Drinking Water Microflora Biofilms and Chlorine Susceptibility

University of Calgary PRISM: University of Calgary's Digital Repository

Graduate Studies The Vault: Electronic Theses and Dissertations

2012-07-19 Drinking water microflora biofilms and chlorine susceptibility

Schwering, Monika C.

Schwering, M. C. (2012). Drinking water microflora biofilms and chlorine susceptibility (Unpublished master's thesis). University of Calgary, Calgary, AB. doi:10.11575/PRISM/25749 http://hdl.handle.net/11023/128 master thesis

University of Calgary graduate students retain copyright ownership and moral rights for their thesis. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission. Downloaded from PRISM: https://prism.ucalgary.ca UNIVERSITY OF CALGARY

Drinking water microflora biofilms and chlorine susceptibility

Monika C. Schwering

A THESIS

SUBMITTED TO THE FACULTY OF GRADUATE STUDIES

IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE

DEGREE OF MASTER OF SCIENCE

DEPARTMENT OF BIOLOGICAL SCIENCE

CALGARY, ALBERTA

JULY, 2012

FACULTY OF GRADUATE STUDIES

The undersigned certify that they have read, and recommend to the Faculty of Graduate

Studies for acceptance, a thesis entitled "Drinking Water Microflora Biofilms and

Chlorine Susceptibility" submitted by Monika C. Schwering in partial fulfilment of the requirements of the degree of Master of Science.

Supervisor, Dr. Howard Ceri, Department of Biological Sciences

Interim Supervisor, Dr. Raymond J. Turner, Department of Biological Sciences

Dr, Lisa M. Gieg, Department of Biological Sciences

Dr. Marie Louie, Departments of Microbiology, Immunology and Infectious Diseases and Pathology and Laboratory Medicine

Dr. Glen D. Armstrong, Department of Microbiology, Immunology and Infectious Diseases

Date

ii Abstract

Waterborne disease outbreaks are especially dangerous in immunocompromised individuals and can be caused by biofilm formation in water systems. The aim of this work was to collect a group of environmental isolates, including opportunistic pathogens, from treated water systems with the purpose of creating a model drinking water system biofilm. This model biofilm would be used to explore the resistance of biofilms to chlorine at levels typical of a water distribution system. Isolates for the model biofilm were collected from Calgary and Ontario, sequenced and then as single and multi-species biofilms exposed to chlorine. The resistance, biofilm structure and microbial community were examined. It was found that biofilm organisms are consistently more resistant than planktonic and that multi-species biofilms even more so. Little change was seen in biofilm communities after treatment. The 3D structure of the biofilm appeared to have a role in resistance by limiting diffusion and protecting inner cells.

iii Acknowledgements

I would first of all like to thank my supervisors, Dr. Howard Ceri and Dr. Ray

Turner. I really appreciate the guidance you both have given me over the past few years.

Your support and belief in me has meant a lot during this project. I would also like to thank my committee members, Dr. Marie Louie and Dr. Lisa Gieg, you have both encouraged me and provided unique perspectives on my project that have helped me to get the most out of my research and my time at the U of C. I would also like to thank

Rhonda Schop of the Ontario MoE. Working with you gave me a real appreciation for the importance of this research start on the path to where I am today.

Secondly I would like to acknowledge the members of the Ceri lab and the

Biofilm Research Group for their help and friendship. Carol Stremick for your help with so many things in the lab; my summer students, Joanna Song and Jessica Leong, you were invaluable; Suzie Golby for proofreading and SEM help; Sean Booth for your help as a sounding board for ideas; and to all my other lab members: Michelle Stan, Lisa

Bieuker, Mark Stanton, Kim Whilloughby and Marc Demeter for making some of the long days in the lab more fun.

Finally I would like to thank those people in my life who have helped me in this project in a less technical, but no less important, way. First of all my wonderful parents

Frances and Rob Schwering and my brilliant sister Taralyn for all your love and encouragement; my good friend Jessica Phillips for the long distance support and laughs; and to my wonderful extended family, both back home in Toronto and the family that welcomed me as one of their own here in Calgary. To everyone, thank you all so much, I would never have been able to accomplish what I have without you.

iv Table of Contents

Approval Page...... ii Abstract...... iii Acknowledgements...... iv Table of Contents...... v List of Tables ...... ix List of Figures and Illustrations ...... x List of Symbols, Abbreviations and Nomenclature...... xii

CHAPTER ONE: INTRODUCTION...... 1 1.1 Introduction...... 1 1.1.1 Hypothesis ...... 5 1.1.2 Project objective and aims...... 5 1.2 Drinking water systems: treatment and monitoring...... 6 1.2.1 Source water and the multi-barrier approach to water safety...... 6 1.2.2 Primary treatment ...... 9 1.2.2.1 UV disinfection...... 10 1.2.2.2 Disinfection by-products...... 11 1.2.3 Distribution systems and secondary disinfection ...... 12 1.2.4 Water quality monitoring ...... 13 1.3 Bacterial biofilms...... 14 1.3.1 Biofilm formation...... 15 1.3.2 Phenotypic variation between biofilm and planktonic cells...... 16 1.3.3 Bacterial advantages of biofilm growth ...... 17 1.4 Biofilm growth in water systems...... 18 1.4.1 The distribution system environment ...... 19 1.4.2 Disinfectant effects on planktonic and biofilm bacteria...... 21 1.5 Drinking water microflora ...... 23 1.5.1 Drinking water distribution system native flora...... 23 1.5.2 Pathogenic organisms in drinking water ...... 24 1.5.2.1 Indicator organisms: Escherichia coli and Total Coliforms...... 26 1.5.2.2 Pseudomonas aeruginosa ...... 28 1.5.2.3 Stenotrophomonas maltophilia ...... 29 1.5.2.4 Legionella pneumophila ...... 29 1.5.3 Eukaryotic distribution system organisms...... 30

CHAPTER TWO: MATERIALS AND METHODS ...... 32 2.1 Media ...... 32 2.1.1 Growth media ...... 32 2.1.2 Selective and differential media ...... 33 2.2 Collection of bacterial isolates from drinking water ...... 34 2.2.1 Samples provided by laboratories ...... 35 2.2.1.1 Infection Prevention and Control Laboratory...... 35 2.2.1.2 Provincial Laboratory for Public Health, (ProvLab), Calgary, AB...... 35

v 2.2.2 Collection of samples for new isolates...... 36 2.2.2.1 Ontario water systems...... 36 2.2.2.2 University of Calgary Biological Sciences building...... 37 2.2.2.3 Alberta Children's Hospital...... 37 2.3 Isolate identification and molecular analysis...... 39 2.3.1 Single-species DNA extraction ...... 39 2.3.2 16S rRNA gene amplification and sequencing ...... 40 2.3.2.1 16S rRNA PCR amplification...... 40 2.3.2.2 Sequencing and identification...... 41 2.4 Biofilm analysis ...... 42 2.4.1 Biofilm inoculation...... 42 2.4.1.1 Legionella pneumophila biofilms ...... 43 2.4.2 Single-species biofilm growth evaluation ...... 46 2.4.3 Single-species minimum biofilm eradication concentration assay ...... 47 2.4.4 Planktonic single-species minimum inhibitory concentration assay...... 48 2.4.5 Multi-species minimum biofilm eradication concentration assay...... 49 2.5 Community analysis of multi-species biofilms ...... 50 2.5.1 Biofilm challenge ...... 50 2.5.2 DNA extraction from pegs ...... 50 2.5.3 Amplification with labelled primer ...... 51 2.5.4 Restriction enzyme digest...... 51 2.6 Biofilm microscopy ...... 53 2.6.1 Confocal scanning laser microscopy (CSLM) ...... 53 2.6.2 Scanning electron microscopy (SEM)...... 53

CHAPTER THREE: ISOLATION AND IDENTIFICATION OF BACTERIA FROM DRINKING WATER DISTRIBUTION SYSTEMS...... 55 3.1 Aims...... 55 3.2 Isolate collection...... 56 3.2.1 Isolates collected at the Alberta Children's Hospital...... 56 3.2.2 Isolates provided by the Infection Prevention and Control laboratory...... 56 3.2.3 Isolates collected at the Ontario Ministry of the Environment...... 58 3.2.4 Isolates collected from University of Calgary Biological Sciences building..60 3.3 Isolate identification ...... 60 3.3.1 Duplicate genera and species within sampling sets...... 66 3.3.2 Duplicate genera and species between sampling sets...... 68 3.4 Terminal Restriction Fragment Length Polymorphism Library construction ...... 68 3.4.1 Selection of isolates for multi-species biofilm model ...... 68 3.4.2 Evaluating restriction enzymes in silico...... 71 3.4.3 Establishing a library of actual terminal restriction fragment lengths for model multi-species biofilm ...... 74 3.5 Summary...... 80

CHAPTER FOUR: ISOLATE GROWTH UNDER BIOFILM FORMING CONDITIONS AND PLANKTONIC AND BIOFILM SUSCEPTIBILITY TO CHLORINATION ...... 81 4.1 Aims...... 81

vi 4.2 Biofilm growth...... 81 4.2.1 Non-biofilm forming isolates ...... 87 4.3 Biofilm equivalency...... 89 4.4 Biofilm susceptibility...... 92 4.5 Summary...... 96

CHAPTER FIVE: MULTI-SPECIES BIOFILMS: GROWTH, STRUCTURE, CHLORINE SUSCEPTIBILITY AND COMMUNITY DYNAMICS...... 98 5.1 Aims...... 98 5.2 Results...... 99 5.2.1 Selecting a multi-species biofilm model inoculation schedule ...... 99 5.2.2 Scanning electron microscopy of model drinking water system biofilm ...... 102 5.2.3 Multi-species biofilm chlorine exposure...... 106 5.2.3.1 Escherichia coli MEC-8 ...... 106 5.2.3.2 Enterobacter cloacae MTC-21...... 111 5.2.3.3 Pseudomonas aeruginosa PAE-1 ...... 118 5.2.3.4 Stenotrophomonas maltophilia FH-W1...... 123 5.2.4 Legionella pneumophila in the model multi-species biofilm ...... 130 5.2.5 Other multi-species biofilm methods investigated ...... 131 5.3 Summary...... 133

CHAPTER SIX: DISCUSSION, CONCLUSIONS AND FUTURE DIRECTIONS .....135 6.1 Summary of results ...... 135 6.2 Discussion of results ...... 138 6.2.1 Isolation of environmental organisms ...... 138 6.2.2 Creating a mixed-species model water system biofilm...... 140 6.2.3 Single species biofilm resistance to chlorine disinfection...... 141 6.2.4 Multi-species biofilm resistance to chlorine disinfection...... 143 6.2.4.1 Comparison between multi-species biofilms...... 143 6.2.5 Microscopy...... 145 6.2.5.1 Scanning electron microscopy...... 145 6.2.5.2 Confocal scanning laser microscopy ...... 146 6.2.6 T-RFLP as a molecular method to study bacterial communities ...... 148 6.2.7 T-RFLP analysis of chlorine treated multi-species biofilms...... 149 6.2.8 Absence of Legionella pneumophila in multi-species biofilms ...... 151 6.2.9 Continuous-flow biofilm reactor ...... 154 6.3 Future directions ...... 155 6.3.1 Multi-species biofilm model challenge with further biocidal agents ...... 156 6.3.2 Addition of organisms to the multi-species community...... 156 6.3.3 Localization of opportunistic pathogens within the biofilm...... 156 6.3.4 Upscale of CBD bioreactor to a continuous-flow model system...... 157 6.4 Conclusions...... 157

REFERENCES ...... 159

APPENDIX I ...... 169

vii APPENDIX II ...... 172

APPENDIX III...... 175

viii List of Tables

Table 2.1: Single-species biofilms growth time and media change schedule...... 44

Table 2.2: Multi-species biofilm inoculation schedules for trials 1-3 and final ...... 45

Table 3.1: Bacterial isolates recovered from the water system of the Alberta Children's Hospital...... 57

Table 3.2: Bacterial isolates recovered from water system of the Foothills Medical Centre (planktonic organisms) in 2004 by the IPC laboratory...... 59

Table 3.3: Bacterial isolates recovered from various water systems within southern Ontario and processed at the Ontario Ministry of the Environment (MoE) Lab...... 61

Table 3.4: Identification of 29 isolates using three 16S rRNA sequence databases: NCBI, RDP and Green Gene. Results shown indicate the most closely related organisms in each of the databases...... 63

Table 3.5: Isolates used in the multi-species drinking water biofilm model...... 72

Table 3.6: Theoretical predicted TRFLs for each isolates based on the in silico digests of sequences in Appendix III with the restriction enzymes listed below...... 73

Table 3.7: Actual versus estimated terminal restriction fragment lengths...... 75

ix List of Figures and Illustrations

Figure 1.1: Components within a traditional drinking water system...... 7

Figure 2.1: Alberta Children's Hospital water system coupons...... 38

Figure 3.1: Colony isolation from the University of Calgary Biological Sciences water system...... 62

Figure 3.2: Phylogenetic tree of sequenced environmental isolates...... 70

Figure 3.3: Combined Peak Scanner data of terminal restriction fragment lengths of 15 species...... 77

Figure 3.4: Peak Scanner data for the family Sphingomonadaceae...... 79

Figure 4.1: Mean viable cell count over time for Foothills Medical Centre isolates...... 84

Figure 4.2: Mean viable cell count over time for Alberta Children's Hospital isolates.... 86

Figure 4.3: Mean viable cell count over time for University of Calgary isolates...... 88

Figure 4.4: Mean viable cell count of each isolate in a single-species biofilm ...... 91

Figure 4.5: MBEC and MIC values for single-species biofilms...... 95

Figure 5.1: Terminal restriction fragment length polymorphism assay of three inoculation schedules ...... 101

Figure 5.2 Scanning electron microscopy of model and native drinking water system biofilms ...... 105

Figure 5.3: Single and multi-species minimum biofilm eradication concentration assays with Escherichia coli ...... 107

Figure 5.4: Confocal scanning laser microscopy of model multi-species biofilm with Escherichia coli ...... 109

Figure 5.5: Terminal restriction fragment length polymorphism assay of model multi- species biofilm with Escherichia coli ...... 112

Figure 5.6: Single and multi-species minimum biofilm eradication concentration assays with Enterobacter cloacae ...... 113

Figure 5.7: Confocal scanning laser microscopy of model multi-species biofilm with Enterobacter cloacae ...... 116

Figure 5.8: Terminal restriction fragment length polymorphism assay of model multi- species biofilm with Enterobacter cloacae ...... 117

x Figure 5.9: Single and multi-species minimum biofilm eradication concentration assays with Pseudomonas aeruginosa ...... 119

Figure 5.10: Confocal scanning laser microscopy of model multi-species biofilm with Pseudomonas aeruginosa ...... 121

Figure 5.11: Terminal restriction fragment length polymorphism assay of model multi-species biofilm with Pseudomonas aeruginosa ...... 122

Figure 5.12: Single and multi-species minimum biofilm eradication concentration assays with Stenotrophomonas maltophilia ...... 124

Figure 5.13: Confocal scanning laser microscopy of model multi-species biofilm with Stenotrophomonas maltophilia ...... 127

Figure 5.14: Terminal restriction fragment length polymorphism assay of model multi-species biofilm with Stenotrophomonas maltophilia ...... 128

Figure 5.15: Terminal restriction fragment length polymorphism assay of control biofilm...... 129

Figure 5.16: Terminal restriction fragment length polymorphism assay of model multi-species biofilm inoculated with L. pneumophila ...... 132

Figure 6.1: Comparison or relative species abundance in multi-species biofilms...... 153

xi List of Symbols, Abbreviations and Nomenclature

Symbol Definition ACH Alberta Children's Hospital AHS Alberta Health Services ATP adenosine triphosphate AWQI adverse water quality incident BCYE buffered charcoal yeast extract BLASTn basic local alignment search tool (nucleotide) BOC biodegradable organic compound BOM biodegradable organic matter BRG Biofilm Research Group BYE buffered yeast extract CBD Calgary biofilm device CFU colony forming units Cl chlorine CSLM confocal scanning laser microscopy CT contact time DBP disinfection by-product DC differential coliform DGGE denaturing gradient gel electrophoresis DNA deoxyribonucleic acid DOM dissolved organic matter EPS extracellular polymeric substance ER emergency room FISH fluorescence in situ hybridization FMC Foothills Medical Centre GAC granular activated carbon HAA haloacetic acids HPC heterotrophic plate count ICU intensive care unit ID identifier IPC Infection Prevention and Control LaSB Laboratory Services Branch LD50 lethal dose 50% MAC maximum allowable concentration MBEC minimum biofilm eradication concentration mg/L milligrams per litre MIC minimum inhibitory concentration MoE Ministry of the Environment NCBI National Center for Biotechnology Information Ng no growth nm nanometers OFL oxidation fermentation of lactose

OTU operational taxonomic unit xii PA presence absence PAGE polyacrylamide gel electrophoresis PBS phosphate buffered saline PCA plate count agar PCR polymerase chain reaction PE polyethylene PVC polyvinyl chloride QAC quaternary ammonia compounds R2A(B) Reasoner's 2 A media (broth) RDP Ribosomal Database Project RNA ribonucleic acid rRNA ribosomal RNA RO reverse osmosis SRB sulfate-reducing bacteria T-RFL terminal restriction fragment length T-RFLP terminal restriction fragment length polymorphism TDS total dissolved solids THM trihalomethane UV ultraviolet WTP water treatment plant X times µL microlitres °C degrees Celsius

xiii 1

Chapter One: Introduction

1.1 Introduction

In July 2010 the United Nations General Assembly recognized safe, clean drinking water and sanitation as a basic human right and acknowledged that access to reliably safe drinking water is integral in the realization of all human rights (UN General Assembly

2010). Although the number is decreasing there are still ~780 million people across the globe without access to improved drinking water sources (WHO and Unicef 2012). Even in Canada, among First Nations communities, 39% of water systems were classified as high overall risk; and of that group, 161 were under drinking water advisories by Health

Canada as of February 2011 (Department of Indian Affairs and Northern Development

2011). In a survey of Canadian water and wastewater systems by Environment Canada, municipalities representing 33% of the responding population (~6.7 million people) indicated that there had been water quality issues in their municipality between 2007 and

2009. These ranged from aesthetic issues such as taste or odour to full-blown contamination events (Environment Canada 2011). Two of the most notable water contamination events causing illness in Canada occurred in 2000 in Walkerton, Ontario and 2001 in North Battleford Saskatchewan. The drinking water system of Walkerton,

Ontario, serving approximately 4,800 residents, became contaminated with Escherichia coli O157:H7 and Campylobacter jejuni in May of 2000, causing 7 deaths and 2,300 individuals to become sick. The source of the organisms was traced back to agricultural manure runoff into source water. A lack of operator training and expertise, shortcomings in provincial regulations regarding results reporting and inspection programs and grossly

2 improper operating procedures were recognized as the cause of the outbreak (O'Connor

2002). The contamination event in North Battleford, Saskatchewan, a community of approximately 15,000, was caused by Cryptosporidium parvum and occurred in March and April of 2001. Over 6,000 individuals became ill and a boil water advisory was initiated which lasted three months. No one specific factor was identified as leading to the contamination, although better training of plant operators, more provincial legislation and source water protection were all suggestions made during an inquiry into the event

(Laing 2002). Events like those in Walkerton and North Battleford have led to increased national awareness to the necessity of safe drinking water management and treatment practices and increased research and education to assist operators, legislators, and the public.

Three specific contamination events have had a bearing on the creation and directions of the research presented here, two of which occurred within two different hospital environments in the city of Calgary. In 2003 a nosocomial case of Legionellosis contributed to the death of a patient at a Calgary hospital. Legionella pneumophila was isolated from bronchoalveolar lavage specimens from the patient as well as from water samples taken in the ICU and ER. Using sequence-based typing and amplified fragment length polymorphism methods, it was determined that the environmental isolates had the same sequence type profile as the patient isolate. L. pneumophila was also isolated from three additional hospital water systems in the city. To resolve the issue, water in the system was super heated to 71°C and flushed for 30 minutes, and a copper-silver ionizer was installed although L. pneumophila still persisted in some locations (Wong et al.

2006). The second event occurred between in 2003-2004 and was caused by

Pseudomonas aeruginosa. Originally, the outbreak began in the ICU, and persisted for three months, after which the same organisms were detected in the Oncology unit, where sporadic cases of infection were detected every 4-7 weeks over a 16 month period. A total of 27 deaths were linked to this particular strain of P. aeruginosa . The outbreak was attributed to colonization and establishment of biofilms in cold water lines. In the ICU, the outbreak was resolved by removing faucet aerators in automatic hand washing sinks, and in the Oncology unit the source of the biofilm was eventually traced back to the sink shut off valve in a patient room. This outbreak demonstrated that even low level water contamination from one or two isolated sources can result in widespread outbreak, especially in locations with high numbers of immunocompromised individuals (Laupland et al. 2005, Pauling-Shepard et al. 2006). Due to these events, when the construction of a new hospital facility in the Calgary area was undertaken, extraneous effort was made to reduce the risks of water system contamination as well as colonization with biofilms harbouring potentially pathogenic organisms. This included engineering the water system to have no dead ends or stagnation and installing UV disinfection units that would treat incoming water from the city distribution system. The new system also provided the opportunity to study water quality and biofilm formation in a highly engineered water system, where coupons were installed in the water system that could be removed and analyzed for biofilm colonization, structure and community identification.

The third water system contamination event, which led more directly to the present study, occurred in a large municipality in Ontario in 2008. The site had reported adverse water

4 quality incidents (AWQIs) on recurring occasions, triggered by low levels of coliforms found at a number of sampling sites throughout the distribution system. The positive bacteriological tests prompted a boil water advisory to be put in place for the municipality by the Medical Officer of Health. It was hypothesized that the recurring results were due to the growth of biofilms containing the coliform bacteria at the sampling sites. Isolated bacteria were speciated using biochemical tests, and identified to be the same species of Enterobacter as detected from new sites which at the time had not been connected to the existing water mains. The situation prompted a joint project between the Ontario Ministry of the Environment and the University of Calgary Biofilm

Research Group to examine biofilm formation capabilities of drinking water system isolates and their sensitivity to disinfectants as well as to identify a disinfection protocol that may mitigate the risks associated with biofilm growth in water systems. The recurring contamination events were eventually resolved within the municipality after training and a review of sampling protocol (including bleach pre-treatment of taps) for staff as well as replacing faucet aerators (Schop 2012).

These contamination events are just three examples of why understanding the community, structure and survival of biofilms composed of bacteria from treated water systems is important. If we are able to prevent the formation of or respond faster to contamination of water systems with biofilms which harbour pathogenic organisms it would help ensure the safety of consumers, especially those with a weakened immune status in hospital environments, and to prevent the implementation of long term boil water orders and the associated treatment involved.

1.1.1 Hypothesis

The hypothesis that was formed at the beginning of this study was that organisms isolated from treated drinking water systems would be less susceptible to disinfection by chlorine at residual levels; additionally, mixed species biofilms of flora indigenous to drinking water systems may provide increased protection to clinically relevant microorganisms when incorporated into the mixed species biofilms.

1.1.2 Project objective and aims

The overall objective of the project was to examine growth of single and multi-species biofilms of environmental organisms, both with and without clinical importance, isolated from drinking water systems. Specifically the effects of chlorination at residual levels, as present in drinking water distribution systems, including survival, biofilm structure and community dynamics were to be evaluated. In order to achieve this goal four aims were established:

1. Isolate and identify indigenous microorganisms from treated drinking water

systems, including both non-pathogenic and opportunistically pathogenic

organisms (Chapter 3).

2. Create a terminal-restriction fragment length polymorphism library of isolated

organisms for use in studying model water distribution system multi-species

bacterial communities (Chapter 3).

3. Evaluate the growth of isolates under biofilm formation conditions and measure

their susceptibility to chlorination as single-species planktonic and biofilm

populations (Chapter 4)

4. Establish a model multi-species water distribution system biofilm of collected

isolates in order to examine the chlorine susceptibility, structure and community

dynamics of the biofilm (Chapter 5).

1.2 Drinking water systems: treatment and monitoring

A drinking water system is comprised of three main components: the source water, the

primary treatment plant and the distribution system. Although the research presented here

pertains mainly to the conditions within the last of these three components, it is important

to have an understanding of all three in order to understand their impact on the organisms

within the distribution system environment, as well as the significance of the results.

Figure 1.1 shows an overview of the entire treatment process, as described in greater

detail in the following sections.

1.2.1 Source water and the multi-barrier approach to water safety

In Canada, close to 90% of the population has their water provided through distribution systems (as opposed to private wells) and of that group, 94% receive water which has undergone treatment (Environment Canada 2011). The provincial governments are responsible for managing natural resources which includes water sources and for providing drinking water services to consumers, although the federal government does mandate national standards that must be met or exceeded in provincial regulations.

Drinking water treatment methods in Canada are not standardized across the entire country as they depend on the characteristics of the source water and can vary greatly between municipalities. A more recent approach to providing quality water more

Figure 1.1: Components within a traditional drinking water system The traditional drinking water system is comprised of three distinct sections: the source water, primary treatment facility and the distribution system. Source water may be derived from surface water such as lakes or rivers, or from ground water aquifers.

Primary treatment of the source water depends on its characteristics but can include a number of steps including filtration and disinfection. Within the distribution system, secondary disinfection occurs by the maintenance of residual chlorine in the lines.

Biofilm formation does occur within the distribution system lines, despite secondary disinfection.

8 efficiently to consumers is by using a multi-barrier approach, rather than the traditional method of compliance monitoring using endpoint tests. Disadvantages to the traditional end point tests include the lag time between sampling and tests results during which time consumers are potentially exposed to contaminants, as well as the fact that tests can only be performed for a limited number of contaminants, as methods may not be available for all. The multi-barrier approach uses a number of different processes to ensure the safety of the drinking water, which could be physical barriers such as source water management and primary water treatment, or procedural barriers such as operator training and legislation. Compliance monitoring is used as only one step in the multi-barrier approach

(CCME Water Quality Task Group and Federal-Provincial-Territorial Committee on

Drinking Water 2004).

The most important factor in determining primary treatment processes necessary is the quality and characteristics of the source water for the water system (CCME Water

Quality Task Group and Federal-Provincial-Territorial Committee on Drinking Water

2004). Source water is grouped into two main categories, ground water, from underground aquifers for example, and surface water from rivers and lakes. The majority of Canadians, ~90%, receive water from surface water sources (Environment Canada

2011) and the city of Calgary alone draws 177,00 megalitres of river water annually (City of Calgary Water Services 2012). These surface water sources may come under the influence of runoff from agriculture, such as pesticides or fertilizers, from industry, or be effected by nearby livestock or wildlife. Source water protection aims to limit the effect of these and other factors on water quality (Davies and Mazumder 2003). By providing

9 better quality water into the primary treatment system there are a number of advantages, including lower acute and chronic health risks; more efficient, less expensive treatment; and less disinfectant needed, which would lead to less disinfection by-products (DBP) present in the final water (Davies and Mazumder 2003).

1.2.2 Primary treatment

Once entering the primary treatment plant, water may undergo a number of treatment steps depending on the quality of the source water. Pristine water may require only a disinfection step (Davies and Mazumder 2003), although conventional treatment includes screening, chemical coagulation, flocculation, sedimentation, filtration and disinfection

(Alberta Environment Drinking Water Branch 2006). Filtration can be performed using a number of different methods including diatomaceous earth filtration, membrane filtration

(such as reverse osmosis or nanofiltration) or slow sand filtration which utilizes a biologically active top layer to biodegrade organic molecules by microorganisms (CCME

Water Quality Task Group and Federal-Provincial-Territorial Committee on Drinking

Water 2004) . Disinfection of the produced water can be performed using one or a combination of methods each with different advantages or drawbacks. The amount and type of disinfectant, like the entire treatment process used, is determined based on the quality and composition of the source water. The most commonly used disinfection method is chlorination, applied as solid or liquid sodium hypochlorite. This method provides free chlorine residual that is necessary for secondary disinfection as discussed below. Chlorine functions as a disinfectant by its reaction with organic molecules in the cell membrane and also within the cell. This causes a number of detrimental effects,

10 including increasing the permeability of the cell membrane and interfering with enzyme activity, ultimately leading to cell death (Venkobachar, Iyengar and Rao 1977). In water treatment, hypochlorous acid is the main reactive form of chlorine present, and can undergo three types of reactions with organic compounds: oxidation reactions, addition reactions with unsaturated bonds or electrophilic substitution reactions at nucleophilic sites (Deborde and von Gunten 2008). At the concentrations present in drinking water, free chlorine has a relatively low toxicity, the LD50 of a 5% sodium hypochlorite solution (~10 4 times greater than would be present in drinking water) by oral dose in rats

is 13 g/kg body weight (Horizon Chemical Co 2004)

Monochloramine is another compound used for disinfection and is less reactive than free

chlorine which allows for the creation of less DBPs (disinfection by-products) and a

longer residual, but it also requires a longer contact time. Chlorine dioxide disinfectant is

more reactive than both chlorine and monochloramine, it is therefore very effective but

has little residual disinfection. Ozone is a powerful disinfectant but may increase the

biodegradable organics in water as well as create dangerous DBPs (CCME Water Quality

Task Group and Federal-Provincial-Territorial Committee on Drinking Water 2004).

1.2.2.1 UV disinfection

UV disinfection is becoming more common in treatment facilities, especially as it is

relatively easier and less expensive to install and operate, although the lamps can become

fouled and they do not provide a residual, so low levels of disinfectant must still be added

to the finished water (Alberta Environment Drinking Water Branch 2006, Dykstra et al.

2007). Recent research has also shown that UV treatment in combination with chlorination, added as sodium hypochlorite solution, caused an increase chlorine demand and the formation potential of the DBP trihalomethane (THM), which was believed to be due to modification of dissolved organic matter (DOM) structure (Choi and Choi 2010).

1.2.2.2 Disinfection by-products

Several hundred types of disinfection by-products have been reported in the literature, with THMs being the largest contributor, followed by haloacetic acids (HAAs) (Krasner et al. 2006). Exposure to DBPs is 50% higher in hot water (Weisel and Chen 1994) and can occur not only through ingestion, but also through inhalation and dermally, for example while showering (Miles et al. 2002). THMs are formed when chlorine reacts with organic material, and most commonly include the following four compounds: chloroform (CHC l3 ), bromodichloromethane (CHCl 2Br), dibromochloromethane

(CHClBr 2), and bromoform (CHBr 3) (Miles et al. 2002). The rate of THM production

largely depends on the original concentration of chlorine present, but is also affected by

temperature and pH of the water as well as the types of organic molecules reacting with

the chlorine to produce THM (Adin et al. 1991). Health Canada regulates the presence

the these THMs to a maximum acceptable concentration (MAC) of 0.100 mg/L. (Federal-

Provincial-Territorial Committee on Drinking Water 2006). THMs at levels higher than

the guideline values are considered possibly carcinogenic, with links to bladder, liver,

kidney, intestinal and colorectal cancers. Exposure to high levels of THMs have also

been linked to possible reproductive problems, such as increased risk of stillbirth or

12 spontaneous abortion (Federal-Provincial-Territorial Committee on Drinking Water 2006,

Karanfil et al. 2008, Krasner 2009).

1.2.3 Distribution systems and secondary disinfection

The final third of the water system is comprised of the distribution system, maintaining the integrity of treated water as it is transported from the treatment plant to the consumer.

The maintenance of a residual disinfectant (also known as secondary disinfection) is required throughout the distribution system. This is designed to prevent re-growth of bacteria, eliminate any organisms that may enter a system and to retard the growth of biofilms within a system (CCME Water Quality Task Group and Federal-Provincial-

Territorial Committee on Drinking Water 2004, Ontario Ministry of the Environment

2006). In Canada, regulations for residual chlorine levels are dependent on provincial legislation. In Alberta for example, a minimum of 0.1mg/L of free or combined chlorine must be maintained within a system at all times. Maximum levels allowable of free chlorine are 4.0 mg/L, and 3.0 mg/L of combined chlorine (Alberta Environment

Drinking Water Branch 2006). As continual monitoring for very small numbers of pathogenic microorganisms can be difficult, a CT value of the residual disinfectant is also used to show adequate treatment of the system where C is the concentration of the disinfectant in mg/L and T is the effective contact time in minutes (Alberta Environment

Drinking Water Branch 2006).

1.2.4 Water quality monitoring

Routine monitoring of drinking water within the distribution system for various contaminants is very important in maintaining water quality. Testing procedures and limits for contaminants are highly regulated. The Guidelines for Canadian Drinking

Water Quality (Federal-Provincial-Territorial Committee on Drinking Water 2010) outline limits for a number of different chemical, physical, radiological and microbiological contaminants. Chemical or physical contaminants include the disinfection by-product THMs as previously discussed, but also close to 90 other contaminants, including physical parameters such as total dissolved solids (TDS; ≤500 mg/L). On-line monitoring technologies have made significant advances in recent years in the area of real-time monitoring of basic water parameters. Instrumentation for monitoring of parameters such as pH, chlorine, total organic carbon (TOC) and conductivity are commercially available (Storey, van der Gaag and Burns 2011).

To insure the microbiological safety of a system, the Guidelines for Canadian Drinking

Water Quality provide limits for two groups of indicator organisms, Escherichia coli and

Total Coliforms, for which the maximum acceptable concentration (MAC) for both is

none detectable per 100 mL of water (Federal-Provincial-Territorial Committee on

Drinking Water 2010). Heterotrophic plate count bacteria are not given a specified MAC,

although increased levels above baseline are considered undesirable and indicate a

deterioration of water quality. Although protozoa such as Giardia and Cryptosporidium as well as enteric viruses are recognized as contaminants responsible for severe illness, reliable testing methods suitable for routine analysis are not available for these organisms

14 and so no MAC has yet been set (Federal-Provincial-Territorial Committee on Drinking

Water 2010). Instead, E. coli and Total Coliforms are to be used to indicate contamination events. As one of the aims for the project was to collect and identify a variety of organisms from drinking water systems, Section 1.5 will examine microorganisms present within treated water systems that are not routinely monitored, including opportunistic pathogens and other heterotrophic organisms that colonize water distribution systems.

At present, microbiological monitoring only includes testing of planktonic organisms, and does not account for biofilms present within the system, within which it is estimated that up to 95% of the bacterial biomass of the system may be found (Flemming 2002).

The following sections will discuss the growth characteristics and increased resistance to treatment of bacterial biofilms both in general as well as within drinking water distribution systems

1.3 Bacterial biofilms

Biofilms are communities of bacteria that grow encased in a layer of EPS (extracellular polymeric substance) composed of polysaccharides, DNA and proteins. They most commonly form attached to a surface, although they also occur as aggregates and floccules in environments such as water treatment facilities (Costerton et al. 1995,

Harrison et al. 2005). It has been widely accepted that biofilms can be found ubiquitously throughout nature, and in aquatic environments can comprise as much as 99.9% of the bacterial community (Costerton et al. 1995). Biofilms can be found as single-species

15 populations but more often exist in the natural environment as multi-species communities. Mixed populations and oxygen and nutrient gradients within the biofilm lead to the formation of heterogeneous microenvironments (Harrison et al. 2005). Growth within a biofilm offers cells numerous advantages over solitary planktonic life including increased protection from antimicrobials, environmental toxins and host immune response; opportunity for interactions with neighboring cells including intercellular communication such as quorum sensing, sharing of metabolic by-products and horizontal gene-transfer (Harrison et al. 2005). In in vitro tests, biofilm formation by various E. coli strains was shown to have significant dependence on environmental conditions and nutrient media composition (Reisner et al. 2006, Pratt and Kolter 1998).

1.3.1 Biofilm formation

Biofilm formation follows a generalized cycle of five stages, any of which may be present within the same system at one time. Before formation of a biofilm, cells exist planktonically within the environment. The first stage in the biofilm life cycle is the adsorption and reversible attachment of planktonic cells to a surface, such as a pipe wall in a distribution system (i). Cells become irreversibly attached and begin to produce EPS

(ii) and begin the maturation-1 phase (iii) wherein cells begin to multiply and communicate via cell to cell signaling. The fourth phase, maturation-2 (iv), involves the formation of a more developed 3D biofilm structure which may include channels within the structure and different microenvironments. The final stage of biofilm development (v) is the shedding of planktonic cells or pieces of biofilm and their dispersion back into the bulk media phase. These cells may then in turn begin the cycle again by attaching to

16 another point within the system and colonizing a new biofilm (Sauer et al. 2002, Harrison et al. 2005). In environments with hydrodynamic conditions, such as a water distribution system, the amount of sheer force the biofilm is under can affect physical characteristics of the biofilm such as thickness and density. In general as the amount of sheer force increases so does the cell density within the biofilm while the thickness decreases. It can also affect the sloughing of cells which is more likely to happen in thicker biofilms, although under high sheer force, biofilms have been shown to form long streamers or tails that may break off from the main biofilm body (Liu and Tay 2002, Stoodley et al.

2002a). Mutualism and synergistic interactions between species in dual- and multi- species biofilms have also been identified as enhancing biofilm growth. One study identified an interspecies signal between two species ( Actinomyces sp. and Streptococcus sp. ) that when present in the correct concentration, increased the biofilm biomass of both species by 10 fold (Rickard et al. 2006) . Another study demonstrated that an adhesion deficient E. coli strain, when incubated with bacterial communities isolated from drinking water systems, was able to grow as a multi-species biofilm on glass slides (Castonguay et al. 2006).

1.3.2 Phenotypic variation between biofilm and planktonic cells

Phenotypic differences are often observed between biofilm and planktonic cells of the same species as each growth form has distinct physiological characteristics, both structural and metabolic. One study involving Pseudomonas aeruginosa biofilms found that the maturation phase had the greatest difference in protein profile from the planktonic phase, with over half the proteome (>800 proteins) having a 6-fold or greater

17 difference in protein expression, either induced or repressed (Sauer et al. 2002). A study comparing the transcriptome for biofilm and planktonic growth of a laboratory strain of

E. coli showed that 5.5% of the genome (223 genes) was differently expressed between the two growth phases. Genes that were induced in the biofilm included those involved in stress response, type I fimbriae and metabolism (Ren et al. 2004). Another transcriptome analysis, in this case in a study of sulfate reducing bacteria (SRB), indicated that 472 genes were either induced or repressed between the biofilm and planktonic growth phases

(Zhang et al. 2007). In addition to differences in the transcriptome and the proteome between the two forms of growth it follows that downstream the metabolome of the cultures would also change, which has been shown to be true in the case of Pseudomonas fluorescens exposed to copper ions (Booth et al. 2011). In multi-species biofilms there are also integrated metabolic activities between species and the communities have been compared to a multi-cellular organism (Stoodley et al. 2002b).

1.3.3 Bacterial advantages of biofilm growth

As mentioned above, growth as a biofilm can confer a number of advantages to microorganisms. These include protection from or increased resistance to a number of anti-microbial agents at concentrations anywhere from 10 to 1000 fold higher (Ceri,

Stremick and Nalepa 2000, Mah and O'Toole 2001, Hoiby 2010). Biofilms have also been shown to have increased resistance and/or tolerance to many different antibiotics

(Hoyle and Costerton 1991, Stewart 2002), environmental toxins such as metals

(Harrison, Ceri and Turner 2007) and disinfectants like quaternary ammonia compounds

(QACs) (Campanac et al. 2002) and chlorine (LeChevallier, Cawthon and Lee 1988a).

Biofilm resistance to disinfection methods involved in water treatment are discussed in more detail in Section 1.4.2. Mechanisms of biofilm resistance have been well reviewed in the literature. Mah and O'Toole (2001) identified four biofilm characteristics that contributed to antimicrobial resistance including physiochemical diffusion barrier created by the EPS, the reduced growth rate characteristic of biofilm bacteria due to oxygen and nutrient limitation, activation of general stress response genes, and the biofilm phenotype induced by heterogeneity within the biofilm. More recent reviews identified the same resistance mechanisms, but also discussed newer research into resistance conferred by mutations and horizontal gene transfer, which both occur at higher rates within some biofilm environments, as well as via the production of persister cells within the biofilm

(Bridier et al. 2011), which are a phenotypic adaptation of a small proportion of the biofilm population (~1%) to a dormant, non-dividing state (Lewis 2007). Protection of cells within multi-species biofilms is also presented, possibly by synergistic effects between species, changes in biofilm structure or possible enzyme interactions (Bridier et al. 2011). As reported throughout the literature, it is clear that resistance mechanisms to antimicrobials greatly depends on the organism present and type of antimicrobial, and that there is no universal factor that indicates increased resistance.

1.4 Biofilm growth in water systems

It has been well established that despite residual disinfection and low nutrient content in water distribution systems, bacterial biofilms do make up the majority of microbial biomass in a distribution system (Flemming 2002). The formation, structure, resistance and microflora of these biofilms have been studied in depth in the past 25 years. Biofilms

19 begin to establish on new distribution system materials after only a few weeks of use, although growth rate is dependent on water system conditions (Obst and Schwartz 2007).

After approximately 6 months to 2 years the biofilm growth and community within water lines reach a steady state, where detachment is approximately equal to net growth rate, independent of surface material (Obst and Schwartz 2007, Wingender and Flemming

2004, Boe-Hansen et al. 2002, Lehtola et al. 2004). Biofilms within water systems are often thin and non-uniform across pipe walls, with mature biofilms covering ~76% of the pipe surface. A high degree of biofilm heterogeneity with microcolony formation has also been described in the literature (Wingender and Flemming 2004, Boe-Hansen et al. 2003,

Percival et al. 1998, Martiny et al. 2003). As has been observed with other types of environmental biofilms, in terms of cultivability and ATP content, the bulk water planktonic organisms showed higher levels than the biofilm cells (Boe-Hansen et al.

2002).

1.4.1 The distribution system environment

The environment within the distribution system, such as the biodegradable organic carbon (BOC) content of the water, the flow rate in the system and the pipe materials used have varying effects on the biofilm. The availability and source of carbon are limiting factors for bacterial growth in water systems (Hu et al. 2005). It has been shown that increasing amounts of BOC increases the heterotrophic plate count (HPC) bacterial counts within a biofilm and that higher phosphorus levels increases the physiological function and biomass in the biofilm, although phosphorus has no effect on HPC bacteria counts (Park and Hu 2009). Higher BOC values are usually associated with conventional

20 water treatment whereas lower BOC levels are associated with biologically treated water, which passes through a biologically active filter to assist in the degradation and removal of organic matter (Norton and LeChevallier 2000). An increase in flow velocity through the distribution system increased the formation of biofilms. Changes in velocity resulted in dispersion of bacteria from the biofilm and an increase in turbidity (Lehtola et al. 2006,

Manuel, Nunes and Melo 2007). A variety of pipe materials, including polyethylene (PE), polyvinyl chloride (PVC), iron, cement, copper and stainless steel, have been evaluated for biofilm growth in operational drinking water systems, model water system pilot scale bioreactors and on coupons. Differing biofilm initial growth rates depending on pipe material was noted by some researchers over the initial biofilm colonization period of one week to four months. There was a consensus that iron has the highest initial biofilm growth rate of the materials tested and that biofilm development occurs on PVC, PE, stainless steel or copper, although at a slightly lower rate. Of these latter materials, there was not a consistent hierarchy between the studies of growth rate. After long term exposure (>18 months) to the drinking water system, biofilms were found to reach a plateau for all material types. This suggests that initially, pipe materials do have an effect on biofilm growth, but on a long term basis, biofilm growth is more site-specific than material-specific (Niquette, Servais and Savoir 2000, Lehtola et al. 2004, Wingender and

Flemming 2004, Buswell, Nicholl and Walker 2001, Camper et al. 2003, Norton and

LeChevallier 2000). After the biofilm had reached a mature state (approximately 6 months -2 years) there was a plateau in growth and no difference between the materials tested (Lehtola et al. 2004, Wingender and Flemming 2004). Within distribution systems there can be the presence of materials, such as rubber coated valves, that are able to

21 support larger, more structured biofilms. These communities may act as a point source for contamination and coliform organisms (Flemming 2002, Schmeisser et al. 2003, Kilb

2003).

1.4.2 Disinfectant effects on planktonic and biofilm bacteria

A variety of treatments and disinfectants are applied to water systems to prevent,

eliminate or control the growth of bacteria in both the bulk water and biofilm phase of

distribution systems, as discussed in Section 1.2. The effect of filter and clarification

methods on biofilm growth was evaluated and it was found that granulated activated

carbon (GAC) filtration of water was more effective than conventional coagulation and

flocculation methods at increasing the biological stability of the water and slowed biofilm

growth. However, in one study the GAC biofilm reached a similar state to the

conventional treatment biofilm after a few weeks of growth in a biofilm reactor (Sharp et

al. 2001, Norton and LeChevallier 2000). It was also found that nanofiltration was more

effective than GAC at removing protozoa from the bulk water and biofilm and it also

decreased the biofilm density within water systems (Sibille et al. 1998). The main

difference between the filtration methods was the removal of BOM from the water

system with biologically active GAC, and size exclusion of protozoa using nanofiltration.

After source water has undergone primary conventional or filtration treatment, a

disinfection step is applied. The three most commonly studied disinfection methods are

via chlorine, monochloramine and UV disinfection. One study looked at different

characteristics of water systems and bacteria (age, attachment, nutrients, surface) that

22 might change the disinfection efficiency of chlorine and monochloramine and found that attachment to a surface as a biofilm was the most important factor for chlorine and the only factor for monochloramine that increased disinfection resistance (LeChevallier et al.

1988a). The same group was able to show that, although for both disinfectants the biofilm was more resistant than the planktonic cells, it was less resistant to disinfection with monochloramine due to its increased penetration into the biofilm (LeChevallier, Cawthon and Lee 1988b) whereas, the more reactive chlorine based treatments are not able to diffuse into the biofilm (Parent et al. 1996). Monochloramine also maintains a better residual throughout the system although it does increase the risk of nitrification of the water system by nitrifying bacteria, effectively decreasing the amount of disinfectant residual in the system, and so additional monitoring needs to be performed (Norton and

LeChevallier 1997). In contradiction to the above, Camper et al. (2003) found that for bioreactor biofilms, residual disinfectant (chlorine or monochloramine) had no effect on biofilm re-growth, and the lowest free chlorine levels actually had the least re-growth.

Reports of the efficacy of added UV disinfection are somewhat mixed. Dykstra et al.

(2007) concluded that UV pretreatment significantly reduced planktonic HPC bacteria and at low residual disinfectant concentrations it increased the susceptibility of both planktonic and biofilm bacteria, although another study found no difference in biofilm biomass or pathogen incidence with or without UV pretreatment (Långmark et al. 2007).

Some negative attributes that have been proposed for UV treatment are that although it showed good elimination of planktonic organisms, for the biofilm phase in some cases it caused longer E. coli persistence (Murphy, Payne and Gagnon 2008) as well as decreased culturability and therefore decreased detection of L. pneumophila (Langmark et al. 2005).

Additionally, in one study UV treatment caused a change in the DOM structure which increased the chlorine demand of the system (Choi and Choi 2010).

1.5 Drinking water microflora

1.5.1 Drinking water distribution system native flora

A number of studies have been performed to identify the microflora present within drinking water systems using different techniques such as classical culture dependent methods, 16S rRNA gene fingerprinting, RNA and DNA probes (including fluorescent in situ hybridization, FISH) and 454 pyrosequencing. One research group (Eichler et al.

2006) performed community fingerprints of two raw water sources and the combined finished treated water. The raw water sources were shown to have very different community profiles, and both had significant influence on the community composition of the finished water although chlorination strongly affected the community structure.

Keinänen et al. (2004) reported similar findings of variation in distribution system biofilm communities between two different systems each with a unique raw water source.

These studies demonstrate the importance of the microbial community of a raw water source to the community present when the finished water is delivered to the end user.

Even within the same distribution system there can be changes in the microbial community. For example, Lautenschlager et al. (2010) used denaturing gradient gel electrophoresis (DGGE) to demonstrate a change in community profile after overnight stagnation of the water within a small portion of a water system. Additionally, a biofilm may have a slightly different community from the bulk water phase in a distribution system (Santo Domingo et al. 2003) and even within the biofilm there are temporal

24 differences in the community, as biofilm communities mature, different species are present in different proportions (Martiny et al. 2003).

A number of researchers have reported on the identification of organisms present within various drinking water systems (Eichler et al. 2006, Williams et al. 2004, Tokajian et al.

2005, Schmeisser et al. 2003, Simoes, Simoes and Vieira 2007, Martiny et al. 2003,

Kalmbach, Manz and Szewzyk 1997, Kalmbach et al. 1999, Hong et al. 2010,

Bereschenko et al. 2010). Organisms routinely identified include: Alpha-, Beta- and

Gammaproteobacteria, specifically the Alphaproteobacteria Sphingomonadaceae,

Methylobacterium and Bradyrhizobiaceae and Betaproteobacteria Acidovorax and

Burkholderia ; Actinobacteria including Mycobacterium spp; Nitrospira spp;

Bacteroidetes and Planctomycetes. Some studies reported that ~70-80% of the community of microorganisms present within a water system could be represented by only 6-8 operational taxonomic units (OTUs) (Kalmbach et al. 1997, Simoes et al. 2007).

The non-indicator organisms present in drinking water systems are often referred to as heterotrophic plate count bacteria (HPC bacteria), although using the standard method for isolation of HPC bacteria from water systems some of the organisms mentioned above likely would not be cultured due to selective effects of media type, incubation length and incubation temperature (Allen 2004).

1.5.2 Pathogenic organisms in drinking water

A number of opportunistic pathogens can be isolated from contaminated drinking water systems, including E. coli, L. pneumophila, P. aeruginosa , Aeromonas spp.,

Mycobacterium avium, Klebsiella pneumoniae, Burkholderia spp., Campylobacter spp.,

Salmonella spp., Vibrio cholerae and Helicobacter pylori (Simoes, Simoes and Vieira

2010, Pavlov 2004) . Within water systems it has been shown that a number of pathogenic

strains are able to survive within biofilms, for anywhere from a few days to several weeks

in biofilm reactor studies (4 days for E. coli, 31 days for H. pylori, weeks for L. pneumophila and P. aeruginosa ) even under high sheer stress conditions (Lehtola et al.

2007, Giao et al. 2008, Moritz et al. 2010). Adhesion deficient strains are also able to

form more proficient biofilms in the presence of heterotrophic organisms (Castonguay et

al. 2006, Giao et al. 2008, Buswell et al. 2001). Detection and monitoring methods for

pathogens can have very different sensitivity. Traditional methods rely on culture-based

techniques. When non culture-based methods were used, such as FISH or 16S rRNA

gene amplification, detection of pathogens was significantly higher than by culture based

methods (Lehtola et al. 2007, Moritz et al. 2010, Williams and Braun-Howland 2003,

Obst and Schwartz 2007). For example, one study used FISH and PCR in addition to

culture based methods to identify E. coli in pipe samples from actual water systems and

coupons from model water systems. This involved the hybridization of cells with

fluorescently labeled species specific oligonucleotide probes and visualization of the

samples using epifluorescence microscopy. The organism was identified in all six pipe

samples and 56% of the coupons using culture-independent methods, whereas with the

culture method, E. coli was only isolated from one pipe sample and none of the coupons

(Williams and Braun-Howland 2003). LeChevallier et al. (1987) suggested that growth of

E. coli , or other coliform organisms, on the surface of pipe walls could be responsible for

repeat occurrences of positive water monitoring tests due to the resistance of the biofilms

26 to chlorination. For the research presented here, three of the above mentioned opportunistic pathogens were isolated for study, E. coli, L. pneumophila and P. aeruginosa , as well as two other opportunistic pathogens, Enterobacter cloacae and

Stenotrophomonas maltophilia.

In this study, the term opportunistic pathogen is used when referring to E. coli, L. pneumophila, P. aeruginosa , E. cloacae or S. maltophilia. Other organisms being evaluated are referred to as normal water system microflora, or normal flora.

1.5.2.1 Indicator organisms: Escherichia coli and Total Coliforms

Of the five Gamma Proteobacteria opportunistic pathogens studied, only the two coliforms, E. coli and E. cloacae, would be detected during routine monitoring of water systems. These organisms, in addition to being potentially pathogenic, play an important role in water safety by acting as indicators of contamination or failure in treatment within a drinking water distribution system. The species E. coli contains numerous pathogenic, opportunistically pathogenic and non-pathogenic strains. Likely the most well known pathogenic strain to the general population is E. coli O157:H7, the enterohemorrhagic strain that caused seven deaths and illness in close to half the population of Walkerton

Ontario. This strain causes not only bloody diarrhoea, but can also lead to hemolytic uremic syndrome and possible long term affects such as post-infectious irritable bowel syndrome (Marshall 2009) . E. coli O157:H7 is asymptomatic in the gut of cattle and can be a member of the normal microbiota of the animal (Money et al. 2010). This gives its presence in water a twofold significance, in that it can indicate the contamination of a

27 water system with agricultural manure runoff and more importantly the probability of illness for consumers exposed to the pathogen. Due to the huge variability between members of the E. coli species, there are also non-pathogenic strains that don't pose the same risk to the population. Whether or not the E. coli strain present in a water distribution system is found to have the pathogenicity of O157:H7 or is benign, its presence will still cause the occurrence of an adverse water quality incident (AWQI), as is its purpose as an indicator organism.

Total coliforms are defined in the Standard Methods for the Examination of Water and

Wastewater as facultatively anaerobic, Gram-negative, non-spore-forming, rod-shaped bacteria that ferment lactose to produce gas and acid within 48 hours at 35°C (APHA

1998). Similar to E. coli, they act as indicators of contamination within the drinking water distribution system and can be either frank or opportunistic pathogens. In this study this group is represented by Enterobacter cloacae, which is often isolated from soil and untreated water. This organism is a common commensal organism but can also be opportunistically pathogenic. Enterobacter spp. were identified in one study as one of the top nine pathogens isolated from patients with nosocomial pneumonia, accounting for just over 8% of cases (Gastmeier et al. 2009). Of the Enterobacter species, E. cloacae has the highest pathogenic potential and causes 60-75% of all Enterobacter infections

(Hennigs et al. 2011). It is an important opportunistic pathogen among immunocompromised populations and has been recognized as an emerging pathogen of neonatal patients with 26 outbreaks described by one survey of the literature (Dalben et al. 2008).

1.5.2.2 Pseudomonas aeruginosa

Pseudomonas aeruginosa is a ubiquitous organism, found throughout natural and man- made environments such as soil and water, but also on many food products. For healthy individuals P. aeruginosa rarely causes infection, although immunocompromised hosts,

such as those with cystic fibrosis, neutropenia, severe burns, or foreign device installation

are at risk for serious infection. It is a very effective opportunistic pathogen and reported

as one of the most prevalent sources of nosocomial infection (Spencer 1996). P.

aeruginosa is often enumerated as part of the heterotrophic plate count in routine

monitoring and there is discussion in the literature regarding whether or not there should

be limits set on the maximum allowable concentration of HPC organisms and P.

aeruginosa in water systems (Allen 2004). The view expressed by a number of authors is

that the basis for classifying P. aeruginosa as an opportunistic water pathogen lies in its

importance in nosocomial infection, and not based on ingestion of contaminated water. It

has been expressed that inactivation of P. aerguginosa in water distribution systems

should not be a priority and for two main reasons, first that exposure to the organisms

would still regularly occur through other routes and secondly that the amount of

disinfectant required to eliminate the organism would be so high as to produce DBPs that

in themselves would be more hazardous to the general population than the organism itself

(Hardalo and Edberg 1997, Allen 2004). In a hospital environment this is not the case as

a larger number of the population are risk for infection as demonstrated by the outbreak

described by Pauling-Shepard et al. (2006) and so the standards on monitoring P.

29 aeruginosa in sensitive water systems may need to be different than for the general population.

1.5.2.3 Stenotrophomonas maltophilia

Stenotrophomonas maltophilia was originally described in 1943 and has undergone a number of taxonomic changes since that time, most recently being reclassified from

Xanthomonas maltophilia to the genus Stenotrophomonas. It is found ubiquitously in the

environment and can be isolated from water, but is most often found in soils, specifically

in association with the rhizosphere of plants (Bollet, Davinregli and Demicco 1995). It is

also an opportunistic pathogen of immunocompromised individuals. S. maltophilia has a high level of intrinsic antibiotic resistance to imipenem and other beta-lactam antibiotics, and has increasingly acquired antibiotic resistance (Bollet et al. 1995, Nyc and Matejkova

2010). It is a serious opportunistic pathogen in many immunocompromised individuals, such as those with HIV or post organ transplantation and it has specifically been identified as one of the characteristic members of the lung microbiota of adult patients with cystic fibrosis lung disease (Talmaciu et al. 2000).

1.5.2.4 Legionella pneumophila

Legionella pneumophila is a well-known opportunistic pathogen. It is the most common water pathogen in US water systems in recent years and since the early 1990s the incidence of outbreaks has been rising (Lau and Ashbolt 2009). The bacterium causes

Legionellosis, also known as Legionnaires' disease, a severe pneumonia, or the milder form of illness, Pontiac fever. It is contracted via inhalation of contaminated aerosols,

30 such as might be created by a shower or an air conditioner (Breiman et al. 1990). L. pneumophila is unique from the other opportunistic pathogens discussed above in its ability to persist and replicate within protozoan predators in water. It has been detected in at least 21 species of Amoeba (Lau and Ashbolt 2009). This growth mode confers a number of advantages to the organism including conferring biocide (chlorine) resistance, increased virulence and a means of dissemination of the bacteria. L. pneumophila are also able to replicate intracellularly in human phagocytic cells (Lau and Ashbolt 2009).

1.5.3 Eukaryotic distribution system organisms

Although they are not the focus of this study, a number of eukaryotic microorganisms are also present within the water distribution system community and can have a significant impact on the growth and survival of other organisms within the biofilm. As discussed above, a number of species of amoeba act as reservoirs for L. pneumophila growth (Lau and Ashbolt 2009). Eukaryotic organisms are also important waterborne pathogens, such as Giardia and Cryptosporidium , as was the case during the outbreak in the North

Battleford water system discussed in Section 1.1. The Guidelines for Canadian Drinking

Water Quality (Federal-Provincial-Territorial Committee on Drinking Water 2010) stipulate that source water treatment should produce a 3-log reduction or inactivation of cysts and oocysts. LeChevallier et al. (1991) showed that in water tested from 66 different treatment plants, cysts or oocysts were present in 39% of the samples. New filtration technologies such as reverse osmosis, nano-, ultra- and microfiltration have more effective removal of cysts than traditional filtration methods (Canadian Council of

Ministers of the Environment, Water Quality Task Group and Federal-Provincial-

Territorial Committee on Drinking Water 2004). Fungi are also found within drinking water system biofilms. Doggett (2000) was able to isolate 39 different species from sampling biofilms on drinking water system pipes. Filamentous species were more prevalent that yeasts, and spores were the predominant form found rather than hyphal or vegetative cells.

The work presented below aims to describe the collection of environmental isolates and development of a model drinking water distribution system biofilm community for use in examining the susceptibility of single and multi-species biofilms to chlorination, as well as the effects on multi-species biofilm structure and community composition. The presence of biofilms in treated water distribution systems has been well established, but only by understanding more about the community, structure and disinfectant resistance will we be able to better respond to and possibly eliminate contamination events caused by these biofilms.

Chapter Two: Materials and Methods

2.1 Media

Several types of solid and liquid media were used throughout the study including five selective and/or differential media as well as two general growth media. These were used both for isolation of organisms as well as for growth and recovery of biofilm cells. Media were selected based on their use in the literature as well as in standard water testing procedures.

2.1.1 Growth media

In order to recover the greatest variety of microorganisms, three types of general growth media were used within the project. Reasoner's 2A media (R2A) (EMD, Gibbstown NJ,

USA) is a low nutrient growth medium specifically designed for maximum recovery of organisms from chlorinated potable water samples using both spread plate and membrane filtration techniques (Reasoner and Geldreich 1985). R2B medium is the broth equivalent of the R2A agar and was used, as described below, in both 1x and 2x concentrations. R2A and R2B were used throughout the project for isolation, biofilm growth and recovery.

Luria-Bertani agar (LB) (EMD) was used along with R2A medium for isolation of organisms using membrane filtration, as described in Section 2.2.2.2. Plate count agar

(PCA) (EMD) was also used, although only for the isolation of heterotrophic bacteria from one source, as described in Section 2.2.2.3. As discussed in Section 1.5.1, standard methods for isolation of heterotrophic plate count (HPC) organisms can be selective

33 based on medium used and incubation conditions. For this reason, to reduce selectivity and recover the greatest diversity of organisms, the above three media were selected for use, as well as incubation at two temperatures (25°C and 37°C) and for a total of 14 days for isolation of HPC bacteria, as discussed in Sections 2.2.2.2 and 2.2.2.3.

2.1.2 Selective and differential media

In addition to general growth media, a variety of selective/differential media was used.

Differential Coliform (DC) agar (Oxoid, Nepean, ON, Canada) was used for the collection of isolates and biofilm recovery. On this medium, E. coli is differentiated from other coliform bacteria by the purple coloured colonies whereas, the total coliform bacteria grow as pink colonies. The medium was amended after being autoclaved with

Cefsulodin (Sigma-Aldrich), a β-lactam antibiotic, to a final concentration of 12 mg/L, in

order to inhibit growth of non-coliform organisms such as P. aeruginosa . DC agar was

found to be more efficient at detecting and enumerating both E. coli and total coliform

bacteria as well as suppression of other background organisms when compared to two

other E. coli and total coliform selective media (m-Coliblue 24 and chromocult coliform

Agar) (Wang and Wanda 2008). MacConkey agar (EMD) was also used for isolation of

Gram negative organisms. Buffered charcoal yeast extract (BCYE) agar (BBL, Becton,

Dickinson and Company, Sparks, MD, USA) was used for isolation of L. pneumophila from water as well as for biofilm recovery. The medium was made and supplemented with L-cysteine HCl (Sigma-Aldrich) to a final concentration of 0.04%, as per manufacturer’s instructions. For biofilm growth, buffered yeast extract (BYE) broth was used. BYE broth differs from the BCYE agar in that it does not contain charcoal which is

34 present in order to neutralize toxins produced when the medium is autoclaved (Feeley et al. 1979). As the charcoal would not stay suspended in a broth medium it was excluded and the broth was not autoclaved but instead heated to boiling and filter sterilized (0.22

µm pore size filter) (Pine et al. 1979). BCYE supplemented with antibiotics was also used as a selective variation of the medium.

BCYE PV was prepared by supplementing BCYE medium with polymyxin B (Sigma-

Aldrich; 40,000 units/L) and Vancomycin (Sigma-Aldrich; 0.5 mg/L). BCYE PAC did not require supplementation, but contained polymyxin B (80,000 units/L), cefamandole

(4.0 mg/L) and anisomycin (80.0 mg/L). Oxidation-Fermentation of Lactose (OFL) agar

(Difco, Becton, Dickinson and Company, Sparks, MD, USA; Sigma-Aldrich, St Louis

MO, USA) was used for recovery and differentiation of P. aeruginosa and S. maltophilia from mixed-species biofilms. These organisms are able to oxidize lactose and therefore are differentiated by a blue colour forming around the colonies in the medium. Presence-

Absence (PA) broth (EMD) is both selective against many background organisms and differential for lactose fermenting organisms and will turn from purple to yellow in their presence. Other opportunistic pathogens, such as P aeruginosa and S. maltophilia , were differentiated by causing rapid growth causing turbidity in the broth medium, between 18 and 24 hours of growth (Acumedia 2010).

2.2 Collection of bacterial isolates from drinking water

All bacterial isolates that were used in the research project were obtained from treated and regulated drinking water systems. A portion of the isolates were provided by two

35 different laboratories, Foothills Medical Centre Infection Prevention and Control laboratory and the Provincial Laboratory for Public Health (ProvLab), Calgary, AB. The remaining isolates were collected during the research period, as described below. All of the normal water system microflora isolates were obtained from sites supplied by the

Bearspaw Water Treatment Plant in northern Calgary, which draws water from the Bow

River. The opportunistic pathogen isolates were obtained from water sampling sites in both north and south Calgary as well as from treated municipal water systems and private well sites in Ontario.

2.2.1 Samples provided by laboratories

2.2.1.1 Infection Prevention and Control Laboratory

The Infection Prevention and Control laboratory at Foothills Medical Centre (Dr. Tom

Louie and Linda Ward) provided a number of isolates, four of which were incorporated into the multi-species biofilm model (Sections 2.4.5 and 2.5). They were originally collected as part of a hospital wide water monitoring program in 2007. The group of isolates provided included normal flora organisms and the opportunistic pathogen

Stenotrophomonas maltophilia.

2.2.1.2 Provincial Laboratory for Public Health, (ProvLab), Calgary, AB

The ProvLab provided two opportunistic pathogen isolates for use in the project. The

strain of Pseudomonas aeruginosa was collected from the Alberta Children's Hospital in

April 2009 during routine water monitoring. It was collected from the water intake to the

west wing of the complex, before the on-line UV treatment apparatus. The strain of

Legionella pneumophila used in this project was isolated from the water system at

Rockeyview General Hospital in 2003, as described in section 1.1 and by Wong et al.

(2006).

2.2.2 Collection of samples for new isolates

New isolates were obtained using two methodologies. Membrane filtration, which is standard practice in water testing, was used to collect bacteria living planktonically within the water system (Sections 2.2.2.1 and 2.2.2.2). The second isolate collection method was by directly sampling biofilms growing within pipes of a drinking water system (Section 2.2.2.3).

2.2.2.1 Ontario water systems

Membrane filtration was used in the collection of E. coli and other total coliform isolates at the Ontario Ministry of Environment Laboratory Services Branch (MoE LaSB) as per their accredited laboratory methods. Water samples were delivered to the laboratory from across southern Ontario and 100 mL of the sample was filtered through a 0.45 µm pore size filter (4.5 cm diameter)(Millipore, Etobicoke, ON, CA). This filter was then placed on a plate of DC agar, incubated at 37°C and evaluated for growth at 24 and 48 hours.

Colonies taken from these plates were then characterized using api20E strips

(bioMérieux, St Laurent, QC, CA) which use a series of biochemical tests to identify organisms within the family Enterobacteriaceae (Holmes, Willcox and Lapage 1978).

2.2.2.2 University of Calgary Biological Sciences building

Isolates collected from the Biological Sciences building at the University of Calgary were also obtained using membrane filtration. For these samples, 5 L of water was filtered through a 0.45 µm pore size filter (14 cm diameter) for recovery of maximum numbers of isolates. Filters were then placed on an R2A and LB agar plates and incubated for fourteen days at 25°C and 37°C. Colonies were selected for further study based on differing morphologies.

2.2.2.3 Alberta Children's Hospital

Collection of isolates directly from water system biofilm biomass was performed as part of a larger scale study performed by the ProvLab, the Infection Prevention and Control lab, the Alberta Children's Hospital and the Biofilm Research Group at the University of

Calgary (BRG). During construction and installation of the water system in the Alberta

Children's Hospital a number of pipes were installed in parallel with the main water system in locations throughout the hospital. Each pipe site contained 12 removable brass coupons (Figure 1). The objective of such installation was that after the water system had been in operation for a period of time, biofilms would form on these coupons, which could then be removed and used for further analysis. Sampling for strains used in this study was performed in September and October 2010. Coupons were removed from three sites including utility rooms in both the medical day treatment area (A3-701) and the paediatric ICU (B1-007) and in the nourishment centre of the rehabilitation area (C0-

328). Before removal, the area of the biofilm coupons exposed to the environment as well as the surrounding pipe and tools used were sterilized with bleach and rinsed with sterile

Figure 2.1: Alberta Children's Hospital water system coupons (A) Coupon manifold installed within the water distribution system of the Alberta

Children's hospital. Pipes containing coupons were installed parallel to main water pipes

of both the 43°C and 38°C systems. (B) Brass coupon removed from coupon manifold.

When coupons were removed during sampling they were replaced with new coupons.

After sampling, biofilms were sonicated off of coupons and spread plated on R2A, PCA,

MacKonkey's and BCYE media.

39 water. Coupons were removed from the pipes and placed in 12 mL of phosphate buffered saline (PBS) pH 7.2 with 1% Tween 20 (Sigma-Aldrich) and sonicated (Branson 5510

Ultrasonic Cleaner; Danbury, CT, USA) for 10 minutes to remove the biofilms from the coupons. The sonicate was spread plated on R2A, PCA, MacConkey's and BCYE agar and incubated for fourteen days at 25°C and 37°C,

2.3 Isolate identification and molecular analysis

2.3.1 Single-species DNA extraction

Strains were grown on R2A medium for 24 hours -14 days (depending on the growth rate of organism) until colonies were clearly discernible (1-2 mm diameter). Cells were suspended in 1 mL of purified water (DNase and RNase free) to the same turbidity as a

1.0 McFarland standard, representing approximately 3x 10 8 cells/mL (Dalynn

Biologicals, Calgary AB, CA), centrifuged for 10 minutes at 12,000 x g and the

supernatant discarded. The pellet was then re-suspended in 200 µL lysozyme solution, pulse vortexed for 10 seconds on medium speed and incubated for 30 minutes at 37ºC with agitation. Lysozyme solution consisted of 20 mg/mL lysozyme (lyophilized, -20ºC;

Sigma-Aldrich) in lysis buffer (20 mM Tris, 2 mM EDTA, 1% TritonX-100, pH 8). After incubation the 200 µL lysozyme solution and cells were transferred to a lysing matrix A

tube from the FastDNA spin kit DNA extraction kit (MP biomedicals, Solon, OH, USA)

containing 20 µL proteinase K (Qiagen, Mississauga, ON, CA) and 880 µL CLS-TC

(FastDNA spin kit, MP biomedicals). Tubes were placed in a bead beater (MP biomedicals) and run at speed 6 for 40 seconds then incubated at 56ºC for another 30 minutes. From this point the protocol included in the FastDNA spin kit was followed

40 from steps 4-11. After the addition of 100 µL purified water (DES, from the FastDNA spin kit) the sample was incubated for 20-30 minutes at 55ºC and underwent a final elution step by centrifugation at 14,000 x g for 2 minutes. Concentrations of double stranded DNA were quantified post extraction (whole genomic DNA) and again post

PCR cleanup (pure PCR product, see Section 2.3.2) using the PicoGreen assay kit

(Invitrogen, Burlington, ON, CA). In a black 96 well microtiter plate, 2 µL DNA product was added to 98 µL of 1x TE (Tris/EDTA) buffer (Invitrogen). Once DNA was added for all samples, 100 µL of 1x PicoGreen reagent (Invitrogen; 200x PicoGreen reagent diluted in TE buffer) was added to each well and the fluorescence of each sample was read and data was recorded using a fluorescent plate reader (excitation 485 nm, emission 538 nm).

Fluorescence readings were then converted to double stranded DNA concentration

(ng/µL) using the following equation generated from a standard curve (dsDNA concentration=((fluorimeter reading-15.06)/4.481)*0.05). DNA samples were stored at -

20ºC.

2.3.2 16S rRNA gene amplification and sequencing

2.3.2.1 16S rRNA PCR amplification

Extracted DNA was PCR-amplified using the universal eubacterial primers 27f

(5'AGAGTTTGATCMTGGCTCAG3') (Lane 1991) and 907r

(5'CCGTCAATTCMTTTRAGTTT3') (Muyzer et al. 1995), which amplify an ~880 bp region of the 16S rRNA gene. Both primers contain degenerate or wobble bases, represented by M and R. These sites contain either an A or C or an A or G respectively, in approximately equal ratios. The presence of these bases, along with a lower initial

41 annealing temperature during the PCR reaction allow for less specific binding to the template DNA therefore allowing amplification of a larger number of bacterial strains

(Telenius et al. 1992). In order to ensure each variation of the primer is present in sufficient quantities, a high initial primer concentration was used. The composition of the

PCR mixture per 100 µL reaction was as follows: 10 µL PCR reaction buffer (10x); 3 µL

MgCl 2 (50 mM); 2 µL dNTP mix (10 mM each dATP, dDTP, dGTP and dCTP); 4 µL

forward primer 27f (20 pmol/µL); 4 µL reverse primer 907r (20 pmol/µL); and 0.5 µL

recombinant Taq polymerase (all reagents obtained from Invitrogen). The reaction

proceeded at 95°C for 3 minutes for the initial template denaturation step; this was

followed by 35 cycles of 95°C for 45 seconds, 52°C for 45 seconds and 70°C for 1

minute; the final elongation step proceeded for 10 minutes at 70°C. PCR product was

then purified using the QIAquick PCR purification kit (Qiagen). The protocol included

with the kit was modified so that the sample was incubated before the final elution step

for 30 minutes at room temperature and was eluted in only 30 µL of purified water.

2.3.2.2 Sequencing and identification

Sequencing of the purified PCR product was performed by the UCDNA services

(University Core DNA Services, Calgary, AB, CA) on an Applied Biosystems 3730xl 96

capillary DNA Analyzer. Polyacrylamide gel electrophoresis (PAGE) purified 27f and

907r primers were used for sequencing. Purified PCR product was required to have a

DNA concentration 100 ng/1000 base pairs in DNA fragment submitted in 10 µL volume

(8.8 ng/µL) in order to be sequenced. Sequences were then analyzed using the NCBI

BLASTN 2.2.24+ application (Zhang et al. 2000) , the Ribosomal Database Project

(RDP) Sequence Match application (Cole et al. 2007, Cole et al. 2009) and the greengenes 16S rRNA gene database (DeSantis et al. 2006a, DeSantis et al. 2006b).

2.4 Biofilm analysis

For this study biofilms were grown using the Calgary Biofilm Device (CBD) (Ceri et al.

1999). The device utilizes a 96 well microtiter plate lid with corresponding polystyrene pegs on which, when placed in inoculated broth medium and incubated on a gyratory shaker, biofilms are formed. The biofilm cultures can then undergo a number of analyses from cell enumeration, to challenge with a biocide or genetic analyses.

2.4.1 Biofilm inoculation

The CBD was inoculated with both single and multi-species biofilms. For both single and multi-species biofilm assays, bacterial strains were sub-cultured from freezer stocks onto

R2A, and then re-streaked. This second subculture was then used to make the inoculum.

A swab of bacterial cells was suspended in 2-3 mL of sterile saline in a glass culture tube to the same turbidity as the 1.0 McFarland standard, measured visually. This was then diluted 1:30 in R2B for single-species biofilm inoculation. For multi-species biofilms, each bacterial strain suspension was added to the R2B broth at a 1:100 dilution. 150 µL of the diluted inoculum was added to each well of the microtiter plate, then the plate was incubated on a gyratory shaker at 125 rpm at 25ºC. For single-species biofilms grown longer than 24 hours, growth medium was replenished by adding 150 µL of R2B to a fresh microtiter plate and moving the biofilms (without rinsing) into the new medium, with no new inoculum added. Table 2.1 summarizes all of the strains used in the assay as

43 well as specific growth times and medium change schedule for each. Lengths of biofilm incubation were dependant on growth rate of the individual strains. Final inoculation schedule for multi-species biofilms was determined by testing three different trial inoculation schedules over a 12 day period. A terminal-restriction fragment length polymorphism (T-RFLP; see Section 2.5) assay was performed at days 6 and 12 to identify which schedule produced a biofilm with the highest proportion of species without any species specifically dominating. From these three trials, a final inoculation schedule was designed. The three trial schedules and final selected inoculation order are described in Table 2.2. Total number of cells added to the biofilm was equal for each inoculation schedule and fresh media was supplied to the biofilm as described in Table

2.2. Biofilms were grown for 24 hours after inoculation of the opportunistic pathogen, except for the multi-species biofilms containing P. aeruginosa which were only incubated an additional 20 hours.

2.4.1.1 Legionella pneumophila biofilms

Inoculation and biofilm growth of L. pneumophila biofilms was similar to inoculation of

other strains except for the following important differences. All subcultures of L.

pneumophila were made on BCYE agar and incubated at 37ºC, rather than on R2A at

25ºC. Before inoculation, pegs of the CBD were coated with BYE media by filling each

well of a microtiter plate with 200 µL of BYE and placing the CBD peg lid in the media

for 24 hours. For the inoculation of the biofilm, the McFarland standard was prepared in

BYE rather than saline and then again diluted in the same media, rather than R2B. The

plate lid was not rinsed between coating and being placed in the inoculum. Additionally,

Table 2.1: Single-species biofilms growth time and media change schedule Strain Total incubation Media changed

Opportunistic pathogens

MEC-8 Escherichia coli 24h na

MTC-21 Enterobacteriaceae cloacae 24h na

PAE-1 Pseudomonas aeruginosa 20h na

FH-W1 Stenotrophomonas maltophilia 24h na

LPR-1 Legionella pneumophila 48h na

Normal Flora

A3-1 Variovorax sp. 3d 48h

A3-2 Cupriavidus respiraculi 48h na

B1-1 Bradyrhizobium sp. 9d Every 72h

C0-3 Sediminibacterium sp. 4d 48h

C0-6 Kocuria rhizophila 4d 48h

FH-D Novosphingobium subterraneum 3d 48h

FH-G Sphingomonas sp. 3d 48h

FH-J Blastomonas natatoria 3d 48h

MWI-1 Methylobacterium isbiliense 6d Every 48h

MWI-2 Mycobacterium sp. 6d Every 48h

Table 2.2: Multi-species biofilm inoculation schedules for trials 1-3 and final Test inoculation schedules Final inoculation Day A B C schedule 0 All Strains except MWI-1; MWI-2 MWI-1; MWI-2; B1-1 MEC-8 C0-3 1 - - - - 2 Media changed C0-3; FH-D; FH-G; - FH-J; A3-1; C0-6

3 Media changed FH-D; FH-G; FH-A; FH-W1; MWI-1; MWI-2 FH-J; A3-1; C0-6 A3-2 4 MEC-8 FH-A; FH-W1; MEC-8 - A3-2; MEC-8 5 Media changed Media changed Media changed C0-3 6 Biofilm Recovery Biofilm Recovery Biofilm Recovery - 1 1 1 7 FH-D; FH-G; FH-J; A3-1 8 Media changed Media changed A3-2; C0-6 9 - - MEC-8; MTC-21; PAE-1 or FH-W1 10 Media changed Media changed Biofilm Recovery 11 12 Biofilm Recovery Biofilm Recovery 2 2 MEC-8 = Escherichia coli ; MTC-21 = Enterobacter cloacae ; PAE-1 = Pseudomonas aeruginosa; FH-W1 = Stenotrophomonas maltophilia; A3-1 = Variovorax sp. ; A3-2 = Cupriavidus respiraculi; B1-1 = Bradyrhizobium sp ; C0-3 = Sediminibacterium sp. ; C0-6 = Kocuria rhizophila ; FH-D = Novosphingobium subterraneum ; FH-G = Sphingomonas sp. ; FH-J = Blastomonas natatoria ; MWI-1 = Methylobacterium isbiliense; MWI-2 = Mycobacterium sp.

L. pneumophila biofilms were incubated at 37ºC.

2.4.2 Single-species biofilm growth evaluation

After biofilm incubation, cell growth on the pegs was evaluated to ensure equivalent biofilm growth across the plate. For the equivalent growth assay one half of the pegs on the CBD were evaluated (48 biofilms). The plate was removed from the shaker and the

CBD lid was removed from the spent medium into a microtiter plate containing 200 µL of sterile saline in each well for one minute to rinse off loosely adherent/planktonic cells.

The plate lid was then moved into the recovery plate, which contained 200 µL R2B with

0.5% Tween 20 per well. Biofilms of the FH-D isolate were found to be sensitive to

Tween 20 and therefore were recovered in R2B alone, without detergent. The recovery plate with biofilm lid was then sonicated for 10 minutes to further remove the biofilm cells from the pegs. After sonication the biofilm plate lid was discarded and the sonicate serially diluted in saline and spot plated onto R2A plates for enumeration of viable cells.

Again, changes were made for the L. pneumophila biofilms. (Pine et al. 1979) showed that adding even 0.05% Tween 80 to L. pneumophila media inhibited growth. Based on this, as well as the fact that sufficient cell growth was recoverable without the Tween, it was not used in the recovery of L. pneumophila biofilms. They were recovered in BYE broth, and then spot plated on BCYE agar. Statistical analysis was performed on the data to evaluate whether the biofilm growth was statistically different between the pegs on the

CBD. A one way analysis of variance and Tukey's multiple comparison test (comparison of the mean of each treatment, row or column, to the means of every other treatment) were applied to the mean CFU/peg values between each of the rows and again between

47 each of the columns (P< 0.05 for both tests).

2.4.3 Single-species minimum biofilm eradication concentration assay

Biofilm inoculation and growth was performed as described above. Mature biofilms were then challenged with increasing concentrations of free chlorine from sodium hypochlorite

(Clorox, Brampton, ON, CA) in a minimum biofilm eradication concentration assay

(MBEC). This assay is used to determine the MBEC value, the minimum concentration of a substance, in this case chlorine, required to eradicate a biofilm population or community. For low level chlorine challenge of the single-species biofilms a stock challenge solution was prepared by adding 2 mL chlorine bleach to 500 mL sterile water.

The concentration of free chlorine in the stock was measured in triplicate using a Hach

DR/700 hand held colorimeter (Hach, Loveland, CO, USA; method 52.05.1) with Hach

DPD free chlorine reagent pouches. The stock was then diluted in PBS to obtain a working solution equal to the highest concentration of chlorine to be used in the assay

(32-124 mg/L free Chlorine, depending on the organism under evaluation). The 96 well microtiter challenge plate was created by performing one half dilutions in PBS from the highest chlorine level to the lowest in columns 1-11. The last column contained 200 µL

PBS, as the zero chlorine control. Biofilms on the CBD plate lids were rinsed twice in saline for 1 minute/rinse and then placed in the challenge solution for 10 minutes. Excess medium must be fully rinsed off the biofilms as it will neutralize the chlorine and give a false negative result for chlorine disinfection efficacy. After the 10 minute challenge, the pegs were again rinsed twice for 1 minute/rinse in saline and then placed in the recovery plate. The recovery plate contained 200 µL R2B with 0.5% Tween 20, except again for

48 the FH-D and L. pneumophila isolates. The R2B broth acted to neutralize the chlorine for these experiments so no additional neutralizer was needed. As with the growth assay, biofilms were sonicated in the recovery broth to remove cells from the pegs for 10 minutes. The growth check biofilm wells (column 12, no chlorine) were serially diluted and 20 µL were spot plated to enumerate the number of cells in the initial biofilm. For all other wells 40 µL of the sonicate (1/5 of total volume) was spot plated on R2A agar

(BCYE for L. pneumophila ) to check for +/- survival.

2.4.4 Planktonic single-species minimum inhibitory concentration assay

For each isolate, an MIC (minimum inhibitory concentration) assay was performed on

suspended planktonic cells, in addition to the MBEC assay performed on the biofilm.

Cells were suspended in PBS to the equivalent of a 1.0 McFarland standard (as above)

and then 50 µL of the culture was diluted in 20 mL PBS (approximately 10 5 CFU/well).

The challenge plate was prepared as described above for the MBEC assay with the

exception that only 100 µL was added to each well and the concentrations of free

chlorine were 2X the desired final concentration. 100 µL of the bacterial culture in PBS

was added to each challenge well (to bring the total volume to 200 µL and dilute the

chlorine to the correct concentration) and left for 10 minutes. After that time, 100 µL was

moved from each challenge well and added to a new microtiter plate containing 100 µL

2X R2B per well to neutralize the chlorine. Wells were then spot plated on R2A as

described above to obtain total initial bacterial numbers and survival at increasing

chlorine concentrations.

2.4.5 Multi-species minimum biofilm eradication concentration assay

For high level chlorine challenge of multi-species biofilms, the challenge plates were prepared as follows. A stock chlorine solution was made by adding 2 mL bleach to 38 mL of PBS and the concentration was measured as above (Section 2.4.3). The stock was then diluted with PBS to the desired working concentrations of free chlorine for the assay

(1000, 2000 or 3000 mg/L). The challenge plates were created by adding an amount of the working challenge solution to PBS to achieve 200 µL of solution at the desired concentration (as opposed to ½ dilutions of the stock solution as in Section 2.4.3).

Biofilms are rinsed twice in saline, challenged for 10 minutes, then rinsed twice more before recovery. Recovery of multi-species biofilms was performed using three different culture media: the growth medium R2A, one selective/differential plate medium for the specific opportunistic pathogen under observation and PA broth. DC agar was the selective medium used for Escherichia coli MEC-8 and Enterobacter cloacae MTC-21 and OFL for Pseudomonas aeruginosa PAE-1 and Stenotrophomonas maltophilia FH-

W1. The first two recoveries were performed as described above (Section 2.4.3) by sonicating in 200 µL R2B with 0.5% Tween 20 and spot plating. For the recovery in PA broth, after the biofilms had been challenged and rinsed they were placed in a plate with

200 µL R2B for 10 minutes to neutralize chlorine, the same length of time used for the sonication of the other biofilm recovery methods. They were then placed in a plate with

200 µL of PA broth per well and incubated for 24-48 hours at 35ºC. The plate was scored

+/- for growth, and therefore survival of the biofilm, based on colour change (MEC-8 and

MTC-21) or increased turbidity (PAE-1, FH-W1). Recovery in the three media types allowed the establishment of an MBEC value for the entire multi-species community as a

50 whole (R2A recovery) as well as an MBEC value for the population of the specific opportunistic pathogen under study within the community of normal flora using the selective and differential recovery media.

2.5 Community analysis of multi-species biofilms

2.5.1 Biofilm challenge

Community analysis of strains within the multi-species biofilms was performed using

Terminal-restriction fragment length polymorphism (T-RFLP). The multi-species biofilms were treated with low (16 or 50 mg/L) and high (500 or 1000 mg/L) free chlorine concentrations for 10 minutes (stock and working solutions created as described in Section 2.4.3). E. coli MEC-8, P. aeruginosa PAE-1 and control biofilms (no opportunistic pathogen added) were treated with 50 and 500 mg/L free chlorine and E. cloacae MTC-21 and S. maltophilia FH-W1 were treated with 16 and 1000 mg/L free chlorine. This is discussed further in Section 6.2.7. After the challenge, biofilm- associated cells were recovered in 200 µL R2B for 24 hours. Treated pegs as well as control, non chlorine-treated pegs were broken off from the CBD lid and attached biofilms underwent DNA extraction.

2.5.2 DNA extraction from pegs

For this assay DNA was extracted from cells directly adhered to the CBD pegs, rather than from suspension, as described in Section 2.3.1. For each chlorine treatment (low, high and control) six pegs were removed and placed in 1.5 mL microcentrifuge tubes with 300 µL lysozyme solution with 2 pegs/tube. The DNA extraction procedure

51 continued as with the single species DNA extraction, with the only difference being that

780 µL CLS-TC (included with MPBio FastDNA spin kit) was used instead of 880 µL.

The pegs and lysozyme solution were transferred to the Lysing Matrix A tubes and the

pegs were discarded after the 56ºC incubation for 30 minutes and 10 minute

centrifugation step at 14,000 x g (step 4 of the FastDNA spin kit protocol). The final

eluted genomic DNA was again quantified with the PicoGreen assay.

2.5.3 Amplification with labelled primer

For this reaction the same 16S rRNA gene fragment was amplified as in the sequencing

reaction described above using the 27f and 907r primers (Section 2.3.2.1), although in

this case the forward 27f primer was labelled with a 6FAM fluorescent label (6 -

Carboxyfluorescein). All of the PCR reagents were present in the same ratios as in the

sequencing reaction although for this assay a 50 µL reaction was used and 0.1 µL of

extracted DNA was added to each. The reaction proceeded at 95°C for 5 minutes;

followed by 25 cycles of 95°C for 1 minute, 52°C for 1 minute and 70°C for 1.5 minutes;

and a final elongation step for 10 minutes at 70°C. PCR product was purified using the

QIAquick PCR purification kit as in Section 2.3.2.1.

2.5.4 Restriction enzyme digest

Amplified DNA fragments were restriction enzyme digested using HhaI (Invitrogen)

which has a four base pair recognition site of GCG/C. This enzyme was selected for this

assay by performing an in silico digest of sequences obtained using the method presented

in Section 2.232 with six different restriction enzymes. The in silico digest involves the

52 identification of restriction enzyme cut site sequences within the amplified sequence from each isolate, and calculating the theoretical terminal restriction fragment length for each sequence for all six of the restriction enzymes tested. The ideal restriction enzyme for the

T-RFLP assay would cut at a different location along the 16S rRNA gene for each isolate, giving each one a unique terminal restriction fragment length (T-RFL). Performing the theoretical in silico digest allows the identification of the most appropriate restriction enzyme for use in the TRFLP assay. Once HhaI was selected, reactions were carried out in a total volume of 20 µL containing 1 µL HhaI, 2 µL buffer M (Invitrogen), 2 µL pure water and 15 µL purified PCR product. Reactions proceeded for 4 hours at 37°C on a gyratory shaker and the reaction was stopped by incubating tubes at 65°C for 20 minutes.

Samples were PCR purified and diluted by one half before being sent to UCDNA services for fragment analysis using the Applied Biosystems 3730xl 96 capillary DNA

Analyzer. Samples that contained large amounts of DNA and registered above the

detection limits were diluted another five fold (to a total of a 1:10 dilution) and re-

analysed. Results for samples as well as standard curves were checked for quality and

analysed using Peak Scanner Software v1.0 (Applied Biosystems). Raw data was

inputted into the software, which was able to calculate peak height and area as well as

identify off scale samples. Data from the Peak Scanner software was exported and

opened in the T-REX online tool (T-RFLP analysis EXpedited, Computational Biology

Service Unit, Cornell University) (Culman et al. 2009). This software was used to align

peaks between samples using a clustering threshold of 0.5 to account for slight TRFL

drift between samples. The software was also used to identifying true peaks and filtering

background noise from the sample, using a factor 1.5 inputted into the software as a

53 multiplication of the standard deviation for the calculation. Relative abundance of each of the peaks within a sample were calculated by dividing an individual peak area by the total area of all peaks present within a sample and multiplying by 100%.

2.6 Biofilm microscopy

2.6.1 Confocal scanning laser microscopy (CSLM)

Multi-species biofilms containing opportunistic pathogens, both chlorine treated and untreated, were visualized using confocal scanning laser microscopy (CSLM). Biofilms were stained with the Live/Dead BacLight bacterial viability kit (Invitrogen) containing propidium iodide and SYTO 9. Pegs were removed from the CBD lid, rinsed for 1 minute with saline, challenged with 16 mg/L chlorine for 10 minutes (if applicable), rinsed again, stained for 10 minutes with the Live/Dead stain and then rinsed a final time with saline for 1 minute. They were then visualized on the Leica Microsystems (Concord ON, CA)

DM IRE2 spectral confocal and multiphoton microscope with the Leica TCS SP2 acoustic optical beam splitter (AOBS). A stack of images of the biofilms was taken from the centre section of the peg to create a 3D rendering of the biofilm. Images were then compiled and analyzed using Imaris x64 version 7.0.0 software (Bitplane Inc. Scientific

Software, South Windsor, CT, USA).

2.6.2 Scanning electron microscopy (SEM)

Scanning electron microscopy (SEM) was performed on untreated control multi-species biofilms (no opportunistic pathogens added). Pegs were removed from the CBD, and rinsed in saline. Biofilms were fixed in 200 µL 2.5% glutaraldehyde in 0.1M cacodylic

54 acid buffer (pH 7.2) for 3 hours at room temperature. After fixing, biofilms were rinsed for 10 minutes each in 0.1M cacodylic acid buffer and then distilled H 2O (200 µL). Pegs were placed in 70% ethanol for 20 minutes to dehydrate and then air dried for a minimum of 24 hours before mounting and visualization by SEM using a Philips ESEM XL-30

(Microscopy and Imaging Facility, University of Calgary).

Chapter Three: Isolation and identification of bacteria from drinking water distribution systems

3.1 Aims

Creation of a model drinking water distribution system biofilm was one of the ultimate goals of the project outlined in Section 1.1.2. In order to do this it was ideal to collect organisms with which to build the model community directly from that environment. It is possible these isolates may be more adapted to water distribution system conditions, specifically chlorination, than laboratory strains. It was also ideal to develop other, non- culture based, methods to evaluate the species and genus populations within the community. With these goals in mind, the following aims were undertaken for this portion of the project:

• Collection of bacterial isolates that would be representative of a native

community of bacteria within a water system

• Use 16S rRNA gene sequencing to identify the isolates to the genus or species

level

• Evaluate and choose a subset of isolates to form a multi-species biofilm model,

including a group of opportunistic pathogens and indicator organisms

• Create a T-RFLP library of these strains for use in characterizing multi-species

biofilms.

3.2 Isolate collection

Isolates for this study were either obtained from collected environmental samples or supplied from a number of different laboratories. Initially, close to 100 isolates were evaluated for the project, although not all of the isolates could be characterized by 16S rRNA gene sequencing and in the single and multi-species biofilm assays within the time and resource constraints of the project. This section describes the samples provided by or collected from each source and which, if not all, isolates were selected for further analysis from each source.

3.2.1 Isolates collected at the Alberta Children's Hospital

In total, 12 isolates with different colony morphologies were obtained from various

media agar plates for use in this project from the Alberta Children's Hospital (ACH) (as

described in Section 2.2.2.3. Additional sampling occurred at the site, but the isolates

were not considered for use in this study as the sampling dates were later into the project.

Table 3.1 lists the isolate ID's, the sampling source, colony morphologies and the media

types on which each isolate was initially cultured. All isolates were identified using 16S

rRNA gene sequencing (Section 3.3), as was required as part of the larger project with

the ProvLab, the Infection Prevention and Control (IPC) laboratory and the Biofilm

Research Group (BRG).

3.2.2 Isolates provided by the Infection Prevention and Control laboratory

The IPC laboratory at Foothills Medical Center (FMC) provided a total of forty-seven

isolates. Appendix I contains a list of the isolates provided by isolate ID, the media on

Table 3.1: Bacterial isolates recovered from the water system of the Alberta Children's Hospital. Isolate Originally Sampling Location Sampling date Colony morphology ID isolated on (dd/mm/yyyy) on R2A at 25°C B1-1 25/35°C B1-007 PICU soiled 27/09/2010 tiny/white PCA utility 25/35°C R2A C0-1 25/35°C C0-328 Nourishment 19/10/2010 m/cream PCA center 25/35°C R2A 35°C BCYE C0-2 35°C PCA s/cream C0-3 25/35°C s/orange/trans PCA C0-4 25°C PCA m/cream C0-5 25°C R2A m/yellow C0-6 35°C m/light yellow BCYE C0-7 25°C PCA m/yellow C0-8 25/35°C tiny/white PCA 25/35°C R2A A3-1 25°C R2A A3-701 Day surgery 19/10/2010 s/yellow soiled utility A3-2 35°C PCA m/beige

35°C BCYE Epost-1 25/35°C East Post-UV 25/01/2011 tiny/white R2A wing treatment unit Isolates originate from the biofilm phase of the water system. All sampling sites are on the 43°C water system except EPost-1 which is not heated.

58 which they were isolated, the colony morphology and growth characteristics on different media and at different temperatures. Based on this information, isolates with duplicate characteristics were considered to have similar identification and only one was evaluated further. Of the 47 original isolates provided, 15 unique colony types were observed and thus15 isolates were selected for further evaluation using 16S rRNA gene sequencing and

biofilm growth assays. These isolates and the information listed above are presented in

Table 3.2.

3.2.3 Isolates collected at the Ontario Ministry of the Environment

As part of routine monitoring of a number of various water systems in southern Ontario,

water samples were sent to the Ontario Ministry of the Environment (MoE) laboratory

services branch (LaSB). From these samples, a total of 36 isolates were collected for the

project, 13 of which were originally identified in that laboratory as E. coli using api20E

strips. The remaining 23 isolates represented different species of indicator organisms

from the family Enterobacteriaceae, including species from the genera Enterobacter,

Serratia, Citrobacter and Rahnella . Appendix II summarizes the isolates collected in

southern Ontario, including the regions within the province where the sample was

collected as well as the api20E identification for each. Seven strains from this group were

selected for sequencing to confirm the identification assigned using the api20E strips: 5

isolates of presumptive E. coli and 2 total coliform isolates. One isolate from each of

these groups was used for the multi-species biofilm model as a representative

opportunistic pathogen and indicator organism. The region from which the sample was

Table 3.2: Bacterial isolates recovered from water system of the Foothills Medical Centre (planktonic organisms) in 2004 by the IPC laboratory. ID Originally Sampling Sampling date Sept 2007 Growth Colony Isolated Location (dd/mm/yyyy) sub on R2A (R2A) morphology on on (37°C) 37°C 25°C R2A at 25°C A BCYE s/ PCU61 Room 20/10/2004 white/milky ++ ++ m/cream/trans sink D BCYE yellow ++ ++ s/light yellow/trans G BCYE GWHC Utility 20/10/2004 yellow +++ +++ s/yellow/trans room H BCYE NG +++ +++ s/yellow

gold/trans

J1* BCYE s/light yellow +++ +++ s/gold

K1* BCYE small ++ +++ variable yellow** size/pink L BCYE yellow haze + ++ tiny/trans P BCYE Patient 20/10/2004 NG +++ ++ s/yellow room Y BCYE ER Minor 13/10/2004 white +++ + tiny/cream/trans surgery B1 BCYE s/white +++ ++ lrg/bright +haze white/opaque J1 BCYE s/ Minor 14/10/2004 peach/trans +++ +++ tiny/trans surgery Q1 R2A37° CVICU Med 20/10/2004 peach ++ + tiny/trans room S1 R2A37° NG ++ - NG sink W1 BCYE s/ PCU61 Patient 20/10/2004 white/mucoid +++ +++ m/cream yellow room Z1 BCYE s/ GWHC Exam 20/10/2004 white/mucoid +++ +++ s/cream/trans room NG = no growth; trans = translucent *greater than 1 colony morphology was obtained from original sample. ** Original isolate of FH-K on BCYE was a pink organism. PCU 61 (Patient Care Unit 61); GWHC (Grace Women's Health Centre); ER (Emergency); TBCC (Tom Baker Cancer Centre); CVICU (Cardiovascular Intensive Care Unit); BCYE s/ (BCYE with antibiotics). The grey section of the table represents data obtained from the IPC lab staff. Isolates listed above underwent further analysis, see Appendix I for a full list of isolates provided by the IPC lab.

60 collected and percent identity determined by the api20E test for each of the 7 selected strains is included in Table 3.3.

3.2.4 Isolates collected from University of Calgary Biological Sciences building

Two distinct colony morphology types were isolated from the University of Calgary

Biological Sciences building water system using membrane filtration. Both isolates were

relatively slow growing (~10-14 days to culture). The first isolate, given the code MWI-

1, was a small (~1 mm) bright pink colony and grew only at 25°C on low nutrient R2A

media. The second isolate, MWI-2, was a slightly larger (1.5-2mm) yellow-orange

colony and was able to grow at both 25 and 35°C (Figure 3.1). Both isolates were

identified using 16S rRNA gene sequencing.

3.3 Isolate identification

The large group of isolates obtained was narrowed down to a manageable number first by

eliminating strains with the same colony morphology, growth rates and then by using 16S

rRNA gene sequencing to ensure there was not more than one strain representing the

same species within the multi-species model drinking-water biofilm. Of all the strains

isolated or provided, 29 isolates were identified via 16S rRNA gene sequencing (as

described in Section 2.2). Table 3.4 identifies the 29 isolates used for this project by the

isolate ID and the putative identification as well as gives a percent identity based on the

most closely related strain as found using the 16S rRNA gene sequence databases from

the National Centre for Biotechnology Information (NCBI), Ribosomal Database Project

(RDP), and Green Gene. Appendix III contains the full sequences obtained from the

Table 3.3: Bacterial isolates recovered from various water systems within southern Ontario and processed at the Ontario Ministry of the Environment (MoE) Lab. Isolate Sampling Water system Date sub'd for Preliminary MoE ID Location type api test (~2-4 identification sample days post using api20E accession sampling) biochemical tests number (dd/mm/yyyy) MEC-1 Waterloo private well 25/02/09 Escherichia coli 166332- Regional (99.8%) 6-12 Municipality MEC-7 Muskoka Treated 30/05/09 Escherichia coli 168232- District distribution (98.4%) 1-24 Municipality system (AWQI) MEC-8 Lennox and private (raw) 06/18/09 Escherichia coli 168808- Addington well (99.9%) 1-49 County MEC-12 Niagara Treated 10/08/09 Escherichia coli ES2-1 Regional municipal (94.0%) Municipality distribution (Citrobacter system (AWQI) freundii 4.2%) MEC-13 Niagara Treated 10/08/09 Escherichia coli ES2-2 Regional municipal (94.0%) Municipality distribution (Citrobacter system (AWQI) freundii 4.2%) MTC-21 Peterborough Treated 07/08/09 Enterobacter ES17-2 County municipal cloacae (91.5%) distribution system (AWQI) MTC-23 Elgin County Treated 07/08/09 Enterobacter ES20-1 municipal intermedius distribution (93.5%) system (AWQI) AWQI (Adverse water quality incident) sample was taken from a regulated drinking water system and the positive result was reported to the local public health unit and corrective actions taken. Isolates listed above underwent further analysis, a full list of isolates from the MoE is provided in Appendix II

Figure 3.1: Colony isolation from the University of Calgary Biological Sciences water system Colonies are growing on a 0.45 µm pore size membrane filter on R2A media after 14 days at 25°C. 5 L of tap water was filtered through the membrane before plating and two colony morphologies were isolated MWI-1 (pink pigment) and MWI-2 (orange pigment).

Table 3.4: Identification of 29 isolates using three 16S rRNA sequence databases: NCBI, RDP and Green Gene. Results shown indicate the most closely related organisms in each of the databases. Strain Putative 16S rRNA gene sequence database ID identification NCBI RDP Green Gene Alberta Children's Hospital Biofilm Isolates A3-1 Variovorax sp. Variovorax boronicumulans =99% Variovorax paradoxus =99.9% Variovorax paradoxus =99.40% Variovorax paradoxus =98% Variovorax sp. =99.9% Variovorax sp. =99.16% A3-2 Cupriavidus Cupriavidus necator =98% Cupriavidus respiraculi =99.9% Cupriavidus respiraculi =99.76% respiraculi Cupriavidus respiraculi =98% Cupriavidus golardii =98.82% B1-1 Bradyrhizobium sp. Blastobacter denitrificans =99% Bradyrhizobium sp. =100% Bradyrhizobium sp. =100.00% Blastobacter denitrificans =99% Bradyrhizobium japonicum =100% C0-1 Cupriavidus Cupriavidus necator =98% Cupriavidus respiraculi =99.9% Cupriavidus respiraculi =99.76% respiraculi Cupriavidus respiraculi =98% Cupriavidus sp. =98.93% C0-2 Cupriavidus Cupriavidus necator =98% Cupriavidus respiraculi =99.7% Cupriavidus respiraculi =99.65% respiraculi Cupriavidus respiraculi =98% Cupriavidus golardii =98.70% C0-3 Sediminibacterium sp. Sediminibacterium salmoneum 91% Sediminibacterium sp. =99.6% Sediminibacterium sp. =99.29% Chitinophaga niabensis =90% Hydrotalea flava =99.6% C0-4 Cupriavidus Cupriavidus necator =98% Cupriavidus respiraculi =99.9% Cupriavidus respiraculi =99.88% respiraculi Cupriavidus respiraculi =98% Cupriavidus golardii =98.93% C0-5 Rathayibacter tritici Rathayibacter tritici =99% Rathayibacter tritici =99.9% Rathayibacter tritici =99.52% Rathayibacter festucae =99% Rathayibacter sp. =99.7% C0-6 Kocuria rhizophila Kocuria rhizophila =100% Kocuria rhizophila =100% Kocuria rhizophila =100.00% Kocuria varians =99% C0-7 Micrococcus luteus Micrococcus luteus =99% Micrococcus luteus =100% Micrococcus luteus =99.52% Micrococcus endophyticus =99% Micrococcus sp. =99.52% C0-8 Bradyrhizobium sp. Blastobacter denitrificans =99% Bradyrhizobium sp. =99.9% Bradyrhizobium sp. =99.87% Bradyrhizobium betae =99% Bradyrhizobium japonicum =99.9% Bradyrhizobium liaoningense =99% EPost-1 Acidovorax sp. Acidovorax temperans =99% Acidovorax sp. =100% Acidovorax sp. =100.00% Acidovorax delafieldii =99% Acidovorax temperans =100% Acidovorax delafieldii =99.76% Acidovorax temperans =99.64% 63

Foothills Medical Centre Isolates

FH-A Acidovorax temperans Acidovorax temperans =99% Acidovorax sp. =99.7% Acidovorax sp . =99.88% Acidovorax delafiedii =98% Acidovorax temperans =99.5% Acidovorax temperans =99.64% FH-D Novosphingobium Novosphingobium subterraneum Novosphingobium subterraneum Novosphingobium yangbajingensis subterraneum =96% =98.8% =99.71% Novosphingobium hassiacum =96% Novosphingobium yangbajingensis Sphingomonas sp. =94.62% =97.2% FH- G Sphingomonas sp. Sphingomonas sanguinis =97% Sphingomonas sp. =98.7% Sphingomonas sp. =98.62 Sphingomonas pseudosanguinis =97% Sphingomonas ginsenosidimutans Sphingomonas adhaesiva =98.12 =98.6% Sphingomonas yunnanensis =98.6% FH-J Blastomonas Blastomonas natatoria =99% Blastomonas natatoria =99.9% Blastomonas natatoria =99.87% natatoria/ Sphingomonas ursincola =99% Sphingomonas ursincola =99.9% Sphingomonas ursincola =99.87% Sphingomonas ursincola FH-K Methylobacterium Methylobacterium populi =100% Methylobacterium populi =100% Methylobacterium populi =100.00% populi Methylobacterium rhodesianum =99% FH-P Blastomonas Blastomonas natatoria =99% Blastomonas natatoria =99.9% Blastomonas natatoria =99.75% natatoria/ Sphingomonas ursincola =99% Sphingomonas ursincola =99.9% Sphingomonas ursincola =99.75% Sphingomonas ursincola FH-Y Novosphingobium Novosphingobium subterraneum =96% Novosphingobium subterraneum Novosphingobium yangbajingensis subterraneum Novosphingobium hassiacum =96% =98.7% =99.55% Sphingomonas aromaticivorans Sphingomonas sp. =94.39% =97.3% FH-J1 Sediminibacterium sp Sediminibacterium salmoneum =91% Sediminibacterium sp =97.7% Sediminibacterium sp. =95.90% Chitinophaga niabensis =90% Hydrotalea flava =97.7% FH-Q1 Sediminibacterium sp Sediminibacterium salmoneum =91% Sediminibacterium sp =97.8% Sediminibacterium sp. =95.87% Chitinophaga niabensis =90% Hydrotalea flava =97.8% FH-W1 Stenotrophomonas Stenotrophomonas maltophilia =99% Stenotrophomonas maltophilia =99.9% Stenotrophomonas maltophilia =100% maltophilia 64

Ontario MoE Isolates

MEC-1 Escherichia or Shigella flexneri =99% Escherichia coli =100% Escherichia coli =1.000 Shigella sp. Escherichia fergusonii =99% Shigella sp. =100% Shigella dysenteriae =99% Escherichia coli =99% Escherichia albertii =99% MEC-7 Escherichia or Shigella flexneri =99% Escherichia coli =100% Shigella sp. =100.00% Shigella sp. Escherichia fergusonii =99% Shigella flexneri =100% Escherichia coli =100.00% Shigella dysenteriae =99% Escherichia coli =99% Escherichia albertii =99% MEC-8 Escherichia coli Shigella flexneri =99% Escherichia coli =100% Escherichia coli =99.88 Escherichia fergusonii =99% Escherichia coli O103:H2 =99.77 Shigella dysenteriae =99% Escherichia coli =99% Escherichia albertii =99% MTC- Enterobacter cloacae Enterobacter hormaechei =99% Enterobacter cloacae =99.9% Enterobacter cloacae =100.00% 21 (Enterobacter cloacae complex) Enterobacter sp. =99.9% Enterobacter sp. =99.77% Enterobacter cowanii =99% Enterobacter hormaechei =99.65%

MTC- Kluyvera intermedia Kluyvera intermedia =99% Kluyvera intermedia =99.7% Kluyvera intermedia =99.88 23 Kluyvera cryocrescens =99% Enterobacter intermedius =99.88 Citrobacter braakii =99% Kluyvera cochleae =99.88 University of Calgary Biological Sciences Building Isolates

MWI-1 Methylobacterium Methylobacterium isbiliense =100% Methylobacterium isbiliense =100% Methylobacterium isbiliense =100.00% isbiliense Methylobacterium nodulans =98% Methylobacterium sp. =100.00% MWI-2 Mycobacterium sp. Mycobacterium frederiksbergense Mycobacterium frederiksbergense Mycobacterium sacrum =99.88% =99% =100% Mycobacterium sp. =99.88% Mycobacterium diernhoferi =99% Mycobacterium sacrum =100% Mycobacterium frederiksbergense =99.88% Full sequences obtained from the UCDNA sequencing lab can be found in Appendix III 65

University Core DNA Services laboratory for each organism. Of the strains used in the multi-species biofilm model, the two strains provided by the AHS ProvLab,

Pseudomonas aeruginosa PAE-1 and Legionella pneumophilia LPR-1, did not undergo

16S rRNA gene sequencing during this project as they had previously been identified by the AHS ProvLab. Within the group of 29 strains, the majority belonged to the phyla

Proteobacteria (classes alpha, beta and gamma) and only 4 of the strains were identified as Gram positive, all belonging to the phylum Actinobacteria. Of the 29 sequenced isolates, there were a number which had the same species identification, both within the same sampling sets and between different sampling sites.

3.3.1 Duplicate genera and species within sampling sets

Within groups of strains originating from the same source there were a number of duplicates of species types. Of the samples obtained from the ACH five of the isolates were identified as Cuprividius respiraculi (A3-2, C0-1, C0-2, C0-4), with 99% sequence similarity between all the isolates . Additionally, two strains were identified as

Bradyrhizobium sp. (B1-1, C0-8; 99% sequence similarity) and two species belonged to the family Comamonadaceae ( Variovorax sp. A3-1, Acidovorax sp. Epost-1). Further sampling from this site has subsequently been performed. Samples taken from the FMC also had a number of isolates with the same identification. Five isolates were classified as belonging to the family Sphingomonadaceae, including two each of the species

Blastomonas natator (or Sphingomonas ursincola ) (FH-J, FH-P; 99% sequence

similarity) and Novosphingobium subterraneum (FH-D, FH-Y; 99% sequence similarity).

Two isolates were also classified as Sediminibacterium spp. (FH-J1, FH-Q1; 100%

sequence similarity).

The samples obtained from the Ontario MoE were previously identified using the api20E

test as described in section 2.2.2.1. This identification is based on a battery of

biochemical tests, and provided a percentage likelihood of identification for each isolate.

For the isolates obtained in this project, percent likelihood of identification ranged from

59.7% (MTC-3; Citrobacter braakii ) to 99.9% (MEC-6 and MEC-8; Escherichia coli )

Although a the api20E test is an effective assay, 16S rRNA gene sequencing has been

reported to occasionally provide more definitive species identification (Bosshard et al.

2006). Therefore, further verification of identifications was performed using 16S rRNA

gene sequencing. Of the 7 isolates sequenced, five were identified as E. coli using the

api20E test. MEC-1 and MEC-7 were identified as either Escherichia or Shigella spp. by all of the databases used. The NCBI database provided the same classification for the strain MEC-8 as MEC-1 and MEC-7, although both the RDP and Green Gene databases indicated that the closest identification for MEC-8 was E. coli , and not Shigella spp . Two additional isolates which were also tested, MEC-12 and MEC-13, were reported to have multiple sequencing products present within each reaction and so identification was not possible. Extraction, amplification and sequencing were performed in triplicate for MEC-

12 and once for MEC-13, all providing the same results. For this reason there were excluded from further analyses. MTC-21 and MTC-23 were also both previously identified using the api20E test as Enterobacter cloacae ( 91.5%) and Enterobacter intermedius (93.5%) respectively. MTC-21 was confirmed as Enterobacter cloacae as

68 originally identified. Alternatively, MTC-23 was identified as Kluyvera intermedia using sequencing and not Enterobacter intermedius.

3.3.2 Duplicate genera and species between sampling sets

Recurrence of species was also found between sampling sets as well as within sets.

Sediminibacterium sp. was isolated from both the ACH (C0-3) as well as the FMH (FH-

J1, FH-Q1). There was a 96% sequence similarity between C0-3 and both FH-J1 and FH-

Q1. Similarly, Acidovorax spp. were also isolated from both these sites (EPost-1 from the

ACH, FH-A from the FMC; 99% sequence similarity). Methylobacterium spp. were

isolated from both the University of Calgary Biological Sciences building (MWI-1) and

the FMC (FH-K) although in this case they were different species: Methylobacterium

isbiliense and Methylobacterium populi, respectively, and only a 95% sequence similarity. Figure 3.2 shows the phylogenetic grouping of all of the sequenced strains, with 12 distinct branches or groups on the tree.

3.4 Terminal Restriction Fragment Length Polymorphism Library construction

3.4.1 Selection of isolates for multi-species biofilm model

The selection of the subset of isolates for the multi-species biofilm model tests was done using the information gained through sequencing as well as data from single species biofilm tests (Chapter 4). Using Figure 3.2, one isolate was chosen to represent each of the phylogenetic branches. Although the genus groups within the family

Sphingomonadaceae are more closely related (represented by the shorter arms on the phylogenetic diagram), one isolate was chosen from each of the genera Blastomonas,

Figure 3.2: Phylogenetic tree of sequenced environmental isolates Tree displays the phylogenetic grouping of aligned 16S rRNA gene sequences for the 29 isolates sequenced successfully. Arrows indicate which 11 isolates were used for the multi-species biofilm model. It is a neighbour joining tree, and bootstrap values calculated from 1000 replicates.

Novosphingobium and Sphingomonas . As discussed in section 1.5.1 it has been suggested that organisms within this family occur frequently within drinking water system biofilms (Tokajian et al. 2005). Of the sequenced Gram positive isolates, even though they do not belong to the same genera, two isolates (C0-5, C0-7) were excluded from the multi-species biofilm model based on poor biofilm formation ability in the CBD which is discussed further in Chapter 4. The final group of isolates selected for use in the multi-species model are indicated on Figure 3.2, distinguished with an arrow and are also listed in Table 3.5.

3.4.2 Evaluating restriction enzymes in silico

In order to select the most appropriate restriction enzyme for the T-RFLP analysis, a number of enzymes were evaluated in silico against the sequences reported in Section 3.3

Table 3.6 shows the theoretical T-RFLs for each of the organisms for six different restriction enzymes: HhaI, MboI, AluI, HpaII, RsaI and HaeII. These six were selected based on their frequent use in the literature for T-RFLP assays. The specific 4bp cut site for each of these enzymes is also noted in Table 3.6. Although more than one restriction enzyme may have been a successful choice, all enzymes had at least two or more isolates with the same predicted T-RFL. Based on good location of cut sites and availability, HhaI was chosen to be evaluated further.

Table 3.5: Isolates used in the multi-species drinking water biofilm model. Strain ID Genus/ species identification Sampling location Opportunistic Pathogens

MEC-8 Escherichia coli Lennox and Addington County, private (raw) well MTC-21 Enterobacter cloacae Peterborough County, treated municipal distribution system PAE-1 Pseudomonas aeruginosa ACH FH-W1 Stenotrophomonas maltophilia FMC, PCU61, Patient room LPR-1 Legionella pneumophila RMC Normal Distribution System Flora

A3-1 Variovorax sp. ACH, Nourishment center A3-2 Cupriavidus respiraculi ACH, Nourishment center B1-1 Bradyrhizobium sp. ACH, PICU soiled utility C0-3 Sediminibacterium sp. ACH, Day surgery soiled utility C0-6 Kocuria rhizophila ACH, Day surgery soiled utility FH-D Novosphingobium subterraneum FMC, PCU61, Room sink FH- G Sphingomonas sp. FMC, GWHC, Utility room FH-J Blastomonas natatoria FMC, GWHC, Utility room MWI-1 Methylobacterium isbiliense UofC, BioSci MWI-2 Mycobacterium sp. UofC, BioSci The 10 normal distribution system flora comprised the base biofilm to which one of the opportunistic pathogens or indicator organisms was added. Only one opportunistic pathogen was present in each biofilm.

Table 3.6: Theoretical predicted TRFLs for each isolates based on the in silico digests of sequences in Appendix III with the restriction enzymes listed below. Strain ID Possible Restriction Enzyme Choices and respective cut sites HhaI MboI AluI HpaII RsaI HaeIII GCG_C _GATC AG_CT C_CGG GT_AC GG_CC Opportunistic Pathogens MEC-8 Escherichia coli 325 225 27 448 379 158 MTC-21 Enterobacter 325 225 27 448 379 158 cloacae FH-W1 Stenotrophomonas 165 148 27 415 431 211 maltophilia Normal Distribution System Flora A3-1 Variovorax sp. 159 212 102 442 381 152 A3-2 Cupriavidus respiraculi 19 219 107 382 425 29 B1-1 Bradyrhizobium sp. 747 141 162 104 62 147 C0-3 Sediminibacterium sp. 844 132 237 45 68 165 C0-6 Kocuria rhizophila 429 836 19 237 413 186 FH-D Novosphingobium 35 207 216 102 398 23 subterraneum FH-G Sphingomonas sp. 35 193 202 102 374 23 FH-J Blastomonas natatoria 35 193 160 102 374 23 MWI-1 Methylobacterium 293 139 160 102 374 15 isbiliense MWI-2 Mycobacterium sp. 126 63 148 226 30 18

3.4.3 Establishing a library of actual terminal restriction fragment lengths for model multi-species biofilm

In order to determine the actual T-RFL for each strain, 16S rRNA amplified DNA from each individually grown organism was digested with HhaI. The actual T-RFL for each of the strains in the multi-species model when run individually, compared to the estimated value based on the in silico digest is given in Table 3.7. For all of the strains, the estimated value was consistently approximately 40-50 bp less than the actual measured

T-RFL. Figure 3.3 shows the peak position, corresponding to a particular fragment size in base pairs, for each of the strains overlaid on the same chromatogram. The figure, as well as the data in Table 3.7, indicates that in a mixed species biofilm, some of the strains may be grouped as the same fragment size and would therefore be indistinguishable by the assay. This occurs for the more closely related strains such as Sphingomonas,

Blastomonas and Novosphingobium species (fragment size all ~76 bp), E. coli and E. cloacae (370 bp and 369 bp respectively) and the S. maltophilia and L. pneumophila (209 bp). As the opportunistic pathogens would only be run individually within the multi- species model, an overlap in the binned size would not matter as only one of the strains would be present at a time. For the Sphingomonadacea family isolates, DNA

was combined from the three isolates, amplified, digested and underwent fragment analysis. The three species formed one peak with a clear maximum, rather than three separate but condensed peaks (Figure 3.4). For this reason, T-FRLP multi-species biofilm results would not be able to distinguish the three species separately but would be able to give an accurate value for the total abundance of Sphingomonadacea present in the

Table 3.7: Actual versus estimated terminal restriction fragment lengths Strain ID Genus/ species identification Estimated TRFL Measured TFRL with HhaI (bp) with HhaI (bp) Opportunistic Pathogens MEC-8 Escherichia coli 325 369.5808 MTC-21 Enterobacter cloacae 325 368.8352 PAE-1 Pseudomonas aeruginosa - 150.6104 FH-W1 Legionella pneumophila 165 209.3944 LPR-1 Stenotrophomonas maltophilia - 209.1514 Normal Distribution System Flora A3-1 Variovorax sp. 159 201.1295 A3-2 Cupriavidus respiraculi 19 59.8996 B1-1 Bradyrhizobium sp. 747 789.6171 C0-3 Sediminibacterium sp. 844 909.5890 C0-6 Kocuria rhizophila 429 472.9894 FH-D Novosphingobium 35 76.0849 subterraneum FH-G Sphingomonas sp. 35 76.1129 FH-J Blastomonas natatoria 35 76.1020 MWI-1 Methylobacterium isbiliense 293 337.1035 MWI-2 Mycobacterium sp. 126 171.4845

Figure 3.3: Combined Peak Scanner data of terminal restriction fragment lengths of 15 species. The top X-axis represents fragment length in base pairs and the peak height along the y- axis shows the relative abundance of each fragment. Each peak is labelled with its terminal restriction fragment length.

Figure 3.4: Peak Scanner data for the family Sphingomonadaceae A: Chromatogram of the three Sphingomonadaceae: Novosphingobium sp FH-D,

Sphingomonas sp. FH-G, and Blastomonas sp. FH-J, run individually overlaid on top of one another. B: Chromatogram of the combined DNA of the three species mentioned above, amplified, digested and analyzed showing a single peak.

80 sample. The results presented in Table 3.7 and Figure 3.3 form the basis of the T-RFLP library used to analyze experimental mixed species biofilms. The results and discussion of these analyses are presented in Chapter 5, Section 5.2.3.

3.5 Summary

Close to 100 organisms were collected from drinking water systems in Calgary and

Ontario, from regulated water systems, both residential, institutional and hospital environments. The group contains environmental organisms including normal flora as well as opportunistically pathogenic organisms. Of these isolates, 29 were identified by

16S rRNA gene sequencing. The majority of strains were identified as belonging to the phyla Proteobacteria (alpha, beta and gamma classes), as well as four isolates from the

Gram positive phylum Actinobacteria. There were overlapping identification within and between sampling sites and so one organism was chosen from each group on a phylogenetic tree to represent it in the multi-species model. A T-RFL library was constructed for the model by performing an in silico digest of the sequences with a number of restriction enzyme cut sites and then finding the actual terminal restriction fragment length for each organism using one restriction enzyme, HhaI.

Chapter Four: Isolate growth under biofilm forming conditions and planktonic and biofilm susceptibility to chlorination

4.1 Aims

In order to gain more insight into each of the isolates used, as well as to select appropriate strains to incorporate into the multi-species biofilm model, it was necessary to know the biofilm growth ability, characteristics and susceptibility to chlorination of each of the organisms under consideration. With the goal of accomplishing this, the following aims were undertaken:

• Evaluate biofilm growth of the isolates on the Calgary Biofilm Device (CBD) and

asses which organisms would be suited for use within the multi-species biofilm

model.

• Test the susceptibility of organisms to chlorine in both the planktonic and biofilm

growth states.

4.2 Biofilm growth

Based on observation of growth of the isolates on minimal recovery medium agar plates, it was recognized that different organisms would require considerably different durations of incubation (<24 hours to over 1 week) in order to produce biofilms of approximately the same viable cell density, or colony forming units per peg (CFU/peg). A time course experiment was run for each normal flora organism where the viable cell density was measured at regular time intervals for each organism. This was also done to ensure that

82 all strains would be able form biofilms in the Calgary biofilm device under the conditions tested.

Twelve of the isolates received from the Infection Prevention and Control (IPC) laboratory at Foothills Medical Center were inoculated individually into the Calgary

Biofilm Device (CBD) and viable cell counts (CFU/peg) were evaluated after 24, 48 and

72 hours of incubation. Four of the strains (FH-A, FH-J, FH-P, FH-W1) had growth above 10 5 CFU/peg at 24 hours of growth, and stayed relatively constant over the three

time points. FH-K also stayed relatively constant throughout the incubation period at

approximately 10 4 CFU/peg (Figure 4.1A). FH-G and FH-J1 both increased gradually throughout the incubation by one and two log units, respectively. Two isolates, FH-D and

FH-Q1 both had low CFU/peg initially and increased dramatically (~4 log units) by 48 hours, and then decreased slightly at 72 hours. The isolate FH-Y was not detectable at 24 hours and was only present at ~10 3 at 48 and 72 hours (Figure 4.1B).

Four strains isolated from the Alberta Children's Hospital (ACH) (A3-1, A3-2, C0-3 and

C0-5) were grown on the CBD for 48, 72 and 96 hours. By the first time point A3-2 had

already reached 10 7 CFU/peg and remained approximately constant over the next two time points. C0-5 similarity remained constant over the entire time course between 2.7 and 8.3x10 4 CFU/peg. By the 48 hour time point A3-1 had reached 1.3x10 5 CFU/peg,

which increased slightly at 96 hours to 5.3x10 5 CFU/peg. Initially, C0-3 had much lower

growth than the other three strains, but by 72 and 96 hours, it had reached the same level

as A3-1 and C0-5 (Figure 4.2A). C0-8 was a much slower growing strain and therefore

Figure 4.1: Mean viable cell count over time for Foothills Medical Centre isolates Mean viable cell counts over 72 hours for biofilms of ten bacterial isolates obtained from the Infection Prevention and Control laboratory at Foothills Medical Centre. A) Five isolates (FH-A, FH-J, FH-K, FH-P and FH-W1) that had only a small or no change in mean viable cell counts (CFU/peg) over the 72 hour period. B) Remaining five isolates

(FH-D, FH-G, FH-Y, FH-J1 and FH-Q1) which either showed an increase and/or a dramatic change in mean viable cell counts (CFU/peg) over the time course (n=3, error bars represent the standard deviation).

Figure 4.2: Mean viable cell count over time for Alberta Children's Hospital isolates Mean viable cell counts over a time course of incubation for biofilms of five bacterial isolates obtained from biofilm sampling at the Alberta Children's Hospital.

A) Mean viable cell count (CFU/peg) of four isolates (A3-1, A3-2, C0-3 and C0-5) examined over a 96 hour time course (n=6). B) Remaining isolate (C0-8) was a much slower growing organism and mean viable cell count (CFU/peg) was examined at 9 and

12 days after inoculation (n=24) (error bars represent standard deviation).

87 was evaluated at 9 and 12 days growth. At the first time point growth had reached

4.4x10 4 CFU/peg, and then decreased slightly at the 12 day time point to 2.0x10 4

CFU/peg (Figure 4.2B).

MWI-1 and MWI-2, isolates obtained from the University of Calgary Biological Sciences

building water system, were grown in the CBD for 14 days and CFU/peg was evaluated

at days 6 through 14. For MWI-1 initial biofilm counts at day 6 were 1.5x10 5 CFU/peg and other than a spike in growth at day 11 to 1.2x10 6 CFU/peg the growth remained

constant. At day 6 the initial biofilm counts for MWI-2 were 7.1x10 5 CFU/peg, but from

there dropped sharply at day seven, and then gradually increased to 4.0x10 6 CFU/peg by

day 14 (Figure 4.3). Slight fluctuations in the CFU values may be due to the media

change schedule for the experiment, which occurred every 48 hours.

Based on the above results, optimal incubation times and media replenishing schedule

were determined for each isolate (Table 2.1) and used for further analysis of the single

species biofilms.

4.2.1 Non-biofilm forming isolates

Although the majority of isolates evaluated were able to form biofilms on the CBD, there

were a small number of isolates which were unable to establish single species biofilms

under the conditions tested. Of the 12 isolates received from the IPC Laboratory and

inoculated into the CBD, two isolates were unable to form biofilms under the conditions

tested, FH-B1 and FH-M1. Both were evaluated at 24, 48 and 72 hours. FH-B1 had

Figure 4.3: Mean viable cell count over time for University of Calgary isolates Mean viable cell count over a time course of 14 days for biofilms of two organisms

(MWI-1 and MWI-2) isolated from the water system of the University of Calgary

Biological Sciences building water system (n=6, error bars represent standard deviation).

89 viable cells in the inoculum but no cells were recoverable from the pegs after 72 hours of growth. FH-M1 was not detectable in the inoculum or from the biofilm pegs over the 72 hours. One of the isolates from the ACH (C0-7) was also unable to establish a biofilm under the tested conditions. After 72 hours of growth, cells were only recoverable from

~44% of the CBD pegs (mean CFU/peg 2.0x10 2 ± 2.9x102). After 96 hours of incubation no cells were recoverable from the pegs. Although these three isolates: FH-B1, FH-M1 and C0-7 were unable to form biofilms in the CBD under the tested conditions, they may have the ability to form biofilms under other circumstances. Even so, they were excluded from further analysis in this study.

4.3 Biofilm equivalency

In order for the minimum biofilm eradication concentration (MBEC) assay to accurately and reproducibly predict the susceptibility of biofilm organisms, it is necessary that all the biofilms in the CBD are grown to an equivalent viable cell density (CFU/peg). To ensure that growth on each peg was equal across the CBD, a growth assay was performed and statistical analysis done to determine if the viable cell count was statistically different between each of the rows and each of the columns on the device, as described in Section

2.4.2. All opportunistic pathogen strains were found to form statistically equivalent biofilms across the pegs in the CBD. Figure 4.4A shows the average CFU/peg across half of the CBD plate for all of the five isolates. All values fall within the range of 2.0x10 5

(MEC-8) to 7.4 x10 6 (MTC-21) CFU/peg. Of the normal flora isolates, the majority formed equivalent biofilms in the CBD with the exception of C0-5 and C0-7 (not shown).

Both were found to have low and inconsistent growth on the device and statistical

Figure 4.4: Mean viable cell count of each isolate in a single-species biofilm Mean viable cell count (CFU/peg) on the Calgary Biofilm Device in colony forming units per peg for A) opportunistic pathogen isolates and B) normal drinking water flora isolates

(n=48; error bars represent standard deviation; A3-1 = Variovorax sp.; A3-2 =

Cupriavidus respiraculi; B1-1 = Bradyrhizobium sp; C0-3 = Sediminibacterium sp.; FH-

D = Novosphingobium subterraneum ; FH-G = Sphingomonas sp.; FH-J = Blastomonas natatoria ; MWI-1 = Methylobacterium isbiliense; MWI-2 = Mycobacterium sp.).

92 differences between the growth on pegs between each of the rows and columns. For the isolate C0-6, despite the fact that the measured values ranged between 8.0x10 2 to 2.6x10 6

CFU/peg, when the statistical tests mentioned above were applied to data, values for mean CFU/peg were found to be not statistically different across the CBD. For this reason it was found that the growth of the organisms in the CBD was inconsistent and therefore would not be considered equivalent. The remaining 9 normal flora strains were found to have equivalent biofilm formation and average CFU/peg ranged from 3.0x10 4

(B1-1) to 4.2x10 6 (A3-2) CFU/peg (Figure 4.4B).

4.4 Biofilm susceptibility

Once it was established that isolates were able to form equivalent biofilms on the CBD under the tested conditions, evaluation of biofilm and planktonic susceptibility to chlorination was performed. Chlorination of single species cultures was evaluated at relatively low levels, approximately equal to those present within water distribution systems. As described in Section 1.2.3, low levels of free chlorine are maintained throughout drinking water distribution systems, referred to as residual chlorine disinfection, or secondary disinfection. The levels of chlorine used in these susceptibility experiments were originally designed to cover a slightly larger range of residual chlorine levels than are stipulated in regulatory legislation by federal and provincial governments.

The Alberta Government regulates that residual chlorine levels must be between 0.1 and

4.0 mg/L of free chlorine (Alberta Environment Drinking Water Branch 2006). The experiment was first attempted with levels of free chlorine between 0.0156 and 16 mg/L exposed for 10 minutes. It became evident that higher concentrations of chlorine would

93 be required for complete biofilm disinfection for a number of strains, as described below, and the maximum chlorine dose was increased to 128 mg/L with the same exposure time.

The level of chlorine required to eliminate a population of planktonic bacteria versus that required to eliminate biofilm bacteria for the five opportunistic pathogen strains is shown in Figure 4.5A. Consistently, for all isolates minimum inhibitory concentrations (MIC) of chlorine for planktonic cells fell within the range of limits set by provincial and federal legislation for residual chlorination, as would be desired and expected. Values were in the center of this range, between 0.8125 mg/L (PAE-1) to 1.0mg/L (FH-W1 and LPR-1) chlorine to eliminate the population. Alternatively, MBEC values for these strains fell either just within or higher than the legislated maximum level of chlorine permissible.

Biofilms of both E. coli MEC-8 and L pneumophila LPR-1 were able to be eradicated with free chlorine levels below 4mg/L (3.25 and 3.19 mg/L respectively). Although still within the legislated range, the biofilms were both able to survive chlorine levels greater than 3 fold higher than their planktonic counterparts. The remaining three organisms required greater than twice the legislated maximum level of chlorine to eliminate the biofilm population, up to 27 mg/L for S. maltophilia FH-W1.

Results from the MIC and MBEC assays on normal flora isolates followed a similar pattern as discussed above. All MIC values fell within the middle range of the legislated allowable chlorine concentrations, between 1.0 and 2.0 mg/L free chlorine (Figure 4.5B).

As well, when grown as biofilms, only 2 of the nine isolates were able to be eradicated using chlorine levels lower than the maximum 4.0 mg/L (3.25 mg/L for B1-1 and 2.59 mg/L for C0-3). The difference between the planktonic and biofilm survival was the

Figure 4.5: MBEC and MIC values for single-species biofilms Minimum biofilm eradication concentration (MBEC) compared to the minimum inhibitory concentration (MIC, planktonic) for single species biofilms of A) opportunistic pathogen isolates and B) normal drinking water flora isolates. Shaded area represents the legislated range of allowable residual chlorine concentrations in drinking water distribution systems in Alberta (0.1-4.0 mg/L free chlorine) (n=14; error bars represent standard deviation; A3-1 = Variovorax sp.; A3-2 = Cupriavidus respiraculi; B1-1 =

Bradyrhizobium sp; C0-3 = Sediminibacterium sp.; FH-D = Novosphingobium

subterraneum ; FH-G = Sphingomonas sp.; FH-J = Blastomonas natatoria ; MWI-1 =

Methylobacterium isbiliense; MWI-2 = Mycobacterium sp.).

96 smallest for B1-1, at only 1.6 fold greater. The remaining seven isolates were able to survive chlorine levels above the maximum legislated limit. The three isolates from the family Sphingomonadaceae (FH-D, FH-G, FH-J) all had MBEC values well above

(greater than 3.5 fold) the legislated maximum, as did A3-1. MWI-1 biofilms had by far the highest survival level, and required 52 mg/L free chlorine (13 times the maximum level and greater than 40 fold higher than its planktonic counterpart). The overall trend over both Figures 4.5A and 4.5B showed that the biofilm mode of growth consistently allowed the bacterial populations to survive free chlorine concentrations higher than those tolerable than their planktonic counter parts, between 1.6 to 40 fold, and in many cases higher than the legislated maximum value of 4 mg/L.

4.5 Summary

Initially, the ability of an organism to form a biofilm in the Calgary Biofilm Device under the experimental conditions and the amount of growth time required was established for

21 isolates. Three of these (FH-B1, FH-M1, C0-7) were unable to form biofilms in the

CBD and another two (C0-5, C0-6) were unable to form equivalent biofilms across the device. Growth time for single species biofilms in the CBD ranged from 20 hours ( P aeruginosa PAE-1) to 9 days ( Bradyrhizobium sp. B1-1). The minimum inhibitory concentration (MIC, planktonic cells) and minimum biofilm eradication concentration

(MBEC, biofilm cells) of chlorine was found for five opportunistic pathogen isolates and nine heterotrophic organisms. All planktonic cells were killed by chlorine concentrations lower than 4.0 mg/L which is the maximum allowable concentration according to provincial legislation. Biofilm populations of bacteria all required higher chlorine

97 concentrations in order to be eradicated than their planktonic counterparts, between 1.6 and 40 fold greater ( Bradyrhizobium sp. B1-1 and Methylobacterium isbiliense MWI-1,

respectively). Of the 14 isolates tested, only four biofilms were eradicated by 4.0mg/L

chlorine or less ( E. coli MEC-8, L. pneumophila LPR-1, Bradyrhizobium sp. B1-1 and

Sediminibacterium sp. C0-3).

Chapter Five: Multi-species biofilms: growth, structure, chlorine susceptibility and community dynamics

5.1 Aims

The majority of environmental biofilms are colonized by greater than one species and therefore, it would give a much more accurate representation of a true distribution system biofilm to culture isolated organisms as a mixed-species biofilm model. It would also allow the observation of any advantages that may exist for these species growing as a community. To this end the following aims were identified for this portion of the project:

• Create a representative mixed species community of normal flora microorganisms

in a reproducible manner and inoculate this normal biofilm with different

opportunistic pathogens, including Escherichia coli, Enterobacter cloacae,

Pseudomonas aeruginosa, Stenotrophomonas maltophilia and Legionella

pneumophila

• Examine the susceptibility of the community of bacteria and the population of an

opportunistic pathogen within that community to chlorination and the effect on

community dynamics using different methods including the MBEC (minimum

biofilm eradication concentration) assay, microscopy and a T-RFLP (terminal

restriction length polymorphism) assay

• Identify other possible biofilm growth models that could be used to further

confirm the results obtained using the CBD

5.2 Results

5.2.1 Selecting a multi-species biofilm model inoculation schedule

There were a number of challenges and goals that needed to be addressed in order to design a representative multi-species biofilm that could be used to study the effects of chlorine exposure on the community. These included allowing each isolate to establish and grow, without one organism outgrowing the rest, having a reproducible inoculation schedule, and being able to incorporate different opportunistic pathogens into the established biofilm.

In order to overcome these challenges three different inoculation schedules were tested, as described in Section 2.4.1 and in Table 2.2. The three trial schedules were incubated under the same conditions as single species biofilms and DNA was extracted from the biofilm and analyzed by terminal restriction fragment length polymorphism (T-RFLP) assay at 6 and 12 days. This allowed enough time for species to grow and establish themselves within the biofilm. The first of these inoculation schedules (A) was the simplest, inoculation of all the strains except for E. coli MEC-8 on the same day and incubate for 4 days prior to inoculating E. coli MEC-8. For the second inoculation schedule (B), each strain was added into the multi-species biofilm with the same number of days remaining as would normally take for its single species biofilm to grow to confluence (i.e. strains were added such that at the time of recovery all species would be reaching their maximum growth level. The third schedule (C) involved grouping the strains into slow, medium and fast growing biofilm groups based on single species biofilm growth times and each group was inoculated sequentially into the multi-species

100 biofilm. In order to establish if all species were equally represented, T-RFLP was used to identify which species were present within the biofilm and if any one strain was specifically dominating or overgrowing the rest. Figure 5.1 demonstrates the species diversity of each of the three biofilms at day 6 and at day 12 for the second two inoculation schedules. The 12 day time point for the schedule (A) biofilm had become visibly overgrown with the fastest growing isolates. It should be noted that this test was done relatively early in the project, and some small changes were made to the final group of species in the model multi-species biofilm. Initially B1-1 was not selected to be part of the multi-species biofilm as it was such a slow growing strain and at the time it was unknown if it would be able to form a biofilm in the CBD, and was therefore not included in this test. In later biofilms B1-1 was inoculated first and allowed to grow for three days before the addition of other species, and was part of the biofilm for a total of 9 days, the same amount of time as its single species biofilm. FH-A was initially included in this test but after analysis of the sequencing data (Section 3.3.2) it was observed that it was closely related to the A3-1 isolate, which was chosen to represent the group. Finally, this biofilm included both opportunistically pathogenic isolates E. coli MEC-8 and S. maltophilia FH-W1 and it was later decided for clarity of the study, to include only one opportunistic pathogen at a time in the model biofilm. The final inoculation selected was closets to that in schedule B. Slight changes in the final inoculation order reflect the items discussed above as well as altering the inoculation of C0-6 into the biofilm to 48 hours before harvest, rather than 72 hours, as in schedule B, in order to prevent biofilm overgrowth with the organism.

Figure 5.1: Terminal restriction fragment length polymorphism assay of three inoculation schedules Terminal-restriction fragment length polymorphism normalized data for three multi-species biofilm inoculation schedules. Treatment

A is shown after 6 days of growth, and treatments B and C are shown after 6 and 12 days of growth (n=3). 101

102

5.2.2 Scanning electron microscopy of model drinking water system biofilm

Scanning electron microscopy was performed on the model normal flora biofilm (no opportunistic pathogen added), formed on a CBD peg (Figure 5.2A and B) in order to compare its structure to that of a native water system biofilm (Figure 5.2C and D).

Although the fixation process does dehydrate the extracellular polymeric substance (EPS) and therefore changes the appearance of the native biofilm to some degree, it is possible to discern a number of interesting features in the biofilm. The first of these is the distinct, dense structures formed in some sections of the model biofilm, while other areas show thinner, monolayer growth. It is also possible to identify a number of different colony morphologies including cocci, large thick rods and much thinner, tightly packed bacilli.

Although in some areas the distinct cell morphologies are grouped together, there is some mixed species contact. Another visible feature of the biofilm is what may possibly be a layer of EPS coating the cells, notably the area with long, thin bacilli in figure 5.2A, and on the microcolonies in figure 5.2B. There also are what appear to be cracks in the layer on figure 5.2A, that may have been caused by dehydration during the fixing process.

These images can be compared to figures 5.2C and D (Images courtesy of Carol

Stremick, unpublished data) which are native biofilms from the water distribution system in the Alberta Children's Hospital, formed on brass coupons which had been in place for approximately 4-5 years. The most striking difference between the two sets of biofilms

(native and model) is the presence of materials and mineralization crystals from the water deposited on the surface of the coupons, which is not unexpected in a distribution system.

There is also variation in the coverage of the biofilms on the coupons between figure 5.2

C and D, the biofilm in 5.2C appears to have consistent growth across the area scanned

103

104

105

Figure 5.2 Scanning electron microscopy of model and native drinking water system biofilms A and B: Scanning electron micrographs (SEM) of model drinking water system biofilm with no opportunistically pathogenic organisms inoculated, after 9 days of growth on the

Calgary Biofilm Device peg at 5000x and 3000x magnification respectively. C and D:

SEM of two native drinking water system biofilms grown on brass coupons obtained from the water system at the Alberta Children`s Hospital after 4-5 years of growth at

10000x and 5000x magnification respectively. (Figures 5.2C and D courtesy of Carol

Stremick, unpublished data)

106

whereas the organisms in 5.2D are more dispersed and form a few separate clusters. The

other difference between the native and model biofilms is the visibility of distinct cell

morphologies which are immediately evident in the model biofilms but not as discernible

in the native biofilms. Despite these differences, some similarities were observed

between the model and native biofilms, most significant to this study would be the

formation of 3D microcolony structures within the all four of the biofilm images.

5.2.3 Multi-species biofilm chlorine exposure

5.2.3.1 Escherichia coli MEC-8

Of the five opportunistic pathogen strains, E. coli (MEC-8) had the lowest single species

biofilm chlorine tolerance, although, as discussed in Section 4.4, it was still over 3 fold

greater than the planktonic population susceptibility. When MEC-8 was incorporated into

the model multi-species biofilm, the organism was found to be nearly 200 fold more

resistant to chlorine disinfection than the single species biofilm (Figure 5.3). The multi-

species biofilm survival was examined using three media types. Firstly the survival of the

opportunistic pathogen in question was evaluated with a selective agar media, in this case

Differential Coliform agar, which prevented the growth of organisms other than E. coli .

The concentration of chlorine required to reduce MEC-8 to below detectable levels (<5

CFU/peg) using this media was 643 (±43) mg/L. The second method of evaluating opportunistic pathogen survival was by using a differential broth media, Presence

Absence (PA broth). Using this method, MEC-8 was detectable (1CFU/peg) at chlorine concentrations up to 928 (±31) mg/L. This is over 230 times the legislated maximum level of chlorine permitted in a distribution system. The biofilm was also tested for the

107

Figure 5.3: Single and multi-species minimum biofilm eradication concentration assays with Escherichia coli Minimal biofilm eradication concentration (MBEC) assay of the model multi-species

biofilm containing Escherichia coli MEC-8. Chlorine concentrations (mg/L) required to eliminate single-species planktonic (minimum inhibitory concentration, MIC assay) and biofilm cultures (MBEC assay) of E. coli alone are also shown for comparison. Grey

shaded region indicates the range of chlorine concentrations required in drinking water

distribution systems in Alberta (0.1-4.0 mg/L chlorine) (n=14, error bars represent

standard deviation).

108

survival of the normal flora to high level chlorination by plating on R2A, the non-

selective low nutrient medium. In the MEC-8 multi-species biofilm the normal flora were

eradicated only after treatment with 1000 (±74) mg/L chlorine, slightly lower than the

control normal flora biofilm, which required 1333 (±56) mg/L chlorine. The same three

assays were performed on the other multi-species biofilms as discussed in the following

sections.

In addition to chlorine challenge survival MBEC assays, multi-species biofilms were

examined using confocal laser scanning microscopy (CLSM) with a live/dead stain in

order to evaluate which areas of the biofilm were most affected by the chlorine. Biofilms

were examined after treatment with low levels of chlorine (16 mg/L) as well as untreated

controls. It was found that high chlorine treatment (500 mg/L) actually removed a large

portion of the biomass from the peg and was therefore not as useful in visualizing the

pattern of disinfection and survival on the peg. Figure 5.4A shows a cross section image

of the model multi-species biofilm, inoculated with E. coli MEC-8, before any chlorine treatment. Although some areas of dead cells (red) are evident, the majority of the cells are alive (green). A number of different cell types and microcolonies are also evident, including cocci grouped into tetrads, larger groups of cocci which have formed into less organized circular patterns, and bacilli that appear to be growing dispersed around the microcolonies. When treated with chlorine, as shown in Figure 5.4B, the majority of cells appear as red, and much of the diversity is lost. It is important to note in this image that some microcolonies are still present and the live cells that are visible on the pegs appear to be contained within these groups, Limitations of the assay do not allow the

109

110

Figure 5.4: Confocal scanning laser microscopy of model multi-species biofilm with Escherichia coli Confocal scanning laser microscopy images of a multi-species biofilm inoculated with E. coli MEC-8 and stained with a live/dead (BacLight bacterial viability kit, Invitrogen) stain . A) The biofilm with no chlorine treatment. A variety of different cell types are visible, along with micro-colony formation (indicated on figure). B) The biofilm after treatment with 16 mg/L of chlorine for 10 minutes. Much of the cell morphology diversity seen in 5.4A is lost, but live cells (green) are still visible and appear to have been protected within the micro-colonies.

111 identification of which isolates are represented in the group of surviving cells and whether or not they are opportunistic pathogens.

The terminal-restriction fragment length polymorphism (T-RFLP) data for the model multi-species biofilm with MEC-8 does not show a great amount of change in the community dynamics within the biofilm (Figure 5.5) other than the fact that the cumulative peak height, which directly relates to the total amount of DNA in the biofilm, is much lower in the high chlorine treated biofilm, and there is little change between the other two treatments. One notable difference between the non- treated biofilm and the low chlorine (50 mg/L) treatment is that DNA from C0-3 ( Sediminibacterium sp.) is no longer detectible. Additionally, a slight increase in the proportion of MEC-8 DNA is seen after treatment. This may be explained by the fact that samples for the treated biofilms are taken after a 24 hour recovery time, rather than right after treatment, when the control biofilm is sampled. This is in order not to extract DNA from cells that are killed but still present within the biofilm.

5.2.3.2 Enterobacter cloacae MTC-21

E. cloacae MTC-21 was the next opportunistic pathogen indicator strain to be added to the model multi-species water system biofilm. Survival of chlorination of MTC-21 planktonically, as a single species biofilm, and as a member of multi-species consortium is displayed in Figure 5.6. Similarly to E. coli MEC-8, the survival of the isolate when incorporated into a multi-species biofilm dramatically increased. As a single species biofilm MTC-21 was able to survive up to 8.5 (±1.2) mg/L of chlorine, already over

Figure 5.5: Terminal restriction fragment length polymorphism assay of model multi-species biofilm with Escherichia coli Terminal-restriction fragment length polymorphism data for a multi-species biofilm inoculated with E. coli MEC-8. Each bar represents the average cumulative peak area for that treatment, either 0, 50 or 500 mg/L of chlorine, and is divided to show the proportion of each isolate within the biofilm (n=3). 112

113

Figure 5.6: Single and multi-species minimum biofilm eradication concentration assays with Enterobacter cloacae Minimal biofilm eradication concentration (MBEC) assay of the model multi-species biofilm containing Enterobacter cloacae MTC-21. Chlorine concentrations (mg/L) required to eliminate single-species planktonic (minimum inhibitory concentration, MIC assay) and biofilm cultures (MBEC assay) of E. coli alone are also shown for

comparison. Grey shaded region indicates the range of chlorine concentrations required

in drinking water distribution systems in Alberta (0.1-4.0 mg/L chlorine) (n=14, error

bars represent standard deviation).

114 twice the legislated maximum limit, but when incorporated into the model biofilm, the survival increased 97 fold (when evaluated with selective media plating on DC agar) and

116 fold (with PA broth) to 821.4 (±31.3) mg/L and 982.1 (±41.1) mg/L chlorine, respectively. The normal flora strains in this biofilm were able to survive up to the same chlorine concentrations, 1357.1 (±43.2) mg/L, as the control multi-species biofilm, 1333

(±56) mg/L).

Confocal scanning laser microscopy (CSLM) images of the untreated MTC-21 multi- species biofilm, like the images for MEC-8, contain mostly live cells and a number of cell morphologies are discernible (Figure 5.7A). It is also possible to see via the cross section of the biofilm that the microcolonies give the biofilm a 3D structure and the cells appear more densely packed together within these groups. Figure 5.7B shows the treated biofilm. A large number of cells in the biofilm have been killed, but more of the cells within the microcolonies seemed to have survived than the surrounding cells.

T-RFLP results, similar to the MEC-8 multi-species biofilm, mainly show that there was significantly less DNA present in the samples after treatment with 1000 mg/L chlorine

(Figure 5.8). Untreated biofilms and those treated with the low chlorine dose (16 mg/L) appear very similar, and unlike the MEC-8 biofilm, DNA from C0-3 was still detectable after low chlorine treatment, although not after high chlorine treatment.

115

116

Figure 5.7: Confocal scanning laser microscopy of model multi-species biofilm with Enterobacter cloacae Confocal scanning laser microscopy images of a multi-species water flora biofilm

inoculated with E. cloacae MTC-21 and stained with a live/dead stain . A) The biofilm with no chlorine treatment. There is a 3D structure is visible within the biofilm as well as some micro-colony formation where the biofilm appears to be thicker than the surrounding area. B) The biofilm after treatment with 16 mg/L of chlorine for 10 minutes.

The main concentration of live cells appears to be within the micro-colonies.

Additionally some of the smaller cell clusters appear to be surrounded by red which could indicate either a layer of dead cells, or possibly DNA that is encased in the extracellular polymeric substance and surrounding the cells. Either of these may act as a barrier between the live cells and the chlorine.

Figure 5.8: Terminal restriction fragment length polymorphism assay of model multi-species biofilm with Enterobacter cloacae Terminal-restriction fragment length polymorphism data for a multi-species biofilm inoculated with E. cloacae MTC-21. Each bar represents the average cumulative peak area for that treatment, either 0, 16 or 1000 mg/L of chlorine, and is divided to show the proportion of each isolate within the biofilm (n=3). 117

118

5.2.3.3 Pseudomonas aeruginosa PAE-1

The third opportunistic pathogen to be incorporated into the model multi-species biofilm was P. aeruginosa PAE-1. The same trend of survival as was observed for E. coli MEC-8 and E. cloacae MTC-21 was seen for this isolate (Figure 5.9). The single species biofilm of PAE-1 required treatment with 13.0 (±1.5) mg/L free chlorine to be eradicated, slightly higher than both MEC-8 and MTC-21 (4 and 1.5 fold greater respectively). The multi- species biofilm required treatment with 607.1 (±43.2) mg/L chlorine to reduce PAE-1 survival to below detectable levels using plating on selective media, and 696.4 (±38.7) mg/L using PA broth detection, a 47 and 54 fold increase in survival, respectively, and

152 and 174 fold greater than the legislated limit. Similarly to MEC-8, the normal flora of this multi-species biofilm required treatment with 1017.9 (±48.8) mg/L chlorine to be eliminated, slightly lower than the control biofilm, 1333 (±56) mg/L chlorine).

Images of the untreated multi-species with PAE-1 biofilm show that it has a distinct 3D biofilm structure, which is especially evident in the cross section portion of the image

(Figure 5.10A). Multiple microcolony types are also visible. The treated biofilm, shown in figure 5.10B, still maintains the 3D structure seen in the untreated image, and a high number of live cells. It is also visible that within the microcolonies all the cells are living, as opposed to the surrounding areas of the biofilm.

The multi-species biofilm with PAE-1 showed the most interesting community dynamics of the multi-species biofilms evaluated using T-RFLP (Figure 5.11). After the low chlorine treatment (5 0mg/L) the proportion of PAE-1 present within the biofilm

119

Figure 5.9: Single and multi-species minimum biofilm eradication concentration assays with Pseudomonas aeruginosa Minimal biofilm eradication concentration (MBEC) assay of the model multi-species biofilm containing Pseudomonas aeruginosa PAE-1. Chlorine concentrations (mg/L) required to eliminate single-species planktonic (minimum inhibitory concentration, MIC assay) and biofilm cultures (MBEC assay) of P. aeruginosa alone are also shown for comparison. Grey shaded region indicates the range of chlorine concentrations required in drinking water distribution systems in Alberta (0.1-4.0 mg/L chlorine) (n=14, error bars represent standard deviation)..

120

121

Figure 5.10: Confocal scanning laser microscopy of model multi-species biofilm with Pseudomonas aeruginosa Confocal scanning laser microscopy images of a multi-species water flora biofilm inoculated with P. aeruginosa PAE-1 and stained with a live/dead stain . A) The biofilm with no chlorine treatment, which shows quite a bit of 3D structure, as is evident in the cross section cut away. There is again a variety in cell morphologies visible in the biofilm

B) The biofilm after treatment with 16 mg/L of chlorine for 10 minutes. Again, similarly to the other biofilms, the thicker micro-colony areas (as observed in the cross section of the biofilm images) contain the highest proportion of live cells.

Figure 5.11: Terminal restriction fragment length polymorphism assay of model multi-species biofilm with Pseudomonas aeruginosa Terminal-restriction fragment length polymorphism data for a multi-species biofilm inoculated with P. aeruginosa PAE-1. Each bar represents the average cumulative peak area for that treatment, either 0, 50 or 500 mg/L of chlorine, and is divided to show the proportion of each isolate within the biofilm (n=3). 122

123

increased by approximately two fold, whereas the amount of Sphingomonadaceae (FH-D,

FH-G and FH-J) decreased by almost five and a half fold. This may be due to the faster

growth rate of PAE-1 out-competing the other species in the biofilm after the low

chlorine challenge. When treated with high levels of chlorine (500 mg/L) however, the

proportion of PAE-1 present decreases dramatically to be closer to that of a number of

the other species within the biofilm, while the amount of DNA detected from the

Sphingomonadaceae is slightly higher than that after low chlorine treatment. One other

item to note about the multi-species and PAE-1 biofilm is that DNA from C0-3 is no

longer detected after treatment with 50 mg/L chlorine.

5.2.3.4 Stenotrophomonas maltophilia FH-W1

The final opportunistically pathogenic organism that was evaluated for chlorine susceptibility within a multi-species community was S. maltophilia FH-W1, and once

again the same trend was seen as with the previous three organisms. Of the four

opportunistic pathogen isolates, FH-W1 was able to survive the highest chlorine

concentrations, both as a single species biofilm (13.0±1.5 mg/L, equal to PAE-1) and

when incorporated into the model multi-species biofilm (Figure 5.12). As part of the

multi-species consortia, 964.3 (±51.5) mg/L of chlorine was required to reduce the

population of FH-W1 below detection levels for the selective media agar MBEC assay or

1089.3 (±33.2) mg/L for the PA broth assay. This shows that the isolate was able to

survive chlorine concentrations over 270 fold greater than the legislated maximum

allowable level. The normal flora organisms of this biofilm were also able to survive

124

Figure 5.12: Single and multi-species minimum biofilm eradication concentration assays with Stenotrophomonas maltophilia Minimal biofilm eradication concentration (MBEC) assay of the model multi-species biofilm containing Stenotrophomonas maltophilia FH-W1. Chlorine concentrations

(mg/L) required to eliminate single-species planktonic (minimum inhibitory concentration, MIC assay) and biofilm cultures (MBEC assay) of S. maltophilia alone are also shown for comparison. Grey shaded region indicates the range of chlorine concentrations required in drinking water distribution systems in Alberta (0.1-4.0 mg/L chlorine) (n=14, error bars represent standard deviation).

125 slightly higher chlorine concentrations than the control biofilm, 1500 (±58.6) mg/L versus 1333 (±56) mg/L of chlorine.

CLSM images of the untreated biofilm (Figure 5.13A) show that although the 3D structure of the biofilm is not as complex as seen in the multi-species with PAE-1 biofilm, there is still a variety of cell types present which appear to be packed tightly together. Some microcolonies even seem to be surrounded by a layer of red. Images from the treated multispecies biofilm with FH-W1 are taken from slightly lower on the peg, which causes the shape seen in Figure 5.13B. There are approximately equal amounts of live and dead cells visible and a number of different cell types are still visible, although it is interesting that within the microcolonies on the peg, there seems to also be equal chlorine disinfection, as opposed to the other treated biofilm images.

The T-RFLP data of the non-chlorine treated and low chlorine (16 mg/L) treated biofilms for the FH-W1 multi-species biofilm are very similar, although as seen in the previous three biofilms, the mean peak area of the high chlorine (1000 mg/L) treated biofilm is much lower than the other two treatments (Figure 5.14). Similarly to the MTC-21 multi- species biofilm, DNA from C0-3 is still detectible after treatment with low levels of chlorine (16 mg/L), but not after the high chlorine treatment. The multi-species biofilm with FH-W1 is unique from the other three biofilms discussed in that it has not overgrown the other species in the biofilm, and the C0-6 strain has increased to levels similar to that of the control biofilm (multi-species biofilm with no opportunistic pathogen added (Figure 5.15).

126

127

Figure 5.13: Confocal scanning laser microscopy of model multi-species biofilm with Stenotrophomonas maltophilia Confocal scanning laser microscopy images of a multi-species water flora biofilm inoculated with S. maltophilia FH-W1 and stained with a live/dead stain . A) The biofilm with no chlorine treatment. Although the biofilm layer is thinner than the PAE-1 multi- species biofilm, there is fairly consistent coverage across the visible area. Additionally, a variety of cell morphologies and micro-colonies are observed. B) The biofilm after treatment with 16 mg/L of chlorine for 10 minutes. The survival of cells within microcolonies is not as evident in the multi-species biofilm with FH-W1, although there does seem to be some variation in thickness of the biofilm across this area of the peg.

Figure 5.14: Terminal restriction fragment length polymorphism assay of model multi-species biofilm with Stenotrophomonas maltophilia Terminal-restriction fragment length polymorphism data for a multi-species biofilm inoculated with S. maltophilia FH-W1. Each bar represents the average cumulative peak area for that treatment, either 0, 16 or 1000 mg/L of chlorine, and is divided to show the proportion of each isolate within the biofilm (n=3). 128

Figure 5.15: Terminal restriction fragment length polymorphism assay of control biofilm Terminal-restriction fragment length polymorphism data for a model drinking water multi-species biofilm which has not been inoculated with any additional opportunistic pathogen strains as a control. Each bar represents the average cumulative peak area for that treatment, either 0, 50 or 500 mg/L of chlorine, and is divided to show the proportion of each species comprising that biofilm

(n=3) 129

130

5.2.4 Legionella pneumophila in the model multi-species biofilm

When performing single species growth assays and susceptibility tests for L. pneumophilia LPR-1 it became evident that the isolate would not grow on either the R2A agar plates, or in the R2B broth as a biofilm and the assays were eventually performed with BCYE agar and BYE broth. This posed a problem with the incorporation of LPR-1 into the model multi-species water biofilm. All of the other opportunistic pathogens or indicator organisms used in this study were able to grow relatively robustly on both R2A plates as well as form biofilms in R2B broth. Additionally, selective or differential media were available for each of these strains which would make colonies discernible from the background micro flora.

It was hypothesized that although the LPR-1 isolate was not able to grow in R2B in the

CBD alone, it may be able to incorporate into and survive (if not necessarily grow) within a previously established multi-species biofilm. In order to test this hypothesis, two selective media were tested to attempt to distinguish the LPR-1 colonies from the other isolates present within the multispecies consortia. The media used were BCYE PV

(BCYE supplemented with polymyxin B and Vancomycin) and BCYE PAC (BCYE supplemented with polymyxin B, cefamandole and anisomycin). Although the antibiotics were able to inhibit growth of most other species present, FH-D ( Novosphingobium subterraneum ) was still able to grow on both agar types. Despite this, a number of multi-

species inoculation and growth procedures were attempted. One of the changes in

inoculation procedure that was attempted was adding BYE broth during the inoculation

of LPR-1 after growing the initial biofilm in R2B. The second change was attempting to

131 inoculate LPR-1, either in BYE or R2B, at different time points in the growth of the multi-species biofilm, day 8 or 9. Each of these inoculations was incubated at 25°C, as usual as well as moving the biofilm to 37°C after LPR-1 inoculation. The first strategy was unsuccessful as it over-enhanced the growth of other species within the biofilm to the point that LPR-1 was undetectable. When LPR-1 was inoculated in R2B it didn't seem to change the proportion of the other organisms within the biofilm, although LPR-1 was not detected when the biofilm was removed and plated. The increased incubation temperature did not produce markedly different results than the 25°C incubation. To confirm this result T-RFLP analysis was performed on the biofilm grown in R2B at 25°C. Results were similar to the control biofilm with no added opportunistic pathogen, in that no fragments corresponding to LPR-1 were detected, although there was a slight difference in average relative peak area between the two samples (Figure 5.16).

5.2.5 Other multi-species biofilm methods investigated

Two alternative methods of growing multispecies biofilms were attempted during the course of this project. The first of these was attempted early in the course of the research and was a very straight forward method. A section of silicone tubing 30 cm in length was fixed to a sterilized lab tap and water was run through it for 25 days at a rate of 2.5

L/min. At the end of this time the tube was clamped at either end, filled with saline plus

1% Tween 20 and sonicated for 10 minutes. This sonicate was then plated onto LB plates. The tubing was also swabbed and this was plated on R2A. This method of biofilm growth and isolate recovery was unsuccessful, and no organisms were recovered.

Figure 5.16: Terminal restriction fragment length polymorphism assay of model multi-species biofilm inoculated with L. pneumophila Terminal-restriction fragment length polymorphism data for a multi-species biofilm inoculated with L. pneumophila and grown in

R2B at 25°C. L. pneumophila was not detected in the biofilm and therefore not included in the legend (n=3). 132

133

The second attempt of alternative multi-species biofilm growth was using a flow through

biofilm reactor. The method involved pumping R2B medium through silicone tubing at a

slow rate (1 mL/min) and sequentially inoculating each organism (or group of organisms)

using a syringe, into the growth tube. Although initially this method seemed promising

and growth was visible within the silicone tubes there were two significant problems.

Very large volumes of medium were required to maintain the system over the 10 day

incubation time, requiring multiple medium changes which exposed the system to

contamination. Additionally, it was very difficult to maintain sterility and ensure that no

unwanted organisms were added to the system over multiple inoculations. For these

reasons, it was decided that this method was not feasible without a significant amount of

optimization which unfortunately was not possible in the time allotted to this project.

5.3 Summary

Using the organisms selected and the T-RFLP library created in Chapter Three, an

inoculation schedule was selected to culture the model multi-species water system the

biofilm. Scanning electron microscopy was performed on the model biofilm to compare it

to a natural water system biofilm which showed that more distinct cell morphologies

were visible, but no mineral deposits were present on the model biofilm. The model water

system biofilm was inoculated with five different opportunistic pathogens, each added to

a separate biofilm rather than mixed together, and the biofilms were then exposed to

chlorine. The results were analyzed by viable cell recovery, confocal laser scanning

microscopy (CLSM) and T-RFLP. The trend seen was similar for E. coli, E. cloacae, P. aeruginosa and S. maltophilia. Viable cell recovery using Presence-Absence broth

134

showed that within the biofilm community the population of the respective opportunistic

pathogen was anywhere from 54 to 286 fold more resistant to chlorine disinfection than

in a single-species biofilm ( P. aeruginosa and E. coli respectively). Confocal laser

scanning microscopy demonstrated the formation of microcolonies within the biofilms

and that when challenged with chlorine often the cells within the microcolonies had more

survival than the surrounding cells, with the exception of S. maltophilia, which had

approximately equal killing across the biofilm. T-RFLP of the challenged biofilms

showed that all organisms but one were present in the biofilm even after treatment with

high levels of chlorine ( ≥ 500 mg/L). Sediminibacterium sp. C0-3 was eliminated from

the biofilm between 16 and 50 mg/L chlorine. Additionally, the amount of DNA

recovered and therefore the number of cells present was much lower for biofilms treated

with high doses of chlorine. L. pneumophila was not able to be cultured from the multi- species biofilm or identified as persisting in it using T-RFLP, despite a number of different culture techniques attempted. In addition to the Calgary Biofilm Device, two additional multi-species biofilm cultivation methods were attempted. Both were based on continuous flow, but unfortunately neither was successful.

135

Chapter Six: Discussion, conclusions and future directions

6.1 Summary of results

In section 1.1 four aims for the project were presented. Each of these was successfully addressed during different stages of the project:

• Isolate and identify both indigenous and opportunistically pathogenic microorganisms from treated drinking water systems.

Close to 100 isolates were collected and provided from water systems in Calgary and

Ontario, including three hospital water systems. Of these, 31 isolates were sequenced and identified using 16S rRNA gene sequencing and a number of groups of closely related or the same organisms were identified from within and between sample sites. Of the samples collected a number of E. coli and potentially pathogenic coliform bacteria were isolated. Three opportunistic pathogens ( L. pneumophila LPR-1, P. aeruginosa PAE-1 and S. maltophilia FH-W1) were provided for the study.

• Create a Terminal-Restriction Fragment Length Polymorphism (T-RFLP) library of isolated organisms for use in studying multi-species communities.

One organism from each of the taxonomic groups sequenced was chosen to represent the group in a model drinking water distribution system biofilm. In silico digests were performed on 16S rRNA gene sequences for each organism and the restriction enzyme

HhaI was chosen for use in the T-RFLP assay. Amplified 16S rDNA from each organism was digested separately to find the actual terminal restriction fragment length for that isolate when digested with HhaI. These values were combined to create the T-RFLP library, which was able to distinguish unique peaks for all the isolates (except for those in the family Sphingomanadaceae, and so they are referred to by the family rather than

136 species name). Fragment sizes were also the same for E. coli MEC-8 and E. cloacae

MTC-21 as well as S. maltophilia FH-W1 and L. pneumophila LPR-1. Although as none of these four would be inoculated into the model biofilm together the fragment size duplication was not an issue.

• Evaluate the growth of isolates under biofilm formation conditions and measure their susceptibility to chlorination as single-species planktonic and biofilm populations.

Twenty-one organisms were evaluated for growth ability and optimal incubation length for biofilm formation under the tested conditions in the Calgary Biofilm Device.

Seventeen isolates were able to form equivalent biofilms with incubation lengths ranging from 20 hours ( P. aeruginosa PAE-1) to 9 days ( Bradyrhizobium sp. B1-1). Three isolates were not able to form biofilms and one isolate was not able to form equivalent biofilms across the CBD. Assays for planktonic susceptibility (minimum inhibitory concentration, MIC) and biofilm susceptibility (minimum biofilm eradication concentration, MBEC) to chlorination at distribution system levels were performed individually for 9 heterotrophic organisms and 5 opportunistic pathogen strains. It was found that all the planktonic populations were killed within concentrations allowable by

Alberta legislation (0.1 mg/L to 4.0 mg/L). Single-species biofilm populations were consistently more resistant (1.6 to 40 fold) and 10 of the 14 biofilms were able to withstand chlorine concentrations higher than 4.0 mg/L, up to 52 (±5.9) mg/L for the most chlorine resistant strain Methylobacterium isbiliense MWI-1.

• Establish multi-species biofilms of isolates to examine the biofilm chlorine susceptibility, structure and community dynamics.

137

Using the T-RFLP library developed in Chapter 3 as well as information on biofilm

growth rate from Chapter 4, a model distribution system biofilm inoculation schedule

was developed. Using scanning electron microscopy (SEM) it was observed that in the

model system differentiation between cell morphologies was greater than in natural

biofilms, which also had much greater mineralization around the biofilm. Resistance of

multi-species communities containing a single opportunistic pathogen population ( E. coli

MEC-8, E. cloacae MTC-21 , P. aeruginosa PAE-1 or S. maltophilia FH-W1) was found to be much greater than that of single-species biofilms by 54-286 fold. Visualization of the pattern of disinfection using confocal scanning laser microscopy (CSLM) and live/dead staining showed that for three of the biofilms, more organisms within microcolonies survived while in the surrounding biofilm the proportion of killed cells was higher. Molecular examination of the treated biofilms with T-RFLP showed that only after treatment with high chlorine concentrations ≥500 mg/L was the amount of DNA, and therefore the number of cells, significantly reduced. Additionally, the isolate C0-3

(Sediminibacterium sp.) was lost from the biofilm at treatment levels between 16 and 50 mg/L chlorine. Other than C0-3, all of the organisms of the multi-species community were detectable in the biofilm even after experiencing high levels of chlorination. L. pneumophila LPR-1 was found to not be viable in the multi-species biofilm environment used in these experiments as it was not detectable by either culture-based methods or T-

RFLP. Other attempts at cultivating multi-species biofilms using continuous flow reactors were unsuccessful.

138

6.2 Discussion of results

6.2.1 Isolation of environmental organisms

Organisms living in and isolated from treated drinking water distribution systems are exposed to chlorinated, oligotrophic environments and so have more selective pressures than laboratory strains to develop resistance to these stresses. For this reason, in this project it was ideal to use organisms which had been directly isolated from the drinking water environment, including opportunistically pathogenic, indicator, and normal flora bacteria, rather than using laboratory strains. It was also desired to take a large sampling of organisms from the environment and from more than one location and growth phase

(planktonic and biofilm). This was in order to ensure that a true representation of the organisms present in natural water systems was collected and present in the final model water system biofilm. As discussed earlier, a water system may have a variation in microflora based on the growth phase, any stagnation which may occur, or temporal variation as the water system and biofilm matures (Santo Domingo et al. 2003,

Lautenschlager et al. 2010, Martiny et al. 2003). Using samples taken from both the biofilm and planktonic phase of the same distribution system (fed by the Bears Paw water treatment plant) spaced out over ~3-4 years, appeared to produce a good representation of the total community present in the distribution system based on expected species outlined in section 1.5.5.

The sample size of close to 100 original isolates was slightly low compared to other similar sampling studies (Martiny et al. 2003, Kalmbach et al. 1997, Williams et al.

2004). As not all of the isolates could reasonably be characterized by 16S rRNA gene

139

sequencing and in the single and multi-species biofilm assays within the time and

resource constraints of the project, the larger sample set was reduced to 29 isolates to be

sequenced. Of the isolates found and sequenced, the sample group generally had many of

the same species as were identified numerous times in the literature including Alpha-,

Beta- and Gammaproteobacteria as well as Actinobacteria. All organisms identified from

the Calgary distribution system had all previously been described in the literature as

occurring in water distribution systems. Organisms not included in the sample set but

which have been discussed in the literature included Burkholderia spp; Nitrospira spp ; and Planctomycetes.

Some important species to highlight that were found in the Calgary distribution system samples include: species in the family Sphingomonadaceae; Methylobacterium spp. and

Mycobacterium spp. The Sphingomonadaceae are important mainly due to the fact that they were present in the majority of studies of drinking water microflora examined

(Eichler et al. 2006, Williams et al. 2004, Tokajian et al. 2005, Schmeisser et al. 2003,

Simoes et al. 2007, Hong et al. 2010, Bereschenko et al. 2010) and in some cases noted as main colonizers (Tokajian et al. 2005, Bereschenko et al. 2010). Methylobacterium spp. are significant organisms to identify in a water system biofilm as some isolates have been shown to be chlorine resistant and are often found in potable water. In one study, approximately 30-40% of environmental isolates were chlorine resistant, except those isolated from raw water treatment systems and potable water storage tanks where between 90-100% of isolates were found to have chlorine resistance (Hiraishi et al.

1995). Additionally, the species Methylobacterium isbiliense was first isolated from a

140 treated water system (Gallego 2005). The presence of an undefined species of

Mycobacterium in the water samples may initially seem extremely undesirable, but it is the case that Mycobacterium spp. are regularly found in drinking water systems. Some of the commonly seen species include Mycobacterium frederiksbergense ; M. porcinum, M. arupense and M. mucogenicum (Williams et al. 2004, Tokajian et al. 2005, Simoes et al.

2007, Schmeisser et al. 2003, Liu et al. 2012). Even within the monitoring project at the

Alberta Children's Hospital (ACH) Mycobacterium spp, including M. gordonae and M. avium have been detected within the water system (C. Ferrato and B. Crago, unpublished data). Some of these strains can become nosocomial opportunistic pathogens in immunosuppressed or immunocompromised individuals (Liu et al. 2012). In this study, as these Mycobacterium spp are often found as native members of the water system biofilms, they were treated as such.

6.2.2 Creating a mixed-species model water system biofilm

In order to create the model water system multi-species biofilm it was advantageous to obtain a good representation of species and therefore collect a large number of isolates as discussed above. Alternatively, in the model biofilm it was not practical to have a very large number of species within the community as the biofilm would be harder to reproducibly inoculate and evaluate. (Kalmbach et al. 1997, Simoes et al. 2007) found that 70-80% of the community of a given water system could be represented with only 6-

8 bacterial strains. For this research the group of strains was decreased from close to 100 isolates initially evaluated to only 10 strains of background normal distribution system microflora and 5 species of opportunistic pathogens. As only 29 organisms were

141

identified by sequencing and the decision of which organisms to sequence was not

consistent across sample sets (For the ACH isolates all organisms were sequenced, for the

Foothills Medical Centre (FMC) isolates only unique colony morphologies were

sequenced), it is difficult to conclude that certain organisms are more or less prevalent in

the distribution system based solely on those results. It is important to note however the

occurrence of some genus groups in more than one location. For example,

Sediminibacterium sp. and Acidovorax sp. were both isolated from both the ACH as well as the FMH. Methylobacterium spp. were isolated from both the University of Calgary

Biological Sciences building (MWI-1) and the FMC (FH-K), but not from the ACH water

system. Being able to isolate the same organisms from multiple locations would indicate

that there is some continuity throughout the system and that the organisms are not only

present in the system from endpoint contamination.

6.2.3 Single species biofilm resistance to chlorine disinfection

The pattern of single species resistance to chlorination was found to be similar for all the

strains tested in the minimum biofilm eradication concentration (MBEC) and minimum

inhibitory concentration (MIC) assays, as organisms were consistently more resistant to

chlorination as biofilms than when grown planktonically. This result is supported by

numerous reports in the literature, notably that attachment to surfaces is the greatest

factor in enhancing bacterial survival in water systems (LeChevallier et al. 1988a). One

interesting thing to note about the data is that the levels of planktonic resistance are

similar for all of the organisms, but the biofilm resistance is quite variable between

different organisms. One possible explanation for this is that different microbial species

142 have different biofilm-forming abilities. For chlorination, diffusion into the biofilm is the limiting factor for disinfection (De Beer, Srinivasan and Stewart 1994) and so it is possible that organisms with a higher susceptibility are forming biofilms that are thinner, less dense or with less EPS production. In support of this the Bradyrhizobium sp. B1-1 was a very slow growing strain and after the 9 day incubation no biofilm was visible on the CBD peg. For other, more resistant strains such as Methylobacterium isbiliense MWI-

1 and the Sphingomonadacae (FH-D, FH-D and FH-J), biofilm growth was visible on the surface of the peg.

As might be expected, the organism able to survive the highest chlorine concentration was the isolate Methylobacterium isbiliense MWI-1, which belongs to the same genus as a number of known chlorine resistant strains, as discussed in Section 6.2.1. It is interesting to note though, that as a planktonic culture, it does not have significantly higher resistance to chlorination than other organisms. It may be the case that growth as a biofilm enhances the natural resistance present in the organism by increasing its biofilm biomass. Simoes et al. (2007) observed that Methylobacterium sp. was able to form the largest single species biofilm of six water isolates evaluated.

The most important item to note about the pattern of chlorine resistance in the single species populations is that, while the free living organisms are easily eliminated in chlorine concentrations within legislated limits (0.1-4.0 mg/L) (Alberta Environment

Drinking Water Branch 2006), in order to be eradicated the biofilm populations require chlorine concentrations to be close to, or greater than legislated levels, up to 13 times the

143 limit. This does not bode well for the use of residual chlorination to eliminate biofilm growth in water distribution systems.

6.2.4 Multi-species biofilm resistance to chlorine disinfection

The result seen in the single-species biofilm challenge of biofilm population resistance to chlorine being greater than planktonic cells is also seen in the multi-species MBEC assay, although at a much higher level. Multi-species biofilm communities of background microflora were able to survive chlorine concentrations up to 375 times the maximum legislated limit, both with and without the incorporation of an opportunistic pathogen population. The presence of the mixed-species community additionally increased the resistance of the population of the opportunistic pathogen isolates to far above their single species resistance to chlorination. This suggests, even more so than the results from the single species MIC and MBEC assays, that in a natural environment, residual chlorination would be ineffective at eliminating pathogens growing as a biofilm. Results in the literature support the findings here, that often multi-species biofilms have increased resistance than each of their individual counterparts (Klayman et al. 2009).

6.2.4.1 Comparison between multi-species biofilms

Although the same trend was observed with all four opportunistic pathogens introduced into established model multi-species biofilms, with the survival of the opportunistic pathogen greatly increasing over that of the single species biofilm survival, there is the difference in opportunistic pathogen detection between the plating and broth methods.

For most strains ( E. cloacae MTC-21 , P. aeruginosa PAE-1 and S. maltophilia FH-W1)

144

the difference between these two values is an increased detection of between 13 and 20%

for the PA broth method. E. coli MEC-8 has a 44% increase in detection between the plating and broth methods, much greater than the other opportunistic pathogen strains.

The main reason for the discrepancy between the detection of surviving opportunistic pathogens using plate versus broth methodologies could be explained simply that by using the broth it is possible to recover all of the cells within the biofilm. When recovering biofilms onto solid medium, cells can be lost or killed during the re- suspension process of sonicating the biofilm to remove it from the peg. In addition to this, only 40 µL is plated of the total 200 µL of re-suspended biofilm. In this way, the use of PA broth gives a much more accurate representation of the survival of opportunistic pathogens in the biofilm. Clark (1980) found that using the PA broth test for detection of coliforms from water systems had twice the recovery of the membrane filtration method.

The mathematical differences in biofilm recovery efficiency are able to account for the lower increase in resistance values for three of the opportunistic pathogen populations recovered in PA broth, but the 44% increase in resistance seen for the E. coli MEC-8 population is notably higher. This may be due to the use of PA broth as the recovery medium. When tested with individual cultures of all four opportunistic pathogens, all were able to cause a high amount of turbidity in the medium in <24 hours. Despite this, the literature suggests that PA broth as it is used in water testing may have higher recovery of coliform organisms, such as E. coli and E. cloacae, than other opportunistic pathogens, as it was originally designed for their recovery (Acumedia 2010).

145

6.2.5 Microscopy

The use of microscopy, both scanning electron microscopy (SEM) and confocal scanning laser microscopy (CSLM), to analyze the multi-species biofilms was very informative in terms of visualizing the differences between cultured and natural biofilms, determining the 3D structure of the biofilms and determining the pattern of disinfection of the biofilm, all on a CBD peg.

6.2.5.1 Scanning electron microscopy

What appears to be the most important difference between the cultured and natural biofilms visualized using SEM is the abundance of material deposits and mineralization crystals with the natural biofilms on the surface of the coupons. It was determined using

Energy Dispersion X-ray spectroscopy that the deposited materials and mineralization crystals were mainly composed of copper and carbon (Carol Stremick, Unpublished

Data). The presence of these materials on coupon surfaces is not unexpected after being exposed to the drinking water system for ~3-4 years. By changing the surface properties of the coupons, it is possible that the attachment of biofilm microorganisms may be altered in the natural biofilm. Other differences seen between the natural and cultured biofilms could be due to the different ages of the biofilms as well as the nutrient environment. The cultured biofilms are young and exposed to a (comparatively) high nutrient environment whereas the natural biofilms are grown in a very oligotrophic environment and are more mature. Despite these temporal and environmental differences and disregarding the presence of surface deposits, there are a number of similarities in the biofilm 3D structure with cells clumping together in microcolonies on some areas of the

146

peg and coupon, while other areas are bare or only sparsely populated by

microorganisms. As will be discussed below, the structure of the biofilm appears to have

an effect on the disinfection pattern. Therefore, the observation that similar structures are

present both in the natural and cultured biofilms would suggest that results regarding

patterns of disinfection on biofilms found in this study, may be applicable for natural

biofilms as well.

6.2.5.2 Confocal scanning laser microscopy

Images obtained by CSLM were very effective in displaying the structure and grouping

of cells of the multi-species community in the untreated biofilm as well as the pattern of

disinfection in the treated biofilm. In all of the untreated biofilms it was possible to see

different colony morphologies present, although this was most evident in the multi-

species biofilm with E. coli MEC-8 added. In other biofilms the same variation is not

immediately visible, which may be due to a number of reasons. In the biofilm including

S. maltophilia FH-W1 there is a higher proportion of live cells and so the exact cell

morphologies are not as discernible. In the biofilm containing P. aeruginosa PAE-1 the different cell morphologies are localized in different depths of the biofilm and not all are visible within the slice shown in Figure 5.10A. The cross sections of the biofilm to the bottom and right of the main image show the distinct 3D structure that is formed by the biofilm. Additionally in the untreated biofilm it can be observed on the cross sections of the images that the cell density is not constant throughout. Specifically in Figure 5.7A of the untreated multi-species biofilm with E. cloacae MTC-21 , the cross section passes directly through a microcolony formation and the cell density is notably higher in this

147

area. As one of the main limitations to chlorine disinfection on drinking water system

biofilms has been reported to be the ability of chlorine to diffuse into the biofilm (Mah

and O'Toole 2001, LeChevallier et al. 1988b, Parent et al. 1996), this higher cell density

could act as a diffusion barrier. This appears to be what is happening in the CSLM

images of the treated biofilms.

In these figures, some microcolonies are still present after treatment with low level

chlorination and the live (green) cells that are visible on the pegs appear to be either

contained within these groups ( E. coli MEC-8 multi-species biofilm) or the proportion of live cells is higher ( E. cloacae MTC-21 and P. aeruginosa PAE-1 multi-species biofilms). This suggests that it could be possible that the cells on the outer areas of the group play a protective role for the inner cells. Although the multi-species biofilm with S. maltophilia FH-W1 does appear to have increased cell density in some areas, whether the proportion of live cells is higher or equal cannot be discerned. In one of the treated biofilms ( E. cloacae MTC-21 multi-species) groups of live cells are visible within a layer

of red. While this colour would normally indicate a dead cell, the propidium iodide dye

will bind any DNA that is not protected by an intact membrane. It is therefore possible

that the red dye has attached to DNA that is part of the extracellular polymeric substance

surrounding the cells, which would serve to further protect the cells from chlorination.

The penetration of chlorine into the biofilm matrix and ability of the disinfectant to reach

the cells is a limiting factor for the disinfection efficiency of chlorine on biofilm

communities. Penetration is a factor of the reaction of the chlorine with organic

148 molecules in the biofilm EPS as well as the diffusion of chlorine through the matrix

(LeChevallier et al. 1988b, De Beer et al. 1994).

6.2.6 T-RFLP as a molecular method to study bacterial communities

A number of methods are available to study microorganisms in communities, such as classical culture dependent methods, 16S rRNA gene fingerprinting, RNA and DNA probes (including fluorescent in situ hybridization, FISH), denaturing gradient gel electrophoresis (DGGE), qPCR, 454 pyrosequencing and T-RFLP. Each methodology has relative advantages and disadvantages and specific limitations. For culture-dependent methods the obvious disadvantage is the selectivity of all media types and the presence of viable but non-culturable cells. This factor likely played a role in the community of microorganisms we were able to recover from the water system. Molecular based methods such as 454 pyrosequencing or T-RFLP are able to overcome the selectivity of growth media to give a good representation of organisms present by using universal primers designed to amplify and represent all organisms in a community. But in turn there are also disadvantages to these methodologies. The main limiting factor with molecular based methods is their ability to amplify DNA from dead cells, which may give a misrepresentation of the community. There is also some amplification bias associated with PCR methods and eubacterial primers (Aird et al. 2011).

For this project T-RFLP was selected as a molecular tool to examine the model multi- species biofilm community. Although 454 pyrosequencing is able to give a more accurate result, with less interpretation necessary and can detect species at very low number, at

149 this time the technology is still expensive and more suited to identifying uncharacterised, environmental communities (Hong et al. 2010). For this study the community being used was a set group of known organisms that had already been sequenced and so T-RFLP was an appropriate and useful tool. Often the main drawback noted for the T-RFLP assay is the difficulty of associating the peaks produced during raw fragment analysis to specific species, especially in large diverse communities. By creating a small library of

15 isolates and inoculating the community with those exact strains it was much easier to correlate fragment lengths (peaks) with specific species. Methods such as qPCR may also be appropriate for this type of community, but would require species specific PCR primers for each isolate. As all of the members in the community had already undergone

16S rRNA gene sequencing and the same primers were also able to be used for T-RFLP, it was the more logical choice.

6.2.7 T-RFLP analysis of chlorine treated multi-species biofilms

Direct comparisons between the T-RFLP data for the four multi-species biofilms with opportunistic pathogens, and the control biofilm with no opportunistic pathogen added are somewhat difficult due to the slight differences in treatment described in Section 2.5.

Originally all biofilms were to be treated with the same concentrations of chlorine, a low dose of 50 mg/L and high dose of 500 mg/L, both for 10 minutes. This would have allowed direct comparison between the five biofilms. Unfortunately it came to light late in the project that the samples for the E. cloacae MTC-21 and S. maltophilia FH-W1 multi-species biofilm had been compromised at one point in the procedure. This was observed via the T-RFLP profiles obtained from these samples, where the DNA from E.

150

cloacae MTC-21 and S. maltophilia FH-W1 was detected in both samples, Fortunately,

extracted DNA was available from an earlier trial of the procedure involving a low dose

of 16 mg/L and a high dose of 1000 mg/L chlorine, again for 10 minutes. Due to the

change in chlorine dose and initial difficulty with PCR amplification which was later

overcome, these samples were originally not intended for analysis with T-RFLP. But as

the 50 and 500 mg/L chlorine treated samples were unusable, the compromise was made

to use the older DNA extractions in order to have T-RFLP data for these strains, even if it

may not be directly comparable to the results for E. coli MEC-8, P. aeruginosa PAE-1 and control (no opportunistic pathogen) multi-species biofilms.

Despite the above difficulty, some patterns do arise in the data. The first of these is the survival (based in the detection or absence of DNA) of the isolate Sediminibacterium sp.

C0-3. In both the multi-species biofilms treated with 16 mg/L, C0-3 is present in small amounts, whereas in the biofilms treated with 50 mg/L it is not detectable, except for the control biofilm where it is only present in very low numbers. Examining the single species biofilm susceptibility of the organism in Chapter 4, it has the least amount of resistance to chlorine treatment, well below the legislated upper limit of 4.0 mg/L. This suggests that it is possible that the C0-3 population has actually been eliminated from the multi-species biofilm by approximately 50 mg/L of chlorine, which is still much higher that its single-species resistance. It is also interesting to observe that only in the multi- species biofilm with S. maltophilia FH-W1 does the Kocuria rhizophila C0-6 isolate grow to a similar proportion as in the control biofilm. This suggests the growth of C0-6 may be tempered by the presence of these other opportunistic pathogen strains.

151

Comparing the relative peak areas, the percent of the total peak area that each species

population contributes to the total biofilm community peak area (Figure 6.1), it is

interesting to note that the ratio of many of the strains in the biofilm remains relatively

constant, even when the total DNA extracted from the biofilm is lower. Additionally,

except for the C0-3 strain, all the other populations are present in the biofilm, even at the

highest concentrations of chlorine. This result was not expected as it was thought that

with high chlorine levels the communities would show a greater change. One other aspect

that can be noted in Figure 5.11 as well as Figure 6.1 is the large increase in proportion

of the community comprised of P. aeruginosa PAE-1 after treatment of 50 mg/L chlorine. Although there is a possibility that the population of P. aeruginosa PAE-1 has a much greater resistance to chlorination, this is not supported by the data from the 500 mg/L chlorine treatment. A likely explanation for the rise in the proportion of the population is due to the 24 hour recovery time between the chlorine treatment and biofilm recovery. As the P. aeruginosa PAE-1 is the fastest growing organism present in the

multi-species biofilm, it is possible that after 50 mg/L treatment, the cells are able to

recover and re-grow at a faster rate than the other organisms within the community.

Alternatively, at the high chlorine concentration of 500 mg/L the cells take longer to

recover and are not able to dominate the biofilm within the 24 hours.

6.2.8 Absence of Legionella pneumophila in multi-species biofilms

The inability to culture L. pneumophila LPR-1 in the multi-species biofilm model was

disappointing but not completely unexpected for a number of reasons. The first of which

152

153

Figure 6.1: Comparison or relative species abundance in multi-species biofilms Comparison of the relative peak areas (as a percent of total peak area) of each of the species within one of 15 biofilm treatments found using terminal-restriction fragment length polymorphism analysis (n=3).

154

was the difficulty in culturing the organism as a single species biofilm. A number of

combinations of inoculation, peg pre-treatment, incubation length, incubation

temperature, growth media, and recovery plates were tested to successfully grow the

single-species biofilm, many of which were not possible in the multi-species biofilm.

Many authors have reported trouble in growing L. pneumophila in bacterial biofilms,

although it was shown to persist over a period time in some studies (Moritz et al. 2010,

Lehtola et al. 2007). As discussed in Section 1.5.2.4, the ability of L. pneumophila to

persist and replicate within protozoa confers some advantages. A number of studies

which were successful in culturing L. pneumophila in biofilms, co-cultured it with amoeba (Kuiper et al. 2004, Murga et al. 2001), which was beyond the scope of this project.

6.2.9 Continuous-flow biofilm reactor

The unsuccessful outcome of both trial flow-through biofilm models was particularly disappointing as the models would give a stronger representation of actual conditions within a water distribution system. The flow-through reactors have a number of advantages in modeling water systems over the currently used biofilm growth model.

First of all by having a flow of media through the reactor rather than incubating in a set volume it enables the use of a lower nutrient formula with a constant inflow of new nutrients, as would be the case in a distribution system. It could be hypothesized that this may change the structure and community composition of the biofilm. To take this concept another step forward, the flow-through reactor could change the way in which the biofilm is challenged. With the current method the biofilm is challenged with an

155 absolute amount of chlorine for 10 minutes. Within the well of the CBD that chlorine reacts with the biofilm matrix and cells and can deplete over time. In a flow through system the biofilm can be exposed to a constant concentration of chlorine, which is continuously added to the biofilm reactor. This also allows for the examination of differences in biofilm growth, structure and community when the biofilm is sequentially or concurrently exposed to chlorination.

6.3 Future directions

Throughout this research there were a number of avenues which would have been interesting to explore further, although time, resources and expertise were limiting factors. Additionally, the review of literature and analysis of results have revealed further research questions and experiments that would further expand upon the research objectives proposed in Chapter 1. Although the original aims outlined for the project were achieved, one study may open the door to many other experimental questions. The logical progression of the study and the new questions that have emerged lie in two broad directions, first using the existing model to study the effects of different biocidal agents and incorporation of additional organisms on biofilm susceptibility, structure or composition. The second area that would be extremely interesting would be scaling up the biofilm model to an improved flow-through bioreactor, as was attempted and briefly described in Section 5.2.5.

156

6.3.1 Multi-species biofilm model challenge with further biocidal agents

One direction that would be interesting to go forward to from the work described here as

well as a logical next step, is performing MBEC assays on the model biofilm described

above, using a number of different water or health care relevant disinfectants. One of the

logical first choices for alternate disinfectants is monochloramine. Discussed in Section

1.2.3, monochloramine is often added to distribution system water as a residual

disinfectant after primary treatment. It is less reactive than free chlorine, as well as

having a longer persistence. It is also reported to have better penetration into biofilms.

Using the multi-species MBEC assay, it would be interesting to observe if

monochloramine provides different results in terms of survival, biofilm structure and

community.

6.3.2 Addition of organisms to the multi-species community

As discussed above, in order to gain a proper insight into the occurrence of L.

pneumophila in drinking water system biofilms, their co-culture with amoeba is required,

as occurs in the environment. It would also be interesting to conduct MBEC and MIC

assays with other water or hospital pathogens or opportunistic pathogens to observe if the

trends are maintained with a wider variety of species.

6.3.3 Localization of opportunistic pathogens within the biofilm

Fluorescent in situ hybridization (FISH) is utilized in a number of drinking water microbiota studies as a method for detecting cells which may be present in a non culturable state in an environmental biofilm. Although this method would not be well

157 suited to identification of all the strains in a multi-species biofilm, it does provide a better understanding of the biofilm structure. FISH enables the detection of where within a biofilm an organism is located, if it is found mostly sparsely on the surface or within microcolony or 3D formations.

6.3.4 Upscale of CBD bioreactor to a continuous-flow model system.

One direction that the research could be further pursued is the adaptation of a flow through biofilm reactor for the low nutrient environment of a water system biofilm, as discussed in Section 6.2.9. Although unsuccessful here, many studies of drinking water biofilms are performed using bioreactors to grow water system biofilms (Lehtola et al.

2007, Buswell et al. 2001). In order to fully relate the results found here to and to test possible solutions to the problem of opportunistic pathogens incorporating into water system biofilms, a flow-through bioreactor is a necessary next step.

6.4 Conclusions

The work presented here describes the collection of environmental isolates, the development of a model drinking water distribution system biofilm community, and its use in examining the effects of chlorination on the survival of opportunistic pathogens and normal community flora, variation in structure and patterns of disinfection in the biofilm, and finally the changes in species populations within the multi-species biofilm community. Within this project it has been successfully shown that environmentally isolated opportunistic pathogens and normal water system microflora, when grown as biofilms, are able to survive higher chlorine concentrations than their planktonic

158 counterparts and often higher than the maximum allowable concentration as legislated by the Government of Alberta. In turn, when grown as multi-species biofilms, it has also been shown that these biofilm populations are able to resist disinfection by chlorine levels ~50-290 fold higher than allowable in distribution systems. The source of this resistance is likely in part due to the 3D and microcolony structure of the biofilm which under low level chlorine stress only cells on the outer areas of the colonies are killed. The community makeup of the biofilm is not changed drastically, even with treatment with high chlorine concentration, other than to lower the total number of cells in the biofilm.

Only one strain is eliminated from the biofilm by treatment with ~50 mg/L of chlorine.

The fact that these biofilms are able to persist, and within them that opportunistic pathogens are able grow may be extremely harmful for those with a compromised immune status, especially in a hospital environment. If the opportunistic pathogens are able to become aerosolized (such as in a shower) or a sufficient quantity consumed, it could cause serious illness. Outside of the hospital environment, these results suggest that it would be possible for an indicator organism such as E. coli or E. cloacae to grow as part of a multi-species biofilm, and that the levels of chlorine within the distribution system would be unable to affect the community. If the biofilm were to periodically shed biofilm clusters it could cause positive water quality monitoring tests, a boil water advisory or at worst illness in a human population. Knowledge of the characteristics and susceptibility of the multi-species water system biofilms is important in order to respond quickly and efficiently in case of a contamination event, or ideally to prevent that occurrence altogether.

159

References

Acumedia, M. I. 2010. Presence-Absence Broth. ed. N. Corporation. Michigan. Adin, A., J. Katzhendler, D. Alkaslassy & C. Ravacha (1991) Trihalomethane formation in chlorinated drinking water: a kinetic model. Water Research, 25 , 797-805. Aird, D., M. G. Ross, W. S. Chen, M. Danielsson, T. Fennell, C. Russ, D. B. Jaffe, C. Nusbaum & A. Gnirke (2011) Analyzing and minimizing PCR amplification bias in Illumina sequencing libraries. Genome Biology, 12 , 14. Alberta Environment Drinking Water Branch, Environmental Protection Branch, Environmental Assurance Division (2006) Standards and Guidelines for Municipal Waterworks, Wastewater and Storm Drainage Systems. ed. Alberta Environment. Allen, M. (2004) Heterotrophic plate count bacteria—what is their significance in drinking water? International Journal of Food Microbiology, 92 , 265-274. American Public Health Agency (1998) Standard methods for the examination of water and wastewater. American Water Works Association. Washington, DC. Bereschenko, L. A., A. J. M. Stams, G. J. W. Euverink & M. C. M. van Loosdrecht (2010) Biofilm Formation on Reverse Osmosis Membranes Is Initiated and Dominated by Sphingomonas spp. Applied and Environmental Microbiology, 76 , 2623-2632. Boe-Hansen, R., H. J. Albrechtsen, E. Arvin & C. Jorgensen (2002) Bulk water phase and biofilm growth in drinking water at low nutrient conditions. Water Research, 36 , 4477-4486. Boe-Hansen, R., A. C. Martiny, E. Arvin & H. J. Albrechtsen (2003) Monitoring biofilm formation and activity in drinking water distribution networks under oligotrophic conditions. Water Science and Technology, 47 , 91-97. Bollet, C., A. Davinregli & P. Demicco (1995) A Simple Method for Selective Isolation of Stenotrophomonas maltophilia from environmental samples. Applied and Environmental Microbiology, 61 , 1653-1654. Booth, S., M. Workentine, J. Wen, R. Shaykhutdinov, H. Vogel, H. Ceri, R. Turner & A. Weljie (2011) Differences in Metabolism between the Biofilm and Planktonic Response to Metal Stress. Journal of Proteome Research, 10 , 3190-3199. Bosshard, P. P., R. Zbinden, S. Abels, B. Boddinghaus, M. Altwegg & E. C. Bottger (2006) 16S rRNA gene sequencing versus the API 20 NE system and the VITEK 2 ID-GNB card for identification of nonfermenting gram-negative bacteria in the clinical laboratory. Journal of Clinical Microbiology, 44 , 1359-1366. Breiman, R. F., B. S. Fields, G. N. Sanden, L. Volmer, A. Meier & J. S. Spika (1990) Association of shower use with Legionnaires disease: Possible role of amebas. Jama-Journal of the American Medical Association, 263 , 2924-2926. Bridier, A., R. Briandet, V. Thomas & F. Dubois-Brissonnet (2011) Resistance of bacterial biofilms to disinfectants: a review. Biofouling, 27 , 1017-1032.

160

Buswell, C. M., H. S. Nicholl & J. T. Walker (2001) Use of continuous culture bioreactors for the study of pathogens such as Campylobacter jejuni and Escherichia coli O157 in biofilms. Methods in Enzymology, 337 , 70-78. Campanac, C., L. Pineau, A. Payard, G. Baziard-Mouysset & C. Roques (2002) Interactions between Biocide Cationic Agents and Bacterial Biofilms. Antimicrobial Agents and Chemotherapy, 46 , 1469-1474. Camper, A. K., K. Brastrup, A. Sandvig & J. Clement (2003) Effect of distribution system materials on bacterial regrowth. American Water Works Association Journal, 95 , 107-121. Castonguay, M., S. Vanderschaaf, W. Koester, J. Krooneman, W. Vandermeer, H. Harmsen & P. Landini (2006) Biofilm formation by Escherichia coli is stimulated by synergistic interactions and co-adhesion mechanisms with adherence- proficient bacteria. Research in Microbiology, 157 , 471-478. Canadian Council of Ministers of the Enrironment Water Quality Task Group & Federal- Provincial-Territorial Committee on Drinking Water (2004) From Source to Tap: Guidance on the Multi-Barrier Approach to Safe Drinking Water. ed. Canadian Council of Ministers of the Environment. Ceri, H., M. E. Olson, C. Stremick, R. R. Read, D. Morck & A. Buret (1999) The Calgary Biofilm Device: New technology for rapid determination of antibiotic susceptibilities of bacterial biofilms. Journal of Clinical Microbiology, 37 , 1771- 1776. Ceri, H., C. A. Stremick & C. J. Nalepa (2000) A novel technique for evaluating the activity of biocides against biofilm bacteria. Corrosion 2000 NACE, 347. Choi, Y. & Y.-j. Choi (2010) The effects of UV disinfection on drinking water quality in distribution systems. Water Research, 44 , 115-122. City of Calgary Water Services (2012) 2011 Water Report. ed. City of Calgary. Clark, J. A. (1980) The influence of increasing numbers of nonindicator organisms upon the detection of indicator organisms by the membrane filter and presence-absence tests. Canadian Journal of Microbiology, 26 , 827-832. Cole, J. R., B. Chai, R. J. Farris, Q. Wang, A. S. Kulam-Syed-Mohideen, D. M. McGarrell, A. M. Bandela, E. Cardenas, G. M. Garrity & J. M. Tiedje (2007) The ribosomal database project (RDP-II): introducing myRDP space and quality controlled public data. Nucleic Acids Research, 35 , D169-D172. Cole, J. R., Q. Wang, E. Cardenas, J. Fish, B. Chai, R. J. Farris, A. S. Kulam-Syed- Mohideen, D. M. McGarrell, T. Marsh, G. M. Garrity & J. M. Tiedje (2009) The Ribosomal Database Project: improved alignments and new tools for rRNA analysis. Nucleic Acids Research, 37 , D141-D145. Costerton, J. W., Z. Lewandowski, D. E. Caldwell, D. R. Korber & H. M. Lappin-Scott (1995) Microbial biofilms. Annu Rev Microbiol, 49 , 711-45. Culman, S. W., R. Bukowski, H. G. Gauch, H. Cadillo-Quiroz & D. H. Buckley (2009) T-REX: software for the processing and analysis of T-RFLP data. Bmc Bioinformatics, 10. Dalben, M., G. Varkulja, M. Basso, V. L. J. Krebs, M. A. Gibelli, I. Van der Heijden, F. Rossi, G. Duboc, A. S. Levin & S. F. Costa (2008) Investigation of an outbreak of

161

Enterobacter cloacae in a neonatal unit and review of the literature. Journal of Hospital Infection, 70 , 7-14. Davies, J.-M. & A. Mazumder (2003) Health and environmental policy issues in Canada: the role of watershed management in sustaining clean drinking water quality at surface sources. Journal of Environmental Management, 68 , 273-286. De Beer, D., R. Srinivasan & P. S. Stewart (1994) Direct measurement of chlorine penetration into biofilms during disinfection. Applied and Environmental Microbiology, 60 , 4339-4344. Deborde, M. & U. von Gunten (2008) Reactions of chlorine with inorganic and organic compounds during water treatment - Kinetics and mechanisms: A critical review. Water Research, 42 , 13-51. Department of Indian Affairs and Northern Development (2011) National Assessment of First Nations Water and Wastewater Systems: National Roll-Up Report. ed. Government of Canada. DeSantis, T. Z., P. Hugenholtz, K. Keller, E. L. Brodie, N. Larsen, Y. M. Piceno, R. Phan & G. L. Andersen (2006a) NAST: a multiple sequence alignment server for comparative analysis of 16S rRNA genes. Nucleic Acids Research, 34 , W394- W399. DeSantis, T. Z., P. Hugenholtz, N. Larsen, M. Rojas, E. L. Brodie, K. Keller, T. Huber, D. Dalevi, P. Hu & G. L. Andersen (2006b) Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. Applied and Environmental Microbiology, 72 , 5069-5072. Doggett, M. S. (2000) Characterization of fungal biofilms within a municipal water distribution system. Applied and Environmental Microbiology, 66 , 1249-1251. Dykstra, T. S., K. C. O'Leary, C. Chauret, R. C. Andrews & G. A. Gagnon (2007) Impact of UV and secondary disinfection on microbial control in a model distribution system. Journal of Environmental Engineering Science, 6 , 147-155. Eichler, S., R. Christen, C. Holtje, P. Westphal, J. Botel, I. Brettar, A. Mehling & M. G. Hofle (2006) Composition and Dynamics of Bacterial Communities of a Drinking Water Supply System as Assessed by RNA- and DNA-Based 16S rRNA Gene Fingerprinting. Applied and Environmental Microbiology, 72 , 1858-1872. Environment Canada (2011) 2011 Municipal Water Use Report: Municipal water use 2009 statistics. ed. Government of Canada. Federal-Provincial-Territorial Committee on Drinking Water (2006) Guidelines for Canadian Drinking Water Quality: Guideline Technical Document- Trihalomethanes (With 2009 addendum). ed. Health Canada. Federal-Provincial- Territorial Committee on Health and the Environment. Ottawa, Ontario. --- (2010) Guidelines for Canadian Drinking Water Quality: Summary Table. ed. Health Canada. Federal-Provincial-Territorial Committee on Health and the Environment. Ottawa, Ontario. Feeley, J. C., R. J. Gibson, G. W. Gorman, N. C. Langford, J. K. Rasheed, D. C. Mackel & W. B. Baine (1979) Charcoal yeast extract agar: Primary isolation medium for Legionella pneumophila . Journal of Clinical Microbiology, 10 , 437-441. Flemming, H.-C (2002) Biofouling in water systems: Cases, causes and countermeasures. In Applied Microbiology and Biotechnology , 629-640.

162

Gallego, V. (2005) Methylobacterium isbiliense sp. nov., isolated from the drinking water system of Sevilla, Spain. International Journal of Systematic and Evolutionary Microbiology, 55 , 2333-2337. Gastmeier, P., D. Sohr, C. Geffers, H. Rueden, R.-P. Vonberg & T. Welte (2009) Early- and Late-Onset Pneumonia: Is This Still a Useful Classification? Antimicrobial Agents and Chemotherapy, 53 , 2714-2718. Giao, M. S., N. F. Azevedo, S. A. Wilks, M. J. Vieira & C. W. Keevil (2008) Persistence of Helicobacter pylori in Heterotrophic Drinking-Water Biofilms. Applied and Environmental Microbiology, 74 , 5898-5904. Hardalo, C. & S. C. Edberg (1997) Pseudomonas aeruginosa: Assessment of risk from drinking water. Critical Reviews in Microbiology, 23 , 47-75. Harrison, J. J., H. Ceri & R. J. Turner (2007) Multimetal resistance and tolerance in microbial biofilms. Nature Reviews Microbiology, 5 , 928-938. Harrison, J. J., R. J. Turner, L. L. R. Marques & H. Ceri (2005) Biofilms- A new understanding of these microbial communities is driving a revolution that may transform the science of microbiology. American Scientist, 93 , 508-515. Hennigs, J. K., H. J. Baumann, S. Schmiedel, P. Tennstedt, I. Sobottka, C. Bokemeyer, S. Kluge & H. Klose (2011) Characterization of Enterobacter cloacae Pneumonia: A Single-Center Retrospective Analysis. Lung, 189 , 475-483. Hiraishi, A., K. Furuhata, A. Matsumoto, K. A. Koike, M. Fukuyama & K. Tabuchi (1995) Phenotypic and genetic diversity of chlorine-resistant Methylobacterium strains isolated from various environments. Applied and Environmental Microbiology, 61 , 2099-2107. Hoiby, N (2010) Antibiotic resistance of bacterial biofilms. In International Journal of Antimicrobial Agents , 322-332. Holmes, B., W. R. Willcox & S. P. Lapage (1978) Identification of Enterobacteriaceae by the API 20E system. J Clin Pathol, 31 , 22-30. Hong, P. Y., C. Hwang, F. Ling, G. L. Andersen, M. W. LeChevallier & W. T. Liu (2010) Pyrosequencing Analysis of Bacterial Biofilm Communities in Water Meters of a Drinking Water Distribution System. Applied and Environmental Microbiology, 76 , 5631-5635. Horizon Chemical Co (2004) Material Safety Data Sheet: Sodium Hypochlorite 6-16%. St.Paul, MN, USA. Hoyle, B. D. & J. W. Costerton (1991) Bacterial resistance to antibiotics: The role of biofilms. In Progress in Drug Research, ed. E. Jucker, 91-105. Hu, J. Y., B. Yu, Y. Y. Feng, X. L. Tan, S. L. Ong, W. J. Ng & W. C. Hoe (2005) Investigation into biofilms in a local drinking water distribution system. Biofilms, 2, 19-25. Kalmbach, S., W. Manz & U. Szewzyk (1997) Isolation of New Bacterial Species from Drinking Water Biofilms and Proof of Their In Situ Dominance with Highly Specific 16S rRNA Probes. Applied and Environmental Microbiology, 63 , 4164- 4170. Kalmbach, S., W. Manz, J. Wecke & U. Szewzyk (1999) Aquabacterium gen. nov., with description of Aquabacterium citratiphilum sp. nov., Aquabacterium parvum sp. nov. and Aquabacterium commune sp. nov., three in situ dominant bacterial

163

species from the Berlin drinking water system. Journal of Systematic Bacteriology, 49 , 769-777. Karanfil, T., S. W. Krasner, P. Westerhoff & Y. Xie (2008) Disinfection By-Products in Drinking Water: Occurence, Formation, Health Effects and Control. American Chemical Society. Keinänen, M. M., P. J. Martikainen & M. H. Kontro (2004) Microbial community structure and biomass in developing drinking water biofilms. Canadian Journal of Microbiology, 50 , 183-191. Kilb, B. (2003) Contamination of drinking water by coliforms from biofilms grown on rubber-coated valves. International Journal of Hygiene and Environmental Health, 206 , 563-573. Klayman, B. J., P. A. Volden, P. S. Stewart & A. K. Camper (2009) Escherichia coli O157:H7 Requires Colonizing Partner to Adhere and Persist in a Capillary Flow Cell. Environmental Science & Technology, 43 , 2105-2111. Krasner, S. W. (2009) The formation and control of emerging disinfection by-products of health concern. Philosophical Transactions of the Royal Society a-Mathematical Physical and Engineering Sciences, 367 , 4077-4095. Krasner, S. W., H. S. Weinberg, S. D. Richardson, S. J. Pastor, R. Chinn, M. J. Sclimenti, G. D. Onstad & A. D. Thruston (2006) Occurrence of a New Generation of Disinfection Byproducts. Environmental Science & Technology, 40 , 7175-7185. Kuiper, M. W., B. A. Wullings, A. D. L. Akkermans, R. R. Beumer & D. van der Kooij (2004) Intracellular proliferation of Legionella pneumophila in Hartmannella vermiformis in aquatic biofilms grown on plasticized polyvinyl chloride. Applied and Environmental Microbiology, 70 , 6826-6833. Laing, R (2002) Report of the Commission of Inquiry into matters relating to the safety of the public drinking water in the City of North Battleford, Saskatchewan. In North Battleford Water Inquiry . Government of Saskatchewan. Lane, D. J (1991) 16S/23S rRNA sequencing. In Nucleic acid techniques in bacterial systematics, ed. M. G. Erko Stackebrandt, 115-175. New York NY: John Wiley and Sons. Langmark, J., M. V. Storey, N. J. Ashbolt & T. A. Stenstrom (2005) Accumulation and Fate of Microorganisms and Microspheres in Biofilms Formed in a Pilot-Scale Water Distribution System. Applied and Environmental Microbiology, 71 , 706- 712. Lau, H. Y. & N. J. Ashbolt (2009) The role of biofilms and protozoa in Legionella pathogenesis: implications for drinking water. Journal of Applied Microbiology, 107 , 368-378. Laupland, Kevin B., Michael D. Parkins, Deirdre L. Church, Daniel B. Gregson, Thomas J. Louie, John M. Conly, S. Elsayed & Johann D. D. Pitout (2005)

Population ‐Based Epidemiological Study of Infections Caused by

164

Carbapenem ‐Resistant Pseudomonas aeruginosa in the Calgary Health Region:

Importance of Metallo ‐β‐Lactamase (MBL)–Producing Strains. Journal of

Infectious Diseases, 192 , 1606-1612. Lautenschlager, K., N. Boon, Y. Wang, T. Egli & F. Hammes (2010) Overnight stagnation of drinking water in household taps induces microbial growth and changes in community composition. Water Research, 44 , 4868-4877. LeChevallier, M. W., T. M. Babcock & R. G. Lee (1987) Examination and characterization of drinking water system biofilms. Applied and Environmental Microbiology, 53 , 2714-2724. LeChevallier, M. W., C. D. Cawthon & R. G. Lee (1988a) Factors promoting survival of bacteria in chlorinated water supplies. Applied and Environmental Microbiology, 54 , 649-654. --- (1988b) Inactivation of biofilm bacteria. Applied and Environmental Microbiology, 54 , 2492-2499. LeChevallier, M. W., W. D. Norton & R. G. Lee (1991) Giardia and Cryptosporidium spp. in filtered drinking water supplies. Applied and Environmental Microbiology, 57 , 2617-2621. Lehtola, M., M. Laxander, I. Miettinen, A. Hirvonen, T. Vartiainen & P. Martikainen (2006) The effects of changing water flow velocity on the formation of biofilms and water quality in pilot distribution system consisting of copper or polyethylene pipes. Water Research, 40 , 2151-2160. Lehtola, M. J., I. T. Miettinen, M. M. Keinänen, T. K. Kekki, O. Laine, A. Hirvonen, T. Vartiainen & P. J. Martikainen (2004) Microbiology, chemistry and biofilm development in a pilot drinking water distribution system with copper and plastic pipes. Water Research, 38 , 3769-3779. Lehtola, M. J., E. Torvinen, J. Kusnetsov, T. Pitkanen, L. Maunula, C. H. von Bonsdorff, P. J. Martikainen, S. A. Wilks, C. W. Keevil & I. T. Miettinen (2007) Survival of Mycobacterium avium, Legionella pneumophila, Escherichia coli, and Caliciviruses in Drinking Water-Associated Biofilms Grown under High-Shear Turbulent Flow. Applied and Environmental Microbiology, 73 , 2854-2859. Lewis, K (2007) Persister cells, dormancy and infectious disease. In Nature Reviews Microbiology , 48-56. Liu, R. Y., Z. S. Yu, H. X. Zhang, M. Yang, B. Y. Shi & X. C. Liu (2012) Diversity of bacteria and mycobacteria in biofilms of two urban drinking water distribution systems. Canadian Journal of Microbiology, 58 , 261-270. Liu, Y. & J. Tay (2002) The essential role of hydrodynamic shear force in the formation of biofilm and granular sludge. Water Research, 36 , 1653-1665.

165

Långmark, J., M. V. Storey, N. J. Ashbolt & T.-A. Stenström (2007) The effects of UV disinfection on distribution pipe biofilm growth and pathogen incidence within the greater Stockholm area, Sweden. Water Research, 41 , 3327-3336. Mah, T. F. C. & G. A. O'Toole (2001) Mechanisms of biofilm resistance to antimicrobial agents. Trends in Microbiology, 9 , 34-39. Manuel, C. M., O. C. Nunes & L. F. Melo (2007) Dynamics of drinking water biofilm in flow/non-flow conditions. Water Research, 41 , 551-562. Marshall, J. K. 2009. Post-infectious irritable bowel syndrome following water contamination. In Kidney International , S42-S43. Martiny, A. C., T. M. Jorgensen, H. J. Albrechtsen, E. Arvin & S. Molin (2003) Long- Term Succession of Structure and Diversity of a Biofilm Formed in a Model Drinking Water Distribution System. Applied and Environmental Microbiology, 69 , 6899-6907. Miles, A., P. Singer, D. Ashley, M. Lynberg, P. Mendola, P. Langlois & J. Nuckols (2002) Comparison of trihalomethanes in tap water and blood. Environmental Science & Technology, 36 , 1692-1698. Money, P., A. F. Kelly, S. W. J. Gould, J. Denholm-Price, E. J. Threlfall & M. D. Fielder (2010) Cattle, weather and water: mapping Escherichia coli O157:H7 infections in humans in England and Scotland. Environmental Microbiology, 12 , 2633-2644. Moritz, M. M., H.-C. Flemming & J. Wingender (2010) Integration of Pseudomonas aeruginosa and Legionella pneumophila in drinking water biofilms grown on domestic plumbing materials. International Journal of Hygiene and Environmental Health, 213 , 190-197. Murga, R., T. S. Forster, E. Brown, J. M. Pruckler, B. S. Fields & R. M. Donlan (2001) Role of biofilms in the survival of Legionella pneumophila in a model potable- water system. Microbiology-Sgm, 147 , 3121-3126. Murphy, H. M., S. J. Payne & G. A. Gagnon (2008) Sequential UV- and chlorine-based disinfection to mitigate Escherichia coli in drinking water biofilms. Water Research, 42 , 2083-2092. Muyzer, G., A. Teske, C. O. Wirsen & H. W. Jannasch (1995) Phylogenetic relationships of Thiomicrospira species and their identification in deep-sea hydrothermal vent samples by denaturing gradient gel electrophoresis of 16S rDNA fragments. Archives of Microbiology, 164 , 165-172. Niquette, P., P. Servais & R. Savoir (2000) Impacts of pipe materials on densities of fixed bacterial biomass in a drinking water distribution system. Water Research, 34 , 1952-1956. Norton, C. D. & M. W. LeChevallier (1997) Chloramination: Its effect on distribution system water quality. Journal / American Water Works Association, 89 , 66-77. --- (2000) A pilot study of bacteriological population changes through potable water treatment and distribution. Applied and Environmental Microbiology, 66 , 268- 276. Nyc, O. & J. Matejkova (2010) Stenotrophomonas maltophilia: Significant contemporary hospital pathogen - review. Folia Microbiologica, 55 , 286-294. O'Connor, D. R (2002) Report of the Walkerton Inquiry: The Events of May 2000 and Related Issues. Ontario Ministry of the Attorney General.

166

Obst, U. & T. Schwartz (2007) Microbial Characteristics of Water Distribution: Compiled Investigations in a German Drinking Water Distribution System. Practice Periodical of Hazardous, Toxic and Radioactive Waste Management, 11 , 78-82. Ontario Ministry of the Environment (2006) Procedure for Disinfection of Drinking Water in Ontario. ed. Government of Ontario. Parent, A., S. Fass, M. L. Dincher, D. Reasoner, D. Gatel & J. C. Block (1996) Control of coliform growth in drinking water distribution systems. Journal of the Chartered Institution of Water and Environmental Management, 10 , 442-445. Park, S. K. & J. Y. Hu (2009) Interaction between phosphorus and biodegradable organic carbon on drinking water biofilm subject to chlorination. Journal of Applied Microbiology . Pauling-Shepard, K., L. Ward, M. Bouchard, J. Leavitt, S. Houshmand, J. Russell & T. Louie (2006) The amazing race - To MBL hell and back. An outbreak of metallo- beta-lactamase producing Pseudomonas aeruginosa (MBL-Psa) on a bone marrow transplant unit. Biology of Blood and Marrow Transplantation, 12 , 154-155. Pavlov, D. (2004) Potentially pathogenic features of heterotrophic plate count bacteria isolated from treated and untreated drinking water. International Journal of Food Microbiology, 92 , 275-287. Percival, S. L., J. S. Knapp, R. Edyvean & D. S. Wales (1998) Biofilm development on stainless steel in mains water. Water Research, 32 , 243-253. Pine, L., J. R. George, M. W. Reeves & W. K. Harrell (1979) Development of a chemically defined liquid medium for growth of Legionella pneumophila . Journal of Clinical Microbiology, 9 , 615-626. Pratt, L. A. & R. Kolter (1998) Genetic analysis of Escherichia coli biofilm formation: Roles of flagella, motility, chemotaxis and type I pili. Molecular Microbiology, 30 , 285-293. Reasoner, D. J. & E. E. Geldreich (1985) A new medium for the enumeration and subculture of bacteria from potable water. Appl. Environ. Microbiol., 49 , 1-7. Reisner, A., K. A. Krogfelt, B. M. Klein, E. L. Zechner & S. Molin (2006) In Vitro Biofilm Formation of Commensal and Pathogenic Escherichia coli Strains: Impact of Environmental and Genetic Factors. Journal of Bacteriology, 188 , 3572-3581. Ren, D., L. A. Bedzyk, S. M. Thomas, R. W. Ye & T. K. Wood (2004) Gene expression in Escherichia coli biofilms. Applied Microbiology and Biotechnology, 64 , 515- 524. Rickard, A. H., R. J. Palmer, D. S. Blehert, S. R. Campagna, M. F. Semmelhack, P. G. Egland, B. L. Bassler & P. E. Kolenbrander (2006) Autoinducer 2: a concentration-dependent signal for mutualistic bacterial biofilm growth. Molecular Microbiology, 60 , 1446-1456. Santo Domingo, J. W., M. C. Meckes, J. M. Simpson, B. Sloss & D. J. Reasoner (2003) Molecular characterization of bacteria inhabiting a water distribution system simulator. 149-154.

167

Sauer, K., A. K. Camper, G. D. Ehrlich, J. W. Costerton & D. G. Davies (2002) Pseudomonas aeruginosa Displays Multiple Phenotypes during Development as a Biofilm. Journal of Bacteriology, 184 , 1140-1154. Schmeisser, C., C. Stockigt, C. Raasch, J. Wingender, K. N. Timmis, D. F. Wenderoth, H. C. Flemming, H. Liesegang, R. A. Schmitz, K. E. Jaeger & W. R. Streit (2003) Metagenome Survey of Biofilms in Drinking-Water Networks. Applied and Environmental Microbiology, 69 , 7298-7309. Schop, R (2012) Personal Communication; M. Schwering. Sharp, R. R., A. K. Camper, J. J. Crippen, O. D. Schneider & S. Leggiero (2001) Evaluation of drinking water biostability using biofilm methods. Journal of Environmental Engineering, 127 , 403-410. Sibille, I., T. Sime-Ngando, L. Mathieu & J. C. Block (1998) Protozoan bacterivory and Escherichia coli survival in drinking water distribution systems. Applied and Environmental Microbiology, 64 , 197-202. Simoes, L. C., M. Simoes & M. J. Vieira (2007) Biofilm Interactions between Distinct Bacterial Genera Isolated from Drinking Water. Applied and Environmental Microbiology, 73 , 6192-6200. --- (2010) Influence of the Diversity of Bacterial Isolates from Drinking Water on Resistance of Biofilms to Disinfection. Applied and Environmental Microbiology, 76 , 6673-6679. Spencer, R. C. (1996) Predominant pathogens found in the European prevalence of infection in intensive care study. European Journal of Clinical Microbiology & Infectious Diseases, 15 , 281-285. Stewart, P. S (2002) Mechanisms of antibiotic resistance in bacterial biofilms. In IJMM International Journal of Medical Microbiology , 107-113. Stoodley, P., R. Cargo, C. Rupp, S. Wilson & I. Klapper (2002a) Biofilm material properties as related to shear-induced deformation and detachment phenomena. In Journal of Industrial Microbiology & Biotechnology , 361-367. Stoodley, P., K. Sauer, D. Davies & J. Costerton (2002b) Biofilms as complex differentiated communities. Annual Review of Microbiology, 56 , 187-209. Storey, M. V., B. van der Gaag & B. P. Burns (2011) Advances in on-line drinking water quality monitoring and early warning systems. Water Research, 45 , 741-747. Talmaciu, I., L. Varlotta, J. Mortensen & D. V. Schidlow (2000) Risk factors for emergence of Stenotrophomonas maltophilia in cystic fibrosis. Pediatric Pulmonology, 30 , 10-15. Telenius, H., N. P. Carter, C. E. Bebb, M. Nordenskjold, B. A. Ponder & A. Tunnacliffe (1992) Degenerate oligonucleotide-primed PCR: general amplification of target DNA by a single degenerate primer. Genomics, 13 , 718-25. Tokajian, S. T., F. A. Hashwa, I. C. Hancock & P. A. Zalloua (2005) Phylogenetic assessment of heterotrophic bacteria from a water distribution system using 16S rDNA sequencing. Canadian Journal of Microbiology, 51 , 325-335. United Nations General Assembly (2010) Resolution 64/292. The human right to water and sanitation. Venkobachar, C., L. Iyengar & A. Rao (1977) Mechanism of disinfection - effect of chlorine on cell-membrane functions. Water Research, 11 , 727-729.

168

Wang, D. L. & F. Wanda (2008) Evaluation of media for simultaneous enumeration of total coliform and Eschetichia coli in drinking water supplies by membrane filtration techniques. Journal of Environmental Sciences-China, 20 , 273-277. Weisel, C. P. & W. J. Chen (1994) - Exposure to Chlorination By-Products from Hot Water Uses. Risk Analysis, - 14 , - 106. World Health Organization & Unicef (2012) Progress on Drinking Water and Sanitation 2012 Update. ed. WHO and UNICEF Joint Monitoring Programme for Water Supply and Sanitation. Williams, M. M. & E. B. Braun-Howland (2003) Growth of Escherichia coli in model distribution system biofilms exposed to hypochlorous acid or monochloramine. Appl Environ Microbiol, 69 , 5463-71. Williams, M. M., J. W. S. Domingo, M. C. Meckes, C. A. Kelty & H. S. Rochon (2004) Phylogenetic diversity of drinking water bacteria in a distribution system simulator. Journal of Applied Microbiology, 96 , 954-964. Wingender, J. & H. C. Flemming (2004) Contamination potential of drinking water distribution network biofilms. 277-286. Wong, S., K. Pabbaraju, V. F. Burk, G. C. Broukhanski, J. Fox, T. Louie, M. W. Mah, K. Bernard & P. A. G. Tilley (2006) Use of sequence-based typing for investigation of a case of nosocomial legionellosis. Journal of Medical Microbiology, 55 , 1707- 1710. Zhang, W., D. E. Culley, L. Nie & J. C. M. Scholten (2007) Comparative transcriptome analysis of Desulfovibrio vulgaris grown in planktonic culture and mature biofilm on a steel surface. Applied Microbiology and Biotechnology, 76 , 447-457. Zhang, Z., S. Schwartz, L. Wagner & W. Miller (2000) A greedy algorithm for aligning DNA sequences. Journal of Computational Biology, 7 , 203-214.

169

APPENDIX I

Complete list of bacterial isolates recovered from the water system of the Foothills Medical Centre (planktonic organisms) in 2004 by the Infection Prevention and Control (IPC) laboratory. ID Originally Sampling Location Sampling date Gram Sept 2007 sub on Growth Colony morphology on Notes isolated on (dd/mm/yyyy) stain R2A (37°C) (R2A) R2A at 25°C 37°C 25°C

A BCYE s/ PCU61 Room sink 20/10/2004 G– white/milky ++ ++ m/cream/trans B BCYE PCU61 Room sink 20/10/2004 G– tiny/white - - Subculture did not grow C BCYE PCU61 Room sink 20/10/2004 G– yellow ++ ++ s/cream/trans Suspected contamination D BCYE PCU61 Room sink 20/10/2004 G– yellow ++ ++ s/light yellow/trans E BCYE PCU61 Room sink 20/10/2004 G– ng - - Subculture did not grow F BCYE s/ GWHC Utility room 20/10/2004 G– white/trans ++ +++ s/white/trans G BCYE GWHC Utility room 20/10/2004 G+ yellow +++ +++ s/yellow/trans H BCYE GWHC Utility room 20/10/2004 G– ng +++ +++ s/yellow gold/trans

J1 BCYE GWHC Utility room 20/10/2004 G– s/light yellow +++ +++ s/gold >1 colony morph obtained from sample J2 +++ +++ s/light yellow

K1 BCYE GWHC Utility room 20/10/2004 G+/– small yellow* ++ +++ variable size/pink >1 colony morph obtained from sample K2 ++ +++ s/yellow (some pink) L BCYE GWHC Utility room 20/10/2004 G– yellow haze + ++ tiny/trans N BCYE GWHC Patient room 20/10/2004 G+/– yellow haze +++ +++ s/light yellow P BCYE GWHC Patient room 20/10/2004 G+/– ng +++ ++ s/yellow Q R2A37° GWHC Patient room 20/10/2004 G– ng +++ ++ s/yellow R R2A37° GWHC Patient room 20/10/2004 G– ng + ++ tiny/trans S BCYE s/ GWHC Exam room 20/10/2004 G– ng ++ +++ tiny/cream/trans

170

Growth (R2A) T R2A37° GWHC Utility room 20/10/2004 G– ng ++ + tiny/trans U BCYE GWHC Mammogram room 20/10/2004 G– ng ++ - ng No growth at 25°C V R2A37° GWHC Mammogram room 20/10/2004 G– yellow +++ + s/cream/trans W R2A37° GWHC Mammogram room 13/10/2004 G– yellow +++ +++ s/cream/trans X BCYE ER Minor surgery 13/10/2004 G– haze + - ng No growth at 25°C Y BCYE ER Minor surgery 13/10/2004 G– white +++ + tiny/cream/trans Z BCYE ER Minor surgery 13/10/2004 G– yellow +++ ++ s/light yellow/trans A1 BCYE ER Minor surgery 13/10/2004 G– haze + - ng No growth at 25°C B1 BCYE ER Minor surgery 13/10/2004 G– s/white +haze +++ ++ lrg/bright white/opaque C1 BCYE s/ ER Minor surgery 13/10/2004 G+ Pink haze ++ ++ variable size/light pink D1 BCYE s/ ER Minor surgery 13/10/2004 G– white - - Subculture did not grow E1 BCYE s/ ER Minor surgery 13/10/2004 G– white +++ + tiny/trans F1 BCYE s/ ER Across room 27 14/10/2004 G– white +++ ++ tiny/trans G1 BCYE s/ ER Across room 27 14/10/2004 G– s/white +++ + tiny/trans H1 BCYE s/ ER Across room 27 14/10/2004 G– haze ++ +++ tiny/trans J1 BCYE s/ ER Minor surgery 14/10/2004 G– peach/trans +++ +++ tiny/trans L1 BCYE TBCC Day care bed sink 04/10/2004 G– white/trans - - Subculture did not grow M1 BCYE TBCC Day care bed sink 04/10/2004 G– haze +++ - ng N1 BCYE TBCC Pharmacy dispensing 04/10/2004 na ng - - Subculture did not grow Q1 R2A37° CVICU Med room sink 20/10/2004 G– peach ++ + tiny/trans R1 R2A37° CVICU Med room sink 20/10/2004 G– peach ++ ++ tiny/trans S1 R2A37° CVICU Med room sink 20/10/2004 G– ng ++ - ng No growth at 25°C T1 R2A37° CVICU Med room sink 20/10/2004 G– peach ++ + tiny/trans

171

Growth (R2A) U1 BCYE CVICU Utility room 20/10/2004 G– peach ++ + s/cream/trans V1 R2A37° CVICU Resp sink area 20/10/2004 G– peach/trans +++ +++ m/cream yellow Suspected contamination W1 BCYE s/ PCU61 Patient room 20/10/2004 G+ white/mucoid +++ +++ m/cream yellow Y1 R2A37° GWHC Pt exam room sink 19/10/2004 G– ng + - ng No growth at 25°C Z1 BCYE s/ GWHC Pt exam room sink 20/10/2004 G– white/mucoid +++ +++ s/cream/trans * Original isolate of FH-K on BCYE was a pink organism. PCU 61 (Patient Care Unit 61); GWHC (Grace Women's Health Centre); ER (Emergency); TBCC (Tom Baker Cancer Centre); CVICU (Cardiovascular Intensive Care Unit); BCYE s/ (BCYE with antibiotics). The grey section of the table represents data obtained from the IPC lab staff.

172

APPENDIX II

Complete list of bacterial isolates recovered from various water systems within southern Ontario and processed at the Ontario Ministry of the Environment (MoE) Lab. Isolate ID Sampling Location Water system type Date sub'd for api test (~2-4 days Preliminary identification using MoE sample post-sampling) (dd/mm/yyyy) api20E test accession number MEC-1 Waterloo Regional private well 25/02/09 Escherichia coli (99.8%) 166332-6-12 Municipality MEC-2 Frontenac County private well 27/02/09 Escherichia coli (99.8%) 166395-1-82 MEC-3 Frontenac County private well 27/02/09 Escherichia coli (99.8%) 166395-2-83 MEC-4 Waterloo Regional private well 27/02/09 Escherichia coli (99.8%) 166414-4-115 Municipality MEC-5 Frontenac County private (raw) well 08/04/09 Escherichia coli (99.5%) 167146-1-3 MEC-6 Frontenac County private (raw) well 08/04/09 Escherichia coli (99.9%) 167146-2-4 MEC-7 Muskoka District Treated distribution system 30/05/09 Escherichia coli (98.4%) 168232-1-24 Municipality (AWQI)* MEC-8 Lennox and Addington private (raw) well 06/18/09 Escherichia coli (99.9%) 168808-1-49 County MEC-9 Frontenac County private (raw) well 25/06/09 Escherichia coli (99.5%) 169014-2-45 MEC-10 Frontenac County private (raw) well 25/06/09 Escherichia coli (99.5%) 169014-3-46 MEC-11 Frontenac County private (raw) well 25/06/09 Escherichia coli (97.7%) 169014-1-44 MEC-12 Niagara Regional Treated municipal 10/08/09 Escherichia coli (94.0%) ES2-1 Municipality distribution system (Citrobacter freundii 4.2%) (AWQI) MEC-13 Niagara Regional Treated municipal 10/08/09 Escherichia coli (94.0%) ES2-2 Municipality distribution system (Citrobacter freundii 4.2%) (AWQI) MTC-1 Waterloo Regional private well 25/02/09 Serratia fonticola (99.3%) 166332-6-12

173

Isolate ID Sampling Location Water system type Date sub'd for api test (~2-4 days Preliminary identification using MoE sample post-sampling) (dd/mm/yyyy) api20E test accession number Municipality MTC -2 Frontenac County private (raw) well 27/02/09 Enterobacter amnigenus (83.3%) 166395-1-82 MTC-3 Frontenac County private (raw) well 27/02/09 Citrobacter braakii (59.7%) 166395-2-83 Citrobacter freundii (38.2%) MTC-4 Frontenac County private (raw) well 27/02/09 Enterobacter aerogenes (96.0%) 166395-3-84 MTC-5 Waterloo Regional private well 27/02/09 Serratia fonticola (99.3%) 166414-4-115 Municipality MTC-6 Waterloo Regional private well 27/02/09 Citrobacter braakii (99.8%) 166414-5-116 Municipality MTC-7 Durham Regional private (raw) well 28/02/09 Klebsiella pneumoniae spp 166434-3-124 Municipality pneumoniae (97.6 %) MTC-8 Waterloo Regional private (raw) well 11/03/09 Citrobacter braakii (99.8%) 166643-1A-25 Municipality MTC-9 Waterloo Regional private (raw) well 11/03/09 Rahnella aquatillis (98.9%) 166643-1B-25 Municipality MTC-10 Frontenac County private (raw) well 08/04/09 Rahnella aquatillis (78.4%) 167146-1-3 Pantoea ssp 2 (17.2%) MTC-11 Frontenac County private (raw) well 08/04/09 Serratia fonticola (99.3%) 167146-2-4 MTC-12 Frontenac County private (raw) well 08/04/09 Rahnella aquatillis (78.4%) 167146-3-5 Pantoes ssp 2 (17.2%) MTC-13 Ottawa region private (raw) well 25/04/09 Serratia fonticola (99.3%) 167491-2-78 MTC-14 Lanark County private (raw) well 05/05/09 Serratia fonticola (60.7%) 168513-1-64 Enterobacter aerogenes (30.5%) MTC-15 Lanark County private (raw) well 05/05/09 Enterobacter intermedius (93.5%) 168513-2-65 MTC-16 Lennox and Addington private (raw) well 18/06/09 Serratia fonticola (97.4%) 168808-1-49 County

174

Isolate ID Sampling Location Water system type Date sub'd for api test (~2-4 days Preliminary identification using MoE sample post-sampling) (dd/mm/yyyy) api20E test accession number MTC-17 private (raw) well 20/06/09 Enterobacter intermedius (93.5%) 168908-1-94 MTC-18 Frontenac County private (raw) well 25/06/09 Serratia fonticola (99.3%) 169014-1-44 MTC-19 Frontenac County private (raw) well 25/06/09 Rahnella aquatillis (98.9%) 169014-2-45 MTC-20 Frontenac County private (raw) well 25/06/09 Pantoea ssp 3 (68.0%) 169014-3-46 Rahnella aquatillis (19.9%) MTC-21 Peterborough County Treated municipal 07/08/09 Enterobacter cloacae (91.5%) ES17-2 distribution system (AWQI) MTC-22 Niagara Regional Treated municipal 07/08/09 Buttiauxella agrestis (88.0%) WNF4-1 Municipality distribution system Enterobacter intermedius (10.7%) (AWQI) MTC-23 Elgin County Treated municipal 07/08/09 Enterobacter intermedius (93.5%) ES20-1 distribution system (AWQI) UI-1 Elgin County Treated municipal 07/08/09 Api all negative, unknown ES20-2 distribution system (AWQI) *AWQI (Adverse water quality incident) sample was taken from a regulated drinking water system, and the positive result was reported to the local public health unit and corrective actions taken.

175

APPENDIX III

Full sequences of amplified 16S rRNA gene fragments for 31 isolates as obtained from the UCDNA sequencing Lab.

Strain and Partial 16S rRNA gene sequences Identification

Ontario MoE Isolates MEC-1 TGCAAGTCGAACGGTAACAGGAAGCAGCTTGCTGNTTCGCTGACGAGTGGCGGACGGGTGAGTAATGTCTGGGAAACTGCCTGATGGAG GGGGATAACTACTGGAAACGGTAGCTAATACCGCATAACGTCGCAAGACCAAAGAGGGGGACCTTCGGGCCTCTTGCCATCGGATGTGC Escherichia/ CCAGATGGGATTAGCTTGTTGGTGGGGTAACGGCTCACCTAGGCGACGATCCCTAGCTGGTCTGAGAGGATGACCAGCCACACTGGAAC Shigella sp. TGAGACACGGTCCAGACTCCTACGGGAGGCAGCAGTGGGGAATATTGCACAATGGGCGCAAGCCTGATGCAGCCATGCCGCGTGTATGA AGAAGGCCTTCGGGTTGTAAAGTACTTTCAGCGGGGAGGAAGGGAGTAAAGTTAATACCTTTGCTCATTGACGTTACCCGCAGAAGAAG CACCGGCTAACTCCGTGCCAGCAGCCGCGGTAATACGGAGGGTGCAAGCGTTAATCGGAATTACTGGGCGTAAAGCGCACGCAGGCGGT TTGTTAAGTCAGATGTGAAATCCCCGGGCTCAACCTGGGAACTGCATCTGATACTGGCAAGCTTGAGTCTCGTAGAGGGGGGTAGAATT CCAGGTGTAGCGGTGAAATGCGTAGAGATCTGGAGGAATACCGGTGGCGAAGGCGGCCCCCTGGACGAAGACTGACGCTCAGGTGCGAA AGCGTGGGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGCCGTAAACGATGTCGACTTGGAGGTTGTGCCCTTGAGGCGTGGCTTC CGGAGCTAACGCGTTAAGTCGACCGCCTGGGGAGTACGGCCGCAAGGTTAA

MEC-7 TGCAAGTCGAACGGTAACAGGAAGCAGCTTGCTGCTTCGCTGACGAGTGGCGGACGGGTGAGTAATGTCTGGGAAACTGCCTGATGGAG GGGGATAACTACTGGAAACGGTAGCTAATACCGCATAACGTCGCAAGACCAAAGAGGGGGACCTTCGGGCCTCTTGCCATCGGATGTGC Escherichia/ CCAGATGGGATTAGCTAGTAGGTGGGGTAACGGCTCACCTAGGCGACGATCCCTAGCTGGTCTGAGAGGATGACCAGCCACACTGGAAC Shigella sp. TGAGACACGGTCCAGACTCCTACGGGAGGCAGCAGTGGGGAATATTGCACAATGGGCGCAAGCCTGATGCAGCCATGCCGCGTGTATGA AGAAGGCCTTCGGGTTGTAAAGTACTTTCAGCGGGGAGGAAGGGAGTAAAGTTAATACCTTTGCTCATTGACGTTACCCGCAGAAGAAG CACCGGCTAACTCCGTGCCAGCAGCCGCGGTAATACGGAGGGTGCAAGCGTTAATCGGAATTACTGGGCGTAAAGCGCACGCAGGCGGT TTGTTAAGTCAGATGTGAAATCCCCGGGCTCAACCTGGGAACTGCATCTGATACTGGCAAGCTTGAGTCTCGTAGAGGGGGGTAGAATT CCAGGTGTAGCGGTGAAATGCGTAGAGATCTGGAGGAATACCGGTGGCGAAGGCGGCCCCCTGGACGAAGACTGACGCTCAGGTGCGAA AGCGTGGGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGCCGTAAACGATGTCGACTTGGAGGTTGTGCCCTTGAGGCGTGGCTTC CGGAGCTAACGCGTTAAGTCGACCGCCTGGGGAGTACGGCCGCAAGGTTAA

176

Strain and Partial 16S rRNA gene sequences Identification MEC-8 TGCAAGTCGAACGGTAACAGGAAGAAGCTTGCTTCTTTGCTGACGAGTGGCGGACGGGTGAGTAATGTCTGGGAAACTGCCTGATGGAG GGGGATAACTACTGGAAACGGTAGCTAATACCGCATAACGTCGCAAGACCAAAGAGGGGGACCTTCGGGCCTCTTGCCATCGGATGTGC Escherichia coli CCAGATGGGATTAGCTTGTTGGTGGGGTAACGGCTCACCTAGGCGACGATCCCTAGCTGGTCTGAGAGGATGACCAGCCACACTGGAAC TGAGACACGGTCCAGACTCCTACGGGAGGCAGCAGTGGGGAATATTGCACAATGGGCGCAAGCCTGATGCAGCCATGCCGCGTGTATGA AGAAGGCCTTCGGGTTGTAAAGTACTTTCAGCGGGGAGGAAGGGAGTAAAGTTAATACCTTTGCTCATTGACGTTACCCGCAGAAGAAG CACCGGCTAACTCCGTGCCAGCAGCCGCGGTAATACGGAGGGTGCAAGCGTTAATCGGAATTACTGGGCGTAAAGCGCACGCAGGCGGT TTGTTAAGTCAGATGTGAAATCCCCGGGCTCAACCTGGGAACTGCATCTGATACTGGCAAGCTTGAGTCTCGTAGAGGGGGGTAGAATT CCAGGTGTAGCGGTGAAATGCGTAGAGATCTGGAGGAATACCGGTGGCGAAGGCGGCCCCCTGGACGAAGACTGACGCTCAGGTGCGAA AGCGTGGGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGCCGTAACGATGTCGACTTGGAGGTTGTGCCCTTGAGGCGTGGCTTCC GGAGCTAACGCGTTAAGTCGACCGCCTGGGGAGTACGGCCGCAAGGTTAA

TC-21 TGCAAGTCGAACGGTAACAGGAAGCAGCTTGCTGCTTCGCTGACGAGTGGCGGACGGGTGAGTAATGTCTGGGAAACTGCCTGATGGAG GGGGATAACTACTGGAAACGGTAGCTAATACCGCATAACGTCGCAAGACCAAAGAGGGGGACCTTCGGGCCTCTTGCCATCGGATGTGC Enterobacter CCAGATGGGATTAGCTAGTAGGTGGGGTAACGGCTCACCTAGGCGACGATCCCTAGCTGGTCTGAGAGGATGACCAGCCACACTGGAAC cloacae TGAGACACGGTCCAGACTCCTACGGGAGGCAGCAGTGGGGAATATTGCACAATGGGCGCAAGCCTGATGCAGCCATGCCGCGTGTATGA AGAAGGCCTTCGGGTTGTAAAGTACTTTCAGCGGGGAGGAAGGCGATGAGGTTAATAACCTCAtCGATTGACGTTACCCGCAGAAGAAG CACCGGCTAACTCCGTGCCAGCAGCCGCGGTAATACGGAGGGTGCAAGCGTTAATCGGAATTACTGGGCGTAAAGCGCACGCAGGCGGT CTGTCAAGTCGGATGTGAAATCCCCGGGCTCAACCTGGGAACTGCATTCGAAACTGGCAGGCTAGAGTCTTGTAGAGGGGGGTAGAATT CCAGGTGTAGCGGTGAAATGCGTAGAGATCTGGAGGAATACCGGTGGCGAAGGCGGCCCCCTGGACAAAGACTGACGCTCAGGTGCGAA AGCGTGGGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGCCGTAAACGATGTCGACTTGGAGGTTGTGCCCTTGAGGCGTGGCTTC CGGAGCTAACGCGTTAAGTCGACCGCCTGGGGAGTACGGCCGCAAGGTTAAAACTCAG

TC-23 TGCAAGTCGAACGGTAGCACAGAGAGCTTGCTCTTGGGTGACGAGTGGCGGACGGGTGAGTAATGTCTGGGAAACTGCCCGATGGAGGG GGATAACTACTGGAAACGGTAGCTAATACCGCATAACGTCGCAAGACCAAAGTGGGGGACCTTCGGGCCTCACACCATCGGATGTGCCC Kluyvera intermedia AGATGGGATTAGCTAGTAGGTGGGGTAATGGCTCACCTAGGCGACGATCCCTAGCTGGTCTGAGAGGATGACCAGCCACACTGGAACTG AGACACGGTCCAGACTCCTACGGGAGGCAGCAGTGGGGAATATTGCACAATGGGCGCAAGCCTGATGCAGCCATGCCGCGTGTATGAAG AAGGCCTTCGGGTTGTAAAGTACTTTCAGCGAGGAGGAAGGCATTGTGGTTAATAACCGCAGTGATTGACGTTACTCGCAGAAGAAGCA CCGGCTAACTCCGTGCCAGCAGCCGCGGTAATACGGAGGGTGCAAGCGTTAATCGGAATTACTGGGCGTAAAGCGCACGCAGGCGGTCT GTCAAGTCGGATGTGAAATCCCCGGGCTCAACCTGGGAACTGCATTCGAAACTGGCAGGCTAGAGTCTTGTAGAGGGGGGTAGAATTCC AGGTGTAGCGGTGAAATGCGTAGAGATCTGGAGGAATACCGGTGGCGAAGGCGGCCCCCTGGACAAAGACTGACGCTCAGGTGCGAAAG CGTGGGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGCCGTAAACGATGTCGACTTGGAGGTTGTGCCCTTGAGGAGTGGCTTCCG GAGCTAACGCGTTAAGTCGACCGCCTGGGGAGTACGGCCGCAAGGTTAAAACTCAG

177

Strain and Partial 16S rRNA gene sequences Identification

Alberta Children's Hospital Biofilm Isolates A3-1 TGCAAGTCGAACGGCAGCACGGGAGCAATCCTGGTGGCGAGTGGCGAACGGGTGAGTAATACATCGGAACGTGCCCAATCGTGGGGGAT AACGCAGCGAAAGCTGTGCTAATACCGCATACGATCTACGGATGAAAGCAGGGGACCGCAAGGCCTTGCGCGAGTGGAGCGGCCGATGG Variovorax sp. CAGATTAGGTAGTTGGTGAGGTAAAGGCTCACCAAGCCTTCGATCTGTAGCTGGTCTGAGAGGACGACCAGCCACACTGGGACTGAGAC ACGGCCCAGACTCCTACGGGAGGCAGCAGTGGGGAATTTTGGACAATGGGCGAAAGCCTGATCCAGCAATGCCGCGTGCAGGATGAAGG CCTTCGGGTTGTAAACTGCTTTTGTACGGAACGAAACGGTCCTTTCTAATACAGAGGGCTAATGACGGTACCGTAAGAATAAGCACCGG CTAACTACGTGCCAGCAGCCGCGGTAATACGTAGGGTGCAAGCGTTAATCGGAATTACTGGGCGTAAAGCGTGCGCAGGCGGTTATGTA AGACAGTTGTGAAATCCCCGGGCTCAACCTGGGAATTGCATCTGTGACTGCATAGCTAGAGTACGGTAGAGGGGGATGGAATTCCGCGT GTAGCAGTGAAATGCGTAGATATGCGGAGGAACACCGATGGCGAAGGCAATCCCCTGGACCTGTACTGACGCTCATGCACGAAAGCGTG GGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGCCCTAAACGATGTCAACTGGTTGTTGGGTCTTCACTGACTCAGTAACGAAGCT AACGCGTGAAGTTGACCGCCTGGGGAGTACGGCCGC

A3-2 TGCAAGTCGAACGGCAGCGCGGGCTTCGGCCTGGCGGCGAGTGGCGAACGGGTGAGTAATACATCGGAACGTGCCCTGTTGTGGGGGAT AACTAGTCGAAAGATTAGCTAATACCGCATACGACCTGAGGGTGAAAGCGGGGGACCGCAAGGCCTCGCGCAATAGGAGCGGCCGATGT Cupriavidus CTGATTAGCTAGTTGGTGGGGTAAAGGCCTACCAAGGCGACGATCAGTAGCTGGTCTGAGAGGACGATCAGCCACACTGGGACTGAGAC respiraculi ACGGCCCAGACTCCTACGGGAGGCAGCAGTGGGGAATTTTGGACAATGGGGGCAACCCTGATCCAGCAATGCCGCGTGTGTGAAGAAGG CCTTCGGGTTGTAAAGCACTTTTGTCCGGAAAGAAATCGCGCTGGCTAATACCTGGCGTGGATGACGGTACCGGAAGAATAAGCACCGG CTAACTACGTGCCAGCAGCCGCGGTAATACGTAGGGTGCGAGCGTTAATCGGAATTACTGGGCGTAAAGCGTGCGCAGGCGGTTTTGTA AGACAGGCGTGAAATCCCCGAGCTCAACTTGGGAATGGCGCTTGTGACTGCAAGGCTAGAGTATGTCAGAGGGGGGTAGAATTCCACGT GTAGCAGTGAAATGCGTAGAGATGTGGAGGAATACCGATGGCGAAGGCAGCCCCCTGGGACGTCACTGACGCTCATGCACGAAAGCGTG GGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGCCCTAAACGATGTCAACTAGTTGTTGGGGATTCATTTCCTCAGTAACGTAGCT AACGNGGTGAAGTTGACCGCCTGGGGAGTACGGTCGCAAGATTAA

B1-1 TGCAAGTCGAGCGGGCGTAGCAATACGTCAGCGGCAGACGGGTGAGTAACGCGTGGGAACGTACCTTTTGGTTCGGAACAACACAGGGA AACTTGTGCTAATACCGGATAAGCCCTTACGGGGAAAGATTTATCGCCGAAAGATCGGCCCGCGTCTGATTAGCTAGTTGGTAGGGTAA Bradyrhizobium sp. TGGCCTACCAAGGCGACGATCAGTAGCTGGTCTGAGAGGATGATCAGCCACATTGGGACTGAGACACGGCCCAAACTCCTACGGGAGGC AGCAGTGGGGAATATTGGACAATGGGGGCAACCCTGATCCAGCCATGCCGCGTGAGTGATGAAGGCCCTAGGGTTGTAAAGCTCTTTTG TGCGGGAAGATAATGACGGTACCGCAAGAATAAGCCCCGGCTAACTTCGTGCCAGCAGCCGCGGTAATACGAAGGGGGCTAGCGTTGCT CGGAATCACTGGGCGTAAAGGGTGCGTAGGCGGGTCTTTAAGTCAGGGGTGAAATCCTGGAGCTCAACTCCAGAACTGCCTTTGATACT GAAGATCTTGAGTTCGGGAGAGGTGAGTGGAACTGCGAGTGTAGAGGTGAAATTCGTAGATATTCGCAAGAACACCAGTGGCGAAGGCG GCTCACTGGCCCGATACTGACGCTGAGGCACGAAAGCGTGGGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGCCGTAAACGATGA ATGCCAGCCGTTAGTGGGTTTACTCACTAGTGGCGCAGCTAACGCTTTAAGCATTCCGCCTGGGGAGTACGGTCGCAAGATTAA

178

Strain and Partial 16S rRNA gene sequences Identification C0-1 TGCAAGTCGAACGGCAGCGCGGGCTTCGGCCTGGCGGCGAGTGGCGAACGGGTGAGTAATACATCGGAACGTGCCCTGTTGTGGGGGAT AACTAGTCGAAAGATTAGCTAATACCGCATACGACCTGAGGGTGAAAGCGGGGGACCGCAAGGCCTCGCGCAATAGGAGCGGCCGATGT Cupriavidus CTGATTAGCTAGTTGGTGGGGTAAAGGCCTACCAAGGCGACGATCAGTAGCTGGTCTGAGAGGACGATCAGCCACACTGGGACTGAGAC respiraculi ACGGCCCAGACTCCTACGGGAGGCAGCAGTGGGGAATTTTGGACAATGGGGGCAACCCTGATCCAGCAATGCCGCGTGTGTGAAGAAGG CCTTCGGGTTGTAAAGCACTTTTGTCCGGAAAGAAATCGCGCTGGCTAATACCTGGCGTGGATGACGGTACCGGAAGAATAAGCACCGG CTAACTACGTGCCAGCAGCCGCGGTAATACGTAGGGTGCGAGCGTTAATCGGAATTACTGGGCGTAAAGCGTGCGCAGGCGGTTTTGTA AGACAGGCGTGAAATCCCCGAGCTCAACTTGGGAATGGCGCTTGTGACTGCAAGGCTAGAGTATGTCAGAGGGGGGTAGAATTCCACGT GTAGCAGTGAAATGCGTAGAGATGTGGAGGAATACCGATGGCGAAAGGCAGCCCCCTGGGACGTCACTGACGCTCATGCACGAAAGCGT GGGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGCCCTAAACGATGTCAACTAGTTGTTGGGGATTCATTTCCTCAGTAACGTAGC TACGCGTGAAGTTGACCGCCTGGGGAGTACGGTCGCAGATAAACTCAG

C0-2 TGCAAGTCGANCGGCAGCGCGGGCTTCGGCCTGGCGGCGAGTGGCGAACGGGTGAGTAATACATCGGAACGTGCCCTGTTGTGGGGGAT AACTAGTCGAAAGATTAGCTAATACCGCATACGACCTGAGGGTGAAAGCGGGGGACCGCAAGGCCTCGCGCAATAGGAGCGGCCGATGT Cupriavidus CTGATTAGCTAGTTGGTGGGGTAAAGGCCTACCAAGGCGACGATCAGTAGCTGGTCTGAGAGGACGATCAGCCACACTGGGACTGAGAC respiraculi ACGGCCCAGACTCCTACGGGAGGCAGCAGTGGGGAATTTTGGACAATGGGGGCAACCCTGATCCAGCAATGCCGCGTGTGTGAAGAAGG CCTTCGGGTTGTAAAGCACTTTTGTCCGGAAAGAAATCGCGCTGGCTAATACCTGGCGTGGATGACGGTACCGGAAGAATAAGCACCGG CTAACTACGTGCCAGCAGCCGCGGTAATACGTAGGGTGCGAGCGTTAATCGGAATTACTGGGCGTAAAGCGTGCGCAGGCGGTTTTGTA AGACAGGCGTGAAATCCCCGAGCTCAACTTGGGAATGGCGCTTGTGACTGCAAGGCTAGAGTATGTCAGAGGGGGGGTAGAATTCCACG TGTAGCAGTGAAATGCGTAGAGATGTGGAGGAATACCGATGGCGAAAGCAGCCCCCTGGGACGTCACTGACGCTCATGCACGAAAGCGT GGGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGCCCTAAACGATGTCAACTAGTTGTTGGGGATTCATTTCCTCAGTAACGTAGC TAACGCGTGAAGTTGACCGCCTGGGGAGTACGGTCGCAAGATTAAACTCAG

C0-3 TGCAAGTCGAGGGGCATCACAGTATAGCAATATATGGGTGGCGACCGGCAAACGGGTGCGGAACACGTACACAACCTTCCGGTGAGTGG GGGATAGCCCAGAGAAATTTGGATTAATACCCCATACTATAATGATCAGGCATCTGGTTATTATCAAAGGCTTCGGCCGCTTATTGATG Sediminibacterium GGTGTGCGTCTGATTAGGTAGTTGGCGGGGTAGAGGCCCACCAAGCCTACGATCAGTAGCTGATGTGAGAGCATGATCAGCCACACGGG sp. CACTGAGACACGGGCCCGACTCCTACGGGAGGCAGCAGTAAGGAATATTGGACAATGGACGCAAGTCTGATCCAGCCATGCCGCGTGAA GGATGACTGCCCTCTGGGTTGTAAACTTCTTTTATAGGGGAAGAAAGTTATCTTTTTTAGGATATTTGACGGTACCCTATGAATAAGCA CCGGCTAACTCCGTGCCAGCAGCCGCGGTAATACGGAGGGTGCAAGCGTTATCCGGATTCACTGGGTTTAAAGGGTGCGTAGGCGGGTA TGTAAGTCAGTGGTGAAATACCGGAGCTTAACTTCGGAACTGCCATTGATACTATATATCTTGAATATTGTGGAGGTAAGCGGAATATG TCATGTAGCGGTGAAATGCTTAGAGATGACATAGAACACCGATTGCGAAGGCAGCTTGCTACGCAAATATTGACGCTGAGGCACGAAAG CGTGGGGATCAAACAGGATTAGATACCCTGGTAGTCCACGCCCTAAACGATGATTACTCGACATTAGCGATACACAGTTAGTGTCTGAG CGAAAGCATTAAGTAATCCACCTGGGAAGTACGACCGCAAGGT

179

Strain and Partial 16S rRNA gene sequences Identification C0-4 TGCAAGTCGAACGGCAGCGCGGGCTTCGGCCTGGCGGCGAGTGGCGAACGGGTGAGTAATACATCGGAACGTGCCCTGTTGTGGGGGAT AACTAGTCGAAAGATTAGCTAATACCGCATACGACCTGAGGGTGAAAGCGGGGGACCGCAAGGCCTCGCGCAATAGGAGCGGCCGATGT Cupriavidus CTGATTAGCTAGTTGGTGGGGTAAAGGCCTACCAAGGCGACGATCAGTAGCTGGTCTGAGAGGACGATCAGCCACACTGGGACTGAGAC respiraculi ACGGCCCAGACTCCTACGGGAGGCAGCAGTGGGGAATTTTGGACAATGGGGGCAACCCTGATCCAGCAATGCCGCGTGTGTGAAGAAGG CCTTCGGGTTGTAAAGCACTTTTGTCCGGAAAGAAATCGCGCTGGCTAATACCTGGCGTGGATGACGGTACCGGAAGAATAAGCACCGG CTAACTACGTGCCAGCAGCCGCGGTAATACGTAGGGTGCGAGCGTTAATCGGAATTACTGGGCGTAAAGCGTGCGCAGGCGGTTTTGTA AGACAGGCGTGAAATCCCCGAGCTCAACTTGGGAATGGCGCTTGTGACTGCAAGGCTAGAGTATGTCAGAGGGGGGTAGAATTCCACGT GTAGCAGTGAAATGCGTAGAGATGTGGAGGAATACCGATGGCGAAGGCAGCCCCCTGGGACGTCACTGACGCTCATGCACGAAAGCGTG GGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGCCCTAACGATGTCAACTAGTTGTTGGGGATTCATTTCCTCAGTAACGTAGCTA ACGCGTGAAGTTGACCGCCTGGGGAGTACGGTCGCAGATTAAACTCAG

C0-5 TGCAAGTCGAACGATGAAGCCCAGCTTGCTGGGTGGATTAGTGGCGAACGGGTGAGTAACACGTGAGTAACCTGCCCTTGACTCTGGGA TAAGCGCTGGAAACGGCGTCTAATACCGGATACGACCTGCCCCGGCATCGGGTGCGGGTGGAAAGTTTTTCGGTCAAGGATGGACTCGC Rathayibacter tritici GGCCTATCAGCTAGTTGGTGAGGTAATGGCTCACCAAGGCGACGACGGGTAGCCGGCCTGAGAGGGTGACCGGCCACACTGGGACTGAG ACACGGCCCAGACTCCTACGGGAGGCAGCAGTGGGGAATATTGCACAATGGGCGAAAGCCTGATGCAGCAACGCCGCGTGGGGGATGAC GGCCTTCGGGTTGTAAACCTCTTTTAGTAGGGAAGAAGGGCTTCGGCTTGACGGTACCTGCAGAAAAAGCACCGGCTAACTACGTGCCA GCAGCCGCGGTAATACGTAGGGTGCAAGCGTTGTCCGGAATTATTGGGCGTAAAGAGCTCGTAGGCGGTTTGTCGCGTCTGCTGTGAAA ACCCGAGGCTCAACCTCGGGCCTGCAGTGGGTACGGGCAGACTAGAGTGCGGTAGGGGAGAATGGAATTCCTGGTGTAGCGGTGGAATG CGCAGATATCAGGAGGAACACCGATGGCGAAAGGCAGTTCTCTGGGCCGTTACTGACGCTGAGGAGCGAAAGCGTGGGGAGCGAACAGG ATTAGATACCCTGGTAGTCCACGCCGTAACGTTGGGAACTAGATGTGGGGACCTTTCCACGGTCTCCGTGTCGCAGCTAACGCATTAGT TCCCCGCCTGGGGAGTACGGCCGCAGGCTAAACTCA

C0-6 TGCAAGTCGAACGCTGAAGCTTGGTGCTTGCACTGGGTGGATGAGTGGCGAACGGGTGAGTAATACGTGAGTAACCTGCCCTTGACTCT GGGATAAGCCTGGGAAACTGGGTCTAATACTGGATACGACATGTCACCGCATGGTGGTGTGTGGAAAGGGTTTTACTGGTTTTGGATGG Kocuria rhizophila GCTCACGGCCTATCAGCTTGTTGGTGGGGTAATGGCTCACCAAGGCGACGACGGGTAGCCGGCCTGAGAGGGTGACCGGCCACACTGGG ACTGAGACACGGCCCAGACTCCTACGGGAGGCAGCAGTGGGGAATATTGCACAATGGGCGAAAGCCTGATGCAGCGACGCCGCGTGAGG GATGACGGCCTTCGGGTTGTAAACCTCTTTCAGCACGGAAGAAGCGAAAGTGACGGTACGTGCAGAAGAAGCGCCGGCTAACTACGTGC CAGCAGCCGCGGTAATACGTAGGGCGCAAGCGTTGTCCGGAATTATTGGGCGTAAAGAGCTCGTAGGCGGTTTGTCGCGTCTGCTGTGA AAGCCCGGGGCTTAACCCCGGGTGTGCAGTGGGTACGGGCAGACTTGAGTGCAGTAGGGGAGACTGGAATTCCTGGTGTAGCGGTGAAA TGCGCAGATATCAGGAGGAACACCGATGGCGAAGGCAGGTCTCTGGGCTGTTACTGACGCTGAGGAGCGAAAGCATGGGGAGCGAACAG GATTAGATACCCTGGTAGTCCATGCCGTAAACGTTGGGCACTAGGTGTGGGGAACATTCCACGTTTTCCGCGCCGTAGCTAACGCATTA AGTGCCCCGCCTGGGGAGTACGGCCGCAAGGCTAA

180

Strain and Partial 16S rRNA gene sequences Identification C0-7 TGCAAGTCGAACGATGAAGCCCAGCTTGCTGGGTGGATTAGTGGCGAACGGGTGAGTAACACGTGAGTAACCTGCCCTTAACTCTGGGA TAAGCCTGGGAAACTGGGTCTAATACCGGATAGGAGCGTCCACCGCATGGTGGGTGTTGGAAAGATTTATCGGTTTTGGATGGACTCGC Micrococcus luteus GGCCTATCAGCTTGTTGGTGAGGTAATGGCTCACCAAGGCGACGACGGGTAGCCGGCCTGAGAGGGTGACCGGCCACACTGGGACTGAG ACACGGCCCAGACTCCTACGGGAGGCAGCAGTGGGGAATATTGCACAATGGGCGCAAGCCTGATGCAGCGACGCCGCGTGAGGGATGAC GGCCTTCGGGTTGTAAACCTCTTTCAGTAGGGAAGAAGCGAAAGTGACGGTACCTGCAGAAGAAGCACCGGCTAACTACGTGCCAGCAG CCGCGGTAATACGTAGGGTGCGAGCGTTATCCGGAATTATTGGGCGTAAAGAGCTCGTAGGCGGTTTGTCGCGTCTGTCGTGAAAGTCC GGGGCTTAACCCCGGATCTGCGGTGGGTACGGGCAGACTAGAGTGCAGTAGGGGAGACTGGAATTCCTGGTGTAGCGGTGGAATGCGCA GATATCAGGAGGAACACCGATGGCGAAGGCAGGTCTCTGGGCTGTAACTGACGCTGAGGAGCGAAAGCATGGGGAGCGAACAGGATTAG ATACCCTGGTAGTCCATGCCGTAAACGTTGGGCACTAGGTGTGGGGACCATTCCACGGTTTCCGCGCCGCAGCTAACGCATTAAGTGCN NGNCTGGGGAGTACGGCCGCAAGGCTA

C0-8 TGCAAGTCGAGCGGGCGTAGCAATACGTCAGCGGCAGACGGGTGAGTAACGCGTGGGAACGTACCTTTTGGTTCGGAACAACACAGGGA AACTTGTGCTAATACCGGATAAGCCCTTACGGGGAAAGATTTATCGCCGAAAGATCGGCCCGCGTCTGATTAGCTAGTTGGTAGGGTAA Bradyrhizobium sp. TGGCCTACCAAGGCGACGATCAGTAGCTGGTCTGAGAGGATGATCAGCCACATTGGGACTGAGACACGGCCCAAACTCCTACGGGAGGC AGCAGTGGGGAATATTGGACAATGGGGGCAACCCTGATCCAGCCATGCCGCGTGAGTGATGAAGGCCCTAGGGTTGTAAAGCTCTTTTG TGCGGGAAGATAATGACGGTACCGCAAGAATAAGCCCCGGCTAACTTCGTGCCAGCAGCCGCGGTAATACGAAGGGGGCTAGCGTTGCT CGGAATCACTGGGCGTAAAGGGTGCGTAGGCGGGTCTTTAAGTCAGGGGTGAAATCCTGGAGCTCAACTCCAGAACTGCCTTTGATACT GAAGATCTTGAGTTCGGGAGAGGTGAGTGGAACTGCGAGTGTAGAGGTGAAATTCGTAGATATTCGCAAGAACACCAGTGGCGAAGGCG GCTCACTGGCCCGATACTGACGCTGANGCACGAAAGCGTGGGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGCCGTAAACGATGA ATGCCAGCCGTTAGTGGGTTTACTCACTAGTGGCGCAGCTAACGCTTTAAGCATTCCGCCTGGGGAGTACGGTCGCAAGATTAAACTCA G

Epost-1 TGCAAGTCGAACGGTAACAGGTCTTCGGATGCTGACGAGTGGCGAACGGGTGAGTAATACATCGGAACGTGCCCGATCGTGGGGGATAA CGAAGCGAAAGCTTTGCTAATACCGCATACGATCTACGGATGAAAGCAGGGGACCGCAAGGCCTTGCGCGAACGGAGCGGCCGATGGCA Acidovorax sp. GATTAGGTAGTTGGTGGGATAAAAGCTTACCAAGCCGACGATCTGTAGCTGGTCTGAGAGGACGACCAGCCACACTGGGACTGAGACAC GGCCCAGACTCCTACGGGAGGCAGCAGTGGGGAATTTTGGACAATGGGCGAAAGCCTGATCCAGCCATGCCGCGTGCAGGATGAAGGCC TTCGGGTTGTAAACTGCTTTTGTACGGAACGAAAAGACTCTGGTTAATACCTGGGGTCCATGACGGTACCGTAAGAATAAGCACCGGCT AACTACGTGCCAGCAGCCGCGGTAATACGTAGGGTGCAAGCGTTAATCGGAATTACTGGGCGTAAAGCGTGCGCAGGCGGTTATATAAG ACAGATGTGAAATCCCCGGGCTCAACCTGGGAACTGCATTTGTGACTGTATAGCTAGAGTACGGCAGAGGGGGATGGAATTCCGCGTGT AGCAGTGAAATGCGTAGATATGCGGAGGAACACCGATGGCGAAGGCAATCCCCTGGGCCTGTACTGACGCTCATGCACGAAAGCGTGGG GAGCAAACAGGATTAGATACCCTGGTAGTCCACGCCCTAAACGATGTCAACTGGTTGTTGGGTCTTCACTGACTCAGTAACGAAGCTAA CGCGTGAAGTTGACCGCCTGGGGAGTACGGCCGCAAGGT

181

Strain and Partial 16S rRNA gene sequences Identification

FMC IPC lab Isolates FH-A TGCAAGTCGAACGGTAACAGGTCTTCGGATGCTGACGAGTGGCGAACGGGTGAGTAATACATCGGAACGTGCCCGATCGTGGGGGATAA CGAAGCGAAAGCTTTGCTAATACCGCATAAGATCTACGGATGAAAGCAGGGGACCGCAAGGCCTTGCGCGAACGGAGCGGCCGATGGCA Acidovorax GATTAGGTAGTTGGTGGGGTAAAAGCTTACCAAGCCGACGATCTGTAGCTGGTCTGAGAGGACGACCAGCCACACTGGGACTGAGACAC temperans GGCCCAGACTCCTACGGGAGGCAGCAGTGGGGAATTTTGGACAATGGGCGAAAGCCTGATCCAGCCATGCCGCGTGCAGGATGAAGGCC TTCGGGTTGTAAACTGCTTTTGTACGGAACGAAAAGACTCTGGTTAATACCTGGGGTCCATGACGGTACCGTAAGAATAAGCACCGGCT AACTACGTGCCAGCAGCCGCGGTAATACGTAGGGTGCAAGCGTTAATCGGAATTACTGGGCGTAAAGCGTGCGCAGGCGGTTATATAAG ACAGATGTGAAATCCCCGGGCTCAACCTGGGAACTGCATTTGTGACTGTATAGCTAGAGTACGGCAGAGGGGGATGGAATTCCGCGTGT AGCAGTGAAATGCGTAGATATGCGGAGGAACACCGATGGCGAAGGCAATCCCCTGGGCCTGTACTGACGCTCATGCACGAAAGCGTGGG GAGCAAACAGGATTAGATACCCTGGTAGTCCACGCCCTAAACGATGTCAACTGGTTGTTGGGTCTTCACTGACTCAGTAACGAAGCTAA CGCGTGAAGTTGACCGCCTGGGGAGTACGGCCGCAAGGTTGAACTCAG

FH-D TGCAAGTCGAACGAAGGCTTCGGCCTTAGTGGCGCACGGGTGCGTAACGCGTGGGAATCTGCCCTTGGGTTCGGAATAACAGTTAGAAA TGACTGCTAATACCGGATGATGTCCCCTTTAGAGATATTGGGGACCAAAGATTTATCGCCCAGGGATGAGCCCGCGTAGGATTAGGTAG Novosphingobium TTGGTGGGGTAAAGGCCTACCAAGCCGACGATCCTTAGCTGGTCTGAGAGGATGATCAGCCACACTGGGACTGAGACACGGCCCAGACT subterraneum CCTACGGGAGGCAGCAGTGGGGAATATTGGACAATGGGGGCAACCCTGATCCAGCAATGCCGCGTGAGTGATGAAGGCCTTCGGGTCGT AAAGCTCTTTTACCAGGGATGATAATGACAGTACCTGGAGAATAAGCTCCGGCTAACTCCGTGCCAGCAGCCGCGGTAATACGGAGGGA GCTAGCGTTGTTCGGAATTACTGGGCGTAAAGCGCGCGTAGGCGGCTACTCAAGTCAGAGGTGAAAGCCCGGGGCTCAACCCCGGAACT GCCTTTGAAACTAGGTGGCTAGAATCTTGGAGAGGTCAGTGGAATTCCGAGTGTAGAGGTGAAATTCGTAGATATTCGGAAGAACACCA GTGGCGAAGGCGACTGACTGGACAAGTATTGACGCTGAGGTGCGAAAGCGTGGGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGC CGTAAACGATGATAACTAGCTGTCCGGGCACATGGTGTCTGGGTGGCGCAGCTAACGCATTAAGTTATCCGCCTGGGGAGTACGGTCGC AAGATTAAAACTCAG

FH- G TGCAAGTCGAACGAAGGCTTCGGCCTTAGTGGCGCACGGGTGCGTAACGCGTGGGAATCTGCCCTTGGGTTCGGAATAACAGCGAGAAA TTGCTGCTAATACCGGATGATGTCGCGAGACCAAAGATTTATCGCCTGAGGATGAGCCCGCGTTGGATTAGGTAGTTGGTGGGGTAAAG Sphingomonas sp. GCCTACCAAGCCGACGATCCATAGCTGGTCTGAGAGGATGATCAGCCACACTGGGACTGAGACACGGCCCAGACTCCTACGGGAGGCAG CAGTGGGGAATATTGGACAATGGGCGAAAGCCTGATCCAGCAATGCCGCGTGAGTGATGAAGGCCTTAGGGTTGTAAAGCTCTTTTACC CGGGATGATAATGACAGTACCGGGAGAATAAGCTCCGGCTAACTCCGTGCCAGCAGCCGCGGTAATACGGAGGGAGCTAGCGTTGTTCG GAATTACTGGGCGTAAAGCGCACGTAGGCGGCTTTGTAAGTCAGGGGTGAAAGCCTGGAGCTCAACTCCAGAACTGCCTTTGAGACTGC ATCGCTTGAATCCGGGAGAGGTGAGTGGAATTCCGAGTGTAGAGGTGAAATTCGTAGATATTCGGAAGAACACCAGTGGCGAAGGCGGC TCACTGGACCGGTATTGACGCTGAGGTGCGAAAGCGTGGGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGCCGTAAACGATGATA ACTAGCTGTCCGGGCACTTGGTGCTTGGGTGGCGCAGCTAACGCATTAAGTTATCCGCCTGGGGAGTACGGCCGCAAGGTTAAACTCA

182

Strain and Partial 16S rRNA gene sequences Identification FH-J TGCAAGTCGAACGAAGGCTTCGGCCTTAGTGGCGCACGGGTGCGTAACGCGTGGGAATCTGCCCTTGGGTTCGGAATAACAGTGAGAAA TTACTGCTAATACCGGATGATGACTTCGGTCCAAAGATTTATCGCCCAAGGATGAGCCCGCGTAAGATTAGCTAGTTGGTGAGGTAAAG Blastomonas GCTCACCAAGGCGACGATCTTTAGCTGGTCTGAGAGGATGATCAGCCACACTGGGACTGAGACACGGCCCAGACTCCTACGGGAGGCAG natatoria CAGTGGGGAATATTGGACAATGGGCGAAAGCCTGATCCAGCAATGCCGCGTGAGTGATGAAGGCCTTAGGGTTGTAAAGCTCTTTTACC (Sphingomonas AGGGATGATAATGACAGTACCTGGAGAATAAGCTCCGGCTAACTCCGTGCCAGCAGCCGCGGTAATACGGAGGGAGCTAGCGTTGTTCG ursincola) GAATTACTGGGCGTAAAGCGCACGTAGGCGGCCATTCAAGTCAGAGGTGAAAGCCCGGGGCTCAACCCCGGAACTGCCTTTGAAACTAG ATGGCTTGAATCTTGGAGAGGCGAGTGGAATTCCGAGTGTAGAGGTGAAATTCGTAGATATTCGGAAGAACACCAGTGGCGAAAGGCGA CTCGCTGGACAAGTATTGACGCTGAGGTGCGAAAGCGTGGGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGCCGTAAACGATGAT AACTAGCTGTCCGGGTTCATGGAACTTGGGTGGCGCAGCTAACGCATTAAGTTATCCGCCTGGGGAGTACGGTCGCAAGATTAAACTCA G

FH-K TGCAAGTCGAACGGGCTTCTTCGGAAGTCAGTGGCAGACGGGTGAGTAACACGTGGGAACGTGCCCTTCGGTTCGGAATAACTCAGGGA AACTTGAGCTAATACCGGATACGCCCTTACGGGGAAAGGTTTACTGCCGAAGGATCGGCCCGCGTCTGATTAGCTTGTTGGTGGGGTAA Methylobacterium CGGCCTACCAAGGCGACGATCAGTAGCTGGTCTGAGAGGATGATCAGCCACACTGGGACTGAGACACGGCCCAGACTCCTACGGGAGGC populi AGCAGTGGGGAATATTGGACAATGGGCGCAAGCCTGATCCAGCCATGCCGCGTGAGTGATGAAGGCCTTAGGGTTGTAAAGCTCTTTTG TCCGGGACGATAATGACGGTACCGGAAGAATAAGCCCCGGCTAACTTCGTGCCAGCAGCCGCGGTAATACGAAGGGGGCTAGCGTTGCT CGGAATCACTGGGCGTAAAGGGCGCGTAGGCGGCCGATTAAGTCGGGGGTGAAAGCCTGTGGCTCAACCACAGAATTGCCTTCGATACT GGTTGGCTTGAGACCGGAAGAGGACAGCGGAACTGCGAGTGTAGAGGTGAAATTCGTAGATATTCGCAAGAACACCAGTGGCGAAGGCG GCTGTCTGGTCCGGTTCTGACGCTGAGGCGCGAAAGCGTGGGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGCCGTAAACGATGA ATGCCAGCCGTTGGCCTGCTTGCAGGTCAGTGGCGCCGCTAACGCATTAAGCATTCCGCCTGGGGAGTACGGTCGCAAGATTA

FH-P TGCAAGTCGAACGAAGGCTTCGGCCTTAGTGGCGCACGGGTGCGTAACGCGTGGGAATCTGCCCTTGGGTTCGGAATAACAGTGAGAAA TTACTGCTAATACCGGATGATGACTTCGGTCCAAAGATTTATCGCCCAAGGATGAGCCCGCGTAAGATTAGCTAGTTGGTGAGGTAAAG Blastomonas GCTCACCAAGGCGACGATCTTTAGCTGGTCTGAGAGGATGATCAGCCACACTGGGACTGAGACACGGCCCAGACTCCTACGGGAGGCAG natatoria CAGTGGGGAATATTGGACAATGGGCGAAAGCCTGATCCAGCAATGCCGCGTGAGTGATGAAGGCCTTAGGGTTGTAAAGCTCTTTTACC (Sphingomonas AGGGATGATAATGACAGTACCTGGAGAATAAGCTCCGGCTAACTCCGTGCCAGCAGCCGCGGTAATACGGAGGGAGCTAGCGTTGTTCG ursincola) GAATTACTGGGCGTAAAGCGCACGTAGGCGGCCATTCAAGTCAGAGGTGAAAGCCCGGGGCTCAACCCCGGAACTGCCTTTGAAACTAG ATGGCTTGAATCTTGGAGAGGCGAGTGGAATTCCGAGTGTAGAGGTGAAATTCGTAGATATTCGGAAGAACACCAGTGGCGAAAGCGAC TCGCTGGACAAGTATTGACGCTGAGGTGCGAAAGCGTGGGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGCCGTAAACGATGATA ACTAGCTGTCCGGGTTCATGGAACTTGGGTGGCGCAGCTAACGCATTAAGTTATCCGCCTGGGGAGTACGGTCGCAAGATTAATACTCA AG

183

Strain and Partial 16S rRNA gene sequences Identification FH-Y TGCAAGTCGAACGAAGGCTTCGGCCTTAGTGGCGCACGGGTGCGTAACGCGTGGGAATCTGCCCTTGGGTTCGGAATAACAGTTAGAAA TGACTGCTAATACCGGATGATGTCCCCTTTAGAGATATTGGGGACCAAAGATTTATCGCCCAGGGATGAGCCCGCGTAGGATTAGGTAG Novosphingobium TTGGTGGGGTAAAGGCCTACCAAGCCGACGATCCTTAGCTGGTCTGAGAGGATGATCAGCCACACTGGGACTGAGACACGGCCCAGACT subterraneum CCTACGGGAGGCAGCAGTGGGGAATATTGGACAATGGGGGCAACCCTGATCCAGCAATGCCGCGTGAGTGATGAAGGCCTTCGGGTCGT AAAGCTCTTTTACCAGGGATGATAATGACAGTACCTGGAGAATAAGCTCCGGCTAACTCCGTGCCAGCAGCCGCGGTAATACGGAGGGA GCTAGCGTTGTTCGGAATTACTGGGCGTAAAGCGCGCGTAGGCGGCTACTCAAGTCAGAGGTGAAAGCCCGGGGCTCAACCCCGGAACT GCCTTTGAAACTAGGTGGCTAGAATCTTGGAGAGGTCAGTGGAATTCCGAGTGTAGAGGTGAAATTCGTAGATATTCGGAAGAACACCA GTGGCGAAGGCGACTGACTGGACAAGTATTGACGCTGAGGTGCGAAAGCGTGGGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGC CGTAAACGATGATAACTAGCTGTCCGGCACATGGTGTCTGGGTGGCGCAGCTAACGCATTAAGTTATCCGCCTGGGGAGTACGGTCGCA GATTAAACTCAG

FH-J1 TGCAAGTCGAGGGGCAGCACAGTATAGCAATATATGGGTGGCGACCGGCAAACGGGTGCGGAACACGTACACAACCTTCCGATAAGAGG GGGATAGCCCAGAGAAATTTGGATTAATACCCCGTAAGATAAAGTTCAGGCATCTGGATTTTATGAGAGGCGCGAGCCGCTTATTGATG Sediminibacterium GGTGTGCGTCTGATTAGGTAGTTGGCGGGGTAAAGGCCCACCAAGCCTACGATCAGTAGCTGATGTGAGAGCATGATCAGCCACACGGG sp CACTGAGACACGGGCCCGACTCCTACGGGAGGCAGCAGTAAGGAATATTGGACAATGGACGCAAGTCTGATCCAGCCATGCCGCGTGAA GGATGAATGCCCTCTGGGTTGTAAACTTCTTTTATAGGGGAAGAAAGATACTTTTTTGAAGGTAGTTGACGGTACCCTATGAATAAGCA CCGGCTAACTCCGTGCCAGCAGCCGCGGTAATACGGAGGGTGCAAGCGTTATCCGGATTTACTGGGTTTAAAGGGTGCGTAGGCGGGCA TGTAAGTCAGTGGTGAAATACCGGAGCTTAACTTCGGAACTGCCATTGATACTATATGTCTTGAATATTGTGGAGGTAAGCGGAATATG TCATGTAGCGGTGAAATGCATAGATATGACATAGAACACCGATTGCGCAGGCAGCTTGCTACACAAATATTGACGCTGAGGCACGAAAG CGTGGGGATCAAACAGGATTAGATACCCTGGTAGTCCACGCCCTAAACGATGATTACTCGACATTAGCGATAAACAGTTAGTGTCTGAG CGAAAGCATTAAGTAATCCACCTGGGAAGTACGACCGCAAGGTTGAAACTCAG

FH-Q1 TGCAAGTCGAGGGGCAGCACAGTATAGCAATATATGGGTGGCGACCGGCAAACGGGTGCGGAACACGTACACAACCTTCCGATAAGAGG GGGATAGCCCAGAGAAATTTGGATTAATACCCCGTAAGATAAAGTTCAGGCATCTGGATTTTATGAGAGGCGCGAGCCGCTTATTGATG Sediminibacterium GGTGTGCGTCTGATTAGGTAGTTGGCGGGGTAAAGGCCCACCAAGCCTACGATCAGTAGCTGATGTGAGAGCATGATCAGCCACACGGG sp CACTGAGACACGGGCCCGACTCCTACGGGAGGCAGCAGTAAGGAATATTGGACAATGGACGCAAGTCTGATCCAGCCATGCCGCGTGAA GGATGAATGCCCTCTGGGTTGTAAACTTCTTTTATAGGGGAAGAAAGATACTTTTTTGAAGGTAGTTGACGGTACCCTATGAATAAGCA CCGGCTAACTCCGTGCCAGCAGCCGCGGTAATACGGAGGGTGCAAGCGTTATCCGGATTTACTGGGTTTAAAGGGTGCGTAGGCGGGCA TGTAAGTCAGTGGTGAAATACCGGAGCTTAACTTCGGAACTGCCATTGATACTATATGTCTTGAATATTGTGGAGGTAAGCGGAATATG TCATGTAGCGGTGAAATGCATAGATATGACATAGAACACCGATTGCGCAGGCAGCTTGCTACACAAATATTGACGCTGAGGCACGAAAG CGTGGGGATCAAACAGGATTAGATACCCTGGTAGTCCACGCCCTAAACGATGATTACTCGACATTAGCGATAAACAGTTAGTGTCTGAG CGAAAGCATTAAGTAATCCACCTGGGAAGTACGACCGCAAGGTTGA

184

Strain and Partial 16S rRNA gene sequences Identification FH-W1 TGCAAGTCGAACGGCAGCACAGGAGAGCTTGCTCTCTGGGTGGCGAGTGGCGGACGGGTGAGGAATACATCGGAATCTACTCTGTCGTG GGGGATAACGTAGGGAAACTTACGCTAATACCGCATACGACCTACGGGTGAAAGCAGGGGATCTTCGGACCTTGCGCGATTGAATGAGC Stenotrophomonas CGATGTCGGATTAGCTAGTTGGCGGGGTAAAGGCCCACCAAGGCGACGATCCGTAGCTGGTCTGAGAGGATGATCAGCCACACTGGAAC maltophilia TGAGACACGGTCCAGACTCCTACGGGAGGCAGCAGTGGGGAATATTGGACAATGGGCGCAAGCCTGATCCAGCCATACCGCGTGGGTGA AGAAGGCCTTCGGGTTGTAAAGCCCTTTTGTTGGGAAAGAAATCCAGCTGGCTAATACCCGGTTGGGATGACGGTACCCAAAGAATAAG CACCGGCTAACTTCGTGCCAGCAGCCGCGGTAATACGAAGGGTGCAAGCGTTACTCGGAATTACTGGGCGTAAAGCGTGCGTAGGTGGT CGTTTAAGTCCGTTGTGAAAGCCCTGGGCTCAACCTGGGAACTGCAGTGGATACTGGGCGACTAGAGTGTGGTAGAGGGTAGCGGAATT CCTGGTGTAGCAGTGAAATGCGTAGAGATCAGGAGGAACATCCATGGCGAAGGCAGCTACCTGGACCAACACTGACACTGAGGCACGAA AGCGTGGGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGCCCTAAACGATGCGAACTGGATGTTGGGTGCAATTTGGCACGCAGTA TCGAAGCTAACGCGTTAAGTTCGCCGCCTGGGGAGTACGGTCGCAAGACTGAAACTCAG

Bio Sci Isolates MWI-1 TGCAAGTCGAGCGGGCCCTTCGGGGTCAGCGGCAGACGGGTGAGTAACGCGTGGGAACGTGCCCTTCGGTTCGGAATAACCCTGGGAAA CTAGGGCTAATACCGGATACGCCCCTTGGGGGAAAGGTTTACTGCCGAAGGATCGGCCCGCGTCTGATTAGCTAGTTGGTGAGGTAACG Methylobacterium GCTCACCAAGGCGACGATCAGTAGCTGGTCTGAGAGGATGATCAGCCACACTGGGACTGAGACACGGCCCAGACTCCTACGGGAGGCAG isbiliense CAGTGGGGAATATTGGACAATGGGCGCAAGCCTGATCCAGCCATGCCGCGTGAGTGATGACGGCCTTAGGGTTGTAAAGCTCTTTTGTC CGGGACGATAATGACGGTACCGGAAGAATAAGCCCCGGCTAACTTCGTGCCAGCAGCCGCGGTAATACGAAGGGGGCTAGCGTTGCTCG GAATCACTGGGCGTAAAGGGCGCGTAGGCGGCTTGCCAAGTCGGGGGTGAAAGCCCGTGGCTCAACCACGGAATGGCCTTCGATACTGG CAGGCTTGAGACCGGAAGAGGACAGCGGAACTGCGAGTGTAGAGGTGAAATTCGTAGATATTCGCAAGAACACCAGTGGCGAAGGCGGC TGTCTGGTCCGGTTCTGACGCTGAGGCGCGAAAGCGTGGGGAGCAAACAGGATTAGATACCCTGGTAGTCCACGCTGTAAACGATGAAT GCTAGCCGTTGGGGTGCATGCACCTCAGTGGCGCCGCTAACGCATTAAGCATTCCGCCTGGGGAGTACGGTCGCAAGATTAA

MWI-2 TGCAAGTCGAACGGAAGGCCCTTCGGGGTACTCGAGTGGCGAACGGGTGAGTAACACGTGGGTGATCTGCCCTGCACTTTGGGATAAGC CTGGGAAACTGGGTCTAATACCGAATATGACCATGCGCCTCCTGGTGTGTGGTGGAAAGCTTTTGCGGTGTGGGATGGGCCCGCGGCCT Mycobacterium sp. ATCAGCTTGTTGGTGGGGTAATGGCCTACCAAGGCGACGACGGGTAGCCGGCCTGAGAGGGTGACCGGCCACACTGGGACTGAGATACG GCCCAGACTCCTACGGGAGGCAGCAGTGGGGAATATTGCACAATGGGCGCAAGCCTGATGCAGCGACGCCGCGTGAGGGATGACGGCCT TCGGGTTGTAAACCTCTTTCAGCACAGACGAAGCGCAAGTGACGGTATGTGCAGAAGAAGGACCGGCCAACTACGTGCCAGCAGCCGCG GTAATACGTAGGGTCCGAGCGTTGTCCGGAATTACTGGGCGTAAAGAGCTCGTAGGTGGTTTGTCGCGTTGTTCGTGAAAACTCACAGC TTAACTGTGGGCGTGCGGGCGATACGGGCAGACTTGAGTACTGCAGGGGAGACTGGAATTCCTGGTGTAGCGGTGGAATGCGCAGATAT CAGGAGGAACACCGGTGGCGAAGGCGGGTCTCTGGGCAGTAACTGACGCTGAGGAGCGAAAGCGTGGGGAGCGAACAGGATTAGATACC CTGGTAGTCCACGCCGTAAACGGTGGGTACTAGGTGTGGGTTTCCTTCCTTGGGATCCGTGCCGTAGCTAACGCATTAAGTACCCCGCC TGGGGAGTACGGCCGCAAGGCTAAACTCA