mediated by the presence of protein in vivo.

Albumin, the most prevalent protein in serum, generally inhibits bacterial adhesion

(Pringle and Fletcher 1986; Keogh et al. 1992; Keogh and Eaton 1994), while fibronectin reportedly promotes bacterial adhesion (Mohammad et al. 1988).

135

Materials & Methods

Chlorine dioxide-generating material: The ClO2-generating material used in these

experiments was a humidity-activated formulation. The polymer was a blend of

polyethylene and ethylvinyl alcohol, coextruded with microspheres as a polymer rod

(diameter 0.65cm). The microspheres were coated with sodium chlorite (NaClO2) and an acid precursor, the chemical reagents responsible for the generation of ClO2. A polymer

rod extrusion with the same chemical composition, but without the NaClO2-coated microspheres was used as a control material. Segments of control material samples

(1.5cm) were prepared and sterilized using ethylene oxide gas. Segments of ClO2-

generating material samples (1.5cm) were prepared aseptically just prior to use. Control

and ClO2-generating material samples were exposed to one of three environmental

conditions prior to being exposed to the bacterial solution: no protein, BSA protein, or

fibronectin protein. Three control and three ClO2-generating material samples were

exposed to each environmental condition.

Bacteria: Staphylococcus epidermidis 10-16 was provided by the Department of

Veterinary Science and Microbiology at the University of Arizona. The strain was

maintained on tryptic soy agar (TSA) (EMD, Gibbstown, NJ) plates at 4°C.

Polysaccharide production: A Congo red agar assay was used to assess production of

polysaccahride by Staphylococcus epidermidis 10-16. Brain heart infusion (BHI) agar

(BD, Sparks, MD) was made with 0.8% (w/v) Congo red (Calbiochem, La Jolla, CA). A

single colony of Staphylococcus epidermidis 10-16 was streaked across the congo red

agar plate and incubated overnight at 37°C. Colonies were assessed according to the 136

method outlined by Freeman et al. (1989). Positive production of polysaccharide is

indicated by the growth of black colonies, while reduced polysaccharide production results in pink colonies.

Antimicrobial activity of ClO2-generating material: A zone of inhibition assay (ZOI)

was used to assess the duration of antimicrobial activity of the ClO2-generating material.

Staphylococcus epidermidis 10-16 was plated on Mueller Hinton agar (MHA) (BD,

Sparks, MD). Segments of the ClO2-generating material (1.5cm; n = 3) were placed on

separate MHA plates. Plates were incubated at 37°C for 24 hours and the ZOI assessed.

Each sample was placed on a fresh MHA plate of Staphylococcus epidermidis 10-16, re-

incubated for 24 hours, and the new ZOI assessed. The process was repeated until no

ZOI formed around the ClO2-generating material sample.

Protein adsorption: Control and ClO2-generating materials were incubated at 37°C in either a 1% (w/v) solution of unmodified bovine serum albumin (BSA) made up in DCF-

PBS or a 1% (w/v) solution of fibronectin (BD Biosciences, Bedford MA) made up in

DCF-PBS. To ensure that total coverage of the material samples was achieved, control material samples were incubated for selected time points and protein coverage assessed.

Segments of control material sample (1.5cm) were incubated in a 1% albumin or fibronectin solution for 1, 2, or 4 hours. At selected time points, the material was removed from the protein solution, rinsed with DCF-PBS and placed into 400µl of sample buffer and boiled for 10 minutes. The sample buffer was concentrated using a centrifugal filter device (MWCO 50kDa for albumin and 100kDa for fibronectin)

(Millipore Corporation, Bedford, MA). The concentrated sample buffer was run on a 137

6%, 1.5mm SDS-PAGE gel for 75 minutes at 20mA, and then silver stained. The protein

bands were assessed for consistency in their intensity between consecutive time points to

ascertain the optimal incubation time required for equilibrium in the protein adsorbed to

the material surface.

Assessment of bacterial adhesion: Bacterial adhesion was assessed using

Staphylococcus epidermidis 10-16 labeled with Indium oxine-111 (111In) (Mohammad

and Ardehali 2000). Bacterial cultures were established by inoculating tryptic soy broth

(TSB) with 1 colony and incubating the solution for 12 hours at 37°C in an incubator

shaker (Model G25, New Brunswick Scientific Company Inc., Edison, NJ). The bacterial

culture was centrifuged at 16,000×g for 15 minutes and the bacteria pellet resuspended in

10ml of TSB with 75µCi of 111In. The solution was incubated for 2 hours at 37°C in the

incubator shaker. The radiolabeled culture was centrifuged at 16,000×g for 15 minutes

and residual 111In was removed by washing the 111In bacterial pellet with 0.85% NaCl.

The washed bacterial pellet was resuspended in 10ml TSB. Eighteen cultures with 111In labeled bacteria were established by inoculating 9.5ml TSB with 500µl of the labeled bacteria. Control and ClO2-generating material samples were incubated in either DCF-

PBS (no protein), 1% (w/v) albumin, or 1% (w/v) fibronectin solutions for 2 hours and

were then incubated in the 111In labeled bacterial culture (1 sample per culture) at 37°C in the incubator shaker. Every 6 hours, the material samples were transferred into new 111In labeled bacterial cultures. After incubation times of 3, 6, 12, 18, and 24 hours, material samples were analyzed for bacterial adhesion using a gamma counter (Gamma 5500,

Beckman Instruments, Fullerton CA) to measure radioactivity. After each assessment the 138

samples were returned to an 111In labeled bacterial culture until the next assessment. The radioactivity was calculated, accounting for isotope decay during the experiment. After the final measurement for radioactivity, the material samples were stored in vials until the

111In isotope decayed. The samples were then prepared for imaging with scanning

electron microscopy (SEM).

Scanning electron microscopy (SEM): SEM was used to visualize bacterial adhesion

on the surface of control and ClO2-generating materials, and to assess the effect of

protein on bacterial adhesion. Samples used in the bacterial adhesion assay were fixed in

3% gluteraldehyde for >24 hours. They were then dehydrated using decreasing dilutions

of alcohol, critical point dried, and sputter coated with gold. Samples were imaged using

a JEOL JSM-820 scanning electron microscope (JEOL Ltd., Peabody, MA).

Assessing bacterial viability: Cultures of Staphylococcus epidermidis 10-16 were established by inoculating 11ml of tryptic soy broth (TSB) with 1 colony and incubating the solution for 12 hours at 37°C in the incubator shaker. Eighteen fresh bacterial cultures were established by inoculating 9.5ml of TSB with 500µl of the 12 hour bacterial culture. Control and ClO2-generating material samples were incubated in either DCF-

PBS (no protein), 1% (w/v) albumin or 1% (w/v) fibronectin solutions for 2 hours and

were then incubated in the fresh bacterial culture (1 sample per culture) at 37°C in the

incubator shaker. Every 6 hours, the material samples were transferred into new bacterial

cultures. After incubation times of 3, 6, 12, 18, and 24 hours, material samples were

removed from the bacterial cultures and prepared for sonication to remove adherent

bacteria from the surface of the material. 139

Sonication: Material samples were placed in sterile 15ml test tubes with 3ml of TSB and

sonicated for 3 hours at 130W/47kHz in a bath sonicator (Branson 2200, Branson

Ultrasonics Corporation, Danbury, CT). Each material sample was rinsed with 5ml of

TSB which was added to the sonicated TSB for the specific sample. The total volume for

each sample was transferred into separate centrifuge tubes and prepared for the

Live/Dead® BacLight™ bacterial viability assay (Molecular Probes, Eugene OR).

Live/Dead® BacLight™ bacterial viability assay: The sonicated bacteria solutions

were centrifuged at 16,000×g for 15 minutes and the bacterial pellet was resuspended in

3ml of 0.85% NaCl. Equal volumes of a green-fluorescent nucleic acid stain, SYTO®9, and a red-fluorescent nucleic acid stain, propidium iodide, were mixed, and 9µl of the combined reagents were added to the 3ml resuspended bacterial solutions. The solutions were incubated at room temperature in the dark for ≥15 minutes and the fluorescence was measured using a double excitation emission fluorometer (Fluorolog 3-22, JY Horiba,

Edison, NJ). The bacterial suspensions were excited at 470nm and the excitation spectrum measured (490-700nm). The data were assembled and corrected using Matlab and exported in .ascii format. The integrated intensities between 510-540nm (live bacteria) and 620-650nm (dead bacteria) were calculated and the ratio of live to total bacteria determined to give a measure of relative bacteria viability.

Statistics: Data are presented as the mean ± the standard error mean. Significance between the different conditions was determined using a 2-tailed unpaired student t-test.

A p ≤ 0.05 was considered statistically significant.

140

Results

Antimicrobial activity of ClO2-generating material: Results from the zone of

inhibition (ZOI) assay indicate that the duration of bactericidal activity produced by the

ClO2-generating material is 5 days (Figure 6.1). Comparable levels of bactericidal

activity were observed for the initial 72 hours, but began to diminish thereafter.

Protein adsorption: Control material samples were assessed to determine the duration of incubation in a 1% protein solution required for protein adsorption to the material surface to reach equilibrium. SDS-PAGE gels suggested that adsorption of protein onto the surface of material sample reached equilibrium after 2 hours (Figure 6.2). Based on these results, for the experiments assessing bacterial adhesion and colonization on the surface of control and ClO2-generating materials, all samples were incubated at 37°C for

2 hours in either a 1% (w/v) albumin solution, a 1% (w/v) fibronectin solution, or a

control solution with no protein.

Polysaccharide production: The extracellular production of polysaccharide by

Staphylococcus epidermidis 10-16 was assessed using a Congo red agar assay. The

growth of smooth pink/red colonies rather than dark colonies with a dry appearance,

indicate that 10-16 does not produce significant quantities of extracellular polysaccharide

(Figure 6.3).

Assessing bacterial adhesion: Analysis of the adhesion of bacteria to control and ClO2- generating materials was assessed using radiolabeled bacteria. The results showed that the presence of albumin or fibronectin did not significantly (p > 0.05) inhibit or enhance adhesion of Staphylococcus epidermidis 10-16 to control or ClO2-generating materials 141

Figure 6.1: Zone of inhibition (ZOI) assay assessing bactericidal activity of ClO2- generating material after the material has been activated for: (A) 24 hours; (B) 48 hours;

(A) (B)

ClO2-generating material

(E) (F) 143

Figure 6.2: SDS-PAGE gels indicating the incubation time required to achieve equilibrium in protein adsorption to a control material segment from a: (A) 1% (w/v) albumin solution; (B) 1% (w/v) fibronectin solution. 144

(A)

66kDa

1 hour 2 hours 4 hours

(B)

250kDa

1 hour 2 hours 4 hours 145

Figure 6.3: Congo red agar assay used to assess the production of extracellular polysaccharide by Staphyloccocus epidermidis 10-16. Pink/red colonies indicate that 10-

16 does not produce significant extracellular polysaccharide.

146

147

(Figure 6.4). The data also indicates that bacterial adhesion to the ClO2-generating material was significantly greater (p ≤ 0.05) than bacterial adhesion to the control material. Scanning electron microscopy images revealed areas of bacterial adhesion on the surface of the material samples (Figure 6.5). All control material samples had bacteria adhered to them (Figure 6.5A, C & E). There were substantially larger areas of bacterial biofilm on the control material samples with fibronectin adsorbed to the surface

(Figure 6.5E). ClO2-generating material samples with fibronectin adsorbed to the surface also had considerably larger areas of bacterial biofilm than samples with no protein, or albumin adsorbed to the surface (Figure 6.5F). ClO2-generating material samples with no protein or albumin adsorbed to the material surface had less bacteria adhered to the surface than control material samples.

Assessing bacteria viability: Results from the Live/Dead® BacLight™ bacterial viability assay are presented in Figure 6.6. The data indicate that there are significantly less (p ≤ 0.05) viable bacteria associated with the ClO2-generating material than the control material, independent of whether any protein was adsorbed to the surface of the materials. The presence of protein adsorbed on the ClO2-generating material did not impede the antimicrobial effects of ClO2, with no significant difference (p > 0.05) in the level of viable versus non-viable bacteria detected on any of the ClO2-generating material samples. However, bacteria viability was affected by the adsorption of albumin or fibronectin to the surface of the control material. The presence of fibronectin caused a significant increase (p ≤ 0.05) in the ratio of viable to nonviable bacteria adhered control material during the first 6 hours of bacterial adhesion/colonization. This effect was 148

Figure 6.4: 111In labeled bacterial adhesion assay. Lines of best fit (logarithmic

regression) for each data set depicted.

C: Control material; AC: Control material with albumin absorbed to the material surface;

FC: Control material with fibronectin adsorbed to the material surface; ClO2: ClO2- generating material; AClO2: ClO2-generating material with albumin absorbed to the

material surface; FClO2: ClO2-generating material with fibronectin adsorbed to the

material surface 149

6 ClO2 AClO2 5 FClO2 AC C 4 FC

C Radioactivity (nCi) Radioactivity 2 ClO2 AC

AClO2 1 FC

FClO2 0 0 5 10 15 20 25 30

Incubation Time (hrs)

150

Figure 6.5: Scanning electron microscopy images of bacterial adhesion to: (A) Control

material (no absorbed protein); (B) ClO2-generating material (no absorbed protein); (C)

Control with albumin absorbed to the surface; (D) ClO2-generating material with albumin

absorbed to the surface; (E) Control with fibronectin absorbed to the surface; (D) ClO2- generating material with fibronectin absorbed to the surface

151

(A) (B)

(E) (F) 152

Figure 6.6: Live/Dead® BacLight™ bacterial viability assay. Lines of best fit

(logarithmic regression) for each data set depicted.

C: Control material; AC: Control material with albumin absorbed to the material surface;

FC: Control material with fibronectin adsorbed to the material surface; ClO2: ClO2- generating material; AClO2: ClO2-generating material with albumin absorbed to the

material surface; FClO2: ClO2-generating material with fibronectin adsorbed to the

material surface

153

0.8

0.7 FC C 0.6 AC

0.5 D ClO L 0.4 2 FClO2

L AClO2 0.3 C

Live/(Live+Dead) ClO2 0.2 AC

AClO2 (fluorescence emission spectra) 0.1 FC

FClO2 0.0 0 5 10 15 20 25 30

Incubation Time (hrs)

154

temporary, with no significant differences (p > 0.05) in bacteria viability detected for incubation times greater than 12 hours. The presence of albumin caused no significant difference (p > 0.05) in bacteria viability during the first 18 hours of incubation in the

Staphylococcus epidermidis 10-16 bacteria culture. After 24 hours the presence of albumin on the control material caused a significant reduction (p ≤ 0.05) in bacteria viability relative to the control material with no protein adsorbed to the material surface.

155

Discussion

Adhesion of bacteria to the surface of a material is the initial step in the

development of biomaterial-associated infections. In these studies, the adhesion and

initial colonization of Staphylococcus epidermidis 10-16 to the surface of a control and a

ClO2-generating material was assessed. Since proteins have been shown to mediate

bacterial adhesion, and because all implanted biomaterials become coated with proteins

in situ, an anionic (albumin) and cationic (fibronectin) protein were adsorbed to separate

material samples to assess their influence on bacterial adhesion.

Whether the production of extracellular polysaccharide would influence bacterial adhesion in these experiments was determined. Prior research has indicated that the production of extracellular polysaccharide does not enhance bacteria adhesion, but may be important for accumulation (Galliani et al. 1994). The results from the Congo red agar assay indicated that Staphylococcus epidermidis 10-16 does not produce excess extracellular polysaccharide. The duration of bactericidal activity of the ClO2-generating material against Staphylococcus epidermidis 10-16 was assessed using zone of inhibition

(ZOI) experiments. The results indicate that the material produces bactericidal concentrations of ClO2 for 5 days.

The adhesion of Staphylococcus epidermidis 10-16 to the surface of control and

111 ClO2-generating material was assessed using bacteria labeled with In. The results from

the assay indicate that more bacteria were adhered to the surface of the ClO2-generating material than to the control material. One possible explanation for these results relates to the fact that prior to specific ligand-mediated adhesion the initial attachment of bacteria 156

to the surface of a substrate is reversible. While the process of reversible binding may

have been transpiring unimpeded for the control material, it may have been interrupted

for the ClO2-generating material. With the rapid bactericidal action of ClO2 upon contact

of the bacteria with the surface of the ClO2-generating material, dead bacteria may have

remained weakly adhered to the surface of the material, providing an accumulating

substratum containing specific bacterial adhesins, thereby promoting bacterial adhesion.

Alternatively, there may have been differences in the physicochemical properties of the

surface of the control and ClO2-generating materials. For example, polymers with greater surface tension and hydrophobicity (Jansen et al. 1989) and reduced polarity (Kitano, et

al. 1996) reduce bacteria adhesion.

The adsorption of albumin or fibronectin on the surface of the control and ClO2-

generating materials generally did not significantly affect the adhesion of Staphylococcus epidermidis 10-16 to the material. However, after being incubated for 24 hours there was a significant reduction in bacterial adhesion to both control and ClO2-generating material

samples with fibronectin adsorbed to the surface, when compared to shorter incubation

times. Studies addressing the effects of protein adsorbed to the surfaces of biomaterials on the adhesion of Staphylococcus aureus and other pathogens have produced relatively

consistent results. These studies found that albumin inhibits bacterial adhesion (Espersen et al. 1990; Carballo et al. 1991; Muller et al. 1991; Zdanowski et al. 1993; Galliani et al.

1994; Keogh and Eaton 1994), while fibronectin enhances bacterial adhesion (Proctor et al. 1982; Ryden et al. 1983; Kuusela et al. 1985; Wadstrom et al. 1985; Muller et al.

1991; Yanagisawa et al. 2004). Whether the presence of various proteins inhibits or 157

promotes the adhesion of Staphylococcus epidermidis, however, remains controversial.

There are some investigators who have reported that albumin inhibits the adhesion of

Staphylococcus epidermidis to the surfaces of selected materials (Espersen et al. 1990;

Carballo et al. 1991; Zdanowski et al. 1993; Galliani et al. 1994; Keogh and Eaton 1994),

while fibronectin enhances bacterial adhesion (Jansen 1987; Herrmann et al. 1988;

Carballo et al. 1991; Arciola et al. 2003). Other investigators report that adhesion of

Staphylococcus epidermidis is not enhanced by the adsorption of fibronectin on the

surface of the material (Naylor et al. 1989; Espersen et al. 1990; Muller et al. 1991;

Dexter et al. 2001), and in some instances, bacterial adherence is actually inhibited

(Muller et al. 1991; An and Friedman 1998).

In addition to the type of protein used, consideration also needs to be given to the

properties of the material substrate. The physicochemical properties of the material can

influence the tertiary structure of the protein when it adsorbs to the material surface

(Kitano et al. 1996). Adsorption of protein to some material surfaces may result in

conformational alterations, causing bacterial adhesin binding sites on the protein to

become inaccessible. This may alter the bacterial adhesion properties usually associated

with the protein. The strain of the bacteria can also influence the degree of bacterial

adhesion observed with certain materials and proteins (Muller et al. 1991; Galliani et al.

1994; Yanagisawa et al. 2004). Galliani et al. (1994) assessed 20 strains of

Staphylococcus epidermidis and found variations in the degree of bacterial adhesion

among the different strains. 158

In contrast to the results from the 111In adhesion assay, the SEM images suggest

that more bacteria are adhered to the surface of the control material with no protein or

albumin absorbed to the surface, relative to the ClO2-generating material samples.

However, the differences observed may be an artifact of the sample preparation required

for imaging with SEM. The adhesion of bacteria to the surface of the ClO2-generating material may not have been as durable as adhesion of bacteria to the control material,

since the former bacteria were killed upon initial contact with the material surface. This

would leave the bacterial biofilms vulnerable to the effects of shear stress and the process

of preparing the samples for SEM imaging may have been sufficient to disrupt the

adhered bacteria. The presence of fibronectin absorbed to the surface of the ClO2-

generating material appeared to prevent this disruption from occurring, since substantial

areas of bacterial biofilm were observed. This suggests that while fibronectin does not

enhance the adhesion of Staphylococcus epidermidis 10-16 to the surface of materials, it may augment the strength of the connections that do occur.

The results from the viability experiment indicated that there were significantly less viable bacteria associated with the surface of the ClO2-generating material than the

control material. The presence of albumin or fibronectin did not affect the bactericidal

activity associated with the ClO2-generating material. However, while the ClO2- generating material reduced the number of viable adherent bacteria, some viable bacteria were adhered to the surface of the material, with the amount increasing with the duration of time in the bacterial solution. One theory for the increasing number of viable bacteria associated with the ClO2-generating material concerns maintaining a bactericidal 159

concentration of ClO2 as the distance from the material surface increases. As bacteria adhered to one another, rather than directly onto the surface of the ClO2-generating material, the diffusion distance for the ClO2 increased, creating a reduced concentration at the outer bacterial layers. The concentration of ClO2 at the external layers of adhered bacteria may have been insufficient to cause bactericidal activity, resulting in the survival of the adhered bacteria. Increasing the concentrations of ClO2 produced by the material could alleviate this issue.

While the results from the adhesion experiment indicate that the presence of albumin or fibronectin did not impact bacterial adhesion, the results from the viability experiment indicate that bacteria viability was affected by the presence of adsorbed protein on the surface of the control material. Albumin adsorbed to the surface of the material resulted in a decrease in the viability of the bacteria adhered to the surface of the control material, relative to the control material with no protein adsorbed to the material surface. Fibronectin absorbed to the surface of the control material resulted in an increase in the viability of the bacteria adhered to the surface of the material, relative to the control material with no protein adsorbed to the material surface. The role that select proteins play in impacting bacteria viability has yet to be fully elucidated. Foreman-

Wykert et al. (2000) found that the presence of albumin enhanced the bactericidal activity of phosopholipases A2 against Staphylococcus epidermidis, but studies have yet to be conducted assessing the enhancement of bacterial viability by proteins such as fibronectin. 160

The results from these experiments indicate that while more bacteria attach to the

surface of the ClO2-generating material, the viability of the adhered bacteria is

significantly reduced, as compared to the control material. The adsorption of albumin

and fibronectin did not affect bacterial adhesion to either material surface, but did impact

bacteria viability on the surface of the control material. The application of a ClO2- generating material as a coating for biomaterials could have a positive impact on the prevention of biomaterial-associated infections.

161

7. ANTIMICROBIAL EFFICACY OF A CHLORINE DIOXIDE

GENERATING BIOMATERIAL IN VITRO AND IN VIVO

Introduction

Infections associated with biomaterials remain a significant cause of device

failure. The treatment of biomaterial-associated infections with systemic antibiotics is

complicated by the formation of bacterial biofilms on the surface of the implant

(Costerton et al. 1999; Xu et al. 2000). The bacteria within the biofilms are frequently

insusceptible to the application of antibiotics, resulting in the removal of the device in

order to resolve the infection.

Mechanisms to reduce device related infections include modifications to the

physicochemical surface characteristics of the biomaterial to prevent bacterial adhesion

from occurring (Gottenbos et al. 2003), the incorporation or attachment of antibiotics

(Dunn et al. 1994; Gaonkar et al. 2003; Hendriks et al. 2004), and the utilization of

materials with inherent antimicrobial properties, such as silver (Tweden et al. 1997b;

Illingworth et al. 1998; Langanki et al. 1998). However, with each of the aforementioned

approaches there are limitations and potential complications. When materials are

exposed to the in vivo environment they become coated with host proteins, thereby

altering the surface characteristics of a material which may have limited bacterial

adherence in vitro (Kitano et al. 1996). The release characteristics of antibiotics

associated with biomaterials can be difficult to control with rapid depletion of the antimicrobial agent and consequent concerns about creating an environment condusive to 162

the development of antibiotic resistant strains of bacteria (van de Belt et al. 1999). The

incorporation of biocides, such as silver, in biomaterials may also be associated with

altered bactericidal activity in vivo, rendering the biocides ineffective (Schierholz et al.

1999).

An antimicrobial polymer that generates chlorine dioxide (ClO2) gas was developed by a Chicago based company, Bernard Technologies Incorporated (Chicago,

IL). The production of this antimicrobial gas is dependent on attaining a specific level of atmospheric humidity. The bactericidal properties of this material have been assessed using an in vitro microbial adhesion assay (Chapter 6). The following studies were

designed to further elucidate the production of ClO2 by the material in vitro and to assess

the efficacy of the ClO2-generating material in an in vivo environment.

163

Materials and Methods

Chlorine dioxide-generating material: The ClO2-generating material used in these

experiments was a humidity-activated formulation. The polymer was a blend of

polyethylene and ethylvinyl alcohol, coextruded with microspheres as a polymer rod

(diameter 0.65cm). The microspheres were coated with sodium chlorite (NaClO2) and an acid precursor, the chemical reagents responsible for the generation of ClO2. A polymer

rod extrusion with the same chemical composition, but without the NaClO2-coated microspheres was used as a control material.

Bacteria: Staphylococcus epidermidis 10-16 was provided by the Department of

Veterinary Science and Microbiology at the University of Arizona. The strain was maintained on tryptic soy agar (TSA) (EMD, Gibbstown, NJ) plates at 4°C.

ClO2 production: Samples of ClO2-generating materials (2.5cm; n = 3) were placed in

separate 40ml vials with gauze wet with 5ml of deionized water for 24 hours at 37°C.

After 24 hours, each vial was connected to a pump, which fed into a solution of 1%

potassium iodide (KI) (Sigma Aldrich, Milwaukee, WI) and acetic acid (Mallinckrodt

Baker Inc., Phillipsburg, NJ) (Figure 7.1). The KI and acetic acid solution was combined

with the deionized water collected from the wet gauze and titrated with 0.001N sodium

thiosulfate (Sigma Aldrich, Milwaukee, WI). The amount of ClO2-generated by the

material was calculated (complete method outlined in the Appendices) and each ClO2-

generating sample was placed in a new vial. This process was repeated until no

measurable quantity of ClO2 was produced.

164

Figure 7.1: System used to quantitatively assess ClO2 production from segments of a

ClO2-generating material 165

1% KI + acetic acid

Pump

ClO2-generating material 166

Antimicrobial activity of ClO2-generating material: A zone of inhibition assay (ZOI)

was used to assess the duration of antimicrobial activity of the ClO2-generating material.

Staphylococcus epidermidis 10-16 was plated on Mueller Hinton agar (MHA) (BD,

Sparks, MD). Segments of ClO2-generating material (2.5cm; n = 3) were placed on

separate MHA plates. Plates were incubated at 37°C for 24 hours and the ZOI assessed.

Each sample was placed on a fresh MHA plate of Staphylococcus epidermidis 10-16, re-

incubated for 24 hours, and the new ZOI was determined. This process was repeated until no ZOI formed around the ClO2-generating material sample.

Assessing efficacy of ClO2-generating material in vivo: This animal study was only

performed after the protocol was approved by the Institutional Animal Care and Use

Committee (IACUC). National Institutes of Health (NIH) Guidelines for the Care and

Use of Laboratory Animals (NIH publication 85-23, revised 1985) were observed.

Animals were housed in American Association for the Accreditation of Laboratory

Animal Care approved facilities. The in vivo experimental model used was developed

by Buret et al. (1991). Four New Zealand white rabbits had ClO2-generating material

samples (one sample per animal), inoculated with Staphylococcus epidermidis 10-16, implanted into the peritoneal cavity. Four New Zealand white rabbits had control material samples (one sample per animal), inoculated with Staphylococcus epidermidis

10-16, implanted into the peritoneal cavity. Samples were explanted after 7 days, and the

sample and local host tissue evaluated.

Material preparation: All ClO2-generating material samples were prepared aseptically, while control samples were sterilized with ethylene oxide. The ClO2-generating and 167

control materials were cut such that each had a surface area of 4cm2. The samples had a

1 /16 inch hole drilled at each end of the samples for suture anchor points. All material

samples were incubated in a bacterial culture of Staphylococcus epidermidis 10-16

(7.8×107 CFU/ml) for 3 hours at 37°C prior to implantation.

Implant procedure: New Zealand white rabbits were anesthetized with a 1ml/kg intramuscular injection of a mixture of ketamine (100mg/kg), xylazine (920mg/kg), and

acepromazine (10mg/ml). The abdomen was shaved and the area cleaned with Nolvasan.

A midline incision was performed through all layers. Each animal received one implant,

inserted into the peritoneal cavity and sutured to the internal peritoneal fascia. The

abdomen was closed and the animals were observed for 7 days post-surgery for signs of

peritonitis.

Explant procedure: The animals were anesthetized with a 1ml/kg intramuscular

injection of a mixture of ketamine (100mg/kg), xylazine (920mg/kg), and acepromazine

(10mg/ml), and the peritoneal cavity was reopened. The implants were located and the

area of implantation was visually evaluated for gross indications of infection and tissue necrosis. The material samples were removed and prepared for analysis. Each sample had 0.5cm removed and fixed in 3% gluteraldehyde for analysis with scanning electron microscopy (SEM). The remainder of the sample was analyzed for the presence of viable bacteria using a viability counts assay.

Histological analysis: The tissue local to the implant site was removed and fixed in

Histochoice (Amresco, Solon, OH). Tissue samples were processed and paraffin embedded. Sections of the embedded samples (6µm) were cut and were stained with 168

hematoxylin and eosin to assess the tissue response to the implants.

Immunohistochemical analysis of the local tissue for the presence of macrophages was

performed using RAM-11 (DakoCytomation, Carpinteria, CA), a cytoplasmic antigen specific for rabbit macrophage cells. Sections of embedded samples (6µm) were

deparaffinized and incubated with the RAM-11 antibody (1:200 in 2% BSA) for 60 minutes at 37°C. After incubation with a peroxidase conjugated goat anti-mouse IgG secondary antibody (1:200 in DCF-PBS) (Chemicon, Temecula, CA), the sections were incubated with streptavidine HRP (DakoCytomation, Carpinteria, CA) for 10 minutes and

the macrophages stained with DAB chromogen substrate (DakoCytomation, Carpinteria,

CA). The sections were counterstained with methyl green.

Macrophage Quantification: The number of macrophages present in the tissue

associated with the material implants was quantified. Using a 6 × 6 grid, with each grid

being 27µm × 27µm, counts of the number of macrophages per unit area were performed

on RAM-11 stained tissue sections. For each implant three separate segments of tissue

were assessed, with counts performed in five regions within each segment.

Viability counts analysis: The material samples were scraped and vortexed in 0.1M

PBS to remove adhered bacteria. A dilution series of the bacteria solution was plated on

TSA plates and incubated for 24 hours at 37°C. Counts of viable colonies were made.

Scanning electron microscopy (SEM): SEM was used to visualize bacterial adhesion

on the surface of control and ClO2-generating materials and to assess the effects of ClO2 production on localized host cells. After fixing the material samples in 3% gluteraldehyde, they were dehydrated using decreasing dilutions of alcohol, critical point 169

dried and sputter coated with gold. Samples were imaged using a JEOL JSM-820 scanning electron microscope (JEOL Ltd., Peabody, MA).

Statistics: Data are presented as the mean ± the standard error mean. Significance between the different conditions was determined using a 2-tailed unpaired student t-test.

A p ≤ 0.05 was considered statistically significant. 170

Results

ClO2 production: Results from the quantitative assay of ClO2 production indicate that

measurable quantities of ClO2 were produced for 7 days (Figure 7.2). In the first 24

-6 hours 10.3×10 moles of ClO2 was produced, with decreased production thereafter.

Antimicrobial activity of ClO2-generating material: Results from the zone of

inhibition (ZOI) study show that the bactericidal activity of the ClO2-generating material

is 5 days, with reduced bactericidal activity after 24hrs (Figure 7.3). After 3 days, no

bactericidal activity was observed at the ends of the ClO2-generating material.

Gross analysis of explants: The implants and surrounding tissue were visually assessed

in situ prior to being explanted for indications of infection and tissue injury. One ClO2- generating material sample exhibiting gross signs of infection at the suture anchor points, with pus at the ends of the implant (Figure 7.4). There were no gross indications of tissue injury associated with the material implants.

Histological analysis: Relative to control tissue samples of the internal peritoneal wall

(no material samples implanted), tissue that had been associated with either the control or

ClO2-generating material exhibited indications of inflammation (Figure 7.5). Comparing

the tissue response between control and ClO2-generating material samples suggested no

differences in the mesothelial lining of the peritoneal cavity. Visual assessment of the

tissue samples stained with RAM-11 suggested an increased number of macrophages

were associated with the areas of inflammation for tissue associated with the ClO2- 171

Figure 7.2: Production of ClO2 in vitro by samples of a ClO2-generating material

172

8 ) -6

(x10 6 2

4 moles ClO moles 2

012345678 Time (days)

173

Figure 7.3: Zone of inhibition (ZOI) assay assessing bactericidal activity of ClO2- generating material has been activated for: (A) 24 hours; (B) 48 hours; (C) 72 hours; (D)

96 hours; (E) 5 days; (F) 6 days.

174

(A) (B)

ClO2-generating material

(E) (F) 175

Figure 7.4: Visual assessment of the material samples in situ at the time of explantation:

(A) Control material; (B) ClO2-generating material; (C) ClO2-generating sample that exhibited gross indications of infection at the suture anchor points. 176

(A) Control material

Peritoneal wall

(B)

ClO -generating material 2

177

Figure 7.5: Histological assessment (hematoxylin and eosin stain) of peritoneal wall tissue that was in contact with: (A) Control material (1×); (B) Control material (10×); (C)

ClO2-generating material (1×); (D) ClO2-generating material (10×); (E) Negative control

tissue (no implant) (10×)

178

(A) Inflammation (B)

(E) 179

generating material samples, relative to the control material (Figure 7.6). Quantitative

assessment of the number of macrophages present in the area of tissue inflammation

showed that significantly more (p < 0.05) macrophages were associated with the tissue

from the ClO2-generating material implants than the control material implants (Figure

7.7).

Viability counts analysis: Counts of colony forming units of viable bacteria removed

from each sample were performed. The results are summarized in Table 7.1. Viable

bacteria were recovered from all but one control sample. The only ClO2-generating material sample from which viable bacteria were recovered was the sample which exhibited gross indications of infections at the suture anchor points.

Scanning Electron Microscopy (SEM): Evaluation of the explanted materials with

SEM revealed both viable and nonviable host cells on the surface of the ClO2-generating material (Figure 7.8). Viable macrophages were present on the surface of the ClO2- generating material phagocytozing bacteria adhered to the surface. Viable host cells were present on the surface of the control implants.

180

Figure 7.6: Immunohistochemical staining with RAM-11 of macrophages associated with areas of tissue inflammation in the peritoneal wall tissue that were in contact with:

(A) Control material (10×); (B) ClO2-generating material (10×); (C) Negative control (no

implant) (10×)

181

(A)

RAM-11 positive cell

(B)

Figure 7.7: Macrophage densities in tissue contacting control and ClO2-generating material samples. There was a significantly higher density of macrophages in the tissue

associated with the ClO2-generating material (p < 0.05).

183

250 * ) 2 200

150

100

50 Macrophage Density (cells/mmMacrophage Density

0 Control ClO2-generating material 184

Material Sample Number CFU/cm2 Control C-1 218 C-2 94 C-3 125 C-4 0 ClO2-generating material ClO2-1 0 ClO2-2 0 ClO2-3 0 * ClO2-4 2563

Table 7.1: Colony forming unit (CFU) counts of viable bacteria recovered from explanted material samples

* Sample that exhibited gross evidence of infection at suture anchor site 185

Figure 7.8: Scanning electron microscopy images of the surfaces of explanted material samples: (A) Control material; (B) ClO2-generating material 186

(A1) (A2)

(B1) (B2) Nonviable host cell

Viable host cell 187

Discussion

A series of in vitro and in vivo experiments were performed to assess the efficacy

of a ClO2-generating material. In Chapter 6, the in vitro bactericidal activity of the ClO2-

generating material on adhered bacteria was assessed. The in vitro studies presented in this chapter were designed to further elucidate the production of ClO2 and the maintenance of bactericidal concentrations of ClO2. The in vivo study was designed to determine the efficacy of the bactericidal activity of the ClO2-generating material in an in

vivo environment and to assess the localized tissue response to ClO2, since ClO2 exhibits indiscriminate biocidal activity.

Analysis of ClO2 production from the ClO2-generating material indicates a rapid

production of ClO2 within the first 24 hours, with decreasing amounts produced for the following 6 days. However, while the duration of ClO2 production by the materials was

7 days, results from ZOI assay indicate that bactericidal concentrations were only

maintained for 5 days. Since the smaller amounts of ClO2 being produced from 2-4 days

(7.0×10-5 - 4.8×10-6 moles) was sufficient for bactericidal activity, the duration of

bactericidal activity of the ClO2-generating material could be extended without the

incorporation of additional precursor chemistries. The stable production of a lower

amount of ClO2, rather than the rapid production of ClO2 within the first 24 hours, could

extend duration of bactericidal activity by the ClO2-generating material.

Several different approaches towards controlling the release of antimicrobials incorporated into biomaterials have been investigated. Currently, the most prevalent approaches are to alter the methods of attaching or incorporating the antibiotics onto or 188

into the biomaterial (Dunn et al. 1994; Trecant et al. 1997; Duran 2000) or to apply

external coatings to lengthen the release profile of the drug (Kwok et al. 1999a). To

preserve the current technologies used to produce ClO2 from a polymer, the latter method

would be an appropriate technique to investigate. Various hydrogel formulations, such as

gelatin (Ginalska et al. 2005) and poly(acrylic) acid and gelatin (Changez et al. 2004),

have been used to alter the release characteristics of antibiotics from material surface.

Other hydrogel formulations, such as sodium alginate (Kulkarni et al. 2001) and

polyacrylamide-chitosan (Risbud and Bhonde 2000), have been tested for the controlled

release of antibiotics, although not in association with release from biomaterials. The

development of a hydrogel coating to control the release of ClO2 by the polymer would

be appropriate, since the ClO2-generating material is humidity activated and hydrogels readily absorb water. One potential complication could be that the water absorbed in the

hydrogel may retain the ClO2, since ClO2 is highly soluble in water, thereby affecting the

concentration of ClO2 actually released. An alternative material to hydrogels would be a

thin polymer overlay, such as a plasma-deposited film (PDF) (Kwok et al. 1999a). One

constraint in the use of PDFs to control the release of ClO2 is that they must also be

hydrophilic and absorb threshold levels of moisture such that ClO2 production occurs.

From the histological results from the in vivo experiment, no adverse tissue response was observed in association with the ClO2-generating material. An inflammatory response was observed in the tissue contacting both the control and ClO2- generating materials and quantitative assessment of the macrophages present in the tissue showed that significantly more (p < 0.05) macrophages were present in the tissue from 189

the ClO2-generating material samples, relative to the control material samples. The

increased number of macrophages indicates that the ClO2 produced by the material,

and/or possible leaching of the precursor chemistries, contributed to the acute

inflammatory response observed. Longer-term implant studies would determine whether

the inflammatory response resolves or becomes a chronic response. The inflammatory

response observed with the control material suggests that the chemistry of the polymer

also contributed to the tissue response. To reduce the inflammatory response observed,

alternative polymers could be considered. When selecting an appropriate bulk polymer

to contain the microspheres, consideration needs to be given to the processing and

polymerization conditions. Low moisture levels are required to limit activation of the

sodium chlorite and acid precursor chemistries, and the degradation temperature of

sodium chlorite (160°C) needs to be considered during the polymerization process.

One ClO2-generating material sample had viable bacteria associated with it at the time of explantation. This sample exhibited gross evidence of infection at the suture anchor points. One hypothesis is that the infection was actually a secondary infection that originated from the suture, since in situ observation of the site of infection indicated that it was localized to the suture anchor points of the material. Since the ClO2 concentrations produced by the material would not be capable of sterilizing a contaminated suture, and the in vitro data indicate that the ClO2-generating implants only

produce ClO2 for ≤ 3 days from the ends of the implant, the infection could have

originated and become established on the suture, before progressing to the implant surface. Images from SEM of the infected sample provided no indication of significant 190

bacteria adhesion to the specific ClO2-generating material sample, further substantiating

that the infection likely did not originate from the ClO2-generating material. SEM

analysis of all material samples indicated the presence of host macrophages. The ClO2- generating material samples had evidence of both viable and nonviable host cells adhered to the surface. Since the production of bactericidal concentrations of ClO2 by the material samples lasts 5 days and the implantation time was 7 days, it is not possible to determine when viable host cells began associating with the surface of the material.

Shorter implant times would help establish the short-term effects of ClO2 on the host cells

associated with the material surface.

When analyzing the bactericidal efficacy of antimicrobial biomaterials, it is

necessary to select an appropriate in vivo infection model. While there are established

models of infection, often specific to the final device that the antimicrobial material is a

part of, there is no concensus amongst investigators as to whether to inoculate the

material with bacteria pre- or post-implantation (Buret et al. 1991; Darouiche et al. 2002;

Ghiselli et al. 2002; Gottenbos et al. 2002; Hernandez-Richter et al. 2003). Inoculating the materials pre-implantation results in antimicrobial materials being implanted with less viable bacteria associated with them. However, inoculating materials post-implantation makes it impossible to ensure that each material sample is exposed to the same bacterial load since the innate host immune response will contribute to the bactericidal activity of planktonic bacteria that have not adhered to the material. Since both methods have advantages and disadvantages, it is probably beneficial to use both experimental models. 191

These in vitro experiments indicate that the current formulations of ClO2-

generating materials produce bactericidal concentrations of ClO2 for 5 days, with the bulk

of the ClO2 being released within the first 24 hours. The addition of an external polymer

layer to prolong the release of ClO2 from the material could lengthen the duration of bactericidal activity achieved by the material. Results from the in vivo study suggest that

the ClO2-generating material provides short-term bactericidal activity against

Staphylococcus epidermidis 10-16, with no evidence of an adverse tissue response at the

material/tissue interface in response to the ClO2-generating materials. These studies suggest that a ClO2-generating material could be used as a surface coating for

biomaterials for the prevention of biomaterial-associated infections. 192

8. INTRODUCTION: THE DEVELOPMENT OF MICROBIAL RESISTANCE

TO ANTIBIOTICS AND BIOCIDES

While the discovery and utilization of antibiotics and biocides is generally

considered to be an innovation of the 20th century, historical evidence indicates that both

were used much earlier. Burdon, Sanderson, Lister, Roberts, and Tyndall were

investigating the antimicrobial affects of mould in the 19th century (Selwyn 1980), long

before they were first utilized clinically in the 1940’s. Reports of biocide utilization date

back centuries, with alcohol first being utilized in early AD (Russell 2002). The

antimicrobial effects of silver and copper, for storing water, and honey and vinegar for

the cleansing of wounds, were recognized, although not understood (Russell 2002).

Antibiotics

The primary modern antibiotics used in the treatment of bacterial infections are β-

lactams, glycopeptides, aminoglycoside-aminocyclitols, tetracyclines, chloroamphenicol, macrolides, mupirocin, lincosamides, sulfonamides, diaminopyrimidines, rifamycins, and

quinolones. The antimicrobial mechanisms of antibiotics are well defined and specific

(Russell 1996). One mode of bactericidal action of antibiotics, common to β-lactams and

glycopeptides, is the inhibition of peptidoglycan synthesis, which interferes with the

integrity of the bacteria cell wall. Alternatively, antibiotics can interfere with: protein

synthesis (tetracyclines, chloramphenicol, mupirocin, macrolides, aminoglycoside-

aminocyclitols) or nucleic acid synthesis (sulphonamides, diaminopyrimidines), or can

inhibit: RNA polymerase (rifamycins) or DNA gyrase (quinolones). 193

Biocides

The major biocides currently in use are phenols, organic acids, halogens,

aldehydes, quaternary ammonium compounds (QACs), bisbiguanides, and organomercurials. How biocides cause bactericidal activity is less clearly defined.

Biocides are known to interact with bacterial cell walls or envelopes (gluteraldehyde), disrupt cytoplasmic membrane integrity (cationic agents), dissipate the proton-motive force (organic acids and esters), inhibit membrane enzymes (thiols), act as alkylating agents (ethylene oxide), cross-linking agents (aldehydes) and intercalating agents

(acridines), or interact with other chemical groups within the bacteria (Denyer 1998).

Unlike antibiotics, which have specific target sites within the bacteria, most biocides are believed to have multiple target sites (Denyer 1998), which makes it difficult to determine exact mechanisms of bactericidal activity.

Acquired microbial resistance

Bacterial resistance to both antibiotics and biocides can be either intrinsic, an innate property of the organism, or acquired. Acquired resistance results from mutation of a target site or acquisition of additional genes via a plasmid or transposon that encodes a specified resistance mechanism (Russell 1991; Russell and Chopra 1996; Sefton 2002).

The mechanisms of bacterial resistance to antibiotics have been extensively researched, and are generally well understood. How bacteria develop resistance to biocides is less clearly defined.

194

Mechanisms of antibiotic resistance

Antibiotic resistance may be due to alterations to the antibiotic target within the cell, thereby limiting the interactions of the antibiotic and eliminating the bactericidal effect. Such mechanisms are common in strains of bacteria resistant to macrolides and sulfonamides. Bacterial mutations that result in the enzymatic modification or destruction of the antibiotic are another common mechanism resulting in bacterial resistance to antibiotics. For example, chemical modification to aminoglycosides compromises the ability of the antibiotic to bind to the ribosomal target within the bacteria (Wright et al. 1998; Wright 1999), while the production of β-lactamases inactivates β-lactams by hydrolyzing the β-lactam ring in the molecule (Thomson and

Smith Moland 2000), thereby imparting resistance to the bacteria. Active efflux of the antibiotic out of the bacterial cell, limiting the concentration of the drug within the cell, is a common mutation observed with tetracyclines (Chopra and Roberts 2001) and quinolenes (Poole 2000a; Poole 2000b) in both gram-negative and gram-positive bacteria. Alterations in the permeability of the bacterial cell to the antibiotic will affect the antimicrobial activity, since interactions with an intracellular target is required to achieve the bactericidal effects. Such mutations are common in gram-negative bacteria, against β-lactam antibiotics, where alterations to the outer membrane cause a significant change in the cell permeability (Hancock and Woodruff 1988; Chen and Livermore 1993;

Charrel et al. 1996; Martinez-Martinez et al. 1996). This form of resistance appears to only be significant, however, when working synergistically with additional mechanisms 195

of resistance, such as active efflux and enzymatic activity (e.g. β-lactamases) (Nikaido

1994; Hancock 1997).

Mechanisms of biocide resistance

Biocide resistance is generally believed to be uncommon, possibly because of the

multiple modes of bactericidal action of most biocides (Russell 1998). While the issue of

biocide resistance is generally believed to be a recent development, initial concerns were

expressed as early as 1887, when Kossiakoff discovered that bacteria could develop

resistance to increasing concentrations of some biocides (boric acid and mercuric

chloride) (Kossiakoff 1887). In the 1950’s evidence of biocide resistance in gram- negative bacteria was discovered (Davies 1954) and, specifically, Pseudomonas aeruginosa was found to develop resistance to cationic biocides (Lowbury 1951). The most common forms of biocide resistance are due to alterations in cell permeability or enhanced biocide efflux (McDonnell and Russell 1999; Schweizer 2001).

The structural organization and composition of the outer membrane for gram- negative and gram-positive bacteria are substantially different (Figure 8.1). For example, gram-positive bacteria possess a permeable peptidoglycan layer protecting the cell membrane, whereas gram-negative bacteria have an external lipopolysaccharide and phospholipid bilayer. The development of resistance in gram-positive bacteria due to permeability changes is uncommon. However, there is a significant amount of research that suggests permeability changes are responsible for biocide resistance in gram- negative bacteria (McDonnell and Russell 1999). The mechanisms suggested include changes in: bacterial cell surface hydrophobicity (Tattawasart et al. 1999), outer 196

Figure 8.1: The cell structure for: (A) Gram-positive bacteria; (B) Gram-negative bacteria (In: Molecular Medicine©. Copyright, Daniele Focosi. 2001-2005. Available at http://www.mm.interhealth.info: Educational Materials: Physiology Of Bacteria.

Accessed 2005)

197

(A) (B) 198

membrane ultra structure (Tattawasart et al. 2000b), outer membrane protein composition

(Gandhi et al. 1993; Winder et al. 2000), and outer membrane fatty acid composition

(Jones et al. 1989; Guerin-Mechin et al. 1999; Mechin et al. 1999; Guerin-Mechin et al.

2000). However, no genetic mutations have been identified that could be attributed to the

physical changes observed, nor has a specific outer membrane permeability defect been

isolated (Poole 2002). Resistance due to alterations in cell permeability is often acting

synergistically with an efflux system.

Efflux systems allow bacteria to remove a variety of structurally unrelated substances, including antibiotics, resulting in a broad spectrum of resistance. There are currently five classes of recognized multidrug efflux systems: the small multidrug resistance (SMR) family (Chung and Saier 2001), the major facilitator superfamily

(MFS) (Pao et al. 1998), the ATP-binding cassette (ABC) family (van Veen and Konings

1998), the resistance-nodulution-division (RND) family (Nikaido 1998; Zgurskaya and

Nikaido 2000), and the multidrug and toxic compound extrusion (MATE) family (Brown

et al. 1999) (Figure 8.2). The SMR family efflux pumps are composed of small peptides

(approximately 110 amino acid residues), with four predicted transmembrane domains

(Grinius et al. 1992; Paulsen et al. 1995), and are responsible for the efflux of quaternary

ammonium compounds. The MFS family efflux pumps contains at least 18 families

(Paulsen et al. 1998) and are considerably larger than the SMR family, being 400-600

amino acid residues in length, with 12 or 14 putative transmembrane domains. The RND

family pumps are primarily responsible for mediating efflux driven resistance for gram-

negative bacteria. These large pumps (1000 amino acids long) have 12 transmembrane 199

Figure 8.2: Illustration of the main types of bacterial efflux pumps: (A) MFS family; (B)

RND family; (C) ABC family. The MATE and SMR families look structurally similar to

MFS, but are designated as distinct families based on phylogenic diversity (MATE) or

size (SMR) (reprinted from Genetics and Molecular Research, 2(1), Schweizer, H.P.,

Efflux as a mechanism of resistance to antimicrobials in Pseudomonas aeruginosa

(A) (B) (C) 201

domains with two periplasmic loops (Saier et al. 1994; Guan et al. 1999). Efflux pumps

are either activated in response to an environmental signal, or a mutation in a regulatory

gene (Levy 2002). While the contributing role of some of these multidrug efflux systems

in biocide resistance has yet to be established, several different families have been identified. In Staphylococcus aureus the efflux systems tend to be members of the SMR and MFS superfamilies, and include QacA/B (Littlejohn et al. 1992; Paulsen et al. 1996),

Smr (Littlejohn et al. 1992; Paulsen et al. 1996), QacG (Heir et al. 1999), QacE∆1

(Kazama et al. 1998), and NorA (Kaatz and Seo 1995; Noguchi et al. 1999). Active

efflux is recognized as one of the specific forms of biocide resistance in Escherichia coli, where the RND family efflux system, AcrAB, removes biocides such as triclosan, chlorhexidine, QACs, as well as numerous antibiotics. Gupta et al. (1999) discovered a plasmid-encoded silver specific efflux system in Salmonella typhimuriuim.

Biocide resistance due to inactivation of the chemical agent is comparatively rare, although it has been reported with organomercurials (McDonnell and Russell 1999;

Miller 1999). While antibiotics have specific targets within the bacterial cell, biocides appear to affect multiple cellular components, which is one reason why mutations to the biocide target site(s) as a cause of biocide resistance is uncommon (McDonnell and

Russell 1999), although target site mutations resulting in a resistance to triclosan have been reported in Escherichia coli (Heath et al. 1998; McMurry et al. 1998).

A correlation between antibiotic and biocide resistance?

There have been suggestions about a possible link between biocide and antibiotic resistance (Levy 2000). Since the modes of bactericidal action for the two groups of 202

antimicrobials are dissimilar with antibiotics having very specific target sites within the

bacterial cell, while biocides generally appear to have multiple possible target sites, it

seems unlikely that the development of antibiotic and biocide resistance are associated.

However, when we consider that the modes of resistance are similar (altered bacterial cell

permeability, active agent efflux etc.), the possibility of a correlation of antibiotic and

biocide resistance becomes a viable hypothesis. Research has indicated that antibiotic

resistant bacteria do not typically show a decreased susceptibility to biocides (Russell

1999; Russell 2000) and neither do biocide resistant bacteria generally show an increased

resistance to antibiotics (Tattawasart et al. 1999; Suller and Russell 2000; Tattawasart et

al. 2000a; Tattawasart et al. 2000b). However, there does seem to be a commonality with

active efflux pumps that remove both biocides and antibiotics from bacteria, discovered

in Escherichia coli and Pseudomonas aeruginosa (Chuanchuen et al. 2001; Levy 2002).

Resistance to chlorine dioxide

Chlorine dioxide (ClO2) is a halogen-releasing biocide being investigated for its

potential to prevent infections associated with biomaterials. ClO2 gas has being assessed

as a new method of sterilization for biomaterials (Chapter 3 and Chapter 4) and a

selection of polymers that produce ClO2 under specific environmental conditions have

been assessed for the prevention of biomaterial-associated infections in situ (Chapter 6

and Chapter 7). While the specific mechanisms of antimicrobial activity of ClO2 remain unclear, its bactericidal activity is believed to stem from the fact that the ClO2 exists as a free radical and is a strong oxidizing agent that interacts with electron-rich amino acids,

including cysteine, tryptophan, and tyrosine, causing protein degradation (Noss et al. 203

1983; Noss et al. 1986). There have been no reported cases of acquired resistance to

ClO2, although some strains of yeast, such as Debaryomyces hansenii, have an innate

reduced susceptibility to the biocidal effects of ClO2 (Ramirez-Orozco et al. 2001).

While assessing the bactericidal activity of a ClO2-generating biomaterial, a spontaneous

mutant of Staphylococcus epidermidis 10-16, with a reduced susceptibility to the

bactericidal effects of ClO2, was isolated.

Research Plan

Since acquired resistance to chlorine dioxide has not previously been reported, the

experiments presented in the following specific aim will investigate possible mechanisms

of resistance behind this new strain. Determining the mechanism of reduced

susceptibility of the mutant to the effects of chlorine dioxide will further elucidate the

area of biocide resistance and aid in the development of methods to avoid acquired

biocide resistance. These studies are presented in Chapter 9.

Specific Aim #5: To investigate a potential mechanism of resistance to chlorine dioxide of a Staphylococcus epidermidis mutant with a reduced susceptibility to the bactericidal effects of chlorine dioxide. Hypothesis #5: The mechanism of resistance to chlorine

dioxide for the Staphylococcus epidermidis mutant is through the over-production of a

molecule that is inhibiting the effects of chlorine dioxide. 204

9. ACQUIRED MICROBIAL RESISTANCE TO CHLORINE DIOXIDE

Introduction

While the antimicrobial mechanisms of antibiotics are well defined and specific

(Russell and Chopra 1996), how biocides cause antimicrobial activity is more ambiguous.

Unlike antibiotics, which have specific target sites within the bacteria, most biocides are

believed to have multiple target sites (Denyer 1998), which makes it difficult to

determine the exact mechanisms of bactericidal activity and the exact mechanisms of

resistance.

Biocide resistance is generally believed to be uncommon, possibly because of the

multiple modes of bactericidal action of most biocides (Russell 1998), but is becoming

more recognized as occurrences of biocide resistance are being reported (Dance et al.

1987). The most common forms of biocide resistance are due to alterations in cell

permeability or enhanced biocide efflux (McDonnell and Russell 1999; Schweizer 2001).

Biocide resistance due to inactivation of the biocide is comparatively rare, although it has

been reported (McDonnell and Russell 1999; Miller 1999). Since biocides appear to

affect multiple cellular components, mutations to the biocide target site(s) as a cause of

biocide resistance is also uncommon (McDonnell and Russell 1999).

Chlorine dioxide (ClO2) is a halogen-releasing biocide that is being investigated

for its potential for the prevention of biomaterial-associated infections. One form of ClO2 technology being assessed is a polymer that produces ClO2 under specific environmental

conditions, which could be used as a surface coating for biomaterials for the prevention 205

of infections in situ. During the course of testing the bactericidal activity of a selection of

ClO2-generating materials (Chapter 6 and Chapter 7), a spontaneous mutant of

Staphylococcus epidermidis with a reduced susceptibility to ClO2 was isolated

(SP030724) (Figure 9.1). A review of the work by other investigators regarding acquired microbial resistance to biocides yielded no evidence that microbes develop resistance to

ClO2. As such, the primary focus of these experiments was to investigate the underlying mechanism of resistance to ClO2. 206

Figure 9.1: Zone of inhibition (ZOI) assay assessing the bactericidal activity of a ClO2- generating material against: (A) Staphylococcus epidermidis 10-16, a wildtype strain; (B)

One of the possible mutant colonies isolated from the edge of the ZOI from experiments with 10-16 (SP030724)

207

(A) Example of a possible (B) mutant colony

ClO2-generating material 208

Materials and Methods

Chlorine dioxide-generating material: Bernard Technologies Incorporated (Chicago,

IL) has manufactured a polymer that produces ClO2 under specific environmental conditions. The formulation used for these experiments was a polymer rod extrusion with precursor chemistries that were humidity activated, with ClO2 being generated when a threshold environmental humidity level was attained.

Zone of inhibition (ZOI) Assay: This assay is commonly used to assess the bactericidal efficacy of antimicrobial agents. Briefly, Staphylococcus epidermidis 10-16 (wildtype

Staphylococcus epidermidis) was spread across a Mueller-Hinton agar (MHA) plate

(vWR Scientific, West Chester, PA). A 2mm segment of a ClO2-generating material was placed on the plate and the plates were incubated overnight at 37°C. The ZOI, or area of no bacteria growth that formed around the ClO2-generating material, was assessed.

Involvement of extracellular factors: The supernatant from SP030724 was assessed to determine whether the mechanism of resistance involved extracelluar factors produced by the mutant. To determine this, a modified ZOI (M-ZOI) assay was used (Figure 9.2).

Briefly, tryptic soy broth (TSB) (vWR Scientific, West Chester, PA) was inoculated with

SP030724 and cultured overnight at 37°C. The bacteria were isolated from the overnight culture by centrifugation (16,000×g for 15 minutes) and the supernatant was collected.

Between 1 and 3 colonies of 10-16 were spread across a Mueller-Hinton agar plate to form a lawn of bacteria. A 50µl sample of the SP030724 supernatant (SP030724S) was placed on the plate and allowed to dry. A 2mm sample of the ClO2- generating material was then placed on the dry supernatant. The plates were incubated at 37°C overnight and 209

Figure 9.2: Schematic diagram representing the modified ZOI (M-ZOI) assay used for testing the supernatant produced by SP030724

210

211

the ZOI was assessed.

Degree of molecular hydrophobicity: The degree of hydrophobicity of the molecule(s) produced by SP030724 was investigated. SP030724 supernatant was incubated with hexane (Sigma Aldrich, St. Louis MO) at 37°C for 2 hours. The hexane and aqueous phase were separated and the hexane was allowed to evaporate for > 24 hours. Any residues were then resuspended in TSB. The separate phases were each assessed using the M-ZOI assay. The plates were incubated at 37°C overnight and the ZOI was assessed.

Temperature stability of molecule(s): The temperature stability of the molecule(s) present in the SP030724 supernatant was assessed by maintaining the supernatant at 4°C or 21°C for ≥ 24 hours, or by heating the supernatant to 60, 80 or 100°C for 15 minutes.

The treated supernatant samples were each assessed using the M-ZOI assay. The plates were incubated at 37°C overnight and the ZOI was assessed.

Molecular weight range of molecule(s): SP030724 supernatant was filtered using centrifugal devices (Millipore Corporation, Bedford, MA) or dialysis membranes

(Spectrum Laboratories Inc., Savannah, GA) to separate molecules of different molecular weights into solutions that contained molecules that were: <5kDa; >5kDa; >10kDa;

>30kDa; >50kDa; and >100kDa. The supernatant solutions with the different molecular weight ranges were individually assessed using the M-ZOI assay. The plates were incubated at 37°C overnight and the ZOI was assessed.

Testing the “protected” colonies: The bacteria colonies that grew in the ZOI when the

ClO2-generating material was tested against 10-16 with SP030724 supernatant were isolated and their phenotypical response to ClO2 assessed. Between 1 and 3 of the 212

isolated colonies were spread across a Mueller-Hinton agar plate to form a lawn of

bacteria. A 2mm segment of a ClO2-generating material was placed on the plate and the

plate was incubated overnight at 37°C. The ZOI that formed around the ClO2-generating material was assessed.

Analysis of SP030724 supernatant: The supernatant isolated from an SP030724 culture was analyzed for the presence of contaminating SP030724 bacteria. A 50µl sample of the SP030724 supernatant was placed on a TSA plate and allowed to dry. The plates were incubated at 37°C overnight and assessed for the formation of bacteria colonies.

The supernatant collected from an SP030724 solution was also tested to see whether filtration of the supernatant affected the protection of 10-16 bacteria when they are exposed to ClO2 with the filtered SP030724 supernatant present. A sample of SP030724

supernatant was filtered through a 0.2µm filter. Filtered and non-filtered SP030724

supernatant were both assessed using the M-ZOI assay. The plates were incubated at

37°C overnight and the ZOI was assessed.

SP030724 susceptibility to increasing concentrations of ClO2: The susceptibility of

SP030724 to the effects of increased concentrations of ClO2 was assessed. Between 1

and 3 colonies of the SP030724 were spread across Mueller-Hinton agar plates to form a

lawn of bacteria. A 2mm and 4mm segment of a ClO2-generating material was placed on

separate plates and the plates were incubated overnight at 37°C. The ZOI that formed

around the ClO2-generating material was assessed.

213

Results

Involvement of extracellular factors: Supernatant produced by the mutant SP030724 was collected and analyzed to determine whether it would afford protection against ClO2 to a wildtype strain of Staphylococcus epidermidis. Results from the M-ZOI assay showed that a similar phenotypical response to the one observed for SP030724 when tested against the ClO2-generating material was seen for 10-16 proveided with the

SP030724 supernatant (Figure 9.3). This suggests that SP030724 is producing

extracellular molecule(s) that are released into the supernatant and inhibiting/inactivating

the oxidative effects of ClO2.

Degree of molecular hydrophobicity: The SP030724 supernatant was analyzed to

assess the degree of hydrophobicity of the molecule(s) responsible for the reduced

susceptibility to ClO2. SP030724 supernatant was incubated with hexane in which

hydrophobic molecules would have increased solubility, relative to the aqueous TSB.

Results from the M-ZOI assay indicate that the molecule(s) responsible for the reduced

susceptibility to ClO2 remained in the aqueous phase (Figure 9.4), suggesting that the

molecule(s) have a greater degree of hydrophilicity, and therefore does not have a lipid-

like structure.

Temperature stability of molecule(s): The temperature stability of the molecule(s) was

analyzed by exposing the SP030724 supernatant to either reduced or elevated

temperatures. The results from the M-ZOI assay indicate that inactivation of the

molecule(s) of interest occurs with temperatures ≥ 60°C (Figure 9.5), but is maintained at

4°C and 21°C (Figure 9.4B). This suggests that the molecule(s) may be a protein, rather 214

Figure 9.3: ZOI formed for: (A) 10-16 with 10-16 supernatant; (B) 10-16 with SP030724 supernatant 215

(A) (B)

Area covered by supernatant 216

Figure 9.4: Degree of molecular hydrophobicity of the molecule(s) of interest in the

SP030724 supernatant was assessed by diffusion of the molecule(s) from the aqueous supernatant into hexane. (A) Hexane phase; (B) Aqueous phase

217

(A) (B) 218

Figure 9.5: Temperature stability of the molecule(s) of interest in the SP030724 supernatant was assessed at: (A) 4°C; (B) 21°C; (C) > 60°C 219

(A)

(B)

than a carbohydrate, since carbohydrates are generally more stable at elevated

temperatures, while proteins typically denature at temperatures exceeding 50°C.

Molecular weight range of molecule(s): SP030724 supernatant was separated into solutions with molecular weight ranges of: <5kDa, >5kDa, >10kDa, >30kDa, >50kDa, and >100kDa, which were analyzed to determine the molecular weight range of the molecule(s) of interest. The results from the M-ZOI assay show that the molecule(s) that resulted in the reduced susceptibility to ClO2 is >100kDa (Figure 9.6), indicating that if

the molecule is a protein, it is not a peptide, since the molecular weight of peptides is

typically less than 5kDa.

Testing “protected” colony forming units: The bacteria colonies that grow in the ZOI

when a ClO2-generating material is tested against 10-16 with SP030724 supernatant were

tested against ClO2 to determine whether their phenotype was similar to 10-16 or

SP030724. The results from the ZOI assay indicate that the phenotypical response of the

isolated colonies to ClO2 was similar to the phenotype observed for SP030724 (Figure

9.7).

Analysis of SP030724 supernatant: The supernatant isolated from an SP030724 culture

was analyzed for the presence of contaminating SP030724 bacteria. Visual assessment of

the TSA plate with SP030724 supernatant plated showed colonies present in the area

where the supernatant was placed (Figure 9.8A). The supernatant collected from an

SP030724 solution was also tested to see whether filtration of the supernatant affected the protection of 10-16 bacteria when exposed to ClO2. The results from the M-ZOI assay

indicate that filtration of the SP030724 supernatant removed the molecule(s) that were 221

Figure 9.6: Separation of SP030724 supernatant into solutions with different molecular weights: (A) < 100kDa; (B) > 100kDa

222

(A) (B)

223

Figure 9.7: ZOI assay assessing the bactericidal activity of a ClO2-generating material

against: (A) 10-16; (B) SP030724; (C) Colonies isolated from the ZOI when a ClO2- generating material is tested against 10-16 with SP030724 supernatant

224

(A)

(B)

(C 225

Figure 9.8: Analysis of the SP030724 supernatant. (A) A sample of the SP030724 supernatant on a TSA plate; (B) M-ZOI assay assessing the bactericidal activity of a

ClO2-generating material against 10-16 with non-filtered SP030724 supernatant; (C) M-

ZOI assay assessing the bactericidal activity of a ClO2-generating material against 10-16 with SP030724 supernatant filtered through a 0.2µm filter 226

(A)

(B) (C) 227

affording protection against the bactericidal effects of ClO2 to the 10-16 strain of

Staphylococcus epidermidis (Figure 9.8C).

SP030724 susceptibility to increasing concentrations of ClO2: The susceptibility of

SP030724 to the effects of increased concentrations of ClO2 was assessed. The results

from the ZOI assay indicate that increasing the concentration of ClO2 causes an increase

in bactericidal activity against SP030724, shown by the increasing size of the ZOI

(Figure 9.9), although the ZOI remains smaller than the ZOI for 10-16.

228

Figure 9.9: Susceptibility of SP030724 to increasing concentrations of ClO2 was assessed using a ZOI assay: (A) 10-16 with a 2mm sample of ClO2-generating material;

(B) SP030724 with a 2mm sample of ClO2-generating material; (C) SP030724 with a

4mm sample of ClO2-generating material 229

(A)

(B)

(C)

230

Discussion

While assessing the bactericidal activity of an antimicrobial material that

produces ClO2 using a zone of inhibition (ZOI) assay, the question of resistance to ClO2 was raised in response to the observation of some colonies at the edge of the ZOI that were notably larger than the surrounding colonies. Subsequent testing of several isolated colonies against the bactericidal effects of ClO2 revealed one spontaneous mutant with a

reduced susceptibility to ClO2 (SP030724). To date, there have been no reported cases of

acquired microbial resistance to ClO2, although some strains of yeast, such as

Debaryomyces hansenii, have an innate reduced susceptibility to the biocidal effects of

ClO2 (Ramirez-Orozco et al. 2001). In these studies, possible underlying mechanisms of

resistance to ClO2 were investigated.

The potential mechanisms of resistance to ClO2 being employed by the mutant

were investigated, beginning with the possibility that extracellular factors may be

inhibiting or inactivating the effects of ClO2. Analysis of the supernatant produced from

the culturing of SP030724 was conducted using a modified ZOI (M-ZOI) assay. The

results from this assay suggest that an extracellular molecule or molecules was at least

partially responsible for the mechanism of resistance. These results did not preclude the

involvement of alternative mechanisms of resistance, such as intracellular or

transmembrane factors. Further studies, focused on the identification of the molecule(s)

of interest in the SP030724 supernatant, suggested that the molecule(s) did not have a

lipid-like structure, and was potentially a large protein. 231

Concurrent with the studies designed to identify the potential protein further

investigations into the mechanisms of actions associated with the molecule(s) were being conducted. An analysis of the colonies that grow in the ZOI where SP030724

supernatant was placed on Staphylococcus epidermidis 10-16 determined that the bacteria

exhibited the same phenotype as SP030724 when tested against a sample of ClO2-

generating material. This result suggested that the protective molecule(s) produced by

SP030724 might actually be contaminating SP030724 in the supernatant. This

hypothesis was confirmed when colonies of bacteria grew from a sample of SP030724

supernatant plated on a TSA plate and when filtration of the SP030724 supernatant

resulted in 10-16 once again being susceptible to the effects of ClO2.

Once it was established that the mechanism of resistance to ClO2 was likely not

an extracellular molecule, other mechanisms of biocide resistance were more closely

examined. Changes in cell permeability and active efflux are the most common forms of

biocide resistance (McDonnell and Russell 1999; Schweizer 2001), while inactivation of

the biocide (McDonnell and Russell 1999; Miller 1999) and specific mutations to target sites (McDonnell and Russell 1999) are rare.

One of the complicating factors in hypothesizing a mechanism of resistance for

ClO2 is that its precise mechanism of biocidal action has yet to be established. However,

when considering the established mechanisms of resistance to biocides, it becomes easier

to propose viable hypotheses. Since this research has established that the mechanism of

resistace is not due to the production of an extracellular antioxidant, there are two

alternative mechanisms of resistance that are feasible: an intracellular antioxidant or an 232

extracellular membrane-bound molecule. Since ClO2 is a gas, and could diffuse across

the bacteria cell membrane, it is unlikely that active efflux is the mechanism responsible

for ClO2 resistance. The remaining established mechanisms of biocide resistance are

inactivation of the agent and mutation of the target site. While the precise mechanism of biocidal action of ClO2 is unknown, research has indicated that is does cause protein

degradation through the oxidation of electron-rich amino acids (Noss et al. 1983; Noss et

al. 1986). For mutation of the target site to be the mechanism of resistance to ClO2, multiple proteins would have to simultaneously mutate to inhibit the effects of ClO2

without having lethal effects on the mutant, since biocides commonly have multiple taget

sites. Given that this is an improbable sequence of events, the remaining mechanism is inactivation of ClO2, either via an intracellular molecule or through the expression of a

transmembrane molecule that interacts with ClO2.

Since cells do produce antioxidants, the intracellular inactivation of ClO2 as a mechanism of resistance seems plausible. Given that these experiments indicate that the mechanism of resistance for SP030724 against ClO2 is not through the production of an

extracellular molecule, perhaps an intracellular molecule is responsible for the

inactivation of ClO2. Investigation into the identification of potential proteins with established antioxidizing activities revealed one group of molecules produced by

bacteria. Glutathione (GSH) is an intracellular thiol produced by many bacteria and has

been shown to offer protection of bacteria against the oxidizing effects of chlorine and

hydrogen peroxide (Chesney et al. 1996). Ingram et al. (2003) showed that sodium

chlorite is highly effective at oxidizing GSH, causing complete oxidation at an oxidant to 233

GSH ratio of 1:4. Studies have also indicated that hydrogen peroxide appears to preferentially oxidize GSH, suggesting that GSH may be acting as a trap for oxidizing agents to protect more vital cell components (Chesney et al. 1996). Research has also shown that chlorine dioxide causes a reduction in intracellular GSH (Abdel-Rahman et al.

1980; Abdel-Rahman et al. 1985; Abdel-Rahman and Scatina 1985). It is possible that the over production of a molecule, such as GSH, could be the mechanism responsible for the reduced susceptibility of SP030724 to ClO2. Determining whether an intracellular molecule is responsible for the mechanism of resistance of SP030724 could be achieved by isolating the intracellular contents of the cells. Capillary electrophoresis could be used to identify differences in the concentrations of the intracellular contents of SP030724 and

10-16. If the intracellular contents could be suspended in a small enough volume, such that dilution effects do not reduce the effects of any antioxidants, the M-ZOI assay could be used to assess the presence of a protective molecule(s). Alternatively, the intracellular contents of SP030724 could be analyzed using gel electrophoresis, by comparing its intracellular protein concentrations to those of 10-16.

Investigations assessing alterations in permeability as a mechanism of biocide resistance have emphasized the importance of the outer lipopolysccharide and protein layer in achieving resistance via this mechanism (McDonnell and Russell 1999). Since this layer is only present in gram-negative bacteria, and Staphylococcus epidermidis is a gram-positive strain of bacteria, it would seem unlikely that an alteration to the cell membrane structure is responsible for the resistance to ClO2. Howeverm, recent reports have indicated that structural changes in the cell wall of gram-positive bacteria can be a 234

mechanism of antimicrobial resistance (Lambert 2002). As such, the expression of a

tansmembrane molecule that inactivates or exhausts the antimicrobial effects of ClO2 would be a viable mechanism of resistance. While the interactions of ClO2 and other halogen-releasing compounds with the cell wall or membrane has not been established

(McDonnell and Russell 1999; Maillard 2002), the expression of a molecule with an appropriate structure could prevent biocidal activity. For example, the over-expression of a non-essential protein with electron-rich amino acid residues could exhaust the available

ClO2 before it interacts with essential cell structures. Cell wall and/or membrane

components, isolated by lysing the bacteria and using graded centrifugation, could be

analyzed for the presence of an antioxidant using the M-ZOI assay or by mass

spectroscopy for the over-expression of cell wall/membrane components.

While the mechanisms of resistance of SP030724 to ClO2 remain undetermined

the fact that the mechanism can be overcome has been discovered. Exposing SP030724

to increasing concentrations of ClO2 causes an increase in bactericidal activity. This

suggests that the mechanism of resistance to ClO2 can be overcome and that the reponse mechanism is saturable. Both hypotheses proposed here would be supported by this data.

For the production of an intracellular antioxidant, protein synthesis will reach a maximum production rate, after which the protein concentration would be at a steady state. Exposing bacteria to a concentration of ClO2 that exceeds the concentration of

available antioxidant would result in bactericidal activity. Similarly, a transmembrane

molecule will have a maximum expression which would be overwhelmed at higher

concentrations of ClO2, with bactericidal activity ensuing. 235

Determining the precise mechanism of resistance of SP030724 to ClO2 is

complicated since the cause of ClO2 bactericidal activity is unknown. The results from

these experiments have determined that the mechanism of resistance to ClO2 is not through the production of an extracellular antioxidant. Based upon the mechanisms of biocide resistance that have been established, our two alternative hypotheses are that the reduced susceptibility of the spontaneous Staphylococcus epidermidis mutant, SP030724, is either through the production of an intracellular antioxidant, such as glutathione, or through the expression of a transmembrane molecule that inhibits or inactivates the bactericidal effects of ClO2.

236

10. CONCLUSIONS

The studies presented in this body of work address the prevention of biomaterial-

associated infections through the application of chlorine dioxide (ClO2) gas. While ClO2 has been utilized as a liquid for disinfection purposes in the food and water industry, it has not previously been assessed as a method for preventing infections associated with biomaterials.

Chlorine dioxide gas was investigated for its potential as a chemical sterilization agent for biomaterials in Chapter 3 and Chapter 4. While high concentrations of ClO2 have been deemed successful in achieving sterilization in previous works (Jeng and

Woodworth 1990a; Han et al. 2003), the studies presented here focused on the utilization of low concentrations of chlorine dioxide. Using low concentrations of chemical agents during sterilization cycles is considered important since it has been determined that high concentrations increase the presence of chemical residuals associated with the materials post-sterilization (Lyarskii et al. 1984). In Chapter 3, a ClO2 sterilization system

comprised of a ClO2 generator and a sterilization chamber was developed and used to assess the optimal environmental parameters required to achieve sporicidal activity in

Bacillus atrophaeus spores. The controlled environmental conditions were ClO2 concentration (25-1500ppm), relative humidity (<25% or >75%), and exposure time (0.5-

6 hours). These conditions were tested against both dry and humidified spores. The results from these experiments isolated several combinations of environmental conditions that caused sterilization of the spore strips. That a range of environmental conditions can 237

be determined is ideal, since it allows for further optimization specific to the type of

material being sterilized. For example, sterilization of select metals may be best conducted in a low humidity environment with longer exposure times to reduce the potential effects of corrosion, while polymers may benefit from exposure to a lower

concentration of ClO2 that requires humidity to be effective, to limit the amount of ClO2 absorbed by the material. However, the importance of some level of humidity has been previously identified (Gilbert et al. 1964; Dadd et al. 1985) and was supported in these experiments when it was determined that dry spores could not be sterilized using low relative humidity ClO2. Given the small percentage of viable spores that remained in the

low humidity studies, increasing either the concentration of ClO2 or the exposure time

could result in successful sterilization under these conditions. ClO2 as a sterilization

technology presents several advantages over current methods used for chemical

sterilization, namely ethylene oxide. Unlike ethylene oxide, sterilization with ClO2 can be achieved at ambient temperature and pressure (Kowalski 1998). When at the same relative humidity and temperature (30°C), ClO2 is 1075 times more efficient than

ethylene oxide (Jeng and Woodworth 1990a), which allows materials to be exposed to

lower concentrations of the sterilizing agent for shorter durations.

In Chapter 4, a selection of polymers were sterilized with ClO2 based on the

environmental conditions determined in Chapter 3 and were assessed for the presence of

cytotoxic chemical residuals. Combinations of environmental parameters were chosen

based on the extremes of either ClO2 concentration or exposure time and were used to

sterilize polyurethane (PU) and polyvinyl chloride (PVC). High density polyethylene 238

(HDPE) and organo-tin stabilized polyvinyl chloride (Sn-PVC) were used as negative

and positive controls, respectively. These experiments were designed to asertain which of the ClO2 sterilization conditions minimized the presence of chemical residuals

associated with the polymers and to determine if post-sterilization processing is required.

Visual assessment of cytotoxicity suggested that sterilization of humidified samples with

high relative humidity ClO2 at a concentration of 25ppm for 6 hours caused the least

cytotoxicity. These results support prior observations that exposure to lower

concentrations of a sterilizing agent reduces that presence of chemical residues associated

with the materials (Berth and Wolffbrandt 1992). However, an MTT assay suggested

that for the test polymers, PU and PVC, sterilization of humidified polymer samples with

high relative humidity ClO2 at a concentration of 500ppm for 30 minutes was the least

toxic method of sterilization. This discrepency is likely due to the different criteria

considered by each assay. While the visual assessment accounts for evidence of

nonviable cells, the MTT assay only considers metabolically active cells. Also, the MTT

assay is an indirect method for assessing cell viability, since it is actually a measure of

mitochondrial activity within the cell. Research has suggested that for sterilization with

ethylene oxide, increased mitochondrial activity may occur (Seguy et al. 1994). As such,

the MTT assay results may only be indicating increased cell metabolic activity, rather

than providing an accurate measurement of cell viability. The results from the

cytotoxicity experiments suggest that post-sterilization processing for materials sterilized

with ClO2 is warranted. 239

The results presented in Chapter 3 and Chapter 4 indicates that the utilization of

ClO2 gas for the sterilization of biomaterials is both feasible and effective. While the

results from the Chapter 3 have determined the exposure time required to sterilize

Bacillus atrophaeous spore strips under specific environmental conditions, a level of sterility assurance needs to be decided upon, and this will affect the exposure time. In addition, EPA and FDA regulations outline specific experimental standards that will need to be completed to ensure that the sterilization conditions meet the regulatory requirements. A post-sterilization processing system needs to be developed to improve the cytotoxic response of materials sterilized with ClO2. Aerating the materials under

pressure or by heat, as with ethylene oxide sterilization, may reduce the level of chemical

residuals. An experiment also needs to be conducted to determine whether the extracted

residuals are simply absorbed ClO2, or if they are also ClO2 decomposition products, and

in what quantities these residuals are present. The optimal environmental conditions for sterilization of metals should also be analyzed, since there may be significant differences

in the response of metals to sterilization with ClO2 than with the polymers tested thus far.

Once a decision has been reached as to which combination of environmental parameters

for ClO2 sterilization is going to be used, substantially more materials with different

properties should be tested for the effects of the ClO2 sterilization cycle. These

assessments need to include both the biocompatibility assessments outlined in ISO 10993

and the impact of the both sterilization and resterilization on the physical/bulk properties

of the materials. 240

There are many factors that need to be considered for the design of a ClO2 sterilization chamber. Whether the sterilization system should have the versatility to sterilize using only one or several of the different combinations of environmental conditions needs to be decided. The ability of the system to sterilize materials using several combinations of environmental parameters may offer versatility as to the types of materials that can be sterilized using ClO2. When designing the prototype ClO2 sterilization system, consideration needs to be given to the materials used to construct the chamber since certain materials may degrade with constant exposure to ClO2. The

impact of humidity also needs to be factored into the material choice. How ClO2 is

generated and delivered into the sterilization system must be determined. For the

development of a portable sterilization system, the utilization of disposable cartridges that

contain the appropriate precursor chemistries would add versatility to the system. It

would also allow the system to incorporate the aforementioned different sterilization conditions, since the concentration of ClO2 produced could easily be controlled by the

concentrations and amounts of the precursor chemistry available. A method for

monitoring the concentration of ClO2 within the sterilization system must be developed.

Since the ultraviolet absorption spectrum of ClO2 exhibits a broad band near 360nm

(Gordon et al. 1972), an ultraviolet based detection system could be a reproducible and accurate measurement of ClO2 concentration within the sterilization system. The system

for monitoring the concentration of ClO2 within the chamber would feed into a control

system that determines if the conditions for sterilization have been maintained for each

sterilization cycle. If a system other than disposable cartridges is being used for the 241

production of ClO2, the output from the system monitoring the concentration of ClO2 would feed into a control system responsible for the production of ClO2, ensuring that the

concentration within the sterilization chamber is held constant. The control system for

measuring ClO2 concentration would also have to feed into the system that controls

access into the sterilization system to ensure that users cannot open the system until the

ClO2 concentrations are below the occupational exposure limits. A monitoring system

for the relative humidity within the sterilization chamber must be designed for

sterilization cycles that require humidity (either humidification of the articles being

sterilized or environmental humidity during the sterilization cycle). The incorporation of

a vacuum into the system design will improve the penetration of the ClO2 gas into the

material being sterilized, and ensure that all surfaces are exposed to the concentrations of

chlorine dioxide gas required for sterilization to occur (e.g. device lumens). These design

considerations are presented schematically in Figure 10.1. Peripheral equipment, such as

chemical indicators, will have to be developed to provide the user with a visual assurance

that the contents from each sterilization cycle were exposed to the environmental

conditions required for sterilization.

In Chapter 6 and Chapter 7, in vitro and in vivo assessments of the bactericidal

efficacy of a ClO2-generating material were performed. Since bacterial adhesion is the first step towards the development of infection, the adhesion and survival of

Staphylococcus epidermidis 10-16 to the surface of a ClO2-generating material were

analyzed in Chapter 6. Since proteins absorb onto the surfaces of biomaterials when they

are implanted, the moderation of bacterial adhesion and the bactericidal activity by 242

Figure 10.1: Schematic diagram of the control systems required in a chlorine dioxide sterilization system 243

ClO2 [ClO2] Generator Control

System access ClO2 Sterilization System Vacuum control

Humidity Humidity generator control 244

albumin and fibronectin adsorbed to the surface of the materials was incorporated into the

design of the experiments. Albumin was selected since prior research has indicated that

it generally inhibits bacterial adhesion (Pringle and Fletcher 1986; Keogh et al. 1992;

Keogh and Eaton 1994), while fibronectin is believed to enhance bacterial adhesion

(Mohammad et al. 1988). The adhesion of Staphylococcus epidermidis to the material

surface was assessed using bacteria labeled with the radioisotope Indium oxine-111

(111In). Measurements of material radioactivity indicated that more bacteria adhered to

the surface of the ClO2-generating material than the control material. There are several

possibilities explaining the difference in bacterial adhesion between the control and ClO2- generating material surfaces, including differences in the physicochemical properties of the material and the more rapid development of an accumulating substratum of bacteria

(nonviable) on the surface of the ClO2-generating material.

The results also suggested that the presence of either protein adsorbed to the

surface of the material did not significantly alter the adhesion of Staphylococcus epidermidis. One theory is that the physicochemical properties of the material may have caused conformational changes to the tertiary structure of the proteins as they adsorbed to the surface of the material such that the adhesin sites responsible for influencing bacterial adhesion were no longer active or accessible (Kitano et al. 1996). Alternatively, the fact that albumin and fibronectin did not influence bacterial adhesion may have been due to

an innate property of the strain of bacteria being used for these experiments, since the

influence of proteins on the adhesion behavior of Staphylococcus epidermidis is

controversial. While the adhesion assay indicated that more bacteria were adhered to the 245

surface of the ClO2-generating material relative to the control and that the adsorption of

proteins on the surface of either material did not influence bacterial adhesion, imaging of

the samples with scanning electron microscopy (SEM) suggested otherwise. The SEM

images indicated more bacteria were adhered to the surface of the control material with

albumin adsorbed to the surface than the ClO2-generating material with albumin adsorbed

to the surface, and that the presence of fibronectin resulted in a substantial increase in the

adhesion of bacteria to both material surfaces. The disparity between these results suggests differences in the strength of bacterial adhesion to the surfaces of the different materials, and the influence of fibronectin, not necessarily on promoting bacterial adhesion, but maintaining it.

The results from the viability assay indicate that the ClO2-generating material significantly reduces the number of viable bacteria adhered to the surface of the material, relative to the control material, and that the presence of proteins adsorbed to the material did not affect the bactericidal activity of the ClO2 produced by the material. The

presence of albumin and fibronectin did impact the viability of bacteria adhered to the

surface of the control material. There were less viable bacteria adhered to the surface of

the control material with albumin adsorbed to the surface and more viable bacteria

adhered to the control material with fibronectin adsorbed to the surface, relative to

control samples with no protein adsorbed to the material surface. This suggests that

albumin and fibronectin may influence the survival of Staphylococcus epidermidis, rather

than adhesion. 246

The in vitro studies in Chapter 7 were designed to assess ClO2 production by a

ClO2-generating material and to compare it to the duration of bactericidal activity

associated with the material. The ClO2-generating material produced measurable

quantities of ClO2 for 7 days although bactericidal activity only lasted for 5 days. There

was a rapid release of ClO2 by the material within the first 24 hours, with decreasing

amounts produced thereafter. The rapid depletion of antimicrobial agents incorporated

into biomaterials is a recognized issue (Avramovic and Fletcher 1991; Goeau-

Brissonniere et al. 1994) that affects duration of protection against infection and

introduces the risk of resistance developing amongst colonies of bacteria exposed to

lower than minimal inhibition concentrations of the antimicrobial agent (van de Belt et al.

1999). The application of an additional coating, such as a hydrogel (Changez et al. 2004)

or a thin polymer film (Kwok et al. 1999a), could be used to control the release of ClO2 from the material, thereby extending the duration of bactericidal activity of the material.

Care needs to be taken when selecting a method for prolonging the release of ClO2 from the material. Since the generation of ClO2 by these materials is humidity activated, the

coating used needs to be sufficiently hydrophilic to absorb enough moisture to activate

the ClO2 precursor chemistries. However, if the coating retains too much moisture there

is the risk that the ClO2 will rapidly diffuse into the extra coating, creating the same

release profile observed in the in vitro study presented here.

An alternative to using additional layers to prolong the release of ClO2 from the polymer would be the development of a delivery system that is activated by ultrasound.

Ultrasonic delivery of drugs via micelles and microbubbles has been both investigated 247

(Munshi et al. 1997) and marketed (ImaRx Therapeutics, Inc., Tucson, AZ). If the

precursor chemistries (sodium chlorite and acid) could be integrated into separate

microbubble or micelle structures, which were then incorporated into or attached onto a

biomaterial, the generation of ClO2 would occur only when the microbubbles were burst

using ultrasound. This could be done at the discretion of the clinician if concerns about

infection arise, thereby preventing the prophylactic administration of antibiotics and

biocides.

The in vivo study performed in Chapter 7 was used to assess the bactericidal

efficacy of the ClO2-generating material in an in vivo environment and to determine the

localized tissue response to the concentrations of ClO2 being produced by the material.

The results from this study indicate that the ClO2-generating material does prevent

bacterial infections in vivo and that the production of ClO2 in an in vivo environment does

not cause an adverse tissue response at the implant site. An inflammatory response in the

local tissue was observed at the implant sites for both control and ClO2-generating material samples. While the production of ClO2 did increase the number of macrophages

present in the tissue associated with the material, the polymer blend of polyethylene and

ethylvinyl alcohol also contributed to the inflammatory response observed. It may be

necessary to select an alternative base polymer or polymer blend to reduce the

inflammatory response. The polymer would need to be hydrophilic and the processing

conditions would have to be insensitive to the sodium chlorite and acid precursor

chemistries on the microspheres, to ensure that they are not altered or prematurely

activated. 248

In testing the bactericidal efficacy of any antimicrobial material in an in vivo

environment, a decision as to when to inoculate the material with bacteria needs to be made. While this study and others inoculated the material samples prior to implantantion

(Buret et al. 1991; Darouiche et al. 2002), there are studies where inoculation with a bolus dose of bacteria after the materials are implanted is done (Ghiselli et al. 2002;

Hernandez-Richter et al. 2003). For each method there are advantages and disadvantages. Inoculation prior to implantation ensures that the antimicrobial activity of the material is primarily responsible for the bactericidal response observed, rather than the host immune response to the presence of the planktonic bacteria that may or may not interact with the material surface. However, inoculation prior to implantation means that there will be less viable bacteria associated with the antimicrobial material at the time of implantation. It is worth performing both methods of in vivo assessment to ensure that no bias is created through experimental design.

In Chapter 6, the viability assay indicated that viable bacteria were still associated with the surface of the ClO2-generating material, despite a significant decrease in bacteria

viability as compared with control material samples. While increasing the concentration

of ClO2 produced by the material may further reduce the presence of viable bacteria, the

ZOI assay in Chapter 7 indicates that bactericidal concentrations of ClO2 are produced

for 5 days, suggesting that the concentrations currently generated by the material are

adequate. In addition, the in vivo study in Chapter 7 indicates that the bactericidal

activity of the material is sufficient to prevent infections from developing. Consideration

needs to be given to the environmental conditions that the material is being tested under 249

and the applicability of these conditions in a clinical environment. In the in vitro

adhesion study the material samples were incubated in solutions of Staphylococcus

epidermidis with 106 viable bacteria, which increased to 108 – 109 bacteria after 6 hours

of incubation. These levels of contamination do not occur clinically. When considering

the viability data obtained from the adhesion assay, there is a significant reduction of

viable bacteria adhered to the ClO2-generating material, although viable bacteria do

remain associated with the material. Similar results have been observed with other

antimicrobial materials, which remain as valid antimicrobial materials for the prevention

of biomaterial-associated infections (Rees et al. 1998; Darouiche and Mansouri 2004).

As such, modifications to the concentrations of ClO2 produced by the material should not

be based solely on the results from the in vitro adhesion study where the initial bacterial

contamination exceeds what would occur in a clinical setting.

The experiments performed in Chapter 9 were designed to investigate the

mechanism of resistance of a spontaneous Staphylococcus epidermidis mutant

(SP030724) to the antimicrobial effects of ClO2. The hypothesis that was investigated proposed that SP030724 was producing a molecule that was affording it protection

against the oxidative effects of ClO2. While initial studies suggested that the molecule

was extracellular, further investigation determined that it was not.

The established mechanisms of biocide resistance include permeability changes,

efflux pumps, mutations of the biocide target site, and inactivation of the biocide. By

assessing these mechanisms and the properties of ClO2, it was determined that the

synthesis of an antioxidant that was inactivating the ClO2 remained the most plausible 250

hypothesis. Since these studies indicated that the molecule was not extracellular, the

possibility of an intracellular molecule was investigated. Published research has

indicated that oxidizing agents, including hydrogen peroxide, preferentially oxidize glutathione (GSH), suggesting that oxidization of GSH may serve as a protective mechanism for other more vital cell components (Chesney et al. 1996). The oxidative effects of sodium chlorite and ClO2 on glutathione have also been established (Abdel-

Rahman et al. 1980; Abdel-Rahman et al. 1985; Abdel-Rahman and Scatina 1985). To

determine whether SP030724 is over synthesizing glutathione, or other intracellular

antioxidants, the intracellular contents of SP030724 would need to be isolated and

compared with intracellular contents of the wild type strain of Staphylococcus

epidermidis. An alternative hypothesis for the mechanism of resistance is that SP030724

is over expressing a tansmembrane molecule that is either inactivating or inhibiting the

bactericidal effects of ClO2. Comparing the composition of cell wall/membrane of

SP030724 to 10-16 would elucidate this theory.

Determining the mechanism of resistance to ClO2 is complicated by the fact that

the precise mechanism of ClO2 biocidal activity has not yet been established. However,

based upon the bacterial cell targets of ClO2 and an understanding of the properties of

ClO2 gas, the production of an antioxidant is a viable mechanism to explain ClO2 resistance. Considering that exposing SP030724 to increasing concentrations of ClO2 causes an increase in bactericidal activity, suggesting that the mechanism of ClO2 resistance may be overcome, lends further weight to both of the proposed mechanisms of resistance. The fact that the mechanism of resistance can be overcome through the 251

application of increasing ClO2 concentrations also suggests that increasing the concentrations of ClO2 produced by the materials may prevent ClO2 resistance from

developing, although consideration would need to be given to the effects in vivo.

This body of work has assessed the applicability and efficacy of ClO2 gas for the

prevention of biomaterial-associated infections. The environmental parameters required

for the utilization of low concentrations of ClO2 gas as a sterilizing agent for biomaterials

have been determined and the advantages of ClO2 over the current chemical sterilization

standard, ethylene oxide, have been established. A ClO2-generating material was

assessed for its validity as a method for the in situ prevention of biomaterial-associated

infections and the results indicated that the current formulation of the material could

provide short-term protection against bacterial infections. Assessing the development of

acquired resistance to ClO2 has suggested that the over-synthesis of an antioxidant

(intracellular or transmembrane) may be responsible for the reduced susceptibility of a

spontaneous ClO2 resistant mutant and that increasing the concentration of ClO2 overwhelms the mechanism of resistance, causing bactericidal activity in the mutant. The data presented here suggests that ClO2 gas can be utilized for the prevention of

biomaterial associated infection, both pre-implantation and in situ.

• Low concentrations of ClO2 gas can achieve sterilization of Bacillus atrophaeus

spore strips

• Post-sterilization processing of polymers sterilized with low concentration ClO2 is

required to remove potential cytotoxic residuals remain associated with the

material 252

• A ClO2-generating material that produces bactericidal concentrations of ClO2 for

5 days causes bactericidal activity in adhered bacteria, but does not inhibit

bacterial adhesion

• The ClO2-generating material produces a substantial amount of ClO2 in the first

24 hours, with decreasing concentrations produced thereafter

• The production of ClO2 in an in vivo environment does not cause an adverse

tissue response to the local tissue

• The mechanism of resistance to ClO2 of a spontaneous bacteria mutant is not

through the production of an extracellular scavenging molecule, but may be the

overproduction of an intracellular molecule or the expression of a transmembrane

molecule 253

APPENDICES

DESCRIPTION AND OPERATION OF A CHLORINE DIOXIDE GENERATOR

Background

The following is a description of the design of the chlorine dioxide (ClO2) generator and its operation. This protocol is derived from a standard operating procedure written by Bernard Technologies Incorporated (BTI) (Chicago, IL). The ClO2 generator

is capable of producing constant ClO2 concentrations ranging from <1ppm to

>80,000ppm for several hours. The relative humidity and rate that the ClO2 gas is generated are also variable. The flexibility of the generator output is achieved through controlling the reagent concentrations, the rate at which they are combined, and three independent air dilution streams. The system is robust and requires little maintenance.

The use of manual control valves and wet chemical processes avoids costly electronic components, which are prone to corrosion, failure, and require periodic maintenance and calibration.

ClO2 is produced by the oxidation of sodium chlorite (NaClO2) by sodium

persulfate (Na2S2O8).

2 NaClO2 + Na2S2O8 → 2 ClO2 + 2 Na2SO4 [Equation A.1]

This method of synthesis provides good yields of essentially pure ClO2 without volatile

by-products such as Cl2. Down stream dilution air is then used to achieve the final desired ClO2 concentration, flow rate, and relative humidity. The approximate generation

parameters (NaClO2 concentration, flow rate, and dilution air flow rates) for the 254

production of a specific ClO2 concentration and flow rate can be calculated based on the stoichiometry of the above reaction (detailed in Calculating ClO2 generator

parameters). The actual concentration is then determined by iodometric titration, and

the generator parameters adjusted as necessary to achieve the desired generator output.

The ClO2 generator is schematically presented in Figure A.1. The production of

ClO2 is controlled by the slow fixed rate addition of NaClO2 to an excess of Na2S2O8 while constantly sweeping the gas product from the reactor.

Procedure for chlorine dioxide production

The following general procedure, referring to the Figure A.1, is a description of how to produce ClO2. The Na2S2O solution in the reactor is a concentrated solution. This insures immediate and complete oxidation of the NaClO2 as it is added, thus removing

any concerns with respect to the ClO2 generation rate due to co-reactant concentration or

mass transport concerns.

Attach the 0.45µm syringe filter (B) to the bottom of the 60 ml disposable syringe

body (A), and the transfer line from the peristaltic pump (C) to the outlet of the filter.

Prepare the appropriate concentration NaClO2 solution (based on calculations) and

transfer into the syringe. Make a concentrated Na2S2O8 solution in the round bottom

reaction flask (D) and use a magnetic stir bar to stir the contents of the reaction flask

using the magnetic stir plate (E). Attach a compressed air supply to the regulator (N) to

provide the three downstream flow meters (F, G, H) with air at a constant pressure of

approximately 10psi. Sparge the reactor with air using the flow meter control valve (H)

255

Figure A.1: Schematic of ClO2 generator: (A) 60ml disposable syringe; (B) 0.45 µm filter; (C) Peristaltic pump and tubing; (D) 250ml round bottom flask; (E) Magnetic stir plate; (F, G) Flowmeter (0-2500 ml/min); (H, L) Flowmeter (0-500 ml/min); (I, J) Gas washing bottles; (K) Rubber septum; (M) Flowmeter (0-5000 ml/min); (N) Air pressure regulator

256

Compressed ClO2 N Air

F G H L M

I J C D

B E

257

1 to set the flow rate. The /8 inch Teflon line used to deliver the sparge air should be

positioned to extend beneath the liquid surface, but not interfere with the magnetic stir

bar. The desired dilution airflow is then set using the flow meter control valves for wet

(G) and dry (F) air. The wet air dilution flow passes through the gas washing bottle (I)

1 which should be approximately /2 full of de-ionized water. The flow of each flow meter

should be balanced to produce the desired final relative humidity, while maintaining the

proper total flow rate to achieve the target ClO2 concentration. Gas generation is initiated

by activation of the peristaltic pump at the desired flow rate to deliver the NaClO2 solution into the reactor. The ClO2 generated by the oxidation of NaClO2 is sparged from

1 the reactor solution and passes through a /4 inch (outer diameter) Teflon gas transfer line

into a ballast chamber (J) where the dilution air streams are added and fluctuations in gas

concentration are dampened. The diluted ClO2 gas mixture then passes the sampling port

(K), which is capped by a rubber septum. This septum should be replaced daily prior to gas generation. The gas stream exits the generator through two flow meters (L, M), each equipped with an outlet pressure gauge to assist the operator in balancing flow rates against any downstream back pressure. The volume of gas flowing through each leg of the outlet is controlled by using the control valves on each flow meter. The generator is equipped with high and low flow outlets, allowing the operator to deliver only the

volume of gas needed depending on the downstream gas requirements.

As the ClO2 generation rate increases, the sparge airflow rate should also be

increased to avoid buildup of ClO2 in the reactor prior to dilution down stream.

Increasing either the concentration or the flow rate of the NaClO2 solution will result in 258

an increased generation rate. As the quantity of ClO2 increases in the liquid phase, more

sparge air is required to purge the gas out of solution and into the gas stream.

The relative humidity of the generator output is variable over a finite range.

Under normal operating conditions, a range of 16 to 95% relative humidity can be

achieved. By varying the relative amounts of wet and dry dilution air, any relative

humidity within that range can be produced. The absolute limits of the humidity range

are a function of the airflow rate through the water filled gas washing bottle and the

relative humidity of the compressed air supplied to the generator. When relative

humidities outside the standard range are required, the generator can be modified to

extend the range of potential output moisture content. If lower humidity is required, a

drying column containing an appropriate desiccant can be installed after the gas mixing

bottle (J). If a higher relative humidity is required, a heating coil can be wrapped around

the gas-washing bottle (I) to gently heat the water reservoir to increase the vapor pressure of the water.

Calculations for the ClO2 generator parameters

While the operating parameters and output of the generator are variable over a

wide range, there are certain operating guidelines. The output concentration and flow

rates are not completely independent. Extremely low concentrations may require

substantial dilution, resulting in high flow rates. Conversely, a generator of this size is

limited to lower flow rates to produce extremely high ClO2 concentrations, since dilution

airflow must be minimized. The recommended operating parameter ranges are outline in

Table A.1. 259

Parameter Recommended operating range NaClO2 flow rate 0.006-0.3ml/min [NaClO2] 0.1-30% Sparge air flow rate 100-500ml/min Dilution air flow rate 100-4000ml/min

Table A.1: The recommended working range for the generator parameters that control the concentration of the ClO2 produced

260

The following equations are provided to approximate the relationship between the operating parameters and the generator output. The equations can be used to estimate the concentration of ClO2 expected for a given set of operating conditions or to estimate the appropriate operating parameters required to generate a particular ClO2 concentration and flow rate. For accurate measurements of the ClO2 concentration being produced by the generator, see the Appendix: MEASURING CHLORINE DIOXIDE

CONCENTRATION USING IODOMETRIC TITRATION.

Output gas flow: The total airflow rate used to dilute the ClO2 is a summation of the independent sparge, wet, and dry air streams:

Fg = (S + Dw + Dd)/1000 [Equation A.2]

where: Fg is the total air flow rate in (l/min)

S is the reactor sparge airflow rate (ml/min)

Dw is the wet dilution airflow rate (ml/min)

Dd is the dry dilution airflow rate (ml/min)

ClO2 concentration: The relationship between operating parameters and the output

ClO2 concentration is:

3 Cg = 1.9812×10 × Fs × Cs/Fg [Equation A.3]

where: Cg is the theoretical output gas stream [ClO2] (ppm)

Fs is the chlorite solution flow rate (ml/min)

Cs is the chlorite solution concentration (%)

Fg is the total airflow rate in (l/min) 261

The values calculated for the output ClO2 concentration (Cg) could deviate as the

concentration of the NaClO2 solution increases, particularly above 10%, due to the

increase in solution density. The equation assumes a density of 1g/ml over the entire

range. If a more accurate estimate of theoretical ClO2 concentration is required, the value

should be multiplied by the actual NaClO2 solution density.

Based on Equation A.3 and the generator parameter ranges outlined in Table A.1,

the following extremes in generator output concentration and flow can be produced:

Minimum [ClO2]: 0.3ppm at 4.5l/min

Maximum [ClO2]: 8.9% at 200ml/min

Solutions: The amount of reagents needed to achieve the desired ClO2 concentration and

flow rate for the required time period should be estimated.

NaClO2: The total amount of NaClO2 solution required for a specific ClO2 concentration, (Cs), based on the output goal, is calculated by the following equation:

Vs = 0.03 × Cg × t × Fg / Cs [Equation A.4]

where: Vs is the volume of NaClO2 solution required (ml)

Cg is the output gas stream [ClO2] (ppm)

Cs is the chlorite solution concentration (%)

t is the total gas generation time (hours)

Fg is the total airflow rate (l/min)

The solution volume may also be calculated based on the solution feed rate and

generation time:

262

Vs = 60 × Fs × t [Equation A.5]

where: Vs is the volume of NaClO2 solution required (ml)

Fs is the NaClO2 solution flow rate (ml/min)

t is the total gas generation time (hours)

Na2S2O8: The minimum amount of Na2S2O8 required for a reaction with a given volume of NaClO2 solution is estimated by Equation A.6. The operator should confirm that the amount Na2S2O8 exceeds the minimum requirements, particularly for high concentration or long duration runs.

Minimum mass Na2S2O8 = 0.79 × Fs × Cs × t [Equation A.6]

where: Fs is the NaClO2 solution flow rate (ml/min)

Cs is the [NaClO2] solution (%)

t is the total gas generation time (hours) 263

MEASURING CHLORINE DIOXIDE CONCENTRATION USING

IODOMETRIC TITRATION

Background

The following is a procedure to determine the concentration of chlorine dioxide

(ClO2) gas being produced by a ClO2 generator. This protocol is derived from a standard

operating procedure written by Bernard Technologies Incorporated (BTI) (Chicago, IL).

This procedure described how to measure ClO2 gas concentrations using two methods.

For concentrations of ClO2 < 100ppm, a dual impinger containing a reducing solution to

remove the reactive gas from a diluent gas stream is used (acidic potassium iodide). For

concentrations of ClO2 ≥ 100ppm, a sample of gas taken from the ClO2 generator

sampling port using a gas syringe is injected into a reducing solution. For both methods,

the reducing solution is titrated iodometrically to determine the ClO2 concentration.

Solutions

1% potassium iodide

Acetic acid

0.01/0.001N sodium thiosulfate solution (see STANDARDIZATION OF 0.01N

SODIUM THIOSULFATE SOLUTION for the protocol)

1% soluble starch indicator

Chemistry

The concentration of ClO2 produced by the generator is determed using

iodometric titration of a known volume of reduced ClO2 with standardized sodium 264

thiosulfate (NTS). Adding ClO2 to an acidic potassium iodide solution (1% KI + acetic

acid) causes the iodide to be oxidized to iodine by ClO2 (Equation A.7). The iodine

solution is then titrated with NTS using starch as an indicator (Equation A.8), reducing the iodine back to iodide:

- + - 2ClO2 + 10I + 8H → 5I2 + 2Cl + 4H2O [Equation A.7]

- + 2Na2S2O3 + I2 → 2I + Na2S4O6 + 2Na [Equation A.8]

Procedure for [ClO2] < 100ppm

The reactive gas must first be completely absorbed from the gas sample into the aqueous phase to be titrated. A dual impinger set-up is used to absorb the ClO2. The reservoirs of the impingers contain an acidic potassium iodide solution (10ml 1% potassium iodide + 5ml acetic acid). This reducing solution causes a complete reaction of the ClO2 to produce iodine, which is subsequently titrated. The flow from the

calibration gas source is partially diverted through the impinger. The exact volume of

calibrant gas passing through the impinger is measured by taking the difference between

the generator out flow and the volume of gas not passing through the impinger using a

soap film bubble flow meter. The equipment set-up for this procedure is shown in Figure

A.2.

Set up the impinger with the valves in the “bypass” position (Figure A2).

Measure the total gas flow from the ClO2 generator using the bubble flow meter. Record

this value (F1). With the needle valve in the closed position, open the shut-off valve to

the vacuum source. Slowly open the needle valve, while continuously checking the flow

rate until approximately 70-90% of the total gas flow (F1) is diverted to the vacuum 265

Figure A.2: Impinger system for measuring [ClO2] < 100ppm (A). The positions for measuring generator flow output and sampling the gas flow are indicated in (B) 266

(A)

Needle Valve Calibration Gas Source

Vacuum Source Bubble Flow Meter

(B)

“Bypass” “Measuring” 267

source. Transfer 10ml of the 1% KI solution into each of the impinger reservoirs, add

5ml acetic acid to each of the impinger reservoirs, and carefully attach each of the

reservoirs to the impinger. Rotate the T-bore stopcocks simultaneously to the “measure”

position and start the stopwatch. Watch the color of the solutions in each of the

impingers and allow the ClO2 gas stream to flow through the impinger until the color in

the second impinger begins to turn yellow. During this time, measure the flow rate of the

gas passing through the bubbler several times (F2). When the second impinger begins to

turn yellow, rotate the T-bore stopcocks to the “bypass” position and record the total time

the gas stream was diverted through the impinger (T). Transfer the contents of both

impinger reservoirs to an Erlenmeyer flask containing a magnetic stir bar, and using deionized water, rinse the sample from the reservoirs into the Erlenmeyer flask. Titrate the

solution using the standardized NTS until it is a faint yellow color. Add approximately

1ml of starch indicator solution. An intense blue (purple or black) color will develop due

to the formation of a starch-iodine complex. Continue the titration until the solution becomes clear. Record the total volume of NTS consumed (V3).

Calculations

The calculations are based on the stoichiometries described by Equations A.7 and

Equations A.8. The concentration of the ClO2 gas stream is reported in parts per million,

or microliters ClO2 per liter of gas. The number of microliters of ClO2 absorbed from the gas stream (V4) is:

V4 = V3 × C2 × 22.4 × 1000/5 [Equation A.9]

where: V3 is the volume of titrant used on the sample (ml) 268

C2 is the concentration of the NTS (normalized)

The volume of gas (in liters) that passed through the impinger is calculated by

difference from the flow rate values recorded:

V5 = (F1 – F2) × T [Equation A.10]

where: F1 is the flow rate of the total output of the gas generator (l/min)

F2 is the flow rate of the gas stream not diverted through the

impinger (l/min)

T is total time the gas was diverted through the impinger (min)

The concentration of ClO2 in the gas stream (in parts per million) from the generator is:

[ClO2] = V4/V5 [Equation A.11]

Procedure for [ClO2] ≥ 100ppm

If the concentration of the ClO2 gas stream is greater than 100ppm, it may be

sampled by gas tight syringe and iodomertically titrated. The required gas sample size is

dependent on the ClO2 concentration in the gas stream, but is generally between 25-

100ml to allow sufficient iodine production required for accurate and measurable volumes of NTS titrant (recommended NTS concentration: 0.001N). Using a gas tight syringe, withdraw a sample of the ClO2 gas stream from sample port (K on Figure A.1),

and immediately inject it into a septum capped vial containing 20ml of 1% KI solution and 5ml acetic acid. A small needle should be used to vent the vial as the gas is slowly injected below the liquid surface, taking care not to allow liquid to enter the needle. The solution should turn a shade of yellow. Remove all needles from the vial and shake the vial for approximately 5 seconds. Pour the contents into a 125ml Erlenmeyer flask 269

containing a magnetic stir bar, and rinse the vial with de-ionized water and add to the

Erlenmeyer flask. Titrate the sample using NTS with constant stirring until the sample is

faint yellow. Add approximately 1ml starch indicator solution (solution should turn blue), and continue titrating until the blue solution becomes colorless.

Calculations

The ClO2 concentration in parts per million (v/v) is then calculated by the following equation:

6 [ClO2] = 4.48x10 × VT × CT/Vg [Equation A.12]

where: VT is the volume of NTS titrant used (ml)

CT is the concentration of the NTS (normality)

Vg is the volume of the ClO2 gas sample (ml)

270

STANDARDIZATION OF 0.01N SODIUM THIOSULFATE SOLUTION

This procedure is for the standardization of a sodium thiosulfate solution utilized

for iodometric titrations in a protocol for measuring ClO2 concentrations.

Solutions

1% potassium iodide (KI)

Acetic acid

Potassium iodate-iodide solution (0.002083 M)

0.01 (0.001N) sodium thiosulfate solution (NTS) (to be standardized)

1% soluble starch indicator solution

Procedure

Add 20ml of 1% KI and 5ml of acetic acid solution into an Erlenmeyer flask.

Add 5ml (1ml) of the potassium iodate-iodide solution for the 0.01N (0.001N) sodium

thiosulfate solution with a volumetric pipette. Since this is the standard that is used to

standardize the NTS, accuracy is essential. The solution will turn dark yellow. Titrate

the sample with the 0.01N (0.001N) NTS. Use the magnetic stirrer and stir bar to ensure adequate mixing of the sample throughout the titrating procedure. Titrate the sample until only a faint yellow color remains and then add 1 ml of the starch indicator solution.

The solution will turn dark blue/purple. Continue the titration until the color is no longer visible. Record the final amounts of NTS utilized for sample. Repeat three times.

Calculations

NTS normality = A × B × 6 / C [Equation A.13]

where: A = Molarity of potassium iodate-iodide solution 271

B = Volume of potassium iodate-iodide solution (ml)

C = Volume of sodium thiosulfate utilized (ml) 272

REFERENCES

- Abdel-Rahman, M. S., D. Couri and R. J. Bull (1980). "Kinetics of ClO2, ClO2 , and - ClO3 in drinking water on blood glutathione and hemolysis in rat and chicken." J Environ Pathol Toxicol 3: 431-449.

Abdel-Rahman, M. S., D. Couri and R. J. Bull (1982). "Metabolism and pharmacokinetics of alternate drinking water disinfectants." Environ Health Perspect 46: 19-23.

Abdel-Rahman, M. S., D. Couri and R. J. Bull (1985). "Toxicity of chlorine dioxide in drinking water." J Environ Pathol Toxicol Oncol 6(1): 105-13.

Abdel-Rahman, M. S. and J. Scatina (1985). "The effect of Alcide, a new antimicrobial drug, on rat blood glutathione and erythrocyte osmotic fragility, in vitro." J Appl Toxicol 5(3): 178-81.

Adams, A. P., E. M. Santschi and M. A. Mellencamp (1999). "Antibacterial properties of a silver chloride-coated nylon wound dressing." Vet Surg 28(4): 219-25.

Aieta, E. M. and J. D. Berg (1986). "A review of chlorine dioxide in drinking water treatment." Journal of the American Water Works Association 78: 62-72.

Alder, V. G., A. M. Brown and W. A. Gillespie (1966). "Disinfection of heat-sensitive material by low-temperature steam and formaldehyde." J Clin Pathol 19(1): 83-9.

Alfred Wegener Institute (2005). http://www.awi-bremerhaven.de/TT/antifouling/index- e.html. 2005.

Alvarez, M. E. and R. T. O'Brien (1982). "Mechanisms of inactivation of poliovirus by chlorine dioxide and iodine." Appl Environ Microbiol 44(5): 1064-71.

American Conference of Governmental Industrial Hygenists, I. (1990). Industrial Hygenic Guide. Cincinnati, OH.

An, Y. H. and R. J. Friedman (1998). "Concise review of mechanisms of bacterial adhesion to biomaterial surfaces." J Biomed Mater Res 43(3): 338-48.

An, Y. H., R. J. Friedman, R. A. Draughm, E. A. Smith, J. Nicholson and J. F. John (1995). "Rapid quantification of staphylococci adhered to titanium surfaces using image analyzed epifluorescence microscopy." J Microbiol Methods 24(1): 29-40.

273

Anderson, R. M. (1999). "The pandemic of antibiotic resistance." Nat Med 5(2): 147-9.

Archer, G. L. (1995). Staphylococcus epidermidis and other coagulase negative staphylococci. Mandell, Douglas, and Bennett's Principles and practice of infectious diseases. New York, Churchill Livingstone: 1777-1784.

Arciola, C. R., Y. Bustanji, M. Conti, D. Campoccia, L. Baldassarri, B. Samori and L. Montanaro (2003). "Staphylococcus epidermidis-fibronectin binding and its inhibition by heparin." Biomaterials 24(18): 3013-9.

Avramovic, J. R. and J. P. Fletcher (1991). "Rifampicin impregnation of a protein-sealed Dacron graft: an infection-resistant prosthetic vascular graft." Aust N Z J Surg 61(6): 436-40.

Aydin, M., I. Gunay, A. Pelit and M. S. Serin (1997). "Deposition profile of antibacterial anodic silver in root canal systems of teeth." J Biomed Mater Res 38(1): 49-54.

Ayliffe, G. (2000). "Decontamination of minimally invasive surgical endoscopes and accessories." J Hosp Infect 45(4): 263-77.

Bacon, F. (1635). Sylva sylvarum or a natural history, 2nd Ed. London, Century IV.

Baker, A. S. and L. W. Greenham (1988). "Release of gentamicin from acrylic bone cement. Elution and diffusion studies." J Bone Joint Surg Am 70(10): 1551-7.

Becker, R. O. and J. A. Spadaro (1978). "Treatment of orthopaedic infections with electrically generated silver ions. A preliminary report." J Bone Joint Surg Am 60(7): 871-81.

Bercz, J. P., L. Jones, L. Garner, D. Murray, D. A. Ludwig and J. Boston (1982). "Subchronic toxicity of chlorine dioxide and related compounds in drinking water in the nonhuman primate." Environ Health Perspect 46: 47-55.

Berger, T. J., J. A. Spadaro, S. E. Chapin and R. O. Becker (1976). "Electrically generated silver ions: quantitative effects on bacterial and mammalian cells." Antimicrob Agents Chemother 9(2): 357-8.

Berman, D. and J. C. Hoff (1984). "Inactivation of simian rotavirus SA11 by chlorine, chlorine dioxide, and monochloramine." Appl Environ Microbiol 48(2): 317-23.

Bernarde, M. A., W. B. Snow and V. P. Olivieri (1967). "Kinetics and mechanism of bacterial disinfection by chlorine dioxide." Applied Microbiology 15: 257-265.

274

Berth, L. and K. H. Wolffbrandt (1992). "Improving the ethylene oxide sterilization process." Med Device Technol 3(5): 38-44.

Bisno, A. L. (1995). "Molecular aspects of bacterial colonization." Infect Control Hosp Epidemiol 16(11): 648-57.

Bolognesi, C., A. Buschini, E. Branchi, P. Carboni, M. Furlini, A. Martino, M. Monteverde, P. Poli and C. Rossi (2004). "Comet and micronucleus assays in zebra mussel cells for genotoxicity assessment of surface drinking water treated with three different disinfectants." Sci Total Environ 333(1-3): 127-36.

Botzenhart, K., G. M. Tarcson and M. Ostruschka (1993). "Inactivation of bacteria and coliphages by ozone and chlorine dioxide in a continuous flow reactor." Water Science and Technology 27: 363-370.

Bragg, P. D. and D. J. Rainnie (1974). "The effect of silver ions on the respiratory chain of Escherichia coli." Can J Microbiol 20(6): 883-9.

Brannen, J. P. (1970). "On the role of DNA in wet heat sterilization of micro-organisms." J Theor Biol 27(3): 425-32.

Bren, L. (2002). "Battle of the Bugs: Fighting Antibiotic Resistance." FDA Consumer 36(4): http://www.fda.gov/fdac/features/2002/402_bugs.html.

Brown, M. H., I. T. Paulsen and R. A. Skurray (1999). "The multidrug efflux protein NorM is a prototype of a new family of transporters." Mol Microbiol 31(1): 394-5.

Brown, P., C. J. Gibbs, Jr., H. L. Amyx, D. T. Kingsbury, R. G. Rohwer, M. P. Sulima and D. C. Gajdusek (1982). "Chemical disinfection of Creutzfeldt-Jakob disease virus." N Engl J Med 306(21): 1279-82.

Bruch, C. W. (1973). Sterilization of Plastics: Toxicity of Ethylene Oxide Residues. Industrial Sterilization. G. B. Philips, Miller, W.S. Durham, North Carolina, Duke University Press: 49-77.

Bruck, S. D. and E. P. Mueller (1988). "Radiation sterilization of polymeric implant materials." J Biomed Mater Res 22(A2): 133.

Brun-Buisson, C., F. Doyon, J. P. Sollet, J. F. Cochard, Y. Cohen and G. Nitenberg (2004). "Prevention of intravascular catheter-related infection with newer chlorhexidine- silver sulfadiazine-coated catheters: a randomized controlled trial." Intensive Care Med 30(5): 837-43.

275

Bumgardner, J. D., P. D. Gerard, W. Geurtsen and G. Leyhausen (2002). "Cytotoxicity of precious and nonprecious alloys - experimental comparison of in vitro data from two laboratories." J Biomed Mater Res 63(2): 214-9.

Buret, A., K. H. Ward, M. E. Olson and J. W. Costerton (1991). "An in vivo model to study the pathobiology of infectious biofilms on biomaterial surfaces." J Biomed Mater Res 25(7): 865-74.

Burrows, L. L. and A. E. Khoury (1999). "Issues surrounding the prevention and management of device-related infections." World J Urol 17(6): 402-9.

Burttschell, R. H., A. A. Rosen, F. M. Middleton and M. B. Ettinger (1959). "Chlorine derivatives of phenol causing taste and odor." Journal of the American Water Works Association 51: 205-214.

Cao, T., T. Y. Saw, B. C. Heng, H. Liu, A. U. Yap and M. L. Ng (2005). "Comparison of different test models for the assessment of cytotoxicity of composite resins." J Appl Toxicol 25(2): 101-8.

Carballo, J., C. M. Ferreiros and M. T. Criado (1991). "Importance of experimental design in the evaluation of the influence of proteins in bacterial adherence to polymers." Med Microbiol Immunol (Berl) 180(3): 149-55.

Carlton, B. D., A. H. Basaran, L. E. Mezza, E. L. George and M. K. Smith (1991). "Reproductive effects in Long-Evans rats exposed to chlorine dioxide." Environ Res 56(2): 170-7.

CBS News (2002). DC Postal Facility Anthrax Cleanup, http://cbsnews.cbs.com/stories/2002/12/14/attack/main533065.shtml.

Changez, M., V. Koul, B. Krishna, A. K. Dinda and V. Choudhary (2004). "Studies on biodegradation and release of gentamicin sulphate from interpenetrating network hydrogels based on poly(acrylic acid) and gelatin: in vitro and in vivo." Biomaterials 25(1): 139-46.

Chapiro, A. (1962). Radiation Chemistry of Polymeric Systems. New York, Wiley- Interscience.

Chapman, J., W. Lee, E. Youkilis and L. Martis (1986). "Animal Model for Ethylene Oxide (EtO) Associated Hypersensitivity Reactions." Trans Am Soc Artif Intern Organs 32: 482-485.

276

Charrel, R. N., J. M. Pages, P. De Micco and M. Mallea (1996). "Prevalence of outer membrane porin alteration in beta-lactam-antibiotic-resistant Enterobacter aerogenes." Antimicrob Agents Chemother 40(12): 2854-8.

Cheesman, G. H. (1930). "Parachor of chlorine dioxide." Journal of the Chemical Society (London): 35-37.

Chemical Industries Association (2003). http://www.cia.org.uk/newsite/talking_points/tparticle.asp?id=235. 2005.

Chen, H. Y. and D. M. Livermore (1993). "Activity of cefepime and other beta-lactam antibiotics against permeability mutants of Escherichia coli and Klebsiella pneumoniae." J Antimicrob Chemother 32 Suppl B: 63-74.

Chen, Y. S. and J. M. Vaughn (1990). "Inactivation of human and simian rotaviruses by chlorine dioxide." Appl Environ Microbiol 56(5): 1363-6.

Cheng, S. and D. R. Kerluke (2003). Radiation processing for modification of polymers. Annual Technical Conference of the Society of Plastic Engineering, Nashville, Tennessee.

Chesney, J. A., J. W. Eaton and J. R. Mahoney, Jr. (1996). "Bacterial glutathione: a sacrificial defense against chlorine compounds." J Bacteriol 178(7): 2131-5.

Chick, H. and C. J. Martin (1908). "The principles involved in the standardization of disinfectants and the influence of organic matter upon germicidal value." Journal of Hygiene 8: 654-697.

Chopra, I. and M. Roberts (2001). "Tetracycline antibiotics: mode of action, applications, molecular biology, and epidemiology of bacterial resistance." Microbiol Mol Biol Rev 65(2): 232-60; second page, table of contents.

Christensen, G., L. Baldassarri and A. Simpson (1994). Colonization of Medical Devices by Coagulase Negative Staphylococci. Infections Associated with Medical Devices. A. L. Bisno, Waldvogel F.A. Washington D.C., American Society for Microbiology: 45-78.

Chuanchuen, R., K. Beinlich, T. T. Hoang, A. Becher, R. R. Karkhoff-Schweizer and H. P. Schweizer (2001). "Cross-resistance between triclosan and antibiotics in Pseudomonas aeruginosa is mediated by multidrug efflux pumps: exposure of a susceptible mutant strain to triclosan selects nfxB mutants overexpressing MexCD-OprJ." Antimicrob Agents Chemother 45(2): 428-32.

277

Chuard, C., J. C. Lucet, P. Rohner, M. Herrmann, R. Auckenthaler, F. A. Waldvogel and D. P. Lew (1991). "Resistance of Staphylococcus aureus recovered from infected foreign body in vivo to killing by antimicrobials." J Infect Dis 163(6): 1369-73.

Chung, Y. J. and M. H. Saier, Jr. (2001). "SMR-type multidrug resistance pumps." Curr Opin Drug Discov Devel 4(2): 237-45.

Coates, D. (2001). "An evaluation of the use of chlorine dioxide (Tristel One-Shot) in an automated washer/disinfector (Medivator) fitted with a chlorine dioxide generator for decontamination of flexible endoscopes." J Hosp Infect 48(1): 55-65.

Cocheci, V., R. Taubert and A. Martin (1971). "Removal of cyanide from waste water." Chem Abst 74(79296k): 201.

Costerton, J. W., P. S. Stewart and E. P. Greenberg (1999). "Bacterial biofilms: a common cause of persistent infections." Science 284(5418): 1318-22.

Couri, D., C. H. Miller, Jr., R. J. Bull, J. M. Delphia and E. M. Ammar (1982). "Assessment of maternal toxicity, embryotoxicity and teratogenic potential of sodium chlorite in Sprague-Dawley rats." Environ Health Perspect 46: 25-9.

Coward, J. E., H. S. Carr and H. S. Rosenkranz (1973). "Silver sulfadiazine: effect on the ultrastructure of Pseudomonas aeruginosa." Antimicrob Agents Chemother 3(5): 621-4.

Cowlishaw, J., J. A. Spadaro and R. A. Becker (1982). "Inhibition of enzyme induction in E.coli by anodic silver." Journal of Bioelectricity 3: 295-304.

Dadd, A. H., M. M. Town and K. E. McCormick (1985). "The influence of water on the resistance of spores to inactivation by gaseous ethylene oxide." J Appl Bacteriol 58(6): 613-21.

Dance, D. A., A. D. Pearson, D. V. Seal and J. A. Lowes (1987). "A hospital outbreak caused by a chlorhexidine and antibiotic-resistant Proteus mirabilis." J Hosp Infect 10(1): 10-6.

Darouiche, R. O. and M. D. Mansouri (2004). "In vitro activity and in vivo efficacy of antimicrobial-coated vascular grafts." Ann Vasc Surg 18(4): 497-501.

Darouiche, R. O., R. Meade, M. D. Mansouri and D. T. Netscher (2002). "In vivo efficacy of antimicrobe-impregnated saline-filled silicone implants." Plast Reconstr Surg 109(4): 1352-7.

Davies, D. J., J. Francis, A. R. Margin, F. L. Rose and G. Swain (1954). "1:6-di-4- Chlorophenyldiguanido-hexane (Hibitane): a laboratory investigation of a new 278

antibacterial agent of high potency." British Journal of Pharmacology and Chemotherapy 9: 192-196.

Davy, H. (1811). "On a combination of oxymuriatic gas and oxygen gas." Philos Trans 101: 155-161.

Dempsey, D. J. and R. R. Thirucote (1989). "Sterilization of medical devices: a review." J Biomater Appl 3(3): 454-523.

Denyer, S. P. and G. S. A. B. Stewart (1998). "Mechanisms of action of disinfectants." International Biodeterioration and Biodegradation 41: 261-268.

Dexter, S. J., R. G. Pearson, M. C. Davies, M. Camara and K. M. Shakesheff (2001). "A comparison of the adhesion of mammalian cells and Staphylococcus epidermidis on fibronectin-modified polymer surfaces." J Biomed Mater Res 56(2): 222-7.

Dolovich, J., S. Evans, U. Baurmeister, H. Schulze, M. Ali and A. Shimizu (1987). "Antibody responses to hemodialysis-related antigens in chronic hemodialysis patients." Artif Organs 11(2): 93-6.

Dolovich, J., C. P. Marshall, E. K. Smith, A. Shimizu, F. C. Pearson, M. A. Sugona and W. Lee (1984). "Allergy to ethylene oxide in chronic hemodialysis patients." Artif Organs 8(3): 334-7.

Downes, A. and T. P. Blunt (1877). "Researchers on the effect of light upon bacteria and other organisms." Proceedings of the Royal Society 26: 488-500.

Doyle, J. E. and R. R. Ernst (1967). "Resistance of Bacillus subtilis var. niger spores occluded in water-insoluble crystals to three sterilization agents." Appl Microbiol 15(4): 726-30.

Drenkard, E. (2003). "Antimicrobial resistance of Pseudomonas aeruginosa biofilms." Microbes Infect 5(13): 1213-9.

Dunn, D. S., S. Raghavan and R. G. Volz (1994). "Ciprofloxacin attachment to porous- coated titanium surfaces." J Appl Biomater 5(4): 325-31.

Duran, L. W. (2000). "Preventing medical device related infections." Med Device Technol 11(6): 14-7.

Dziabo, A. J., J. Y. Park and P. S. Ripley (1994). Contact lens disinfection with liquid medium comprising stabilized chlorine dioxide and activator to release free chlorine dioxide, US Patent 5 336 434.

279

Dziabo, A. J. and P. S. Ripley (1994). Disinfecting contact lenses by passing current to chlorine dioxide from a precursor, US Patent 5 320 806.

Dziabo, A. J. and P. S. Ripley (1995). Aqueous ophthalmic composition especially for cleaning contact lenses containing chlorine dioxide as sole preservative, US Patent 5 424 078.

Ehrenberg, L., K. D. Hiesche, S. Osterman-Golkar and I. Wenneberg (1974). "Evaluation of genetic risks of alkylating agents: tissue doses in the mouse from air contaminated with ethylene oxide." Mutat Res 24(2): 83-103.

Elek, S. D. and P. E. Conen (1957). "The virulence of Staphylococcus pyogenes for man; a study of the problems of wound infection." Br J Exp Pathol 38(6): 573-86.

Engelhard Corporation (1997). Scentrex with Aseptrol, Product Bulletin EC7099.

Enger, M. (1960). "Treatment of water with chlorine dioxide to improve taste." Gas Wasserfach 14: 330-336.

Espersen, F., B. J. Wilkinson, B. Gahrn-Hansen, V. Thamdrup Rosdahl and I. Clemmensen (1990). "Attachment of staphylococci to silicone catheters in vitro." APMIS 98(5): 471-8.

Farr, R. W. and C. Walton (1993). "Inactivation of human immunodeficiency virus by a medical waste disposal process using chlorine dioxide." Infect Control Hosp Epidemiol 14(9): 527-9.

Feldman, L. A. and H. K. Hui (1997). "Compatibility of medical devices and materials with low-temperature hydrogen peroxide gas plasma." Medical Devices and Diagnostic Industry 19(57-62).

Ferreiros, C. M., J. Carballo, M. T. Criado, V. Sainz and M. C. del Rio (1989). "Surface free energy and interaction of Staphylococcus epidermidis with biomaterials." FEMS Microbiol Lett 51(1): 89-94.

Fleer, A. and J. Verhoef (1989). "An evaluation of the role of surface hydrophobicity and extracellular slime in the pathogenesis of foreign-body-related infections due to coagulase-negative staphylococci." J Invest Surg 2(4): 391-6.

Flemming, R. G., C. C. Capelli, S. L. Cooper and R. A. Proctor (2000). "Bacterial colonization of functionalized polyurethanes." Biomaterials 21(3): 273-81.

280

Foreman-Wykert, A. K., J. Weiss and P. Elsbach (2000). "Phospholipid synthesis by Staphylococcus aureus during (Sub)Lethal attack by mammalian 14-kilodalton group IIA phospholipase A2." Infect Immun 68(3): 1259-64.

Fox, C. L., Jr. (1983). "Topical therapy and the development of silver sulfadiazine." Surg Gynecol Obstet 157(1): 82-8.

Fox, C. L., Jr. and S. M. Modak (1974). "Mechanism of silver sulfadiazine action on burn wound infections." Antimicrob Agents Chemother 5(6): 582-8.

Fox, C. L., Jr., B. W. Rappole and W. Stanford (1969). "Control of pseudomonas infection in burns by silver sulfadiazine." Surg Gynecol Obstet 128(5): 1021-6.

Fracastoro, G. (1546). Contagion, contagious diseases, and their treatment. W. Wright, translator. New York, Putnam, 1930.

Fraise, A. P. (2002). "Susceptibility of antibiotic-resistant cocci to biocides." J Appl Microbiol 92 Suppl: 158S-62S.

Freeman, D. J., F. R. Falkiner and C. T. Keane (1989). "New method for detecting slime production by coagulase negative staphylococci." J Clin Pathol 42(8): 872-4.

Fujioka, R. S., M. A. Dow and B. S. Yoneyama (1986). "Comparative disinfection of indicator bacteria and poliovirus by chlorine dioxide." Water Science and Technology 18: 125-132.

Gahtan, V., G. E. Esses, D. F. Bandyk, R. T. Nelson, E. Dupont and J. L. Mills (1995). "Antistaphylococcal activity of rifampin-bonded gelatin-impregnated Dacron grafts." J Surg Res 58(1): 105-10.

Galliani, S., M. Viot, A. Cremieux and P. Van der Auwera (1994). "Early adhesion of bacteremic strains of Staphylococcus epidermidis to polystyrene: influence of hydrophobicity, slime production, plasma, albumin, fibrinogen, and fibronectin." J Lab Clin Med 123(5): 685-92.

Gandhi, P. A., A. D. Sawant, L. A. Wilson and D. G. Ahearn (1993). "Adaptation and growth of Serratia marcescens in contact lens disinfectant solutions containing chlorhexidine gluconate." Appl Environ Microbiol 59(1): 183-8.

Gaonkar, T. A., L. A. Sampath and S. M. Modak (2003). "Evaluation of the antimicrobial efficacy of urinary catheters impregnated with antiseptics in an in vitro urinary tract model." Infect Control Hosp Epidemiol 24(7): 506-13.

281

Gardner, J. F. and M. M. Peel (1986). Principles of Heat Sterilization. Introduction to Sterilization and Disinfection, Churchill Livingstone: 45-56.

Geiss, H. K., H. Heid and R. Hirth (1994). "Plasma sterilization - an alternative low- temperature sterilization method." Zentral Sterilization Central Service 2: 263-269.

Gentry, L. O. (1997). Osteomyelitis and other infections of bones and human joints. The staphylococci in human disease. K. B. Crossley, Archer, G.L. New York, Churchill Livingstone: 455-473.

Ghanbari, H. A., W. B. Wheeler and J. R. Kirk (1982). "Reactions of aqueous chlorine and chlorine dioxide with lipids: chlorine incorporation." J Food Sci 147: 482-485.

Ghiselli, R., A. Giacometti, O. Cirioni, F. Mocchegiani, F. Orlando, M. Del Prete, G. D'Amato, G. Scalise and V. Saba (2002). "Quinupristin/dalfopristin bonding in combination with intraperitoneal antibiotics prevent infection of knitted polyester graft material in a subcutaneous rat pouch model infected with resistant Staphylococcus epidermidis." Eur J Vasc Endovasc Surg 24(3): 230-4.

Gibson, C. (1987). Microwave applications in the sterilization of heat sensitive medical supplies. PhD thesis, University of Wales.

Gibson, G. L., H. P. Johnston and V. E. Turkington (1968). "Residual formaldehyde after low-temperature steam and formaldehyde sterilization." J Clin Pathol 21(6): 771-5.

Gilbert, G. L., V. M. Gambill, D. R. Spiner, R. K. Hoffman and C. R. Phillips (1964). "Effect of Moisture on Ethylene Oxide Sterilization." Appl Microbiol 12: 496-503.

Ginalska, G., M. Osinska, A. Uryniak, T. Urbanik-Sypniewska, A. Belcarz, W. Rzeski and A. Wolski (2005). "Antibacterial activity of gentamicin-bonded gelatin-sealed polyethylene terephthalate vascular prostheses." Eur J Vasc Endovasc Surg 29(4): 419- 24.

Goeau-Brissonniere, O., F. Mercier, M. H. Nicolas, F. Bacourt, M. Coggia, C. Lebrault and J. C. Pechere (1994). "Treatment of vascular graft infection by in situ replacement with a rifampin-bonded gelatin-sealed Dacron graft." J Vasc Surg 19(4): 739-41.

Goetz, A. (1943). "Water sanitation with silver." Journal of the American Water Works Association 35: 579.

Goldman, R. C. and D. J. Tipper (1978). "Bacillus subtilis spore coats: complexity and purification of a unique polypeptide component." J Bacteriol 135(3): 1091-106.

282

Gordon, G. (1993). Water chemistry of the oxy-chlorine species: chlorine dioxide drinking water issues. Proceedings of the 2nd International Symposium, May 7-9, 1992, Denver, AWWA Research Foundation.

Gordon, G., R. G. Keiffer and D. H. Rosenblatt (1972). The chemistry of chlorine dioxide. Progress in inorganic chemistry. S. J. Lippaer. New York, Wiley Interscience. 15: 201-287.

Gordon, G. and A. A. Rosenblatt (1995). Gaseous, chlorine-free chlorine dioxide for drinking water. American Water Works Association Water Quality Technology Conference, New Orleans, LA.

Gottenbos, B., D. W. Grijpma, H. C. van der Mei, J. Feijen and H. J. Busscher (2001). "Antimicrobial effects of positively charged surfaces on adhering Gram-positive and Gram-negative bacteria." J Antimicrob Chemother 48(1): 7-13.

Gottenbos, B., H. C. van der Mei, F. Klatter, D. W. Grijpma, J. Feijen, P. Nieuwenhuis and H. J. Busscher (2003). "Positively charged biomaterials exert antimicrobial effects on gram-negative bacilli in rats." Biomaterials 24(16): 2707-10.

Gottenbos, B., H. C. van der Mei, F. Klatter, P. Nieuwenhuis and H. J. Busscher (2002). "In vitro and in vivo antimicrobial activity of covalently coupled quaternary ammonium silane coatings on silicone rubber." Biomaterials 23(6): 1417-23.

Grinius, L., G. Dreguniene, E. B. Goldberg, C. H. Liao and S. J. Projan (1992). "A staphylococcal multidrug resistance gene product is a member of a new protein family." Plasmid 27(2): 119-29.

Guan, L., M. Ehrmann, H. Yoneyama and T. Nakae (1999). "Membrane topology of the xenobiotic-exporting subunit, MexB, of the MexA,B-OprM extrusion pump in Pseudomonas aeruginosa." J Biol Chem 274(15): 10517-22.

Guerin-Mechin, L., F. Dubois-Brissonnet, B. Heyd and J. Y. Leveau (1999). "Specific variations of fatty acid composition of Pseudomonas aeruginosa ATCC 15442 induced by quaternary ammonium compounds and relation with resistance to bactericidal activity." J Appl Microbiol 87(5): 735-42.

Guerin-Mechin, L., F. Dubois-Brissonnet, B. Heyd and J. Y. Leveau (2000). "Quaternary ammonium compound stresses induce specific variations in fatty acid composition of Pseudomonas aeruginosa." Int J Food Microbiol 55(1-3): 157-9.

Gunther, D. A. (1980). "The chemistry and biology of EtO sterilization." Medical Devices and Diagnostic Industry 2(12): 31-35.

283

Gupta, A., K. Matsui, J. F. Lo and S. Silver (1999). "Molecular basis for resistance to silver cations in Salmonella." Nat Med 5(2): 183-8.

Guzzella, L., S. Monarca, C. Zani, D. Feretti, I. Zerbini, A. Buschini, P. Poli, C. Rossi and S. D. Richardson (2004). "In vitro potential genotoxic effects of surface drinking water treated with chlorine and alternative disinfectants." Mutat Res 564(2): 179-93.

Han, Y., B. Applegate, R. H. Linton and P. E. Nelson (2003). "Decontamination of Bacillus thuringiensis spores on selected surfaces by chlorine dioxide gas." J Environ Health 66(4): 16-21.

Hancock, R. E. (1997). "The bacterial outer membrane as a drug barrier." Trends Microbiol 5(1): 37-42.

Hancock, R. E. and W. A. Woodruff (1988). "Roles of porin and beta-lactamase in beta- lactam resistance of Pseudomonas aeruginosa." Rev Infect Dis 10(4): 770-5.

Handlos, V. (1977). "Formaldehyde sterilization. I. Determination of formaldehyde residuals in autoclaved-sterilised materials." Arch. Pharm. Chem. Sci. Ed. 5: 163.

Handlos, V. (1979). "Formaldehyde sterilization. II. Formaldehyde-steam sterilization: the process and its influence on the formaldehyde residuals." Arch. Pharm. Chem. Sci. Ed. 7: 1.

Harakeh, M. S., J. D. Berg, J. C. Hoff and A. Matin (1985). "Susceptibility of chemostat- grown Yersinia enterocolitica and Klebsiella pneumoniae to chlorine dioxide." Appl Environ Microbiol 49(1): 69-72.

Harakeh, M. S., A. Illescas and A. Martin (1988). "Inactivation of bacteria by Purogene." J Appl Bacteriol 64: 459-463.

Hartford, C. E. (1981). "The bequests of Moncrief and Moyer: an appraisal of topical therapy of burns - 1981 American Burn Association Presidential Address." J Trauma 21(10): 827-34.

Harvey, S. C. (1980). Antiseptics & Disinfectants; Fungicides; Ectoparasitides. Goodman & Gilman's The Pharmacological Basis of Therapeutics, 6th Ed. A. Goodman Gilman, L. Goodman and A. Gilman. New York, MacMillan Publishing: 976-977.

Heath, R. J., Y. T. Yu, M. A. Shapiro, E. Olson and C. O. Rock (1998). "Broad spectrum antimicrobial biocides target the FabI component of fatty acid synthesis." J Biol Chem 273(46): 30316-20.

284

Heffernan, W. P., C. Guion and R. J. Bull (1979). "Oxidative damage to the erythrocyte induced by sodium chlorite, in vitro." J. Environ. Pathol. Toxicol. 2: 1501-1510.

Heir, E., G. Sundheim and A. L. Holck (1999). "The qacG gene on plasmid pST94 confers resistance to quaternary ammonium compounds in staphylococci isolated from the food industry." J Appl Microbiol 86(3): 378-88.

Hendriks, J. G., J. R. van Horn, H. C. van der Mei and H. J. Busscher (2004). "Backgrounds of antibiotic-loaded bone cement and prosthesis-related infection." Biomaterials 25(3): 545-56.

Hernandez-Richter, T., H. M. Schardey, F. Wittmann, S. Mayr, M. Schmitt-Sody, S. Blasenbreu, M. M. Heiss, C. Gabka and M. K. Angele (2003). "Rifampin and Triclosan but not silver is effective in preventing bacterial infection of vascular dacron graft material." Eur J Vasc Endovasc Surg 26(5): 550-7.

Herrmann, M., P. E. Vaudaux, D. Pittet, R. Auckenthaler, P. D. Lew, F. Schumacher- Perdreau, G. Peters and F. A. Waldvogel (1988). "Fibronectin, fibrinogen, and laminin act as mediators of adherence of clinical staphylococcal isolates to foreign material." J Infect Dis 158(4): 693-701.

Hogt, A. H., J. Dankert, J. A. de Vries and J. Feijen (1983). "Adhesion of coagulase- negative staphylococci to biomaterials." J Gen Microbiol 129(9): 2959-68.

Hogt, A. H., J. Dankert and J. Feijen (1985). "Adhesion of Staphylococcus epidermidis and Staphylococcus saprophyticus to a hydrophobic biomaterial." J Gen Microbiol 131(9): 2485-91.

Hutchinson, W. S. and R. I. Derby (1947). "Improved flour bleaching process using chlorine dioxide produced by an electric arc method." Cereal Chemistry 24: 372-376.

Illingworth, B. L., K. Tweden, R. F. Schroeder and J. D. Cameron (1998). "In vivo efficacy of silver-coated (Silzone) infection-resistant polyester fabric against a biofilm- producing bacteria, Staphylococcus epidermidis." J Heart Valve Dis 7(5): 524-30.

Ingram, P. R., N. Z. Homer, R. A. Smith, A. R. Pitt, C. G. Wilson, O. Olejnik and C. M. Spickett (2003). "The interaction of sodium chlorite with phospholipids and glutathione: a comparison of effects in vitro, in mammalian and in microbial cells." Arch Biochem Biophys 410(1): 121-33.

Jacobs, P. T. and S. Lin (1987). Hydrogen peroxide plasma sterilization system. US Patent 4 643 876.

285

Jacobs, P. T. and S. Lin (1996). Gas-plasma sterilization. Irradiation of polymers: fundamentals and technological applications. S. W. S. R.L. Clough. Washington D.C., Americal Chemical Society: 21-239.

Jaeger, K., S. Zenz, B. Juttner, H. Ruschulte, E. Kuse, J. Heine, S. Piepenbrock, A. Ganser and M. Karthaus (2005). "Reduction of catheter-related infections in neutropenic patients: a prospective controlled randomized trial using a chlorhexidine and silver sulfadiazine-impregnated central venous catheter." Ann Hematol 84(4): 258-62.

Jansen, B., K. G. Kristinsson, S. Jansen, G. Peters and G. Pulverer (1992). "In vitro efficacy of a central venous catheter complexed with iodine to prevent bacterial colonization." J Antimicrob Chemother 30(2): 135-9.

Jansen, B., F. Schumacher-Perdreau, G. Peters, et al. (1987). Adhesion and Survival of S. epidermidis on Polyurethanes. ASM Annual Meeting, Abstract D-62.

Jansen, B., F. Schumacher-Perdreau, G. Peters and G. Pulverer (1989). "New aspects in the pathogenesis and prevention of polymer-associated foreign-body infections caused by coagulase-negative staphylococci." J Invest Surg 2(4): 361-80.

Jeng, D. K. and A. G. Woodworth (1990a). "Chlorine dioxide gas sterilization under square-wave conditions." Appl Environ Microbiol 56(2): 514-519.

Jeng, D. K. and A. G. Woodworth (1990b). "Chlorine dioxide gas sterilization of oxygenators in an industrial scale sterilizer: a successful model." Artif Organs 14(5): 361- 8.

Johnson & Johnson (1998). History of medical product sterilization. New York, Jonhson & Johnson.

Jones, M. V., T. M. Herd and H. J. Christie (1989). "Resistance of Pseudomonas aeruginosa to amphoteric and quaternary ammonium biocides." Microbios 58(234): 49- 61.

Kaatz, G. W. and S. M. Seo (1995). "Inducible NorA-mediated multidrug resistance in Staphylococcus aureus." Antimicrob Agents Chemother 39(12): 2650-5.

Kazama, H., H. Hamashima, M. Sasatsu and T. Arai (1998). "Distribution of the antiseptic-resistance gene qacE delta 1 in gram-positive bacteria." FEMS Microbiol Lett 165(2): 295-9.

Kenawy el, R., F. I. Abdel-Hay, A. el-Raheem, R. el-Shanshoury and M. H. el-Newehy (1998). "Biologically active polymers: synthesis and antimicrobial activity of modified 286

glycidyl methacrylate polymers having a quaternary ammonium and phosphonium groups." J Control Release 50(1-3): 145-52.

Keogh, J. R. and J. W. Eaton (1994). "Albumin binding surfaces for biomaterials." J Lab Clin Med 124(4): 537-45.

Keogh, J. R., F. F. Velander and J. W. Eaton (1992). "Albumin-binding surfaces for implantable devices." J Biomed Mater Res 26(4): 441-56.

Kim, J., M. R. Marshall, W. X. Du, W. S. Otwell and C. I. Wei (1999). "Determination of chlorate and chlorite and mutagenicity of seafood treated with aqueous chlorine dioxide." J Agric Food Chem 47(9): 3586-91.

Kitano, T., Y. Yutani, A. Shimazu, I. Yano, H. Ohashi and Y. Yamano (1996). "The role of physicochemical properties of biomaterials and bacterial cell adhesion in vitro." Int J Artif Organs 19(6): 353-8.

Klueh, U., V. Wagner, S. Kelly, A. Johnson and J. D. Bryers (2000). "Efficacy of silver- coated fabric to prevent bacterial colonization and subsequent device-based biofilm formation." J Biomed Mater Res 53(6): 621-31.

Koch, R. (1881). "On disinfection." Mittheilungen aus dem Kaiserlichen Gesunheitsamte 1: 234-282.

Kojima, Y., M. Tojo, D. A. Goldmann, T. D. Tosteson and G. B. Pier (1990). "Antibody to the capsular polysaccharide/adhesin protects rabbits against catheter-related bacteremia due to coagulase-negative staphylococci." J Infect Dis 162(2): 435-41.

Korich, D. G., J. R. Mead, M. S. Madore, N. A. Sinclair and C. R. Sterling (1990). "Effects of ozone, chlorine dioxide, chlorine, and monochloramine on Cryptosporidium parvum oocyst viability." Appl Environ Microbiol 56(5): 1423-8.

Kossiakoff, M. G. (1887). "De la propriété que possèdent les microbes de saccommoder aux milieux antiseptiques." Ann Inst Pasteur 1: 465-476.

Kovtunovitch, G. P. and L. A. Chernaya (1936). "Bactericidal effects of chlorine solutions from the interaction of potassium chlorate and hydrochloric acid." Sovetskaya Khirurgiya 2: 214-221.

Kowalski, J. B. (1998). Sterilization of medical devices, pharmaceutical components and barrier isolation systems with gaseous chlorine dioxide. Sterilization of medical products, vol VII. R. F. Morrisey and J. B. Kowalski. Champlain, N.Y., Polyscience Publications.

287

Kronig, B. and T. L. Paul (1897). The chemical foundations of the study of disinfection and of the action of poisons. Milestones in Microbiology. T. Brock, translator. Englewood Cliffs, Prentice-Hall, 1965: 163-176.

Kross, A. D. (1995). Treatment or prevention of microbial infections in mammals, PTC Patent WO 9 523 511.

Kulkarni, A. R., K. S. Soppimath, T. M. Aminabhavi and W. E. Rudzinski (2001). "In vitro release kinetics of cefadroxil-loaded sodium alginate interpenetrating network beads." Eur J Pharm Biopharm 51(2): 127-33.

Kuusela, P., T. Vartio, M. Vuento and E. B. Myhre (1985). "Attachment of staphylococci and streptococci on fibronectin, fibronectin fragments, and fibrinogen bound to a solid phase." Infect Immun 50(1): 77-81.

Kwok, C. S., T. A. Horbett and B. D. Ratner (1999a). "Design of infection-resistant antibiotic-releasing polymers. II. Controlled release of antibiotics through a plasma- deposited thin film barrier." J Control Release 62(3): 301-11.

Kwok, C. S., C. Wan, S. Hendricks, J. D. Bryers, T. A. Horbett and B. D. Ratner (1999b). "Design of infection-resistant antibiotic-releasing polymers: I. Fabrication and formulation." J Control Release 62(3): 289-99.

Lambert, P. A. (2002). "Cellular impermeability and uptake of biocides and antibiotics in gram-positive bacteria and mycobacteria." Symp Ser Soc Appl Microbiol(31): 46S-54S

Langanki, D., M. F. Ogle, J. D. Cameron, R. A. Lirtzman, R. F. Schroeder and M. W. Mirsch (1998). "Evaluation of a novel bioprosthetic heart valve incorporating anticalcification and antimicrobial technology in a sheep model." J Heart Valve Dis 7(6): 633-8.

Langlais, B., D. A. Reckow and D. R. Brink (1991). Ozone in water treatment: application and engineering. Chelsea, Lewis Publishers: 106-109.

Levy, S. B. (2000). "Antibiotic and antiseptic resistance: impact on public health." Pediatr Infect Dis J 19(10 Suppl): S120-2. Levy, S. B. (2002). "Active efflux, a common mechanism for biocide and antibiotic resistance." J Appl Microbiol 92 Suppl: 65S-71S.

Li, D. Q., F. Lundberg and A. Ljungh (2000). "Binding of von Willebrand factor by coagulase-negative staphylococci." J Med Microbiol 49(3): 217-25.

Littlejohn, T. G., I. T. Paulsen, M. T. Gillespie, J. M. Tennent, M. Midgley, I. G. Jones, A. S. Purewal and R. A. Skurray (1992). "Substrate specificity and energetics of 288

antiseptic and disinfectant resistance in Staphylococcus aureus." FEMS Microbiol Lett 74(2-3): 259-65.

Locci, R., G. Peters and G. Pulverer (1981). "Microbial colonization of prosthetic devices. I. Microtopographical characteristics of intravenous catheters as detected by scanning electron microscopy." Zentralbl Bakteriol Mikrobiol Hyg [B] 173(5): 285-92.

Lowbury, E. J. (1951). "Contamination of cetrimide and other fluids with Pseudomonas pyocyanea." Br J Ind Med 8(1): 22-5.

Lubbers, J. R., S. Chauan and J. R. Bianchine (1982). "Controlled clinical evaluations of chlorine dioxide, chlorite and chlorate in man." Environ Health Perspect 46: 57-62.

Lyarskii, P. P., T. V. Likhtman, N. V. Kareyev, N. I. Komarkova and S. E. Gleiberman (1984). "Physico-chemical and toxicologo-hygienic aspects of using ethylene oxide for the sterilization of medical appliances. Part I. Sorption and diffusion of ethylene oxide in polymeric materials." J Hyg Epidemiol Microbiol Immunol 28(2): 119-28.

Maillard, J. Y. (2002). "Bacterial target sites for biocide action." Symp Ser Soc Appl Microbiol(31): 16S-27S.

Marshall, C. P., F. C. Pearson, M. A. Sagona, W. Lee, R. L. Wathen, R. A. Ward and J. Dolovich (1985). "Reactions during hemodialysis caused by allergy to ethylene oxide gas sterilization." J Allergy Clin Immunol 75(5): 563-7.

Marshall, K. C. (1985). Mechanisms of bacterial adhesion at solid-water interfaces. Bacterial Adhesion. Mechanisms and Physiological Significance. D. C. Savage and M. Fletcher. New York, Plenum Press: 133-61.

Marshall, K. C., R. Stout and R. Mitchell (1971). "Mechanism of initial events in the sorption of marine bacteria to surfaces." J Gen Microbiol 68: 337-48.

Martinez-Martinez, L., S. Hernandez-Alles, S. Alberti, J. M. Tomas, V. J. Benedi and G. A. Jacoby (1996). "In vivo selection of porin-deficient mutants of Klebsiella pneumoniae with increased resistance to cefoxitin and expanded-spectrum-cephalosporins." Antimicrob Agents Chemother 40(2): 342-8.

Masschelein, W. L., ed. (1982). Ozonization manual for water and wastewater treatment. New York, John Wiley & Sons.

Masschelein, W. L. and R. G. Rice (1979). Chlorine dioxide chemistry and the environmental impact of oxychlorine compounds. Ann Arbor, MI, Ann Arbor Science Publishers: 111-145.

289

Matthews, I. P., C. Gibson and A. H. Samuel (1989). "Enhancement of the kinetics of the aeration of ethylene oxide sterilised polymers using microwave radiation." J Biomed Mater Res 23: 143.

Matthews, I. P., C. Gibson and A. H. Samuel (1994). "Sterilisation of implantable devices." Clin Mater 15(3): 191-215.

McAllister, E. W., L. C. Carey, P. G. Brady, R. Heller and S. G. Kovacs (1993). "The role of polymeric surface smoothness of biliary stents in bacterial adherence, biofilm deposition, and stent occlusion." Gastrointest Endosc 39(3): 422-5.

McCulloch, E. C. (1945). Disinfection and sterilization. Philadelphia, Lea & Febiger: 57- 68.

McDonnell, G. and A. D. Russell (1999). "Antiseptics and disinfectants: activity, action, and resistance." Clin Microbiol Rev 12(1): 147-79.

McGunnigle, R. G., J. A. Renner, S. J. Romano and R. A. Abodeely, Jr. (1975). "Residual ethylene oxide: levels in medical grade tubing and effects on an in vitro biologic system." J Biomed Mater Res 9(3): 273-83.

McMurry, L. M., M. Oethinger and S. B. Levy (1998). "Triclosan targets lipid synthesis." Nature 394(6693): 531-2.

Mechin, L., F. Dubois-Brissonnet, B. Heyd and J. Y. Leveau (1999). "Adaptation of Pseudomonas aeruginosa ATCC 15442 to didecyldimethylammonium bromide induces changes in membrane fatty acid composition and in resistance of cells." J Appl Microbiol 86(5): 859-66.

Mermel, L. A., S. M. Stolz and D. G. Maki (1993). "Surface antimicrobial activity of heparin-bonded and antiseptic-impregnated vascular catheters." J Infect Dis 167(4): 920- 4.

Miller, S. M. (1999). "Bacterial detoxification of Hg(II) and organomercurials." Essays Biochem 34: 17-30.

Modak, S. M. and C. L. Fox, Jr. (1973). "Binding of silver sulfadiazine to the cellular components of Pseudomonas aeruginosa." Biochem Pharmacol 22(19): 2391-404.

Mohammad, S. F. and R. Ardehali (2000). Quantitation of Bacterial Adhesion to Biomaterials using Radiolabeling Techniques. Handbook of Bacterial Adhesion: Principles, Methods, and Applications. F. An Y.H. and R.J. Totowa, Humana Press: 259- 272.

290

Mohammad, S. F., N. S. Topham, G. L. Burns and D. B. Olsen (1988). "Enhanced bacterial adhesion on surfaces pretreated with fibrinogen and fibronectin." ASAIO Trans 34(3): 573-7.

Mokdad, A. H., J. S. Marks, D. F. Stroup and J. L. Gerberding (2004). "Actual causes of death in the United States, 2000." JAMA 291(10): 1238-45.

Moore, G. S. and E. J. Calabrese (1982). "Toxicological effects of chlorite in the mouse." Environ Health Perspect 46: 31-7.

Muller, E., S. Takeda, D. A. Goldmann and G. B. Pier (1991). "Blood proteins do not promote adherence of coagulase-negative staphylococci to biomaterials." Infect Immun 59(9): 3323-6.

Munshi, N., N. Rapoport and W. G. Pitt (1997). "Ultrasonic activated drug delivery from Pluronic P-105 micelles." Cancer Lett 118(1): 13-9.

Naylor, P. T., D. Ruch and C. Browlon (1989). Fibronectin binding to orthopedic biomaterials and its subsequent role in bacterial adherence. Trans Ortho Res Soc.

Neu, H. C. (1992). "The crisis in antibiotic resistance." Science 257(5073): 1064-73.

Nikaido, H. (1994). "Prevention of drug access to bacterial targets: permeability barriers and active efflux." Science 264(5157): 382-8.

Nikaido, H. (1998). "Antibiotic resistance caused by gram-negative multidrug efflux pumps." Clin Infect Dis 27 Suppl 1: S32-41.

Noguchi, N., M. Hase, M. Kitta, M. Sasatsu, K. Deguchi and M. Kono (1999). "Antiseptic susceptibility and distribution of antiseptic-resistance genes in methicillin- resistant Staphylococcus aureus." FEMS Microbiol Lett 172(2): 247-53.

Noss, C. I., W. H. Dennis and V. P. Olivieri (1983). Reactivity of chlorine dioxide with nucleic acids and proteins. Water chlorination chemistry: environmental impact and health effects. Chelsea, MI, Lewis Publishers. 5: 1077-1086.

Noss, C. I., F. S. Hauchman and V. P. Olivieri (1986). "Chlorine dioxide reactivity with proteins." Water Research 20(3): 351-356.

O'Gara, J. P. and H. Humphreys (2001). "Staphylococcus epidermidis biofilms: importance and implications." J Med Microbiol 50(7): 582-7.

291

Okpara-Hofmann, J., M. Knoll, M. Durr, B. Schmitt and M. Borneff-Lipp (2005). "Comparison of low-temperature hydrogen peroxide gas plasma sterilization for endoscopes using various Sterrad models." J Hosp Infect 59(4): 280-5.

Olivieri, V. P., F. S. Hauchman and C. I. Noss (1985). Mode of action of chlorine dioxide on selected viruses. Water chlorination chemistry: environmental impact and health effects. R. L. Jolley. Chelsea, Lewis Publishers. 5: 619-634.

Olson, W. P. (1997). "Synergistic sterilization: a brief history." PDA J Pharm Sci Technol 51(3): 116-8.

Orme, J., D. H. Taylor, R. D. Laurie and R. J. Bull (1985). "Effects of chlorine dioxide on thyroid function in neonatal rats." J Toxicol Environ Health 15(2): 315-22.

Oshima, H. and M. Nakamura (1994). "A study on reference standard for cytotoxicity assay of biomaterials." Biomed Mater Eng 4(4): 327-32.

Ostendorf, T., A. Meinhold, C. Harter, H. Salwender, G. Egerer, H. K. Geiss, A. D. Ho and H. Goldschmidt (2005). "Chlorhexidine and silver-sulfadiazine coated central venous catheters in haematological patients - a double-blind, randomised, prospective, controlled trial." Support Care Cancer.

Pao, S. S., I. T. Paulsen and M. H. Saier, Jr. (1998). "Major facilitator superfamily." Microbiol Mol Biol Rev 62(1): 1-34.

Pariente, J. L., B. S. Kim and A. Atala (2002). "In vitro biocompatibility evaluation of naturally derived and synthetic biomaterials using normal human bladder smooth muscle cells." J Urol 167(4): 1867-71.

Paulsen, I. T., M. H. Brown, S. J. Dunstan and R. A. Skurray (1995). "Molecular characterization of the staphylococcal multidrug resistance export protein QacC." J Bacteriol 177(10): 2827-33.

Paulsen, I. T., M. H. Brown and R. A. Skurray (1996). "Proton-dependent multidrug efflux systems." Microbiol Rev 60(4): 575-608.

Paulsen, I. T., M. K. Sliwinski and M. H. Saier, Jr. (1998). "Microbial genome analyses: global comparisons of transport capabilities based on phylogenies, bioenergetics and substrate specificities." J Mol Biol 277(3): 573-92.

Peeters, J. E., E. A. Mazas, W. J. Masschelein, I. Villacorta Martiez de Maturana and E. Debacker (1989). "Effect of disinfection of drinking water with ozone or chlorine dioxide on survival of Cryptosporidium parvum oocysts." Appl Environ Microbiol 55(6): 1519- 22. 292

Phillips, C. R. (1977). Gaseous sterilization. Disinfection, Sterilization and Preservation. S. S. Block. Philadelphia, Lea & Febiger.

Poole, K. (2000a). "Efflux-mediated resistance to fluoroquinolones in gram-negative bacteria." Antimicrob Agents Chemother 44(9): 2233-41.

Poole, K. (2000b). "Efflux-mediated resistance to fluoroquinolones in gram-positive bacteria and the mycobacteria." Antimicrob Agents Chemother 44(10): 2595-9.

Poole, K. (2002). "Mechanisms of bacterial biocide and antibiotic resistance." J Appl Microbiol 92 Suppl: 55S-64S.

Pringle, J. H. and M. Fletcher (1986). "Influence of substratum hydration and adsorbed macromolecules on bacterial attachment to surfaces." Appl Environ Microbiol 51(6): 1321-5.

Proctor, R. A., D. F. Mosher and P. J. Olbrantz (1982). "Fibronectin binding to Staphylococcus aureus." J Biol Chem 257(24): 14788-94.

Quirynen, M., H. C. van der Mei, C. M. Bollen, A. Schotte, M. Marechal, G. I. Doornbusch, I. Naert, H. J. Busscher and D. van Steenberghe (1993). "An in vivo study of the influence of the surface roughness of implants on the microbiology of supra- and subgingival plaque." J Dent Res 72(9): 1304-9.

Raad, I., R. Hachem, A. Zermeno, M. Dumo and G. P. Bodey (1996). "In vitro antimicrobial efficacy of silver iontophoretic catheter." Biomaterials 17(11): 1055-9.

Raad, I. and H. Hanna (1999). "Intravascular catheters impregnated with antimicrobial agents: a milestone in the prevention of bloodstream infections." Support Care Cancer 7(6): 386-90.

Ramirez-Orozco, M., N. Y. Hernandez-Saavedra and J. L. Ochoa (2001). "Debaryomyces hansenii growth in nonsterile seawater ClO2-peptone-containing medium." Can J Microbiol 47(7): 676-9.

Rees, E. N., S. E. Tebbs and T. S. Elliott (1998). "Role of antimicrobial-impregnated polymer and Teflon in the prevention of biliary stent blockage." J Hosp Infect 39(4): 323- 9.

Revis, N. W., P. McCauley, R. Bull and G. Holdsworth (1986). "Relationship of drinking water disinfectants to plasma cholesterol and thyroid hormone levels in experimental studies." Proc Natl Acad Sci U S A 83(5): 1485-9.

293

Reynolds, E. C. and A. Wong (1983). "Effect of adsorbed protein on hydroxyapatite zeta potential and Streptococcus mutans adherence." Infect Immun 39(3): 1285-90.

Rice, R. G. (1986). Application of ozone in water and wastewater treatment. Analytical aspects of ozone: treatment of water and wastewater. R. G. Rice. Chelsea, Lewis Publishers: 7-26.

Rice, R. G., D. M. Graham and W. H. Glaze (1997). Ozone preservation of foods and foodstuffs. Proceedings of the 13th Ozone World Congress: International Ozone Association, Kyoto, Japan.

Ricketts, C. R., E. J. Lowbury, J. C. Lawrence, M. Hall and M. D. Wilkins (1970). "Mechanism of prophylaxis by silver compounds against infection of burns." Br Med J 1(707): 444-6.

Rideal, S. and J. T. A. Walker (1903). "The standardization of disinfectants." Journal of the Royal Sanitary Insitute 24: 424-441.

Ridenour, G. M. and E. H. Armbruster (1949). "Bactericidal effect of chlorine dioxide." Journal of the American Water Works Association 41: 537-550.

Ridenour, G. M. and R. S. Ingols (1947). "Bactericidal properties of chlorine dioxide." Journal of the American Water Works Association 39: 561-567. Risbud, M. V. and R. R. Bhonde (2000). "Polyacrylamide-chitosan hydrogels: in vitro biocompatibility and sustained antibiotic release studies." Drug Deliv 7(2): 69-75.

Rochart, C. and K. Uzdins (1947). "Katadyn (Silver Preparation): Clinical Applications." Schweiz Med Wochenschr 77: 1100-4.

Roller, S. D., V. P. Olivieri and K. Kawata (1980). "Mode of bacterial inactivation by chlorine dioxide." Water Research 14: 635-642.

Rosato, A. E., S. M. Tallent, M. B. Edmond and G. M. Bearman (2004). "Susceptibility of coagulase-negative staphylococcal nosocomial bloodstream isolates to the chlorhexidine/silver sulfadiazine-impregnated central venous catheter." Am J Infect Control 32(8): 486-8.

Rosenblatt, A. A. and J. E. Knapp (1988). Chlorine dioxide gas sterilization. HIMA Conference Proceedings.

Rosenblatt, D. H., A. A. Rosenblatt and J. E. Knapp (1985). Use of chlorine dioxide as a chemosterilizing agent. US Patent 4 504 442.

294

Rosenblatt, D. H., A. A. Rosenblatt and J. E. Knapp (1987). Use of chlorine dioxide as a chemosterilizing agent. US Patent 4 681 739.

Ross, W. C. (1953). Chemistry of Cytotoxic Alkylating Agents, IV. 1,2-Epoxides. Advances in Cancer Research. A. E. H. J.P. Greenstein. New York, NY Academic Press, Inc. 1: 429-435.

Russell, A. D. (1991). "Mechanisms of bacterial resistance to non-antibiotics: food additives and food and pharmaceutical preservatives." J Appl Bacteriol 71(3): 191-201.

Russell, A. D. (1998). Mechanisms of bacterial resistance to antibiotics and biocides. Progress in Medicinal Chemistry. G. P. Ellis, Luscombe, D.K., Oxford, A.W., Elsevier.

Russell, A. D. (1999). "Bacterial resistance to disinfectants: present knowledge and future problems." J Hosp Infect 43 Suppl: S57-68.

Russell, A. D. (2000). "Do biocides select for antibiotic resistance?" J Pharm Pharmacol 52(2): 227-33.

Russell, A. D. (2002). "Introduction of biocides into clinical practice and the impact on antibiotic-resistant bacteria." J Appl Microbiol 92 Suppl: 121S-35S.

Russell, A. D., Chopra, I. (1996). Understanding Antibacterial Action and Resistance. Chichester, Ellis Horwood.

Russell, A. D., J. R. Furr and J. Y. Maillard (1997). "Synergistic sterilization." PDA J Pharm Sci Technol 51(5): 174-5.

Ryden, C., K. Rubin, P. Speziale, M. Hook, M. Lindberg and T. Wadstrom (1983). "Fibronectin receptors from Staphylococcus aureus." J Biol Chem 258(5): 3396-401.

Saier, M. H., Jr., R. Tam, A. Reizer and J. Reizer (1994). "Two novel families of bacterial membrane proteins concerned with nodulation, cell division and transport." Mol Microbiol 11(5): 841-7.

Samuel, A. H. and I. P. Matthews (1985). International patent application no. PCT/GB85/0050.

Sansebastiano, G., C. Cesari and E. Bellelli (1986). "Further investigations on water disinfection by chlorine dioxide." Igiene Moderna (Parma) 85: 358-380.

Sansebastiano, G., G. Mori and M. L. Tanzi (1983). "Chlorine dioxide: methods of analysis and kinetics of inactivation of poliovirus type I." Igiene Moderna (Parma) 79: 61-91. 295

Satou, N., J. Satou, H. Shintani and K. Okuda (1988). "Adherence of streptococci to surface-modified glass." J Gen Microbiol 134(5): 1299-305.

Scatina, J., M. S. Abdel-Rahman and E. Goldman (1985). "The inhibitory effect of Alcide, an antimicrobial drug, on protein synthesis in Escherichia coli." J Appl Toxicol 5(6): 388-94.

Schaufler, C. (1933). "Antiseptic effects of chlorine solutions from interactions of potassium chlorate and hydrochloric acid." Zentralbl Chir 60: 2497-2500.

Schierholz, J. M. and J. Beuth (2001). "Implant infections: a haven for opportunistic bacteria." J Hosp Infect 49(2): 87-93.

Schierholz, J. M., J. Beuth, G. Pulverer and D. P. Konig (1999). "Silver-containing polymers." Antimicrob Agents Chemother 43(11): 2819-20; author reply 2820-1.

Schierholz, J. M., L. J. Lucas, A. Rump and G. Pulverer (1998a). "Efficacy of silver- coated medical devices." J Hosp Infect 40(4): 257-62.

Schierholz, J. M., H. Steinhauser, A. F. Rump, R. Berkels and G. Pulverer (1997). "Controlled release of antibiotics from biomedical polyurethanes: morphological and structural features." Biomaterials 18(12): 839-44.

Schierholz, J. M., Z. Wachol-Drewek, L. J. Lucas and G. Pulverer (1998b). "Activity of silver ions in different media." Zentralbl Bakteriol 287(4): 411-20.

Schweizer, H. P. (2001). "Triclosan: a widely used biocide and its link to antibiotics." FEMS Microbiol Lett 202(1): 1-7.

Sefton, A. M. (2002). "Mechanisms of antimicrobial resistance: their clinical relevance in the new millennium." Drugs 62(4): 557-66.

Seguy, N., H. F. Hildebrand and J. M. Haguenoer (1994). "Toxic action of ethylene oxide on pulmonary cells (L132) cultured under aerobic conditions." Toxicol Lett 70(1): 23-32

Selwyn, S. (1980). Chapter 1. The discovery and evolution of the penicillins and cephalosporins. The Beta-lactam Antibiotics: Penicillins and Cephalosporins in Perspective. London, Hodder and Stoughton: 1-55.

Shikama, K. and K. I. Miura (1976). "Equilibrium studies on the formaldehyde reaction with native DNA." Eur J Biochem 63(1): 39-46.

296

Sidgwick, N. V. (1951). Halogen oxides. The chemical elements and their compounds. Oxford, Oxford University Press. II: 1202-1218.

Sieradzki, K., R. B. Roberts, D. Serur, J. Hargrave and A. Tomasz (1998). "Recurrent peritonitis in a patient on dialysis and prophylactic vancomycin." Lancet 351(9106): 880- 1.

Silverman, G. J. (1983). Sterilization by ionizing radiation. Disinfection, Sterilization, and Preservation, 3rd Ed. S. S. Block. Philadelphia, Lea & Febiger: 91-92.

Skowronski, G. A., M. S. Abdel-Rahman, S. E. Gerges and K. M. Klein (1985). "Teratologic evaluation of Alcide liquid in rats and mice. I." J Appl Toxicol 5(2): 97-103.

Spallanzani, L. (1776). Observations and experience about the animalcules of the infusion. Writings on the natural animals and vegetables of the Abbey Spallanzani. Modena. Paris, J. Duplain, 1787.

Stanley, P., E. Bertranou, F. Forest and L. Langevin (1971). "Toxicity of ethylene oxide sterilization of polyvinyl chloride in open-heart surgery." J Thorac Cardiovasc Surg 61(2): 309-14.

Strachan, C. J., S. W. Newsom and T. R. Ashton (1991). "The clinical use of an antibiotic-bonded graft." Eur J Vasc Surg 5(6): 627-32.

Suh, D. H., M. S. Abdel-Rahman and R. J. Bull (1983). "Effect of chlorine dioxide and its metabolites in drinking water on fetal development in rats." J Appl Toxicol 3(2): 75-9.

Suller, M. T. and A. D. Russell (2000). "Triclosan and antibiotic resistance in Staphylococcus aureus." J Antimicrob Chemother 46(1): 11-8.

Suzuki, Y., M. Tanihara, Y. Nishimura, K. Suzuki, Y. Kakimaru and Y. Shimizu (1998). "A new drug delivery system with controlled release of antibiotic only in the presence of infection." J Biomed Mater Res 42(1): 112-6.

Symons, J. M. et al. (1976). Interim guide for the control of chlorform and other trihalomethanes. Cincinnati, OH, Water Supply Research Division, Municipal Environmental Research Laboratory, US Environmental Protection Agency.

Tambyah, P. A., K. T. Halvorson and D. G. Maki (1999). "A prospective study of pathogenesis of catheter-associated urinary tract infections." Mayo Clin Proc 74(2): 131- 6.

297

Tanihara, M., Y. Suzuki, Y. Nishimura, K. Suzuki, Y. Kakimaru and Y. Fukunishi (1999). "A novel microbial infection-responsive drug release system." J Pharm Sci 88(5): 510-4.

Tattawasart, U., A. C. Hann, J. Y. Maillard, J. R. Furr and A. D. Russell (2000a). "Cytological changes in chlorhexidine-resistant isolates of Pseudomonas stutzeri." J Antimicrob Chemother 45(2): 145-52.

Tattawasart, U., J. Y. Maillard, J. R. Furr and A. D. Russell (1999). "Development of resistance to chlorhexidine diacetate and cetylpyridinium chloride in Pseudomonas stutzeri and changes in antibiotic susceptibility." J Hosp Infect 42(3): 219-29.

Tattawasart, U., J. Y. Maillard, J. R. Furr and A. D. Russell (2000b). "Outer membrane changes in Pseudomonas stutzeri resistant to chlorhexidine diacetate and cetylpyridinium chloride." Int J Antimicrob Agents 16(3): 233-8.

Thomson, K. S. and E. Smith Moland (2000). "Version 2000: the new beta-lactamases of Gram-negative bacteria at the dawn of the new millennium." Microbes Infect 2(10): 1225-35.

Timm, D. and D. Gonzales (1997). "Effect of sterilization on microstructure and function of microsurgical scissors." Surgical Services Management 3: 47-49.

Timmerman, C. P., A. Fleer, J. M. Besnier, L. De Graaf, F. Cremers and J. Verhoef (1991). "Characterization of a proteinaceous adhesin of Staphylococcus epidermidis which mediates attachment to polystyrene." Infect Immun 59(11): 4187-92.

Tojo, M., N. Yamashita, D. A. Goldmann and G. B. Pier (1988). "Isolation and characterization of a capsular polysaccharide adhesin from Staphylococcus epidermidis." J Infect Dis 157(4): 713-22.

Trecant, M., J. Guicheux, G. Grimandi, M. Leroy and G. Daculsi (1997). "Dynamic compaction: a new process to compact therapeutic agent-loaded calcium phosphates." Biomaterials 18(2): 141-5.

TSO3 (2005). Frequently asked questions, http://www.tso3.com/en/faq.htm. 2005.

Tuthill, R. W., R. A. Giusti, G. S. Moore and E. J. Calabrese (1982). "Health effects among newborns after prenatal exposure to ClO2-disinfected drinking water." Environ Health Perspect 46: 39-45.

Tweden, K. S., J. D. Cameron, A. J. Razzouk, R. W. Bianco, W. R. Holmberg, R. J. Bricault, J. E. Barry and E. Tobin (1997a). "Silver modification of polyethylene terephthalate textiles for antimicrobial protection." ASAIO J 43(5): M475-81. 298

Tweden, K. S., J. D. Cameron, A. J. Razzouk, W. R. Holmberg and S. J. Kelly (1997b). "Biocompatibility of silver-modified polyester for antimicrobial protection of prosthetic valves." J Heart Valve Dis 6(5): 553-61.

US Environmental Protection Agency (1977). Ozone, chlorine dioxide, and chloroamines as alternatives to chlorine for disinfection of drinking water. State of the art. Cincinnato, OH, Water Supply Research, Office of Research and Development, US Environmental Protection Agency.

van de Belt, H., D. Neut, W. Schenk, J. R. van Horn, H. C. van Der Mei and H. J. Busscher (2001). "Staphylococcus aureus biofilm formation on different gentamicin- loaded polymethylmethacrylate bone cements." Biomaterials 22(12): 1607-11.

van de Belt, H., D. Neut, J. R. van Horn, H. C. van der Mei, W. Schenk and H. J. Busscher (1999). "...or not to treat?" Nat Med 5(4): 358-9.

van Leeuwenhoek, A. (1684). "An abstract of a letter from Mr. Anthony Leeuwenhoek at Delft dated Sept. 17, 1683, about some microscopical observations about animals in the scurf of the teeth." Philosophical Transactions of the Royal Society of London 14: 568- 574. van Loosdrecht, M. C., J. Lyklema, W. Norde, G. Schraa and A. J. Zehnder (1987). "The role of bacterial cell wall hydrophobicity in adhesion." Appl Environ Microbiol 53(8): 1893-7. van Veen, H. W. and W. N. Konings (1998). "The ABC family of multidrug transporters in microorganisms." Biochim Biophys Acta 1365(1-2): 31-6.

Veenstra, G. J., F. F. Cremers, H. van Dijk and A. Fleer (1996). "Ultrastructural organization and regulation of a biomaterial adhesin of Staphylococcus epidermidis." J Bacteriol 178(2): 537-41.

Venngopalan, M. and A. Shih (1981). "Reaction of hydrogen peroxide vapor dissociated in a microwave plasma." Plasma Chemistry and Plasma Processing 1: 191-200.

Vicaretti, M., W. J. Hawthorne, P. Y. Ao and J. P. Fletcher (1998). "An increased concentration of rifampicin bonded to gelatin-sealed Dacron reduces the incidence of subsequent graft infections following a staphylococcal challenge." Cardiovasc Surg 6(3): 268-73.

Vink, P. (1986). "Residual formaldehyde in steam formaldehyde sterilised materials." Biomaterials 7: 221.

299

Vologodskii, A. V. and M. D. Frank-Kamenetskii (1975). "Theoretical Study of DNA unwinding under the action of formaldehyde." J Theor Biol 55(1): 153-66.

Wadstrom, T., L. Switalski, P. Speciale et al. (1985). Binding of microbial pathogens to connective tissue fibronectin: An early step in localized and invasive infections. Bayer Symp III: The Pathogenesis of Bacterial Infections. Berlin, Springer: 193-207.

Walker, J. T., C. W. Mackerness, D. Mallon, T. Makin, T. Williets and C. W. Keevil (1995). "Control of Legionella pneumophila in a hospital water system by chlorine dioxide." J Ind Microbiol 15(4): 384-90.

Ward, H. M. (1893). "Further experiments on the action of light on Bacillus anthracis." Proceedings of the Royal Society 53: 23-45.

Washington Post (2002). Brentwood cleanup successful so far, http://www.washingtonpost.com/ac2/wp-dyn?pagename=article&contentId=A49567- 2002Aug22¬Found=true.

Webster, D. A., J. A. Spadaro, R. O. Becker and S. Kramer (1981). "Silver anode treatment of chronic osteomyelitis." Clin Orthop Relat Res(161): 105-14.

Whalen, R. L., C. Cai, L. M. Thompson, M. J. Sarrasin, D. J. Dempsey and M. A. Bowen (1997). "An infection inhibiting urinary catheter material." ASAIO J 43(5): M842-7.

White, G. C. (1972). Chlorine dioxide. Handbook of chlorination. New York, Van Nostrand: 596-616.

Whitmore, T. N. and S. Denny (1992). "The effect of disinfectants on a geosmin- producing strain of Streptomyces griseus." J Appl Bacteriol 72(2): 160-5.

Winder, C. L., I. S. Al-Adham, S. M. Abdel Malek, T. E. Buultjens, A. J. Horrocks and P. J. Collier (2000). "Outer membrane protein shifts in biocide-resistant Pseudomonas aeruginosa PAO1." J Appl Microbiol 89(2): 289-95.

Windholz, M. (1983). Meck index, 10th edition. Rahway, NJ, Merck and Co.: 324.

Wright, G. D. (1999). "Aminoglycoside-modifying enzymes." Curr Opin Microbiol 2(5): 499-503.

Wright, G. D., A. M. Berghuis and S. Mobashery (1998). "Aminoglycoside antibiotics. Structures, functions, and resistance." Adv Exp Med Biol 456: 27-69.

Xu, K. D., G. A. McFeters and P. S. Stewart (2000). "Biofilm resistance to antimicrobial agents." Microbiology 146 ( Pt 3): 547-9. 300

Yanagisawa, N., D. Q. Li and A. Ljungh (2004). "Protein adsorption on ex vivo catheters and polymers exposed to peritoneal dialysis effluent." Perit Dial Int 24(3): 264-73.

Yokose, Y., K. Uchida, D. Nakae, K. Shiraiwa, K. Yamamoto and Y. Konishi (1987). "Studies of carcinogenicity of sodium chlorite in B6C3F1 mice." Environ Health Perspect 76: 205-10.

Young, S. B. and P. Setlow (2003). "Mechanisms of killing of Bacillus subtilis spores by hypochlorite and chlorine dioxide." J Appl Microbiol 95(1): 54-67.

Zdanowski, Z., E. Ribbe and C. Schalen (1993). "Influence of some plasma proteins on in vitro bacterial adherence to PTFE and Dacron vascular prostheses." APMIS 101(12): 926-32.

Zgurskaya, H. I. and H. Nikaido (2000). "Multidrug resistance mechanisms: drug efflux across two membranes." Mol Microbiol 37(2): 219-25.

Zhou, W., S. M. Kuebler, K. L. Braun, T. Yu, J. K. Cammack, C. K. Ober, J. W. Perry and S. R. Marder (2002). "An efficient two-photon-generated photoacid applied to positive-tone 3D microfabrication." Science 296(5570): 1106-9.

Zoeteman, B. C., J. Hrubec, E. de Greef and H. J. Kool (1982). "Mutagenic activity associated with by-products of drinking water disinfection by chlorine, chlorine dioxide, ozone and UV-irradiation." Environ Health Perspect 46: 197-205.